Stp1123-Eb 19568 PDF

Copyright by ASTM Int'l (all rights reserved); Mon Feb 24 23:07:03 EST 2020
Downloaded/printed by
Universidad Industrial de Santander (Universidad Industrial de Santander) pursuant to License Agreement. No further reproductions authorized.
STP 1123
Stabilization and Solidification of

Hazardous, Radioactive, and
Mixed Wastes, 2nd Volume
T. Michael Gdliam and Carlton C. Wiles
ASTM Publication Code Number (PCN)

04-011230-56
AsTM
1916 Race Street
Philadelphia, PA 19103
Library of Congress Cataloging in Publication Data
ASTM Pubhcat~on Code Number (PCN) 04-011230-56

ISBN 0-8031-1443-5
ISSN 1059-423X
Copyright 9 1992 AMERICANSOCIETYFOR TESTINGAND MATERIALS, Phdadelph~a, PA Pnor edition copy-

righted 1989 by the Amencan Society for Testing and Matenals All rights reserved Th~smaterial may not
be reproduced or copsed, in whole or ~n part, sn any printed, mechanical, electronic, film, or other dlstrs-
button and storage media, w~thout the wntten consent of the pubhsher
Photocopy Rights
Authorization to photocopy items for ~nternal or personal use, or the tnternal or personal use of specific
clients, is granted by the AMERICAN SOCIETY FOR TESTING AND MATERIALS for users registered
w~th the Copyrsght Clearance Center (CCC) Transactional Reporting Service, provided that the base fee
of $2 50 per copy, plus $0 50 per page is paid directly to CCC, 27 Congress St, Salem, MA 01970, (508)
774-3350 For those organizations that have been granted a photocopy Itcense by CCC, a separate sys-
tem of payment has been arranged The fee code for users of the Transactional Reporting Service ~s 0-
8031-1443-5/91 $2 50 § 50
Peer Revmw Policy
Each paper published ~nthis volume was evaluated by three peer reviewers The authors addressed all
of the reviewers' comments to the satisfaction of both the techntcal editor(s) and the ASTM Committee
on Pubhcat~ons
The quality of the papers In this pubhcat~on reflects not only the obwous efforts of the authors and the
technical editor(s), but also the work of these peer reviewers The ASTM Committee on Pubhcat~ons
acknowledges with apprectatton their dedtcatson and contribution to t~me and effort on behalf of ASTM
Pnnted LnBaltimore,MD
January 1992
Foreword
This publication, Stabtllzatlon and Sohdtficatton of Hazardous, Radloactlve, and Mixed

Wastes, 2nd Volume, contains papers presented at the Second International Symposium on
Stabihzatlon/SohdlfiCatlon of Hazardous, Radioactive, and M~xed Wastes, which was held in
Wllhamsburg, Virginia, 29 May to 1 June 1990 The symposium was sponsored by ASTM
Committee D-34 on Waste Disposal, Oak Ridge National Laboratory, U S Environmental
Protection Agency, Hazardous Substance Management Research Center of the New Jersey
Institute of Technology, Environment Canada, and Alberta Enwronmental Centre T
Michael Gllllam, Oak Ridge National Laboratory, was symposium chairman The editors of
this book, T Michael Gllham and Carlton C Wiles, wish to express their appreciation to the
symposium attendees, authors, and ASTM staff who have contributed to this pubhcatlon A
special note of gratitude as due to those lndlwduals who served as peer reviewers for the papers
presented in this publication
Contents
Overview 1X
REGULATORY AND TECHNICAL GUIDANCE
Low-Level Radioactive Waste Disposal in the United S t a t e s - -

RICHARD L BANGARTAND MICHAELTOKAR
Issues Related to the Public and Regulatory Acceptance of Performance Standards

for Wastes Remaining on Hazardous Waste Sites--JOHN J BARICH,III AND
B J MASON 12
Solidification/Stabilization:Is It Always Appropriate?--CARLTON C WILESAND

EDWIN BARTH 18
Guidelines for Evaluation of the Permanence of a Stabilization/Solidification

TechnoIogy--RENGARAJAN SOUNDARARAJAN 33
SPECIALTY WASTES ORGANICS, ASHES, AND RESINS
A Critical Review of the Effectiveness of Stabilization and Solidification of

Hazardous Organic W a s t e s - - R O B E R T E BROWN, BIPENDER S JINDAL, AND
JOSEPH D BULZAN 43
Stabilization/Solidificationof Wastes Containing Volatile Organic Compounds in

Commercial Cementitious Waste Forms--ROGER D SPENCE,
T MICHAEL GILLIAM, IVAN L MORGAN, AND STEVEN C. OSBORNE 61
Solidification and Stabilization of Phenol and Chlorinated Phenol Contaminated

S o i l s - - N A N C Y J SELL, MARK A. REVALL, WILLIAM BENTLEY, AND
THOMAS H McINTOSH 73
Incorporation of Radioactive Ashes into Ceramic-Clay Matrices--A S ALOY,

A S VISHNEVSKII, T I KOLYCHEVA, B A KUZNETSOV, AND
E A. SHASHUKOV 86
Fixation of Residues from Special Hazardous Waste Incinerators for Shallow Land
Disposal--DORIS
Copyright by ASTM Int'l (all rights reserved); ASSMANN AND
Mon Feb 24 23:07:03 BERTHOLD-GUNTER
EST 2020 BRODDA 90
Chemical Fixation of Flue Gas Desulfurization Sludges and Fly A s h - -
P A U L L BISHOP, T I M C K E E N E R , D A V I D C D U S I N G , R U N Z H I G O N G , A N D
C H O N G - L E LI 103
Evaluation of the Impact of Ocean Disposal of Stabilized/Solidified Municipal

Solid Waste Incinerator Ash on Marine SedimentS--NANCY E KINNER,
D I A N A B D U R L I N G , W B E R R Y L Y O N S , D A V I D L GRESS, A N D
M ROBIN COLLINS 119
Performance Testing of Grout-Based Waste Forms for the Solidification of Anion

Exchange Resins--IVAN L MORGANAND WILLIAMD BOSTICK 133
LABORATORY-SCALE LEACHABILITY STUDIES
Determination of Leaching Characteristics of Waste Materials Leading to

Environmental Product Certification--GERARD J DE GROOT AND
H A N S A. V A N D E R S L O O T 149
Release of Toxic Elements from Solidified Wastes: A Mathematical M o d e l - -

F R A N C O MEDICI, C A R L O M E R L I , G I A N C A R L O SCOCCIA, A N D
ROBERTO VOLPE 171
Leach Studies: Influence of Various Parameters on the Leachability of Cesium

from Cemented BWR Evaporator Concentrates--GIORGIO DE ANGELIS,
ANTONIA MARCHETTI, AND STEPHANIA BALZAMO 182
Effects of Selected Waste Constituents on Solidified/Stabilized Waste

LeachabilityhLARRY w J O N E S , R M A R K B R I C K A , A N D
M JOHN CULLINANE, JR 193
Short-Term Leaching Behavior of Co-60, Sr-85, and Cs-137 from Cemented Ion-
Exchange Resin Waste F o r m s - - J A E - O W A N LEE, K Y O N G - W O N H A N , A N D
LEO P BUCKLEY 204
Effect of Hydrating Water on the Physical Characteristics and the Diffusion

Release of Cesium Nitrate Immobilized in Cement--EDMONDO ZAMORANI
AND GIORGIO SERRINI 217
A Model of the Attack of Pure Water or Undersaturated Lime Solutions on

Cement--MICHEL BUIL, E L I A N E R E V E R T E G A T , A N D J E A N O L I V E R 227
Long-Term Leaching Performance for Commercially Stabilized W a s t e - -

K I M B E R L Y J P E R R Y , N E I L E, P R A N G E , A N D W I L L I A M F G A R V E Y 242
LABORATORY-SCALEPROCESS DEVELOPMENT
Application of Radioactive Waste Solidification Methods to Hazardous Wastes in

Italy--ALDO DONATOAND GIUSEPPEA RICCI 255
Stabilization and In Situ Management of Radioactive Contaminated Sediments of
the Port Hope H a r b o r - - G E O R G E M DOLINAR, R W DOUG KILLEY, AND
KEN E PHILIPOSE 266
Mixture Design Optimization Considering Three Disposal Options for Solidified

Heavy Metal SIudges--HANG-SIKSHIN, JONG-OH KIM, JA-KONGKOO, AND
EUNG-BAISHIN 283
Solidification of Dried Borate Waste with PoIyester--HWAN-YOUNG KIM,

K I - H O N G K I M , H U N - H W E E P A R K , A N D I N - S U K SUH 297
Laboratory Evaluation of Stabilization/Solidification Technology for Reducing the

Mobility of Heavy Metals in New Bedford Harbor Superfund Site
Sediment--TOMMY E M Y E R S A N D M A R K E Z A P P I 304
Cement Conditioning of Waste Materials and Polluted Soil Using the GEODUR
ProcesS--KNUD BRODERSEN, OLE HJELMAR, AND SVEND MORTENSEN 320
Cementation of Borate Waste by Adding Slaked Lime--JOON-HYUNG KIM,

H W A N - Y O U N G K I M , H U N - H W E E P A R K , A N D I N - S U K SUH 338
The Development of Cement Formulations for Radioactive Waste Encapsulation--

DAVID J LEE AND ANDREWFENTON 348
TEST METHOD DEVELOPMENT
Generalized Acid Neutralization Capacity Test--JERRY ISENBURGAND

MICHAEL MOORE 361
Methods of Evaluating Explosive Reactivity of Explosive-Contaminated Solid

Waste Substances--T s B A J P A Y E E A N D R I C H A R D J M A I N I E R O 378
An Assessment of Short-Term Penetration Testing as an Indicator of Strength

Development in Solidified/StabilizedHazardous W a s t e - -
M JOHN CULLINANE, JR AND LARRY W JONES 385
Development of an In-Line Grout Meter for Improved Quality Control--

GUILLERMO D DEL CUL AND T MICHAEL GILLIAM 394
LARGE-SCALE EVALUATION OR DEMONSTRATION
Summary of Solidification/StabilizationSITE Demonstrations at Uncontrolled

Hazardous Waste SiteS--EDWIN F. BARTH 409
Field Demonstration of In Situ Grouting of Radioactive Solid Waste Burial

Trenches with Polyacrylamide--BRIANP SPALDINGAND
THOMAS A FONTAINE 415
Stabilization/Solidification of Metal-Contaminated Soils: Two Case Histories--
J A M E S H K L E P P E , M A R K T. O T T E N , A N D J O H N T F I N N 426
Dynamic Compaction of Nuclear Waste Deposits--ROBERT G LUKASAND

CLIFF SCHEXNAYDER 440
Contributions of Lysimeter Data to the Development of Site Specific Performance

Assessment P l a n s - - R O B E R T D R O G E R S , J O H N W M c C O N N E L L , J R . ,
JULIE D JASTROW, AND DIANA S WICKLIFF 448
Laboratory Treatment and Field Demonstration Lessons Learned from a Waste

Stabilization Project--JEFFREY L MEANS, J E F F E R Y C H E A T H , A N D
W I L L I A M J. M c L A U G H L I N 466
Performance Appraisal of Full-Size Radioactive Waste Packages--J c NOMINE

AND A BILLON 479
Indexes 495
Overview
Stabilization and sohdlfication (S/S) technologies have been used for decades as a final treat-
ment step prior to the disposal of both radioactive and chemically hazardous wastes Stablh-
zation refers to the alteration of waste contaminants to a more chemically stable form (1 e , a
form which results m a greater difficulty of contaminant release to the environment), thereby
resulting in a more environmentally acceptable waste form Solidification refers to the physical
alteration of the waste to restrict water release from or access to the waste and thereby resulting
in a more environmentally acceptable waste form Typically, stabilization processes also
involve some form of physical solidification
This Special Technical Publication contains 39 peer-reviewed papers out of the 70 that were
presented at the Second International Symposium on Stabdlzatlon/Sohdificatlon of Hazard-
ous, Radioactive, and Mixed Wastes, held in Wllhamsburg, Virginia, 29 May-1 June 1990.
This symposium was sponsored by ASTM Committee D-34 on Waste Disposal, Oak Ridge
National Laboratory, Risk Reduction Engineering Laboratory, Office of Research and Devel-
opment of the U.S. Environmental Protection Agency (EPA), Hazardous Substance Manage-
ment Research Center of the New Jersey Institute of Technology, Environment Canada, and
Alberta Environmental Centre Cooperating organizations include the Commission of the
European Communities (CEC), Brussels, Belgmm, ENEA (Comltato Nazlonale per la Ricerca
e per lo Svlluppo dell'Energia Nucleare e delle Energ~c Alternative), Rome, Italy, Netherlands
Energy Research Foundation (ECN), Petten, The Netherlands, Imperial College of Science,
Technology & Medicine, University of London, London, England; VKI (Water Quality Insti-
tute), Horsholm, Denmark, Studiecentrum voor Kernergie/Centre d'etude de l'Energie
Nucleaire (SCK/CEN), Mol, Belgium; Korea Advanced Institute of Science and Technology
(KAIST), Seoul, Korea, KAERI (Korea Atomic Energy Research Institute, Chung-Nam,
Korea, Agence Natlonale pour la Recuperation et l'Ehmlnation des Dechets-Les Transfor-
meurs, Angers, France; and RISO National Laboratory, Roskllde, Denmark
This symposium series (scheduled once approximately every three years) provides a forum
for technical exchange between researchers working with S/S technologies from both the low-
level radioactive and chemically hazardous waste communities Although the two scientific
communities are faced with similar problems and basically work with the same technologies,
this symposium series presents a unique forum for technical exchange between the two com-
munities. This meeting is the second in the series Papers from the first meeting in the series
were published in STP 1033, Envlronmental Aspects of Stabth zatton and Sohd~catton of Haz-
ardous and Radtoactlve Wastes
Land disposal restrictions are becoming ever increasingly stringent, driven by technical, reg-
ulatory, and pohtlcal considerations. To the largest extent practicable, alternatives to land dis-
posal are desirable, such as waste minimization, recycling, and destruction (e g , incineration)
In many instances, however, these alternatives are unreahstlc due to the physical nature or
location of the waste, the type and concentration of contaminants that it contains, or technical
and economic issues. In such cases, S/S technology is a viable technical option which has his-
torically proven to be cost effective
A wide variety of both radioactive and chemically hazardous wastes are amenable to S/S
These include liquids, sludges, filter cakes, contaminated soils, and ash S/S technologies are
IX
X OVERVIEW
often used with chemically hazardous wastes contaminated with metals (e g , lead, nickel,
chromium) and are routinely used with low-level radioactive wastes of all types Clearly, S/S
is not the preferred technology for orgamc-nch wastes, but research is increasingly focusing on
its application to wastes containing trace levels of organic contaminants
S/S technologies are often classified on the basis of the principal additives used to obtain a
solid matrix Various systems based on organic and inorganic additives are listed below
Inorganic Based Systems Organic Based Systems
portland cement b~tumen

soluble sdlcates-cement urea formaldehyde
pozzolan-hme polybutad~ene
pozzolan-cement polyester
clay-cement epoxy
gypsum polyethylene
The specific technology used is based on several factors including volume Increase, perfor-
mance requirements, economics, logistic constraints, and material availability
Inorganic-based systems may involve both stabilization and solidification They generally
utilize a hydraulic binder which reacts with water to form a solid product Portland cement is
the most commonly used binder, Cement-based systems are widely used due to the availablhty
of raw materials, compatiblhty with existing process equipment, and adaptability to a variety
of process configurations and waste-form performance requirements Pure cement systems are
becoming increasingly rare Additives such as fly ash, granulated blast furnace slag, and spe-
cialty clays are used to improve waste-form performance and to minimize the cost Compared
with a pure cement system, these additives can reduce the resulting volume Increase, reduce
porosity (and, thus, reduce leachability), and increase compressive strength
Organic-based systems consist of encapsulating the waste in a matrix that is essentially
impervious to water As such, they are designated as sohdlfiCatlon processes Many of these
systems require water removal from the waste prior to solidification and thus result in an over-
all volume decrease Products from these systems require a container to provide the necessary
structural integrity to support overburden pressure Use of such systems has been hmlted to
special types of high-hazard, relatively low-volume, radioactive wastes due to the increased
costs generally associated with these systems
The scientific community has been focusing attention on understanding and predicting the
long-term containment prospects of waste treated with this technology Consequently, the
majority of papers presented m this pubhcatlon address to some degree one or both of the two
principal issues associated with long-term containment, leachability and durability Leach-
ability refers to the release of contaminants from the waste form upon exposure to an aqueous
media Durablhty refers to the ability of the waste form to maintain its structural integrity
upon exposure to expected environmental conditions
Attendees from 14 countries participated in this meeting, and the papers presented in this
publication represent a cross section of research activities being conducted in this field world-
wide The papers are grouped into six sections to deal with (1) Regulatory and Technical
Guidance; (2) Speciality Wastes Organics, Ashes, and Resins, (3) Laboratory-Scale Leach-
ability Studies, (4) Laboratory-Scale Process Development, (5) Test Method Development,
and (6) Large-Scale Evaluation or Demonstration
Significant overlap among the papers is evident For example, papers grouped under Lab-
oratory-Scale Process Development may well contain information pertinent to Laboratory-
OVERVIEW Xl
Scale Leachabihty Studies This is to be expected as the papers are grouped pnmarlly based on
the apphcation of the data and conclusions as presented by the authors rather than on the data
themselves Moreover, the papers were purposely not grouped according to waste category
(hazardous, radioactive, and mixed) The purpose of this meeting was to provide information
exchange between the vanous technical communities The presentations clearly showed that
materials, processes, performance requirements, methods of experimentation, and subse-
quent data analyses were, for the most part, waste-category independent Thus, only by read-
ing papers from all represented technical communities can one get a true understanding of
what this technology can and cannot do
Regulatory and Technical Guidance

When contemplating the use of S/S technology for a particular application, the first ques-
tions asked are" What are the pertinent regulatory requirements 9 Has this technology been
used successfully on similar wastes 9 What test methods can be used to assess the effectiveness
of this technology 9 The four papers presented in this section deal directly with these questions
The first two papers address pertinent regulatory issues, the third addresses previous successes
and failures; and the fourth addresses available examination techniques
Specialty Wastes: Organics, Ashes, and Resins

S/S technology is used routinely in the nuclear industry for the treatment of low-level radio-
active wastes In addition, the EPA has specified S/S technology as the best demonstrated and
available technology for a variety of wastes containing metals, such as cadmium, chromium,
and lead Clearly, one of the roles of the scientific community is to define the limits of existing
technology and expand those limits where appropnate Indeed, the majority of the papers pre-
sented in this publication could be viewed as contributing to this endeavor Eight papers are
presented in this section which directly address the application of this technology to three
waste types of particular interest during the meeting organics, combustion residues (e g ,
ashes), and resins The first three papers deal with the issue of organic-containing wastes and
range from a review of existing literature to on-going experamental studies Four of the papers
deal with combustion residues such as incinerator ash, fly ash, and/or desulfunzation sludges
One paper deals with anion exchange resins
Laboratory-Scale Leachability Studies

Understanding the mechanisms involved in the leaching of a waste form is crucial to pre-
dicting its long-term performance m the environment At present, the prediction of long-term
behavior must rely on leaching studies and their associated leach-model development due to
(1) a lack of understanding of the chemistry and morphology of the waste form, resulting in a
limited ability to quantitatively predict its long-term performance and (2) both the lack of and
difficulty in mterpreting field-scale leach data, resulting in limited venficatlon of laboratory-
scale leach studies and derived models Eight papers are presented in this section which deal
with the development of leach models and/or determinmg parameters that affect waste-form
leachability. Collectively, these papers indicate the similarity between the performance of
waste forms containing hazardous and radioactive wastes The editors beheve that this subject
more than any other shows the benefits and need for information exchange between the tech-
nical communities
xll OVERVIEW
Laboratory-Scale Process Development

Numerous research and development efforts are being expended m formulating stabfl~za-
Uon and sohdlfiCatlon processes The e~ght papers presented m this section provide an excel-
lent cross section of these development efforts The majority of the papers deal w~th cement-
based systems (the most commonly used) and clearly show the benefits of the use of add~tlves
such as fly ash, granulated blast furnace slag, and slaked hme Four of the papers address alter-
nahves to the cement-based systems including compaction ( a form ofsohdlficatmn), sodmm
silicates, polyesters, and urea formaldehyde
Test Method Development

Test methods are useful tools m evaluating S/S technologies Appropriate test methods have
the potentml to increase one's knowledge of the process or product, decrease the t~me needed
to acqmre this knowledge, or ~mprove the product/process The four papers presented m this
section address all of these potentml uses and range from laboratory-scale apphcatmn to
~mproved process control
Large-Scale Evaluation or Demonstration

The papers presented m this pubhcatlon are predominantly related to laboratory-scale stud-
les, which is an accurate representauon of the level of effort apphed m the scientific commu-
mty Laboratory-scale studies are dominant for a variety of reasons including the ability to
control variables of interest, which facflttates the determination of cause-and-effect relat~on-
ships It is this control which necessitates the need for large-scale evaluaUon or demonstration
Large-scale studtes are needed to determine performance under less-controlled condmons that
are inherent to field-scale operations, such as variability of the waste feed (both chemical and
physical), vanablhty xn ablhty to control process parameters, and vanablhty m environmental
exposure conditions on the product The seven papers presented m this sectmn address a van-
ety of S/S techmques ranging from dynamtc compaction to m s~tu stabfllzahon Two of the
papers provide case h~story studies which md~cate the mterrelat~onshlp between laboratory
and full-scale studies The first paper m this section provides a summary of demonstrations
conducted under the EPA's Superfund Innovatxve Technology Evaluation Program and can
serve as a grade to reformation available on this subject from other sources
T. Michael Gilham
Oak Ridge National Laboratory, Oak Ridge,
TN 37831-7273,symposmm chairman and
editor
Carlton C. Wiles
Environmental Protection Agency, Cincin-
nati, OH, editor
Regulatory and Technical Guidance
Richard L. Bangart Land Michael Tokar z
Low-Level Radioactive Waste Disposal in the

United States
REFERENCE: Bangart, R L and Tokar, M, "Low-Level Radioactive Waste Dtsposal in the
United States," Stabdtzatlon and Sohdtficatzon of Hazardous, Radtoacttve, and Mtxed Wastes,
2nd Volume, ASTMSTP 1123, T M Gflham and C C Wries, Eds, American Sooety for Test-
mg and Materials, Phdadelphla, 1992, pp 3-11
ABSTRACt. In this paper, we &scuss the U S Nuclear Regulatory Commlssmn's (NRC)

reqmrements for structural stablhty for Classes B and C low-level ra&oactlve wastes (LLW), as
expressed m Part 61 to Title 10 of the Code of Federal Regulatmns ( 10 CFR Part 61 ) Also dis-
cussed ~sthe gmdance developed by the NRC staff for the staNhzat~on methods of major current
interest These include the use of (1) sohdlficatlon/stablllzatmn media, (2) h~gh integrity con-
tamers, and (3) engineered alternatives, such as below-ground vaults and earth-mounded con-
crete bunkers, for disposal of LLW
KEY WORDS. low-level radioactive waste, cement, bitumen, stabdlzatlon, waste form, engi-
neered alternatives, high-integrity containers
Basic Concepts
Disposal of low-level radioactive waste (LLW) is regulated in the United States under the
provisions o f U S Nuclear Regulatory Commission (NRC) regulation, Part 61 to Title 10 of
the Code of Federal Regulations ( 10 C F R Part 61 ) [ 1 ] Part 61 establishes the procedures, cri-
teria, and the terms and conditions upon which the Commission issues licenses for the disposal
of radioactive wastes containing byproduct, source, and special nuclear matenal received from
other persons Section 61 55 of Part 61 establishes three categories or classes of wastes" namely,
Classes A, B, and C, in a generally ascending order with regard to degree of hazard (based upon
type and concentration) of radlonuchdes All three classes of waste are required to meet certain
m i n i m u m requirements regarding packaging, treatment of hqulds, avoidance of explosive,
pyrophonc, or flammable materials, and so on In addition, Classes B and C wastes (and Class
A wastes that are comlngled with Classes B or C wastes) are required to be structurally stable
The reqmred structural stability may be provided by" (1) the waste form itself (as in the case
of an activated metal component), (2) processing the waste to a stable waste form (for example,
by mixing and solidifying the waste with a cementltiOUS material), or (3) placing the waste in
a suitable disposal container or structure This paper will concentrate on the opUons available
under Part 61 for stabilizing low-level radioactive wastes and the actions bemg taken by the
N R C to ensure that these wastes possess the long-term stability required by the regulation
1Director, Division of Waste Management and Decommlssmmng, Office of Nuclear Material Safety
and Safeguards, U S Nuclear Regulatory Commission, Washington, DC 20555
2 Sectmn leader, Englneenng Section, Technical Branch, Dlvlslon of Waste Management and Decom-
missioning, Office of Nuclear Material Safety and Safeguards, U S Nuclear Regulatory Commission,
Washington, DC 20555
3
Copyright9 1992 by ASTM International www.astm.org

4 HAZARDOUSWASTES
Structural Stability
The importance of long-term stability in the disposal of low-level radioactive waste becomes
apparent in the reading of Part 61 As stated in Section 61 7, titled Concepts, "A cornerstone
of the system is stability--stability of the waste and the disposal site so that once emplaced and
covered, the access of water to the waste can be minimized Migration ofradlonuchdes is thus
minimized, long-term active maintenance can be avoided, and potential exposures to intrud-
ers r e d u c e d . . . To the extent that it 1s practicable, Class B and C waste forms or containers
should be designed to be stable, i e , maintain gross physical properties and identity, over 300
y e a r s " Further explanation of the need to provide stability for the waste is furnished in 10
CFR 61.56(b) [ 1 ], which states, "Stability is intended to ensure that the waste does not struc-
turally degrade and affect overall stability of the site through slumping, collapse, or other fail-
ure of the disposal unit and thereby lead to water infiltration Stability is also a factor in hm-
iting exposure to an inadvertent intruder, since it provides a recognizable and nondisperslble
w a s t e " The concern about loss of stability of the waste form and site (and the concomitant
slumping and collapse of the disposal unit that could lead to failure of the trench cap, increased
water infiltration, and subsequent leaching of radionuchdes from the waste) stemmed from
experience at some pre-Part 61 disposal sites where such slumping and subsidence had been
encountered Consequently, one of the performance objectives in Part 61 (Section 61.44)
states that "The disposal facility must be sited, designed, used, operated, and closed to achieve
long-term stability of the disposal site and to eliminate to the extent practicable the need for
ongoing active maintenance of the disposal site following closure so that only surveillance,
monitoring, or minor custodial care are r e q u i r e d "
Waste Form Solidification and High-Integrity Containers

Techmcal Position on Waste Form
Background--Although 10 CFR Part 61 provides the basic licensing requirements for low-
level radioactive waste (LLW) structural stability, the regulation does not include detail on
how those requirements can or should be met, That type of guidance is provided in a "Tech-
nical Posmon on Waste F o r m " [2], which was formulated and issued (as "Rev 0") in 1983
For solidified waste forms, the tests involve subjecting the waste specimens to conditions of
compression, irradiation, blodegradatxon, leaching, immersion, and thermal cycling. Most of
the tests, which were selected for their relative slmphclty and reproducibility, are based on
American Society for Testing and Materials (ASTM) or American Nuclear Society (ANS)
standard methods of test [3-9] that were originally developed for specific nonradioactive
material applications Though not explicitly stated in the technical position (TP), these meth-
ods of test are intended to provide confidence, by exposing specimens to relatively short-term
(minutes to weeks) conditions, that low-level radioactive waste forms will have the desired
long-term (over 300-year) structural stability Significantly, there is a major difference in the
periods of time allotted for the tests and the period of time of concern for L L W disposal There-
fore, the test conditions cannot match or duplicate the exact conditions that might actually
exist in the disposal facility at the time of disposal or in the future following placement of the
waste in the facility For example, the irradiation test calls for the specimens to be exposed to
a m i n i m u m of 10E + 6 grays (10E + 8 rads), which is the m a x i m u m level of exposure
expected for the waste forms after (300 years o0 disposal. This test requires the specimens to
be exposed to a much higher gamma flux than would actually be encountered under real ser-
vice conditions Thus, some of the TP tests can be considered to be accelerated tests, while in
a more fundamental sense they are actually screening tests that are used to separate out mate-
rlal formulations
Copyright that
by ASTM Int'l (all rights do not
reserved); provide
Mon Feb sufficient
24 23:07:03 EST 2020 assurance of long-term structural stability
BANGART AND TOKAR ON WASTE DISPOSAL 5
Test Parameters--The 1983 TP on waste form deals with the type of short-term testing that
should be performed to demonstrate long-term stability as well as the acceptance cntena for
the tests As indicated in the TP, there are eight types of tests or test conditions for sohdlfied
waste forms called out in the 1983 TP The pnncipal parameter used as an acceptance cntenon
for most of these tests is compressive strength The compressive strength cntenon and the tests
are related to Part 61 through the statement [in Section 61 56(b) Ref 1] that "a structurally
stable waste form will generally maintain its physical dimensions and its form, under expected
disposal conditions, such as weight of overburden and compaction equipment, the presence
of moisture [a rationale for immersion and leaching tests] and microbial activity [a rationale
for blodegradation tests] and internal factors such as radiation effects [a rationale for radiation
stability tests] and chemical changes " I n the 1983 TP, a cover matenal density of 1920 kg/m 3
( 120 lbs/ft 3) is assumed, which yields a pressure of approximately 259 kPa (37 5 psi) at a bunal
depth of 13 72 m (45 It) (the then m a x i m u m burial depth at the Hanford, WA burial site)
Taking into consideration potential additional loads from trench compaction equipment,
waste contents, and so on, the compressive strength c n t e n o n was set at 345 kPa (50 psi), which
was raised to 414 kPa (60 psi) when the depth of the trenches at Hanford was increased to 16 76
m (55 ft) Thus, the compressive strength cntenon was not established as a result of a direct
correlation of an lntnnsic matenal property to long-term structural stability, but was instead
intended to accommodate the environmental or in SltU loads at the bottom of a disposal
trench F o r certain types o f sohdlfication media, (for example, portland cement or vinyl ester
styrene), which typically have (in the unadulterated form) compressive strengths on the order
of tens of thousands of kPa (several thousand psi), a 414-kPa (60-psi) compressive strength
c n t e n o n does not appear to have a strong correlation to long-term structural stablhty There-
fore, for this and other reasons (discussed below), the NRC staff is in the process of clanfylng
the acceptance criterion for cement-stabilized waste forms
Toptcal Report Reviews

As noted earlier, the purpose of the TP of waste form is to provide guidance on an acceptable
approach for demonstrating compliance with 10 CFR Part 61 requirements for LLW-form
structural stability In accordance with the current regulatory procedure for qualifying specific
waste form solidification media and hlgh-mtegnty containers (HICs), NRC performs technical
reviews of topical reports (TRs) that describe the charactenstlcs and design of the solidified
waste forms and HICs and the results of the qualification tests Under an agreement reached
with the sited states in 1983, NRC provides a centralized generic review, with input from the
states The results of the review are intended to be applicable to all the disposal sites The dis-
posal site agreement state authorities implement the NRC staffs technical decisions regarding
the adequacy o f the sohdificatlon medium or container
The current status of the topical report reviews is shown in Table 1 As Indicated, 34 topical
reports have been submitted since the beginning of the program Of the seven TRs that have
received approval, three deal with sohdlfiCatlon media and four concern high-integrity con-
tamers. Of the thirteen reports that are under review, eight deal with the use of cement as a
sohdlfication/stablllzation medium, while no cement stabilization reports have yet received
an approval The reason for this situation is that problems have been encountered with cement
stabilization in field applications as well as in laboratory testing The sited states and NRC have
agreed, however, that continued use of cement as a solidification/stabilization medium is
acceptable d u n n g this review period. What should also be noted is that NRC has endorsed the
Department of Energy's cement-stablhzed waste forum for decontaminated supernatent waste
Copyright
at the by ASTM
WestInt'lValley
(all rightsDemonstration
reserved); Mon Feb 24 23:07:03 EST 2020
ProJect.
6 HAZARDOUSWASTES
TABLE l--Toptcal report revww status summary, sohdtfied waste forms and htgh-mtegrtty containers,
30 May 1990, Office of Nuclear Materzal Safety and Safeguards, U S Nuclear Regulatory
Commission
Vendor Docket No Type Disposition
Waste Chem WM-90 Sohdlfication (bitumen) Approved

General Elec WM-88 SohdlfiCatlon (polymer) Approved
DOW WM-82 Solidification (polymer) Approved
Chlch~bu WM-81 HIC (polyethylene impregnated/concrete) Approved
(Rev 2 1)
Nuclear Pkg WM-45 HIC (ferrahum/FL-50) Approved
Nuclear Pkg WM-85 HIC (ferrahum/famdy) Approved
LN Technologms WM-93 HIC (stainless/polyethylene) Approved
Chem-Nuclear WM- 18 HIC (polyethylene) Not Approved
Westinghouse WM-80 HIC (polyethylene) Not Approved
TFC Nuclear WM-76 HIC (polyethylene) Not Approved
U S Gypsum WM-51 Sohdxficatlon (gypsum) Not Approved
U S Ecology WM-91 Solidification (bitumen) Discontinued
VIKEM WM- 13 Sohdafication/oll (cement) Discontinued
Stock WM-92 Solidification (cement) Discontinued
Nuclear Pkg WM-71 Solld/Encap (cement/gypsum) Withdrawn
Chem-Nuclear WM- 19 Solidification (cement) Withdrawn
Chem-Nuclear WM-96 Solidification (cement) Withdrawn
Westinghouse WM-79 Solidification (SG-95) Withdrawn
Nuclear Pkg WM-87 HIC (316-staxnless/SDS) Withdrawn
LN Tech WM-57 HIC (polyethylene) Withdrawn
Chem-Nuclear WM-47 HIC (fiberglass/polyethylene) Withdrawn
Chem-Nuclear WM- 101 Solidification (cement # 1) Under review
Chem-Nuclear WM-97 Solidification (cement #2) Under review
Chem-Nuclear WM-97 Solidification (cement #2) Under review
(Rev 1)
Chem-Nuclear WM-98 Sohdlfication (cement # 1) Under review
LN Tech WM-20 Sohdification (cement) Under review
LN Tech WM-99 Sohdificatlon (cement/decon) " Under review
Westinghouse WM-46 Solidification (cement) Under review
U S Ecology WM-100 Solidification (NSI bitumen) Under review
Bondico WM-94 HIC (fiberglass/polyethylene) Under review
Babcock & Wilcox WM-95 HIC (coated carbon steel) Under review
U S Ecology WM-102 Solidification (LLW bitumen) Under review
Pacific Nuclear WM- 103 HIC (enviroglass) Under review
JGC Corp WM-104 Solidification (cement) Under review
Waste Soh&ficatlon and H I C Problem Areas--Since the TP on waste form was developed
in 1983, there has been considerable field experience obtained and research conducted on
waste sohdlficatlon media and HICs This reformation has been factored into and in some
cases has been inltmted by the topical report reviews The most notable issues have been
encountered in the use of high-density polyethylene (HDPE) containers, asphalt-solidified
(bituminized) wastes, and cement-sohdlfied wastes These problems and their resolution are
addressed below
H D P E containers Before the promulgation and Implementation of 10 CFR Part 61 (in

1982), H D P E containers were being accepted for disposal at the Barnwell, SC low-level waste
disposal facdlty Beginning m 1979, the state of South Carolina Issued certificates o f c o m p h -
ance for H D P E containers indicating that the containers met South Carohna's regulatory
requirements
Copyright for
by ASTM Int'l (all high-integrity
rights containers
reserved); Mon Feb 24 Cerhficates of comphance for H D P E containers
23:07:03 EST 2020
were issued to HDPE container vendors as well as to nuclear reactor utlhty hcensees With the
implementation of 10 CFR Part 61, the state of South Carohna and NRC agreed that the con-
tamers would continue to be accepted for disposal while NRC reviewed vendor-submitted top-
~cal reports describing the design, testing, and analys~s of the containers
As part of the topical report review effort, NRC contracted with Brookhaven National Lab-
oratory (BNL) to analyze existing data on creep of polyethylene and to develop a model and
criteria that could be used to evaluate the structural stablhty of the containers Brookhaven
recommended [10] that the containers be shown to be able to resist buckhng, to not undergo
tertiary creep, and to not exceed allowable membrane stresses Prehmmary calculations, using
the BNL model, indicated that large HDPE containers might not satisfy the criteria The ven-
dors were told the results of the BNL studies and were requested to show via (1) analyses, (2)
testing, (3) admlmstratlve procedures, and/or (4) redesign that their containers could meet the
proposed acceptance cnterla The vendors' reformation regarding the structural stablhty of the
container designs was reviewed with the assistance of a staff consultant from Brown Univer-
sity The report [11] containing the consultant's findings outhned several technical issues
involving the use of HDPE for 300-year structural stabdlty purposes. The ldennfied problem
areas included (1) a ductde/bnttle transmon (extending over decades of time) that could lead
to crack m m a n o n and growth, (2) creep, (3) buckhng (aggravated by creep), and (4) lrradm-
non-Induced embnttlement After further consultation with the states, the vendors and their
consultants, the Advisory Committee on Nuclear Waste (ACNW), and others regardmg the
techmcal issues, the staff concluded [12] that the HDPE containers would not meet the
reqmrements for 300-year structural stablhty that are put forth in 10 CFR Part 61 Since that
decision was reached, regulatory authontxes m the current s~ted states of South Carohna and
Washington have decided to permit the disposal of HDPE containers at their low-level radio-
active disposal faclllnes by placing the containers in overpacks constructed of reinforced con-
crete In this approach the thick-walled concrete overpacks are intended to provide the long-
term structural stablhty required by 10 CFR Part 61
Bitumlmzed waste The NRC has recewed four topical reports describing the quahficatlon
testing of bituminized waste forms The pnnclpal techmcal issue m the use ofb~tumen to sta-
blhze L L W waste is that, as a viscoelastic material, bitumen continuously deforms under load
Of the two mare types of bitumen that have been proposed for stablhzanon purposes, namely,
distilled versus oxxd~zed bitumen, the latter possesses generally higher strength and wscos~ty
Therefore, during the review of one of the b~tumlnlzed waste reports (Docket No 91, see Table
1) NRC staff concluded that the bitumen formulanons described m the report could not be
approved for low-level waste disposal because the type of bitumen used (a "distilled" version)
did not possess sufficient strength As a result, the review of the topical report was d~scontlnued
and the report was returned to the vendor Another report (Docket No 90) was approved,
however, because the bitumen used for the formulations descnbed m that report was of an
oxidized type that produced bltumlmzed waste forms with adequate strength (414 kPa at 10%
deformation) when used m conjunction with an administrative backfill procedure Both of the
two b~tumen reports currently under review address the use of an oxidized bitumen whose
composmon has been adjusted to provide waste forms with compresswe.strengths at room
temperature that are about an order of magmtude (that is, approximately 4000 kPa) above
previous formulations
Cement-sohdtfied waste Over the years there have been several incidents at power reactors
where problems have been encountered with cement-sohdlfied low-level radioactive wastes
These incidents include problems w~th expansion, disintegration, nonsohd~ficatlon, and
overly rapid sohdlfiCatlon ofcement-sohdlfied waste forms Examples are (1) the cracking and
sphttlng of two hners at Three Mile Island (caused by expansion and dlsmtegranon of the
cement-sohdlfied
Copyright by ASTM Int'l (all rightsbead resin
reserved); waste),
Mon Feb (2)EST
24 23:07:03 expansion
2020 of a liner containing cement-sohdlfied low
8 HAZARDOUSWASTES
oxidation state metal ion (LOMI)-decontamlnatxon waste on bead resin at Millstone, (3) non-
sohdificatlon of LOMI-decontammaUon waste (on resin beads) at Fitzpatnck, (4) too-rapid
solidification (occumng before the prescribed amount o f cement could be added to the m~x)
at Quad Cities, and (5) excesswe foaming (due to detergent "impunties") at Sequoyah [13]
Contnbutlng factors in these cases appear to include
1 Excessive loading (as a fraction of the total volume of the waste form) of bead resin
2 The presence of decontamination solution chemicals (for example, oxahc or plcollmc
acid).
3 The presence of ingredients (considered to be "impurities") m the waste stream that were
not tested m the laboratory development of the waste formulation or recipe
4 A lack of knowledge by the waste generator/processor of what speofic composition of
matenals and chemicals was being processed
The types of observed problems that have been encountered in the field have been corrob-
orated by data obtained through numerous testing programs [14] conducted at Natlonal Lab-
oratories for NRC as well as by testing conducted by radioactive waste processing vendors
In an attempt to obtain an improved understanding of the techmcal concerns revolved m
the use o f c e m e n t m o u s materials to solidify and stablhze low-level radioactive wastes, a work-
shop was held from 31 May to 2 June 1989 in Ga~tbersburg, MD. In attendance were federal
and state regulators, cement system vendors, utility representatives, consultants, and national
laboratory researchers The results of the workshop are reported in an NRC conference pro-
ceedings report [ 15 ].
Technical Posttton Revzszons--Because of the problems that have been observed with
cement-soh&fied wastes, N R C staff is in the process of revising the 1983 TP on waste form
Changes under consideration for the TP include the development of a new appen&x for
cement-stabilized waste that would provide mo&ficatlons to the recommendations for qual-
ification testing as well as new and improved guidance on waste characterization, specimen
preparation, surveillance speomens, and reporting of mishaps
Alternattve Methods of Dtsposal

Low-Level Radtoacttve Waste Pohcy Amendments Act (LLR WPAA) Amendments Act of
1985--Although 10 C F R Part 61 (through the techmcal requirements contained in Subpart
D of the regulation) focuses p n m a n l y on improved (trench-type) shallow land bunal, those
technical reqmrements and performance objectives (in Subpart C) apply to all methods of
near-surface disposal, including those that would utlhze addmonal engineered barriers or
structures As noted earlier, the regulation allows such structures to be used for prowdlng long-
term stability for Classes B and C wastes With the passage of the Low-Level Radioactive Waste
Policy Amendments Act (LLRWPAA) of 1985 [ 16], a number of states are interested m devel-
oping disposal facilities that will incorporate an engineered structure (that ~s, alternative
method of disposal) of some type
The L L R W P A A of 1985 addressed alternatwe methods of &sposal by estabhshmg sched-
ules for development of hcenslng procedures and technical guidance Specifically, NRC was,
by January 1987, required to (1) identify methods of disposal other than shallow land burial
(SLB) and (2) establish the procedures and technical capability for processing a license appli-
cation within 15 months of receipt Further, the Act reqmred that by January 1988, NRC had
to publish all relevant technical information regarding such altenat~ve methods of disposal
The NRC met the requirements of the L L R W P A A of 1985 by a variety ofacuons, discussed
below
Copyright In fact,
by ASTM Int'l (allthe
rightsagency hadFebbeen
reserved); Mon engaged
24 23:07:03 in work on alternatives for quite some time before
EST 2020
the L L R W P A A ' s enactment For example, in 1983 a study was initiated with the U S Corps
of Engineers (COE) that evaluated several generic disposal concepts That study resulted in a
multivolume report [17,18] that identified five basic types of alternative disposal facilities (1)
above-ground vaults (AGVs); (2) below-ground vaults (BGVs), (3) earth-mounded concrete
bunkers (EMCBs), (4) augured holes (shaft disposal); and (5) mined cavities The study cul-
minated in the publication of a branch technical position [ 19] that, combined with the above-
mentioned NRC contractor report, fulfilled the requirements of Section 8(a) of L L R W P A A
with regard to identifying viable alternative disposal methods and providing technical infor-
mation concerning such methods
Some further details on the NRC staff position on alternatives are contained in a paper
titled, " N R C Perspective on Alternative Disposal Methods" [20], which was delivered at the
Department of Energy (DOE) 8th Annual Forum on Low-Level Waste, in Denver, in Septem-
ber 1986 In that paper the following two points were emphasized
1 Although all five alternative methods of disposal identified in the COE study are work-
able and licensable, N R C will focus its resources on soil-covered options such as BGVs
and EMCBs (Thus, AGVs, augured holes, and mined cavities would receive minimal
attention, at least for the present )
2 The N R C would encourage standardization of designs
The focus on soil-covered designs was based on the fact that engineered structures for low-
level waste facilities would be required to perform for a m i n i m u m of 300 to 500 years Such
structures will likely be constructed of portland cement-based concrete, which from a long-
term materials durability standpoint requires protection from the potentially deletenous
effects of various chmatologlcal phenomena such as normal erosional forces, freeze/thaw
cycles, and acid rain Thus, the earthen cover would provide protection for the concrete from
adverse climatological effects and would also provide an additional barrier for radionuchde
release and intruder protection
The emphasis on standardization was intended to foster a uniform approach to the national
program The proposed approach acknowledged that technical resources were limited and that
providing guidance on all conceivable options and still meeting the milestones and schedules
established by the amendments act was not possible
Meetmg the Amendments Act Mdestones--In addition to identifying viable methods of dis-
posal and publishing technical guidance on such methods by January 1987, NRC was also
required by that date to establish procedures and to develop the technical capablhty to com-
plete the review of a licensing application within 15 months of receipt The review procedures
and capability requirements were met by NRC (for disposal by trench-type burial) through the
issuance of two guidance documents: (1) a Standard Format and Content Guide (SF&C) [21]
and (2) a Standard Review Plan (SRP) [22] The SF&C describes for an applicant the types of
information that should be submitted with an application for a disposal facility license, while
SRP provides NRC staffrevlewers with guidance on how to review the information provided
with the application Because the (Revision 0) versions of SF&C and SRP that were pubhshed
in January 1987 primarily address Part 61-type trench burial, they had to be augmented with
further information concerning engineered structures and barriers This augmentation was
was accomplished in two steps (1) the publishing of a two-volume report [23] that provided
recommendations from COE to NRC for acceptance criteria for BGVs and EMCBs, (2) the
augmentation of SRP and SF&C by incorporating additional information and criteria from
the COE study into a Revision 1 version of those N R C licensing guidance documents These
steps satisfied the Section 8(b) requirements of the LLRWPA, namely, to identify and publish
Copyright
"all by ASTM Int'l technical
relevant (all rights reserved); Mon Feb 24 23:07:03
information" ESTmust
that 2020 be submitted to the NRC to pursue an alter-
10 HAZARDOUSWASTES
naUve disposal method, together with the techmcal reqmrements for hcensmg alternative dxs-
posal facd~Ues."
Technical Guidance and Review Crzterla for EMCBs and BGVs--ln developing the tech-
mcal gmdance and review cntena that are contained in SRP, the staff addressed eight major
categories of concern (1) loads and load combinations; (2) structural design and analysis, (3)
construchon materials quahty and durabd~ty, (4) constructmn and operations, (5) quahty
assurance, (6) structural performance monitoring, (7) filter and drainage systems, and (8)
waste cover systems For each of these categories, specific design rewew criteria were developed
and incorporated into SRPs to prowde NRC staff reviewers with guidance on performing a
safety rewew of an EMCB or BGV design.
The N R C staff is aware of the concerns for safe LLW disposal that have resulted m addi-
tional (by some state authontxes) techmcal reqmrements and features that go beyond the tech-
mcal reqmrements and performance objectwes of 10 CFR Part 61 Examples of these reqmre-
ments include the capabdmes of monltonng mdlwdual waste containers for radmact~ve
matenal releases and recovenng the emplaced waste from a disposal unit for periods of time
after placement that would extend through the lnst~tutmnal control penod The staff recog-
nizes that the objectives of these addmonal measures are to reqmre a greater degree of protec-
Uon and conservatism than is inherent in the requxrements of 10 CFR Part 61 Although pro-
posals and suggestions for lmprowng the level of protection from the potentml hazards of low-
level waste d~sposal are always welcome, the staff has been consistent m ~ts determmahon not
to expend resources to develop techmcal gmdance and cntena for engmeenng enhancements
and addmons that are m excess of 10 CFR Part 61 reqmrements Th~s posmon stems from the
fact that, m the view of the staff, the reqmrements of 10 CFR Part 61, as supplemented by the
referenced gmdance, are sufficient to assure the safe d~sposal of waste m an EMCB or BGV
Therefore, the posmon has consistently been [24] that the staff, m reviewing a LLW hcense
apphcatlon, would address the ~mpact of state regulatory reqmrements that exceed 10 CFR
Part 61 solely to ensure that such reqmrements would not result in detrimental conditions that
could adversely affect the abdlty of the disposal facd~ty to safely contain the LLW
Summary
Nuclear Regulatory Commission regulation 10 CFR Part 61 estabhshes the basic reqmre-
ments for long-term structural stabihty of low-level radioactive wastes The regulation indi-
cates that the reqmred structural stabd~ty can be prowded m a variety of ways The methods
most w~dely used at present revolve the use ofsohdlficatlon media and high-integrity contain-
ers With the passage of the L L R W P A A of 1985, however, several potenUal new waste disposal
facdmes may address the stability reqmrements through the use of engineered structures such
as below-ground vaults The regulatmn ~s complemented w~th guidance m the form of tech-
nical posmons, such as the "Techmcal Posmon on Waste F o r m " (for sohdlficatmn/stablllza-
tlon medm and HICs) and v i n o u s N U R E G reports (for engineered structures)
References
[1 ] "Part 6 l--Licensing Requirements for Land Disposal of Radioactive Waste," Code of FederaIReg-
ulattons, Title 10, "Energy," U S Government Printing Office, Washington, DC, 1 Jan 1990
[2] "Technical Position on Waste Form," Rev 0, U S Nuclear Regulatory Commission, Washington,
DC, May 1983
[3] ASTM Test Method for Compressive Strength of Cyhndrlcal Concrete Specimens (C 39-84)
[4] ASTM Test Method for Compressive Strength of Bituminous Mixtures (D 1074-83)
[5] ASTM Test Method for Thermal Cycling of Electroplated Plastics (B 553-79)
[6] byASTM
Copyright Recommended
ASTM Int'l Practice
(all rights reserved); Mon for Determining
Feb 24 23:07:03 EST 2020 Resistance of Synthetic Polymeric Materials to
Fungi by
Downloaded/printed (G 21-70)
[7] ASTM Recommended Practice for Determining Resistance of Plasncs to Bacteria (G 22-76
(1985)]
[8] American Nuclear Society, Standard for Measurement of the Leachablhty of Sohdlfied Low-Level
Radloact~ve Wastes by a Short-Term Test Procedure, ANS 16 1, 1986
[9] American Nuclear Society, Standard for Ra&oacUve Waste Processing System for Light Water
Cooled Reactor Plants, ANS 55 1, 1979
[10] P~res, J , "Review of the High Integrity Cask Structural Evaluation Program (HICSEP)," report,
Brookhaven NaUonal Laboratory, Upton, LI, NY, 6 April 1987
[11] Sllhng, S A, "Review of the Structural Designs of Polyethylene High Integrity Containers," Brown
Umvers~ty, Providence, RI, June 1988
[12] "Techmcal Evaluation Report Related to the Topical Reports on H~gh Integrity Containers made
w~th Hxgh-Denslty Polyethylene," Nuclear Regulatory Commission, Washington, DC, December
1988
[13] "Chemical Reactions with Radioactive Waste Soh&ficatlon Agents," NRC Informatmn Notice No
88, Nuclear Regulatory Commission, Washington, DC, 14 March 1988
[14] Plcmlo, P, et al, "The Effect of Cure CondlUons on the StaNhty of Cement Waste Forms After
Immersion m Water," Report BNL-FIN A-3171/13, Brookhaven National Laboratory, September
1987
[15] "Proceedings of the Workshop on Cement Stabilization of Low-Level Radioactive Waste," Report
NUREG/CP-0103, Nuclear Regulatory Com mlssmn, Washington, DC, October 1989
[16] Public Law 99-240, Title l--Low-Level Ra&oactxve Waste Pohcy Amendments Act of 1985
[17] Bennett, R D, et al, "AlternaUve Methods of Disposal of Low-Level Radmact~ve Waste,"
NUREG/CR-3774, Vols 2-5, Nuclear Regulatory Commlssmn, Washington, DC, October 1985
[18] McAneny, C C, "Alternative Methods of Disposal of Low-Level Ra&oacUve Waste," NUREG/
GR-3774, Vol 6, Nuclear Regulatory Commxsslon, Washington, DC, October 1986
[19] Hlggenbotham, L B, et al, "Licensing of Alternative Methods of Disposal of Low-Level Ra&oac-
Uve Waste," Report NUREG- 1241, Nuclear Regulatory Commission, Washington, DC, December
1986
[20] Plttlgho, C L and Tokar, M, "NRC Perspective on AlternaUve Disposal Methods," presented at
8th Annual DOE Low-Level Radtoacttve Waste Forum, Denver, September 1986, Nuclear Regu-
latory Commission, Washington, DC
[21] "Standard Format and Content for a License Apphcatlon for a Low-Level Radmoctlve Waste Dis-
posal Faclhty," NUREG-1199, Nuclear Regulatory Commlssmn, Washington, DC, January 1987
[22] "Standard Review Plan for the Review of a License Apphcatmn for a Low-Level Ra&oact~ve Waste
Dxsposal Facility," NUREG-1200, Nuclear Regulatory Commlssmn, Washington, DC, January
1987
[23] Denson, R H, Bennett, R D, Wamsley, R M, Bean, D L, and Amsworth, D L, "Recommen-
daUons to the NRC for Revxew Criteria for Alternative Methods of Low-Level Radioactive Waste
D~sposal," Vol I--Task 2a Below-ground Vaults, December 1987, Vol II--Task 2b Earth-
Mounded Concrete Bunkers, NUREG/CR-5041, Nuclear Regulatory Commlssmn, Washington,
DC, January 1988
[24] Kane, J D andTokar, M ,"ConslderatmnsandGmdancemtheDeslgnandConstructmnofAlter-
native LLW Disposal Facllmes," presented at the Nmth Annual DOE Radioactive Waste Manage-
ment Conference, Denver, CO, 25 Aug 1987, Nuclear Regulatory Comm~ssmn, Washington, DC
John J. Barich, l i p and B. J. Mason 2
Issues Related to the Public and Regulatory

Acceptance of Performance Standards for
Wastes Remaining on Hazardous Waste Sites
REFERENCE" Banch, J J , Ill and Mason, B J , "Issues Related to the Public and Regulatory
Acceptance of Performance Standards for Wastes Remaining on Hazardous Waste Sites," Sta-
bthzatlon and Sohdtficatton ofHazardous, Radloacuve, and Mlxed Wastes, 2nd Volume, ASTM
STP 1123 T M Gllham and C C Wiles, Eds, American Society for Testing and Materials,
Philadelphia, 1992, pp 12-17
ABSTRACT" Stablhzation/sohdlfiCatlon is suitable as a permanent remedial measure for haz-

ardous waste sites whenever the performance standards fulfill regulatory requirements and pub-
lic expectations The acceptability of performance standards is a function of the comphance
points selected for the site, the test methods specified to evaluate the performance of the stabl-
hzed waste, and the analytical procedures (models) used to relate test data to performance stan-
dards at the compliance points
Performance standards derive from natural resource or health nsk assessments and from the
identification and analysis of regulatory requirements Performance standards also include the
minimal acceptable engineering and geotechnical properhes of the stablhzed waste The com-
phance points are the locations where and frequencies when monitonng occurs Test methods
range from fully peer reviewed and approved to nonstandard Data acquired with approved
methods Is easier to defend Nonstandard test methods may address more directly site-specific
reqmrements Analytical models are used frequently to interpret data Models range from simple
(for example, mulUpllcatmn by unity) to complex
Issues related to the regulatory and public acceptance of stabilization or soh&fication perfor-
mance standards are &scussed
KEY WORDS: stabilization, soh&ficatlon, comphance, performance standards, risk assess-

ment, regulatory acceptance
A n i m p o r t a n t class of hazardous waste sites in the U n i t e d States includes those with signif-
icant v o l u m e s o f accessible, c o n t a m i n a t e d soils Stabilization is a p e r m a n e n t remedial mea-
sure specified frequently for these sites Stabilization/solidification is a suitable p e r m a n e n t
remedial measure w h e n e v e r the performance standards established for the site fulfill both reg-
ulatory requirements and public expectations Performance standards are associated with
defined c o m p l i a n c e points Test protocols are specified at the compliance points Analytical
procedures (defined as models) relate test data to performance standards at the compliance
points Models must address the complete p e r m a n e n t remedy, of which the performance o f
the stabilized waste is one element. Issues related to the regulatory and public acceptance of
stabihzation/solidification performance standards are discussed.
P e r f o r m a n c e standards, c o m p l i a n c e points, test methods, analytical models, and the total
remedial design concept are independent factors. Because there is considerable latitude in the
site-specific specification of these factors, an opportunity exists at any site to propose different
l U S Environmental Protection Agency, Seattle, WA 98101

2 ETHURA, Grants Pass, OR 97526
12
Copyright 9 1992 by ASTM International www.astm.org

BARICH AND MASON ON PERFORMANCESTANDARDS 13
remedial actions based on stabxhzatlon. This provides both flexibility and uncertainty. Flexl-
blhty increases the probability that site owners, regulators, and the interested public can agree
on a permanent remedy. Uncertainty increases the probability that these principals will hold
different ideas as to what the permanent remedy entails. Flexibility is exploited while uncer-
tainty is avoided ffperformance standards, compliance points, test methods, analyhcal mod-
els, and the total remedial design concept are defined precisely at the beginning of a project
Although this paper discusses hazardous waste sites, especially those regulated by the United
States Comprehensive Environmental Response, Compensation, and Liability Act (CER-
CLA, or Superfund), the discussion of issues also applies to other environmental problem
classes such as pits, ponds, and lagoons where stablllzatlon/sohdlfiCatlon may be considered
as best demonstrated available technology (BDAT) Similarly, since equivalent problems can
occur at sites characterized by radioactive contamination, consideration of these issues may
also be useful in gaming acceptance of solidification/stabilization for radioactive sites.
Discussion
Performance Standards
Performance standards derive from the well-documented nsk assessment process and the
identification and analysis of regulatory requirements [1 ]. In the federal Superfund program
in the United States, the Applicable or Relevant and Appropnate Requirements (ARARs) pro-
cess documents regulatory requirements [2] Performance standards also include the minimal
acceptable englneenng and geotechnlcal properties of the stabilized waste
ARARs include all requirements in federal or state law that have direct applicability at a
site They include requirements which are similar enough to the elements of the remedial mea-
sure to be considered as "relevant and appropnate" performance standards [3] ARARs can
be numerous There is, however, a procedure to simplify the translation of ARARs into per-
formance standards.
Federal and state requirements often address the same concerns A dnnking water require-
ment, for example, might be similar in its federal and state form. All candidate performance
standards which derive from ARARs are evaluated for overlap (commonly between govern-
mental junsdictlons) "Nonoverlapping performance standards" are the smallest set of per-
formance standards for which there is no overlap These are analyzed for Instances where by
meeting one performance standard, another will be met This situation occurs most frequently
in sites with multiple contaminants. If by controlling one heavy metal out of several confirmed
at a site, for example, all will be controlled, then the performance standard associated with the
first metal is a "design-controlhng performance s t a n d a r d "
The desired set of performance standards which denve from ARARs is the set of nonover-
lapping, design-controlling performance standards The complexity of hazardous waste sites
is such that considerable effort should be allocated to simplifying the performance standards
by defining exactly what constitutes the nonoverlapping, design-controlling set of ARARs
Each nonoverlappmg, deslgn-controlhng A R A R is analyzed further to quantify the A R A R
as a performance standard Three cases are possible First, the A R A R may itself be a well-
defined, unambiguous performance standard An example is the waste listing threshold that
softs exceeding the Toxicity Characteristic Leaching Procedure (TCLP) are hazardous The
test method and the meaning of the test result are known, alternative interpretations and site-
specific apphcations are neither desired nor permitted
Second, the A R A R may contain a well-defined endpomt, but alternative methods of spec-
ifying the endpolnt at a site may exist. An example is a water quality criterion for the protection
of aquatic
Copyright resources
by ASTM Int'l (cadmium
(all rights reserved); less
Mon Feb 24 thatEST8 2020
23:07:03 3E - 5 mol/m 3 to be protective of m a n n e aquatic
14 HAZARDOUSWASTES
resources from chronic toxioty). Although the endpomt is well-defined, the relationship to the
stabilized waste is not. How much cadmium will be released from the waste to the water pro-
tected by the chronic toxicity ARAR? Will a zone of mixing be permitted 9 How large is the
zone of mixing9 Although answers to these questions are obtained without reference to the
stabilized waste, they are cntical to the determination of the performance to be achieved by
the stabilized waste The questions embody both management and technical components
How large a zone of mixing is primarily a management issue The configuration of the zone
of mixing is more a technical issue, with alternative technical tools available to the analyst
Third, both the endpoint and the method of application to the site may be subject to inter-
pretation An example is a requirement that groundwater at the site be returned to as close to
the background condition as possible A study must be performed to answer the questmn,
"What is backgroundg" Once available, what contaminant flux can be tolerated and still not
wolate the background requirement9 As w~th the previous case, answers to these questions are
obtained without reference to the stabilized waste They are pnmary determinants, however,
of performance to be achieved by the stabilized waste
Performance standards which denve from natural resource or health nsk assessments are
similar to the ARARs case where alternative endpoints and alternative methods ofapphcation
to sites are permitted In this case the natural resource assessment or risk assessment may
require a study as involved as any associated with the site Once these assessments are avail-
able, the acceptable flux of contaminants is determined Natural resource and health nsk
assessments, as m the earlier cases, are conducted without reference to the stabilized waste
They, too, are primary determinants of performance to be achieved by the stabilized waste
Englneenng and geotechnical properties also define performance standards Included are
requirements such as m i n i m u m compressive strength, bulking, pumpablllty, set time, etc
These tend to be of more interest to site owners and stabilization mix designers than to regu-
lators and interested pubhcs
Stablhzatlon/sohdlfication is suitable as a permanent remedy if it fulfills all performance
standards established for the site Estabhshlng performance standards is an involved process
which can result in different standards for the same site The regulator is in no position to
approve stabilization as suitable until the performance standards are complete The stabili-
zation vendor is unable to propose a stablhzatlon mix until the performance standards are
agreed upon Finally, interested publics have no assurance that the remedial measures will
fulfill their expectations until performance measures are defined for the site
Comphance Points
Many performance standards are evaluated at compliance points Compliance points are
locations where and frequencies when monItonng occurs When remedial decisions are com-
pleted for Superfund sites In the Untied States, compliance points must be identified The
requirement states "Performance shall be measured at appropnate locations in the ground
water, surface water, soils, air, and other affected environmental media Measurement relating
to the performance of the treatment processes and the engmeenng controls may also be lden-
tiffed [4] " J u s t as there is flexibility in the identification of performance standards, there
IS flexibility in the identification ofcomphance points Since the achievement of performance
standards is determined at the compliance points, the suitability of stablhzatlon as a perma-
nent remedy cannot be ascertained without consideration of the compliance points
Compliance points exhibit the following properties location, frequency of testing, method
of sampling, and test methods Each must be specified precisely before a valid compliance
point exists
The
Copyright location
by ASTM of areserved);
Int'l (all rights compliance
Mon Feb 24point
23:07:03 can be established within the waste form, adJacent to
EST 2020
BARICH AND MASON ON PERFORMANCE STANDARDS 15
the waste form, at a property hne or other legally defined boundary (such as a site boundary),
at a discharge point, or at a point of use Multiple compliance points can be specified, for exam-
ple, at the edge of the waste form and at the property line The closer the compliance point to
the waste form, the more direct is the measure of waste behavior, but the less d~rect is the rela-
tionship of comphance point m o n l t o n n g data to performance standards derived from natural
resource and risk assessments Conversely, the closer the compliance point to the point of use,
the more direct is the relationship of compliance point momtoring data to performance stan-
dards, but the less direct is the measure of waste behavior
The location of a compliance point is not a point but a dimensioned volume The dimen-
sions of the volume may or may not be important An example ofd~menslons of the comph-
ance point volume being important is a compliance point which is a well at the property line
at a site where the primary contaminant is a dense nonaqueous phase llqmd (DNAPL) The
size and placement of well screens with respect to formanons which constrain the DNAPL can
play a significant role in determining whether a given sample meets the performance
standards.
At each compliance point a frequency of testing is specified The shorter the frequency with
respect to anticipated changes in the stabilized waste, the smaller will be the expected changes
m the performance standard parameters and the more difficult the task of confirming that real
changes are occurnng Shorter frequencies improve the chances that failures can be detected,
and hopefully corrected, before they become severe Longer frequencies may slmphfy the task
of confirming that real changes are occurring Failures, however, may be more advanced and
more d~fflcult to correct.
The method of samphng can simplify or exacerbate the confirmation at the compliance
points that performance standards are achieved If the compliance point is within the waste
form, for example, and the performance standard is one of permeability, then it may be desir-
able to develop a stock of samples (or "coupons") during construction for future permeabdlty
testing. The alternative may be to gain access periodically to the stabilized waste and attempt
to obtain samples that are appropriate for permeabihty testing
Performance standards are expressed whenever possible as quanntanve characteristics, test
methods transform samples acqmred at the compliance points into results which are related
through models to the quantitative characteristics Leaching is an example of a characteristic
of a waste form Leaching of a coupon by TCLP is an example of a test method
Test Methods
Alternative test methods are available to evaluate characteristics of performance standards
Leachablhty is an important charactenstlc ofstablhzed wastes Several test methods for leach-
ability are available The U S Environmental Protection Agency in a recent guidance docu-
ment, for example, ldennfies twelve separate leach tests [5]
The availability of several leaching tests illustrates three characteristics of test methods.
First, there are often several test methods addressing a performance charactenstlc which has
either regulatory status or is fully peer reviewed and approved Whenever these are specified
for use at a compliance point, the user need only defend the selection of the method and the
interpretation of the data The requirement to defend only method selection and data Inter-
pretation can be an advantage whenever stablhzanon IS used in a regulatory or litigation set-
ring The peer review process or the administrative procedures which precede the adoption of
a regulatory test provide the primary defense for approved methods
Second, whenever nonstandard or nonpeer-reviewed test methods are specified, the meth-
ods themselves must be defended This is in addition to the defense of the application and the
interpretation
Copyright ofrights
by ASTM Int'l (all data A nonstandard
reserved); orEST
Mon Feb 24 23:07:03 nonpeer-revlewed
2020 method may meet the engineering
16 HAZARDOUSWASTES
requirements of the site better than the approved test methods The user must decide whether
the improvement is sufficient to justify the additional requirement to defend the method
Third, whenever several candidate tests are available, they can be rank ordered from most
to least aggresswe. The most aggressive test method is the one least likely to produce data which
meets the performance standards, the least aggressive test method ~s the one most hkely to
meet the performance standards The decision on how aggressive a test method to specify is
often based on the management decision of how conservatively the site is managed
Models'
Data acquired at compliance points are not the same as the performance standards Data
are related to performance standards through the use of models Models range from elemen-
tary (e g , multiplication of test results by unity) to complex (e g , those which allow numerous
complex phenomena to act on the test results) Leachate modeling provides a good example
Perhaps the most important property of a stabihzed waste which relates to performance
standards is leachability The U S Environmental Protecnon Agency has published several
guidances on models associated with contaminant release to groundwater from hazardous
waste sites [6-8] Models produce dilution attenuation factors (DAFs) or adjustment that
must be made to leaching test data before a comparison is made to performance standards at
compliance points Two characteristics of models are important as they relate to DAFs
First, for any slte-speofic application, candidate models can be ordered according to the
D A F they are expected to produce Candidate models are only those which reflect reasonably
well conditions at the site. Models which might be expected to provide larger DAFs are those
which simulate processes such as dispersion, diffusion, sorptIon, chemical degradation, etc
Cote and Bridle have developed simphfied mathematical models to describe leaching of lnor-
ganlcs [9] They evaluate eight scenanos which assume different contaminant solublhties,
waste form permeabllmes, groundwater velocities, and groundwater chemistry Their results,
presented as leaching rates m m m o l / m - 2 / d a y - ' versus time, display as eight dlStlnct bands
Leaching rates vary over eight orders of magnitude However since the variables which make
up the scenarios are continuous rather than discreet, leaching from actual sites is not hmlted
to the eight bands derived from their analyses Eight orders-of-magnitude vanablhty is shown
by them to be reasonable If leaching rates can be expected to vary this much, then DAFs can
also be expected to vary greatly
Second, leaching model results are sensitive to basic input parameters such as permeabihty
Half of the total benefit of stablhzing soils at the Western Processing Superfund site, for exam-
ple, was modeled to be attributable to an improvement m sod permeablhty [10] Hydraulic
conductlvitaes were assumed m the model to be reduced by one order of magnitude m the
stablhzed waste Stablhzed coupons from three stabilization vendors vaned in permeabdlty by
three orders of magnitude Permeabilitles of the untreated, contaminated soils varied over two
orders of magnitude Thus, the modeling effort could have jushfied the use of a hydraulic con-
ductivity improvement of much more or less than the one order of magnitude selected
Models are necessary to interpret data acquired at compliance points Users must recognize
that there is considerable freedom m the selection of appropriate models and in the speclfi-
carton of input data
Conclusion
The question of whether stablhzatlon/sohdlficatlon is an appropriate permanent remedy for
a hazardous waste site is answered by confirming the regulatory and public acceptance of the
performance
Copyright by ASTM Int'lstandards established
(all rights reserved); for the
Mon Feb 24 23:07:03 ESTsite
2020 There is considerable flexibility in establishing
BARICH AND MASON ON PERFORMANCE STANDARDS 17
performance standards Performance standards are evaluated at comphance points There IS

also flexibility in establishing compliance points Finally, compliance point data is related
through models to the performance standards Once again, there is flexlblhty in the selection
and application of models
Flexibility increases the likelihood that regulators, stabilization technology providers, site
owners, and interested publics can agree on an appropriate stablhzatlon response However,
flexibility increases the potential for miscommunlcatlon Flexibility is exploited while mIs-
communication is avoided if the performance standards, compliance points, and analytical
models are agreed upon early in the hfe of a project These items require regulatory agreement,
whenever possible, prior to selection and implementation of stabilization/solidificationas a
permanent remedy
References
[1] "Summary of Ecological Risks, Assessment Methods, and Risk Management Decisions in Super-
fund and RCRA," EPA-230-03-89-046,USEPA, Office of Pohcy, Planning, and Evaluation, Wash-
lngton, DC, June 1989
[2] Section 121, Superfund Amendments and Reauthonzatlon Act of 1986, Public Law 99-499
[3] "CERCLA Compliance with other Environmental Laws Draft Guidance," EPA/540/6-89/006,
USEPA, Washington, DC, 1989
[4] 40 CFR 300 430(f)(5)(nl)(A)
[5] "Stablhzatlon/SohdlfiCatlon of CERCLA and RCRA Wastes, Physical Tests, Chemical Procedures,
Technology Screening, and Field Activities," Section 5 2, EPA-625/6-89/022, USEPA, Washington,
DC, May 1989
[6] "Modeling Remedial Actions at Hazardous Waste Sites," EPA/540/2-85/001, USEPA, Office of
Solid Waste and Emergency Response, Washington, DC, April 1985
[ 7] "Selectmn Cntena for Mathematical Models Used in Exposure Assessments Ground Water Mod-
els," EPA/600/8-88/075, USEPA, Washington, DC, 1988
[8] "Predicting Subsurface Contaminant Transport and Transformation Considerations for Model
Selection and Field Vahdatlon," EPA/600/2-89/045, USEPA, Washington, DC, 1989
[9] Cote, P L and Bridle, T R, "Long-Term Leaching Scenarios for Cement-Based Waste Forms,"
Waste Management & Research, Vol 5, 1987, pp 55-66
[10] Banch, J J, Greene, J, and Bond, R, "Soil Stabilization Treatability Study at the Western Pro-
cessing Superfund Site," SUPERFUND "87 Proceedings of the 8th National Conference, The Haz-
ardous Materials Control Research Institute, Washington, DC, November 1987
Carlton C. Wiles La n d Edwin Barth ~
Solidification/Stabilization: Is it Always
Appropriate?
REFERENCE" Wiles, C C and Barth, E "Solidification/Stabilization Is It Always Appropri-
ate ~'' Stabdtzatton and Sohd#catton of Hazardous, Radzoacttve, and Mixed Wastes, 2nd Vol-
ume, ASTMSTP 1123, T M GIlham and C C Wiles, Eds American Socwty for Testing and
Materials, Philadelphia, 1992, pp 18-32
ABSTRACT: The findings of recent research and evaluatmn efforts are assessed to determine
whether sohdlficatlon/stabfllzatlon (S/S) has been properly and appropriately applied for differ-
ent types of hazardous wastes Results from these studies are mixed and, as a result, the need for
proper test procedures and for treatabihty stu&es is emphasized Technical gmdance ts given for
assessingthe appropriateness of using S/S for any specificwastes
KEY WORDS. sohdLficatlon,stabdlzation, Immobilization, hazardous waste, toxic waste, waste

treatment, remedmtton, treataNhty studies
The increasing use of sohdlficatlon/stablhzatlon (S/S) technologies in the United States,

especially for remedlation of sites under the Superfund program, has raised several questions
about the overall appropriateness of S/S For many types of hazardous waste, notably for
heavy metals, S/S usually gives excellent results for long-term immobilization, as measured by
existing physical and chemical protocols However, results of several studies, as well as data
from remedlatlon of several Superfund sites, have raised concerns about whether S/S is a valid
technology for treating orgamc-bearlng wastes Even when applied to heavy metals, S/S
requires careful choice of proper binders (recognizing the amphotenc behavior of certain met-
als) and good quality control throughout the process Lack of good Investigative procedures
has &mmlshed the value of data for evaluating S/S for some metals Furthermore, studies also
provide evidence that tests other than the regulatory extraction tests [for example, toxicity
characteristic leaching procedure (TCLP)] will be required to evaluate the effectiveness of
S/S, especially when applied to organic wastes Suggestions are offered for improving treat-
ability studies used for evaluating S/S applied to selected metals Approaches are also provided
for determining the appropriateness of S/S applied to organic contaminants This paper will
focus on evaluating chemical leaching behavior as a measure of S/S effectLveness
Mechanisms for Containing Contaminants by Solidification/Stabilization

At present, the most prevalently used S/S technologies are cement-based or pozzolan-based
Such processes result in several chemical and physical mechanisms that combine, capture,
and/or lmmoblhze the waste contaminant The chemical mechanisms can include chemical
form change (such as the chemical change of a soluble salt of a hazardous metal to a relatively
insoluble silicate, hydroxide, or carbonate form), chemical incorporation within a crystal, sire-
U S Environmental Protection Agency, Risk Reduction Engmeenng Laboratory, Cincinnati, OH

45268
18
Copyright9 1992byASTM International www.astm.org

WILES AND BARTH ON SOLIDIFICATION/STABILIZATION 19
ple adsorption mechanisms, and similar chemical processes The physical mechanisms
involve the capture (mlcroencapsulatlon) of hazardous material within the resulting physical
structure of the solidified waste form. Release of physically contained contaminants is
impeded by effects such as a decrease in the permeability and porosity compared to that of the
"native" soil, an increase in the tortuosity of escape paths, and decreased surface area These
physical characteristics aid in eliminating or controlling water transport through the solidified
product and thus the transport of the contaminant to the environment However, if the waste
exists (or can exist) in a liquid or gaseous form at ambient temperatures and pressures, then
the possibility of contaminant escape in a form other than in an aqueous solution must also
be considered
Solidification/stabilization has been repeatedly demonstrated as an acceptable method to
immobilize heavy metal wastes, as well as soils and sludges contaminated with heavy metals
This treatment technique is not without potential pitfalls Technologies such as pozzolanlc
solidification, in which the hazardous metal is incorporated into an insoluble crystalline struc-
ture of the solid product, are probably the most reliable for long-term immobilization Other
techniques, such as lime stabilization, may not be as reliable Careful consideration of waste
characteristics versus binder mechanism is absolutely necessary when consldenng any S/S
process
The opportunities to capture and contain organic materials in a cement-based solidification
process are limited Most organics do not form insoluble precipitates Moreover, most organic
molecules are too large and do not readily enter into the crystal structure of most crystalhne
solids Therefore, sorptlon processes and various physical containment mechanisms are the
major mechanisms that immobilize organics It should be noted that at least one S/S vendor
claims to have a proprietary chemical that incorporates large organic molecules into a clath-
rate, a structural capture analogous to capture within a crystal
Evaluations of whether a contaminant has been successfully immobilized by a particular
S/S process are usually based on how much of the contaminant is removed by an extraction
test For metals, such extraction tests have been based on aqueous or weak acid extractions
Test procedures for the evaluation of the immobilization of organic materials must take into
consideration the m a n n e r in which organic materials can escape Most hazardous organic
materials are only slightly soluble in water or in the weak acids used in the common regulatory
extraction test procedures Thus, these tests may measure the ability of the material to escape
in an aqueous medium but not ItS ability to escape in its liquid or gaseous form Many organics
are not readily mobile in an aqueous medium and yet can be quite mobile in liquid or gaseous
form The U S Environmental Protection Agency (EPA) is sponsoring studies of test and eval-
uation procedures that may lead to improved technical evaluations of S/S applied to a broad
range of waste types, both organic and inorganic
S / S Applied to Toxic Metals

The chemical mechanisms for binding and incorporating hazardous constituents into S/S
products are exceedingly complex Review of data from several S/S apphcatlons reveals a lack
of consistency In reporting which makes it difficult to interpret and compare results Further-
more, many S/S technologies on the market include proprietary additives whose function can
vary from useless to absolutely necessary The EPA is supporting research to further the under-
standing of S/S approaches Even without a complete understanding of S/S chemistry, it is
obvious that the chemical character of the waste constituents and possible chemical interac-
tions between the waste, the binder materials, and any interfering chemicals are of paramount
Importance
For
Copyright example,
by ASTM consider
Int'l (all rights three
reserved); Mon Feb toxic metals
24 23:07:03 EST 2020of c o m m o n concern arsenic, chrome, and lead
20 HAZARDOUS WASTES
Arsenic under oxidizing conditions will be in the pentavalent form and will be soluble over a
broad range o f p H values (pH from = 2 to 14). Thus, in order to stabilize arsenic chemically,
it may be necessary to reduce it to the trivalent or elemental form. In this case, the pH must
be maintained below 11, as As(III) becomes soluble at a p H of > 11.
Chromium is similar to arsenic in its chemical behavior. The oxidized, soluble form of
chrome is the hexavalent form Cr(VI). In order to apply many of the c o m m o n SIS technolo-
gies, the operator first must chemically reduce Cr(VI) to Cr(III), which is relatively insoluble
over a p H range of about 5 to 13. Again, controlling final pH in the process is necessary because
Cr(III) is soluble outside this range.
Lead is amphoteric, that is, it is soluble under both basic and acidic conditions. Lead is fairly
insoluble at pH values between 7 and 12 (Fig. 1). At a pH below about 7, lead is soluble as
Pb ++, while at a pH above 12, it can be soluble as anionic plumbate ion. Chemical reduction
of the lead is ineffective; proper control of pH is usually more important.
100 k I I I I
m 10-
E
z
O
m
l-
et
z 1.0
w
c~
z
0
r
.--I
,<
I-.
uJ 0.1
w
._1
rn
,_1
0
u~ 0 . 0 1 -
0
0
"o
0.001 -
I 1 ] t t J
6 7 8 9 10 11 12
SOLUTION pH
FIG. 1--Solubilities of metal hydroxides as a function of pH.
While these shortcomings do not preclude the use of S/S for these metals, they do emphasize
the importance of performing carefully controlled treatablhty studies before applying specific
S/S techniques to specific waste streams Furthermore, to improve the interpretation of treat-
ability study results, certain guidelines must be followed Preliminary rules of thumb for assess-
ing S/S technology applied to metals, at the bench-scale level, are as follows
1 Characterize the original waste as well as possible This charactenzatlon should include
chemical characteristics, mineral composition, physical properties, and so on, as well as
the normal analyses for pollutant levels
2 Identify the target contaminants and tailor the binder system to low solubility regions of
the metal of interest
3 Perform appropnate extraction tests on the untreated waste to serve as a benchmark for
subsequent data on the treated waste Report extraction test data for both treated and
untreated forms Consider incorporating leaching data into a site-specific groundwater
model
4 Analyze and record p H on the extracts from the untreated waste and treated product
5 Analyze additives for presence of hazardous constituents Identify critical characteristics
of all additives to estabhsh a quality control program Carefully control and record all
additives dunng treatablhty tests Small variations in additives can greatly affect
leachability
6 When reporting leachate data on treated samples, show the total percent reduction
achieved and the extent to which simple dilution has contributed to such reduction
These are not complete gmdehnes for performing treatability studies, merely some rules to
follow that will make S/S data easier to interpret and transfer The importance of conducting
properly designed and controlled treatability studies for any waste cannot be overemphasized
The EPA is currently evaluating protocols for conducting such treatability studies
S / S Applied to Wastes Containing Volatile Organics

It IS c o m m o n to distinguish organics as volatile, semivolatlle, and nonvolatile Typically,
volatile organic compounds (VOC) are considered to be those having boiling points below
150~ (301 ~ When one considers the operational steps in the S/S process, it is obvious that
during the mixing of binder, water, and waste, as well as during the cunng process, VOCs are
lost because of the mixing and because of any rise in temperature The EPA's Office of Air
Quality Planning and Standards (OAQPS) is presently developing guidance for the Resource
Conservation and Recovery Act (RCRA) Treatment, Storage and Disposal (TSD) facilities,
indicating that solidification processes, even for very low levels of volatile constituents, must
incorporate capture and control mechanisms to ensure that there is no release of volatile con-
stituents to the atmosphere Furthermore, since the capture mechanisms of cement-based pro-
cesses depend so heavily on physical containment, it is almost a certainty that volatiles can
escape from the solidified form, no matter how great the retardation introduced by reductions
In pore space or increases in tortuos~ty Thus, as a general rule, sites contaminated with only
volatile constituents should not be considered as candidates for S/S Sites contaminated with
a mix of materials, including volatiles as well as other organics and lnorganlcs, will require
careful characterization before selecting S/S as the pnncipal technology Whether or not S/S
should be used will depend upon the relative levels of volatiles and other constituents present
If the site contains metals or other constituents above specific levels of concern, then additional
treatment by S/S may be reqmred
22 HAZARDOUS WASTES
The quantlhes of such organics acceptable for S/S should be based on a nsk assessment for
the given site and/or on the result of a treatability study that includes a mass balance on the
organics before, during, and after treatment The treatability study should include a "harsh"
extraction procedure (for example, total waste analysis) to evaluate the migration of organics
m both treated and untreated material because many organics are not soluble m water or weak
acid media. The risk assessment should assume that none of the highest risk compounds will
be retained by the S/S process and/or that all such compounds will be lost via mr emissions
d u n n g the S/S processing, unless an air pollution control mechanism ~sin place This scenario
is very conservative as it does not assume any retardation of the compound that can occur by
physical mechanisms m the solidified waste form Conversely, the mass balance approach will
give some credit for physical characteristics or other bonding mechanisms in the solidified
waste form Even so, this approach Is also conservative m that a total waste analysis (TWA) or
s~mflar extraction procedure reqmres gnndmg of the sample This grinding results m destroy-
ing or altenng and reducing the effectiveness of some physical properties that can impede the
release of the organics under actual field conditions However, we beheve these conservative
approaches are necessary based upon available S/S data as applied to organics and the appar-
ent inadequacies in the samphng, analysis, and extraction tests used to evaluate the treatment
of organics at Superfund sites
Therefore, except for the case when 100% hazardous volatile constituents are present, the
amount of organics present may not be the most important factor in deodmg whether or not
to evaluate S/S as part of the total treatment process The important factors are the concen-
tration and characteristics of remaining constituents of concern that will require treatment, ff
all the volatlles were removed and/or destroyed In other words, pretreatment, removal, and/
or capture and treatment of volatile organic constituents will be required prior to or d u n n g S/
S
S / S Applied to Wastes Containing Semivolatile Organics

Evaluating the effectiveness of S/S processes to capture and immobilize semlvolatilc organ-
lCS is more complicated After mixing, the sohdlficatlon process that takes place over a period
of time is called " c u r i n g " During the curing process, the water present IS used in hydration
reactions The final solidified waste form, especially for cement-based processes, is usually a
monolithic structure incorporating hydrated silicates and carbonates in an agglomeration of
crystalline structures that incorporate and/or mlcroencapsulate the hazardous constituents,
the original soils, and any aggregate that may have been added to enhance the product's
strength Solidification processes are normally exothermIc because of the hydration reactions
Typically, the heat given off is sufficient to raise the temperature of the matrix by 30 to 40~
m the early stages o f c u n n g In cases where kiln dust or other lime-based products [that is, those
containing large amounts of quicklime (CaO)] are used, temperature rises can be significantly
higher yet Thus, it is possible that some of the low-bolhng semivolatile compounds volatilize
and escape d u n n g the c u n n g process
Once cunng is complete, the effectiveness of cement-based solidification processes to con-
taln semivolatile organic compounds is currently unclear The apphcatlon of the TCLP test to
organically contaminated soils before and after treatment with cement-based solidification
processes m the Superfund Innovative Technology Evaluation (SITE) program has yielded
inconclusive results This situation is due largely to the fact that the solubility of the organics
m the extraction fluid is so low that even small quantities of these organic contaminants
quickly saturate the TCLP liquid Thus, no measurable difference between the leachability of
organic contaminants from untreated soils and from the solidified soil product can be deter-
mined
Copyright While
by ASTM the
Int'l (all TCLP
rights reserved);test may
Mon Feb be anEST
24 23:07:03 adequate
2020 measure of the probable mobility of these
contaminants m aqueous-phase leaching processes, it does not address the potential escape of
organics in liquid or gaseous form Since the hquld and gaseous forms of these organics can
flow, it is not unreasonable to expect that escape will occur
Thus, the same basic approach is probably necessary for semlvolatfle orgamcs, as is sug-
gested for volatile organics If the level of nonvolatile and/or inorganic constituents is present
above some agreed-upon acceptable threshold, then S/S processes will probably be required
as part of a complete treatment train StabfllZatlon/sohdlfiCatlon would follow some earlier
stage of treatment for the removal and/or destruction of the volatile and semlvolatde constit-
uents The levels of residual semlvolatile compounds remaining m the waste pnor to a solid-
ificat~on process must be determined based on m a x i m u m concentration hmits allowable for
each such constituent Note that in this context, S/S technologies should be deemed inappro-
priate for sites contaminated only with volatile and/or semivolatfle organic compounds, unless
some speemhzed S/S process can be demonstrated to be effective At present, such demon-
stratlon should be based on a TWA and should include a mass balance for the waste constit-
uents As stated earlier, protocols for evaluating S/S apphed to organics are presently being
evaluated by EPA
Discussion of Extraction Tests

Extraction tests are used to determine or estimate the levels of targeted contaminants that
can be extracted from waste under selected conditions None of the presently existing leach
tests were developed for, nor validated for use in, assessing the leachability of orgamc com-
pounds Table l lists several leach tests and some comments on their apphcablhty for orgamc
contaminants Techmcally, the TCLP and similar extraction test methods are not adequate
for identifying all classes of orgamcs that might be available for escape to the environment
The harsh extractions, such as total waste analysis, do not represent realistic field conditions
nor do they recognize immoblhzatlon due to the physical characteristics of the sohdlfied waste
form What is needed is a leaching method sensitive to all phases of organics that are not ade-
quately bound in the solidified matrix Th~s test also needs to represent reahstic field condi-
tions and be relatively easy to perform Unfortunately, no such test currently exists However,
EPA is investigating potentml test methods for this purpose
Caldwell, C6t6, and Chao [ 1 ] have developed leaching procedures based on sequential batch
leaching, with the development of a partition coefficient between the solid and liquid phases
The technique appears to work well for low levels of orgamc contaminants (up to 1000 #g per
gram of solidified waste), but does not fully address the concern about escape m a nonaqueous
medium. Through EPA support, the U S Army Corps of Engineers, Waterways Experiment
Station and others have performed several studies aimed at evaluating different leach proce-
dures for use in solidification of organic contaminated sods and sediments [2-4] In general,
this work indicates that a sequentml batch extracuon process gwes a better measure & t h e total
leachable quantity of contaminants present than do the single-exposure tests
Summary of Selected Results for Organic Wastes

Several studies have been performed that strongly Indicate the madvlsablhty of using S/S as
the prmopal remedtation technology for organic wastes Kolvltes and Bishop [5] conducted
column leaching tests on cement-based solidified samples containing known amounts of phe-
nol and tnchloroethylene (TCE) So much of the TCE was lost in preparation and mixing that
the results were inconclusive However, the results for phenol give a good measure of the
a m o u n t of this semlvolatile organic compound that is retained In the solidified product Table
2 gwes
Copyright the Int'l
by ASTM leach testreserved);
(all rights results Monon
Feb a
24 sample that
23:07:03 EST 2020 was allowed to cure for three days and Table 3
24 HAZARDOUS WASTES
TABLE 1--Some extraction tests used for evaluating stabthzatton/sohdlficatmn of organtc-

contammated waste
Peer Reviewed
Extraction for Organic Utilized for Evaluating
Tests Stablhzatlon Organic Stabilization Comments
TCLP No Several Superfund Gnnding of monolith,

Sites Codlsposal scenario only may not
show organics
Most organics not soluble in
leachate
TCLP Cage No ASTM Committee No gnndlng
D-34 Erosion caused by cage does not
represent "real world"
Poor reproduclblhty noted by
ASTM
Most organics not soluble in
leachate
TCLP Acid No Limited number Used synthetic acid rain with TCLP
Rain of Superfund apparatus
sites
TCLP Acetone No Limited number Acetone to extract PCBs not
of Superfund realistic
sites Conservative worst case does not
represent "'real world"
MEP No Dehstlng (metals) Somewhat realistic
Simulates acid rain
Total Waste No R&D Methylene chlonde or hexane
Analysis Conservative worst case, does not
(TWA) represent "real world"
ANSI 16 1 No Several Superfund May not show organics
sites
ANSI 16 1 plus No Limited number May not show organics
groundwater of Superfund
sites
Acid Rain No Unknown May not be representative of real
(1312) acid rain
May not be apphcable for organics
for a sample that was allowed to cure 28 days before leach testing These tables show a com-
plete mass balance indicating the amount of phenol retained m the sohdlfied product The
enhanced retenhon in the 28-day cured product is probably due to a decrease m pore size in
the product as cunng proceeds Twenty-eight days ~s a typical curing time for cement-based
sohdlficaUon processes. Even so, only about 40% of the phenol is retained m the sohd~fied
product.
Brlcka, Holmes, and Culhnane [6] of the U S. Army Corps of Engineers, Waterways Exper-
Iment StaUon, in a study performed for EPA, showed a companson of the EP and TCLP leach-
mg extracts for two sohd~fied sludges and one nonsohdlfied, orgamc-based sludge (Table 4) A
metal hydroxide sludge (identified as WES) was stabdlzed by addmon of cement A second
matenal (ldenUfied as WTC) was a metal solution containing c h r o m m m chloride, c a d m m m
nitrate, lead mtrate, sodmm arsenlte, and phenol at a pH of 2 5. Th~s syntheUc metal solution
was sohdlfied by adding portland cement, Type F fly ash, and soil in equal quantmes to the
metal solution The third matenal, a PCE sludge, was a sUll-bottom waste resulting from the
TABLE 2--Mass balance leach test No 1 [5]
Phenol Leachant
Leaching Period Released, mg Volume, mL
Days 1 and 2 leachate 27 42 245

vapor 0 00
vapor 0 00
vapor 0 00
vapor 0 00
Total phenol leached from 30 g of cement product 37 13
Phenol adsorbed to hner of cement mold 0 05
Phenol mixed into 30 g of cement product 40 00
Phenol retained m cement product (calculated by &fference) 2 74
&stdlatlon of"dwty" perchloroethylene for solvent recovery Each of these wastes was spiked
with 1000 to 10 000 mg/L of twelve orgamc compounds and then treated The treated mate-
rials were then subjected to the two extraction procedures and their leachate analyzed The
data presented in Table 4 show that, except for the polar compounds butanone and 4-methyl-
pentanone, less than 3% of the spiking contaminant was recovered m either extract, even from
the nonsoll&fied PCE orgamc-based sludge This result clearly ln&cates that aqueous-phase
leach tests are not truly measunng the potentml for escape of low-solubdlty orgamc
compounds.
In an attempt to enhance the capture and retention of organic materials m cement-based
soh&fied waste, Sheriff et al [ 7] of the Imperial College of London added actwated charcoal
and various clays to specific organic waste streams prior to cement-based solidification Figure
2 is an example of the results of m~xmg 40 g of cement with 1 g of charcoal and 40 mg of phenol
and two chlonnated phenols The controls indicated on the graph are the same concentration
of organic contaminants m cement alone with no charcoal added
In a study evaluating the loss of volatile organic matenal d u n n g sohdtficaUon, We~tzman
et al. [8] presented several graphs showing the loss of specific volatile orgamc compounds dur-
ing mixing and c u n n g Figure 3 is typical of their results and shows the results for two soh&-
TABLE 3--Mass balance leach test No 2 [5]
Phenol Leachant
Leaching Penod Released, mg Volume, mL

vapor 0 00
vapor 0 00
vapor 0 00
vapor 0 00
Total phenol leached from 30 g of cement product 23 30
Phenol adsorbed to hner of cement mold 0 02
Phenol mixed into 30 g of cement product 40 00
Phenol retained in cement product (calculated by &fference)
16 68
26 HAZARDOUSWASTES
TABLE 4--Study B average toxtctty charactertsttc leaching procedure (TCLP) and extractton
procedure (EP) extract concentrattonfor the organtc splke contaminants (averaged over the three
rephcate extract samples) [6])
Extract Concentration,
Level of mg/L Organic Recovery, %
Spike Added,
Sludge mg/L EP TCLP EP TCLP
CHLOROFORM
WES l 000 0 88 l 40 0 088 0 140
10 000 13 97 27 27 0 140 0 273
PCE 1 000 1 01 1 56 0 101 0 156
10 000 23 77 32 70 0 238 0 327
WTC 1 000 0 22 0 20 0 022 0 020
10 000 8 98 9 13 0 090 0 091
1,2-DICHLOROETHANE
WES 1 000 1 57 1 27 0 157 0 127
10 000 38 70 61 37 0 387 0 614
PCE 1 000 3 61 4 23 0 361 0 423
10 000 57 30 71 40 0 573 0 714
WTC 1 000 0 76 0 49 0 076 0 049
10 000 45 03 44 23 0 450 0 442
1,1, I-TRICHLOROETHANE
WES 1 000 0 96 1 93 0 096 0 193
10 000 18 33 46 80 0 183 0 468
PCE 1 000 0 55 4 80 0 055 0 480
10 000 15 07 25 07 0 151 0 251
WTC 1 000 0 29 0 45 0 029 0 045
10 000 15 07 24 83 0 151 0 248
CARBON TETRACHLORIDE
WES 1 000 0 42 0 89 0 042 0 089
10 000 3 93 7 60 0 039 0 076
PCE 1 000 0 23 0 50 0 023 0 050
10 000 10 00 10 00 0 100 0 100
WTC 1 000 0 10 0 20 0 010 0 020
10 000 5 00 5 00 0 050 0 050
TRICHLOROETHENE
WES 1 000 3 47 6 90 0 347 0 690
10 000 64 63 134 33 0 646 1 343
PCE 1 000 1 48 3 54 0 148 0 354
10 000 33 73 39 97 0 337 0 400
WTC 1 000 2 32 2 55 0 232 0 255
10 000 98 07 135 67 0 981 1 357
BENZENE
WES 1 000 1 60 2 30 0 160 0 230
10 000 42 97 85 33 0 430 0 853
PCE 1 000 2 62 5 29 0 262 0 529
10 000 54 17 76 57 0 542 0 766
WTC 1 000 0 91 0 79 0 091 0 079
10 000 55 23 62 40 0 552 0 624
l, 1,2,2-TRICHLOROETHANE
WES 1 000 0 25 0 22 0 025 0 022
l0 000 l 00 5 00 0 010 0 050
PCE by ASTM Int'l (all rights reserved);
Copyright 1 000 Mon Feb 24 23:07:03 EST
7 2020
31 9 04 0 731 0 904
Downloaded/printed by l0 000 92 70 79 63 0 927 0 796
TABLE 4--Contmued
Extract Concentration,
Level of mg/L Organic Recovery, %
Spike Added,
Sludge mg/L EP TCLP EP TCLP
WTC 1 000 0 10 0 20 0 010 0 020

10 000 5 00 5 00 0 050 0 050
TETRACHLOROETHENE
WES 1 000 3 10 7 00 0 310 0 700
10 000 25 97 38 67 0 260 0 387
PCE 1 000 3 03 3 19 0 303 0 319
10 000 28 30 13 37 0 283 0 134
WTC 1 000 1 00 1 60 0 100 0 160
1887 39 87 0189 0399
TOLUENE
WES 1 000 3 03 4 43 0 303 0 443
10 000 55 43 93 67 0 554 0 937
PCE 1 000 1 37 2 50 0 137 0 250
36 67 35 77 0 367 0 358
WTC 1 000 1 24 1 39 0 124 0 139
10 000 65 67 89 57 0 657 0 896
ETHYLBENZENE
WES 1 000 5 27 17 33 0 527 1 733
10 000 33 83 47 33 0 338 0 473
PCE 1 000 2 03 2 33 0 203 0 233
10 000 34 53 20 93 0 345 0 209
WTC 1 000 2 93 3 94 0 293 0 394
10 000 36 10 95 60 0 361 0 956
BUTANONE
WES 1 000 35 80 17 00 3 580 1 700
10 000 188 00 256 67 1 880 2 567
PCE 1 000 5 19 5 39 0 519 0 539
10 000 133 33 134 33 1 333 1 343
WTC 1 000 9 59 6 29 0 959 0 629
10 000 163 00 165 67 1 630 1 657
4-METHYL-2-PENTANONE
WES 1 000 41 33 13 33 4 133 1 333
10 000 192 67 313 33 1 927 3 133
PCE l 000 11 63 10 63 1 163 1 063
10 000 233 00 247 00 2 330 2 470
WTC 1 000 7 67 4 88 0 767 0 488
298 00 306 00 2 980 3 060
ficatlon processes, o n e b a s e d o n p o r t l a n d c e m e n t a n d fly ash, the o t h e r o n lime kiln dust a n d

fly ash B o t h are c o m p a r e d with a b l a n k w h e r e i n t h e organic materials were m i x e d with n o n -
reactive solid m a t e r i a l to s i m u l a t e the solidification m i x i n g process T h i s work a n d a d d i t i o n a l
field tests are b e i n g used b y E P A to d e v e l o p regulations a n d g u i d a n c e o n the control o f V O C
e m i s s i o n s d u r i n g S/S t r e a t m e n t processes used at R C R A T S D facilities
Recently, research with m o d i f i e d clay has s h o w n p r o m i s e for a strong b o n d i n g b e t w e e n
semivolatlle c o m p o u n d s a n d o r g a n o p h l h c clay m a t e r i a l [9] T h e e x t r a c t i o n data presented in
T a b l e 5 a n d b o n d i n g e v a l u a t i o n t e c h n i q u e s Indicated strong b o n d i n g
28 HAZARDOUS WASTES
9 phenol
40- 9 3-chloropheno[ "lO0
9 2 , 3 dichorophenol
E m
-80 .~
ua 3 0
o
:60 0
z
/
i
Z
20
I11
94 0 "n
O -11
Q.
oo IO m
92 0 z
.J
O
Z
W I' I I I I I I I I
.1-
Q. I 2 3 4 5 6 7 8 9
TI ME/DAYS
F I G . 2--Leach tests on cement containing charcoal with adsorbed phenol, 3-chlorophenol, and 2,3-
dichlorophenoL
Recommendations for Determining Appropriateness of S / S for Organics

The following recommendations are based upon current knowledge about using S/S for
treating organics. Generally, S/S should not be used to treat a site containing only organic
waste. Alternative technologies (for example, incineration, steam stripping, vacuum extrac-
tion, and so on) should be used to remove and/or destroy the organics. If residues remain after
this primary treatment that still require additional treatment, S/S can be effectively used. How-
ever, a well-designed and controlled treatability study should be conducted to assess S/S effec-
tiveness and to select and design a proper S/S process.
I00
90- Lime Kiln D u s t ~
B0-
Z 70
0
60
50-
UJ 40
30
20
I0
0
30 60
M I NUTES
Copyright by ASTM Int'l (all rights reserved); Mon F
Feb
I G24. 23:07:03 EST 2020
3--Cumulative acetone emissions.
TABLE 5--Extraction data upon waste stabthzed with organophthc clays
Raw Waste Mean

Concentration, Treated Waste Mean
Compound ~g/kg Concentration, ~g/kg
Bls-(2-chlorolsopropyl) ether 8 528 ND a
Naphthalene 18 060 1445b
Phenanthrene 20 184 ND
Benzo (a) anthracene 30 460 ND
a ND = not detected
b Value corrected for dilution
On strictly technical considerations, there may be exceptions to this general approach For
example, if the organic IS generally not mobile through air, soil, and water [for example, poly-
chlorinated biphenyls (PCBs)] then some approaches to S/S may be acceptable cost-effective
treatment alternatives for a given site, depending on the remedlation goals Careful attention
must be paid to any existing state and federal environmental regulations concerning the par-
tlcular organic contaminant (for example, dloxIns, and so on) Treatability studies must be
performed that incorporate appropriate test methods to evaluate the organic waste's potential
for escape
Based on existing data, volatile organic compounds cannot be immobilized by current S/S
technology Whether or not a site containing VOCs can be treated by S/S will depend upon
specific conditions existing at the site. Available data also indicate that semlvolatlle organic
compounds cannot be effectively treated by current S/S techniques Whether or not a site con-
taining semlvolatlle organics, along with other wastes, should be treated using S/S also depends
upon site-specific factors.
Figure 4 provides a general decision tree for determining if S/S should be considered for
treating a site containing organics The approach is as follows
1 Determine the amount of organics (relative to inorganic contaminants and other mate-
rials) and their chemical and physical characteristics Determine whether the organics of
concern are above acceptable levels (for example, above regulatory limits or above levels
causing unacceptable risks to exposed populations)
2 Based upon this Information and other site characterization data, further determine what
inorganic compounds, materials, and so on, would remain if all the organics were
destroyed or removed
3 Determine the concentrations and the chemical and physical characteristics of the resid-
uals remaining if organics were removed/destroyed.
4, If the site contains only organics, do not consider S/S unless the organics are not mobile
(in air, water, or soil) and do not require special handling~disposal~treatment because of
special regulations (for example, TOSCA, and so on) or the organics are below some pre-
determined acceptable level based upon a risk assessment.
a. If organics are not mobile and do not require special treatment and management
because of environmental laws, unacceptable risks, or other special considerations,
consider S/S.
b If organics are volatile or semlvolatlle, are hazardous and are above levels of concern,
then they must be destroyed and/or removed prior to S/S However, depending upon
the amounts of volatile and/or semivolatile organics present, it may be feasible to col-
Copyright by ASTM Int'land
lect (all rights
treatreserved); Mon Feb 24 23:07:03
the organics as partESTof
2020
the S/S treatment train.
30 HAZARDOUSWASTES
Does The SIl"e [-.n-~ S/S I

Con'rain Organics ? [ l "
t yes
Are Orgonlcs Above [ no ]
Levels of Concern' ?I "
yes J "rf Orgonlcs Are J
Does The Sll"e I Removed Wlll l
I Con'raln Only no ~ Reslduols Requlre; no =:S/S NOt NeededJ
JOrgonlcs ? Treol"ment ? l
yes
I y.,
JAre The Organics J ,,,, ]' ] Consider SIS Wlth
IMobile,Volol"ile, ' "" ~ ~; Proper Pre'rrea1"men'r
I Semi-Vola'l'ile, J ]" J And/Or Corrtrols
I or Soluble ? I I ye' po. [ As Required
JAre There Candldo're Ires I Per'~orm Treo*ob.i*yJ
IS/S Processes For ~ Study For Specl'flc J
ITreol"lng Such I J S/S Process(es) I
JOrgonics
no
LDo No1- Consider S/S
FIG 4--General decision treefor S/S apphed to orgamcs
c Iforgamcs are mobile and/or volatile but not hazardous (which is unhkely at Super-
fund sites) then S/S is a potentially feasible treatment
Determining Levels of Volatile/Semivolatile Organics Acceptable to S / S

There are two approaches for determining the m a x i m u m amounts of volatde/sem~volatfle
orgamcs acceptable for S/S as the whole treatment or as part of a treatment tram One ~s based
on results of risk assessment and an assumption that S/S will not prowde treatment The sec-
ond Is based on results of a treatabdlty study and a comparison of orgamc levels m the
untreated waste w~th those m the sohd]fied waste using a total waste analys~s or s~mdar tests
The first approach, based on risk assessment, ]s as follows
1 Determine the concentration of the orgamcs present m the waste to be treated

2 ldenUfy the compound that poses the h~ghest health or environmental risk (quantxty and
tox]c]ty)
3 Determine a threshold concentration of the highest risk compound below which It will
not pose a health or environmental risk at the given s~te
4 Assume that
a The S/S process will not treat or contain the selected compound and therefore all of
it will be released from the sohd]fied waste, and/or
b That all of the compound will be released as air emlss]ons during the S/S process
5 Ifthese concentrations are above those levels determined to pose a health/environmental
risk at the given site, then pretreatment to remove or destroy the orgamcs will be reqmred
and/or mr em]sslon controls and treatment wdl be needed on the S/S treatment tram
Regulatory hmits have not yet been established but could be based on some muluples of
the d n n k m g water standards, on limits being established by EPA for aw emission con-
trols, or on some other standard
This approach is very conservatwe and does not account for the retardation of the release
of organics caused by the physical charactenstlcs of the sohdified waste form A second
approach, based on "harsh" extraction test concepts, can be envisioned as follows
1 Use a total waste analysis, methylene chloride, or similar harsh extraction procedure to
determine the organic content of the untreated waste
2. Perform the S/S treatability study, but ensure that proper controls are used to collect all
organic air emissions Continue collection of air emissions d u n n g the cunng phase
3 Perform extractmn/leachmg tests on sohdified waste forms after curing Collect organic
air emissions
4 By companng the organic levels m the untreated waste, the collected air emissions, the
sohdlfied waste form, and in the extract from the extractmns test, one can determine if
acceptable treatment has been achieved and also whether VOC emission controls will be
necessary d u n n g full-scale apphcation of the S/S process
5 Compare the levels oforgamcs in the extract from the solidified waste form to some pre-
determined level (for example, 100 umes drinking water standards) estabhshed as accept-
able for the s~te If the levels are acceptable, then S/S can be used However, one must
also determine ffcontrols are necessary to collect and treat VOC em~ssmns d u n n g the S/
S process
This second approach 1s also conservative, but will give some credit to any S/S processes
which claim to be able to treat organics Because of the grinding reqmred for the harsh extrac-
tion tests, this approach provides little credit for physical retardaUon of releases of the organics
that may result from the improved physical properties of the sohdified waste form
Summary
This paper discussed some approaches for determining whether or not organic contami-
nated soils should be treated by S/S technologies The approaches are conservative and give
little recogmtlon to the physical characteristics of the solidified waste forms in the lmmoblh-
zatlon process. These approaches are also based upon technical rather than regulatory consid-
erations after reviewing available information on the S/S of organic wastes Several instances
have been reported where processors have claimed treatment of organics by S/S In most, if
not all of these, the experimental approach was too limited Measunng organic content before
treatment and after treatment without controls to collect and analyze air emissions ~s not
acceptable Many, if not all, of the volatile and semlvolatile organics will "disappear" d u n n g
the process because they volatilize Much more sound scientific evidence is reqmred before S/
S oforgamc contaminated waste can become routine practice The EPA is addressing this issue
as well as the development of reasonably quick and mexpenswe protocols for conducting S/S
treatability studies mvolwng organics
Also presented were some general rules for conducting treatability studies for metal-beanng
wastes The need for this has become evident in revlewing many studies, including data from
the conduct of treatability studies at Superfund sites The purpose of these rules is to improve
the quality and apphcablhty of the data
32 HAZARDOUSWASTES
Acknowledgments
The authors wish to thank Dr Joseph T Swartzbaugh, PEER Consultants, P C , and Dr
Jeffrey Means, Battelle Memorial InsUtute, for their assistance m prepanng th~s paper
References
[1] C6t6, P L,Caldwell, R,andChao, C C,"PhyslcalandChemlcalContammentofOrgamcContam-
mants m Sohdtfied Waste," Waste Management, Vol 10, 1990, pp 95-102
[2] Shlvely, W E and Crawford, M A m EnvtronmentalAspects of Stablhzatlon and Sohdtficatton of
Hazardous and Radtoacttve Wastes, ASTM STP 1033, P L C6t6 and T M Gllllam, Eds, American
Society for Testing and Materials, Phdadelphla, 1989, pp t 50-169
[3] Stegemann, J and C6t& P L, "Investlgahon of Test Methods for Sohdlfied Waste Evaluatlon--A
Cooperative Program," Report EPS 3/HA/8, Wastewater Technology Center, EnwronmentCanada,
Burhngton, Ontario, January 1991
[4] Bncka, R M, Holmes, T , and Culhnane, M J, "An Evaluation of Stablllzatlon/Sohdlficatlon of
Metal Hydroxxde Sludge (F006)," Techmcal Report EL-88-XX, U S Army Engineer Waterways
Experiment Station, V~cksburg, MS, 1987
[5] Kolvltes, B and Bishop, P, "Column Leach Testing of Phenol and Tnchloroethylene Stabdtzed/
Sohdtfied with Portland Cement," Envlronmental Aspects of Stabthzatton and Sohdtficatton of Haz-
ardous and Radtoaetzve Wastes, ASTM STP 1033, P C6t6 and M Gflllam, Eds American Society
for Testing and Materials, Philadelphia, 1989, pp 238-250
[6] Brlcka, R M, Holmes, T T , and Culhnane, M J, "A Comparative Evaluatmn of Two Extraction
Procedures The TCLP and the EP," USEPA, Cincinnati,OH, m press, 1990
[ 7] Sheriff, T S, Sollars, C J, Montgomery, D, and Perry, R, "The Use of Activated Charcoal and
Tetra-Alkyl Ammomum-SubstltutedClays m Cement-Based Stabfl~zat~on/Soh&ficatlonof Phenols
and Chlonnated Phenols," Envtronmental Aspects of Stabthzatton and Sohdtficatton of Hazardous
andRadloacttve Wastes, ASTMSTP1033, P L C6t6and T M Gflham, Eds, ASTM, Phfladelphm,
1989, pp 273-286
[8] Weltzman, L, Hamel, L, and Cadmus, S, "Volatile Emissions from Stabilized Waste," final report,
U S EPA Contract 69-02-3993, WA 32 and 37, Risk Reduction EneAneenngLaboratory, Environ-
mental Protection Agency, Cincinnati, OH, 1988
[9] Soundararajan, R and Barth, E, "Using an Organophlhc Clay to Chemically Stabilize Waste Con-
taming Orgamc Compounds," Hazardous Matertals Control Journal, Vol 3, No 1, 1990
Rengarajan Soundararajan ~
Guidelines for Evaluation of the Permanence

of a Stabilization/Solidification Technology
REFERENCE: Soundararajan, R , "Guidelines for Evaluation of the Permanence of a Stabili-
zation/Solidification Technology," Stabtllzatton and Sohdtficatwn of Hazardous, Radtoacttve,
andMzxed Wastes, 2nd Volume, ASTMSTP 1123, T M Gflham and C C Wiles, Eds, Amer-
ican Society for Testing and Matenals, Phdadelphm, 1992, pp 33-39
ABSTRACT: Stabfllzatmn/sohdtficatmn technologms are used widely for treating both lnor-
gamc and orgamc waste matenals Unfortunately, some of these processes have turned out to be
simple adsorption/dilution phenomena For a true stablhzatlon/solldlficatlon process, the
binder and the waste must interact chemically to create chemical bonding Based on our past
lnvestlgatmns in this area, some guldehnes for evaluating the permanence of a stabilization/
solidification process are presented m this paper
KEY WORDS: stablhzatlon, sohdlficatlon, adsorptmn, ddutmn, chemical bonding, Fourier

transform infrared spectra (FTIR), thermogravlmetnc analysis (TGA), &fferentlal scanningcal-
onmetry (DSC), bond energy, heat ofsolutmn
Stabdlzatlon/soh&ficatmn (S/S) technologies are emerging as viable alternatives for other

waste management processes. However, there have been a number of questmns rinsed about
the technology's effectweness and permanence by the Office of Technology Assessment. Sev-
eral reports critical of the "science" revolved m th~s process have been pubhshed m the recent
past. It becomes lmperatwe to take a hard look at the technology from a purely scmnUfic stand-
point and to gather evidence about the permanence of this process in order to preserve the
integrity of the technology. In the past, we have conducted a series of mvestlgauons on various
aspects of S/S technology. Based on our findings, we present a few protocols that would assess
the permanence of an S/S process Most of the research work presented in this paper was
funded by the Office of Research and Development of the U S Environmental Protectmn
Agency (EPA), CmcmnatL OH.
Discussion
It is worth mentioning that all S/S technologies are governed by the fundamental laws of
chemistry. A thorough understanding of the chemistry of the waste to be stabdized results m
a better staNhzatlon process In general, the S/S process of morgamcs and organics may be
explained as follows In the inorganic S/S process, the metallic 1on to be fixed is converted into
its most insoluble form and then placed reside a cementmous matrix [ 1]
In the case of staNllzatlon of lead, for example, the following facts were considered The
soluNhty product of lead hydroxide [Pb(OHh] is known to be 1 2 • 10-15 We need this lnfor-
l Director, Research and Development, RMC Environmentaland Analytical Laboratories, 214 West
Mare Plaza, West Plains, MO 65775
33
Copyright9 1992byASTM International www.astm.org

34 HAZARDOUSWASTES
marion since m an alkaline medmm the logical end product of any lead compound would be
Pb(OH)2
The pH of most stabilized waste forms m a cementmous matrix has been found to be around
11 to 12 Assuming that m a gwen snuatlon the pH is 11, then the pOH can be calculated from
the relationship
pH + pOH = 14
pOH = 1 4 - 11 = 3
The concentratmn o f ( O H - ) ions can be found from pOH, which is 3 as 0 001 mol/L
The solubility product constant may be defined as a product of concentrations of ions
(where the n u m b e r of moles are raised to the power of concentration) that are m equlhbrlum
wxth a precipitate Hence, ffthe solubility product is exceeded m some way, say by the addition
of more hydroxyl (OH-) Ions, all the lead can be precipitated as Pb(OH)/
According to the e q m h b n u m
Pb(OH)2 ~- Pb 2+ + 2OH-
The solubdlty product K, v can be written as
Ksp = [Pb2+1 [OH-] 2
where
K,p= 1 2 • 10 -~5,and
O H - --- [0 001]
1 2 X 10 -15
therefore, Ksp (Pb 2+) (0 001 )2 and [Pb 2+] - - 1 2 X 1 0 -9 mol/kg
[0 001] 2
The atomic weight ofPb ~s 207 2, therefore the amount ofPb m the solution will be
1 2 X 10 -9 mol • 207 2 = 2 48 X 10 7 g/L --~ 0.25 parts per bilhon
This is an acceptable value at the present time by any regulatory agency Based on the above
discussion, a portland cement-based binder that had a pH of 12 was used to successfully sta-
bihze lead A major constderat~on in lead S/S chemistry is that if the pH reaches values beyond
13, then the Pb ions will transform mto anionic PbO 2- as shown below
Pb(OH)2 + 2OH- ~ [Pb(OHh] 2-

[Pb(OHh] 2- ~ PbO~ + 2H20
Most alkah metal plumbates are water soluble Due to this phenomenon, formulations that
contained alkali metal slhcates failed the leach tests such as the extraction procedure toxicity
(EP TOX), toxicity characteristic leaching procedure (TCLP), multiple extraction procedure
(MEP), and so on
On the other hand, m the organic S/S process, the binder contains one or more organic com-
pounds and these compounds preferentially extract the waste from any medium such as sod,
water, and so on into the binder matrix. This process is somewhat similar to a chromato-
graphic process where the analytes are preferentially held m the stationary phase (binder
SOUNDARARAJAN ON STABILIZATION/SOLIDIFICATIONPERMANENCE 35
matnx) Once the wastes are drawn into the binder, they can chemically bond with various
components of the binder [2] The cementltlOUS matter in the binder helps to solidify the
binder-waste matter into a tough solid with low permeabihtles, minimizing the leaching of
waste matenals Formation of the most insoluble form of the metal and creations of bonds
between the waste and the binder ensure that an inorganic/organic waste form is stabilized
There are a number of modern analytical tools available today to verify these facts They are"
(1) X-ray diffraction, (2) Fourier transform infrared (FTIR), (3) thermogravlmetrlc analysis
(TGA), and (4) differential scanning calorimetry (DSC) Two leach tests, namely the TCLP
and MEP, can be used to understand (to a limited extent) the leaching behavior ofinorganics 2
Drastic extraction by a solvent such as hexane or methylene chloride can reveal the extent of
stabilization for organic wastes The application of each technique mentioned above is dis-
cussed in subsequent paragraphs
X-Ray Diffraction
X-ray diffraction patterns are used as fingerpnnts for identifying most inorganic com-
pounds In an inorganic S/S process, the metal is converted into its most insoluble form Such
conversion can be confirmed by the X-ray diffraction patterns of the stabilized waste and com-
paring it with a standard
Fourier Transform Infrared

All the molecules have vibrational frequencies in the Infrared region of the electromagnetic
spectrum However, when a molecule interacts with some other molecule such as an organo-
phlllc binder, the vibrational frequencies of the bound molecule differ from that of the free
molecule The extent of the frequency shift IS a measure of the interaction between the haz-
ardous waste molecule and the binder We beheve that at least a 30-cm- ~shift would be nec-
essary to indicate a reasonable interaction since smaller shifts could indicate simple Vander
Waal's attractions, which are not true chemical bonds The experimental aspects are discussed
elsewhere [5] The following two examples from our recent investigations best illustrate this
concept
Trlchloroethylene
There is again a posmve shift in the C- -~CI absorption frequencies, as was seen in the case
ofchloroanlhne The characteristic ethylene frequency and the C---~H stretching frequencies
are both lowered Apparently, intercalation causes the former, and residual hydrogen bonding
causes the latter (Fig. l, Table 1)
2 In 1984, Congress modified the Resource Conservation and Recovery Act (RCRA), passed m 1976,
with the Hazardous and Sohd Waste Amendments (HSWA) These amendments provide an analyhcal
method that would serve to monitor the success ofdetoxlficatlon techmques This method involves mul-
tiple extractions with solvents and IS called the toxicity characteristic leaching procedure (TCLP) EPA
reqmres the TCLP to be performed on each type ofsohd waste to determine whether treatment is required
The rule establishes maximum allowable concentrations for 36 chemicals that may be contained in the
extracts If test results meet TCLP standards, the waste involved will not be restncted or regulated and
may be disposed of in an RCRA-permltted facility IfTCLP results exceed standards, the waste must first
be treated by the technology specified m the rule (or by an equivalent method) and the residue must again
be subjected to TCLP
36 HAZARDOUS WASTES
"k 7"
..../ SlO 2
k:',
FIG l--Posstble mteracttons between trtchloroethylene and clay matrtx
Trlethanolamme
In the case of tnethanolamlne, we are able to establish the presence of a coordinate covalent
bond between the lone pair of electrons on the nitrogen and the matrix (aluminum) The C --~
N stretching frequency has dramatically shifted by 193 cm -~ toward the positive direction,
which is very characteristic of amine salt formation Protonatlon of organic compounds by
clay minerals is a well-known reaction It appears that there is a combined effect of this coor-
dination bond with aluminum as well as with a proton, which produces such a large shift We
also see a shift in the negative direction in the alcohol-OH stretching frequency This phenom-
enon is due to the hydrogen bonding of the hydroxyl hydrogen with the substrate oxygens (See
Table 1 and Fig 2) In both cases, the compounds in discussion were mixed with a binder
containing organophihc clays in the absence of water
Thermogravimetric Analysis and Differential Scanning Calorimetry

These two thermoanalytical techniques provide us with the information on the energetIcs
revolved in evaporating a stabilized waste from Its matrix
Should the stablhzatIon be due to adsorption/dilution there will not be any significant
change in the heats of evaporation (AHv.p) and the respective boiling points On the other
TABLE 1--Infrared frequencies before and after treatment, cm-l
Tnethanolamme +
Tnethanolamlne Treatment Material Shift Peak Assignment
2104 2297 193 Amine salt formation

TRICHLOROETHYLENE TRICHLOROETHYLENE
-1- T R E A T M E N T
MATERIAL
641 660 19 C'--~CI stretch
2286 2281 -5
2648 2611 --37 C---~H stretch
3057 3047
- 10 Ethylene (characteristic)
SOUNDARARAJAN ON STABILIZATION/SOLIDIFICATION PERMANENCE 37
Ab03 LAYER
A I . , H*
II 9 ii,e oi
HO .... C H 2 ~ C H 2 ' N -- CH2--CH2-- OH

O0
" I
CH2
I
CH2
1 80
H.O.
SIO2 LAYER
FIG. 2 - - Triethanolamine.
hand, if there are chemical interactions there will be a considerable increase in the heats of
vaporization and also changes in the boiling points. These concepts can be illustrated by the
following examples. We believe that a m i n i m u m of 30 to 50% energy increase will be required
to indicate a true stabilization process, as several chemical bonds between the binder and the
waste will have to be broken before the waste could evaporate by the application of heat. In
the case of nitrobenzene (Fig. 3), four endotherms are observed. The endotherms do not in
wT, 13.~2 m9
SCAN RATE, I0.00 dag/mim
RP-OI-23/NITROBENZENE (#~
PEAK FROM: 81.3O TC: 18~.75

ONSET: 9 8 . 2 9
CAL/GRA~: 5.51
HAX/HI~ I I 0 . 9 e 9
LJ
L~J
I. 5B PEAE FR~MI 110.85 TO: 1 8 8 . 7 ~ / f
~NSET: 112.42 /
CAL/GRAHs .E8
MAX/MIN: I i g . 415 ~
PEAK FROMa I~8.78fO= 153.67
ONSET= 129.2S /
PEAK FROM. 189.15 TO: 2 3 4 . B7

ONSET~ 197.39
CALICRAM~ 12.7
MAX/HIN~ 218.453
IiO0
320. r~n 3 ~ r*n 400. nn a~O.
075. ,v2 , 4 F M FILE, O5OZS.. DA TEMPERATURE (C) OSC
FIG. 3--Nitrobenzene.
38 HAZARDOUSWASTES
TABLE 2--Dfferentml scanmng calortmetry data

AH of Highest
Vaporlzatmn ObservedAH of Percentage Bolhng Endothermm
Endotherms, (Literature), Vaporization, Increasein P o i n t , Temperature,
Compound *C Kcal/mol Kcal/mol Energy *C *C
Nltrobenzene 110 99
119 42 12 17 18 60 52 8 210 2 218 45
138 87
218 45
Tnethanolamme 150 97
337 30 12 78 24 16 89 0 335 4 337 30
Tnchloroethylene 105 05
127 92 9 18 34 43 275 9 113 8 204 26
204 26
any way correspond with that of the treatment material itself, and the total energy revolved in
the treated matrix is higher than the heat of vaporization of mtrobenzene The same thing is
true m the case oftrlchloroethylene (See Table 2 and Fig 3)
Gas Chromatography/Mass Spectrometry Analysis on Volatile Emissions during an S / S

Process
In several instances, adding water to cementmous materials has been observed to bnng
about exothermlc reactions, and a number of volatile compounds are lost during mixing itself
Also, during an S/S process, a number of polynuclear aromatics, polychlorlnated blphenyls
(PCBs), and so on are converted into volatile aromatics and are evaporated into the atmo-
sphere After the S/S process, the original compounds were not found in the matrix, not
because they were stablhzed but because they were lost due to volatlhzatlon This can be con-
firmed by running the S/S process on a laboratory scale reside a glove bag and analyzing the
effluent gases by G C / M S
Differential Scanning Calorimetry/Gas Chromatography/Mass Spectroscopy

These techmques confirm unequivocally that if a truly stablhzed waste is driven from the
matrix by any agency such as heat it breaks down to smaller fragments D u n n g heating, the
weakest bond is broken first (lowest temperature), followed by stronger bonds In the case of
nltrobenzene, whmh is depicted here as an example, neither the molecular ions ( m / z = 123)
nor the benzene ring ( m / z = 77) were seen This suggests that once the hazardous waste is
bonded, it does not reappear m its own original form, instead, smaller fragments result when
forced out At higher temperatures, some rather unusual long chain nitrated compounds were
detected These compounds, of course, were never detected in the D S C / G C / M S run of the
pure treatment matermls before treating them with mtrobenzene
Leach Tests
Leach Test for Inorgamcs
A c o m b m a h o n of TCLP and MEP seems to be adequate to understand the leaching behav-
ior o f a stablhzed Inorganic waste The multiple extraction procedure revolves the use of both
mechanical tumbling and a drastic solvent (mixture of nitric and sulfuric acids) Further, the
ninebyextractmns,
Copyright filtration
ASTM Int'l (all rights ofFeb
reserved); Mon the24 residue,
23:07:03 ESTre-extraction,
2020 and so on predicts the behavior of the
SOUNDARARAJAN ON STABILIZATION/SOLIDIFICATIONPERMANENCE 39
sohdlfied matrix (whether it crumbles or stays intact) as well as the leaching behavior of the
pollutants Although these tests do not represent the site conditions, these two seem to provide
more useful information than the other leach tests such as the American Nuclear Society
(ANS) 16 l, Batch Extraction Procedure, that are available at this hme.
Leach Test for Orgamcs

As has been observed on many occasions, a chemically bonded waste is not easily leached
from the matrix On the other hand, if the waste is simply adsorbed on the surface, any good
solvent such as hexane or methylene chloride can extract the waste very easily Such a leach
test can be used as an indicator for the permanence ofa S/S process From our experience, we
feel that a good binder should retain at least 30% of the waste even d u n n g a drastic solvent
extraction Thls is because when the leaching process takes place there are two conflicting
chemical forces They are (1) the heat of solution (AH~0 of the waste when it dissolves in the
solvent and (2) the bond energy When there is true bonding, the heat of solution is too small
to break the bond so that the waste can be leached In good stabilization/solidificationpro-
cesses, we have seen up to 98% retention of the waste in the matrix d u n n g solvent extraction
[6] The 30% threshold is more of a passing grade requirement for a stabilized organic waste
Conclusions
During the evaluation of a stablhzation/sohdlficatIon binder, the following guidelines may
be followed to ensure the permanence of the process
1 The Fourier transform Infrared spectra (FTIR) of the treated and untreated waste should
show a shift to a lower frequency of at least 30 cm-1 to indicate modest chemical bonding
after treatment
2 Thermogravlmetrlc analysts and differential scanning calorimetry should show a mini-
m u m of 30 to 50% energy increase in their AH values to prove chemical interaction.
3 Differential scanning calorimetry/gas chromatography/mass spectrometry studies can
further confirm the positive interaction between the binder and the waste
4 For inorganic waste, TCLP may be used in conjunction with the MEP
5 For organics, drastic extraction tests should be used, and the binder should retain at least
30% of the orgamc waste
References
[1] Soundararajan, R and Dixon, W, "Final Report of Sapp Battery Site-Fixation of High Level Lead
m the Soil," U S EnvironmentalProtection Agency, Region VI, Atlanta, GA, September 1989
[2] Cotton, F A and Wilkinson, G , Advancedlnorgamc Chemistry, 4th ed, John Wiley & Sons, New
York, 1980
[3] Wendlandt, W W, ThermalMethods ofAnalysts, 2nd ed, John Wdey & Sons, New York, 1974
[4] Test Methodsfor Evaluatmg Sohd Waste, SW-846 Methods, 3rd ed, U S EPA, Washington, DC,
U S EnvironmentalProtection Agency, Nov 1986
[5] Gibbons, J J and Soundararajan, R, "Nature of Chemical Bonding between Modified Clay Minerals
and Organic Waste Materials," Part I, American Laboratory, July 1988, "The Nature of Chemical
Bonding between Organic Wastes and Organophdlc Binders," Part II, American Laboratory, July
1989
[6] Soundararajan, R, Barth, E F, and Gibbons, J J, "Using an Organophlhc Clay to Chemically Sta-
bilize Waste Containing Organic Compounds," Hazardous Materials Control Research Institute
Magazine, January/February 1990
Specialty Wastes: Organics, Ashes, and
Resins
Robert E. Brown, 1 Bipender S. Jindal, ~ a n d Joseph D. B u l z a n I
A Critical Review of the Effectiveness of

Stabilization and Solidification of Hazardous
Organic Wastes
REFERENCE. Brown, R E , Jindal, B S, and Bulzan, J D , "A Critical Review of the Effec-
tiveness of Stabilization and Solidification of Hazardous Organic Wastes," Stablhzatlon and
Sohdlficatgon of Hazardous, Radioactive, and Mtxed Wastes, 2nd Volume, ASTM STP 1123, T
M Gilham and C C Wiles, Eds, American Society for Testing and Materials, Philadelphia,
1992, pp 43-60
ABSTRACT: A cntlcal review is presented on the stabdlzatxon/sohdlficatmn (S/S) of hazardous

organic wastes (WASTES) The hterature on the S/S of WASTES with cementing agents con-
raining expandable lattice clay binders modified by catmn exchange with quaternary ammonium
catmns (MODIFIED CLAYS) suggests that hazardous organic waste constituents (ORGANICS)
may form permanent chemical bonds preventing future releases of ORGANICS into the envi-
ronment This conclusmn is compared with basic studies on the sorptmn of ORGANICS from
aqueous solutmns by soils and MODIFIED CLAYS This latter hterature would suggest that
sorptmn of ORGANICS from aqueous solutmns IS an equlhbnum process wherein the MODI-
FIED CLAYS buffer the concentratmn of ORGANICS m the equlhbnum aqueous solutmn
rather than a process forming permanent chemical bonds, thereby preventing the movement of
ORGANICS mto the enwronment Often the buffered concentratmn of ORGANICS m the
equlhbnum aqueous solution is far m excess of dnnklng water standards and therefore S/S
WASTES are not necessarily environmentally safe Dlscussmns are presented on difficulties m
relating data on the sorptlon of ORGANICS by MODIFIED CLAYS to conclusmns regarding
possible retentmn of ORGANICS against releases to the envxronment Overall consxderatlons
are then g~ven to the practice of S/S of WASTES at treatment, storage, and disposal (TSD)
faclhtles
KEY WORDS: stablhzatlon and solidification, hazardous organic wastes, quaternary ammo-
n m m modified clays, sorptlon and desorptlon eqmhbrlum
T h e O h i o H a z a r d o u s W a s t e Faclhty Board, with t h e O h i o E n v i r o n m e n t a l P r o t e c t i o n

Agency, h a n d l e s t h e O h i o h a z a r d o u s waste p e r m i t process F o r a p e r m i t to b e issued, t h e Board
m u s t d e t e r m m e w h e t h e r t h e p r o p o s e d o p e r a t i o n o f a t r e a t m e n t , storage, a n d disposal ( T S D )
f a o l l t y represents t h e m l m m u m adverse e n w r o n m e n t a l impact, c o n s l d e n n g the state o f avail-
able technology, t h e n a t u r e a n d e c o n o m i c s o f various alternatives, a n d o t h e r p e r t i n e n t con-
sideratlons T h e fact t h a t m a n y p e r m i t a p p h c a t l o n s request t h e a u t h o r i t y to treat h a z a r d o u s
organic wastes ( W A S T E S ) by s t a b l h z a t l o n / s o h d t f i c a t l o n (S/S) necessitated a n i n - d e p t h review
o f S/S by t h e B o a r d staff
As will be & s c u s s e d in m o r e detail below, the S/S o f W A S T E S is receiving considerable
a t t e n t i o n as a possible m e t h o d for r e c l a m a t i o n o f S u p e r f u n d s~tes u n d e r t h e U S E n v i r o n -
m e n t a l P r o t e c t i o n A g e n c y ' s (EPA) S u p e r f u n d I n n o v a t i v e T e c h n o l o g y E v a l u a U o n P r o g r a m
(SITE) In general, t h e v o l u m e s o f m a t e n a l s to be treated w i t h o u t g e n e r a t i n g toxic fugitive dust
1 Chief technical advisor, environmental manager, and environmental manager, respectively, State of
Ohio, Hazardous Waste Faclhty Board, P O Box 1049, Columbus, OH 43266-0149
43
Copyright 9 1992 by ASTM International www.aslm.org

44 HAZARDOUSWASTES
or vapor emissions during a Superfund site remedlanon greatly restrict the options available
for cleanup of such sites. In sltu S/S treatments and on-site mixing of S/S treatments with
contaminated smls is an opnon that presents some advantages m materials handhng and con-
trolling fugmve dust emissions during the site treatment Certmnly, S/S is an option that must
be evaluated m the search for effecnve and economical ways for Superfund site remedtation
Questions anse, however, regarding the extension of S/S expenences at Superfund sites to
treatment processes at TSD facilities under the Resource Conservation and Recovery Act
(RCRA) regulations and the land-ban restncnons Many land-ban restrictions are concentra-
tion-based treatment standards that have been based upon results achieved with the best dem-
onstrated available technology (BDAT) Other land-ban restncnons are based upon the haz-
ardous organic waste constituents (ORGANICS) content of the tOXloty charactensnc leaching
procedure (TCLP) extract of the WASTES or specified treatment technologies. Stabihzatlon/
sohdlficatlon has not been specified as a technology for the treatment of WASTES containing
O R G A N I C S The question ~salso raised In this paper regarding the abihty of the S/S treatment
to render WASTES in comphance with land-ban O R G A N I C S concentranon hm~ts or TCLP
extract concentration hmlts
For those WASTES that are land banned because of the O R G A N I C S concentration, the
rules do not require that the BDAT be used for processing WASTES, however, the O R G A N -
ICS content of treatment residues cannot exceed those obtained using BDAT In contrast, the
objecnve of an S/S treatment is to bind O R G A N I C S in a manner that would restnet their
release to the environment, but it ~s not generally expected to reduce the concentration of
O R G A N I C S in waste residues other than by dllunon or by the possible loss of ORGANICS
vapors into the air d u n n g the treatment For this reason, if a waste falls land-ban restnctions
because of the concentration of ORGANICS, S/S would be unlikely to render the waste in
compliance with the land-ban restnctions
Some possible mechanisms by which the S/S of O R G A N I C S might render the WASTES m
comphance w~th land-ban rules include the following: 1 the S/S of WASTES might reduce
the O R G A N I C S content o f a TCLP extract by ddution of the waste, by causing the vaponza-
non of O R G A N I C S from the waste, or by binding the waste m a manner that would restnct
the release of O R G A N I C S into the TCLP extract 2 the S/S of WASTES that are land banned
because of the O R G A N I C S concentration limits might satisfy the intent of the land-ban
restnctlons if the bond between O R G A N I C S and S/S binders was sufficient to prevent per-
manently the movement of O R G A N I C S from S/S WASTES into the enwronment The sur-
faces of the SIS binders might actually break down the O R G A N I C S to less toxic materials, or
the S/S treatment might encapsulate the O R G A N I C S such that their presence would not be
detected by a total O R G A N I C S concentration analysis. In consideration of these posslbdines,
a critical review of the literature on S/S of WASTES is presented A cntlcal review is also pre-
sented of basic studies on the sorpt~on of organic compounds by soils and quaternary ammo-
m u m cations ( M O D I F I E D CLAYS) Conclusmns are then based upon these two fields of
hterature
A second issue in the S/S of WASTES at TSD facilities is the increase in volume of waste
solids that must be disposed of m landfills The Increases of waste volumes observed in the
papers to be reviewed will be briefly noted Volume increases will be evaluated in terms of the
general goal of minimization of waste volumes
An Overview of the Literature on S/S of Organic Wastes

A critical review of the literature up to 1985 regarding the S/S of organic liquid wastes and
sludges was completed by Tlttlebaum et al [1], they cite 125 references of both the research
BROWN ET AL ON S/S EFFECTIVENESS 45
hterature and the claims of vendors Their effort will not be repeated in this paper However,
two pertinent conclusions of their paper are as follows.
1 "Few adequately documented studies have been reported on the performance (physical
and chemical stablhty) of solidified and/or stabilized mixtures containing hazardous
organic wastes"
2 "Almost no published information exists on the nature, strength, and permanence of the
bonds formed in the solidification and/or stablhzatlon process."
More recent data are available from the SITE program [2] Under this program, the EPA
sponsored six projects to evaluate the S/S of WASTES The results of only three projects were
available at the time of this writing
1 The results of the Hazcon SITE project were reported by de Percm [3] The results for
volatile and semlvolatlle organic TCLP tests were the same for both stabilized and unsta-
bdlzed soils According to the project report, the interpretation of the data was difficult
because the mmal concentrations of semlvolatlle and volatile organics m sod were less
than l mg/L
2 The results of the Sohdltech SITE project were reported by Grube [4] The objective of
this project was to stablhze a Superfund site contaminated with polychlonnated blphen-
yls (PCBs), lead (with various other metals), and oll and grease using proprietary addi-
tives, pozzolanlc matenals, and water in a batch mixer
The PCB results were inconclusive The PCB content of TCLP extracts were below
detectable limits both before and after the S/S of the WASTES Also, there was uncer-
tainty regarding the fixation of volatile and semlvolatlle organics. Initial concentrations
of volatile and semlvolatlle organics in wastes were very low The maximum volatile
organic concentratlon reported for the untreated waste was only 32 mg/kg No volatile
organics were detected in analysis of treated wastes In a personal communication with
Grube 2, he stated that the concentrations of volatile organics in this study were simply
too low for any defimte interpretation Since most cementaclous and pozzolamc reac-
tions are exothermlc, the low concentrations of volatile orgamcs could have easily been
lost to the air dunng mixing without being detected by the alr-monltonng equipment
which was employed
3 Sawyer et al [5] reported the results of the International Waste Technologies (IWT)
SITE project The objective of the IWT project was to immobilize PCBs by in situ mixing
of the service shop soil contaminated with PCBs with a cement-organo clay mix referred
to as HWT-20 Observations were also made regarding the lmmoblhzatlon of volatile
organics; however, the HWT-20 additive was not designed to immobilize volatile organ-
lcs The concentration of PCBs in the untreated soils ranged from 150 to 950 mg/kg
They concluded that the HWT-20 appeared to immobilize PCBs, but because of the low
values In the TCLP extracts, that could not be confirmed by this project
The concentration ofvolatde organics in the soils vaned from 160 to t485 mg/kg The
concentrations of volatile organics in untreated-soil TCLP extracts ranged from 2 49 to
7 89 mg/kg The concentrations of volatile organics In S/S-treated soils' TCLP extracts
ranged from 0 325 to 0 605 mg/kg They concluded that this reduction in volatile organic
compound (VOC) concentrations may have been due to a combination of factors, the
2 Telephone conversation with Dr Grube on 23 March 1990

46 HAZARDOUSWASTES
main one being the reduction of concentration due to the horizontal and vertical mixing.
This is to be expected in In SltU mixing operation In addition, as stated earher, since
cementatlous and pozzolomc reactions are exothermic, the loss of VOC to the air or dilu-
tion due to the large amounts of additives is also to be expected
The effectiveness of O R G A N I C S S/S is in question not only due to the lack of published
information on this subject but also due to the fact that many available studies did not include
the effect of volatlhzatlon of O R G A N I C S dunng mixing Many of the reactions In the sohdl-
fication of a waste are exothermlc and would therefore provide heat for the vaponzatlon of
O R G A N I C S dunng the mixing of the S/S reagents with WASTES The studies that consider
the effect of volatilization of O R G A N I C S dunng mixing clearly show that volatlhzatlon
greatly impacts the effectiveness of S/S of WASTES Weitzman et al [6] reported that from
80 to 90% of the volatile O R G A N I C S were lost to the air d u n n g S/S of WASTES In a more
detalledeffort, Weltzman et al [7] studIedthe volatdeemlsslonsofacetone, tnchloroethylene,
and n-pentanol during S/S D u n n g a 1-h penod in which these O R G A N I C S were mixed with
lime/fly ash or portland cement/fly ash treatment, 20 to 80% of the ORGANICS were lost to
the mr by vaponzatlon Further losses from the S/S WASTES were also observed over a 28-
day cunng penod
The possible loss of O R G A N I C S by volatdizatlon was also raised as an issue in an evaluation
of the Hazcon process conducted by the American Petroleum Institute [8] The Hazcon S/S
process was compared with a number of treatment technologies Again, it was concluded that
interpretation of Hazcon S/S data was uncertain because volatdes may have been lost to the
air dunng S/S of WASTES
A few recent studies appear to show that the permanent bonding of ORGANICS with
M O D I F I E D CLAYS is possible. The use o f the commercial product HWT-22 provided by
International Waste Technologies and clays modified with quaternary a m m o n i u m com-
pounds of the form R3ArN + (At IS a phenyl group) in the S/S of ORGANICS has been studied
by Gibbons and Soundararajan [9,10] By the use of Fourier transform infrared spectroscopy,
thermogravlmetrlc analysis, and differential scanning calorimetry, a wide range of bonds
between M O D I F I E D CLAYS and a number of ORGANICS were observed Furthermore, it
was reported that when the sorbed molecules were thermally forced from their sorpt~on sites,
the molecules were broken down into simpler substances Soundararajan et al [I 1 ] also com-
pared extractions of O R G A N I C S from M O D I F I E D CLAYS using the TCLP test with extrac-
tions by methylene chlonde They suggested that methylene chlonde extractions provided a
more meaningful Indication than TCLP extracts of the bond between a M O D I F I E D CLAY
and O R G A N I C S
Conclusions Regarding the Literature on the S/S of Organic Wastes

With the exception of work by Gibbons and Soundararajan [9, I0] and Soundararajan et al
[I1], there is very little data to support the claim that S/S can effectwely treat WASTES and
thus provide unqualified protection of the environment by preventing their releases Further-
more, the Weltzman et a l [ 7] observation ofvolatihzatIon of O R G A N I C S dunng mixing and
cunng of treated WASTES is contrary to claims that S/S solid matnx encapsulates O R G A N -
ICS and isolates them from the environment At least in systems commonly employed at TSD
facdlnes, where the fixating agents are usually hme/fly ash or portland cement/fly ash
mixtures, the solid matrix is porous enough for a significant vaporization of ORGANICS dur-
ing mixing and cunng
VolaUhzatlon of O R G A N I C S d u n n g the mixing of WASTES with S/S reagents is not in
itself a concern since such emissions could be captured by appropriate air pollution control
devices However, it is not clear how one could control this feature of the reaction to achieve
a specific ORGANICS concentration in WASTES in order to satisfy land-ban restriction.
Also, there remains the concern regarding further volatilization of ORGANICS from the
treated WASTES long after the waste has been placed in a final disposal site.
It should also be noted that m all the above studies, with the exception of Weltzman et al
[6, 7], the initial concentrations of ORGANICS in WASTES were very low, that is, less than
2000 mg/L In contrast, many TSD facilines seek permits from the Board for the S/S of
WASTES with no limitations on the concentrations of ORGANICS If the results of S/S stud-
]es containing low concentrations are suspect, one would question the effectweness of these
treatments at much higher concentrations, as would be the case at TSD facilities
A Review of Basic Studies Regarding the Sorption of Organic Compounds by Natural Soils
and MODIFIED CLAYS
Equthbrlum Systems
The sorption of pesticides by soils is a major source of information m studying the interac-
tion of ORGANICS and sods. Hamker and Thompson [12] d~scuss the research and theoret-
ical basis for sorptxon of organic constituents in soil systems Some of their findings, relevant
to the topic of this paper, are as follows
1 In the soil system, the most hkely bonding IS of physical sorption/desorptlon rather than
a chemical bonding that could permanently hold the constituent to a clay-organo
complex
2 Sorption and desorpnon IS considered to be an equllibnum process whereto the amount
of organic compound sorbed is m equilibnum with its concentration in the soft pore
water Sorbed matenals may be desorbed if the equilibnum concentration m the soil pore
water is reduced The initial approach to e q u i h b n u m may be a rap~d sorpnon/desorption
followed by a much slower rate requiring several days or longer to reach a true
equihbrlum.
3 There is a high positive correlation between sorption oforganIc compounds and the soil's
humic content Some data were also presented on the sorptlon of organics by clay min-
erals of soils. However, the primary interaction is between organic consntuents and the
soil's humlc contents
In more recent studies, Chiou et al. [13,14] hypothesized that the uptake of nonionic
organic chemicals by soft organic matter from water solutions is essentially a process of par-
tmonlng They studied the equihbna of several nonIonlc orgamc constituents between soil
particles and soil-water solutions They described the equihbna as a partlnon between an aque-
ous phase and an orgamc phase
It IS beyond the scope of this paper to consider whether bonding of ORGANICS by natural
softs and by MODIFIED CLAYS ~s an adsorption or partitioning reaction The word sorption
has been used to represent the Interaction between ORGANICS and binders w~thout any
attempt to define the actual mechanism of the interaction Instead, the direction of this paper
will be to observe that there IS an e q u l h b n u m between ORGANICS sorbed onto the surfaces
of MODIFIED CLAYS or a clay phase and the concentration of ORGANICS in water solu-
n o n or a water phase The environmental sigmficance of this equilibrium for WASTES treated
by S/S is evaluated
48 HAZARDOUSWASTES
Environmental Constderattons
The ability of the natural sotls of Chlou's studies to b m d O R G A N I C S m an environmentally
safe manner is evaluated in Table 1 by companng the concentrattons retained on the solid
phase wRh the land-ban restrictions and companng the concentrations tn equihbnum water
solutions with land-ban restncUons or d n n k m g water standards. A total of eleven nonionic
orgamc consUtuents were the subject of Chmu's study; however, selected data are presented in
Table 1 for only those O R G A N I C S covered under land-ban restnctmns
For example, the m a x t m u m sorptmn of 1,2-dlchlorobenzene by the soil exceeds the land-
ban ltmR of thts O R G A N I C by 816 times as based upon K086 waste Likewise, the equilib-
rium water phase concentration for this O R G A N I C exceeds the land-ban restnctton of F005
wastes by 228 times and d n n k m g water standards by 246 times The m a x i m u m sorption of
1,2,4-tnchlorobenzene by the soil exceeds the land-ban llmtt of this O R G A N I C as based upon
K030 waste by 7 6 ttmes Ltkewtse, the equilibrium water phase concentration for this
O R G A N I C exceeds the d n n k m g water taste and odor limits by 533 times. Fmally, the maxi-
m u m sorpt~on of benzene by the sod exceeds the land-ban hmlt of benzene as based upon
K 104 waste by 87 umes, and the eqmhbrmm water phase concentraUon for benzene exceeds
dnnking water standards by 320 000 t~mes Thus, even though the natural sod sorbs these
ORGANICS, the bonds that are formed do not render the wastes environmentally safe
The sorptlon of pentachlorophenyl (PCP) by hexadecyltnmethyl-ammonmm ( H D T M A +)
modified smectlte from dxstilled water was studied by Boyd et al [15] The results are pre-
sented m Table 2. The sorption of the PCP was as htgh as 31 960 mg/kg, and the equthbnum
concentration m the water phase was 5 3 mg]L. The PCP concentration tn the H D T M A +
modified clay was 864 times land-ban restrictions for this consmuent as defined for wastes
K001, and the concentratton of the e q m h b n u m water phase exceeded dnnking water stan-
dards by 26 times
The sorptmn of benzene and phenol from water solution by smectite modified with
H D T M A + was studted by McBnde et al. [16] Thetr data are used to calculate benzene and
phenol sorption and equtlibnum water phase concentratmns in the respecttve units of mg/kg
and mg/L as presented m Table 2 Details of these calculattons are presented as a footnote to
Table 2 Benzene sorption by the H D T M A + M O D I F I E D CLAYS was about 31 500 mg/kg,
and the equlhbnum water phase concentratton was 287 mg/L. The concentration of benzene
on the H D T M A § M O D I F I E D CLAYS was 5250 times the land-ban restrictions for K I 0 4
waste, and the concentration in the equihbnum water phase was 57 400 t~mes d n n k m g water
standards
T A B L E I - - A compartson o f the concentratton o f ORGANICS sorbed on sod clay-organic complexes

and m assoctated equdtbrlum water phase wtth land-ban restrtcttons and drmkmg water standards
Land-Ban Restrictions
Eqmhbnum Dnnkmg
Constituent Water Phase Total Water
Sorbed, ConcentraUon, Concentration, TCLP, Standards,
Waste Constttuent mg/kga mg/La mg/kg mg/L mg/L Reference
1,2-dichlorobenzene 400 148 0 49 K086 0 65 F005 06 EPAb

1,2,4-tnchlorobenzene 145 16 19 K030 0 03 c
Benzene 522 1600 6 KI04 0 005 OEPA a
a Data are points of highest sorptlon from Fig 1 ofChlou et al [13] and Fig 1 ofChiou et al [141
b U S Environmental Protection Agency, 40 CFR Parts 141-143, Proposed Rules, Federal Register, 22 May 1989,
p 22064
c Based upon taste and odor Verschueren, K, Handbook of Environmental Data on Orgamc Chemwals, Van Nos-
trand Reinhold Co, New York, 1977, p 604
Copyright by ASTM
g OEPA Int'l (all
(Ohio rights reserved); Mon
Environmental Feb 24 23:07:03
ProtecUon EST 2020
Agency), Regulations, Dnnkmg Water Rules, 3745-81-12
TABLE 2--A compartson of the concentrattons of ORGANICS sorbed on HDTMA + MODIFIED

CLAYS and m assoctated equthbrlum water phase with land-ban restrtcttons and drmkmg water
standards
Equlhbnum Land-Ban
Water Restrictions, Drinking
Constituent Phase Total Water
Sorbed, Concentration, Concentration, Standards,
Waste Constituent mg/kg mg/L mg/kg mg/L Reference
Benzene 31 500 a 287 a 6 K 104 0 005 OEPA e

Phenol 14 000 ~ 322 a 12 K022 0 001 OEPA e
Phenol 0b 94 b 12 K022 0 001 OEPA ~
Phenol 18 000 c 320 c 12 K022 0 001 OEPA e
Pentachlorophenol 31 960 d 5 3d 37 K001 02 USEPAf
a Table 2 of McBride et al. [16] contains the following data Respective sorptlon of 18 and 8% of
benzene and phenol from 25 mL volumes of 0 035% (by weight) aqueous soluUons of benzene and
phenol by 50 mg of sorbent The sorptlon concentration of mg/kg and equdtbnum water phase concen-
tration m mg/L is calculated as follows
1 Total mg of benzene m a system = (25 mL) (1000 mg/mL) (0 035%/100) = 8 75 mg
2 Quantity benzene adsorbed = (8 75 mg) (18%/100) = 1 57 mg benzene
3 Concentration of benzene on binder = (1 57 mg benzene/50 mg binder) = 31 500 mg benzene/
kg binder
4 Concentratton of benzene in the equilibrium water phase = (8 75 -- 1 57 mg in solution)/(25 mL)
-- 0 287 mg/mL or 287 mg/L
b Data are taken from Fig 1 of Mortland et al [17]
c Table 1 of Sheriff et a l [ 18] gwes the amount of clay in the system, the amount of waste constituent
sorbed, the sorpUon etfic~ency, and the concentration of waste constituent m solution before sorptlon
The data for an mltml aqueous phenol concentration of 500 rag/L, 1 g of clay, 18 mg phenol sorbed,
and 36% sorpt~on were selected for comparison m Table 2 above
1 The concentraUon of phenol on the binder = 18 mg phenol/g binder = 18 000 mg/kg
2 The concentration of phenol in the equlhbnum water phase = (500 mg/L) (100 - 36%)/100 =
320 mg/L
d Data are taken from Fig 1 of Boyd et al [15] with the conversion of sorpUon units from mmol/g
to mg/kg and e q m h b n u m water phase concentration umts from mmol/100 mL to mg/L
e OEPA, Regulations, Drinking Water Rules, 3745-1-07
f U S Environmental Protection Agency, 40 CFR Parts 141-143, Proposed Rules, Federal Regtster,
22 May 1989, p 22064
A d d i t i o n a l d a t a m T a b l e 2 are o b t a i n e d f r o m M o r t l a n d et al [17] a n d S h e r l f f e t al [18] o n

t h e s o r p U o n o f p h e n o l b y W y o m i n g b e n t o n i t e m o & f i e d w i t h H D T M A + f r o m water. S h e n W s
d a t a o n s o r p U o n b y H D T M A + M O D I F I E D C L A Y S are used to calculate p h e n o l s o r p t l o n a n d
e q u d l b n u m w a t e r p h a s e c o n c e n t r a t i o n s . D e t a d s o f these c a l c u l a t i o n s are p r e s e n t e d as foot-
n o t e s to T a b l e 2 In a h k e m a n n e r , S h e n W s d a t a for s o r p t l o n o f p h e n o l f r o m w a t e r by clays
t o o & f l e d b y b e n z y l & m e t h y l t e t r a d e c y l a m m o m u m ion ( B D T D A § are p r e s e n t e d in T a b l e 3
The phenol sorptlon from water vanes considerably among the three observations presented
m T a b l e 2 M o r t l a n d & d n o t observe a n y s o r p U o n o f p h e n o l b y t h e H D T M A § M O D I F I E D
C L A Y S M c B n d e ' s d a t a yield a p h e n o l c o n c e n t r a t i o n in t h e H D T M A + M O D I F I E D C L A Y S
o f 14 0 0 0 m g / k g a n d a n e q u l h b r l u m w a t e r p h a s e c o n c e n t r a t i o n o f 322 m g / L ShenlYs data
s h o w e d a p h e n o l c o n c e n t r a t i o n m t h e M O D I F I E D C L A Y S o f 18 000 m g / k g a n d a n e q u d l b -
h u m w a t e r p h a s e c o n c e n t r a t i o n o f 320 m g / L .
Again, M c B r l d e ' s a n d SherifFs c o n c e n t r a t i o n s o f p h e n o l s o r b e d f r o m w a t e r b y H D T M A +
M O D I F I E D C L A Y S exceed l a n d - b a n restrictions b y 1167 a n d 1500 t i m e s respectively, as
Copyright
b a s e by
d ASTM
u p o nInt'lt (all
h e rights
h mreserved);
l t s forMon
K 0Feb
2 224 23:07:03
wastes. ESTAlso,
2020 the e q m l i b r l u m w a t e r p h a s e c o n c e n t r a t a o n s
50 HAZARDOUSWASTES
TABLE 3--A comparison of the concentratwns of ORGANICS sorbed on BDTDA + MODIFIED

CLAYS and m associated equlhbrtum water phase wtth land-ban restrictions and drinking water
standards
Land-Ban
Eqmhbnum Restncnons, Dnnkmg
Constituent Water Phase Total Water
Waste Clay, Sorbed, Concentration, Concentration, Standards,
Constituent g mg/kga mg/La mg/kg mg/L Reference
phenol 4 8 000 180 12K022 0 001 OEPA b

2 12 000 260
1 17 000 310
a The data of the second and third column are calculated m the same manner as defined in Note c to
Table 2 of this paper using data from Table 2 of Shenff et al [18]

b OEPA, Regulations, Dnnkmg Water Rules, 3745-1-07
exceeded drinking water standards by 320 000 Likewise, ShenlTs phenol sorptlon from water
by B D T D A + M O D I F I E D CLAYS exceeded land-ban restrictions and dnnkmg water stan-
dards by slmdar orders of magnitude (Table 3)
Concentratton Effects
Mortland's failure to observe sorptlon of phenol from water by H D T M A § M O D I F I E D
CLAYS may have been due to the low concentrations of phenol and M O D I F I E D CLAYS
The experimental parameters for the Mortland, McBride, and Shenffstudles are presented in
Table 4 In Mortland's study, the total phenol was one third to one fifth of the phenol used by
McBride and Sheriff Also, Mortland's M O D I F I E D CLAYS concentration was one fourth
that of McBnde's and one twentieth that of ShenWs studies One could refer that at the con-
centranons of Mortland's study, equlhbrlum greatly favored phenol m the water phase over
phenol sorphon One would not refer that there was no sorphon Rather the M O D I F I E D
CLAY concentration may have been too low to cause a detectable reducuon of the phenol
concentration of the e q m h b n u m water phase
The influence of concentration on experimental results may also be observed m McBnde's
and Sheriff's studies Both systems are near the same equlhbnum posmon as defined by the
c o m m o n concentranon of 320 to 322 mg/L phenol in the equlhbrlum water phase However,
TABLE 4--A compartson of the parameters of expertmental systems of McBrMe, Mortland, and
Sherff m the sorptton of phenol by HDTMA + MODIFIED CLAYS
Parameters McBride Mortland Sheriff
Clay Smecnte Bentonite Bentomte

Binder concentration 2000 mg/L 500 rng/L 10 000 mg/L
Phenol concentration
Initial 350 mg/L 94 mg/L 500 mg/L
Equlhbnum water
phase 322 mg/L 94 mg/L 320 mg/L
Sorpnon of phenol
Concentration on
binder 18 000 mg/kg Not detected 14 000 mg/kg
Percent sorptlon 8% 36%
the percent sorpnon of phenol by the MODIFIED CLAYS was 8% m McBrlde's study and
36% in Shenf?s study As shown m Table 4, the only difference between the experimental sys-
tems are concentration parameters
Further consideration of the properties of equlhbrlum can be used to demonstrate that the
parameter "percent sorpnon" is a function of concentration as well as the strength of bonds
between MODIFIED CLAYS and ORGANICS For example, consider a system at eqmhb-
rlum where the H D T M A § MODIFIED CLAYS have been centrifuged to the bottom of a ves-
sel with a water phase supernatant above The volume of supernatant could be vaned at will
w~thout disrupting the equihbrlum between the concentration of ORGANICS m the super-
natant solution and the ORGANICS held on the surfaces of the MODIFIED CLAYS Reduc-
ing the supernatant volume, shaking the remaining system, and recentrlfugmg would yield a
system near the same eqmhbrlum point but with an increased percent sorption L~kewlse, if
the volume ofsupernatant were held constant, portions of the H D T M A + MODIFIED CLAYS
could be removed without disrupting the equilibrium In this case, shaking the remaining sys-
tem and recentrlfuglng would yield a system near the same equilibrium point but with a
decreased percent sorpt~on These considerations show that for a given equlhbrlum condmon
there are a continuum of possible selections of ORGANICS and MODIFIED CLAYS con-
centratlons that would come to the same equlhbrlum condmon
For these reasons care must be taken m the interpretation of the parameter percent sorption
Many systems could be m equd~brlum with the same water phase and have percent sorptlons
ranging from below detectable limits to near 100% In such systems the percent sorption would
depend upon the concentration parameters that were selected for the experiment Thus, per-
cent sorptlon may not be a clear indicator of permanence or strength ofsorption bond between
a MODIFIED CLAY and ORGANICS
Desorptton Data
To judge the effectiveness of the S/S of WASTES one should also look at desorptlon or
extraction data to characterize possible leachates from S/S WASTES All of the data discussed
so far in this section consider the sorptlon side of the equilibrium phenomenon A good exam-
ple ofdesorptlon data is also found in the Shenffet al [18] paper In this study, phenolic com-
pounds, water, MODIFIED CLAYS, and cementing agents were mixed and allowed to cure
for 28 days The S/S samples were then ground and subjected to nine 24-h extractions with
100 mL of distilled water. Portions of Sheriff's accumulative leaching data are presented in
Figs 1 and 2. The curves are charactenzed by a relatively rapid desorptlon in the first few
extractions followed by a much slower desorption rate for later extractions
Although the concentrations of phenolic compounds in a given water extract are not pro-
vided by Shenffet a l [ 18], these can be estimated from Figs 1 and 2 First, accumulative leach-
lng data points are read from Figs 1 and 2 and presented in Table 5 The quantities leached
for a given extraction can then be estimated by determining the difference between data points
as presented in Table 6 The concentration of phenolic compounds in a given extraction is
then based upon an extractant volume of 100 mL and is also presented m Table 6 We rec-
ogmze that this approach is only a first approximately and that working directly with Sheriff's
original data would have been preferable However, even with the inaccuracies of th~s
approach, the general concept of desorptlon can be demonstrated
The estimated concentration of phenol in the desorptlon solution varied from 37 to 52 mg/
L m the lmtlal extractions and decreased to 9 6 to 22 6 mg/L m the final extractions Therefore,
even after several extractions, the sorbed phenol was still leaching into the extractant solution
In a hke manner, 3-chlorophenol and 2,3-dlchlorophenol sorbed by the MODIFIED CLAYS
Copyright
also bydesorbed
ASTM Int'l (all rightsthe
into reserved); Mon Feb 24 soluhon
extractant 23:07:03 EST 2020
52 HAZARDOUSWASTES
9 Phenol
9 3 - Chlorophenol
9 2. 3-Oichlorophenol
-- 10 " - " "
| I I' l' ' "I 9 T ~ | w
1 2 3 4 S 6 7 II 9
TIME/DAYS
FIG. l--Leaching tests on cement containing Wyoming bentonite modified with HDTMA + with
adsorbed phenol, 3-chlorophenol, and 2,3-dichlorophenol.
Interpretation of Data Near the Analytical Detection Limits

The data of Sheriff et al. [18] also show that total sorption of an organic constituent by
MODIFIED CLAYS from a water does not imply a permanent and irreversible bonding
between constituent and the binder. In systems containing only 2 g of clay and 50 mg ofphen-
olics, as shown in Table 7, the equilibrium water phase concentrations were not detected. This
may lead one to infer that the total sorption ofphenolics by MODIFIED CLAYS has occurred.
However, when the same systems were solidified and subjected to extractions, it was observed
9 Phenol
9 3 - Chiorophenol
9 2, 3 Dichlorophenol
ii
*t
10
1 2 3 4 S 6 7 $ 9
TIME/DAYS
FIG. 2--Leaching tests on cement containing Wyoming bentonite modified with BDTMA + with
Copyright
adsorbedby ASTM Int'l (all3-chlorophenol,
phenol, rights reserved); Mon Feb
and24 2,3-dichlorophenol.
23:07:03 EST 2020
TABLE 5 - - W a t e r extractton o f O R G A N I C S from ground S / S

sohds contammg W y o m m g bentontte m o d r e d wtth H D T M A +
or BDTDA + a
Accumulative ConsUtuent Extracted, mg
Consutuent Day 1 Day 2 Day 4 Day 7 Day 9
Phenol
HDTMA + 8 77 12 56 13 87 14 60 15 25
BDTDA + 12 12 17 86 21 16 24 30 26 30
3-Chlorophenol
HDTMA § 7 50 8 83 10 06 b 11 29
BDTDA § 2 21 2 92 4 22 5 45 6 49
2,3-dlchlorophenol
HDTMA § 2 19 3 44 4 84 5 65 6 81
BDTDA § 097 097 1 61 1 61 227
a Data are read from F~gs 1 and 2

b Data from Day 6 instead of Day 7
that these p h e n o h c s were readily leached f r o m the M O D I F I E D C L A Y S by repeated water

extracUons, as s h o w n in Table 6 This latter observation w o u l d suggest that m all cases the
e q m h b r l u m water phase c o n t a i n e d s o m e o f the phenolics T h e c o n c e n t r a t i o n m a y have been
below the d e t e c t i o n limit m two cases, b u t it did n o t i m p l y a p e r m a n e n t b o n d
Conclusions Based upon Basic Studies of the Sorption of Organic Compounds from Water
by Natural Soils and MODIFIED CLAYS
In all o f the papers o t e d a b o v e f r o m the sod s o e n c e a n d clay mineralogy hterature, the sorp-
tlon o f O R G A N I C S by M O D I F I E D C L A Y S is described as an e q u d l b n u m p h e n o m e n o n
O R G A N I C S c o n t i n u a l l y transfer b e t w e e n surfaces o f M O D I F I E D C L A Y S a n d the e q m h b -
TABLE 6 - - T h e concentratlon o f O R G A N I C S m successive water extracts o f ground S / S sohds

contamlng W y o m m g bentomte modtfied with H D T M A + or BDTDA + a
Increment of Constituent Extracted, mg, and Concentratmn of Constituent m

the Eqmhbnum Extractant Solution, mg/L
Concen- Concen- Concen- Concen-

Constituent D 2-1 b tratlon D 4-2 tratlon D 7-4 tratlon D 9-7 tratmn
Phenol
HDTMA + 3 69 37 1 31 13 065 6 096 9
BDTDA § 5 16 52 38 38 2 65 26 2 26 22
3-Chlorophenol
HDTMA + 1 45 14 5 1 19c 11 9 1 23 d 12 3
BDTDA § 096 96 097 97 129 129 1 17 117
2,3-d~chlorophenol
HDTMC + 1 04 104 1 61 16 1 097 97 096 96
BDTDA + 0 16 16 048 48 0 0 0 65 65
a Data are based upon Table 5 and an extractant volume of 100 mL

b D 2-1 represents the increment between Days 2 and 1 of Table 5
c Data for the interval between Days 6 and 4 (D 6-4) instead of D 7-4
Copyright by ASTM
d Data for Int'l
the(allinterval
rights reserved);
D 9-6 Mon Feb 24 23:07:03
instead of D EST
9-72020
54 HAZARDOUS WASTES
TABLE 7--The eqmhbrmm concentrations of phenohc compounds sorbed by BDTDA + MODIFIED

CLA YS from aqueous solunons
Constituent Eqmhbrlum Water Phase

Waste Constituent Clay, g Sorbed, mg/kg~ Concentration, mg/L
3-chlorophenol 2 25 000 ND b
1 48 000 20
05 78 000 135
2,3-dlchlorophenol 2 25 000 ND b
1 48 000 20
05 86 000 7O
a The data of the second and third column are calculated m the same manner as defined in Note c to
Table 2 of this paper using data from Table 2 of Sheriff et al [18]

b ND = below detecnon hm~t
rlum water phase High concentrations of ORGANICS in the water phase favor sorptlon by
the MODIFIED CLAYS while the lower concentrations in the water phase favor desorptlon
Even the total sorptlon of ORGANICS by MODIFIED CLAYS from water does not imply
either permanent or Irreversible bonding between ORGANICS and binders
The return of ORGANICS from the surface of MODIFIED CLAYS to the equlllbnum
water phase shows that S/S treatments containing MODIFIED CLAYS do not provide
unqualified protection from the release of ORGANICS into the environment when such
wastes are placed in landfills Instead, as water leaches through a landfill, ORGANICS sorbed
on the surfaces of MODIFIED CLAYS would tend to buffer the concentration of ORGANICS
in leachates As observed above, the concentration of ORGANICS in leachates could be sev-
eral orders of magnitude greater than d n n k l n g water standards
Finally, In all papers cited, the concentration of ORGANICS on the surfaces of MODIFIED
CLAYS was in excess of typical land-ban restrictions This causes additional concern regard-
ing the use of MODIFIED CLAYS and the S/S of WASTES at TSD faclhties.
A Comparison of Basic Studies with the Literature on the S / S of W A S T E S

A comparison is possible of phenol sorptlon by MODIFIED CLAYS in the dry systems
studied by Gibbons and Soundararajan [9] and aqueous systems studied by Sheriff et al.
[18] Gibbons and Soundararajan studied sorptlon of phenol by HWT-22 provided by Inter-
national Wastes Technologies, I n c , m water-flee systems The form of quaternary ammo-
n i u m employed in HWT-22 is not specified Sheriff studied phenol sorption from water by
bentonite modified with H D T M A § and also BDTDA §
Using differential scanning calorimetry to determine the kind of bonding involved and the
a m o u n t of energy necessary to release the compounds from the matrix, Gibbons and Soun-
dararajan observed the breakdown of phenol to simpler compounds and also that the total
energy to drive organic constituent out of the m a t n x is much higher than its normal heat of
vaporization This would lead one to Infer that there is strong bonding between the MODI-
FIED CLAYS and phenol However, in Sheriff's aqueous systems, phenol was readily
extracted by water from MODIFIED CLAYS (see Table 6) Thus, many of the stronger chem-
ical bonds between phenol and MODIFIED CLAYS that were observed by Gibbons and
Soundararajan in dry systems may not exist in aqueous systems and may not prevent the
leaching of phenol from S/S WASTES Into the environment
General Discussion
Interpretation of ExtracUon Data from the Papers Clted Above
Observations on e q m h b n u m cited above lead to the conclusion that studies on extractions
of O R G A N I C S from M O D I F I E D CLAYS or S/S WASTES that contain M O D I F I E D CLAYS
must be included m any evaluation of the environmental protection offered by a given treat-
ment On the other hand, extraction data are not always easy to interpret Consider the data
of Soundararajan et al [11] on methylene chloride and TCLP extractions of O R G A N I C S
from S/S WASTES treated with M O D I F I E D CLAYS Two questions arise (1) does the meth-
ylene chloride test verify that the treatment has formed strong permanent bonds with the
O RG A NI C S , and (2) is the methylene chloride test a more rehable indicator of bond strength
than is the TCLP test9
In regard to the first question, extraction data for methylene chloride are presented in Table
8 The concentration of three of four O R G A N I C S m the methylene chloride extract were
below detectable hmlts This may suggest that retention of the O R G A N I C S by the MODI-
FIED CLAYS was effective However, consider the use of the detection limit instead o f " n o t
detected" in the presentation of the data The quanUty of O R G A N I C S extracted could have
vaned from 2 to 8% of the total mass of the O R G A N I C S m the system Also, the posslbflaty
exists that ffthe extraction had been repeated, another 2 to 8% of the O R G A N I C S may have
been extracted Thin would be similar to results of repeated extracUons from water systems
reported m Table 6, where the quantity of 3-chlorophenol and 2,3-d~chlorophenol extracted
vaned from 0 to 1 61 mg, which is eqmvalent to as much as 3 0% of these constituents m the
study These conslderaUons suggest that Soundararajan et a l ' s methylene chloride extraction
data are not sufficient for a conclusion on the permanence of the bond between the MODI-
FIED CLAYS and O R G A N I C S In their study, the sensmvlty of the analytical procedures Is
not sufficient to evaluate the posslblhty of desorptlon of O R G A N I C S from the M O D I F I E D
CLAYS
In regard to the second question, the TCLP test was below the detection hmlt for all four
O R G A N I C S considered This was to be expected because of the low concentration of
O R G A N I C S in the raw wastes and the low solublhty of the benzo(a)anthracene Th~s point is
further emphasized in Table 9 where it is shown that even ff all of the first three constituents
have been extracted from the S/S WASTES, the concentratmn of O R G A N I C S in the TCLP
TABLE 8--The extractton of ORGANICS from MODIFIED CLAYS using methylene chlortde
Mean
Concentration
Mean Total Extracted with
ConcentraUon ORGANICS Methylene Percent of
m Raw Waste, per MODIFIED Chloride, Constituent
Constituent #g/kg CLAYS, mg/kga /~g/kg Extracted
Bls-(2-chloro-lsopropyl)ether 8 528 212 ND or <660 b <7 7

Naphthalene 18 060 451 1445c 80
Phenanthrene 20 184 500 ND or <660 <3 2
Benzo(a)anthracene 30 460 761 ND or <660 <2 1
a Calculated using 6% as the MODIFIED CLAY content of the binder and the binder to waste ratio
of 40 60
b ND = below detection limit
c Value corrected for ddutlon
56 HAZARDOUS WASTES
TABLE 9 - - The extraction of ORGANICS from MODIFIED CLA YS using the TCLP test
Maximum Percent
Mean Concentration Extraction
Concentration Possible in Limit of Possible
in Raw Waste, TCLP Test, Detection, without
Constituent ug/kg ~tg/kga ~g/kg Detection
Bis-(2-chloro-isopropyl)ether 8 528 256 660 100

Naphthalene 18 060 542 660 100
Phenanthrene 20 184 605 660 100
Benzo(a)anthracene 30 460 14 660 N/Ab
a This calculated value for the first three constituents is based upon a reduction m concentration of
the raw waste to 60% of the original by the addition of the stabilization reactants and then a further
dilution of 20 times by the TCLP extractant The maximum concentration possible is based upon the
possible total extraction of the constituent from the stabilized waste The value for the fourth constituent
is limited by its very low solubility in water of 14 ug/kg
b N/A In this case the limiting factor is the solubility of the constituent in water rather than the detec-
tion limit
extract would still have been below detectable limits On the other hand, even without S/S, the
concentration of benzo(a)anthracene in the TCLP extracts would have been below detectable
limits. Again, these data point out that the limit of detection is too high to permit any conclu-
sions regarding the retention of O R G A N I C S by M O D I F I E D CLAYS
One might pose the argument that If O R G A N I C S are not extracted by methylene chloride,
then there is no need for an aqueous extraction such as the TCLP test However, these data
would not support such an argument The only possible conclusion from these data is that the
concentrations of O R G A N I C S were too low to be detected If however, the TCLP extract had
been evaporated down to about the same mass as the methylene chloride extract without the
loss of the ORGANICS, then a comparison of the aqueous extractions and methylene chloride
extractions would have been possible
Other Constderattons tn the lnterpretatton of Extraction Data

Another issue in interpreting extraction data is the difficulty in devising a test method that
simulates the environment in some reliable manner For example, the TCLP test has been
adopted by USEPA as the basis for some of the land-ban regulations However, from a tech-
nology point of view the question still remains can the TCLP test, which requires samples to
be reduced to a granular form, be used to evaluate the effectiveness of treatments that contain
waste in monohthlc sohds 9 In the papers cited in this review, experimental procedures
included the TCLP test on S/S WASTES, distilled water extracts of S/S ORGANICS, and
sorptlon of O R G A N I C S from distilled water solutions by suspensions of M O D I F I E D CLAYS
and natural soils The solids in these experimental systems all had a surface area per unit
weight considerably greater than the surface area of monolithic solids. Thus, the issue arises,
do the experimental systems cited in this paper to demonstrate the importance of an equihb-
rlum between sorbed O R G A N I C S on the surfaces of M O D I F I E D CLAYS and O R G A N I C S
in water solution imply that such an equihbrium is also important in field situations where
WASTES are bound by M O D I F I E D CLAYS and are contained in monohthlc solids9
We did not find any research on this issue regarding the S/S of O R G A N I C S However,
Bishop [19], in studies of the mechanisms of metal binding in the S/S process, did study the
effect of particle size on the rate of leaching of metals from S/S WASTES Of Importance to
our discussion
Copyright by ASTM Int'l (allIS thereserved);
rights observation
Mon Feb 24 that
23:07:03when the S/S metal wastes are finely ground, the alka-
EST 2020
hnity of the S/S treatment tends to quickly neutralize the acidity ofleachates and retards sol-
ublhzatlon of metals until all of the alkalinity of the S/S treatment has been consumed by the
leachate In contrast, in systems that contained coarser particles, the acidity ofleachates more
aggressively attacked the surfaces of the particles and much more readily extracted metals from
the S/S metal waste In this case the alkalinity on the surfaces ofsohds was exhausted quickly,
while the alkalinity on the intenor of the solids was much more slowly exposed to the acidity
ofleachates Based upon these observations, a possible model for the monohthlc solids of S/S
ORGANICS bound by MODIFIED CLAYS might be lntenor pores containing water satu-
rated with calcium hydroxide Ca(OH)2 or other alkalinity from the S/S reagents and a surface
exposed to more acidic leachate conditions
Of the papers cited in this review, the data of Sheriff et al [18] most easily correspond to
possible bonds of ORGANICS by MODIFIED CLAYS within the pore waters of a monolithic
solid Shenffetal ground thelr S/S organlcs with a mortar and pestle to a fine powder and then
conducted leach tests using water They did not report the pH of their leach solutions at the
end of 24-h periods, however, it seems reasonable to assume an alkaline pH, since the samples
were only leached with water If these experimental conditions are comparable to the pore
water within a monolithic sohd of S/S WASTES, it then follows that the equilibrium reactions
that have been described above and that were observed in the data ofShenffet al would also
occur within the alkaline pore waters of monolithic solids
Another observation of Sheriffet al [18] was that H D T M A + and BDTMA + MODIFIED
CLAYS both began to slowly desorb H D T M A + and BDTMA + after 24-h sorptlon studies
The actual quantity of the alkylammonlum compounds desorbed was not measured, however,
the formation of foam was observed One possible interpretation of this observation is that the
Ca ++ ions resulting from the dissolving of the alkalinity of the S/S reagents have displaced the
alkylammonium compounds from the surface of the clays via ion exchange reactions. In the
preparation of the MODIFIED CLAYS, the alkylammonium compounds displaced Na + ions
from the surface of the clays In the same manner it would appear that Ca ++ has displaced
some alkylammonium ions from the surface of the clay
The possibility of Ca + + displacing alkylammonlum ions from the surface of clays by ion
exchange reactions leads to the question of how permanent are the bonds between the al-
k y l a m m o n i u m ions and the clays Ion exchange reactions are equilibrium reactions between
ions in solution and ions sorbed upon the surface of clays Detailed discussions of ion exchange
reactions ofclays can be f o u n d i n Marshall [20] In aleachlngsituatlonwhereonecomponent
is being leached from the system even slowly, the more abundant ion may eventually displace
almost completely another ion that was more strongly bound by the clay This property of the
ion exchange reactions of clays and the observation of foam In the Shenffet al studies suggests
the possibility that over a time period of years, Ca ++ could displace the majority of alkylam-
m o n l u m ions from the surfaces of MODIFIED CLAYS
The possibility of displacement of alkylammonlum ions from the surfaces of MODIFIED
CLAYS contained in the surfaces of monolithic solids must also be considered As indicated
above with the discussion of Bishop's data, the model we are considenng is a monolithic sohd
whose surface may be exposed to acid leachates and whose Interior pores contain alkaline
water saturated with Ca(OH)2 The dissolving of S/S reagent alkalinity by acid leachates will
result in a greater Ca ++ concentration on the surface of monolithic solids than would occur
within the pores of the solids where the Ca ++ concentration is largely determined by the soi-
ublhty of Ca(OH)2 Also, a lower pH will occur on surface of the monolithic solid than will be
found in the interior pore waters These conditions are much more conducive to the displace-
ment of alkylammonium ions from the surfaces of clays by ion exchange than were the con-
dltions of Sheriffet a l ' s experiments or conditions within the interior pore space ofS/S mono-
hthicbysohds
Copyright ASTM Int'l (all rights reserved); Mon Feb 24 23:07:03 EST 2020
58 HAZARDOUSWASTES
To return to the inmal quesnon of this discussion, does the equfllbnum between ORGAN-
ICS sorbed upon MODIFIED CLAYS and ORGANICS in water solution play a role m S/S
WASTES that will exist largely as monohthic solids The Shenffet al data would indicate that
this equilibrium mechamsm Is very important in characterizing the bonding of ORGANICS
by MODIFIED CLAYS contained within monolithic sohds. The Sheriffdata also suggest that
the ion exchange equfllbnum must also be considered m regard to the durability of the MOD-
IFIED CLAYS, which are used to brad ORGANICS into the S/S monolithic sohds matnx.
Commercial Applications of Modtfied Clays

The possible commercial use of MODIFIED CLAYS in S/S of ORGANICS at TSD facili-
ties also raises the question of what mass ratio of ORGANICS to MODIFIED CLAYS is
reqmred for the desired level of protection to the environment As expected, the data of Tables
3 and 7 show that as the concentration of MODIFIED CLAYS increases, the concentration
of ORGANICS in the equlhbrlum aqueous solution decreases If one assumes total sorption
m the Soundararajan et al. [I 1 ] study, as shown m Table 8, the concentrations of ORGANICS
sorbed by MODIFIED CLAYS were relatively low, varying from 212 to 761 mg/kg In con-
trast, the concentrataon of ORGANICS sorbed on MODIFIED CLAYS for other studies cited
above [15-18] were all an order of magnitude greater The low ratios studied by Soundara-
rajan et al. would favor retention; however, the question still remains Would Soundararajan
have observed environmentally slgmficant concentrations in TCLP or methylene chlonde
extracts of these same ORGANICS if his test waste had been typical of the higher ORGANICS
concentrations encountered at TSD facdmes?
Waste Volume Constderattons

The question anses whether S/S of WASTES is consistent with the goal of minimization of
waste volumes in order to extend the life of available hazardous waste landfill space In general,
treating hazardous wastes by S/S increases the volume of the waste For example, the Hazcon
process increased the volume of waste by 120% in the SITE demonstration [3] Likewise, in
the American Petroleum Institute (API) evaluation [8], the ratio of Hazcon reagent to waste
was 1 5 In the Sohdltech SITE [4] evaluation the ratio of reagent to waste was about 1 to 1
In the International Wastes Technologies SITE [5] evaluation, the mass of the treated soil
increased on an average of 32%
The volume consideration is even more important when sludges are to be treated by S/S in
order to bind free liquids. In the API report [8], mechanical presses, thermal treatments, and
extraction processes were evaluated and compared with the S/S of refinery waste Pretreat-
ments such as mechamcal presses, thermal treatments, and extraction processes reduced the
mass of typical refinery waste by 84 to 96% In contrast, as cited above, the S/S of refinery
wastes increased the mass up to 150% Therefore in this case, the mass of waste solids after
treatment by S/S was as much as an order of magnitude greater than the mass of waste solids
from the other treatments considered In this case, it is difficult to see how the S/S treatment
would be consistent with efforts to minimize the volume of hazardous waste to be placed m
landfills for final disposal
Overall Conclusions
The available data regarding the S/S of ORGANICS do not demonstrate the effectiveness
of this treatment method The use of MODIFIED CLAYS as an additive in S/S treatments
will still require extensive studies The sorption of the ORGANICS from aqueous systems
should be evaluated at a number of ORGANICS and MODIFIED CLAYS concentrations.
Repeated aqueous extractions will be required to define the desorptlon tendency and the
nature of leachates that might pass through the S/S WASTES Studies should consider the
mixtures and the concentrations of ORGANICS that are typically encountered at TSD
facilities
The initial questions regarding the ability of S/S treatments at TSD facilities to render
organic wastes in compliance with land-ban restrictions still remain For those WASTES wlth
land-ban restnctlons on total ORGANICS concentration, there was no example found where
S/S bound the ORGANIC sufficientlyto prevent detection of ORGANICS by analytical meth-
ods, thereby reducing the apparent concentration to compliance with land-ban restrictions
Likewise, for those ORGANICS with land-ban restrictions on TCLP extracts of WASTES, no
example was found where S/S reduced the ORGANICS content of the TCLP extract to com-
pliance with land-ban restrictions
The question of what degree of environmental protection is offered by the sorptlon of
ORGANICS by MODIFIED CLAYS also remains unanswered The available literature does
not provide any clear example of a situation where the S/S of WASTES has resulted in bonding
of ORGANICS sufficiently to prevent the movement of ORGANICS from S/S WASTES back
into the environment via leaching or vaporization Gibbons and Soundararajan [9,10] may
have achieved permanent bonds in water-free systems, however, what must be recognized is
that most waste streams contain water or water is added to the waste as a part of the sohdlfi-
cation treatment Furthermore, in aqueous system data presented m this paper, the concen-
trations of ORGANICS m solutmn in equilibrium with ORGANICS on MODIFIED CLAYS
was far m excess o f d n n k m g water standards For this reason we concluded that S/S ORGAN-
ICS are not necessarily environmentally safe
The concern regarding the possible return of ORGANICS from the surfaces of MODIFIED
CLAYS back to the environment is not reduced by the posslbdity that such releases will be
diluted in the environment One could not directly use the data of ORGANICS concentrations
in solution In equilibrium with MODIFIED CLAYS to predict the concentration of ORGAN-
ICS in leachate from a landfill containing S/S WASTES Possible dlluhons of leachate con-
centratlons cannot be projected further without defining specific details of the disposal site
For this reason, the more important observation may be that the data available show that the
ORGANICS sorbed on MODIFIED CLAYS can be desorbed back to the environment
Therefore, the use of S/S of organic wastes at TSD facditles may not be appropriate because
other treatment options exist that will destroy ORGANICS and eliminate any concern about
the return of the ORGANICS to the environment after treatment
Concerns regarding the conflict between goals for mlmmlzat~on of waste volumes and the
inherent increase m waste volumes associated w~th S/S of WASTES still remain There is no
way to eliminate the volume increase assocmted with S/S of WASTES Therefore, if the S/S
of orgamc hazardous wastes is to be practiced at TSD faclhtles, the WASTES should be pre-
treated to reduce the volume of WASTE solids to a m i n i m u m before S/S
References
[1] Tlttlebaum, M E et al, "State of the Art of StabIhzatlon of Hazardous Organic Liquid Wastes and
Sludges," Critical Revtews In Environmental Control, Vol 15, Issue 2, 1985, Chemical Rubber Co
Press, Cleveland, OH, pp 179-211
[2] Herrmann, J G , "The Superfund Innovative Technology Evaluation Program Technology Pro-
files," EPA/540/5-88/003, U S Environmental Protection Agency, Washington, DC, November
1988
60 HAZARDOUSWASTES
[3] De Percm, P R and Sawyer, S, Proceedmgs of Research Symposium on Land Disposal Remedial
Action, Inclneratton and Treatment of Hazardous Waste, Vol 14, EPA/600/9-88/021, U S Envl-
ronmentalProtectlonAgency,Cmcmnatl, OH, July 1988, pp 136-144
[4] Grube, W E, Jr, "EvaluaUon of the Sohdltech Sohdlficat~onand Stablhzatlon Technology," EPA/
600/9-89/072, U S EnvironmentalProtection Agency, Cincinnati,OH, August 1989
[5] Sawyer, S and Stlnson, M, "Technology Evaluation Report SITE Program Demonstration Test
International Waste Technologies In Sltu StablhzaUon and Soh&ficat~on Hlaleah, Florida," EPA/
540/5-89/004a, U S EnvironmentalProtection Agency, Cincinnati,OH, June 1989
[6] Wotzman, L, Hamel, L E, and Barth, E, Proceedings of Research Symposium, Land Disposal,
Remedial Action, Incineration and Treatment of Hazardous Waste, Vol 14, EPA/600/9-88/021,
U S Environmental Protection Agency, Cincinnati,OH, July 1988, pp 542-557
[7] Weltzman, L, Hamel, L E, De Percm, P, and Blaney, B, Proceedings of Research Symposium,
Land Disposal, Remedial Action, Inctneratlon and Treatment of Hazardous Waste, Vol 15, EPA/
600-9-90/006, U S EnvironmentalProtection Agency, Cincinnati,OH, Feb 1990, pp 448-458
[8] Rowe, G , "Evaluation of Treatment Technologies for Listed Petroleum Refinery Wastes, Final
Report," No 4465, American Petroleum Institute, Washington, DC, May 1988
[9] Gibbons, J J and Soundararajan, R, Amerwan Laboratory, Vol 20, No 7, July 1988, pp 38-46
[10] Gibbons, J J and Soundararajan, R, American Laboratory, Vol 21, No 7, July 1989, pp 70-79
[11] Soundararajan, R, Barth, E, and Gibbons, J J, Hazardous Materials Control, Vol 3, No l, 1989,
pp 42-45
[12] Hamaker, J W and Thompson, J M, m Orgamc Chemicals m the Sod Envlronment, Marcel Dek-
ker, Inc, New York, 1972, pp 49-143
[13] Chlou, C T, Peters, L J, and Freed, V H, Science, Vol 206, Nov 1979, pp 831-832
[14] Chlou, C T, Porter, P E, and Schmeddmg, D W, Enwronmental Science and Technology, Vol
17, No 4, 1983, pp 227-230
[15] Boyd, S A, Shaobal, S, Lee, J -F, and Mortland, M M, Clays and Clay Mlnerals, Vol 36, No 2,
1988, pp 125-130
[16] McBride, M B, Pmnavala, T J, and Mortland, M M, m Fate of Pollutants in the Air and Water
Environments, Part 1, John Wdey and Sons, New York, 1977, pp 145-154
[17] Mortland, M M, Shaobm, S, and Boyd, S, Clays and Clay Minerals, Vol 34, No 5, 1986, pp
581-585
[18] Sheriff,T S, Sollars, C J, Montgomery, D, and Perry R, Environmental Technology Letters, Vol
8, 1987, pp 501-514
[19] Bishop, P L, Hazardous Waste and Hazardous Matenals, Vol 5, No 2, 1988, pp 129-143
[20] Marshall, C E, The Physical Chemistry and Mmeralogy of Soils, Volume I Sod Materials, John
Wiley and Sons, New York, 1964, pp 260-287
Roger D. Spence, 1 r. Michael Gilliam, ~Ivan L. Morgan, ~and
&even C. Osborne ~
Stabilization/Solidification of Wastes
Containing Volatile Organic Compounds in
Commercial Cementitious Waste Forms
REFERENCE: Spence, R D , Gllham, T M , Morgan, I L , and Osborne, S C, "Stabilization/
Solidification of Wastes Containing Volatile Organic Compounds in Commercial Cementitious
Waste Forms," Stabthzatton and Sohdlficatton of Hazardous, Radtoacttve, and Mtxed Wastes,
2nd Volume, ASTMSTP 1123, T M Gllham and C C Wdes, Eds, American Society for Test-
lng and Materials, Philadelphia, 1992, pp 61-72
ABSTRACT" Stabilization/solidification (S/S) is one of the most widely used techniques for
waste treatment and remedial actions, but does not have regulatory approval for treating organ-
lCS Apphcatlon with volatile organic compounds (VOCs) is particularly controversial since it
was believed that the necessary mechanical mixing and exothermlc cementmous reactions
would vaporize the VOCs The objective of this study was to establish whether S/S is a viable
alternative for a sludge heawly contaminated (about 1%) with relatively immobile metals, but
lightly contaminated ( < 0 04%) with VOCs that were contaminating the groundwater The mass
balance indicated that > 50% of the VOCs were retained m the laboratory preparation ofcemen-
tmous samples cured for 28 days The performance tests m&cated the commercial products
could attain leachability indices from 7 to > 9 for the eight VOCs stu&ed and distribution coef-
ficients of > l0 could be attained for all eight and > 100 for some compounds
KEY WORDS stabilization, sohdlfiCatlon, VOCs, organic, immobilization, cement, grout,

waste form, tnchloroethene, acetone, methyl ethyl ketone, 1,2-dlchloroethene, chloroform, ben-
zene, chlorobenzene, perchloroethene
Stabilization/solidification (S/S) is one o f the most widely used techniques for the treatment
and ultimate disposal o f both radioactive and chemically hazardous wastes because o f low pro-
cesslng costs, compatibility with a variety o f disposal scenarios, and ability to meet stringent
processing and performance requirements S/S is accepted as the best demonstrated available
technology ( B D A T ) for m a n y applications involving metals contamination, but not involving
organic c o n t a m i n a t i o n [1,2]. The sludge used in this study was c o n t a m i n a t e d (see Table 1)
mainly with metals, to a lesser extent with semivolatlle organics, and to an even lesser extent
volatile organic c o m p o u n d s (VOCs) Only the V O C s were observed to be migrating in the
groundwater at the site from which the sludge was excavated, hence, V O C s were the focal point
o f the study. In SltU S/S IS an attractive remedial action alternative in this case because incin-
eration will result in little or no v o l u m e reduction for this sludge, incineration m a y result in
m o r e mobile metallic species, excavation m a y result in evaporation o f most of the VOCs, and
in SltU S/S m a y achieve the e n v i r o n m e n t a l protection desired with the least disturbance to the
site and e n v i r o n m e n t and the m a x i m u m protection to operating personnel and residents. To
pursue this alternative, the sponsor needed to demonstrate the l m m o b i h z a t i o n potential o f the
i Chemical Technology Division, Oak Ridge National Laboratory, Oak Ridge, TN 37831-7273
61

62 HAZARDOUS WASTES
TABLE l--Measured concentrations tn the unsplked sludge

(Note Some volatdes were lost m the process o f excavating and
homogemzmg the large sample The table values reflect what was
present tn the laboratory prior to spiking) a
Concentration, Site
Compound mg/kg Maximum
VOLATILE ORGANIC COMPOUNDS (VOC)

Acetone 17 9
1,2-DCE 0 02 100
Chloroform 17 17
MEK 56 4
TCE 15 130
Benzene 0 55 3
PERC 75 59
Chlorobenzene 39 20
BASE~NEUTRAL/AcID ORGANIC COMPOUNDS (BNA)
Phenol 37
1,3-Dlchlorobenzene 130
1,4-D~chlorobenzene 120
1,2-Dlchlorobenzene <6 1
2,4-Dlmethylphenol 110
1,2,4-Tnchlorobenzene 100
2-Methylnaphthalene 190
Dl-n-butylphthalate 160
Bls(2-ethylhexyl)phthalate 190
METALS
Barium 140
Cadmium 560
Chromium 6 200
Lead 760
Silver 47
Nickel 210
Aluminum 28 000
Caloum 3 000
Iron 82 000
Phosphorus 5 400
Slhcon 1 500
a A 46 4 wt% loss was observed upon drying the unsp~ked
sludge
commercially available waste forms Fortunately, recent investigations have indicated true
bonding between organic wastes and some binders as required in the Environmental Protec-
tion Agency's (EPA) interpretation of the Hazardous and Sohd Waste Amendments (HSWA)
to the Resource Conservation and Recovery Act (RCRA) [1,3,4]
This paper presents the results of a study done at Oak Ridge National Laboratory (ORNL)
on the VOC immobilization potential of commercial cementitlOUS waste forms through the
Hazardous Waste Remedial Actions Program (HAZWRAP) in support of the sponsor
Although the toxicity characteristic leaching procedure (TCLP) IS the regulatory test for these
organics, it was not considered a satisfactory test of the immobilization potential for these
waste forms because this test was not designed for waste that had been stablhzed/sohdified,
and, at the time of this study, regulatory guidance was not clear on sample handling and prep-
SPENCE ET AL ON VOLATILE ORGANIC COMPOUNDS 63
aration when VOCs were revolved (that IS, VOC losses prior to the initiation of the extraction
procedure may have given erroneously good results) A static leach test was utilized that
enabled the calculation of both the mass transfer resistance of the waste forms and the affinity
of the waste forms for the VOCs Crucial to the accuracy of this test was the development of
techniques to estimate the VOCs actually retained in the sample at the start of leaching This
was accomplished indirectly by measunng the VOCs that evaporated dunng sample prepa-
ration Although the TCLP was done and is reported in this paper, the main measure o f l m m o -
bihzatlon potential is considered to be the leachability index and distribution coefficient
reported for each VOC and each of the following four vendors Vendor A, RMC Environ-
mental and Analytical Laboratories C o , Vendor B, Wastech, I n c , Vendor C, International
Waste Technologies, and Vendor D, Silicate Technology Corp
Procedures
It is standard practice in studies of this type to compare the TCLP performance of the waste
before and after treatment This approach was not used in this study because much of the
VOCs were lost in the process of taking the large sample needed, and the objective of the study
was measurement of the VOC immobilization potential In the former approach, it is quite all
right, desirable even, to have leachate, or extract, concentrations below the detection limits as
a gross measure of effective treatment for a specific site, but one learns little about the true
immobilization potential of a particular waste form that can be extrapolated to other condi-
tions or sites Since the VOC concentration of the waste sample in the laboratory was not rep-
resentative of the site VOC concentration, spiking of the laboratory sample was required, and,
to ensure obtaining measurable concentrations in the leachates, spiking was designed to g~ve
concentrations not only higher than the site average concentrations, but also higher than site
m a x i m u m concentrations Eight VOCs--acetone, 1,2-dichloroethene (1,2-DCE), chloro-
form, methyl ethyl ketone (MEK), benzene, trlchloroethene (TCE), chlorobenzene, and
perchloroethene (PERC)--were selected among the possible candidates to be spiked into the
sludge and studied
Sample Preparation
The procedure consisted of spiking the sludge using a Hobart mixer outside the glove box,
taking three sludge samples for analysis by standard U S. Environmental Protection Agency
(EPA) Contract Laboratory Program (CLP) protocols (EPA CLP Statement of Work for
Orgamcs Analysis Multimedia, Multlcomponent), immediately placing the spiked sludge
inside the glove box, mixing the spiked sludge with the vendor materials according to the ven-
dor Instructions (grout), filling the stainless steel molds with the grout and covenng with stain-
less steel endplates, removing the molds from the glove box and quickly sealing inside the
stainless steel curing pipe, measuring the VOC concentration of the glove box air using a gas
chromatograph with a flame ionization detector (GC-FID), cunng the samples for 28 days,
measuring the VOC concentration of the cunng pipe air with the GC-FID, removing the cured
samples from the pipe, and quickly removing the endplates, weighing, and sealing inside a
zero-headspace extraction vessel (ZHEV) (Grout means the cementltlOUS waste form in this
document ) The amount of VOCs retained in the sample prior to leaching was estimated by
subtracting the amount of VOCs that evaporated into the glove box and curing pipe from
the amount measured in the sludge by the EPA protocol (Rubbers and plastics, Including
Teflon | should be avoided )
64 HAZARDOUS WASTES
Static Leachmg
The static leaching procedure consisted of suspending each cured sample in 603 g ofdelon-
lzed water reside a ZHEV Each sample consisted of a fiat disk (6 2-cm dmmeter by 1 6-cm
thick) reside a stainless steel n n g with no covenng on either fiat face (the leaching surfaces)
Two types of static leaching were done First, three samples were leached separately for the
entire t~me period w~th periodic m-house leachate analysis Second, five separate samples were
leached, each for a different time period At the end of the time period, the leachate was
removed and submitted to an EPA approved laboratory for analysis
Leachate Analysts
The in-house analysis was done using a Tekmar liqmd sample concentrator (LSC) in con-
j unction with a Perkm-Elmer GC ion trap detector (ITD) capable of measuring concentrations
as low as 1 mg/Mg (PPB) within 15% of the true value The submitted samples were measured
using an LSC with GC mass spectrometer following EPA contract laboratory procedures
(CLP), capable of measunng concentrations as low as 1 mg/Mg within 10% of the true value.
Results
Grout Composition and Properties
Only about 10 to 50% of the VOCs added d u n n g spiking was mixed into and retained In the
sludge, presumably, the remainder evaporated d u n n g the process of mixing the spike into the
sludge This was expected based on the development tests and was the reason analysis of
the sludge VOC concentration was required after spiking The sludge analysis technique con-
slsted of extrachon followed by analysis of the extract The extractant (typically hexane) did
not always extract all of the VOCs in the sludge, thus, the VOC content of the spiked sludge
tended to be underestimated for this study. Tables 2 and 3 list the average VOC concentra-
tions, and their standard deviations, measured In the spiked sludge used to make the static
leach samples and the TCLP samples, respectwely The acetone concentration was left as an
unknown for the Vendor A static leach samples because its concentration was measured to be
negligible, well below the amount that leached d u n n g the static leaching The concentrations
listed in Table 3 were more representative of m a x i m u m site concentrations, as Intended. The
concentrations hsted in Table 2 were designed to give measurable leachate concentrations dur-
ing static leaching
TABLE 2--The VOC sludge concentrations measured after spgkmgfor the static leach samples, mg/kg
(average of three with the standard deviation in parentheses)
Compound Vendor A Vendor B Vendor C Vendor D
Acetone a 123 (25) 223 ( 9) 210 (119)

1,2-DCE 927 (139) 170 (16) 485 (304) 953 (143)
Chloroform 1033 (125) 190 (16) 313 (180) 1150 (147)
MEK 253 (69) 137 ( 9 ) 430 (59) 257 (74)
TCE 623 (56) 193 (12) 184 (94) 573 (82)
Benzene 787 (98) 173 (17) 327 (188) 820 (91)
PERC 963 (119) 537 (115) 767 (201) 887 (246)
Chlorobenzene 550(51) 547 (52) 220(64) 477 (37)
a Left as an unknown
TABLE 3--The VOC sludge concentratzons measured after sptkmg for the TCLP samples, mg/kg
(average of three with the standard devlatton m parentheses)
Acetone 148 (74) 180 (28) 183 (24) 163 (26)

1,2-DCE 457 (127) 313 (19) 303 (47) 363 ( 9 )
Chloroform 31 ( 1) 29 ( 2 ) 27 ( 1) 27 ( 3 )
MEK 763 (29) 198 (128) 210 (262) 637 (202)
TCE 101(63) 180(24) 183(21) 103(12)
Benzene 6( 2) 6(03) 4(04) 8(05)
PERC 233 (66) 197 (17) 106 (21) 170 (36)
Chlorobenzene 48 ( 4) 38 ( 4) 36 ( 4) 40( 3)
than while c u n n g for 28 days inside the c u n n g pipe Although the vapor VOC concentration
reside the curing pipe was significant, little mass was lost to the small headspace of these pxpes.
The concentrahon retained m the cured grout was estimated from the concentraUon measured
m the spiked sludge correcting for the evaporation losses measured d u n n g mixing and curing
and for the dilution of the spiked sludge m the grout product Table 4 hsts the grout compo-
SlUOn for each vendor's product along with some other physical properties The estimated
VOC concentration retained m the samples after curing and percent retention ~s hsted m
Tables 5 and 6 for the static leach samples and the TCLP samples, respectwely
Static Leaching
Figure 1 illustrates the scatter in the leaching data Despite the scatter, the general trend in
Fig 1 was obvious and was typical of this data This trend was analyzed using the average
leachate concentration for all the samples at a gwen time The leachate for Vendor B's product
foamed d u n n g analysis, resulting m mvahdaUon of much of the early data for in-house anal-
ysis until this problem was solved As a consequence, only the single set of data for the sub-
m~tted samples exist for this early t~me period for Vendor B
TABLE 4--Cementttlous waste form composttton and physical properties
Vendor A Vendor B Vendor C Vendor D
COMPOSITION,WT%
Sludge~ 39 8 91 62 5 73 5
Water added 159 185 69 81
Solid additives 44 3 63 3 13 9 18 4
Liquid additives 00 91 167 00
PROPERTIES
Density, kg/L 1 61 1 62 1 22 1 39
Volume increase, % 121 784 93 40
28-day unconfined compressive 6447 12 797 83 414
strength, kPa
This reflects the wt% of the unsplked sludge The spike was added as a water emulsion, but this extra
mass of water is included under "water added "'
66 HAZARDOUS WASTES
TABLE 5--Esttrnated VOC concentrauon (mg/kg) retained m the cured cementlttOUS samples for the
stattc leach test (The value m parentheses ts the wt% o f the VOC that was esttmated to be retained
through mixing and curmg )
Acetone a 9 (75%) 148 (98%) 108 (63%)

1,2-DCE 277 (67%) 8 (44%) 245 (75%) 482 (62%)
Chloroform 352 (76%) 13 (66%) 144 (68%) 710 (76%)
MEK 83 (73%) 11 (76%) 286 (98%) 146 (70%)
TCE 226 (81%) 13 (64%) 101 (81%) 392 (84%)
Benzene 270 (77%) 12 (67%) 174 (79%) 480 (72%)
PERC 412 (95%) 39 (72%) 510 (98%) 702 (97%)
Chlorobenzene 233 (95%) 50 (90%) 122 (82%) 370 (95%)
Left as an unknown
TCLP
Table 7 lists the results of the TCLP test (Table 7 lists only the VOCs that were spiked, but
the other compounds were below their TCLP limits ) The extract for Vendor A did not exceed
any TCLP hmlts The TCE limit was exceeded for Vendors B, C, and D, and the PERC hmlt
was exceeded for Vendor C Refernng to Tables 1 and 3, the spiked concentration exceeded
the site m a x i m u m concentration in all cases, except for TCE for Vendors A and D The ratio
of the site m a x i m u m to the spiked sludge concentration of TCE for Vendor A was 1 29 Mul-
nplymg the TCE extract concentranon for Vendor A by this factor yields 0 032 rag/L, still
below the hmlt of 0 07 mg/L The eqmvalent factor for Vendors B and C was 0 72, giving a
reduced TCE concentration of 0 051 mg/L (below the TCLP hmlt) for Vendor B and 0 22
mg/L (still above the TCLP limit) for Vendor C A similar pro rata reduction for PERC and
Vendor C gave 0 06 mg/L compared to the limit of 0 1 mg/L Thus, correcting the TCLP
extract concentrations by the ratio of the site m a x i m u m concentration to the measured sludge
concentranon resulted m concentrations below the TCLP hmits for Vendors A and B, but
TCE was still above the TCLP limit for Vendors C and D The ratio of the TCE limit to the
measured extract concentration was 0 23 and 0 37, respectively, for Vendors C and D Assum-
Ing the extract concentration is propomonal to the measured sludge concentranon, the sludge
concentranon would have to be lower than 43 mg/kg (33% of the site maximum) for Vendor
C and 38 mg/kg (29% of the s~te maximum) for Vendor D The average of the TCE concen-
tranons measured for the samples taken from the b u n e d lagoon d u n n g site charactenza~lon
TABLE 6--Esttmated VOC concentratton (mg/kg) retamed m the cured cementutous samples for the
TCLP test (The value m parentheses ts the wt% of the VOC that was estimated retamed through
mlxmg and curing)
Acetone 68 (98%) 18 (92%) 124 (99%) 131 (98%)

1,2-DCE 146 (68%) 25 (73%) 160 (78%) 212 (71%)
Chloroform 13 (85%) 2 (74%) 16 (89%) 19 (87%)
MEK 360 (100%) 22 (99%) 142 (100%) 520 (100%)
TCE 28 (58%) 10 (52%) 98 (79%) 56 (66%)
Benzene 2 (67%) 0 4 (36%) 3 (76%) 5 (83%)
PERC 107 (97%) 19 (88%) 67 (93%) 135 (97%)
Chlorobenzene 21 (94%) 4 (93%)
23 (9%) 31 (95%)
SPENCE ET AL. ON VOLATILE ORGANIC COMPOUNDS 67
ORNL DWG 9 0 A - 1 3 6 7
I I
E 7
t I [ [] I I
0
6
7-
rO
MODEL
5 -- ZX ~ --
z
4
g
=< 3
,e, z
(/3
d o3
'< 1
A MCC-1
[] I I I I I I
20 4O 60
TIME (a)
FIG. 1--Static leaching of TCE from Vendor A "sproduct.
was 5.8 mg/kg (4% of the site maximum). Thus, all of the vendors may pass the TCLP test
depending on what basis of comparison is allowed by the regulatory agencies.
Regardless of these number games about the TCLP limits for this particular site, it is impor-
tant to note that this level of sludge concentration (10 to l000 mg/kg) did lead to measurable
extract concentrations (0.004 to 0.3 mg/L), in some cases in the vicinity of the TCLP limits
for these compounds. Undoubtedly, the performance of these products could be improved,
for example, by targeting the additive for a specific compound or compounds, but it does seem
to imply that application to some organic species present at as high as, or higher than, 0.1 wt%
would be questionable, at least for this sludge. On the other hand, the performance was impres-
sive for some compounds (for example, MEK was spiked in the sludge as high as 763 mg/kg,
but never exceeded 0.5 mg/L in the extract, despite being a hydrophilic compound) compared
to others (for example, the extract TCE concentration was uncomfortably close to or greater
than the TCLP limit, despite being a hydrophobic compound). An additive such as organo-
TABLE 7--Concentrations measured in the extracts for the TCLP test (mg/L).
Component Vendor A Vendor B Vendor C Vendor D TCLP Limit
Acetone 0.15 0.18 0.25 0.18 0.59

1,2-DCE 0.13 0.08 0.10 0.19 . .a
.
Chloroform 0.005 0.004 0.009 0.004 0.07

MEK 0.014 0.008 0.048 0.032 7.2
TCE 0.025 0.071 0.30 0.19 0.07
Benzene 0.002 0.009 0.005 0.007 0.07
PERC 0.026 0.028 0.11 0.083 0.1
Chlorobenzene 0.074 0.027 0.11 0.14 1.4
CopyrightNo
by ASTM
limitInt'l
has(allbeen
rights reserved);
set for Mon
thisFeb 24 23:07:03 EST 2020
compound.
68 HAZARDOUSWASTES
phlhc clay could be targeted for TCE to xmprove performance, In fact, the clay could be mod-
ified into two or more different forms to target different types of orgamc molecules The per-
for~aance of such an approach needs to be studied and the economics compared to other
alternatives
Analysis of Results
This section concerns estimation of the effective diffusion coefficients and distribution coef-
ficient from the static leach results The static leach test involved leaching from the surfaces of
a slab The initial concentration was assumed to be uniformly distributed in the slab and zero
for the leachate The leaching is assumed to be simple diffusion from the slab as defined by
Flck's second law, with the diffusion coefficient for the slab much less than the leachate film
diffusion coefficient The leachates were not replaced (static leaching), and, hence, leaching is
assumed to approach an equlhbrium distribution of each specie between the slab and the
leachate Crank gives the following solution for this diffusion problem [5]
A ~ 2~(1 + a ) e_Oq~/t2
(1)
(Ao -- A f) - 1 -- = 1 - + - ~ + a2q~n
where the q,'s are the nonzero positive roots of
t a n q. = - ~q.
and
A = total mass leached in time t, mg,
Ao - initial mass In slab, rag,
Af -final mass in slab in e q u l h b n u m with leachate, mg,
D =effective diffusion coefficient, cmZ/s,
t =time, s,
l =half the thickness of the slab, cm,
O~ = VI/KV s = (Ao -- Af)lAf, and
K = e q u i h b n u m distribution coefficient, slab concentration/leachate concentration
The dlstnbutlon coefficient was defined as
K = ( A I V O / [ V,(Ao -- AI)] (2)
where
Vl ----geometric volume of leachate outside the slab, cm 3, and
V, = geometric volume of slab, cm 3.
By basing the dlstnbutlon coefficient on the geometric volume, the porosity, or void vol-
ume, of the porous sohd body of the slab can remain unknown Since most, if not all, of this
void volume will be occupied by the leachate, AI can approach the specie mass m the pore
solution, but not zero Mathematically, this means K can approach the ratio of the void vol-
ume (occupied by pore solution) to the leachate volume, but the variability of the data made
such a low K statistically indistinguishable from zero. Thus, the least squares estimates of A~
and hence K, were allowed to go to zero The reader may substitute a nonzero value represen-
tative of this ratio (say, 0.04) for the K values of zero in Table 8, if It IS desired to keep K math-
ematlcally correct as defined for this paper For the purposes of this paper and considering the
TABLE 8 - - N E W B O X esttmates of the mass transfer parameters (the ttme delay for Vendor B 1s given
in parenthests)
K,
m g / L grout
Leachabdlty per mg/L
Vendor D, cm2/s .4: mg Ao, mg Index water
ACETONE
A 3 0 >( 10 7 000 a 50 a 65 00 a
B(176days) 3 5 • 10 -8 006 08 75 10
C b 7 00 72 b 4397
D 30• 10 7 000 a 160 a 65 00 a
1,2-DICHLOROETHENE
A 70X 10 -8 100 240 72 90
B(16 7 days) 60 • 10 -8 000 0 7 72 00
C 10• 10 -6 400 120 60 63
D 10X 10 -7 130 360 70 71
CHLOROFORM
A c 30 0 a 30 6 c 6281 a
B(18 8days) 4 0 • 10 -8 000 11 74 00
C 1 0 X 10 -6 070 e 70 60 14 e
D 2 0 • 10 -8 46.0 540 77 750
METHYL ETHYL KETONE
A 40 X 10 -7 130 72 64 28
B(122days) 40• 10 s 055 09 74 197
C 10 • 10 -s 800 140 80 168
D 10X 10 6 125 110 60 16
TRICHLOROETHENE
A 20)< 10 -8 130 200 77 233
B(174days) 1 3)< 10 -s 000 11 79 00
C 50>( 10 7 170 49 63 67
D 1 5)< 10 -s 170 300 78 164
BENZENE
A 70>( 10 8 110 230 72 115
B(129days) 3 5 >( 10 s 000 10 75 00
C 80>( 10 -7 140 85 61 25
D 80)< 10 -s 140 360 70 80
PERCHLOROETHENE
A 6 0 >( 10 -1~ 3350 36 0 92 1683
B(12 8days) 20 X 10 -9 000 34 87 00
C 10>( 10 9 220 250 90 921
D 4 0 >( 10 -1~ 5150 530 94 4313
CHLOROBENZENE
A 48)< 10 9 1650 200 83 592
B(05days) 10>( 10 9 090 43 90 33
C 60>( 10 -s 000 59 72 00
D 1 8 >( 10 -9 2100 280 87 377
a The a m o u n t leached exceeded the esUmated Ao, thus, Af was set as neghglble and N E W B O X est~-
mated Ao and D The low acetone concentration measured m the sludge used to prepare Vendor A's
samples was suspected of being erroneous
b This data set was particularly poor for evaluating D Using all the data, N E W B O X selected a D of
1 • 10 5 which ~s not reahst~c, especially combined w~th the high K This value ~s left u n k n o w n untd
Copyright
a more by ASTM Int'l (all analysis
thorough rights reserved);
wdlMon Feb 24 23:07:03
hopefully gweEST 2020
a defensible value
Downloaded/printed
c N E W B ObyX estimated a suspicIously high D of 1 X 10 -l~ using the graphical estimate for A: but
N E W B O X ' s estimate for both (a D of 1 • 10 -12 and a neghglble Af) was considered even worse
a Th~s Afwas estimated graphically and used as input for N E W B O X to estmate a D
e Parameters estimated using only the data from the first 14 days of leaching
70 HAZARDOUS WASTES
data scatter, a value of zero to a few units for K means little or no significant interaction of the
specie with the solid body A value of about 10 probably means there was significant interac-
tion with the solid body The equivalent of the analytical solution is included in NEWBOX,
a computer program that estimates the least squares fit for up to five parameters for leaching
problems and that includes the equivalent for several analytical solutions to Flck's second law
[6] In general, the average mass measured in the leachate for each vendor and specie at each
time and the mass of each specie estimated retained in the slab (Table 5) were used as input to
NEWBOX to estimate two parameters, D and AI The distribution coefficient, K, was calcu-
lated from this Aiand the mass estimated retained in the sample after curing, Ao One exception
to this approach was for the acetone in Vendor A's product In this case Ao was left unknown
and estimated by NEWBOX, and Aiwas assumed to be zero
The trend for Vendor B's product was to leach little or no mass for a few days, followed by
leach rates comparable to the other vendors This was not a simple diffusion process and could
not be handled by NEWBOX The NEWBOX model was modified for Vendor B by assuming
no leaching for a set time, followed by a simple diffusion, that is, a time offset in Eq 1 The
time delay was estimated as the time intercept of the least squares fit for the hnearized form of
the data The hnearized form of the data comes from the much simpler solution of the diffu-
sion model for dynamic leaching of semi-infinite medium From this solution, a plot of the
total a m o u n t leached versus the square root of time is linear Finite geometries and static leach-
ing will deviate significantly from hneanty, but the early leaching behavior will mimic this
model Thus, the quantity leached was plotted against the square root of time, the linear por-
tion subjectively selected, and least squares analysis used on this portion to find the time inter-
cept for each specie These time delays were subtracted from the times for Vendor B's data and
these corrected times used as input to NEWBOX to get estimates for D and K. Table 8 lists
NEWBOX's estimates for D and K, as well as the leachability index defined as
L = - log (D) (3)
where
L = leachability Index with D in cm2/s
Discussion
The leachability index for these commercial products vaned from 6 0 to 9.4, and the distri-
bution coefficient varied from 0 to 628 From past experience with mobile species such as
nitrate and some radlonuchdes, a leachability index of 6 to 7 is not very good for these waste
forms and indicates that unless a strong interaction exists the specie will readily leach out of
the porous solid body (usually there is a correlation between the leachability Index and the
distribution coefficient) Cementmous waste forms usually can achieve a teachablhty index of
7 to 8 and the better ones can get as high as 9 or better for the more difficult species Some
species are more readily lmmoblhzed in cementltlOUS waste forms, or are just not very mobile,
and have a leachability index of 10 or higher
Taking each specie individually, one would expect acetone to be quite difficult to control
because of its affinity for water, and this indeed proved to be the case. The leachability index
and distribution coefficient were low for three (although Vendor B had a fair leachability index
of 7 5) out of the four vendors, but Vendor C's product never leached much of the acetone
estimated retained The leachate concentration stabilized at a low level quickly, indicating
both a high interaction and low leachability index (In fact, it stabilized so quickly that NEW-
Copyright
BOXby ASTM Int'l (all rights
estimated thereserved); Mon Feb 24
leachability 23:07:03as
index EST52020
This seemed unreasonably low because water has
a leachability index of about 5 and the distribution coefficient was estimated so high )
Although NEWBOX estimates the two as independent parameters, D and K are expected to
be related as follows
D = Dp/g(1 + IQ (4)
where
Dp = true dlffUSlVltyin the pore solution, cm2/s, and
g = geometric factor, dimensionless
This expectation raised questions about why the acetone concentratmn stabilized at this low
value so quickly for Vendor C's product Several explanations are possible, including the
a m o u n t retained was m error and very httle was actually m the sample, the original acetone
was dlv[ded into two or more species, one of which was "unleachable" (strongly sorbed or
disappeared by reaction), and the concentration was actually slowly increasing but the ana-
lytical technique was unable to detect this (the analytmal variability has already been men-
tloned and acetone was one of the harder compounds to follow at these concentrations) There
is not enough ewdence at this time to rule out or verify any of these explanations or other
possible explanatmns All that is known is that httle of the acetone estimated retained in the
Vendor C sample leached into the water according to analysis of the leachate
For chloroform, although three of the leachability indices were still less than 8, Vendor D
had a high K of 75 and Vendor A was even better at 628 NEWBOX estimated the leachability
index at 11 for Vendor A, qmte high for cementmous waste forms This case was similar to
the one for acetone with Vendor C in that not much leached Considenng the variability and
low concentrations, the index estimates for these two cases may be m considerable error
The best performance for 1,2-dichlorocthene was observed for Vendor A's product with a
leachability index of 7 2 and a K of 9 0 Methyl ethyl ketone, hke acetone, has an affinity for
water, but had a better performance than 1,2-dlchloroethene with an index of 8 and a K of
16 8 for Vendor C The best performance for trlchloroethene was for Vendor A with an Index
of 7 7 and K o f 2 3 3 Vendor B had an index of 7 5 for benzene and a K o f 0 , but Vendor A
had an index of 7 2 and a K o f 11 5 The performance for perchloroethene and chlorobenzene
was better than for the other compounds, up to 9 4 and 9 0, respectively, for the index and 431
and 59 for K
Conclusions
The TCLP was judged not satisfactory for testing the immobilization potential of cemen-
tltlOUS waste forms Consequently, the immoblhzation potential was tested by static leaching
and subsequent estimation of the leachability Index and distribution coefficient
Although open mechanical mixers were required to mix the sludge and treatment solids, the
resulting reactions were exothermlc, and the products were cured for 28 days, most of the
VOCs were retained in the cured samples The large uncertainty In VOC analysis means
the error bars were large for the parameter estimation The estimated error for leachate analysis
was from 5 to 15%, depending on the compound, but the error in the parameter estimation
was also affected by the errors in spiking and subsequent sludge analysis Part of the problem
was the inability of the standard extraction techmque to extract some compounds completely
from the sludge for analysis As a consequence, the VOCs tended to be underestimated m the
sludge and the subsequent testing and parameter est~matmn tended to be conservative because
more VOCs were actually present in the samples than estimated It is not unusual for the leach-
ability
Copyright index
by ASTM Int'l to varyreserved);
(all rights by 0 Mon2 without such
Feb 24 23:07:03 ESTlarge
2020 analyhcal errors. The leachability index range
72 HAZARDOUSWASTES
for this study may be 1 0 or more, w~th the reported estimates at the low end, and the estimates
for K should probably be considered order of magnitude estimates
Strong lnteractmns were estimated for acetone, chloroform, and perchloroethene and slg-
mficant interactions for 1,2-dichloroethene, methyl ethyl ketone, tnchlorocthene, benzene,
and chlorobenzene, but no single product appears to interact significantly for all eight com-
pounds Th~s lmphes that it is possible for the vendors to formulate for a combination of VOCs
using a mixture of ad&tlves unless the different addmves used by the vendors interfere w~th
each other. Despite the evidence of interactmn, the estimated leachability ln&ces were &s-
appolntmgly low except for perchloroethene and chlorobenzene It is possible that some, but
not all, of a given VOC was strongly sorbed and that the "free" VOCs qmckly leached out,
giving a relatively low index but a high or moderate K Nevertheless, these low or moderate
m&ces agree with the observation that the sludge concentratmn must be hmlted at about the
spiked levels or less ( < 1000 mg/kg) to pass the TCLP test The TCLP results proved that a
commercmlly available cementltmUS waste form can pass the TCLP cntena for sludge con-
centratmns indicative of this site Based on these results, stabd~zatmn/sohdlficat~on ~s a vmble
alternative for a sludge or sod that ~s heavdy contaminated w~th metals and hghtly contami-
nated with VOCs In such a case, incineration is an expensive alternative that wall hkely con-
vert the metals Into a more mobile form requmng further treatment of the incinerated sludge/
soil. Currently available ad&tlves proved capable of handhng VOCs with hmlted success,
albeit good enough to pass TCLP for this s~te Perhaps better ad&tlves will be developed that
will handle even higher orgamc concentrations ff S/S is accepted for such apphcaUons
Acknowledgment
This research was sponsored by the Office of Defense Waste and Transportation Manage-
ment, Defense Programs, U S Department of Energy, under Contract DE-AC05-840R21400
References
[ 1] "Stablhzation/SohdlfiCatlonof CERCLA and RCRA Wastes, Physical Tests, Chemical Testing Pro-
cedures, Technology Screening, and Field Actlvmes," EPA/625/6-89/002, Center for Environmental
Research Information and Risk Reduction Engineenng Laboratory, U S Environmental Protection
Agency, Cincinnati, OH, May 1989
[2] Weitzman, L, Hamel, L E, and Barth, E, "EvaluaUon of SohdIfiCatlon/Stablhzatlon as a Best Dem-
onstrated Available Technology," 14th Annual Hazardous Waste Englneenng Laboratory Confer-
ence, CmclnnaU, OH, May 1988
[3] G~bbons, J J and Soundararajan, R, Amertcan Laboratory, Vol 20, No 2, 1988, pp 38-46
[4] "Prohibition on the D~sposal of Bulk Liquid Hazardous Waste m Landfills--Statutory Interpretive
Guidance," OSWER Policy DlrecUve No 9487 002A, EPA/530-SW-016, U S Environmental Pro-
tectlon Agency, CIncmnatl, OH, June 1986
[5] Crank, J , The Mathematws of Dtffuston, Oxford University Press, London, 1956, pp 52-56
[6] Nestor, C W , Jr, Godbee, H W , and Joy, D S, NEWBOX, A Computer Codefor Parameter Estt-
, matlon m Diffusion Problems, ORNL/TM-10910, Oak Ridge National Laboratory, Oak Ridge, TN
(in preparation)
Nancy J. Sefl, ~ Mark A. Revall, 2 William Bentley, ~and
Thomas H. Mclntosh 1
Solidification and Stabilization of Phenol and

Chlorinated Phenol Contaminated Soils
REFERENCE: Sell, N J , Revall, M A , Bentley, W , and Mclntosh, T H , "Solidification and
Stabilization of Phenol and Chlorinated Phenol Contaminated Soils," Stabthzatlon and Sohdt-
ficatton of Hazardous, Radloactwe, and Mtxed Wastes, 2nd Volume, ASTM STP 1123, T M
Gdham and C C Wries, Eds, American Society for Testing and Matenals, Philadelphia, 1992,
pp 73-85
ABSTRACT: Phenol and chlonnated phenols are toxic compounds found m many treated wood
products such as fence posts and railroad t~es, hence sods near such treatment facdmes or near
old radroad yards can be quite contaminated Th~s study lnvestxgated the results of using sodium
bentomte clay modified w~th dlmethyl dl(hydrogenated tallow) ammonium chlonde m~xed w~th
Type I portland cement to immobilize a senes of phenols (phenol, 2,4,6-tnchlorophenol, and
pentachlorophenol) from a sandy Shawano B honzon (typic Udlpsamment) soil
The test sod was m~xed with the vanous phenols at a contaminant concentratxon of 1000 mg/
kg An admixture of contaminated soil and modified clay (0 to 10% w/w organoclay) was pre-
pared Final sohdlficatlon was done by adding cement at concentratmns of 16 to 39% w/w Tox-
xc~ty characteristic leaching procedure (TCLP) and unconfined compresswe strength determl-
natxons were made The results of these tests indicate potential for using these organoclay
admixtures for treating soils contaminated w~th vanous chlonnated phenols
KEY WORDS" organoclay, phenols, chlorinated phenols, stablhzatlon, sohdlficatlon, s~te reme-
diatlon, toxic wastes
U n i n t e n t i o n a l c o n t a m i n a t i o n o f soil with toxic chemicals is a continuing problem The U S

General A c c o u n t i n g Office estimates that as m a n y as half o f the 5000 hazardous waste sites
regulated by the U S E n v i r o n m e n t a l Protection Agency (EPA) m a y be leaking hazardous
materials into local earth materials and then to groundwater [I ].
M a n y o f these toxic chemicals are organic in nature, some with potentially severe adverse
effects. T h e y can be carcinogenic or mutagenlc or both in m a n and animals, toxic to aquatic
life, and generally degrade the quality o f water for h u m a n c o n s u m p t i o n Hence, there is a need
to develop new and alternate control technologies capable o f preventing these toxic chemicals
from dispersing through soil into aquifers
A variety o f different stabilization and solidification technologies have been developed in
recent years [2] Suggestion has been m a d e that site remediatlon can be accomphshed by using
an organically modified clay (organoclay) [3] This study focuses on the possibility o f using an
organoclay to stabilize phenol and c h l o n n a t e d phenols in c o n t a m i n a t e d soils
To be feasible for stabilizing and sohdifying contaminants, a treatment procedure must pro-
duce a waste that (1) withstands a pressure of 0 344 M P a (50 psi) when applied in an uncon-
i Professor, research assistant, and professor, respectively, University of Wisconsin--Green Bay, 2420
Nlcolet Dnve, Green Bay, WI 54311-7001
2 Vice-president, Research and Development, J V Manufactunng Co, P O Box 370, DePere, WI,
54115
73

74 HAZARDOUSWASTES
fined compressive strength test (USEPA Office of Solid Waste and Emergency Response
(OWSER) Directive, No 9437.00-2A) and (2) meets certain m a x i m u m leachate concentra-
tions when subject to a toxicity characteristic leaching procedure (TCLP) determination The
allowable TCLP results are specified according to the type of waste (40 CFR Part 268, Subpart
D)
Phenol and chlorinated phenols are toxic compounds found in many treated wood products
such as fence posts and railroad ties Soils near this type of treatment faohty or near old rail-
road yards can be qmte contaminated Site remedmtion is necessary to prevent this contami-
nation from spreading and causing environmental damage
Organoclays
Structure
The types of clay minerals used for organically modified sorbents are primarily layer lattice
types such as the montmorlllonites and the inverted nbbon types such as attapulgltes These
two clay types differ in their crystal structure
Montmonllonltes [4] (Fig 1) are one of the smetlc group of clay minerals Montmorlllonltes
are characterized as 2 1 silicon aluminum type expanding clays They are comprised of two
sheets of silica (5102) tetrahedra linked to one sheet of alumina [A1 (OH)3] octahedra The
montmorlllonites differ from their theoretical structure (2 1 silicon aluminum) in that within
the lattice some of the alumina ions are replaced by other cations of similar size dunng for-
mation of the clay mineral When the substituted cation has a lower charge than the Ion It
replaces, this process of lsomorphous substitution of other metal ions for the aluminum ions
results in net negative charges within the crystal structure; that is, a charge deficiency occurs
within the clay crystal This negative charge deficiency is balanced by additional cations
attracted to the external surfaces of the clay crystal These cations adsorbed external to the
sheet structure are located mainly between the layers, with some around their edges Since the
permanent negative charge due to isomorphous substitution is located m the AI(OH)3 layer a
s~gmficant distance from the balancing positive ions external to the crystal unit, the clay crystal
and the balancing ions readily hydrate In an aqueous environment the clay may expand three
to ten times its dry volume and become gelatinous These cations are loosely bound and may
be readily exchanged if they are physically accessible to the surrounding interstitial clay/water
system [5 ] The quantity of exchangeable cations required to balance the net negative charge
in the clay surface IS called the cation exchange capacity For the various montmorlllonltes,
the cation exchange capacity ranges from 50 to 120 cmol/kg (meq/100 g) of clay
Attapulgite is a hydrated magnesium aluminum sihcate; ~t consists of three-dimensional
chains with a specific cleavage and lath-like crystalline repetitions. Each of the individual laths
is made of rectangular straps of rectangular cross-section joined at their corners, which lends a
needle-like or rod-like structure to the mineral For attapulgltes, the cation exchange capaoty
is typically 5 to 20 cmol/kg of clay
Modlficatton with Organic Compounds

An untreated clay is hydrophihc because water molecules orient with the silica tetrahedron
via hydrogen bonding These sheets of water molecules are interrupted by the hydrated metal
cations on the surface However, clays can be modified by ion exchange with selected organic
compounds that have a positive charged site, that is, are ionic, hence rendenng the clay/organo
complex hydrophobIc.
Frequently,
Copyright by ASTM Int'l (allthe
rightsorganic molecule
reserved); Mon used
Feb 24 23:07:03 EST is a quaternary amine, an a m m o n i u m salt substi-
2020
SELL ET AL. ON CONTAMINATED SOILS 75
SMECTITE STRUCTURE
~.1=:: : : : : : : :: : : : ::.* j ~ ~ .~'~m[onS~ir~tm~
-~il-iiii!?!-:~!!!" ....... (_.':'._'......

:'.
- ~ - ~ "-':::::='=::::::'-~
v ~ii::iiiiiiiF.ii-:
00|174
K+ 02- OH- AI 3+ Si 4+ H20 Mg 2+
FIG. 1--Structure o f smectite-type clay. The K + are surrounded by water molecules and the S i 4+ a r e
surrounded by the oxygen to form the silica tetrahedra, hence are not visible in this figure.
tuted with long-chain carbon groups. The quaternary amine can then replace the exchangeable
sodium, calcium, and magnesium ions located on the clay surface. A variety of organic com-
pounds may thereby be sorbed on the organoclay.
The increased hydrophobic nature of the organoclay is due to the structure of quaternary
amines. The quaternary amines have both polar and nonpolar character. The ionic portion of
these substituted a m m o n i u m salts is attracted to the negatively charged clay surface. The non-
polar hydrocarbon chains can attract nonpolar organic contaminants.
When the quaternary amine is sorbed on a montmorillonite, it forces the layers further
apart, to perhaps 2.5 nm. This phenomenon, called pillaring, thus provides additional porosity
which permits the sorption of organic contaminants.
Data first reported by Jordan [6] indicate that the organophilic properties of a clay are neg-
ligible unless an amine with a carbon chain length of at least ten is reached, and that maximum
swelling capacity is attained when a component with twelve carbon atoms is used to modify
the clay. Though it is dependent on the number of charged sites/area, an amine with twelve
carbon atoms typically covers somewhat more than half of the available surface area of the
clay by
Copyright [6].
ASTM Int'l (all rights reserved); Mon Feb 24 23:07:03 EST 2020
76 HAZARDOUSWASTES
@
I
C;
FIG. 2--Structure of dimethTl di(hydrogenated tallow) ammonium chloride.
Experimental Procedure
This study comprises two separate experiments. Initially, the sorption efficiencies of acti-
vated charcoal, untreated clays, and a commercially available organoclay were compared for
sorbing phenol and chlorinated phenol compounds from aqueous solutions. In the second
phase of the study, varying amounts of the organoclay were mixed with a sandy Shawano B
Horizon (typic Udipsamment) soil and then solidified with Type I portland cement. The
unconfined compressive strength and the TCLP leachates were evaluated for the various
admixtures.
The particular organoclay used in these tests was a sodium bentonite clay modified with
dimethyl di(hydrogenated tallow) ammonium chloride (Fig. 2), manufactured by J.V. Man-
ufacturing Co., Inc. Each of the quaternary amine compounds contains two fatty acid groups.
These fatty acids consist of 5% myristyl (q4), 30% palmityl (q6), and 65% stearyl (q8). Ninety-
eight and a half percent passes through a 150 U.S. mesh sieve (that is, is smaller than 0.10-mm
in diameter), and 92 to 97% passes through a 200 U.S. mesh sieve (smaller than 0.074 mm).
Removal of Organic Contaminants from Water

Previous studies [ 7-9] indicated the potential use of organoclays to remove contaminants
from aqueous solutions. This set of experiments was designed to determine the effectiveness
of this particular organoclay as a sorbent for phenol, 3-chlorophenol, 2,4,6-trichlorophenol
(TCP), and pentachlorophenol (PCP). These results are compared to those obtained for
untreated clay and activated charcoal. The activated charcoal used for these tests was Type
203C, obtained from Barnebey and SutcliffCorp. Ninety percent of the charcoal can be passed
by a 12-mesh screen but retained by a 30-mesh screen; that is, 90% is between 0.595 and 1.68
mm in diameter.
The efficiencies of the three sorbents were measured at 20"C. A 1.00-g sample of phenol or
the various chlorophenols was dissolved in 1.00 L of water. Twenty-millilitre aliquots of the
various contaminant solutions were poured into 50-mL centrifuge tubes. The solutions were
placed on a magnetic stirrer and set into motion. Samples 250 mg in size, of either activated
carbon, unmodified bentonite clay, or the organoclay sorbent, were poured into the mixing
cones of the contaminant solutions. The sorbent]contaminant mixtures were stirred for 1800
s to attain
Copyright by ASTM equilibrium andMon
Int'l (all rights reserved); then
Feb 24centrifuged for 600 s. The solutions were then decanted, and
23:07:03 EST 2020
SELL ET AL ON CONTAMINATED SOILS 77
the clear centrifuged supernatants were analyzed by a Beckman DU6 UV/visible spectropho-
tometer to determine the amounts of phenohc compounds sorbed The absorbances were
determined at 252, 274, 293, and 318 nm for phenol, chlorophenol, TCP, and PCP, respec-
tively, and compared w~th a standard calibration curve To compensate for any potentml inter-
ferences, a blank of distilled water was used
Stablhzatlon and SohdlfiCatlon

A recent study [10] Investigated the use of organoclays to stabilize petroleum sludges In
contrast, th~s investigation focused on the stabilization and solidification of contaminated soils
with a relatwely low concentration (1000 mg/kg and, in a few cases, 10 000 mg/kg) of organic
contaminants
Various admixtures containing 0 to 7 8% organoclay, 16 to 39% portland cement, and 60
to 85% contaminated soft were prepared by the following procedure.
Sufficient soil was mixed with the various contaminants (phenol, 2,4,6-TCP or PCP) so that
the contaminant concentration was 1 00 g/kg or, m a few cases, 10 00 g/kg E~ther 0 00, 5 00,
or 10 00 g of organoclay was added to 100 g of the contaminated soft The sohdffying agent
(portland cement) was added at sohdlfier/contammated soil ratios varying from 1 1 to 1 5
Water was added at a ratio of 350 m L of water to 1 00 kg of sohd mixture
These admixtures were poured into plastic containers 5 0 cm m dmmeter and 10 0-cm high
and allowed to harden After 28 days, the unconfined compressive strength determinations
[ASTM Test Method for Compresswe Strength of Molded Soft-Cement Cylinders (D 1633-
84)] were conducted using a Carver pellet press
The Toxicity Characteristic Leaching Procedure (TCLP, Federal Regtster 1986) was used
to leach the pulverized, solidified mass, and the amount of phenols leached was measured by
a Beckman DU6 UV/visible spectrophotometer as in the first set of tests.
In each case, but particularly for the TCP and phenol, the soil and, to some extent, the
cement provides significant interferences at the wavelengths of interest To compensate for
these interferences, blanks were run consisting of admixtures of pure cement and sod contain-
ing 32, 25, and 16% cement
Results and Discussion

Removal of Contaminants from Water
Figure 3 compares the sorption efficiencies of activated carbon and the dimethyl
dffhydrogenated tallow) a m m o n i u m chloride modified sodium bentonite clay, or LOCKIT |
The unmodified bentonite clay did not sorb a measurable amount of any of these
contaminants.
Though the activated carbon is twice as effective as this particular organoclay for removing
phenol from an aqueous solution, the organoclay is significantly more effective when remov-
ing pentachlorophenol. Since activated charcoal and organoclays sorb contaminants through
a different mechanism, these results are not surpnstng. Activated charcoal sorbs organics
through a surface mechanism Organoclays, on the other hand, operate through a partitioning
phenomenon, that is, the organics are separated from the aqueous solution by the organoclay
in a ratio independent of the concentration. The effectiveness of organoclays for removing
dissolved organics is reversely related to the water solubihty of the organic contaminant
The solubilities of the phenol and chlonnated phenol compounds in water are presented in
Table 1 As can be seen, this relationship is verified for this organoclay/organic contaminant
system
78 HAZARDOUS WASTES
1 =PHENOL
2=3-CHLOROPHENOL
80- 5= 2,4,6-TRICHLOROPHENOL
ACTIVATED C A R B O N 4-=PENTACHLOROPHENOL
d3 7 0 - ~ ORGANOCLAY
Ld
rn
Od
0 60-
(.f)
Z 50-
<
Z
4-0-
.<
F-
Z 30-
0
0
h_ 2 0 - .
0
10-
0 . . . .
0 I 2 5 4
TYPE OF C O N T A M I N A N T
C O M P A R I S O N OF S O R P T I O N E F F I C I E N C I E S
ACTIVATED C A R B O N AND O R G A N O C L A Y
FIG. 3--Comparison of sorption efficiencies of activated carbon and organoclay.
Stabilization and Solidification

The results of both the unconfined compressive strength (UCS) tests and the TCLP tests are
in Table 2.
Figures 4 to 6 summarize the results of the unconfined compressive strength tests. All data
are averages of three to five replicates. As can be seen, the compressive strengths in all cases
significantly exceed the m i n i m u m 0.344 MPa (50 psi). In no case was any liquid expelled dur-
ing the compression.
The percentage of organoclay does not affect the compressive strengths of the admixtures,
The organic contaminants are at too low a concentration to interfere with the solidification of
the cement. Therefore, sorbing the contaminants in the organoclay does not increase the over-
all strength as it might if the contaminant concentrations were higher, which others have
observed for phenol 3. Similarly, at these relatively low concentrations the organoclay itself
does not contribute either positively or negatively to the strength development.
3 Personal communication.
TABLE l--Solubilities of phenol and chlorinated phenols in

water.
Solubility at
Substance 20~ mg/kg
Phenol 77 500
3-chlorophenol 25 000
2,4,6-tfichlorophenol 800
pentachlorophenol
14
I t~
<
[..
E,
t-,
80 HAZARDOUS WASTES
%-
n
v 25-
I
I--
(..9
Z
w 2o!
rY
I.-
09 ,i
ILl 15-
>
w
n~
0_ 10
0
0
5- 39% PORTLAND CEMENT
w 9 32% PORTLAND CEMENT
Z
9 25% PORTLAND CEMENT
Z 9 20% PORTLAND CEMENT
0 0 I 1 I / I I I
0 o 1 2 4 5 6 7
Z
% ORGANOCLAY ADDED
1000 m g / k 9 PHENOL
UNCONFINED COMPRESSIVE STRENGTH TESTS
FIG. 4--Unconfined compressive strength." 1000 mg/kg phenol.
%-
n
20-
I
F-
0
Z
W
n~ 15-
F-
lO~k
w
nI
EL D
0
C) 5
n
W 9 32% PORTLAND CEMENT
Z
LL 9 25% PORTLAND CEMENT
Z 9 16% PORTLAND CEMENT
0 I I / I I I I
C) 0 1 2 3 4 5 6 7
Z
% ORGANOC~Y ADDED
1000 m g / k g TCP
FIG. 5--Unconfinedcompressive strength: 1000 mg/kg TCP.
SELL ET AL. ON CONTAMINATED SOILS 81
"6"
n
20-
T
I--
CD
Z
Ld
rY 15-
F--
U')
LIJ Ai
>
(/3
9 9 9
(Y) 10~I,
LIJ
r I
Cl
O P
C) 5
C3 9 32X PORTLAND CEMENT
Ld
9 25X PORTLAND CEMENT
Z
LI_ 9 20~ PORTLAND CEMENT
Z 9 16X PORTLAND CEMENT
O I I / | I / I
(D 0 1 2 3 4 5 6 7
Z
% ORGANOCLAY ADDED
1000 mg/kg PCP
FIG. 6--Unconfined compressive strength. lO00 mg/kg PCP.
50-
9 20~ PORTLAND CEMENT
* 32~ PORTLAND CEMENT
40
E~
E
~-~ 30.
O0
9 A 9
d
CO 2O-
Ld
rF
n
d
(D 10.
m m
I II I
2 4 5
% ORGANOC~Y ADDED
1000 mg/kg PHENOL
TCLP LEACHING TEST RESULTS
FIG. ?--TCPL leaching tests: 1000 mg/kg phenoL
82 HAZARDOUS WASTES
50-
32% PORTLAND CEMENT
40"
E
30
o3
ZD
CO 20
I_d
i"1/
13_
_.1
0 10.
I--
A. a ~ A
0 I I ~t" ii ~ I I
0 2 3 4 5 6 7 8
% ORGANOCLAY ADDED
1000 m g / k g TCP
F I G . 8--TCLP leaching tests: 1000 mg/kg TCP.
5O
32% PORTLAND CEMENT
o'~ 40
(3n
E @
301
co
~1.1_1 2 0 .
a2
n
_..1
O 10
0 I I I 9
0 1 89 -3 4" 5 6 7 A
% ORGANOCLAY ADDED
1000 mg/kg PCP
FIG. 9--TCLP leaching tests: 1000 mg/kg PCP.
S E L L ET AL. ON C O N T A M I N A T E D SOILS 83
As might be expected, generally the compressive strength increases slightly with an increase
in the percentage of portland cement. The one exception to this is that the PCP-contaminated
samples containing 25 and 32% cement are inverted for all percentages of organoclay. There
is no known reason for this reversal of compressive strengths.
Since the high compressive strengths indicate that these samples will have no difficulty in
meeting Environmental Protection Agency (EPA) unconfined compressive strength guide-
lines, no further work was done to verify the differing trends observed for phenol, 2,4,6-TCP,
and PCP-contaminated soils.
Figures 7 to 9 summarize the immobilization efficiency of the organoclay for 1000 mg/kg
phenol, TCP, and PCP. As indicated in these figures, increasing the amount of organoclay
generally increases the amount of contaminant that is bound in the solidified admixture until
all is immobilized. Depending on the relative amounts of organoclay, soil, cement, and con-
taminant, essentially all of the PCP and TCP can be immobilized with the percentage oforgan-
oclay used in these trials, as seen in Fig. 10. The phenol is never as efficiently bound in the
solidified mass, except for the trials using 20% portland cement. This entire set of runs exhib-
ited unexpectedly high phenol immobilization.
Contrary to the trend observed when removing the three contaminants from an aqueous
solution, however, TCP is more efficiently immobilized than is PCP. This trend is more clearly
illustrated in Fig. 11, indicating the immobilization of 10 000 mg]kg of each contaminant at
one particular cement:organoclay:soilratio (35-g portland cement: 10-g organoclay: 100-g con-
taminated soil).
60-
TCP
PCP NO LOCKIT
50- ~ PHENOL
ETI
E 40-
U3
I-'-- 30-
._J
(,9 10 g LOCKIT
ILl
fY 20-
0--
_J
(.D
t'--- 10-
0
0
COMPARISON OF SORPTION EFFICIENCIES (TCLP)
35 g PORTLAND CEMENT PER 100 g SOIL

1000 m c j / k g CONTAMINANT
0 to 10 g LOCKIT PER 100 g SOIL
FIG. lO--Comparison of sorption efficiencies: 1000 mg/kg TCP, PCP, phenol.
84 HAZARDOUS WASTES
250-
I TCP
PCP
~--~ PHENOL
~,,,,.
~ 200.
o-i
E
~-" 150.
(23
o")
Ia.J 100.
n-"
EL
.__J
5O.
I---
O. I
COMPARISON OF SORPTION EFFICIENCIES (TCLP)
,.35 g PORTLAND CEMENT PER 100 g SOIL
10,000 mg/kg CONTAMINANT
1 0 g L O C K I T PER 1 0 0 g S O I L
FIG. 11--Comparison of sorption efficiencies: 10 000 mg/kg TCP, PCP, phenol.
Conclusions
Sodium bentonite clay, modified with dimethyl di(hydrogenated tallow) a m m o n i u m chlo-
ride (LOCKIT| can be used to remove phenol and chlorinated phenols from aqueous solu-
tion. Its effectiveness is inversely related to the water solubility of the organic contaminant,
which is as predicted, since organoclays operate through a partitioning phenomenon.
When the organoclay is mixed with portland cement, it shows potential for stabilization and
solidification of chlorinated phenol contaminants. Proper mixtures of organoclay and cement
can immobilize these contaminants sufficiently so that TCLP leaching procedures provide 0
mg/kg contaminant leached. These tests do not, however, exhibit the same trends as were
observed when the contaminants were removed from water solution. All of the admixtures
tested exhibit high unconfined compressive strengths ( > 5 MPa). Phenol appears not to be
immobilized by this particular organoclay.
References
[1] National Wildlife, Vol. 27, No. 2, February-March 1989, pp. 33-40.
[2] Sell, N., Pollution Engineering, Vol. 20, No. 8, August 1988, pp. 44-49.
[3] Soundararajan, R., Barth, E. F., and Gibbons, J. J., Hazardous Materials Control, Vol. 3, No. 1,
January-February 1990, pp. 42-45.
[4] Schulze, D. G. and Hess, P. J., "Claypot--An Interactive Program for Displaying Clay Mineral
Structures," Purdue University, West Lafayette, IN, April 1989.
[5] Miller, R. M. and Donahue, R., Soils, 6th ed., Prentice Hall, Englewood Cliffs, N J, 1990.
[6] Jordan, J. W., Journal of Physicaland Colloidal Chemistry, Vol. 53, No. 2, February 1949, pp. 294-
306.
[ 7] byBoyd,
Copyright S. A.,
ASTM Int'l Shaobai,
(all rights S.,Mon
reserved); Lee,
Feb J.-F., andEST
24 23:07:03 Mortland,
2020 M. M., Clay and Clay Minerals', Vol. 36, No. 2,
1988, pp.
Downloaded/printed by 125-130.
[8] Mortland, M M, Shaobal, S, and Boyd, S A, Clay and Clay Mtnerals, Vol 34, No 5, 1986, pp
581-585
[9] Wolfe,T A, Demlrel, T , and Baumann, E R, Journal of the Water Pollutton Control Federatton,
Vol 58, No 1, 1986, pp 68-76
[10] Evans, J C and Pancoskl, S, "Stabilization of Petroleum Sludges," Superfund 1989, Proceedings,
10th NaUonal Conference, Hazardous Materials Control Research Institute, Sdver Spnng, MD, 27-
29 November 1989, pp 292-297
A. S. Aloy, ~A. S. Vishnevskii, ~ T. I. Kolycheva, ~B. S. Kuznetsov, ~ a n d
E. A. S h a s h u k o v ~
Incorporation of Radioactive Ashes into

Ceramic-Clay Matrices
REFERENCE" Aloy, A S, Vishnevskn, A S, Kolycheva, T I, Kuznetsov, B S, and Shashu-
kov, E A, "Incorporation of Radioactive Ashes into Ceramic-Clay Matrices," Stabdtzatton and
SohdLficatmn of Hazardous, Radmacttve, and Mixed Wastes, 2nd Volume, STP 1123, T M
Gflham and C C Wiles, Eds, American Society for Testing and Materials, Phlladclpla, 1992,
pp 86-89
ABSTRACt: The possibility for Immobilization of ashes ansmg from incineration of radioac-
tive wastes generated in research centers was investigated on a laboratory scale Ceramic mate-
rials based on natural clays were made by routine ceramic techniques It was shown that ceramics
on the basis of Chasovyar clay may incorporate up to 30 vol% of ash and that the products
obtained at 1050*C have a high chemical stability and mechanical strength It is concluded that
the proposed method deserves further study aimed at the development of process equipment
KEY WORDS. radioactive wastes, ceramic-clay matrices, clays, ceramic techniques
Incineration is acknowledged to be among the most efficient methods for the management
of combustible radioactive wastes, providing drastic reduction of their volume The remaining
radioactive ash, because of its solublhty and dusty nature, is unsuitable for long-term storage
and even more so for final disposal without additional barriers between the waste and bio-
sphere In technological flowsheets under development for the management of such wastes,
the incorporation into a matrix and fabrication of a solid product are commonly envisaged as
the first b a m e r The choice of matrix and technological process for ash immobilization is
determined greatly by the requirements for mechanical strength and chemical stablhty of the
obtained product, the requirements being determined by the radlologlcal hazard of the ash as
well as by subsequent handling of the product Presently there is no method in c o m m o n use
for ash conditioning Some investigations of incorporation of ash into organic binders, for
example bitumen [1], into lnorgamc binders (cements) [2], or into clay-based ceramics, for
example kaoline and bentonite [3], are available Other possibilities have been considered as
well [4,5]
In this report results are presented for incorporation of ashes arising from the incineration
of real radioactive wastes from research centers into ceramic matrices based on natural clays.
The work was aimed at demonstrating safe lmmoblhzatlon of ash in these widely spread and
easily available materials by means of routine ceramic technology Properties such as chemical
stabdlty and mechanical strength of the obtained ash-contmnmg ceramics were studied
1V G Khlopm Radmm Institute, Leningrad, USSR
86

ALOY ET AL ON CERAMIC-CLAY MATRICES 87
f00,),(0o)
,r
0 30 25 20 15 10 5
FIG 1--X-ray dtffractlon pattern of typtcal ash

Determination of the ash composition has shown that the mare content ( ~ 50 mass %) is
constituted by calcmm and phosphate, 2the remainder 35 to 40 mass % is provided equally by
aluminum, slhcmm, sodmm, potassmm, and iron oxides The total sum of other elements
does not exceed 10 mass % Soot content in the ash is neghg~ble X-ray phase analys~s of typical
ash xs shown on Fig 1 It indicates that hydroxoapat~te [Cas(PO4)3OH] is the mare crystalhne
phase m ash Other phosphates are presented to a lesser degree, parhcularly B-Ca3(PO4)2 and
B-CA2P207. Cambrian and Chasovyar clays were used for the ash lmmobfllzat~on, ~ts chemical
composition is presented m Table 1 Using these clays, it is possible to obtain compact prod-
ucts (open porosity is below 5 vol%) at relatwely low smtenng temperatures (1000 to 1100~
Chemically the clays are characterized by a low content of A1203(~ 19 mass %) along with hxgh
content of alkahne oxides m the case of the Cambrian clay and by elevated content of AIzO3
( ~ 3 3 mass %) for the Chasovyar clay The fraction ofSIO2 m both clays is the same and con-
stltutes 56 to 58 mass % Air-dried clays were subjected to gnndmg m a vlbromfll before
use
2 The presence of phosphorus ~sdue to biological substances m rad~oactwe wastes
TABLE 1--Chemical composition of clays
Composmon, wt%
Type of Clay $102 A1203 Fe203 MgO CaO Na20 + K20
Cambrian (Leningrad 57 5-59 5 18 0-20 5 5 7-8 0 1 8-2 9 0 7-2 8 2 8-6 2

region, USSR)
Chasovyar (Ukraine, USSR) 52 9-56 0 30 5-32 6 1 2-1 6 0 2-0 8 0 6-2 5 2 8-3 3
88 HAZARDOUSWASTES
The fabrication process for the radioactive ceramics included the stages for mixing the com-
ponents with 10 mass % water (as related to air-dried charge), for molding the obtained mix-
ture, and drying and slnterlng the mold products The mixing process was performed in a
vlbrostIrrer with a 400-cm 3 working chamber The molding was made by semldry pressing or
by extrusion using a hydraulic press or an extruder of throughput 7.5 kg ceramic material per
hour The pressed specimens had a form of tablet 26 0 to 26 2 m m in diameter and 9 to 12
m m in thickness The pressure was 40 MPa The extruded specimens had a cylindrical form
of 8 m m in diameter and 15 to 20 m m in length Small-size specimens were chosen to facilitate
the drying and sintenng process made using a resistance furnace. The specimens were dried at
150~ for 1 h The end temperature of slntenng was vaned within 950 to 1100~ and the
specimens were kept at the end temperature for 0 5 to 2 h
The apparent density p (g/cm 3) and open poroslte Po(vol%) were determined for specimens
based on both clays with l0 to 50 mass % ash content, the dependence of these characteristics
on end sintenng temperature, exposure time at the end temperature, and on molding tech-
nique was investigated
The analysis of the obtained experimental data has revealed that, irrespective of the matrix
material and relative ash content in the ceramic specimens, the increase in sintering temper-
ature from 950 to 1050~ leads to a high degree of increase of apparent density and decrease
of open porosity of the specimens Further increase of slntenng temperature up to 1100*C
induces no improvements in the characteristics of specimens The slntering at I050~ for 1 h
is optimum for the selected sizes of specimens The plastic molding process furnishes some-
what better specimens as compared with the semldry pressing under the same sintering con-
ditions The ceramic row mixture for the plastic molding may comprise up to 40 mass % ashes
for both clays. When the ash content in the mixtures was increased above this upper limit, the
production of a plastic paste failed
Apparent density of the best specimens is about 2 3 g/cm 3, and open porosity is 5 to 7 vol%
All these specimens were made from the Chasovyar clay at sintering temperatures not below
1050~ The ash content was rather h i g h - - u p to 30 mass % It should be noted that both mold-
ing techniques may be used, therefore, different technological solutions would be acceptable
for the fabrication The evaluation of chemical stability and mechanical strength of these
specimens was carried out by the determination of sodium leach rate and compressive
strength
The sodium concentration in distilled water after contact with specimens at 20 + 3~ for a
day was measured by flame photometry It follows from the data obtained that the leach rate
depends primarily on sintering temperature and to a lesser degree on ash content For the spec-
imens sIntering at 1050"C, the influence of the latter factor is less pronounced than in the case
of specimens sintering at 950~ The sodmm leach rate for the specimens with open porosity
5 to 7 vol% is (2 to 5) 10 5g cm 2 day-~ Compressive strength was determined for cylindri-
cal specimens of 26 0 to 26 2 m m in diameter and 9 to 12 m m in height The tests were per-
formed on a mechanical press with a strength-measuring device using continuously increasing
load up to complete destruction of the specimen It was found that ceramic specimens sintered
at temperatures not below 1050~ irrespective of the clay type, had a compressive strength
limit of no less than 60 MPa
X-ray investigation of phase composition of the specimens sintered at 1050*C shows high-
temperature interaction between clay and ash This involves, in particular, the formation of a
new crystalline compound ofleucite-type KA1SI206,where potassium may be isomorphously
substituted by sodium, cesium, and calcium, and a l u m i n u m by iron In the specimens slntered
at 950~ the formation of this phase is not observed Only a transition of hydroxoapatite, an
ash component, into the dehydrated form vitloklt could be seen
ALOY ET AL ON CERAMIC-CLAY MATRICES 89
Summary
The investigation indicates that, based on natural clays and routine ceramic technology and
eqmpment, xt Is possible to make a product which can incorporate considerable amounts of
ashes artsmg from incineration of radioactive wastes from research centers A Chasovyar clay
product smtered at 1050~ is characterized by high content of ash (30 mass %), good mechan-
ical strength (the compressive strength hmlt is three times as much as for the best building
brick), and good chemical resistance [sodmm leach rate = (2 to 5) 10 -5 g cm -2 day-~ at 20
+ 3~ and SA/V ~<4 cm-~] The proposed method could be of particular importance for man-
agement of ashes containing considerable amounts of long-lived a-emitters
References
[1] "Treatments of Spent Ion-Exchange Resins for Storage and Disposal," Technical Reports Series No
254, InternationalAtomic Energy Agency, Vienna, Austria, 1985, p 73
[2] Sobolev, I A et al, Nuclear Power Statton, m Russian, No 3, 1980, pp 14-19
[3] Rlege, U, Dlppel, T, and Kartes, H in Management ofAlpha-Contammated Wastes, proceedings of
a symposxum, V~enna,Austria, 2-6 June 1980, International Atomic Energy Agency, V~enna,Aus-
tria, 1981,p 355
[4] Patent 57-41000 (Japan)
[5] Patent 2048555 (Great Britain)
Doris Assmann I and Berthold-Gimther Brodda 1
Fixation of Residues from Special Hazardous

Waste Incinerators for Shallow Land Disposal
REFERENCE" Assmann, D and Brodda, B -G, "Fixation of Residues from Special Hazardous
Waste Incinerators for Shallow Land Disposal," Stabglzatlon and Sohdlficatlon of Hazardous,
Radioactive, and Mlxed Wastes, 2nd Volume, ASTM STP 1123, T M Gilham and C C Wiles,
Eds, American Society for Testing and Materials, Philadelphia, 1992, pp 90-102
ABSTRACT- The residues from special waste incineration (slag, filter dust, and the reaction
products from off-gas purification--the salts) must be stored at speoal dumps since they are
highly polluted with heavy metals The extraction rate of heavy metals out of unfixed filter dusts
and salts is very high Therefore, salts can only be stored underground On the other hand, getting
licenses to build such special dumps is very difficult A fixation of boiler ash (a special filter dust)
or salts with water glass and aluminum phosphate (AIPO4) hardener reduces the leachability of
heavy metal so well that these fixed residues can be dumped together with municipal or com-
mercial and industrial refuse rather than in an underground repository
KEY WORDS. special waste incineration, slag, filter dust, flue ash, boiler ash, salts, special
dumps, residential refuse, commercial refuse, lndustnal refuse, fixation, leachablhty, water glass,
aluminum phosphate hardener
The disposal of household, c o m m e r o a l , and industrial waste is becoming an ever more

important task In the Federal Republic of Germany The quantities of waste each year [1]
amount to about 260 mllhon tonnes, of which 230 mllhon tonnes originates from manufac-
turing industries and 30 million tonnes is accounted for by household refuse and similar waste
These 260 mllhon tonnes of waste include 4 8 to about 15 mllhon tonnes of accountable waste
requlnng special safeguards pursuant to the "Act on the Avoidance and Disposal of Waste"
as amended on 27 Aug 1986 This waste is generally known as special waste and IS character-
lzed by the fact that it cannot be disposed of together with residential waste and, moreover, is
subject to accountability (Table l)
The vague quantities specified, 4 5 to about 15 million tonnes, are based on the evaluation
of shipping notes c a m e d out by the Bundesumweltamt (Federal Environmental Agency) in
1983 for special wastes throughout the Federal Republic of Germany The list of substances
included under the heading of special waste was considerably expanded in the period from
1983 to 1987 There were only 86 types of waste in 1983, whereas in the second draft of the
waste catalogue prepared by the Working Association of the Federal States in 1987 there were
600 types of waste, of which 357 had to be declared
This draft divides the special wastes into three categories
1 Predominantly toxic waste

2 Industrial and commercial waste with restricted risk potential
3 Industrial bulk waste
Scientist and group leader, respectively, Nuclear and Chemical Waste Management, Institute of
Chemical Technology, KFA-Juhch, P O Box 1917, D-5170 Juhch, Federal Republic of Germany
9O

ASSMANN AND BRODDA ON FIXATION OF RESIDUES 91
TABLE 1--Specml wastes to be dumped at underground

reposttortes for spectal waste
Filter dusts from special waste incinerationplants

Solid reaction products from the off-gas scrubbing of refuse
incineration plants and special waste incinerationplants
Hardener sludges containingCN , NO~, NO~
Berylhum-contalnmg wastes
Mercury-containing wastes and also fluorescent tubes
Dry batteries
Arsenic compounds
Old stock/production wastes of pestlodes
Pharmaceutical waste products
Gas purification masses (from cokmg plants and gas works)
Polychlonnated blphenyls (PCBs) or PCB waste
Inorganic pigments
Vanadium, borax residues
Heavy metal chlorides, sulfates, sulfides
Tanning agents and corroswes, as well as their residues
Chemicals from laboratories, industry, trade
Catalysts
North Rhine-Westphalia, a state of the Federal Republic of Germany, is, of our country,
the most strongly charactenzed by industry and trade The volume of special waste produced
in this state was 10 5 million tonnes in 1984 Of this, 33% can be recycled by outside firms
Since North Rhine-Westphalia accounts for about 50% of the special waste ansing in the Fed-
eral Republic of Germany, a nationwide special waste volume of 15 million tonnes is to be
expected Of the special waste arising, 8% is incinerated and the rest is dumped or reprocessed
in waste treatment plants
The residues from special waste incineration (slag, filter dust, and the reaction products
from off-gas purification--the salts) must be disposed of at special waste dumps since they are
highly polluted with heavy metals The salts can only be disposed of underground Approm-
mately 140 kg of flue dust and slag, approximately 2 kg of electric precipitator dust, and
apprommately 60 kg of salts from flue gas washing arise per tonne of incinerated special waste
[2] Since the capacity of the special waste dumps is very limited and m the densely populated
Federal Repubhc of Germany obtaining a hcense for waste dumps is difficult, particularly for
special hazardous wastes, a process has been developed permitting salts and filter dusts to be
fixed in such a way that only minor amounts of heavy metals are leached out in the event of
an ingress of water In this process [3], the residues from a special waste incineration plant are
fixed with sodmm water glasses with the addition of an a l u m i n u m phosphate (A1PO4)
hardener
Water Glass
Sodmm and potassium water glasses [4,5] are produced by melting quartz sand and soda or
potash in oll or gas-fired furnaces at 1300 to 1500~ Water glass belongs to the class of glasses,
but in contrast to conventional glasses, under speoal conditions water glass is soluble in water,
"~ence the name "water glass " S o d i u m water glass, for example, is only water-soluble if slhcon
dioxide (SIO2) sodium oxide (Na20) --< 4 The most common SIO2 Na20 ratio in water glass
IS3 4
The glass is dissolved in special pressure vessels The solution thus obtained is filtered and
Copyright by ASTM Int'l
the desired SlOE (all rights
NaEO reserved);
ratioMonadjusted
Feb 24 23:07:03
by EST 2020 sodium hydroxide (NaOH) and concentrating
adding
92 HAZARDOUS WASTES
OH H
l
HO\ OH HoxO_,oH HO\ /OH
/
/
si \ /Six ~ Six OH
HO OH HO ~ OH /OH HO/ 0 /
HO OH H /
Si \ /
Si \
\ /
/Si x HO OH HO OH
HO OH
. . . . . . . .
FIG. , --Condensation reactions of silicic acid.
the mixture. Water glass may be purchased both as a solution and also as a readily soluble
powder.
If an A1PO4 hardener is added to a water glass solution, then the product is hydrolyzed by
the alkalinity of the water glass. The acid thus released causes a release of silica gel, which
finally leads to the solidification of such masses. The solidification is based on the polycon-
densation of the silica gel solution (the water glass) with the formation not only of linear struc-
tures but also chains and branched three-dimensional structures.
Figures 1 and 2 show diagrammatically the process from polycondensation to polymeriza-
tion. Flocculation follows as a further step. These branched silicon-oxygen structures can be
used in solidifying the salts and boiler ash from special waste incineration plants in order to
fix in part very high heavy metal concentrations in an unleachable form.
Fixation and Leaching

Salts and boiler ash from the special waste incineration plant HIM (Hessische Industriemiill
G m b H ) were available for our experiments. This municipal special waste incineration plant,
with a throughput of 60 000 tons per year, incinerates special waste that cannot be disposed
(•11 O1-1 OH OH OH OH OH OH
l I I I i
IIO-Si-OH + HO- Si-O- Si-O-Si-OH + ...- O- Si- O- Si - O- Si - O- Si- O-...
I t I
oH OH o. 8H 6. 8H 6H OH '
H4SiO4 H8Si3010 (H2SiO3)n
OH OH OH OH
H O - Si - O - S i - O - S i - O - Si- OH
I I I
OI
Og O 6
...0- Si - O- Si - O- Si- O- Si- 0...
OH OH OH OH
FIG.
Copyright by ASTM Int'l (all rights reserved); Mon Feb 2--Polymerization
24 23:07:03 EST 2020 of silicic acid.
TABLE 2--Chemical composition of the boder ash (a special

flue ash) and salt at the HIM special waste mcmeratton plant
Boiler Ash Salt
WEIGHT PERCENT
Na 15 3 34 5
CI 1 18 45 3
SO4 31 90 12 0
Ca 1 89 0 307
Fe 3 99 0 536
Zn 2 45 0 772
St 6 84 1 54
Tt 1 24 1158
Cu 1 49 0 547
Pb 1 93 0 405
K 2 94 0 48
AI 2 23
Sn 28 0 255
Cr 0 201
Br 0 727
PARTS PER M I L L I O N
Cd 351 56
As 209 189
N1 965 209
Sb 388 266
Sn
Cr 639
o f m the m-house plants of the industry or commercial branch involved Table 2 gives a survey
of the chemical composition of the boiler ash or salt used in the expenments
In solidifying the boiler ash with sodium water glass solution and A1PO4 hardener, 20 to 30
wt% of the hardener is mixed with the boiler ash to be solidified and 30 to 35% sodmm water
glass (calculated as solid) is added to the mixture, the ingredients are then mixed thoroughly
and poured into molds. In solidifying the salts, the hardener fraction amounted to 25 to 30
wt% The mixture was subsequently hardened with 38 to 54% sodium water glass (calculated
as a sohd).
The specimens were removed from the molds after 24 h, and after 7 or 28 days, respectively,
and subjected to nine months of leaching in water D u n n g the leaching period, the specimens
were hung in a sealed vessel filled with water (n20) The speomen/leaching medium volume
ratio was 1:40.
Leaching of Fixed Boiler Ashes

In examining the influence of the different preliminary storage periods (7 or 28 days) before
beginning leaching of the solidified boiler ash specimens, it became apparent that the different
preliminary storage times in air only had an effect m the case of lead (Fig 3) In the case of
lead, the 28-day specimens display a leaching rate lower by 15 to 30% over a period from the
eighth to fortieth week, with a m a x i m u m lead concentration of 1 38 ppm after 40 weeks In
companson, the specimens with a preliminary storage period of seven days reached only 1 9
ppm after 40 weeks
In the case of all other elements (copper, zinc, nickel, cadmium, arsenic, antimony, cobalt,
Copyright by ASTM Int'l (all rights
and chromium), thereserved);
prehmmaryMon Feb 24storage
23:07:03 EST 2020is not important. No sigmficant difference in the
time
94 HAZARDOUS WASTES
2.0 I I ' I ' I ' I ' I 9 I ' I
E
s
s
t..-a
t.5-
._E
c-
O
4-,
2 1.0-
0"51
(9
C
0
Z)
AKF 7 Days
Pb o
>, 9 28 Days
>
o
n-
0.0
0 5 110 115 2i0 215 310 315 410 45
Leaching Time [weeks]
FIG. 3--Leaching of leadfrom fzxed boiler ash, depending on different hardening times.
leaching behavior of the specimens can be determined between a preliminary storage period
of 7 and 28 days, as shown by the diagrams in Figs. 4 and 5.
The influence of the different silicon oxide and sodium oxide (SiO2/Na20) ratios in the water
glass and the different hardener contents in the solidified boiler ash specimens are considered
in the following. Table 3 gives a survey of the mixing ratios used. With the exception of lead,
zinc, and copper, the different boiler ash:hardener:SiO2:Na20:H20 ratios in the specimens
studied do not have a particularly great influence on the leaching behavior (Fig. 6). The ele-
ments arsenic, nickel, cadmium, cobalt, and chromium show this in a characteristic way. The
elements cadmium, cobalt, chromium (Fig. 7) and arsenic and nickel show this in a charac-
teristic way.
Only in the case of the elements copper, lead, and zinc, which are also present in the highest
concentrations in the boiler ash, do influences arise that can be attributed on the one hand to
the proportion of hardener in the solidified product and on the other hand to the
SiO2:NazO:H20 ratio and the water glass quantity. If solidified with 43 to 50% A1PO4 relative
to the boiler ash content (Samples A K F and BKF), copper and zinc show clearly higher leach-
ing rates than with 28 to 35% AIPO4 hardener (samples A K F M 1 and BKFM 1) (see Fig. 8 and
Table 4). The element lead displays precisely the reverse behavior; in this case the higher A1PO4
TABLE 3--Composition of the solidified boiler ash specimens.
Sodium Water Glass
Composition
A1PO4
Boiler Ash, g Hardener, g Quantity, g SiO2,wt% Na20, wt% H20, wt%
AKF 70 30 75 29.63 14.76 55.61

AKFM 1 70 20 80 29.63 14.76 55.61
BKF 70 35 100 23.53 9.05 67.42
BKFM 1 70 25 100
23.53 9.05 67.42
0-151 ' I ' I ' I ' I ' I I ' I ' I '

E t
o_
o_ O.lO-t
0.05
0.00 J JAKF
Cr Co Cod 7
Days
9 9 9 28 Days
~) O.0OJ
O 0.15 t
1
j...-....O~ 0 O-
~ O lO~
>>" 0.o5]
-v o.oo/
o ~ 1'o l'S 2'o 2'5 ~ 3'5 go 45
FIG. 4--Leaching of chromium, cobalt, and cadmium from fixed boiler ash, depending on different
hardening times.
t--1 3.5 I I I I I | I i
E
D_
5.0- A
O
2.5.
E A
.--
E
O 2.0-
e .
"~ 1.5-
o
E
0
0
1.0"
NID 9
0 . 5 - z~ Cu & 7 Days
>
0 9 28 Doys
m 0.0
0 ~ 1'0 l'S 2'0 2's 3'0 3'5 ,~ 4s
FIG. 5--Leaching of copperfrom fixed boiler ash depending on different hardening times.
96 HAZARDOUS WASTES
,--, 1.2 , I ' I ' I ' I I I ! I
1.oi AS
0.s-.
0.6- "- -
~= 0.4:
.E
oc
0.2- As, Ni
0.0 &: AKF o: BKF
.+., 1.2- A: AKFM1 ~ BKF'M1
~
@ 1.0- Ni
o
c~ 0.8- 9 9
0.4-"
0.6 - ~ n
>, o.2-~ e ~ ~ n A
>
O
o 0.0
T 0 :5 110 115 210 215 .310 315 4~3 45

FIG. 6--Leaching of arsenic and nickel from different fixed boiler ashes.
T 0.15 , , ' i i I I I I
4 cd
o.1o~~T_
i~ '-
~
o.o t
0 00 J
Ccl, Co, Cr
,,, AKF o: BKF
.E 0.10- *; AKFM1 " BKFM1
=o 0.08" Co
io
~o.o~ 1
0.02-
o.1o7 ~ ~= ...
"~ 0.061
9"r 0.02| . i ' I ' I ' i ' I ' ~ t

0 5 10 15 20 25 30 35 4~0 45
Leaching Time rweeks]
FIG. 7--Leaching of cadmium, cobalt, and chromium from different fixed boiler ashes.
,3.5 I I I I I I I I
E
O..
l..._a A
3.0-
IO
2.5.
t"-"
o_
c-
O 2.0-
A 9 A
E(D
1.5
O
c-
O
c)
1.0-
s
A [] [] Cu
0.5-
> - IA: AKFM1 m: BKFM1
-r
0.0
0 ~ 1'0 1'5 2'0 2'5 3'0 3'5 go 45

FIG. 8--Leaching of copperfrom different fixed boiler ashes.
hardener fraction results in clearly lower lead leaching than with hardener contents of 28 to
35% (Fig. 9).
A possible explanation for this behavior could be that in the case of specimens with 43 to
50% hardener additive, mainly short-chain or small-skeleton silicon-oxygen condensation
products are rapidly formed, in which case during their further storage large pore spaces and
channels result under the influence of condensation, in which copper, zinc, and lead are incor-
porated. Since at 1.21 A Pb > has the largest ionic radius of the three elements (Cu > 0.73 A,
Zn > 0.74 A,), the probability that it can be leached out of the pores in the case of an ingress of
water is somewhat lower than for the smaller Cu > and Zn > ions.
TABLE 4--Heavy metal concentrations in the eluate of solidified and unsolidified boiler ash after
storage in water for nine months compared with the maximum permissible values
for Class 3 landfill sites9
Maximum
Concentration
in Eluate for
Solidified Boiler Ash, ppm Boiler Ash, Class 3 Sites,
Unsolidified, after Ref 7,
Elements AKF AKFM 1 BKF BKFM 1 ppm ppm
Cu 3.06 2~23 2.38 1.10 262.0 10.0

Pb 1.90 2.55 2.52 3.18 3.5 2.0
Zn 2.02 1.74 2.20 1.92 1221.0 10.0
As 1.07 0.73 0.80 0.85 1.0 1.0
Cd 0.12 0.11 0.13 0.11 26.9 0.5
Co 0.07 0.06 0.06 0.06 6.9 2.0
Cr 0.08 0.08 0.05 0.05 0.38 10.0
Ni 0.46 0.91 0.90 0.77 16.7 10.0
98 HAZARDOUSWASTES
3.5 i I I i ! i i i
E
13_
(D_
3.0-
O
2.5-
E
c-
O 2.0-
A
8
c-
9) 1.5-
O
E
O
(J
1.0-
0.5-
>
0
-r-
0.0
1'o 1'~ 2'o 2'5 3'0 3'~ 4'o 45
Leeching Time [weeks]

FIG. 9--Leaching of leadfrom differentfixed boiler ashes.
In contrast, condensation would proceed more slowly for specimens with 28 to 35% hard-
ener contents since the hardener fraction is lower. However, the liquid content of the specimen
is higher since the boiler ash/water glass solution ratio was not altered in comparison to the
previous case. Due to this slower condensation more regular silicate structures could result in
the form of chains, bands, or skeletons in whose smaller pores and channels copper and zinc
could be incorporated. The element lead, which has an ionic radius 1.65 times greater than
copper and zinc, will be more difficult to incorporate; it is therefore much more easily leached
out from the specimens with lower hardener contents.
What are the leaching results of the boiler ash fixed with sodium water glass and A1PO4 hard-
ener in comparison to unfixed boiler ash, and how do they stand up to German limits and
requirements for Class 3 and 4 landfill sites as specified in Ref 7? The fixation of boiler ash
with water glass and hardener enabled the leaching rates to be considerably reduced in com-
parison to unsolidified boiler ash, with one exception (Table 4). Only arsenic displayed hardly
any difference in leachability with and without fixation. As far as the disposability of the resi-
dues thus fixed is concerned, the values for lead were in part above the permissible limits for
Class 3 sites (landfill site for residential waste), although for lead the mixture AKF was below
the limit in question. For this reason, such wastes should be dumped at a Class 4 site (landfill
site for commercial and industrial wastes). Although these sites are not distinguished by limits,
as in the case of Class 3 sites, nevertheless they are subject to special structural requirements
that are, however, considerably less strict than for Class 5 (landfill site for special wastes) or
Class 6 (underground repository for special wastes) sites. Boiler ash from special waste incin-
eration plants is stored at the latter sites.
Leaching of Fixed Salts

The results of salt solidification with sodium water glass and AIPO4 hardener will be pre-
sented in the
Downloaded/printed by following. The different preliminary storage periods in air of the specimens
(7 or
1.2 I ! | I I I I I
E
o. 1.o-
Cu
o.s-
o.o-
~: 0.4z
0.2-
i; F j
k~ 0.0
1.2-
C
Cu Pb
o~ 1.0-" Pb o 7 Doys
o
co 0.8-" 9 28 Days I
0.6-
-6
~ 0.4-
.,
>
o.2-'o1"~176 c o Q o
~ 0.0
o ; 1'o 1'5 2'o 2'5 3'o 3'5 45
Leeching Time rweeks]
FIG. lO--Leaching of copper and teadfrom fixed salt, depending on different hardening times.
28 days) before the start of leaching has practically no influence on the leachability of the ele-
ments. Figures 10 and 11 show, by way of example, the leaching behavior of the solidified salt
specimens.
In contrast, the influence of the different hardener or water glass quantities with a constant
salt content is in part quite considerable. The elements lead and zinc display this influence
most clearly. The mixture ASF with 42.9% hardener content (relative to the salt quantity) is
considerably more resistant to leaching than the ASFM2 mixture with 28.6% hardener con-
tent. The salt and water glass quantities are constant for both mixtures, as shown in Table 5.
The lead and zinc contents in the leaching medium after 33 or 40 weeks of leaching, respec-
tively, are 2.4 to 2.6 times higher for the ASFM2 mixture than for the ASF mixture.
If the proportion of salt and hardener is kept constant and the sodium water content varied,
that is, the (salt plus hardener):sodium water glass ratio in the mixture reduced from 1.06 to
0.8, then the lead and zinc contents in the leaching solution are increased by 2 to 1.76 times
[Fig. t 2 (copper, lead) and Fig. 13 (zinc, arsenic)]. In the case of the other elements, these two
described effects occur to a much smaller extent.
TABLE 5--Composition of the solidified salt specimens.
Sodium Water Glass

Composition
A1PO4 SiO2, Na20, H20,

Sample Salt, g Hardener, g Quantity, g wt% wt% wt%
ASF 70 30 94 29.63 14.76 55.61

ASFM 1 70 20 125 29.63 14.76 55.61
ASFM2
Copyright 70 reserved); Mon Feb 20
by ASTM Int'l (all rights 24 23:07:03 EST 2020 94 29.63 14.76 55.61
100 HAZARDOUS WASTES
'--" 1.2 I I I I I I | I
E
ca_ 1.O-
ct_ Zn
="-' 0.8-"
"~ 0.6-
o.+ S
o
E
_
f A
,- 0.2-
o
I; F
9+-' o.o
8 1.2-
Zn As
,~ 1.o-" ^o . 7 Oar~
0 ,L"~ 1.3
* 28 Days
r0 03.[
__r 0.6-"
0 C ~
~ o.4-
>>, o.2 L.
0
0.0
-1- o ~ 1'o l's ~o ~5 do d5 go +.s

FIG. 11--Leaching of zinc and arsenicfrom fixed salt, depending on different hardening times.
What then are the results in comparison to unsolidified salt, or rather how can they be
assigned to the landfill site classification according to Ref 7?. In the case of the unsolidified salt,
the content of leachable heavy metals was significantly higher for all elements after 33 or 40
weeks of leaching than for the solidified salt. Furthermore, the contents of leachable heavy
metals are clearly below the limits for Class 3 landfill sites for the elements studied iri the solid-
ified specimens.
,-'-, 1.2 I ' I I I I I + 1 ' 1

E
o_ 1.O-
n Cu
o.8'
0.6-
._ 0.4-
g 0.2- Cu, Pb
:'~ 0.0 A: ASF a:. ASFId2
1.2-
Pb I ": AsFu,
9, 1.0 .it
g o.85
8 0.6L
S~
g. 0.+--
>>, 0.2--.
o
o.o
,- o ~ 1'o l's 2'o 2'5 3'o 3'5 4o 45

Downloaded/printed by FIG. 12--Leaching of copper and leadfrom different fixed salts.
1.6 I ' I ' I I

I"--1
E
o_
1.4'
Zn
O. 1.2'
1.0-
o 0.8,
3=
~c 0.6,
E 0.4'
O /x
0.2' A a .', A
C
Q) 0.01 Zn, As
O 1.2-
c n: ASF B. ASFM2
O
O 1.0' As &: ASFId I
-6 o.8' 9Ik "l&
2~ 0.6' & - .,,.~------'A

0.4'
(1)
-r 0.2'
0.0
0 110 1=5 210 21,5 310 3t5 40 45
FIG. 13--Leaching of zinc and arsenicfrom differentfixed salts.
TABLE 6--Heavy metal concentrations in the water after 33 or 40 weeks of water leaching offixed
salt with sodium water glass and hardener compared with resultsfor unsolidified salt and limits for
Class 3 landfill sites [7].
Maximum Value
ASFM 1,a ASFM2,= in Eluate for
Elements ASF, ppm ppm ppm Salt, ppm Class 3 Site, ppm
Cu 0.34 0.58 0.47 354 10.0

Pb 0.21 0.98 0.50 40.2 2.0
Zn 0.35 1.58 0.90 520 10.0
As 0,58 0.75 0.90 0.9 1.0
Cd DL b DL b DL b 10.6 0.5
Co DL b DL b DL b 3.6 2.0
Cr DL b DL b DL b 0.31 10.0
Ni 0.50 0.60 0.84 7.89 10.0
a After 33 weeks.
b DL (detection limit) = 0.05 ppm.
Conclusions
A c o m p a r i s o n o f the leaching results o f the solidified salts (Table 6) with the solidified boiler
ash specimens (Table 4) shows that fixation with sodium water glass and AIPO4 hardener pro-
vides significantly better results for salt and boiler ash. W h a t should be considered is whether
s o d i ubymASTM
Copyright water glass
Int'l (all rightssolidification or23:07:03
reserved); Mon Feb 24 a different
EST 2020fixation m e t h o d (cementation, vitrification, and
so on) should
Downloaded/printed by be preferred for boiler ash.
102 HAZARDOUSWASTES
In considering these leaching results, the proposal immediately suggests itself that residues
from special waste incineration, after specific fixation with sodium water glass and hardener,
should be dumped together with residential or commercial and industrial refuse rather than
in an underground repository The corresponding limits pursuant to a "Guideline from the
Office for Water and Waste of the Federal State of North Rhine-Westphalia" are fulfilled for
both types of waste disposal site This could considerably relieve the limited underground stor-
age space available for hazardous special waste from incineration residues This space would
then be available for toxic wastes from the chemical industry, for which there are currently no
processing possibilities
References
[1] "Mare Draft for Planningof Special Hazardous Waste Plants," m German, Ministry of the Environ-
ment, Area Planningand Agriculture of North Rhlne-Westphaha, October 1987
[2] Rmgel, H, "Untersuchungen zur Verbesserung der Umweltvertraghchkelt be1 der hochthermlschen
Behandlung yon Abfall," Internal Report KFA-1CT-I-509/87, February 1987
[3] KFA-Juhch, D Assmann, German Patent Apphcatlon P 39 38 858 9, Verfahren zur Verfestlgung
Korn- und pulverformlger Abfalle 24 I l 1989
[4] Slhcates, m German, Henkel KG a A, company brochure, Henkel Dusseldorf, Germany, 1988
[5] Hmz, W, Slhkate, Vols 1 and 2, VEB-Verlagfur Bauwesen, Berhn, 1970
[6] Iler, R K, 7"heColloid Chemtstry of Sdtca and Slhcates, Cornell Umverslty Press, Cornell, Ithaca,
NY, 1955
[ 7] "Examiningand Estimating of Waste," Part 2, m German, Office for Water and Waste of the Federal
State of North Rhme-Westphaha, June 1987
Paul L. Bishop, 1 Tim C. Keener, 2 David C. Dusing, 3 Runzhi Gong, 3
and Chong-Le Ll 3
Chemical Fixation of Flue Gas Desulfurization

Sludges and Fly Ash
REFERENCE: Bishop, P L, Keener, T C, Duslng, D C, Gong, R, and LI, C -L, "Chemical
Fixation of Flue Gas Desulfurization Sludges and Fly Ash," Stabdtzatton and Soltdlficatmn of
Hazardous, Radtoactlve, and Mlxed Wastes, 2nd Volume, ASTM STP 1123, T M Gllham and
C C Wries, Eds, American Sooety for Testing and Matenals, Philadelphm, 1992, pp 103-118
ABSTRACT. A study is under way to investigate the heavy metal binding mechanisms m sohd-
lfied/stabihzed high sulfur coal fly ash and flue gas desulfunzation (FGD) sludges Fly ash and
sludge from a typical wet FGD process and dry flue gas desulfurlzatlon by-product from a dem-
onstratmn lime mjectmn mulustage burner (LIMB) process are being evaluated The latter mate-
nal contains substantml portions of available lime and may prove amenable as a sohdifyingagent
with the fly ash This paper descnbes mltml results on the waste charactenzatlon, soh&fied/sta-
blhzed waste form formulation, and influence of liquid/solid ratm on metal leaching from the
waste forms
KEY WORDS: stablhzation/sohdlfiCatlon, fixation, fly ash, flue gas desulfunzatxon sludge,
leaching, metals, LIMB ash
More than one half of this nation's electricity is generated from coal In Ohio, over 95% of
the electricity comes from coal-fired utility boilers [1] The United States has vast supplies of
high heat content coal to feed these boilers, but unfortunately much of the coal has a high
sulfur content, averaging about 3 5% sulfur by weight Enforcement of environmental laws
dealing with air pollution, which severely restnct sulfur dioxide (SO2) emissions, has led to a
significant decline in the a m o u n t of high sulfur coal used in Ohio and elsewhere The new
Clean Air Act A m e n d m e n t could compound this trend dramatically
In order to minimize potential environmental problems from power production, public
utilities have been required to Install highly efficient pamculate control and flue gas desulfur-
lzatlon (FGD) facilities in new plants Acid rain legislation would further require the instal-
latIon of F G D systems on existing, as yet uncontrolled sources of SOx [predominantly SO2 but
including other sulfur compounds, for example, sulfur trioxide (SO3)], which would signifi-
cantly Increase the volume of waste produced
Flue gas emission control waste residues from the combustmn of coal or other fossil fuels,
including fly ash and F G D sludges, are currently excluded from Resource Conservation and
Recovery Act (RCRA) Subtitle C classification as a hazardous waste and can be disposed of
under Subtitle D (nonhazardous sohd waste) provisions This situation greatly reduces the
requirements for &sposal but does not eliminate the utfllty's responslbflmes If improperly
William Thorns Professor and head, Department of Civil and EnvironmentalEngineering, University
of Cincinnati, Cincinnati, OH 45221
2 Associate professor, Department of Civil and EnvironmentalEnglneenng, University of Cincinnati,
Cincinnati, OH 45221
3Graduate research assistants, Department of Civil and EnvironmentalEngineering, University of Cin-
cinnati, Cincinnati, OH 45221
103
Copyright9 1992byASTM International www.aslm.org

managed, these wastes can cause serious environmental damage, pnncipally due to associated
toxic metals that can leach into water supphes
The amounts of flue gas wastes to be disposed of are huge Utilities currently produce about
64 million metric tons of ash and about 6 million metric tons of F G D sludge per year [2]. The
total is expected to increase to about 125 to 150 million tons per year by the year 2000 [3-5].
Utilities have found markets for about 25% of their coal ash, mainly as a cement additive or
as fill material in road embankments and bases [6], but the remainder of their waste is disposed
Into ponds and landfills About two thirds of the ash and sludge is placed in ponds in the form
of a wet slurry, while the remainder is placed in landfills in a dry or dewatered form
All of these wastes have the potential to Impact the environment adversely Both ashes and
sludges contain trace quantities of heavy metals (including lead, mercury, cadmium, chro-
mium, zinc, selenium, and arsenic) and polycyclic aromatic hydrocarbons These chemicals
can leach from the waste and potentially enter a groundwater aquifer.
One way to minimize leaching problems is to fix chemically, or solidify/stabilize, the wastes
before landfilhng. This process sohdlfies the waste, reduces permeability, converts many
potentially hazardous constituents into less soluble forms, and reduces leaching potential One
commonly used fixation process involves the mixing of F G D sludge with fly ash and lime,
resulting in a stable pozzolanic material Several leaching studies have indicated that chemi-
cally fixed F G D sludges pass federal standards for heavy metals leaching, but there is still a
large degree of concern because the leaching tests used are not intended to represent the con-
ditions the wastes will face over long time periods and do not truly indicate potential long-
term impacts from disposal New dry scrubbing processes for removing SO2 may pose an
increased risk for leaching of heavy metals due to increased fly ash In the waste mixture in a
matrix form that may prove to be less stable than the conventional wet scrubbing waste matrix.
What is needed is more fundamental information on the binding mechanisms involved in
F G D sludge fixation and the effects of long-term environmental exposure on these mecha-
nisms Future use of high sulfur coal may be dependent on assurances that the fixed wastes are
truly environmentally safe.
The primary objective of this research is to determine heavy metal binding mechanisms
occurring in model fixed wastes by using a leaching procedure that provides leaching rate data
(rather than just a single pass-fail concentration) and that uses a series of graded eluants to
indicate the form of the metal present One of the main concerns with the leaching tests cur-
rently used is that they do not indicate the effects of long-term leaching, dunng which time
portions of the fixed waste matrix may dissolve, increasing pore space and available surface
area and leading to increased rates of leaching This concern is being studied by evaluating
changes in pore size, pore distribution, and available surface area as the wastes go through the
sequential leaching procedure
This paper presents initial findings from this long-term research program In particular, the
paper addresses waste characterization, sohdlfied/stablhzed waste form formulation, and
effects of hquid/sohd ratio on metal leaching from the waste form
Previous Research
Both fly ash and flue gas desulfunzatlon wastes may pose serious disposal problems.
H u e gas desulfurlzation sludges are generally the result of scrubbing the flue gas with a sus-
pension of lime or limestone to remove SO2 The SO2 is absorbed from the flue gas and
removed from solution by a chemical reaction that forms a calcium sulfite (CaSO3 2H/O) and
calcium sulfate (CaSO4) salt solution, along with insoluble calcium sulfate dihydrate (gypsum)
These salts precipitate in a holding tank, forming the F G D sludge [4]. The sludge also contains
someby fly
Copyright ash
ASTM Int'land a number
(all rights of Feb
reserved); Mon heavy metals
24 23:07:03 at trace levels (see Table 1) The physical properties
EST 2020
BISHOP ET AL ON CHEMICAL FIXATION 105
TABLE l--Typtcal FGD by-product and fly ash characterlstzcs
FGD Sohds, FGD Waste Fly Ash,

ppm [3] Liquor, ppm [3] ppm [8]
AnUrnony 0 09 to 1 6
Arsemc 0 6 to 63 <0 004 to 1 8
Banum 896 to 1253
Berylhum 0 05 to 11 <0 001 to 0 18 13 to 17
Boron 218 to 770
Cadmmm 0 08 to 350 0 004 to 0 11
Chrommm 3 to 250 0 001 to 0 5 169 to 193
Cobalt 64 to 105
Copper 1 to 76 0 002 to 0 6 293 to 313
Lead 0 2 to 21 0 001 to 0 55 96 to 183
Manganese 170 to 1432
Mercury 0 001 to 6 < 0 001 to 0 07
Molybdenum 0 9 to 5 3
Nickel 6 to 27 0 005 to 1 5 154 to 263
Selemum 0 2 to 19 <0 001 to 2 7
Tm 75 to 171
Vanadium 249 to 390
Zinc 10 to 430 0 01 to 27 195 to 310
of the sludge vary widely, depending on the makeup of the matenal The CaSO3 2H20 and
CaSO4 particles both range in size between 1 and 100 um, but the sulfite wastes form as plate-
lets, whereas the sulfates are blocky in shape [ 7] Sludges that contain much CaSO3 2H20 tend
to be thixotropIc and difficult to dewater, so current practice is to oxidize the waste to convert
the sulfites.
Hy ash is the fine residue that results from coal combustion The particles are generally
spherical in shape with diameters on the order of 1 to 80 #m [8] They are composed primarily
of silica and a l u m i n u m plus carbon, and oxides of iron, calcium, magnesium, and sulfur [9]
In addition, they contain trace quantities of many toxic metals Table l indicates typical con-
centratlons of these contaminants Hy ash is classified as an artificial pozzolan; the finely
divided siliceous or alumlnoslhceous materials will react in the presence of lime and water to
form a hard solid [10].
The metals associated with fly ash and FGD sludge can leach and threaten groundwater
aquifers Van der Sloot et al [11] found, using shake tests, that most of the metals in fly ash
readily leach A major study by Michael Baker, Jr, Inc. [12] showed that scrubber liquors and
elutriates commonly have metal concentrations in excess often times the drinking water stan-
dards, but generally less than RCRA standards, which for most metals are 100 times the drink-
ing water standard. Similar results have been reported by others [ 7,13] A three-year field study
of six F G D sludge disposal sites found that metals did leach readily, often in excess of RCRA
limits, but were rapidly attenuated by the soil so that monitoring wells were not severely
impacted [4] A recent study of five samples of coal combustion by-products from power
plants m Pennsylvania showed that pore water and leachate metal concentrations were gen-
erally very low except for boron, strontium, nickel, and, sometimes, vanadium [14]
Many utilities are now chemically fixing their F G D sludges as a way to improve the char-
acteristics of the sludge for disposal and to mimmize leaching The most widely used process
involves the addition of fly ash and lime to the F G D sludge These materials undergo a poz-
zolanic reaction, resulting m reduced moisture content, hardening of the mass into a solid with
reduced permeability, and chemical conversion of many of the contaminants to more Insol-
uble forms During sohdlfication, the heavy metals become entrapped in the gel matrix or
106 HAZARDOUSWASTES
adsorbed to the surfaces of the mlcropores in the gel, thus reducing their mobility The overall
result is that leaching is greatly reduced.
IU Conversion Systems has a proprietary process of this type, which they call "Poz-O-Tec,"
in which a mixture of 0 5 to 1 part or more of fly ash and 1 part F G D sludge is used along with
up to 4% lime, all on a dry weight basis [12]
Fly ash is composed largely of alumina and silica When mixed with water (in this case from
the F G D sludge) and lime, it experiences alkali attack resulting m the formation of the follow-
ing products
3 CaO 2 SIO 2 3 H20

3 CaO A1203 6 H20
These products are similar to those from the reaction of portland cement and water, are insol-
uble, and solidify Into a monolithic mass [12,13,15]
The fixation of flue gas wastes described above is very similar to the stablhzatIon/sohdlfi-
cation process used for hazardous wastes Recent research has shown that many of the metals
are intimately bound into the paste matrix as silicates and carbonates and do not easily leach,
while others are trapped in the pores as hydroxides and can leach more readily [16-19] U S
Environmental Protection Agency (EPA) extraction procedure (EP) and toxicity charactenstlc
leaching procedure (TCLP) leaching tests have also been found not to be very suitable for these
types of wastes because the initial leaching that they mimic does not provide an indication of
the long-term leaching potential [20-22] It has been suggested that over time the lime and
other matenals filling the pores of the paste matrix leach away, decreasing pore water pH and
Increasing the permeability of the waste and the dlffUSlVlty of the waste constituents Thus,
leaching rates may increase after long time exposure [18,20] Existing metal leaching models
for fixed wastes are based on external surface area only [23-25] and cannot account for these
internal pore changes Most leaching tests only consider the effects o f p H on leaching, but other
factors such as oxidation-reduction potential or presence of complexlng or chelating agents
may be just as important [26] These are not accounted for in the EP or TCLP tests and have
not been properly studied A recent research needs workshop on h ~ a r d o u s waste manage-
ment found that considerable research is needed on the stablhzatlon/sohdlfiCatlon process
[27]
Experimental Procedures
Model fixed fly ash and F G D waste formulations were prepared using both wet and dry F G D
by-products and fly ash from local utilities. The wet scrubbing F G D by-products and fly ash
were obtained from the Conesvllle Station, Columbus Southern Power, which uses 3 5% sulfur
coat The dry scrubbing by-product was obtained from the LIMB demonstration project at
Edgewater Station, Ohio Edison, in Loraln, OH Numerous F G D sludge-fly ash-lime formu-
lations were prepared, cured, and tested for ultimate compressive strength to determine suit-
able formulations for the model fixed wastes Samples were prepared and placed into cyhn-
drlcal molds using ASTM Standard Method for Mechanical Mixing of Hydraulic Cement
Pastes and Mortars of Plastic Consistency (C 305-65) The fixed samples were cured at 100%
humidity for a m i n i m u m penod of 28 days before testing Prior to solidification, the wastes
were charactenzed as to pH, alkahnlty, organic carbon content, metal content, and sulfate and
sulfite concentration to establish baseline conditions
Solidified waste sample cylinders were analyzed for compressive strength using ASTM Stan-
dard Test Method for Compressive Strength of Cyhndncal Concrete Specimens (C 39-86).
Permeablhties are being determined using a tnaxlal cell permeameter by ASTM Standard Test
Method for Permeability of Granular Soils [Constant Head (D 2434-74] Pore size distribu-
tions were determined by a mercury intrusion poroslmeter The pH stability of the waste was
determined by measunng acid neutralizing capacity
Samples were manually crushed and sieved to obtain particles in the 4 8 to 9 5-mm size
range for use in the extraction tests This size range has been shown to be the most effective
for the multiple extraction procedure Crushed samples were used because the leaching pro-
cedure proposed is based on a surface diffusion model Crushing the waste increases the surface
area and accelerates leaching However, the diffusion rates remain the same and can be used
to predict leaching rates from intact specimens
The effects of liquid solid ratio on metals leaching was evaluated The leaching procedure
used is a modification of that proposed by the American Nuclear Society for solidified low-
level radioactive wastes in Standard ANSI/ANS 16 1, adopted in 1986 The pnncipal differ-
ences are that a crushed waste (4 3 to 9 5 mm) is used in place of a monolithic one, and several
leachants are used to measure the intrinsic properties of the solidified wastes rather than the
delonized water leachant used in ANS 16. l The latter has been suggested by researchers at
Environment Canada [26,28] At appropriate intervals over a 30-day period, the leachate was
decanted from the extractors for analysis, fresh leachant was added, and extraction continued
Small (l-g) samples of the solids were taken at each sampling time for porosity measurement
The leachant used in the portion of the research reported here was delonlzed water at three
liquid-to-solid ratios (L S = 4, 20, 100) These ratios were selected to determine whether the
rate of metal leaching is controlled by solubility product or diffusion All aqueous samples were
analyzed for pH, alkalinity, oxidation-reduction potential, total dissolved solids, and metals
found to be of interest from the F G D sludge and fly ash characterizations Standard analytical
methods were used throughout [29] Metals were analyzed by graphite furnace atomic absorp-
tion spectrophotometry (AAS). Solid samples from each leaching period were also analyzed
for porosity and pore size distribution using a 0 to 206 840 kPa (0 to 30 000 psi) mercury intru-
sion porosimeter Pore surface area was measured by mtrogen adsorption using the Brunauer,
Emmet, and Teller (BET) procedure
Results
Samples of fly ash and F G D sludge were obtained from ConesvIlle Station, and LIMB ash
and coal ash were obtained from the Edgewater plant. Samples of each were analyzed for min-
eral ash content. Table 2 presents results of these analyses The mineral analysis is typical of
ash from high volatile bituminous coal found in western Pennsylvania and eastern Ohio.
These coals typically have an ash content of 10 to 14% by weight (dry basis) and generally have
a sulfur content from 3 3 to 3 5%
The samples were also analyzed for metal content by inductively coupled plasma emission
spectroscopy and by flame atomic absorption spectrophotometry (AAS) Table 3 presents the
metal concentrations The two analytical procedures gave comparable results In nearly all
cases, but there were a few discrepancies These discrepancies are still under investigation. The
AAS procedure resulted in higher concentrations of cadmium, nickel, and silicon, which were
closer to expected concentrations, and a lower concentration of arsenic. A comparison of these
results with typical fly ash and F G D sludge charactenstIcs given in Table 1 shows that the
materials being evaluated in this research are fairly typical.
The Conesville fly ash has high concentrations of several regulated compounds, including
arsenic (457 mg/kg), boron (139 000 mg/kg), barium (2530 mg/kg), chromium (123 mg/kg),
nickel (448 mg/kg), lead (232 mg/kg), vanadmm (254 mg/kg), and zinc (359 mg/kg). This fly
TABLE 2--Mmeral ash analysts offly ash, LIMB ash, and FGD sludge
Conesvflle Conesvdle Edgewater Edgewater

Fly Ash, FGD LIMB Coal Ash,
%a Sludge, % Ash, %a %b
Sdlcon dioxide 34 76 1 32 21 86 48 85
Calcmm oxide 2 38 40 30 33 80 1 61
Sulfur tnoxlde 1 52 11 58
Ferric oxide 3 49 0 17 1 07 19 01
Titanium oxide 0 37 0 01 0 69 1 06
Aluminum oxide 52 81 0 40 21 94 22 84
Magnesium oxide 0 74 1 44 1 00 0 73
Sodmm oxxde 2 46 0 08 1 89 0 16
Potassium oxxde 0 40 0 19 0 31 2 20
Phosphorus pentoxlde 0 21 <0 01 0 18 0 35
Sulfates (as SO3) 4 61 1 86
Sulfites (as SO2) 40 80
Loss of ,gnltlon 1 47 10 89 4 59
pH 3 44 12 78
Alkahmty, mg/L, as CaCO3 826 204
a Analyses by Broeman Laboratories, Inc

b Analyses by Consohdat~on Coal Co
ash also has very high concentrations of iron, aluminum, calcium, potassmm, and magne-
sium The Edgewater LIMB ash is also high in the same elements.
Based on these analyses, it was decided to limit metals analyses in the leachates generated
from solidified/stabilized wastes to arsenic, boron, chromium, lead, vanadium, zinc, calcium,
and iron
Several formulations were evaluated for the sohdification/stabihzatlonof fly ash, LIMB ash,
and F G D sludge Table 4 shows a hsting of formulations studied The Conesvllle fly ash and
Edgewater LIMB ash formulations were prepared in the laboratory The Conesville FGD
sludge had pozzolanlc properties and would set up quickly, so mixes were prepared at the
power plant as the sludge came from the vacuum filters.
All solidified samples were cured for 28 days and then tested for unconfined compressive
strength Results are shown in Table 4. Many of the formulations tested passed the suggested
criterion of at least 344 74-kPa (50-psi) compressive strength. It is apparent that the Conesvllle
and Edgewater fly ashes have pozzolamc properties Very high compressive strengths were
achievable with the Edgewater LIMB ash. Figure 1 shows contour plots of compressive
strength as a function of composition. These plots Indicate that additional lime is not needed
with the LIMB ash, but lime is needed to increase the compressive strength of sohdlfied/sta-
blllzed fly ash
Based on these results, four formulations were selected to evaluate their ablhty to contain
heavy metals and prevent them from leaching. Mix A contained 45% F G D sludge, 50% fly
ash, and 5% lime The sludge contained all necessary water MIX B was composed of 55% F G D
sludge, 43% fly ash, and 2% lime This is the formulation used at Conesvllle Mix C contained
75% fly ash, 10% lime, and 15% water No F G D sludge was added Mix D contained 75%
LIMB ash and 25% water
Initial leaching studies were conducted using rotary mixers as specified by the EP toxicity
procedure However, the abrasion resulting from the tumbling action caused the samples to
disintegrate quickly even though they had suitable compressive strengths Consequently, sub-
sequent leaching tests were performed in a static manner
Samples
Copyright of(all
by ASTM Int'l each
rights mix were
reserved); Mon set up23:07:03
Feb 24 for leachate
EST 2020 testing using three different hquld/sohd ratios,
TABLE 3--Metals analysis offly ash, LIMB ash, and FGD sludge, mg/kg a
Conesvdle Fly Ash Conesvdle FGD LIMB Ash
Element ICP AA ICP AA ICP AA
Ag < 0 01 < 0 01
AI 131 000 111 131 10631 51 000 64097
As 457 24 <2 384 39
Au < 0 005 < 0 005
B 139 000 330 000
Ba 2530 2050
Be 29 2 25 4
Ca 17 800 207 000
Cd < 0 001 16 < 0 001 22
Co < 0 001 < 0 001
Cr 123 136 10 61 0 77
Cu 138 46 5
Fe 181 000 74 000
Hg < 0 01 < 0 01
I <0 1 <0 1
In <0 1 <0 1
K 12 800 12 200
Ll 152 54 7
Mg 4480 4180
Mn 278 164
Mo 53 2 201
Na < 0 01 < 0 01
N1 < 0 005 448 62 < 0 005 113
P < 0 01 < 0 01
Pb 232 204 671 690 413
Pt < 0 01 169
S 8440 58 800
Sb < 0 005 < 0 005
Se < 0 01 < 0 01
S~ < 0 005 646 2307 < 0 005 6734
Sn <0 1 <0 1
Sr 654 374
Tl 5760 3040
TI < 0 05 < 0 05
V 254 267 29 94 3 83
Zn 359 329 147 286 668
a AA analysis by method of standard ad&tlons AA = atomic absorption, ICP = inductively coupled

plasma
with delonlzed water as the leachant Leachant was decanted and replaced with fresh leachant
on a defined schedule.
Table 5 shows leachate p H values for the first 14 days o f leaching for each mix Figure 2
shows trends in p H changes with t i m e for the 20" l L ' S samples The leachates from the F G D
sludge and fly ash mixes were always slightly to moderately alkahne. Leachates from the L I M B
ash mixes were always highly alkahne, even after five extractions This alkalinity was to be
expected since the L I M B ash had a large excess o f h m e The relatively high p H values should
help m i n i m i z e metals leaching for most metals
Figure 3 shows plots o f c u m u l a t i v e alkalinity leached as calcium carbonate (CaCO3) versus
the square root o f t i m e for Mixes A and D. The plots are essentially hnear, indicating that
leaching is diffusion controlled, but the liquid.solid ratio had a significant effect on the a m o u n t
o f leachmg o c c u m n g
TABLE 4--Formulations evaluated for sohdtficatton/

stabdtzatton offly ash, LIMB ash, and FGD sludge, and
resultmg unconfined compressive strengths (UCS)
CONESV1LLE FLY ASH
M~x NO FA, % Lime, % Water, % UCS, psi
1 90 0 10 145
2 85 0 15 103
3 65 10 25 437
4 75 10 15 1055
5 85 5 10 Too thick
6 80 10 10 Too thick
CONESVILLE FGD SLUDGE
Mlx NO FGD, % FA, % Lime, % UCS, psi
1 55 43 2 10
2 45 50 5 230
3 50 47 3 130
4 55 43 2 73
5a 55 43 2 24
EDGEWATER LIMB ASH
Mix No LIMB Ash,% Lime,% Water,% UCS, ps~
1 75 0 25 1614
2 65 0 35 38
3 55 0 45 Too thick
4 70 5 25 40
5 60 5 35 856
6 50 5 45 167
7 65 10 25 Too thick
8 65 10 35 635
9 65 10 45 121
Field sample
TABLE 5--Leachate pH values
Leaching Time, Days

Llqmd/Solld
Mix Formulation Ratio 1 3 5 7 14
A 45% FGD 4 1 98 85 94 91 106

50% FA 20 l 92 79 80 80 90
5% Lime 100 1 77 70 69 65 73
B 55% FGD 4 1 89 82 87 86 98
43% FA 20 1 90 87 89 88 92
2% Lime 100 1 7 1 65 66 65 76
C 75% FA 4 1 83 82 86 85 97
10% Lime 20 1 84 74 77 80 93
15% Water 100 1 7 1 65 66 65 76
D 75% LIMB 4 1 117 11 8 125 124 124
25% Water 20 1 11 1 12 3 12 5 12 2 12 4
100 1
10 2 100 11 8 11 6 11 8
BISHOP ET AL. ON CHEMICAL FIXATION 111
e.
,q
,o
V
o9~-" 1 o
,," z s d ) H/._.ON__T6'/,9
13
12
11
10
:32
EL
9-
8-
7-
6 i r i i i r i i ~11 r ~ I I I J El I I J I ~ I II I I I ~ I I E I I I J r E
0 4 8 12 16
Time, days
FIG. 2--Effect of leaching time on leachate pH for various fly ash, FGD sludge, and LIMB ash admix-
tureformulations leached at an L:S of 20.l.
Tables 6 and 7 show leachate c h r o m i u m and lead concentrations, respectively, for each mix.
In all cases, leachate metals concentrations were very low, ranging from less than 1 to 50 #g/
L for c h r o m i u m and from less than 1 to 233 # g / L for lead. This represents less than 1% of the
metals present in the waste.
Figures 4 and 5 show plots of cumulative c h r o m i u m leached and cumulative lead leached
versus the square root o f time, respectively, for Mixes A and D. The plots for the three liq-
TABLE 6--Effect of liquid~solids ratio on leaching of chromium from solidified~stabilized FGD sludge
and fly ash (ug/L).
Liquid:Solid
Mix Formulation Ratio Day 1 Day 3 Day 5 Day 7 Day 14
A 45% FGD 4:1 11.7 2.9 13.1 5.8 7.9

50% FA 20:1 11.7 2.1 12.7 4.1 2.5
5% Lime 100:1 1.9 1.5 1.7 0.3 1.9
B 55% FGD 4:1 1.9 2.1 11.2 7.3 15.4
43% FA 20:1 45.0 1.5 21.9 49.5 13.0
2% Lime 100:1 18.8 1.5 9.2 12.5 2.9
C 75% FA 4:1 33.5 12.0 15.7 7.1 8.6
10% Lime 20:1 19.1 2.8 9.8 5.0 7.3
15% Water 100:1 4.5 1.2 11.3 1.5 4.9
D 75% LIMB 4:1 3.1 1.9 4.8 1.8 8.3
25% Water 20:1 1.7 0.1 5.9 4.2 9.7
100:1 0.I 0.6 4.7 0.8 3.7
BISHOP ET AL. ON CHEMICAL FIXATION 113
10000
A o
oh
oh
E 8000
(1)
(-
o 6000
o
q.)
>,,
E
4000
o
,..r
o
@
> 2000
._
E
(..) 0 iill I ~ l l l l l ~ l l ~ l l E l ~ f ~ l l l l ~ l l r l l l l l r l l I
1 2 5 4
Square root of time (days)
80000 o
100:1
o~
B
o~
E
~J60000
273
ff:E
9
0 o " 20:1
9
~40000
9
4:1
9 20000
>
9
3
E
0 0 i i i i I
0 1 2 5 4
FIG. 3--Cumulative alkalinity leached versus the square root of time: (a) Mix A at 3 L.'S ratios. (b) Mix
D at 3byL:S
Copyright ASTMratios.
Int'l (all rights reserved); Mon Feb 24 23:07:03 EST 2020
---,8OO
cm
9 600
_s
9
o
9
E
400
E
o
k~
(D
9> 2 O O
~ 4:1 / o
9 D/ /
/
/
E
(D
0 rll]ll]l[llllrl~lrlLi-]l~3~-W-lllll~l~llEi I
1 2 5 4
~1000
c~
/
800
nO
//
9
0
~100:1
@
-- 600
E
o 4O0
LD
9
>
'~ 2OO
o
E
0
0
3 2 3 4
FIG, 4 - - C u m u l a t i v e chromium leached versus the square root of time: (a) M i x A at 3 L:S ratios. (b)
M i x by
Copyright D ASTM
at 3 L:S ratios.
B I S H O P ET AL. O N C H E M I C A L F I X A T I O N 115
2500-
o7
_Y_
2000
A H
o/100:1
/
@
1500
9
9
~D
1000
9
>
4--'
9
5oo
E ~ 4:1
O G ~
~ a
0 I [ I I I I ~ i i i E t i i i i i F iiI i i [ i t ~ i i i i i i i E i i i I
0 1 2 3 4
12000
s
cy~ o
~10000
o7
7:~ 8000
@
9
9
@ 20:1
-- 6000
7D
9
|
9 4000
>
9
E 2000
4:1
O
i i i i i [ I F ] i i i i i i i ii]
0 1 2 5 4
FIG. 5 - - C u m u l a t i v e lead leached versus the square root o f time. (a) M i x A at 3 L:S ratios. (b) M i x D
at 3 L:S
Copyright ratios.
by ASTM Int'l (all rights reserved); Mon Feb 24 23:07:03 EST 2020
116 HAZARDOUSWASTES
TABLE 7--Effect of hquM/sohd ratio on leaching of leadfrom sohdtfied/stabthzed FGD sludge and
fly ash (ug/L)
Llquld/Sohd
Mix Formulation Ratio Day 1 Day 3 Day 5 Day 7 Day 14
A 45%FGD 4 1 328 78 366 214 318

50%FA 20 1 226 36 160 123 186
5% Lime 100 1 49 20 83 22 59
B 55% FGD 4 1 14 1 142 55 4 50 2 125 7
43% FA 20 l 233 7 l0 7 69 8 204 3 64 5
2% Lime 100 1 74 2 45 31 2 72 9 16 9
C 75%FA 4 1 282 73 481 431 519
10% Lime 20 1 35 5 39 22 0 21 4 22 1
15% Water 100 1 15 <1 0 68 30 54
D 75% LIMB 4 1 39 8 30 6 56 5 62 3 153 1
25% Water 20 1 34 6 50 74 2 79 1 176 4
100 1 208 40 268 144 398
uld solids ratios are linear, indicating that diffusion does control leaching However, the 4" 1
ratio samples leached considerably less chromium than the 20. l and 100:1 samples, indicating
that solubility product constraints may have been in effect Similar results were obtained with
the other mixes and with lead, but there was more scatter of the data and the trends weren't as
clear
Summary
These preliminary results indicate that coal fly ash, F G D sludges, and LIMB ash can easily
be sohdlfied/stabilized to achieve a product that meets environmental regulatory standards.
Formulations can be developed that result in high ultimate compressive strengths, suitable for
landfilhng. However, abrasion resistance is low and the long-term stability of the material is
uncertain
Only a very small fraction of the heavy metals contained in the solidified/stabilized wastes
is leached when exposed to multiple extractions with delonized water The amount of leaching
was influenced by the liquid'sollds ratio used, though A L.S ratio of 20 1 was found to leach
a great deal more heavy metals from most S/S samples than a 4:1 mix, while smaller differ-
ences existed between 20:1 and 100 1 mixes
Acknowledgments
Grant RI3-88-013 from the Ohio Coal Development Office of the Ohio Department of
Development is greatly acknowledged for partial support of this project.
References
[1] "1988 Ohio Coal Development Agenda-Draft," Ohio Coal Development Otfice, Columbus, OH,
1987
[2] Shepard, M, "Toxic Research The Real World," EPRIJournal, Vol 12, No 6, 1987, pp 17-21
[3] Santhanam, C, Balasco, A, Bodek, I, and Cooper, C, "Full-Scale Field Evaluahon of Waste Dis-
posal from Coal-Fired Electric Generating Plants," EPA 600/7-85/028 A-F, EnvironmentalProtec-
tion Agency, Washington, DC, 1985
[4] "Sulfur Oxides Control Technology Series Flue Gas Desulfunzatlon, Lime/Limestone Processes,"
EPA 625/8-81-006, EnvironmentalProtection Agency, Washington, DC, 1981
[5] Moore, T , "Groundwater Examininga Resource at Risk," EPRIJournal, Vol 10, No 8, 1985, pp
6-9
[6] Shepard, M, "Utility Solid Waste Managmgthe By-Products of Coal Combustion," EPRIJournal,
Vol 10, No 8, 1985, pp 20-35
[7] "Disposal of Flue Gas Desulfurlzatlon Wastes Shawnee Field Evaluation," EPA 625/2-80-028,
EnvironmentalProtection Agency, Washington, DC, 1980
[8] Torvey, S, CoalAsh Utdlzatton, Noyes Data Corp, Park Ridge, NJ, 1978
[9] "Lime-Fly Ash-Stabilized Bases and Subbases," National Cooperative Highway Research Program
Synthesis 37, Transportation Research Board, Washington, DC, 1976
[10] LechIch, A and Roethel, F, "Manne Disposal of Stablhzed Metal Processing Waste," Journalof
the WaterPollutlon ControlFederatton, Vol 60, 1988, pp 93-99
[11] Van der Sloot, H, Wijkstra, J, van Stlgt, C, and Hoede, D, "Leaching of Trace Elements from
Coal Ash and Coal Ash Products," Wastes tn the Ocean, Vol 4, I Duedall, et al, Eds, John Wiley
and Sons, 1985, pp 467-497
[12] Baker, M, Jr, "FGD Sludge Disposal Manual," FP-977, Electric Power Research Institute, Palo
Alto, CA, 1979
[13] Tlckanen, L and Fitzpatrick, J, "Chemistry of Leachates from Raw and Stabilized Coal Fired
Wastes," Proceedings of the 39th Purdue Industrial Waste Conference, Vol 39, Butterworth Pub-
hshers, Stoneham, MA, 1985, pp 187-199
[14] "Field Data of Leachates Provide Additional Basis for Modifications of the FOWL Code," Land and
Water Quahty News, Vol 5, No l, 1990, pp 3-5
[15] Czernln, W, Cement Chemistry and Phystcsfor Civil Engineers, Foreign Pubhcatlons, New York,
1980
[16] Brown, T , Bishop, P, and Gress, D, "Use of Upflow Column Leaching Test to Study the Release
Patterns of Heavy Metals from Stabilized/Solidified Heavy Metal Sludges," Hazardous and Indus-
trtal Sohd Waste Testing and Dtsposal Stxth Volume, ASTM STP 933, D Lorenzen, Ed, Amencan
Society for Testing and Materials, Philadelphia, 1986, pp 79-92
[17] Bishop, P, Brown, T, and Shively,W, "Alkahnity Releasesand the Leaching of Heavy Metals from
Stablhzed/Sohdlfied Wastes," Chemistry for Protectton of the Envlronment, L Pawlowski, G
Alaerts, and W Lacy, Eds, Elsevier Science Publishers, Amsterdam, The Netherlands, 1986, pp
217-233
[18] Bishop, P and Gress, D, "Evaluation of the Potential for Manne Disposal of Sohdified/Stabihzed
Inorganic Hazardous Wastes," completion report submitted to the Office of Sea Grant, National
Oceanic and Atmospheric Administration, Silver Spnng, MD, 1986
[19] Bishop, P, "Development of a Dynamic Leaching Test for Stablhzed/Sohdlfied Inorganic Hazard-
ous Wastes," completion report submitted to the New Hampshire Office of Waste Management,
Concord, NH, 1985
[20] Bishop, P, "Prediction of Heavy Metal Leaching Rates from Stablhzed/Sohdlfied Hazardous
Wastes," Toxtc and Hazardous Wastes, G Boardman, Ed, Technomic Pubhshmg Co, Lancaster,
PA, 1986, pp 236-252
[21] C6t~, P, "Contaminant Leaching from Cement-Based Waste Forms Under Acidic Conditions,"
Ph D dissertation, McMaster University, Hamilton, Ont, Canada, 1986
[22] C6t6, P, "Evaluation of the Long Term Stability of Solidified Wastes," Proceedings of the Waste
Stabthzatlon/Sohdtficatton Workshop, Alberta Environmental Centre, Vegrvevllle,Ont, Canada,
1983, pp 82-117
[23] Godbee, H and Joy, D, "Assessment of the Loss of Radioactive Isotopes from Waste Sohds to the
Environment Part I Background and Theory," ORNL-TM-4333, Oak Ridge National Laboratory,
Oak Ridge, TN, 1974
[24] Godbee, H, Compere, E, Joy, D, Kabbey,A, Moore, J, Nestor, C, Anders, O, and Nellson, R,
"Apphcatlon of Mass Transport Theory to the Leaching of Radlonuchdes from Waste Solids,"
Nuclear and Chemtcal Waste Management, Vol 1, 1980, pp 29-35
[25] C6t~, P, Bridle, T, and Benedek, A, "An Approach for Evaluating Long-Term Leachability from
Measurement of Intnnslc Waste Properties," Hazardous and Industrml Sohd Waste Testing and
Disposal Sixth Volume, ASTM STP 933, D Lorenzen et al, Eds, American Society for Testing
and Materials, Philadelphia, 1986, pp 63-78
[26] C6t6, P, "Stablhzatlon/Sohdlficatlon of Hazardous Waste Seminar," Environmental Protection
Agency, Cincinnati, OH, 16 Feb 1988
[27] Culhnane, M, Bmhop, P, Jones, L, Pitt, W, and Tlttlebaum, M, "Physical Treatment," Research
Needs Workshop Hazardous Waste Treatment and Disposal, R Loerner, G Parkln, and W P~pes,
Eds, National Science Foundation, Washington, DC, 1986, pp 35-42
[28] "Test Methods for SolidifiedWaste Characterization," Draft January 1986, EnwronmentalCanada,
Burhngton, Ontario, Canada
[29] Standard Methods for the Exammatlon ofWaterand Wastewater, 16th ed, American Pubhc Health
Assn, Washington, DC, 1985
Nancy E. Kinner, ~Diana B. Durling, 2 W. Berry Lyons, 3 Davtd L.
Gress, l and M. Robin Collins 2
Evaluation of the Impact of Ocean Disposal of

Stabilized/Solidified Municipal Solid Waste
Incinerator Ash on Marine Sediments
REFERENCE" Klnner, N E , Durhng, D B , Lyons, W B , Gress, D L , and Colhns, M R ,
"Evaluation of the Impact of Ocean Disposal of Stabilized/Solidified Municipal Solid Waste
Incinerator Ash on Marine Sediments," Stabthzatton and Sohdtficatlon of Hazardous, Radm-
active, andMlxed Wastes, 2nd Volume, ASTMSTP 1123, T M Gllham and C C Wiles, Eds,
American Society for Testing and Materials, Philadelphia, 1992, pp 119-132
ABSTRACT" This research explored the feasibility of ocean &sposal of stabxhzed/sohdlfied

(S/S) municipal solid waste (MSW) incinerator ash in orgamc-nch near-shore m a n n e sediments
A mixture of MSW bottom ash (92%) and fly ash (8%) from a starved-air mass burn MSW incin-
erator was combined with Type II portland cement in a 4 54 1 ash cement ratio After cunng,
specimens were cut and placed in centrifuge tubes filled with sediment Controls containing con-
crete (cement and sand) specimens and containing sediments alone were also prepared All sam-
ples were incubated in a flow-through seawater system Over the first ten months of the experi-
ment, the sediment controls reflected conditions similar to those m northern New England
estuaries with respect to pH (7 8 + 0 1), alkahmty (8 7 + 1 0 meq/L), total dissolved iron ( 16 7
+ 4 0 mg Fe/L) and SO42 /C1- (mM) ratios (0 034 _+ 0 0025) The S/S ash and concrete control
samples exhibited a higher pH (8 9 to 9 6) and more reducing conditions (SO2-/C1 - ratio =
0 015 to 0 020, alkahnlty = 3 0 to 5 6 meq/L, total &ssolved iron = 0 08 to 0 67 mg/L) In their
sediments than the sediment controls No statistically slgmficant [confidence interval (CI) =
90%] amounts of metals (arsenic, copper, lead, cadmium) were detected in the se&ments adja-
cent to the S/S ash specimens as compared to the controls Either ( 1) the metals were not being
released from the S/S ash into the surrounding sediment, or (2) they were released but at con-
centratlons below our detection capabilities
KEY WORDS. stabdxzed/solldified municipal sohd waste incinerator ash, organlc-nch m a n n e

sediments, blogeochemlstry, arsenic, cadmium, copper, lead
T h e disposal o f m u n i c i p a l solid waste ( M S W ) has r e a c h e d crisis p r o p o r t i o n s in m a n y areas

o f t h e U n i t e d States a n d E u r o p e In the U n i t e d States, estimates are t h a t t h e r e will be a n aver-
age 1 8% increase p e r year in t h e v o l u m e o f M S W g e n e r a t e d u p to the year 2000 [ 1 ] As a result,
t h e U n i t e d States will p r o d u c e 180 m i l l i o n t o n s / y e a r o f M S W at the b e g i n n i n g o f the 21 st
c e n t u r y T h e r e are presently t h r e e alternatives for M S W disposal landfilhng, i n c i n e r a t i o n , a n d
recycling I n c i n e r a t i o n is used p r i m a r i l y because it results in a 70% r e d u c t i o n in weight a n d
90% v o l u m e r e d u c t i o n [2] a n d h a s s o m e i n h e r e n t h e a t i n g value w h i c h c a n b e used to generate
s t e a m o r electricity
i Associate professor of civil engmeenng, Environmental Research Group, University of New Hamp-
shire, Durham, NH 03824
2 Graduate student and assistant professor of civil engmeenng, respectively, Environmental Research
Group, University of New Hampshire, Durham, NH 03824
3 Professor of geochemistry, Hydrology/Hydrogeology Program, Mackay School of Mines, University
of Nevada, Reno, NV 89557
119

One of the primary concerns associated with operating MSW incinerators is the problem of
disposal of the fly and bottom ashes generated Estimates indicate that 2 5 to 4 million tons of
incinerator ash are generated each year in the United States, and that figure is projected to
increase to 17 million tons per year by the year 2000 [3] Disposal of the ash is difficult because
it contains relatively high concentrations of metals and noncombustible organics Currently,
most ash is placed in lined landfills designed with leachate collection and groundwater mon-
itoring systems
Several innovative alternatives to the ash disposal problem have been studied by the Envi-
ronmental Research Group (University of New Hampshire) including (1) codlsposal with
wastewater sludge [4], (2) stabilization with sodium carbonate (Na2CO3) and trona amend-
ments [5], and stabilization/solidification (S/S) with cement, lime, or asphalt amendments
[6]. All of these solutions require final disposal of the modified ash in secure landfills In some
European countries, the feasibility of deep storage of incinerator ash in salt mines is being
explored
Another option is to create an S/S monolith and place it in the marine environment In the
marine environment, monoliths would not be subject to freeze-thaw cycling or acid precipi-
tation In addition, the ocean is very alkaline, which should limit dissolution of metal hydrox-
ides and metal carbonates Placement in anaerobic sediments containing large quantities of
EHES (total dissolved hydrogen sulfides species) could result m precipitation of any metals
released from the S/S monohth
Marine sediments consist of a series of blogeochemlcal zones where various microbial reac-
tions, consuming orgamc carbon, control the redox chemistry The sequence of these zones is
well documented, but the depth of each zone is primarily dependent on the sedimentation rate
and, hence, the burial rate of organic carbon These zones are termed OXlC, suboxlc, and
anox~c In the OXlCzone, microbes as well as larger mfaunal organisms oxidize organic matter
using oxygen (O2) as a terminal electron acceptor In the suboxlc zone, NO~, Mn 4+, and Fe 3+
are used by the microorganisms as electron acceptors In the anoxlc zone, reduction occurs
until SOl is exhausted, and then methanogenesls becomes the predominant mode of bacterial
energy generation The higher the sedimentation rate, the narrower the zones become In estu-
a n n e and coastal environments in northern New England, where sedimentation rates are
between 0 1 to 0 2 cm/year, the depth of both the OXlCand the suboxlc zone may be less than
1 cm during the warmer periods of the year In this environment, SOl reduction begins just
below the sediment-water interface and continues until SO]- is depleted When SOl reducing
bacteria oxidize organic carbon, they produce 2CO2 and ~HES as by-products These com-
pounds can increase to very high (that is, mM) concentrations in the sediments Of all the
biogeochemical zones present in coastal marine sediments, the SO~ reduction zone theoret-
Ically provides the best location for placement of S/S incinerator ash, because most of the
potentially hazardous metals present in the ash will form very insoluble sulfide minerals
Therefore, any metals leaching from the S/S ash speomens should be rapidly removed from
the dissolved phase as authlgenlc sulfides and stabilized within the sediment column
In less organic-rich sediments, where SO 2- reduction is minimal, metals leaching from the
S/S ash specimens could be more mobile and hence available to enter the marine ecosystem
An understanding of the blogeochemlcal regime m the sediments where the S/S ash monohths
are placed ~s, therefore, extremely important ffthey are to be disposed of in an enwronmentally
acceptable m a n n e r The momtoring of such parameters are pore water S O ] , C1 , Z;HzS, Fe z+,
pH, and alkahnlty are needed to better estabhsh the ultimate fate of metals ffthey are leached
into the sediments
The objective of this research was to determine if metals were released from small speomens
ofS/S MSW incinerator ash into organic-rich marine sediments The specimens placed in the
sediments had (all rights
a largereserved); Mon Febarea
surface 24 23:07:03 EST 2020 ratio to maximize the potential leaching that
to volume
KINNER ET AL ON OCEAN DISPOSAL 121
might occur d u n n g the two-year study By simultaneously m o m t o n n g other geochemically

important parameters such as pH, alkahmty, CI-, SO4 2, and total &ssolved iron (Fe), the
redox status of the sediment pore water was determined and used to pre&ct the mobility of
the metals in the environment The metals arsenic (As), cadmium (Cd), lead (Pb), and copper
(Cu) were monitored because they typically have the highest enrichment in MSW ash when
compared with naturally occurring se&ment concentrations This paper reports the results
from the first ten months of the project
Methods and Materials

A detailed description of the methods and materials used m this research is available [7]
S/S Specimen Preparatton

A mixture of 92% bottom ash and 8% fly ash (by volume) was created from ash collected at
a local starved-air mass burn MSW incinerator Incineration at the facility occurs m a lower
chamber under starved-a~r, low turbulence condmons, and the hot gases are relgmted at 980~
m an upper chamber where excess air ~s added The bottom ash obtained from the incinerator
was sieved prior to mixing w~th the fly ash to remove all objects unable to pass a 1 3-cm mesh
The 92%/8% mixture was used because it represents the relatwe proportions of the two ashes
produced by the incinerator
Various ash cement and water cement ratios were tested to determine the m~xture produc-
ing S/S specimens that had adequate compresswe strength and workability and contained a
m a x i m u m a m o u n t of ash Based on evaluations using a umversal compresswe strength testing
machine and a standard Atterberg limits test [8], an ash cement (portland Type II) ratio of
4 54 1 and a water cement ratio of 1 75 1 was chosen [ 7] The mixture's compressive strength
and workability were 1080 psi (7446 kPa) (after. 30 days of cunng) and three blows, respec-
twely A control was prepared using Ottawa silica sand (ASTM Test Method for Compressive
Strength of Hydrauhc Cement Mortars (Using 2-1n or 50-mm Cube Specimens) m place of
the MSW ash The sand cement and water'cement ratios of the concrete controls were 4 54 1
and 1 75 1, respectively The ash and control pastes were placed into molds (38 by 7 6 by 7 6
cm) filled m one-third volume increments and compacted 60 times with a standard steel com-
paction rod The beams were allowed to cure for 28 days (100% huml&ty) before being cut
into 2 5 by 7 6 by 0 6-cm specimens using a masonry saw fitted with a diamond blade Th~s
specimen size was the m a x i m u m that would fit m 50-mL polypropylene centrifuge tubes
packed with an mtemdal organic-rich m a n n e se&ment These specimens also had a large sur-
face area per umt volume to maximize their contact w~th the sediments
The centrifuge tubes containing the S/S ash and control specimens were submerged m 28
cm of seawater that was flowing ( ~ 1 L/mm) through plastic tanks located at the Umversay's
Coastal M a n n e Laboratory In addition, some tubes were filled with sediment alone to serve
as controls to ensure that experimental condmons were similar to those normally observed m
northern New England coastal sediment The S/S ash samples and controls were placed m
separate tanks All tanks were covered to hm~t photosynthetic actwity
Samphng Protocol
Each month, four S/S ash, two concrete control, and two sediment control tubes were
capped, removed from the seawater tank, and processed m glove boxes sparged with nitrogen
gas to maintain anaerobic conditions The S/S ash and concrete control specimens were
Copyright
removed by ASTM Int'l (alltheir
from rights reserved);
tubes Mon Feb 24 23:07:03
Attached EST 2020 was collected along with the sediment remain-
sediment
122 HAZARDOUSWASTES
lng in the tube that had been adjacent to the speomens during incubation Subsequently, this
sediment was processed for metals analyses The remaining sediment was centrifuged ( ~ 5000
rpm for 15 min) to obtain the pore water that was analyzed after filtration (GF/C) for pH,
alkahnlty, Cl-, SO 2-, and total dissolved Fe The specimens were placed in 0 1% sodmm pyro-
phosphate and sonicated (two 30-s on/offcycles per side) to remove the blofilm associated with
their surfaces
Analyttcal Methods
The sediment was dried at 50~ and then subjected to a 10% hot (60"C) nitnc acid (HNO3)
extraction for 20 to 24 h [9] Cu, Pb, and Cd were analyzed by flame atomic adsorption [10],
while As was analyzed using a graphite furnace according to the method developed by Chor-
m a n n [11 ] The detection hmlts for Cu, Pb, Cd and As were 8 65, 24 8, 1 03, and 1 7 tsg/g dry
weight of sediment, respectively 4 The bacterial extract was filtered and stained with 1-mM
acrlflavlne and examined using eplfluorescence microscopy pH was measured in the pore
water samples using a combination mlcroelectrode [10] followed by an alkalinity titration
with 0 15 N hydrochloric acid (HC1) (The CI- added during the titration was accounted for
In the C1- analysis ) The C1- was analyzed on a Dlonex 430011on chromatograph eqmpped
with amon separator and suppressor columns and a conductiwty detector Total &ssolved Fe
was determined using a ferrozlne method [12] and SOl- was analyzed using the turbidlmetnc
method [13]
Data Analysis
Statistical analyses were performed with respect to each chemical parameter to determine if
there were (1) slgmficant differences between the data obtained from the S/S ash and sediment
and concrete control samples over time, or (2) trends in the concentrations observed over time
SAS| [ 14] was used to determine If any significant differences existed among the various spec-
Imens at each sampling period using Scheff6's test Scheff~'s test was used since the sample sizes
were unequal A 90% confidence interval (CI) was chosen as acceptable for this study Mann-
Kendall [15] trend analysis was performed using a 90% CI on each sample type over the ten-
month sampling period This test was chosen since missing values could be handled and the
data did not need to conform to a particular distnbution The Mann-Kendall test uses relaUve
magmtudes of the data rather than their measured values

Pore Water Condlttons
The pore water SO4z /C1- ratios (mM/mM) are shown in Fig 1 The ratios were plotted in
order to normalize any changes in SOl- concentrations that may have been due only to
changes in salinity (that is, C1-) The SOl-/C1- relationship is commonly monitored because
C1 behaves conservatively, whde SO]- can be biologically "active" If microbial sulfate reduc -
tion is o c c u m n g
The sediment control SOl-/C1 ratios averaged 0 034 + 0 0025 and exhibited no overall
increasing or decreasing trend with time The seawater ratio is normally 0 052 Hines et al
[16] observed SO] /CI ratios ranging from 0 038 to 0 048 in tidal salt marsh sediments in the
Great Bay, NH estuary when SOl reduction was occurnng. The concrete control pore water
4 All ash and se&ment concentrations are expressed as dry weights

0.050
I 1 - - 1 1 Ash Exprmntl
0.045 O O Concrete cntl
Z~--Zk Sediment cntl
0.040
zE 0 . 0 3 5 -
E
0.030
E 0.025
I
0.020
Io 0.015.
0.010.
0.005.
0.000 I I I I I I I I I I
0 .30 60 90 120 150 180 210 240 270 300 330
Time (days)
Sept Oct Nov Dec Jan Feb Mar Apr May June
FIG I--SO~4-/CI- rattos (mM/mM) m the pore water of the S/S ash and concrete and sedzment control
samples
SO2-/C1 ratios averaged 0 020 + 0 0036 The lower initial ratio for the concrete control indi-
cated that SO42- was being removed from the pore water e~ther by incorporation into the con-
crete specimens or by microbial reduction An increasing trend in the SO2-/C1 - over time
existed for the concrete controls This may have reflected a decreasing rate of SOl- incorpo-
ration into the concrete specimens, thereby allowing more SO]- to remain in the pore water
The S/S ash SO2-/C1 - ratios averaged 0 0154 _+ 0 0056 Even with the high vanablllty
observed m the ratio, the S/S ash values were slgmficantly lower than those for the sediment
control pore water from September through January The low initial SO] /C1 ratio for the
ash specimens and the increasing trend was similar to the data obtained for the concrete con-
trol specimens.
The concentration of d~ssolved Fe present m the pore water in the sediment controls aver-
aged 16 72 _+ 3.96 mg/L (Fig 2) and was in the range observed by Hines [17] (0 21- to 23 7-
mg Fe/L) in Great Bay, NH sediments Though the Fe concentration fluctuated d u n n g the
ten-month sampling period, the values for the sediment controls over time were usually not
significantly different from one another
The h~gh pore water Fe concentrations m September may have occurred because the sedi-
ments were mixed before they were packed into the tubes Hines et al [18] noted that lnfaunal
reworklng and irrigation actlvmes (bloturbatlon) tend to maintain high dissolved Fe concen-
trations m the presence of rapid EH2S production by enhancing Fe and S cycling B~oturbatlon
enhances oxidation, which promotes the formation of dissolved F e m two addmonal ways (1)
burrowing activity allows the transport of Fe 3+ deeper into the sediments where it is reduced
to Fe 2+ and then may or may not form ferrous sulfide (FeS), and (2) redox reactions occumng
along the walls ofthe burrows are a source ofFe 2+ [17] In these expenments, t h e l n m a l m i x i n g
of the sediments may have created conditions s~mllar to those of bloturbatlon
The low Fe concentrations In October and November may have represented transition to
diffusion-dominated processes where mmroorgamsms were still active, but bloturbatlon or
effects from artificial perturbation due to sediment mixing decreased Infauna were not
observed in any of the samples Therefore, the Fe may have been precipitated as FeS if reduc-
ing condmons existed In October, the SO 2 concentratxon and SO] /C1 ratio of the sediment
35"
T T 9 9 Ash exprrnntl
l 0 0 C o n c r e t e cntl
~_ J" 30. cntl
~n
E 25.
t-
9..i,J
s 20.
15.
O
t-
O
0
10"
t-
O 5.
0 ! ,~ "! m ,m .~ m
0 30 60 90 120 150 180 210 24-0
Time (days)
Sept Oct Nov Dec don Feb Mar
FIG 2--Total dtssolved tron (mg Fe/L) m the pore water of the S/S ash and concrete and sediment
control samples
control decreased indicating SO 2 reduction was occumng The increased pore water Fe con-
centrations m December may have represented the remobihzatlon of Fe by the oxidation and
dissolution of metal sulfides Although Hines et al [28] observed low winter dissolved Fe lev-
els, Elderfield et al [19] and Lyons and Fitzgerald [20] noted that metals may be remoblllzed
during the winter by the oxidation and dissolution of metal sulfides Visual observation of the
depth of iron oxyhydroxide precipitates in the sample tubes (Fig. 3) indicated that the oxygen
was diffusing further into the sediment controls than into the concrete and S/S ash sediments.
From September through December, the apparent oxygen penetration m the tubes increased,
corresponding to an increase in dissolved Fe in December The apparent oxygen depth then
decreased through February, which may explain the concurrent decrease m Fe
The Fe data appeared to increase again from February to March Hines et al [18] noted that
SO~- reduction activity m Great Bay, N H sediments did not usually increase until mid-April.
Hines [17] also found that SO 2- reduction rates in Great Bay did not increase until the sedi-
ments had reached 12.5"C In our case, this temperature was not reached unnl June There
was no decrease in SO 2- concentrations or m the SO~-/C1- ratios through June, indicating a
lag m the onset of SO 2- reduction and sulfide precipitation dunng the spnng, which may have
allowed the remoblllzatlon of Fe to occur The increasing water temperatures m March and
April may have stimulated heterotrophic activity other than SO24- reduction (e g , fermenta-
tion), thereby consuming organic carbon in the sediment controls, producing reduced by-
products, lowering redox potential (Eh) and remobfllzmg Fe by the reduction and dissolution
oflron oxyhydroxldesto Fe 2§ [17] Thls reduced Fe would not tend to p r e o p l t a t e a s F e S since
microbial SO~- reducnon probably had not begun and therefore EH2S was not available
The dissolved Fe content of the pore water in the concrete control sediments was negligible
(Fig 2) throughout the experimental period (0 08 + 0 04 mg/L), and the apparent depth of
oxygen penetration was also small (Fig 3) The concrete control pore water Fe concentrations
were usually significantly lower than the sediment control concentrations In order for Fe to
precipitate as FeS in the reducing environment, there would have to be more SO~- reducnon
o c c u m n g in the concrete control than m the sediment control The v~sual apparent oxygen
penetration
Copyright by ASTM Int'lwas low,
(all rights indicating
reserved); Mon Feb 24that there
23:07:03 was a more reduced environment in the concrete
EST 2020
0.0
-1.0.
-2.0-
f=%
E
tJ -3.0-
t-
t~
q)
-4.0-
r-~
-5.0
II--ll Ash exprmntl
-6.0 0~0 Concrete cntl
~--A Sediment cntl
-7.0 I I I I I I I I I I
0 30 60 90 120 150 180 210 240 270 300 330
Time (days)
Sept Oct Nov Dec Jan Feb Mar Apt May June
FIG 3--Apparent depth (cm) of oxygen penetratton m the S/S ash and concrete and sediment control
samples as )udged by vtsual observatton of the depth of the red color in the sedtments
control If microbial SOl- reduction was not enhanced in the concrete control, then another
reducing agent must have been present in the concrete controls that was not present in the
sediment controls
The dissolved Fe content in the pore waters of the S/S ash samples averaged 0 67 _+ 1 79
mg/L The apparent oxygen penetration was also less than that observed in the sediment con-
trol, Indicating more reducing conditions were occurnng in the S/S ash samples.
The S/S ash pore water Fe concentrations were significantly lower than the sediment control
values; however, they were never significantly different from the concrete controls. Since the
condmons appeared to be more reducing in the concrete and S/S ash samples, any Fe present
in the pore water should have reacted with the available EHzS and precipitated as FeS. This
notion was supported by the visual observation that the sediments surrounding the concrete
control and S/S ash specimens were very black (FeS is a black precipitate). If the sediment
environment was highly reducing, most metals leached from the S/S ash specimens would
probably have precipitated or copreclpltated as highly insoluble metal sulfides.
The p H (Fig 4) of the sediment control pore water averaged 7.8 + 0 1 throughout the study,
which was expected since the pH of a coastal sediment is usually 7.5 or higher [21]. The sed-
iment control pore water pH exhibited no trend over time, which is typical for Great Bay, N H
sediments Hines et al [16] observed that the p H in the pore waters of salt marsh sediments
in Great Bay vaned from 6 0 to 7 5
The pore water associated with the concrete control had a pH averag3ng 9 6 + 0.3 (Fig. 4).
The high p H observed was probably the result of excess hydroxide concentrations from the
release of lime [Ca(OH)2] present in the anginal cement. The concrete controls m this study
were visibly very porous and easily fractured
Roethel et al [22], who performed seawater shaking leaching tests on S/S concrete blocks,
measured pHs of "-~10.6 They also observed high initial calcium fluxes from an S/S block
submerged into a seawater bath Using X-ray diffraction analysis, they found that calcium
compounds existed in the S/S blocks, which were soluble in seawater [Ca(OH)2 > gypsum
(CaSO4 2H20) > aragonite (CaCO3)] They postulated that the higher calcium flux in the
early stage of the leach tests was due to the dissolution of the free lime and gypsum present at
the surface
Copyright by ASTM of
Int'lthe blocks.
(all rights After
reserved); their
Mon Feb concrete
24 23:07:03 blocks were exposed to a seawater bath (no flow)
EST 2020
126 HAZARDOUSWASTES
11
10
"1-
8
7
9 9 expermtl
O O Concrete entl
6 A A Sediment cntl
5 I I I I I I I I I I
0 30 60 90
120 150 180 210 240 270 300 330
"Hme (days)
FIG 4--pH in the pore water of the S/S ash and concrete and sediment control samples
for 30 to 60 days, brucxte [Mg(OH)2] and aragomte (CaCO3) were formed, which slowed or
obstructed the release of calcmm by plugging the pore spaces
Shlvely [23] noted that calcium, known to exist as calcmm hydroxide, calcium silicate
hydrate, and other crystalhne structures in hydrated cement paste, ~s susceptible to leaching
even m a noncorrosive medium such as delomzed water Melchmger [24] observed a high pH
(approximately 9 5) in her S/S heavy metals sludges leached w~th seawater She postulated that
this was due to a flux of hydroxide from the paste to the leachant She observed a gradual
decrease (over 40 days to pH 8 5) in the pH of the leachant, indicating that a precipitate had
formed on the surface and sealed the concrete's pores
The concrete control in this research exhibited a decreasing trend m pH with values ranging
from pH 10 to 9 This decreasing trend may have been the result of the slowed calcium leach-
mg or the plugging of the pores by bruclte or aragomte The pH of the concrete control pore
water was significantly h~gher than the sediment control pore water in all cases from Septem-
ber through April
The S/S ash sediment pore water pH averaged 8.9 _+ 1 0, which was anticipated because of
the combined effect of the high pH from the hme, cement, and possibly the ash According to
Bagchl and Sopclch [25], the p H of M S W incinerator ash ranges from 8 9 to 9 9 The pH of
the ash pore water over time showed no trend, probably because the standard deviation was
very large However, the pH of the S/S ash pore water was slgmficantly h~gher than the sedi-
ment control pore water throughout most of the experimental period Again, th~s occurrence
was probably due to the influence of the cement The pH of the S/S ash pore water was only
slgmficantly lower than the concrete control dunng three months The shghtly lower pH may
have been due to predominance of the ash over the concrete or the less porous construction
of the S/S ash specimens
The sediment control pore water had alkallnmes of 8 7 _+ 1 0 meq/L (Fig 5) Hines et al
[16] found alkahmty values of 1 5 to > 10 meq/L m salt marsh sediments of Great Bay, NH,
which vaned seasonally depending on temperature and SO]- reduction rates Alkahnlty vanes
inversely with SO 2- concentration because, as SO ] is reduced by microorganisms, ECO2
(alkahmty) is(all rights reserved);
produced TheMon Feb 24 23:07:03control
sediment EST 2020 pore water alkahnlty showed a decreasing trend
KINNER ET AL ON OCEANDISPOSAL 127
16
/~ 9 9 Ash expermtl
O O Concrete cntl
14- tl
~-, 12-
~IO-
E
8-
0
v
4-
0 I I I I I I I I I I
0 30 60 90
120 150 180 210 240 270 300 330
Time (days)
FIG 5--Alkahnzty (meq/L) m the pore water of the S/S ash and concrete and sedzment control samples
over time This trend was probably due to the low water temperatures throughout the exper-
imental penod, which slowed the amount of SO] reduction activity and therefore decreased
the a m o u n t of corresponding alkahmty production
There was a sharp increase in alkahmty in October in the se&ment control The alkalinity/
SOl- ratios increased during th~s period This increase was probably due to SO] reduction
Normally, the m M alkallmty/SO] ratio averages 0 5 in seawater, but can range from 0 1 to
1 3 in m a n n e sediments, increasing with depth as SO 2- xs reduced and alkalinity increases
[26] The increase in alkahmty associated with the sediment control observed in October was
probably due to increased SO 2- reduction caused by the artificial perturbation of the se&ments
that occurred dunng the expenmental setup The alkalinity/SOl- ratio of 1 5 was near the
range expected for SO]- reducing conditions
The concrete control pore water alkalinity values (Fig 5) averaged 5 64 +_ 0 76 meq/L and
were not consistently &fferent from the sediment controls The concrete controls also showed
a decreasing trend over time A sharp increase in alkahmty in October was observed for the
concrete controls, and the alkahmty/SO] ratio approached 5 0 The increase in alkallmty
may have been due to (1) a corresponding increase in SO]- reduction as postulated for the
sediment control, or (2) the leaching of hydrated hme [Ca(OH)2] from the cement, adding
hydroxide alkalinity, since the pH also Increased After the first month, precipitates ofcalcmm
carbonate (CaCO3) and magnesium hydroxide may have begun to form on the surface of the
concrete control specimens, clogging the pores and reducing the amount ofalkahnlty released
It is also possible that SO]- was incorporated into the concrete This, or microbial SO] reduc-
tion, could have reduced the amount of SO]- available to be measured in the pore water
Therefore, the high alkalinity SO 2 ratio in October could have been caused by (1) an increase
m alkalinity (hme leaching), and/or (2) a decrease m SO]- surface incorporation (possibly as
CaSO4 2H20)
The alkahmty values m the pore water (Fig 5) of the S/S ash samples averaged 2 95 +_ 0 75
meq/L These values were slgmficantly less than the sediment controls throughout the entire
expenment,
Copyright by ASTM Int'lprobably as a Mon
(all rights reserved); result
Feb 24of precipitation
23:07:03 EST 2020 of CaCO~ The high pH of the sediment pore
128 HAZARDOUSWASTES
water for the S/S ash samples could have converted the bicarbonate ions (HCO;) normally
present in seawater to carbonate ions (COl-), which could have combined with the Ca 2§ in the
seawater or that Ca 2§ released with the lime The S/S ash alkahnitles were significantly lower
than the concrete control alkahnities during five months The reason for this is not known
The ash alkahnitles did not experience a sharp increase in October (Fig 5) and showed a
decreasing trend with time This may also have been due to the pore plugging phenomena
discussed earlier. As the specimens aged, the pores may have been plugged by precipitates so
that less leaching of lime occurred In addition, one would expect to see a decrease with time
m alkalinity since the difference (gradient) between the amount of lime in the specimen and
in the pore water should decrease
B~ofilm
The bacterial density on the S/S ash and concrete controls averaged 1 7 • 105 + 2 3 • 105
and 1 6 • 105 + 1 7 • 105 mlcroorganlsms/cm2, respectively (Fig 6) These values only rep-
resented the bacteria associated with the specimen surface and any attached sediment that
could be removed by the sodium pyrophosphate/sonlcatlon procedures There was no statis-
tical difference between the bacterial numbers associated with the S/S ash and concrete control
specimens, indicating that the S/S ash specimens did not inhibit bacterial fouling more than
the concrete controls An epifluorescent enumeration was performed on the bacteria associ-
ated with the sediments in each S/S ash, and concrete and sediment control sample d u n n g
June There was no significant difference in bacterial numbers among the three sample types,
which averaged 3 3 )< 10 6 + 2 7 X l 0 6 bactena/g sediment These data also indicated that
conditions in the S/S ash and concrete control specimens were not affecting bactenal numbers.
However, it should be noted that eplfluorescence does not indicate the level of microbial activ-
ity occurring, so the total impact of the S/S ash and concrete controls on sediment microor-
ganisms could not be assessed
-~ 1.2E6
O
-~ 1.1E6- 9 9 Ash Exprmntl
O - - O Concrete cntl
e~E 1.0E6-
o
O3 9.0E5-
E 8.0E5-
=
o
7.0E5-
6.0E5-
O
,9
~ 5.0E5-
O
4.0E5-
3.0ES-
2.0E5-
"~ 1.0E5-
z 0.0
0 30 60 90
120 150 180 210 240 270 300 330
"Hme (days)
FIG 6--Bactertal numbers associated wtth the S / S ash a n d concrete control specimens per cm e surface
area based on eplfluorescence microscopy
Metals
Arsemc--The average arsenic concentration of the MSW ash used in the S/S ash specimens
was 6 89 • 0 58 ~g As/g [8o/orelative standard deviation (RSD)] Though the As content was
low, it was within the range previously observed for this ash (3 to 1300 ~g/g) The As concen-
tration in the sediment controls averaged 4 04 • 0 12 #g As/g sediment (Fig 7), which IS typ-
ical of uncontaminated sediments [27] Not surprisingly, no significant increase in As was
found in the sediment associated with the S/S ash specimens over time. The low sediment
concentrations (4 11 • 0 38 #g As/g) were probably observed in the S/S ash samples because
the e n n c h m e n t factor of As in the S/S ash specimens versus the control sediments was rela-
tively small The average As concentration in the concrete control sediments (4 7 • 0 43 ~g/
g) was higher than the other two samples and increased over time However, there was never
a significant difference among the three types of samples
Copper--The values of Cu in the sediment controls averaged 54.3 • 20 6 #g/g, which was
within the range 6 3 to 201 #g/g reported for local estuanne sediments (X = 42 0 • 29 6 #g/
g) [28] (Fig 7) The copper concentrations in the sediments adjacent to the S/S ash specimens
(49 6 • 6 3 #g/g) were not significantly different from those of the concrete (46 1 • 5 6 ~g/g)
or the sediment controls, except in November and December when the S/S ash sediments were
significantly higher than the concrete controls There were no trends over time in any of the
Cu data The enrichment factor of the Cu in the MSW ash versus the sediments was 26 1
Lead---The concentration of lead in the MSW ash (1748 + 662 ~g/g) was highly vanable
and lower than expected This concentration was only twelve times greater than the sediment
control concentration ( 163 4 __+ 16.0 ug/g) Lead concentrations in New Hampshire estuanne
sediments have been found over the range 0 80 to 145 ug/g [28,29] Generally, the lead content
of the sediment control was either not significantly different or was significantly higher than
the S/S ash and concrete control sediments (Fig 7) over the ten-month incubation period and
exhibited no trend with time.
Cadmmm--The cadmium content in the MSW ash was 36 3 _+ 0 71 #g/g, which was sim-
ilar to expected concentrations The cadmium in the sediment controls averaged 1 13 + 0 09
~tg Cd/g and hence provided the greatest differential in contaminant concentration between
the S/S ash samples and sediment controls of any of the metals monitored The concrete con-
trol sediments contained an average 1 41 + 0 15/~g Cd/g and showed an increasing concen-
tratmn with time For four months (September, November, March, April), the Cd concentra-
tions in the concrete controls were significantly higher than those observed in the sediment
controls The S/S ash sediments averaged I 51 +_ 0 15 #g Cd/g and also exhibited an increasing
trend with time The Cd concentrations in the S/S ash sediments were significantly higher than
the sediment controls in five of the ten months (November, December, March, April, May).
However, the S/S ash results were never significantly higher than both the concrete and sedi-
ment controls in any month Note that all of these concentrations were near our analytical
detection hmlt for cadmium (1 03 + 0 71 tsg/g)
Since the S/S ash and concrete control Cd concentrations were both significantly higher
than the sediment controls, cadmium possibly was being leached from the cement Based on
studies of Type II portland cement by the National Bureau of Standards, 5the total amount of
cadmium in each of our cement-based specimens was probably < 2 0 0 fzg Whatever the cause
of the trends in the cadmium data, this metal warrants further scrutiny d u n n g the remaining
14 months of m o n l t o n n g Based on our results, the concentration and location of the cad-
mium in the concrete specimens need to be documented
5 R McKenzle, Office of Standard Reference Materials, National Bureau of Standards, Gmthersburg,

MD, 1989
o 3 o =
I o ~,~ .i.o .e
-== / o I.= u "~
lu~,"~!pom6/p 0 6n
lumwlpmm 5/~'~ 6n
o
!
o ~, ~
F-, =~
"SE~
io '= .,
i ~
z
O~
ii"I 1 : I I I I o I I I I I ~ : I I o ~
0 0 0 0 0 0 0 0 0 0 0
%uawlp0s6/qd6n )U~LUlpas 6/n:D 6n

r--,~
Conclusions
The mare effect of placement of the S/S ash and concrete control specimens in the organic-
rich m a n n e sediments was the increase in pore water p H and the creation o f a h~ghly reducing
environment This occurrence was probably the result of the leaching o f h m e from the speci-
mens in conjunction with the initial htghly porous nature of the concrete controls Initml
leaching o f h m e also apparently Increased alkalinity, and there appeared to be increased SO42-
incorporation in the concrete. However, these trends moderated after the first two months of
incubation The se&ments in the S/S ash and concrete controls remained h~ghly reducing
throughout the ten-month incubation p e n o d either because of increased microbial SO42-
reduction or abiotlc chemical reduction
The reducing environment in the sediments may be beneficial, because any metals released
from the ash specimens would probably be scavenged by metal sulfide precipitates, with the
exception of Cd, which may become solubdlzed by complexatlon with sulfide lIgands
At no time dunng the ten-month mcubatton were the metal concentrations m the S/S ash
se&ments significantly higher than those in both the concrete and sediment controls Two
scenarios may explain the results (1) no significant amounts of metals were released from the
S/S ash specimens into the surrounding se&ment, or (2) metals were released into the sedi-
ments, but in such small amounts that they were not detectable by our analytical techmques.
The results of the next 14 months may help clarify which scenario ts occumng and deter-
mine the long-term feasIbdlty of&sposing of S/S M S W ash in the manne environment.
Acknowledgments
This project was funded by the Office of Sea Grant, NOAA, U S. Department of Commerce
through Grant No NA89AA-D-SG020 to the University of New Hampshire/University of
Maine Sea Grant College Program We also acknowledge the help of Steve Truchon of the
U N H Coastal M a n n e Laboratory and Nancy Chenm of the U N H Electron Microscopy
Faclhty
References
[1] Goldstem, G , MechamcalEngmeenng, Vol 111, 1989, pp 44-50
[2] Hjelmar, O, Proceedings of the Internatwnal Workshop on Mumctpal Waste Incmeratwn, Mon-
treal, Canada, Environment Canada, Ottawa, Canada, 1987, pp 287-319
[3] Gough, M and Ujlhara, A, "Management of Mumclpal Waste Combustor Ash," Center for Risk
Management Resources for the Future, Washington, DC, 1989
[4] Elghmy, T , Kroner, N, and Ballestero, T , "Final Report--Co&sposal of Lamprey Regional Sohd
Waste Cooperative Incinerator Bottom Ash and Somersworth Wastewater Sludge," Lamprey
Regional Sohd Waste Cooperatwe, Durham, NH, 1988
[5] Thompson, P H, "A Feaslbdlty Study on the Use of Sodium Carbonate Amendments to Promote
the Chemical Stablhzatlon of Base and Heavy Metals Leached from a Mumclpal Sohd Waste Incin-
erator Combined Ash," master's thesis, Department ofCwd Engmeenng, Umverslty of New Hamp-
shire, Durham, NH, 1989
[6] Holland, P J, "Evaluation of Cement, Lime, and Asphalt Amended Mumc~pal Sohd Waste Incin-
erator Residues," master's thesis, Department ofClvd Engmeenng, Umvers~ty of New Hampshire,
Durham, NH, 1989
[ 7] Durhng, D B, "Evaluating the Feaslbdlty of Ocean Disposal of Stabfllzed/Soh&fied Mumclpal
Sohd Waste Incinerator Ash," master's thes~s, Department of Cwd Engmeenng, Umverslty of New
Hampshire, Durham, NH, 1989
[8] Tattersall, G , The Workabthty of Concrete, Viewpoint Pubhcatlons, New York, 1976
[9] Lessard, C E, "An Evaluation of the Degree of Metals ContammaUon m a Small Rwer," master's
thesis, Department OfClvd Engmeenng, University of New Hampshire, Durham, NH, 1987
[10] byAmencan
Copyright Pubhc
ASTM Int'l (all HealthMon
rights reserved); AssocmUon,
Feb 24 23:07:03Standard
EST 2020 Methods for the Exarnmatton of Water and Waste-
water, by16th ed, APHA, Washington, DC, 1985
Downloaded/printed
[11] Chormann, R, "The Occurrence of Arsenic in Soils and Stream Sediments, Town of Hudson, New
Hampshire," master's thesis, Umverstty of New Hampshire, Durham, NH, 1976
[12]Stookey, L L,AnalytlcalChemtstry, Vol 42, 1970, pp 779-781
[13]Tabatabax, M A, Sulfur lnstttute Journal, Vol 10, 1974, pp 11-13
[14]SAS Institute, Inc, SAS~ User's Grade Statistics, Verston 5, SAS Institute, Cary, NC, 1985
[15]Gilbert, R O, Statlsttcal Methods for Envtronmental Pollutton Monttormg, Van Nostrand Rein-
hold, New York, 1987
[16] Hines, M E, Knollmeyer, S L, and Tugel, J B, Llmnology and Oceanography, Vol 34, 1989, pp
578-590
[17] Hines, M E, "Seasonal Blogeochemxstrym the Sedxments of the Great Bay Estuanne Complex,
New Hampshire,'" Ph D dissertation, Department of Microbiology, Umverslty of New Hampshire,
Durham, NH, 1981
[18] Hines, M E, Lyons, W B, Armstrong, P B, Orem, W H, Spencer, M, Gaudette, H E, and Jones,
G E, MartneChemlstry, Vol 15, 1984, pp 173-187
[19] Elderfield, H, McCaffrey, R J, Luedtke, N, Bender, M, and Truesdale, V W, American Journal
ofSctence, Vol 281, 1981, pp 1021-1055
[20] Lyons, W B and F~tzgerald, W F, m Trace Metals m Seawater, C S Wong, E Boyle, K W Bru-
land, J D Burton, and E D Goldberg, Eds, Plenum Press, New York, 1983, pp 621-642
[21] Kolke, I and Hattore, A, EnvtronmentalMtcroblology, Vol 35, 1978, pp 278-282
[22] Roethel, F J, et al, "Phase II The Fixation of Incinerator Residues," prepared for the New York
State Legislative Commission on the Water Resources Needs of Long Island, Waste Management
Institute, State University of New York, Stony Brook, NY, 1987
[23] Shlvely, W E, "The Chemistry and Binding Mechamsms Involved with Leaching Tests of Heavy
Metals Sohd~fiedand Stabd~zedwith Portland Cement," master's thes~s, Department of C~vllEng~-
neenng, Umverslty of New Hampshire, Durham, NH, 1984
[24] Melchmger, K L, "An Investigation into the Seawater Leachabdlty of Cement Stabdlzed/Sohdlfied
Heavy Metal Sludges," master's thes~s, Department of Cwd Engineering, Umvers~tyof New Hamp-
shire, Durham, NH, 1987
[25] Bagchl, A and Sopclch, D, Journal of Environmental Engineering, Vol 115, 1989, pp 447-452
[26] Sholkowtz, E, Geochlrmca CosmochtmlcaActa, Vol 37, 1973, pp 2043-2073
[27] Peterson, M L and Carpenter, R, Geochtmtca CosmochlmcaActa, Vol 50, 1986, pp 353-369
[28] Armstrong, P B, Hanson, G M, and Gaudette, H E, Envtronmental Geology, Vol 1, 1976, pp
207-214
[29] Nelson, J I, "The Presence of Mercury, Chrommm, Lead, N~ckel, Copper and Zinc m the Great
Bay Estuanne System, New Hampshire," master's project paper, Department of C~vllEnglneenng,
Umverslty of New Hampshire, Durham, NH, 1986
Ivan L. Morgan I and William D. Bostick 2
Performance Testing of Grout-Based Waste

Forms for the Solidification of Anion Exchange
Resins
REFERENCE. Morgan, I L and Bostlck, W D , "Performance Testing of Grout-Based Waste
Forms for the Solidification of Anion Exchange Resins," Stabthzatlon and Sohdlficatlon of Haz-
ardous, Radioactive, and Mlxed Wastes, 2nd Volume, ASTM STP 1123, T M Gllham and C
C Wiles, Eds, Amencan Sooety for Testing and Matenals, Phdadelphm, 1992, pp 133-145
ABSTRACT: The sohdlficatlon of spent ion exchange resins m a grout matrix as a means of
disposing of spent orgamc resins produced m the nuclear fuel cycle has many advantages in terms
of process slmphoty and economy, but associated with the process is the potential for water/
cement/resins to interact and degrade the integrity of the soh&fied waste form Described in this
paper is one possible solution to preserving the integrity of these soll&fied waste forms the
encapsulatmn of beaded amon exchange resins m grout formulations containing ground gran-
ulated blast furnace slag, Type MI (mixed) portland cement, and ad&twes (clays, amorphous
slhca, slhca fume, and fly ash) The results of the study reported herein show that the cured waste
form tested has a low leach rate for nitrate ion from the resin [and a low leach rate is inferred for
Technetmm-99 (Tc-99)] and acceptable durablhty as assessed by the water Immersion and freez-
ing/thawing test protocols (but often not as assessed by a wetting/drying test protocol) The
results also suggest that a tested surrogate waste form prepared m wnyl ester styrene binder per-
forms satisfactorily against the wetting/drying cntenon, and ~t should offer addmonal insight
into future work on the soh&ficatlon of spent organic resins
KEY WORDS: cement waste forms, durablhty testing, ion exchange resins, solidification, vinyl
ester styrene waste forms, wetting/drying test
Organic-based ion exchange resins (IERs) are frequently used in the nuclear power industry
for r e m o v i n g radioactive impurities in aqueous process and waste streams They also are the
most i m p o r t a n t waste in terms o f " b e c q u e r e l s " generated in nuclear power plants--especially
pressurized water r e a c t o r s - - a n d they contribute a large annual waste v o l u m e for dis-
posal [ 1 ]
The direct Incineration o f spent organic resins would appear to be an attractive disposal
option (especially for v o l u m e and mass reduction) [2], but this process results in a higher dose
rate in the ash, and it imposes a higher purification efficiency for the exhaust gas for m i n i m i z -
Ing radioactive emissions [3] The scarcity of permitted radioactive waste (radwaste) inciner-
ators also severely limits the application o f this option. In practice, IERs are generally con&-
tioned into stable waste forms by incorporation in suitable inert matrices [e g., cement,
bitumen/asphalt, or (to a lesser extent, due to process economics) synthetic organic p o l y m e r
(including epoxy resins, polyesters, styrene polymer, polyethylene, polyvlnyl chloride, urea-
formaldehyde, phenol-formaldehyde, polyurethane, and polymethyl methacrylate)] [ 1,4]
J Senior engineenng assistant, Chemical Technology Dw~slon, Oak Ridge National Laboratory, Martin
Manetta Energy Systems, Oak Ridge, TN 37831
2 Head, Chemistry Department, Technical Division, Applied Technology Organization, Martin Mar-
letta Energy Systems, Oak Ridge, TN 37831
133

134 HAZARDOUSWASTES
Cementation is the simplest and most frequently used method and often the most adequate
method for stablhzlng (solidifying) spent IERs [5]. Cementation has numerous advantages in
terms of cost, convenience, and performance, but it is not without cntIclsm [6] Problems with
the concrete encapsulation of spent IERs are related to both the cement solidification process
itself(i.e, interactions between the hydraulic binder and water in the resins and chemical inter-
actions between the immobilized Ions and the ions in the cement paste) [6-8] and to the util-
ization of the process (mixing, process control, variable waste feed composition, etc ) [8]
Cementation process Interactions have, on occasion, resulted in poor mechanical proper-
ties, waste form destruction upon immersion in water, and altered setting rates. An undesir-
able water/cement/resin interaction is the potential for drying out (and contraction) of the
resin d u n n g excessively exothermic cement hydration reactions, with subsequent destructive
expansive swelling of the resins upon Immersion of the waste form in water [8] This problem
is more prevalent for cation resins and mixed bed resins solidified in an ordinary portland
cement (OPC) binder [9] and is enhanced by the continuous replacement of monovalent ions
on the cation resin by Ca +2 and A1+3 from the cement that may cause a further volume increase
in the resin [10]
Displacement from the spent resin of chemicals that alter cement set rate (e g., borates,
which retard set) and certain organic chelates (e g , citrate or picollnate) that can complex free
Ca +2 from the grout or enhance radionuchde mobility can also cause problems when cemen-
tation is the method used to stabilize/solidify spent organic resins
With the proper selection of a binder, adequate knowledge of the waste process chemistry,
and quality control guidance obtained through laboratory-scale testing and pilot scale confir-
mation, most undesirable water/cement/resin interactions can be eliminated or controlled
within acceptable limits. Beneficial admixtures can be applied in many cases (e g , soluble sil-
ica to accelerate set, a fluidizing agent to facilitate mixing, clays to control free water or impart
selective sorptlve properties, etc ), and binder blends can be selected that evolve less heat or
evolve heat more slowly [8] Binder blends in which a large proportion of the OPC is replaced
with ground granulated blast furnace slag (BFS) have been demonstrated to minimize the
potential for the formation of expansive phases [ 1, 7,8, I I ] and they allow relatively high vol-
umetnc incorporation of resin waste in the waste form [e g , as great as 50 to 75% (v/v)]
[1,8,11]
Purpose and Scope of the Study

For several years, spent organic resins have been stored at the U S Department of Energy's
Portsmouth Gaseous Diffusion Plant (PORTS) in Plketon, OH, pending the selection of an
acceptable option for disposing of this waste [12,13] The resin waste that needs to be disposed
of is beaded strongly basic anion exchange resin (Dowex-SBR) used at the Portsmouth plant
for removing the Technetium-99 (Tc-99) isotope (as the pertechnetate ion, TcO4) from neu-
tralized decontamination waste containing nitrate ion [14,15] Tc-99 undergoes radioactive
decay by low-energy (0 29 MeV) beta emission, with a half-life of 213 000 years Spent resin
from the PORTS process typically contains 0 1 to 1 0-mg Tc-99 per gram of wet resin (about
0 8 to 8 0 uCI/cm3); a small sample of spent resin from the PORTS process obtained for the
study had a surface gross beta activity of about 154 000 dislntegrations/mln/100 cm 2.
Our efforts early on in this study were Influenced by other work being reported on the solid-
lfiCatlon of spent organic resins Preliminary work had been reported by Dyer and McGlnnes
[16], who demonstrated that technetium-laden organic anion exchange resins could be suc-
cessfully encapsulated In binders containing 90 BFS 10 OPC and 50 BFS 50 OPC (w/w) and
in a polymer-modified cement The chief performance cntenon used in their investigation was
the loss of compressive strength after immersion in simulated seawater for one month, there
MORGAN AND BOSTICK ON GROUT-BASED WASTE 135
was little change noted in the structural integrity of the cement forms after immersion,
although small amounts of surface beads and cement flakes were found to be lost m the waste
forms containing 60% resin waste loadings
We also knew that grout containing BFS has a residual reduction potential [ 17] and has been
demonstrated to decrease significantly the leachablhty of Tc-99, presumably by its ablhty to
reduce Tc(VII) to the less soluble Tc(IV) form, with possible mmeralogacal fixation [14,18-
20]. Thus, we believed that a BFS grout might have an addmonal benefit for the fixation of
the Tc-laden resin on-site at the Portsmouth plant
The sohdlfled resin waste form properties of concern to us m our study are summanzed in
Table 1 [from Ref 6]. Specifically, we were Interested in the waste form's abihty to (1) form a
monolithic waste form with adequate compresswe strength for a hypothesized disposal sce-
nano, (2) have adequate stability toward radmtion, (3) resist blodegradatlon, (4) retard ra&o-
nuclide leachability, (5) demonstrate probable long-term durability as assessed by water
immersion and thermal cycling testing procedures, and (6) contain free water associated with
the resin
After a review of the literature, we determined that IER waste form durability would be a
key issue We therefore focused our testing on the durability testing of a surrogate waste
(Dowex SRB, converted to the nitrate form) solidified in binder blends containing various pro-
portions of BFS, OPC, ASTM Type F fly ash, amorphous silica, and clays (bentomte or atta-
pulglte). After a 43-day cure time, our samples were subjected to the American National Stan-
dards Instltute/Amencan Nuclear Society (ANSI/ANS)-16 1 leach test procedure [21] (to
determine the mass transport propemes for nitrate ~on), a 90-day (and beyond) immersion
test using distilled water, and tests to evaluate resistance to cycles of freezing/thawing and wet-
tlng/drymg procedures The latter two testing procedures were developed at the Alberta Envi-
ronmental Centre, Canada, and are outhned in draft testing protocols developed by ASTM
Task Group D34 02.07 on Physical Analytical Methods.
In ad&tlon to testing the cured waste form we had prepared for the study, we also decided
to test a waste form prepared by a commercial vendor by sohdffymg surrogate IER waste in a
vinyl ester styrene matnx. Sections prepared from this waste form were subjected to wetting/
TABLE 1--Sohdtfied product guMance
Tests Methods Cntena
Compresswe strength ASTM C 39 or D 1074 414 kPa (60 psi)

Radiation stablhty See 1983 TP a 414 kPa compressive strength after 10E + 8 rads
Biodegradation ASTM G 21 and G 22 No growth b and compressive strength >414 kPa
Leachablhty ANS 16 1 Leach index of >--6
Immersion See 1983 TP a 414 kPa compressive strength after 90 days
Thermal cychng ASTM B 553 414 kPa compressive strength after 30 cycles
Free liquid ANS 55 1 05%
Full-scale See 1983 TP a Homogeneous and correlates to lab size test results
a The 1983 Branch Technical Position (TP) paper calls for a minimum compressive strength of 345
kPa (50 PS0 This has been raised to 414 kPa (60 psi) to accommodate an increased maximum bunal
depth at Hanford of 55 ft (from 45 ft) as defined by the U S Nuclear Regulatory Commission publica-
tion, Draft Regulatory Gutde on Low-Level Waste Form Stabthty (a proposed revision to 10 CFR 61,
"RG LLW Form Stability," Rev 4, October 1986)
b The 1983 Branch TP calls for a multlstep procedure for bxodegradatlon testing If observed culture
growth rated "greater than 1" is observed following a repeated ASTM G 21 test, or any growth ~s
observed following a repeated ASTM G 22 test, longer term testing (for at least six months) is called for,
using the Bartha-Pramer Method From this test, a total weight loss extrapolated for full-size waste forms
to 300
Copyright yearsInt'l
by ASTM should produce
(all rights lessFeb
reserved); Mon than a 10%EST
24 23:07:03 loss of total carbon in the sample
2020
136 HAZARDOUSWASTES
drying testing and fire accident scenario testing by immersing the waste form m burning fuel
oil for an interval of 10 m m
Materials Tested
Virgin Dowex SRB-OH resin beads (the hydroxide form ofDowex- 1XS) were received from
the PORTS facility as a subsample of the matenal used in their Tc-removal process The as-
received drained resin had approximately a Gausslan particle size dmtnbution, with a mean
diameter of 610 + 180 #m (mean + 1 standard deviation) and a moisture content of about
54% (w/w) The drained matenal that was converted to the nitrate form had a mean diameter
of 540 + 220 tzm
The attapulglte- 150 dnlhng mud (a fibrous, adsorbent clay) was obtained from Mountain-
eer Mud and Chemical C o , Charleston, WV And bentonite, a gel-forming smectlte clay, was
obtained from Fisher Scientific C o , Fair Lawn, NJ (Product No B-235)
The ground granulated BFS was obtained from Standard Slag C o , Canfield, OH Portland
Type I-II cement (ASTM Specification for Portland Cement, C 150-85) was obtained from
Ash Grove West, Seattle, WA ASTM Class F fly ash was obtained from A m e n c a n Hy Ash
C o , Des Plalnes, IL The typical composition of these materials is given in the pubhcatlon
entitled "Blast Furnace Slag-Cement Blends for the Immobilization of Technetium-Contain-
ing Wastes" [18]
Amorphous silica Grade 1240 was obtained from Illinois Minerals C o , Cairo, IL This
material has a lognormal particle size distribution, with a median diameter of 6 4 um and a
spread factor of 2 9 [22] Silica fume mineral admixture, with pozzolanic and void-filling
properties said to produce a relatively dense and impermeable grout product, was obtained
from Master Builders, Cleveland, OH (Product No. MB-SF*)
Sodium silicate solution (43 ~ Baum6) was obtained from Fisher Scientific (Product No
SS338-1)
Methods Employed
Preparatton of a Surrogate Waste
The sample of spent resin that we received from PORTS was not sufficient for our use in an
extensive development effort Instead, what we needed was a larger quantity of a surrogate
waste with primarily the same characteristics as the PORTS sample and leach rate properties
no less conservative, for convenience, we elected to use a "cold" surrogate prepared using
sodium nitrate (NaNO3) solution, reasoning that nitrate ion would be more readily leached
from the solidified resin than pertechnetate ion (see below).
We stirred 1 L of 1 25 N s o d i u m nitrate solution and 500 g of moist Dowex SRB-OH resin
for 2 h to achieve an equillbnum distnbution of nitrate ion The resin slurry was transferred
to a Buchner funnel to remove excess nitrate solution Nitrate ion in the filtrate was analyzed
by ion chromatography to determine (by mass balance) the nitrate ion loading on the resin
(typically, 0 1-g mtrate/g wet resin) The "cold" surrogate waste was stored m capped Teflon
jars until ready for use
Preparatton of the Grout Samples

Predetermined amounts of the binder dry-blend ingredients (as listed in Table 2) were
mixed in a "V" (or twin-shell) blender for 4 h to ensure a homogeneous mixture After binder
homogenatlon, a known a m o u n t of dry-blend was placed in the mixing bowl (5-L capacity) of
MORGAN AND BOSTICK ON GROUT-BASEDWASTE 137
TABLE 2--Grout dry-blendformulations
Binder Blend Composltmn, Wt%
Blend Blend Blend Blend Blend Blend Blend

Component No 1 No 2 No 3 No 4 No 5 No 6 No 7
Blast furnace slag 75 69 70 75 69 84 5 90

Type I-II cement 10 10 10 10 10 95 10
Bentonite clay 15 15
Attapulglte clay 15 15 15
Amorphous silica 6 5
Class F fly ash 6
Sxhca fume
a standard Hobart| Model N-50 mixer Next, a measured mass of the surrogate waste "wet"
resin (containing about 50% moisture) and the reqmred amount of reagent water were added,
in that order The contents of the bowl were mixed at low speed (140 rpm) for 1 mm, then
mixmgwas switched to m e d m m speed (285 rpm) and continued for an addmonal minute The
mass ranos (w/w) for constituents added to the as-poured grout were as follows, dry-mix blend,
43 3%, reagent water, 39.4%, and drained resin, 17 3% On a volumetric basis, the drained
resin compnsed approximately 40% (v/v) of the waste form
After the m~xmg process was completed, the resulting waste loaded grout paste was trans-
ferred to either stainless steel [50 8 m m (2-1n)] cube molds (prewously coated with sflmone
lubrxcant to faclhtate removal of the cubes after set) or cyhnder molds fabncated from Teflon
for use m the ANSI/ANS- 16 1 procedure Grout samples were removed from their mold form
after three days and placed m a c u n n g chamber (with 100% relanve humidity at ambient tem-
perature) for about 45 days to complete the c u n n g process
Preparatton of the Vinyl Ester Styrene Waste Form

The vinyl ester styrene process for radwaste disposal was developed by Dow Chemical Co
and ~s currently marketed by Dwerslfied Technologies, Chestertown, MD, as the VERI| pro-
cess The waste form tested m this mvesnganon was prepared by the vendor using "cold" sur-
rogate waste IER (mtrate form) The volumetnc waste loading accomphshed m this process
approaches 100%, since the binder s~mply fills the lnterstmal voids between the resin beads
Testing Procedures
Displacement of Ions from the Resin Beads
In order to mlmmlze contaminant leachablhty, we knew it would be desirable that the con-
tammant(s) of interest remain sorbed on the IER and not be displaced into the grout matrix
pore water. Strongly basic amon exchange resin has a high afflmty for mtrate and pertechnetate
ions [14]. Pertechnetate m n is especially strongly adsorbed and can be eluted only by runs with
very high affinity for the resin, such as perchlorate and sahcylate [23]
We performed some prehmlnary investigations to determine the difficulty of displacing per-
technetate m n from spent Dowex SRB resin Two-mdhhtre volumes of packed resin were pre-
pared from slurried virgin resin loaded into 0 76-cm ID d~sposable chromatographic columns
[15], and 1 m L of a surrogate waste (contmnmg about 11 #g Tc-99 and 62 mg NO3)
[14,15] was added and the column nnsed with dlstdled water. The resin retained > 98 5% of
138 HAZARDOUSWASTES
the applied Tc-99 [equivalent loading, about 11 #g Tc-99/g resin (wet weight basis) or 25 ug
Tc-99/g resin (dry weight basis), with a surface beta activity of about 18 000 disintegrations/
m i n / 1 0 0 c m 2] F o u r b e d - v o l u m e a h q u o t s o f v a r l o u s s o l u t i o n s w e r e a p p h e d t o t h e c o l u m n ( f o l -
lowed by 1 bed volume of distilled water nnse), and the eluted solutions were analyzed by
liquid scintillation counting [ 14] to determine the amount of Tc-99 displaced by the solutions
The solutions tested were as follows 2 N perchlonc acid (HC104), 1 N HC104, 1 N NaC104, 1
N sulfuric acid (H2SO4), 1 N sodium citrate, and 2 N sodium hydroxide (NaOH)
We know that nitrate ion is displaced from the resin dunng service by traces ofpertechnetate
ion [ 15] and might be labile enough to undergo a small amount of ion exhange by the hydrox-
ide or soluble silicate ions in the waste form pore water In a batch test, 10 g of"cold" surrogate
waste resin (containing 1 03-g sorbed nitrate ion) were exposed to 100 mL of dilute sodium
silicate (Na2SIO3) solution (containing 0 36-g soluble silicate) to determine the difficulty of dis-
placing the nitrate ion from the resin
Leachabdlty Testmg Using the Cured Grout Specimens

The leachability of the nitrate in the solidified surrogate waste form was assessed using the
ANSI/ANS-16 1 procedure [21] at ten replacement leachate collection intervals (a 30-day
cumulative leach in deminerahzed water), This standard expresses the leaching data in terms
of basic mass transport theory (diffusion from a semi-Infinite medium), and then an "effective
diffusivlty" IS determined for the constituents of interest
D = ~ I- o] [ ]
(-2~).
2
~
2
(r) (1)
where
a n activity of a nuchde released from the sample during leaching interval n, corrected
for radioactive decay,
A o = total activity of a given radionuchde in the sample at the beginning of the leach test
(1 e , after the initial 30-s nnse),
(a~-)n = r . - r._t, duration of the nth leaching interval, s,
D= effective d l f f U S l V l t y , c m 2 / s ,
V= volume of sample, cm 3,
S= geometric surface area of the sample as calculated from measured dimensions, cm 2,
and
T= [0 5 (r ~5 + ~o51)]2, leaching time, representing the mean time of the leaching inter-
val, s
The negative logarithm of the average effective dlffUSivlty (in cm2/s) is the defined leachability
index
L, = -
,• [log (fl/D,)]. (2)
n l
where/3 is a defined constant (1 cm2/s) and D, is the effective diffUSlVlty of nuchde 1 calculated
from the test data Larger values for the leachability index we knew would be desirable, since
they represent smaller values for contaminant dlffUSlVlty (1 e , a lower value for contaminant
release rate).
Freezmg/Thawlng Testing and Wetting~Drying Testmg

Triplicate samples from each grout formulation were subjected to testing procedures
designed to evaluate waste form durability Specifically, the tests we performed were the freez-
ing/thawing and wetting/drying test procedures developed at the Alberta Environmental Cen-
tre and subsequently outhned in draft testing protocols developed by ASTM Task Group
D34 02 07.
The freezing/thawing test involved freezing the sample at -- 20~ for 24 h, storing the sam-
ple at room temperature for 1 h, and then ~mmersmg the sample m water chilled to 4~ The
sample was allowed to reach room temperature while being immersed in the water for a period
of 23 h After this immersion phase, the sample was removed from the water, any loose mate-
rial washed from the sample into the water, and the sample transferred to a clean tared beaker
for the start of the next cycle of testing The mass loss of the sample was determined by evap-
orating the water from the beaker after removing the waste form and weighing the material
left behind The sample was subjected to twelve cycles of this test
The wetting/drying test was s~milar to the freezing]thawing test procedure, except the sam-
ple was first d n e d at 60~ allowed to cool for 1 h, then covered wtth room-temperature water
for23 h
A waste form's failing either the freezing/thawing or wetting/drying tests was based upon a
criterion of a cumulative weight loss of greater than 20% (dry-weight basis) at any point in the
procedure
Immerston Testing
The cyhndncal cured grout specimens were immersed in disnlled water for a recommended
90 days (the time interval recommended in a 1983 Nuclear Regulatory Commission branch
technical posmon paper, see Table 1) and then for a longer interval to determine any visible
signs of degradation such as swelling or flaking After apprommately 200 days of immersion,
duplicate specxmens were subjected to compressive strength testing [ASTM Test Method for
Compressive Strength of Cylindrical Concrete Specimens (C 39)]
Fire Accident Scenarto Testmg

Due to the fact that the surrogate waste we studied contained a relatively high concentration
of nitrate ion (a strong oxidant and a potential accelerant for combustion), we performed lim-
ited testing of the waste forms by subjecting them to a fuel oil fire for a 10-min interval to
simulate a possible transportation accident scenario
The waste forms tested included a 50 8-mm (2-m) grout/resin cube (with a waste form den-
slty of 1 44 g/cm 3) and cyhndncal sections (approximately 4 6 cm m diameter by 3 4 cm m
height and with a waste form density of 0 89 g/cm 3) of solidified VERI| (vendor-sohdl-
fled samples of mixed-bed resins and SRB-NO3)

Displacement of lons from the Resin Beads
The results of our preliminary testing to determine the difficulty of displacing pertechnetate
ion from spent Dowex SRB resin showed these percents of Tc-99 displaced 2 N HC104 (53%),
1 NHC104 (26%), 1 NNaC104 (14%), 1 NH2SO4 (0 27%), 1 N s o d i u m citrate (0 13%), and 2
N N a O H (0 13%). Sodium silicate solution (0 3 N) displaced neghglble Tc-99 These data lndl-
140 HAZARDOUSWASTES
cate that sorbed pertechnetate Ion is very difficult to displace from the resin and that little ion
exchange with Tc should occur in the grout waste form pore water (pH about 13)
The results of our testing to determine the difficulty of displacing nitrate 1on from spent resin
showed 4 7E-3 mol of alkaline soluble silicate displaced only 3.7E-4 mol of nitrate ion (6 3%
of the nitrate applied to the resin), as determined by ion chromatographic analysis This result
suggests that nitrate ion would not readily be displaced from the resin in the grout waste form
pore water.
The presence of sorbed mtrate ion (a strong oxidant) on an amine-based anion exchange
resin may be subject to gradual exothermic gas-evolving decomposition reactions, this effect
has been noted for dried resin from strongly acidic medium [24,25] but should be mimmal
for resin produced under alkahne conditions Diluting the resin in the alkaline water-filled,
mlcroporous structure of grout should allow small amounts of gases and heat from chemical
and radlolysis reactions (if any) to dissipate safely without damaging overpressunzation of the
binder matrix [11 ]
Leachablhty Testing Usmg the Cured Grout Spectmens

The ANSI/ANS- 16 1 leachablhty index for nitrate 1on determined for each grout formula-
tion after its 43-day cure is given in Table 3 The leachability index was similar for all five grout
preparations (mean _+ standard deviation, 9 8 _+ 0 3) and greatly exceeded the guidance value
of a leachability index of 6 (see Table 1)
Similar testing for nitrate-containing sludge waste solidified in OPC/BFS/fly ash grout
yielded leachability indices of 6 to 7 [15,18], the greater leachability indices for nitrate ion
from ion-exchange resin solidified in grout suggests that nitrate ion is not readily displaced
from the resin beads within the grout matrix
Representative data for nitrate ion leaching from a cured grout-based IER waste form are
given in Fig. 1 Since a plot of the cumulative fraction of contaminant leached (a./Ao) versus
the square root of time (Eq 1) should be linear for a simple diffusion-limited process, we con-
cluded that the main transport phenomenon for nitrate ion is consistent with a dlffUsional
process
Due to the greater affinity of the anion exchange resin for TcO~ and the inherent low leach-
ability of Tc in BFS-based grout, we concluded the release rate for Tc-99 should be very small
Freezing~Thawing Testing
Weight loss data from the freezing/thawing test protocol are given in Table 3 for the grout
specimens All specimens passed the c n t e n o n ( < 2 0 % weight loss), with a m a x i m u m loss of
TABLE 3--Results of the leachablhty and freezmg/thawlng tests
Leachabdlty Index Cumulative Mass Loss Grout,

Blend Number~ for Nitrate Ionb (Freezing/Thawing Procedure)
1 9 7 4- 0 1 0 92% (w/w)
2 98 + 01 179%(w/w)
3 99 + 02 0 57%(w/w)
4 10 1 _+ 0 1 4 19%(w/w)
5 95 + 00 6 24% (w/w)
a Table 2 defines the dry-blend mix composmons used m the preparation of these grout blends
b These values are a mean + l standard deviation for three rephcate samples
Cumulative Fraction Released

0.07
f
0.06
0.05
0.04
0.03
0.02
0.01
I f
0
0 500 1000 1500 2000 2500 3000 3500
S q u a r e R o o t , E l a p s e d T i m e (s)
FIG. 1--The cumulativefraction of nitrate ion leached from grout blend No. 2 versus the square root of
time. (See also Table 2.)
only 6% for formulation No. 5. After the testing protocol, only minor surface cracks were
observed with no significant waste form deterioration.
Immersion Testing
The cylindrical cured grout specimens immersed in distilled water for the recommended 90
days (and then for a longer period of time) showed no visible signs of deterioration. After
approximately 200 days of immersion, the duplicate specimens subjected to compressive
strength testing (ASTM C 39) indicated the following columnar compressive strengths: blend
No. 1 (4560 and 3170 kPa), blend No. 2 (4150 and 2930 kPa), blend No. 3 (4530 and 2820
kPa), blend No. 4 (5520 and 3920 kPa), and blend No. 5 (6110 and 3770 kPa). These values
are similar for all specimens (average value 4080 kPa) and easily exceeded the 414-kPa (60-
psi) guidance value listed earlier in Table 1.
Wetting~Drying Testing
Data from the wetting/drying testing on the grout preparations are presented in Table 4.
Blend Nos. i and 2 containing bentonite failed the mass loss criterion (a cumulative weight
loss of less than 20%) within one or two testing cycles. Figure 2 illustrates the physical deteri-
oration o f a 50.8-mm (2-in.) cube of grout prepared from blend No. 1 after water immersion
in the first cycle of the wetting/drying procedure. Blends containing attapulgite performed bet-
ter in this testing, although blend No. 3~(with added amorphous silica) and blend No. 5 (with
added ASTM Class F fly ash) failed after several cycles (eleven cycles for blend No. 3 and
between cycles seven and eight for blend No. 5). The performance for blend No. 4 (and a repeat
pour, Blend 4R) was superior, with a cumulative weight loss of about 1.6% after twelve cycles
of testing. Blend No. 7, without added silica or clay, demonstrated an intermediate cumulative
weight loss.
TABLE 4--Results of the wetting~drying testing (waste form cumulative weight loss after each
cycle, %).a
Blend Number
Cycle
Number 1 2 3 4 4R b 5 6 7 VES
20.70 17.62 0.02 0.07 0.19 0.11 1.42 0.07 0.22

62.85 50.86 0.15 0.10 0.36 0.21 5.29 1.03 0.47
)5.1 l 68.34 0.71 0.12 0.52 0.39 7.31 1.57 0.64
0.86 0.12 0.66 1.62 10.47 2.03 0.70
o o 1.08 0.15 0.78 4.10 14.51 2.57 0.79
6 1.99 0.18 0.78 9.46 17.14 3.78 0.86
7 4.27 0.21 0.87 15.61 22.07 4.49 0.90
8 8.09 0.23 0.96 23.44 26.70 5.59 0.92
9 11.09 0.38 1.05 30.79 31.54 7.29 0.93
10 15.58 0.53 1.17 36.04 34.79 10.50 0.95
11 o . 20.06 1.01 1.40 41.71 40.75 11.87 0.98
12 24.14 1.56 1.66 45.41 46.62 14.65 1.07
These data are average values for three replicate samples. The defined failure criterion was a cumu-
lative weight loss of > 20% at any point in the testing procedure9
b This is a repeat pour using Blend No. 4.
FIG. 2 - - T h e physical deterioration of a 25.4 mm (2-in.) cube of grout prepared from blend No. 1, after
Copyright
waterbyimmersion
ASTM Int'l (all in
rights
thereserved); Mon Feb
first cycle of24wetting~drying
23:07:03 EST 2020 testing. (See also Table 2.)
TABLE 5--Results of the fire accident scenario testing (wasteform weight loss after lO-min fuel oil
fire).
Sample Starting Weight, g Final Weight, g Weight Loss, %
Grout/SBR-NO3 resin 189.07 161.38 14.6

VERI| resin 52.47 47.58 9.3
VERI| bed resin 47.88 15.10 68.3
The wetting/drying procedure was applied to a section prepared from the VERI~ resin sta-
bilized waste form; the cumulative weight loss recorded was 1.1% after twelve cycles of testing.
Fire Accident Scenario Testing

The results of our fire accident scenario testing are summarized in Table 5; most of the
observed weight loss was attributed to waste form dehydration, although some binder may
have melted and been lost from the VERI| mixed bed resin sample (see
Fig. 3).
FIG. 3--The appearance of cross-sectional disks of nitrated resin in vinyl ester sytrene matrix both
before and after subjection to afuel oilfire for an interval of lO min.
144 HAZARDOUSWASTES
FIG. 4--The appearance of a 25.4-mm (2-in.) cube of grout-stabilized nitrated resin after subjection to
a fuel oilfirefor an interval oflO min.
Conclusions
We demonstrated that beaded anion exchange resins can be incorporated with a moderate
waste loading (approximately 40%, by volume) into a grout matrix with good waste form integ-
rity, as judged by our freezing/thawing and immersion testing. (No specimens failed.) The wet-
ting/drying procedure we employed was also a more severe performance test for these waste
forms; only two formulations tested in this investigation passed the success criterion (namely,
a grout prepared from a dry binder containing 90% BFS and 10% Type I-II OPC and a grout
prepared from a dry binder blend containing 75% BFS, 10% Type MI OPC, and 15% attapul-
gite- 150 drilling mud). The improved performance toward the wetting/drying protocol dem-
onstrated by the grouts containing the attapulgite suggests that the fibrous morphology of this
additive may have imparted a physical advantage when the waste forms were subjected to the
swelling and shrinking of the matrices.
Limited testing using resin solidified in vinyl ester styrene suggests that the commercially
prepared surrogate waste form performs very satisfactorily against the wetting/drying test cri-
teflon. Important properties for resin embedded in this type of organic matrix included fire
resistance, leachability, radiation stability, compressive strength, and swelling behavior [5].
Vendor-supplied information suggests that the VERI| process performs well by these criteria
[26]. VERI| IER loaded with nitrate resin was subjected to exposure to a fuel oil fire
for 10 min; no significant acceleration of the fire was observed, although an occasional surface
bead was observed to flare. The waste form showed surface charring, but the bulk material did
not burn. A grout waste form subjected to the burn test remained intact but showed surface
cracks (see Fig. 4).
Acknowledgments
We would like to thank E. W. McDaniel of the Oak Ridge National Laboratory for his valu-
able input on the selection and behavior of clay additives in grout. We also want to thank R.
A. Bruce, H. E. Filter, and C. E. Jenson for demonstrating the VERI| process with our resin
sample. by
Downloaded/printed
This work is supported by the U S. Department of Energy under Contract DE-ACO5-

84OR21400
References
[1] Le Bescop, P, Boumol, P, and Jorda, M, Sctenttfic Basis for Nuclear Waste Management XIII, V
M Oversby and P W Brown, Eds, Materials Research Society, Pittsburgh, PA, 1990,13P 183-189
[2] Sandrelh, G , Terrain, SI, and Sangalh, G , Proceedings of the 1989 Joint International Waste Man-
agement Conference, American Society of Chemical Engineers, NY, 1989
[3] Otoda, T , et al, Waste Management "87, R G Post, Ed, Vol 3, Umvers~tyof Arizona, College of
Eng'.neerlngand Mines, Tucson, AZ, 1987, 1313 117-122
[4] Ghattas, N K, Eskander, S B, and Ikladlous, N E, Radioactive Waste Management and the
Nuclear Fud Cycle, Vol 7, No 1, 1986, 13 13
[5] Patek, P R M, et al, Radioactive Waste Management, Vol 2, International Atomic Energy
Agency, Vienna, 1984, P13 171-187
[6] Workshop on Cement Stabdlzatlon of Low-Level Radtoacttve Waste, Report NUREG/CP-0103,
U S Nuclear Regulatory Commission, Washington, DC, 1989
[ 7] Duquesne, J and Jaouen, C, Proceedings of the Internattonal Conference on Nuclear Fuel Repro-
cessmg and Waste Management (RECOD '87), August 1987
[8] Joauen, C and Vlgreus, B, Proceedmgs of Nuclear and Hazardous Waste Management (Spectrum
'88), American Nuclear Society, La Grange Park, IL, 1988, 1313 105-108
[9] Arnold, G, et al, Sohdtficatton orlon-Exchange Resm Wastes, Report BNL-51615, Brookhaven
National Laboratory, Upton, NY, 1982
[10] Soo, P, Mlhan, L W, and Plclulo, P L, The Leachablhty andMechamcal Integrity of Stmulated
Deeontammatton Resin Wastes Solidified In Cement and Vinyl Ester-Styrene, Report NUREG/CR-
5153, U S Nuclear Regulatory Commission, Washington, DC, 1988
[11] Lee, D J, Gntfiths, C J, and Fenton, A, "The Development of Cement Formulations for Radlo-
actwe Waste Encapsulation," th~s13ubllcat~on
[12] Acox, T A, Proceedmgs of the Fourth DOE Environmental Protection Information Meetmgs
(CONF-821215-DE83017117), August 1983, 131377-85
[13] Saraceno, A J, "The Control of Technetmm at the Portsmouth Gaseous D~ffus~onPlant," Report
GAT-2010, Goodyear Atomic Corporauon, Plketon, OH, Nov 1981
[14] Bostlck, W D and Evans-Brown, B S, "Sorptwe Removal of Technetmm from Heavy Metals
Sludge Filtrate Containing Nitrate Ion," Report K/QT-160, Martin Marietta Energy Systems, Inc,
Oak Ridge, TN, 1988
[15] Bost~ck,W D, et al, Emerging Technologws m Hazardous Waste Management, D W Tedder and
F G Pohland, Eds, American Chemical Society, Washington, DC, 1990, P13 345-367
[16] Dyer, A andMcGmnes, D F,"OrganlcReslnAmonExchangersfortheTreatmentofRadloactwe
Wastes," Report No DOE/RW/88-106, Department of the Enwronment, London, U K 1988
[17] Rahman, A A and Glasser, F P, "Cements in Radloactwe Waste Management Charactenzatmn
Requirements of Cement Products for Acceptance and Quahty Assurance Purposes," Report EUR-
1083-EN, Commission of the European Communmes, Luxembourg, Belgium, 1987
[18] Bostlck, W D, et al, "Blast Furnace Slag-Cement Blends for the Immobdlzatlon of Technetmm-
Containing Wastes," Report K/QT-203, Martin Marietta Energy Systems, Inc, OaK Ridge, TN,
1988
[19] Tallent, O K, et al, Proceedings of the Materlals Research SocwtySymposlum, Vol 112, Materials
Research Society, Pittsburgh, PA, 1988, p 23
[20] Brodda, B G , Scwnce of the TotalEnvlronment, Vol 69, No 69, 1988
[21] "Measurement of the Leachabd~ty of Sohdlfied Low-Level Radmactwe Wastes by a Short-Term
Test Procedure," ANSI/ANS-16 1, American Nuclear Society, La Grange Park, IL, 1986
[22] MacKay, D and Paterson, S, EnvlronmentalScwnce Technology, Vol 18, 1984, 13 207A
[23] Anders, E, The Radlochemlstry of Technetium, Monograph NAS-NS 3021, National Academy of
Sciences, Washington, DC, Nov 1960
[24] Van Slyke, W J, "Final Report of Chemical Safety for Technetmm-99 Recovery by Amon
Exchange," Report BNWL-476, Battelle Pacific Northwest Laboratory, R~chland,WA, 1967
[25] Cleveland, J M, The Chemtstry of Plutomum, Gordon and Breach ScientificPubhshers, New York,
1970,19 157
[26] Fdter, Harold, et al, "The Dow Waste Sohdlficatlon Process for Low-Level Radmactlve Waste,"
U S NRC To131calReport DNS-RSS-200-P, U S Nuclear Regulatory Commission, Washington,
DC, 1990
Laboratory-Scale Leachability Studies
Gerard J. de Groot 1 and Hans A. van der Sloof
Determination of Leaching Characteristics of

Waste Materials Leading to Environmental
Product Certification
REFERENCE: Groot, G J de and Sloot, H A van der, "Determination of Leaching Charac-
teristics of Waste Materials Leading to Environmental Product Certification," Stablhzatlon and
Sohdtficatzon of Hazardous, Radtoactlve, and Mixed Wastes, 2nd Volume, STP 1123, T M
Gdham and C C Wiles, Eds, American Society for Testing and Materials, Phlladelphm, 1992,
pp 149-170
ABSTRACT: Quantifying the environmental impact of potentially hazardous materials In a

wide variety of applications and disposal scenarios is essential to proper management of these
materials In an environmental assessment, the release of contaminants from the waste body
with time has to be determined rather than aiming at concentration limits as currently used In
this approach, test results should yield information on leach rates This implies the use of test
methods other than those currently applied in the regulatory framework (EP, TCLP, DIN 38414-
S4)
The tank leaching procedure proved to be a proper method for charactenzlng the leaching
behavior of trace contaminants from construction materials and stabilized waste products Dis-
tlnCtlOn between different leaching mechanisms is possible dissolution, surface washoff, and
matrix diffusion Leach parameters, whlLh allow estxmatlng release at much longer time scales
than the duration of the actual experiment, can be obtained from the tank leaching test These
parameters [avallabdlty for leaching, physical retardation (tortuosIty), and chemical retention]
offer control over the environmental quahty of materials A calculation method, different from
the methods that have been published so far m the hterature, is proposed for determlnatmg the
leaching mechamsm rn the material
The paper shows that the traditionally used one-dlmensmnal model cannot predict long-term
release w~th sufficient accuracy m actual apphcatlons because it does not take into account the
limited leachable amount and the dlmensmns of the material Since a one-dimensional model
cannot describe the release with time as a function of different material shapes, even when deple-
tmn of the source is accounted for, at Netherlands Energy Research Foundation (ECN) a three-
dlmensmnal model is currently being developed that does offer these capablhtles Some prehm-
mary results are presented The vanatmn found m the test results within one category of mate-
rials is relatively small and indicates that possibilities for developing short tests exist For certi-
fication activities and quality control, this would be useful in creating a framework for
environmental control parallel to technical product quality
At ECN a database for leach data has been developed Storing detailed leach data on different
construction materials and stablhzed waste products in a database offers opportumtles for a more
effioent use of costly leaching studies The knowledge gamed by gathering mformatmn on a wide
variety of materials may prove particularly useful in estabhshlng systematic trends in leaching
and selecting new dlrecUons for reusing waste materials Also, a database will be a very useful
tool m the environmental certification process for construction materials containing waste mate-
rials It can be concluded that test methods tailored to a utihzatlon optmn offer better possibilities
than a single shake test for making proper chmces in the wide variety of materials m an even
wider range of possible applicatmns
i Project manager and senior sclenUst, respectively, Netherlands Energy Research Foundation, Petten,
The Netherlands
149

KEY WORDS. hazardous elements, waste materials, construction matenals, leaching mecha-
rosins, tank leach test, diffusion, diffusion models, certification, database
In recent years, a strong emphasis has been placed on the beneficial use of bulk waste mate-
rials Siting problems for new disposal faclhties has been a dnvlng force In addition, the dis-
covery of beneficial properties of waste materials that are being disposed of today certainly
emphasizes reuse The long-term environmental impacts resulting from using waste materials
in construction is the primary concern
Regulations are needed to avoid soil and groundwater pollution through reusing waste
materials In the Netherlands, the Soil Protection Act was passed In 1988, which will contain
regulations for using waste materials in construction (Bouwstoffenbeslult) and regulations for
disposal (Stortbeslmt) The regulatory framework is being developed now, and the first draft
of the Regulation for Utilization of (Waste) Matenals in Construction has been sent out for
comments Since soil protection is a goal, the release of contaminants to the soil system over
time is the crucial factor to control This means test methods have to address this problem,
which implies the use of other test methods than currently applied in the regulatory framework
(extraction procedure, EP [ ! ], toxicity characteristic leaching procedure, TCLP [2], Deutsches
Institut fur Normung (DIN) 38414-$4, German Standard Method for Water, Wastewater, and
Sludge Investigation [3], since these methods are aimed at controlling leachate concentration.
Within the framework of the new regulations, an environmental assessment of a wide van-
ety of waste materials in a wide variety of potential applications is addressed The flexibility in
testing procedures necessary to manage this broad field requires a systematic approach, which
is based on a fundamental understanding of the mechanisms controlling the environmental
impact in a given situation A detailed evaluation of a wide variety of materials has shown
systematic behavior of groups of materials In terms of leaching properties [4] Materials need
to be classified based on technical properties, leaching characteristics, and potential applica-
tions, which would allow standardlzaton of sampling, analytical, and leaching procedures for
specific waste materials
A major distinction is the classification of powder/granular materials which are dominated
by a percolation regime, and waste products and monoliths (4~ > 40 ram), which are domi-
nated by slow (diffusion-controlled) release After a short overview of the merits of the most
important leaching test, currently applied, this paper will focus on intact products, for which
the leaching is mainly governed by diffusion-controlled release Several aspects regarding tank
leaching test procedures and processing of experimental results, to describe the systematlcs in
leaching behavior, will be discussed
Currently, a one-dimensional diffusion model, assuming constant source, IS used to assess
the leaching parameters from the results of the tank leaching experiment (similar to American
Nuclear Society, ANS 16.1 [5] In this paper, the relation between the logarithm of the cumu-
lative release (in m g / m 2) and the logarithm of leaching time is used to identify the controlling
leaching mechanisms such as initial wash-off, diffusion, and dissolution For diffusion-con-
trolled release, estimates of the chemical and physical retardation of product components are
given, and leaching test results of few materials will be discussed Estimates of contaminant
release from specimens of different spatial dimensions and a limited amount of leachable com-
ponents are possible for different exposure times once the release mechanism is estabhshed
Using a constant source diffusion model to predict component release in practice is limited,
and the model does not take depletion into account, causing inaccurate estimates of long-term
leaching Depletion can be implemented in a one-dimensional model However, one-dimen-
sional models cannot account for the influence of the different material shapes (for example,
thinbylayers
Copyright versus
ASTM Int'l (all rightscubes)
reserved); on
Monthe release
Feb 24 as a2020
23:07:03 EST function of time. A three-dimensional model can
DE GROOT AND VAN DER SLOOT ON LEACHING 151
Some preliminary results of a simple three-dimensional model, currently under investigation,

to explore the effect of different shapes of waste products on release rates, will be discussed
For better quality control, certifying construction materials and stabilized waste materials
is needed A database with extended leaching test results plays an important role In establishing
the systematic leaching characteristics of materials on which short verification tests have to be
based The routing of the certification process will be briefly outlined
Test Methods
Comparison of Test Methods
To compare the various leaching tests for waste materials, it is important to realize what the
tests are meant to assess The regulatory EP [1], TCLP [2] and DIN 38414-$4 [3] are designed
as screening tests to classify materials as being environmentally acceptable or not These tests
have been applied inappropriately by using them for environmental assessment purposes
Relating the results of these tests to the wide variety of materials in an even wider range of
possible field situations is impossible In Table 1 a comparison of regulatory test methods is
given In the Netherlands, more extensive procedures have been applied for several years now
to judge the environmental impact of waste materials In these procedures a column test at
low liquid/solid ratios (L/S) and a senal batch test at medium and high L / S are applied to
granular matenals (largest size < 3 ram) [6] Some results obtained by these tests are reported
m Refs 7 and 8.The tank leaching test procedure for waste products and monoliths (size > 40
mm) will be described m more detail in this paper.
The amount of information gained from the more extensive procedures applied in the Neth-
erlands leads to a better understanding of mechanisms and factors controlling leachability,
thus resulting in better management tools for control over release of undesired contaminants.
TABLE l--Comparison of leaching tests
Column + Senal Batch [6], Granular Tank Leaching Test [7],

EP/TCLP/S4 [3-5] Materials Products >40 mm
Concentration control Release control Release control

pH control with a Material dictates its own chemical environ-
maximum acid ment
consumption No external pH control
No predictions to At low L/S ratio a close correlation with Mechanistic information
environmental field leachate composition (short Surface dissolution
conditions possible term) Washoff
Diffusion
Time dependent release through the L~
S ratio a and infiltration rate
Mechanisms washout, solubility Parameters obtained
hmltation, long-term release Physical retardation
Chemical retention
Chemical fixation
Relatively cheap per Imtlally more costly, but systematic trends
sample allow a substantial reduction in critical
components and conditions to be tested
Frequent testing Development of short tests tailored to a
required problem results in a more flexible
control system
a L/S
Copyright = liquid-to-solid
by ASTM ratio
152 HAZARDOUSWASTES
Intrmstc Leachmg Parameters

The leaching behavior of construction materials can be described by the following param-
eters. These are
1 The fraction of each element tied up in silicate and poorly soluble mineral phases, which
is assumed to be environmentally inert (availability test [6])
2 The physical retardation (tortuoslty), which is derived from measunng the release of an
inert component from the product matrix (often sodium can be used for this purpose).
3 The chemical retention, which is derived from the measured release of a given compo-
nent, its free mobility in water, and the tortuoslty as obtained for the inert component
(for example, sodium)
The latter two factors are calculated from data obtained in a tank leaching experiment
[4,9,10] The availability test quantifies the fraction of an element present in the material
that is not tied up in low soluble mineral phases and can potentially be released to the envi-
ronment
Avagablhty Test--The avadablhty for leaching is measured on fine-grained material (95%
< 125 ~m) The procedure consists of extracting the crushed and ground specimen at a liquid-
to-sohd ratio (L/S) of 100 (L/kg) at pH = 4 (controlled) In this procedure, low-soluble min-
eral phases, silicates and the hke, are not dissolved The soluble fraction reflects the quantity
that can eventually be leached on the very long term after complete disintegration of the mate-
rial In addmon, the test yields information on the acld-neutrahzlng capacity of the material
To relate this test result to a time scale, two factors are important the rate of material neu-
trahzatlon by acid rain and the material's durabihty (carbonation, abrasion, hardness)
Tank Leaching Test--The leaching behavior of construction materials is characterized best
by a tank leaching test (Draft NVN 5432 [9] that has similarities to ANS 16 1 [5] In short,
the test proceeds as follows the specimen to be tested is submersed in a tank containing five
times its water displacement, ensuring permanent contact with water The contact solution IS
renewed at regular intervals, and the extracts are analyzed among others by Induced coupled
plasma emission spectrometry (ICP) and atomic absorption spectrometry (AAS) for major ele-
ments and trace constituents of concern The information gained from this test can be used to
identify the major release mechanism(s) and leaching parameters such as physical retardation
and chemical retention
Leachmg Mechamsms
By plotting the release as a function of contact time, different leaching mechanisms can be
identified Figure 1 gives a schematic representation In some cases, a combination of two
mechanisms can be Isolated The following mechanisms can be distinguished" dissolution, sur-
face washoff, and matrix diffusion
Dtssolutlon--The solubility of the product matrix can be such that dissolution of material
from the surface proceeds faster than diffusion through the pores of the matrix This phenom-
enon has been observed in products containing very high gypsum loadings (ECN, unpublished
results) The calcium sulfate (CaSO,) solubility is relatively high, which results in a theoretical
slope + 1 in the release-time plot (log-log) At longer time intervals the slope decreases as the
contact solution gets saturated with respect to CaSO4. Figure 2 illustrates an actual example of
this release mechanism (stabilized phosphoric acid gypsum, with 15~ coal fly ash and 3%
lime)
RELEASE CURVE
"E 1OO, 9 +1 DISSOLUTION

/
/
/
E /
/
/
/
r
_e 10 / +o.5 DIFFUSION
/ s'*
r f s JS
.m
as s S
,~s s
0 WASH-OFF
E 1
O
0.3
10 100
Time [days]
TANK LEACHING EXPERIMENT
> DISSOLUTION
DIFFUSION
WA S N - O F F
I WATER PHASE ]
FIG. 1--Release mechanismsfrom intactproducts.
Surface Washoff--Due to process conditions or condensation processes a material can be

covered with a relatively soluble surface coating. In the initial phase of the leaching experi-
ment, this layer will dissolve. In many cases, the subsequent release is diffusion controlled.
This type of release is c o m m o n for slag-type materials. Figure 3 shows data of a traditional
concrete block as an example of this release mechanism and the resolution of the washoffand
diffusion component of the overall release [4].
Diffusion Control--In most cement-based materials the leaching process is diffusion con-
trolled. Figure 4 gives an example of strictly diffusion-controlled release as well (asphalt con-
crete with municipal solid waste-fly ash) [4]. Intrinsic leach parameters, as described before
and further discussed in the next paragraph, can be derived, allowing predictions about release
at times considerably longer than the experiment's duration.
154 HAZARDOUSWASTES
10 a
E
10 2
Slope = 0.8
i_
1O'
-5
E
O
10 0
0.2 1 10 100
Time [days]
FIG.2--Surface dissolution of calcium from a stabilized phosphoric acid gypsum containing 15% coal
fly ash and 3% lime.
10 ~
ga
+
t=
Slope = 0.02
E
10'
t~
9 wash-off: diffusion of
component (slope = 0.46)
100
-5 I
E
0
1 0 ~ . . . . . . . i . . . . . . . . i . . . . . . . . i
0.2 1 10 IO0
Time [days]
FIG. 3--1nitial surface washoff of barium from traditional concrete: plus-sign = measured cumulative
release; solid-dot = calculated cumulative release according to the method described in section entitled
"'Diffusion
Copyright by ASTMModel, Data
Int'l (all rights Handling,
reserved); Mon Feband Evaluation,
24 23:07:03 EST 2020 ""Eq 5.
F
E
E
10'
|
k_ Slope 0 49
-5
E 1Q ~
O
0.2 1 10 ~O0
Time [days]
FIG. 4--Diffusion-controlled release offluorine from asphalt concrete with 2% municipal solid waste
fly ash.
The ion flux through the geometrical surface area of the product is described by Fick's sec-
ond law
6C/6t = L 62C/6x2
where L is the overall leach constant with dimensions of a diffusion coefficient.

In the case of one-dimensional diffusion, the property L can be obtained from
L = f z . Do
R'r
where
L = the overall leach constant in m2/s,
f = availability factor defined as the ratio between the leachable concentration and the
total concentration for each element in the material ( - ) ,
Do = the free mobility of the component in water (m2/s),
R = the chemical retention factor ( - ) , and
r = the physical retardation or tortuosity factor ( - ) .
Because the driving force for the diffusion is determined by the leachable and not the total
a m o u n t of each element in the material, L equals the effective diffusion coefficient f o r f = 1.
By transforming these parameters to log values by means o f p D = --log D, a new relation
is obtained showing the individual contributions to the release of components from construc-
tion materials and stabilized waste materials
Copyright by ASTM Int'l (all rights reserved); Mon Feb p

24L23:07:03
-- pDoEST 2020
+ 2pf - - pR -- pr
156 HAZARDOUSWASTES
After correction for the leachable fraction, f an effective diffusion coefficient, De, is obtained
that will be used In subsequent descriptions of release from waste forms
p D e = pDo -- p R -- p r
The larger the pDe, the slower the release from the matenal. Measures to improve the envi-
ronmental quahty of products containing secondary materials can be aimed at any of the three
factorsf R, and r Changes in technical production parameters generally can lead to a denser
product and consequently influence the physical retardation factor However, within one cat-
egory of materials, the range in tortuoslty is relatively small This imphes that changes in chem-
ical properties, which influence both R a n d f have more effect on ultimate release rates
Diffusion Model, Data Handling, and Evaluation

The diffusion model used for interpreting the tank leaching results can be derived from the
equation presented by Crank [11] for the diffusion from a product with semi-infinite dimen-
sions in which the initial concentration is uniformly distributed in the product and the con-
centratlon on the surface between the product and the leachate is constant in time
C - C, _ e r f x
C o - C~ 2 (V~ t) (1)
where
C = C ( x , t ) concentration as function of place and time,
C~ = constant concentration at x = 0 (surface),
Co = initial concentration at t = 0 in the product (uniformly distributed),
D = diffusion coefficient,
t = time,
x = distance to surface (positive values), and
err = standard error function.
In a tank leaching, the surface concentration will only be constant as long as no material
depletion occurs, and the mean concentration In the solution does not deviate slgmficantly
from zero These requirements are met by using a product for which the smallest dimension
is greater than 5 cm, preventing in most cases depletion within the time frame of the experi-
ment (64 days), and by refreshing the leachate at regular time intervals
Assuming that the surface concentration Cl is equal to zero at all times seems reasonable
when the solution is well mixed and regularly refreshed during the leaching experiment When
the solution is not perfectly mixed, as in our tank leaching experiments, the concentration near
the product surface can be significantly h~gher than zero Since diffusion In the leachate ~s at
least as fast as diffusion in the product, the upper limit of CI will be 89 Co Prehmlnary labo-
ratory experiments using radio tracers indeed indicate a value for C1 larger than zero, even
close to ~A Co The resulting diffusion equatmns from Eq 1 for the boundary conditions Cl --
0 and C~ = 89 Co will be
For C~ = O"
7r- Bt2
De =EST
Copyright by ASTM Int'l (all rights reserved); Mon Feb 24 23:07:03 4 2020t (Urn.x- d) 2 (2a)
For C1 = ~A- CO
De = (2b)
t (Um~ d) 2
where
De = the effective diffusion coefficient for component x in the product, in m2/s,
B, = the cumulative release of the component in mg/m 2,
t = the contact time m seconds,
Um~, -- the m a x i m u m leachable quantity in mg/kg (-- avallabihtyftlmes total concentra-
tion In product),
d = the bulk density of the product, in kg/m 3
The difference between Eq 2a and 2b is a factor of 4 The effective diffusion coefficient De
calculated for the boundary conditions C1 -- 0 is four times lower than calculated for the
boundary condition Ca = xA Co
For calculating releases from De it is of no consequence as long as the same model is used
However, i f a diffusion coefficient is related to the free mobility in water for the estimation of
the physical retardation (tortuosity), a difference of a factor of four wdl also be found for the
two boundary conditions Because of the prehmlnary results in our already mentioned ra&o
tracer experiments, in our work Eq 2b is used rather than the more commonly applied equa-
tion 2a [12,13]. This topic is currently under investigation more extensively
Determlnatton of Leachmg Mechamsm from the Cumulative Release Pattern versus Ttme
To determine the leaching mechanism the logarithm of the cumulative release has been
plotted versus the logarithm of time. Rearranging equation 2b yields
8,=U.=.d V5 (3)
and after transformation to logarithm
log(Bt)=~A log (t) + log [Um~x d . ~ ]
From the tank leaching test results the release of each component per time interval can be
calculated with the formula
c, v,
B,- 1000----~ (4)
where
B, -- the release m period l, expressed In m g / m 2,
c, = the concentration of the component in the t-th penod, expressed in #g/L,
V, = the volume of the contact solution, In htres, and
A = the geometrical surface area of the specimen in m 2.
For the measured cumulative release, the measured release from previous penods is summed
This implies that deviations in a given period accumulate In the subsequent periods, which
may hamper interpretation. There are two methods currently used to avoid this problem,
which was published by Duedall [13] and C6t6 [12]. The approximations they use have a dis-
advantage in that they deviate in the early leach fractions from the onglnal one-dimensional
diffusion model We apply a different m e t h o d by calculating the cumulative release until the
l-th period only from the release in the l-th p e n o d assuming dlffHSlOn control in the l-th a n d
previous time periods These values can be used to check whether the release is diffusion con-
trolled The measured a n d by our m e t h o d calculated cumulative releases are equal when the
release is indeed controlled by diffusion
The calculated cumulative release we use for all N periods can be d e n v e d from
B,,, = B~ ~for/ -- 1 t o N (5)

Vt, l
where
Bt,, = the cumulative release of the c o m p o n e n t until period l, in m g / m 2,
t, = the contact time after the p e n o d i in seconds, a n d
t,_ ~ = the contact time after the p e n o d ~ -- 1 in seconds
After plotting the logarithm of the calculated cumulative release (B,,,) against the logarithm
of the time (t,) for t = 1 to N, the slope of the relation for all points a n d for three time intervals
of three fractions each can be determined
0 Total range Fractions 1 to 8

1 Initial leach range e g , Fractions 1 to 3
2 Intermediate range e g , Fractions 3 to 6
3 Last range e g , Fractions 5 to 8
F r o m the slope of the graph the m e c h a n i s m of leaching can be denved, components dis-
solving from the surface (slope > ~ 0 8), short initial release of surface-deposited c o m p o n e n t s
(initial stage slope < ~ 0 4), and diffusion-controlled release (slope ~ 0 5) The m e a n i n g of the
change in the slopes at different time intervals is s u m m a n z e d in Table 2 Only in the case where
the slope in the intermediate a n d last range is about 0 5 can an effective diffusion coefficient
be calculated (De) W h e n the slope m the entire plot is 0 5 _+ 0 1 and the slope m the last range
(3) is smaller than 0 6, all fractions can be used for the calculation of the effective diffusion
coefficient If this is not the case, only those subsequent fractions can be used for De calculation
in which the slope is 0 5 + 0 1
The effective diffusion coefficient (De) for the c o m p o n e n t of interest is then calculated from
each p e n o d for the release per p e n o d (B,) using only those data points for which the slope is
0 5 + 0 1 with a deviation of less than 50% by
D ..... = (Um~, d) 2 (~ - tx/~,_~) (6)
where
D ..... = the effective diffusion coefficient of c o m p o n e n t x calculated from the release in
p e n o d l, in m2/s,
B, = the release in p e n o d i m m g / m 2,
d = the bulk density of the product, m kg/m 3,
Um~x = the m a x i m u m leachable q u a n h t y in mg/kg,
t, = the contact-time until p e n o d t m s, a n d
Copyrightt,by 1 = the
ASTM contact-time
Int'l (all until
rights reserved); Mon period
Feb 24 l --2020
23:07:03 EST 1 in s.
TABLE 2--Meanmg of slope changes at dzfferent ttme mtervals
Slope
Range <0 4 0 4-0 6 >0 6
Begin Surface washoff Diffusion Lag tlme/dlssolunon

Middle Depletion D~ffus,on Dissolution
End Deplenon D~ffusmn D~ssolutIon
After expressing the effective diffusion coefficient in the negative logarithm pDe,x -- - l o g
(De,,,x), the average effective diffusion coefficient and its standard deviation for c o m p o n e n t x
can be calculated
Comparison of the Moblhty of a Specimen and Its Free Moblhty in Water

The general equation for the effective diffusion coefficient as a function of the physical retar-
dation (tortuoslty) and chemical interaction is
1
De = D0~ R T
where
De = effective diffusion coefficient of c o m p o n e n t x in the product, in m2/s,
Do,x = diffusion coefficient of c o m p o n e n t x in water, in m2/s,
R = chemical retention factor o f c o m p o n e n t x in the product ( - ) , and
T = physical retardation in the product ( - )
For the calculation o f the physical retardation (tortuoslty), an ion should be chosen which
does not interact with the m a t n x (R = 1) In m a n y cases, sodium is an a p p r o p n a t e choice
The tortuoslty of the product can be calculated with the formula
DNa -/- __
Ee,Na
where
T = the tortuosity o f the product ( - ) ,
DN. = the diffusion coefficient o f sodium m water (pDNa = 8 91 at 22~ m m2/s, and
DeN. = the effective diffusion coefficient o f sodium in the product in m2/s.
The chemical retardation factor (R) for the c o m p o n e n t of interest is then calculated from
DX
R-
De~ T
where
R = the chemical retardation factor (--),
Dx = the diffusion coefficient for c o m p o n e n t x in water, in m2/s,
De,x = the effective diffusion coefficient for c o m p o n e n t x in the product, in m2/s, and
r ASTM
Copyright by = the
Int'l tortuoslty of Mon
(all rights reserved); theFeb
product
24 23:07:03( EST
- ) 2020
Results
Relevance o f the A vmlabdtty Test
The s]gmficance of the avaflablhty test for judging the maximum leachability of a given
component from waste materials ~s best illustrated by the tank leaching test results obtained
for a coarse granular material containing coal fly ash The leach data obtained m a tank leach-
mg experiment are expressed as release data (mg element leached per m 2 surface area) against
time The avadabdlty test results are recalculated to the same umts by taking the size and shape
of the specimen into account The plots m Fig 5 show that the release ofsodmm, arsemc, zinc,
and fluorine assymptotmally approach the release hm]t dictated by the avadabd]ty test Similar
observat]ons have also been made m column leaching experiments (on granular matenal)
[7,8], rod]caring the practical s]gmficance of this test method as a screening tool for ultimate
release at the very long term
Dtffuslon Coefficwnt
From the concentrations measured m each penod (fraction) in the tank leaching test, the
leaching mechanism can be determined according to the procedure descnbed m the previous
...... Leachable a m o u n t according to the 'Availability test'
Na As
r ~ 9 . . . . . . . . . ;__~_._.
w
E
E
IO0
LU
r~ 70 OC
<
01 1 10 1OO 01 1 10 11(30
U.I
_1
I11
n"
W
Zn Se
_>
I-
< _,t.
.J 10
IE
0
01 1 10 100
~176
0.003
01 1 10 100
TIME [days]
FIG 5--Slgmficance of the "avallabthty test ""A comparison of the tank leachlng test results with the
Copyright
m a x iby
mASTM Int'l (all rightsamount,
u m leachable reserved); Mon Feb 24 23:07:03
according EST"'avatlablhty
to the 2020 test ""
TABLE 3--Leach parameters for stabthzed coal ash for apphcatton m road stabzhzatton
Element Concentration, mg/kg Avadabfllty, mg/kga Tb pOe, m2/s c Rd
Na 4140 297 19 l0 2 1
Mo 17 04 12 2 90
Cu 56 80 13 0 450
Sb 31 05 12 4 100
As 75 0 03 10 5 17
a AvaIlabihty = availability for leaching at L / S = 100 and pH = 4

b ~. = physical retardation (tortuoslty) factor
c pDe = -- log D,, De is the effecUvediffusion coefficient
d R = chemical retention factor
paragraph If diffusion-controlled release is the predominant mechanism, a diffusion coeffi-

cient can be calculated, also, the leach parameters' physical retardation and chemical retention
can be estimated In Table 3 the leach parameters for a stabilized coal ash specimen are given
as an example
Physical Retardation
The physical retardation or tortuosity reflects the extended path length of a diffusing ion in
the pore structure of a product relative to its free mobility in water Figure 6 gives the tortuosity
values obtained for sodium in a vanety of construction matenals and stabilized coal fly ash for
road base application The high tortuosity values for asphalt concrete are stnklng (r = 2000
to 50 000), and it is caused by the hydrophoblc nature of bituminous mixtures, which results
in partial wetting of the material Small pores may not be wetted at all The difference in phys-
ical retardation between asphalt concrete and ordinary concrete amounts to at least one order
of magnitude
In contrast, the stabilized coal ash has a relatively low tortuosity (T = 5 to 20) Within one
category of matenals, the tortuosity factor IS often correlated with density and strength of the
product. This implies that Improvement of technical product quality is associated with an
improvement in environmental quality
C h e m t c a l Retentton
The chemical retention reflects the retardation in the release of a component relative to that
of an inert species (for example, sodium) by chemical interactions of the diffusing ion with the
product matrix The pore water composition and the pore water pH are important factors in
this respect Reactive surface sites may also affect the diffUSlVlty of components in the pore
solution. Figure 7 g~ves the chemical retention o f b a n u m and fluonne in eight types of mate-
nals studied. The retention factors range from about 1 to 100 000, which indicates that sub-
stantlal reductions in release rates can be obtained by chemical modifications When measures
are taken to modify the chemical retention of specific components, it is important to ensure
that as a result of these changes other components are not mobilized
Chemical retention is influenced strongly by pore water pH and redox potential For a
proper quantification of these influences, matenals with smaller variations in total composi-
tion need to be studied In earlier work, relations between the chemical retention factor and
pH [10]
Copyright andInt'lbetween
by ASTM leachability
(all rights reserved); Mon Feb 24 and pH
23:07:03 EST[14]
2020 have been presented
Physical Retardation (Tortuoldty)

10 o
+ 'r t~lh P.,oWIly u h
" ++ . v~h MSW Inc:lnerlor ly mh
10 4 + ~ * l MSW Inclnem~ m
0
.2
e-
1 0 ~,
.+ d':
t
10 =, +**** qt~~
9+,+++ +
101,
10 ~ I I I I I I i i i i i i I i i i i i i i i i i i i i I i i i i I I
BP C PB CSB LC SP IB GM
Types of Products:
BP BituminousProducts LC Light weight Concrete

C Concrete IB Insulation bricks
PB Pavement Blocks SP Stabilized Products
CSB Calcium Silicate Brics GM Granular Material
FIG. 6--Physical retardationfactors (tortuosity)for eight types of products.
Reproducibility of Testing
To illustrate the reproducibility of the tank leaching tests, experiments were performed with
one product (calcium silicate bricks with 37% coal fly ash [4]), which was tested in different
sizes. This procedure resulted in different surface areas for the same volume of material tested.
The tortuosity in the three cases was found to be very similar (58 _+ 10). In an appendix the
results of the tank leaching experiments are summarized. The chemical retention factors
derived from these results are given in Fig. 8. The reproducibility is good when taking into
account that the surface area was very different in each case.
Long- Term Predictions

Within certain boundary conditions, the leach parameters obtained from tank leaching
studies can be used to predict contaminant release in a given time period.
A first approximation is to apply a simple one-dimensional diffusion model with a constant
source. For the short term, this approach is valid because the concentration in the material has
not decreased substantially to violate the boundary conditions of the simple model used. How-
ever, long-term predictions should take into account the limited source strength (fraction
Copyright by ASTM
available Int'l leaching)
for (all rights reserved);
andMonthe
Feb 24 23:07:03 EST 2020
dimensions of the material. A three-dimensional diffusion
Badum
10 7
0 Reference product
9tr with Coal Ily ash
10 a ,k with MSW incin~a~ot fly ash
+ wire MSW incinerator slag
"5
C
.2
10" +'+++++'+ +
+
tr 10 ~ +
+
i | i * i | i i
Fluorine
10 ~
10 ~
._o
.~ 10 =
q
10'
l O `) I I I a I I l i i i l i i i l I l i i i i l i I I l I | I I l | I
BP C PB CSB LC SP IB GM
FIG. 7--Chemical retention factors of barium and fluorine for eight types of products (for X-axis labels
see Fig. 6).
model is currently being developed to enable long-term predictions. The correction for the
available fraction on the long-term leaching could be addressed by an adapted one-dimen-
sional (l-D) model, but the three dimensional (3-D) model has the advantage of taking the
actual dimensions into account, so differences in leaching from a product with a cubic versus
a flat rectangular shape can be described.
Some preliminary results of this model are presented in Fig. 9. In this figure the release-time
curve is shown illustrating the deviation between the 1-D and 3-D models. With the 3-D
model, release from different product shapes or release from one side only of the material can
be modeled also. The long-term release for stabilized coal ash has been calculated based on
standard specimen size (10-cm cube) for proper comparison. Table 4 gives the release
expressed
Copyright by ASTMin
Int'lmg/m
(all rights2reserved);
at 64 days (duration
Mon Feb of experiment),
24 23:07:03 EST 2020 1 and 100-years' continuous exposure
164 HAZARDOUSWASTES
KZEN KZEN1 ~ KZEN2

0.092 m' 0.14 m' 0.17 m'
105
10"
_=
10 3
t-
10 2
oOu
e-
.= 101
J
0
rr
10 ~
10-1
Cs PO, 80, F Ba Cu Mo Sb As
FIG. 8--Reproducibility of chemical retention, calculated from the tank leaching test results.
to water. The m a x i m u m quantity, which can be released at extremely long term, the total con-
centration, and the effective diffusion coefficient in the product are given for comparison.
The release data in Table 4 show a large error for the 1-D estimation, in which depletion is
not accounted for, of the long-term release. The 1D-values marked with an exclamation mark
are higher than the maximal leachable amounts. The mobility of arsenic in the coal ash
amended mix is reduced by a factor of 4 compared with the traditional product. The percent-
age leachable of the total concentration for both products are comparable and low. In field
//
/
I
E I
/
10 4 /
/
E . . . . . . . . . . . . . . . . . . .
MEASURED
RELEASE
O9 ------ 1-Dimens.
O~ z
CALCULATED
m
- - 3-Dimens.
103 CALCULATED
@
> 100%
.m
Leached
E +
0
1O 2 ........ , ......... . ........ . ....... , . , ,
0.1 1 10 100 1000
Time [days]
FIG. 9--Cumulative release versus time; difference between the infinite source I D and the 3D diffusion
Copyright
model. by ASTM Int'l (all rights reserved); Mon Feb 24 23:07:03 EST 2020
TABLE 4--Calculated long-term arsemc release from stabthzed

coal ash (mg/m 2)
Stabilized Sand/Coal Ash
Traditional Coal Ash
Time ID 3D ID 3D
64 days 1 16 0 87 3 65 3 25
1 year 2 76t 1 30 8 72 6 55
100 years 27 64t 1 42 87 19t 10 61
Maximum leachable 1 42 10 63
Concentration m product 4 55 mg/kg 43 34 mg/kg
Percentage leachable 0 89% 0 87%
pDe [-log(mZ/s)] 10 0 10 7
apphcatlons, the temperature is lower than in the laboratory, causing the diffusion to slow
down, the shape and size are important for the release rate (surface-to-volume raUo), and the
materials are not continuously exposed to contact with water.
In contrast to the factors dimimshlng the release of elements, the carbonation of the stabi-
lized product does Increase the release to the environment. Estimates for reahstlc carbonation
rates are needed to quantify the impact by that exposure route To include durability aspects
m modeling actIvmes is an area of future research
Certtficatlon and Database

Since the a m o u n t of testing required to characterize a material m sufficient detail to identify
systematic leaching behavior and leaching mechanisms IS expensive and time consuming, it is
important to insure that m a x i m u m reformation is gained from these tests. This means the data
should be more assessable than lust from reports or publications Also, reformation is needed
which does not necessarily fit in a publication or report One powerful tool to extract useful
information from raw data is a well-organized database The U D N (Leaching Database Neth-
erlands) [15] compiles results from many different leach tests on many wastes and waste
products
The data base contains the following type of data
la General reformation about the origin, production process, manufacturer, and so on.
lb Information about the sampling procedure, sample preparation, and the institute/lab-
oratory that performed the leaching experiments
2. Physical and chemical characterization of the sample, including size analysis of the
sample and specific physical parameters for different types of products.
3 Type of test performed, general leaching conditions; the standard test methods are the
Dutch standards NVN 2508 and NVN 5432, and other tests are also supported
4. Total element concentration and leaching results.
No mean values oftne experimental data are stored Since multiple (for example, analyhcal)
results of the same sample can be entered separately in one sample record, a statistical evalu-
ation on any subset of the experimental data is possible. The database has extended password-
protection capabdmes to prevent improper use of sensitive Information like data type la
Effectwe
Copyright use
by ASTM Int'l (allof this
rights tool Mon
reserved); allows
Feb 24systematic leaching behavior to be identified Knowledge of
23:07:03 EST 2020
WASTE
MATERIALS/PRODUCTS
!
IS IT PRESENT IN DATABASE ? /
HAS COMPLETETEST PROCE- I - - ~'
DURE BEEN EXECUTED ?
ARE CRITICAL COMPONENTS

AND LEACHING SYSTEMATICS
KNOWN ?
T I
PERFORM COMPLETE TEST,
ALL ELEMENTS/COMPONENTS
PROCESS DEMONSTRATION,
SUBMISSION FOR CERTIFICATION
~ iill
PERFORMSHORTTEST, [
ONLY CRITICAL j -- --
ELEMENTS/COMPONENTS
CERTIFICATION TESTS
PRODUCT~PROCESS CERTIFICATION
REGULAR PRODUCT~PROCESS CONTROL
FIG. l O--Schematic representation of the role of a database, containing systematic leaching data, in
the certification process.
these systematics will ultimately reduce costs associated with environmental testing by direct-
ing emphasis to the most important constituents to be measured and parameters to be
assessed.
Environmental certification of construction materials containing bulk waste materials and
of stabilized waste materials becomes a realistic possibility using the systematics described.
Since the properties of a stabilized product need to fulfill certain requirements, this route leads
to better control over waste materials to begin with and will ultimately limit uncontrolled mix-
ing of wastes. The leaching database will also be a very useful tool in the certification process.
Its role in relation to certification is outlined in Fig. 10. Currently, projects have been started
to implement this approach in daily practice.
Acknowledgment
This work was partly funded by the Netherlands Agency for Energy and the Environment
(NOVEM) within the framework of the National Coal Research Programme and the National
Reuse of Waste Research Programme.
,.-1
VVV V
v
R o
o
8 -a ~ 8:=
V
r' O ~ O ~
u
.~ ~ ~ =-~ VVVVVVVV
C
~E
{,.} 2
I-a
c~
,-.1
{'xI
o-, ,v
O
C~
~o
O O ~ O O ~
Z
N
C
Z 0
e-
e-
0
r 1
r
oo
- - ~ - - 0 ~ ~ V V VVVVVVVV ~o- o o-o V
I ~ V
-8
,., 0
~ ~ 0 ~ ~ 0
o
~ 0 ~ 0 0 ~ 0 V
A V VVVVVVVV A o ~V
o'gg
Z
,--1
< r-.- ~ t..
1--
o4o~ Vv d
A 0 A Vv
o
oo'~o~ -.. 8 z
m
-~'~ ~
d
0
0 ~1~'~-~ CI~
> 0 .
I ~ -- ~,1~ ~1 = 1 ~ ~
d
8
o o~
v v
0 --
0o
0 0 0 0 0 0 0 0 ~ ~1 o o
--0-- 0 o - o V
VVVVVVV A V
r~ - s
U~
o0
0
V V Z O ~ o ~ ~~ V V
0 0 ~ ~ 0 ~ 1
VVVVVVVV A o ~V
r
4- v%t~l
~q
z
oo
t~ ~ . - ~ ~ _ o , ~ o~ oo Oc-.I
~A d ~A
0 A Vv 0
cq o~
o0
e~
Z
m
0
d
0 "0 :~
s -.
0
~ ~ ..~ . ~ ,.C~ __ t=I
o
c- 0
Copyright
O by ASTM Int'l (all rights reserved); Mon Feb 24 23:07:03 EST 2020
References
[1] "Hazardous Waste Proposal Guldehnes and Regulations," Section 250 13 Paragraphs C to E, U S
Environmental Protection Agency, Washington, DC, 1978
[2] "Toxicity Characteristic Leaching Procedure," Federal Regtster, Vol 5 l, No 114, 1986
[3] "German Standard Method for Water, Wastewater and Sludge Investigation," Part 4 "Auswasch-
barkelt mIt wasser" ($4), Group S, DIN 38414, Deutsches InstltUt fur Normung, Berlin, Germany,
1984
[4] de Groot, G J , van der Sloot, H A , Bonouvne, P, and Wljkstra, J "'Mammoet Report No 09
Charactenzauon of the Leaching Behavior of Intact Products" (Karaktensenng van bet ultloogge-
drag van lntacte produkten), Netherlands Energy Research Foundation (ECN), Petten, The Neth-
erlands, ECN Report No ECN-C-90-007, 1990
[5] American National Standard for the Measurements of the Leachability of Solidified Low-Level
Radioactive Wastes by a Short-Term Test Procedure, ANS1/ANS-16 l, 1986, American National
Standards Institute, New York, NY, 1986
[6] Prenorm NVN 2508 "Determination of Leaching Characteristics of Coal Combustion Wastes,"
Netherlands Normahsatlon Institute (NNI), Delft, The Netherlands UDC 662 62/67 543 2, 1987
Also published earlier in English as van der Sloot, H A , Piepers, O , and Kok, A , "A Standard
Leaching Test for Combustion Residues, Technical Report Bureau Energy Research, Netherlands
Energy Research Foundation (ECN), Petten, The Netherlands Projects BEOP-31, 1984
[7] Aalbers, Th G and Gerntsen, R , Leaching Behavior of Inorganic Parameters from Primary and
Secondary Materials (Ultlooggedrag van anorganlsche parameters uit prlmalre en secundalre grond-
stoffen), National Institute of Pubhc Health and Environmental Protection (RIVM), Bilthoven, and
Environmental and Energy Research--TNO, Apeldoorn, The Netherlands, Mammoet Report No
07, 1990
[8] Gerrltsen, R , "Characterization of the Leaching Behavlour of Primary and Secundary Materials In
the Post-Usage Stadium," (Karaktensenng van bet uitlooggedrag van verklelnde produkten), Envi-
ronmental and Energy Research--TNO, Apeldoorn, The Netherlands, Mammoet Report No 08,
1990
[9] ConceptprenormNVN5432 "Determlnation ofthe Release oflnorganic Contaminants from Con-
strucUon Materials, Monohtic Waste Materials and Stabilized Waste Products of Mainly Inorganic
Character," (Bepahng van de maximaal uItloogbare hoeveelheld en afglfte van potentieel schadehjke
componenten vanuit constructiematenalen, monohtlsche reststoffen en gestablhseerde reststofpro-
dukten), Netherlands Normalisatlon Institute, Delft, The Netherlands, September 1989
[10] van der Sloot, H A , de Groot, G J , and Wljkstra, J , "Leaching CharactensUcs of Construction
Materials and Stabilization Products Containing Waste Materials," Environmental Aspects of Sta-
blhzatton and Sohdtficatton of Hazardous and Radtoacttve Wastes, ASTM STP 1033, P L C6t6
and T M Gilham, Eds, American Sooety for Testing and Materials, Philadelphia, 1989, pp 125-
149
[11] Crank, J , TheMathematlcs ofDtffuston, 2nd ed, Clarendon Press, Oxford, 1975
[12] C6t6, P L and Isabel, D , "Application of a Dynamic Leaching Test to Solidified Hazardous
Wastes," Hazardous and Industrial Waste Management and Testmg Third Sympostum, ASTM
STP 851, L P Jackson, A R Rohhk, and R A Conway, Eds, American Society for Testing and
Materials, Phdadelphla, 1984, pp 48-60
[13] Duedall, I W, Buyer, J S, Heaton, M G , Oakley, S A , Okubo, A , Dayal, R , Tatro, M , Roethel,
F J , Wdke, R J , and Hershey, J P, "Diffusion of Calcium and Sulfate Ions in Stabilized Coal
Wastes," Wastes m the Oceans, Vol 1, I W Duedall, B H Ketchum, P K Park, and D R Kester,
Eds, Wiley, New York, 1983, pp 375-395
[14] de Groot, G J , van der Sloot, H A , and Wijkstra, J , "Leaching Characteristics of Hazardous Ele-
ments from Coal Fly Ash as a Function of the Acidity of the Contact Solution and the Llqmd/Sohd
Ratio, Envzronmental Aspects of Stablhzatton and Sohdtficatlon of Hazardous and Radioactive
Wastes, ASTM STP 1033, P L C6t6 and T M Gilham, Eds, American Society for Testing and
Materials, Phdadelphm, 1989, pp 170-183
[15] de Groot, G J , Mammoet Report No 12 "The Organisatlon and Structure of the Leaching Test
Database 'UITLOOG'," (Opzet en innchting database ultlooggegevens), Netherlands Energy
Research Foundation (ECN), Petten, The Netherlands ECN Report No ECN-C-90-008, 1990
Franco Medtci, ~ Carlo Merli, ~ Giancarlo Scoccia, 2 a n d Roberto Volpe 2
Release of Toxic Elements from Solidified

Wastes: A Mathematical Model
REFERENCE: Medic1, F, Merh, C, Scoccm, G, and Volpe, R, "Release of Toxic Elements
from Solidified Wastes. A Mathematical Model," Stabthzatton andSohdlficatzon of Hazardous,
Radtoacttve, and Mlxed Wastes, 2nd Volume, ASTM STP 1123, T M Gllham and C C Wdes,
Eds, American Sooety for Testing and Materials, Philadelphia, 1992, pp 171-181
ABSTRACT: A mathematical model to pre&ct the release of toxic elements lmmoNhzed m dif-
ferent cementmous matrices ~spresented Th~s model has been obtained on the bas~s ofprevmus
experimental tests canned out adopting a dynamic leaching test proposed by the American
Nuclear Sooety Dunng the tests, hthmm and bivalent copper have been used as tracers, &fferent
types of cementmous matrices and hme-pozzolan mortars have been tested To correlate the
release of these two elements with the mtnnslc properties of the matrix and of the ~mmobfllzed
element system, a statlsUcal approach has been used The final semlempmcal model obtained
has been very satisfactory for the pre&ctmn of contaminant loss from cementltmUS immobdlz-
mg matrices
KEY WORDS: retentlon/lmmoblhzatlon, release, cementltlous matrices, hme-pozzolan matri-

ces, hthlum, bivalent copper, modehng
One of the most debated issues in the field of toxic hazardous waste disposal through
cement-based stablhzatlon/sohdlfiCatlon systems is describing the kinetics of the release of
lmmoblhzed toxic elements The problem is remarkably complex as several interrelated fac-
tors influence the release process, such as the characteristics of the leaching medium, of the
immobilized toxic element, and of the matrix which immobilizes it
Several studies carried out on the subject [1,2] have provided both physical and empirical
models, more or less complex and/or able to thoroughly describe only particular and partial
aspects of the release kinetics, which are still the object of discussion and deepening
This study contributes to the solution of this problem by giving a definition of and adopting
a statistical approach to a model which describes the release kinetics of the immobilized toxic
elements in cementltlOUS matrices of various kinds This model results from previous experi-
mental tests [3-5 ], the aim of which was to evaluate the dependence of the modahtles and of
the entity of the release mechanism from the chemical and physical characteristics of the
Immoblhzing matrices and from the lmmoblhzlng mechanism of the tracers In such tests,
lithium and bivalent copper have been used as tracers, which, as is known, reflect different
retentlon/lmmoblhzatlon mechanisms Eleven different kinds of cementltlOUS and immobl-
hzlng matrices have been used, two of which are hme-pozzolan mortars
The experimental release tests have been carried out using the leaching dynamic test pro-
posed by the American Nuclear Society (ANS) [6]
Researcher and professor, respectively, Department of Chemical Engineering, Materials, Raw Mate-
rials and Metallurgy, Umverslty of Rome, Rome, Italy 00184
2 Professor, Department of Chemistry, Chemical Engmeenng and Materials, University of L'Aquda,
L'Aqulla, Italy 67100
171

172 HAZARDOUSWASTES
Background
Several mathematical models [1,2] based on different hypotheses and consldenng different
controlhng factors [2] have been proposed to predict medium- and long-term leachability
from cementitlOUS matrices. Some leaching models adopt bulk diffusion as the controlhng fac-
tor; others consider both bulk diffusion and dissolution chemical reactions, others are based
on mterfaclal mass resistance [ 1]
A well-known expression (Eq 1), denved from Fick's second law, considers diffusion as the
release controlling factor [ 7]. To formulate this equation, a semi-infinite medium and a zero-
surface concentration have been assumed dunng the leaching
S {DI 1/2
Ea, = 2 t t/2
Ao V \~- ] (1)
where
Ea, = contaminant loss dunng leaching penod n,
A0 = initial amount of contaminant present in the specimen,
S = surface area of the specimen,
V -- volume o f the specimen,
D = effective diffusion coefficient, and
t = time.
Th~s equation does not properly interpret expenmental data when other hmmng factors,
rather than a diffusive mechanism, control the leaching
Therefore, P. L C6t6 et al. [8] proposed a general equation (Eq 2) to take into consideration
the chemical and physical reactions occumng dunng the leaching
OC D 02C
Ot - ~ + E(R,) (2)
where
x = distance from the interface,
C = concentration, and
E(R,) = net effect of all reactions
There are many solutions to this general equation which take into account the different
mathematical expressions that the term E(R,) may assume [9]. The proposed solutions, very
complex in mathematical terms, are not really adequate to interpret leaching data.
Another mathematical model, which uses a kinetic approach, has been proposed by C. Pes-
catore et al [10]
E a , = kl[l - e -(kd+aka)t] (3)
where
kd, ka = coefficients describing the kinetics of the processes revolved from or onto the
surface,
3 = ratio of the sohd surface area to the aqueous solution volume, and
k~ = constant.
This
Copyright model
by ASTM xsrights
Int'l (all rather difficult
reserved); Mon Febto applyEST
24 23:07:03 since
2020 the kinetic coefficients, ka and ka, are complex
functions
Universidad ofdethe
Industrial physical
Santander andIndustrial
(Universidad chemical properties
de Santander) pursuant toof theAgreement.
License surfaceNoand ofreproductions
further the aqueous solution.
authorized.
MEDICI ET AL ON SOLIDIFIED WASTES 173
A satisfactory semiempincal approach (Eq 4) has been proposed by P. L C6t6 et al. [ 1 ]. such
a model singles out the different phases of the leaching, although it does not find out how the
intrinsic propemes of the matrix influence the release
E a . = kl(l - e -k2t) + k3t '/2 + k j (4)
where k~, k2, k3, and k4 are constants

Considenng the difficulties in the evaluation of the chemical reactions (Eq 2) and of the
kinetic parameters (Eq 3), many authors [11-13] agree that easy empirical models such as Eq
5 can be convemently employed in order to describe the release m time of the lmmobdlzed
toxic
y = af (5)
where
y = cumulatwe contaminant loss, and
a,b = constants
Such a function interpolates the experimental data obtained from leaching tests with very
high correlation coefficients, independently from the retention/immobihzatlon mechanism
which occurs in the cement-based stabihzation/sohdification process
In previous papers [3-5] the results of experimental tests on hthlum [3,4] and bivalent cop-
per release [5] from different cementltlOUS matrices have been presented.
The characteristics of the experimented cementltiOus matrices are the following
1--portland cement + sand

2 - - p o r t l a n d cement + slhca fume + sand
3 - - p o z z o l a m c cement + sand
4 - - p o r t l a n d cement + superplastlcizer + sand
5,6,7--portland cement + silica fume + superplasticlzer + sand
8 - - p o r t l a n d cement + air-entraining agent + sand
9 - - p o r t l a n d cement + retarder + sand
10,11--pozzolan + lime
For the mixing of Matrices l, 2, 3, 8, and 9, the following ratios were used
1 Water/cement ratio 0.5 by wt

2 Aggregate/cement ratio 3 by wt
Silica fume, air-entraining and retarder agents dosages were, respectwely, 15, 0 15, and 0.6%
in weight of cement
To manufacture Matrices 4, 5, 6, and 7, the following ratios and dosages were used
Matrix 4:
water/cement ratio: 0.35 by wt
Copyright byaggregate/cement ratio:
ASTM Int'l (all rights reserved); 3 by
Mon Feb wt EST 2020
24 23:07:03
Universidad superplasticlzer dosageIndustrial
Industrial de Santander (Universidad 2.5%demSantander)
wt of pursuant
cement to License Agreement. No further reproductions authorized.
174 HAZARDOUSWASTES
Matrices 5, 6, and 7"

water/(cement + silica fume) ratio 0 3 by wt
aggregate/(cement + slhca fume) ratio 3 by wt
silica fume dosage 10% in wt of c e m e n t - - M a t r i x 5
15% in wt of c e m e n t - - M a t r i x 6
25% in wt of c e m e n t - - M a t r i x 7
superptastlclzer dosage 4 5% in wt of(cement + slhca fume)
To manufacture Matrices 10 and 11, the following ratios and dosages have been used
1 Llme/pozzolan ratio 1/5 5

2 Lime/water ratio 1/1 9
adopting two different mixing modallties with regard to the constituents adding order (water-
hme-pozzolan, water-pozzolan-hme)
In all the matrices, different concentratmn levels ofhthium were immobihzed, while differ-
ent concentration levels of bivalent copper were immobilized in the following matrices l, 2,
3, 4, 10, and 11
On the whole, 37 d~fferent mixes were prepared for as many leaching tests, each of which
was repeated twice
To evaluate the leachability of the different matrices, a dynamic leaching test proposed by
ANS [6] was adopted The leaching test was conducted by immersing the specimen in delon-
ized water using a specimen surface area to leachant volume ratio of 0 064 cm i. All the spec-
imens were leached for 28 days according to a standard leachant renewal schedule
During the experimental tests the following parameters were measured
1 Compressive and flexural strength, permeability coefficient, and porosity of the matrix
2 Conductivity, pH, and hthium and copper concentration in the vanous fractions of the
eluates
The coefficient of permeabihty was determined according to the Bntish Standards Institu-
tion Methods of Testing Concrete (BS 1881/1970)
Proposed Model
In order to describe the release In time of the immobilized contaminant, in agreement with
what was proposed by several other authors, the following model has been chosen
y = at" (6)
where
y = cumulative contaminant loss, mg,
t = time, h, and
a , n = constants
To this purpose, the expenmental results of all the leaching tests carried out were interpo-
l a t e d - u s i n g the least squares m e t h o d - - w i t h a function of the aforesaid kind Thus, the values
o f " a " and "n" relative to each release test were calculated
Table 1 shows the values o f " a " and " n " relative to each release test (two iterates average)
with the respective correlation coefficients, r In this table, the permeability coefficient, K, and
initial
Copyright contaminant
by ASTM amount,
Int'l (all rights reserved); Mon A0,
Feb 24are alsoESTreported
23:07:03 2020
TABLE l--Leaching tests experimental values
Tracer
Element Matrix "a . . . . n" r K, cm/s Ao, mg
LI 5 0 242 0 331 0 986 5 70E-7 13 06

L~ 7 0 290 0 293 0 989 3 26E-7 13 18
Lx 7 0 865 0 250 0 982 3 26E-7 25 86
LI 5 0 842 0 283 0 975 5 70E-7 25 86
Ll 1 1 530 0 334 0 994 1 45E-6 43 20
Lx 2 1 806 0 276 0 959 6 52E-7 43 97
L1 3 1 309 0 333 0 982 1 25E-6 45 12
LI 4 1 310 0 305 0 999 8 15E-7 32 70
L~ 6 0 728 0 260 0 999 4 89E-7 28 67
L~ 1 1 117 0 361 0 999 1 45E-6 44 48
Lt 1 0 017 0 448 0 999 1 14E-6 4 40
Lx 2 0 029 0 435 0 992 6 52E-7 4 47
LI 3 0 019 0 468 0 998 1 25E-6 4 47
Ll 4 0 614 0 305 0 998 8 15E-7 16 70
Lx 6 0 343 0 260 0 999 4 89E-7 14 46
L~ 9 0 917 0 425 0 988 1 79E-6 44 74
L~ 8 1 209 0 361 0 987 1 55E-6 44 35
L1 9 0 379 0 428 0 990 1 79E-6 19 54
L1 8 0 572 0 370 0 889 1 55E-6 22 46
LI 1 0 611 0 365 0 986 1 45E-6 22 34
Lx 2 0 515 0 326 0 976 6 52E-7 21 95
L~ 3 0 276 0 431 0 989 1 25E-6 22 53
LI 10 1 062 0 512 0 987 1 05E-5 31 60
L~ 11 1 139 0 515 0 992 3 05E-5 31 60
L~ 10 2 204 0 472 0 969 1 05E-5 63 80
L~ 11 2 754 0 444 0 969 3 05E-5 63 10
Cu 3 0 003 47 0 304 0 993 1 25E-6 413 06
Cu 1 0 012 50 0 335 0 973 1 45E-6 395 39
Cu 2 0 006 40 0 305 0 991 6 52E-7 402 43
Cu 2 0 003 77 0 244 0 979 7 34E-7 409 09
Cu 1 0 013 40 0 303 0 979 1 45E-6 204 74
Cu 2 0 008 19 0 312 0 964 6 52E-7 201 22
Cu 3 0 001 20 0 283 0 980 1 45E-6 206 53
Cu 10 0 002 75 0 450 0 983 1 05E-5 587 40
Cu 11 0 006 25 0 418 0 894 3 05E-5 586 90
Cu 10 0 002 16 0 452 0 983 1 05E-5 289 10
Cu 11 0 002 18 0 470 0 988 3 05E-5 289 90
As s h o w n , all t h e c o r r e l a t i o n coefficients are h i g h e r t h a n 0.9, t h u s c o n f i r m i n g t h a t t h e m o d e l

used suits t h e e x p e r i m e n t a l d a t a i n d e p e n d e n t l y o f t h e k i n d o f l m m o b l h z e d c o n t a m i n a n t and,
therefore, o f t h e r e t e n t i o n / I m m o b i l i z a t i o n m e c h a m s m w h i c h occurs m the m a t r i x
T a b l e 1 shows also t h a t t h e " a " values r u n b e t w e e n 0.0012 a n d 2 754, the lower values b e i n g
in c o r r e s p o n d e n c e w i t h t h e c o p p e r release O n t h e o t h e r h a n d , " n " values r u n b e t w e e n 0 244
a n d 0.515, h o m o g e n e o u s l y d i s t r i b u t e d in all the release tests.
O n t h e basis o f t h e results s h o w n in T a b l e 1 a n d o n t h o s e o b t a i n e d in t h e e x p e r i m e n t a l tests
previously carried out, a m a t h e m a t i c a l m o d e l able to calculate t h e c o n s t a n t s " a " a n d " n " o f
t h e expression (Eq 6) has b e e n defined
A n a t t e m p t h a s b e e n m a d e to e v a l u a t e statistically in a d v a n c e t h e average effects o n the
c o n s t a n t s " a " a n d " n " o f t h o s e factors which, d u r i n g t h e e x p e r i m e n t a l tests, v a n e d A m o n g
t h e m , t h e m o s t significant factors are t h e h y d r o x i d e solubility o f the tracers used {lithium
h y d r by
Copyright o xASTM
i d e Int'l
( L I(all
OH ) reserved);
rights = 12 8Mong /Feb1 024
0 23:07:03
g H 2 0EST
, 2020
c o p p e r h y d r o x i d e [Cu(OH)2] = 0 003 m g / 1 0 0 g
H20}, lmtlal amount of the tracer immobilized m the matnx, A0, the matrix permeablhty, K,
which, as known, is correlated with other physlco-mechanical properties
Since in all the leaching tests c a m e d out, the specimen total surface/volume ratio was kept
constant, it was not possible to evaluate the effect on " a " and " n " of such a factor. However,
previous stu&es [11] prove that such a factor basically does not bear any influence on the
exponent "n"
The results shown m Table 1, along with the solublhty values of the copper and hthmm
hydroxide, allow the statistical evaluation of the main effects of A0, K, and S (solubility) on
" a " and " n " The " m a r e effect" of a factor is the change in response produced by a change in
the level o f the factor when other factors are averaged [14]
D u n n g th~s experiment, the S factor was considered at two levels, the A0 factor at three lev-
els, and the K factor at three levels. Such levels correspond, basically, for S to the solubility of
the two hydroxides used as tracers, for A0 to the three concentrations of the hthlum chloride
and the copper chloride solutions used in the tests (0 04, 0 2, and 0 4 M), and for the perme-
ablhty coefficients to the three orders of the exponents in which all the values obtained during
experimentation are comprised (10 -7, 10 -6, 10 -5)
Figures 1 and 2 show the mare effects of the factors A0, K, and S on " a " and "n", respec-
tively. All the factors have a more or less significant influence on "a", as for "n", K and A0
have a significant influence, while remarkably less influential is the effect of the factor S
In particular, these two figures emphasize how the influence of the type of trace metal used
is remarkable on " a " and significantly smaller on "n", as the gradient analysis concerning the
main effects of the S factor on " a " and "n" shows
1.0
ooau
X ~ Ao
O.8 O~S
0.6
0.4
0.2
0.0
K! K2 K3
L A I
Ao1 02 Ao3
) i
S1 S2
FIG l --Mare effects of the factors K, Ao, and S on "a"
0.6
O m g
,sn0*
X N
0.5
0.4
0.3
0.2 !
K1 g2 t(3
k~
t
k~ /~:~3
S1 S2
FIG 2--Main effects of the factors K, Ao, and S on "'n"
Using these results, semlemplrlcal linear mathematical models of the following kind were
defined
y = ~,f~(x) + ~ ( x ) + + ~(x) (7)
where y represents " a " or "n", a n d f ( x ) various functions of the independent vanables K, A0,
and S Such functions were chosen on the basis of theoretical and purely mathematical
assumptions and on the basis of the behavior analysis of the main effects shown in Figs 1
and 2.
A s f ( x ) , the following functions of the variables K, A0, and S were considered
f, f2 f3 f4 f5 f6 f7 f8 f9 f,o fn
K 13
Ao K S KS KAo AoS AoKS log K K/Ao log 70o7 AoK~176
The proposed models have been determined by the statistical mathematical method of step-
wise multiple regression [15], choosing as the best model that in whose correspondence the
square average and the standard error of the esumate reached a m i n i m u m value. As known,
according to this method, the various independent variables are introduced in the model in
function of the partial correlation coefficient produced; that is, the variable which produces
178 HAZARDOUSWASTES
the highest value of the partial correlation coefficient is introduced first, then the variable
among the remaining ones which produces the highest partial regression coefficient, and so
forth Using this method, for " a " the best model is the following
a = 0 003 46AoK~ ~ -- 640 295 86K/Ao + 0.0772 (8)
In correspondence with each step, the significance of the introduction of the respective
group of vanables has been evaluated by an F-test at significance level a = 0 01
At Step 2, in which the sequential research stopped since the error m i n i m u m value had been
reached, the following results have been obtained
1 Multiple correlation coefficient 0 953

2 F-value for analysis of variance 173 7
3 Standard error of estimate 0 216
4 Standard error for regression coeffioent 0 003 46 0 000 20
5 Standard error for regression coefficient 640 295 86. 220 225 39
Using a similar procedure, the best model for "n" resulted in the following
g 13
n = 0 1 6 3 1 o g ~--v~-- 1 0 - 5 S - - 5 1 4 6 5 6 K + 1856 (9)
A~'
Here, also, the introduction of each variable has been verified at a significance level ~ =
0 01 The results obtained at Step 3 of the best model sequential research, in correspondence
of which the m i n i m u m value of the standard error of estimate has been reached, are the
following
1 Multiple correlation coefficient. 0 930

2 F-value for analysis of variance 75 7
3 Standard error of estimate 0 030
4 Standard error for regression coefficient 0 163" 0 015.
5 Standard error for regression coefficient 10 -5. = 10 -6
6 Standard error for regression coefficient 5146 56 1100.66
Table 2 shows, for " a " and "n", the expenmental values, those estimated by the proposed
models, and the corresponding residuals In order to test even further the model rehablhty, the
analysis of residuals has been performed [ 15 ], plotting the values of the normalized residuals
in function of the calculated values of the independent variable ( " a " or "n") The normalized
residuals are defined as follows
d_Y,-f ~,
MSR (10)
where
d, = normalized residual,
y, = experimental value of the dependent variable,
)3 = calculated values of the dependent variable, and
Mby
Copyright S RASTM
= Int'l
mean square
(all rights reserved);residual
Mon Feb 24 23:07:03 EST 2020
TABLE 2--Table of residuals
" a " Value " a " EsUmate Residual " n " Value " n " Esumate Residual
0 242 0 337 --0095 0 331 0 331 0000

0 290 0 321 --0031 0 293 0 280 0013
0 865 0 579 0 286 0 250 0 247 0 003
0 842 0 633 0 209 0 283 0 297 - 0 014
1 530 1 203 0 327 0 334 0 353 -0019
1 806 1 063 0 743 0 276 0 283 - 0 007
1 309 1 223 0 086 0 333 0 338 - 0 005
1 310 0 835 0 475 0 305 0 317 -0012
0 728 0 679 0 049 0 260 0 278 - 0 018
1 117 1238 --0121 0361 0351 0010
0017 0 017 0000 0 448 0 466 -0018
0 029 0 084 --0 055 0 435 0 396 0 039
0019 0 014 0005 0 468 0 453 0015
0 614 0 441 0 173 0 305 0 350 -0045
0 343 0 365 --0 022 0 260 0 312 - 0 052
0 917 1 291 --0 374 0 425 0 369 0056
1 209 1 249 --0 040 0 361 0 357 0 004
0 379 0 559 --0 180 0 428 0410 0018
0 572 0 638 --0 066 0 370 0 391 - 0 021
0 611 0 629 --0018 0 365 0 386 -0021
0 515 0 555 --0040 0 326 0 317 0009
0 276 0 623 --0 347 0 431 0 373 0 058
1 062 1 112 --0 050 0 512 0 504 0 008
1 139 1 003 0 136 0515 0499 0016
2 204 2 490 --0 286 0 472 0 469 0 003
2 754 2 851 --0 097 0 444 0 465 - 0 021
0 003 47 0 0804 --0 077 0 304 0 301 0 003
0 012 5 0 0799 --0067 0 335 0 315 0020
0 006 4 0 0805 --0 074 0 395 0 245 0 060
0 003 8 0 0806 --0 077 0 244 0 255 -00l 1
0 013 4 0 0753 --0 062 0 303 0 348 - 0 045
0 008 2 0 0773 --0 069 0 312 0 279 0 033
0 001 2 0 0753 --0 074 0 283 0 347 - 0 064
0 002 8 0 0769 --0 074 0 450 0 431 0 019
0 006 3 0 0578 --0 052 0 418 0 427 - 0 009
0 002 2 0 0595 --0 057 0 452 0 466 - 0 014
0 002 2 0 0167 --0 014 0 470 0 461 0 009
Statistically, a n d for a sufficiently high n u m b e r o f tests, a m o d e l is c o n s i d e r e d to be satisfac-

tory i f 9 5 % o f t h e value is i n c l u d e d in a r a n g e o f _+ 1 96 In Figs 3 a n d 4, the analyses o f the
residuals for " a " a n d " n " are reported, respectively
T h e s e figures show h o w b o t h t h e p r o p o s e d m o d e l s c a n be c o n s i d e r e d adequate, because only
for " a " a n e x p e r i m e n t a l p o i n t is q u i t e far f r o m the line d, = + 1 96 Actually, whereas there
IS basically only o n e residual for " n " , w h i c h exceeds, t h o u g h shortly, line d, = - 1 96, for " a "
t h e r e are two p o i n t s w h i c h exceed line d, = + 1 96 (see Fig 3), a m o n g w h i c h o n e IS r a t h e r far
f r o m s u c h a line Such exceedence w o u l d require f u r t h e r investigations in o r d e r to verify if the
e x p e n m e n t a l test is w r o n g or i f t h e influence o f o t h e r variables is to be c o n s i d e r e d
Conclusions
T h e m o d e l p r o p o s e d to describe t h e l i t h i u m a n d b i v a l e n t c o p p e r release from cementitlOUS
m a t r i c e s is q u i t e easy, satisfactory, a n d a l m o s t always overestimates the release values with
Copyright
respectby ASTMto tInt'l
h e (all
e xrights
p e rreserved);
i m e n t aMon Feb 24 23:07:03 EST 2020
l ones
180 HAZARDOUSWASTES
d
I
+1.96
i ~9 o 99 9 9
-1.96
* I i I , I
1 2 3
"a" c a I c u I ot..ed
FIG 3--Analysts of residual for "'a'"
d
I
+1.96
9 0 O 0 9
9 9 9 9
9 9
~176 |
-1.96
I I I I I
0 0.1 0.2 0,3 0.4 0.5
"n" calculated
FIG
Copyright by ASTM Int'l (all rights reserved); Mon Feb 4--Analysts
24 23:07:03 EST 2020 of reslduals for "n "'
Obviously, the values of the constants ofEqs 8 and 9 are valid for the adopted experimental
conditions.
The model, as all the semlempincal models, does not claim to describe the microscopic
aspect of the problem However, it is qmte successful in describing, thanks to the rehablhty
g~ven by the statistical evaluation, the macroscopic aspect, which has its practical importance
The model can certainly be improved by extending the experimentation to other elements
and by introducing further variables, such as the solid surface/volume ratio, which surely has
its influence on the release
Moreover, the model can be a valid starting point to obtain physically based models.
Acknowledgments
Financial support for this research was provided by Ministry of Scientific Research
(MURST 40%, 1989)
References
[1] C6t~, P L and Constable, T W, Nuclear and Chemtcal Waste Management, Vol 7, 1987,pp 129-
139
[2] Pooh, C S m Environmental Aspects of Stabdlzatton and Solld~catton of Hazardous and Radio-
active Wastes, ASTMSTP 1033, P L C6t~ and T M Gdham, Eds, American Sooety for Testing
and Materials, Phdadelphla, 1989, pp 114-124
[3] Medlo, F, Merh, C, Scocoa, G , and Volpe, R m EnvtronmentalAspects ofStabdtzatlon andSohd-
lficauon of Hazardous and Radzoactlve Wastes, ASTM STP 1033, P L C6t~ and T M Gdham,
Eds, American Sooety for Testing and Materials, Phdadelphla, 1989, pp 229-237
[4] Medlo, F, Merh, C, Scocoa, G, and Volpe, R, m Proceedings, Fifth InternaUonal Conference on
Sohd Wastes, Sludge and Residual Matenals Charactenzat~on, Technology, Management, Pubhc
Pohcy, Roma, 26-29 Apr 1989, Antomo Delfino Edition, Rome, pp 546-555
[5] Medic1,F, Scoccm, G , and Volpe, R, Ingegnerta Samtarla, Vol 36, 1988, pp 77-83
[6] "Measurement of the Leachabdlty of Sohdlfied Low-Level Radioactive Waste," Working Group A
N S 16 1, Amencan Nuclear Socwty, La Grange Park, IL, 1981
[7] Godbee, H W e t al, Nuclear and Chemtcal Waste Management, Vol 1, 1980, pp 29-35
[8] C6t~, P L, Bndle, T R, and Benedek, A, m Hazardous and Industrtal Sohd Wastes Testing and
Disposal Stxth Volume, ASTM STP 933, D Lorenzen et al, Eds, American Sooety for Testing
and Materials, Phdadelphm, 1986, pp 63-78
[9] Crank, J, The Mathematics of Dtffuston, Oxford Umverslty Press, London, 1956
[10] Pescatore, C, Slmonson, S A, and Machlels, A J, "Predlctxve Modelhng of Long-Term Leaching
Behawours," presented at Waste Management 82, Tucson, 8-11 March 1982
[11] Subramanmn, R V and Nahahngam, R m Toxw and Hazardous Waste Dlsposal, R B Pojasek,
Ed, Vol 1, Chap 14, Ann Arbor Soence, Ann Arbor, 1982, pp 271-276
[12] Godbee, H W and Joy, D S, "Assessment of the Loss of Radioactive Isotopes from Waste Sohds
to the Enwronment, Part 1 Background and Theory," TM 4333, Oak Ridge NaUonal Laboratory,
Oak Ridge, TN, 1974
[13] Zamoram, E, Sernm, G , and Blanchard, H, Cement and Concrete Research, Vol 16, 1986, pp
394-398
[14] Dawes, O L, The Design and Analysis of Industrial Experiments," Imperial Chemical Industnes,
Ltd, Edinburgh, Scotland, 1971
[15] Buzzl,F G,"Anahsleldentlficazlonedelmodelh,"CooperatwaLlbranaUnwersltanaPohtecnlco,
Mdano, Italy, 1975
Giorglo D e Angelis, ~Antonia Marchetti, ~ and Stephania B a l z a m o ~
Leach Studies" Influence of Various

Parameters on the Leachability of Cesium
from Cemented BWR Evaporator
Concentrates
REFERENCE" De Angehs, G , Marchettl, A , and Balzamo, S, "Leach Studies Influence of
Various Parameters on the Leachabihty of Cesium from Cemented BWR Evaporator Concen-
trates," Stabdtzatton and Sohdtficatlon of Hazardous, Radtoactwe, and Mtxed Wastes, 2nd Vol-
ume, STP 1123, T M Gdham and C C Wdes, Eds, Amencan Society for Testing and Mate-
nals, Phdadelphm, 1992, pp 182-192
ABSTRACT: The influence of different parameters on the leaching of cesmm from cemented
boding water reactor (BWR) evaporator concentrates (sulfates) was tested In particular, the van-
atlon of the following parameters was evaluated temperature, mmal tracer concentratmn,
renewal and chemical composmon of the leachant, and specimen size In a further senes of tests,
the following effects on leaching were also taken into account leachant volume, addmon of mr-
entraining agent to the mix, shorter cunng Umes (3 days instead of 28 days), different ways of
carrying out the test (static, instead of dynamic condmons) Demonstratmn is gwen that only a
few parameters, hke temperature and chemical composmon of the leachant, have an effect on
leaching A strong influence, however, is caused by the way of preparing the samples, which, in
general, can strongly modify the final propemes of cemented waste matenals
KEY WORDS" cesium, chemical tracers, diffusion, dynamic test, leaching, sulfate wastes, radio-
active wastes, release, standard test, static test
Leach studies are the m o s t suitable for evaluating the retention capacity o f solidified waste
forms D u n n g the past m a n y efforts have been devoted to understanding basic leaching mech-
anisms and long-term leaching behavior, which are not yet completely clarified Studies have
also d e m o n s t r a t e d that, apart from pressure, which seems to have little effect, leach rate is
strongly influenced by a senes o f parameters ranging from temperature to the leachant's chem-
ical c o m p o s i t i o n [ I]
This paper, w n t t e n with the financial support o f the European C o m m u n i t i e s [Contract N o
FI 1W-0101-I(A)], summarizes the results related to the release o f cesium from c e m e n t e d boil-
ing water reactor (BWR) evaporator concentrates (sulfates) in v a n o u s experimental condi-
tions. In particular, the variation o f the following parameters has been evaluated temperature,
initial tracer concentration; leachant renewal based on two different standard methods Inter-
national Organization for Standardization (ISO) 6961-1982 and A m e n c a n National Stan-
dards I n s t l t u t e / A m e n c a n Nuclear Society ( A N S I / A N S ) 16 1 [2,3], the leachant's chemical
composition, and specimen size.
Further experiments were also performed to evaluate the following influences on leaching
leachant volume, adding an air-entraining agent to the mix to increase the sample's porosity,
ENEA, C R E Casaccla, Rome, Italy
182

DE ANGELIS ET AL ON LEACH STUDIES 183
shorter curing times, and different ways of carrying out the test (test under static condltxons,
without any leachant renewal)
Sample Preparatton
Cyhndncal samples were prepared in each case, with a diameter/height ratio of 1 Diameter
was generally 5.0 cm, except with the specimens used to test the influence of V/F ratio (the
ratio between volume and external surface area of the sample) In that case, samples with
increasing diameters of 3 0, 5 0, and 7 5 cm were prepared The final weight of the specimens
was about 200 g
The following formulation for prepanng the mixes was usually adopted. Italian 425 poz-
zolanic cement, 65 5 g; and BWR evaporator concentrates (sulfates), 34 5 g; for a total of 100 0
g, with 66.0 mg ofcesmm chloride (CsC1) added to 100 g of mix
In the experiments testing the influence of the initial chemical tracer concentration, the fol-
lowing amounts of CsC1 were used" 33, 66, and 660 mg.
The composition of the evaporator concentrates was as follows water 78 5%, sodium sulfate
(Na2SO4), 21 1%, barium chloride (BaC12), 0 25%, nickel sulfate (N1SO4), 0 12%, and potas-
sium manganate (KMnO4), 0 03% The influence of increased porosity on leaching was yen-
fled by adding a commercial air-entraining agent to the mix Each sample was cured in a
humid atmosphere (relative humidity over 90%) for 28 days, except when the evaluation of a
shorter c u n n g period (three days) was performed.
In the two series of experiments the leaching behavior of the various samples was compared
to that of a reference sample, of cyhndncal shape, with both a diameter and height of 5 0 cm,
containing 100 mg of cesium, cured at a relative humidity not less than 90% for 28 days, leach
tested at ambient temperature by deiomzed water, and whose renewal was made according to
the time schedule reported in ANSI/ANS 16 1 Two different batch preparations were made
for the two series of experiments, each including the reference samples Distinct preparations
were made for samples containing a different a m o u n t of tracer and for those that were added
with an air-entraining agent The opportunity of using a reference sample for each experiment
series was due to the strong influence that the mixing method can have on the final properties
of the specimens (Fig 1)
§ PZ/S-ANSI O PZ/S-ANSI
40 i i i i i i i i I
35 _...-o-
30
?
o 25 .......-41.--
N 20
~
r..) 15
10
5
0 , 9 , . , . , . , 9
0 10 20 30 40 50 60 70 80 90 100
TIME (days)
FIG 1--Cesmm leachingfrom two tdentwal referencesamples of pozzolamc-cemented sulfatesfrom
two different preparattons
184 HAZARDOUSWASTES
Leach Experiments
Two different standard leaching methods were followed to evaluate the influence of a dif-
ferent leachant renewal: ANSI/ANS 16.1 and ISO 6961-1982 The difference between the
dynamic and static tests (the last one made without any leachant renewal) was also evaluated
and its consequences on leach data were checked using the ANSI method
The experiments were usually conducted at ambient temperature (varying in the range 295
to 298 K; 22 to 250C), except when the influence of that parameter was verified In that case,
Increasing temperatures of 313 K _+ 1 and 343 K _+ 1 were chosen
The influence of the leachant's chemical composition was demonstrated by using the fol-
lowing leaching waters delonlzed water (conductivity less than 150 #S/m), tap water, and syn-
thetic seawater The leachant was generally used in the volume of 250 mL [m that case the
ratio between the volume of the leachant and external surface area of the specimen (V/F) was
about 2 cm] To evaluate the influence of leachant volume in these expenments, increasing
volumes of 500 mL ( V/F = about 4 2 cm) and 750 mL ( V/F = about 6 2 cm) were used The
leachate, after any leachant renewal, was filtered under vacuum by using a filter of 0 45 ~m,
then acidified to a pH of I with concentrated hydrochloric acid The concentratmn of cesium
In the leachate was determined by atomic absorption spectrophotometry

The results of the leach tests are reported in Tables I and 2, and in F~gs 1 to 10 In order to
have a useful comparison with other jobs, the test data are reported, unlike ANSI/ANS spec-
ifications, in the following different ways
1 As incremental leaching rate, R, (kg/m 2 • s)
R, = a,/(Ao • F • t.) (1)
where
a, -- weight in kilograms of each constituent leached d u n n g each leaching interval,
Ao = ratio between weight of the lth constituent initially present in the specimen and weight
of the sample,
F = exposed surface area of the specimen, in square metres, and
t, = duration of the nth leaching period, in seconds.
2 As leaching factor, v, (m/s)
v, = R , / p (2)
where
o = density of the specimen (kg/m 3)
3 As cumulative fraction released, CFR% (m)
(Z(a,/Ao) • (V/F)) • 100 (3)
where
ao -- weight in kilograms of the lth constituent initially present in the specimen, and
V -- volume of the specimen, in cubic metres.
T A B L E 1--Cesium leachmg from pozzolantc-cemented BWR

evaporator concentrates (average of two values calculated after 90
days of leaching)
Sample a Rt vz CFR
ANSI 2 26 10 -7 1 15 10 - l ~ 35 3 10 -2
ANSI-40 3 78 10 -7 1 90 10 -1~ 35 1 10 -2
ANSI-70 6 29 10 -7 3 28 10 - l ~ 72 2 10 -2
50-ANSI 2 10 10 -7 I 10 10 - j ~ 36 6 10 -2
100-ANSI 2 26 10 -7 1 15 10 - l ~ 35 3 10 -2
1000-ANSI 2 40 10 7 1 22 10 - l ~ 36 5 10 -2
ANSI 2 26 10 -7 1 15 10 1o 35 3 10 -2
ISO 3 57 10 -7 1 86 10 - l ~ 42 9 10 2
DW 2 26 10 -7 1 15 10 - l ~ 35 3 10 2
TW 2 37 10 -7 1 15 10 lo 35 3 10 -2
SW 3 62 10 -7 1 81 10 - l ~ 31 4 10 2
s 1 77 10 -7 0 88 10 ~o 26 9 10 -2
m 2 26 10 -7 1 15 10 -1~ 35 3 10 -2
B 3 12 10 -7 1 57 10 -1~ 35 6 10 2
V 2 10 10 -7 1 07 10 - l ~ 25 4 10 -2
2V 2 40 10 7 1 20 10 - l ~ 27 2 10 -2
3V 2 20 10 -7 1 09 10 - l ~ 26 1 10 -2
ANSI 2 10 10 -7 1 07 10 - l ~ 25 4 10 2
ANSI-aer 2 01 10 -7 1 01 10 -~~ 25 2 10 -2
3D 1 81 10 -7 0 9 1 10 - l ~ 2 4 6 10 -2
2 8D 2 10 l 0 -7 1 07 l 0 -10 25 4 10 -2
ST 2 51 10 -7 1 31 10 lo 25 2 10 -2
DY 2 10 10 -7 1 07 10 -I~ 25 4 10 -2
a Key to abbreviations
S = B W R evaporator concentrates (sulfates)
PC = Italian high-resistance port l a nd c e m e n t
PZ = Italian high-resistance pozzolanic c e m e n t
50, 100, 1000 = initial c o n c e n t r a t i o n (mg) of cesium in the
specimen
60, 70 = 40 and 70~ t e m p e r a t u r e of the test
D W = deionized water
T W = tap water
SW = seawater
s = small sample (diameter 3 cm, height 3 cm)
m = m e d i u m sample (diameter 5 cm, height 5 cm)
B = big sample (diameter 7 5 cm, height 7 5 cm)
ISO = standard m e t h o d ISO 6961-1982
ANSI = standard m e t h o d A N S I / A N S 16 1
V = 250 m L o f l e a c h a n t
2V --- 500 m L o f l e a c h a n t
3V = 750 m L o f l e a c h a n t
aer = sample prepared with a ddi t i on of an air
e n t r a i n i n g agent
3D = specimen cured in h u m i d a t m o s p h e r e for
three days
28D = specimen cured in h u m i d a t mos phe re for 28
days
D Y = test m a d e u n d e r d y n a m i c c ondi t i ons (accord-
lng to A N S I / A N S )
ST = test m a d e under static co ndi t i ons (without any
leachant
Copyright by ASTM Int'l (all rights reserved); Mon Feb 24 23:07:03 EST 2020renewal)
T A B L E 2--Diffusion coefficients of cesium leached from

pozzolanic cemented B WR evaporator concentrates (average of
two values calculated after five days of leaching).
Sample a D Li vi Percent Release
ANSI 5.1 10 -8 7.3 1.6 10 -9 20.0

ANSI-40 3.6 10 -8 7.4 1.4 10 9 16.2
ANSI-70 ...o 6.7 3.0 10 -9 35.5
50-ANSI 5.5 10 -8 7.3 2.2 10 -9 20.8
100-ANSI 5.1 10 -8 7.3 1.6 10 -9 20.0
1000-ANSI 6.6 10 8 7.2 2.2 10 -9 22.4
DW 5.1 10 8 7.3 1.6 10 -9 20.0
TW 2.3 10 -8 7.6 1.4 10 -9 13.8
SW 0.9 10 -8 8.0 0.8 10 -9 8.2
s 2.5 10 -8 7.6 1.2 10 9 21.0
m 5.1 10 -8 7.3 1.6 10 -9 20.0
B 3.1 10 -8 7.5 1.4 10 -9 10.2
V 1.2 10 -8 7.9 1.1 10 -9 10.6
2V 1.3 10 8 7.9 1.1 10 -9 11.0
3V 1.2 10 -8 7.9 1.1 10 -9 10.6
ANSI 1.2 10 -8 7.9 1.1 10 -9 10.6
ANSI-aer 1.4 10 -8 7.8 1.2 l0 9 11.3
3D 1.1 10 -8 7.9 1.1 10 -9 10.0
28D 1.2 10 -8 7.9 1.1 10 -9 10.6
ST 1.3 10 -8 7.9 1.6 10 -9 11.2
DY 1.2 10 -8 7.9 1.1 10 -9 10.6
For abbreviations, see Table 1.

b Semi-infinite diffusion model not applicable (percentage of
release over 20).
[] P Z / S 0 PC/FS 9 PZ/IER
o . , . , . , . i . , i i . i , i , I , i , , - ,
-.6~
o -,8
r..)
~. -1.
u -1,2.
-1.4,
-1,6.
-1.8,
-2
-2.2.
-2,4
0 ,'2 " ,4 " .6 " ,~} ' 1 " 112'114" 116"118' ~- '212 214 2,6
Log TIME (days) _

FIG. 2--Diagram of log C/Co versus log time related to the leaching of cesium from B WR evaporator
concentrates, flter sludges, and bead ion exchange resins.
4 As percentage of release (dimensionless)
(s • 100 (4)
Cesium is leached from the sohd phase according to a diffusion controlled mechanism, as
demonstrated by leaching kinetics, which can be described by the emplncal relationship
C/Co = Bt" (5)
where
C = cesium concentration in the liquid phase,
Co cesium concentration in the solid phase,
=
B = value of C/Co at t = 1 day [4], and

n = time exponent, n = 0 5 stays for a leaching diffusive mechanism, and n = 0 5 IS for
a better retention of the contaminant
The curves reported in Fig 2 demonstrate that cesium is leached according to a diffusion-
controlled phenomenon when BWR evaporator concentrates are cemented On the contrary,
filter sludges and bead Ion exchange resins retain cesium with different mechanisms (adsorp-
tion and cationic exchange, respectively)
With n = 0 5 a relationship drawn from the Flck's law under the condition o f a semnnfinlte
medium can be used to describe the leaching mechanism
(C/Co) (V/F)= 2(D t/Tr) '/2 (6)
The diffusion coefficient D can be calculated from Eq 6 under the condition of a cesium release
less than 20%, a condition venfied until five days of leaching (Table 2)
The leachability index L, is also reported in Table 2, and it has been calculated according to
C6t6 et al [5], as
t, = (l/m) ~, log(fl/D) (7)

1
9 25 ~ O 40 ~ A 70 ~
8O
7oi
c~ 6O
o 50
40
~ 30
U 20]
10
O, 9 . , . , . . . . . . . , . , . ,
10 20 30 40 50 60 70 80 90 100
TIME (days)
FIG 3--Cumulattve fractlons of cesmm released at vartous temperatures
188 H A Z A R D O U S WASTES
9 25 ~ O 40 ~ 9 70 ~
40 , , , , , , ,
5
o 35 ~
~ 30
~ 25
~I 20
o
U 5
0
0 Ib 2; 3; 4; 5; 6; 7; 80
-2
C F R x 10 of C e s l u m
FIG 4--Temperatu~ effect on cesium leachlng O~t made accordlng to ANSl m~ho~
As expected [6], the leachability is strongly influenced by the increase m temperature (Fig 3),
not only as total leached a m o u n t but mainly as leach rate at increasing leach times Moreover,
at elevated temperatures (70"C) the leaching mechanisms seem to change, as shown by plotting
the CFR at a given temperature against the CFR at ambient temperature. A straight line is for
a constant leaching mechanism as it is during the first leaching Intervals, while a different slope
of the curve holds for a different leaching mechanism (Fig 4)
The initial chemical tracer concentration does not seem to have any Influence on the leach-
lng data, at least in the range of concentrations that were investigated 50, 100, and 1000 mg
of cesium m the cylindrical samples (Fig 5)
The leachant renewal also Infuences the leach rate (Fig. 6) Both the ANS 16 1 and ISO tests
call for leachant replacement at specified times. Actually, as has been demonstrated [6], sat-
uratIon effects may occur for Inadequate leachant replacement frequency, with the conse-
quence of questionable data analysis The discrepancies in the trend of the curve according to
§ 50-ANSI A 100-ANSI O 1000-ANSI

40 | i i i i , i . . . .
35
3O
c~
i 25
o
x 20
u
lo
5
o , . , . , . . . , . . . , . , , , .
o 10 20 30 40 50 60 70 80 90 100
TIME (days)
FIG 5--Cumulattvefracttons of cestum releasedfrom samples contamlng mcreasmg amounts of chem-
tcal tracer
Copyright by ASTM (CsCI)
DE A N G E L I S ET AL. ON L E A C H STUDIES 189
ANSI-40 O ISO-40
8 0 - 9 - , 9 , . , . . . . . , . , . , . = . = . , . ,
7o ~
~ 60 ~
I 50
N40~ "
i~ 30
o 20
10
0
0 " 2'0 " 4'0 ' 6'0 " 8'0 " 1 6 0 ' li0'1'40' 1 ; 0 " 1 ; 0 ' 2 ; 0 ' 210 " 2'40 " 260
TIME (days)
FIG. 6--Cumulative fractions of cesium released from samples of pozzolanic cemented sulfates leach-
tested according to two different test methods.
the ISO method (Fig. 6) can be presumably attributed to deep crack formation in the sample,
which comes into evidence after about four months (Fig. 7). Not surprisingly, the different
leachant chemical compositions relate to different quantities of cesium released (Fig. 8). In
particular, as evidenced in Table 2, an increasing ionic strength corresponds to a decreasing
diffusion coefficient.
The opportunity of testing the influence on leaching due to shape and size of the samples is
derived from the specifications of the standard methods: ISO indicates cylindrical specimens
with a surface area in the range of 1 9 10 -3 to 5 • 10 -~ m 2 and a diameter/height ratio of 1;
ANSI on its own says that samples can be monolithic cylinders (with a length/diameter ratio
in the range of 0.2 to 5), parallelepipeds or spheres with dimensions included within a given
interval.
Our experiments were carried out with cylindrical samples having the same concentration
of cesium and increasing diameter. The diameter/height ratio was kept constant, as well as the
ratio between volume of the leachant and external surface area of the specimen.
FIG.
Copyright 7--Deep
by ASTM Int'l (allcracks in cylindrical
rights reserved); Mon Feb 24samples after
23:07:03 EST 2020four months of leaching in deionized water at 40"C.
I"I A N S I - T W @ ANSI-SW 9 ANSI-DW

. . . . . . . , . , . , . , , | . , . i
,04
35
30.
! 251
~ 201
N
15s
0 1 0s
s2
0' 9 , . , . 0 . , , , . , . , 9 ) . , 9
10 20 30 40 50 60 70 80 90 100
TIME (days)
FIG 8--Cumulatlve fractmns of cestum released from samples tmmersed mdtfferent leachmg waters
(detonzzed water, DW, tap water, TP, synthettc seawater, SW)
The dimensions of the specimens tested were as follows
1 diameter 3 cm, height 3 cm for the small sample

2 diameter 5 cm, height 5 cm for the medium sample.
3. diameter 7.5 cm, height 7 5 cm for the big sample
As shown in Fig. 9, the cumulative fraction released is practically the same for the samples
with a diameter m the range 5.0 to 7.5 cm. On the contrary, lower values have been found for
samples having a diameter of 3 0 cm, showing that such parameters can influence the leach
test beyond a gaven size of the specimen Different volumes of the leachant (increasing volumes
of 250, 500, and 750 mL) do not affect the leach data at all (such as is expected in the case of
a diffusion-controlled phenomenon)
Finally, the presence of an air-entraining agent in the sample (about 0 1% by weight with
respect to cement) (Fig 10), the shorter curing times of the specimen (3 days, instead of 28
O ANSI-s § ANSI-m 9 ANSI-B

40* 9 , . . . . , i , i , i ! , i , |
3O
?o 2sl . / / -o-
9 o 9 . 9 o 9 , 9 i 9 , , i 9 o . i 9
0 10 20 30 40 50 60 70 80 90 100
TIME (days)
CopyrightFIG 9--Cumulattve
by ASTM fracttons
Int'l (all rights reserved); Mon Febof
24cesium released
23:07:03 EST 2020 from cylindrical samples of mcreasmg diameter
§ ANSI O ANSI-aer
3O
25-
cq
I 20"
o
X 15-
(3 10-
5-
0 9 , . , . , 9 , - , 9 T . , 9 , .
0 10 20 30 40 50 60 70 80 90 100
TIME (days)
FIG lO--Cumulattve fracttons of cesmm releasedfrom samples of pozzolantc cemented sulfates pre-
pared wzth or wtthout the addttton of an alr-entrammg agent
days) and the execution of the test in static conditions (by analyzing a small a m o u n t of
leachant at the time intervals specified in ANSI/ANS method), do not seem to be so important
in the case considered here.
Conclusions
The general conclusions are that the leaching of cesium from cemented BWR evaporator
concentrates is diffusion-controlled
Many parameters among the ones tested do not influence the release of cesium at all This
is true for initial concentration of the tracer, volume of the leachant, presence of an air-entrain-
Ing agent in the sample, curing time, and conditions of the test (static or dynamic) The diam-
eter ofcyhndrlcal specimens on its own can influence the leach test only beyond a given Inter-
val of values O n the basis of the results obtained, however, the dimensions of the sample
reported In both the ISO and ANSI methods could need to be revised On the contrary, the
release of the tracer is particularly influenced by a strong temperature Increase and by the
chemical composition of the leachant and its ionic strength. Much more influence, however,
is shown by the method used to prepare the mix, which, in general, can strongly modify the
final properties of cemented waste materials (mechanical strength, resistance to external
agents, and so on) Consequently the procedures followed to prepare the samples need a careful
control with the aim of obtaining reproducible data
References
[1] "Characterization of Radloactwe Waste Forms," Vol 1, K Brodersen and K Nllsson, Eds, Progress
Report for 1987, EUR 12077 EN 1989, Commission of the European Commumtles, Luxembourg
[2] ISO 6961 Long-Term Leach Testing of Sohdlfied Radloactwe Waste Forms, International Orgam-
zaUon for Standardization, Geneva, Swltz, 1982
[3] ANSI/ANS 16 1 Proposed American National Standard Measurement of the Leachabihty of Sohd-
ified Low-Level Radioactive Wastes, Amencan National Standards Institute, New York, NY, 1981
[4] Zamoram, E, Serum, G , and Blanchard, H, "Calcium Release from Cement Samples of Different
CopyrightSize,"
by ASTM Int'l (all rights
Cement andreserved);
Concrete Mon Research,
Feb 24 23:07:03 EST 2020
Vol 16, 1986, pp 394-398
[5] C6t6, L and Isabel, D, "Apphcatlon of Dynamic Leaching Test to Sohdlfied Hazardous Wastes,"
Hazardous and Industrial Waste Management 3rd Symposium, ASTM STP 856, American Sooety
[6] Colombo, P, et al, "Accelerated Leach Test(s) Program Annual Report," BNL 51955, Brookhaven
National Laboratory, Sept 1985
Larry IV. Jones, 1R. Mark Bricka, 2 and M. John Cullinane, Jr. 2
Effects of Selected Waste Constituents on

Solidified/Stabilized Waste Leachability
REFERENCE: Jones, L W , Brlcka, R M , and Culhnane, M J , J r , "Effects of Selected Waste
Constituents on Solidified/Stabilized Waste Leachability," Stabthzatton and Sohdlficatlon of
Hazardous, Radtoacttve, and Mtxed Wastes, 2nd Volume, ASTM STP 1123, T M Gilham and
C C Wiles, Eds, American Society for Testing and Materials, Philadelphia, 1992, pp 193-203
ABSTRACT: The effects of ten common waste constituents that have the potential to interfere
with the strength and contaminant lmmobihzatlon propertaes of sohdlfied/stabihzed metal fin-
lshlng wastes were studied A synthetic, metal finishing waste treatment sludge containing high
levels of cadmium, chromium, mercury, and nickel was prepared Interference materials, which
were individually added to the sludge dunng treatment, included three heavy metals (copper,
lead, and zinc as nitrate salts), sodium hydroxide (NaOH), sodmm sulfate (Na2SO4), and five
organic substances (grease, oil, trlchloroethene, hexachlorobenzene, and phenol) Solidification/
stablhzation was accomplished using one of three binder systems portland cement (CEM), port-
land cement plus fly ash (CFA), and a lime/fly ash (LFA) Specimens were tested in duplicate
using the U.S Environmental Protection Agency (EPA) extraction procedure
Leachate concentrations of the three metals in the sludge (cadmmm, chromium, and nickel)
and the three metal interference matenals (copper, lead, and zinc) were prlmanly related to the
final pH of the extraction solution Of the interference materials tested, the metal mtrates
reduced the leachate pH values to the greatest extent and thus had the greatest effect on the metal
concentrations in the leachate The addition of NaOH as an mterfenng material increased the
final pH of the leachate and thus tended to suppress leachate metal concentrations, especially in
the less buffered CFA specimens On the other hand, the addition of Na2SO4 reduced leachate
pH only slightly, again mainly in the CFA specimens, and thus tended to slightly increase the
metal concentration in the leachates The addition of up to 8% by weight of the organic interfer-
ence materials had no significant effect on metal concentrations in the leachates, nor on the
leachate pH Phenol, the most polar of the organic interference materials used in this study, was
itself extracted at appreciable levels from all three binder systems However, phenol addition had
little effect on metal leaching Leachate mercury concentrations were not affected by the addition
of the interference materials, except perhaps by phenol Nearly equal mercury concentrations
were found in leachates from the raw sludge and all treated samples
In general, the effects of the added interference matenals on the short-term leachate metal con-
centrations could be explained by differences in the leachate final pH None of the lnterfenng
materials appeared to act as carriers or to change the solubility of the metal contaminates other
than through their effects on the leachate pH This research confirms the need for waste-binder
specific studies before selecting a chemical stablhZatlon/sohdlfiCatlon process for the treatment
of hazardous waste.
KEY WORDS: sohdlfiCatlon, stabilization, solidification interference, stabilization interference,

leaching, heavy metal stabilization, metal finishing waste treatment
1 Professor, Waste Management Research and Education Institute, University of Tennessee, Knoxville,
TN 37996
2 Chemical engineer and research cavil engineer, respectively, U S Army Engineer Waterways Experi-
ment Station, Vicksburg, MS 39180
193

194 HAZARDOUSWASTES
Water
Waste )- Mixing ______•. Solidified/Stabilized

Product
l
Primary Binder Admixture
FIG 1--Schematw of typwalsohdlficatton/stabdlzatton process
The sohdification/stabihzatlon (S/S) of hazardous waste involves mixing the waste with a
binder material to enhance the physical properties of the waste and to immobilize contami-
nants that may be detrimental to the environment [1] Figure 1 is a schematic of the S/S pro-
cess Many hazardous wastes contain materials that are known to inhibit or interfere with the
setting and strength development properties of the cement and pozzolan binding agents com-
monly used in S/S processes Unfortunately, the effects of these materials on the contaminant-
containment properties of the matrix are not well documented
This paper presents an evaluation of the effects of ten materials that have the potential to
interfere with the strength and contaminant-containment properties of treated wastes. The
interference materials selected for study were three heavy metals (nitrate salts of copper, lead,
and zinc), sodium hydroxide (NaOH), sodium sulfate (Na2SO4), and five organic substances
(grease, oil, trichloroethene (TCE), hexachlorobenzene (HCB), and phenol) [2] Three binder
materials, portland cement (CEM), portland cement/fly ash (CFA), and hme/fly ash (LFA),
were used to solldlfy/stablhze a specially prepared sludge containing substantial concentra-
tions of four metals (cadmium, chromium, nickel, and mercury) The effects of these Interfer-
ing materials, at 2, 5, and 8% by weight, on contaminant leaching properties of the solidified/
stabilized sludge, were evaluated using the U S Environmental Protection Agency's (EPA)
extraction procedure (EP)

The project described by this paper was designed using the recommendations of a panel of
experts made up of representatives from academia, government and regulatory agencies, land-
fill operators, and solidification/stabilization vendors [3]. Project execution adhered to the
panel's recommendations as closely as possible
Synthetw Sludge Productton

The metal sludge was produced by hydrated lime precipitation of the metal hydroxides from
a solution of reagent-grade mercury nitrate [Hg(NO3)2 H20], chromium nitrate [Cr(NO3)3
9H20], cadmium nitrate [Cd(NO3)2 4H20], and nickel nitrate [NI(NO3)2 6H20]. The syn-
thetic metal solution was prepared by dissolving 23 1 g/L of Cr(NO3)3 9H20, 14 9 g/L of
NI(NO3)2.6H20, 1.6 g/L of Cd(NO3)2 4H20, and 0.02 g/L of [Hg(NOs)2 H20] in tap water
Copyright
A lime by ASTM Int'l (allof
dosage rights
20reserved); Mon Feb 24 23:07:03
g/L produced EST 2020
a sludge with optimal settling characteristics and a super-
JONES ET AL ON WASTE CONSTITUENTS 195
natant with the lowest turbidity This hme dosage was twice the calculated stolchiometnc
a m o u n t required for metal precipitation. The sludge contained approximately 9% sohds and
had a density of 1.08 g/mL and a pH of 11 5
The sludge was withdrawn from the setthng vessel and dewatered using a rotary drum vac-
u u m filter. The sohds content of the dewatered sludge ranged from 30 to 35% by weight. Before
use in the sohdlficatlon/stabdlzatlon stu&es, the dewatered sludge was diluted with tap water
to a constant 25% ( + 0 5%) by weight sohds content This procedure allowed ngld control of
the water-to-binder ratio m the sohdlfied/stabdlzed waste
Selection of Interference Chemlcals

Several categories of substances that might cause a decrease in strength, durability, and con-
stituent containment were selected for study [2]. Interfermg chemicals were added to freshly
sohdlfied/stablhzed sludge at 2, 5, and 8% by weight A control specimen, to which no inter-
fenng matenal was added (the 0% interference addmon speomen), was also prepared The
categories of interferences desired, the lnterfenng agent used, and the weight-correction factor
are listed in Table 1.
Sludge Treatment and Binder Selection

Three generic binder systems, representing the basic technology of the majority of com-
mercially available S/S processes, were selected for study [1] Binder-to-sludge ratios, which
were formulated to produce a 28-day unconfined compressive strength (UCS) of at least 100
psi, are shown m Table 2. One interference chemical was evaluated for all three binder systems
using a single batch of sludge The heavy metal sludge used m this study was consistent in
composmon and age and was free of u n k n o w n and unwanted contaminants that might have
effects upon the final product or test results.
Extractton Procedure (EP) Toxwlty Test

All specimens, including the control specimens, were subjected to the EP (USEPA Method
1310) after 28 days of curing at 98% humidity and 23 _+ I*C All specimens were ground to
pass a 9 5-mm sieve. Two separate specimens of each binder mixture at each interference con-
centration were tested The EP extracts were filtered and analyzed for cadmium, chromium,
mercury, nickel, the interference material of interest, and the final pH of the leachate
TABLE 1--Descrlptton of interference agents used m the study
Desired Interference Interference Reagent Weight Correction Factor~
Oll 30 weight motor od 1 00

Grease Axle grease 1 00
Phenol Phenol 1 00
Degreaser Tnchloroethytene (TCE) 1 00
Pestlode Hexachlorobenzene (HCB) l 00
Strong base Sodmm hydroxide (NaOH) 2 35
Sulfate Sodmm sulfate (Na2SO4) 1 48
Lead Lead mtrate 1 60
Copper Copper mtrate tnhydrate 3 80
Zinc Zinc mtrate trihydrate 4 55
a Amount of reagent to be added to get desired concentraUon of interference material, for example,
Copyright
1 6 g by
ofASTM
leadInt'l (all rights
mtrate is reserved);
neededMon to Feb
add24 l23:07:03 EST 2020
g of lead
TABLE 2--Bmder systems and rattos used m thts study
Binder Binder/Sludge Ratio Dilution Factors
Portland cement (Type I) 03 1 Cement sludge 1 3a 5 2b

Fly ash (Class F)/portland 02 1 Cement sludge and l7 68
cement (Type I) 05 1 Fly ash sludge
Lime/fly ash (Class C) 05 1 Fly ash sludge and 18 72
03 1 Lime sludge
a Weight of final mixture/weight of wet sludge
b Weight of final mixture/weight of dry sludge sohds
Standard quality control and assurance activities and protocols were followed throughout
the study. Statistical analyses were performed using the Statistical Analysis System [4].
Table 3 summarizes the control speomen leaching data for the four sludge contaminants
Table 3 also compares the total sludge concentration with the EP leachate concentraUons for
the raw sludge and specimens treated with all three reagent systems The EP leachate final pHs
for CEM and LFA were both quite basic (medians of 11 6 and 11 9), indicating an appreciable
a o d neutralization (buffenng) capacity m the specimens that offset EP acid addition The CFA
specimens, with a median final pH of 9 13, apparently had a lower buffenng capacity
For both cadmium and chromium, CEM and LFA gave median leachate concentrations
between 1 and 10 ppb due to the low metal solubility at the more basic pHs of these leachates
The CFA leachate concentrations of these metals from the control samples were about ten
times greater than from the other binder systems, again probably due to the more acidic pH
of the CFA leachates Median nickel concentrations were near 10 ppb for LFA leachates, but
both CEM and CFA leachates contained about l0 times higher nickel concentrations Mer-
cury concentrations in the leachates from treated specimens were near those ofleachates from
the raw waste for all three binder systems, indicating that these treatment methods did not
effectively contain mercury
Interference materials found at appreciable levels in the control leachates were sodium
(averaging 135 mg/L) and sulfate (40 mg/L), and phenol (found in all leachates at 30 to 60
mg/L). Lead concentrations were especially elevated in all CFA leachates, perhaps due to its
presence in the Class F fly ash
Comparison of metal concentrations in the EP leachates from the raw and treated sludge
indicates excellent contaminant immobilization, considering the high levels of metals con-
TABLE 3--Summary of total analysis and EP leachmg data for raw sludge and treated control
spectmens
Parameter Cadmium Chromium Mercury Nickel
Sludge concentrations, mg/kg dry wt 6330 0 3170 0 44 0 29 350 0

EP of raw sludge, mg/L leachate 57 9 242 0 0 840 149 0
EP of CEM control, ~ mg/L, final pH 11 6 0 0021 0 010 0 95 0 083
EP of CFA control, a mg/L, final pH 9 13 0 028 0 068 0 29 0 068
EP of LFA controlf mg/L, final pH 11 9 0 0009 0 007 0 69 0 009
a Median of (all
20 rights reserved); Mon Feb
determinations 24 23:07:03
from EST 2020 sludge batches
ten separate
talned in the sludge and raw sludge EP leachates. Metal concentrations in the EP leachates
were reduced by three to five orders of magnitude by all binders even though all specimens
were ground before the EP However, mercury concentrations in the EP leachates from the
treated sludge were about the same as those in the EP leachate from the raw sludge
Effects of EP Leachate Fmal p H on Sludge Metal Release

The primary effect of the different interferences on the leaching of sludge metals appears to
be tbelr impact on the final pH of the EP leachates This effect is especially true for the samples
solidified with CFA, which had a lower buffenng capacity than the CEM and LFA specimens
This was most clearly evldent for the metal ion interferences (copper, lead, and zinc) as seen
in Table 4 and illustrated in Fig. 2. The metals were added as their nitrate salts, which may
account for the pronounced reduction in the final pH of the EP leachate. Copper nitrate had
the greatest effect on the final pH Lead and zinc nitrates had similar but less pronounced
effects The dramatic effects of metal nitrate addition on the final leachate pH and sludge metal
leachate concentrations for the CFA specimens are shown in Fig 3
The inverse relationship between metal concentration and leachate final pH was consistent
for all of the interferences and binder systems and was especially significant for the leaching of
the sludge metals The relationship between final pH and the log of the cadmium concentra-
tion in the leaching fluid is illustrated for all samples from all interferences and binder systems
in Fig. 4. The amphotenc nature of cadmium solubility is evident in the plot as cadmium
concentrations increase as the leachate final pH rises above about 10 5 The concentration of
cadmium leached can be predicted very closely from the final pH of the EP leaching solution,
regardless of the interference material added or binder reagents used
As illustrated in Fig 5, leachate nickel concentrations followed similar, but less pronounced,
dependence on leachate final pH Leachates contained fairly constant nickel concentrations
above a pH of about 7 5 The nickel concentrations increased about two orders of magnitude
at pHs below about 7.5 Deviations from this pattern may be caused by differences in the
effects of the individual interferences interacting with the nickel in solution, but no specific
Interactions were found
As shown in Fig 6, chromium concentrations also tended to increase as the pH of the leach-
TABLE 4--Summary of EP leaching data for each binder averaged over the metal mterferents only
(copper, lead, zmc)
Concentratmn m Leachate, mg/L
Final Moisture,
Bmde~ Percentage pH 28-day % Cadmium Chrommm Mercury Nickel
CEM 0% 11 5 28 8 0 0014 0 0065 0 620 0 0367

CEM 2% 11 1 28 0 0 0006 0 0173 0 598 0 0388
CEM 5% 10 2 27 0 0 0121 0 0553 0 368 0 0458
CEM 8% 10 3 29 9 0 0362 0 0313 0 0466 0 0463
CFA 0% 93 23 1 0 0204 0 0622 0 0163 0 0383
CFA 2% 94 21 5 00712 00872 00522 00318
CFA 5% 71 26 4 32 8 0 0245 0 0239 1 74
CFA 8% 64 26 1 41 8 0 2912 0 245 3 22
LFA 0% 11 7 29 2 0 0005 0 0046 0 762 0 0285
LFA 2% 11 4 18 9 0 0008 0 0095 0 901 0 0295
LFA 5% 10 9 22 1 0 0065 0 0133 0 602 0 0357
LFA 8% 10 0 19 2 0 0049 0 0255 0 412 0 0362
a CEM
Copyright = Int'l
by ASTM portland
(all rightscement, CFA
reserved); Mon = 23:07:03
Feb 24 portland cement plus fly ash, LFA = hme/fly ash
EST 2020
EFFECTS OF INTERFERENCE ON LEACHATE pH

METAL INTERFERENTS ONLY
12-
11
z
w 10
..r
<
U.I
<
z 8-
,T
7-
T "1 r
0% 20 5"/0 8%
P E R C E N T METAL I N T E R F E R E N C E
FIG. 2--Effects of metal interference materials on EP leachate final pH.
ate decreased, but much less dramatically than for the other metals. As shown on Fig. 7, mer-
cury concentrations in the leachates varied considerably in different leachates but showed little
or no apparent pH dependence.
As expected, metal interference concentrations in the EP leachates also increased as their
level in the waste product increased. Copper and zinc interference concentrations were held
to low levels in the CEM and LFA samples' leachates, but were leached at very high levels from
LEACHING BY METAL INTERFERENT CONC.

FOR CEMENT/FLYASH BINDER SAMPLES
O)
E 100: -9.5
z
_o -9
n" 10 3::
I.- -8.5 n
Z LU
LU I--
<
o -8 T
z (..)
O <
L) ~.5 L4
_J
U.I
0.1
-7 Z
E
W
pH 6.5
T
< 0.01
W
..J
P E R C E N T METAL NITRATE INTERFERENCE
--e- CADMIUM i CHROMIUM [] NICKEL )( pH

FIG. 3--Leachate metal concentration and final pH as a function of metal interference material and
percent
Copyright addition
by ASTM Int'l (allfor CFA
rights specimens.
J O N E S ET AL. ON W A S T E C O N S T I T U E N T S 199
EP LEACHATE NI CONC. BY pH FOR

ALL SAMPLES
10=
[]
[]
133 []
[]
[]
Z
O
rr []
I'--
zUJ 0.1
(.3
Z []
[]D
o []
o []
,_1 []
w 0.01 [][]
_o []
Z [2 [] []
[] []
[]
0.001
5 e ~ 9 i'o 1'1 1'2 la
EP LEACHATE pH
FIG. 4--Summary of leachate cadmium concentrations as a function of pH for all specimens.
the CFA samples. Both copper and zinc concentrations exceeded 1000 mg/L in leachates from
the CFA specimens prepared using 8% by weight interference addition.
Strong Base and Sulfate Interference on Leaching

Sodium hydroxide generally affected the final leachate in an inverse manner compared to
the metal nitrates in that the higher levels of addition raised the leachate final pH (Table 5).
The higher final pH is especially noticeable in the CFA samples with low acid-neutralization-
capacity, where it results in lower cadmium and nickel concentrations in the leachates. The
EP LEACHATE CD CONC. BY FINAL pH FOR

ALL SAMPLES
100=
,-....
E 1(]
Z
I-- 1
[]
[]
I-- [] [ ]
Z
'" 0.1
o
Z Eli
O
o 0.01
[]
a< 0.001 []
o []
L~
0.0001
5 "7 9 1'0 1'1 1'2 13
EP LEACHATEpH
FIG.Int'l
Copyright by ASTM 5--Summary ofMon
(all rights reserved); leachate nickelEST
Feb 24 23:07:03 concentrations
2020 as a function ofpH for all specimens.
T0P
EP LEACHATE CR CONC. BY pH FOR
ALL SAMPLES
10=
O3
E
Z
o I
[]
[] []
rr
I-- []
Z
ILl
~o~ []
0 0.1 [] J~ooS o~ [] [] []
Z 000
0
0 [] [] []
[] []
0.01 [] [] []
on -
.1-
0 []
0.001
EP LEACHATE pH
FIG. 6--Summary ofleachate chromium concentrations as a function ofpH for all specimens.
high levels of sodium in the leachates reflects the NaOH additions and the generally high sol-
ubility of sodium salts.
Sodium sulfate addition did not substantially affect the concentrations of metals in the EP
leachates. The most significant effect of Na2SO4 addition is its own high concentration in the
leachates from all three binder systems.
Effects of Organic Interference on EP Leaching

The nonpolar, organic interferences (those with low water solubilities, that is, grease, oil,
TCE, and HCB) had no significant effect on the final pH or the metal concentrations in the
EP LEACHATE HG CONC. BY pH FOR

ALL SAMPLES
10:
[] E3
Z [] []
o [] []
1
tr
i,- [] o d~ o9
z
ILl
o []
[]
[] [] [] []
Z
o
o 0.1: [] []
>.-
[]
r.r
[]
O
tr
LU
0.01
5 # ~ 6 1'0 1'1 1~ 13
EP LEACHATE pH
FIG. 7--Summary
Downloaded/printed by of leachate mercury concentrations as a function of pH for all specimens.
TABLE 5--Summary of EP leachate concentratzonsfrom sodmm hydroxide and sulfate tnterferences,

averaged over all bmder systems
EP Leachate Concentration, mg/L

Moisture, Final
Interference Percentage a 28-day, % pH Cadmium Chromium Mercury Nickel Interference
NaOH 0% 0 348 10 49 0 0412 0 087 0 258 0 067 133 0 b

NaOH 2% 0 360 10 59 0 0305 0 065 0 210 0 051 662 0
NaOH 5% 0 363 10 94 0 0044 0 097 0 243 0 037 1281 7
NaOH 8% 0.358 11 39 0 0017 0 273 0 264 0 026 1761 7
Sulfate 0% 0 200 10 90 0 0217 0 018 0 724 0 005 40 6 ~
Sulfate 2% 0.224 10 78 0 0323 0 038 0 581 0 004 475 0
Sulfate 5% 0 236 10 38 0 4920 0 105 0 528 0 014 1074 3
Sulfate 8% 0 247 10 04 0 3234 0 161 0 236 0 011 1643 3
a Percent of interference matenal added

b As s o d i u m
As sulfate
E P leachates ( T a b l e 6) Averages f r o m the grease, off, HCB, a n d T C E s a m p l e leachates follow

closely t h e results o f t h e c o n t r o l s a l o n e In a d d m o n , o n l y T C E was f o u n d a b o v e t h e d e t e c t i o n
hm~ts, e v e n f r o m t h e s a m p l e s w~th a n 8% by weight a d d m o n
A d d m o n o f t h e m o r e w a t e r soluble p h e n o l also d~d n o t effect leachate c a d m i u m , c h r o m i u m ,
a n d m c k e l c o n c e n t r a t i o n s H o w e v e r , leachates c o n t a i n i n g p h e n o l a p p e a r e d to h a v e higher
m e r c u r y levels t h a n t h o s e o f t h e o t h e r organles (Table 7) P h e n o l was detected at very high
levels, b e t w e e n 0.2 a n d 0 3%, m leachates f r o m all t h r e e b i n d e r s w i t h 8% p h e n o l a d d m o n s
P h e n o l has a p r o n o u n c e d effect o n S/S p r o d u c t s t r e n g t h a n d d u r a b l h t y [3], b u t a p p e a r s to h a v e
o n l y a m i n o r effect o n t h e leaching o f sludge c o n s t i t u e n t s P h e n o l itself was n o t effectwely
c o n t a i n e d b y a n y o f t h e b i n d e r systems
N o c o n s i s t e n t r e l a t i o n s h i p b e t w e e n m o i s t u r e c o n t e n t a n d leachate c o n c e n t r a U o n s o f a n y o f
t h e sludge m e t a l s or interferences was f o u n d
TABLE 6--Average leachate concentrattonfrom all organws except phenol by binder system a
EP Concentration in Leachate, mg/L

Moisture,
Binderb Percentage r 28-day % pH Cadmium Chromium Mercury Nickel
CEM 0% 0331 11 66 00112 0019 0446 0099

CEM 2% 0 278 11 71 0 0087 0 019 0 828 0 092
CEM 5% 0 271 11 71 0 0086 0 015 0 649 0 094
CEM 8% 0 263 11 70 0 0094 0 011 0 0590 0 093
CFA 0% 0 347 08 91 0 3310 0 782 1 306 0 105
CFA 20/o 0 334 09 01 0 0835 0 086 0 691 0 071
CFA 5% 0 320 08 89 0 1403 0 122 0 540 0 073
CFA 8% 0 311 09 09 0 0924 0 083 0 688 0 074
LFA 0% 0 218 11 89 0 0020 0 008 0 429 0 030
LFA 2% 0217 11 95 00051 0023 0510 0047
LFA 5% 0 220 11 89 0 0055 0 006 0 364 0 034
LFA 8% 0 215 11 89 0 0030 0 009 0 365 0 035
a Grease, o11, tncholorethylene, and hexachlorobenzene

Copyright by ASTM
b CEM = Int'l (all rights reserved);
portland cement,Mon Feb 24=23:07:03
CFA EST 2020
portland cement plus fly ash, LFA = lime/fly ash
c Percent of interference matenal added
202 HAZARDOUSWASTES
TABLE 7--EP leachate characterlsttcs from spectmens with phenol mterference
Concentration xn Leachate, mg/L

Final Moisture,
Binder a Percentage b pH 28-day % Cadmmm Chrommm Mercury Nickel Phenol
CEM 0% 11 70 0 507 0 0013 0 008 2 140 0 070 34

CEM 2% 11 56 0 542 0 0014 0 010 6 580 0 064 746
CEM 5% 11 59 0509 00059 0011 6560 0075 1570
CEM 8% 11 24 0503 00023 0011 7 175 0073 2710
CFA 0% 9 06 0 378 0 0190 0 139 2 540 0 075 65
CFA 2% 9 50 0 414 0 0066 0 062 7 230 0 074 573
CFA 5% 9 57 0 398 0 0085 0 033 3 755 0 073 1310
CFA 8% 9 48 0 414 0 0123 0 031 3 860 0 077 2350
LFA 0% 12 05 NAc 0 0050 0 006 1 660 0 008 30
LFA 2% 12 22 NA 0 0003 0 006 3 920 0 006 676
LFA 5% 12 17 NA 0 0003 0 006 2 290 0 005 1640
LFA 8% 12 06 NA 0 0001 0 007 2 705 0 010 2650
a CEM = portland cement, CFA = portland cement plus fly ash, LFA = hme/fly ash
b Percentage of interference material added
c Moisture contents were not determined
Conclusions
In general, the differences in the leachate concentrations of cadmium, chromium, and
nickel for the control and interference containing specimens can be largely explained by dif-
ferences in the final pH of the EP leachate As reported elsewhere, this contrasts to the pro-
nounced effects of the interfenng materials on strength and durability [3] None of the inter-
feting materials appeared to act as careers nor to change the solubility or mobility of the sludge
metals
Leachate mercury concentrations were not significantly affected by leachate pH Nearly
equal mercury concentrations were found in leachates from all treated samples as well as the
raw sludge. None of the treatment processes and none of the interference matenals, except
perhaps phenol, appeared to affect the extraction of mercury
The highly water-soluble interferences (phenol, sodium, and sulfates) were extracted from
specimens prepared with all three binder systems and in direct proportion to tbe~r addition
rates All three binder systems were equally ineffective in containing very soluble materials
This research confirms the need for waste-binder specific studies prior to the selection of a
chemical S/S process for the treatment of hazardous waste
Acknowledgment
The tests descnbed and the resulting information presented herein, unless otherwise noted,
were obtained from research conducted by the U S. Army Engineer Waterways Experiment
Station, Vicksburg, MS The study was sponsored by EPA's Risk Reduction Engmeenng Lab-
oratory, Cincinnati, OH. Mr. Carlton Wiles was the USEPA project manager Permission to
publish this information was granted by the Chief of Engineers
References
[1] CullInane, M J , Jr, Jones, L W, and Malone, P G , "Handbook for Stabxhzatlon/SohdlfiCatlon of
Hazardous Waste," EPA/540/2-86-001, Hazardous Waste Enganeenng Research Laboratory, Office
Copyright of
by ASTM Int'l (alland
Research rightsDevelopment,
reserved); Mon Feb 24
U23:07:03 EST 2020
S Environmental Protection Agency, Cincinnati, OH, 1986
[2] Jones, L W, "Interference Mechanisms in Waste Sohdlfication/Stablhzatlon Processes Literature

Review," Interim Report Interagency Agreement DW219306080-01-0 between Waterways Experi-
ment Station, U S Army Corps of Engineers, Vicksburg, MS, and Risk Reduction EnglneenngLab-
oratory, U S EnvironmentalProtection Agency, Cincinnati,OH, 1988
[3] Bncka, R M and Jones, L W, "An Evaluation of Factors Affectingthe Stabilization/Solidification
of a Heavy Metal Sludge," final report, Interagency Agreement DW930146-01-01 between Environ-
mental Laboratory, Waterways Experiment Station, U S Army Corps of Engineers, and Risk Reduc-
tion Engmeenng Laboratory, U S EnvironmentalProtection Agency, Cincinnati,OH, 1990
[4] SAS User's Gutde Statzstlcs, Version 5 edition, SAS Institute, Cary, NC, 1987
JaeoOwan Lee, 1 Kyong- Won Han, 1 and Leo P. Buckle): 2
Short-Term Leaching Behavior of Co-60, Sr-

85, and Cs-137 from Cemented Ion-Exchange
Resin Waste Forms
REFERENCE: Lee, J -O, Han, K -W, and Buckley, L P , "Short-Term Leaching Behavior of
Co-60, Sr-85, and Cs-137 from Cemented Ion-Exchange Resin Waste Forms," Stabthzatton and
Sohdtficauon of Hazardous, Radloacttve, and Mtxed Wastes, 2nd Volume, ASTM STP 1123, T
M Gllham and C C Wiles, Eds, Amencan Society for Testing and Matenals, Philadelphia,
1992, pp 204-216
ABSTRACT. The leaching characteristics of Co-60, Sr-85, and Cs-137 from cemented ran-
exchange resin waste forms containing water-extend~ble polyester resin and Zeolon 900-sodmm
addmves were investigated using three d~fferent types of leach tests, that is, the American Nuclear
Society (ANS) 16 1, the Battelle Matenals Characteristics Centre (MCC) MCC-4S and modified
MCC-4S methods Specimens used m the leach tests were fabricated with cement and 1on-
exchange bead resins present m the LI+-OH - form The ran-exchange resins were doped w~th
Co-60, Sr-85, and Cs-137 to simulate the spent resin waste from nuclear power plants
The presence of Zeolon 900-sodium addmve to the waste form formulation prowded d~s-
tlnctly improved adsorptmn for Cs- 137 and seemed not to reduce effectwely the releases of Co-
60 and Sr-85 The leaching pr~ ~ the three radI~ fr~ the waste f~ were d~
mated by diffusion The leaching rates of Co-60 and Sr-85 were both apprecmbly affected by the
leachant flow rate, while Cs-137 was not apparently sensmve The presence of nonsorbmg back-
fill material was hkely to induce the radlonuchdes to follow longer transport paths than for spec-
imens only m the presence of water The chnoptdohte selected as a sorbmg backfill material did
not retard the Co-60 release, but there was strong absorptmn of Sr-85 and Cs-137 onto the
chnoptllohte
KEY WORDS: leaching, cemented ton-exchange resin waste form, addltwe (Zeolon 900-Na),
ANS 16 1 test, diffusion process, waste disposal, flow-around leach tests, low flow rate, backfill
material (chnoptllohte)
A waste f o r m c o n t a i n i n g r a d i o n u c l l d e s i n c o r p o r a t e d in a s o h d m a t e r i a l is c o n s i d e r e d t h e
initial b a r r i e r in a disposal system It b e c o m e s t h e source for r a d l o n u c h d e s to m i g r a t e into the
s u r r o u n d i n g e n v i r o n m e n t E s t i m a t i o n o f t h e q u a n t i t i e s o f r a d l o n u c h d e s released f r o m waste
f o r m s is t h u s a critical c o m p o n e n t in d e v e l o p i n g safety assessments for t h e disposal o f radio-
active wastes.
T h e release e s t i m a t i o n o f r a d l o n u c h d e s f r o m waste f o r m s requires i n f o r m a t i o n o n leaching
m e c h a n i s m s u n d e r ~r disposal c o n d m o n s a n d a m o d e l to predict r a d i o n u c h d e releases
based on the mechamsms
T h i s study, in t h e a b o v e c o n n e c t i o n , selected c e m e n t e d i o n - e x c h a n g e resin p r o d u c t s as the
l Senior researcher and head of Radwaste Disposal Dept, respectively, Korea Atomic Energy Research
Institute, P O Box 7, Daeduk-Danj1, Daejon, Korea
2 Soentlst of Waste Management Technology Division, Atomic Energy of Canada Ltd, Chalk River,
Ontario, Canada
204

LEE ET AL ON LEACHING BEHAVIOR 205
waste form of interest because ion-exchange resins are one of the most important wastes in
terms of both the activity and quantities of low- and intermediate-level wastes from nuclear
power plants Also, little is known about the leaching characteristics of radionuchdes from
immobilized ion-exchange resin products
For the purpose of improving upon the information available, the leaching behavior ofCo-
60, Sr-85, and Cs-137 from cemented ion-exchange resin waste forms with water-extendible
polyester resin (WEP) and Zeolon 900-sodium was investigated in a possible disposal envi-
ronment using three types of leach tests, that is, the American Nuclear Society (ANS) 16 1 test,
the Battelle Materials Characteristics Centre (MCC) MCC-4S test, and a modified leach test
to the standard MCC-4S method which incorporated backfill material around a waste form
In the leach tests, the following conditions were initially assumed
1 The concentration ofradionuchdes in the hquld phase in contact with the surface of the
waste form is zero,
2 The leachant flow rate is maintained low enough to approximate water movement m a
waste repository
3 The presence ofbackfill material surrounding the waste form alters the releases observed
The experimental results obtained during short-term leaching periods were examined,
focusing on (1) leaching characteristics of Co-60, Sr-85, and Cs- 137 under the leachant con-
dition that assumes the leachate has an infinite dilution for the radlonuchdes, (2) the effect of
low flow rate and backfill material on leaching behavior, and (3) the controlling mechanisms
of leaching under such conditions
Mechanistic Consideration in Radionuclide Release

The leaching of radlonuclides from a waste form to the surrounding leachant IS acknowl-
edged to proceed via chemical interaction between the waste form containing radlonuchdes
and the leachant and mass transport mechanisms of the mobilized radIonuchdes
The chemical interaction involves the mobilization of radlonuchdes incorporated in the
waste form matrix and the sorptlon of the radlonuchdes mobilized on solidifying agent/addi-
tives in the process of their mass transport The Interaction can be instantaneous (that is, the
radlonuchde distribution is described using equilibrium chemistry) or a rate-limited reaction,
which is described by kinetic expression
The mass transfer ofradionuclldes dissolved into the leachant takes place in terms of at least
three mechanisms (1) convection, (2) diffusion through pores within a waste form, and (3)
diffusion at the interface of the waste form and the surrounding leachant
Convection refers to the bulk movement of leachant through the matrix It IS generally
assumed that the mechanism can be negligible in cement-based waste forms However, if there
are some physical changes such as swelling, shrinking, or cracking in the waste form of con-
cern, ItS role in the release of radionuclldes becomes important because the convection
decreases the positive role of the waste form as an initial b a m e r against radionuchde release
Pore diffusion is induced by concentration gradients within a waste matrix. Many Investi-
gators have suggested that the diffusion process can be expected to be the most Important
mechanism that controls the release of radIonuchdes since the cemented waste form may be
described as a porous solid with water-filled pores [1-4]
Interface diffusion subsequent to the pore diffusion takes place between the waste form sur-
face and the surrounding leachant This diffusion is subject to mass transfer resistance through
a film layer formed at the interface as a result of the different chemical potential ofradlonu-
clldes on the waste form surface and in the leachant flowing around the waste form The film
layer is mainly affected by a hydraulic regime of the flowing leachant since it establishes con-
centratlon gradients around the waste form
The relative importance of the above leaching mechanisms depends on the waste form prop-
erties and the disposal environment within the repository where the waste forms will be placed.
The waste form properties affecting leaching involve chemical composition, porosity, crack-
ing, durability, and surface condition They are determined by waste type, solidifying agent
and additives, and their dosage For the cemented ion-exchange resin waste form in this study,
the WEP and Zeolon 900-sodium additives may cause physical and/or chemical changes to
the waste form product and in consequence influence the release of radlonuchdes
The release of radlonuchdes from disposed waste may also dep,~nd on the disposal environ-
ment, including the disposal system and the chemistry and flow of groundwater through the
repository The disposal system in the study is assumed to consist of the waste form and backfill
without a container, adding additional conservatism to the safety assessment Groundwater
chemistry is not considered here since deIonlzed water is used in place of actual groundwater
The flow rate of groundwater is expected to be in the range of 0 5 to 500 m/year as cited in the
literature [5] This flow rate IS so low that a film layer may form at the interface of waste form
and provide mass transfer resistance to the release of radionuchdes If a waste form is sur-
rounded by backfill material, the abdity of this backfill to bind leached radlonuchdes chemi-
cally, in addition to its Influence on the hydraulic regime of leachant, may alter the boundary
condition between waste form and leachant to reduce the mass transport of radionuchdes
For the development of a suitable model to predict the release of radlonuchdes, it is essential
to understand from leach tests how the waste form properties and disposal environment affect
the leaching of radionuchdes from waste forms
Experimental
Specimen Preparation
Specimens used for the leach tests were fabricated with portland Type III cement and the
LI+-OH - form of Rohm and Haas Amberhte IRN- 154 bead resin, together with additives such
as WEP and Zeolon 900-sodium to prevent swelling and disintegration upon immersion of
the product in water and to reduce Cs-137 release, respectively The ion-exchange resins were
doped with radlonuchdes (Co-60, Sr-85, and Cs-137) to simulate the spent resin waste from
nuclear power plants Each specimen had a cylindrical shape, the dimension of which mea-
sured 2 53 cm in diameter and 5 80 to 6 40 cm in height
Test specimen preparation was conducted according to the formulation in Table 1 The rec-
ipe is an optimum suggested by Buckley [6] All specimens were cured in a humid atmosphere
for seven days
Leach Tests
The following three types of leach tests were conducted to investigate the leaching charac-
tenstics of the radlonuchdes, Co-60, Sr-85, and Cs-137, under various conditions from
cemented ion-exchange resin waste forms containing WEP and Zeolon 900-sodium additives
ANS 16 1 Method--The ANS 16 1 test is Intended to serve as a basis for indexing radio-
nuclide release from solidified low-level radioactive waste forms in a short-term test under con-
trolled conditions in a well-defined leachant This standard assumes that internal bulk diffu-
sion is the most likely rate-determining mechanism during the initial phases of the leaching
process, and that the concentration of the species being leached is zero at the surface of the
waste form byafter leaching commences In this test, the leachate is sampled and replaced at des-
Downloaded/printed
TABLE 1--Description of the formulation of cemented ton exchange resm specimens, umts m wetght
percent a
Dry IRN- 154 Water-Extendible H20 with
Resin Polyester Resin Isotopesb Cement Zeolon 900-Na
110 37 346 432 75

a Weight ratms (1) H20/dry IRN-154 --- 1 67, (2) H20/dry Zeolon 900-Na = 0 32, (3) H20/cement
= 0 32, and (4) WEP/dry IRN-154 = 0 33
b The total amount of water includes that absorbed by dry IRN-154 resin, by dry Zeolon, and for the
hydratmn of cement
lgnated time Intervals (2, 7, 24, 48, 72, 96, 120, 456, 1128, and 2136 h) The ratio of the
leachant volume to the specimen external geometric surface area is maintained within fixed
bounds (10 + 0 2 cm) d u n n g the leaching interval The method utilizes periodic leachant
replacement to simulate the conditions of a continuous-flow system
M C C - 4 S M e t h o d - - T h e MCC-4S method is a continuous, low flow test which, in compar-
ison with ANS 16 1, ehmmates the effects of the sampling interval and of large arbitrary
changes in the leachate concentration, providing better control over the boundary conditions
of the leachant In this study, attempts were made with the MCC-4S method to simulate the
flowing conditions of a real repository and thus to investigate how a low flow rate affects radio-
nuclide release from cemented 1on-exchange resin waste forms containing WEP and Zeolon
900-sodium additives Flow rates in this study were maintained at an average of 0 04 m L / m i n
using a peristaltic pump The samphng was conducted at time intervals of 2, 4, 8, 24, 48, 72,
96, 120, 144, 168, 336, 672, 1008, 1344, 1680, 2018, and 2352 h
M o d l f i e d M C C - 4 S M e t h o d - - I f a waste form is surrounded by backfill material, the hydrau-
hc regime induced by backfill material and the matenal's ability to bind leached radionuclldes
chemically should affect the mass transport of radionuchdes from a waste form To confirm
this hypothesis some modifications of the MCC-4S test were made by including in the test
program backfill matenals such as nonsorblng glass beads or a known adsorbing material, ch-
noptilohte, around the waste form The flow rates were again maintained at an average of 0 04
m L / m m , and the sampling was taken according to the time schedule of the MCC-4S test
The leachant used in the leach tests was deminerahzed water with an electrical conductivity
of less than 5 ~mhos/cm at 25~ and a total organic carbon content of less than 3 ppm The
temperature of the leachant was maintained in the range of 17 5 to 27 5~ d u n n g the tests
The average temperature was 25~ Each leach test was performed in tnplIcate to improve the
reliability of data obtained from the experiment Activity levels of Co-60, Sr-85, and Cs-137
in the leachates were determined by gamma ray spectroscopy using an lntnnsic germanium
detector
The experimental data obtained in triplicate were arithmetically averaged, and the results
were expressed as a plot of cumulative fraction of each specific radionucllde leached from the
waste form as a function of the total time of leaching

Leaching Behavtor o f Co-60, Sr-85, and Cs-137 under the Leachant Condmon M a m t a m e d
at Infimte Dtlutlon
The ANS 16 1 test was conducted to investigate the leaching characteristics of Co-60, Sr-85,
and Cs- 137 from cemented ion exchange resin waste forms with WEP and Zeolon 900-sodium
as addmvesbyunder the leachate condition of Infinite dilution
Downloaded/printed
208 HAZARDOUSWASTES
Figure 1 displays the cumulative fractions leached of the three radionuclides as a function
of the total leaching time. The triplicate specimens for each radionuclide gave nearly identical
release data. Typically, the variational range of the data obtained for Sr-85 was plotted while
the data for Cs-137 and Co-60 were plotted with arithmetically averaged values.
As shown in this figure, Cs- 137 release was reduced by a large extent in comparison with the
results of ordinary cemented ion-exchange resin products having no additives. The loss of Cs-
137 was only 1.4% after 25 days, and after 90 days had risen only to 4.5%, while a previous
investigation [6] indicated that about 94% of the Cs-137 had been released after 25 days from
the specimen without the additives. Sr-85 showed an 18% release of the total activity at 90
days. The release of Co-60 is, on the whole, small relative to Cs-137 and Sr-85 releases. Only
0.6% of the total activity was released in 90 days, and most of the initial activity came out
within ten days. The rapid loss of Co-60 at the early stage of leaching is probably because the
Co-60 activity released from these specimens was close to the surface and was readily washed
off.
For ordinary cemented ion-exchange resin waste forms, Cs- 137 had rapid release, Sr-85 was
leached substantially less readily, and the Co-60 leaching rate was the lowest among the three
radionuclides [ 7-9]. The leach rates of the three radionuclides from the improved waste form
used in this study were ranked according to the following order: Sr-85 > Cs- 137 > Co-60.
This suggests that the Zeolon 900-sodium used as an additive plays an important role in
reducing the release of the radionuclide, Cs- 137, from the waste forms. That is, the cesium was
tightly adsorbed and retarded by the Zeolon 900-sodium additive. On the other hand, Zeolon
900-sodium additive did not seem to reduce the Sr-85 and Co-60 releases effectively.
The addition of the WEP was also expected to lower the release of the radionuclides. How-
ever, the resin pretreatment with the WEP was shown not to alter the release appreciably,
although the WEP barrier physically reduced water diffusion through the porous product
[6]. The major contribution of WEP is that the product remains intact, preventing exposure
of additional surface, compared to the swelling and cracking that take place in untreated
cement ion-exchange resin waste forms.
o.o ] F o.o
z~ C o - 6 0
] Iz S r - 8 5 ]- L
o -
0.06 0.06 m
0.04 0.04
g
2 CL
Lu_
X
0.02 0.02
J
0.00 0.00
0.0 50.0 100.0 150.0 200.0
Leaching Time (Days)
Copyright byFIG. 1- -(all
ASTM Int'l The leaching
rights reserved); behavior of Co-dO,
Mon Feb 24 23:07:03 Sr-85, and Cs- 137 under A NS 16.1 test conditions.
EST 2020
This study employed the ANS 16 1 method for analyzing leaching behavior The leach rate
observed with a uniform, regularly shaped waste form is controlled by diffusion, and if less
than 20% of a leachable species is leached from the waste form, ~ts leaching behawor approx-
imates that o f a semi-mfimte medium In a semi-infinite waste form, the amount of a diffusing
radmnuchde A, per unit surface area dunng t~me t is given by
A, = 2Co(Dt/Tr) 'n (1)
where
Co = the mmal concentration of a diffusing radlonucllde, and
D = the diffusion coefficient (cm2/day)
Thus, the cumulative fraction leached, F, from a specimen having volume V (cm3), surface
area S (cm2), and the initial amount of a diffusing radlonuchde Ao is gwen by
2S(Ao/ V)( Dt/Tr) '/2

F = A,S]Ao = = 2(S/V)(Dt/Tr) 'n (2)
Ao
If the diffusion coefficient remains unchanged, the cumulanve fraction leached is hnear with
respect to the square root of the leaching t~me, w~th a slope
m = 2(S/V)(DDr) '/2 (3)
The value of D can be calculated from the slope of the hnear relationship between F a n d k/t
Figures 2 and 3 show the cumulative fractions leached for Co-60, Sr-85, and Cs- 137 versus
the square root leaching time. As seen m the F~g 2, the leaching curve for the Cs-137 was
0 020
E
O
"~ 0 015
-ID
.l:E
O
ID
9
c 0010
o
0
-> o 0 0 5
ID
5
E
(D
0 0 0 0 . . . . . . . . . , . . . . . . . . . i .........
O0 50 100 150
( L e a c h i n g Tqme ( d a y s ) ) * * O 5
Copyright by ASTM
FIG Int'l (all rightsleaching
2--The reserved); Mon Feb 24 23:07:03
behavtor of Cs-13EST 7
2020
plotted as a functton of square root leachmg ttme
0.08 0.08 oc
Iz, Co-60 I __ c3
&
,--4-
E --'
~" 0.06 0.06 m

no
r-
0.04- 0.04
g
2 o_
IJ_
x
.-> 0.02 0.02 -4

~6 :.4
-5 m
E I
8
9
0.00 ......... i ......... i ......... 0.00 "-1
0.0 5.0 10.0 15.0
(Leoching Time (days))**0.5
FIG. 3--The leaching behavior of Co-dO and Sr-85 plotted as afunction of square root leaching time.
observed to be linear between F and k/t. The leaching process for the cesium used here can be
reduced to the simple form
F - - mV/t (4)
where m is the slope of the linear relation defined in Eq 3. The diffusion coefficient was cal-
culated to be 1.04 • 10 -6 (cmZ/day). For the Co-60 and Sr-85 in Fig. 3, on the other hand, a
straight line was not obtained, possibly because the removal up to about 20 days was more
likely to be by washofffrom the surface of the specimen. The leaching curves could be regarded
as satisfying the requirement of Eq 2, that is, the linearity relationship between F and k/t,
except for the initial period. Thus, the leaching processes for cobalt and strontium can be sim-
ulated by
F= mk/t+a (5)
where a, the intercept of the F a x i s at t = 0, is the representation of surface radionuclides which

are readily washed off the specimens. This term depends on both the rate and length of the
initial leaching process. For the linear portion of the leaching curve, the diffusion coefficients
of the Co-60 and Sr-85 were 2.30 X 10 -9 and 1.28 X 10 -5 (cm2/day), respectively.
The Effect o f Low Flow Rate on the Leaching Rate

If the leachant around a waste form has a high flow rate, then the concentration of radio-
nuclides leached at the surface of the waste forms may be regarded as approximately zero. The
ANS 16.1 leach test has been widely used to measure the leaching rate of radionuclides from
waste forms under such a condition. However, groundwater flow in an actual underground
repository is expected to occur in a low range. To more accurately estimate leaching behavior
LEE ET AL. ON LEACHING BEHAVIOR 211
under realistic conditions, it is necessary to investigate the effect of the low flow rate on the
leaching rate.
Figures 4 and 5 represent the comparison of the results of ANS 16.1 test with those of an
MCC-4S test with a flow rate of 0.04 m L / m i n for Co-60, Sr-85, and Cs-137, respectively. It
was found in these figures that the leach rates of Co-60 and Sr-85 were influenced to an appre-
ciable extent by the leachant flow rate, while Cs-137 was not sensitive to the flow rate of
leachant. This appears to suggest that diffusion resistance through a film layer formed around
the surface of a waste form should be larger for strontium and cobalt than cesium. That is, the
decrease ofleachant flow rate increases the concentrations of Co-60 and Sr-85 at the interface
between leachant and waste form, and thus the accumulation of such radionuclide concentra-
tions may not only impede the diffusion of radionuclides from the waste form but may also
result in a solubility limitation for Co-60 and Sr-85 in the leachant.
Cobalt-60 and Sr-85 exhibit different behaviors with time when one examines the effect of
leachant flow rate on the leaching rate. The flow rate effect on the leaching rate of Co-60 was
reduced in proportion to leaching time, but Sr-85 showed a small increase in the leaching rate.
This is probably because the very low solubility of Co-60 induces accumulation at the interface
and then the effect ofleachant flow rate on the leaching rate becomes gradually less significant
after the removal of the Co-60 which existed near the surface of the test specimen in the early
leaching period. Some increase of the Sr-85 leaching rate is likely to be attributed to its higher
solubility and more complex leaching processes than attributed to Co-60.
The Effect o f Backfill Material on the Leaching Rate

In general, backfill material surrounding waste forms has hydrodynamic and sorptive effects
on the leaching behavior of radionuclides. The presence of backfill around waste forms may
induce a hydraulic variation ofleachant flow and eventually may affect the mass transport of
radionuclides from waste forms. The sorptive effect of backfill material takes place by chem-
f-)
0.08 0.08 c-
z~ C o - 6 0 in ANS 16.1 3
9 C o - 6 0 in MCC-4S F_.
[] S r - 8 5 in ANS 16,1 /
E
o
13
0.06 0.06 7q
o 3
o
6"
f--
0.04 0.04
o
o
(o
2
la_
O_
0.02 0.02 -.4

o 88
-5 m
I
E t-o
S"
3
0.00 I]lll]~[{illlflll[lllllll]lt [lilillill 0.00
0.0 50.0 100.0 150.0 200.0
Leoching Time (DGys)
FIG. 4--The effect offlow rate on the leaching of Co-60 and Sr-85.
212 HAZARDOUSWASTES
0 020
o ANS 16 1
9 MCC-4S
E
0
0015
7D
&)
c-
O
9
J
c- 0010
O
"5
L
>
4-J
I
0 005
E
(D
0 000
0 0' ........ 50 ' 0......... 100"d ....... 150"d 200 0
. . . . . . .
Leoch)ng Time (Days)

FIG 5--The effect offlow rate on the leachmg of Cs-137
lcally binding any radionuchdes leached to effectively retard the migration The resulting accu-
mulation of radionuchde concentrations in the VlClmty of waste forms may substantially alter
the release
To investigate such effects of backfill material, the following two types of modified MCC-4S
leach tests were performed in this study the space in the leaching container was filled for one
test with clinoptllolite having high sorptlve capacity to simulate a backfill material, and for the
other test the space was filled with nonsorblng material (that is, glass beads) consisting of the
same sized particles as chnoptllohte The results obtained from these leach tests are shown in
Figs 6 through 8
As shown in Fig 6, the leaching of Co-60 was not significantly affected by the presence of
backfill Conversely, the leach rates of Sr-85 and Cs- 137 (Figs 7 and 8) were hydrodynamically
influenced to an appreciable extent by the backfill material In addition, part of the leached
Sr-85 seems to have been adsorbed on the glass beads used as a nonsorblng backfill material.
In Fig 9, Sr-85 had a higher cumulative fraction leached m the MCC-4S test without backfill
material, while the Sr-85 cumulative fraction leached In the modified MCC-4S test filled with
glass beads was comparable to that ofCs- 137 The comparison of these two cases suggests the
posslblhtythat Sr-85 may be adsorbed on glassbeads For a clearer inference, further testsare
needed to check for the possibility of Sr-85 adsorption
The sorptlve effect of the chnoptllohte on Co-60, Sr-85, and Cs-137 releases is also shown
In Figs 6, 7, and 8 As shown in Fig 6, Co-60 release was not retarded by the chnoptIlohte, on
the contrary, it increased, although the increase was not appreciable Serne et al reported that
some Co-60 migrated together with soil at the same rate as water [lO] Torok showed that
mobile Co-60 was associated with colloids of the clay in lysimeter and released as a component
of the colloid particles [11] The previous works are thought to explain the shght increase of
Co-60 in terms of the clinoptilohte There was strong adsorption of Sr-85 and Cs- 137 onto the
chnoptilohte,
Copyright asrights
by ASTM Int'l (all shown inMon
reserved); theFebFigs. 7 and
24 23:07:03 EST82020
In particular, the Sr-85 release in the leachate was
0 0020
o NO BACKFILL
a GLASS BEADS
z~ CLINOPTILOLITE
E
.~o 0 0015
00010
2
L
o 0005
0 0000 T i i i i i I i i i I 1 i i I I i i I I I i i i I I i i l I
0 0 40 0 80 0 120 0
FIG 6--The hydrodynamtc and sorpttve effects of backfill matertal on the leachmg of Co-60
0 040
o NO BACKFILL
a G L A S S BEADS
~-- ~ CLINOPTILOLITE
E
0
"J 0 030
"I0
t-
o
__J
c 0 020
_o
"6
ta_
I1,
> 0010
E
(..)
0.000 .-,--, ,-', ~ , , ,~, ,, , ' , , ,", , , ,",, ,", , , ~,'

0 0 40 0 80 0 120 0
FIG 7 - - The hydrodynamic and sorpttve effects of backfill material on the leachmg of Sr-85
0012
o NO BACKFILL
[] GLASS BEADS
A CLINOPTILOLITE
E
0
c- 0 008
O
E
0
'0
L
0 004
>
E
(D
0 000
0 0 40 0 80.0 120.0
FI G 8--The hydrodynamlc and sorptlve effects of backfill matertal on the leachlng of Cs-137
0 040 0 040
NO BACKFILL GLASS BEADS
o Sr-85 o Sr-85 C)
E 9 Cs-157 9 Cs-137 C
3
c
"0 0 030 0 0.30
Q)
r <
o
(~
D -q
J
8
0
t- 0 020 0 020
O 5
S
o f--
CI
I1) (b
> D"
0010 0010 (1)
a.
0
E"1
q)
3
0 000 . . . . . . . . . . . . . . , . . . . . . . . . ~l,lr,,,ilWlll~,l,ll~ll,l,, 0 000

00 1000 15000 500 1000 1500
keach,ng Time (Days) Leaching T,me (Days)
Copyright by ASTM 9--The
FIG Int'l (all rightshydrodynamm effect
reserved); Mon Feb 24 of backjWl
23:07:03 EST 2020 matertal on the leaching of Sr-85 and Cs-137
below the detection hmlt used for the analysis of the radlonuchdes The adsorption capacity
of the clmoptllohte to the three ra&onuchdes was found to be in the following order Sr-85 >
Cs-137 > Co-60
Conclusions
The results obtained here were highly reproducible as demonstrated by the trlphcate speci-
mens evaluated for each leach test
The leaching rates of the three radlonuchdes under a leachant condition maintained at raft-
rote dilution were ranked in the following order Sr-85 > Cs-137 > Co-60 The presence of
Zeolon 900-sodium addmve to the waste product formulation distinctly provided an adsorp-
tion for Cs- 137 and thus compared with previous stu&es, minimized Cs- 137 release from the
specimens However, the Zeolon 900-sodmm additive did not seem to reduce the Co-60 and
Sr-85 releases effectively
The leaching of Cs- 137 from the cemented ion exchange resin waste forms was dominated
by &ffUSlOn, and for Co-60 and Sr-85, the diffusion-controlled leaching process was vahd
except for the initial period, where it was thought to be by wash-off from the surface of the
specimens For Co-60, in particular, a solublhty effect was also strongly present m llmmng its
release
The flow leach tests with and without backfill were an attempt to examine the release under
more realistic disposal conditions The leaching rates of Co-60 and Sr-85 were both apprecm-
bly influenced by the leachant flow rate, while Cs- 137 was apparently not sensmve to the flow
rate of leachant. This appears to suggest that diffusion resistance through a film layer around
a waste form could be larger for strontium and cobalt than cesium
The presence of nonsorbmg backfill material caused radlonuchdes to follow longer paths
than for specimens leached only in the presence of water A natural zeolite, chnoptflohte
selected as an adsorbing backfill material did not retard Co-60 release, a result observed with
the nonsorbmg backfill material It is strongly suspected that the Co-60 is released as colloids
rather than as dissolved ions from the waste forms to the surrounding leachant The presence
of the chnoptflohte restricted the releases of Sr-85 and Cs-137 to the lowest values observed
for all the test condmons evaluated in this study
The study did point out that additional work is needed to develop a model that quantita-
tively accounts for the effect of backfill material on the release of radlonuchde from waste
forms.
Acknowledgment
The authors wish to acknowledge the support of the International Atomic Energy Assn
(IAEA) and Atomic Energy of Canada Limited (AECL) in this work project We are also
indebted to B L Woods and V T Le for their advice and kindness during J. O Lee's attach-
ment at Atomic Energy of Canada Limited
References
[1] Andersson, K, Torstenfelt, B J, and Allard, B, "Sorptlon and Diffusion Studies of Cs and I in
Concrete," Report 5-412-96, Chalmers University of Technology, Goteburg, Sweden, 1981
[2] Mendel, J E, et al, "A State-of-the-Art Review of Matenals Properties of Nuclear Waste Forms,"
PNL-3802, Battelle Pacific Northwest Laboratory, Richland, WA, 1981
[3] Godbee, H W and Joy, D S, "Assessment of the Loss of Radioactive Isotopes from Waste Solids
to the Environment,Part 1 Background and Theory," ORNL-TM-4333, Oak Ridge National Lab,
Copyright byOak
ASTMRidge, TN,reserved);
Int'l (all rights 1974 Mon Feb 24 23:07:03 EST 2020
[4] Anders, O U , Bartel, J F, and Altschuler, S J , "Determination of the Leachability of Solids,'"

Analyttcal Chemlstry, Vol 50, No 4, 1978, pp 564-569
[5] Macedo, P B, Barkatt, A, and Simmons, J H , "A Flow Model for the Kanetics of Dissolution of
Nuclear Waste Glasses," Nuclear and Chemtcal Waste Management, Vol 3, 1982, pp 13-2 l
[6] Buckley, L P and Ward, W J , "Evaluation of Cement as an Immobilization Matrix for Ion-
Exchange Resins," CRNL-2320, Chalk River Nuclear Lab, Chalk River, Ont, 1983
[ 7] Buckley, L P and SperanzInl, R A, "Evaluation of Matrices for Immobilizing Ion-Exchange Res-
ins," AECL-6971, Atomic Energy of Canada Limited, Chalk River, Ont, 1980
[8] Atkinson, A, Nlckerson, A K, and Valentine, T M, "The Mechanism of Leaching from Some
Cement-Based Nuclear Waste Forms," Radloacttve Waste Management and the Nuclear Fuel
Cycle, Vol 4, No 4, 1984, pp 357-378
[9] Colombo, P, et al, "An Experimental Survey of the Factors that Effect Leaching from Low-Level
Radioactive Waste Forms," BNL-52125, Brookhaven National Laboratory, Upton, NY, 1988
[10] Serne, R J, Walter, M B, Jones, T L, and McLaunne, S B, "Chemical Characterization, Leach,
and Adsorption Studies of Solidified Low-Level Wastes," Battelle Pacific Northwest Laboratory,
Rlchland, WA, PNL-6047, 1986
[11] Torok, J , Buckley, L P, Woods, B L, and Chol, H J, "The Mechanisms of Radlonuchde Release
from Waste Forms and Migration through Buffer Layers in Laboratory Lyslmeters," AECL Tech-
nical Report TR-466, Atomic Energy of Canada Limited, Chalk River, Ont, 1988
E d m o n d o Z a m o r a n i ~ a n d Giorgio Serrini ~
Effect of Hydrating Water on the Physical

Characteristics and the Diffusion Release of
Cesium Nitrate Immobilized in Cement
REFERENCE: Zamoram, E and Sernm, G , "Effect of Hydrating Water on the Physical Char-
acteristics and the Diffusion Release of Cesium Nitrate Immobilized in Cement," Stabthzatton
and Sohdtficatton of Hazardous, Radtoactlve, and Mixed Wastes, 2nd Volume, ASTM STP
1123, T M Gdham and C C Wiles, Eds, Amencan Society for Testing and Matenals, Phlla-
delphm, 1992, pp 217-226
ABSTRACT: The effect of the preparation procedure--(vanable amounts of hydrating water on

the leachablhty of a soluble contaminant (CsNO3) from cement samples immersed in water, as
well as the physical charactensUcs of cement samples such as density, pore volume, and com-
press]ve strength--were considered in this work Leach tests of cement specimens containing 2%
m weight ofcesmm nitrate (CsNO3), prepared at different water-to-cement ratios, W/C = 0 35,
0 40, 0 45, and 0 55, were cured for 60"C, 98% RH, for eleven days The leaching standard pro-
cedure suggested by the International OrganIzatmn for Standardization (ISO) was used to follow
the kineUcs for the Cs +, NO~-, and Ca 2+ releases from cement nght cylinders, 20 mm in diameter
and 20 m m high, leached m water at a solid surface-to-liqmd volume (S/V) = 0 1/cm The
results of the cumulative fraction release (CFR) versus time (t) in days indicate that the leaching
of soluble elements like NO~- and cesium follows a diffusion mechamsm The cumulative frac-
tion release (CFR) values were found to depend on a porosity factor
/3 (Vp)l,2,3
( V.)o
where (Vp) 1,2,3 represents the total porosity of cement at the W/C considered, and ( Vp)o
represents the porosity at W/C = 0 35 A diffusion coetfioent, De = l 4 )< l0 -7 cm2/s, calcu-
lated by the values of the cumulative fraction release at 50~ and W/C = 0 35, confirms our
previous results obtained on cement samples containing CsNO3 and prepared under the same
conditions
KEY WORDS. hydrating water, leach tests, porosity, diffusion coefficient
T h e p r e s e n t work, c a m e d o u t at t h e E n v i r o n m e n t a l Institute o f t h e J R C - I s p r a (Italy) o f t h e

C o m m i s s i o n o f the E u r o p e a n C o m m u n i t i e s , deals w i t h t h e p r o g r a m o n t h e s t a b l h z a t i o n / s o h d -
lfication o f m e d i u m level waste ( M L W ) a n d toxic h a z a r d o u s industrial waste P r e v i o u s works
were c o n c e r n e d w i t h t h e investigation o f the influence o f several p a r a m e t e r s , such as curing
c o n d i t i o n s , f o r m factor, a m o u n t o f chemicals, t e m p e r a t u r e , a n d w a t e r flow rate, o n a n y pos-
sible m o d i f i c a t i o n o f t h e leachability o f l m m o b l h z e d waste [1-3] C e m e n t - b a s e d materials in
w h i c h e l e m e n t s r e p r e s e n t a t i v e o f n u c l e a r waste a n d structural materials (cesium, c h r o m i u m ,
mckel, a n d so on), l n d u s t n a l toxic wastes, a n d p o l y c h l o n n a t e d organic c o m p o u n d s h a d b e e n
c o n f i n e d were also investigated to define a t i m e - b a s e d release i n t o t h e water. T h e leaching
b e h a v i o r o f t h e matrices, In w h i c h t h e n u c l e a r M L W or industrial h a z a r d o u s waste are l m m o -
1 Commission of the European Communities, Joint Research Center, 21020 Ispra (Va), Italy
217

218 HAZARDOUSWASTES
blhzed, represents one of the most important properties that characterizes the retention capac-
ity of the host material. For soluble elements (such as cesium) lmmoblhzed in cement, the
release by water action is governed by a diffusion mechanism based on Flck's diffusion law,
according to which a diffusion coefficient can be calculated from the values of the cumulative
fraction leached out of the matrix as a function of the total leaching period
Our previous works on matrices containing chemical compounds have confirmed the influ-
ence of the preparation conditions, such as water-to-cement rano (W/C) and c u n n g condi-
nons, on the physical characteristics of hardened cement specimens In some cases a surplus
of hydrating water had to be used to modify the rheologlcal consistency of the cement paste
In fact, the addmon of chemical compounds affects the set of the cement paste A value of W/
C = 0 35 (normally used for the preparation of neat cement specimens) could not be applied
because of rapid set when Cr 3+, Cd 2+, N12+, AS 3+, and pentachlorophenol compounds were
blended m cement Therefore, in those cases, the a m o u n t of water had to be increased to values
of W/C in the range 0 65 to 0 80 [4-6] Through the effect of the increasing W/C ratio, the
properties of hardened cement samples, as well as the leaching resistance, were found to
change
Extended investigations have been done on the preparation conditions on the diffusion of
elements lmmobihzed in the cement matrix Several expenmental conditions (such as diffu-
sion through a slab, diffusion by electric field, and so on, Including the effect of matrix com-
position) were used by previous authors [ 7-10] A general diffusion model that accounts for
the diversity and complexity of the interacting mechanisms has not yet been applied [ 11 ] The
effect of the preparanon procedure (variable amounts of hydrating water W/C) on the leach-
ability of a contaminant (CsNO3) and of a cement consntuent (calcium) from cement speci-
mens immersed m water, as well as on the physical characteristics of cement such as density,
pore volume, and compressive strength, were considered in this work
Laboratory Procedure
Sample Preparatton
Pure hydrated cement paste and cement paste containing 2% of CsNO3 were fabricated
under identical conditions The chemical additives were blended with clinker cement powder
of ordinary portland cement (OPC 325) to reach a sohd-to-cement ratio S/C = 0 027 and
water-to-cement ratios W/C = 0 35, 0 40, 0 45, and 0 55 The paste, in plastic molds, was
cured at 60~ and 98% RH for eleven days At the end of the c u n n g period, the hardened
speomens were cut by the "Isomet" cutting machine from Buehler, Illinois, with parallel
blades The specimens, 20 m m in diameter and 20 m m high, were prepared for subsequent
physical tests, chemical analysis, and leach tests
Measurement of Physical Propertws

Taking into account the influence of hydrating water on the physical and chemical behavior
of the cement matnx, physical properties such as compressive strength (CS), apparent (bulk)
density (Dapp), pycnometnc density (Dpyc)and pore-size distnbution were measured at the end
of the c u n n g penod The Dpyc,pore volume (Vp), and pore-size distnbution were measured on
sample granules (2.83 to 3 36 mm) dned overnight and maintained in a silica gel container A
hydraulic press from Scientific Instrument (Mllano, Italy) was used in the measurement of the
compressive strength The cumulative pore volume and pore-size dlstnbutlon were deter-
mined by a modified AG 60 mercury intrusion poroslmeter (MIP) from Carlo-Erba (Mllano,
Copyright
Italy) by ASTM
at a Int'l
m a(allx irights
m ureserved); Mon Feb of
m pressure 24 23:07:03
100 MPa EST 2020
(MN/m 2) In the calculation of the pore radii, a
ZAMORANI AND SERRINI ON HYDRATING WATER 219
TABLE 1--Results of the measurement of the phystcal

characterzsttcs on cement samples prepared and cured under the
spectfied condzuons
W/ C CS D apo D pyc Vp
CEMENT WITH CsNO3

035 38 205 273 0 13
040 31 1 99 2 70 0 14
0 45 19 1 84 2 69 0 21
0 55 11 1 82 2 68 0 22
PURE CEMENT
035 44 200 270 013
0 40 30 1 92 2 66 0 15
045 24 186 267 019
0 55 13 1 73 2 62 0 25
NOTE W/C = water to cement ratio

CS = compressive strength (MPa or MN/m 2)
Dapp = apparent density (g/cm 3 or l03 kg/m 3)
Dpyc = pycnometrlc density (g/cm 3 or l03 kg/m 3)
Vv = total pore volume (cm3/g or l0 -a m3/kg)
mercury contact angle o f 141 ~ was assumed All physical properties were measured on neat
c e m e n t samples and c e m e n t samples containing c e s m m nitrate according to the procedure
previously described [12,13] T h e results are reported In Table 1 and Fig 1
Procedures o f the Leach Tests

Leach tests on c e m e n t specimens containing cesium nitrate and cured under the conditions
reported in the previous section were c a m e d out at 50~ in closed plastic bottles at a solid
surface-to-liquid v o l u m e ratio S~ VL -- 0 1/cm T h e temperature of 50~ used in the leach tests
was selected assuming a scenario for a possible waste-confining repository as well as to com-
pare the leaching results previously obtained at th~s temperature In the duplicate leach test,
the frequency o f l e a c h a n t renewal was adopted according to the ISO procedure [14] This pro-
cedure suggests a wide degree of freedom in the leaching temperature and in the renewal fre-
quency At the end o f the teaching period, the leachate was cooled to r o o m temperature and
analyzed to d e t e r m i n e cesium, N O ; , and calcium by a procedure previously described [1 ]
Discussion
The experience gamed in the immobihzatlon/sohdlfiCatlon investigations put into evidence
the effect o f chemlcal compounds on the cement rheology [4-6,15-17] In order to counteract
the rapid set o f the c e m e n t paste and the consequent modification of physical properties such
as strengthening of the hardened samples, the a m o u n t of water had to be increased to W / C
values greater than 0 35 as normally e m p l o y e d in neat c e m e n t preparation All interferences
in the hydration process forming a calcium silicate hydrate (C-S-H) and calcium alumlnate
hydrate (C-A-H) gel modify the strength, which depends on the setting of the c e m e n t paste
Changes in the pore structure can occur due to modification of C-S-H, which is primarily
responsible for the mechanical strength and the capillary voids [18,19].
T o counteract the modification of physical characteristics, which can occur when the
c e m ebynASTM
Copyright t specimens
Int'l (all rightsare i m mMon
reserved); e r sFeb
e d24in23:07:03
water ESTfor
2020a long period o f time, the dimensions o f the spec-
CS o Previous results
(MPa) on neat cement
+ Th~s work
50 (Z)Thls work cement + C s N O 3
40
t o
30
20
10
I I I I I I I
01 02 03 04 05
018 06 07
W/C
FIG l--Compresstvestrength (MPa)versusW/C
lmens (20 m m diameter by 20 mm) have been chosen In such a way, total release ofcesmm
and NO~, not influenced by structural modification, can rapidly occur
When the release of soluble elements follows a dependence on the square root of time, the
calculation of diffusion coefficient is justified Due to the complex multlphase structure, the
physlochem~cal properties of cement and the ~mmobdlzed elements or compounds can evolve
d u n n g the water lmmers~on occurnng m an accidental event m a repository So that, m con-
necUon with a r~sk analys~s, a release mechamsm must be investigated case by case taking into
account the boundary cond~uons.
Modtficatton of the Cement Structure

The increase of hydrating water leads to a decrease m the compresswe strength and an
increase m the total porosity of hardened cement samples. The results, reported m Table I and
Fig 1, show the influence of hydrating water W/C on physical properties
Leachmg Experiments
The direct influence of a structure mod~ficaUon can result m an increasing mass transfer
(diffusion mechamsm) from the bulk to the outside of the specimen by water penetration into
the pores. The time-dependent release of d~fferent elements through the effect of water, that
is, calcium as the principal cement constituent and cesium and NO{ as immobilized com-
pounds, was determined by the relatwe amount of a considered element or compound
reported versus time (t) m days on a log-log diagram [20]. A straight hne is obtained m&catmg
that the leaching can be represented by the relation
CFR = c/co = B r (1)
where the cumulative fraction release (CFR) is gwen by the ratio o f the amount (C) m the
liquid phase and the amount C0 m the sohd. The B value represents the CFR value at t = 1
day. When the slope of the straight hne assumes a value of the time exponent n = 0 5, the
release is governed by a &ffusmn mechanism In that way it ~s permissible to calculate a dlf-
fusmn coefficient De given by
(2)
in which Vand S represent the geometric volume and surface of the samples, respectwely The
results of the leach tests at 50"C on speomens containing 2% of CsNO3 are reported in Fig 2
Cesmm Leaching
The results o f C F R = C/Co, obtained in the leach tests of cement containing cesmm mtrate
prepared at different water-to-cement ratios, were mathematically treated with the help of Eqs
1 and 2 From Eq 1 the values of C/Co versus time in a log-log dmgram, a slope n = 0 5 from
the linear part, four to eleven days, can be calculated (Fig. 2) Beyond this time, cesium is
completely leached out. From Fig 2 it appears that the value of coefficient B (Eq 1) apparently
I I I l I [ i ) I l I ' I
CFR
100
~ W t" " ,
/C = 0 35
2) W/C = 0 40
3) W/C = 0 45
4) W/C = 0 55
10-1 i i i I t i i I I i l i i
100 10 ~
t (days)
FIG 2--Cumulatzvefractlon release (CFR) of cestum versus time (days) T = 50"C
I I I I I
De
10-6
10-71
i
1
O3
Y 04
I
07 05
I
06
I I
W/C
FIG 3--Cesium dffuston coefficient De calculated by Eq 2 versus W/C T = 50~
increases with increasing W/C, which influences the porosity and compressive strength of
cement specimens This increase appears to be proportional to a porosity factor
_ (Vp)l,2,3 (3)
( Vpo
in which ( Vp)o represents the total pore volume at W / C = 0 35, and (Vp) 1,2,3 the porosity
at W / C > 0 35 (Table 2) In Table 2 an agreement between the coefficients Bexpand B035 •
at W / C ratios 0 35 to 0 55 is observed Nevertheless, the CFR values calculated by B035 •
CFR
NO3 release
1) W/C = 0 35
2) W/C = 0 55
10 o
10-1 I I I
10 o 101
t (days)
Copyright by ASTM Int'l (all
FIGrights4--Cumulatlve
reserved); Mon Feb 24 23:07:03 EST
fractton 2020 of N03 versus tzme (days) T = 50"C
release
ZAMORANI AND SERRINI ON HYDRATINGWATER 223
TABLE 2--Comparison between experimental and calculated

values of CFR at t = 1 day
WIG Vp Vff( Vv)o = 3 Bexp Bo35 )< 3
TOTAL PORE VOLUME MEASURED BY MIP

035 013 1 40• 10 i 40• 10
040 014 108 45 • 10 -I 43 • 10 -1
045 021 161 56)< 10 -1 64)< I0 -1
055 022 169 61 )< 10 -1 68 • 10 -j
TOTAL PORE VOLUME CALCULATED BY DENSITY RESULTS
035 012 1 4 0 ) < 10 I 4 0 ) < 10 l
040 013 108 4 5 ) < 10 I 4 3 ) < 10-1
045 017 142 5 6 ) < 1 0 -~ 5 7 ) < 1 0 -1
055 018 1 50 6 1 )< 10 -I 6 0 ) < 10 -1
coefficient show a departure from that obtained in the leaching experiments at W/C ratios of
0 45 and 0 55 W h e n new values o f the total pore v o l u m e in the samples were calculated from
the results o f the density determinations (Dpyc, Dapp), a good agreement between the Bexp and
that of B035 X 3 will result Therefore, the porosities calculated by the density measurements
seem to represent best the leaching m e c h a n i s m affected by the cement specimen
characteristics
In the context of the leach tests carried out at 50"C, diffusion coefficients De have been cal-
culated with the help o f E q 2 The value of De = 1 4 X l0 7 cm2/s at W/C = 0 35 (500C)
confirms the value that we previously obtained with c e m e n t samples containing CsNO3 and
prepared under similar conditions. Taking into account the activation energy E = 9400 cal
in the Arrhenius equation, it is possible to calculate a value of(De)25*C = 4 2 • l0 -s cm2/s,
which is in good agreement with the values obtained in the previous works at r o o m tempera-
ture [1,21,22] The calculated values of De versus the W/C values obtained in the leach tests
at 50~ are reported in Fig 3 The p o o r alignment of the last point concerning the diffusion
coefficient De at W/C = 0 55 In Fig 3 and the curves relative to the pore v o l u m e distribution
at different W/C ratios indicate that the draining off o f cesium from the bulk o f the cement
specimens is influenced by the pore structure. The v o l u m e of the macroporosity tends to
increase and then reaches a constant value, indicating a behavior similar to that o f the c o m -
pressive strength in the region o f W/C = 0 55 (Fig l)
Results of NO; Leachmg

The fraction release o f N O 3 versus t i m e (t) (days) shows a leaching m e c h a n i s m controlled
by diffusion T h e slope, calculated from the values of C]Co versus time at 1 to 20 days with the
help o f Eq 1, gwes a value o f n = 0 5 The a m o u n t of N O 3 , i m m o b i l i z e d in the c e m e n t matrix
as CsNO3, is not completely released in the leachate in the almost 30 days of leaching A m e a n
value o f about 15% is retained in the solid sample, as shown in Fig 4 in which only the bound-
ary lines o f the region o f C F R s between W/C -- 0 35 to 0.55 are reported The situation is
similar to that o f previous results concerning the release of C1- [4] An a m o u n t o f about 10 to
20% o f C l - , retained in the solid c e m e n t samples, was attributed to a c o m p o u n d C3A 3CaY2
m H 2 0 in which Y -- m o n o v a l e n t anion In the present work, a distribution o f NO~- between
the solid and the aqueous solution seems also to occur, which determines the departure o f the
(C/Co)NO~ ratio from unity As in the case o f cesium, a proportionality of the C/Co at one
day by
Copyright (Factor B(all
ASTM Int'l inrights
Eq reserved);
1) at different W/C EST
Mon Feb 24 23:07:03 is observed
2020 to depend on the values of(Vv) 1, 2, 3 ./
224 HAZARDOUSWASTES
CFR
Ca2+ release
;)) W//C __--;:35 / /
10 -2
10 3 i
10o
1
101 102
I
t (days)
FIG. 5--Cumulative fraction release (CFR) of Ca2+ versus time (days). T = 500C.
( Vp)o (Eq 3). The values of De at 50~ calculated with the help of Eq 2, lie in the range 5.6 X
I0 -8 and 1.2 X 10 -7 cmY/s at W / C = 0.35 and 0.55, respectively.
Matrix Degradation
Among the chemical and physical parameters, the characteristics of cement such as density
and porosity will facilitate water penetration into the pores and capillaries of the cement
matrix. The matrix degradation, measured by the calcium release, shows a combination of two
processes: a surface washoff followed by a diffusion stage [21,23]. The results, reported in Fig.
5 in a similar way as in Fig. 4, show a dependence of the CFR on the value of the hydrating
water W / C used in the sample preparations. Beyond a period of three days of leaching, the
calcium release follows a diffusion mechanism (n = 0.5 in Eq 1). As in the case o f C s + and
NO3 release, the matrix degradation seems to depend on the relative value of the total pore
volume. Values of diffusion coefficient De have been calculated to be 7.9 X 10-" and 1.5 X
10 -~~ cmY/s in the range W / C = 0.35 and 0.55, respectively.
Conclusions
The results of the present investigation, dealing with the immobilization of CsNO3 in
cement prepared at W / C = 0.35, 0.40, 0.45, and 0.55, indicate the following.
1. At increasing W/C, a decrease in compressive strength, an increase in pore volume, and

a decrease in density are observed.
2. The cumulative fraction release (CFR) of cesium and N O ; during the first days of leach-
Copyright bying follows
ASTM a ~reserved);
Int'l (all rights dependence, as predicted
Mon Feb 24 23:07:03 EST 2020 by Fick's law of diffusion.
3 The kinetic of the cesium release depends on the water-to-cement ratios (W/C) used in
the specimen preparation, and it increases with increasing W / C (Fig 2)
4 The diffusion coefficients, De, for cesium calculated by Eq 2 increase with increasing W~
C (Fig. 3)
5 The slope of the calculated De versus W / C for cesium is in agreement with previous
results taking into account the difference in temperature of the leach tests and the differ-
ent experimental procedures (diffusion through a cement slab) used by previous authors
[10,24]
6 The N O ; release differs from the behavior of cesium About 15% remains in the solid
matrix, probably due to a compound like C3A 3Ca(NO3)2 nH20 [16]
7 In the case of cesium, encapsulation is the pnncipal mechanism of fixation in cement.
Therefore, for industrial purposes, a low W / C ratio has to be used in order to denslfy the
matrix and decrease its porosity.
8 Comparison with previous results indicates that the matrix degradation, measured by the
c a l o u m release, is much higher under dynamic conditions, typical for an accidental
event in a repository, than in stagnant water conditions In dynamic flow, or in water
renewal, the slope of the calcium release versus time assumes a value n = 0 5, indicating
a diffusion mechanism On the other hand, n assume a value of 0 25 in stagnant water
conditions [25].
9 Extrapolation of the data to a long leaching period of time would be possible when
boundary conditions are established on water flow rate, pH, temperature, and water
chemical composition affecting the cement matrix durablhty
References
[1] Zamoranl, E and Sernnl, G , Waste Management and Nuclear Fuel Cycle, Vol 14, No 3, 1990, p
239
[2] Zamoranl, E, et al, Proceedings of the 5th Internatlonal Conference on Sohd Wastes, Sludges and
ResldualMatertals, A Delfino, Ed, Rome, Italy, 26-29 April 1989
[3] Zamoranl, E and Rlaz, M, Proceedings of the 5th lnternattonal Conference on Sohd Wastes,
Sludges andResldualMatertals, A Delfino, Ed, Rome, 26-29 April 1989
[4] Zamorani, E, Semnl, G , Sheikh, I, and della Rossa, M in Proceedmgs of the 12th Internattonal
Symposmm on ScwntlfiC Basis for Nuclear Waste Management, ORA PRO 34110, MRS, Berlin,
Germany, 10-13 Oct 1988
[5] Zamoranl, E, Semnl, G , and Sheikh, I, "Characterization of Cement ContainingArsenic Triox-
ide," EUR 11926 EN, Commission of the European Communities, Luxembourg, 1988
[6] Rlaz, M and Zamoranl, E, "Testing of Ordinary Portland Cement for the Immobilization of Pen-
tachlorophenol," EUR 12044 EN, Commission of the European Communmes, Luxembourg, 1988
[7] Shxgun,L and Roy, D M, Cement and Concrete Research, Vol 16, 1986, p 749
[8] Selshl, G and Roy, D M, Cement and Concrete Research, Vol 11, 1981, p 751
[9] Westlk, J H and Peters, R D, Sctenttfic Basis for Nuclear Waste Management, Vol 3, Plenum
Press, New York, 1981
[10] Atkmson, A and Nlckerson, A K, Journal of Materials Science, Vol 19, 1984, p 3068
[11] Gllham, T M and Dole, L R, paper presented at the Boston Summer AIChE Meeting, 24-27
August 1986
[12] Zamoranl, E and Mazzocchl, N, European Apphed Research Report Nuclear Scwnce Technology,
Vol 5, 1984, p 741
[13] Zamoranl, E and Blanchard, H, "Pore Volume and Pore Size Distribution of Cement Samples
Measured by a Modified Mercury Intrusion Poroslmetry," EUR 11069 EN, Commission of the
European Communmes, Luxembourg, 1987
[14] "Long-Term Leach Testing of Solidified Radioactive Waste Forms," ISO-6961- 1982, 1st ed, Inter-
national Organization for Standardization, Geneva, Switzerland, 1982
[15] Glasser, F P, et al, Aberdeen University Work Performed in the Frame of the 3rd Radioactive
Waste Management Research Programme of the CEC-DG XII, Brussels, CEC number F1
IW0025UK(H),
Copyright by May 1986
ASTM Int'l (all rights reserved); to December
Mon Feb 24 23:07:03 EST1986
2020
[16] Lea, F M , The Chemtstry of Cement and Concrete, 3rd Ed, Edwards Arnold Ltd, Great Britain,
1980
[17] Zamoram, E, et al, Nuclearand Chemtcal Waste Management, Vol 8, 1988, p 239
[18] Stgn~cance of Tests and Properttes of Concrete and Concrete-Making Materials, ASTM 169-A,
American Sooety for Testing and Matenals, Philadelphia, 1966
[19] Zamoranl, E, et al, Radloactlve Waste Management and the Nuclear Fuel Cycle, Vol 7, No l,
1986, p 29
[20] Katayama, Y B, "Leaching of Irradiated LWP Fuel Pellets m Delomzed and Typical Ground
Water," BNWL-2057, Battelle Northwest Laboratories, 1976
[21] Crawford, R W., et al, Radloacttve Waste Management and the Nuclear Fuel Cycle, Vol 1, March
1985, p 1
[22] "CharactenzaUon of Radioactive Waste Form," F P Glasser and C McCullogh, Eds, Progress
Report CEC, EUR 11354 EN, Vols 1 and 2, Commission of the European Communmes, Luxem-
bourg, 1986
[23] C6t6, P L and Constable, T W , Nuclear and Chemwal Waste Management, Vol 7, 1987, p 129
[24] Page, C L, et al, Cement Concrete Research, Vol 1 l, No 3, 1981, p 395
[25] Zamoranl, E, Nuclear Technology, Vol 77, June 1987, p 313
Michel Buil, 1 Eliane Revertegat, ~ and Jean Oliver ~
A Model of the Attack of Pure Water or

Undersaturated Lime Solutions on Cement
REFERENCE. Bull, M , Revertagat, E, and Ohver, J , "A Model of the Attack of Pure Water
or Undersaturated Lime Solutions on Cement," Stabthzatlon and Sohdlficatton of Hazardous,
Radloactlve, andMtxed Wastes, 2nd Volume, STP 1123, T M Gdham and C C Wdes, Eds,
American Sooety for Testing and Materials, Philadelphia, 1992, pp 227-241
ABSTRACI': A slmphfied model of cement attack by pure water or undersaturated hme solu-
tmns ~sproposed for long-term predictions of concrete degradation The degradation ~smodeled
by lomc dlffusmn m the cement pores coupled w~th mass exchange between the sohd and aque-
ous phases The local equlhbnum hypothes~s is d~scussed and justified for the present apphca-
Uon The only sohd phases taken into account are portlandlte [Ca(OH)2] and calcmm-sdlcate-
hydrate (C-S-H) The model predicts the existence of an undamaged core separated from the free
surface being attacked by a sharp portlandlte dissolutmn front w~th a progressive C-S-H decal-
ClfiCatlon zone located between this front and the surface Good agreement has been found
between the model and experimental results for the degradatmn of thin portland cement discs
kept for three years in an undersaturated lime solution (1 6 mM/L) The model is completely
determined with a few parameters measurable by simple tests (porosity, calcium and portlandlte
contents, diffusmn coefficient) The model predicts a ~/t kinetics for the cement degradation (as
long as an undamaged core remains)
KEY WORDS: cement, water, degradation, model, diffusion, hydrates, calcium, dissolution,
calcium hydroxide, calcium sihcate hydrates
Nomenclature
C C a l c i u m concentration in the pore solution of the outer hydrates layer o f the c e m e n t
grains, in m o l / L
Co C a l c i u m concentration in solution outside o f the c e m e n t paste, in m o l / L
C~ Initial calcium concentration in the pore solution o f the c e m e n t paste, in m o l / L
C v Solid calcium concentration o f the c e m e n t paste, in moles for 1 L of paste
Cp0 Initial total solid calcium concentration in the c e m e n t paste, in m o l / L
Cpj Initial solid calcium concentration contained In the C-S-H phase o f the cement paste, in
mol/L
Cs C a l c i u m equilibrium concentration in the pores at m l d d e p t h of the inner hydrates layer,
in m o l / L
dc C e m e n t density, in g / c m 3
riCH Density o f p o r t l a n d l t e , in g/L
De Effective diffusion coefficient through the c e m e n t paste, in m2/s
Dee Effective diffusion coefficient through a c e m e n t paste containing no capillary porosity,
in m2/s
Dex Diffusion coefficient in the zone located outside of the inner layers, in mE/s
Commissariat g l'Energle Atomlque, France
227

228 HAZARDOUSWASTES
Do D~ffuslon coefficient in the outer hydrates layer of the cement grains, in mE/s
J, Flow o f calcmm ions crossing the boundary between the inner hydrates layer and the
outer hydrates layer, in mol-s -~ m -E
k Rate coeffioent of the calcmm d~ssolutlon in the outer pore solution, m s-
McH Molecular weight ofportlandite, in g
& Surface of the boundary between the tuner and outer layers, in a unit cement paste
volume, In m - i
W/C Water-to-cement raUo of the cement paste, in weights
Ax Half-thickness of the inner hydrates layer, in m
X
Boltzmann variable, )1 = - ~
0 Total porosity of the cement paste

0e Porosity located outside o f the inner layers
Inltml porosity located outside of the inner layers
O, Porosity located inside the tuner layers
pc Volume density of the space in the cement paste located outside of the original hmits
of the cement grains
Introduction
In a surface repository of low-level radioactive wastes, water attack ~s the most probable
degradation mechanism o f the concrete bamers, concrete containers, and cement matrixes
used for the waste solidification. The structural concrete corrosion can result from an acciden-
tal flow or slow seepage of water through the clay cover protecting the concrete barrier The
concrete waste containers located inside the concrete structural barrier can also be corroded
by the flow of water that has crossed the concrete bamer. In this case, the corrosive water
remains for a very long time a basic solution containing alkalis and hme leached from the
concrete barrier [ 1,2]
To predict the extent of concrete and cement matrix degradation dunng long-term waste
storage, modehng is necessary. We present here a simphfied model that deals with the action
of pure water or undersaturated hme water The model is restricted to calcium release from
the two principal phases of hydrated cement, portlandlte [Ca(OH)E]and the calcium silicate
hydrate phase CaO-SiOE-H:O (or C-S-H) The dissolution of C-S-H is described by an empir-
ical relationship adjusted using experimental data. This approach avoids a theoretical treat-
ment of the incongruent dissolution o f the C-S-H phase, for which there is no complete ther-
modynamic theory (see for example Berner [3], Gartner and Jennmgs [4] The model's
predictions are then compared with experimental results for the attack on an ordinary port-
land cement paste by an undersaturated lime solution.
Basic Assumptions of the Model

Hydrated cement degradation by pure water results mainly from two phenomena: (1) an
exchange o f matter between the aqueous interstitial solution and the hydrated solid phases;
and (2) a transport of the species in solution through the pores of the hydrated cement
The exchange of matter between the aqueous solution and the hydrates can result from dis-
solution or transient precipitation of intermediate less soluble hydrates (or both) In the pres-
ent model, only the two main hydrated phases of ordinary portland cements have been con-
sldered, that is, C-S-H and Ca(OH): Other minor constituents such as hydrates containing
aluminate and sulfate have been disregarded. We also exclude the presence m the water of
species other
Downloaded/printed by than calcium ions (and O H - ) that could interact with the hydrated phases.
BUlL ET AL ON MODELFOR ATTACKON CEMENT 229
In order to obtain quantitative values for the mass exchange between the solid phases and
the interstitial solution, we have assumed local equilibrium between the composition of the
pore solution and the hydrates On the basis of this assumption, the rate of this mass exchange
(by dissolution or precipitation) is high enough so that equilibnum can be reached locally
within a time interval where the vanation of the solutlon's composition by Ionic diffusion 1s
neghglble. This assumption is justified in more detail m the next section.
The transport of ions m the interstitial solution of cement hydrates has been considered as
resulting from a diffusion process. No convection transport has been taken into account. This
hypothesis is a simplification that can be partly justified by the fact that the pores of hydrated
cement are small enough to prevent thermal convection and also by the fact that the hydraulic
pressure gradient is very low under the clay cover at the waste disposal site
Justification of the Local Equilibrium Hypothesis

Local equlhbnum is a generally accepted hypothesis for very slow processes like geochem-
ical phenomena. However, in the case of cements, and for a time scale of a few years, the valid-
ity of this hypothesis can be discussed on the basis of our present knowledge of the microstruc-
ture of hydrated cement paste In fact, the hydration of the main constituent of portland
cement, tncalclum silicate (C3S), proceeds by the development of a compact hydrated layer
around the unhydrated C3S core According to Pommershelm and Chiton [5], the hydrated
layer consists of an inner layer (formed inside the ongmal boundary of C3S grams) and an outer
layer, the inner layer being much more diffusion-resistant than the outer layer (Fig. 1). Pom-
mershelm and Clifton [5,6] have developed a mathematical model of C3S hydration showing
that, after a certain time, the hydrated layer controls the hydration kinetics by limiting ion
diffusion through its thickness. This effect is important m spite of the small thickness of the
hydrated layer (an average of a few microns after one month [ 7]) because of the very low dif-
fusivlty In this layer. This very low dlffuswlty could induce a small amount of calcium
exchange between the hydrated phase [portlandlte Ca(OH)2 and C-S-H] and the aqueous solu-
tion filling the pores between the hydrated grains. This could invalidate the local eqmhbnum
hypothesis.
We calculate m the following section a rough estimate of the extent of the noneqmhbnum
zone that can result from this slow exchange effect
H y d r a t i o n reaction (r=rl)
HYDRATING CaS + oH20 --~- b(C-S-H) + cColOH)z
TRICALs a, b, c - s t o l c h i o m e t r l c coefficients
SILICATE
PARTICLE
~ /_ Innerhydrahon
~~produc,
H,dd(e or ~ s I-'~ CO
~/.~,~ / ~',rl boundary

Oufer producf ~ ~ / ~ _ j ~
ro
FIG l--Representation
Downloaded/printed by of the hydratton of a C3Sparttcle accordmg to Pomershetm and Chfton [5]
Ftow o "~/~
FIG 2--Representatton of the structure of hydrated cement as a packing of spheres of hydrates
Derlvatton o f an Equatton for the Rate o f Calcium Release by Diffusion in the Hydrated
Layers
The hydrated cement paste structure is represented as a packing of monodimenslonal
spheres of hydrates (Fig 2) These spheres are composed of an inner core of the same size as
the onglnal cement grains and an outer layer (created by the total solid volume increase during
hydration) According to the model of Clifton and Pommershelm [5], the effective diffusion
coefficient of ions (diffusion through unit surface of material, including both solid and pores)
is approximately 150 times lower in the inner core than in the outer layer We conclude that
the order of magmtude of the diffusion coefficient in the outer layer Do can be obtained by the
following relation
Dec
Do - (1)
pc
where Dec 1s the effective diffusion coefficient of a cement paste where the pores have been
completely filled by hydrates (that is, for a water-to-cement ratio W / C <_ 0 38, according to
the Powers model [8], pc being the volume density of the "outer" space (space outside of the
original limits of the cement grains).
From measurements by Bernaudat and Revertegat [9] on cement pastes with W / C = 0 37,
we have Dec ~ 10 -12 to 10 -13 m2/s, and the lower estimation of the diffusion coefficient in the
inner core is
Dec
D, -- - - -- 10 -~5 m2/s (2)
pc 150
with Pc --~ 0.55

BUlL ET AL. ON MODEL FOR ATTACK ON CEMENT 231
The flow of calcium ions, Je, crossing the boundary between the inner core and the outer layer
is, in a linear approximation and for a unit volume of cement paste
C~-C
3",. = D g . - - . Sa (3)
Ax
where
C = the calcium concentration in the pores of the outer layer,
C, = the calcium equilibrium concentration at middepth of the inner core,
Ax = the half-thickness of the inner core, and
,5', = the surface of the boundary between the inner and outer layers, in a unit cement paste
volume.
We evaluate the coefficient k in the following equation, which gives the rate of ionic con-
centration increase (for example, calcium ions) in the external pore solution as a function of
the concentration difference C, - C
dC
--~ = k ( C s - C) (4)
We define an "outer" porosity 0, as the porosity located outside of the limits of the original
cement grains
0e = 0 - Oi (5)
where 0, 0i = total and inner porosity, respectively.

With Eq 2, we derive
dC J~ k(C,- C) = D , . C , - - C S~ k= D,. S.
dt - Oe - - - - - ~ - " 07 ~ A x . 0--~ (6)
As an example, we can make an estimation of k for W / C = 0.37. The volume density of

original cement grains is approximately 45%. According to the Powers model, the total poros-
ity of the hydrated cement is 28% (see for example, Ref6) and the outer porosity is
0e = 0.28oc = 0.28(1 -- 0.45) --~ 0.15
Taking a mean cement grain diameter d --~ 15 #m, Sa is given by
~d ~ 6
Sa = 0.45 . - - = 0.45 9 ~--~ 0.18 • 10 6 m 1
1 a.d3
6
and a very rough estimate o f k is
.0.18 X 10 6
k ~ 10 -15 9 3 X 10 -4 S -1
15
-- X 10 -6 X 0.15
4
Extent m Ttme and Dtstance of the Nonequlhbrlum Zone

Approximating the order of magnitude of the effect of the calcium release coefficient k on
the n o n e q u l h b n u m zone's extent can be obtained by solving the following transport equatmn
with a first-order kinetics term
OC D~ 02C
Ot = 0-~ + k ( C - C) (7)
where Dex IS the &ffUSlOncoefficient m the outer porous structure Dex is related to the effective
&ffUSlOn coefficient through the cement paste, D~, by the approximate equatmn
De
Dex ~ - - (8)
O~
assuming that no transport takes place through the inner layer, since D, << De
From the general solution ofEq 7 for a semi-infinite m e d m m (with the boundary located at
x = 0, t~me starting at t = 0) and a constant concentration at the boundary and uniform lmtial
condlUons (given for example by Crank [10], one can state that local e q u l h b n u m is reached
(that is, C --~ Cs) when
t >> ~ a n d x >> (9)
Taking De = 10 -j2 m2/s, 0e = 0 15 (gwmg Dex = 6 7 X 10-J2), and k --- 3 10 -4 s -l, the
con&tlons (Eq 9) give
t >> 3333 s and x >> 150 ~m (10)
Apparently, the existence ofa n o n e q u d l b n u m zone Is restricted to a thin layer of material near
the surface that is exposed to the attack Thus, the local equilibrium hypothes~s seems to be a
reasonable assumption m the present study
Numerical Modeling
Lime Exchange between the Sohd Phases and the Solution
As indicated before, the only solid phases considered are C-S-H and portlandite The C-S-
H solublhty data for various CaO/SIO2 ratios show that the slhca concentrations are neghg~ble
compared to the calcmm concentrations for almost the complete range of the latter (see for
example Berner [3]) Accordingly, we have neglected the sihcate ions in solution This hypoth-
esis can be restated m the following way C-S-H is assumed to dissolve mcongruently in pure
water or undersaturated lime solution, hberatmg mainly Ca(OH)2 in solution, with all of the
slhca remaining in the solid state Th~s hypothesis ~s in accordance with the solublhty model
proposed by Fujx and Kondo [11], where C-S-H is considered as a sohd solution of caloum
hydroxide Ca(OH)2 and silica gel SIO2 9 0 8H20, the chemical formula of C-S-H being written
as
xCaO SIO2 (x + 0 8)H20

BUlL ET AL ON MODEL FOR ATTACK ON CEMENT 233
The partml d~ssolutlon of C-S-H can be written
xCaO 9 SIO2 9 (x + 0 8)H20 --" ( x - - y)CaO S102 (x - y + 0 8)H20 + yCa(OH)2

solid sohd in solution
With regard to the portlandlte, the solubility of this phase is determined by a single calcium
concentration (approximately 20 mM/L), since the only ions taken into account are Ca +§ and
OH-
As a consequence, the exchange of matter between the solid phases [C-S-H and Ca(OH)2]
and the solution can be completely fitted by a single curve Cp(C) where Cp is the sohd calcium
concentration (m moles for 1 L of paste) and Cthe calcium concentration in solution (in mll-
hmoles per htre).
Figure 3a shows the curve Cp(C) chosen for the numerical calculations This curve is
deduced from the solublhty curve giving the CaO/SiO2 ratio of C-S-H as a funcUon of C by
multiplying by the constant factor So (So total silica content in moles for 1 L of paste) The
shape of this C-S-H solublhty curve (Fig 3b) has been taken as slmdar to the empmcal curve
proposed by Glasser [12] Figure 3b shows a companson with the experimental data available
in the literature on the solublhty of C-S-H This curve can be seen to be made up of three
distinct sections: (1) the vertical segment AB, which corresponds to the portlandlte dissolution
(the solution is saturated with respect to lime), (2) the hne BC, which corresponds to the pro-
gressive calcium depletion of the C-S-H phase--the curve is fitted by a function C TM (n = 3
for the curve of Fig. 3); and (3) the vertical hne CD corresponds to the decomposition of C-S-
H into slhca gel
Mass Conservatton Equation

Assuming that the transport phenomenon occurs only in the outer porous structure (since
D, << Dex), the mass balance equation is m the unld~mensmnal case
OC 0 (Dex(C) OC~ 10Cp (10)

ot-~\ El o, ot
Equation 10 can be wntten
0--7 = 1 + Oe~ dCp(C) -Ox Dex(C) (11)

dC
This equation is not eqmvalent to a diffusion equation with a variable diffusion coeffioent,
but the properties of the solution are qmte similar to the diffusion solution in certain cases, as
is dlustrated for the kinetic properties of this solution in a later section
Numertcal Calculations
Numencal solutions ofEq 11 have been calculated with a Crank-Nlcolson implicit method
The calculations were carried out on a 386 microcomputer for a half-layer of width L/2,
divided into a 200-point grid. A constant value of the concentration Co is imposed at the sur-
face, with, inside the layer, a uniform mmal solutmn concentration C~ and solid composition
given by the concentrations Cp0 and Cp~.
Cp
Cpo A
Ca from
. . . . . B
Ca(0H) 2
Cpl
C v3
Ca from
'[-S-H
DI Co%,I [(mM/t )
0 15 20
b
26 §
" + I ~
2)1- ,tk,,
f " o 0 o 0 ~
~
c 7 ~ 141- . § /~-oO%~ ~ ~.~**~ro
/ o ^--,,ff.o'~'~~ x~e., ~ t
o I- oO &.~@.~.~~-~ ,, o
~06 *"
"5
02
O0 2 l. 6 8 10 12 14 16 18 20 22
[ {reMit )
FIG 3--(a) The curve Cp(C) m the model, and (b) the solubtltty data of C-S-H m hme water represented
by the CaO/Sl02 ratio of C-S-H as a function of the calczum concentration zn Solutton C (experimental
points collected by Berner [3])
The initial outer porosity 0~0is deduced from the Powers model, assuming a porosity in the
tuner layer equal to the gel porosity (28%)
0~o= 0 - 0 . 2 8 dc W/C + 1
where d~ is the cement density

The equation for the dependence of 0e on Cp was determined by assuming that the porosity
created by leaching a given quantity of calcium hydroxide [from C-S-H or Ca(OH)2] is equal
to the volume of the same quantity of portlandlte (this hypothesis is supported by data from
Feldman [I 3] and also by our own data). This gives the following relation
MCH
Oe(Cp) = 0~o + ~ pr -- Cp) (13)
where M c n and dcn are the molecular weight and density of portlandite, and Cp0 being the
initial solid calcium content. If Cp is given in moles/litre, Eq 13 becomes (with M c n = 74 g
and dcn = 2230 g/L)
0e(Cp) = 0~o + 0 033p~(Cpo - Cp)
The diffusion coefficient Dex IS assumed to be constant throughout the calcium-depleted

zone where a C-S-H phase remains This assumption is supported by the results of Feldman
et al [13], which show that the mIcrostructure of C-S-H is only slightly modified by calcium
depletion Thus, Dex can be determined experimentally by a single measurement of De for the
reference unleached sample (with the relation Dex = De/Oe )
In the silica gel zone (where a complete calcium leaching has occurred and no C-S-H
remains), we assume a complete collapse of the material structure, and Dex IS taken equal to
the ionic diffusion coefficient in bulk water (approximately 10 -9 m2/s)
Figure 4 gives an example of the calculated solid calcium profile as a function of time for a
cement paste layer exposed to pure water attack (taking Dex = 5 • 10 ~2m2/s for the reference
cement paste) This profile can be divided into three parts
1 An undamaged core with a sharp dissolution front of the portlandlte

2 A zone of progressive calcium depletion of C-S-H ahead of the portlandlte dissolution
front.
3 A third zone between the surface and a second front of decomposition of C-S-H, remain-
ing very close to the surface being attacked. The zone between this front and the free
surface is composed of silica gel.
.T
LL " ~'porf[arld,fe
Cp 2rnm
~p[el'ed - III I........ f om
Sll,ca get tayer

FIG 4--Calculated calctum profiles m a cement paste layer exposed to pure water (Dex = 5 • 10 -~2
m2/s), evolution up to three-years" exposure (curves are gtven every stx months)
236 HAZARDOUSWASTES
The Degradation Kinetics According to the Model

One can show that the degradation kinetics follow a simple diffusmn law in V~ in the case
of a semi-infinite medium with a constant surface condition and a uniform mmal condition
(thts case represents a c o m m o n situation in the degradation of massive elements submttted to
an aggressive solution with a constant compositmn) This point can be proved by testing the
compatibility ofEq 11 with the Boltzmann transformatmn defined by the change of variables
x
- ~ (14)
Assuming that the solution C is a function only of n, Eq 11 becomes
2 d~ dCp(C)
1 + Og'(Cp) dC
The partml differential Eq 11 15 thus transformed into an ordinary dlfferentml equation in

C(~), confirming the compatibility with the Boltzmann transform [10]. Also easily verified is
that the initial and boundary condmons can be expressed as functions of ~ only
C = Co for ~ = 0 (boundary condition)
and C = C~ for 77 = oo (initial condition)
What follows from this result is that, m the case of a semi-infinite medmm and uniform
initial and constant surface condmons, the concentration profiles C(x,t) for different gwen
times are superposable by the simple transform
x---~ x
This means that the evolution of the process follows a V~ law; such an evolutlon concerns, for
example, the rate of progression of the fronts or the rate of calcium loss.
Comparison of the Model Predictions with Experimental Results for a Portland Cement
Exposed to Undersaturated Lime Water
A senes of portland cement paste specimens have been kept for three years m an undersat-
urated lime solution. We gwe here a companson between the expenmental results and the
model predmtlons
Spectmen Preparatton and Conservalton

Table 1 gwes the chemical composition oftbe ordinary portland cement used in th~s study
The water-to-cement ratio was 0.37 The paste was molded m 7-era-diameter cyhnders and
kept at 20"C under saturated lime water for three months. Before testing, the cylinders were
cut into 4-ram-thick d~scs.
The disc speomens were then immersed m a 60-L volume of undersaturated hme solution
TABLE 1--Chemtcal analysts of the portland cement used m the

expertmental study
Constituent %, In Weight
SIO2 20 5
A1203 52
Fe203 34
CaO 64 0
MgO 08
K20 09
SO3 30
TIO2 0 28
MnO 01
Na20 01
LOI 14
Free CaO 12
Insoluble 02
( 1 6 mM/L, corresponding to a pH "- 11.5) at 20"C. The specimens were fixed to a rod moving
slowly up and down for the whole duration of the test This constant motion was imposed to
prevent the formation of a stagnant layer of saturated solution near the surface of the
specimens
The lime solution was kept under a nitrogen atmosphere to prevent carbonation and
renewed by means of ion exchange resins when the pH increased by more than 0 2 pH umts.
Speomens were taken out of the conservation container after 1, 3, 6, 12, 18, 24, 30, and 36
months and subjected to chemical analysis
Expertmental Measurements
After each storage penod, the total water and mercury porosity and the chemical compo-
sition of the specimens were determined. Chemical analysis was performed by X-ray fluores-
cence, with a precision of + 0.2% for calcium and silicon and between + 0 1 and _+0.05% for
the other elements The calcium hydroxide content was measured by differential thermal anal-
ysis. Table 2 gxves the compositions of the specimens. The calcium content profiles through
the thickness of the specimens were determined by electron mlcrobeam analysis.
TABLE 2--Evolutton of the compostuon and porostty of the spectmens from 1 to 36 months after
exposure to a 16 mM/L hme solutton All chemzcal contents are gtven as percentages of the mtttal
weight of the samples (at ttme 0 and before desstcatton)
Porosity,
Time, SiO2, A1203, Fe203, CaO, SO3, 1(20, L O I, Water, Porosity, Ca(OH)2,
Months % % % % % % % % Hg, % DTA, %
Reference 152 42 24 452 23 046 169 25 1 226 134

1 141 38 25 416 23 009 161 31.0 249 96
3 149 40 26 403 24 009 160 33.1 245 86
6 150 41 25 382 23 009 154 362 261 83
12 16 2 43 27 35 7 2.3 009 14 9 380 279 73
18 151 42 25 366 2.3 005 156 384 306 75
24 146 40 23 349 2.3 001 148 395 278 81
30 158 40 23 326 23 001 158 382 255 74
36 16 8 39 23 30 6 23
001 15 7 405 286 77
The diffusion coeffioent of ions m the cement paste was determined for the reference spec-
imen by means oftntmted water diffusion Trltiated water was chosen as the diffusing element
because of its low adsorption coefficient on the sohd surfaces in hydrated cement [9], results
from three specimens gave an average effective diffusion coeffioent De --~ 1 6 X 10 -12 m2/s
L i m e Content Evolutton
Table 2 shows that the pnnclpal chemical modification concerns the calcium content This
observation supports, for this type of chemical attack, the simplified hypothesis of the model
with respect to material exchange between the solid phases and the solution Figure 5 gwes the
measured and calculated total calcmm oxide content loss as a function of t~me
Calcium Profiles
Figure 6 shows the calculated and measured calcium oxide profiles for different conserva-
tion times. The experimental profiles exhibit a fluctuating shape with numerous peaks due to
the fact that they are measured along a single line with an electron beam spot a few microns
in diameter When the electron beam hits a pore, a gap appears on the profile, when the beam
hits a portlandlte crystal (giving a calcium density higher than the mean value), a peak appears
on the profile.
Figure 7 gives the calculated depths of the portlandlte and C-S-H decomposition fronts as a
function of the square root of time
Discussion
Reasonably good agreement is found between the respective shapes of the measured and
calculated calcium profiles Although the correct order of magmtude of the portlandlte front
depth is predicted by the model, quite large deviations have been observed between the mea-
sured and the calculated depths. Those deviations may be explained, at least in part, by the
fact that the experimental profiles are very localized and thus subject to large fluctuations
The experimental calcium profiles show, as m the model, the presence of a zone ofprogres-
40
t-
O
I.d
30- o o
o
U
o 20
t.J
10
o
=- [ l l 1
O( I 2 3 years
FIGInt'l5--Measured
Copyright by ASTM (all rights reserved);and
Mon calculated
Feb 24 23:07:03total calctum oxide content loss as a functton of time
EST 2020
%CaO
5o~~
40-
30-
20-
BUlL ET AL ON MODEL FOR ATTACK ON CEMENT
1 year
239
~o-~1
o l~r.V 2r~m.
40-~
30- 2 years
20- ,/'-
10- ',2ram
o ' I"
50-
40- 'i
f I
30 3 years
20-
10. 1 mm !2 mm
~ ' i(L)
FIG 6--Compartson between calculated and expenmental calcmm profiles at one, two, and three years
(the profiles are gtven as CaO content m percent wetght of the sohd phase)
2 mm -
15mm
Imm
05 m m / / C-S\H front
0 , i J J
6 months 1year 2 3 years
Vfime
FIG 7--Calculated depths of the portlandlte and C-S-H decomposttlon fronts as afunctton of the square
root of timeby
Downloaded/printed
240 HAZARDOUSWASTES
sive decalcification between the surface and the portlandite front The experimental depth of
a superficial silica gel zone, flit emsts, is very low, as m the model (in fact, according to Fig 3,
a pH of 11 5 corresponds approximately to the lower limit of stabdity for C-S-H)
The predictions of the model for the calcium loss evolution are in acceptable agreement with
the experimental results (Fig 5) What should be emphasized is that these results have been
obtained with a model that is dependent only on a few parameters, all easily determined exper-
Imentally These parameters are.
t Total porosity
2 Total calcium oxide content [assumed to be totally contained in C-S-H and Ca(OH)2 ]
3 Portlandlte content
4 Effective diffusion coefficient
An arbitrary adjustment was made for the curve Cp(C) (Fig 3) in particular for the choice
of the power law C v", owing to the large scatter of the solubility data for C-S-H given in the
hterature Different values of n have been tested (between 2 and 4) they reduce small varia-
tions (no more than 10%) m final results, either for the poslhon o f the portlandlte front or for
the total calcium loss
Figure 7 shows that the model predicts a V/t kmencs law for the duranon of the tests This
result can be explained by the fact that for up to three years an undamaged core (limited by
two portlandlte fronts) is still present m the specimen In this case, it can be shown that a layer
behaves as a semI-mfimte medium as long as an undamaged core remains (see above section
on the degradation kinetics according to the model)
A deviation from the k/t law would occur for a cumulative calcmm leached fraction higher
than approximately 35%, corresponding to the disappearance of the undamaged core. In the
case of a simple diffusion process, this limit for the cumulative leached fracnon is approxi-
mately 20% This difference arises from the particular shape of the calculated calcium profiles,
very different from the simple diffusion profiles m a plane sheet [10]
The experimental results are not m contradiction with a V~ kinetics law, but the accuracy
of the data is not sufficient to gave a definitive confirmation of this important point
Conclusions
The present simplified model can fairly accurately describe the phenomenon of cement
attack by an undersaturated lime solution The advantage of this model is that it needs only a
few parameters, all measurable by simple experiments This model can be easily applied to
different types of cements where the major hydrated constituent is C-S-H, such as portland,
pozzolamc, or fly ash cements
A major drawback is that the model is limited to pure water or lime solution attacks Com-
mon aggressive groundwaters such as sulfate solutions need more complex models that take
into account the diffusion of several species m a multlmmeral system
The diffusion-controlled kinetics of the cement attack (V~ law) predicted by the model is
not directly confirmed because of the expenmental data's lack of accuracy from a klnencs
point of view More experimental work is needed on the precise klnencs of cement attack
Nevertheless, the correct prediction of both degradation depth and total calcium loss by the
present diffusion-controlled model can be considered as an indirect element of evidence of a
diffusion-controlled phenomenon
If this point could receive a more defimte experimental proof, simple extrapolations could
be made for the prediction of the depth of degradation of a concrete or a cement matrix for
Copyright
certain by ASTM Int'l (all rights
constant reserved); Mon
aggressive Feb 24 23:07:03 of
condmons ESTthe
2020 environment (such as pure water attack) For
example, a predlcUon of the degradation profile of a concrete element after 300 years could be
made by multiplying the depths by ten from the results of three years' laboratory tests
References
[1] Atkmson, A, Eventt, N M, and Guppy, R M, Matertals Research Socwty, Symposium Proceed-
mgs, Scwnttfic Baszs for Nuclear Waste Management XII, Vol 127, 1989, pp 439-446
[2] Haworth, A, Sharland, S M, and Tweed, C J, Materials Research Socwty, Symposium Proceed-
rags, Scientific Baszs for Nuclear Waste Management XII, Vol 127, 1989, pp 447-454
[3] Berner, U R, RadtochlmlcaActa, Vol 44-45, 1988, pp 387-393
[4] Gartner, E M and Jennmgs, H M, Journal oftheAmerlcan CeramwSoctety, Vol 70, 1987, pp
743-749
[5] Pommershelm, J M and Chfton, J R, Cement and Concrete Research, Vol 9, 1979, pp 765-770
[6] Pommershelm, J M and Chfton, J R, Cement and ConcreteResearch, Vol 12, 1982, pp 765-
772
[7] Osbaek, B and Johansen, V, Journal of the Amerlcan CeramtcSocwty, Vol 72, 1989, pp 197-201
[8] Hansen, T C, Materlals and Structures, Vol 19, No 114, 1986, pp 423-436
[9] Bernaudat, F and Revertegat E, "Mod6hsation des ph~nom~nes de transport dans les m6camsmes
d'alt6ratlon d'une pgRede oment," 8th International Congress on the Chemistry of Cement, Rio de
Janewo, Proceedings, Vol V, 1986, pp 41-46
[10] Crank, J, The Mathematics of Dlffuston, Clarendon Press, Oxford, England, 1983
[11 ] Fuji, K and Kondo, W, Journal of the Ceramtc Socwty, Dalton TransacUons, Vol 2, 1981, pp 645-
651
[12] Glasser, F P, Report DOE/RW/84159, United Kingdom Department of Environment, 31 Oct
1989
[13] Feldman, R F and Ramachandran, V S, Cement and Concrete Research, Vol 12, 1982, pp 179-
189
Kimberly J. Perry, 1 Neil E. Prange, 2 and William F. Garvey 2
Long-Term Leaching Performance for

Commercially Stabilized Waste
REFERENCE" Perry, K J , Prange, N E, and Garvey, W F, "Long-Term Leaching Perfor-
mance for Commercially Stablhzed Wastes," Stablllzatton and Sohdtficatton of Hazardous,
Radloacuve, and Mtxed Wastes, 2nd Volume, ASTM STP 1123, T M Gllham and C C Wdes,
Eds, American Society for Testing and Matenals, Philadelphia, 1992, pp 242-251
ABSTRACT: Waste stablhzatlon has emerged as one of the frequently required best demon-
strated available technology (BDAT) treatment methods for land disposal of hazardous wastes
as mandated under the Hazardous and Solid Waste Amendments (HSWA) of 1984 Despite this
increasing reliance on stabilization as a final treatment for hazardous wastes, there is little pub-
hshed Information on the long-term leaching performance for thxs type of treatment Recogmz-
lng this, Monsanto began a study in early 1988 to evaluate the effect of elapsed time after treat-
ment for a selected group of four Monsanto wastes, which included metallic and reactive solids
The wastes were sent to six stabilization vendors, who returned treated waste samples to Mon-
santo for analysis A Toxicity Characteristic Leaching Procedure (TCLP) extraction was per-
formed on each raw waste sample and on each stablhzed waste sample at 28 days (in duplicate)
and at 90, 200, 470, and 650 days (single samples) after treatment The results demonstrated a
highly waste-dependent effect of time on the TCLP results Leachate values for some metallic
wastes remained stable over time, and for other metallic wastes, leachate concentraUons
increased over time These results suggest that additional scientific and regulatory evaluation of
stabilization technologies and treated samples may be required to ensure long-term leaching per-
formance confidence
KEY WORDS: stabilization, hazardous waste treatment, TCLP, long-term stablhty, leachability
Regulations promulgated by the U S E n v i r o n m e n t a l Protection Agency (EPA) under the

Hazardous and Solid Waste A m e n d m e n t s ( H S W A ) to the Resource Conservation and Recov-
ery Act ( R C R A ) restrict the land disposal o f untreated hazardous wastes The only hazardous
waste land disposal permitted must be "protective of h u m a n health and the e n v i r o n m e n t for
as long as the waste remains hazardous " [1]
Stabihzatlon has emerged as a t r e a t m e n t option for m a n y of these hazardous wastes The
E n v i r o n m e n t a l Protection Agency has specified stablhzatlon as the best demonstrated avail-
able technology ( B D A T ) for a n u m b e r ofwastes regulated under R C R A By i m m o b d i z i n g haz-
ardous constituents prior to landfill disposal, there is s o m e assurance that long-term protection
o f h u m a n health and the e n v i r o n m e n t has been provided as required by the U S. Congress
Despite the increasing reliance on stablhzatlon as a final treatment for hazardous wastes,
there is little published i n f o r m a t i o n on this treatment's long-term leaching performance Rec-
ogmzlng the emerging i m p o r t a n c e of waste stablhzatlon and lack o f pubhshed long-term
leaching data, M o n s a n t o began a study in early 1988 to evaluate the effect of elapsed Ume after
t r e a t m e n t for selected M o n s a n t o wastes for which stabihzatlon m a y b e c o m e the preferred
1 Environmental specmhst, Monsanto Co, Krummnch Plant, 500 Monsanto Ave, Sauget, IL 62206
2 Senior environmental consultant and environmental technician, Monsanto Co, 800 N Lindbergh
Blvd, St Louis, MO 63167
242

PERRY ET AL ON LONG-TERM LEACHING PERFORMANCE 243
method of treatment Samples of five Monsanto plant wastes were treated by s~x commercial
stabilization vendors and evaluated using the Toxicity Charactensnc Leaching Procedure
(TCLP) at several rimes over a nearly two-year penod to evaluate the effect of rime on hazard-
ous constituent stabfl~zanon for these wastes
Materials and Methods

Plant Wastes
The following five Monsanto plant wastes were chosen for the study"
1 Waste A a process waste water treatment filter cake containing chromium

2 Waste B a process waste sludge containing cadmium
3 Waste C. an electroplatmg waste water treatment sludge containing nickel, and
4 Waste D a process waste filter cake containing morgamc arsenic
All o f these wastes are either classified as hazardous or contain contaminants that might
classify them as hazardous under future regulations. In addmon, Waste B is classified as haz-
ardous due to reacnvlty, and Waste D is classified as hazardous due to corroswity An indicator
contaminant was selected for TCLP analysis in each waste
Vendors
Six commercial hazardous waste stablhzat~on vendors were chosen for the study A vanety
of stabilization process methods that included the use of fly ash, portland cement, and sd~ceous
matenals were used
Samphng
Two-gallon samples of each plant waste were sent from the plants to Monsanto laboratories
in St Lores, then sent to the vendors Vendors were sent only those wastes they expressed
wllhngness to treat Each waste was accompanied by a waste profile that hsted the composi-
tion, contaminants, properhes, safety precautions, RCRA and Department of Transportanon
(DOT) classifications, and attachments where apphcable, such as Matenal Safety Data Sheets
(MSDS) Vendors were also sent a waste stabilization form, containing requirements for the
treated waste samples, such as the amount to be returned, and mformanon to be supplied, such
as date of stabilization. Vendors then returned their samples of treated wastes to the Monsanto
Environmental Sciences Center (ESC) m St Louis
Analysts
Samples of the treated wastes were extracted using TCLP and analyzed by SW-846 methods
[2], TCLP extracts were prepared by the ESC and analyzed by commercml laboratories located
m St Louis
Separate samples of the treated wastes or allquots of the treated hazardous waste samples
were removed, crushed, and extracted according to the following schedule 28 days (1 month),
90 days (3 months), 200 days (6 months), 470 days (15 months), and 650 days (21 months)
The mmal nine, 28 days, was chosen because it is the commonly accepted time to cure con-
crete, and the remaining txmes were selected to provide as lengthy and evenly spaced a penod
as feasible.
Copyright The
by ASTM Int'l schedule
(all rights reserved);was adjusted
Mon Feb 24 23:07:03slightly
EST 2020 so that samples could be extracted and analyzed
244 HAZARDOUSWASTES
together Duphcate samples of the 28-day poststabxhzatlon sample were extracted and ana-
lyzed to evaluate the homogeneity of the waste samples. One sample of each raw waste was
also extracted and analyzed Testing for one sample of Waste B was discontinued after 200
days because of msufficmnt sample volume
For quahty control purposes, appropriate blanks, splits, and spikes were included w~th each
set of samples sent for analys~s. Method blanks were spiked at approximately 10% of the reg-
ulatory level, and actual sample extracts were spiked at approximately 60% of the regulatory
level Atomm absorpUon was specified for all metal analyses
Results
Tables 1 through 4 present the analytical results of TCLP extractions of the raw and treated
plant wastes for the 28-, 90-, 200-, 470-, and 650-day poststabdlzatlon tests F~gures 1 through
4 present the TCLP results graphed against time Numerical identification of vendors is used
throughout this report Specific ldenttficaUon of vendors ts restricted under confidentmhty
TABLE 1--Analyses of TCLP extracts of Waste A (chromium

waste)
Chrom|um
Vendor Sample Age,Days ConcentraUon, mg/L
1 1 36 0 032
2 36 0 030
l 96 0 030
1 205 0 10
1 485 0 29
1 646 0 37
2 1 29 0 057
2 29 0 052
1 100 0 028
1 210 0 049
1 476 0 062
1 637 0 04
3 ! 29 0 20
2 29 0 11
1 97 0 12
1 211 0 15
1 478 0 21
1 629 0 14
4 1 34 1 50
2 34 1 50
1 99 1 73
1 209 1 72
1 475 1 28
1 636 1 02
5 1 39 0 19
2 39 0 19
1 102 0 20
1 205 0.21
1 450 0 22
1 611 0 23
Method blank < 0 02
Raw waste
6 02
agreements. Results of the analys~s of the method blanks are included. Results of the other
quality control measures are not presented, but they affirmed the vahdlty of the reported data.
Discussion
Homogeneity
The abdlty to evaluate with confidence the effect of the time after treating stabdlzed waste
samples ~s dependent on the homogeneity of the waste samples. Plants were requested to pro-
vide homogeneous samples for the project, but the samples possibly may not have been homo-
geneous at the time of treatment or sampling. To examine the homogeneity of the treated
waste samples and, by ~mpllcatmn, the homogeneity of the raw waste sample and the repro-
duoblhty of the TCLP, two separate samples of the 28-day poststablhzatlon samples were
crushed, extracted, and analyzed. The results, shown m Tables 1-4, indicate that wastes were
reasonably homogeneous.
TABLE 2--Analyses of TCLP extracts for Waste B (cadmtum

waste)
Cadmmm
Vendor Sample Age, Days ConcentraUon, mg/L
1 I 32 0 05
2 32 < 0 02
1 92 < 0 02
1 201 <0 02
1 483 < 0 02
1 644 < 0 02
2 1 28 < 0 02
2 28 < 0 02
1 96 < 0 02
1 210 < 0 02
1 470 < 0 02
630 < 0 02
3 1 30 0 04
2 30 <0 02
1 95 0 28
1 205 2 98
1 474 23 6
1 634 31
4 1 37 <0 02
2 37 <0 02
1 100 < 0 02
1 203 <0 02
1 451 < 0 02
1 611 < 0 02
5a 1 30 <0 02
2 30 <0 02
I 93 0 02
1 196 62
Method blank <0 02
Raw waste <0 02
a Insuflioent treated sample volume precluded 470- and 650-
dayreserved);
tests Mon Feb 24 23:07:03 EST 2020
TABLE 3--Analyses of TCLP extracts for Waste C (mckel

waste)
Nickel
Vendor Sample Age, Days Concentration, mg/L
l 1 32 0 06
2 32 0 07
1 92 <0 05
1 201 <0 05
1 489 <0 05
1 649 <0 05
2 1 27 0 06
2 27 0 06
1 100 0 13
1 210 0 10
1 484 0 I0
1 644 0 08
3 1 30 0 49
2 30 0 50
1 98 4 94
1 212 2 04
1 477 2 14
1 637 5 1
4 1 29 11 8
2 29 13 0
1 94 11 3
1 213 107
1 478 10 8
1 638 10
5 1 28 0 07
2 27 0 07
1 112 0 11
1 215 0 16
1 469 0 10
1 629 <0 05
Method blank < 0 05
Raw waste 28 0
TABLE 4--Analyses of TCLP extracts for Waste D (arsemc

waste)
Arsenic
Vendor Sample Age, Days Concentration, mg/L
1 1 37 0 91
2 37 0 67
1 95 0 30
1 204 0 54
1 494 0 51
1 654 0 36
2 1 31 46
2 31 46
1 93 49
1 207 48
1 474 50
1 634 48
Method
Universidad Industrial de Santander blank
(Universidad Industrial de Santander) pursuant to License Agreement. No<further
0 01reproductions authorized.
Raw waste 30 0
P E R R Y ET AL. ON L O N G - T E R M LEACHING P E R F O R M A N C E 247
C 2
h
r
m
0
u
i
1.5
/
C
0
n
C
e
n
t
r
a
t
i 0.5
0
n
m
g
/
I 0 I
0 100 200 300 400 500 600 700
Age,Days
Vendor 1 ~ Vendor 2 -~- Vendor 3

-E~ Vendor 4 " ~ - Vendor 5
Raw Waste Value Was 6.02 mg/I

FIG. 1--Analyses of TCLP extractsfor Waste A (chromium waste).
The reproducibility of the TCLP method has been studied by several researchers, includ-
ing an interindustry group, with varying conclusions. Most results are expressed as percent
relative standard deviation (RSD) or as coefficient of variation. The relative standard devia-
tions calculated for this study are presented in Table 5. Williams et al. [3] considered the
TCLP method precise for single-laboratory analyses if the variability of replicate samples
analyses did not exceed 25% RSD. Precision of the TCLP for metals in their study was con-
cluded to be "quite good" as long as concentrations did not approach the detection limit or
the level found in the method blank. A single-laboratory study found RSDs of 10 to 20%
for metals [4].
The relative standard deviations calculated for this study present only two data and thus
provide only an indication of homogeneity or no homogeneity. Three of the twelve values
approach or exceed the upper limits considered acceptable in the above-referenced studies. All
three concentrations for these samples are relatively close to the method blank. The conclu-
sion, based on this qualitative comparison of the data from this study with TCLP precision
studies, is that the results are within the reported limits of acceptability and that the stabilized
wasteby samples
Copyright can
ASTM Int'l (all rightsbe considered
reserved); Mon Feb 24to be homogeneous.
23:07:03 EST 2020
C 70
a
d
m 60
i
u
m
50
C
o
n
40
c
e
n
t 30
r
a
t
i 20
o
n
10
m
g
I
I
0 100 200 300 400 500 600 700
Age,Days
Vendor 1 -~ Vendor 2 -~- Vendor 3

-E~ Vendor 4 -G~ Vendor 5
Raw W a s t e Value Was ~.02 mg/I

FIG. 2--Analyses of TCLP extractsfor Waste B (cadmium waste).
TABLE 5--Percent RSD for duplicate TCLP waste extraction analyses.
Mean Concentration, Method Blank,

Constituent Vendor mg/L RSD a mg]L
Chromium 1 0.031 4.6 <0.02

2 0.054 6.5
3 0.16 40.0
4 1.5 0
5 1.9 0
Nickel 1 0.06 12.0 <0.05
2 0.06 0
3 0.50 1.4
4 12.4 6.8
5 0.07 0
Arsenic 1 0.79 21 <0.01
2 46 0
a Relative
Copyright by ASTM Int'lstandard deviation
(all rights reserved); (percent).
Mon Feb 24 23:07:03 EST 2020
P E R R Y ET AL. ON L O N G - T E R M LEACHING P E R F O R M A N C E 249
14
N
i
C
k 12
e
I -g
lO ------0
C
O
n
c
e
13
t
r
a
t
i
O
n
m
g
/
I
o ~. I I j ol~,-
o 100 200 300 400 500 600 700
Age,Days
Vendor 1 ~/~ Vendor 2 ~- Vendor 3

Vendor 4 -0- Vendor 5
R a w W a s t e V a l u e W a s 2 8 mg/I
FIG. 3--Analyses of TCLP extractsfor Waste C (nickel waste).
Time
Hazardous waste treatment standards are presently based on the assumption that the leach-
ing properties of treated wastes do not change with time. Stabilization technology assumes that
once the mixture is "set" (typically after 28 days), the waste is stable and may be landfilled.
The purpose of this study was to evaluate this assumption using Monsanto plant wastes and
commercial vendors.
The results as presented in Figs. 1 through 4 show that leachate concentrations for some
stabilization process/waste combinations do change over time. Let us examine each of these
in detail.
Waste A (Chromium Waste)--As shown in Fig. 1, most of the stabilization techniques
appear to render a treated waste system that is stable over time. However, results for Vendor
1 indicate that there has been about one order of magnitude increase in chromium leachate
values over the 650-day test period. The apparent change in performance was first seen at the
nominal 200-day test and appeared to be an ongoing phenomenon through the final 650-day
test. There also appears to be a steady decrease in leachate concentration over time for the
Vendor
Copyright 4 samples,
by ASTM Int'l (all rightsstarting after
reserved); Mon the
Feb 24 200-day
23:07:03 EST 2020test.
250 HAZARDOUSWASTES
60
A
r
s
e 50
n
i
C
C 40
0
n
C
e 30
n
t
r
a
t 20
i
0
n
10
m
g
/
I
100 200 300 400 500 600 700
Age,Days
Vendor 1 - - ~ Vendor 2
R a w W a s t e Value W a s 3 0 m g / I
FIG. 4--Analyses of TCLP extractsfor Waste D (arsenic waste).
Waste B (Cadmium Waste)--As shown in Fig. 2, the stabilized waste product from three of
the five vendors showed no detectable leaching of cadmium, and the raw waste also showed
no detectable cadmium leaching. (Cadmium analysis, whole waste basis, has been performed
on this waste and indicates a cadmium content of approximately 15%.) However, the Vendor
3 sample showed a steady increase in leaching beginning at the nominal 90-day test and con-
tinuing up through the 650-day conclusion of the program. The Vendor 5 sample showed a
major leaching increase at the nominal 200-day test, but, due to depletion of the sample, test-
ing of that sample was terminated. This cadmium waste exhibits the RCRA characteristic of
reactivity due to its sulfide content. The increase in leaching for the two treated waste samples
may reflect a delayed chemical reaction between the waste and the stabilizing agent.
Waste C (Nickel Waste)--As seen in Fig. 3, the nickel leaching levels in four of the five
vendors' stabilization systems are virtually unchanged over the 650-day test. Vendor 3 data
indicate a significant increase in leaching at the nominal 90-day test, which then seems to per-
sist at the same level through the conclusion of the 650-day aging period.
Waste D (Arsenic Waste)--As shown in Fig. 4, the arsenic leaching levels from the two sta-
bilized
Copyright waste
by ASTM systems
Int'l (all wereMonessentially
rights reserved); Feb 24 23:07:03unchanged
EST 2020 over the entire period of the test. However,
Vendor
Universidad 2 leachate
Industrial de Santanderhad an arsenic
(Universidad content
Industrial de Santander) at about
pursuant the Agreement.
to License same level as the
No further untreated
reproductions waste.
authorized.
The increased leaching seen m some of the staNllzed wastes lmphes that a chemical or phys-
ical change m the stabilized waste matrix is occumng that affects leaching This change can
begin many months after the waste was initially stabdlzed. Thus, leaching tests performed on
staNhzed wastes after only short cunng times, typically 28 days or less, may not be a true m&-
cator of long-term leaching propemes for some combinations of wastes and staNhzatmn
systems
The scope ofthxs study &d not include exploratmn of the mechamsms causing this increased
leaching m staNhzed wastes over time But rather, the study was conducted to determine ff
such a s~tuaUon could m fact occur The study data emphatically accomphshed this, ratslng
serious questxons about current regulations and predtsposal testmg practices for stabdlzed
wastes.
Conclusions
1 The stabdlzed waste samples may be consxdered to be homogeneous with the leaching
results of duphcate samples within the normally accepted hmits of the TCLP
2. Some stabilized wastes may exhlNt increased TCLP leachate levels w~th the passage of
time
3 The onset of change In TCLP extract concentration seen in some of the staNhzed wastes
did not occur untd 90 to 200 days after stabilization
References
[1] Resource Conservation and Recovery Act of 1976, Section 3004, P1 98-616 (Hazardous and Sohd
Waste Amendments of 1984 enacted 11/8/84)
[2] Test MethodsJbr Evaluatmg Sohd Waste, 3rd ed, U S EPA, Office of Sohd Waste and Emergency
Response, Washington, DC, November 1986
[3] Williams, L R , Francis, C W , Maskannec, M P, Taylor, D R, and Rothman, N, "Single-Labo-
ratory Evaluation of MoNhty Procedure for Sohd Waste," Environmental Momtonng Systems Lab-
oratory, Otfice of Research and Development, Environmental Protection Agency, Las Vegas, NV,
undated
[4] Taylor, D R , "Single Laboratory Testing of the ToxxcxtyCharacteristic Leaching Procedure (TCLP)
Using Conventmnal Apparatus," Tasks 4, 9, and 14, Contract 68-03-1958, Enwronmental Protectton
Agency, Washmgton, DC, February 1986
Laboratory-Scale Process Development
Aldo Donato I and Gmseppe A. Ricci I
Application of Radioactive Waste Solidification

Methods to Hazardous Wastes in Italy
REFERENCE: Donato, A and Rlccl, G A , "Application of Radioactive Waste Sohdification
Methods to Hazardous Wastes in Italy," Stablhzatlon and Solidification of Hazardous, Radio-
active, and Mtxed Wastes, 2nd Volume, ASTM STP 1123, T M Gilham, and C C Wiles, Eds,
American Society for Testing and Matenals, Phlladelphm, 1992, pp 255-265
ABSTRACI" The apphcatlon of radioactive waste sohdlfieatlon processes to industrial hazard-

ous waste (HDW) has been studied An mvestigatxon to identify the most significant HDW types
currently produced in Italy was earned out, and samples were collected from various producing
industries to use in an expenrnental sohdification program The sohdlficatmn processes have
been dxwded into two categories (1) processes using polymers as binders (polyester resins and
urea-formaldehyde), apphcable to the sohdlfiCatlon of HDW of mixed organic-inorganic nature,
and (2) processes using cement, applicable to HDW of inorganic nature
The apphcabfllty of the mentioned sohd~ficaUon processes to some representatxve HDW was
studied by estabhshing for each the best waste-to-binder ratio Mechanical strengths and heavy
metal releases were then determined HDW incorporation m cement and m polyester resins ~s
feasible, and the sohdxfied final product satisfies the regulatory reqmrements On the contrary,
the incorporation in urea-formaldehyde resins is not apphcable The sohdlfiCatlon costs have
been calculated
KEY WORDS. hazardous wastes, solld~ficatxon, leachability, polymer sohd~ficatmn, polyester

resins, cement, urea-formaldehyde resin
After the Chernobyl accident and its consequences for Itahan nuclear programs, the Italian
State C o m m i t t e e for Nuclear and Alternative Energy (ENEA) has given m u c h consideration
to the possible application o f experience acquired in the nuclear field to nonnuclear techno-
logical and industrial activities T h e sohdificatlon o f intermediate radioactive wastes is con-
sldered one o f these possiblhtles, as it has been studied for several years at the E N E A C R E -
Casaccia Research Centre
In principle, applying solidification processes used for radioactive wastes to the solidification
o f industrial hazardous wastes ( H D W ) produced by nonnuclear industries is possible The
requirements are in fact the same. the need to encapsulate the wastes in a stable solid form and
the need to m i n i m i z e the release of certain chemical elements o f radioisotopes to the
environment
In practice, the simple application o f such processes is not always possible The costs in the
nuclear field in fact are generally relatively high because they are related to the safety and qual-
ity of the solidified product In addition, radioactive waste volumes are generally m u c h lower
than industrial hazardous wastes, whose characteristics are extremely various and often very
difficult to know.
According to the laws that currently regulate n o n n u c l e a r waste m a n a g e m e n t in Italy
Research manager and chemist researcher, respectively, Itahan State Committee for Nuclear and
Alternative Energy (ENEA), Nuclear Fuel Cycle Division, 00060 Rome, Italy
255

[1-5], H D W disposal can be performed in sites of different charactenstics, namely of the 2B,
2C, and 3rd categories, each of them being authorized to accept H D W of increasing harmful-
ness, the third category disposal sites being designed to accept the most harmful ones, which
may be in liquid form The disposal costs, of course, strongly increase as the disposal site com-
plexity increases
According to the law, the HDW, if immobilized, can be disposed of m disposal facilmes of
the lowest category, 2B, that is, at lower costs This disposal can be done only ffthe heavy metal
release, as determined according to a standard test, is demonstrated to be below the established
limits The H D W sohdlficatlon can reduce the environmental ~mpact, reducing at the same
time the disposal costs.
HDW Selection
In Italy, neither the total yearly amount, nor the composition of H D W production is cur-
rently well known Estimates are on the order of 5000 to 10 000 kg per year out of a total
production of 35 000 to 40 000 per year of lndustnal wastes A more precise evaluation will
be possible only when the Register of Indusmal Wastes at the Minister of the Environment is
fully operational
From the immobilization point of view, H D W can be divided into the three following cat-
egones according to their chemical nature
1 Inorganic wastes, composed of inorganic substances with negligible organics In this cat-
egory the inorganic sludges coming from industnes operating in practically all fields are
included" mechanics, mines, foods, ceramics, galvanic, and so on These wastes could be
disposed of after immobilization with inorganic binders
2. Organic wastes, conslstmg of organic substances with negligible inorganic components
The most applied treatment system for these wastes is volume reduction by incineration
3 Inorganic-organic mixed H D W containing both organic and inorganic components in
vanous percentages. This waste type includes, for example, effluents coming from the
Kraft process for paper production, which contain about 30 to 40% inorganic salts and
60 to 70% of organics, the effluents from metal surfaces preparation such as removing
grease or phosphat~zing operations, and so on
The treatment of the above wastes has been known to give nse to many difficulties, their
incineration IS m fact not always possible because inorganic salts can often severely attack oven
walls while, on the other hand, the immobilization in cement or other inorganic binders can
be prevented by the orgamc substances' presence. In all these cases immobihzation takes place
by means of organic binders.
Selecting the H D W for the expenmental study took into consideration the following
requirements
1. The H D W should be significant from the technical and economic point of view
2. Their yearly produchon rate should not be very high" in this case, in fact, the recovery of
valuable materials or the recycling should preferably be investigated due to the lack of
disposal sites in Italy
3 The H D W should be produced by industrial activmes that are significant in terms of
spread o f H D W in the Italian territory
After a careful investigation, the mixed H D W reported m Table 1 and the inorganic H D W
reported
Copyright by ASTM mInt'lTable
(all rights2reserved);
have Mon been identified,
Feb 24 and significant samples have been acqmred.
23:07:03 EST 2020 The
DONATO AND RICCI ON SOLIDIFICATION METHODS 257
TABLE 1--Types and sources of mtxed HD W constdered for the experimental immobtllzatton
program
Predominant Typical
HDW Type Chemicals Producers Production Rate Present Treatment
A Soda added w~th Cellulose producers 50 m 3 / h Concentration and/

Paper mtll lye CO2 for the straw or mcmeraUon
washing digestion
B Tnchloroethane Researchlabs 0 1 to 1 m3/month No treatments
Chlonnated chloroform 2,
solvents chlorophenol,
and so on
C Water, emulsifying Metal or About 250 m3/ Chemical
Painting agents, paints mechamcal month precipitation
operations industries sludge separation
effluents
D Phosphoric acid Metal or About 1 5 m3/ Chemical
Phosphonc aqueous solutions mechamcal month precipitation
process for with emulsifying industries sludge separation
grease agents and
removal orgamcs
E Exhausted Alodme Metal or About 1 5 m3/ Chemical
Chrommm- (trade name) Cr mechanical month preop~tat~on
plating VI soluUons industries sludge separatton
effluents
sohdlfiCatlon of mixed H D W Types C, D, and E and morgamc H D W Types D and F is

reported in this paper.
Immobilization Processes
The sohdlficatlon processes developed for radioactive wastes are of two main types, accord-
ing to the chemical charactenstxcs of the binding agents, which can be morgamc or organic.
Cement (portland, pozzolamc, or slag) is certainly the most w~dely used morgamc binder.
Other morgamc binders that have been proposed are the Envlronstone gypsum cement [6],
Synrock [ 7], and borosdlcate glass [8] Only the glass is at present currently used in Europe
for high-level radioactive waste solidification, but the technological cost of its use will probably
prevent it from being apphed to H D W sohdlficatlon. In this study the cement lmmoblhzatlon
process has been selected for the morgamc H D W sohdlficatlon.
Concerning the orgamc binders, several matrices have been developed, bitumen [9], poly-
ester resms [ l O], polystyrene [l l ], polyethylene [12 ], and urea formaldehyde resms [13] Only
some of them have been lndustnally apphed, hke bitumen, polyester resins, or urea formal-
dehyde resins.
For mixed wastes lmmoblhzatlon, two polymenr binders have been selected on the grounds
of process slmphclty, cost, and final product characteristics urea-formaldehyde resins, which
could be considered the best polymeric binder from the cost point of view, although thetr phys-
lcochemical characteristics have been demonstrated to be only relatively satisfying m the case
of radioactive wastes [14], and polyester resins, which on the contrary produce solidified
blocks with excellent physlcochemlcal charactensUcs through a simple process but at relatively
Copyright
highbycosts
ASTM Int'l (all rights reserved); Mon Feb 24 23:07:03 EST 2020
TABLE 2 - - Types and source of the morganlc l i d W constdered for the experimental lmmobthzatton
program
Predominant Typical Present

HDW Type Chemicals Producers Production Rate Treatment
A l Fe, Zn, Pb, Mn, Mumopal 400 000 to Storage, studies

Fly ash from Cr, Cd, Cu, incinerators 1 200 000 kg/ for
electron oxides, year utxhzatmn
filters sulfates, or going on
carbonates
B 1 Similar as above Municipal About 1 600 000 Storage, studies
Residual ashes with less Zn, mcmerators kg/year for
Cd, Hg, and utlhzatlon
more Cr, Cu, going on
Fe
C 1 V205 containing Sulfuric aod 1000 to 2000 Storage, high
Exhausted 646 ppm Hg, production plant kg/year cost disposal
catalysts 1653 ppm Pb, necessary
39 ppm As,
others
D l Sohd Fe2(OH)3 Treatment of 3 000 000 to Storage, high
Sludges from containing effluents from 4 000 000 kg/ cost disposal
effluent Cd 129 7 zinc-plating year necessary
chemical ppm, Cr 2284 mdustnes
treatment ppm, Zn
12 891 ppm,
PB 33 2 ppm,
Cu 41 3 ppm
E l $1 19% w/w, Al Electncity production ~ 1 5 kg/year Storage, high
Ashes from coal 12% w/w, Fe plants cost disposal
incineration 5% w/w, Ca necessary
3% w/w,
others
F 1 Fe 29 8% w/w, Gold jewelry 60 000 to Storage, high
Sludges from Cu 0 7% w/w, manufactunng 600 000 kg/ cost disposal
effluent Zn 3 4% w/w, industries year necessary
chemical A1 1 5% w/w,
treatment Cd 0 15% w/
w, Pb 01%
w/w, others
G 1 Narates, sulfates, Au, Ag, and Pt 10 000 to Storage, high
Crystalhzatmn and chlondes recovery from 20 000 kg/ cost disposal
salts of Na, Pb, Cu, jewelry year necessary
Cd, Cr, K manufactunng
mdustnes
Processes using polyethylene and polystyrene as well as bltumlnlzatlon have been excluded
because they operate at high temperatures (from 80 to 200"C)
Solidified H D W Characterization
According to Italian law, sohdlfied H D W waste, to be accepted in a Type 2B disposal site,
must not release the following metal chemical elements in concentrations higher than those
estabhshed in the water protection against pollution law arsenic, mercury, c a d m m m , thai-
h u m , berylhum, Cr VI, lead, antimony, selenium, t e l l u n u m , and copper [1].
Determlnatlng metal element release must be according to a test [4] performed contacting
the solid or solidified waste with acetic acid 0 5 M at ph _< 5 for 24 h and with carbon dioxide
(CO2) saturated distilled water for 6 h in a suitable mixing apparatus, at room temperature (in
the range 15 to 30~
The samples should be in granular form (dimensions --<9 5 mm) The solidified waste sam-
ples could be in a monolithic form only if they would keep their structural integrity while
under the disposal conditions. Nevertheless, the tests to be used for such a demonstration
unfortunately are not yet specified In our study, therefore, the structural integrity demonstra-
tion has been tested according to the methods established for nuclear wastes [ 17] and assuming
the same requirements
The compressive strength of the solidified H D W waste monoliths should be therefore at
least 50 kg/cm 2, and this mechanical strength must be kept after 30 daily weathenng cycles
from - 4 0 to 60~ and after three months of immersion m tap water
Mixed HDW Solidification with Polyester Resin

The polyester resin types SNIATRON 35144 and 3560/S produced by the SNIAL Italian
firm were used for the immobilization experiments The H D W were emulsified in a mixture
of these resins (70 30 ratio) after alkallZatlon with sodium hydroxide (NaOH) up to pH -- 11
The emulsion was then polymerlzed using benzoyl peroxide as catalyst and dimethylparato-
hdlne as promoter At pH = 11 some ferric hydroxide [Fe(OH)3] precipitate appeared in the
D and E wastes without major influence on the polymerization kinetics or on the product
characteristics In Table 3 some characteristics of the solidified D and E waste types are
reported.
The leaching resistance of the D and E waste types immobilized in polyester resins has been
tested using solid cyhndrlcal blocks 50 m m both in diameter and height The leachability
results, both in acetic acid and in CO2-saturated distilled water, are reported in Table 4
The release of copper, cadmium, lead, antimony, and Cr VI from the solidified waste Type
D is below the acceptability limits for 2B disposal In the case of the solidified waste Type E,
on the contrary, the Cr VI release is above the concentration limits, while the release of other
metal elements is acceptable for disposal The immobilization in polyester resins, therefore, as
directly applied to this waste type without preliminary treatments, cannot guarantee in all
cases an effective trapping of Cr VI ions and cannot prevent its diffusion into the water The
chemical reduction of Cr VI to Cr III, which produces chromium hydroxide at the pH of the
emulsion formation (pH = 11), could be very effective from this point of view, just as in the
case of the a l u m i n u m retention in the matrix (see Table 4)
Mixed HDW Solidification with Urea-Formaldehyde Resin

For the experimental immobilization of H D W in urea-formaldehyde resins, the Xllocolla
L 1570 by Montedlson has been used, the polymerization being camed out in about 30 s by
TABLE 3--Preparation characteristics o f some H D W tmmoblhzed m polyester resms
Compressive
Waste Waste/Resin Catalyst, Promoter, Gel Time, Hardening Strength, kg/
Type Ratio, w/w % % Mm Time, Mm cm 2
D 70/30 06 03 7 20 25
D 60/40 12 03 5 20 130
E 50/50 08 03 10 40 70
E
60/40 08 03 10 30 15
260 HAZARDOUS
WASTES
~ 0 0 0 0 0 0 0 0
0 0 0 0 0 0 0 0
VVVVVVVV
oZ~ ~ ~
=~ VVVVVVVV
g~ ~o~ooooo
~ VVVVVVV
VVVVVVVV
~ Q
VVVVV VV
.<
VVVV
VVVVVVV
~.~ o o o o o o o o
VVVVVVV
VVVVVVVV
<
/ / ,60
[] [] Kg/sq.cm
V/V 140
1.5' 120
100
1.0 ' 80
60
0.5 ' 40
20
0.0 0
C C C D D D E E E
HDW Waste Types
FIG. 1--Preparation characteristics of HDW immobilized in urea-formaldehyde resins.
means of concentrated sulfuric acid. The mechanical strength of some HDW-solidified sam-
ples is shown in Fig. 1, while in Table 5 the leachability resistance in CO2 and in acetic acid is
reported.
While the mechanical strength of the solidified wastes could be considered acceptable, at
least for short times and for some waste:polymer ratios, the leaching resistance of urea-form-
aldehyde solidified wastes is not satisfactory for H D W Type E. Moreover, most of the solidified
samples do not overcome the weathering test [15].
Inorganic HDW Immobilization in Cement

The cement solidification was carried out mixing the HDW, which were in the form of par-
tially dried sludges, with water in the ratio 0.3:0.35 w/w and then with pozzolan cement. Cylin-
drically shaped specimens, 60-mm in diameter and height, have been used for the compressive
strength and leachability measurements. The main results obtained in the case of H D W Types
D1 and F1 after 28 days of cement setting time are reported in Table 6. The chemical elements
to be taken in consideration concerning leachability are copper, cadmium, lead, and chro-
mium. In all cases the concentration of these elements in the leachate is below the maximum
allowed value. The heavy metal release in the case of the acetic acid leachability test is generally
higher than in the case of CO2 test, mainly in the case of granulated samples, when in some
cases the concentration in the leachate is higher than the maximum allowed value.
The chemical elements' retention effectiveness in cement is strongly affected by the cement
setting time: after seven days of setting, only samples at waste:cement = 1 of the F1 H D W
type and at waste:cement = ~ of the D 1 H D W type comply with the required leachability
values. This means that in the practice the solidified H D W cannot be discharged in open air
disposal sites before 28 days from the preparation.
Economic Considerations
CopyrightThe industrial
by ASTM Int'l (all rightsapplication
reserved); Mon Febof
24H D W EST
23:07:03 solidification
2020 is largely conditioned by economic con-
siderations,
Downloaded/printed by which is different than in the nuclear field, where long-term safety aspects and
ts~
<
un
e~ e~
t~
<
4~
~ o o - o ,~ o o
V V
eq t~. m ~D
0 ~, 0 e,~ 0 0
~a
V V
,.o
--~z z o o z z V
el
- ~ z z o o ~ z
~kO o o o ~ Z Z
V
el
o o ~ o cq o Z
V
eO"O r 0'~
I
-"-8<~ Z ~ o o Z o Z
.a d
<
[- eo , . 0 .~
~ Z o o o Z Z
<
.E-o
zoz~z~z~
e~
0 ~ 0 0 0 0 0 ~
0 0 0 0 0 0 0 0
V V V V
0 0 0 0 0 0 0 0
VVV V V
CN
0 0 0 0 0 0 0 0
VVVVVVV
5 ~ o ~0 0O00 5 0
0 0 0 0 0 0 0 0
VVVVVVVV
8. ~ ~
0 0 0 0 0 0 0 0
VVVVVVVV
o
0 0 0 0 0 0 0 0
V V V V
0 0 0 0 0 0 0 0
V
,.,< ~ <
0 ~ 0 ~
I
L~
H
<
Z
health physics considerations are of outstanding importance The cost of solidification with
urea-formaldehyde and polyester resins or cement has been therefore evaluated assuming the
construction and the exploitation ofa solldificahon plant at the H D W production site More-
over, the cost evaluation has been based on the following assumptions
1 H D W dally production 2 m 3
2 No. 2 plant operators
3. Plant operation time. 60 days per year
4. Plant amortization time ten years, at a capital investment interest of 16%
5 Urea-formaldehyde present cost about $400 per 1000 kg.
6 Polyester resins present cost" about $4000 per 1000 kg
7 Inflation rate 5% per year
8 Disposal site distance about 100 km
The cost calculatmn results are of $600 to $700 per 1000 kg of solidified HDW In the case
of urea-formaldehyde and of $6000 to $7000 per 1000 kg of sohdlfied HDW for polyester
resin The cost of cement sohdlfication is calculated to be $120 to $160 per 1000 kg ofsohdIfied
HDW, that is, approximately 30 to 40% less ff compared to the direct disposal cost in a 2C
disposal site (see beginning of the paper)
Conclusion
Sohdlficatlon of H D W using processes currently applied to radioactwe wastes must take
into account H D W chemical charactensUcs Inorganic HDW can be solidified at a reasonable
cost using pozzolanic cement, and produced blocks satisfy the physlcochemlcal requirements
established by Italian law
In the case of mixed inorganic-organic wastes, when they cannot be burned in incinerators,
the solidification with polyester resins can produce solidified blocks of good charactenshcs,
while the urea-formaldehyde resin cannot be used due to poor leachability and weathering
resistance The cost of polyester polymers nevertheless seems not to be acceptable from an
industrial point of view
Acknowledgments
The authors wish to thank all colleagues of the ENEA--Nuclear Fuel Cycle Division, who
participated in the research work reported in this paper
References
[1] Itahan rules for water protection from pollution (Merh law), 10 May 1976 (Law no 319) (Norme
per la tutela delle acque dall'mqumamento Legge 10 Magglo 1976 n 319 Legge Merll )
[2] Accomplishment of European Community directives on wastes (No 75/442), on polychlorodl-
phenylandpolychlorotnphenyldlsposal(No 76/403),andontoxlc/hazardouswastes(No 78/319)
Italian law by decree No 915, l0 Sept 1982 (Attuazlone delle dlrettlve CEE n 75/442 relatlva al
nfiuti, n 76/403 relativa aUo smaltlmento del pollclorodlfenlh e del pohcloro trlfenlh e n 78/319
relatlva al nfiuti tossicl e nocivi DPR 10 Settembre 1982 n 915 )
[3] Provisions for the Italian Law by Decree No 915 Article 4 (10 Sept 1982) accomplishment con-
cernlng the waste disposal Itahan Ministry Coordination Committee Resolution of 27 July 1984
(Dlsposlzlonl per la prima apphcazlone dell'art 4 del DPR 10 Settembre 1982 concernente 1o smal-
timento de1 nfiutl Comitato Intermlnlstenale Dehberazlone 27 Lugho 1984 )
[4] Modifications of provisions for the Italian law by decree, No 915, article 4 ( 10 Sept 1982) accom-
phshment
Copyright by concerning
ASTM Int'l (all the
rights reserved); waste
Mon Feb 24disposal, Italian
23:07:03 EST 2020 Ministry Coordination Committee of 14 July 1986
(Modlficazlonl e lntegiazlonl alle dlspos1zlonl per la prima apphcazlone dell'art 4 del DPR 10 Set-
tembre 1982 concernente lo smaltlmente del rlfiUtl Comltato mtermlnlstenale, Dehberazlone del
14/7/1986 )
[5] Italian Law No 475 of 9 Nov 1988 concerning urgent regulations in the field of industrial waste
disposal (Legge 9/11/1988 n 475 Converslone in legge del D L 9/9/1988 no 397 recante dlsposl-
zlonl urgentl in matena dl smaltlmento di nfiutl mdustrlall )
[6] Rosenstlel, T L and Lange, R G ," Envlronstone Gypsum Cement l0 CFR 61 Compliance Test-
rag," Proceedings, Symposium Waste Management 84, Tucson, AZ, 1984
[7] Rlngwood, A E , Kesson, S E , Ware, N G , Hlbberson, W , and Major, A , "Immobilization of
High-Level Nuclear Reactor Wastes in Synrock," Nature, Vol 278, March 1979, pp 219-223
[8] "Monographlc Number of Nuclear Europe," Journal of ENS (European Nuclear Society), No 2,
February 1985
[9] Day, Y E , Doyle, R D , Rastberger, S B, and Tchemltcheff, E, "Transportable Volume Reduc-
tion Bitumen Sohdlficatlon System from Design to Operation," in Proceedings of American Nuclear
Soctety Conference Spectrum 86, Niagara Falls, NY, September 1986, American Nuclear Sooty, La
Grange Park, IL
[10] Thompson, M L, Miller, G P , Klncald, C B, Caputl, R W , Weech, M E, and Rodrlquez, L F,
"Aztech System and Testing," Proceedings of the Symposium on Waste Management 85, Tucson,
AZ, 1985
[11] Ambros, R J , Baatz, H , Hepp, H , Rlhdcher, D , and Schroeder, H , "Reactor Waste Management
Practices In the FRG," Proceedings of the Symposium on On-Stte Management of Power Reactor
Wastes, Zurich, Switzerland, 1979
[12] Recalde J A , "Operative Experience in Radioactive Waste Treatment in Atucha Nuclear Power
Plant," Proceedings of the Symposium on On-Site Management of Power Reactor Wastes, Zunch,
Switzerland, 1979
[13] "AlternaUve for Managing Wastes from Reactors and Post-fission Operations in the LWR Fuel
Cycle," Vol 2, Report ERDA-76-43, 1976
[14] Donato, A , "Technical Evaluation of Urea-Formaldehyde Resins Incorporating Spent Ion
Exchangers from Nuclear Power Plants," Atomkernernergle Kerntechmk, Bd 33, 1979 Lfg 1
[15] Technical Guide No 26 on Radioactive Waste Disposal, Slcurezza e Protezlone, No 14, 1987, in
Italian
George 34. Dohnar, 1 R. Wo Doug Killey, 1 and Ken E. Phdipose ~
Stabilization and In Situ Management of

Radioactive Contaminated Sediments of Port
Hope Harbor
REFERENCE" Dohnar, G M , Kllley, R W D , and Phd~pose, K E , "Stabihzation and In Situ
Management of Radioactive Contaminated Sediments of Port Hope Harbor," Stabdtzatton and
Sohdlficatton o f Hazardous, Radloacttve, and Mtxed Wastes, 2nd Volume, STP 1123, T M
Gilham and C C Wiles, Eds, Amencan Society for Testing and Materials, Philadelphia, 1992,
pp 266-282
ABSTRACT" Th~s paper presents the findings of laboratory and field studies undertaken to
assess the feasibility of m sttu management of contaminated sedaments in Port Hope harbor The
contaminated sediments stem from historic releases from an adjacent radium and uranium
refinery, and uranium, arsenic, and radium are the most abundant contaminants With
improved emission controls, currently accumulating se&ments have much lower levels of con-
tamlnation, and the harbor waters currently meet water quahty limits for the contaminants of
concern Within a few years, however, the continuing sedimentation will render the harbor unus-
able Field tests have confirmed that dredging will result in incomplete removal of the contam-
inated sediments and that sediment suspension and the release of pores waters during dredging
will produce harbor water contaminant concentrations that would require the treatment of large
volumes of water In addition, no remedial work can start until a site for the dredged material
can be found The local community mqmred whether in sttu burial of the sediments and aban-
donment of the harbor would provide safe disposal Removing the sediment by conventaonat
dredging methods would leave approxamately 5% of the contaminants in the harbor and result
in possible mobihzatlon of the contaminants so that a method of further ~solatlng the contami-
nants in place is investigated The use of a geotextde (filter fabric) as central to the concept of
being able to consohdate the sediment in place by decreasing the water content of the sediment
dunng the appllcatmn of fill materaal while confining the contaminated sediment
KEY WORDS contaminated sediment, in SltU, radioactive waste, stablhzatlon
T h e t o w n o f Port H o p e is located a d j a c e n t to t h e m o u t h o f t h e G a n a r a s k a R i v e r o n the n o r t h

shore o f Lake O n t a r i o (Fig 1) A l t h o u g h once a n active s h i p p i n g t e r m i n a l d u n n g the late 1800s
a n d early 1900s, use o f t h e h a r b o r s u b s i d e d with t h e o p e n i n g o f the St L a w r e n c e Seaway; cur-
rently t h e h a r b o r IS used by pleasure craft. M o s t l a n d s u r r o u n d i n g the h a r b o r is z o n e d indus-
trial, a n d t h e m a j o r i n d u s t r y r e l e v a n t to this study 1s t h e ore refining a n d m a t e r i a l processing
o p e r a t i o n s c a m e d o n b y E l d o r a d o N u c l e a r L t d ( n o w C A M E C O ) from 1933 to t h e present
As a result o f these o p e r a t i o n s ( a n d to a lesser e x t e n t m u n i c i p a l a n d o t h e r Industrial effluents)
the h a r b o r s e d i m e n t s h a v e b e e n c o n t a m i n a t e d with a v a n e t y o f organic a n d i n o r g a n i c c o n t a m -
i n a n t s U r a n i u m , r a d i u m , a n d arsenic (with t h e latter p r e s e n t as various arsenldes in the ores
processed b y E l d o r a d o In t h e 1930s) are the c o n t a m i n a n t s o f p r i m e c o n c e r n T h e C a n a d m n
g o v e r n m e n t h a s classified the P o r t H o p e h a r b o r s e d i m e n t s as historic low-level radioactive
Research Scientist, hydrogeologlst, and research engineer, respectively, Atomic Energy of Canada
Ltd, Chalk River Laboratones, Chalk River, Ontario, Canada
266

DOLINAR ET AL. ON CONTAMINATED SEDIMENTS 267
+d
z
~0
!
i
+o
8
+1
_z
m0
o
vl--
,<z
.~0
wastes As such, the federal government has assumed responsibdity for their management
through the Low Level Radioactive Waste Management Office (LLRWMO) that is managed
by Atomic Energy of Canada Ltd (AECL)
Although waters in the harbor generally meet current water quality criteria, the harbor was
identified by the International Joint Commission (IJC) as one of the 42 areas of concern m the
Great Lakes Basin m 1987 With sediment accumulating at the rate of 3 cm/year, continued
use of the harbor will require sediment removal in the near future Dredged sediments, how-
ever, would be considered too contaminated for open water disposal into Lake Ontano. Dur-
ing 1987 and 1988, L L R W M O commissioned studies of methods for dredging the harbor and
subsequent handling of the sediments and pore waters [1-3] In 1989, the town of Port Hope
asked the L L R W M O to mvesUgate the feasibility of burying the sediments m place
Purpose and Scope

The L L R W M O retained Waste Management Systems AECL for an initial evaluation of the
feasibility o f in SltU management of the contaminated sediments Information on sediment
consohdation characteristics and shear strength is central to the evaluation, and these data
were obtained through laboratory testing of conventional and large-diameter undisturbed
cores, and by field measurements A selection ofgeotextlles was tested for clogging resistance
and air entry pressure Pore waters liberated from the sediment during compaction were ana-
lyzed for arsenic, uranium, and 226Ra Conceptual methods for geotextlle placement and
gmdehnes for filling rates and material selecUon were developed based on test data and on
published and unpubhshed material Finally, environmental considerations were discussed
based on the implementation of the in sttu management option
Sediment Characteristics
Analysis and testing of the harbor sediment was carried out to determine the pore-water
quahty and quantity, load-bearing capacity, and sediment consolidation rates to examine the
feasibility of filling the harbor
Contaminants
Sediment accumulation m the harbor averages 3 cm/year with approximately 2 8 m of

water depth in the turning basin and 3 5-m depth in the approach channel The sediment lay-
enng is shown in Fig 2, with the recent sediment composed of three main layers
1 An upper layer of recently deposited soft sediment with high water and organic content,
slightly contaminated
2 A middle layer of soft silty sediment with significantly elevated levels of contamination
(the layer of concern)
3. A bottom layer o f s d t y sediment, slightly contaminated
The amounts of the three contaminants are estimated as follows [3]
1 226Ra26.7 CI
2. U r a n i u m - - 5 0 7 tonnes
3. Arsenic--58 2 tonnes
Approximately
Copyright 95%reserved);
by ASTM Int'l (all rights of these contaminants
Mon Feb 24 23:07:03 EST 2020 are in the sediments of the turning basin. Profiles
of E26Ra distribution (Fig 3) within the sediment
Universidad Industrial de Santander (Universidad Industrial de Santander) pursuant to suggest that the
License Agreement. contaminants
No further are not being
reproductions authorized.
DOLINAR ET AL, ON CONTAMINATED SEDIMENTS 269
.5
I
c~
0
8
E
n-
rL
1::
12.
9 '
jjii
E o
u~ ~o~
oJ c~
(tu) eoejJeZuTJe$eNV~Ue~J!peS MOleq qtde(:]
DOLINAR ET AL ON CONTAMINATED SEDIMENTS 271
remobflized and transported by hydraulic or chemical gradients This conclusion has been
supported by various studies on fish and biota, and pathways analysis [4-6]
Sediment Shear Strength and Consohdatton

Th~s remedml opnon involves placing cover material over the sedament to further isolate the
contaminated sediment from the human environment. The option is dependent on carefully
controlled fill placement to ensure that the load-bearing capacity of the sediment as not
exceeded at any stage The present water content of the soft sediments ~s about 80%, and the
shear strength of the sediments as estimated to be approximately 1 0 kPa (20 psf)
The parameter controlhng the load-supporting ability of the sediment as the undrained shear
strength. If filhng the harbor were attempted by direct large volume dumping from the cope-
walls into the harbor, the subgrade would be overloaded and large-scale sediment displace-
ment would result Therefore, the sediment must be loaded slowly and carefully by applying
fill-m layers over a geotexnle to slowly compact the sediments The allowable stress on a
subgrade soil, to ensure that local overstresslng in shear does not occur, can be estimated from
the following expressaon [ 7]
qa = 3 0 C (1)
where
qa = allowable vertical stress, and
C = available soft shear strength, 1 kPa
The allowable stress is 3 0 kPa (60 psf), which corresponds to a submerged hft thickness of
300 m m of soil cover The installation of the geotextlle wall further enhance the beanng capac-
ity of the soil, but this additional bearing capacity (apprommated as equivalent to an additional
300-mm hft) will be used as a safety factor for locahzed overloading condlnons
Pore- Water Charactensttcs

Undisturbed cores of harbor sediments 15 cm in diameter and up to 1 m in length were
collected from three locations in the harbor using the methods described below Pore-water
samples were taken from the ports located at the bottom, approximate center, and the sedi-
ment-water interface of each column (Fig 4) After filtration (0 45 urn) and acidification, the
waters were analyzed for arsenic (hydnde generation and atomic absorption spectrometry),
2Z6Ra (liquid scintillation analysis of the radon daughter), and uranium (radlochemical sepa-
ration and alpha spectrometry) [8] Detection limits for arsenic, 226Ra, and uranium were 0 7
ug/L, 0 17 Bq/L, and 3 ug/L, respectively Health and Welfare Canada's maximum permis-
sible concentrations (MPC) for d n n k m g water for these contaminants, in the same sequence,
Is 50 #g/L, 1 Bq/L, and 100 ~g/L Table 1 lists the mean concentrations and 1~ standard devi-
ations for the three elements in the pore waters, harbor water taken at the sediment-water
interface contained 47 ug/L of arsenic, undetectable 226Ra, and 18 ug/L of uranium, while
Table 1 lists the long-term average concentrations in the harbor Arsenic, which has a mean
pore water concentration of 14 500 ug/L, is the limiting contaminant at 290 times the MPC
High arsenic concentrations may reflect moderately reducing conditions in the harbor sed-
iments, with most of the arsenic in the more mobile As(III) valence Both the low radium and
high uranium concentranons m the pore waters could reflect the presence of h~gh concentra-
nons ofb~carbonate and carbonate ion Carbonate species would be expected to be abundant,
givenby the
Copyright ASTMhigh
Int'l (allpartial pressure
rights reserved); of24carbon
Mon Feb dioxide
23:07:03 EST 2020 (CO2) produced by biologic activity Radium
272 HAZARDOUSWASTES
Pressure & Sample Port
Wat
Lead Weig
Sand I~
Filter Fab
Harbour Sedim(
Pressure & Sample Ports
FIG. 4--Large-diameter core used for consolidation testing and evaluation of geotextiles.
carbonate has an extremely low solubility (10-30). Uranium, on the other hand, forms a variety
of dissolved carbonate complexes that enhance its mobility and solution concentration. This
discussion of geochemical controls is, however, speculative, and detailed investigations of the
reasons for the observed pore water contaminant concentration are beyond the scope of this
study.
The column consolidation tests provide information on the rate of pore water release, allow-
ing calculation of contaminant fluxes to the harbor water during filling. From the consolida-
tion behavior of the columns, the load imposed by completely filling the harbor generates a
settlement of about 0.3 m if most consolidation is limited to the upper meter of the existing
harbor sediments. This assumption is reasonable, given the marked increase in penetration
resistance at depths of about 1 m below the sediment-water interface noted during collection
of the large-diameter cores, and other drilling records [1]. Given the turning basin dimensions
of 197 by 126 m, the total volume of pore water that will be released is 7900 m 3. If we assume
similar amounts of consolidation in the west approach channel (300 by 40 m), an additional
3800 m 3 of pore water would be released. From the difficulty experienced in coring in the west
approach channel, it is probable that consolidation of those sediments would be substantially
less than settlement in the turning basin, but for now we will assume identical compaction and
pore waterby compositions for the two areas of the harbor.
Downloaded/printed
TABLE 1--Contaminant levels in pore water
Current Harbor Released Pore Water Harbor

Water Mean + Standard Concentration
Contaminant Concentration Deviation Dunng Fflhng
25 +
U, ~tg/L 15 130 + 75 63
226Ra, Bq/L 003 24 _+ 1 6 023 + 0 13
As, #g/L 20 14 500 + 8300 1200 + 690
The predicted harbor water concentrations shown in Table 1 are simple dilutions of the total
contaminant releases from the sediments d u n n g compaction into the standing water volume
of the harbor and approach channel, with no allowance for mixing and escape into Lake
Ontario, w~th the range provided by the variations m the pore water samples Actual contam-
mant levels and the possible need for water treatment would need to be studied further m e~ther
a large-scale laboratory model or field test Since prewous testing has shown that contaminants
in the approach channel are near the base of the sediment section and present at much lower
concentrations (representing only 5% of the total contaminant mventory), this second
(expected high) calculation almost certainly overpredlcts contaminant releases to the harbor
Geotextile Selection
Loading and dewatenng of the sediments is made technically possible through the use of a
geotexUle fabnc applied over the sediments that separates the fill materml from the sediment
and m this way allows the sediments to be uniformly compressed m place The use ofgeotex-
tiles to improve the load-bearing charactenshcs of soils has been well-documented [9] and is
a combination of soil separation and tensioned membrane effects, it should be noted that the
p n m a r y purpose of the geotextlle m this application is to act as a filter fabnc As the sediment
is being slowly dewatered, the liberated pore water and gases generated m the orgamc sediment
must be allowed to pass up through the geotextlle and fill material and into the harbor water
Careful selection of the geotextfle must be made so that it will contain the soft, silty, organic
sediments and yet not become clogged by them.
Physical Testing
In this feasibility study, there was no intent to carry out comprehenswe tests on the full range
of potentially suitable geotextlles available on the market Samples of fabnc were obtained
from Dominion Texnles (Mlrafi fabrics) and from Du Pont of Canada The Mlrafi samples
included woven polypropylene materials, polyester felts, and spun-bonded polypropylene, m
vanous strengths and pore sizes The Du Pont samples were a range of spun-bonded polypro-
pylene materials of different strengths and pore sizes (Table 2)
Most relevant data of fabric performance are available from the manufacturer Clogging
resistance, however, can only be determined with the actual sohd that will be m contact with
the fabnc This factor was measured usmg the gradient ratio test procedure [ 10,11] The appa-
ratus (Fig 5) is essentially a large-diameter constant head permeameter with five equally
spaced plezometer ports. After filhng the lower chamber, a perforated support plate and a disk
of the fabnc to be tested were clamped m place with the upper portion of the test column Ten
centimeters of saturated matenal from the grab sample (wet sieved through 20 mesh) were
placed on the filter fabric and the column was filled with water, capped, and connected to a
constant head
Downloaded/printed by tank Head differential across the entire column was approximately 70 cm of
E
0
- ~ ~ p ~
1111111++I
I
m
<
C~
55= = 5=~
0
.= 0
0 0 0 0 0 0 0 0 0 0
•215 ~ o <
o,
~o
c~
DOLINAR ET AL. ON CONTAMINATED SEDIMENTS 275
FIG. 5--Gradient ratio test equipment.
water. The distribution of head through the sediment and across the fabric was measured at
20 and 24 h, and the test was extended if steady state had not been achieved. The clogging
resistance tests also allowed us to measure sediment permeability, although the hydraulic gra-
dient across the sediments was much greater than that recommended by ASTM protocols
[12]. The test ratio is the hydraulic gradient between Piezometers 4 and 5 divided by that mea-
sured between Piezometers 3 and 4 (Fig. 5). Ideally, the head drop per unit length of sediment
column, and across the lowermost sediment section and filter fabric, would be uniform. More
head drop across the interval with both sediment and the filter fabric (between Piezometers 4
and 5) will be seen if fabric clogging occurs, reducing the effective permeability of the fabric to
a value lower than that of the sediments, Less head drop would be observed across the lower-
most piezometers if excessive piping of fines through the filter increased the permeability of
the sediments.
Adequate particle retention ability is conventionally assumed if a fabric is selected such that
EOSfabJD85 ~oil< 2 to 3 (2)
where
EOSrabnc = standard test measurement of the equivalent opening size in the fabric, and
[13].
D85 soilby= diameter for which 85% of the soil particles are finer
Downloaded/printed
276 HAZARDOUSWASTES
Partmle retention was also qualitatively evaluated dunng the clogging resistance tests and
the undisturbed core consolidation expenments. Permeability of the fabnc, as measured by
standard constant or falling head test, should be such that
Kfab~c > 10K~o,l (3)
where K is saturated hydrauhc conductwlty

Following the clogging resistance tests, selected fabrics were mounted m the holder illus-
trated m Fig. 4 and used for consolidation testing of the undisturbed cores The piston used to
apply load to the sediment column has a neoprene gasket that prevents sediment migration
up the walls of the core barrel The load required to overcome friction between the piston and
core barrel was measured at the start of the tests. To compress the harbor sediments, the piston
contained the following sequence ofmatenals, from bottom to top the filter fabric to be tested,
a perforated aluminum support plate, a 4-cm-thlck bed of sieved coarse sand, and a variable
stack o f cyhndncal lead weights
Geotexttle Performance
The clogging test results, although not extensive enough to be definitive, provide some gmd-
ance in selecting geotextlles (Table 2) and show that some matenals are indeed satisfactory m
their ability to retain the surfioal harbor sediments but transmit water Both woven fabrics
exhibited substantial clogging and in the test apparatus permitted transmission of readily vis-
Ible quanntles of sediment (although m percentage terms the amount of sediment passing
through the fabnc was extremely small) All of the spun-bonded fabrics were satisfactory m
their resistance to clogging, and one sample of this matenal provided a gradient ratio of only
slightly greater than 1 Performance of the felted matenals was variable, rangmg from very
good in the hghtwelght Mlrafi P50 to the worst performance of any of the matenals tested,
observed with the heavy P550 felt
The combined hydraulic conductivity of the surficlal harbor sediments and filter fabnc was
measured dunng the clogging resistance tests by measunng the flow rate and applying the stan-
dard constant head analysis One anomaly observed m the gradient ratio tests hes in the mea-
surements of hydraulic conductivity Since the procedure ts nothing more than a constant-
head permeameter test, the rate of water discharge from the columns was measured Regard-
less of whether or not the gradient ratio measurements indicated slgmficant fabric clogging,
almost all of the permeability calculations produce a hydraulic conductivity near 2 • 10 -6
cm/s Reasons for this constancy remain unclear, although it may be related to the hydraulic
gradients used for the clogging test, which greatly exceed the maxima recommended for per-
meablhty testing We have assumed that the mean value 2 6 • 10 6 cm/s is representative for
the surficial harbor sediment
The three fabrics used in the consohdatlon tests using the 15-cm-dlameter sediment cores
were all matenals that produced values of 2 5 or less m the gradient ratio tests Two (the Mirafi
T2000 and Du Pont 3407) are spun-bonded polypropylene, while the piston m the third col-
umn was fitted with Mlrafi P50, the lightweight polyester felt In all three cases, at least 50%
of the column consolidation (measured by the downward movement of the piston) was com-
plete an 24 h, and consohdat~on was at least 90% complete wnhln ten days Pore pressure dis-
sipation, on the other hand, revealed substantial differences between the three tests Figure 6
displays the records of pore pressure versus time. On the figure, vertical solid lines mark the
times when weights were added to the columns, and the applied loads are listed along the top
of the graphs Pore pressures are reported as kPa in excess of the head applied by the standing
water
Copyright column
by ASTM Int'l (allabove the sediments.
rights reserved); The
Mon Feb 24 23:07:03 ESTDu
2020 Pont 3407 fabnc provided the most rapid dissi-
pation m pore pressure, qmckly returning to hydrostatic except after the last load application
~0
0
'0
:!ili!ii: . ,
t
: 9
:
;
,
r-
"
"" i~' "'~'*

~ " '" ....
*~'" "" "%*"
: : :
i ~ 9
..... i ....... Ii .......
: : ii : . ..
: . 9 , .I")
9 : 9 9 ; IC~I
::.: : ~
,i,i,l,i, i,i,i,i, i,i,i, z
i
(Ecl~) eJns~eJd eJod (~d~l) eJnsseJd eJ0d (Rd~) GJrl$$gJd 0J0 d
The Mlrafi P50 exhibited slow pore pressure dxsslpatmn, and the column fitted with the T2000
fabric appeared to maintain an excess pressure of 5 kPa All of the columns generated gases
during the compactmn experiments; although d~rect measurements of gas generatmn rates
were not undertaken, we estimated a productmn rate of 0 3 m L day -~ cm 2 In all cases, bub-
bles were observed to escape through the filter fabnc and sand bed of the load piston, but for
the T2000 fabric, there was always a substantial accumulatmn of gas beneath the piston The
5-kPa excess pore pressure maintained by the T2000 fabric represents its air entry pressure
under m sltu condmons Gas transmission capablhty should be considered when selecting fab-
rics for apphcatxon over bmdegradable materials
The most interesting aspect of the results of the consolldatmn tests conducted in the 15-cm-
diameter columns was the influence of mr entry pressure for the three fabrics tested Consoh-
datlon as measured by the downward movement of the load piston was quite similar in all
three cases
Conceptual Design
The following m a description of the m s t t u remedial option involving containment of the
sediment by geotextfles and fill (gravel and general fill) for Port Hope harbor (Fig 2) in the
following sequence A geotexUle (filter fabric) will be placed directly on the harbor sediment
The geotextlle will be prefabricated m sheets as large as can be handled, no more than three
sections The sheets can be pulled out into the harbor by dragging them with ropes that have
been sewn m to assist with placement and anchored with sandbags Accounting for overlap
and seaming, 45 000 m 2 ofgeotextlle will be required to cover the turning basra and approach
channel
After the geotextfle has been placed, the first lift (layer) of gravel can be applied (piped slurry
or by a barge mounted spreader box) The speofied "pea gravel" (or crushed stone between 5
and 20 mm) will allow ready drainage of the pore waters being pressed for the sediments below
and will provide a high shear strength layer for further addmons of fill Slow, even placement
will ensure that dlfferentaal filling does not shear the fabric and underlying harbor sediments
To provide a 300-ram hfe throughout the turning basra and approach channel, 11 000 m 3
(20 000 tonnes) of pea gravel is reqmred
The established base of geotextlle and gravel allows subsequent hft thicknesses to increase
A second layer of sand and gravel, 700 m m (2 fl) think, can be pit run material as long as the
m a x i m u m rock size is less than 150 m m m dmmeter This hft will further consohdate the
underlying sediments while also contributing to a base of good shear strength for the general
fill layer yet to come The volume of fill for this hft is 25 000 m 3 (46 000 tonnes)
The geotextlle and 1 m of gravel will form a base capable of suppomng the load imparted
by the general fill, which will be used to complete the filhng of the harbor Ideally, general fill
will come from construction projects m the Port Hope area Approximately 1 5 m of water
will be present m the turning basra and 2 0 m m the approach channel, and filhng to a height
1 8 m above the water surface is reqmred for final gramng Approximately 141 000 m 3 of gen-
eral fill will be reqmred on top of the gravel, of which about one half (76 000 m 3) will be
emplaced below the current water level This will be apphed by dumping, m staged hits, from
the inland copewall moving across the harbor Once the harbor is filled, final contouring of the
approximately 36 000 m 3 (3 6 ha) of reclmmed land can take place
Environmental Considerations
This section identifies many of the possible impacts on the surrounding environment result-
lng from
Copyright by ASTMthe remedial
Int'l (all action
rights reserved); Mon Febfor Port EST
24 23:07:03 Hope2020 harbor The following impacts are discussed
briefly,
Universidad addressing
Industrial de Santanderseventy
(Universidadand indicating
Industrial de Santander)possible mltlgatmg
pursuant to License Agreement.measures
No further reproductions authorized.
Sediment Suspenston Durmg Fdlmg Operattons

D u n n g the placement o f the xmtial layers of fill over the geotextfle, a significant amount of
fine particulate from the fill will become suspended. These suspended particles will be pre-
vented from leaving the harbor by providing a silt curtain at the channel entrance, therefore,
outside the approach channel entrance, on the lake side of the silt curtain, there will be no
effect from the suspended particulate, only clean fill will be used so that the problem of sus-
pended particulate is of a temporary aesthetic nature within the confined area
Pore Water Release from the Sedtment

During filling operations, contaminated pore water will be released and will mix with the
harbor water The resulting concentrations of contaminants will need to be monitored closely
during operations, however, our laboratory studies indicate that only arsenic concentrations
will be of some concern Dilution of the pore water with the 120 000 m 3 of harbor water
will help to reduce arsenic concentrations It should be added that the pore water released
from the sediments will become pore water in the added fill, so its impact will be greatly
reduced
Disturbance of Sediments
The sediments as they exist will not be disturbed (consolidated in place as opposed to
moved), thereby eliminating possible contaminant mobilization
Worker Exposure
Worker exposure during remedial measures will be minimized since no handling of the con-
tamlnated sediments will take place
Long- Term Effects

Site Control--None of the three remedial options considered here could be considered as
walk-away options, that Ls, some form of institutional control (registered site with Atomic
Energy Control Board, AECB), and an ongoing monltonng program would be ~mplemented
Mlgranon of Contammants--One of the long-term considerations is the impact the con-
tammated sediments will have m the future From a health and safety point of view, even
though the sediments are radioactive, they are low in activity These sediments pose no threat
except the threat of being disturbed during dredging operations if the harbor continues to be
used as a boating facility The contaminated sediment has been confined for many years, a
significant portion o f l t being deposited over 50 years ago, and during that time period it has
remained confined, covered by only a thin layer of sediment The addition of more fill will act
only to further enhance the isolation
Radon--Migration of radon-222 produced by the radium in the sediment will be minimal
because of the thickness of water-saturated fill and due to the soft organic upper layer of sed-
iment, both of these are good bamers (due to the short half-hfe ot radon, 3 8 days) and have
been shown to retard radon movement [14] The slow movement of radon-222 through the
sediment and soil will provide ample time for radioactive decay processes to produce daughter
radionuclides (nongaseous) that will become trapped m the fill and sediment Quantitative
values
Copyright for Int'l
by ASTM radon flux
(all rights ratesMonwould
reserved); require
Feb 24 23:07:03 further study Some restrictions would apply to
EST 2020
postremedlal
Universidad Industrial de uses
Santanderto(Universidad
preventIndustrial
inadvertent intrusion
de Santander) into Agreement.
pursuant to License the contaminated layerauthorized.
No further reproductions
Outfalls and Intakes tn the Harbor

Outfalls and intakes m the harbor wdl need to be rerouted when this remedial action Is
implemented In add~uon to the CAMECO retake and outfalls, a storm water condu~t outfall
(and creek) receive water from the 2-kin2 [3] drainage areajust north of the harbor and can be
rerouted easdy to the Ganaraska River, w~th httle ~mpact aside from marginally increased flow.
(Increased flow to the river would be unnoticeable and the water quahty of the creek and storm
water condmt are the same as other nearby trabutanes, since they are geographically w~thm the
same area drained by the river )
Future Stte Uses

Once the remedial action process ~scompleted, future site uses wdl need to be reviewed since
some uses would not be practical on the site The contaminated sediment wdl be covered by
no less than 4 5 m of fill, which will reduce radioactivity from the confined sediment to back-
ground levels at the surface of the site The reclaimed area will have hm~ted uses because of its
location and close proxlmRy to mdustnal areas Possible uses include parkland with walking
paths, tennis courts, or any nonenclosed building structure that does not require excavation
for foundations of more than approximately 0 75 to 1 m xn depth Such possible buddlngs
might include storage facdmes (dry-dock boat storage, town eqmpment yards) and other s~m-
dar uses
Groundwater Flow
The anticipated flow hnes for groundwater near the harbor follow the path of least reszst-
ance, which, m this case, is towards the Ganaraska River for areas roland of the harbor and
d~rectly to Lake Ontario for areas closer to the lake The setting and apparent lack of contam-
inant transport from the sediments into the harbor water both indicate neghg~ble groundwater
flow under current cond~txons Quantification of the groundwater flow system before and after
remediaUon would require addmonal study Unacceptable groundwater flows through the
contaminated sediments could be controlled by impermeable barraers or drainage systems
Surface Water and Prectpltauon

Surface water can come from two sources, precipitation directly on the site, and surface
water flowing over the site Penetration of surface water into the site should be mlmmlzed to
avoid leaching of contaminants from the site
Surface water flowing towards the s~te will be intercepted by the addition of a french dram
or ditch along the north edge of the harbor, which xn turn wdl dram into the raver Site closure
uses wdl d~ctate some of the water infiltration characterasucs It is important to note that this
~s not a large site, and the contaminated sediments are currently saturated and show httle s~gn
of mobxhzat~on; therefore, a small a m o u n t of water entering the s~te wdl have httle effect To
minimize infiltration, the top layer of fill materaal will be contoured to encourage preclp~tat~on
runoff as opposed to penetration, th~s contoured layer wdl not reqmre special materials but
will consist of the general fill and topsoil A final layer oftopsod seeded with grass to prevent
erosion and promote transp~raUon of water penetrating the surface ~s recommended
Intrusion
One of the major considerations of any site containing potentially dangerous wastes is pro-
tectlon from
Downloaded/printed by inadvertent contact with the waste by man, ammals, and plants Someform of
institutional control will be in place at the harbor site classifying the site as a restricted use area
and the reclaimed harbor site will be a registered/licensed site with the AECB and the Land
Registry Office with estimates of the a m o u n t and type of contained contamination Intrusion
into the site by plants or animals is highly unlikely The primary defense b a m e r is the 4 5- to
5 5-m depth of clean fill cover above the contaminated sediments The 3-m saturated zone
thickness of clean fill above the contaminated sediment provides an effective barrier to root
and animal Intrusion
Seismic Events
Port Hope is situated in an area of moderate to low seismic activity on the west arm of the
St Lawrence Platform, so the probability of the area experiencing seismic activity is corre-
spondingly low The primary consideration in the event of a significant seismic event is the
possible liquefaction of the underlying sediments and the possible consequences liquefaction
might have on structures existing on the reclaimed land Before constructing any structure on
the reclaimed land, the liquefaction scenario would require further study (which would also
be a requirement of the building codes)
Summary and Conclusions

Our initial investigations have indicated that the m sltu management of the Port Hope har-
bor contaminated sediment is a feasible remedial option that should be considered further
with specific regard to those areas that require further study The load-bearing capacity of the
sediment is sufficient (once consolidated) to support the load imparted by the required fill,
provided filling operations take place in the specified manner Furthermore, sediment con-
sohdatlon takes place quickly, giving the project a reasonable time frame Testing geotextlles
for sediment retention and ability to pass pore water d u n n g consohdatlon has shown that suit-
able geotextlles are commercially available
The major environmental concerns are those arising from the liberated pore water The con-
centration of contaminants (arsenic in particular) In the harbor as filling takes place can only
be approximated at this, the feasibility stage A full-scale test of the envisaged methods would
be most useful in verifying the proposed methods as well as providing data on actual contam-
inant levels that would be expected d u n n g the remedial operations, allowing mitigating mea-
sures to be developed and implemented if necessary
References
[1] "Port Hope Remedial Program, Conceptual Engineering Design Report for the Harbour Cleanup--
Report No 3," report for the Low Level Radioactive Waste Management Office, MacLaren Engi-
neers, Toronto, Ontario, Canada, July 1987
[2] "Port Hope Remedial Program, Radlologlcal Assessment and Conceptual Design for the Cleanup
of Major On-land Areas of Contamination--Report No 4," report for the Low Level Radioactive
Waste Management Office, MacLaren Engineers, Toronto, Ontario, Canada, July 1987
[3] "EnvwonmentalAssessment of the Port Hope Remedial Program," report for the Low-LevelRadio-
active Waste Management Office, Senes Consultants Limited, Wfllowdale,Ontario, Canada, Octo-
ber 1987
[4] "Benthologxcal, Chemical, Radlologlcal and Chronological Evaluat,on of the Sediments In Port
Hope Harbour," report for Environmental Protechon Service, Ontario Region, Beak Consultants
LTD, Mlsslssauga,Ontario, Canada, August 1985
[5] "Summary of Investigations into Radloactxve Contamination in the Sediments of the Port Hope
Harbour Turning Basin, Port Hope, Ontario," a brlefsubm,tted to the Federal Provmoal Task Force
on Radioactivity, Environmental Protection Service, Ontario Region, Toronto, Ontario, Canada,
October
Downloaded/printed by 1985
[6] "A Discussion of the Environment Canada Investxgatlon into Radlonuchde Levels in Fish Collected
from Port Hope Harbour," Environmental Protection--Ontario Region, Environment Canada,
Toronto, Ontario, Canada, December 1986
[ 7] Jarret, "Geotextdes m Building Sites and Storage Faohtles," Proceedings, Symposium on Use of
Geotextdes m Practice, Canadian Geotechmcal Society, Toronto, 1984
[8] "Assessment of the Feaslblhty of in sltu Management of Port Hope Harbour Contaminated Sedi-
ments," report prepared for the Low Level Radioactive Waste Management Office, Atomic Energy
of Canada Ltmlted, Chalk Rtver Laboratories, Chalk River, Ontario, Canada, January 1991
[ 9] Fluet, J E m GeotexttIe Testtng and the Destgn Engmeer, ASTM STP 952, American Sooety for
Testing and Materials, Phdadelphla, 1985
[10] Calhoun, C C, "Development of Design Criteria and Acceptance Speoficatlons for Plastic Filter
Cloths," Techmcal Report F- 72-7, U S Army Engmeenng Waterways Experimental Station, Vicks-
burg, MS, pp 19-52, 1972
[11] Hahburton, T A and Wood, P D , "Evaluation of the U S Army Corps of Engineers' Gra&ent
Ratio Test for Geotextde Performance," proceedings of the Second International Conference on
Geotextlles, Las Vegas, NV, Geotextdes, Vol 1, 1982, pp 97-102
[12] ASTM Standard Test Method for Permeabdlty of Granular Soils (Constant Head) (ANSI/ASTM D
2434-68, reapproved 1974)
[13] Carroll, R G , "Hallburton Gradient Ratio Testtng," Mtrafi Techntcal Note 83-01, Mxrafi Inc,
Charlotte, NC, Jan 1983
[ 14] Michel, J, "Sources," Envtronmental Scwnce Research, Vol 35 EnvtronrnentalRadon, Chapter 4,
Plenum Press, New York and London, 1987, pp 81-126
Hang-Sik Shin, 1 Jong-Oh Kim, 2 Ja-Kong Koo, 1 and Eung-Bal Shi n 3
Mixture Design Optimization Considering

Three Disposal Options for Solidified Heavy
Metal Sludges
REFERENCE: Shin, H -S, Kam, J -O, Koo, J -K, and Shin, E -B, "Mixture Design Optimi-
zation Considering Three Disposal Options for Solidified Heavy Metal Sludges," StablhzaHon
and Sohdtficatton of Hazardous, Radioactive, and Mixed Wastes, 2nd Volume, ASTM STP
1123, T M Gllham and C C Wdes, Eds, American Society for Testing and Materials, Phda-
delphm, 1992, pp 283-296
ABSTRACT: This research evaluated the feaslblhty of three different &sposal opUons such as
landfill, reuse m construction, and ocean d~sposal for sohdlfied materials containing synthetic
heavy metal sludges--copper and chromium Three factors--binder/sludge, fly ash/binder, and
sodium silicate (Na2S103) sludge ratios--were investigated m terms of physical and leaching
properties Constdenng the actual &sposal environment, three different leachants, namely,
acl&c water, distilled water, and seawater, were used in the leaching tests
The inhibition of cement hydration by copper sludge slgmficantly affected the physical prop-
ertleS of solidified matenals Fly ash additions deteriorated the overall properties of sohdlfied
material, but the use of a sodium slhcate solution had no significant effect Leachate character-
1StlCSafter 24-h extraction were strongly affected by the nature of leachant, and leached metal
concentrations were well matched w~th theoretical solubihty according to leachate pH A com-
parison of leaching characteristics indicated that acidic water was the most aggressive toward
leaching the sohdlfied matrices and that distilled water was shghtly more aggressive than sea-
water Leachate alkallmty after seawater leaching decreased less than initial seawater alkahmty,
which would he caused by carbonate reflux and magnesmm hydroxide [Mg(OH)2] formatmn
onto sohd~fied matrices
Results of tnphcate tests and statistical analysis in&cared that the hinder/sludge ratlo had the
greatest effect on physical properties while the type ofleachant had the greatest effect on leaching
properties M~xture design optimization using linear programming was attempted for the eco-
nomic comparison of three dmposal options As a result, ocean &sposal as well as reuse in con-
struction were estimated as the preferred disposal alternatives for solidified heavy metal sludge
KEY WORDS: hazardous materials, wastes disposal, leaching, soll&ficatlon, heavy metals, alka-
hmty, m~xture design optimization, statistical analysis
L a n d disposal has b e e n a m a j o r t e c h n o l o g y for s o h d waste m a n a g e m e n t owing to its eco-

n o m i c , p e r m i s s i o n , a n d o p e r a t i o n a d v a n t a g e s However, limited sites a n d s t n c t s t a n d a r d s for
landfilhng are b e i n g b r o u g h t to b e a r o n t h e use o f land for disposing o f h a z a r d o u s wastes
W h e n m d u s t n a l sludge c o n t a i n i n g h e a v y m e t a l s is landfilled, it m i g h t release toxic metals to
Associate professor, Department of Civil Engineering, Korea Advanced Institute of Science and Tech-
nology, P O Box 150, Cheong Ryang, Seoul, Korea
2 Postdoctoral researcher, Waste Control Laboratory, Korea Institute of Science and Technology, P O
Box 131, Cheong Ryang, Seoul, Korea
3 Professor, Department of Cwll Engineering, Hanyang University, Haengdang Dong, Sungdong Ku,
Seoul, Korea
283

the ground environment, contaminating its soil and water. To make the sludge acceptable for
safe land disposal, solidification technology employing various cementitious binders has been
developed. The binders' aim is to improve the handling and physical characteristics of haz-
ardous waste as well as to retard the pollutants from leaching. A detailed review of the existing
processes has been reported [ 1-4]
In addition to landfillxng, the solidified materials can be disposed of in other ways For exam-
ple, the solidified blocks of titanium metal processing waste could be dumped into the ocean
[5] Chemical and physical tests indicated that these blocks were amenable to ocean displace-
ment without adverse environmental impact Sandblasting residues that contained cadmium
and lead were used as an admixture for concrete [6] Up to 15% of the waste was added to
concrete without leaching problems Also, Shin et al. [ 7] investigated the factors of a cement-
based solidification process designed for the hazardous sludge produced from the electroplat-
ing industry. Proper mixing criteria were suggested for the use of the solidified mixture as con-
struction materials.
In using the sohdlfiCatlon process, leaching and strength tests have been developed to sim-
ulate an actual disposal environment For example, acetic acid solution simulates the leaching
situation under codisposal of industrial wastes in municipal landfills and has been used as a
leachant in U S. Environmental Protection Agency (EPA) extraction procedure (EP) toxicity
tests [8] However, the exposure of sohdlfied wastes to weak organic acids has been found to
dissolve major constituents of cement paste, thus accelerating the release of toxic heavy metals
from the solid matrix [9,10] Based on these findings, it is possible that long-term exposure of
solidified wastes to acidic groundwater may result In detrimental leaching of metals
This research attempted to evaluate the feasibility of three different disposal options for
solidified heavy metal sludges: (1) landfill, (2) application as construction materials; and (3)
ocean disposal. To study the leaching characteristics of the solidified waste matrix under the
different disposal scenarios, three leachants--acidlc water, distilled water, and seawater--were
used for the 24-h extraction test Mixture design optimization considenng the ultimate dis-
posal options was attempted using the factorial experimental design of mixture composition
factors Laboratory experiments were conducted with two kinds of synthetic sludges contain-
ing copper and chromium, the dominant heavy metal species generated from the industries.
Materials and Methods

Synthetic Sludge
Two synthetic sludges were prepared, one for copper and one for chrommm Each sludge
was prepared by precipitating the metal hydroxide from an aqueous salt solution, cupric chlo-
nde (CuC12) for copper and chromium trioxide (CrO3) for chromium, by sodium hydroxide
(NaOH) addition. Sodium bIsulfite was added to chromium solution to reduce the chromium
from the hexavalent to the trlvalent state p n o r to precipitation The water content m the sludge
sample was adjusted to 75%. Copper content of one sludge was 74 0 mg/g of wet sludge, and
the chromium content of the other was 27 5 mg/g of wet sludge. A mixed heavy metal sludge
was also prepared by blending these two synthetic sludges in equal proportions
TABLE l - - F l y ash composttton, %
SIO2 A1203 Ee203 CaO IgnlUon Loss

556 264 53 002 49
SHIN ET AL ON HEAVY METAL SLUDGES 285
TABLE 2--Mtxture ratto destgn on wet wetght basts
Mtxture Ratto Level 1 Level 2 Level 3
Bmder/sludge (B/S) 05 1.0 15

Fly ash/bmder (F/B) 00 0 25 05
Slhcate/sludge (St/S) 00 0 025 0 05
Mtxture Design and Leachant

Type I portland cement and a fly ash were used as soh&ficatlon binders while a 25 wt%
so&um silicate (Na2S103) solution was added The fly ash was obtained from a coal power
plant, and its major composmon is presented m Table 1. Heavy metal contributions from the
addmves were negligible.
The mixture ratio design and the leachant type were selected as the major experimental van-
ables. In the mixture design, three factors, B/S (binder/sludge) ratto, F/B (fly ash/binder) rauo,
and S1/S (silicate/sludge) ratio, were investigated Under the respechve condmons as shown m
Table 2, experimental specimens were prepared according to a 3 • 3 • 3 factorial design
Samples not containing heavy metal sludge were also prepared as controls in the water/binder
(W/B) ratio of 0 5. Considering the actual &sposal enwronments, three leachants of ac~&c
water (AW), distilled water (DW), and seawater (SW) were used, and their characteristics are
shown m Table 3
Phystcal and Leachmg Test

All batches were prepared using a mortar mixer by adding the mixture components m the
following order and mixing between each addmon, synthetic sludge or water, sodmm slhcate
solution, fly ash, and cement The m~xtures were poured into 50-ram cubic molds for hard-
enmg and demolded after 14-days cunng m a damp chamber Physical properhes ofsohdlfied
waste, including final setting Ume, bulk density, and unconfined compresswe strength, were
measured The final setting Umes of fresh mixture were determined by the ASTM Test for
Time of Setting of Hydrauhc Cement by Vlcat Needle (C 191-79). All 14-day cured specimens
in cubic shape were weighed to calculate their bulk density Fourteen-day unconfined com-
pressive strengths of the sohdlfied wastes were measured according to the ASTM Test Method
for Compressive Strength of Hydrauhc Cement Mortars (Using 2-m or 50-mm Cube Speci-
mens) (C 109-86)
The leaching test followed the general gmdehnes of the U S EPA EP toxicity test with three
different leachants. Prior to this, the soh&fied samples were crushed with a mortar and pestle
TABLE 3--Leachant characteristics

Charactensttcs Acid Water Distilled Water Sea Water
Composttton (mg/L) 0 1 Nacetlc acid Na + (t0 750) CI (19

solution 340)
Mg +2 (1290) SO~-2 (2710)
Ca +2 (410) HCOj- (140)
K § (400)
pH 28 55 75
Alkahmty (meq/L) 00 02 25
Acl&ty (meq/L) 78 69 72
to a particle size of 2 to 9.5 mm. It was felt this procedure would result in a worst-case scenario
of maximizing surface area available for leaching. In polyethylene bottles, 10 g of sample and
200 m L of leachant were placed. After 24-h of mechanical shaking, the leachates were filtered
through 0.45-#m filters using a vacuum filtration apparatus. The resulting filtrate was then
divided into two subsamples: one was measured for pH and alkalinity content, and the other
was acidified with nitric acid for the calcium and metal analyses by a flame atomic absorption
spectroscopy (AAS). The detection limits of the AAS were 10 mg/L for calcium, 0.05 mg/L
for chromium, and 0.01 mg/L for copper.
All experiments were conducted in triplicate test runs. Collected data were statistically ana-
lyzed using analysis of variance (ANOVA) and multiple linear regression methods with the
S P S S / P C + computer package [11].

Physical Properties
The final setting time of the fresh mixtures is illustrated in Fig. 1. It was observed that copper
sludge considerably retarded cement hydration, which delayed detachment from the mold
form. But the final setting of mixtures containing chromium sludge was completed within 7 h
of the same time as that of control samples. Many researchers [12,14] reported that copper,
lead, and zinc compounds inhibit the initial hardening of cement while cadmium, chromium,
and mercury compounds have a negligible effect on the hardening property of cement. They
suggested that the inhibition effect of copper, lead, and zinc to cement hydration may be
related to the chemical reactions, forming gelatinous coating on the major cement compo-
nents and acting as a diffusional barrier to water. In both copper and chromium sludge, an
increase of the B/S ratio significantly shortened the setting time. When the F/B ratio increased,
the setting time in the low B/S ratio mixture increased while that in the high B/S ratio was
hardly affected. But the addition of sodium silicate had no significant effect on the setting time
of fresh mixtures.
As shown in Fig. 2, the measured bulk density of solidified materials varied between 1.17
and 1.74 g/cm 3 in copper sludge and between 1.37 and 1.83 g/cm 3 in chromium sludge. The
ca
solidified chromium waste was slightly denser than copper waste. The bulk density of control
samples was smaller than that of the solidified waste with the same W/B ratio. An increase of
the B/S ratio had a significant effect on increase of bulk density, while an increase of either the
100
t_
Cr Control
80
~" 60
40
-~ 20
.=_
0
0.5 1.5 0.5 1.5
Binder/Sludge Ratio
(Fly ash/Binder Ratio: [ ] = 0 =0.25 g~ = 0.5)
Copyright by ASTM Int'l (all rights reserved); Mon Feb 24 FIG. EST 2020 setting time.
23:07:031--Final
i
o ~
2.0
1.6-
1.2-
0.8-
Ca Cr]H
SHIN ET AL. ON HEAVY METAL SLUDGES
Control
287
0.4-
0
0.5 1.5 0.5 1.5
Binder/Sludge Ratio
(Fly ash/Binder Ratio: ~ = 0 • =0.25 W'A= 0.5)
FIG. 2--Bulk density.
F/B ratio or the Si/S ratio slightly decreased bulk density. The waste volume change factor
(VCF) could be calculated in the following way for each batch [15]
~d~
VCF = - - - -
wrd,
where
Wr = weight of raw waste used to prepare the batch, g,
W, = weight of the solidified waste batch ( IV, + weight of the additives), g,
dr = bulk denisty of the raw waste, g/cm 3, and
ds = bulk density of the solidified waste, g/cm 3.
Based on the bulk density values of 1.16 and 1.26 g/cm 3 in the synthetic copper and chro-
mium sludge, respectively, the VCFs varied over a range of 1.43 to 1.85 in solidified copper
waste and 1.33 to 1.84 in chromium waste.
Figure 3 shows that the 14-day compressive strengths varied widely over a range of 8 to 340
kg/cm 2. The strength of control samples was smaller than that of the solidified chromium
sludge with the same W/B ratio. Due to the cement hydration inhibition of copper sludge, the
strength of solidified waste was also in about half the range of that of chromium waste. An
Cu Cr Control
300
200
~ 100-
I ,
0.5 1.5 0.5 1.5

Binder/Sludge Ratio
(Fly ash/Binder Ratio: ~ = 0 ml =0.25 W)I= 0.5)
3--Fourteen-day unconfinedcompressivestrengths.
Copyright by ASTM Int'l (all rights reserved);
FIG. Mon Feb 24 23:07:03 EST 2020
288 HAZARDOUSWASTES
increase in the compressive strength corresponded to an increase in the bulk density that
resulted from a decrease of pore volume [16] Accordingly, effects of the B/S, F/B, and S1/S
ratios on the compressive strength had a similar trend as on the bulk density But the addition
of sodium silicate had an adverse effect on strength due to the increase of water content In the
solidified materials
Leachate pH, Alkahmty, Calcmm Concentrations

As shown in Fig 4, leachate characteristics such as final pH, alkahnlty, and calcium con-
centratlon after a 24-h extraction period depended mainly on the type of leachant All these
leachate charactenstlcs were significantly affected by the F/B ratio more than the B/S ratio
The fly ash additions resulted in a decrease ofleachate pH, alkahnlty, and calcium concentra-
tion due to the decreased lime production for cement hydration Effects of the B/S ratio on
leachate characteristics had a different trend according to each leachant, but the SI/S ratio had
a negligible effect
13
T~
11.
e~
9.
79
5-
120 9
---__y
80,
-~
<
40. T
.L /~ x
200~]
100•
.x.~ ~ "' .t.'7-

I I I
0.0 0.25 0 5
Fly ash/Binder Ratio
(Leachant 9 A W , 9 D W , 9
FIG 4--Leachate charactertstws
Leachate pH after 24-h extraction under acidic water was in the range of 5 1 to 6 7, while
the leachate alkalinity and calcium concentrations were 80 to 120 meq/L and 750 to 2300 mg/
L, respectively Due to the inhibition effect of copper sludge on cement hydration, the alkahn-
ity release from solidified copper wastes was relatively smaller than those from solidified chro-
m i u m waste, whereas the leached calcium concentrations were greater It seemed that the low
alkalinity contents combined with higher calcium concentrations were caused by the
decreased lime production and the increase of leachable calcium (unhydrated cement) result-
ing from poor cement hydration of copper sludge For the same reason as above, leachates of
control specimens showed the highest pH and alkalinity contents but the lowest calcium con-
centrations among the leachates under acidic water An increase of the B/S ratio, however,
resulted in an increase ofpH, alkalinity, and calcium concentration due to the enhanced effect
of cement hydration in both solidified chromium and copper wastes
Leachate pH after distilled water extraction varied in the small range of 11 9 to 12 3, whereas
alkalinity and calcium concentrations were 20 to 50 meq/L and 20 to 330 mg/L, respectively
An increase of the B/S ratio resulted in a decrease of alkalinity release because the dis-
tilled water could not attack effectively the solidified materials with a high B/S ratio Heavy
metal sludge type had a similar effect on leachate characteristics as in acidic water extrac-
tion
Leachate pH after seawater extraction was in the range of 7 7 to 9 0, while alkalinity and
calcium concentrations were 0 6 to 1 4 meq/L and 430 to 860 mg/L, respectively Therefore,
leachates showed higher pH and calcium concentrations than the seawater leachant with pH
7 5 and 410 mg/L of calcium, whereas the alkalinity contents decreased less than that of
leachant, 2 25 meq/L Campbell et al [17] had observed similar results from a study on leach-
ing characterlstxcs ofsohdlfied heavy metals under seawater They suggested that the influx of
carbonates, a major constituent contributing to seawater alkalinity, to form calcium carbon-
ates decreases the leachate alkalinity In spite of raising pH In this study, 38 to 73% of seawater
leachant alkalinity decreased after 24-h extraction Increase of the F/B and B/S ratios raised
both leachate alkalinity and calcium concentrations, probably due to a decrease of carbonate
Influx into solidified materials
Acid neutralization capacity (ANC) is defined as the amount of a strong acid required in a
titration to reach an equivalent point One of the most important ANC contributing additives
in waste solidification is portland cement [18] The alkalinity release from solidified waste
could be related to ANC, while the release of calcium could be related to the extent of matrix
decomposition Therefore, it was found that considerable portions of ANC and calcium were
leached out from the solidified matnx under acidic conditions, while the most was contained
m the matrix under distilled and seawater conditions These results indicated that acidic water
is the most aggressive towards leaching the hydrated matrix, and that distilled water is more
or less aggressive than seawater
Heavy Metals Leaching

As shown in Fig 5, soluble metal concentrations in leachate were well matched with their
theoretical solubilities according to leachate pH. Leached chromium concentrations were 0 8
to 7 3 mg/L at the acidic range, but below 0 2 and 0 1 mg/L in distilled water and seawater,
respectively Those chromium concentrations were found to be slightly below solubility of
chromium hydroxides (Fig 5a) Meanwhile, copper concentrations leached In acidic water
were considerably high in the range of 90 to 770 mg/L, indicating that copper easily dissolved
out to leachant from the solidified matrix In the distilled water and seawater, copper concen-
trations were low, in the range of 0 2 to 0 5 mg/L and 0 05 to 0 15 mg/L, respectively Corn-
290 HAZARDOUS
WASTES
g
Hydroxdi~es
-5
A
-7-
g
-5
"", O
ii1 Hydroxides',,... bonates
9"-. . . . . . . . . . '
I I I I I I 1 I I
4 6 8 10 12 14
pH
o : Metal concentration in B/S=0.5

: Metal concentration in B/S=1.0
o : Metal concentration in B/S=1.5
- - : Theoretical solubility considering
chemical speciations in distilled water
FIG. 5--Leached metal concentrations and solubilities.
pared with the copper hydroxide solubility, however, copper concentrations in seawater were
very high. As a result, copper leached under seawater seemed to be in the form of soluble car-
bonate complexes as shown in Fig. 5b [19].
In both solidified chromium and copper wastes, an increase of the B/S ratio resulted in
decreased metal concentrations under all three leachant conditions. However, the F/B ratio
had a negligible effect on the leached metal concentrations in distilled water and seawater,
whereas it had a significantly adverse effect in acidic water. Meanwhile, the Si/S ratio had a
small and inconsistent effect on heavy metal leaching.
SHIN ET AL. ON HEAVY METAL SLUDGES 291
FIG. 6--Scanning electron m icroscopephotograph ofleached specimens: (a) leached in acidic water, fo)
leached in distilled water, and (c) leached in seawater.
Surface Analysis Using Scanning Electron Microscopy (SEM) and X-Ray Diffraction (XRD)
Figure 6 shows microscopic surfaces of leached specimens. Specimens leached in acidic
water showed that only the skeletons of hydrates remained (Fig. 6a). When leached in distilled
water, surfaces of solidified sludge were covered with C-S-H and fly ash grains (Fig. 6b). Solid-
ified matrices leached in seawater resulted in surface coating (Fig. 6c).
Figure 7 shows XRD results after a 24-h extraction test. In Fig. 7a showing leaching results
in acidic water, lime and C-S-H did not exist any longer while only matrix skeltons such as
ferrite (C4AF) remained. When leached under distilled water, lime did not exist while C-S-H
and calcite existed as shown in Fig. 7b. Figure 7c indicated that, in seawater, lime dissolved
out and magnesium hydroxide crystals formed and coated on specimen surfaces.
Statistical Analysis and Engineering Application

Analysis of Variance
In this study, conducted under factorial design with three levels and triplication of test runs
[20], the analysis of variance (ANOVA) was performed to estimate the effects of experimental
factors on physical and leaching properties. From these ANOVA, the main effect was found
to be more significant than the interaction effect. As shown in Table 4, the decreasing order of
TABLE 4--ANOVA results.
F Statisticsa
Measured
Properties A B C D E Main
Setting timesb 96* 85* 4.3 0.1 ...c 31"

Bulk density 953* 2854** 88** 0.7 ... 977**
UC strength 728* 1442"* 75** 43* ... 550**
pH 0.67 29* 98* ... 2102** 637**
Alkalinity 73* 9.4* 164** ... 19 759** 5705"*
Calcium 743** 8.5* 233** 7.9* 15 562** 3596**
Heavy metals 569* 39** 51"* 0.2 585** 213"*
A = heavy metal type, B = B/S ratio, C = F/B ratio, D = Si/S ratio, and E --- leachant type.
b Significance of F statistics (blank = above 0.000, * = 0.000, and ** = 0.0).
c Unknown.
.5K
.15
20.00 a.O. 00 BO. 00
9 5K
CaC(~
.15
CaC03
s-,& ".=~ "- " "/- - -~-

....L L ,
28. BO 4fl. OO 60.00
.51{
~(oH)2
.15 m
2~. 08 4~. 00 60.,00
FIG. 7--X-ray diffraction analysis of leached specimens: (a) leached in acidic water, 00) leached in dis-
tilled water, and (c) leached in seawater.
sigmficant effect for physical properties was the B/S ratio, F/B ratio, heavy metal type, and S1/
S ratio. For leaching properties, the order of significant effect was the leachant type, F/B ratio,
B/S ratio, heavy metal type, and SffS ratio Therefore, leaching properties were found to be
sigmficantly affected by the leachant type rather than mixture ratio design factors
Multiple Linear Regression

Using each weight ratio of solidification agents to heavy metal sludge, multiple linear regres-
sion was performed to project the measured value of several properties into hneanzed equa-
tions. The observations were descnbed by the following multiple linear regression model
Y = ~o + B,X, + BzX2 + [3Q(3 + e
where
B = regression coefficient,
e = random error component N(0, a2E), and
XI, )(2, and X3 = regression vanables for cement/sludge, fly ash/sludge, and Na2SIO3/
sludge, respectively
Obtained experimental models are listed in Table 5 Based on R statistics of regression equa-
tions, the correlation of physical properties with the mixture compositions was better than that
of leached metal concentrations
Mtxture Design Opttmtzatlon

In order to compare the economic feasibility of three disposal options for the solidified
waste, the mixture design optimization for the solidification of 1 ton (1000 kg) of sludge was
performed by a hnear programming method. The objective function was constructed to min-
imize the total cost, which consisted of sohdIfiCatlon material cost and ultimate disposal cost
Operational costs for the solidification were not considered in this study As a basis of disposal
cost, the waste volume after solidification was calculated using the volume change factor Con-
TABLE 5--Results of multiple hnear regression analysis
Regression Coetfioent
Measured Propenxes Bo ~1 ~2 ~3 R
Volume Change Factor (VCF)
Cr 1 152 0 369 0 528 1 220 0 988
Cu 1 260 0 314 0 452 00 0 952
UC Strength, kg/cm2
Cr - 8 9 97 243 8 162 8 -459 6 0 959
Cu - 3 3 41 113 1 4425 -260 6 0 924
Leached Cr, mg/L
AW 5 204 --3 664 00 00 0 753
DW 0 213 --0 061 --0 142 00 0 882
SW 0 104 --0 047 --0 725 00 0615
Leached, Cu, mg/L
AW 610 8 -- 396 7 00 O0 0817
DW 0 576 --0 232 --0 270 00 0 745
SW 0 144 --0 064 --0 049 00 0 664
stralnts were formulated using regression equations and proper disposal guidelines for the
strength and leached metal concentrations. The experimental conditions of mixture design
were also considered as constraints. Linear programming for optimization [21] was as follows
Minimize total cost = material cost + disposal cost

= (MC~XI, + MC2X2, + MC~(s,) + (DCjVCF,)/(d,,)
Constraints
(Regression equation for UC strength), < UCSj
(Regression equation for heavy metal concentration), < HMC U
05<X~<15
00<X2<15
00<X3<005
where
MC = unit cost of each solidification agent, S/ton,
DC -- unit cost of each disposal option, $/m 3,
VCF = regression equation for volume change factor,
UCS = disposal guldehne of unconfined compressive strength, kg/cm 2,
HMC = disposal guideline of heavy metal concentration, mg/L,
dr = bulk density of the raw waste, t o n / m 3,
t = heavy metal types = 1, 2, and
j = ultimate disposal options = 1, 2, 3
Unit cost and disposal guidelines were set up as shown in Table 6 Unit costs were estimated
through the survey of sohdificatlon agent supplier and consignment company. In case of reuse
as construction materials, income from bnck marketing was considered In ocean disposal,
transport to a 100-km disposal site was assumed The strength guideline of sohdIfied waste for
three disposal options was based on the compressive strength standards for landfill and brick
of 1st and 3rd grade in Korea Concentration criteria of the leached heavy metals were referred
to the standards for drinking water and U S. EPA EP toxicity Note that results from these
optlmIzations are specific to the synthetic sludges used in this study and are presented for illus-
tration purposes only
Optzmal Solutzons
Optimal solutions for the solidification and ultimate disposal of 1 metric ton of sludge are
presented in Table 7 Due to high price and poor performance, sodium silicate solution was
not selected in any disposal option As the strength and leaching properties ofsohdlfied matrix
containing copper sludge was much poorer than those containing chromium sludge, the eco-
nomic comparison showed that the total costs for copper waste are more expensive than that
for chromium waste The decreasing order of total unit cost for three disposal options was
landfill, ocean disposal, and construction reuse Although sludge reuse as construction mate-
nals was estimated as the most cost-effective alternative, the unit cost including neglecting the
benefit from doubtful brick marketing resulted in the same level as that of the ocean disposal.
Accordingly, if the impact on public health and environment was assessed carefully before its
apphcatIon, ocean disposal as well as reuse in construction would be considered as the pre-
ferred disposal alternatives for the solidified sludges.
Critical constraints in optimal solution may gwe information useful to understand the three
different disposal options In copper waste, the critical constraint for landfill was copper con-
centration
Copyright by ASTM Int'lcriteria,
(all rights whereas
reserved); MonmFeb
i n24i m u m strength
23:07:03 EST 2020 requirement was the critical constraint for con-
struCtlve
Universidad reuse
Industrial and (Universidad
de Santander ocean disposal
Industrial de In the case
Santander) ofto chromium
pursuant waste,
License Agreement. however,
No further reproductionsstrength
authorized. was the
TABLE 6--Bastc data set for mixture design Optlmtzatton
Index, a k orj*
Cost and Guldehne 1 2 3
Material cost, MCk, S/ton 75 15 150

Disposal cost, DCj, S/ton 43 9 -30 0 17 1
Strength guldehne, UCSj, kg/cm 2 10 80 40
Heavy metal concentration guldehne, HMC U, mg/L
Cr(t = 1) 5 0 05 0 05
Cu (l = 2) 100 1 1
ak = sohdlficatlon agent (1 = cement, 2 = fly ash, and 3 = sodium silicate) j = ultimate d~sposal
option (1 = landfill, 2 = constructive reuse, and 3 = ocean disposal)
critical constraint for landfill, while c h r o m i u m concentration criteria was the requirement for
the other options Especially, reuse in construction had no feasible solution due to the strict
criteria o f c h r o m i u m concentration.
Conclusions
Based on the results obtained, the following conclusions were m a d e
1 D u e to c e m e n t hydration inhibition o f copper sludge, physical and leaching properties

o f solidified waste were observed to be inferior to those of solidified c h r o m i u m waste,
having a negligible effect on c e m e n t hydration
2 Fly ash addition to the solidification process detenorated the solidified waste matrices
and increased metal dissolution accordingly But, use of the sodium silicate solution had
no significant effect
3 Leaching charactenstlcs d e p e n d e d mainly on the type o f leachant, and leached heavy
metal concentrations were well m a t c h e d with their solubilities
4 A c o m p a r i s o n o f leaching characteristics indicated that acidic water is the most aggres-
sive toward leaching solidified matrices, and distilled water is slightly m o r e aggressive
than seawater
5 Leachate alkalinity in seawater decreased less than leachant alkalinity, which would be
caused by carbonate influx and m a g n e s i u m hydroxide formation onto solidified
materials.
TABLE 7--Optimum solutions for three disposal options
Binder, Ton/Ton of
Sludge Unit Cost, S/Ton of Sludge
Metal Disposal Ophons Cement Fly Ash Matenal Disposal Total
Cr Landfill 0 50 00 37 5 46 6 84 1
Constructive reuse No feamble solution
Ocean disposal 0 50 0 10 39 0 19 0 58 0
Cu Landfill 1 29 00 96 6 62 5 159 2
Constructive reuse 0 71 0 75 64 5 --47 0 17 5
Ocean disposal 0 50
0 38 43 2 20 3 66 5
NOTE
Universidad l tonde Santander
Industrial = 1000(Universidad
kg Industrial de Santander) pursuant to License Agreement. No further reproductions authorized.
6 Using staUstlcal analyses such as ANOVA and multiple linear regression, experimental
factors affecting the phystcal and leaching properttes of sohdified waste were able to be
compared quant~tatlvely
7. As a result of mixture design optimization, ocean disposal as well as construction reuse
were estimated as the preferred disposal alternatives for the solidified heavy metal sludge.
References
[1] Conner, J R, ChemtcalFlxattonandSohdlficattonofHazardous Wastes, VanNostrand Reinhold,
New York, 1990
[2] Envlronmental Aspects of Stabdlzatton and Sohdlficatton of Hazardous and Radtoactlve Wastes,
ASTMSTP 1033, P L C6t6 and T M Gdham, Eds, American Society for Testing and Materials,
Philadelphia, 1989
[3] Tlttlebaum, M W, Seals, P K, Cartedge, F K, and Engels, S, "State-of-the-Art on Stabfllzatton
of Hazardous Orgamc Liquid Wastes and Sludges," CRC Crmcat Revtews m Envtronmental Con-
trol, Vol 15, No 1,1985, pp 179-211
[4] "Stabdlzatlon/Sohdlficatlon of CERCLA and RCRA Wastes," EPA/625/6-89/022, U S Environ-
mental ProtecUon Agency, Office of Research and Development, Cincinnati,OH, May 1989
[5] Lechlch, A F and Roethel, F J, "Manne Disposal of Stabdlzed Metal Processing Waste," Journal
of the WaterPollutlon ControlFederatton, Vol 60, No 1, 1988, pp 93-99
[6] Benson, R E, Chandler, H W, and Chacey, K A, "Hazardous Waste Disposal as Concrete Admlx-
tuie," Journal of Environmental Engmeermg, ASCE, Vol 111, No 4, 1985, pp 441-447
[ 7] Shin, H S, Her, N R, and Koo, J K, "Design Optlm~zatlon for Solidification of Hazardous
Wastes," Hazardous Waste and Hazardous Materials, Vol 5, No 3, 1988, pp 239-250
[8] "Test Methods for Evaluating Sohd Waste, Vol IC Laboratory Manual Physical/Chemical Meth-
ods," SW-846, U S Environmental Protection Agency, Office of Sohd Waste and Emergency
Response, Washington, DC, 1986
[9] Malone, P G and Larson, R J, "Soentlfic Basisof Hazardous Waste Immoblhzatlon," Hazardous
and Industrial Sohd Waste Testmg Second Symposium, ASTM STP 805, American Sooety for
Testing and Materials, Phdadelphm, 1983, pp 168-177
[10] Shwely, W, Bishop, P, Gress, D, and Brown, T , "Leaching Tests of Heavy Metals Stablhzed with
Portland Cement," Journal of the Water Pollutton Control Federatton, Vol 58, No 3, 1986, pp
234-241
[11] Norusls, M J, "SPSS/PC+ for the IBM PC/XT/AT, "SPSS Inc, Chicago, IL, 1986
[12] Tashlro, C, Takahashl, H, Kanaya, M, H~rak~da, I, and Yosh~da, R, "Hardening Property of
Cement Mortar Adding Heavy Metal Compounds and Solublhty of Heavy Metal from Hardened
Mortar," Cement and Concrete Research, Vol 7, No 3, 1977, pp 283-290
[13] Thomas, N L, Jameson, D A, and Double, D D, "The Effect of Lead Nitrate on the Early Hydra-
tlon of Portland Cement," Cement and Concrete Research, Vol 11, No 1, 1981, pp 143-153
[14] Cocke, D, Ortego, J D, McWhmney, H, Lee, K, and Shukla, S, "A Model for Lead RetardaUon
of Cement Setting," Cement and Concrete Research, Vol 19, No 1, 1989, pp 156-159
[15] Cft~, P L and Hamilton, D P, "Leachabdtty Comparison of Four Hazardous Waste Sohd~ficatlon
Processes," Proceedings of the 38th Annual Purdue Industrial Waste Conference, Lafayette, IN,
1983, pp 221-231
[16] Zamoram, E, Sheikh, I A, and Serrlnl, G, "Physical Properties Measurement and Leaching
Behavior of Chromium Compounds Sohdlfied m a Cement Matrix," Nuclear and Chemzcal Waste
Management, Vol 8, 1988, pp 239-245
[17] Campbell, K M, El-Korchl, T , Gress, D, and Bishop, P, "Stabdlzatlon of Cadmmm and Lead m
Portland Cement Paste Using a Synthetic Seawater Leachant," Environmental Progress, Vol 6, No
2, 1987, pp 99-103
[18] C6t6, P L, "Contaminant Leaching from Cement-Based Waste Forms under Acidic Condxtlons,"
Ph D dissertation, McMaster Umvers~ty,Hamilton, Ont, 1986, pp 11-16
[ 19] Snoeymk, V L and Jenkins, D, Water Chemwtry, John Wiley and Sons, New York, 1980,pp 220-
222
[20] Montgomery, D C, Design andAnalysls of Expertments, John Wiley and Sons, New York, 1976,
pp 306-449
[21] Hiller, F S and Llberman, G J, Introductlon to Operations Research, Holden-Day Inc, Oakland,
CA, 1986, pp 102-167
Hwan- Young Kim, 1 Ki-Hong Kim, ~Hun-Hwee Park,
and In-Suk Suh 1
Solidification of Dried Borate Waste with

Polyester
REFERENCE: Kam, H -Y, Klm, K -H, Park, H -H, and Suh, I -S, "Solidification of Dried
Borate Waste with Polyester," Stablhzatlon and Sohdlficatton of Hazardous, Radloacttve, and
Mtxed Wastes, 2nd Volume, ASTMSTP 1123, T M Gllham and C C Wiles, Eds, Amencan
Sooety for Testing and Matenals, Philadelphia, 1992, pp 297-303
ABSTRACT: An effective method to dry the sludge produced in a reaction between the evapo-
rator concentrates of a pressurized water reactor (PWR) and calcmm oxide (CaO) has been
sought The optimum reaction condition in this study for avoiding the adhesion of dried prod-
ucts to the heating wall of a dryer and for preventing the agglomeration of dned products was
found to be about 0 75 mol ratio of CaO/bonc acid (H3BO3) In this condition the dried products
were porous spheres, and their true density was 2 04 )< 103 kg/m3
In the immersion test of the sohd~fied forms with unsaturated polyester in demmzed water for
520 days, the weight change of the sohdlfied forms was not more than 1%, and thor volume
change was negligible The specific gravity of the solidified form containing 40% by weight of
dried products was 1 45 and increased to 1 55 with an increase of the dned products to 50% by
weight The compressive strength of a solidified form m the above range was increased from
139 3 to 162 8 MPa The effective diffusion coefficients in the 18-month leaching test for Cs- 137
and Co-60 were 2 08 • 10 -15 m 2 / s and 3 30 • 10 -19 - - 2 26 • 10 18 m 2 / s , respectively
KEY WORDS: volume reduction, solldlficaUon, H3BO3, unsaturated polyester, CaO, mechan-
ical properties, leaching
The hmlt of evaporation of pressurized water reactor (PWR) hquld radioactive wastes con-
taming various radionuclldes and boric acid (H3BO3) depends greatly on the H3BOa concen-
tration rather than on the radlonucllde concentration Therefore, in consldenng the solubility
of H3BO3, these hqmd radioactive wastes have been concentrated to y]eld concentrates of up
to 12% by weight of H3BO3 The concentrated hquld was then sohdlfied in cement However,
this cement solidification process causes a 1.5 times increase of volume over that of the original
hquld wastes.
High-volume reduction processes are now under development One involves drying fol-
lowed by solidification [1-3] Another involves pretreatment of soluble salt to produce insol-
uble sohd through separation and sohdificatlon [4] For instance, the drying of the liquid
radioactive wastes of a boiling water reactor (BWR) is easily carried out using a vertical thin
film evaporator with swing blades [5] When this method is applied to the P W R llqmd wastes
contaimng H3BO3, some difficulties have arisen, such as the adhesion of dried products to the
heating wall of the drying apparatus Several pretreatment methods using additives such as
aluminum sulfate [A12(SO4)3], sodium hydroxide (NaOH), and slaked lime were tested follow-
ing the drying process using either a horizontal thin film evaporator or a special dryer
[1-3,6] These methods continue to have problems, including incomplete drying
Pnncipal researcher, senior researcher, director, and vice president, respectively, Korea Atomic
Energy Institute, P O Box 7, Daeduk-Danjl, Taejon, Korea
297
Copyright9 1992 by ASTM International www.aslm.org

298 HAZARDOUSWASTES
Experiments
Preparatton of Drwd Product
The reaction of H3BO3 solution (12% by weight) with calcium oxide (CaO) ( < 100 mesh)
was carried out at 80"C with 0 75 mol ratio (calcmm/boron) The density of the dned products
was measured using a multivolume pycnometer, and their surfaces were observed with a scan-
ning electron microscope (SEM)
Compressive Strength and Immerston Test

The dried products were added into unsaturated polyester (propylene glycol, dlethylene gly-
col, malelc anhydnde, phthahc anhydnde, and styrene) with 40, 42, 44, 46, 48, 50, and 52%
by weight, respectively, and they were mixed with benzoyl peroxide (0 25 g/100 g of polyester)
and dlmethyl aniline (0 13 mL/100 g of polyester) in cyhndncal aluminum molds 50 m m in
dmmeter The mixing blade was moved up and down The mixtures were set at room tem-
perature for a week, and then solidified forms were taken out from the mold The heights of
the specimens were 50 mm, and their diameters were 49 3 to 49 9 m m
The compressive strength of the specimens was tested using a universal tester (Sh~madzu
UHE-200A) The height and volume changes of the specimens were observed after immersion
in delomzed water for 30, 200, and 520 days, respectively
Leachmg Test
The specimen, composed of 80 g of unsaturated polyester and 80 g of dned products after
the addition of 0 1261 mC1 of Cs-137, showed a diameter of 49.5 m m and a height of 56 1
m m For the Co-60 specimen, the activity was 0.1630 mC1, and the diameter and height were
49 4 and 55 5 ram, respectively The specimen was immersed in 1 L of de~omzed water at
25~ and the procedure for the replacement of the leachant followed the International Atomic
Energy Assn (IAEA) recommended method [ 7] The activity of the leachant was measured
using a low-level alpha/beta counter

Dried Product
Figures 1 and 2 are microscopic pictures of the dned products using SEM The morphology
of the major porhon of dned product surface is shown in Fig 1, and the morphology of an
extremely small remaining portion is shown m Fig 2 When the morphology of Fig 2
appeared, the dned products could be seen to be more adheswe on the heating wall and
agglomeratwe together As shown In Fig 1, these products seemed to have grown m sphencal
and porous forms The density o f d n e d products was 2 04 )< 103 kg/m 3
Water Resistance
The volume changes of the specimens solidified with unsaturated polyester were negligible
d u n n g the immersion test, and any blister and crack of their surface was not observed for 520
days The results of weight change are shown m Fig 3 The weight changes that were measured
after wiping water from the surface were expressed by open symbols After 30 days, specimen
weights decreased by 0.5 wt% This minor weight loss occurred because of the low solubility
KIM ET AL. ON BORATE WASTE SOLIDIFICATION 299
FIG. 1--The morphology of the major portion of dried product surface.
of the dried product through fine pores in the protective polyester layer. After 200 days, the
weight loss was 0.2% by weight, which was less than that after 30 days. One explanation for
this is that the weight of water absorbed into the dried products was greater than the weight of
salts dissolved from the solidified form. This effect becomes more significant in the case of the
520-day immersion test. The closed symbols indicated weight loss after drying the specimens
at 100~ for 6 h. The weight losses were little affected by immersion time. These results might
arise from the fact that either the dissolution of salt occurs only at a thin layer of surface or the
water at inner layers, which remains unevaporated, exceeds the increasing outflow of soluble
salt.
FIG. 2--The morphology of an extremely small remaining portion of dried product surface.
300 HAZARDOUSWASTES
+1.0
0:30. h:Z00. 0:520 DAYS IMMERSED
(AFTER WIPING WATER)
D:30. A:200. 1:520 DAYS IMMERSED

* 0.5 (AFTER DRYING)
,N
)
Z 0
o
~ - - , o
-0.5
-I.0.
I I I I I I I
40 42 44 46 48 50 52
DRIED POWDER IN THE WASTE FORM. wiT.

FIG 3--The wetght change after immersion m demtwater
Denstty
The densities of solidified forms with their varying dried powder compositions are shown m
Fig 4 The density increased proportionally with the composition of dried powder However,
in the case of the composmon of 55% by weight dried powder, the density decreased to 1.52
X 10 3 kg/m 3 This could be caused by the fact that the air introduced for mixing cannot be
removed from high vise6us mixtures during setting and curing
Compressive Strength
Figure 5 shows that the compressive strength of solidified forms vaned with dried powder
content. In Fig 5, the compressve strength increased linearly with the increase m weight com-
position of dried powder m sohdlfied form from 40 to 50% by weight At over 50% by weight
of the dried powder composition, the compressive strength tended to decrease sharply (139 3
MPa at 55% by weight). This was because the homogeneity of sohdlfied forms started to fad
after this mixing ratio.
Consldenng mechanical properties, volume reduction, and ease of mixing, these results
indicate that the proper conditions for the composmon of dried powder are 50 to 52% by
weight
The obtained compressive strength of the form is higher than that of unsaturated polyester
itself (136 3 MPa) since the dried powder acted as a filler, reducing fine thermal cracks and
enhancing the heat release dunng curing
KIM ET AL ON BORATE WASTE SOLIDIFICATION 301
1.60
~0 1.55
\
C0
(3 I. 50
F1
M 1.45
V)
N
N
e~
1.40
I I I I I ; !
40 42 44 46 48 50 52
DRIED POWDER IN THE WASTE ]FORM. wt~

FIG 4--The denszty of the sohdzfiedforms
170
=. 160
150
>
~3
140
m
m
(3
1750
i I I I I I I
40 42 44 46 48 50 52
DRIED POWDER IN THE WASTE FORM. wt%

FIG 5--The compresslvestrength of the sohdtfiedforms
n
0
40
i
0
30.
t5
r<
m
CD
Z
z 20
>
I01
D
D
9e , , i o i
5 10 15 20 25
.,/TIME ( D A Y S )
FIG 6--Leaching curvefor Cs- 137
6.
z"
0
M
N
4
M
N
M
Y i i i i
I0 15 20 25
,~TIME ( DAYS )
F I G 7--Leaching curvefor Co-60
KIM ET AL ON BORATE WASTE SOLIDIFICATION 303
Leachabthty
The cumulative fractions ofCs- 137 and Co-60 leached per day from the sohdlfied form are
given m Figs 6 and 7, respectively In Fig 6, the effective diffusion coefficients ofCs- 137 were
relatively lower through the first three weeks than after those three weeks This phenomena
could be showing that cesmm ion and water were diffused along the pores generated by dis-
solution of calcium borate [Ca(BO2)2 2H20] and calcium hydroxide [Ca(OH)2]. The sum-
mary diffusion coefficient of Cs-137 was 2 08 • 10 15 m2/s For Co-60, because the leaching
hardly appeared for a month, the activities of the leachant could not be measured, being below
the low hmlt of detection This occurrence could be explained in that the retarding effect of
leaching existed because the solubility of cobaltous hydroxide [Co(OH)2] is much lower than
that ofcesmm hydroxide (CsOH) For this reason, the dlffusmn coefficient of Co-60 was much
lower than that of Cs-137 and had two values, that is, 3 30 • 10 19and 2 26 X 10 -~8 m2/s,
hawng an reflection point at about 300 days
Conclusions
The mole ratio of CaO/H3BO3 for avoiding the adhesion of dried products on the heating
wall of a dryer and for preventing the agglomeraUon between dned products was determined
The optimum reaction raUo (calcium/boron) was 0 75 In this condition, the dried products
had a morphology of porous spheres and their true density was 2 04 • 10 3 kg/m 3
From the immersion test results m deionized water for 520 days, it was found that the solid-
ified forms have a high integrity. The density of the sohdlfied form containing from 40 to 50%
by weight of dried powder increased linearly. The compressive strength of a solidified form m
the above range increased from 139 3 to 162 8 MPa The diffusion coefficients for Cs- 137 and
C o - 6 0 m t h e 18-monthleachlngtestwere 2 08 • 10-15 m2/s and 3 30 • 10 19to2.26 )< 10 -18
m2/s, respectively
References
[1] Cordxer, J J etal, U S Patent No 4,086,325, 15 April 1978
[2] Mon, K et al, U S Patent No 4,671,897, 9 June 1987
[3] Kurumata, N, U S Patent 4,800,042, 24 January 1989
[4] Kunbayashl, H et al, JGCNuclearRevtew, No 3, February 1987, p 8
[5] Chino, K et al, Journal of the Japan NuclearSoctety, Vol 25, No 1, 1983, p 69
[6] Mergan, L M etal, U S Patent No 4,409,137, 11 October 1983
[7] Technical Report Senes No 82, P101-105, International Atomic Energy Assn, Vienna, Austria,
1968
Tommy E. Myers IandMark E. Zappi I
Laboratory Evaluation of Stabilization/

Solidification Technology for Reducing the
Mobility of Heavy Metals in New Bedford
Harbor Superfund Site Sediment
REFERENCE: Myers, T E and Zappl, M E, "Laboratory Evaluation of Stabilization/Solid-
ification Technology for Reducing the Mobility of Heavy Metals in New Bedford Harbor Super-
fund Site Sediment," Stabdlzatton and Sohdlficatton of Hazardous, Radioactive, and Mixed
Wastes, 2nd Volume, ASTMSTP 1123, T M Gllham and C C Wiles, Eds, Amencan Society
ABSTRACT. Sediment samples from the New Bedford Harbor Superfund Site, New Bedford,
MA, were stablhzed/sohdified using the following binders portland cement, portland cement
with a proprietary additive, and a proprietary binder system Comparison ofcontammant masses
released during sequential batch leaching of stablhzed/sohdlfied sediment and untreated sedi-
ment showed that stabilization/solidification reduced the leachablhty of some but not all metals
investigated The amounts of metal leached from stablhzed/sohdified sediment relative to the
mass leached from untreated sediment were reduced by factors of 10 to 37, 1 2 to 2 6, 4 3 to 28,
and 9 0 to 14 for cadmium, chromium, lead, and zinc, respectively Copper and nickel, however,
were mobilized by stabilization/solidification The amounts of copper and nickel leached from
stabihzed/sohdlfied sediment relative to the mass leached from untreated sediment were
increased by factors of 20 to 23 and 9 6 to 17 for copper and nickel, respectively Desorptlon
isotherm analysis of sequential batch leach data showed that sohdlfiCatlon/stablhzatIon process-
lng of this sediment altered the mechanisms controlling metal leaching
KEY WORDS: stablhzatlon/sohdlfiCatlon, leaching, sediment, metals, lmmoblhzation, dredged

material
Stabilization/solidification (S/S) is a developing technology for treatment and disposal o f

c o n t a m i n a t e d materials T h e technology has been applied in Japan to b o t t o m sediments con-
talnlng toxic substances [1-3] and in the U n i t e d States to industnal wastes [4-6] Tlttlebaum
et al [ 7] reviewed the current technology and its application
Solidification is the process o f e h m l n a t l n g free water by hydration of a setting agent(s) Typ-
ical setting agents include portland cement, lime, fly ash, kiln dust, slag, and combinations o f
these materials CoaddltlVeS such as bentonite, soluble silicates, and sorbents are sometimes
used with setting agents to give special properties to the sohdlfied product
Stablhzatlon can be both physical and chemical Physical stabthzatlon is the process of sohd-
ificatlon and refers to i m p r o v e d e n g l n e e n n g properties such as b e a n n g capacity, trafflcablhty,
and permeability Solidification usually provides physical stablhzation but not necessarily
chemical stablhzatlon. Physical stabilization (solidification) l m m o b l h z e s contaminants
Environmental engineer and civil engineer, respectwely, Water Supply and Waste Treatment Group,
Environmental Engineering Division, Environmental Laboratory, U S Army Engineer Waterways
Experiment Station, Vicksburg, MS 39180
304

MYERS AND ZAPPI ON LABORATORY EVALUATION 305
through alteratmn of the physical properties of the material Materials such as sediments and
other semlsohds when converted from a plastic to a solid state are less susceptible to leaching
due to reduced accessibility of water to the contaminated solids After solld~ficahon processing,
the matenal may be less permeable, and/or contaminated sohds may be mlcroencapsulated
by hydration products [8]
Chemical stablhzatlon is the alteration of the contaminants' chemical form such that leach-
ability ~s ehmmated or substantially reduced Stablhzation/sohdlfication processes are usually
formulated to minimize the solublhty of metals by controlling pH and alkalinity and/or by
sorptlon to specml process additives [8-14] Anions are typically more difficult to bind m
insoluble compounds [15] Thus, most S/S processes rely on entrapment or mlcroencapsu-
latlon to lmmoblhze anions. Information on immobilization of organic chemicals by S/S is
hmlted, but some work has been done [16-22] and research is continuing
In 1984, the U S Environmental Protection Agency (EPA) reported on the feaslblhty of
remedial action alternatives for the Upper Acushnet River Estuary, New Bedford, MA [23]
The EPA received extensive comments on the proposed remedial action alternatives from
other federal, state, and local officials, potentially responsible parties, and other concerned
individuals Responding to these comments, the EPA chose to conduct additional studies to
better define available cleanup methods Because dredging was assocmted w~th all of the
removal alternatives, the EPA requested the U S Army Corps of Engineers (ACE) to conduct
an engineering feasibility study of dredging and disposal alternatives for the Upper Acushnet
River Estuary
Sohdlficat~on/stablllzatlon was identified by ACE as a technology with potential appllcanon
to treatment of contaminated sediments and dredged matenal from the Upper Acushnet River
Estuary, New Bedford, M A This paper descnbes laboratory-scale applications of selected S/
S processes to Upper Acushnet River Estuary sedxment and an evaluation of the solidified/
stablhzed products on the basis of heavy metal leaching in sequential batch leach tests
Procedures
Materials
The composite sediment sample from the Upper Acushnet River Estuary, New Bedford,
MA, as descnbed by Averett [24] was used m this study The collection procedures are only
briefly described here Sediment samples were obtained with a modified box corer that col-
lected a 0 093-m 2 ( l-ft 2) sample to a depth of 0 6 m (2 It) according to a samphng scheme
developed from reconnaissance sampling Samples were composlted, mixed on site, and
shipped to the U S Army Engineer Waterways Expenment Station (WES), Vicksburg, MS,
by refrigerated truck At the WES, the composite sediment sample was rem~xed and stored at
4~ No other processing (for example, dewatenng) was applied before applying vanous S/S
process additives
Three setting agents (or binders) were used to sohdlfy/stabfllze sediment as follows portland
cement, portland cement with Flrm~x, and a proprietary silicate additive Type I portland
cement was used in the portland cement and portland cement with Firmlx processes Firmlx,
a proprietary additive, was obtained from Trident Engineenng, Baltimore, M D The S/S
reagent for the propnetary slhcate process was provided by Silicate Technology Corporation
(STC), Scottsdale, AZ Interstitial water in the sediment was used to hydrate the setting agents
Stabdtzatzon/Sohdlficauon Processing
Sediment was solidified using the following process addmve-to-sed~ment ratios (by weight)
Copyright by ASTM Int'l cement
0 3 portland (all rights reserved); Mon Feb 24 23:07:03
1 0 sediment, 0 15ESTportland
2020 cement 0 15 Firmlx 1 0 sediment,
and 0 3
306 HAZARDOUSWASTES
STC addltlVe'l 0 sediment Other binder-to-sediment formulations were tested [17] How-
ever, the emphasis in this paper is on formulations that provided the best overall metal immo-
bilization Process additives were mixed with sediment in a Hobart C-100 mixer for 5 mln per
additive. After mixing, the freshly prepared stabilized/solidified sediment was cured at 23"C
and 98% relative humidity in a curing chamber for 28 days After 28 days, stabilized/solidified
sediment was crushed before leach testing to pass a 2 0-mm sieve and retain on a 0 3-mm sieve.
Twenty-eight-day unconfined compressive strengths for the stabilized/solidified sediments as
reported elsewhere [17] were 1 9, 2 6, and 3 3 MPa (277, 380, and 481 psi) for the portland
cement, portland cement with Flrrnlx, and STC products, respectively
Sequentml Batch Leach Tests

Stabilized/solidified sediments were leached for metals using a sequential batch leach test
developed in this laboratory for evaluating the leaching potential of untreated sediments and
dredged matenal [25] The procedure uses a four-to-one ratio of water-to-dry solids by weight
and 24 h o f shaking time Stabilized/solidifiedsediments were leached for metals by contacting
crushed stabilized/solidified sediment with dlstilled-delonlzed water on a mechanical shaker
for 24 h and separating leachate from solids by centrlfugation and 0 45-urn membrane filtra-
tion The solids were then resuspended in fresh distllled-deionlzed water, and the shaking-sep-
aration steps were repeated for a total of four leaching cycles The general procedure is shown
in Fig 1 Stabilized/solidified sediments were leached in tnplicate Sequential batch leach
tests on untreated sediment, reported elsewhere [25], involved four replicates and seven
cycles
The mass leached data were normalized with respect to the amount of sediment that was
stabihzed/sohdlfied in order to account for dilution by S/S additives The normalized mass of
metal leached in the sequential batch leach test is given by
Z = ~-~c, V
,., Ms (1)
where
= normalized mass leached at end ofnth cycle, mg metal/kg dry weight sediment solids,
(7,
= concentration in leachate from tth cycle, mg/L,
l
= cycle index,
V
--- volume of leachate, L,
n
--- number of cycles, and
Ms = mass of dry sediment solids, kg
For untreated sediment, V/M, the liquid sohds ratio is 4 so that Eq 1 can be written as
Lu = 4 ~ C, (2)
t=l
where
Lv = L for untreated sediment, mg/kg

For stabilized/solidified sediment, the amount of dry sediment solids that was solidified/sta-
bilized is obtained from the following relationships
I
CRUSH& SIEVE ]
SAMPLE [
45-
SOLIDS I
,~ ADD
<i> WATER
SHAKE
LEACHATE ~ SEPARATE ] 3
PHASES
<2-
SOLIDS
42-
REPEATSTEPS
1,2, & 3
ASNEEDED
FIG 1--Schemattc of sequenttal batch leach test
Ms/s = MwetR + Ms (3)
where
Msls = total mass of sohdlfied/stablhzed sohds, kg, and
R = dosage of S/S reagents, kg reagent/kg wet sediment sohds,
and
Mwet --- Ms(1 + w) (4)
where
Mwet --- mass of wet sediment solids, kg, and
w = water content of wet sediment, kg water/kg dry sediment sohds
SubstRutlon from Eq 4 for Mwet m Eq 3 yields
Ms/s = RMs(1 + w) + Ms (5)

Rearranging Eq 5 and substituting for Mwet m Eq 1 yields
C,V(1 + R + R,~)
-s z_.., (6)
,=l Ms/s
where
Ls/s = L for stabfllzed/sohdlfied sediment, mg/kg

Since V/Ms/s was also equal to four for leaching stablhzed/sohdlfied sediment, Eq 6 can be
written as follows
-Ls/s = 4(1 + R + Rw) ~ C, (7)

/ffil
Equation 7 was used to calculate normalized mass leached to account for dilution effects due
to additive addition The dilution factor developed in Eq 7 (1 + R + Rw), however, does not
account for solids produced by hydration reactions or bound water in the stablhzed/sohdified
sediments. Problems associated with determining the actual amount of ongmal sample in the
stabilized/solidified material used in a batch leach test have been noted by other workers
[101
Percent immobilization was calculated as follows
Immobilization (%) = 1 - \ Lu J lOO (8)
Data from the sequential batch leach tests were also used to develop desorptlon isotherms:
plots of solid phase contaminant concentration, Q, versus leachate contaminant concentra-
tion, C The term desorption as used in this paper refers to a combination of elementary pro-
cesses, such as sorption, ion-exchange, dissolution, and so on, involved in the transfer of con-
taminant from the sohd phase to the aqueous phase Calculation of the solid phase
concentration at the end of each cycle in sequential leach procedure is gaven by
_ (c,v
e,-- e,-, \-M-~s/s/ (9)
where
Q, = solid phase contaminant concentration at end of the tth cycle, mg/kg
Since V/Ms/s was equal to four for leaching stabllized/sohdlfied se&ment, Eq 9 can be wmten
as follows
Q, = Q,_~ -- 4C, (10)
Equation 10 was used to calculate the sohd phase metal concentration at the end of each leach
cycle The lnmal metal concentration in the sohd phase Q0 needed to calculate the sohd phase
metal concentration after the first step, QI, was obtained by bulk analysm of stabilized/solidi-
fled sediment and untreated sediment
Chemical Analysis of Leachate Samples

Leachates, stablhzed/sohdlfied sediment, and untreated sediment samples were analyzed
for metals using directly coupled plasma emission spectroscopy on a Beckman Spectraspan
IIIB plasma ernlsslon spectrometer or by atomic absorption spectroscopy using a Perkin-
Elmer Model 5000 atomic absorption spectrometer coupled with a Perkin-Elmer Model 500
hot graphite atomizer following appropriate sample digestion procedures [26]
TABLE 1--Physwal properties of estuary composite sediment a
Analysis Result
Percent sand, >50 um 39

Percent silt, 2 to 50 um 61
Percent clay, <2 um 0
Liquid limit 129
Plastic limit 61
Plasticity index 68
Specific gravity 2 35
Water content, kg water/kg dry solids 1 78
USCS classification OH
Soil group Sandy organic silt
a From Ref 24
Results
General Charactertzatton of S / S and Untreated Sedtments
Physical charactenzatmn of the Upper Acushnet River Estuary composite sediment sample
is provided in Table 1, and chemical charactenzatlon is provided in Table 2 The Unified Soil
Classification System (USCS) descnptlon is a dark gray, sandy orgamc silt (OH) Copper, lead,
and zinc, the most abundant metals m the estuary composite sediment, were present m con-
centrations of 1730, 2013, and 3017 mg/kg, respectively These concentraUons exceed those
normally encountered m sediments where dredging is proposed or implemented [27] Because
of the comparatwely low levels of arsemc, mercury, and selenmm, analysis for these metals
was not included m the sequential batch leach tests.
Sequenttal Batch Leach Data

Leachate pH values d u n n g sequential leaching of untreated and stablhzed/solldlfied sedi-
ments are presented m Table 3 Leachate pH for stablhzed/solld~fied sediments ranged from
9 2 to 12 5, and leachate pH for untreated sediment ranged from 7 1 to 7 6
Metal leachate concentrations d u n n g sequential batch leaching of untreated and stablhzed/
TABLE 2--Chemtcal analysts of estuary composite sedtmenf
Parameter Concentration, mg/kgb
Arsenic 8 66 (0 24)
Cadmium 35 4 (0 25)
Chromium 754 (9)
Copper 1730 (21 )
Lead 2013 (239)
Mercury 2 59 (0 03)
Nickel 122 (1 8)
Selenium <0 49
Zinc 3017 (22)
Total organic carbon 26 000
From Ref 25
b Standard error in parentheses
0.020 1 1 1 1 1 1 1 1 I I I I I
LEGEND LEGEND
0.016 __9 UNTREATED T ~\ 9 UNTREATED

VPC 5 \\ V PC
.J
9 Pc/F A\ T \\ 9
~ 0.012 E
4
0.008 E
r o 2
0.004
0.000 m 1 1 1
1 2 3 4 5 6 7 8 0 I 2 3 4 5 G 7
CYCLE NUMBER CYCLE NUMBER
0.4 I I I I I I I 0.5 I I I I I I I
LEGEND _ LEGEND
9 UNTREATED
9 UNTREATED A
0.4 PC
9 PC/F I/\ T -; PD,,
E c~ 0.3
E
~ 0.2
oa:: 0.2
..J
2iz
u 0.1
0.1
0.0 I I I I I 0.0
2 3 4 5 G 8 0 2 3 4 5 G 7 8
1.0 I I I I I I I I I I I I i
LEGEND
0.8 _ LEOEND T I
9 UNTREATED &
0.8
t 9 UNTREATED
~7 PC
v Pc /1\ /I
0.6 -- ~ PC/F //I\ T/~
9 PC/F
0.6 [] STC c~
E E
d~ 0.4
0.4
u N
7
0.2 0.2
0.0 0.0
I 2 3 4 5 G 7 8 0 I 2 3 4 5 6 7 8
FIG 2--Metal concentrations in leachate during sequential batch leaching (PC = portland cement; PC/
F = portland cement with Firmix; S T C = proprietary silicate process; error bars indicate standard error,
missing error bars are embedded in the symbols).
solidified sediment are shown in Fig. 2. The sequential batch leach data presented for
untreated sediment are from another study [25] that as previously discussed involved seven
instead of four leach cycles. As indicated in Fig. 2, the concentrations of cadmium, chromium,
lead, and zinc in leachates from stabilized/solidifiedsediment were reduced relative to the con-
centrations
Copyright in(allleachate
by ASTM Int'l from
rights reserved); Mon untreated
Feb 24 23:07:03 sediment.
EST 2020 Copper and nickel concentrations in leach-
TABLE 3--Composite pH during sequential batch leaching ~
Cycle Number
Matenal 1 2 3 4
Untreated sediment 72 73 7 1 76
pC b 12 5 12 4 12 5 12 5
PC/F c 12 1 120 120 11 9
STC e 10 6 10 5 96 92
a Composite subsamples of rephcates

b 0 3 portland cement 1 0 wet sediment
c 0 15 portland cement 0 15 Flrm~x 1 0 wet sediment
d 0 3 STC ad&twe 1 0 wet se&ment
ates from stabdlzed/sohdtfied se&ments were significantly higher than copper and nickel con-
centratlons m leachates from untreated sediment
Blanks c a m e d through the leaching procedure showed no ewdence o f c o n t a m i n a t i o n except
for the lead data for the portland c e m e n t process and the zinc data for the STC process [17]
T h e high lead concentrations for the portland c e m e n t process relative to the portland c e m e n t
with F l r m l x and STC processes are, therefore, unreliable The STC zinc data are comparable
to the zinc data for portland c e m e n t and portland c e m e n t w~th Flrmlx processes, although the
blanks showed potential c o n t a m i n a t i o n problems with the zinc data for the STC process
Leachate concentrations d u n n g sequential batch leaching suggest that c a d m i u m , chro-
m m m , lead, and zinc were l m m o b l h z e d to s o m e degree by S/S processing Simple compari-
sons o f l e a c h a t e concentrations, however, can be misleading because the a m o u n t of sediment
in the stabilized/solidified sediment was diluted by process addmves T o properly account for
additive dilution, cumulative metal masses leached over four cycles from untreated and sta-
blhzed/sohdlfied sediment were normalized w~th respect to the a m o u n t of dry s e & m e n t sohds
leached using Eqs 2 and 7, respectively N o r m a h z e d masses leached are presented m Table 4.
The a m o u n t s of metal leached from stablllzed/sohdlfied sediment relative to the mass leached
from untreated s e & m e n t were reduced by factors of 10 to 37, 1 2 to 2 6, 4 3 to 28, and 9 0 to
14 for c a d m i u m , c h r o m i u m , lead, and zinc, respectively Thus, the reduced c a d m i u m , chro-
m m m , lead, and zinc concentrations in batch leachates from stablllzed/sohdlfied sediments
relative to the concentrations of these metals in batch leachates from untreated sediment (F~g
2) were not due to simple dilution o f the sediment by S/S additives These data suggest that
some form of l m m o b l h z a t l o n such as mlcroencapsulatlon or conversion to less soluble forms
has occurred
TABLE 4--Metals leached m sequential batch leach test normahzed with respect to the mass of
sediment leached, mg/kg ~
Material Cadmmm Chrommm Copper Lead N~ckel Zinc
Untreated se&ment 0 113 2 99 3 72 3 14 0 57 13 1

PC b <0003 1 13 77 6 8 96 1 44
PC/F c < 0 003 2 47 87 8 0 72 5 50 0 88
STC a 0 011 1 66 79 7 0 11 10 2
a Four-cycle sequential batch leach test

b 0 3 portland cement 1 0 wet sediment
c 0 15 portland cement 0 15 Flrmlx 1 0 wet sediment
e 0 3 STC addmve t 0 wet sediment
312 HAZARDOUSWASTES
The normalized metal masses leached from stabilized/solidified sediment relative to the
mass leached from untreated sediment are presented in Fig. 3 as percent metal immobilization
(Eq 8). Percents cadmium immobilization relative to the mass of cadmium leached from
untreated sediment were 100, 100, and 90 for portland cement, portland cement with Firmix,
and STC processes, respectively. Percents chromium immobilization relative to the mass of
chromium leached from untreated sediment were 62, 17, and 44 for portland cement, port-
land cement with Firmix, and STC processes, respectively. Percents lead immobilization rel-
ative to the mass of lead leached from untreated sediment were 77 and 96 for portland cement
with Firmix and STC processes, respectively. (Lead immobilization for the portland cement
process was not calculated due to contamination in blanks, Ref 17.) Percents zinc immobili-
zation relative to the mass of zinc leached from untreated sediment were 89 and 93 for port-
land cement and portland cement with Firmix processes, respectively. (Zinc immobilization
for the STC process was not calculated due to contamination in blanks, Ref 17.)
The data in Table 4 also show significant mobilization of copper and nickel over a four-cycle
sequential batch leach test. The amounts of copper and nickel leached from stabilized/solidi-
fied sediment relative to the mass leached from untreated sediment were increased by factors
of 20 to 23 and 9.6 to 17 for copper and nickel, respectively. Mobilization is indicated in Fig.
3 as values less than zero. Percents copper mobilization relative to the mass of copper leached
from untreated sediment were 1986, 2260, and 2042 for portland cement, portland cement
with Firmix, and STC processes, respectively. Percents nickel mobilization relative to the mass
of nickel leached from untreated sediment were 1500, 865, and 1690 for portland cement,
portland cement with Firmix, and STC processes, respectively.
Several explanations for mobilization of copper and nickel by S/S processing are possible.
The first to consider is the normalization equation for solidified/stabilized sediment (Eq 7).
loo - I ~ , ~
Cd Cr Zn CdCr IPIZn Cd Cr "//'~,P b ~
N
.5 -1,000
'!l
Y=
13.
-2,000
Cu
-
Cu
i i i Cu i i
piC P C /,F ~ T IC
-2,600 -
PC: PORTLAND CEMENT

PC/F: PORTLAND CEMENT WITH FIRMIX
STC: STC PROPRIETARY ADDITIVE
Copyright by ASTM Int'l (all rights reserved);FIG.
Mon Feb 24 23:07:03 EST
3--Percent 2020 immobilization (Eq 8).
metal
TABLE 5--Bulk metal concentrattons tn sohdtficatton additives mg/kg a
Additive Cadmium Chromium Copper Lead Nickel Zinc
Portland cement 0 11 (0 01) 48 0 (0 9) 20 6 (0 4) 41 6 (0 6) 26 2 (0 3) 426 (3)

Flrmlx 006(001) 20 1 (03) 1 03(006) 0 12(004) 1 33(009) <30
STC 3 24 112 172 132 24 0 556

a Tnphcate analysis for portland cement and Flrmlx, single analysas for the STC additive Standard
error m parentheses
The normahzation factor in Eq 7 (1 + R + Rw), if Incorrect, could lead to errors This source
of error would be particularly important if the copper and nickel concentrations in leachate
from stabflized/sohdified sediment were approximately the same or lower than the copper and
nickel concentrations in leachate from untreated sediment However, the copper and nickel
concentrations an leachates from stabllazed/sohdlfied sediment were significantly higher than
those m the leachates from untreated sediments Hence, mobdizanon rather than ammoblh-
zanon results even if Eq 2 is used instead of Eq 7 to calculate normalized mass leached for
stablhzed/sohdlfied sediments Using Eq 2 for both untreated and stabilized/solidified sedi-
ments is basically the same as assuming no dilunon by S/S addmves, so that leachate concen-
trations can be compared directly Copper and nickel mobilization is clearly established by
direct comparison of metal concentrations an leachates, only the magnitude of the mobiliza-
tion IS affected by accounting for dilution by S/S additives
The S/S additives possibly could have been a source of copper and nickel during leaching
Bulk analysis of the solidification reagents (Table 5) showed the order of metal contamination
in the S/S additives to be as follows STC > portland cement > Firmlx If bulk metal con-
centratlons in the S/S addanves were controlling metal concentrations an leachates from sta-
bflazed/sohdified sediments, then the order of copper and nickel mobilization would be the
same order of copper and nickel concentrations m the S/S additives, and the order of cad-
mlum, lead, and zinc lmmobIhzatIon would be opposite the order of these metal concentra-
tions in S/S additives In general, there as no correlation between the relative order of metal
contamination in Table 5 and the amount of metal leached in Table 4 For example, the STC
addlnve contained 170 times more copper than portland cement but leached about the same
a m o u n t of copper as the portland cement product Accepting an explanation based on con-
tamanated S/S reagents is difficult, therefore, because this explanation is not consistent with
the response an leachable metal content of the stabihzed/sohdafied sediments
Copper and nickel moblhzatlon may also be explained as differences in leaching character-
lSnCS In the untreated sediment leach test, copper and nickel concentrations increased with
each cycle in the sequential leach procedure up to about Cycle 5 and by Cycle 7 had begun to
decrease In the sohdified/stabflazed sediment leach tests, copper and nickel concentrations
decreased with each step in the sequential leach procedure Addmonal discussion of the dif-
ferences m leaching characteristics is provided in the section on desorption isotherm analysis
Desorptzon Isotherm Analysts

Desorptlon isotherms are plots of solid phase contaminant concentration, Q, versus leach-
ate phase concentration, C Desorptlon isotherm analysis is a data reduction technique that
can yield information on the processes governing anterphase transfer of contaminant from the
solid phase to the aqueous phase
Copper
Copyright and(allnickel
by ASTM Int'l desorption
rights reserved); Mon Feb 24isotherms for untreated and stabilized/solidified sediment are
23:07:03 EST 2020
314 HAZARDOUS
WASTES
_o~ COPPER -I
&
0 98
/ I
o 97 - I
0 96
0 95
1 2 3 4 5 6
C, mg/~
lOOq~ [ I I I
0 99 ~ ) ) ~ N I f ~ k ' F I ~"
097
END
0 96 TED
0 95 --
0 02 04- 06 08 10
C, rng/,,~
FIG 4--Copper and mckel desorptlon motherms for untreated and stabthzed/sohdlfied sedzments (PC
= portland cement, P C / F = portland cement wtth Fmmlx, S T C = proprtelary sdtcateprocess)
shown in Fig 4 Copper and nickel desorptlon isotherms for stabihzed/sohdlfied sediments
had classical isotherm shapes that could be modeled with linear, Langmulr, and Freundhch
isotherm equations [17] In a classical desorptlon process, aqueous phase contaminant con-
centrations decrease as the solid phase contaminant concentrations are reduced by sequential
leaching
However, d u n n g sequential batch leaching of untreated sediment, copper and nickel con-
centrations in the leachate tended to increase with each cycle before beginning to decrease
This type of leaching produces a complicated, nonclassical desorptlon isotherm To satisfy
conservation of mass, leachate contaminant concentrations would not be expected to con-
tinue to Increase indefinitely as the solid phase contaminant concentration is decreased, but
would turn toward a classical isotherm shape The expected turn is shown in Fig 5
As shown in Fig 5, the type of metal desorptlon isotherms observed in this study for
untreated sediment imply nonconstant slope d u n n g sequential batch leaching The noncon-
stant portion of the desorption isotherm is related to elutlon of salt [25] As salt is removed
from
Copyright this estuarlne
by ASTM sediment,
Int'l (all rights reserved); the
Mon Feb ionicEST
24 23:07:03 strength
2020 of the aqueous phase is reduced According to
> NONCONSTANT REGION

I:T
/ /K2
O
I /
rr"
p.-
I / / I
Z
LJJ
I / /
Z
I / /K3
O
O I // // / /
I"--
Z I / / /K4
E21
/ / > CONSTANT REGION
LJ / / /
U3
/ / /
2~
f / I /
or"
/ / /
/ / /
O / / /
UJ
/ / /
/ /
EOUILIBRIUM LEACHATE CONCENTRATION,C

FIG 5--Conceptual desorptzon tsotherm showmg nonconstant and constant dtstnbutlon betweensohd
and aqueousphases during sequential batch leachmg (from Ref 25)
the Gouy-Chapman model of charge distribution in double layers, decreasing ionic strength
increases repulsive forces [28] so that flocculated colloidal matter becomes dispersed. Defloc-
culatlon of sediment organic carbon increases the concentration of dissolved organic carbon,
thereby mobilizing metals bound to the colloidal matter [25]. At some point most of the salt
has been removed so that there IS little change m ionic strength and no further deflocculatlon
of se&ment colloidal matter
Desorptlon isotherm analysis shows that S/S converted the nonclassical release character-
iStlCS of the untreated sediment to classical type desorption The conversion of contaminant
interphase transfer to desorptlon processes that show decreased contaminant release as leach-
lng proceeds would be expected to be an improvement over processes that show increased con-
taminant release as leaching proceeds However, conversion from nonclassical to classical
desorption was accompanied by higher concentrations of copper and nickel
Discussion
General Imphcatlons
The physical and chemical mechanisms responsible for contaminant immobilization are
complex, and detailed investigation of the factors affecting immobilization of metals in New
Copyright
Bedfordby ASTM Int'l (all rights
Harbor reserved); Mon
sediment byFeb
S/S24 23:07:03 EST 2020are not within the scope of this study The data
processing
316 HAZARDOUSWASTES
m this study, however, have some important implications Leaching of cadmium, chromium,
lead, and zinc was eliminated or substantmlly reduced, whereas leaching of copper and nickel
was increased by S/S processing These data suggest that stablhzatlon against aqueous leaching
is probably metal speofic, and the mechamsms responsible for immobihzatlon may ddt'er sub-
stantlally among various metals Bishop [ 13] also observed substantial differences in the leach-
ability of various metals after S/S processing Bishop [13], however, found that cadmium was
more readily leached from a synthetic hydroxide sludge than lead or chrommm In this study,
lead and c h r o m m m were more readily leached than c a d m m m Neuwirth, Mlkula, and Han-
nak [29], using scanning and transmission electron microscopy, obtained results indicating
that the containment mechamsms responsible for stablhzatlon o f c a d m m m and chromium In
stablhzed/sohdlfied synthetic sludges were different from those for lead The sequential batch
leach data m this study suggest similar lmmoblllzanon mechanisms for cadmium, chromium,
and lead relative to the mobilization of copper and mckel The work of Bishop [13] and Neu-
Wlrth, Mlkula, and Hannak [29] taken with the data obtained in this study provide little evi-
dence for a c o m m o n chemical stabilization mechanism for metals Immobilization of metals
by S/S processing may, therefore, be highly metal and waste specific
The data in this study showed significant mobilization of copper and mckel by S/S process-
mg of New Bedford Harbor sediment The theoretical implication of these data are not well
understood, but S/S processing clearly increased copper and nickel concentrations and cumu-
lative mass leached by altenng fundamental processes governing mterphase transfer from con-
taminated solids to water Other researchers have reported mobilization of contaminants by
S/S processing [12,30,31] These researchers also reported reduced leaching of most of the
metals they investigated Bncka and Hill [12] reported mercury mobdlZatlon following S/S
processing of a synthetic hydroxide sludge Brlcka, Holmes, and Culllnane [30] reported
moblhzation of banum, chrommm, and iron m incinerator ash following S/S processing,
depending on the S/S process used Mahloch, Averett, and Bartos [31] reported significant
moblhzaUon of sulfate and other ions contnbutmg to electncal conductivity in some of their
experiments with various real world wastes and propnetary S/S processes There is therefore
a potentml for misapplication of S/S technology, and reference leaching of untreated materials
m laboratory tests is needed to fully document anticipated benefits of stabllizatlon/sohdlfica-
tlon processing
Limitations
Several important aspects of leach testing were not addressed in this study In particular, the
effects on metal leachability of cunng stabilized/solidified sediment in an aerobic, humidity-
controlled environment followed by crushing and selvlng is poorly understood Metal mobil-
ity in sediments is sensitive to oxidation-reduction conditions [25], but such effects were not
investigated m this study
Desorptlon Isotherm analysis is best applied when solid and leachate contaminant concen-
trations are at steady state at the end of each cycle Kinetic testing has shown that a 24-h shake
time is sufficient for approximating steady-state when leaching untreated sediment [25]
Kinetic tests on stabilized/solidified sediments were not conducted in this study, but a 24-h
shake rime is often used in batch leach tests [32-36], although not necessanly to approximate
steady-state conditions Thus, possible kinetic effects were not explicitly accounted for in pre-
panng the stabilized/solidified sediment desorption isotherms
Leaching of loose granular material in an agitated system maximizes surface area and allows
intnnslc contaminant release properties to be isolated from interpartlcle pore phenomena that
affect leaching o f m o n o h t h s The purpose of such testing is to determine if chemical alteration
of chemical
Copyright by ASTM Int'l forms has occurred
(all rights reserved); relative
Mon Feb 24 23:07:03 ESTto
2020untreated sediment and not for predicting field
leachate concentrations. Data obtained from leaching loose, granular material could be used
m a permeant-porous medm model as a worst-case scenario However, what should be reahzed
is that such a model may not be reahstlc unless water penetrates the sohd~fied/stab~hzed mate-
hal Mass transport models with other assumptions, such as the sohd phase d~ffuslon approach
[37,38], might be more appropriate for estimating field leachate concentrations, depending on
physical properties of the stablhzed/solldlfied material and site-specific engmeenng-deslgn and
hydrologic factors
Conclusions
S/S processing potentmlly reduces the amount of metal leached for some but not all metals
m contaminated estuarlne sediments Some metals m estuanne sediments, such as copper and
mckel, can be mobilized by S/S processing Mobilization of metals by stabd~zatlon/sohdlfi-
cat~on processing of New Bedford Harbor sediment was related to alteration ofmtnns~c release
properties Desorpt~on isotherm analysm showed that copper and mckel release was converted
from a comphcated, nonclassical desorpt~on process to a classical desorpUon process by S/S
processing.
Metals can differ substantmlly m terms of thew release charactensUcs m batch leach tests
designed to evaluate chemical stabilization The data m this study are suggestive of, but cer-
tainly are not proof for, h~ghly metal specific ~mmobfllzatlon mechamsms for stabxhzat~on/
sohd~ficat~on of contaminated estuanne sediments
The data show the merit of reference leaching of untreated materials Jn laboratory tests to
document the anUclpated benefits ofstabfllzatlon/sohdlfiCatlon processing Addmonal studies
are needed to validate the mobilization effects observed in this study and to help gain addl-
Uonal insight into chemical stabilization and moblhzatlon mechamsms
Acknowledgment
The tests described and the resulting data presented, unless otherwise noted, were obtained
from research conducted by the U.S A r m y Engineer Waterways Experiment Station for the
U S Environmental Protection Agency (EPA), Region I, as a component of the comprehen-
swe EPA feasibility study for the New Bedford Harbor Superfund S~te, New Bedford, MA.
Permission was granted by the Chief of Engineers to pubhsh this reformation
References
[1] Kata, D and Kubo, H, "'Several Sohdlfied Sediment Examples," Proceedings of the Nmth Annual
U S/Japan Experts Meeting, Management of Bottom Sediments Contammg Toxic Substances, U S
Army Engineer Water Resources Support Center, Fort Belvolr, VA, 1983
[2] Nakamura~ M ~``Expenences w1th the Stab~hzatl~n ~fSed~ments~ Pr~ceedmgs ~f the Nmth Annua~
U S/Japan Experts Meeting, Management ofBottom Sedtments Contammg Toxic Substances, U S
Army Engineer Water Resources Support Center, Fort Belvolr, VA, 1983
[3] Otsukl, T and Shlma, M, "Soil Improvement by Deep Cement Continuous Mixing Method and
Its Effect on the Environment," Proceedings of the Etghth Annual U S/Japan Experts Meeting,
Management of Bottom Sediments Contammg Toxtc Substances, U S Army Engineer Water
Resources Support Center, Fort Belvolr, VA, 1982
[4] Pojasek, R B, Toxtc and Hazardous Waste Dtsposal, Ann Arbor Science, Ann Arbor, MI, 1979
[5] Malone, P G , Jones, L W, and Larson, R J, "Guide to the Disposal ofChemlcally Stablhzed and
Sohd~fied Waste," SW-872, U S Envzronmental Protection Agency, CmcmnaU, OH, 1980
[6] Culhnane, M J, Jr, Jones, L W , and Malone, P G , "Handbook for StabdlzaUon/Sohdlficatlon
of Hazardous Waste," EPA/540/2-86/001, U S Enwronmental Protection Agency, Cmclnnah, OH,
1986
[ 7] Tlttlebaum, M E, Seals, R K , Cartledge, F K, and Engles, S, CRC Crtttcal Reviews m Envlron-

mentaIControl, Vol 15, No 2, 1985, pp 1179-1211
[8] Malone, P G and Jones, L W, "Scientific Basis of Hazardous Waste Immobilization,"in Hazard-
ous and Industrlal Sohd Waste Testing Second Sympostum, ASTM STP 805, R A Conway and
W P Gulledge, Eds ,AmerlcanSoclety for Testlng and Materials, Phlladelphla, 1983, pp 168-177
[9] Myers, T E, Francingues, N R, Jr, Thompson, D W, and Hill, D O, "Sorbent Assisted Sohdl-
fication," Proceedings Internatzonal Conference on New Fronttersfor Hazardous Waste Manage-
ment, EPA/600/9-85/025, U S EnvironmentalProtecUon Agency, Cincinnati,OH, 1985
[10] Shlvely, W, Bishop, P, Gress, D and Brown, T , Journal of the WaterPollutlon ControlFederatton,
Vol 58, No 3, 1986, pp 234-241
[11] C6t6, P L, Constable, T W, and Morelra, A, Nuclear and Chemzcal Waste Management, Vol 7,
1987, pp 129-139
[12] Bncka, R M, and Hill, D O, "Metal ImmobilizationBy Solidificationof Hydroxide And Xanthate
Sludges," Environmental Aspects of Stablhzatlon and Sohdtficatton of Hazardous and Radtoacttve
Wastes, ASTM STP 1033, P L C6t6 and T M Gllliam, Eds, American Society For Testing and
[13] Bishop, P L, Hazardous Waste & Hazardous Materials, Vol 5, No 2, 1988, pp 129-143
[14] Ezell, D and Suppa, P, "The Sohroc Process in North America A Stabdizatlon/SohdlfiCatlonTech-
nology for the Treatment of Metal-Beanng Wastes with Reference to ExtracUonProcedure Toxicity
Testing," in Envtronmental Aspects of Stabdtzatton and Sohdtfcatton of Hazardous and Radtoac-
tire Wastes, ASTM STP 1033, P L C6t6 and T M Gilham, Eds, American Society for Testing
[15] Oblath, S B, EnwronmentalSctenceand Technology, Vol 23, No 9, 1989, pp 1098-1102
[16] Myers, T E and Zappl, M E, "Laboratory Investigation of Organic Contaminant Immobilization
by Proprietary Processing of Basin F Liquid Rocky Mountain Arsenal, Denver, Colorado," Tech-
nical Report EL-87-11, U S Army Engineer Waterways Experiment Station, Vicksburg, MS, 1987
[17] Myers, q E and Zappl, M E, "New Bedford Harbor Superfund Project, Acushnet River Estuary
Englneenng Feaslblhty Study of Dredging and Dredged Material Disposal Alternatives Report 9,
Laboratory-Scale Apphcatlon of Sohdification/Stabihzation Technology," Technical Report EI-88-
15, U S Army Engineer Waterways Experiment Station, Vicksburg, MS, 1989
[18] Kowtes, B and Bishop, P, "Column Leach Testing of Phenol and Tnchloroethylene Stablhzed/
Sohdlfied with Portland Cement," Envtronmental Aspects of Stablhzatton and Sohdtficatlon of Haz-
ardous and Radloacttve Wastes, ASTM STP 1033, P L C6t6 and T M Gllham, Eds, American
Society for Testing and Materials, Philadelphia, 1989, pp 238-250
[19] Sheriff, T S, Sollars, C J, Montgomery, D, and Perry, R, "The Use of Activated Charcoal and
Tetra-Alkylammonlum-Substltuted Clays in Cement-Based Stabilization/Solidification of Phenols
and Chlonnated Phenols," Envtronmental Aspects of Stabthzauon and SohdtficaHon of Hazardous
andRadtoacttve Wastes, ASTMSTP 1033, P L C6t6 and T M Gllham, Eds, American Society
[20] Walsh, M B, Eaton, H C, Tlttlebaum, M E, Cartledge, F K, and Chalasani, D, Hazardous
Waste and Hazardous Matertals, Vol 3, No 1, 1986, pp 111-123
[21] Chalasam, D, Cartledge, F L, Eaton, H C, Tlttlebaum, M E, and Walsh, M B, Hazardous
Waste and Hazardous Matertals, Vol 3, No 2, 1986, pp 167-173
[22] Chou, A C, Eaton, H C, Cartledge, F W, and Tlttlebaum, M E, Hazardous Waste and Hazard-
ousMatertals, Vol 5, No 2, 1988, pp 145-153
[23] "Draft Feasibility Study of Remedial Action Alternatives, Acushnet River Estuary Above Coggesh-
all Street Bridge, New Bedford Site, Bristol County Massachusetts," NUS Corp, Pittsburgh, PA,
1984
[24] Averett, D E, "New Bedford Harbor Superfund Project, Acushnet River Estuary Engineering Fea-
sibility Study of Dredging and Dredged Material Disposal Alternatives, Report 3, Characterization
andElutriateTestlngofAcushnetRiverEstuarySediment,"TechmcalReport EL-88-15,U S Army
Engineer Waterways Experiment Station, Vicksburg, MS, 1989
[25] Myers,T E and Brannon, J M, "New Bedford Harbor Superfund Project, Acushnet River Estuary
Engineenng Feaslbihty Study of Dredging and Dredged Material Disposal Alternatives Report 5,
Evaluation of Leachate Quality," Techmcal Report EL-88-15, U S Army Engineer Waterways
Experiment Station, Vacksburg, MS, 1988
[26] "Test Methods for Evaluating Solid Waste," SW-846, U S Enwronmental Protection Agency,
Washington, DC, 1986
[27] Brannon, J M, Plumb, R H, and Smith, I, "Long-Term Release of Heavy Metals from Se&-
ments," in Contaminants In Sediments Volume II, R A Baker, Ed, Ann Arbor Science, Ann
Copyright byArbor,
ASTM Int'l
MI, (all rights
1980, reserved); Mon Feb 24 23:07:03 EST 2020
pp 221-226
[28] Stumm, W and Morgan, J J , Aquattc Chemtstry, J Wiley and Sons, New York, 1981, p 616
[29] Neuwlrth, M, Mlkula, R , and Hannak, P, "Comparatxve Studies of Metal Containment an Sohd-
lfied Matrices by Scanning and Transmission Electron Microscopy," EnvtronmentalAspects of Sta-
bthzatlon and Sohdtficatton of Hazardous and Radtoacttve Wastes, ASTM STP 1033, P L C6t6
and T M Gdham, Eds, American Society For Testing and Matenals, Phlladelphm, 1989, pp 201-
213
[30] Bncka, R M, Holmes, T, and Cullmane, M J, "An Evaluation of Stabfllzat~on/Sohd~ficatlonof
Fluldlzed Bed Incinerator Ash (K048 and K051)," Techmcal Report EI-88-24, U S Army Engineer
Waterways Experiment Station, Vicksburg, MS, 1988
[31] Mahloch, J L, Averett, D E, and Bartos, Jr, M J , "Pollutant PotenUal of Raw and Chemically
F~xed Hazardous Industrial Wastes and Flue Gas Desulfurlzat~on Sludges Intenm Report," EPA-
600/2-76-182, U S Environmental Protection Agency, Cincinnati, OH, 1976
[32] C6t6, P L and Constable, T W, "Development of a Canadian Data Base on Waste Leachablhty,"
Hazardous and lndustrlal Sohd Waste Testing Second Sympostum, ASTM STP 805, R A Conway
and W P Gulledge, Eds, Amencan Society for Testing and Materials, Phdadelphla, 1983, pp 53-
66
[33] Soyle, W C, Ham, R K, Pastene, J , and Stanforth, R, "Leach Testing of Foundry Process
Wastes," Hazardous and IndustrtaI Sohd Waste Testing Second Syrnposmm, ASTM STP 805, R
A Conway and W P Gulledge, Eds, American Society for Testing and Materials, Phdadelphia,
1983. pp 67-88
[34] Maskarinec, M P and Brown, D K, "Batch and Column Extractlon Procedures for Orgamcs m
Sohd Wastes," Hazardous and Industrial Sohd Waste Testing Second Symposium, ASTM STP
805, R A Conway and W P Gulledge, Eds, American Society for Testing and Materials, Phda-
delphm, 1983, pp 141-151
[35] Means, J C, Wood, S G , Hasset, J J , and Banwart, W L, EnvlronmentalSctenceand Technology,
Vol 14, No 12,1980, pp 1524-1528
[36] Griffin, R A, Sack, W A, Roy, W R, Amsworth, C C, and Krapac, I G, "Batch-Type 24-h
Distribution Ratio for Contaminant Adsorption by Soll Materials," Hazardous and Industrtal Sohd
Waste Testing and Disposal Stxth Volume, ASTM STP 933, D Lorenzen, R A Conway, L P
Jackson, A Hamza, C L Perker, and W J Lacy, Eds, American Socxetyfor Testing and Materials,
Phdadelphia, 1986, pp 390-408
[37] C6t~, P L and Isabel, D, "Application of a Dynamic Leaching Test to Sohd~fied Hazardous
Wastes," Hazardous and lndustrtal Waste Management and Testing Third Symposmm, ASTM
STP 851, L P Jackson, A R Rohhk, and R A Conway, Eds, American Society for Testing and
[38] McIntyre, P F, Oblath, S B, and Wllhlte, E L, "Large-Scale Demonstration of Low-Level Waste
Sohd~ficatlon xn Saltstone," Envtronmental Aspects of Stabdlzatzon and Sohdlficatton of Hazardous
andRadtoactlve Wastes, A S T M S T P 1033, P L C6t6 and T M Gdham, Eds, American Society
for Testing and Matermls, Phdadelphia, 1989, pp 392-403
K n u d Brodersen, ~ Ole Hjelmar, 2 and Svend Mortensen 3
Cement Conditioning of Waste Materials and

Polluted Soil Using the GEODUR Process
REFERENCE: Brodersen, K , Hjelmar, O , and Mortensen, S, "Cement Conditioning of Waste
Materials and Polluted Soil Using the GEODUR Process," Stabdlzatzon and Sohdtficatmn of
Hazardous, Radioactive, and Mlxed Wastes, 2nd Volume, STP 1123, T M Gllham and C C
Wries, Eds, American Society for Testing and Materials, Philadelphia, 1992, pp 320-337
ABSTRACT" In this work, two areas of apphcatIon of the GEODUR additive in cement stabl-
hzatlon of waste materials have been investigated stabihzatlon of radioactive contaminated soil
and stabdlzatlon of municipal sohd waste incinerator ash Preliminary experimental work on a
clayey soil contaminated with radloactave cesmm and strontmm has indicated that the
GEODUR process IS a technically feasible method for soil solidificatmn The retarding effects of
humic materials in the sod are eliminated by the additive even at low cement contents The sohd-
lfied soil is not particularly strong, but has satlsfactory water permeablhty Retention ofcesmm
is reasonably good, but not as good as for the untreated soil Retentmn of strontium is not good
but is considerably improved by carbonation The volume stability dunng permanent ~mmer-
stun of the solidified products in water is satisfactory, but crack formaUon dunng dryout cannot
be excluded
In a parallel experimental investigation of the effect of stabihzlng incinerator ash with cement
with and without addition of GEODUR it was found that the addmve reduces the leachable
fractions of copper, zinc, cadmium, and lead Addition of GEODUR was also found to reduce
the rate of release of cadmium, copper, and lead (and possibly zinc) from cement-stabilized
Incinerator ash an a m a n n e (seawater) envaronment Incinerator ash stabdlzed with 7 4 and 13%
sulfate-resistant cement was investigated It seems likely that apphcatmn of the GEODUR pro-
cess m some cases may reduce the amount of cement necessary for stablhzation of incinerator
ash without sacrificing waste form performance The results presented are based on a limited
number of results from continuing experimental programs which are expected to produce a more
complete basis for firmer and more quantitative conclusions
KEY WORDS: additives, cadmium, carbon dioxide, cements, cement stablhzatlon, cesium iso-
topes, copper, diffusion, GEODUR process, heavy metals/trace elements, incinerator ash, lead,
leaching, leach tests, m a n n e disposal, mechanical properties, physical properties, radioactive-
contaminated soft, radlonuchdes, soils, sohdlfication, strontmm isotopes, zinc
T h e G E O D U R process is a m e t h o d for m a k i n g c o h e r e n t a n d m e c h a n i c a l l y r e a s o n a b l y
strong p r o d u c t s f r o m o r d i n a r y sod or o t h e r materials using only a small q u a n t i t y (for example,
8 wt%) p o r t l a n d c e m e n t .
A m o d e r a t e to high c o n t e n t o f h u m l c materials in soils will n o r m a l l y retard t h e h a r d e n i n g
so t h a t satisfactory c e m e n t e d p r o d u c t s c a n n o t be o b t a i n e d e v e n with high cement-to-soil
ratios A small a m o u n t o f the special G E O D U R additive, 4 a n i n e x p e n s i v e a n d n o n t o x i c 1%
solution, e h m l n a t e s t h e r e t a r d a t i o n a n d favors t h e h y d r a u l i c reactions Beneficial effects have
Section head, Rlso National Laboratory (RNL), 4000 Roskilde, Denmark

2 Section head, Water Quahty Institute (VKI), Agern Alle I l, 2970 Horsholm, Denmark
3 Director, GEODUR A/S, Denmark
4 The additive IS a proprietary component Questions concerning properties and so on should be
directed to GEODUR A/S, 2840 Holte, Denmark
320

BRODERSEN ET AL ON CEMENT CONDITIONING 321
also been demonstrated when the additive is used in making concretes from various waste
materials without significant organic content Only conventional but efficient concrete mixing
equipment is needed in the process
The method can be used for construction purposes, for example, road construction [1 ] and
in waste management. It has been tried on several waste matenals, demonstrating the low con-
sumption of cement [2] However, documentation of the retention of the toxic components
and the long-term behavior of the conditioned matenals under disposal conditions is necessary
before the use of the method m practice This has been studied for radioactive contaminated
soft at Rlso Natmnal Laboratory and for mumclpal sohd-waste incinerator ashes at the Danish
Water Quahty Institute
Soil contammated with radlonuchdes may arise from research, from mining and mllhng,
and from accidents of various types, or it may be detected during decommissioning of nuclear
mstallatmns The radionuchdes involved as well as the amounts of soil and the level of activity
may vary within wide hmlts In some cases disposal on site or in low-level repositories of the
material in uncondmoned form is possible However, as an option it is valuable to know
whether condmonmg of radioactive contaminated soil xs feasible and can be motivated by
safety gains In addition to use as a conditioning method for contammated soft and waste
matenals, the G E O D U R process may also be of interest m the construction of barriers or cover
layers for piles of contaminated soft, mill talllngs, etc
The G E O D U R cementation process has therefore been selected for study within a Nordic
cooperative project, which also comprises a general discussion of management modes for
radmactlve contaminated soil. The project IS part of a larger research program financed by the
Nordic Nuclear Safety Authontles [3] The results presented In the first part of this paper are
mainly from a prequahficatmn study [4] for the Nordic project The experimental methods
were developed within the European Commumtxes Research Programmes on Radioactive
Waste Management [5-7]
In the case of municipal solid waste incinerator (MSWI) residues, m a n n e disposal is often
an attractwe waste management option, especially ff an aspect of utilization is included (for
example, land reclamatmn or construction of artificial reefs) However, due to the potential
leachability of contaminants, particularly trace elements/heavy metals [8, 9], from untreated
granular MSWI residues, it may be necessary to dispose of the material m the form ofcement-
stabihzed blocks in order to avoid unacceptable environmental impacts
Results from a study of the potential of the G E O D U R additive for improving the perfor-
mance of cement-stabihzed blocks of combined MSWI bottom ash and fly ash in a marine
environment are presented in the second part of this paper
Stabilization of Radioactive Contaminated Soil

Sample Preparatton
Topsoil from the Rise area was used as a basis for the expenments The soil is developed
from moraine material and is classified as sandy clay with 13% < 0 002 m m and 2 1% hurnlc
materials The 1on exchange capacity is ~ 2 2 meq/100 g soil at pH 8 2, increasing with pH
Immediately before the cemented products were prepared, ~34Csand 85Srplus inactive careers
were mixed into the semidry soil The same was done with the G E O D U R additive, followed
by dry mixing with the cement before addition of water The composition of the five investi-
gated products are given in Table 1 Products with 8 or 16% sulfate-resistant portland cement
(SRPC) and with 0, 2, or 6% of the solution o f G E O D U R addmve were made The high total
water content ~ 2 4 % ( W/C = 1 5 to 3) was necessary to make castable mixtures from this
Copyright
clayey by ASTM
soil Int'l (all rights reserved);
Samples from Mon theFeb 24 23:07:03with
material EST 2020
140/owater were prepared by ramming The mate-
TABLE 1--Compressive strength measured usmg 4 4-cm-dtameter cyhnders Oa = 8 8 cm) after 54

days of hardenmg at high humidity, and hydrauhc conducttvmes measured for ~ 2-cm-thtck slabs of
four of the materials
Product Composition, Density (Water Compressive Hydrauhc

Cement-GEODUR-Water Saturated), g/cm 3 Strength, MPa Conductivity, m/s
16-2-24 1 87 45 06 10-I~
8-6-25 1 83 13 20 10-E~
8-2-24 ! 89 15 23 10-t~
8-2-14 205 12 06 10 10
rials are referred to as, for example, the 8-2-24 or the 8-2 product The first figure gives the
cement content and the second the amount of the 1% solution of G E O D U R additive as per-
cent of the total solids The third figure gives the amount of water as percent of solids plus
water
Physical Properties
Measurements of compression strength and other physical properties of the cement-solidi-
fied soil have been made to supplement previous determinations made for G E O D U R A/S by
other institutes on similar products [2] The values given in Table 1 are in general agreement
with earher determinations
The compressive strength shows the expected increase with cement content The materials
are not strong compared with normal construction concrete, but for waste materials to be dis-
posed of, or for subsurface construction of barriers, they are probably good enough Change
in strength with carbonation, dryout, or long-term leaching should be investigated
The high water contents--with only a minor part bound as hydration water--result in low
densities ( ~ 1 8 to 1 9 g/cm 3) and high porosltes of the hardened materials ( ~ 3 0 to 45 volume
%) Although much of this must be in the form of interconnected pores, the hydraulic con-
ductivlty of the cemented soil is not partlcularlly high (see Table 1) The conductwIty is lowest
in the high-cement or the low-water products Values about 2 10 ~0m/s are also typical for
ordinary construction concrete The soil itself (with ~ 2 4 % water content) has a somewhat
higher conductivity, about 8 10 ~0m/s
Leaching Behavior
The leaching of 134Csand 85Sr from four of the cement-solidified soil materials were inves-
tigated using various procedures All the measurements were made at 20~ The system used
to study leaching from blocks ofcement-sohdlfied material is shown as an insert in Fig 1 The
samples were cast in the bottom of soft polyethylene bottles and hardened for ~ 1 month (with
closed lid) The upper surface ( ~ 4 2 cm 2) was then exposed to water or 3% sodium chloride
(NaC1) solution with and without access to carbon dioxide (CO2) from the atmosphere, that
is, as open or closed systems The leachant was only partially removed at each sampling The
leach curves for the 8-2-24 material are shown as examples in Fig 1a The accumulative activ-
ities removed with the leachant are plotted against the square root of time as the corresponding
equivalent thickness of completely leached material
For 85Sr in the closed system and also for ~34Csin both cases (although this cannot be seen
on the scale of the figure) the curves are nearly straight hnes indicating diffusion-controlled
leaching The diffusion coefficients given m Table 2 are calculated from the slope of the lines
For by85Sr
Copyright ASTMinInt'lthe open
(all rights system,
reserved); the
Mon Feb initialEST
24 23:07:03 rapid
2020 release of strontium, after an initial period of
BRODERSEN ET AL. ON CEMENT CONDITIONING 323
200 c m ~ 200 c m ~
1.0 t ~ ""1='~3 ~ i i i
tt~
0.8
Product 8 - 2 - 2 4 ~ " - .ose~
0.6 in water
0.4
_~ 0.2
~
'~Cs Closed
: : : : -_ : . - . Open I
0 2 4 6 8 10 12
Square root of leaching time in days
pH
14 ! !
pH development
13 Product 8 - 2 - 2 4 in w a t e r
Closed
12
11
10 Open
I i I I I I
0 2 4 6 8 10 12
Square root of leaching time in days
FIG.l--One-dimensional leaching of 134Cs and 85Srfrom cement-solidified soil in water with or without
access to C02fi'om the atmosphere (open or closed systems). The lowerfigure shows the corresponding
development in pH of the leach water (Product 8-2-24).
TABLE 2--Effecttve dlffus10n coeJfictents corresponding to the slope of the leach curves for 134Csand
SSSrfor four of the products m four different leachants Data for the pure sod (wtth somewhat lower
water contenO are also tncluded (unit 10-t2 m2/s)
Leachant
Product Composltxon,
Cement-GEODUR- Closed, Open, Water Closed, 3% NaCI Open, 3% NaC1
Water Water wath C02 Solutaon with C02
34Cs
16-2-24 0 035 0 02 0 21 0 13
8-6-25 0 000 6 0 0003 0 023 0 020
8-2-24 0 010 0 003 0 19 0 12
8-0-24 0 000 5 0 0002 0 024 0 019
0-0-15 0 000 002 02
85Sr
16-2-24 58 0 052 6 0 056
8-6-25 7 0 15 12 ~2
8-2-24 52 0 035 8 ~ 11
8-0-24 64 0 13 13 ~2
0-0-15 02 6
14 days, drops to much lower values However, this latter part of the curve is still reasonably
straight, permitting the calculation of an apparent diffusion coefficient, although the interac-
tion between diffusion and precipitaUon reactions may be quite complicated A complete set
of the curves are given in Ref4.
The diffusion coefficients in Table 2 show that cesium leaching increased in saltwater and
slightly decreased in systems open to the atmosphere The 8-2 material shows increased release
compared with the 0% control as well as the material with 6% G E O D U R The experiments
are too few to decide whether this unsystematic trend is real The higher cement content in the
16-2 material increases the release somewhat further, but due to the clay content in the soil it
still remains much lower than for ordinary concrete, where a typical diffusion coefficient for
cesium would be ~ 5 10-12m2/s [5,6] Results from leaching of the soil without cement or
the G E O D U R additive are also included in the table In water the diffusion coefficients are
even lower than for the best of the cemented products, but the difference disappears in
saltwater.
In the case of the closed systems, Fig 1a and the diffusion coefficients for 85Sr in Table 2
show that strontium is leached much more rapidly than cesium, which is the opposite of what
is normally observed for cemented waste products However, for systems open to the atmo-
sphere the release rates are much decreased, probably due to carbonate preclp~taton (This is
probably also the case for many literature values, indicating low strontium leaching from
cemented waste ) The observed initial delay of some 15 days is caused by the slow diffusion of
C02 in the water layer The improvement in leach rate may only be temporary since, as indi-
cated by the pH development in Fig. 1b, the HCO3 ions begin to be important m the open
systems after about 50 days This will result in renewed increased leaching of strontium as was
found particularly for the systems in saltwater
In another type of leaching experiment a sample consisting of small pieces of hardened
cemented soil (typical dimension ~ 5 mm, surface/volume ratm ~ 1000 m -l) was exposed
to water in open or closed systems as shown in the insert in Fig 2 The accumulative leached
fraction o f ~34Cs activity ~s given in the figure as a function of the volume of water that has
150 c m 3
~ 3
~0 cm;
-20 g
8 I 134 i , '_~w41 Closed system [
I C S leaching in w a t e r ~.,~,~ Open I
6[ -5mmbr~ ~ ~/~'~'~]6"2
,=
2i Z / _ _ 8-o i
0 8-6 and open
0 20 40 60 80
Volume of water contacted with product, cm3/g
FIG. 2--Fractional lZ4Cs leaching in water from small broken pieces offour types of cement-solidified
soil, with and without access to COe.
contacted the material or rather the accumulative liquid-to-solid ratio. All eight investigated
systems are included.
It appears that cesium is leached most easily from the 16-2 product; the 8-2 product is inter-
mediate while the control and the 8-6 material show similar lower values. The influence of
CO2 is slight, which is in agreement with the results given above for the experiments with solid
blocks. However, when the broken pieces were predried for about 50 days in laboratory air
before initiating leaching, cesium releases decreased considerably and the order of the quality
of the materials changed completely (see Fig. 3). Now the high-cement material is the one with
lowest release. These effects are thought to be associated with changes in cesium fixation to the
clay minerals.
The accumulative fractions of 85Sr leached from small broken pieces of Product 8-2-24 (as
hardened or hardened and predried) are shown as examples in Fig. 4. The influence of car-
bonation is clearly seen. The predried material was exposed to CO2 during the ~50 days in
laboratory air. This eliminates the initial rapid strontium release from the freshly prepared
samples. However, the slope of the latter part of the curves for the two closed systems are more
or less the same, and the releases are still quite considerable.
This type of leaching experiment is quite severe and the bulk of the material is leached con-
siderably: up to about 25% of the calcium was removed. There was only a minor difference
between the calcium release from the hardened materials and from the materials predried after
~ 0.5 I I i
I 1341_ a
I ~ s leachingin water
8-2
t~ 0.4
Closed systems
8-6
0.3
8-0
&
0.2
16-2
~ 0.1
0.0 II ~ I I [ I
0 20 40 60 80
Volume of water contacted with product, cm3/g

FIG 3--Predrwd materials fractional leaching of l34Cs m waterfrom small pieces of four types of
cement sohdlfied sod (closed systems)
hardening, although calcmm hydroxide [Ca(OH)2] from the cement hydratmn m the latter
case must have been converted to calcmm carbonate (CaCO3) before the immersion The high
leachablhty for calcmm indicates that internal high pH values could be a rather temporary
phenomena m such products
Some work on 65Znleaching was also done within th~s study The retention was considerably
better than for cesium, and the zinc isotope could not be detected in any of the leachates
80 I I I !
SSSr leachingin water

ordinary and p r e ~ g'i' cioseu
M
60
~ 8-2, openO
& 40
I,}
~ 20
I I I
0
0 20 40 60 80
V o l u m e of w a t e r c o n t a c t e d with p r o d u c t , cm'/g
FIG 4--Fractional leaching ofSSSr m waterfrom small pieces of the 8-2 material, with and without
access
Copyright to C02
by ASTM Results
Int'l (all for predrted
rights reserved); Mon Feb bits of theEST
24 23:07:03 8-22020
product are also included
Volume Stablhty and Water Uptake

For use as a waste-condiuonxng method, and especially if the cemented soil products are to
be used for b a m e r construction, a certain volume stability of the materials ~s required This
posslbdlty has been stu&ed using spherical samples (diameter 4 7 cm) immersed permanently
in water or exposed to wet/dry cycling (see insert above Figs 5 and 6) In the dry period the
samples are simply left hanging m the laboratory air The volume change and the water uptake
in the samples were followed using a combination of weighing immersed In water and in air
[ 7] The leached material was estimated from calcium analyses after aci&ficaUon of the leach
water (the systems were open to the atmosphere).
Figures 5a and 5b show the swelling, water uptake, and leached materials for the perma-
nently immersed samples of products with 2% G E O D U R and 16 and 8% cement, respectwely
The swelling is moderate, ~ 0 2 to 0 3 vol%, and probably without much practical importance
The internal porosity is somewhat increased, about 8 vol% according to the water uptake,
which may possibly have some influence on, for example, hydrauhc conductivity
Figures 6a and 6b show the swelling and water uptake for the same products under wet/dry
cycling The 8-2-24 product contracts about 6 vol% or ~ 1% lineanly when dried The value
for the 16-2-24 product is about halfa percent The water uptake Into coarse pores is very rapid
when the samples are re~mmersed, and the sample swells again to nearly the original volume
The cycling has not damaged the small-size samples used m these experiments, but the con-
traction may give cracks in larger constructions ifdryout is a possibdlty
Stabilization of Municipal Solid Waste Incinerator (MSWI) Ash

Characterization of Granular MSWI Ash
The MSWI ash used in this study was combined bottom ash and fly ash from a moving grate
incinerator equipped with electrostatic precipitators but without scrubbers The ash sample
was collected after quenching and draining as a large number of subsamples dunng a period
of one week Particles larger than 45 m m (mostly scrap metal) were removed from the ash
sample, which was then air dried and crushed in a jawcrusher until it passed a 10-mm screen
Uncrushable p~eces of metal that did not pass the screen were removed The particle-size dls-
t n b u h o n of the crushed sample is shown in Fig 7
A representative subsample of the ash was further crushed and subsequently ground to
< 0 090 m m in a tungsten carbide gnnder The ground sample was subjected to chemical anal-
ysis Most major elements were determined by X-ray fluorescence, trace elements by atomic
absorption spectrophotometnc (AAS) methods after total digestion Sulfur was determined by
combustion followed by titration, and C1- by alkahne fusion followed by acid extraction and
spectrophotometry p H was determined in a 1% ash slurry in demlneralized water after 0 5 h
of stlrnng under cover, acid neutrahzation capacity (ANC) by bolhng with diluted hydrochlo-
ric acid (HC1) and back titration with sodium hydroxide (NaOH) to pH = 7, loss on lgnmon
(LOI) by heating to 550~ for 2 h, and total orgamc carbon (TOC) by treatment with HC1,
oxidaUon at 800~ to CO2, reduction to methane, and detection on a flame ion detector (FID)
The results are presented in Table 3
In order to investigate the leaching properties of the granular MSWI ash in a manne envi-
ronment, the ash was subjected to four consecutive batch extractions with ocean water (from
the Sargasso Sea, salinity = 3 6%, pH = 8 0) Each extraction was performed at a liquid/solid
ratio of 25 L/kg and a contact time of 23 h The filtered extracts were analyzed for arsenic,
cadmium, copper, lead, and zinc by appropnate AAS methods. The accumulative leached
amounts of these elements are presented in Fig 8 as a function of the accumulaUve hquld-to-
solidbyratio
Copyright L/S.
ASTM Int'l (all The pH changed
rights reserved); Mon Feb 24from 8 EST
23:07:03 L/S = 25 to 8 3 at L/S = 100 The results ln&cate
6 at2020
permanently
in water
mm
=,
0
0.4 1 i l l i
0.03
0.3
0.02
0.2
0.01
0.1
0.0 , j , , , , 0.00
0 50 100 150 days
"~..~. 0.10
9 0.08
N o.06
0.04 9 t / 16-2 .
0.02
go 0 50 100 150
Time in water, clays
FIG. 5--Swelling of and water uptake in and leached material from spherical samples of Products 8-2
and 16-2 continuously immersed in water, with C02 access.
f
~ J
Wet Dry
mm
~" 1
0.0
1
~ -1
-0.2
-3
-5 -0.4
-7 I I f t I I
0 50 100 150 d,q Is
c~ 0.1
,'~ 0
2
-0.1
-0.2
wet dry wel dry w 4 dry wet dry wet dry ~:

-0.3
0 50 100 150
Time under wet/dry cycling, days
FIG. 6--Swelling(or rather contraction) and water uptake for spherical samples of Products 8-2 and 16-
2 undergoing wet~dry cycling.
SAND GRAVEL
V 100
"'
,.,J
90
o
~ 80
'~ 70
0u. 60
~ 5o
-~ 40
(-I 30
z 20
MSWI ASH
o 1()
g,l
a. 0 J , , a w i * 1 " , i ' 1
0.1 0.5 1.0 5 10

PARTICLE SIZE, d (ram}
FIG 7--Partlcle-slze dlstrzbutton of crushed granular M S W I ash
that c a d m m m , copper, zinc, a n d particularly lead m a y c o n t i n u e to leach from granular M S W I

ash over a large L / S range m q u a n t i t i e s that m a y well be unacceptable m a m a r i n e e n v i r o n -
m e n t C e m e n t stabilization o f the ash m the f o r m o f blocks with relatively low s u r f a c e / v o l u m e
ratios before e x p o s u r e to the m a n n e e n w r o n m e n t m a y reduce the release o f trace e l e m e n t s to
an acceptable level
TABLE 3--Composttton of combined 34SWI ash used m the

stabthzatton tests
Element g/kg Element mg/kg
$1 218 Ag 83
A1 52 9 As 41
Cl 63 Ba 1800
Fe 117 Cd 23
Ca 82 2 Cr 390
K 12 8 Cu 4800
Mg 12 5 Hg 0 28
Na 24 7 Mo 64
S 46 Nl 150
P 3 66 Pb 5200
T1 6 77 Se 22
Mn 1 46 Sr 100
Zn 9100
NOTE pH, 1% slurry = 9 9

ANC, eqv/kg = 3 1
LOI, g/kg = 19
Copyright by ASTM Int'l (all rights reserved); Mon Feb
TOC, 24 23:07:03
g/kg = 7 8EST 2020
GRANULAR COMBINED MSWI ASH

As Cd Cu Pb Zn
.~. t4
[] & 0 9 A
12
E 10
"5
20 40 60 80 00
L/S
FIG 8--Results of multiple batch leachtng of granular MSW1 ash with ocean water, accumulative
amounts of leached trace elements versus L/S
Stablhzatton o f M S W I Ash
The metal pleces removed on the 10-mm screen (approximately 2% of the total weight) were
returned to the screened MSWI ash that was then used for the stablhzatlon tests The MSWI,
sulfate-resistant low-alkah cement, G E O D U R addmve, and water were m~xed thoroughly m
four different combinations (see Table 4) Test mixtures with two different contents of cement
(7 4 and 13%) were prepared, both without and with addmon of G E O D U R The mixtures
were cast in plastic molds with dmmeters = 10 5 cm, settled by vibration, and cured for seven
days at 40~ and 100% relaUve humidity Cube-shaped specimens (approximately 6 by 6 by 6
cm) were cut out of the cured test materials and used m tank leaching experiments
Table 4 shows the dry densRles of the test matenals after seven days of cunng The densxty
appears to increase slightly when the content of cement is increased and when G E O D U R is
added Due to lack of material, the compressive strengths could only be determined using non-
TABLE 4--Composmon of the stablhzed waste forms All percentages are based on dry weight
Approxtmately 12% (of the total dry wetght) of water was used m all cases
Sample Combined MSWI Sulfate-Resistant Low GEODUR Additive, Dry Density,

Numbers Ash Alkali Cement Dry Material g/cm3
1 92 6% 7 4% None l 80
2 87% 13% None 1 91
3 92 6% 7 4% 0 01% 1 86
4 ASTM Int'l (all rights87%
Copyright by 13%
reserved); Mon Feb 24 23:07:03 EST 2020 001% 1 96
standard cyhnders of height = 5 cm and dmmeter = 10 5 cm For these samples the uncon-
fined compresswe strength was found to increase from ~ 2 5 to ~ 6 0 MPa when the cement
content increased from 7.4 to 13%. The data are not sufficientlyaccurate to determine whether
the addmon of G E O D U R affects the compressive strength
Leachable Fractions
A measure of the leachable fractions, f of trace elements In the stablhzed products were
obtained by grinding each of the products to 95% < 0 090 m m and subjecting them to batch
extraction w~th demmerahzed water at a constant pH of 4 for 4 h at a hquld-to-sohd ratio of
100 Feedback control and automatic addmon of dilute mtnc acid was used to maintain pH
= 4 The leachable fractlonfls defined as the amount of a particular trace element extracted
(determined by AAS analysis after filtenng the extract through a 0 45-~m filter) relative to the
total a m o u n t ofthat element present In the product [ 10] The fractlonsf or the corresponding
leachable amounts, can be regarded as measures of the avaflablhty of the various elements for
leaching from that particular material over a long period of time The test does not s~mulate
normal natural leaching condmons, and the results should never be interpreted an terms of
concentrations of trace elements in the extracts
It can be seen from the results of the avadablhty tests in Table 5 and Fig 9 that the addmon
of G E O D U R reduces the f-values substantially for copper, zmc, cadmmm, and lead The
reduction for copper is most pronounced, 44 to 45%, whereas the reducuon for lead ~ssmallest,
7 to 13%. Except for copper, the largest reduction ofJis seen at the lower cement content Since
the test materials were ground before extraction, the reductions of f apparently must be attrib-
uted to chemical binding reduced by the G E O D U R additive
Tank Leachmg of Stabthzed MSWI Ash with Seawater

The cut-out cubes of the four stablhzed products were subjected to tank leaching wath ocean
water at 20~ In the tank-leaching experiments, each of the monohthac waste forms (surface
area = ~ 0.0216 m 2 and volume = ~ 0 000 216 m 3) was suspended m a closed seawater-filled
polypropylene tank by a nylon stnng At certain tame antervals (0 7, 1, 3, 6, 11, 18, 92, and 47
days), the water (0 0009 m 3) was replaced by new seawater and analyzed for pH and trace ele-
ments after filtration through a 0 45-#m filter
The flux, J(mg]m2/day), of a certain element or ~on out of the stablhzed waste forms d u n n g
each cycle can be calculated from the expression
J = ( C - Co) VsM(S At),
where
C = concentration of the element In question m replaced seawater, mg/L,
Co --- background concentration of element m seawater, rag/L,
Vsw = volume of replaced seawater, m 3,
S = surface area of waste form, m 2, and
At = duration of cycle, days
The log-log plots of J versus Ume (data points represent the average flux d u n n g each cycle
versus the total time at the end of that cycle) and the accumulative leached amounts (mg/m 2)
versus time, which are shown in Figs 10a, b, c, and d, indicate that the addmon of G E O D U R
does reduce the rate of release of cadmium and copper Similar effects were observed for lead
and by
Copyright zinc,
ASTMalthough
Int'l (all rights a n u m bMon
reserved); e rFeb
of24these concentrations
23:07:03 EST 2020 were below the detection hmat
TABLE 5--Percentage reduction zn leachable fracttons f by use

o f the GEODUR addtttve
( l -fwllh GEODUR/f',vlthoutGEODUR) 100

Cement
Content Cadmmm Copper Lead Zinc
7 4% 15% 44% 13% 25%

13% 10% 45% 7% 17%
The slopes of the curves in Fig 10 generally point to a decrease in avadablhty for leaching
(m some cases following lmtml surface d~ssolutlon) rather than simple d~ffUSlOn as the rate-
controlhng mechanism for release o f c a d m m m and copper [ 10] Calculation of d~ffUSlOncoef-
ficients was therefore not attempted.
The accumulative amounts of cadmium, copper, lead, and zinc release from each of the four
blocks after 10 7 days are presented in Table 6 For cadmium and copper the reduction of the
released quantlUes resulting from the addmon o f G E O D U R ~s of the same order of magnitude,
20 to 50% m all cases It can be seen that the mitigating effect on cadmium and copper release
LEACHABLE FRACTION OF CADMIUM LEACHABLE FRACTION OF COPPER

L / S = 100, pH = ~-, grmn raze < 0 0 9 m m L/S = 100, pH = 4, grain size < O O g m m
100 20
8 8O
/
I~ 60
w 10~- NO G4odur
a 4.0 O01:Z Geodur

5 o
~s
= 20
0 0 7 4 % cement 7 4~ cw1"mrrt 1.] X cerne,~t ~3 9 C ~

7 4 ~. clwn~nt 7 4 Z r 1"~ % c l r / ~ t 1.3 ~ r
L E A C H A B L E F R A C T I O N OF" LEAD LEACHABLE FRACTION OF ZINC

L/S = 100, pH = 4, cJrom size < 0 09 mrn
20[
L/S = 100. pH = 4, gram s~ze < 0 0 9 mrn
100
r
o 15
~ 60
~I0 NO C,t~ur
NO G~w:wJ
=e 40
~s
_~ 20
~ 0 0 74 %r 74 X c ~ X c ~ 13 % r
FIG 9--Leachable fractions o f cadmmm, copper, lead, and zinc in cement-stabdlzed M S W I ash,
ground samples in demlnerahzed water maintained at p H 4
t .I-
!
I
"10 v
0
! i i
I i ~I I ,i i o
-o ~ ';o 'o 'o "to ?o "o o o o
(~p/~u~w) r 'xnl.,-I (~w/'l~w) ~,unmuo peqaoe I pelolnuana~v
la
I §
o
I
r~
o~
I::
i i i o
r
O 0 0 0
o ~ b o ~o o o
(~op/~tu/SuJ) [ 'xnl_-I (~uJ/'BLu) I.unowo peq~0e I pelotntun~:~V
TABLE 6--Accumulattve quantttles of trace elements On mg/m 2)

released from cement stabElzed M S W I ash after 10 7 days
exposure to seawater m tank-leachmg expertments
Cement GEODUR Cadmmm, Copper, Lead, Zinc,

Content Addmve mg/m 2 mg/m 2 mg/m 2 mg/m 2
74% - 025 110 037 <17

74% + 015 72 <019 <17
13% - 0039 46 041 <17
13% + 0021 37 <017 <17
by using 13% c e m e n t instead o f 7 4% exceeds t h e effect o f using G E O D U R T h e release o f lead,

however, is r e d u c e d substantially by t h e a d d m o n of G E O D U R m b o t h cases, b u t barely
affected b y t h e increase m c e m e n t c o n t e n t Nearly all c o n c e n t r a t i o n s of zinc were below t h e
d e t e c t i o n limit In each p a r t i c u l a r a p p h c a t l o n , the target values o f trace e l e m e n t release a n d
the relative costs o f c e m e n t a n d G E O D U R will d e t e r m i n e w h e t h e r or n o t the use o f the addi-
tive is technically a n d e c o n o m i c a l l y attractive
Comparison
A c o m p a n s o n b e t w e e n results f r o m t h e two parts o f t h e study is a t t e m p t e d m T a b l e 7 T h e
e q u i v a l e n t t h i c k n e s s o f c o m p l e t e l y leached layers after 10 7 days' e x p o s u r e to s a h n e w a t e r are
given for t h e two types o f m a t e r i a l s a n d for c a d m i u m , copper, a n d cesium, s t r o n t i u m , respec-
tively T h e m u c h b e t t e r r e t e n t i o n o f t h e h e a v y m e t a l s in t h e ash m a t e n a l t h a n t h a t o f the radio-
isotopes in the solidified soil is clearly d e m o n s t r a t e d T h e m a i n reason is p r o b a b l y the different
c h e m i s t r y a n d r e t a r d a u o n m e c h a m s m s for the four p o l l u t a n t s at the high p H of the c e m e n t e d
materials, b u t o t h e r differences such as s t r u c t u r e o f the cement-solidified ash a n d soil m a y also
c o n t r i b u t e A l t h o u g h r e t e n t i o n o f c e s i u m a n d p a r t i c u l a n l y s t r o n t i u m is s o m e w h a t unsatisfac-
tory, c e m e n t sohdlficaUon o f soil p o l l u t e d with h e a v y m e t a l s m a y still give excellent results
Th~s c o n c l u s i o n is s u p p o r t e d by t e n t a t i v e results for zinc, where a leached thickness o f < 0 1
u m after ten days is i n d i c a t e d for t h e c e m e n t e d soil
TABLE 7--Compartson of equivalent thtcknesses On ~m) for pollutants leached from blocks of
cemented MSWI or cemented sotl after 10 7 days m sahne water"
Tank Leaching of 6- by 6- by 6-cm Cubes of One-Dimensional Leaching of Cemented Clayey

Cemented MSWI Topsoil in Flasks
Cement GEODUR Cadmmm, Copper, Cement GEODUR Cesmm, StronUum,

Content Ad&tlve ~m ~m Content Additive um um
7 4% - 7 15 8% -- 210 4070
+ 5 15 + 620 3230
13% -- 11 4 16% -
+ 07 6 + 620 3060
a The leached thicknesses are calculated from the accumulative leached amounts (mg or Bq/m 2)
&vlded by the product densities and the concentration (mg/kg) of cadmium or copper extractable at pH
4 (see F~g 9) or the total concentration (Bq/kg) m case ofcesmm or strontmm
Another important difference between the two matenals is the apparent differences in reten-
tion mechanisms However, it cannot be decided from these short-term experiments whether
the rapid decrease in leach rate with time for the heavy metals in the stablhzed MSWI material
is dependent on the high pH in the matenal so that the situation would change with extensive
leaching of the cement components
Conclusions
Based on these expenmental investigations, the following conclusions can be drawn regard-
lng stabdlzat~on of radioactive contaminated soft
1 The G E O D U R process Is a techmcally feasible method for soil sohdlficatlon The main
advantage ISthe low cement consumption combined with the possibility of cement solid-
lficatlon of soil with a considerable content of humlc material
2 The compressive strength of the cemented soil is not high, but probably sufficient for the
purpose
3 The permeability for water is satisfactory
4 The retention of cesium is reasonably good, but not as good as for the untreated soil
5 The retenUon properties for strontium are not good but are considerably improved by
carbonation of the matenal
6 The products are reasonably volume stable when permanently immersed in water
7 Wet/dry cycling gives somewhat larger volume changes, and drying may lead to crack
formation
The results are preliminary since they are based on relatively few experiments with only one
soil type The main effect of the G E O D U R addlUve is to make soil solidification possible, while
the leaching behavior seems more to be determined by interaction between the pore structure,
retention on the clay particles, and pore water chemistry as modified by the presence of
hydrated cement minerals and interaction with CO2 from the surroundings Some interesting
mechanisms such as the effect of drying on cesium retention were demonstrated
The following conclusions can be made regarding MSWI ash stablhzatlon
1 Addition of G E O D U R reduces the leachable fractions of copper > zinc > cadmium >
lead in cement-stabilized MSWI ash, particularly at low cement contents (7 4% cement).
2 Addition of G E O D U R reduces the flux ofcadmlum and copper (and lead and zinc) from
cement-stabilized MSWI ash In a marine environment The best overall reduction was
seen for the high (13%) cement content
3 The rate of release of cadmium and copper from cement-stabilized MSWI ash (with and
without G E O D U R additive) appears to be controlled by decreased availability rather
than diffusion.
4 The addition of G E O D U R may reduce the amount of cement necessary for stabihzatlon
of MSWI ash without sacrificing waste form performance
Also for MSWI ash, the conclusions are based on a limited number of lnlhal results of an
ongoing study Further work is necessary to produce a more complete basis for firmer and
more quantitative conclusions
In general it should be pointed out that the studies have not been aimed at satisfying any
specific safety requirements, but to obtain an understanding of the behavior of the cemented
materials with or without G E O D U R additive under vanous experimental conditions
References
[1] Mortensen, S, "The Geodur SohdlflcaUon Process," abstracts, Second Forum on InnovatzveHaz-
ardous Waste Treatment Technologws, Phdadelphla, May 1990, U S EPA/540/2-90/009, 1990, p
25
[2] "Introduction to the Geodur System for Waste and Sod," GEODUR A/S, Maaloev, Denmark,
1988
[3] "Det Nordlska Karnsakerhetsprogrammet 1990-1993," Nordic Nuclear Safety Research, NKS
Report No (89) 15, R~so National Laboratory, Denmark, 1989
[4] Brodersen, K and Vmther, A, "Charactenzat~on of Waste Products Prepared from Radioactive
Contaminated Clayey Soil Cemented According to the GEODUR Process," Rlso-M Report No
2909, RlSONational Laboratory, Roskllde, Denmark, 1990
[5] "Characterization of Radloactwe Waste Forms," Progress Report 1987, Nuclear Soence and Tech-
nology Report No EUR 12077 EN/1, Brodersen, K and Nllsson, K, Eds, Commission of the Euro-
pean Commumty, Luxembourg, 1989, Vol 1, pp 18-31 and 166-169
[6] Brodersen, K and Nllsson, K, "Mechanisms and Interaction Phenomena Influencing Releases m
Low- and Medium-Level Waste Disposal Systems," Final Report 1986-1990 for EC contract FI 1W-
0089-DK, Rlso-M Report No 2908, Riso National Laboratory, Denmark, 1990
[ 7] Brodcrsen, K, "Hygroscopic Water Uptake and Dry/Wet Cychng Compared with Normal Water
Exposure of Cemented Waste Forms," Scwnttfic Basis for Nuclear Waste Management XII, Berhn
1988, Material Research Sooety Symposmm Proceedings, Vol 127, Pennsylvama 1989
[8] Hjelmar, O, "Leachate from Incinerator Ash Disposal Sites," Proceedmgs of the International
Workshop on Mumctpal Waste Incmeratlon, Montreal, Canada, 1-2 October 1987, NITEP, Envi-
ronment Canada
[9] Van der Sloot, H A, de Groot, G J, W1jkstra, J, and Leenders, P, "Leaching Charactenstlcs of
Incinerator Residues and Potenhal for Modification of Leaching," Proceedings of the International
Conference on Munlctpal Waste Combustion, Hollywood, FL, 11-14 April 1989, U S EPA/Envl-
ronment Canada, Vol 1, pp 2Bl-13
[10] Van der Sloot, H A, de Groot, G J, and Wijkstra, J, "Leaching Charactenstlcs of Construction
Matenals and Stabilization Products Containing Waste Materials," Environmental Aspects of Sta-
blhzatton and Solldtficatton of Hazardous and Radtoacttve Wastes, ASTM STP 1033, P L C6t6
and T M Gdham, Eds, Amencan Sooety for Testing and Materials, Phdadelphm, 1989, pp 125-
149
Joon-Hyung Kim, ~Hwan- Young Kim, 2 Hun-Hwee Park, 3 and
In-Suk Suh 4
Cementation of Borate Waste by Adding

Slaked Lime
REFERENCE" Kam, J - H , Klm, H -Y, Park, H -H, and Suh, I -S, "Cementation of Borate
Waste by Adding Slaked Lime," Stabzhzatlon and Sohdtficatlon of Hazardous, Radtoacttve, and
Mtxed Wastes, 2nd Volume, ASTMSTP 1123, T M Gdham and C C Wries, Eds, Amencan
Society for Testing and Matenals, Phlladelphm, 1992, pp 338-347
ABSTRACT" With regard to the cementatmn of the evaporator concentrate from pressunzed
water reactors, this study identifies the operating condmons glvmg properly coherent waste
forms of a hxgh volume reduction factor, hxgh compresswe strength, and good workaNhty w~th-
out the development of freestanding water Simulated waste specimens prepared by varying the
mixing ratms of sxmulated borate waste [bonc acid (HjBO3) 12% by weight, pH 9], portland
cement, and slaked lime were tested The results appear in a three-phase dmgram depxctmg the
proper blending necessary to yield final products possessing a compresswe strength of over 15
MPa and a volume reduction factor above 0 65
The soluNhty of sodmm borate is hxgher than that of HjBOj, and cementatmn of a concen-
trated borate solution (24 wt% HjBOj, pH 9) was therefore also tested The region of over 15
MPa of compresswe strength w~th a volume reductmn factor of over 0 6 based on the waste vol-
ume was identified By increasing the HjBO3 concentration m the waste from 12 to 24 wt%, the
actual volume was reduced by an addmonal factor of about two
KEY WORDS: boric acid, pressunzed water reactor, cementatmn, slaked hme, borate waste
With nine nuclear power plants m operation by the end o f 1989, Korea had accumulated
approximately 22 000 d r u m s of sohd radioactive waste, of which 40% were solldxfied hquld
wastes W h a t is essential, therefore, is to develop a m e a n s by which the v o l u m e of the cement-
solidified wastes might be reduced O f the nine nuclear power plants, eight plants are pressur-
ized water reactors (PWR) The m a r e c o m p o n e n t o f the llqmd concentrate is boric acid
(HjBOj)
T h e llqmd radwaste containing HjBO3 generated from the P W R s is generally concentrated
by evaporation and then sohdlfied In Korea the waste is concentrated to 12 wt% HjBO3 and
the concentrate is then sohdlfied with portland c e m e n t As the H3BO3 strongly lnhlblts the
hydration of cement, slaked h m e or s o d m m slhcate is added to e h m l n a t e this effect [1]
Because slaked h m e is used as the additive at most reactors, it was chosen for use in th~s study
Head, Radwaste Treatment Technology Department, Korea Atomic Energy Research Institute, P O
Box 7, Daeduk-Danjl, Taejon, 305-353, Korea
2 Pnnc~pal researcher, Radwaste Treatment Technology Department, Korea Atomxc Energy Research
Institute, P O Box 7, Daeduk-Danjl, Taejon, 305-353, Korea
3 Director, Radloactwe Waste Management Dwlsmn, Korea Atomic Energy Research Instatute, P O
Box 7, Daeduk-Danjl, Taejon, 305-353, Korea
4 Vice-president, RadloacUve Waste Management Projects, Korea Atomic Energy Research Institute,
P O Box 7, Daeduk-Danji, Taejon, 305-353, Korea
338

KIM ET AL ON CEMENTATION OF BORATE WASTE 339
Specifying appropriate operating condmons dunng cementation is needed to obtain high vol-
ume reduction factor and good compressive strength
Th~s study rams to ~dentffy the operating condmons necessary to obtain properly coherent
waste forms with a h~gh volume reducUon factor, high compressive strength, and good work-
ability without freestanding water, Specimens prepared by varying the mixing ratios of sxmu-
lated borate waste (H3BO312 wt%, pH 9), portland cement, and slaked hme were tested Work-
abxhty, generation of freestanding water, compressive strength, and water resistance were
observed, and a three-phase diagram displaying appropriate operating hmlts was drawn from
the results. Adequate operating condmons for the waste forms were suggested to be a com-
pressive strength of over 15 MPa and a volume reducUon factor of above 0 65
The pH of the simulated waste was adjusted to 9 by adding sodmm hydroxide (NaOH) As
the solublhty of sodmm borate ~s higher than H3BO3, cementation of a concentrated borate
solution (corresponding to 24 wt% H3BO3, pH 9) was also tested varying the mixing ratms of
portland cement, slaked hme, and the borate waste CrystalhzaUon temperature of the borate
solution (24 wt%, pH 9) was around 49 to 5 l'C, and there was no nsk of crystalhzaUon above
70~ A three-phase dmgram was also drawn for this case The region above 15 MPa o f c o m -
presswe strength, over 0 6 volume reduction factor, w~th proper workability, and without free-
standing water after cunng for two weeks was suggested as appropnate m~xmg proportions
The volume reducuon factors are calculated on the bas~s of waste volume used in the
mixtures The actual volume reducUon obtained by increasing the H3BO3 concentraUon m the
waste from 12 to 24 wt% was about two while still maintaining good properties of waste forms
The effect of the H3BO3 concentranon m the waste on the solidified form was also tested The
results can be utilized m the sohdlficatlon of P W R llqmd wastes
Research Methods
TheoreticalBackground
Cement pamcles m water undergo hydrolysis, the mare products being calcmm hydroxide
[Ca(OH)2] and a less basic calcium silicate hydrate (CSH). The reactions may be represented
by the following equaUons
2(3CAO S~O2) + 6H20 ---" 3CaO 2SIO2 3H20 + 3Ca(OH)2 (1)

2(2CAO SIO2) + 4H20--" 3CaO 2SaO2 3H20 + Ca(OH)2 (2)
3CaO A1203 + 31H20 + 3CASO4 ~ 3CaO A1203 3CASO4 31H20 (3)
4CaO A1203 Fe203 + 10H20 + 2Ca(OH)2 ---- 6CaO A1203 Fe203 12H20 (4)
Reaction 3 produces ettnngate, and the hardening time is controlled by its physical hin-
drance effect Reactions 1 and 2 produce CSH, which is the mare strengthening component
Also, calcmm hydrox]de is produced and precipitates the formation of crystalhne particles
Reaction 4 depletes calcmm hydroxide The hydration reacUon may be continued for a long
time [2]
Boric acid and borate wdl also react with the Ca(OH)2 to produce nearly water-insoluble
calcmm borate [Ca(BO2)2 2H20] Thin may also become a solld-hquld interface b a m e r pre-
venting cement particles from hydration Bonc acid consumes Ca(OH)2 and may hinder reac-
tion 4 By precipitation of Ca(BO2)2 2H20 as a slurry, the retardation by borate can be pre-
vented [3] The pretreatment reaction between calcmm oxlde(CaO)andH3BO3can bewntten
as follows
6H3BO3 § Feb
Copyright by ASTM Int'l (all rights reserved); Mon 2CaO --" 2CaO
24 23:07:03 EST 2020 3B203 nH20 + (9 -- n)H20 (5)
340 HAZARDOUSWASTES
For example, 2CaO 3BzO3 5HzO is a mineral referred to as colemamte whose solubility is
very low and specific gravity is 2 42, similar to that of cement [4]
Experiments
Preparation of Simulated Borate Waste--The measured H3BO3 was put into distilled water
and the pH of the solution was adjusted by adding a 32 wt% NaOH solution Then the weight
percent of H3BO3 m the simulated solution was adjusted by adding more distilled water The
solution temperature was controlled to 70~ in a water bath When the pH of the solution is
9, the mole ratio of NazO to boron oxide (BzO3) is about 0 41 The pH has not changed sig-
nificantly by regulating the water content This shows that the solution IS a kind of buffer solu-
tion, where the ratio of B(OH)3 over B(OH)4 does not change significantly by water content
The increasing solubility by the addmon of NaOH to H3BO3 is an advantage [4]
Workablhty Test--The quantity of water mixed with cement should be supplied as neces-
sary for hardening and also for easy mixing and transportation To determine the workability
limit of the cement paste, Korean standard KS L 5111 [comparable to ASTM Specification
for Flow Table for use in Tests of Hydraulic Cement (C 230)] was used The mixing ratio of
workability limit was determined when the diameter of extended paste was 21 cm The mixing
of cement paste was done according to the method and apparatus of Korean standard KS L
5109 [ASTM Method for Mechanical Mixing of Hydraulic Cement Pastes and Mortars of
Plastic Consistency (C 305)] To determine workability and freestanding limits, the addition
of Ca(OH)2 was increased m steps of 2 wt% from 0 to 40%, and waste content was increased
in steps of 3 wt% in the cement mixtures The homogeneity of the formulation was observed
after mixing The detailed test diagram is shown in Fig 1 [5,6] Ordinary portland cement
KSA 5201 Type 1 [ASTM Specification for Portland Cement (C 150), Type 1] was used The
purity of the Ca(OH)2 was over 99%
Testfor Freestanding Water--A polyethylene bottle having an inside diameter of 5 cm and
a height of 5 cm was used to test for freestanding water Generally less than 0 5% of freestand-
ing water based on the sohdlfied volume after 24 h is the accepted limit [7] Because of the
H3BO3, hardening takes longer and visual inspection after two weeks was therefore also done
[5,6]
Water Immersion Test--The safety of a disposal site is assured if the waste form maintains
ItS soundness even after immersion In ground or seawater and despite penetration through a
rusted container After hardening for two weeks a specimen was immersed In tap water for a
two-week penod Swelling, crumbhng, and cracks were looked for wsually and the compres-
sive strength was measured [5,6,8]
Compressive Strength Test--After mixing simulated waste, cement, and Ca(OH)z, the mix-
ture was put into a bottle of 5 cm in diameter and 12 cm in length and cured for four weeks
After removing the bottle, the two ends were ground parallel and the compressive strength was
measured by using a manually operated hydraulic press The debns was visually respected for
cracks and voids [5,6]
Volume Reduction Factor and Density--The ends of a specimen were ground parallel and
the weight was measured. Careful attention was prod to prevent water evaporation in the spec-
imen The size of the specimen was measured, and the volume reduction factor and density
were calculated. The volume reduction factor was calculated by dlwdlng the simulated waste
volume with the speomen volume
Leach Test--Generally, long half-life radionuchdes such as cesmm- 137, strontium-90, and
cobalt-60 are used to study the leach behavior of low-level radioactive waste forms Cobalt was
chosen for this study and cobalt-60 added as a radioactive tracer The concentration of non-
Waste (12 or 24 wt% H3BO3,pH9), Cement,SlakedLime ]
Nix!ng
i
Sealed Lurg
i [
] 1DayCuring ] ]2 Weeks Curing 4 Weeks Curing ]
I I
I
Free Standing
I Free Standing
Water Test
Co~ress~ve
StrengthTest
Water Test
I I
Immersion Test] Volum Reduction
Factor & Density
Passed imt 1 '

I
FIG l--Experimental methods and procedure
radioactive cobaltous chloride (CoC12) was 3 87 X 10 -7 to 7 74 X 10 -7 mol/cm 3 of specimen,

and the quantity of tracer in each specimen was adjusted to 46 to 93 MBq
The container used in the test was made of polyethylene, and the volume of the leachant
was 750 mL The specimen was cyhndrlcal with a 5-cm diameter and length The leachant was
distilled water with electric conductivity less than 0 15 m S / m During the leach test the tem-
perature of water was maintained to 25 _+ 0 5"C and water evaporation was suppressed The
leachant was changed after l, 6, 7, 30, and 120 days Whenever the leachant was changed, 5
mL of the leachant was collected and the activity was measured using a low-level alpha/beta
counter (Canbera Model 2210) [5].
The results were expressed by a plot of the incremental leaching rate, L (cm/day), as a func-
tion of time of leaching [ 9]
a. V
L=GxstG
where
L -- leaching rate, cm/day,

a, = ra&oactlvlty leached during the leachant renewal period, n,
Ao = radioactivity initially present in specimen,
V = volume of specimen, cm 3,
S = exposed surface area of specimen, cm 2, and

t, = duration ofleachant renewal period, day.

Workability and Freestanding Water
Figures 2 and 3 show the three-phase diagrams for simulated waste, cement, and Ca(OH)2
in 12 and 24 wt% H3BO3, respectively. The thick solid line shows the workability limit. To the
left side is represented a region capable of a homogeneous mixture of the three components.
The freestanding water limit was determined by visually inspecting water on the specimen
after one day or two weeks. The one-day freestanding water limit is shown as a one-dot slash
line and the two-week limit is shown as a two-dot slash line, at the right side of which is dis-
played a region free of water. A combination of three components should lie in the area
between the lines for workability and freestanding water limits.
Water Resistance
All the specimens whose mixing ratio was within the area mentioned above showed no
symptoms of swelling, crumbling, or cracking when inspected visually after a two-week
immersion in tap water.
OAIO0
WORKABILITY LIMIT
----- LIMIT OF FREE STANDING
WATER AFTER 1 DAY
. . . . LIMIT OF FREE STANDING
WATER AFTER 2 WEEKS
- - COrreCtIVE STRENGIII ( ~ a )
...... VOLUME REDUCTI~ FACTOR 2(
3~
~0
6o
3O
oe
5o
40 ~
f i _
7~ ,:, ' _
30
1 20
10
100 / / / / / /
4.57
0 10 20 30 40 50 60
SLAKED LIME (~%)
FIG. 2--Three-phase diagram of cement, slaked lime, and borate waste (12 wt%).
KIM ET AL. ON CEMENTATION OF BORATE WASTE 343
0^100
WORKABILITYLIMIT
- - - - - LIMITOFFREEST/~IDING
WATERAFTER 1 DAY
10/ --k90
.... LIMIT OF FREE ST~DING
WATERAFI~R 2 WEEKS
- - CO~RESSIVE STRENGTH( ~ a )
...... VOr,U~ REDUCTIONFACTOR
2C
3( ~ 7 0
4o 6O
so 5o~
30
6(
ZO
--30
10
80 --20
90 --10
100 ' "' 0

0 10 20 30 40 50 60
SLAKED LIME ( ~ % )
FIG. 3--Three-phase diagram of cement, slaked lime, and borate waste (24 wt%).
Compressive Strength
The compressive strength is shown in Figs. 2 and 3 in fine contour lines. These lines show
that the compressive strength increases as the amount of simulated waste or the Ca(OH)2
decreases and is heavily dependent on the content of the cement. In the case of 12 wt% H3BO3,
the m i n i m u m content of cement is 37 wt% and the m a x i m u m content of waste 42 wt% to
provide a compressive strength of over 15 MPa. For 24 wt% H3BO3, the m i n i m u m content of
cement is 32 wt% and the m a x i m u m content of waste 41 wt%.
Volume Reduction
The volume reduction factor is defined as the simulated waste volume divided by the
attained solidified volume and is an important factor in calculating waste management cost.
Figures 2 and 3 show that, at a fixed waste fraction, decreasing cement content has a tendency
of decreasing volume reduction factor, that is, the waste form becomes more porous by the
increasing amount of added Ca(OH)2 [2].
Optimal Operation Regtme

Conmdenng the freestanding water limit and compressive strength of over 15 MPa and a
volume reduction factor of over 0.65, the optimal operation region is chosen and dmplayed as
a shaded area in Fig 2 At the midpoint of the area, the weight ratio of the simulated
waste cement slaked hme is 39.55 6 The m i n i m u m requirement of Ca(OH)2 according to
reacnon 5, where the mole ratio of boron to calcmm is equal to 3, is shown as a solid line m
Fig 2 This gwes an idea of Ca(OH)2 addinon
In Fig 3 a similar tendency is shown for the case of 24 wt% H3BO3 The opnmal operation
region ~s chosen as the shaded area where the volume reducUon factor is over 0 6 and the com-
pressive strength is over 15 MPa. At the midpoint of the selected area, the weight ratio of sim-
ulated waste cement slaked hme is 38 54 8
The Effect ofpH

The effect of waste pH is shown in Fig 4 The rano of waste cement lime is fixed to 39.55 6
when H3BO3 content was 12 wt% In the case of 24 wt% H3BO3 the ratio is 38 54 8 Figure 4
shows that as the water-to-cement ratio decreases the compressive strength has a tendency to
increase, but when sodium content exceeds a certain limit the tendency is reversed Thus, the
optimal pH co n d m o n was chosen as between 7 and 9 For more concentrated waste, however,
a lower pH condmon by adding less N a O H is recommended
26
O
24
22
20
O
t
18
16 __ lw-tZOF H3B03 W /C /L
i
o 12 39/55/6
t3 24 38/54/8 o
14
[ I I [ I
5 7 9 11 13
ptl OF WASTE
FIGMon
Copyright by ASTM Int'l (all rights reserved); 4--Compresstve strength with regard to waste pH
Feb 24 23:07:03 EST 2020
KIM ET AL. ON CEMENTATION OF BORATE WASTE 345
19
18"-
17
16
15
I , . I , I , t , 1
8 12 16 20 24
BORIC ACID IN WASTE(~7~)

FIG, 5--Compressive strength with regard to H3BOj concentration.
The Effect of H3B03 Concentration

Figure 5 shows the compressive strength with varying H3BO3 concentration at a fixed
waste:cement:lime ratio of 39:55:6 and a fixed pH of 9. As the water-to-cement ratio
decreases, the compressive strength increases. The last point, however, shows the effect of high
sodium concentration [I0].
Leach Characteristics
Figure 6 shows the cumulative leach fraction for four kinds of products based on 12 wt%
H3BO 3 with a pH of 9. In the worst case, the cumulated leached fraction was less than 0.1% in
120 days. The average leaching rate was 9.0 • 10 -~3 to 2.9 • 10 -~2 m/s.
Conclusions
For use in the cementation of evaporator concentrate from PWRs, a series of tests has
resulted in a proposal for the proper blending ratios for concentrates with 12 and 24 wt%
H3BO3. The proposed optimal conditions are as follows.
1. The influence of mixing proportions on volume reduction factor, compressive strength,

workability, and presence of freestanding water for the simulated borate waste with 12
wt% of H3BO3 at a fixed pH of 9 are shown in Fig. 2. The mixtures giving adequate prop-
erties for the waste forms, a compressive strength of over 15 MPa and a volume reduction
factor of above 0.65, are shown as the shaded area. At the midpoint on the area, the
weight ratio of waste:cement:lime is 39:55:6.
2. The cementation of the concentrated borate (24 wt% H3BO3, pH 9) was also tested and
Copyright bythe
ASTM Int'l (all rights
proper regionreserved); Mon Feb
giving 24 23:07:03with
products EST 2020
a compressive strength of over 15 MPa and a vol-
346 HAZARDOUSWASTES
10.0
I}- C /W /L
9 78/22/ 0
/
' xlO~i
c3 54/36/10 r~
9 45/35/20 /~
O 26/48/26
~T 7.5 ~ /
M
5.0
2.5
LEACH TIME (DAY~

FIG. 6--Cumulative leachfraction ofr~
ume reduction factor of over 0.6 based on the waste volume was shown to be the shaded
area in Fig. 3. At the center of the area the ratio of waste:cement:lime is 38:54:8. The
additional volume reduction obtained by increasing the H3BO3 concentration in the
waste from 12 to 24 wt% means that the actual volume is reduced by a factor of about
two.
3. The compressive strength was not changed significantly in the pH range from 7 to 10 in
the case of 12 wt% H3BO3. For 12 and 24 wt% H3BO3 the optimal pH was chosen to be
between 7 and 9. Because of the sodium content in the more concentrated solution, a
lower pH is recommended to get higher compressive strength.
Acknowledgment
This work was supported by the Korea Radioactive Waste Management Fund.
References
[1] "Preliminary Safety Analysis Report, Kori 3 & 4 and Younggwang 1 & 2," Korea Electric Power
Corporation, Seoul, Korea, p. 11.4-3.
[2] Soroka, I., Portland Cement Paste and Concrete, Chemical Publishing Co., Inc., New York, 1980.
[3] Brownstein, M. and Levesque, R. G., "Experience with Cement Usage as the Binding Agent for Rad-
waste," 78-NE-15, ASME, New York, 1978.
[4] Kirk-Othmer, Encyclopedia of Chemical Technology, 3rd ed., Wiley & Sons, New York, 1982, Vol.
4, pp. 67-110.
[5] Park, H.-H., et al., "A Study on Cement Solidification of Borate Waste (I)," KAERI/RR-700/86,
Korea Atomic Energy Research Institute, Taejon, Korea, 1987.
[6] Kim, Joon-Hyung, et al., "A Study on Cement Solidification of Borate Waste (II)," KAERI/RR-
701/87,
Copyright by Korea
ASTM Int'l Atomic
(all rights reserved); Energy Research
Mon Feb 24 Institute,
23:07:03 EST 2020 Taejon, Korea, 1988.
[ 7] "Final Waste Classification and Waste Form Technical Position Paper," Branch Technical Position
on Waste Form, U S Nuclear Regulatory Commission, Washington, DC, 1983
[8] Colombo, P, et al, "Waste Form Development Program Annual Progress Report," BNL-51614,
Brookhaven National Laboratory, New York, 1982
[9] Hespe, E D , "Leach Testing of Immoblhzed Rad~oactlve Waste Sohds." Atomic Energy Rewew,
Vol 9, No 1, 1971,pp 195-207
10] Nagamo, S, "Counterplans for Radioactive Wastes," Denken Revtew, No 3, Central Research Insti-
tute of Electric Power Industry, Tokyo, Japan, 1982, p 15
D a v i d J. L e e ~ a n d A n d r e w Fenton ~
The Development of Cement Formulations for

Radioactive Waste Encapsulation
R E F E R E N C E : Lee, D J and Fenton, A, "The Development of Cement Formulations for
Radioactive Waste Encapsulation," Stabzhzatlon and Sohdlficatton of Hazardous, Radloacttve,
andMtxed Wastes, 2nd Volume, ASTMSTP 1123, T M Gdham and C C Wries, Eds, Amer-
ican Sooety for Testing and Matenals, Phdadelphm, 1992, pp 348-358
ABSTRACT: Cement formulations that have been developed m the Umted Kingdom (UK) for
treating rad~oactwe wastes are described Primarily, blends of ordinary portland cement wzth
ground granulated blast furnace slag or pulverized fuel as~ have been shown to possess adequate
processing performance and product propemes when used~o encapsulate a wide range ofradm-
active wastes The versatility of these formulations Is demonstrated by reference to the key pro-
cess and product propemes necessary to meet Identified cntena for disposal of radioactive
wastes In particular, the development of the strength, structure, and the sohd and llqmd phases
w~th rime is discussed An lllustratmn ~sgiven of the selectmn and development of a cementation
process for lmmoblllzanon of Wlnfrlth reactor waste sludges
KEY WORDS. cement, radmacnve waste, sohdlfiCatlon, sludges
Radioactive wastes are generated in the U K as a result of reactor operation, fuel reprocess-
Ing, and research and development They arise in a variety of physical forms that are generally
classified as liquids, sludges, and solids. Intermediate level wastes (ILWs), that is, wastes having
a specific activity exceeding 4 GBq/tonne alpha or 12 GBq/tonne f13' that are not significantly
heat generating, are currently stored at their site of origin under a vanety of storage conditions
To improve the safety of storage, it would be desirable to immobilize these wastes to produce
a solid waste form for subsequent storage and disposal.
The choice of the immobilization matrix depends on the process and product fulfilling cer-
tain key criteria As far as the process is concerned, this should be s~mple and cost effective and
should not lead to the production of secondary waste streams The product of this lmmoblh-
zatlon process should be a monolith with acceptable mechanical, chemical, and physical prop-
ertIes that are maintained over an appropriate time such that the release of radioactivity from
the waste form is minimized
A broad range of potential immobilization matrices has been investigated in the U K and
elsewhere [ 1]. These matrices have included bitumen, polymers, cements, and polymer-mod-
ified cements Of these, inorganic cements have been selected as the first choice matrix for
cementing most ILWs in the U K
Cementation is an attractive option from the processing point of view as cement remains
fluid for a sufficient time to allow placement, processing is safe, it is a low-temperature oper-
anon tolerant to water, and the set cement has adjustable properties Chemically, cement pro-
duces a reacting matrix with a porous microstructure Therefore, a knowledge of the chemical
1 Section leader and project manager, Cement Technology, respecUvely,AEA Technology, Dorchester,
Dorset DT28DH, Umted Kingdom
348

LEE AND FENTON ON CEMENT FORMULATIONS 349
TABLE l--Typical cement composluons
OPC BFS PFA
CaO 63 2 39 1 1 55
S102 20 1 37 2 46 7
A1203 5 48 10 6 25 3
MgO 2 56 97 0 95
F%O3 3 12 0 31 16 7
SOs 2 71 3 07 0 83
K20 0 51 0 40 3 46
NazO 0 32 0 45 0 69
TIO2 0 44 0 64 1 04
Mn~Os 0 06 0 75 0 04
C 0 03 0 02 3 66
S 13
Loss on lgnltlOn 08 + 1 24 4 62
environment and developing mlcrostructure is desirable to assess the compatibility of waste

streams with cement and to be able to predict the performance of the waste form d u n n g storage
and dmposal
Cement is therefore a very versatile material for waste encapsulation Cement can tolerate
a wide range of waste types by adaptation of the cement composmon to the different physical,
chemical, and radlochemlcal properties of the waste To do this requires a detailed knowledge
of the characteristics of the waste, which often requires a broad-based experimental sampling
and characterization program, to underwrite the selection of a cementation route for treating
the waste
This paper describes the special cement compositions that have been developed m the U K
for treating radioactive wastes. They are primarily blends of ordinary portland cement (OPC)
with ground granulated blast furnace slag (BFS) or pulverized fuel ash (PFA), chosen for their
beneficml effects on process and product properties Typical chemical compositions are shown
m Table 1 The paper also describes the development of cement properties that result from the
complex chemical hydraUon reactions that occur w~th time. An illustration is gwen of the
charactenzauon and treatment of sludge resulting from operation of the Wmfnth reactor
Choice o f Cement
Processing ConsMerattons
Cement is a low-cost material that ~s processed using s~mple plant operation at room tem-
perature. Simple low-temperature operations assist in mlmm~zlng maintenance requirements
and operator exposure to radmt~on in the encapsulation of radioactive wastes The raw mate-
rials have a long shelf hfe and are readily available to standard spec~ficatlons from a range of
suppliers There ~s considerable experience of the processing of cement m the construction
industry, and the product is well known to everybody. Cement is fire resistant and nontoxic
Cementataon is able to deal w~th a variety of physical forms of waste by the use of two basic
types of process These are (1) in-drum mixing for hqulds and sludges and (2) m-drum grout-
lng for solid wastes. The favored in-drum mixing process m the U K utlhzes a rumple design of
lost paddle m~xer whereby waste is metered into a disposal container to a fixed volume, fol-
lowed by the addmon of cement powder at a controlled rate On completion of m~xlng, the
paddle is decoupled from the mixer drive and remains m the drum This method avoids the
350 HAZARDOUSWASTES
need for decontamination of the mixing paddle, which is required for reusable paddle mixers
The process can be readily monitored by measurements of mixer torque and mixer speed
together with volume and weight measurements of the waste and cement
In-drum grouting ofsohd wastes is achieved by placing the waste into a drum and pumping
a cement/water paste into the container The process may be assisted by using a basket to con-
tain the waste and by using vibration to assist in the lnfilhng of fine voids Grout fluidity and
flow rate and level measurements are the important process parameters which require
monitonng
Product Performance
As well as possessing a range of desirable process properties, cement provides product prop-
ertles that make it an ideal material for radioactive waste encapsulation As a result of cement
hydration, a high internal pH is generated that provides conditions for the precipitation of
many radioactive species, particularly the long-lived actlnldes (for example, uranium and plu-
tonium) and also transition metals (for example, cobalt, mckel, and iron)
Furthermore, the high pH ensures that the majority of metallic containers considered for
immobilization (for example, carbon and stainless steel) are passlvated based on laboratory
measurements of corrosion potential at Harwell The high surface area of the hydration prod-
ucts provide sites for sorption of radionuclldes, further retarding release of radioactivity into
the environment
Further inhibition of radlonuclide release is provided by the low permeability of cement-
based systems, which restrict water movement and the movement of ions through the waste
form
The mechanical strength of cement is sufficient to withstand damage under normal han-
dhng and under extreme (accident) conditions
Cement products have been shown to be durable over extended tlmescales, for example, the
Mediterranean cements that have remained intact for many hundreds of years [2] Such prod-
ucts are also very tolerant to radiation In addition, cement has a history of good performance
as shielding in high radiation fields such as those generated in nuclear reactor pressure vessels
This self shielding offers a further benefit as a matrix for radioactive waste immobilization.
The product properties of cement based systems for waste encapsulauon are discussed m
more detail under cement performance in Table 1
Cement Performance
The chemistry of these hydrating cementitious systems is complex The processes are both
competitive and sequential so that apparently minor changes in the curing regime (for exam-
ple, temperature or dissolved salt concentrations) may result in step changes in the reaction
mechanisms and pathways What should be pointed out at this stage, however, IS that the
majority of these deviations represent no deleterious effect on the product performance
Cement systems have been studied for many years [3,4], particularly OPC, but more
recently blended cements and novel materials This great wealth of research has identified sim-
ple, readily accessible analytical techniques that can be routinely used to follow the progress
of the hydrating systems. The chemical processes that occur are known, and their effects on
the performance of the waste encapsulating matrices are also known, at the very least,
empirically
The progress of hydration may be determined in terms of (1) mechanical, (2) physical, and
(3) chemical properties with time, that is
TABLE 2--Characterlzatton of cement performance

Mechanical Physical Chemxcal
Rheology Gas and lIqmd permeabxhty Exotherm (hydraUon enthalpy)

Compressive strength Pore-size distribution Thermal analysis
Tensile strength Dimensional stablhty X-ray dlffracnon (XRD)
Modulus of elasnclty Pore water chemistry
Modulus of rigidity Radiation stability
Polsson's ratio
Ultrasonic pulse velocity
1 The development of strength with time

2 The development of structure with time
3. The development of sohd and liquid phases with time
Table 2 hsts properties which may be determined to characterize the development of these
aspects of cement performance These broad classifications are, however, heavily interacting,
for example, adverse chemical interactions may mamfest themselves m dimensional instabil-
ity, which may result m mechanical degradation
Mechamcal Propertws
Cement Rheology--The term cement rheology is used here to describe the fluid character-
ist~cs of cement/water or cement/waste pastes from the point of mixing through to setting
Inmally, when cement powder is mixed with the aqueous component of a hquid or sludge
waste, or with water, a paste is produced Too httle liquid or too much cement will produce a
paste that is impossible to mix. Conversely, a paste that possesses a high fluidity may suffer
from segregation leading to unimmoblhzed liquid, free water, or bleed
The instantaneous fluid properties have been characterized in the laboratory usmg viscom-
etry measurements and at process scale by monitonng mixing parameters such as mixer
torque and rotational speed This has enabled upper and lower hmlts to be identified for pro-
cessing liquid waste/cement formulations and cement grouts
Following the initial mixing of cement powder w~th liquid, cement hydration produces an
increase in viscosity with time. This structure development ~s responsible for the stiffening of
the paste and eventually, when the coagulanon-crystalhzatlon network becomes well devel-
oped, for setting The progress of setting has been momtored using penetranon resistance mea-
surements to ensure that specific formulations do not set too rapidly and cause problems with
the mixing operanon or too slowly such that plant throughput cannot be maintained
Mechamcal Strength--Both destructive and nondestructive tests can be used to determine
the mechanical properties of set cement pastes. The m i m m u m strength requirements for full-
scale products can be estabhshed by acceptable performance under handling and drop test
trials [5]
Nondestructive tesnng may be used to ascertain the strength development of cement pastes
indirectly However, such testing cannot alone provide absolute strength data but m associa-
non with destructive testing may be "cahbrated" to yield such data
Methods ofanalysxs include ultrasomc pulse velocity and resonance frequency testing these
techmques provide an accurate means of following the trend of strength development and can
detect any reduction m integrity of the product as well as any deviation from the normal path
of structural development An example is the onset of pozzolanic reacnon or structural deg-
352 HAZARDOUSWASTES
radatlon due to expansive waste matrix interactions such as with high sulfate-beanng wastes
Parameters determined include the dynamic modull of elasticity and rigidity (Ed, Gd), Pols-
son's ratio, and ultrasonic pulse transmission velocity
Destructive testing techniques include compressive and Brazlhan tensile sphtting strength
determination made on cube and cyhnder samples cast from process scale mixes or on speo-
mens cut or cored from full-scale waste packages
Phystcal Properttes
Gas and Ltquld Permeabthty--A product with low permeability to water is desirable to
restrict groundwater flow m a repository environment Permeation by water wdl result m the
leaching out of sodmm, potassium and calcmm hydroxide, and eventually calcmm sdlcate
hydrates These are the predominant speoes maintaining the high pH of the pore water With
the loss of these materials the actmide and transition metal speoes within the product may
become soluble and be transported from the matrix by the passage of the permeant However,
the permeability must be sufficiently high to allow any gases produced, by corrosion for exam-
ple, to dissipate without damaging overpressunzatlon of the matrix [6] L q m d permeabdlty
is taken as that closest to the real permeablhty as this provides the most pessimistic case Prac-
tical techniques to determine gas permeability on "natural" pastes at full scale are being
developed
Porosity and Pore-Stze Dlstrtbutlon--Total porosity by helium pycnometry and pore-size
distribution by mercury intrusion porosimetry are effective ways of determining the progress
of cement paste hydration and of producing a unique characterization of pore structure As
hydration proceeds, the deposition of reacuon products causes a reduction m the total porosity
and also a move toward smaller pore sizes Thus, the development of structure may be fol-
lowed closely by these techmques
Dtmenstonal Stabthty--As a result of their hydration, cements generally undergo dimen-
sional changes, particularly over the initial cunng period Hence, in order to facihtate a better
understanding of the long-term behavior of the cemented product and also to provide an input
to mathematical modeling codes, it is important to monitor any dimensional movements
associated with the hydrating matrix
In addition, adverse chemical interactions between the waste and the matrix may manifest
themselves m dimensional lnstabdity That is, deviation from normal hydration movements
resulting in excessive stress which may cause degradation of the waste form. Factors such as
buildup of corrosion products from encapsulated reactive metals (for example, aluminum),
swelhng of orgamc ion-exchange resins m alkaline cement systems, and development of
expansive phases within the hydrating paste (for example, high sulfate wastes) may all lead to
product degradation
Determination of dimensional changes are performed at laboratory scale using a measunng
frame and ultimately at process scale by the encapsulation of strain gauges within the matrices
Chemtcal Properties
Heat of Hydratzon--The hydration of portland cements may be regarded as a series of inter-
related consecutive exothermic chemical reactions These reactions are modified by the intro-
duction o f BFS or PFA The major components of portland cement are calcium silicate and
aluminate compounds In general terms, the aluminates react more rapidly than the silicate
compounds, but it IS the silicates which are primarily responsible for longer term structural
strength The aluminates contribute to early gel strength and are important in the stiffening
Copyright
and bysetting
ASTM Int'l (all rights reserved);
processes. CementMon Febhydration
24 23:07:03 ESTenthalpy
2020 is determined m the laboratory by the use
of isothermal conduction calorimetry and, at the process scale, by the inclusion of thermo-
couples within the hydrating product By careful control of the composmon of a given waste
form, generation of excessive temperatures (for example, > 100~ during the curing process
may be avoided This is desirable to avoid differential thermal stresses, activity release by steam
release, and generation of steam voids within the matrix
Phase Composition--An understanding of the chemical environment within the cement
paste is essential to predict the long-term performance of the waste package, as it will deter-
mine the corrosion processes that will operate and control the release of activity from the pack-
age In particular, a high pH, greater than 10 5, is desirable for an appropriate period of time
to minimize the solubility ofactlnides and transition metals within the matrix [ 7].
The pore solution chemistry [8] is controlled by its equilibrium with solid phases present in
the bulk matrix The pH, particularly, will be governed initially by equilibrium with Ionic spe-
oes, principally sodmm, potassium, and calcium hydroxides, but ultimately by the calcium
Sdlcate hydrate gel phases present in the hydrated matrix
An initial pH of greater than 12 5 will be achieved by sodmm and potassium hydroxides in
the pore solution released from sodium/potassium oxides in the cement powders This will be
maintained within a closed system Leaching of the waste form would eventually remove these
highly soluble alkalis, whereupon a pH of ~ 12.5 would be maintained by the less soluble cal-
cium hydroxide Eventually, the system would be buffered by calcium silicate hydrate phases
to a pH of ~ 10 5, still sufficiently high to minimize the solubility of any actmlde species pres-
ent. This extended buffering capacity at high pH is the key to the use of lnorgamc cements,
particularly OPC, BFS, and PFA, for the safe immobilization of radioactive waste
Pore solutions are extracted from cementltious products by a hydraulic pore water press
These solutions are then examined using a range of techniques including emission spectros-
copy, ion specific electrode, ion chromatography, potentlometrlC titration, and redox poten-
tial measurement
The solid phase composition of hydrating matrices is determined by the use of X-ray dif-
fraction (XRD) and thermal analysis As previously stated, the solid phase composition is par-
amount In determining the solution chemistry of the system but in addition provides an accu-
rate means of monitoring the progress of hydration and indeed determining any deviations
from the expected reaction pathway that may lead to loss of product performance
Examples of principle areas of interest are factors that may affect the pH buffering capacity
of the matrix (for example, calcmm/sihcon ratio in calcium silicate hydrate gel and degree of
crystalllnlty) and development of phases accompanied by volume changes that will generate
mechanical stress in the waste forms [8]
The high surface area of hydration products, the surface characteristics, and chemical incor-
poration ofspeofic radlonuclldes into hydrated phases reduces the mobdlty of fission products
likely to be soluble under alkaline conditions These characteristics combined with low per-
meabdIty are an effective bamer to activity migration under storage, accident, and disposal
conditions, for example, immersion and repository groundwater movements
Sorptlon may be determined experimentally by analyzing the partitioning of tracer radio-
nuchdes between the solid and aqueous phases, using gamma spectrometry, for example Ref
9. X-ray diffraction has been used to characterize hydrated phases that have incorporated spe-
cific radionuchdes
Radzatlon Stablhty--During their lifetimes, waste forms may be subjected to irradiation
from their own contents and by association with other waste packages Radlolytlc damage can
result from two basic mechanisms. Firstly, volume changes may result from lattice displace-
ment reactions These effects are seen primarily from heavy particle interactions, for example,
a and 0~nradiation The second and most significant effect with respect to ILW encapsulation
in cement
Copyright is radlolysls
by ASTM Int'l of the
(all rights reserved); aqueous
Mon Feb 24 23:07:03phases
EST 2020 and waste inclusions (for example, NO3) by all
forms of lomzmg radiation These processes may generate hydrogen and oxygen that may
cause disruptive overpressunzatlon of the matrices [10]
The propenslty of a matrix to succumb to radlolysls degradation can be considered a func-
tion of the permeabdlty to the product gas (for example, HE) and the dose rate That Is, the
generation of gas is broadly dependent on integrated dose; however, gas release from the
matrix is time dependent Therefore the dose rate ~s cnUcal
Radmtlon stability may be assessed by direct ~, lrradmtlon of specimens and by the incor-
poration of a emitters such as Pu-238 Subsequent analysis of evolved gases may be made and
the performance of the system momtored, with time, as a function of the performance criteria
detaded [ 11 ]
Summary of the Advantages of lnorgantc Cement

In order to fulfill its role as a radloachve waste immobihzing matrix, a cement paste and
waste form must perform well under any conditions considered possible dunng short- and
long-term storage scenanos and thereby minimize the release of activity to the environment
Principally, the advantages shown by inorganic cement waste forms are as follows.
1 G o o d mechamcal strength, allowing early movement under produchon process condi-

tions (~<24 h) and providing good response in transport accidents, for example, impact.
2. Inorganic cements are nonflammable and of low thermal conductivity, which ensures
good resistance to engulfing fires associated with hydrocarbon fuel fires in transport acci-
dents The pnnclpal mode of activity release in such an acodent would be vaporization
of water from the matnx transporting achvlty, causing structural degradation of the
matrix
3 There is good activity retention The high pH and the large buffenng capacity of the
matrices maintains a low actlmde solubihty
4 G o o d radmtion stability The great stability of cements and concretes to ionizing radia-
tion has been demonstrated by their successful use m biological shielding in nuclear reac-
tor containment for many years [12]
Summary
By understanding the hydration of blended cement pastes and the reqmrement of the waste
product, these techniques may be used to determine the suitability of the waste form in terms
of both ~ts process and product performance. In addmon, by relation to pure cement pastes,
any deviation from " n o r m a l " hydration, which may introduce unfavorable charactenstlcs,
may be highlighted and eradicated
This approach has allowed the development of cementation formulations that satisfy pres-
ent immobilization c n t e n a in the short term and, by modeling, the long term
Development of a Cementation Process for the Immobilization of Winfrith Reactor Sludges

Introduction
This section discusses the selection and development of a cement formulation for radioac-
tive sludges which arise from the operation of the Wmfnth reactor The sludges consist mainly
of powdered Ion exchange resins (65% by weight) with small quantmes of filter aids and decon-
TABLE 3--Approxtmate composltton of bulk zngredtents of SGHWR sludge
Relative Amount,
Matenal % by weight
Powdex--darnp cation and anion exchange resins (cation amon ratio 2 1) 65

Hydrated hme 20
Oxidized corrosion products (Fe. Cu, N1, Mn, V oxides) 5
"Stella" filter rods
Cellulose
Dmtamaceous earth
tamlnatlon reagents (Table 3) The methodology developed m earher sections is followed with
specific reference to the characterization of the waste, selection, and development of a cement
formulation, testing of product properties, and an assessment of the radiologlcal impact of the
cemented product
Sludge Characterlzatton and Stmulatton

Samples of homogemzed sludges from a storage tank were analyzed (Table 4) Analyses
showed that the pH of the supernatant liquor was approximately 11 7 due to the presence of
hydrated hme High levels of a m m o m a released from the decontamination reagent (Decon
4521) were also present Further examination of the analytical data revealed that the propor-
tion of Powdex to hydrated hme to filter aid were m good agreement with the inventory of
bulk materials added to the tank Copper and iron m the ash residue are from circuit corrosion
products deposited on the Powdex filter The mean specific actwlty of Cobalt-60 was 1 • 10"
Bq m 3 (2 6 CI m-3), and for cesium-137 it was 6 • 109 Bq m -3 (0 16 C1 m 3) The total bulk
beta and alpha levels were 3 5 • 10 ~~Bq m -3 (0 92 C1 m 3) and 6 • 107 Bq m -3 (1 5 X 10 ~
C1 m-3), respectively A nonradioactive simulant sludge has been developed for practical trials
based on the inventory of bulk materials added to the tank
TABLE 4--Properties of the Wmfrtth reactor sludge
Properties Homogemzed Active Sludge

Chemical propertles--supernatant
pH 11 7
Ca ++, ppm 60
Na +, ppm 150
NH~, ppm 2600
Chemical Properhes--ash (after combustion at 800~
A1, % 0 33
Ca, % 26 8
Cu,% 12
Fe,% 14
S1,% 53
Radlonuclldes--Sludge(30% by woght dry sohds)
6~ Bq m 3 1 • 10"
~37Cs,Bq m -3 6 • 109
Total beta, Bq m -3 3 5 • 10~~
Total alpha, Bq m -3 6 • 10 7
Cement Formulation Selectton, Development, and Product Testmg

Cement formulations for lmmoblhzmg these sludges have been assessed agamst a range of
process- and product-related cratena discussed above. The aim of the development program
was to maximize waste loadmgs for economic purposes and yet provide an ~mmoblllzaUon
route that saUsfied the process- and product-related criteria as discussed above
FormulaUons utlhzmg OPC alone were rejected on two counts. (1) the high water content
of the sludge would reqmre a sagmficant loadmg of OPC to prevent bleeding--the resultant
exotherm would be excessive, possibly resulting in product degradatmn and activity release,
and (2) OPC ~s relatively intolerant of orgamc ~on-exchange resins--their expansion within
the matrix after set results m mechamcal degradatmn above waste loadmgs of 10% by weight
[ 13,14], making this route economically unacceptable
Replacement of OPC with BFS was shown to systematically increase the resin loadmgs
achievable and enhance the performance of the waste form At 9 1 BFS/OPC the optimum
resin tolerance (67% by volume) and long age characteristics were obtained At these high slag
replacement levels, however, waste cement mteractmns resulted m hydration retardaUon (72
to 90 h) and free water formulation (bleed) Analysis of the sludge components highlighted
organic complexing agents and lnorgamc transmon metal species as potential retarding agents
Investigation of the components m lsolauon showed the lnorgamc transition metal compo-
nents to be the cause of the retardation
Set initiation was encouraged by pretreatment of the sludge waste with sodium slhcate (3%
by weight of the dried solids content of the sludge) A comparison of the properties of the
sodmm slhcate pretreatment and nonpretreated cemented sludge is gwen in Table 5
As can be seen, both formulations contain 67% by volume of sludge in the waste form The
dry sohds content of the sludge is 31% by weight The pretreatment produces a substantial
improvement m process properties (vascoslty, set, bleed) resulting in more efficient and con-
trollable mixing (viscosity), economic product throughput (set), and minimization of second-
ary waste streams (that is, no bleed)
TABLE 5--Compartson of the properttes of cemented pretreated and untreated sludge
Pretreated Formulatmn w~thout

Formulation Pretreatment
Cement blend 9 1 BFS/OPC 9 1 BFS/OPC

% additivea 3 0
Waste sohds content, wt% 31 31
Waste/cement ratio by weight 0 80 0 80
Water/cement ratio by weight 0 55 0 55
Setting time, h 20 96
Bleed at 24 h, vol% 0 5
Relatwe viscosity, Pasc 44 58
Compressive strength,d MPa at 28 days 134 165
Dimensional movements, e % at 56 days - 0 082 - 0 138
a Expressed as a percentage of the dry sohds content of the sludge

b Bleed was determined by removing and weighing the dralnable water from the sohd product at 24
h after mixing
c Relatwe wscos~ty was determined using a Contraves RM30 rotatmnal vxscometer fitted w~th a mix-
mg paddle
aUnconfined compresswe strength was measured to British Standard, BS1881 Part 116 1983,
Method for determlnaUon of compresswe strength of concrete cubes
e Dimensional measurements were determined to mo&fied ASTM C 157 Test Method for Length
Copyright
Change by ASTM Int'l (all rights reserved);
of Hardened CementMon Feb 24 23:07:03
Mortar EST 2020
and Concrete
There is a small reduction in strength, but this is not regarded as sigmficant m terms of the
overall waste package program as it is sutficient to allow normal handling and will impart ade-
quate resistance to impact incidents, as evaluated in terms of standard tests [5].
Product Performance
Product properties have been evaluated at laboratory and full (500-L drum) scale. The scope
of the testing program is shown in Table 5 Samples obtained by sectlomng and trepanning
from a full-scale mix showed that the lost paddle m-drum mixing system produced a homo-
geneous product of uniform appearance with no signs of unmixed powder and macrovoidage.
Strength tests showed an average core compressive strength of 22 5 MPa after 90-days curing
and a cylinder strength of 21.0 MPa at the same age, which greatly exceeds adequate strength
for handling All specimens showed shrinkage consistent with typical cement hydration
shnnkage under sealed storage conditions, and there was no ewdence for resin/cement swell-
ing Mold-cast specimens of cemented sludge were subjected to a gamma irradiation dose of
1760 kGy (17 Mrad), which is a factor of 17 greater than the expected hfetime dose due to self-
Irradmtlon. Measurements of dimensional movements, gas generation, and structural lnteg-
n t y showed that the waste form was sound and dimensionally stable, and that hydrogen was
evolved with a typical G value of 0 15 molecules of H2 per 100 eV [10]
Radtologwal Impact
Radiologlcal impact has been assessed by leach testing and by a simple environmental
~mpact assessment The pnnclpal radionuchdes present m the waste form are the gamma emit-
ters cobalt-60 and cesium-137 The level of alpha emitters is very low indeed
Dynamic leach rates were measured using a leachant replacement method [15] based on a
modified International Organization for Standardization (ISO) procedure [16] An equihb-
rlum leach test measured the buildup of activity in solution Results for cobalt-60 showed that
dynamic leach rates fell from l0 -4 cm day -1 to between l0 -6 and l0 -8 cm day -j after one to
two years The e q m h b n u m amount of cobalt-60 was 0 02 to 0 03% after 200 to 300-days leach-
rag. Dynamic leach rate value for cesium-137 were 10 2 cm day-1 initially, falling to 10 -4 to
10-Scm day -j after one to two years The equihbrlum amount ofceslum-1371n solution peaks
at between 50 and 65% after 50 days, after which there is a gradual decrease up to 200 to 300
days
The disposal assessment showed that drums fabricated from ~ 2 5-mm-thlck stainless steel
would contain the waste form for at least 100 years By this time, the waste will have been
placed in a disposal site, cobalt-60 actlwty will have decayed, and cesium-137 activity will
dominate ( ~ l09 Bq tonne -~) An environmental impact assessment has shown that a further
period of decay of 400 years would reduce the specific acUvlty to such a level that someone
drinking 730 L o f water per year (2 L per day) in intimate contact with the waste form in which
all the cesium-137 had been released into the aqueous phase would recewe an annual dose of
1 mSv
Conclusions
A logical framework has been developed for selecting cement matrices and the evaluation
of cement waste forms based on criteria for both process and product performance This eval-
uation process requires a detailed knowledge of the waste characteristics so that suitable
cement matrices can be identified for testing Product performance testing at laboratory and
largebyscale
Copyright has(allevaluated
ASTM Int'l rights reserved);waste
Mon Febfrom properties
24 23:07:03 EST 2020 against criteria for storage, transport, and dis-
posal under normal and abnormal conditions Failure to meet essential cntena requires a
reevaluation of matrix selectmn and product performance
By careful selection of the cement matrix and the waste-to-cement ratio and by modifying
or pretreatlng the waste as appropnate, it is possible to produce waste forms with acceptable
properties from a wide range of waste streams This demonstrates the versatility of cement-
based lmmobdlzatlon techmques
The logic which has been developed has been illustrated by reference to a specific waste
s t r e a m - - W l n f n t h reactor sludge Production of an acceptable waste form from this waste,
which contains organic ion exchange resins, decontammaUon reagents, and heavy metals, is
a further demonstration of the versatlhty of cement for waste immobilization The properties
of the waste form have been shown to provide good protection of the environment from radm-
activity release
References
[1] Falrhall, G A, "ObJectives and Scope of the Joint Funded BNFL/DOE Product Evaluation Devel-
opment Programme," DOE/RW/90/050, UKDOE, March 1989
[2] Malhnson, L G , "Concrete as a Barrier for Nuclear Waste Stu&es of Ancient and Old Concretes
and Thexr ImphcaUons for Long-Term Durability," Concrete for Hazard Protection, Edinburgh,
Scotland, 27-30 Sept 1987, The Concrete Society, Wexhane, Slough, Btrks, SL36PJ, U K
[3] Lea, F M, The Chemistry of Cement and Concrete, Edward Arnold Ltd, London, U K, 1970
[4] Neville, A M, Properties of Concrete, Pitman Pubhshmg Limited, London, U K, 1981
[5] "Regulations for the Safe Transport of Radioactive Material," IAEA Safety Series 6, International
Atomic Energy Agency, Vienna, Austria, 1985
[6] Lever, D A and Rees, J H, "Gas Generation and Migration In Waste Repositories," Proceedings
of an NEA Workshop Near-FteldAssessment of Repositories for Low and Medium Level Radtoactlve
Wastes, Pans, France, 1988, NEA, pp 115-127
[7] Ewart, F T , Howse, R M, Thomason, S J, Wllhams, S J, and Cross, J E, "The Solubility of
Actlmdes in the Near-Field," Materials Research SoctetySymposturn Proceedmgs, Vol 50, 1985
[8] Atkmson, A and Hearne, J A, "An Assessment of the Long-Term Durability of Concrete in Radio-
active Waste Repositories," Technical Report No AERE-R 11465, United Kingdom Atomic
Energy Authority, Harwell, U K, October 1984
[9] Holland, T R and Lee, D J, "Radlonuchde Getters m the Near-Field Chemistry of Reposltones,"
Annual Report-March 1990, Technical Report No AEEW-R 2574, AEA Technology, Wlnfnth,
U K, 1990
[10] Holland, T R, "The Effect of Gamma Irradiation on the Stability of Cemented Wmfnth Reactor
(SGHWR) Sludge," Technical Report No AEEW-R 1924, United Kingdom Atomic Energy
Authority, Wmfrlth, U K, January, 1985
[11] Wlldmg, C R, Phdhps, D C, and Lyon, C E, "The Effects of Radiation on lntermedmte Level
Waste Forms," Proceedmgs of the Twelfth International Symposium on the Sctent~c Basts for
Nuclear Waste Management, West Berlin, Germany, 1988, Matenals Research Society, U S
[12] Kaplan, M F , ConcreteRadtatlonStabthty, Longman, U K, 1989
[13] Bonnene-Svedsen, M, Tallberg, K, Ahtolla, P, and Tollback, H, "Studies on the Incorporatmn of
Spent Ion Exchange Resins from Nuclear Power Stations Into B~tumenCement," IAEA-SM-207/
78, InternationalAtomic Energy Agency, Vienna, Austria, 1976
[14] Arnold, G , et al, "Solidification oflon Exchange Resin Wastes," BNL 51615 UCTO, Brookhaven
National Laboratory, Upton, NY
[15] Holland, T R, "The Leachablhty of Cemented Wlnfnth Reactor Sludge," Techmcal Report No
AEEW-R 1949, United Kingdom Atomic Energy Authority, Wmfnth, U K, August 1985
[16] Smltton, C, et al, "Proposed Standard Leach Test for Low and Medmm Level Sohdlfied Waste,"
ORNL Conference on the Leachabthty of Radioactive Sohds, Oak Ridge National Lab, Oak Ridge,
TN, 9-12 December 1980
Test Method Development
Jerry Isenburg' and Mwhael Moore'
Generalized Acid Neutralization Capacity Test

REFERENCE: Isenburg, J and Moore, M, "Generalized Acid Neutralization Capacity Test,"
Stablhzat~on and Sohdlficatlon of Hazardous, Radloacttve, and Mzxed Wastes, 2nd Volume,
ASTMSTP 1123, T M Gdham and C C Wales, Eds, Amencan Sooety for Testmgand Mate-
nals, Phdadelphm, 1992, pp 361-377
ABSTRACF: A method is developed to measure the neutrahzlng capaoty of various generic

banders The method, termed the generahzed acad neutrahzatmn capacaty (GANC) test, as based
on modaficatlons to Envaronment Canada Method No 7, Acad Neutrahzatlon Capacity, com-
bined wath U S Envaronmental Protectaon Agency (EPA) Method No 1311, Toxacaty Charac-
tenstlc Leaching Procedure The GANC test as a single batch procedure that utdazes leachants
normally ranging an acadtty from 0 to 6 eqmvalents of acid per kilogram ofsohds After 48 h of
contact, the pH of the leachate as measured The leachate can also be analyzed for alkahmty,
hardness, and metals concentrataon Plotting pH values versus eqmvalents per kilogram ofsohds
defines the alkaline behavmr of a sample The more equivalents a bander neutrahzes, the greater
ats buffenng capacaty and effectavenessin damlmshlng metals moblhty, but too much alkahmty
can also be harmful Examples are gwen of the apphcatmn of GANC test results and concepts to
metal-spaked samples and to lead-contaminated soils from a Superfund sate
KEY WORDS" waste, stabdazatmn, portland cement, hme, lead, copper, fly ash, kiln dust, dolo-
mate, leaching
In 1986 Environment Canada published Method No 7, Acid Neutralization Capacity

(ANC) test to measure the capacity of different binders and wastes to resist pH reduction by
exposure to acid leach testing [I]. It was discovered in work done by the University of Cincin-
nati at the U S EPA laboratory at Center Hill on solidification/stabilization (S/S) o f toxic
wastes from a Superfund site (1987-1988) that a series of increasing acidity leach tests gave
useful tnformaUon on the effectiveness of a binder in stabilizing heavy metals. Other research-
ers subsequently reported the use of acid neutralization capaoty of stabilized wastes to predict
leach results [2,3]
A generalized acid neutralization capaoty test (GANC) was developed at Center Hill to
1 Identify the chemistry ofneutrahzatlon, speclatlon, and mobility dunng leaching of toxic
metals from a stabilized waste
2. Characterize the net alkalinity o f candidate binders
3 Engineer a trial stabilization mix
This report presents four stages of the use o f G A N C information The first section describes
the test itself and the results of applying it to standard binders The second section demon-
strates how the reactivity or lack thereof in solids blended with binders changes the G A N C
results. The third section demonstrates how the G A N C procedure can be used to study the
Umvers~ty of Cmcmnata, Department of Cavil and Envaronmental Engmeenng, Cmcmnah, OH

45221 (Machael Moore as an envaronmental engineer wath Newport Steel Corp, Newport, Kentucky
41072
361

condmons needed to stabilize a metal, namely copper The fourth section presents very pre-
liminary results from attempts to use the GANC results quantitatively to design a stabilization
mix for lead on a Superfund site
The G A N C Test and Results on Binders

Introduction
The Canadmn ANC test gwes eleven test values for each sample The ANC is not designed
to parallel the toxicity characteristic leaching procedure (TCLP) [4] in particle size of solids,
leachant acid type used, or hquld-to-sohds ratio Since the statlsncally accurate prediction of
the binder performance and stablhzaUon behavior of mixtures as determined by the TCLP is
of primary concern to the researchers at Center Hill, a new test has been developed The
GANC test parameters are specifically chosen to follow the TCLP test (now the regulatory
standard leach test m the United States), and more duplication is used for statistical accuracy
Procedure
A relatively simple and flexible procedure was developed to study the effects of leaching on
a pure binder, raw waste, or stabilized waste The detailed procedure is presented In Appendix
A and summarized here m four major steps as follows
1. One gram of dry sohds passing a 40-mesh sieve is placed m a series of 125-mL polyeth-
ylene bottles (usually five duplicates are used per sample)
2 Twenty mllhhtres of liquid is added to each bottle A declining amount of ASTM Type
I water is added to each bottle followed by an increasing amount of 2 N acetic acid
(C2H4Oz) This process produces a series of bottles containing increasing equivalents of
acid per kilogram of sohds but the same total hqmd volume The usual series for a sta-
blhzed waste is 0 to 6 equivalents of acid per kilogram of solids (eq/kg) in increments of
0 5 eq/kg
3 Tumble liquid m the standard TCLP rotating extractor for 48 h and then let stand for
15-mm
4 Measure chemical properties of Interest on the supernate (or filtrate ~f setthng is incom-
plete) such as pH, metals concentraUon, or hardness (calcmm content)
Results
Chemtstry Revealed by Patterns--Figure I shows a typical GANC result for a pure binder
containing only one mineral species calcium hydroxlde [Ca(OH)z] The plateau at a pH value
of 12 to 12 5 represents the buffer tendency of the mineral The gradual decline of the graph
above pH 12 is due to the c o m m o n ion effect of dissolved calcium on dissolving Ca(OH)z. The
tailing off below pH 6 of the sharply declining curve from pH 12 is due to the buffering of
calcium acetate combined with the weak acid behavior of C2H402 solutions
Figure 1 also includes a curve for a hydrated lime sample made from dolomltzc (high mag-
nesium carbonate) limestone The gradual dechne of the curve is the result of the common ion
effect, and the second plateau is due to the magnesium oxide from the source rock of the lime,
that is, dolommc limestone (see Appendix B for the chemistry of dolomite and hme
production)
Figures 2 through 4 show the differences obtained when mineral acid (hydrochloric) is used
instead of acetic on three pure binders Center Hill researchers desire to simulate the TCLP
ISENBURG AND M O O R E ON ACID NEUTRALIZATION 363
Q)
2.0! ) ~ o - o @ o o ~ @ ---o-= @ 5 ~ - 6 o ~
\
o _o _o
<>
o o l 1.0 i
(.-
0.0
\
O 0-0-000, >0-0-0...
L d ~ 9.0 0
I-- ,--
ol
<
-I-8 .,_ 8.0 i O--0 Analytical Grade Calcium Hydroxide
(D o
LI
7.0 i O--0 Type S Hydrated Dolomitic Lime
_~o
OO
6.0 i
\0. O-O.
5.0 i 0-.. O-<
LJ~O~O_o_ (
.<
4.0 9 J . i . i , i . i , I , i , i . i . i . r . i . i . i 9 i , i , i , i , i .
0 2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40
Equivolents of Acid per Kilogrom of Dry Somple

FIG. 1--GANCfor analytical grade Ca(OH)2 and hydrated lime from dolomitic limestone.
Thus, C2H402 has been selected for the GANC test. The GANC patterns are similar, with the
mineral acid giving a slightly lower pH. Below a pH of 6.0, the same tailing offas in Fig. 1 is
observed for all samples tested with C2H402.
Typical Binders (Pure)--Figure 5 shows typical results for lime materials. The quicklime
has the greatest acid neutralization capacity tested (40 eq/kg to pH -- 7). The figure also shows
that hydrating the quicklime reduces the capacity on a weight basis (but not on a molar basis).
12.0
~- 11.0
10.0
9.0 v'-e=e=e~e&e=e-$-.z~
T
8.0
\X
I,I 7.0
6.0
< o
I 'S 5.o
0 ~ 4.0 ~----A Acetic Acid
<
Ld 9"r- 3.0
_ J 0o v v Hydrochloric Acid
~- 2.0
v~
~' 1.o
o.o
2 4 6 8 10 12 14 16 18 20 22 24 26 28 30 32 34 36 38 40
Equivolents of Acid per K i l o g r o r n of Dry S o r n p l e

FIG. 2--GANC for dolomitic hydrated lime with different acids.
364 HAZARDOUSWASTES
12.0
_e
n 11.0
"<-3,
-r'.~"
l.d
I---
<
2,-
o
o
10.0
9.0
7.0
), \
\
T o 6.0
~ A~ ~A~ ~
O "6
< ~ 5.o A A
._1 4.0 ~---~ A c e t i c Acid V

0o 3.0
2.0-
1.0-
v v H y d r o c h l o r i c Acid \
v ~ v ~ v ~
0.0 : l : l : l : l ; I I I I I I I ' I a I ' : : l

0 -2 4- 6 8 10 12 14. 16 18 20 22 24.
Equivalents of Acid per Kilogram of Dry Sample

FIG. 3--GANC for portland cement kiln dust with different acids.
The results for a sample of limekiln dust are also shown in Fig. 5. The single plateau pattern
demonstrates that the lime production plant was running nearly pure calcium limestone when
this sample was taken. The acid neutralization capacity was low, which also indicates that
much inert material was present. Such patterns demonstrate the need for testing the specific
binders intended for use in remediation of a waste site.
12.o!
o~ lO.O-
i
z ~ 9.0-
c~
Ld "5
<
~ 7.0-"
c
8.o-
%
z 8 6.0'
o ~ 5.oi n - - - - A Acetic Acid v\xzx~_n~_n_zx_n_A -
<
1J_1 ~ 4.0 v v Hydrochloric Acid \
O
T 3.0
cO v~ v
~- 2.0
1.o
< o.o
0 2 4. 6 8 10 12 14 16 18 20 22 24 26
Equivalents of Acid per Kilogram of Dry Waste
FIG.
Copyright by ASTM Int'l (all rights 4--GANC
reserved); Mon Febfor portland
24 23:07:03 cement Brand 1 with different acids.
EST 2020
ISENBURG AND MOORE ON ACID NEUTRALIZATION 365
9-~ 12.0
E
o 11.0
(/3
I
..E
9~- 10.0 XQX
I %_.~___~a-_.._. a a n/x
O_
o 9.0
l_J 2
t"-
<<
I
C) ~ 7.0
<
bJ ~ 6.0
__] T
oo 5.0
\
0~9 r-1x
O--0 Lime Kiln Dust (Source 1) ~J~O-- 0 r-I~F]~FI-[-I~

A - A Dolomitic Quicklime (Source 2)
~ 4.0
.< []--[] Hydrated Dolomitic Lime (Source 2)
3.0 ' I ' I , I ' I , I , I ' I ' I , I '
0 4 8 12 I6 20 24 28 32 36 40

FIG. 5--GANC for dolomitic quicklime, Type S dolomitic hydrated lime, and limekiln dust.
Figures 6 and 7 show data for portland cement and its waste product, kiln dust. The cement
reacts as the G A N C test is run and produces Ca(OH)2 as a byproduct of its setting/hydration
reactions. The hydration reaction is retarded by adding sugar so that less of a plateau for
Ca(OH)2 is recorded, but not all the plateau is removed by the presence of the sugar. Producing
a sample o f hydrated cement by reaction with excess water in a bottle followed by filtration
removes much of the soluble Ca(OH)2 in the filtrate and gives a generally lower pH than the
-~
O. 12.0:Ik-"~-$ .--'~-~.x = - = X~ v
E
o 11.0.
I 9~ lO.O
W
F-
<s
I
(D
2
o
0
t-
O
9.0
8.0
\ "%
<
Ld :3
0
7.0 Uohdroed \
-r
I:0
6.0 o o Bottle Hydrated ~176 ~-
5.0 z~--z~ Unhydrated Plus Sugar ~)~o-o-o-o_!
4.0 : : : I : I : 1 : I : : : : : l : : : : : I : : :
0 2 4 6 8 10 12 14 16 18 20 22 24 26

Copyright by ASTM Int'l (all rights reserved);
FIG.Mon6 -Feb
- G24AN 23:07:03
C forEST 2020
portland cement Brand 2 bottle.
_e
o_
12.0
E
11.0 \V\\ A
-T-
Q_ "~ 10.0
\vXA
"6
Ld 9.0
F- E
<s O
-t- o 8.0
o \\\\
<s 7.0
I,I o
"1-
6.0
V - - - V Hydrated Cement Kiln D u s t V'~"''v~A~'~-~-V
~, s.o
<
4.0 , , , , , ,
0 2 4 6 8 10 12 14 16 18 20

FIG. 7--GANC for portland cement kiln dust before and after hydration.
unhydrated cement in the GANC test. Likewise, the portland cement kiln dust has a lower
acid neutralization capacity than the cement, and its neutralization capacity is further reduced
by hydration.
Conclusion and Recommendation--The GANC test has proved successful at quantifying
the alkalinity of binders and characterizing their buffering chemistry. The next step in unrav-
eling the behavior with wastes is to test how solids, both reactive and inert, affect GANC
results.
GANC Tests of Binder Blends with Solids

Inert Solid with Ca(OH)2Mixtures and Procedures
Most ingredients in a soil-type waste have little reaction with a binder. The GANC pattern
for dilution of the binder with various proportions of an inert solid is demonstrated in Fig. 8.
The GANC test was run on blends of Ca(OH)2 and silica sand at various typical waste/binder
ratios. The patterns are scaled down on the eq/kg axis to almost exact images of one another.
This shows that a truly inert solid simply dilutes the alkalinity proportionately.
Figure 9 shows the distortions of proportionality that occur when the binder dolomitic lime
reacts with the pozzolanic material Type F fly ash. The water bound into the solid calcium
silicate hydrate formed by the pozzolanic reaction adds to the total weight without adding alka-
linity. This alkalinity dilution effect is also seen in the steam-hydrated lime compared to quick-
lime (Fig. 5) or the bottle-hydrated cement compared to unhydrated cement (Fig. 7).
Conclusion and Recommendation

Simple dilution proportionality can be applied to inert solids blended with an alkaline
binder, but corrections must be applied if the solids react with the binder.
13
~_ 12
E
u
(/3
11
Ca(OH) 2
..E 10
9 9
~2 zx zx 5 0 / 5 0
I-- E 8
<
T 8 7
< 6
9 9 20/80
-~, 5
o o lO/9o
4
L 0 0 0/100
2 I l I I I l I I I
0 3 6 9 12 15 18 21 24 27 30

FIG. 8--GANC for blends qf silica sand and calcium hydroxide.
12.5 _ _
Q) 0
E 11.5
O ~ a ~ k N Hydrated Dolomitic Lime Type "N"
~n 10.5 \ \ N urn~ / ny ~,h \
I "i
~ .,., g.5
U
I - - ~o 8.5
-r-
r._) "6 7.5
LLJ ~ 6.5
_1 o
-1-
oo 5.5
4.5
.<
ii~ I ' { ' : ~ " " ~ I '' -:I " I I , I g I , I , I , I " I "
3.5
2 4 6 8 10 12 14. 16 18 20 22 24 26 28 30 32 34

FIG. 9--GANC for hydrated blends of Type F fly ash and dolomitic hydrated lime.
The next step was to try to apply the GANC test results, at that point incomplete, to binder-
waste systems that contained heavy metals.
Synthetic Soil Spiked with Copper

Mixtures and Procedures
A solidified synthetic waste containing copper was prepared by blending 95% Ottawa silica
sand and 5% Wyoming bentonite. To this was added a sludge containing copper carbonate,
lime, and water to give 1% copper by weight of the synthetic waste solids. Portland cement was
added to the synthetic waste at two ratios: 1:20 and 1:10, cement-to-waste solids. Enough
water beyond that in the sludge to make the mass castable was added during mixing. The cop-
per was made into a sludge first so that its retarding effect on cement set was minimized. The
mortars were cured 28 days sealed in a 100% relative humidity chamber. They were then tested
for unconfined compressive strength and achieved 0.76 MPa ( 110 psi) for 5% cement and 1.01
MPa (146 psi) for 10% cement mixes. The fragments from compression testing were saved in
plastic bags for GANC tests.
Results
When these solidified samples were run on GANC and copper was determined in the leach-
ate by inductively coupled atomic plasma spectrophotometry (ICAP, EPA Method 6010),
Figs. 10 and 11 resulted. The general pattern seen in these figures is predicted by Shively,
Bishop, Gress, and Brown [5]. The pattern interpretation is that heavy metals are immobile
until enough acid has contacted the sample to neutralize most of the alkaline constituents in
the binder. Acid exposure reduces the pH of the leachate to a value, varying with the specific
metal being considered, at which the metal begins to be mobile.
13.0 1000 E
12.0 - _A/A A.-......~ Q..
0O
9.~ 1 "1 0 ~ - ~ A ~ .100 ~-"
oc
1o.o- 9 :~~
C) - t
9.0.
o u
__1 -10 9
,,,_"5 8.0. o
o.~ 7.0- o
-
6.0 - -1.0
L. 5.0 0
I ~ 4.0 9 LO.1 (.9
99 3.0I A A ]
~,
2.0 ' I ' I , I J I , I , I , I , 0.01 <]
0.0 0.3 0.5 0.8 1.0 1.3 1.5 1.8 2.0
Equivalents of Acid per Kilogram of Treated Waste

FIG. 10--GANCfor copper-spiked simulated waste treated with 5% portland cement including copper
concentration
Copyright by ASTM Int'lleached.
(all rights reserved); Mon Feb 24 23:07:03 EST 2020
13.0 lOOO E
120~ ! ~ A AfA--A~
A ~ A -100 o
..= lO.O-
o ~ ' e~ - ,A. o
~
o~ 90, c
_./ o -10 9
,,,.- "6 8.0. o
c
o ~ 7.0-" 9 o
:3::g
13-'- 60- -10 ~
~- 5 0 ~ 9 o
' O
-z- , 4 0 ZX~A \ -01
9 30
2.0
A
: I ' I : ', : I t I ~ I : I : I : ', : ', J t ~ I ) I J -0.01
I
O0 02 04 0.6 0.8 10 12 14 16 18 20 22242628
Equivalents of Acid per KJIogram of Treated Waste

FIG 1l--GANC for copper-splked stmulated waste treated wtth lO~ portland cernent lncludmg copper
concentration leached
TheresultsmFlgs 10and 11 speclfically show that the 5% cement added needs 1 eqmvalent
of acid per kilogram of stablhzed sample to reduce the pH to 6, and 10% cement added needs
2 equivalents of acid per kilogram This exactly corresponds to the proportionally reduced
portland cement GANC curve in Fig 6 where reduction to pH 8 requires 20 equivalents of
acid The copper release to solution at about pH = 6 or 7 is also demonstrated
Conclusmn and Recommendation

A synthenc soil waste composed ofmlica sand and bentonite spiked with preneutrahzed cop-
per sludge acts as an inert solid with respect to prediction of the amount of portland cement
binder required to remst 2 eqmvalents of acid leachant (exactly the amount in the TCLP test)
The next step is to try these concepts in the demgn of trial mtxes for stablhzation of a toxic
waste field sample
Engineered L e a c h T e s t Performance: Lead

Introduction
Predicting leach test performance is thus a two-part problem The first part is to predict the
amount of alkahnlty needed in a stabilized waste sample to overcome the acid added in a par-
ticular leach test Predictive tests should have the same physical form and the same leachant
acid as the leach test under consideration For the TCLP leach test this is 20 to 1 liquid/solid,
C2H402 leachant, 2 eq aod/kg waste (for most binders), a crushed sample, 18-h contact, and
the leachate filtered through 5-~m pores after completion of leaching
The second part of the prediction problem is the response of the particular heavy metal to
the pH changes in the leach test. This includes the chemical speoes formed by the S/S process
Copyright
in the by ASTM Int'l (alland
sample rightsthe
reserved);
m a Mon
x i mFeb
um 24 23:07:03
amount EST 2020
(usually much less than 100%) potentially released
by leaching. It also includes the apparent diffusivity of that metal through the binder residue
remaining after leaching. If leaching time is enough to allow equilibrium conditions to be
obtained, then the diffusivity effect is obliterated.
Mixtures and Procedures

A soil from a Superfund site containing 4 to 11% lead was evaluated at Center Hill to deter-
mine the feasibility of using S/S to treat the waste (results to be reported completely elsewhere).
After preliminary tests indicated that 30% portland cement would prove satisfactory, a 30%
mix and another "engineered" mix were tested further.
The "engineered" mix recipe and procedure was chosen to at least match the cement mix
in alkalinity but to avoid the delay of set observed for this waste. Tests identified the reaction
of the cement with the mobile lead in the waste as the cause of delayed set. Twenty percent
(weight lime/dry soil weight) dolomitic hydrated lime and 2% dolomitic quicklime were added
to the waste, hand mixed, covered, and allowed to stand seven days. These were chosen as the
most alkaline materials available that would precipitate the lead before the cement was added.
At the end of that time, what had been a very damp soil had become a dry, free flowing, crum-
bly soil. Ten percent portland cement (weight cement/dry original soil weight) was added to
the now pretreated soil along with sufficient water for flow, and samples were cast. All physical
properties of the "engineered" mix were superior after 28 days of cure from faster set ( 13 days
versus 30 days) and higher strength at a given time [ 1.65 MPa (240 psi) versus 0.12 MPa (17
psi) at 35 days old] to lower permeability (8.9 X
10 -8 cm/s versus 6.8 • 10 -6 cm/s).
Results
The leach test results shown in Figs. 12 and 13 were unexpected. The "engineered" mix
showed 2000 times higher lead concentrations in the TCLP leachate than the 30% cement mix,
which performed as expected (2 eq/kg data points, Fig. 12). The results of a multiple TCLP
1.0E4
E ]N Condition of the
o_
r-1
[
] TCLP Test
~ O .[
c- I000,0
o
E 100;0
....
)~
O---.9 "u..
- --.+1\ I 0 /
/0/'/"
o
c-
0 I "0\ /
O ~o I
I
-,,[]
/
/
C)
]D 10.0
o 5.0
g3
.d
Limit for Lead T~ E] .... D _ 7 ~ I . . F I _ . . _ . . _ . . /
1.0
(5.0 ppm) I O~ / 0--0 30 % Portland Cement Mix
c-
o I 0~0/0 [] . . . . [] "Engineered" Mix
u
g3 [ 22~ Limes, 10% Cement
J 0.1 , J , I J , ) , ~ , z ~ , f ,
0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 5.5 6.0 6.5 7.0 7.5 8.0
Equivalents of A c e t i c Acid p e r K i l o g r a m of Dry S a m p l e
Copyright
F I Gby
. ASTM
1 2 - - GInt'l
AN (all (metals)
C rights reserved);lead
Mon releases to leachate
Feb 24 23:07:03 EST 2020 from stabilized waste as a function o f acidity added.
f-" 1.0E4
E
Q_ O--O 30~g portland cement
) [] []-- I-I "Engineered" Recipe,
E 1000.0 - " ~ . 22~ Limes, 10~portlond cement
._oo_~ -O\~D D~
.4.a
E 100.0 /
o D
~-O 10.0 TCLP Regulotory Limit ~
/ /
o 5.0 . . . . . . . . . . . . . . . . _-/-- --<9-
/
0) D/DD~D U
-~ 1.0 ~ 0
=
o \ /
o~ O ~ o / O
--J 0.1 ' ~ ~ ' ' ' ' J ' J ' ' '
5.5 6.0 6.5 7.0 7.5 8.0 8.5 9.0 9.5 10.0 10,5 11.0 11.5 12.0 12.5
Leachate pH
FIG. 13--GANC (metals) lead releases to leachate from stabilized waste as a function of p H ofleachate.
(which Center Hill is using to replace the EPA Method No. 1320, Multiple Extraction Proce-
dure (MEP) now that TCLP is the regulatory standard) suggest that we "engineered" too much
alkalinity into the mix since the leachatcs of successive increased acidity stages were actually
lower in lead concentrations. On the acid side, the 30% cement mix released lead at between
3.0 and 3.5 eq/kg of acid while the "engineered" mix required 4.5 to 5.0 eq/kg.
When the two mixes are superimposed on a graph of the leachate pH versus lead concen-
tration in the leachate (Fig. 13), the amphoteric solubility curve for lead in a saturated calcium
ion environment is obtained. This is the first time that researchers at Center Hill have detected
the alkafine end of the pH curve above pH 10.5 in the leachate.
Conclusion
Apparently, there is a "window" of pH values in the leachate of the TCLP test which, by
definition, "stabilizes" lead in a soil. This effect might also be expected in copper, cadmium,
zinc, and certain other heavy metals based on their published pH versus solubility curves
[6]. Despite the lack of passing TCLP in our first attempt at mix design using GANC results,
the GANC test should prove useful in the future at predicting the amount of binder of any
given type needed to place the waste at any particular point in that "window." While the
detailed data needed for engineering other future mix designs are not yet in hand, the patterns
to study and the methods to use to develop those patterns seem to be clear. Meanwhile, there
can be "too much of a good thing" with regard to alkalinity added to stabilize a metal-contam-
inated waste.
Acknowledgments
Although the research described in this article has been funded wholly or in part by EPA
under Contract 68-03-3379 with the University of Cincinnati, it has not been subjected to the
agency's review and therefore does not necessarily reflect the views of the agency, and no offi-
Copyright
cial by ASTM Int'l (all rights
endorsement reserved);be
should Moninferred.
Feb 24 23:07:03 EST 2020
T h e a u t h o r s w o u l d hke to t h a n k S t e p h e n Proflitt, M a r k K e m p e r , A n d r e a s Slckert, a n d

I n g n d K h c h for t h e i r w o r k o n this project.
APPENDIX A
Generalized Acid Neutralizing Capacity Procedure (GANC)

Introduction
Stabdizmg hazardous wastes with highly alkaline, morgamc binders is currently being reviewed as a
possible treatment method for heavy metal-contaminated solid wastes, soils, and sludges, Alkahne binders
reduce leachabihty by surrounding metal wastes m a high pH matrix Within the matrix, metal ~ons are
converted from water-soluble salts into less soluble hydroxide forms However, ff pH is raised too high,
metals will again form water-soluble speoes prone to teaching Statable binders further control leachabd-
lty by releasing addmonal hydroxtdes to buffer water or leachant that comes m contact with treated waste
Thin abihty to buffer acid leachant is important because metal hydroxides become increasingly soluble as
leachate pH drops below 8 5 Determining potentml buffenng capaoty is an integral part m evaluating a
binder's abfltty to tmmobd~ze heavy metals
A procedure was needed to measure buffenng capacity that allowed results to be correlated with
approved standard methods Buffenng capamty is the neutrahzmg reserve of a binder after contact with
an aci&c leachant and is measured as final leachate pH Adapting Environment Canada aod neutrahza-
tlon capacity (ANC) [ 1] with the Environmental Protection Agency's (EPA) toxicity characteristic leach-
mg procedure (TCLP) [4] has led to the development of a generalized a o d neutrahzmg capaoty (GANC)
procedure used at the Center Hill facfllty
Method
The GANC method is a single-batch leaching procedure that utlhzes a series of dried samples extracted
with increasingly aci&c leachant Leachant strength starts out at 0 equivalents of acetic acid (C2H402) and
is increased in proportional Intervals until pH is below 5 for three consecutive equivalents Calcium acetic
buffers leachate pH below 5 and masks the neutrahzlng affect of a sample GANC uses a hquld-to-sohd
ratio of 20 1, approximately a TCLP extraction Sample particle size is reduced to pass through ASTM
sieve No 40 This size maximizes surface area without overly manipulating the sample Approximately
1 00 g of dried sample is wetghed out into 21 sets of 125-mL bottles A set consists of three to five duplicate
speomens for each aod eqmvalent interval Equivalents are generated by increasing the ratio of acid to
water m a total leachant volume of 20 mL An eqmvalents schedule for pure binder is provided below
(Table 1) As the percent of binder decreases, the strength of acid should also decrease proportionally
Water is added first to avoid subjecting dry sample to &rect contact with the 2-N aceUc acid Aod makes
up the remaining leachant volume Speomen bottles are rotary agitated for 48 h and then allowed to settle
for 15 mIn Leachant contact t~me of 48 h allows the sohds to equlhbrate with the hquld and atlowmg t)me
for separation of the two phases reduces interference in pH analysis
Procedure
1 0 Sample Preparatlon
I 1 Sample Drying
1 1 1 Reduce initial sample particle size to <% m
1 l 2 Weigh out about 200 g net weight of sample Into a drying &sh
1 1 3 Dry sample at 105~
1 1 4 Weigh and record dried sample weight In lab book
1 1 5 Return sample to oven for 2 h
Copyright by ASTM1Int'l
1 6(all Rewelgh andMon
rights reserved); record dlaed sample
Feb 24 23:07:03 EST 2020 weight in lab book
Downloaded/printed1by1 7 If weight unchanged < 1%, continue to Step 1 2
TABLE 1--Equivalents schedule, 1 O0 g +_ 0 O1 g dry wetght of sample, 2N acettc acM
Sample Identlficatmn Water, mL Aod, mL Eqmvalents per Kalogram
0 20 0 0
1 19 1 2
2 18 2 4
3 17 3 6
4 16 4 8
5 15 5 10
6 14 6 12
7 13 7 14
8 12 8 16
9 11 9 18
10 10 10 20
11 9 11 22
12 8 12 24
13 7 13 26
14 6 14 28
15 5 15 30
16 4 16 32
17 3 17 34
18 2 18 36
19 1 19 38
20 0 20 40
Reagents
2 N acetic aod, glacml
ASTM D 1193-77 Type I water
116 mL concentrated (99 7%) acetic aod diluted to 1 L with ASTM Type I water
Equtpment
Agitator, rotary capable of 30 _+ 2 rpm
Balance, l kg + 0 005 g
Bottles, 125-mL hlgh-denslty polypropylene
Liquid measuring devtces, either burets, (1) 50 mL _+ 0 05 mL, (1) 25 mL _+ 0 05 mL Replpetters,
(l) l0 mE, (l) 50 mL
Drying oven at 105~
Mortar and pestle
pH meter and buffers
Sieve, ASTM E I 1-87 No 40 and No %
1 1 8 If weight changed > 1%, return to Step 1 1 4 and repeat the following steps until a < 1%
change is observed
1 2 Particle Size Reduction
1 2 1 G n n d the dned sample with mortar and pestle to pass through an ASTM No 40 sieve
1 2 2 Grind particles remaining on the screen until 130 g have been collected
1 2 3 Store sample m desiccated environment and properly &spose unused port~on
1 3 Bottle Labehng
1 3 I Arrange 63 bottles m sets of three and begin labehng
1 3 2 A set of bottles is numbered following the Equivalents Schedule, 0 through 20
1 3 3 Within a set the bottle number is followed by a letter (A or B or C), which represents one
of the tnphcate equwalence
1 4 Weighing Out Sample
1 4 1 Weigh out 1 0 g of sample into each bottle
1 4 2 Record the sample ID and exact weight on an GANC report form
2 0 Leachant Addition
2 1 Using the Equivalents Schedule
Copyright by ASTM 2 Int'l
1 1(allFdl
rightsand labelMon
reserved); theFeb25-mL buret
24 23:07:03 or 10-mL replpetter w~th 2 N acetic aod
EST 2020
Downloaded/printed2by1 2 Fill and label the 50-mL buret or 50-mL rep~petter with ASTM Type I water
2 1 3 Following the schedule, add the a p p r o p n a t e a m o u n t o f water to each set o f bottles

2 1 4 After water is added, add the a p p r o p n a t e a m o u n t o f acid to each set o f bottles
3 0 Sample Extraction
3 1 Agitation
3 1 1 Shake each bottle m a n u a l l y until the contents appear thoroughly m i x e d
3 1 2 Place the bottles into the rotary agitator
3 1 3 T u m b l e for 48 h
3 1 4 R e m o v e the bottles a n d allow to set for a b o u t 15 m i n
4 0 Leachate Analysis
4 1 p H o f Leachate
4 t 1 Calibrate the p H m e t e r following manufacturers' r e c o m m e n d a t m n s
4 1 2 Record pH reading for each sample on G A N C Report F o r m
Analysts
Final leachate p H for each bottle is determined and recorded on a G A N C Report Form an example as
found in Table 2 Plotting p H values on the Y axis a n d acid equivalents per kilogram of dry sample on
the X axis will reveal a point(s) o f Inflection in the resulting curve Inflection points represent the maxi-
m u m n u m b e r o f acid equivalents a binder can neutralize before losing ItS alkaline stability Alkaline
behavior of a binder characterizes ItS chemical resistance to acidic leachants T h e m o r e acld eqmvalents
a binder is capable o f neutrahzing above p H 8 5, the greater its buffenng capacity An o p t i m u m binder
would convert metal ions Into hydroxide species, elevate leachant p H in contact with a treated waste, a n d
m a i n t a i n a l k a h m t y in reserve
Comments
Research was u n d e r t a k e n to determine buffenng capacity of analyl~cal grade calcium hydroxide
[Ca(OH)2] and various hydrated a n d u n h y d r a t e d genene binders, portland c e m e n t Type l, c e m e n t kiln
dust, limekiln dust, dolomltlC limes, a n d dolomitic lime mixed w~th fly ash Type F Each has a slightly
&fferent alkaline characteristic that could m a k e it m o r e or less appropriate as a waste stabilizing agent
Pure binders were also blended with varying a m o u n t s o f inert sohds a n d tested Plotting pH values o f pure
samples w~th those o f blends reveals a proportional decrease m alkaline b e h a w o r An unexpected change
In alkahne curve m a y indicate an interactmn between waste a n d binder Analytical grade Ca(OH)2 was
also used to test solids dilution further and to check reproducibility Examples o f approximate eqmvalence
for v a n o u s genenc binder is shown m Table 3
T h e procedure was initially designed for testing pure binders with high alkalinity To overcome high
p H in pure binders, a stronger concentration of acid was used However, when testing dilute binders or
w h e n m o r e detailed information about an reflection point was needed, a weaker acid concentratmn was
used A weaker concentration allowed for smaller acid equivalent intervals Smaller intervals yielded a
m o r e defined characterization o f a binder blend
APPENDIX B
Chemistry of Binders
Common lon Effect
W h e n two partially soluble materials [that is, c a l e m m hydroxide, Ca(OH)2, a n d calcium carbonate,
CaCO3] are both dissolving in the s a m e water a n d they have one ion in c o m m o n (that is, Ca ++, concen-
tration = [Ca++]), the concentration of that ion is the s u m of the a m o u n t s from each material (see Eq 1
below) Since each solubility is controlled by a solubility product (that is, RseCa(OH)2 = [Ca + +I[OH ]2 a n d
Ksvcaco~
Copyright = [Ca
by ASTM + +][CO;]),
Int'l (all the
rights reserved); other
Mon Feb 24ions concerned
23:07:03 EST 2020 (that IS, O H - ) will show a decrease (that is, lower pH)
TABLE 2--GANC reportform
1 00 g _+ 0 01 g Dry Weight of Sample

Acetic Acid Concentration _ _
Binder Descnptlon
Final pH
Sample Water, Acid Equivalents,
Identification mL mL per kg A B C
0 20 0
1 19 1
2 18 2
3 17 3
4 16 4
5 15 5
6 14 6
7 13 7
8 12 8
9 11 9
10 10 10
11 9 11
12 8 12
13 7 13
14 6 14
15 5 15
16 4 16
17 3 17
18 2 18
19 1 19
20 0 20
Notes
compared to their values without the second material Thls also occurs when part of a substance [that is,
Ca(OH)E] is dissolved by reaction (see Eq 2 below) and part remains to saturate the solution [that IS,
Ca(OH)2 ~ Ca ++ + 2OH-]
mCa(OH)E + nCaCO3 ~ (m + n)Ca ++ + 2mOH- + nCO3 (1)

nCa ++ + mCa(OHh + 2mHCEH3OE ---"(m + n)Ca ++ + 2mCEH3OE- + mHEO (2)
nMg ++ + mMgO + 2mHCEH3OE ~ (m + n)Mg ++ + 2mCEH3OE- + mHEO (3)
TABLE 3--Approxtmate neutrahzmg eqmvalents available with

pure binder
Binder Type Eqmvalents per Kilogram
Quickhme (dolommc) 44
Hydrated Lime (Type N) 34
Hydrated Lime (Type S) 30
Hydrated Lime 27
Portland Cement (Type I) 22
Portland Cement Kiln Dust 16
Limekiln Dust 16
Cement and Lime Binders

The binders used to stabdaze heavy metal toxic wastes are most commonly based on some form of
hmestone Portland cement as made by heating to lUStmolten a blend of ground shale and hmestone Dme
may be one of several forms from one stage of the following process hmestone is calcined to qmckllme,
which is hydrated to Type S or N lime In chem,cal terms
CaCO3 ~ CaO + CO2~then CaO + H20 ~ Ca(OH)2
Dolomite
While portland cement requires hmestone hawng almost no magnesmm, the natural process of hme-
stone precxpltatlon in the ocean frequently produces a double salt ofcaloum and magnesaum carbonates
called dolomite Dolomite or dolom~tlc hmestone is generally unsmtable for cement manufacture, on the
other hand, it xs used extensively to make the various hines The double salt breaks down dunng calcining
to separate phases of magnesmm and calcmm oxides
CaMg(CO3)2 ~ CaO + MgO + 2CO2i'
Hydration, usually by steam, proceeds easdy for calcmm oxide but very slowly for magnesmm oxide The
usual hydrated hme from dolommc hmestone is caloum hydroxide plus magnesium oxide
CaO + MgO + H20 ~ Ca(OH)2 + MgO
Likewise, calcium hydroxide reacts w~th acid easdy at a pH of about 12 45 while magnesium oxide
reacts less vigorously at a pH of about 10
APPENDIX C
Statistical A c c u r a c y o f t h e G A N C T e s t
Twenty duphcates of data points just before, dunng, and .lust after the endpolnt of the aod reaction
with calcium hydroxide [Ca(OH)2] were run The statistical parameters that resulted from fitting a normal
&stnbutlon to the data are shown in Table 4
TABLE 4--GANC stattstlcs
pH pH Standard pH Coeffioent of
eq/kg Average Deviation Variation
22 1194 a 0022 a 019 a

24 t 1 85 0 023 0 19
25 1101 0416 397
26 6 37 0 214 3 36
28 5 44 0 032 0 59
a One data point ehmmated as an outher or operator error by

the method of ASTM Standard Recommended Pracuce for Deal-
ing with Outlying Observations (E 178-80) at the 0 1% probabd~ty
level
References
[ 1] Test Methods for Sohdlfied Waste Characterization, Acid Neutrahzatlon Capactty (ANC), Method
No 7, EnvironmentCanada and Alberta Enwronmental Center, 1986
[2] C6t~, P L and Bridle, T R, "Long-Term Leaching Scenarios for Cement-Based Waste Forms,"
Waste Management and Research, Vol 5, 1987a, pp 55-66
[3] Stegmann, J, C6t~, P L, and Hannak, P, "Prehmmary Results of an Internataonal Government
Study of Waste Stabdlzatlon/SohdlfiCatlon," Hazardous Waste Detectton, Control, Treatment, R
Abbou, Ed, Elsevier Soence Pubhshers B V, Amsterdam, The Netherlands, 1988, pp 1539-1548
[4] Test Methods for Evaluating Sohd Waste, Toxloty Characteristic Leaching Procedure (TCLP),
Method 1311, SW-846, 3rd ed, U S Environmental Protection Agency, Washington, DC, Nov
1986
[5] Shlvely, W, Bmhop, P, Gress, D, and Brown, T, "Leaching Tests of Heavy Metals Stablhzed with
Portland Cement," Journal of the Water Pollution Control Federatton, Vol 58, No 234, 1986
[6] U S EnvironmentalProtection Agency, FederalReglster, Vol 52, No 155, 12 Aug 1987, p 29999
T. S. B a j p a y e e 1 a n d R t c h a r d J . Malniero 1
Methods of Evaluating Explosive Reactivity of

Explosive-Contaminated Solid Waste
Substances
REFERENCE" Bajpayee, T S and Mamlero, R J , "Methods of Evaluating Explosive Reactiv-
ity of Explosive-Contaminated Solid Waste Substances," Stablhzatlon and Sohdlficatlon of
Hazardous, Radtoactgve, and Mlxed Wastes, 2nd Volume, ASTM STP 1123, T M Gilham and
C C Wiles, Eds, American Society for Testing and Matenals, Philadelphia, 1992, pp 378-384
ABSTRACT. The Bureau of Mines has developed test procedures and criteria for evaluating
explosive reactivity of explosive-contaminated solid waste substances generated by U S Army
ammunition plants These substances are produced as explosive-contaminated sludge from
wastewater treatment plants, residues from the burning of munitions and explosives on open
ground, and residues from deactivation furnaces The characterization of explosive reactivity is
a prerequisite for disposal of such waste materials, which may be contaminated with primary
explosives, propellants, or pyrotechnic materials The Bureau has proposed two tests for this pur-
pose These tests were developed to evaluate the explosive reactivity as defined in Title 40, Code
of Federal Regulations, Part 261 23(a)(6) and (7) These are gap and internal ignition tests, which
determine the sensitivity to shock and thermal stimuli, respectively This paper also includes gap
and internal ignition reference data for typical blasting agents, high explosives, propellants, and
marginally reactive substances These reference data were used to establish test criteria The
Bureau has evaluated over 400 samples of contaminated soil, sludge, and burning residues using
these two test methods
KEY WORDS. detonation, explosion, explosive, gap test, internal ignition test, mutations, reac-
tivity, sohd waste
The U S Bureau o f Mines has c o n d u c t e d research to establish methods and c n t e n a to eval-

uate the explosive reactwlty of explosive-contaminated sohd waste substances generated by
U S A r m y a m m u m t l o n plants These substances are c o n t a m i n a t e d with various primary
exploswes, propellants, and pyrotechnic materials Such samples are generated as sludge from
wastewater t r e a t m e n t plants, residue from burning of munitions and explosives on open
ground, and residues from deactivation furnaces U n i f o r m testing methods and assessment
criteria were developed to evaluate the explosive reactivity of substances generated from all
three sources
T h e purpose o f this research initiative was to develop tests suitable for determining the prop-
erhes described in 40 C F R 261 Subpart C "Characteristics of Hazardous Waste", in particular,
Paragraph 261 23(a)(6) and (7), "Charactenstlc o f ReacUvlty," which defines a solid waste as
having the characteristic o f reactivity if it has, a m o n g others, any of the following properties
(a)(6) Capable o f detonation or explosive reaction if subjected to a strong initiation source, or
If heated u n d e r confinement, and (a)(7) Readily capable o f detonation or explosive decom-
position or reaction at standard temperature and pressure
1 Supervisory physical scientist and research supervisor, respectively, Pittsburgh Research Center, U S
Bureau of Mines, Cochrans Mill Rd, Pittsburgh, PA 15236
378

BAJPAYEE AND MAINIERO ON EXPLOSIVE REACTIVITY 379
The Bureau is a recognized expert in the field of explosives and is represented in many
national and international organizations At the request of the U S Environmental Protection
Agency and the U.S Army, the Bureau initiated research to establish testing methods and cri-
teria for characterizing the reactivity of explosive-contaminated solid waste substances The
characterization of explosive reactivity of such substances IS a prerequisite for their disposal
This research was aimed at evaluating detonatablhty or explosive reactivity when subjected to
a strong initiating source or heated under confinement Two separate tests were developed the
Bureau's gap and internal ignition tests
The gap test is an arrangement for determining the response of a substance to a shock stim-
ulus under confinement The sample is contained in a length of steel pipe and subjected to a
shock stimulus from the detonation of a pentollte booster in contact with the sample The
occurrence of detonation is determined by fragmentation of the pipe, the rate of propagation
of the pressure wave in the sample (using an electronic velocity probe), and the perforation of
a steel plate at the end of the pipe opposite the booster
The internal ignition test is an arrangement for determining the response of a sample to
heating by internal lgnmon under confinement The sample ~s placed m a steel pipe capped at
both ends An igniter capsule containing 20 g of commercial grade FFFg (grain size) black
powder inserted in the center of the sample is ignited The response of the sample IS observed
according to vanous degrees of reaction (failure to ignite, partial burning, cap blown off,
pipe bulged, split, or laid open, pipe fragmented, and pipe and caps fragmented) The frag-
mentation of the pipe and/or caps into more than two distinct pieces indicates explosive
reactivity
It is considered necessary to perform both tests since there are materials that are sensitive to
ignition under confinement but not to shock, there are others that are sensitive to shock but
not to igmtion under confinement Th~s paper provides a description of the two test proce-
dures The Bureau has tested over 400 explosive-contaminatedsamples using these procedures
and reported the results to the respective U S Army ammunition plants
The Bureau also conducted validation tests using typical blasting agents, high explosives,
propellants, and marginally reactive substances These validation tests were used to establish
test c n t e n a
Experimental Setup and Test Procedure

Gap Test
The apparatus for the gap test is shown in Fig 1 The test sample is contained in a cylinder
16-1n (40 6 em) long, made of 189 -diameter schedule 80 black seamless-steel pipe The steel
pipe holds a 425-mL sample A mild steel witness plate 6 in (15 24 cm) square and 0 125 in
(0 32 cm) thick is mounted at the upper end of the sample pipe and separated from it by spacers
of 0 062-1n (0 16-cm) thickness The bottom of the cylinder is closed with two layers of 0 003-
in (0 008-cm)-thick polyethylene sheet held In place with gum rubber bands and polyvlnyl
chlonde electncal insulating tape
There xs no other gap between the pentohte booster and the test sample as used in this test
A continuous velocity of detonation [1] probe made of thin a l u m i n u m tubing with an axial
resistance wire having a resistance of 7 62 ohms/in (3 0 ohms/cm) is mounted on the wall of
the sample holder The outer tubing of the probe IS cramped against the inner wire at the lower
end, forming a resistor When this assembly is inserted in a medium that transmits a shock
wave, the outer wall crushes against the inner wire as the wave moves up the tubing, shortening
the effective length and changing the resistance Ifa constant current (usually 0 06 A) is made
Copyright
to flowby ASTM Int'l (all rights
between the reserved);
outer MonandFebinner
24 23:07:03 EST 2020
conductors, the voltage between them is proportional to
Witness plate \
I I
Spacers j 'lil'-~,.
III
Rate probe
!11
iii
];I I -in Schedule 80
"~
II I seamless-steel pipe
III
II I
16" " ~ Sample

iii
III
'I,I',I1
III
III
III
Ill
rill
III
ill
l~=J__~
_ Synch wire
Pentolite ,. ~
pellet ~ I n**,
,, ~Detonator
L ~ holder
" Detonator
FIG. 1--Gap test for solids.
the effective length and can be recorded as a function of time using an oscilloscope. The slope
of the oscilloscope trace is proportional to the velocity of the shock wave.
The apparatus for the gap test for liquids is the same as that for solids except that a method
of injecting bubbles into the liquid sample is provided. The experimental setup is shown in
Fig. 2. The bubbles are injected by means of a 0.925-in. (2.35-cm)-diameter loop of vinyl plas-
tic tubing located at the bottom of the sample. The tubing is the type used for medical cathe-
terization, with an outside diameter of 0.07 in. (0.18 cm) and a wall thickness of 0.016 in. (0.04
cm). The loop is perforated with two rows of holes diametrically opposite each other with the
holes in each row spaced 0.125 in. (0.32 cm) apart. The holes are made by inserting a 0.05-in.
(0.13-cm)-diameter needle through the wall of the tubing. Owing to the elastic nature of the
tubing, the holes contract almost completely when the needle is withdrawn, so the actual hole
diameter is much smaller than 0.04 in. (0.1 cm). The tubing is sealed at one end of the loop
with epoxy cement, and a length of the tubing from the other end of the loop is led outside to
the air supply through a hole in the steel pipe, which is also sealed with epoxy cement. Air is
supplied at a pressure of 0.3 to 1.0 atm (30 to 100 kPa) to obtain a flow rate of 2.5 ft3/h (1.2 L/
min). Where it is suspected that the sample may react with the steel tube, the inside of the tube
is sprayed with a fluorocarbon resin coating.
The sample is loaded into the top of the steel tube. For liquid samples, adequate ullage
should
Copyright be Int'l
by ASTM allowed. Solid samples
(all rights reserved); are loaded
Mon Feb 24 23:07:03 EST 2020 to the density attained by tapping the cylinder
Witness plate--/
I ,-- !
9 9 II
Spacers j '," ~ -,,I "~=:~Rate probe
I 9 9 . .1
I ~ o , - hi
I# e ". # I l l
I * *11 ~ I "
' . . . : 9 :I l Y 2 - m Schedule 80
I 9 * ii
I." " .p.epi"' 'I] se~ steel
P ,0, ~ "1
16'! , .. ~Sample
,.- o ~ *:~
, o . . - ,,
Toair .......
supply i f : o*:';I I/Bubbler
Pentolite/@ ~ ~
pellet [ Synch wire
I ,, / DetOnatOr
~ holder
~ Detonator
FIG. 2--Gap test for liquids.
until further settling becomes imperceptible. The sample, at 25 + 3"C, is subjected to the
shock wave generated by the detonation o f a pentolite (50:50 PETN:TNT) pellet; 2 in. (5.08
cm) diameter by 2 in. (5.08 cm) thick, having a density of 1.6 + 0.05 g/cm 3. The pentolite
pellet is butted against the bottom of the test sample and initiated with a No. 8 strength electric
detonator. The detonator is held in place by a cork detonator holder. Three trials are per-
formed on each sample.
The criteria for detonation propagation (positive result) are as follows:
1. The sample tube is fragmented along its entire length.

2. A hole is punched in the witness plate.
3. A stable propagation velocity greater than 4900 ft/s (1.5 km/s) is observed.
The sample is considered reactive if it satisfies the above criteria in any trial. If a positive
result is observed during the first or second trial, further testing is discontinued and the sample
is considered reactive.
lnternal lgnition Test

The experimental arrangement is shown in Fig. 3. The sample to be tested is contained in
an 18-in. (45.7-cm)-long by 3-in. (76.2-mm) diameter schedule 80 carbon-steel pipe with 2.9-
~ Igniterleads
~,~ Forgedsteel cap
~.f--- 5-in Schedule80

seamlesssteel pipe
-/•
,2 !
f !~ ~/8-in Schedule40
seamless-steelpipe
~ Seal
18" I
~ Igniter assembly
J I \
I I__ \
\
\
~
_ %
_ Forged steel
cap
FIG. 3--Internal ignition test.
in. (7.37-cm) ID, a wall thickness of 0.30 in. (0.76 cm), and capped at both ends with 3000-1b
(13 608-kg) forged steel pipe caps. The pipe holds a 1950-mL sample.
The sample is subjected to the thermal and pressure stimuli generated by an igniter consist-
ing of 0.7 oz (20 g) of commercial grade FFFg (grain size) black powder located at the center
of the sample vessel. The igniter assembly consists of a cylindrical container 0.81 in. (2.06 cm)
in diameter and 2.5 in. (6.4 cm) long, which is made of a 0.01-in. (0.0254-cm)-thick cellulose
acetate sheet held together by two layers of nylon-filament-reinforced cellulose acetate tape.
The igniter capsule contains an ignition source that is a resistance heater. The resistance heater
consists of a small loop formed from a 1-in. (2.54-cm)-long nickel-chromium alloy resistance
wire 0.012 in. (0.030 cm) in diameter having a resistance of 0.343 ohm. This loop is attached
to two insulated tinned copper lead wires 0.026 in. (0.066 cm) in diameter. The overall wire
diameter including insulation is 0.05 in. (0.127 cm). The lead wires for the igniter are fed out
through a H-in. (3.2-cm) schedule 40 seamless-steel pipe attached to one of the pipe caps.
For gelatinous samples, the substance is packed as nearly as possible to its normal shipping
density. For granular samples, the substance is loaded to the density obtained by repeated tap-
ping of the pipe against a hard surface. The igniter is fired by a current of 15 A obtained from
a 20-V transformer. Three trials are performed on each sample. The sample is tested at a tem-
Copyright by ASTMof
perature Int'l
25(all +rights3~reserved); Mon Feb 24 23:07:03 EST 2020
The criterion used for interpretation o f a p o s m v e result is that either the pipe or at least one
o f the end caps be fragmented into at least two distinct pieces In any trial I r a positive result
is observed d u n n g the first or second trial, further testing is discontinued and the sample is
considered reactive Results in which the pipe ~s merely split or laid open or In which the pipe
or caps are distorted to the point where the caps are blown off are considered to be negative
results
Vahdatton Tests
Validation of the above tests and criteria were done as part of the original effort in devel-
oplng these tests The rationale of the vahdation was to set the test stimulus relative to the suite
o f substances, all of which possess, at least theoretically, some capacity for energy-releasing
TABLE 1--Examples of vahdatton tests for vartous reJbrencesubstances
Internal Ignition
Sample Designation Gap Test Test
BLASTING AGENTS
Ammomum mtrate-fuel od P P
Water gel (TNT sensmzed) P P
HIGH EXPLOSIVES
Flake TNT P P
Granular TNT P P
Nltroguamdlne P P
PROPELLANTS (CANNON)
Sample A P P
Sample B P P
Sample C P P
Sample D P P
Sample E P P
Sample F P P
Sample G P P
MARGINALLY REACTIVE
Ammonium mtrate pnlls, low, medmm, high density N N
Ammonium perchlorate
Coarse crystals N P
Free crystals P P
Fine crystals, low density P P
Benzoyl peroxide, powder N N
Guamdme mtrate, granular P N
m-dlmtrobenzene, dry, fine crystals P P
N~trocellulose
11 5% nitrogen, 20% water P P
11 5% mtrogen, 30% ~sopropanol P P
13 2% mtrogen, 30~ ethanol P P
13 3% mtrogen, dry P P
13 3% mtrogen, 20% water P P
Potassmm chlorate-lactose (50 50) P P
Smokeless powder (small arms) P P
Sodmm pmramate N P
2,4 dmltrophenol, granular P N
2,4 dmltrotoluene, granular P N
Copyright
N ObyT ASTM
E N Int'l= (all rights reserved);
negative result,MonPFeb
= 24positive
23:07:03 EST 2020
result
384 HAZARDOUSWASTES
chemical reaction. In addition to materials with known explosive properhes, such as explo-
sives, blasting agents, and propellants, the experlmentatlon included (l) substances that are of
minimal explosive hazard but that still have exhibited some potentml to detonate or violently
burn with sufficiently strong sUmulus and confinement, and (2) substances that, while theo-
retically capable of releasing energy when decomposed, have not been known to burn or det-
onate as a result of involvement m transport or storage (such as dense-prilled a m m o n i u m
nitrate and dmltrotoluene) The stimuli of the two tests were set at values that would give pos-
itive results for as many as possible of the first type of substance, whde giving negative results
for as many as possible of the second type of substance Table 1 lists reference substances and
the test results.

The Bureau has evaluated over 400 samples of contaminated soil, sludge, and burning res-
idue from U S A r m y ammunition plants Only two of these samples showed evidence of
explosive reactivity. One of the samples showed a positive result in the internal lgnmon test.
The sample reacted wolently to the thermal stimulus The end caps were blown off and the
pipe was fragmented into more than three pieces
The other sample reacted violently in the gap and internal ignition tests In the gap test, the
pipe was fragmented into small pieces, a hole of approximately 3-in (76 2-mm) diameter was
punched in the witness plate, and the detonation velocity probe showed a rate in excess of 6600
ft/s (2 km/s) In the internal ignition test, the end caps were blown offand the pipe fragmented
into more than three pieces However, the majority of the contaminated samples were non-
reactive relative to 40 C F R 261.23(a)(6) and (7).
Conclusions
G a p and internal ignition tests were performed to a state of known reactive and marginally
reactive formulations to demonstrate their apphcablllty for evaluation of explosive reactivity
These tests evaluate the reactivity to shock and thermal stlmuh, respectwely The test protocol
included both tests because some reactive formulations are sensmve to shock while others are
sensmve to thermal stimulus These tests are very effective in differentiating reactive materials
from others This test protocol was apphed to evaluate the explosive reactivity of explosive
contaminated solid waste substances relative to 40 CFR 261 23(a)(6) and (7)
Reference
[1] Rlbovlch, J, Watson, R W, and Gibson, F C, "Instrumented Card-Gap Test," AIAA Journal, Vol
6, No 7, 1968, pp 1260-1263
M. John Cullinane, Jr. i and Larry W. Jones 2
An Assessment of Short-Term Penetration

Testing as an Indicator of Strength
Development in Solidified/Stabilized
Hazardous Waste
REFERENCE" Culhnane, M J and Jones, L W , "An Assessment of Short-Term Penetration
Testing as an Indicator of Strength Development in Solidified/Stabilized Hazardous Waste,"
Stabthzauon and Sohd~catlon of Hazardous, Radmacuve, and Mixed Wastes, 2nd Volume,
ASTM STP 1123, T M GIlham and C C Wiles, Eds, American Society for Testing and Mate-
rials, Philadelphia, 1992, pp 385-393
ABSTRACT" The ultimate strength ofsohdlfied/stablhzed waste is an important physical prop-

erty that frequently must be determined for each batch of treated waste The usual test for this
purpose is the unconfined compressive strength (UCS) taken after 28 days of cure An alternaUve
test method, the resistance to penetratmn or cone index (CI), as measured using a cone pene-
trometer, can also gwe meaningful rnformatlon on the settrng and strength development prop-
erties ofcementltlous materials after as httle as one, five, and ten days of cure This paper presents
an analysis of the possible correlatmn of the 28-day UCS with the one-, five-, and ten-day CI for
a vanety of low-strength, sohdlfied/stabfllzed waste matenals
The data used m thls analysm were developed in an evaluation of the effects of ten matenals
that have the potential to interfere with the contaminant contamment and strength development
properties ofa sohdlfied/stabdlzed metal hydroxide sludge Three binder systems (cement, lime/
fly ash, and cement/fly ash) were used to sohdlfy/stabdlze a specmlly prepared metal hydroxide
sludge containing substantial concentratmns of four metal contaminants (cadmmm, chromium,
nickel, and mercury) Ten interfering matemals were lndwldually added to the sludge The effects
of these materials on the physical properties of the treated sludge were evaluated using UCS and
CI testing methods Substantial differences In the settrng and strength development charactens-
tics of the different brnder systems precluded pooling of the total data set from all binder systems
Even variations within the populatmns from each binder system are quite large
The analysis of data generated dunng the study indicated a slgntficant correlatmn (99% con-
fidence level) between the one-, five-, and ten-day CI and the 28-day UCS for specimens from all
three binder systems except the five-day specimens prepared with cement and the one-day spec-
imens prepared with hme/fly ash However, vanablhty m the test values between the different
batches used m the experiment resulted in high standard deviatmns and relatively low correlation
coefficients (around 0 5) This varaaNhty reduces the confidence in the ability to predict 28-day
UCS from the CI in quantltatwe terms
The data set and analys~s procedure is a stnngent test of the predxcuve relatmnsh~p between
CI and UCS In general, the interfenng materials used m the underlying study had greater effect
on the early setting and strength development properties (one to ten days) than on the longer
term (28 days) strength Materials that cause retarded set of cernentltlOUS materials, and thus
strongly affect the one- to five-day CI, often resulted m long-term strengths greater than the con-
trol specimens Based on the data collected dunng this study, a one- to five-day measurement of
the CI ~s shown to prowde a good red,cation of the 28-day strength However, ~t seems unhkely
Research civil engineer, U S Army Engineer Waterways Experiment Station, ATTN CEWES-EE-S
3909 Halls Ferry R d , Vicksburg, MS 39180-6199
2 Professor, Waste Management Research and Educauon Institute, University of Tennessee, Knoxville,
TN 37996-0710

386 HAZARDOUSWASTES
that a short-term test of any type can be used as a rehable quantitative predictor of long-term
strength in the highly variable waste streams and low-strength products typicallyencountered m
the treatment of hazardous wastes
KEY WORDS" hazardous waste, sohdlficat]on/stablhzatlon,waste treatment
Sohdzfication/stablhzatlon (S/S) of hazardous waste is a c o m m o n treatment alternatwe for

many waste streams before placement m a secure landfill Figure 1 illustrates a typical solidi-
ficatlon/stabihzation process In the interest of economy, the typical product of S/S is a low-
strength concrete or pozzolanlc matenal that requires the least amount of S/S binder Current
U S Environmental Protection Agency (EPA) guidance requires that the solidified/stabilized
product have a minimum, 28-day unconfined compressive strength (UCS) of 50 psi [1]
A rapid, inexpensive test method capable of predicting the final strength of the sohdlfied/
stabdlzed waste product would streamline operations at waste treatment and disposal faoh-
ties Myers [2] suggested early measurement of the resistance to penetration using the cone
index test (CI) as an mexpenswe method for predicting the long-term strength development
properties of the sohdlfied/stabllized matrix
The data collected dunng an extensive study of the effects of mterfenng materials on the
final physzcal and contaminant release characteristics o f a sohdlfied/stabihzed waste provided
an opportunity to evaluate the merits of a short-term, cone penetrometer test for predicting
the 28-day UCS of the treated waste. This paper descnbes the correlation between the CI and
the UCS and discusses the problems associated w~th using the CI for quanUtatwely predicting
28-day UCS

Synthetic Sludge Production
The metal sludge was produced by preop]tatlon of metal hydroxides from a solution of
reagent-grade mercury nitrate [Hg(NO3)2 H20], chromxum nitrate [Cr(NO3) 3 9H20], cad-
m m m mtrate [Cd(NO3)2 4H20], and nickel nitrate [Nl(NO3)2 6H20] The synthetic metal
solution was prepared by dmsolvmg 23 1 g/L of Cr(NO3)3 9H20, 14 9 g/L of NI(NO3)2 6H20,
Water
Waste
-I
_[ Mixing Solidified/Stabilized
r Product
FIG
T
Primary Binder
1--Schematic
T
Admixture
of typical sohdlficatlon/stabdlzatton process'
CULLINANE ET AL ON PENETRATION TESTING 387
1 6 g/L ofCd(NO3)2 4H20, and 0 02 g/L of rig(NO3)2 H20 in tap water A lime dosage of 20
g/M produced a sludge with desirable settling characteristics and a supernatant with the lowest
turbidity This lime dosage was twice the calculated stochiometnc amount required for metal
precipitation The sludge contained approximately 9% solids and had a density of 1 08 g/mL
and a p H of l l 5
The sludge was withdrawn from the settling vessel and dewatered using a rotary drum vac-
u u m filter The solids content of the dewatered sludge ranged from 30 to 35% by weight Before
use in the sohdlficatlon/stabihzation studies, the dewatered sludge was diluted with tap water
to a constant 25% ( _+0 5%) by weight solids content This procedure allowed ngld control of
the water-to-blnder ratio in the solldified/stabfllzed waste
Selectzon of Interference Chemicals

Several categones of substances that might cause decreased strength and their constituent
contaminant were selected for study [3,4] Interfenng chemicals were added to freshly solidi-
fled/stabilized sludge at 2, 5, and 8% by weight A control specimen, to which no interfering
material was added (the 0% interference addition specimen), was also prepared The categories
of interferences, the interfering agent used, and the interference agent weight correction factor
are listed in Table 1
Sludge Treatment and Bmder Selection

Three generic binder systems, representing the basic technology of the majority of com-
mercially available S/S processes, were selected for study [1] Binder-to-sludge ratios, which
were formulated to produce a 28-day UCS of at least 100 psi, are shown in Table 2 One inter-
ference chemical was evaluated for all three binder systems using a single batch of sludge The
heavy metal sludge used in this study was consistent in composition and age and was free of
unknown and unwanted contaminants
Test Methods
The 28-day UCS was determined on three separate samples using ASTM Test Method for
Compressive Strength of Hydrauhc Cement Mortars (Using 2-1n or 50-mm Cube Specimens)
TABLE l--Description of interference agents used m the study
Weight Correction
Desired Interference Interference Reagent Factor~
Oil 30 weight motor oll 1 00

Grease Axle grease 1 00
Phenol Phenol 1 00
Degreaser Tnchloroethylene (TCE) 1 00
Pestlode Hexachlorobenzene (HCB) 1 00
Strong Base Sodium hydroxide (NaOH) 2 35
Sulfate Sodmm sulfate (NaESO4] 1 48
Lead Lead nitrate 1 60
Copper Copper nitrate tnhydrate 3 80
Zinc Zinc mtrate hexahydrate 4 55
a Amount of reagent to be added to get desired concentraUon of interference material, for example,
1 6 gbyofASTM
Copyright leadInt'l
mtrate contains
(all rights 1 g Feb
reserved); Mon of 24
lead
23:07:03 EST 2020
TABLE 2 - - B m d e r systems and rattos used m thts study

Dllutmn
Binder Binder/Sludge Ratm Factors
Portland cement (Type I) 0 3 1 cement sludge 1 3~ 5 2b

Fly ash (Class F)/portland 0 2 1 cement sludge and 0 5 1 fly ash sludge 17 68
Cement (Type I)
Lime/fly ash (Class C) 0 5 1 fly ash sludge and 0 3 1 hme sludge 18 72
a Weight of final mixture/weight of wet sludge
b Weight of final mixture/weight of dry sludge sohds
(C 109-86). The only devmtion from this method was vibration of the specimen to remove
voids.
The CI measures resistance to penetration and was determined using either a standard cone
penetrometer or an airfield penetrometer as described in Corps of Engineer TM 5-530 [5]
Both procedures employ apphcaUon of a 30 ~ right o r e u l a r cone to the treated waste material
Specimens of each treated waste were subjected to CI testing at three to five time intervals
ranging from one to twenty days of cure The CI was determined for quadruphcate specimens
at one day of cure and at variable times thereafter until the values were outside the range of
the instrument ( > 750 psi) For consistency, linear regression analysis was used to estimate the
CI at five and ten days of cure These CI values were then correlated with the values for 28-
day UCS
Discussion
C I as a Predzctor o f U C S
The development of short-term tests that would accurately predict the ultimate strength of
solidified/stabilized waste would be of great use to the waste treatment industry The quick,
reproducible CI test method has been suggested as a rapid, inexpensive method for predicting
the long-term properties of sohdxfied/stabihzed waste [2]
The relationship between UCS and CI was evaluated using the pooled UCS and CI data
from the above described study [3]. This pooled data set includes the mean values for all spec-
imens, including both the control specimens and those to which lnterfenng agents were added
The data collected were analyzed to determine the ability of the CI measurements to predict
28-day UCS To compare CI data with the 28-day UCS data, a best-fit regression hne was
determined for the pooled data The values of CI for one, five, and ten days of curing were
compared to the 28-day UCS value
The regression equations, "best-fit" regression line, and the 95% confidence intervals
describing the relationship between CI and UCS for the three binder systems are shown on
Figs 2 through 6 Figures 2 and 3 present the 28-day UCS plotted as a function of the one-day
CI. Figures 4 through 6 present the 28-day UCS plotted as a function of the five-day CI. The
scatter m the data produces moderately wide confidence intervals and low correlation
coefficients.
The correlation coefficients (r 2) are listed m Table 3 The correlation coefficients are rela-
tively low (near 0 5) because of the high vanablhty between the d~fferent S/S batches; however,
the large number of measurements results m highly slgmficant coefficients (slgmficantly dif-
ferent from zero at the 99% level of confidence) These slgmficant correlation coefficients for
the one-, five-, and ten-day CI values support a link between the early CI and 28-day UCS
CULLINANE ET AL ON PENETRATIONTESTING 389
200 m"~"~'' ~
~d ,~ W ~ m " " [] " ...."

It [] 9 ~
H
.,,,. ,,," ~ REGRESSION ANALYSIS
9., " " "" UCS28= 67.6 + (0.34 C l 1)
-100 I I r I I
I()0 2OO 3~)0
CONE INDEX (psi)

FIG 2--Predlcted one-day C/ versus the 28-day UCSfor the cement sohdtfied/stabthzed samples, with
the regresslon hne and the 95% confidence tnterval
1200
95% CONFIDENCE
~ INTERVAL
~ ~ .j ~ ~ *'-" m ~ ~ ~ m ~
REGRESSION ANALYSIS
800
IN 9
am Hi [] 9
iN 9
6OO
[] m [] [] LINEAR REGRESSION LINE
9 U
9 [] 9
[] 9
20 40 60 80 100 120 140
CONE INDEX (psi)

FIG 3--Predtcted one-day CI versus the 28-day UCSfor the hme/fly ash sohd~ed/stabdtzed samples,
with the regresston hne and the 95 % confidence mterval
~2oo
REGRESSION ANALYSIS
UCS=e=351 * (1.16 Cll)
lo00
95% CON~DENCE iNTERVAL __ .,. __ ~ . . . . 2"
9 In 9 Ii 9
"~ SO0 9 9 9
tO .m
B 9
m
95%cONROENCE INTERVAL - - --
0 20 40 60 BO 100 120
CONE INDEX (psi)
FIG 4--Predicted one-da~, C1 versus the 28-day UCS for the cement~fly ash sohdtfed/stabdzzed sam
ples, wtth the regresston hne and the 95% confidence interval
- - v
800 ~ ~" Q
.. o
U
=~ 400 t~"" u o ~ ~ ~176 ~ I" t l
~Q d= . . . . ' , ~ I~176
100 .~P 9
1 [I~_B_E_SSI ON ANALY,S_~
o o= -" / |UCS.= 100 + 1. l ( C l . )
0 '" I ~ j " ~ l ' ~ . . . .
0 200 400 600
C O N E I N D E X (psi)
FIG 5--Predwted five-day CI versus the 28-day UCSfor the cement~fly ash stabEzzed/sohdlfied sam-
ples, wtth the regression hne and the 950/8confidence interval
I
I
800 . 11 1
/ _ / /
I I [] '~ ~' / LINEARREGRESSION LINE I
ta
1 []
500. I u " ~ e ~
o~
0 /// [][] /
400 D [] f~f"
2oo [] .. <_. ;o,*'-

~ []
[] [] [] ~ r R E G R E S S I O N ANALYSIS
o [] , ~ ~
L U C ~ = 79.6 + 1.43(C~)
0 200 400
CONE INDEX (psi)
FIG 6--Predicted five-day CI versus the 28-day UCS for the hme/fly ash sohdlfied/stablhzed samples,
wtth the regresston hne and the 950/o confidence mterval
Because of the wide variety of effects of the interfering agents on the setting and strength
development properties of the sohdlfied/stablhzed products and the inherent differences in the
setting and strength development properties of the three binder systems studied, this study Is
especially rigorous for comparing early CI values with the 28-day UCS Some ofthe mterfenng
agents caused rapid setting, for example, sodium hydroxide (NaOH), while others delayed or
prevented setting altogether, for example, metal nitrates at high concentrations
A major cause of the variation between UCS and CI measurements appears to be due to
differences m measured strength between the different batches for each binder system In fact,
analyzing the control samples alone (those w~thout mterfenng agents) results in similar cor-
TABLE 3--Correlanon between one-, five-, and ten-day cone mdex (CI) and unconfined compresstve
strength (UCS) for each binder system a
Correlation Coefficient Between 28-Day UCS and
Binder System 1-Day CI 5-Day CI 10-Day CI
CEM 0 53 NSb 0 45
CFA 0 52 0 53 0 51
LFA NSb 0 51 0 49
a Pearson correlation coefficients calculated on means All correlation coefficients listed are sxgmfi-
cantly different from zero at the 99% confidence level
Copyright
b NSby ASTM
= not Int'lsignificant
(all rights reserved);
at 99%Mon Feb 24 23:07:03
level EST 2020
of confidence
relation coefficient values of around 0 5. Furthermore, the large differences m setting and
strength development properties between specimens prepared with the different binder sys-
tems precluded poohng data from all three binders A fundamental problem m estimating the
28-day UCS or ultimate product strength based on short-term (one-, five-, or ten-day) pene-
tration tests appears to result from the fact that chemicals that cause delayed set or strength
development in the cementitlOUS product often result in long-term strength that is h~gher than
the control specimens Similarly, chemicals causing rapid setting or early strength develop-
ment (for example, NaOH) often result m products with reduced long-term strength.
These effects are illustrated by remowng the specimens with added N a O H from the pooled
data set Sodium hydroxide accelerates the set of the materials so that relatively high CI values
are obtained after one day However, at 28 days the effect of NaOH on strength development
is much smaller Correlation coefficients between UCS and the one-day CI for the cement-fly
ash data set, from which the NaOH-treated speomens were excluded, increase from 0.52 to
a more respectable 0 81 For the cement specimens the increase is from 0 53 to 0 60 On the
other hand, the correlahon coefficients for lime-fly ash specimens are not increased by remov-
ing the NaOH specimens from the data set
Conclusions
The development of short-term tests that would accurately predict the ultimate strength of
treated wastes would be of great use to the waste treatment industry
Although the correlations between 28-day UCS and CI are statistically significant, vanatlon
between different batches appears to hinder quantitative prediction of the 28-day UCS Resist-
ance to penetration, as measured by the CI, is shown to be a qualitative predictor of 28-day
strength development but not with a precision that would make it more than a broad indtcator
without further development
Contaminants that result in accelerated or delayed setting and/or early or late strength
development of the sohdlfied/stabdlzed waste preclude development o f a generahzed equation
for predicting strength from short-term resistance to penetration measurements
Acknowledgment
The tests described and the resulting information presented herein, unless otherwise noted,
were obtained from research conducted by the USAE Waterways Experiment Station, Vicks-
burg, MS The study was sponsored by the EPA, Risk Reduction Engineering Laboratory, Cin-
cinnati, OH Carlton Wiles was the EPA project manager Permission to publish this Infor-
mation was granted by the Chief of Engineers
References
[1] Culhnane, M J, Jr, Jones, L W, and Malone, P G , "Handbook for Stablhzatlon/Sohdxficatlon of
Hazardous Waste," EPA/540/2-86-00 l, Hazardous Waste Engmeenng Research Laboratory, Otfice
of Research and Development, U S Environmental Protection Agency, Cincinnati, OH, 1986
[2] Myers, T E, "A Simple Procedure for Acceptance Testing of Freshly Prepared Solidified Waste,"
Hazardous and Industrtal Waste Testing Fourth Sympostum, ASTM STP 886, J K Petros, Jr, W
J Lacy, and R A Conway, Eds, Amencan Sooety for Testing and Materials, Philadelphia, 1986
[3] Brlcka, R M and Jones, L W , "An Evaluation of Factors Affectmgthe Stabilization/Solidification
of a Heavy Metal Sludge," Final Report, Interagency Agreement DW930146-01-01 between Envi-
ronmental Laboratory, Waterways Expenment Station, U S Army Corps of Engineers, Vicksburg,
MS, and the Risk Reduction Englneenng Laboratory, U S Environmental Protection Agency, Cln-
cinnah, OH, 1990
[4] Jones, L W, "Interference Mechamsms In Waste SohdlfiCatlon/Stablhzatlon Processes Literature

Review," Intenm Report Interagency Agreement DW219306080-01-0 between Waterways Experi-
ment Station, U S Army Corps of Engineers, Vicksburg, MS, and Risk Reduction EnglneenngLab-
oratory of the U S EnvironmentalProtection Agency, Cincinnati, OH, 1988
[5] Materials Testing, Technical Manual TM 5-530, Office of the Chief of Engineers, U S Army, Wash-
mgton, DC, 1971
Gudlermo D. Del CuP and T. Mtchael Gdliam 2
Development of an In-Line Grout Meter for

Improved Quality Control
REFERENCE: Del Cul, G D and Gdham, T M, "Development of an ln-Lme Grout Meter for
Improved Quality Control," Stabthzatmn and Sohdtficatton of Hazardous, Radmacuve, and
Mtxed Wastes, 2nd Volume, ASTMSTP 1123, T M Gflham and C C Wdes, Eds, Amencan
Sooety for Testing and Matenals, Phdadelphm, 1992, pp 394-406
ABSTRACT" Data have been presented which show that electrical resistance of freshly prepared
cement-based grouts is hnear with respect to grout-mix ratio over a wide range The data serve
to estabhsh the proof of pnnople for apphcatmn to the real time measurement of m~x ratm ut~-
hzmg a meter based on the electrical resistance of the grout The data affirm that special care
must be maintained dunng design and installation to insure that the electrodes are properly
sealed
KEY WORDS: stabdlzauon, sohdlficatlon, quahty assurance/control, electrical resistance,

grout, waste form, cement
Stablhzatlon/sohdlficatlon (S/S) technology IS the most widely used technique for the treat-
ment and ultimate disposal of both radioactive and chemically hazardous waste Cement-
based products, commonly referred to as grouts, are the predominant materials of choice due
to their associated low processing costs, compatlblhty with a wide variety of disposal scenarios,
and ability to meet stnngent processing and performance requirements
It has long been recognized that there is a need for a meter to use with freshly prepared grouts
which would faclhtate improved quality control In the past, efforts in this area have not
proven successful due to the fact that the freshly prepared grout tended to cake on the monitor
probe, greatly reducing its effectiveness This paper documents progress to date on efforts at
Oak Ridge National Laboratory (ORNL) in support of the Westinghouse Hanford Co Grout
Technology Program to develop an in-line meter for mix ratio verification and, hence,
improved quality control in the Westinghouse Hanford Grout Treatment Facility (GTF)
Background
This section describes the expected behavior of a freshly prepared grout, hereafter simply
referred to as grout, when subjected to an alternating electrical field As such, it provides the
basis for the grout-meter development
Theory
When a dry-solids blend containing cement matenals is mixed with an aqueous waste to
form a grout, a series of chemical reactions begin to take place The reactions of the vanous
1Research scientist, Oak R~dge K-25 Site, P O Box 2003, Oak Ridge, TN 37831-7272
2 Manager, Stabd~zatxon/Soh&ficatlon Technology, Oak Ridge Natmnal Laboratory, P O Box 2003,
Oak Ridge, TN 37831-7273
394

DEL CUL AND GILLIAM ON GROUT METER 395
cementltiOUS phases proceed at different rates and involve both hydrolysis and hydration pro-
cesses All of the physlochemical properties are going to change while the system evolves from
the initial plastic state to the hardened state
When an electrical field is applied to a grout, some ions are free to drift, producing a con-
ductlon effect In addition, other charges that are bound to the solid particles and, thus, are
not free to drift, can oscillate under the action of an alternating electric field The electrical
behavior of this system can be modeled as a orcult with a resistive, R, and a capacitive, C,
element in parallel The Impedance, Z, of such a circuit is given by [1]
R
Z = (1)
(2 f C R)2+l
w h e r e f i s the frequency of the applied electrical field At a relatively high frequency (on the
order of 1 kHz), the electrical measurement on the grout is relatively insensitive to the move-
ment of the grout as a whole (e g, flow through a pipe) In addition, the fast alternating current
inhibits the electrolysis of the solution because any reaction induced during the first half of a
cycle will be reversed by the second half of the cycle
Apphcatmn to Grout-Meter Development

When the dry-solids-blend materials are mixed with the aqueous waste, there IS an lmtial
fast adsorption of the liquid on the surface of the solid particles Associated with this adsorp-
tion is the creation of an electncal double layer along the surface of the solid particles as well
as dissolution of soluble components and some hydrolysis This Initial stage, which is virtually
instantaneous, is followed by an interval of nearly constant behavior referred to as the dormant
period This dormant penod should result in a stable electrical response from an applied elec-
trical field for a period of several minutes to several hours, depending on the grout composition
and meter configuration
For a given meter geometry, the resistance of the system will rise with increases in mix ratio
(1 e , the ratio of dry-solids blend to waste) due to the growing fraction of volume occupied by
poorly conducting sohds (as compared with the liquid waste) In addition, the solid particles
can affect the movement of some of the free ions in solution The movement of some of these
ions in a direction other than parallel to the applied electrical field due to the solid particles
dispersed in the hquld phase will further increase the resistance of the grout
The capacitance of the system is due to polanzaton processes following the alternation of
the electrical field (1 e , a separation between centers of positive and negative charges) The
nature of the different polarization processes will strongly depend on the frequency of the alter-
natlng current At a frequency of 1 kHz, the two main processes are (1) the orientation of both
permanent and induced dipoles in the liquid phase in sympathy with the alternating electrical
field, and (2) polarization of the double layer adjacent to the surface of the solid particles As
the mix ratio is increased, the amount of liquid phase present will decrease and the capacitance
from Process 1 will dlmlmsh On the other hand, the total solid particle surface area will grow
as mix ratio increases, and the capacitance from Process 2 will increase
Meter Development
All data presented in this section utilized simulated 101-AW as the reference waste 10 l-
AW is a liquid, low-level radioactive waste presently being stored on the Hanford Reservation
Copyright
The bycomposition
ASTM Int'l (all rights
ofreserved);
simulatedMon Feb101-AW
24 23:07:03 EST 2020
is shown in Table 1 All grouts were prepared using a
TABLE 1 --Composttton of stmulated IO1-AW waste
Component Molanty
NaAlO2 0 46
NaOH 2 49
NaNO2 0 86
NaNO3 1 86
Na2CO3 0 25
Na3PO4 1412H20 0 10
Na2SO4 0 02
NaF 0 08
NaC1 0 16
dry-sohds-blend consisting of 47 wt % ASTM Class F fly ash, 47 wt % granulated blast furnace

slag, and 6 wt % Type I-II LA portland cement
Eqmpment Configuration
The configuration of the grout meter was designed based on two pnncipal criteria (1) the
monitor which would be considered a prototype, be compatible with the G T F (i.e., connecuble
to a plpehne), and (2) the electrodes be as unobtrusive as possible to minimize and hopefully
avoid grout caking during operation The first c n t e n o n was met qmte easily by requiring the
monitor to be flanged for mstallatmn to an existing pipehne In order to meet the second cri-
terion, several geometries were considered The one chosen was the use of electrically ~solated
stainless steel nngs made with the same materials as the p~pehne and configured such that
when flanged into the p~pehne, the nngs would be flush with the pipe Thus, the electrodes
would be completely unobtrusive
As discussed previously, upon application of an electrical field to the grout, some ions are
free to draft through the material, producing a conduction effect The electrical resistivity (the
reoprocal of the conduct~wty) is equal to the electrical resistance offered by the material to the
flow of current times the cross-sect~onal area of current flow per unit length of current path
Thus, the resistivity wdl be a function of the momtor configuration dimensions, such as length
between electrodes In order to address this issue, the monitor design also needed to allow for
variable electrode spacing, which became the third criterion
The design chosen to meet all three criteria is shown in Fig 1 The monitor was made of
PVC in the shape of a right orcular cylinder with a hole drilled through the center In the center
of the hole is a 1-m (2.54-cm)-wide lip with a nominal inside diameter of 2 m (5 08 cm) One-
third-inch-wide rings, either stainless steel or PVC, of nominal 2-m (5 08-cm) dmmeter, were
placed on either side of the hp to form a nominal 2-m (5 08-cm)-diameter "'pipe" through
whmh the fluid of interest will flow The metal rings serve as the electrodes or probes, while the
PVC rings serve as spacers Changing the dispositions of the metal rings and spacers readily
allows the spacing between electrodes to be vaned between 1 m. (2.54 cm) and 2 33 in (5.92
cm). Metal screws contacting the metal rings through the side of the momtor make the con-
nectlon with the Impedance meter A model IET IMF-600 impedance meter with a frequency
of 1 kHz was used to measure e~ther the resistance or capacitance between the electrodes The
momtor in combination with the impedance meter is hereafter referred to as the grout meter
The momtor was connected with flanges (with rubber gaskets between the monitor and
flanges) to a nominal 2-1n (5 08-cm)-diameter bench top pumping loop as shown in Figs 2
and 3. The loop consists of a plastic 10-gal (37.85-L) stirred holding tank, a Warren Rupp
Model
Copyright SB11/2-A
by ASTM Int'l (all rightsType 4 Mon
reserved); mr-powered
Feb 24 23:07:03 diaphragm
EST 2020 pump, and sufficient nominal 2-in. (5.08-
FIG. 1--Disassembled grout monitor (ORNL Photo 10796-88).
ORNL DWG 90A-124R
ITHERMOCOUPLE
• MIXER
MONITOR
I
\
DIAPHRAGM
PUMP
IMPEDANCE
METER ADC 'I
COMPUTER
Copyright by ASTM Int'l (all rights reserved); Mon Feb 24 23:07:03
FIG. EST 2020
2--Schematic of pumping loop.
FIG. 3--Pumping loop (ORNL Photo 1152-90).
cm) diameter stainless steel pipe to close the loop. A Type K thermocouple was inserted
upstream of the monitor to record grout temperature. A data acquisition board Metrabyte
model DAS-8 and IBM-AT personal computer were used to continuously record temperature
and electrical variable of interest (resistance or capacitance).
Evaluation of Electrode Spacing

For a given waste liquid and variable mix ratios, the expected range of the capacitance or
resistance will vary between the values for the pure liquid (mix ratio of zero) and the values
for the hypothetical mix ratio of infinity (pure dry-solids blend). In order to evaluate the sen-
sitivity of electrode spacing to electrical response, capacitance and resistance were measured
at three electrode spacings (chosen arbitrarily) for various dilutions of simulated waste.
The three electrode spacings were obtained by rearranging the spacers and tings as shown
TABLE 2--Labeling of various monitor configurations.

Configuration Electrode Spacing, in. Designation
Ring-spacer-spacer-lip-spacer-spacer-ring 2.33 (5.92 cm) O/O
Ring-spacer-spacer-lip-spacer-ring-spacer 2.00 (5.08 cm) O/M
Spacer-ring-spacer-lip-spacer-ring-spacer 1.67 (4.24 cm) M/M
ORNL DWG 90A-81
I I I I I I I I I
I
,56 9 MONITOR -
CONFIGURATION -
32 9 M/M -'
~., 9 O/M -
E 2 8 - 0/0
.20 .-
24 2
(D
Z
<l:
~-
m 20
c~ 1 6
" - - "
12 ~A~ ,9
~-T
- 9 '----"--" 9
,.A ,A "--''-" A
08 I t I I t t I t I
0 20 50 40 50 60 70 80 90 1 O0
WASTE CONCENTRATION (vol %)

FIG 4--Electrtcal response (resistance) normahzed to 20~ as a functton of time for selected dtlutlons
o f reference waste
in Table 2 Measurements were taken on various waste dilutions with results shown in Figs 4
and 5 In the figures, a waste concentration of 100 vol% corresponds to a liquid of composition
shown in Table 1 A waste concentration of 10 vol% corresponds to a waste consisting of 90
vol% distilled water and 10 vol% hquld of composition shown in Table 1
As shown in Figs 4 and 5, at approximately 70 vol% waste concentration, the electrical
response reaches a m a x i m u m for capacitance and a m i n i m u m for resistance At approxi-
mately 70 vol% waste concentration, the liquid becomes saturated and higher waste concen-
trations result in the appearance of undissolved solids The increasing fraction of undissolved
solids m suspension at h~gher concentrations behaves m a manner similar to that expected by
the addition ofdry-sohds blend (i e , the electrical response is dominated by the liquid fraction
present which, in turn, IS controlled by the mix ratio)
The derlvatwe of the electrical response w~th respect to waste concentration as a function of
waste concentration is shown in Figs 6 and 7 The denvatlves or slopes of the electrical
response versus waste concentrations can be used to evaluate the sensitivity of the electrical
response (1 e , the higher the slope, the greater the sensltlVRy of the electrical response to
changes m waste concentration) Inspection of Figs 6 and 7 m&cates that of the three config-
urations evaluated (1) configuration M/M (the shortest distance between electrodes) exhibited
the highest sensitivity with respect to capacitance, and (2) configuration O/O (the longest dis-
tance between electrodes) exhibited the highest sensitivity w~th respect to resistance
The data would suggest that if both capacitance and resistance were to be measured simul-
taneously, a four electrode probe would be o p t i m u m - - t w o electrodes with spacing O/O for
resistance determinations and two with spacing M/M for separate capacitance measurements
ORNL DWG 9 0 A - 7 9
90
~ u ~ m-m m--,---u~ I
80
7O
~-~ 60 - /= o~ a ~ "_A
I,I
(,3
Z
50
I.--
(.3
< /'/ / MONITOR -
a_ 40
< / / / CONFIGURATION
(.3
//,,-" 9 M/M
30
~'/ 9 O/M
20 / ,, o / o _
10 I I I I I I I I
0 20 50 40 50 60 70 80 90 100
FIG 5--Electncal response (capacltance) normahzed to 20*C as a functton o f tlme for selected ddutlons
o f reference waste
ORNL DWG 9 0 A - 8 0 R
0 ! I ' I ' I ! I ' I i ~
MONITOR CONFIGURATION ~ =""--~J"="~':

-0.02 ~-- M/M ..,,,:~-
,_. - ..... o/M ..,,r
o -0.04 ...... o/o ~ . "
-0.06
W
-0.08 /2.-2"
-0.10
N
Lt.I -0.12
-0.14
~
r~ A/)"
iT, - O. 16 //,,'/
/ /
-0.18 2"
--N.QC}___ i I I I I I ~ i I i I i I i I
10 14 18 22 26 30 34 38
FIG 6--Sensitivity o f reslstance measurements normahzed to 20~ as a function o f selected dilutions
of reference
Copyright waste
DEL CUL AND GILLIAM ON GROUTMETER 401
ORNL DWG 90A-82R

I I I I
3.0 MONITOR CONFIGURATION: -
~" 2.6 . . . . . . O/M

-', \ . . . . . . o/o
_ \\
~" 2.2
1.8
~ 1.4 _
rY
0.6
0.2 I I I I
10 20 30 40 50 60
FIG 7_Senstt1vt•y •f capact•ance measurements n•rmahzed t• 2•~ as a funct••n •f se•ected d1•utt•ns
of reference waste
The configuration O / M was used for the remainder of th~s development effort As shown m
Figs 6 and 7, this configuration gave the medmn sensmvity for both capacitance and
resistance
Grout Preparation
At the beginning of each experiment utd]zmg grout, 3 gal (11 4 L) of reference waste was
placed m the stlrnng tank with the stirrer m operation. The llqmd was circulated through the
p u m p loop for about 5 m m prior to the addition of dry-sohds-blend material The dry-sohds-
blend material was added m stages with the resulting group circulating through the p u m p loop
for approximately 5 m m prior to the introduction o f a d d m o n a l dry-sohds-blend material The
dry-sohds blend was blended for 23 h m a 3-ft3 (0 085-m 3) V-blender prior to addmon to the
waste
Mix Ratio as a Function of Resistance
Grouts were prepared by four sequentml addmons ofdry-sohds-blend material to the waste
to obtain mix ratios of 2 5, 5 l, 7 8, and 10 lb/gal (0 3, 0 61, 0 93, and 1 20 kg/L) designated
as Run 1 Resulting res]stance data are shown m F]g 8
As seen m Fig 8, the response of the detector ~s quite rapid In some instances, there ~s a
sharp m]tlal peak m the response dunng the a d d m o n of the grout Th~s response Is due to the
fact that the grout matenals can enter the p u m p loop before homogenization takes place The
peak qmckly disappears after mixing takes place, which is important because ~t shows that the
meter
Copyright can detect
by ASTM even
Int'l (all rights minor
reserved); temporary
Mon Feb changes
24 23:07:03 EST 2020 m homogeneity
ORNL DWG 90-92
I ~ I i I i I i
2.2
2.0
C-
O
I
1.8
ba
C)
Z
<{
F--
U3
J
(/3
1.6
i,i
Qf
1.4
MONITOR CONFIGURATION:
O/M
1.2 z I I I i I i I i
0 0.2 0.4 0.6 0.8 .0

TIME (h)
FIG. 8--Electrical response (resistance) normalized to 20~ as a function o f time for Run 1.
ORNL DWG 90A-83
~
I I I I I I I I I I
10 MONITOR CONFIGURATION: ~o
o/M
~-, 8
o
n,-
x 4
2
.J
[ ~ i I i I i I i I
I.2 1.4 I.6 1.8 2.0 2.2
RESISTANCE (ohms)
FIG. 9--Grout m i x ratio as a function o f resistance normalized to 20~ for Run l.
TABLE 3--Comparison of calculated and actual mtx ratios for

Run 1
Mix Ratto, lb/gal b
Resistance, a ohm Actual Calculated
1 206 0 0 1
1 445 25 25
1 695 5 1 50
1 982 78 78
2 207 l0 0 10 0
a Normahzed to 20~
b To convert to umts of kg/L multiply by 0 1198
The mix ratio was plotted against the resistance (from Fig 8) as shown In Fig 9 Also shown
is a plot o f the equation obtained by linear regression o f the data pairs
M= 996R- 119 (2)
where
M = mix ratio, lb/gal (to convert to units o f k g / L multiply by 0 1198), and

R = resistance, o h m s
E q u a t i o n 2 was then used to calculate the mix ratio based on the electrical response A com-
parison o f the calculated values with actual values o f mix ratio is shown In Table 3 The linear
equation results in calculated values within + 0 1 lb/gal ( + 0 01 kg/L) of the actual mix ratios
Data Reproductbthty
In an effort to assess the reproducibility of the measurements, another run was performed
at conditions similar to R u n 1, but with five sequential additions ofdry-sohds-blend materials
to obtain mix ratios of 6, 7, 8, 9, and 10 lb/gal (0 7, 0 8, 1.0, 1 1, and 1 2 kg/L) designated as
R u n 2 Mix ratio as a function of resistance is shown in Fig 10 The equation obtained by
linear regression of the data pairs is
M= 962 R- 123 (3)
As with R u n 1, this hnear equation results in calculated values within _+0 1 lb/gal ( + 0 01 kg/
L) o f the actual mix ratios (see Table 4)
The two equations are quite similar, but m i n o r differences do exist These m i n o r differences
are thought to be the result of variations in waste composition The 3-gal (11 4-L) simulated
waste sample used for each run is taken from a larger 55-gal (208-L) d r u m sample As the
simulated waste contains undissolved solids, some composition differences m a y exist upon
subsampllng
Equtpment Modification
Several experiments were performed with the m o n i t o r being dismantled and cleaned fol-
lowing each run Reproducibility was poor, and duplicate runs were inconsistent It was
observed that the rubber gasket did not m a k e a proper seal on the outermost electrode As
ORNL DWG 9 0 A - 8 4
I ~ ! I 1 I I I I I
~,
-6
10
8
MONITOR CONFIGURATION
0/M
.//7 0
~ 6
O
F--
<
cF
X
4
0 i I i I t
1.2 14 1.6 18 20 22 24
RESISTANCE (ohms)
FIG lO--Grout-mtx ratto as a functton of resistance normahzed to 20~ for Run 2
discussed prewously, the configuration used m this effort has one electrode in the outermost
position o f the m o n i t o r and the other surrounded by two PVC nngs Lack of a proper seal on
the o u t e r m o s t n n g by the rubber gasket would allow the electrode area to vary slightly d u n n g
each run, which would affect the electncal response measurements by altenng the electncal
field Consequently, the m o n i t o r was modified to i m p r o v e the seal on the outermost electrode
This modification consisted o f reducing the width o f one o f the PVC spacers between the out-
ermost electrode and lip by ~6 in (0 16 cm) and fabrication of an ~6-1n PVC spacer to be placed
between the o u t e r m o s t electrode and the rubber gasket (1 e , the new m o m t o r configuration is
1~6-1n spacer-rang-spacer with width reduced 'A6-in. spacer-hp-spacer-rlng-spacer) This modi-
fication would ensure that the only area of the electrode nngs exposed to the grout would be
the internal face in contact w~th the grout.
TABLE 4--Comparison o f calculated and actual mtx ratios for

Run 2
Mix Ratio, lb/galb
1 285 00 0 1
1 877 60 58
2 000 70 70
2 106 80 80
2223 90 9 1
2319 100 100
a Normahzed to 20~
Universidad Industrial de Santanderb(Universidad
TO convert to units
Industrial of kg/L
de Santander) multiply
pursuant to Licenseby 0 1198No further reproductions authorized.
Agreement.
ORNL DWG 90A-85

i ~ i i I l I l I
10 MONITOR CONFIGURATION fe -
O/M MODIFIED fO
fe
.-~
-6
_Q
8
of
.I,.I"/"
m
Q/
o 6
rF
x 4
0 I I i I
1.2 1.4 1.6 1.8 2.0 2 2
RES)STANCE ( o h m s )
FIG 1 l--Grout-mix ratio as a function of reststance normahzed to 20*C for Run 4
The new configuration was tested by preparing a grout with an m m a l mix ratio of 6 lb/gal
(0.7 kg/L) and then sequential additions ofdry-sohds-blend materials to increase the mix ratio
to 10 lb/gal ( 1.2 kg/L) in steps o f 0 5 lb/gal (0 06 kg/L) and is designated R u n 4 Linear regres-
sion on the resulting data, shown m Fig. 1 l, resulted in the following equation
M= 965 R-- 11.5 (4)
TABLE 5--Comparison of calculated and actual mtx rauos for

Run 4
Mix RaUo, lb/gal b
1 216 00 0 1
1 810 60 59
1 862 65 64
1913 70 69
1 972 75 74
2027 80 80
2 096 85 86
2 142 90 9 1
2 196 95 96
2 243 10 0 10 1
a Normahzed to 20"C
Universidad Industrial de Santanderb(Universidad
To convert to umts
Industrial of kg/L
de Santander) multiply
pursuant byAgreement.
to License 0 1198No further reproductions authorized.
The reduction m the spacing between the electrodes is thought to account for the differences
m slope of this equation in comparison to Eqs 2 and 3 A comparison between actual and
calculated mix ratios using Eq 4 is shown in Table 5 Calculated and actual values are in agree-
ment to within _+0 1 lb/gal ( + 0 01 kg/L)
Summary and Conclusions

Data have been presented which show that electncal resistance of freshly prepared grouts
containing simulated 101 -AW waste is hnear with respect to grout-mix ratio over a wide range
of mix ratios The data serve to establish the p r o o f o f p r l n o p a l for application to the Westing-
house GTF. The data affirm that special care must be maintained during design and installa-
tion to ensure that the electrodes are properly sealed
The data clearly indicate that the grout meter has the potential for real time measurement
of mix ratio and may offer quahtatlve indications of homogeneity Thus, the meter may serve
as an important quality assurance/control tool Future development efforts will concentrate
on assessing the effects of waste composmon and temperature Prior to implementation in the
field, a larger data base will be reqmred m order to establish the sensmvlty of measurements
to grout characteristics in a statistically sound manner
Acknowledgment
This research was performed for Westinghouse Hanford Company by Oak Ridge National
Laboratory (ORNL) ORNL is managed by Martin Marietta Energy Systems, Inc, for the U S
Department of Energy under contract DE-AC05-84OR21400
Reference
[1] McCarter, W J andCurran, P N, MagazlneofConcreteReseamh, Vol 136, 1984, pp 44-49
Large-Scale Evaluation or Demonstration
E d w i n F. Barth L
Summary of Solidification/Stabilization SITE

Demonstrations at Uncontrolled Hazardous
Waste Sites
REFERENCE Barth, E F, "Summary of Solidification/Stabilization SITE Demonstrations at
Uncontrolled Hazardous Waste Sites," Stabdtzatlon and Sohdlficatton of Hazardous, Radto-
acttve, andMtxed Wastes, 2nd Volume, ASTMSTP 1123, T M Gflham and C C Wiles, Eds,
Amencan Sooety for Testrng and Matenals, Philadelphia, 1992, pp 409-414
ABSTRACT Four large-scale solldlflCatlon/stablhzatlon demonstrations have occurred under

EPA's SITE program In general, physxcal testing results have been acceptable Reduction In
metal leachability, as determined by the TCLP test, has been observed Reduction m organic
leachablhty has not been determined since only low levels of orgamcs have been leached from
the raw waste utlhzmg the TCLP test
KEY WORDS: sohdlfiCatlon, stablhzatlon, metals, lead, orgamcs, PCBs
In response to the Superfund Amendments and Reauthorlzatlon Act of 1986 (SARA), the
United States Environmental Protection Agency (EPA) has established a formal program to
accelerate the development, demonstration, and use of new or innovative technologies that
offer permanent, long-term cleanup solutions at Superfund sites This program is called the
Superfund Innovative Technology Evaluation (SITE) program and is administered by the
Office of Solid Waste and Emergency Response (OSWER) and the Office of Research and
Development (ORD)
The SITE program has four primary goals
1 To conduct demonstrations of the more promising innovative technologies to establish

reliable performance and cost information for site characterization and cleanup decision
making
2 To identify and remove impediments to the development and commercial use of alter-
nate technologies
3. To develop procedures and policies that encourage the selection of available alternative
treatment remedies at Superfund sites as well as other waste sites and commercial
faclhtles
4 To structure a development program that nurtures emerging technologies
Several sohdlfiCatlon/stablhzatlon (S/S) processes have been accepted for detailed evalua-
tion in the SITE program These technologies are to be evaluated using actual waste from
uncontrolled hazardous waste sites Four vendor demonstrations involving full-scale process-
Environmentalengrneer, Center for Envxronmental Research InformaUon, U S EnvironmentalPro-

tecUon Agency, Cincinnati, OH 45268
409
Copyright9 1992byASTM International www.aslm.org

410 HAZARDOUSWASTES
lng units have been completed at the time of this report Several other demonstrations revolv-
ing S/S vendors are planned
Information from these demonstrations ~s important to the Superfund program for evalu-
ating the appropriate use and identifying limitations of th~s treatment technology Approxi-
mately 25% of the Records of Deosions for Superfund sites involved S/S in fiscal year 1988
Several of these sites contained both heavy metals and organic waste constituents The hml-
tatlons of this technology for treating high levels of metals or organics are not well
documented.
The results of the four S/S demonstrations are discussed in the following section
Results
A defendable set of testing methods and techniques (protocols) for evaluating the effective-
ness of S/S processes for treating hazardous waste needed to be developed for the SITE pro-
gram Several meetings between regulatory agencies (EPA, Department of Energy, and Envi-
ronmental Canada) and vendors were held to establish a baseline set ofevaluauon protocols
These protocols 2'3 needed to revolve evaluation parameters such as mixing effectiveness, sim-
ulation leaching test, and indicators of long-term durability An identified problem is that the
majority of these test methods have come from the geotechnical field or from the regulatory
waste testing arena for hazardous waste determination and have not been peer reviewed for
matenals treated by S/S processes Furthermore, acceptance cntena for these tests for S/S
materials have not been formally estabhshed by peer groups A key concern is that no one leach
test has been verified for evaluating the long-term leaching of S/S waste
HAZCON Demonstration
The H A Z C O N demonstration occurred in October 1987 at the Douglasville, PA site This
20 2-ha (50-acre) site was a former oll reprocesslng plant The site contained several areas of
soil contaminated with od and grease, volatde organic compounds (VOCs), base/neutrals
(BNA) [including polychlonnated blphenyls, (PCBs)] and heavy metals (primarily lead) At
the site, 3 8 m 3 (5 yd 3) from five waste areas and 19 m 3 (25 yd 3) from another waste area were
processed The vendor claimed to be able to stabilize organic compounds
Table 1 1s a summary of the results of the demonstration Reduction of the Toxloty Char-
actenstlc Leaching Procedure (TCLP) leaching test concentrations of lead were observed in
the treated waste, indicating immobilization No conclusions on the effectiveness of this pro-
cess for treating organic waste could be drawn from the VOCs and BNA TCLP data process
because their original values were low. Although no leachability index was calculated, Amer-
ican Nuclear Society test (ANS 16 1) data showed an increase m leaching concentratmns of
lead over specified time increments Unconfined compressive strength (UCS) data indicated
sufficient beanng strength Hydraulic conductivity values were low The volume of the solid-
lfied sod was double the original volume Weathered samples showed small weight loss
International Waste Technology Demonstration

The International Waste Technology (IWT) demonstration occurred April 1988 at the
Hlalejah, FL, site This demonstration also involved evaluating the use of in sltu mixing equip-
2 Solldlficatlon/Stabdlzatlon Protocol Development Workshop Summary, EPA, Newark, N J, 1989
(unpublished report)
3 Assessment of Sohdxficatlon/Stabdlzatlon Protocols for SITE Demonstrations, EPA, 1988 (unpub-
hshed
Copyright report)
BARTH ON SITE DEMONSTRATIONS 411
TABLE l--Results of H A Z C O N demonstration"
TCLP b Leacheate
Untreated Sod, mg/L Treated Soil, m g / L Hydrauhc

ConductwIty, UCS, e
Location VOC c BNA d Lead VOC BNA Lead cm/s Pa ( 106)
DSA 0 92 ND 15 0 38 ND 0007 1 8 • 10 -9 77
LAN 002 102 318 006 145 0005 4 0 ) < 10 9 36
FSA 103 281 179 072 279 040 84)< 10 -9 15
LFA 5 10 0 01 27 7 0 37 0 10 0 05 4 5 )< 10 -9 65
PFA 1 10 001 224 084 011 0011 5 0 ) < 10 -1~ 108
LAS 006 001 526 011 073 0051 2 2 ) < 10 9 61
HAZCON demonstration report [1]
b TCLP = toxloty characteristic leaching procedure
VOC = volatde orgamc compounds
d BNA = base/neutrals
e UCS = unconfined cornpresslve strength
m e n t C o n t a m i n a t i o n at t h e site was p r i m a r i l y h e a v y m e t a l s a n d PCBs T w o separate waste

areas, 3 0 b y 6 1 by 5 5 m (10 b y 20 b y 18 ft), were processed T h e v e n d o r c l a i m e d to be able
to stablhze organic c o m p o u n d s
T a b l e 2 is a s u m m a r y o f the results o f th~s d e m o n s t r a t i o n T h e effectiveness o f this process
for i m m o b i l i z i n g c o n t a m i n a n t s was difficult to ascertain because o f t h e low levels o f c o n t a m -
i n a n t s f o u n d m the raw waste T C L P leachate PCBs were n o t detected m the A N S 16 1 leach-
ates o f t r e a t e d waste H y d r a u h c c o n d u c t i v i t y values were low
T h e U C S d a t a r a n g e d f r o m 2 1 • l 0 6 t o 3.5 X 10 6 P a (300 to 500 psi) V o l u m e increase in
t h e treated soft was 8.5% W e a t h e r e d samples utilizing w e t / d r y cycles s h o w e d low weight loss,
however, f r e e z e / t h a w weight losses were large
Soliditech Demonstration
T h e Solldltech d e m o n s t r a t i o n o c c u r r e d m N o v e m b e r 1988 at the I m p e r i a l Oil C o m p a n y

site in N e w Jersey T h i s 2 0 - h a (5 0-acre) s~te i n c l u d e d facilities t h a t p r o d u c e d c h e m i c a l s a n d
r e c o v e r e d waste oil T h i s site c o n t a i n s several areas c o n t a m i n a t e d with VOCs, B N A s (includ-
ing PCBs), a n d h e a v y m e t a l s ( p n m a n l y lead) F r o m three waste areas, 3 8 m 3 (5 yds 3) were
processed in separate b a t c h o p e r a t i o n s T h e v e n d o r c l m m e d to be able to stabilize organic
compounds
TABLE 2--Results o f l W T demonstratton a
TCLP Leachate b
Hydrauhc
Untreated Soil, Treated Sotl, Conductivity,
Location PCB, c ug/L PCB, #g/L cm/s
B-6 12 0 < 10 1 84 • 10 -7
B-7 400 0 < 10 3 0 4 ) < 10 7
B-8 <1 0 <10 27 92 • 10 -7
a IWT demonstration report [2]

b TCLP =tOXlClty characteristic leaching procedure
Copyright by ASTM Int'l (all rightscreserved);
PCB =Mon polychlonnated
Feb 24 23:07:03 ESTblphenyl
2020
412 HAZARDOUSWASTES
TABLE 3--Results o f SOLIDITECH demonstration a
TCLP Leachateb
Untreated Soil, Treated Soil, Hydraulic UCS,C

Location Lead, mg/L Lead, mg/L Conductivity, cm/s Pa ( 106)
A (filter cake) 43 < 0 05 45 X 10 -7 27

B (filter cake/oil) 54 <0 05 89 X 10 -9 59
C (off-site) 05 <0 05 NAa NAa
a SOLIDITECH demonstrauon report [3]

b TCLP = toxicity charactensttc leaching procedure
c UCS = unconfined compressive strength
d NA = not analyzed
Table 3 is a summary of the results Reductmns of TCLP leaching test concentration of lead
were evident m that lead was not detected m the treated waste leachate, although imtlal leach-
lng levels were low Metals were not detected m posttreatment waste leachates produced by
the ANS 16 1 method Unexpectedly, posttreatment samples TCLP leachates contained
higher concentrations of three semlvolatfle organic compounds than the raw waste leachates
The effectweness of this process for immobilizing PCBs could not be determined because of
the low TCLP concentratmns m the raw waste Hydrauhc conductivity values were low The
UCS data indicated sufficient beanng strength Weathered samples showed low weight loss
CHEMFIX Technologies Demonstration

The CHEMFIX demonstrahon occurred In March 1989 at the Portable Equipment Salvage
Company, OR, site. This 0 81-ha (2.0-acre) site was a former metal salvaging operation The
site contained several areas contaminated primarily with lead, copper, and PCBs. From four
different areas, 7 6 m 3 (10 yds3) of waste were processed The vendor claimed to be able to
stabilize metal waste at high processing rates The treatment capacity of the continuous flow
system was 100 tons per hour
Table 4 is a summary of the results of the demonstration Substantial reductions of the
TCLP leaching test concentrations of lead and copper occurred Area C contained up to 14%
total lead There were no conclusions drawn upon the treatment effectweness for PCBs since
the initial TCLP concentratmns were low. The calculated leachable index from the ANS 16.1
test was 12 Hydrauhc conductivity values were low Data on UCS indicated sufficient beanng
strength Weathered samples showed low weight loss. The volume increase for treated exca-
vated waste ranged from 20 to 50%
Lessons Learned
Several problem areas for performing large-scale demonstrations and evaluating S/S treat-
ment data from uncontrolled hazardous waste sites were observed d u n n g these demonstra-
tions The following is a brief summary of general problem areas encountered d u n n g these
demonstrations
1 Debris (metal wire, brick) was encountered and needed to be separated physically from
the soils before treatment. Remaining debris of less than 25 4 m m (1 m ), such as lead
413
TABLE 4----Results o f C H E M F I X demonstration a
TCLP Leachate b
Untreated Soil, mg/L Treated Sod, mg/L Hydrauhc

Conductivity, UCS, Pa c
Location Lead Copper PCBa Lead Copper PCB cm/s (105)
A 610 45 <0 4 <0 5 0 57 <0 1 NAe NAe

C 880 12 NAe 25 0 54 NAe 6 4 >( 10-7 98
E 740 120 NAe 47 0 0 65 NAe NAe NAe
F 390 120 NAe 0 1 0 60 NAe NAe NAe
a CHEMFIX demonstration report

b TCLP = toxicity charactenst~c leaching procedure
c UCS --- unconfined compressive strength
a PCB = polychlonnated blphenyl
e NA = not analyzed
wire, may have impacted leach testing results, although leaching is generally influenced
by fine size particles
2 Regulatory leach test methods may not reflect site-specific risk assessments Leach data
should be coupled w~th groundwater models for quahtatlve risk assessments.
3 Lower contaminant levels m TCLP leachates of raw waste made performance difficult
to ascertain
4 Some demonstrations occurred at lower processing rates than the equipment was
designed to reduce the amount of treated material generated. This would reduce cahbra-
tlon time for the equipment
5 Preliminary samphng at the s~te may not have been consistent w~th concentrations from
the waste encountered during the demonstration, therefore binder matrix may not have
been optimized by the vendor. Heterogeneous waste m~xtures are c o m m o n on uncon-
trolled hazardous waste sites
6. Samphng methods such as conng may impact leaching results by altenng sample surface
or releasing volatfles from heat caused by friction from coring operations
7 Testing methods for evaluating m~xmg effioency not well developed
8. In sltu measurements of hydraulic conductivity and density are difficult to obtain, mak-
ing comparisons to treated values difficult
Conclusions
The H A Z C O N , Solldltech, and C H E M F I X processes were effective for lmmoblhzatlon
based on the reduction of TCLP leaching concentrations for heavy metals Site specific risk
assessments were not performed. Absolute conclusmns regarding the effectiveness of these pro-
cesses and the IWT process for orgamc waste lmmobdlzatlon cannot be made because of the
low levels o f T C L P leachate concentration in the raw waste The sohdlfied samples from each
demonstration exhibited low hydrauhc conductivity, adequate UCS, and good resistance to
wet/dry cycles. Comparison of lmmoblhzatlon performance between two demonstrauons Is
not recommended because of the different waste matrices and soil types processed at the dif-
ferent sites More development or verification work on a long-term leaching test for S/S waste
is necessary, as well as methods for determmlng mixing effectiveness.
414 HAZARDOUSWASTES
References
[1] "HAZCON Demonstration Report," EPA 540/5-89/001A, Environmental Protection Agency,
Office of Research and Development, Cincinnati,OH, 1989
[2] "IWT Demonstration Report," EPA 540/5-89/004A, Environmental Protection Agency, Office of
Research and Development, CmclnnaU, OH, 1989
[3] "SOLIDITECH DemonstraUon Report," EPA 540/5-89/005A, Environmental Protection Agency,
Office of Research and Development, Cincinnati,OH, 1990
[4] Environmental Protection Agency, EPA 540/5-891011A, CHEMFIX Technologies, Inc Sohdlfica-
tlon/Stablhzatlon Process, Office of Research and Development, Cincinnati,OH, 1990
Brian P. Spalding 1 a n d T h o m a s A. Fontaine ~
Field Demonstration of In Situ Grouting of

Radioactive Solid Waste Burial Trenches with
Polyacrylamide*
REFERENCE" Spalding, B P and Fontame, T A , "Field Demonstration of In Situ Grouting
of Radioactive Solid Waste Burial Trenches with Polyacrylamide," Stabthzatlon and Sohdlfi-
cation of Hazardous, Radioactive, and Mzxed Wastes, 2nd Volume, ASTM STP 1123, T M
Gllham and C C Wiles, Eds, Amencan Society for Testing and Matenals, Philadelphia, 1992,
pp 415-425
ABSTRACT. Demonstrations o f m SltU grouting with polyacrylamxde were carried out on two
undisturbed burial trenches and one dynamically compacted bunal trench in Solid Waste Stor-
age Area (SWSA) 6 The injection of polyacrylamide was achieved quite facilely for the two
undisturbed burial trenches, and the trenches were filled with grout at typical pumping rates of
95 L/mm m several batches injected over several days The compacted burial trench, however,
failed to accept grout at more than l 9 L/ram even when pressure was apphed Consequently,
burial trenches that have been staNhzed by dynamic compaction appear to have a permeability
too low to be considered groutable The water table beneath the burial trenches did not respond
to grout injections, Indicating a lack of hydrologic connection between fluid grout and the water
table, which would have been observed if the grout faded to set Because grout set times were
adjusted to less than 60 mln, the lack of hydrologm connection was not surprising Postgroutlng
penetration testing revealed that the grout had improved the stability of the bunal trenches from
further subsidence from only 26 to 79% of that measured in the undisturbed soil formation In
situ permeation tests on the grouted trenches have indicated a significant reduction in hydraulic
conductivity of the trench contents from a mean of 2 1 )< l0 3 to 1 85 )< l0 5cm/s Preliminary
observatmns indicate that grouting with polyacrylamlde is an excellent method for both
improved stability and hydrologic isolation of radioactive waste and its incidental hazardous
constituents
KEY WORDS: in sltu grouting, polyacrylamide, radioactive waste, hydraulic conductivity, pen-
etratlon resistance, stabilization, field demonstration
As p a r t o f a burial g r o u n d s t a b i h z a t l o n a n d closure t e c h n o l o g y d e m o n s t r a t i o n project, a

g r o u p o f 19 b u r i a l t r e n c h e s m O a k Ridge N a t i o n a l L a b o r a t o r y ( O R N L ) Sohd W a s t e Storage
A r e a (SWSA) 6 was selected as a d e m o n s t r a t i o n a n d test area These 19 t r e n c h e s are c o n t a i n e d
w i t h i n a hydrologically isolated area o f S W S A 6, and, for t h a t reason, a n y effects o f stabiliza-
t i o n actLvitles o n site p e r f o r m a n c e a n d g r o u n d w a t e r quality w o u l d be separable f r o m t h e influ-
e n c e o f o t h e r waste disposal u n i t s in S W S A 6. T o obviate t h e c h r o m c p r o b l e m o f burial t r e n c h
s u b s i d e n c e a n d to p r o v i d e s u p p o r t for a n m f i l t r a t m n b a n n e r cap, 5 o f the 19 t r e n c h e s were
d y n a m i c a l l y c o m p a c t e d in July a n d A u g u s t 1988 D y n a m i c c o m p a c t i o n was a c h i e v e d by
Research staff member and research associate, respectively, Oak Ridge National Laboratory, Oak
Ridge, TN 37831
* Research sponsored by the Office of Technology Development, U S Department of Energy, under
contract DE-AC05-84OR 21400 with Martin Marietta Energy Systems, Inc, Publication No 3648, Envi-
ronmental Soences Division
415

416 HAZARDOUSWASTES
repeatedly dropping a 3 6 Mg weight, with a 0.9 m 2 base, onto each trench from heights of 4
to 8 m The penetration resistance of these five trenches was extremely low prior to compac-
tion and was increased to a level equivalent to that of the undisturbed surrounding soil after
c o m p a c t i o n [ / ] Thus, dynamic compaction was found to be very effective In stablhzlng burial
trenches to the extent that no dafferennal land surface settlement should be expected to com-
promise the foundation support of an Infiltranon barrier
However, because of the large inventory of radioactivity in some burial trenches in SWSA
6, additional assurance of hydrologic isolation of trench contents from groundwater intrusion
and/or percolation, beyond that provided by an infiltration barrier, may be warranted for par-
tlcular trenches In SltU grouting is a technique for attaining hydrologic isolation of bunal
trench contents regardless of infiltration barrier integrity or performance Two in sItu grouting
field demonstrations have been completed previously at O R N L [2,3] Both of these demon-
stratlons involved putting polyacrylamlde grout in uncompacted burial trenches Both were
successful in changing the bunal trenches from a condition of extreme permeability (i e , 10 -2
cm/s) and facile water intrusion to a condition of unmeasurable permeabihty ( < 7 • 10 -6 cm/
s) The long-term stability of the polyacrylamIde grout has been established previously through
measunng low rates of microbiological decay [3] The inherent low permeability of neat poly-
acrylamide grout (i e , 10 ~0cm/s) has the potential to sequester radionuchdes to diffusion-
controlled leaching mechanisms, which is the subject of related investigations However, nei-
ther of the previous demonstrations addressed the problem of trench geotechnical stability,
nor, because of their limited scope, did they provide an adequate assessment of grout distn-
butlon and hydraulic properties within trenches These demonstrations did establish that the
cost of materials alone for In SltU grouting with polyacrylamlde was qmte h~gh, i e , about
$50 000 per typical (4- by 15- by 5-m deep) burial trench, due to the large amount of voids per
trench and the cost of the acrylamide grout materials (about $530 per m 3) Dynamic com-
paction, m contrast, is a relatively inexpensive method to eliminate many large voids within
a trench, which potentially might result in a smaller residual void volume to be grouted An
additional benefit of dynamic compaction would be its ability to collapse voids within waste
packages that may not be accessible during grout injection into an uncompacted trench
The objectives of the present demonstrations were to evaluate the effectiveness of grouting
in (1) reducing burial trench permeability, (2) Improving trench stability against further sur-
face subsidence, (3) avoiding groundwater contaminatton by the grout's toxic monomer, acry-
lamide, and (4) achieving hydrologic isolation of dynamically compacted trenches to the same
degree expenenced after grouting undisturbed burial trenches
Site Selection and Preparation

The site selected for the demonstration lies on a small hillock in the northeast corner of
SWSA 6 (Fig 1) The site was selected based p n m a n l y on the following two cntena ( 1) It was
away from operational areas within SWSA 6 and would, therefore, not interfere wlth dally
waste management activities or vice versa, and (2) it was located entirely on high ground and
was, thus, isolated hydrologically from any penpheral recharge areas that would complicate
performing a site water budget and performance monitonng as part of the stabilization and
closure evaluation The water table at the site is at least 9 m below the bottoms of the trenches,
which, therefore, remain unsaturated throughout the majority of the year in contrast to chron-
ically or seasonally inundated trenches in other areas within SWSA 6
Three trenches (6, 7, and 165) w e r e selected for the polyacrylamlde grouting demonstration
Residual grout was also injected into Trench 8, which because of partial sidewall collapse dur-
ing its construction showed hydrologic connection with Trench 7 These specific trenches were
selected
Copyright forInt'l
by ASTM demonstration because
(all rights reserved); Mon of their
Feb 24 23:07:03 proximity to several of the groundwater monitor-
EST 2020
SPALDING AND FONTAINE ON IN SITU GROUTING 417
EME!qGENCY
WASTE
•ASI N
L '~0
.// 84
844
eJ843
369
9 E WELL t'655
TRENCH
a3o
- - - - ROAD
- - -- SWSA 6 INTRUDER FENCE
....... CREEKS
_ _ oDD SURFACE CONTOUR

so cr
S C A L E
FIG. l--Location of demonstration site for in situ grouting at the Oak Ridge National Laboratory,
showing location of burial trenches, topography, and groundwater monitoring wells.
ing wells at the site such that any breakthrough ofacrylamide might be observed with minimal
migration. Trench 6 had been dynamically compacted in 1988, collapsing 66% of its water-
accessible voids [1]. The other trenches (7, 8, and 165) were undisturbed since their closure in
1972 to 1974. Leachate and bottom soil samples from each of these trenches had been tested
previously for compatibility with the polyacrylamide grout polymerization, and no interfer-
ences were observed [ 1]. Five grout delivery and/or trench fluid monitoring wells were inserted
into each trench. Each trench was also surrounded by six 4.5-m deep monitoring wells within
1.5 m of the known trench edges. These monitoring wells were placed in residual penetration
test holes, which were used to obtain penetration resistance measurements on the trenches and
surrounding soil formation. Monitoring wells within trenches consisted of 3.66 m of screen,
with 2.5-ram slots, which was threaded with flush-joints to 1.22 m of solid polyvinylchloride
(PVC) pipe. Monitoring wells outside trenches were identical except that screens with 0.5-mm
slots were employed. The larger slots were used for wells within trenches to avoid grout setting
between
Copyright slots
by ASTM Int'l (allspanned byMon
rights reserved); residual fluids
Feb 24 23:07:03 EST after
2020 daily batch injections. Well screen diameters
were either 3.8 or 3.2 cm depending on which size a particular hole would accept to its full
depth.
To avoid bringing highly contaminated soil and/or waste to the surface, which would have
resulted during a standard penetration test [ASTM Method for Penetration Test and Split-
Barrel Sampling of Soils (D 1586-84)], a nonstandard penetration test was devised that used a
63.5-kg (140-1b) drill-rig-mounted drop h a m m e r to drive a 4.4-cm-diameter drill rod into the
soil. The end of the drill rod was fitted with a 5.1-cm-diameter 60~ point, and the rod was
marked offin 30-cm lengths. At each nonstandard penetration site the rod was hammered into
the soil, and the h a m m e r blows for each 30-cm depth were recorded. This nonstandard pen-
etration is similar to the Dutch cone penetration test (DCPT) except that the test borehole is
not augered out every 46 cm (18 in.) to relieve drill stem friction. Blow count values (penetra-
tion resistance) obtained by the DCPT have been found equivalent to blow count values [4]
obtained with the standard penetration test.
Field Grouting Operations

Field grout operations consisted of mixing the grout and catalyst solutions as they were
pumped into a trench (Fig. 2). Solutions of grout and catalyst were made up in 5678-L (1500
gal) seamless polyethylene tanks. Each tank was fitted with a 5-cm (2-in)-diameter ball valve
at its lower access port. One side of the valve was connected to the intake of a portable gasoline-
powered centrifugal pump. A T-fitting with two valves was placed on the outflow of the pump
to regulate fluid flow either for recirculation to the tank's top access port, when grout com-
ponents were being dissolved in the tank, or for injection into a trench. The delivery lines from
both the acrylamide (grout) and catalyst tanks contained water meters to record volumes
injected and check-valves to prevent backflow and grout set within the hoses. The two lines
(consisting of 2.5-cm-diameter hose) were joined in a Y-fitting and the solutions mixed in a
2.5-cm-diameter by 43-cm-long static mixer containing an array of angled baffles (Cole-
Parmer, Inc.). A bleed valve was also fitted to allow for line drainage, if necessary, and for line
flushing at the end of each batch injection. The delivery line was then connected with a manual
shut-off valve, to the desired grout injection wellhead via quick-connect fittings to facilitate
FIG. 2--Arrangement offluid-handling equipment employed for mixing and injecting polyacrylamide
groutbyinto
Copyright ASTM radioactive
Int'l (all rights solid waste
reserved); burial
Mon Feb trenches.
24 23:07:03 EST 2020
switching of injection wells, if necessary The entire arrangement of pumps, tanks, meters, and
valves was placed within secondary containment consisting of a reinforced PVC liner with
sidewalls maintained by air-inflatable pillows (Aero Tec Laboratories, Inc )
A 10% acrylamlde grout formulation was selected for the field demonstration as had been
used in previous demonstrations Approximately 4500 L (1200 gal) of water from the emer-
gency waste basin to the north of the site (Fig. l) was pumped into the grout tank Fifty double-
lined bags, containing 22 7 kg (50 lb) of Q- l chemical grout (Ques, Inc ) each, were then emp-
tied into the tank followed by 30 L oftnethanolamlne and I 14 kg of potassium ferrocyanide
The tank was made up, as necessary, to Its capacity, 5870 L (1500 gal). In the catalyst tank,
previously filled with water, was placed 57 kg of a m m o n i u m persulfate The solutions in each
tank were then reclrculated until all solids were dissolved, usually in about 30 min The final
formulation, based on a 1 l mixture of the two solutions, would contain 10% acrylamide grout
(9 5% acrylamide plus 0 5% methylenebisacrylamide), 0 3% trlethanolamlne, 0.01% potas-
sium ferrlcyanide, and 0 5% a m m o n i u m persulfate Personnel prepanng the acrylamide grout
solutions employed personal protective equipment including disposable water-resistant suits,
double rubber gloves, and full-face respirators Personal air samplers were also worn by all
personnel to quantify exposure to possible fugitive dust emissions Samples of the solutions
were taken from each tank and mixed 1 l (v v), and the time for grout set was measured to
determine if the batch set time was less than the speclficatmn of 60 min maximum (Table 1)
Grout was delivered to the desired injection well by routing flow from both pumps through
the static mixer to the dehvery hose Grout dehvery rates vaned from 5 to 95 L/mln with the
lower flows occurnng during final batches into trenches after prevmus batches had sealed up
much of the available well screen Generally, 2 to 4 h were required for the delivery of each
batch One individual was assigned to monitor the flow meters of both grout delivery lines and
record cumulative flow readings at 10-min intervals Pump delivery rates were adjusted reg-
ularly to maintain a 1 1 mixture d u n n g dehvery During each batch dehvery, an "as-deliv-
ered" sample was taken from the line by interrupting flow at the wellhead and diverting it to
a sample container, these samples were also monitored for set time and later analyzed for resid-
ual acrylamlde, total solids, and ash-free solids [1] for verification of grout quahty (Table l)
D u n n g injection, one individual was assigned to monitor the mjectmn well for leaks or backup
flow around the well annulus A leak containment collar, made of 40-cm-hlgh steel sheet with
TABLE l--Polymerlzatwn and set ttmes of polyacrylamtde grout used for mjectwns mto burial
trenches
Batch InJection Set Time, Residual Grout Sohds, Degree of

Number Date mm Acrylam~de,ug/g % PolymenzaUon, %
1 28 Aug 89 21 18 1 104 99 98
2 29 Aug 89 23 13 7 14 6 99 99
3 30 Aug 89 23 54 6 82 99 93
4 31 Aug 89 34 11 4 148 9999
5 1 Sep 89 31 NAa NA~ NAa
6 6 Sep 89 34 10 9 14 5 99 99
7 7Sep 89 34 11 3 132 99 99
8 8 Sep 89 30 14 5 14 0 99 99
9 9 Sep 89 35 75 13 3 99 99
10 13 Sep 89 43 74 15 4 99 99
11 14 Sep 89 40 24 0 13 7 99 98
12 15 Sep 89 35 89 16 2 99 99
13 18 Sep 89 41 22 1 14 9 99 99
a NA
Copyright = not
by ASTM analyzed
420 HAZARDOUSWASTES
a beveled edge, was pushed into the ground around the injection well. Another indwidual was
assigned to momtor fired levels and set grout (1 e , the depth to well bottom) within the injec-
tion trench at 10-mm intervals during mjecUons until grout had set within all mtratrench wells.
Fluid levels were measured using an electrical depth sounder (SodInst, Inc.) Another mdwld-
ual was asstgned to monitor fired levels or confirm their absence m the six perimeter mom-
torlng wells surrounding each grouted trench and in wells within surrounding burial trenches.
Another mdtv~dual was asstgned to monitor water levels within the 15 permanent ground-
water m o m t o n n g wells at the site These readings were recorded hourly during rejection pen-
ods and were continued until grout had set m all wells, usually about 1 h after lnjectmn ended
Following completion of a batch delivery, approxlmatley 75 L of water was flushed through
each tank line to the trench by switching the pump retake to water from a third storage tank
A residual of approximately 500 L remained m each tank due to the height (about 10 cm) of
the port valve above the tank bottom. This residual volume was then included m the next
batch and resulted in the increasing sohds content of latter batches of grout. For the final batch,
the tanks were tdted when fluid levels were near the bottom so that the contents were com-
pletely emptied
Field Grouting Results

The total volumes of grout delivered to each trench are summarized in Table 2 The vol-
umes of vanous trenches were computed from their surveyed corner coordinates and their
nominal depth of 4 57 m (15 ft) The total grout volume dehvered was then expressed as a
percentage of the total volume of the trench Although none of these grouted trenches had theLr
void volume measured directly before grouting, previous water filhng tests on six other O R N L
burial trenches found water-accessible volds accounting for an average of 20% (range of 8 to
30%) of total trench volume [1,5] Thus, the grout capacity of Trenches 7 and 165 was con-
slstent with these previous measurements because the acrylamlde grout is very similar in fluid
properties to water Perhaps not too surpnslngly, Trench 6, which had been previously dynam-
ically compacted, accepted very little grout (1500 L). All five of its access wells eventually
refused grout delivery after Initially accepting small volumes, presumably due to access to
minor voids. Once these voids had been filled, further grout delivery was refused Thus, in situ
grouting, as a technique to reduce residual permeability after dynamic compaction, apparently
is not feasible and perhaps not necessary. Therefore, the decision to pursue trench stabilization
or trench hydrologic Isolation will need to be made prior to site closure
The response offlmds and set grout within Trench 7 is depicted m Fig 3 The concomitant
cumulative rejected grout volume is plotted with the intratrench fluid hydrograph to dem-
onstrate the time relationship between grout ,njectlon and fluid responses in other regions of
the trench The intratrench hydrograph was based on observations taken from monitoring
wells within the trench most distant from the point of initial grout injections but was typical
of all wells within both Trenches 7 and 165. Grout injections were started at the lower eleva-
TABLE 2--Polyacrylamtde grout mjectlon summary
Grout Delivered, Trench A r e a , Trench Volume, Fraction Grouted,

Trench Number m3 m2 m3 %
7 679 652 2981 228

165 41 1 29 6 135 5 30 4
6 (compacted) 15 57 6 263 2 06
8 (partial fill) 23 0 52 5 239 9 96
256 100
255
> 80
w
Z -- ~ - - _ j ~ . . . . . ~ . ........ E
254 i
> . . . . . . . . . . . . . . . . . . :
"o
o 60
J~
o
253
E -- Fluids c_
< !': > 40
- 4 - - Bottom oL
o
,-- 252
4.'*
al t.9
>
4) ....... Grout
Volume 20
[] 251
250 i i , 0
-10 0 10 20 30
Days from September 1, 1989

FIG. 3--The elevation of fluids and set grout in a monitoring we//within ORNL solid waste burial
Trench 7 during and after in situ grouting with polyacrylamide.
tion end of the trench, i.e., the northernmost access well. Although the slopes of the trench
bottoms are not known, the elevation of the bottoms of the northern access wells was about 1
m lower than those in the southern end. Thus, there was apparently no fluid response in the
southern access wells to the first batches of grout placed in the northern end of the trenches.
However, once the level of grout fluids had risen during later injections, fluid responses were
observed in direct and undelayed response to grout injections. This response illustrates the
facile hydrologic connections within burial trenches, which can be ascribed to their large voids.
The trench response to grout fluids was very similar to that observed during water filling tests,
where water elevations rise uniformly within the trench during filling [5]. Because of the rel-
atively short time lag for grout to set, i.e. about 1 h, only a few instances where fluid levels
exceeded set grout levels were observed during actual grout injection intervals. An apparent
occurrence ofunset grout fluids, persisting for several days within trenches, can be seen in Fig.
3. However, this occurrence was probably an artifact of the delivery line washing procedures.
When the delivery line was flushed with 75 L of water, this volume was delivered to the injec-
tion well, diluting the grout within the well casing so that the resulting fluid would not set.
However, the grout, which had been displaced around the well casings, circumvallated this
fluid, resulting in a persistent standing fluid within the casing. Support for this interpretation
is provided by the lack of persistent fluid levels in other access wells within the trench during
the same interval.
The response of the water table to the grout fluid injections is depicted in Fig. 4. Well 1 was
used as a reference well because it was distant from the three trenches that received grout (Fig.
1). Responses of other wells should be interpreted relative to this well to account for water table
responses to rainfall during the 22-day interval over which grout injections were carried out.
Although the water table was quite responsive to rainfall during the interval of grout injections,
it was
Copyright unresponsive
by ASTM to the Mon
Int'l (all rights reserved); grout injections,
Feb 24 particularly when normalized to elevations in Well
23:07:03 EST 2020
244 iiiiiiiiiiiiiiiiii~ Qro.t

Injections
Liiiiiiiiiiiiiiiiii 10
> Well 3 .....

c~
z 243
>
5- 8 A
O 7 E
Well 1
242
Well 4 ...... .c_
c-
5 qa
O
tr
~= 241 Well 7 4 >,
>
Well
L
6 ......
.....
_:--:5LS--:---5.2 5
9 I J
3
UJ
~ 240 9 2
Well 367
1
i .~ II d,_,
239 . . . . . . . . . . . i - - - - ,
-20 -15 -10 -5 0 5 10 15 20 25 30
Days from September 1, 1 9 8 9

FIG. 4--Groundwater elevations within permanent monitoring wells and rainfall at the demonstration
site during the period of polyacrylamide grout injections into burial Trenches 6, 7, 8, and 165. (See Fig. 1
for locations of wells relative to grouted trenches.)
1. On a daily and hourly basis, for which resolutions cannot be discerned from Fig. 4, none of
the proximate groundwater monitoring wells responded to grout injections. The time lag for
hydrologic connection of the water table to the trenches had been established in previous water
filling tests where changes in water elevations of several meters were observed within a day of
fluid placement within the burial trenches.
About a month after completing the grouting of Trenches 7 and 165, the nonstandard pen-
etration tests were repeated using five testing points per trench. In addition, further penetration
tests were completed about five months later as part of the installation of in situ permeameters;
this resulted in an additional 12 and 15 testing points in Trenches 165 and 7, respectively. Prior
to grouting, ten penetration tests were completed along longitudinal transects of each trench.
A total of 12 penetration tests were completed on the soil formation immediately surrounding
the grouted trenches as a reference for the ambient stability of the soil formation. Results of
all these penetration tests are summarized in Fig. 5. Prior to grouting, the inherent instability
of the burial trenches was apparent in that the total number of blows required to reach 4.6 m
(15 ft) was approximately one fourth that required within the surrounding undisturbed soil
formation. Following grouting, the cumulative number of blows to reach this same depth
within a burial trench was increased by a factor of three over pregrouting conditions, from 26
to 79% that of the undisturbed soil formation. This increase in penetration resistance is
ascribed to filling trench voids with polyacrylamide grout. Thus, grouting with polyacrylamide
should afford a considerable increase in burial trench stability concomitant with the reduction
in trench permeability.
Hydraulic conductivities of the waste burial trenches were determined before and after
grouting (with newly installed wells) using a constant head pump-in test [6]. The governing
equation for deriving hydraulic conductivity of the material surrounding a well point was
150
125
9 Before
Grouting ~ /~
100 ~" After _~" /
o.,,n0 y--
E
9 Outside .0/ ./ --
~
0. 75
0
N 5o
25
I I [ I
0
0 1 2 3 4 5
Depth below Surface (m)

FIG. 5--Penetration resistance of two radioactive solid waste burial trenches before and after in situ
grouting with polyacrylamide relative to the surrounding undisturbed soilformation.
q In {(mL/D) + [1 + (mL/D)2] m}
K = (1)
27rLHz
where
K = hydraulic conductivity in cm/s,
q = rate of water acceptance in cm3/s,
L = length of screened section in cm,
Hz = constant piezometric head in cm,
D = diameter of well casing in cm, and
m = transformation ratio, assumed to = 1.
Three field techniques were used to collect the pregrouting water acceptance rates to calculate
the hydraulic conductivity. In the first method, the time required for the well to accept 10 L
of water was measured. In most of the intratrench wells, no measurable water level could be
determined in the 25 to 40 s required to deliver this amount of water into the well casing. In
order to use Eq 1, the length of the intake section was assumed to be the entire slotted interval
of the well casing, and the piezometric head was assumed to be equivalent to that depth inter-
val. In the second method, the acceptance rates of the acrylamide grout were used to calculate
pregrouting hydraulic conductivities. In this case, the intake section was the entire length of
the slotted interval or, in wells not used for injection until later batches, the fraction of the
intake section that was free from set grout of previously injected batches. The piezometric head
was taken as the entire height of the well casing below ground, usually 4.27 m, plus the addi-
tional height of grout in the delivery tank. The third method used water acceptance rates
achieved in filling tests on Trenches 2, 3, 4, 5, 6, and 151; the piezometric head and intake
section were assumed to be the entire depth of uncased hole. The estimated hydraulic con-
ductivities by these three methods were very similar, resulting in a geometric mean of 2.10 •
10 3 c m / s , which was based on 35 independent determinations. The average hydraulic con-
ductivity of the undisturbed soil formation has been measured at 3.47 • 10 5cm/s in previous
studies [ 7]. This order of magnitude difference of two in the hydraulic conductivities of burial
trenches and the surrounding soil is a significant factor in the frequent occurrence of standing
or perched water within burial trenches at ORNL.
After grouting, in situ permeameters were inserted into penetration holes in Trenches 7 and
165 extending to depths of 1.8, 2.7, or 4.3 m below ground. However, only the bottom 60 cm
was screened with 0.5-mm slots, with the remaining length being solid pipe. Thirteen such in
situ permeameters were inserted into Trench 165, while 15 were inserted into Trench 7. An
additional five permeameters, which were screened over the bottom 360 cm, were placed in
both Trenches 7 and 165 using the penetration test holes. Water acceptance rates were then
measured for each permeameter using either a pour-in method to maintain a constant head
near the top of the casing or a pump-in method using positive head and flow meter. Equation
1 was used to compute the postgrouting hydraulic conductivities from the water acceptance
rates described above. Analysis of the characteristics of the different in situ permeameters indi-
cated that the different depths of insertion had no effect on the measured hydraulic conduc-
tivity nor did the diameter of the permeameter (both 3.8 and 3.2-cm pipe was used). The geo-
metric mean hydraulic conductivity of the population of measurements was 1.85 • 10 -5 cm/
s for 36 independent determinations. The distribution of hydraulic conductivities in the
groups of pre- and postgrouting measurements is depicted in Fig. 6. Although the mean
hydraulic conductivity of 9.7 • 10 -7 c m / s of the polyacrylamide grouted pilot-scale burial
trench in the previous study [2] was an order of magnitude lower than that in the present study,
the distribution of previous values was much narrower with eight out of nine values at or below
the field detection limit of 7 • 10 -7 cm/s. In that previous study, hydraulic conductivities were
measured in well casings that had remained within the trench during grouting; whether those
1 0 -+
Before ~ After
II Mean m Grouting Grouting
~" 1 0-' lira,- 2.91 x 10 "j
IIIIII,l,,,..
E
o 10"
I
.'=
._>
1 0 -4
"O
O
0 1 0 "s
lillilillilil
.o
I
= 1 0-"
"r" 1 0 "T
10"
Independent Conductivity Measurements

FIG. 6--The distribution of hydraulic conductivity measurements within two radioactive solid waste
burialbytrenches
Copyright before
ASTM Int'l (all rights and after
reserved); inFeb
Mon situ grouting
24 23:07:03 ESTwith
2020 polyacrylamide.
measurements could be ascnbed to grouting of waste and trench contents or to grouting of

well screen could not be determined The present observations, however, are free from such
possible artifacts because the in sltu permeameters were not placed into the trenches until sev-
eral months after grout had set Thus, the hydrologic isolation of burial trench contents
appears to have been achieved by in situ grouting with polyacrylamide The regions of higher
hydraulic conductivity within the trenches may be caused by the insertion of the permeame-
ters into waste packages inaccessible to the grout, or they may be caused by grout fractunng
by waste movement reduced by penetrometer insertion However, the determination of
whether such higher hydraulic conductlvlties by grout segregation within the trenches or by
failure of the grout polymerization in specific regions cannot be determined at present. Like-
wise, the persistence of these hydrologic properties with time needs to be determined
Acknowledgments
The assistance ofC D Farmer, A L Thomas, D S Marshall, A E Hunley, and B D Puett
was essential in completing the many difficult field operations and samphng required d u n n g
this project
References
[1] Spalding, B P, Jacobs, G K, and Davis, E C, "Demonstrattons of Technology for Remedlatlon
and Closure of Oak Ridge National Laboratory Waste Disposal Sites," Report ORNL/TM-l 1286,
Oak Ridge National Laboratory, Oak Ridge, TN, 1989
[2] Spalding, B P, Hyder, L K, and Munro, 1 L, "Grouting as a Remedial Techmque for Buned Low-
Level Radioactive Wastes," Journal ofEnvtronmental Quahty, Vol 14, 1985, pp 389-396
[3] Spalding, B P, Lee, S Y, Farmer, C D, Hyder, L K, and Supaoklt, P, "'In Sltu Grouting of Buraed
Transuranlc Waste with Polyacrylamlde," m Oak Rtdge Model Conference Proceedings, CONF-
871075, Vol l, Part 3, Martin Marietta Energy Systems, Inc, Oak Ridge, TN, 1987, pp 39-76
[4] Solymar, Z V and Reed, D J, "Companson between In-Sltu Test Methods," in Use of the In Sltu
Tests m Geotechmcal Engmeenng, S P Clemence, Ed, Amencan Society of Clvfl Engineers, New
York, 1986, pp 1236-1248
[5] Spalding, B P, "Bunal Trench DynamicCompaction at a Humid Site," in Proceedings of the Seventh
Annual Partwlpants" Informatton Meeting DOE Low-Level Waste Management Program, CONF-
8509121, EG&G Idaho, Inc, Idaho Falls, ID, 1986, pp 676-698
[6] Hvorslev, M J, "Time Lag and Sod Permeability in Ground Water Observations," Bulletin 36,
Waterways Experiment Station, U S Army Corps of Engineers, Vicksburg, MS, 1951
[7] Luxmoore, R J, Spalding, B P, and Munro, I L, "Areal Variation and Chemical Modification of
Weathered Shale Infiltration Charactensucs," Sod Science Socwty of Amerwan Journal, Vol 45,
1981, pp 687-691
J a m e s H. Kleppe, ~M a r k 7". Otten, ~ a n d John T. Finn 2
Stabilization/Solidification of Metal-
Contaminated Soils: Two Case Histories
REFERENCE: Kleppe, J H , Otten, M T , and Finn, J T , "Stabilization/Solidification of
Metal-Contaminated Soils: Two Case Histories," Sta&hzatlon andSohdlficatton of Hazardous,
Radtoactwe, andMlxed Wastes, 2nd Volume, ASTM STP 1123, T M Gllham and C C Wiles,
Eds, American Society for Testing and Materials, Philadelphia, 1992, pp 426-439
ABSTRACT. This paper describes treatablhty studies, design, and remedlatlon for the stablh-
zation of contaminated soll at two hazardous waste sites in Washington State In both sites, the
soil was contaminated with metals The study ofstablhzation/sohdlfieationincluded bench-scale
treatability studies, preparation of plans and specifications for the remedlatlon, and monitoring
stabilization remedlatlon This paper presents the results of the above work with focus on prac-
tical, regulatory, and performance aspects of the stabilization The effect of stabilization agents
on the leaching of metals and the relation of the leaching results to apphcable regulations are
discussed
Site A was a large manufactunng facility covering about 300 000 m 2 (80 acres) Sod contam-
ination included lead, chromium, and arsemc The soil to be stabilized contarned total concen-
trations of up to 19 000 mg/kg lead, 1600 mg/kg chromium, and 180 mg/kg arseme, however,
bulk soil samples obtained for this study contained unexpectedly lower concentratmns of these
metals Extraction procedure toxicity (EP Tox) concentrations were below hazardous waste des-
ignation levels StaNhzatmn will be used on soil with total concentrations greater than 8000 mg/
kg lead, 600 mg/kg chrommm, and 100 mg/kg arsenic Site B covered about 10 000 m 2 (2 acres)
and was used to repair and store machinery As a result, the soil was contaminated with lead
The highest lead concentration was 17 000 mg/kg, with the corresponding EP Tox concentration
of 300 mg/L The selected remedial action was to remove the soil with lead over 500 mg/kg and
place it in a solid waste landfill In order for the soil to be classified as a solid waste, the leachable
lead had to be less than 5 mg/L based on the EP Tox test
At Site A, the stabihzed soil had low compressive strengths, possibly due to petroleum contam-
matlon and natural organics In the soil Leaching tests using a monolith showed that the leachate
from the soil/cement specimen unexpectedly contained higher concentrations of chromium
than the leachate from untreated soil The relative amount of lead leached from sod/cement and
untreated soil varied The results for arsenic were inconclusive because of the low arsenic con-
centratmns in the tested sods and leachates At Site B, proprietary stabilization agents were suc-
cessfully used to reduce leachability below hazardous waste levels as measured by EP Tox
KEY WORDS. stabilization, sohdlfiCatlon, remediation, leaching, treatabthty, stabd~zatIon

agents, compressive strength tests, EP Tox tests, hazardous waste, lead, chrommm, arsenic,
metal-contaminated soil
While sod s t a b l h z a h o n has been used for decades m geotechmcal/clvll construction, con-
t a m m a t e d s o d / w a s t e sohdlficatlon or stablhzatlon is a relatively n e w t e c h m q u e The typical
goal o f solidification/stabilization (S/S) is to p r e v e n t or m i n i m i z e the release o f a c o n t a m i n a n t
into the e n v w o n m e n t by (1) p r o d u c i n g a solid, (2) i m p r o v i n g h a n d h n g charactenstms, (3)
decreasing surface area for c o n t a m i n a n t transport, (4) h m l t m g mobility o f the c o n t a m i n a n t
Geotechnlcal engineer, Hart Crowser, Seattle, WA 98102

2 Chemical engineer, Hart Crowser, Seattle, WA 98102
426
Copyright 9 1992 by ASTM International www aslm org

KLEPPE ET AL ON METAL-CONTAMINATED SOILS 427
when exposed to leaching fluids, or (5) chemically transforming or bonding the contaminant
into a nontoxic form
Each site will have its own goals or objectives for S/S, depending on the contaminants and
site conditions This paper describes two sites with metal-contaminated soils and the S/S
approaches used for remedxatIon These sites have slightly varying objecuves that resulted in
different selections of S/S agents At Site A, the contaminated soil was not classified as a haz-
ardous waste, but was considered a risk to human health and the environment based on total
metal concentrations At Site B, the contaminated soil was classified as a state dangerous waste
(similar to federal hazardous waste) based on toxicity test results
This paper will present results learned from the study and implementation of S/S at these
sites, including treatability studies, regulatory issues and resolutions, design (including plans
and speoficatlons), and implementation and quality assurance (Site B only)
Site A: Initial Treatability Testing of Portland Cement Stabilization

ProJectDescrlptton
Site A covers over 300 000 m 2 (80 acres) in Western Washington and was the site of a pre-
vious manufactunng facility The facility operated from 1907 to the early 1980s and produced
a variety of metal products As a result of site activities, on-site surficial soils became contam-
inated with metals (predominantly arsenic, chromium, and lead) and petroleum hydrocar-
bons (predominantly diesel fuel) Soils became contaminated with metals as a result of on-site
disposal of foundry baghouse dust, slag, and paint waste (including sand-blasting waste) Soils
contaminated with metals were limited to the upper 2 m (7 ft) of portions of the s~te Total
arsenic, chromium, and lead concentrations were up to 180, 1600, and 19 000 ppm, respec-
tively, with 198 of 200 samples hawng extraction procedure toxicity (EP Tox) concentrations
less than hazardous waste levels The metal-contaminated soils also contained concentrations
of petroleum hydrocarbons (about 500 mg/kg)
A remedml investigation and feasibility study (RI/FS) and remediation process were vol-
untarily undertaken by the current property owner As a result of the RI/FS work, remediatIon
of the metal-contaminated soils was proposed in order to reduce risks to human health (pri-
marily by reducing inhalation of dust and ingestion of soil contaminated with metals) Leach-
ing of metals into the groundwater was not an established problem at this site
S/S was selected as the preferred alternatwe for metal-contaminated sods The proposed
remedial action at the site will stabilize about 15 300 m 3 (20 000 yd 3) of soil with coneentra-
Uons exceeding 8000 mg/kg lead, 600 mg/kg chromium, or 100 mg/kg arsenic The cleanup
goals were based on a risk assessment that identified exposure due to dust and d~rect contact
of the metal-contaminated sods as the p n m a r y concern
Objecttves
Table 1 presents the preliminary design criteria developed from the FS and the correspond-
ing tasks of the treatability study Portland cement (with no additives) was selected as a simple,
low-cost S/S agent hkely to achieve the p n m a r y design criteria A preliminary treatability
study was conducted to determine if soil/cement mixes could meet the preliminary design cri-
teria The p n m a r y objective of the treatability study was to test the effectiveness of soft/cement
mixtures in forming a strong material capable of reducing human exposure due to dust and
direct contact with the soil The secondary objective was to test the relative leachability of the
soil/cement mixtures in comparison to the already low leachability of the soil alone
TABLE 1--Prehmtnary destgn criteria and corresponding treatablhty study tasks Stte A
Prehmmary Design Criteria Treatabdlty Study Tasks
Form solid with compressive strength between 345 Test strengths of 3, 6, and 12% soil/
and 6900 kPa (50 psi and 1000 psi) to reduce cement mixtures
human exposure due to dust and direct contact and
prowde for site development
S/S should result m leachabihty below or equivalent to Compare leachablhty of soil/cement
original sod and onganat sod, and pH of soil/
cement and original soil
Compresstve Strength Testing

Sod Samphng--The sods used m this study were collected from the site and were chosen to
include relatwely high metal concentrations combined with relatively high total petroleum
hydrocarbon (TPH) concentrations Although the actual concentrations of metals and TPH
obtained m the sod samples were lower than expected from data obtained during the RI, the
soil samples were acceptable for the purposes of this mmal study The sod used m the study
was described as a gravelly, sdty sand with a TPH concentraUon of 540 mg/kg
Spectmen Preparatton--Bulk samples were screened on the 19 05-mm (~-m.) sieve and
homogemzed prior to preparation of test cyhnders We prepared 10 2-cm (4-m) dlameter sod/
cement cyhnders using 3, 6, and 12% by dry weight of portland cement Type I/II Each cyl-
inder reqmred about 2 3 kg (5 lb) of soft Cyhnders were allowed to cure m a moist condmon
at about 25~
Results--We performed unconfined compresswe strength tests 7 and 28 days after prepa-
rattan of the cyhnders. We used ASTM Standard Test Method Compressive Strength of Cyhn-
dncal Concrete Specimens (C 39) Results of the 7 and 28-day tests are presented m Tables 2
and 3, respectively.
Dtscusston--The mix of sod/cement resulting from the procedures above represents a
strong material m comparison to existing sod The results indicated that w~th addmonal pro-
cess development, sod/cement mixtures could meet the prehmlnary design criterion for
strength
Addmonal process development wdl include testing of cyhnders with a higher percentage
of cement Compresswe strengths m the m~x were approximately proportmnal to the percent
cement, with 12% gwmg the highest compresswe strength of about 1 030 kPa (150 PS0 We
would expect that the 12% sod/cement mix would be effective m reducing human exposure to
metals by dust and direct contact, particularly ff isolated by a structural fill cover as planned
at this s~te.
Compresswe strengths were about 3-fold lower than antlopated; typical compresswe
strengths for similar compacted sod/cement mixes with 120/0cement are on the order of 1700
TABLE 2--Results of seven-day unconfined compresstve strength testmg Stte A

Percent Unconfined Compressive Moisture Content of Soil Dry Density of Compacted
Cement Strength m kPa (psi) in % at Compaction Mix in kg/m 3 (lb/ft 3)
3 186 (27) 21 1 68 (105)

6 448 (65) 20 1 70 (106)
12 903 (131) 19 1 70 (106)
12 690 (100)
15 1 65 (103)
KLEPPE ET AL ON METAL-CONTAMINATEDSOILS 429
TABLE 3--Results of 28-day unconfinedcompressivestrength testing Site A ~

Moisture Dry Density of
Content of Compacted
Range of Unconfined Compressive Soft m % at Mix m kg/m 3
Cement, % Strength m kPa (psx) CompacUon (lb/ft ~)
3 234 to 255 (34 to 37) (2 specimens) 21 1 68 (105)

6 469 to 552 (68 to 80) (3 specimens) 19 1 71 (107)
12 986 to 1096 (143 to 159) (3 specimens) 19 1 71 (107)
a According to ASTM Test Method for Compresswe Strength of Cyhndncal Concrete Specimens
(C 39-86)
result of natural orgamc matter that could interfere with cement curing m the samples tested
Natural organics were about 8% m the soft [as determined by ash content tests, ASTM Stan-
dard Test Method for Moisture, Ash, and Organic Matter of Peat Materials (D 2974), Method
C] The presence of petroleum hydrocarbons may also have had an adverse effect on the soft/
cement strengths.
Leachabthty Testmg of Sotl/Cement Cyhnders

We used a leachablhty testing procedure adapted from the American Nuclear Society (ANS)
Method 16 1 [2]. ExtracUon procedure toxicity (EP Tox) and toxicity characteristic leaching
procedure (TCLP) testing were not required because the contaminated soil was not a hazard-
ous waste The ANS 16 1 method was chosen because, unlike the TCLP procedure, it is a non-
destructive test that more closely simulates the leaching conditions of concern at this site.
These conditions include possible periodic contact of the soil/cement with rainwater infiltra-
tion Sod/cement will be protected at the surface by a structural fill cover
Spectmen Preparatton Soil/Cement Specimens--Three specimens of solidified soils were
used Specimens were approximately 10 2 cm (4 in ) tall and 10 2 cm (4 in ) m diameter. The
top and bottom of the cyhnder had a fresh cut, and a fresh surface was exposed on the sides
All soil/cement specimens used for leachability testing contained 12% portland cement by dry
weight
Specimen Preparation Sod Only--Three compacted sod cylinders were prepared The soil
was molded but without the addition of cement We prepared the soil cylinders at water con-
tents and densities comparable to soil/cement cylinders.
Test Setup--The study consisted of a total of six specimens, three each of sml/cement and
soil only. Each specimen was placed m a woven Teflon basket. The basket fit loosely about the
specimen and covered approximately 10% of its surface area The basket was suspended in the
center of the leachate vessel. The vessel was a 5 7-L stainless steel stock pot Vessels were cov-
ered to prevent disturbance Dlstdled water was used as the leachate solution Sufficient water
was used to cover the suspended specimen by at least 1 cm The sohd-to-hquid volume ratio
was about 1" 1 3
Leachate Sample Collection--The ANS 16 1 method was modified as suggested by C6t6
and Isabel [3] in order to obtain analytical results greater than the method detection limits
The leachate from one specimen of soil/cement and one specimen of untreated soil was sam-
pied on Days 1, 3, 7, and 14 The leachate from two specimens of soil/cement and two speci-
mens of untreated soil were sampled at Day 14 only (These duplicate samples are termed
Copyright by ASTM Int'l (all
"continuous run"rightsm
reserved);
Table Mon4Feb 24 23:07:03 EST 2020
) Additional duplicates would have been advantageous but were
TABLE 4 - - T o t a l metals tn sod and cement Site A ~
Concentration in mg/kg
Constituent Soil Cement
Arsenic 49 46
Chromium 80 58
Lead 530 272
pH 62 12 2
a Environmental Protection Agency methods arsenic--7061,

chromium--7190, and lead--FAA, all are reported on a dry
weight basis
n o t i n c l u d e d in this p r e l i m i n a r y work because o f cost c o n s i d e r a t i o n s In a b r o a d sense, the

m o r e f r e q u e n t s a m p l i n g is r e p r e s e n t a t i v e o f t r a n s i e n t exposure o f soil only or s o i l / c e m e n t to
water, while t h e " c o n t i n u o u s r u n " s a m p l i n g is r e p r e s e n t a t i v e o f c o n t i n u o u s l y s a t u r a t e d expo-
sure o f soil only or s m l / c e m e n t to w a t e r O n e b l a n k s a m p l e o f distilled water was collected from
the a p p a r a t u s p r i o r to i m m e r s i o n o f t h e soil s a m p l e s
E a c h 1000 m L leachate s a m p l e was collected using a peristaltic p u m p to m o v e water from
the vessel t h r o u g h a filter a n d i n t o p r e p r e s e r v e d s a m p l e bottles, a c c o r d i n g to s t a n d a r d g r o u n d -
w a t e r s a m p l i n g procedures. L e a c h a t e was filtered in o r d e r to o b t a i n data for dissolved metals,
w h i c h b e t t e r indicates i m p a c t s to g r o u n d w a t e r t h a n total m e t a l s
Vessels were filled w i t h at least 1000 m L o f distilled w a t e r After the leachate samples were
collected, t h e r e m a i n d e r o f the leachate was discarded T h e vessels were n n s e d with distilled
water. F r e s h distilled w a t e r was t h e n a d d e d to the vessels, a n d the samples were again sus-
p e n d e d in t h e vessels
R e s u l t s - - S a m p l e s were stored in T e f l o n - h n e d , a m b e r bottles in a refrigerator until all sam-
ples were collected. S a m p l e s were t h e n a n a l y z e d for dissolved lead, c h r o m i u m , a n d arsenic
u s i n g low d e t e c t i o n limits' testing m e t h o d s T a b l e 4 shows t h e total metals a n d p H in soil a n d
c e m e n t T a b l e 5 p r e s e n t s t h e results o f t h e leachability testing
TABLE 5--Results o f leachabthty testmg by Amertcan Nuclear Socwty 16 i procedure Stte A a
Leachate Concentration, mg/L b
Constituent Initial Water Blank 1 3 7 14 14Cc 14DE d
SOIL ONLY
Arsenic 5U 5U 5U 5U 5U 5U 5U
Chromium l0 l0 1U IU 20 IU 1U
Lead 5U 5U 5U 5U 14 0 90 5U
SOIL/CEMENT
Arsenic 5U 5U 5U 5U 5U 5U 5U
Chromium 10 30 30 40 60 18 0 15 0
Lead 5U 5U 5U 5U 5U 80 50
a Environmental Protection Agency methods arsenic--7061, chromium--6010, and lead--6010

b U = Below detection limits as listed
c 14C = Continuous run sampled once at the end of 14 days
d 14DC = Duplicate run of 14C
Dtscusslon--The results of this study show relaUvely slmdar leachabd~ty (for lead and arse-
nic) between sod-only and sod/cement specimens Concentrations of metals leached from sod-
only or soft/cement specimens were tow when compared to total metal concentrations in the
soft These results ln&cated that, with addmonal process development, sod/cement mixtures
could meet the prehmmary design criterion with regard to leachablhty
The leachate from the sod/cement specimen actually contained shghtly higher concentra-
tions of chromium than the soil-only specimen The same effect has been observed m numer-
ous other stu&es with various binders and waste types, as summarized in Conner [3] This
study did not produce data to predict the complex chemical interactions occurnng m the pro-
cess No speclaUon of the chromium was done However, one explanation for the h~gher chro-
m m m concentrations m sod/cement is that the pH change from 6 2 to 12 2 y~elded more sol-
uble forms of chromium such as Cr VI Another explanation is that the cement ~tselfmay have
contributed more leachable chromium than the soil The total chromium concentration of the
portland cement used was s~mdar to other reported concentrations in portland cement [4] but
was a full 70% as high as the total chrommm concentration m the soil alone
The concentrations of lead were low in leachate from both the sod/cement and the untreated
soil These low concentranons (near the detection hmlt of 5 mg/L) m&cated neither an
increase or decrease in the leachabd~ty of lead as a result of th~s cement stabilization on these
soils The total arsenic concentration of the tested soils was too low for any conclusions to be
drawn regarding the effects of porttand cement stabilization on arsenic mobd~ty m these soils
Addlttonal Design Criteria and Proposed Process Development Work

Addmonal remedial design criteria for this site were developed d u n n g consideration of
future property use by the owner and negotmtlons with regulatory agencies Table 6 presents
these addmonal criteria and the corresponding proposed tasks for S/S process development
work
The goal of forming a sohd with a strength between 1380 and 6900 kPa (200 and 1000 ps0
was developed to accommodate future use of the site, including truck traffic over the area
where the solidified soil would be placed Destgnatlon of appropriate areas for placement of
TABLE 6--Addttlonal design crtterta and correspondmgproposedprocess development tasks Stte A
Addmonal Design Criteria Proposed Process Development Tasks
Form durable solid with compresswe Test strengths of 15, 20, and 26% sod/cement
strength between 1380 and 6900 kPa (200 mixtures
and 1000 ps0 Test wet/dry durabdlty (freeze/thaw durability not
apphcable)
Sohd should have low hydraulic conductivity Test hydraulic conductivity
S/S should result in leachablhty below or Compare leachabdlty of soil/cement and original soil
eqmvalent to original sod Compare pH of soil/cement and onganal soil
S/S should not result in formation of a Perform TCLP tests
hazardous waste
Ensure complete mixing of sod/cement Use a pug mill, verify using acid/base ln&cator
Treat materials of acceptable particle s~ze Determine largest size feasible and evaluate screening,
shredding, and crushing as pretreatment
alternatives
the solidified soft (beneath parking area, beneath buildings) has become an important issue in
ongoing agency negotiations
We anticipate that the soil particle size criterion will become an important final design con-
slderatlon Separation of and shredding or crushing of oversize debris may be required as pre-
treatment Such pretreatment steps can greatly increase the cost and lengthen the schedule of
full-scale S/S processing [4].
Site B
ProjectDescrtptton
Site B covers about l0 000 m 2 (2 acres) and was used to repair and store machinery, remove
paint from small boats, and store used battenes As a result, the soil was contaminated with
lead The highest lead concentration was 17 000 mg/kg, with the corresponding EP Tox con-
centration of 300 mg/L The objective of our work was to remove and treat the contaminated
soil so that it could be placed in the owner's private solid waste landfill In order to accomplish
that goal, we had to
1 Determine the extent of soil that was designated dangerous waste and therefore required
treatment pnor to disposal as solid waste (in Washington State, material with EP Tox
level >_ 5 mg/L is designated dangerous waste, which is the same as Resource Conser-
vation and Recovery Act (RCRA) c n t e n a for hazardous waste)
2 Determine the type of treatment
3 Prepare bid documents (plans and specifications) for public bidding of the remedlatlon
work
4 Obtain regulatory agency concurrence that the treated soil was solid waste
5 Complete all work before the owner's private landfill would be closed m November 1989
The landfill had to be closed before the effective date of new state solid waste landfill
regulations
The selected remedial'action at Site B included placing stabilized soil into an existing solid
waste landfill The landfill disposal site was located about one mile from the remediation site
The stabilized waste was to be covered with a temporary cover and will be covered with
another low-permeability cover when the landfill is dosed
In this situation, strength, durability, and permeability of the stabilized soil are not impor-
tant The main c n t e n a for using stabihzation at the site was to reduce the leachability to less
than 5 mg/kg as measured by the EP Tox test so that the treated soil could be designated solid
waste and not dangerous waste
In order to meet the above obJectives, we performed a bench-scale treatability study,
reviewed the results with the state regulatory agency, prepared bid documents for stabilization,
and performed quality assurance during stabilization work The following presents the pro-
cedures we used and discusses unantlopated situations
Treatablhty Study
The treatability study was used to meet the first two project objectives, which were (1) to
determine extent of soil that was designated dangerous waste, and therefore, required treat-
ment p n o r to disposal as solid waste, and (2) to determine the type of treatment
The treatability study was started in June 1989 and completed m August 1989 Our first
choice
Copyright wasInt'l
by ASTM to (alluse cement
rights alone
reserved); Mon Feb 24as the stabilization
23:07:03 EST 2020 agent Since we were not sure if cement
alone or cement with fly ash would lower the EP Tox concentration and we had a tight time
schedule, we tested several mixtures at the same t~me Therefore, our treatabdlty study
included two vendor stabdlzatlon agents, as well as several cement and cement/fly ash
mixtures At the start of the treatablhty study, we estimated that about 380 m 3(500 yd 3) of sod
reqmred treatment
We needed to use vendors for the treatabdlty work who would be wdhng to conduct bench-
scale treatablhty stu&es, were wdhng to send us their treated samples to test independently,
and were interested m working as a subcontractor on a rather small j o b We had previously
contacted several vendors o f stabdlzatlon agents, and several had m the past successfully
treated soil with lead We selected two vendors
Although we had planned to conduct the treatabdlty study m one step, we encountered
unexpected results, and ended up with three steps The work in each step and the results are
presented below
The Imtlal Study P l a n - - T h e mitml plan called for samphng the most-contaminated sod at
one 15 by 15-m grid (50 by 50-ft) location, and prepanng cyhnders for strength and chemical
testing In order to obtain a representative sample, 21 subsamples were obtained throughout
the grid of highest contamination The total quantity of sample collected was about 275 kg
(600 lb)
In the lab, we screened the sod on the 19 05-mm (~-m) sieve and &vlded it mto six equal
plies of about 45 kg (100 lb) each Sod from pdes (a) through (f) were tested m the following
mixes.
(a) 3% cement
(b) 6% cement
(c) 12% cement
(d) 3% cement, 9% fly ash
(e) Proprietary treatment " ' C "
(f) Proprietary treatment "S "
Sod from piles (a) through (e) were mixed m our laboratory and placed m 10 2-cm-diameter
by 20 4-cm-hlgh (4 by 8-m) cylinders for cunng Soil from pile (f) was shipped to a vendor for
mixing and testing
After seven clays cunng, the cylinders were transported to the strength testing laboratory m
mdwldual plastic bags They were tested and returned m the bags so that specimens could be
used for chemical testing.
Tables 7 and 8, respectively, present the results of the strength and chemical testing An
important unexpected result was that the total lead m the untreated s p e o m e n was 76 mg/kg
TABLE 7--Inmal study strength test results Stte B

Unconfined Compressive Strength, kPa (psi)a
Mix Test 1 Test 2 Test 3 Average
3% cement 517 (75) 469 (68) 552 (80) 510 (74)

6% cement 1448 (210) 1600 (232) 1331 (193) 1462 (212)
12% cement 2517 (365) 2241 (325) 1993 (289) 2248 (326)
3% cement, 9% fly ash 938 (136) 724 (105) 793 (115) 821 (119)
"C" 903 (131) 924 (134) 876 (127) 903 (131)
"S" 7345 (>50)
Copyright by ASTM Test
a ASTM Int'l (allMethod
rights reserved); Mon Feb 24 23:07:03
for Compressive EST 2020 of Cylindrical Concrete Speomens (C 39)
Strength
TABLE 8--Imtlal study chemical test results Site B
MIX Sample Type EP Tox, mg/L~ Total Lead, mg/kg"
Control (soft) 0 0057 76

3% cement Fragments 0 685
6% cement Fragments 0 1065 72 4
Monohthlc 0 0879
Duphcate 0 778
(Frag)
12% cement Fragments 0 0222
Fragments 0 0283
3% cement, 9% fly ash Monohthm 0 0537
"C'" Fragments 0 0057
"S" 0 083
a Enwronmental Protectmn Agency methods lead--7421, and extractmn procedure toxlclty (EP
Tox)-- 1310/7421
compared to nearly 12 000 mg/kg reported d u n n g the remedial investigation Leachable lead,
as indicated by the EP Tox test, was similarly low at 0 0057 mg/L
Archlved Sample S t u d y - - T h e goal of the study, however, was to test the suitability of the
stabilization techniques on soil specimens that would be designated dangerous waste Since
time was critical at this point, we retrieved an archwed soil specimen from our remedial mves-
tlgatlon specimens for testing The archlved sample was mixed and tested as in the initial
study, however, there was not a sufficient amount of the sample to prepare full-size cylinders
or to submit samples to vendors Table 9 presents the results of the chemical testing on the
stablhzed archlved sample
The unexpected result from this phase of the design study was that the leachability increased
with cement stablhzatlon As discussed above in Site A, the cement itself is one potential
source of lead in the toxicity extract Conner [4] reported that in four tests, cement had total
lead of 57 to 86 mg/kg and TCLP extract concentration of 0 25 to 0 34 mg/L We did not test
the cement used in Site B, but the cement we used at Site A had 272 mg/kg lead (Table 4)
Another possible reason for the increase in leachability is the change in pH d u n n g stablh-
zatmn and EP Tox testing The laboratory reported that the initial pH of the extract was about
1 l, which was lowered to about 5 after addition of the acetic acid during the extraction period
Quantification of the species of lead compounds (such as soluble lead hydroxide) before, dur-
ing, and after the EP Tox testing was not within the scope of this work. However, such analyses
would provide more Information regarding this phenomenon
TABLE 9--Archive sample chemical test results Site B
Sample EP Tox, Lead, mg/L a
Control (soft) 0 471

3% Cement 2 22
6% Cement, fragment 1 684
6% Cement, monolathm 10 165
Environmental Protection Agency methods lead--7421, and

extraction
Copyright by ASTM Int'l (all rights procedure
reserved); Mon toxicity
Feb 24 23:07:03 EST 2020(EP Tox)-- 1310/7421
TABLE lO--Resampled chemlcal test results Stte B
Grid EP Tox, Lead, mg/La Total Lead, mg/kga
14 0 415 214
H4 1 142 826
15 3 619 4 850
J5 < 0 054 102
J6 0 372 429
I6 334 000 17 005
C4 < 0 054 195
C5 0 103 513
C6 0 120 791
H5 0 113 1 454
K9 0 076 57
a Enwronmental ProtecUon Agency methods extracUon pro-
cedure toxicity (EP Tox)-- 1310/7421, lead--7421
Final VendorStudy--Based on the initial mix design work, the spatial variability of the lead
in the soil was much greater than revealed by the remedial investigation As a result, we felt
that the design based on the remedial investigation probably would overstate the amount of
soil requlnng cleanup under the established criteria We therefore recommended that Site B
be resampled using a 16-point composite (as opposed to a five-point composite used in the
remedial investigation) sample in each 15-m 2 (50-ft2) grid The greater number of subsamples
used in the composite provides more representative specimens in each grad In addition, the
16-point composite sample together with backhoe sampling would better model the actual
condition of the soils excavated d u n n g construction
We resampled eleven grids of Site B using a 16-point composite specimen These specimens
were tested for total lead and EP Tox lead Table 10 presents the results of these tests The total
lead levels are shown on a site plan of Site B in Fig 1 The distribution of the lead on the site
was more consistent than that reported during the remedial investigation The levels were
highest near the edge of the pavement and fell off with increasing distance Also, Grad K9,
which showed total lead of over 3000 mg/kg in the remedial investigation, showed total lead
levels near background
Figure 2 shows the areas to be remedlated These areas were based on the results of the latest
round of samphng Soils in these areas were remedmted as follows ( 1) soil with EP Tox above
5 mg/L was stabilized and placed in on-site landfill, and (2) remaining soils were placed in on-
site landfill.
The results of the resamphng revealed there is an area with significant leachable lead (EP
Tox above 5 mg/L) at Grad I6 We therefore completed a mix design study on this soil using
the two vendors previously used We did not use cement or cement/fly ash mixtures since the
archived specimen tests showed higher leachability in the specimen stabilized with cement
Table 11 presents the results of the study Both vendors provided materials that would suitably
stabilize the soil What was interesting to note was that in the samples with an EP Tox less than
5 mg/kg, the extract had a pH of 8 to 11 after the extraction period
The following mix designs were included in the final project specifications
Vendor Dry Soil Agent Water, by Wet Weight
"C" 1 020 030

"S" 1 014 020
436 HAZARDOUSWASTES
FIG. 1--Cleanup zones: Site B.
Treatability Study Conclusions--The principal conclusions and lessons learned from the
Site B treatability study were:
1. The distribution of lead contamination in the soil was highly variable, even within a 15-
m (50-ft) square grid.
2. Cement stabilization did not lower the leachability as measured by the EP Tox test.
3. Both ofthe proprietary vendor stabilization agents reduced the leachability in the EP Tox
test to less than 0.5 mg/L.
Both vendors used enough water to produce a fluid slurry mixture, similar to grout. Since
strength was not critical at Site B, we did not need to limit water content nor compact the
stabilized mixture to increase strength.
Site B Design
The objective of the design phase was to produce a design that met the following: budget;
schedule
Copyright by ASTM deadline;
Int'l (all rights contract
reserved); Monbidding requirements;
Feb 24 23:07:03 EST 2020 and regulatory agency approval of quality
Universidad Industrial de Santander (Universidad Industrial de Santander)the
assurance. Due to the short time schedule, design
pursuant was
to License startedNoinfurther
Agreement. July, beforeauthorized.
reproductions the treatability
KLEPPE ET AL. ON METAL-CONTAMINATED SOILS 437
FIG. 2--Surface soil lead concentrations: Site B.
study was completed (and therefore, before the regulatory agency agreed with the details of the
remediation plan). The design had to be completed by the end of August to allow time for the
remediation to be done by the end of November.
The owner was a state agency that was required to use public bidding to obtain remediation
contractors. In order to provide quality work and meet the project schedule, we recommended
restricting the bidding to prequalified contractors. The prequalification criteria were (1) expe-
rience with hazardous waste; (2) corporate health and safety plans including medical moni-
toring programs; and (3) personnel health and safety training. The bid documents were pre-
pared in the same manner as typical earthwork documents. The specifications required that
the stabilization be done using one of the two agents demonstrated in the treatability study.
The contract called for lump sum payments.
The specifications included quality assurance provisions for the following items: (1) mea-
surement of soil, stabilization agent, and water ratios; (2) thorough soil mixing; (3) testing of
TABLE 11--Final vendor study results: Site B. ~

Sample EP Tox, Lead, mg]L Extract pH, End of Test
C-7% 126.7 5.0
C-15% 29.61 5.1
C-20% 0.243 8.3
S- 14% 0.42 11.4
Duplicate <0.06
Copyright by ASTM Int'l (all rightsareserved);
Environmental Protection
Mon Feb 24 23:07:03 EST 2020 Agency methods: lead--7421;
extraction procedure toxicity (EP Tox)--1310/7421.
_ ~ CoverSand ~ C Membrane
StabilizedSoil ~ ~
. . . . . . .
NativeSoil
FIG. 3--Disposal site cross section: Site B.
soil left at the site after excavation for total and EP Tox lead; (4) controlled placement of the
stabilized soil; and (5) testing of the stabilized soil for EP Tox lead. The specifications required
that the engineer conduct the soil chemical testing and that the contractor be responsible for
controlling the mixing, with the specific details of mixing left up to the contractor. The spec-
ifications required the contractor to measure the soil, stabilization agent, and water in each
batch. They required that the mixing be done in a rotary-type mixer, that the soil could not
set without agitating more than 15 min, and that the soil be placed within 2 h after being
mixed.
Figure 3 shows the plan and section of the stabilized soil at the landfill and the temporary
cover. Since the cover was temporary, only a m i n i m u m thickness of sand was placed over the
0.000 76-m (30-mil) poly(vinyl chloride) (PVC) cover membrane. The purpose of the PVC
was to limit possible leaching of lead from rainfall infiltration through the stabilized soil.
Site B Remediation Work

The cleanup of contaminated soils at Site B was done between 30 Oct. and 22 Nov. 1989.
The contractor's main equipment included: Case 580K rubber-tired backhoe; d u m p truck and
flatbed trailer; Mack 6-m 3 (8-yd) concrete truck; and hopper and portable conveyor.
A combined total of 320 m 3 (415 yd 3) of soil was excavated and removed from the site. The
contaminated soil was placed in the disposal cell located on-site. The disposal cell was staked
out by the field engineer with a base area of 750 m 2 (8100 ft2). The cell consisted of the stabi-
lized and contaminated soil, covered with a 15-cm (6-in.) sand layer, the synthetic membrane,
and finally a 30-cm (12-in.) layer of sand. The cell varied from 1.0 to 1.3 m (3 to 4 ft) in thick-
ness and was sloped for drainage. Hydroseed was placed on the cell to inhibit erosion.
TABLE 12-- Verification sampling results, a

Zone Total Lead, mg/kg EP Tox, Lead, mg/L
1 1777 3.48
1A 121 ...
1B 116
2 365 "0.66
3 281 <0.10
4 70 <0.10
5 378 <0.10
6 270 0.12
a reserved);
Copyright by ASTM Int'l (all rights Environmental Protection
Mon Feb 24 23:07:03 EST 2020Agency methods: lead--7421; and
extraction procedure toxicity (EP Tox)--1310/7421.
TABLE 13--Stabdlzed sotl analytical results a
Sample Total Lead, mg/kg EP Tox, Lead m mg/L
C-1 253 <0 10

C-2 306 <0 10
a Envxronmental Protection Agency methods Lead--7421, and
extraction procedure tOXlClty(EP Tox)-- 1310/7421
The verification samphng confirmed that the soil on-site after remediation had total lead
less than 500 mg/kg and EP Tox lead below 5 mg/L The results are shown in Table 12
Composite samples of the stabilized soil were sent to a laboratory for analysis ofEP Tox lead
and total lead The EP Tox lead analysis indicated that the soil stabilization was successful
because the concentrations were below the detectable limit of 0 1 mg/L. The results are shown
m Table 13
Conclusions
At Site A, we anticipate that a soil/cement mix with no other additives'will provide a simple
and low-cost treatment that meets the remedial design criteria for this site This approach may
be useful for similar sites where the primary goal is to form a strong, durable solid At Site B,
S/S was successfully used to treat soil with lead contamination In this case, the primary cn-
t e n o n was reducing leachability, as measured by the EP Tox test, to less than 5 mg/L
The following is a list of our conclusions, based on our work on the two sites
1 At both sites, the metal concentrations in soil were hxghly variable

2 Strength of soil/cement may be reduced by natural organics and petroleum hydrocar-
bons in the soil
3 Cement itself may contain metals at concentrations high enough to impact leachability
test
4 Soil/cement may not decrease leachability
The specific type of S/S technology used for a given project must be selected to meet the
needs of that project Therefore, the scope of treatability studies, design criteria, and quality
assurance testing must be developed to answer the project-specific questions
References
[1] Felt, E J, "Factors Influencing Physical Properhes of Sod-Cement Mxxtures," Portland Cement Asso-
c~atton Bulletin D-5, 1955, Portland Cement Assooahon (repnnted from Highway Research Board
Bulletin 108, 1955, p 138)
[2] American National Standard Measurement of the Leachabd~tyof Sohdlfied Low-Level Radioactive
Wastes by a Short-Term Test Procedure, 16 1, American Nuclear Society, La Grange Park, IL, 1986
[3] C6t~, P L and Isabel, D, "Apphcatlon of a Dynamic Leaching Test to Sohdlfied Hazardous Wastes,"
Hazardous and lndustrlal Waste Management and Testing Thwd Sympostum, A S T M STP 851, L
P Jackson, A R Rohhk, and R A Conway, Eds, American Socwty for Testing and Materials, Phd-
adelphla, 1984, pp 48-60
[4] Conner, J R, ChemwalFtxatlon andSolld~catton and Hazardous Wastes, Van Nostrand Reinhold,
New York, 1990
[5] Wdes, C C, "Sohdlfication and Stabdxzatlon Technology," Standard Handbook of Hazardous
Waste Treatment andDtsposal, H M Freeman, Ed, McGraw Hdl, lnc, New York, 1989, pp 7 85-
7 101
Robert G. Lukas I and Cliff Schexnayder 2
Dynamic Compaction of Nuclear Waste

Deposits
REFERENCE" Lukas R G and Schexnayder, C, "Dynamic Compaction of Nuclear Waste
Deposits," Stabthzatton and Solldtficatlon of Hazardous, Radloacttve, and Mixed Wastes, 2nd
Volume, A S T M S T P 1123, T M Gllham and C C Wiles, Eds, Amencan Society for Testing
ABSTRACT: This paper discusses the use of dynamic compaction to denslfy miscellaneous
materials that have become contaminated with low-level radioactive materials which were bur-
led m trenches and covered with soil After densificatlon, the ground was compressed approxi-
mately 12 7% and cone penetration resistance readings were approximately four to s~x times
higher than before denslficatlon The type of equipment used for dynamic compaction, the grid
spacing and number of drops at each location, and safety procedures used during the dynamic
compaction operations are discussed in this paper Environmental considerations are excluded
from this paper
KEY WORDS: dynamic compaction, nuclear waste, settlement
Nuclear wastes from the D e p a r t m e n t of Energy's Savannah River nuclear plant were depos-
ited in a series of parallel trenches, which were 6 m (20 ft) wide by 6 m (20 ft) deep Each trench
was separated by a 3 to 6-m (10 to 20-ft) b e r m o f natural soil Within the trenches, the nuclear
waste was d u m p e d until the lower 4 9 m (16 ft) was filled Afterwards, the entire trench was
covered with 1 2 m (4 ft) o f a sandy silt overburden These trenches occupied a 4049-m 2 (58-
acre) site
The nuclear waste consisted o f r a n d o m matenals that had been exposed to nuclear radia-
tion, including clothing, building materials, metal vessels, pipes, construction equipment, and
fluids such as oll m i x e d with absorbent substances and placed in 21-m 3(55-gal) d r u m s In some
areas, the nuclear wastes were placed in metal boxes, known on-site as B25 boxes, which are
similar to conex containers S o m e boxes were r a n d o m l y d u m p e d into trenches, others were
stacked in an orderly matrix In e~ther case, they were covered with 1 2 m (4 ft) o f soil The
radiation is classified as ranging from low-level to intermediate-level beta g a m m a radiation
The deposits were placed within the trenches from before 1976 until 1986
It was observed that the soil cap which had been shaped to shed surface water was settling
and that water was beginning to accumulate in the low spots This was considered undesirable
since there was the likelihood of surface water seeping through the cap and b e c o m i n g contam-
inated from the nuclear deposits and then percolating downward to the groundwater table As
part o f the closure plan, it was decided to denslfy the nuclear waste within the trenches to
reduce future settlement before constructing a new impervious cap D y n a m i c c o m p a c t i o n was
selected to densffy the nuclear wastes
J Senior pnnclpal engineer, STS Consultants, Ltd, Northbrook, IL 60062

z Chief engineer, Nello L Teer Co, Durham, NC 27702
440

LUKAS AND SCHEXNAYDER ON DYNAMIC COMPACTION 441
In-Situ Improvement by Dynamic Compaction

Dynamic compaction is the process of usmg heavy tampers, typically rangmg from 5 5 to
27 3 metric tons (6 to 30 tons), raised and dropped by a single cable with a free spool or allowed
to free-fall through distances ranging from 9 1 to 30 5 m (30 to 100 fi) to ampact the soft,
thereby causing deep denslficatlon Dynamic compaction is described in numerous techmcal
papers [1-7]. The advantages and disadvantages of dynamic compaction are outhned m
Ref 5 This process was considered ideal for compacting the nuclear waste for a number of
reasons
1 Denslfication of the nuclear waste trenches takes place from the existing ground surface
without exposing the buned wastes This was considered essential for the safety of the
operation
2 The weight of the tamper and the drop height can be adjusted to vary the depth and
degree of compactmn attained In general, the greater the weight of the tamper and the
higher the drop height, the greater the depth and degree of compaction
3 Wherever resistant matenals are encountered below grade, addmonal energy can be
apphed to crush drums or displace large objects, thereby collapsing potentml voids within
the waste
4 Th~s process ts generally the most economical s~te ~mprovement process, and for deep
denslficauon it was one of the safest Other methods, such as using stone columns, would
have to penetrate the nuclear waste and expose constructmn personnel to the radmt~on
Excavatmn followed by recompactmn w~th convennonal compaction eqmpment would
also lead to unnecessary exposure
Preconstruction Test Program

E I Du Pont de Nemours & Co was the operator of the Savannah River plant for the
Department of Energy through March 1989 Du Pont conducted a dynamic compaction test
program of hmlted scope to aid m the techmcal design of the dynamic compactton closure
plan The test program prowded the reformation which was used m developing the project
specifications The test program utilized a F M C Link-Belt, L5418A crane with a 27 4-m (90-
fi) boom The Link-Belt machine had a 0 62-m (24 4-m )-diameter drum, power up/power
down, and was equipped with a 25-mm ( l-in )-diameter wire hoist rope
Two d]fferent size weights, 11 8 to 18 2 metric tons (13 to 20 tons) were dropped using a
two-part hne hoist nggmg The drop height was vaned to a m a x i m u m of 12 8 m (42 ft) The
18 2 metnc ton (20-ton) weight, which had a dmmeter of 2 4 m (8 ft), when dropped from 12 8
m (42 fi) was found to produce the greatest tmprovement Th~s result was annclpated since
this represents the greatest amount of energy that could be applied with the equipment on
hand Since the waste was b u n e d m 6 2-m (20-ft)-wlde trenches, a 3 by 3-m (10 by 10-ft) drop
pattern gnd was selected D u n n g both the test program and production denslfiCatlon, mom-
tonng was implemented for possible radiation releases. Air momtors, ITT pump and Eberhne,
and hand sweep surveys using Vlctonne, Thyac and Eberhne, Ludlurn and RO-Z equipment
were employed. No releases were detected
Based on the test program, expectations were that the soft-waste matnx could be compressed
so as to reduce the waste volume in the range of 11 to 13% Using an expected average crater
depth of 1 4 m (4 5 ft) and assuming httle or no heave of the adjacent ground surface, the
average percentage ground compression within the nuclear waste plus the cap computes to be
10 3% by the following expression
Dc • AT
AGC- - - • 100 (1)
GS X Dr
where
AGC = average percentage ground compression,
Dc = depth of crater 1 4 m (4 5 it),
AT = area of tamper 4 7 m 2 (50 2 ft2), for a 2 4-m (8-ft)-diameter tamper,
OF ----depth of fill 6 7 m (22 ft), 0 6-m (2-ft) blanket plus 6 l-m (20-ft) trench, and
GS = grid spacing, 3 by 3 m (10 by 10 ft)
The closure project specifications that were developed from the test program stated the
following
1 Prior to any waste trench being treated by dynamic compaction, a 0 6-m (2-ft)-thlck soil
blanket would be placed without compaction Most burial trenches had an existing 1 2
m (4 ft) of earth cover resulting in 1 8 m (6 ft) of cover
2 Dynamic Compaction"
(a) Tamper--18 2 metric tons (20 tons) with a flat bottom, 2 4-m (8-ft) diameter
(b) Drop height-- 12 8 m (42 ft)
(c) Drop pattern--The trench surface was to be subdivided into 3 by 3-m (10 by 10-ft)
grids designated primary and secondary grid drop locations, and all primary craters
within a work area would be compacted and backfilled prior to dynamic compaction
of the secondary grid drop locations (Fig l)
(d) To facilitate weight recovery, the weight would remain attached to the crane cable
during all drops.
(e) Each crater would be driven using 20 drops, or until a m a x i m u m crater depth of 1 8
m (6 ft) was achieved, this m a x i m u m depth was specified for safety reasons, in order
not to encounter the radioactive waste material It was expected that on the average,
1 4-m (4 5-ft)-deep craters would result from 20 drops
3 Crater Backfilling
(a) Place a uniform 1 2-m (4-ft) + 0 15-m ( _+ 0 5-ft) loose lift of fill material into the
driven crater
(b) Compact the loose fill by dropping the 18 1 metric tons (20-ton) weight, five times
from 12 8 m (42 ft)
(c) Continue backfilling and compacting as outlined in (a) and (b) until a 0 6-m (2-ft)
m a x i m u m crater depth, measured from the surface of the soil blanket, is obtained
(d) Backfill would be compacted to 95% of m a x i m u m dry density [ASTM Test Methods
for Moisture-Density Relations of Soils and Soil-Aggregate Mixtures, Using 5 5-1b
(2 49-kg) R a m m e r and 12-in (304 8-mm) Drop (D 698-79)] at + 2% of optimum
moisture
Dynamic Compaction Equipment

Nello L Teer Company, Durham, NC, was awarded the construction contract to close spec-
ified areas and perform site grading for the new bunal ground Before bidding on the project,
equipment that had been used on similar work was investigated Equipment dealers did not
want to rent or lease full-revolving cranes for use in dynamic compaction jobs The quick
release of the weight and sudden stop when the weight strikes the ground, with the resulting
lack of resisting force on the cable at that moment, severely rocks the boom and upper works
of the
Copyright crane.
by ASTM Int'l This rocking
(all rights places
reserved); Mon Feb 24unusual
23:07:03 ESTstresses
2020 on a crane's undercamage, particularly the
TRENCH WIDTH
F 6.1m (20 ft.) "
f
2.4m (8ft.)
i
'4.6m 2 ( 5 O f t 2)
@
@ E)
@
13m
GRID PA]q-ERN
P E R SPECIFICATIONS
P = PRIMARY D R O P POINTS
S = S E C O N D A R Y D R O P POINTS
FIG. l--Dynamic compaction weight drop pattern.
turntable. The cost of repairs to machines used for dynamic compaction have been as high as
$200 000. Placing a hydraulic spud on the rear was investigated, but not found to be fully
effective in reducing rocking.
A Lampson LDC-350 machine (Fig. 2) was selected. This machine was developed specifi-
cally for dynamic compaction work. In order to turn, this machine must use its tracks as it has
no turntable and the upper works are fixed to the undercarriage. Each track runs off its own
motor. While this does slow the operation of positioning a weight on a specific location,
machine availability, which is defined as productive work time divided by total available work
time, was 86%. The LDC-350 has a large diameter hoist drum with greater line pull and speed,
which greatly enhances cycle times.
With the LDC-350, the boom is raised into its operating position with an erection line and
then tied off with two rear boom pendants. Additionally, there are two front kickback pen-
dants. Once these pendants are connected, the boom angle is fixed and there is no stress on the
boom-raising line drum.
The hoist or drop line works off a drum having a dual brake system. The main brake can
apply sufficient resistance so that the weight can be stopped and held at a desired height. The
second
Copyright brake
by ASTM is rights
Int'l (all more likeMon
reserved); a drag on a fishing
Feb 24 23:07:03 EST 2020 reel. When dropping the weight, the operator
1 ft.= 0.305m
1 ft. 2 = 0.093m 2 E~
1 U.S. ton = 0.91mt ro
.46m
3 6 . 6 m No. 22 M.E.C.
BOOM W/ 4.5" OFFSET
TIP & BOOM POINT
SHEAVE ADAPTER
BOOM PENDENTS
ERECTION PENDANTS
DROP LINE
BOOM RAISING
TIEOFF AFTER
ERECTION -a
WEIGHT
iAREA 4.65 m2
o6
8.7m ~ ~ i
. 12.3m RADIUS _l
FIG. 2--Lampson LDC-350 dynamic compactor.
applies the drag brake just before the mass strikes the ground. This prevents drop line backlash.
If, as on some machines, the operator would have to use the main brake for this purpose, there
can be severe damage to the machine if the brake is applied too soon. Even with its fixed boom,
upper works structurally tied to tracks, and the dual braking system, the LDC-350 is subject
to about 0.3 m (1 ft) of vertical motion at its rearmost point.
To provide backfill material, a self-loading scraper was used to bring the fill to each crater
location. A 335 HP bulldozer would then push the fill into the crater. With this size machine,
one blade load provided all the required fill. Though this meant the bulldozer was idle a large
portion of the time, the effect on total cycle time per crater justified the use of the large bull-
dozer. The use of the bulldozer also enhanced safety since it could fill the crater without having
to get directly
Downloaded/printed by under the hanging weight,
Field Operations
D u n n g initial dynamic compaction operations, STS Consultants, L t d , Northbrook, IL,
provided technical support to the contractor Pins were set in the ground around each crater
and elevation readings taken before and after dynamic compaction. Ground heave adjacent
to the crater locations did occur, but the amount of heave was slight Therefore, almost all of
the energy applied was used in denslfyIng the soil mass
After each drop the depth of crater was measured with a Spectra-Physics EL-1 electronic
level The EL- 1 projected an infrared laser signal in a 360* circle from its setup location One
person using a level eye detector that gave a visual liquid crystal display and a synchronous
audible tone could take accurate level readings up to 305 m ( 1000 ft) from the laser The aver-
age Incremental crater depth following each drop was found to be 0 15 m (0 5 ft) and 0 12 m
(0.4 ft) for primary and secondary craters, respectively Less crater depth occurred at the sec-
ondary grads because the compaction at the p n m a r y grid points created a stiff soil matrix at
the secondary grid locations. Some voids were encountered with resulting incremental depths
as high as 0 5 m (1 73 ft) per drop The result of achieving an average incremental depth of
0 14 m (0.45 ft) per drop was that, on the average, only 13 drops were required to drive the
craters This number is less than the 20 blows required by the specifications, but the test pro-
gram was conducted with a crane with a two-part line, which greatly reduces the energy deliv-
ered to the soil mass The average crater depth for the project was 1 7 m (5 67 ft), resulting in
an average ground compression of 12 9% The craters could easily have been driven deeper,
but as a safety measure, driving was stopped when the crater depth reached the range of 1 5 to
1.8 m (5 to 6 ft)
There were initially some problems with backfill compaction within the craters formed by
dynamic compaction The specifications called for 95% of standard proctor energy according
to ASTM Test Methods for Moisture-Density Relations of Soils and Soil-Aggregate Mixtures,
Using 5.5-1b (2 49-kg) R a m m e r and 12-in (304 8-mm) Drop (D 698), which is 60.5 ton
metres/m 3 ( 12 375 ft-lbs/ft3) Additionally, the specifications limited the backfill lift to a max-
Imum of 1 4 m (4 5 ft) and called for five drops of the tamper. An 18 2 metric ton (20-ton)
tamper free falling from 12 8 m (42 ft) imparts 233 ton metres drop (1 680 000 ft-lbs/drop)
Mechanical losses for the LDC-350 using a single line are calculated to be 11 1% Therefore,
for five drops, the resulting energy IS 1035 ton metres (7 467 600 ft-lbs) A 1 4-m (4 5-ft) lift
2.4 m (8 ft) in diameter is 6 4 m 3 (226.2 ft 3) Thus, dividing the energy imparted by the value
of soil to compact results in a unit compactlve effort of 161 ton metres/m 3 (33 014 ft-lbs/ft3),
which is 2.67 times higher than standard proctor energy Dunng application of this energy,
rebound after impact was experienced in the top part of the compacted backfill With so much
energy, the backfill operation was actually another phase of high-energy application
To correct this situation, a change was Instituted in the crater backfill operations The num-
ber of blows required was adjusted to the depth of backfill hft, and the lift thickness was
increased When the thicker backfill lift was tried, dlificulties were experienced with keeping
the tamper level upon initial impacting The final procedure adopted was to fill the crater
completely, including about 0 3 m (1 ft) of overfill, drop the weight one time from 4 6 m (15
ft) to seat the loose fill, place additional soil again to fill the crater completely; and, finally, at
this point, to apply five drops from 12.8 m (42 ft) The resultant unit energy was about 2 2
times standard proctor Density tests taken at different levels in the compacted backfill con-
firmed that the 95% compaction specification was being achieved for the full depth Once the
procedure was proven, it became the standard for the project
Backfilling normally required two fillings The first step was as described above and nor-
mally resulted in a crater depth of slightly greater than 0 6 m (2 ft) After filling that shallow
depth with loose soil, the second step required only two drops from 12 8 m (42 It) This rep-
resents a unit
Downloaded/printed by energy of 2 4 times standard proctor energy
The majority of the craters were completedy by:
1. Thirteen initial driving blows from 12.8 m (42 ft).

2. An initial full crater depth backfill with a 4.6-m (15-ft)-high seating drop and five 12.8-
m (42-ft)-high compaction drops.
3. A secondary backfill with two 12.8-m (42-ft)-high compaction drops.
With such a procedure, the Lampson machine supported by a bulldozer could tamp and
backfill 22 craters in a seven and one-half hour work period. The remaining half hour was
taken up in dressing and monitoring out of the burial ground.
Dynamic Cone Penetrometer Measurements

To check on the degree and depth of improvement throughout the full depth of the nuclear
waste facility, dynamic cone penetrometer readings were undertaken. The dynamic cone pen-
etrometer consists o f a 51-mm (2-in.)-diameter conical cone with a 60 ~ cone angle that is con-
nected to a drill rod driven by a 63-kg (140-1b) h a m m e r falling 0.8-m (30-in.). This cone is
driven into the nuclear waste, and the number of blows per 0.3 m (i 2 in.) required to advance
the penetrometer is recorded. Readings were taken in the nuclear waste before and after
dynamic compaction. In all cases, there was a significant increase in the penetration resistance
following dynamic compaction. Figure 3 illustrates the range in cone penetrometer values
before and after dynamic compaction for two different test sections. Typically, the cone pen-
etrometer values before dynamic compaction were found to range from about 5 to 20 blows
per 0.3 m (12 in.) with occasional higher values. These high values were assumed to be the
result of encountering large objects within the waste fill. After dynamic compaction, the typ-
ical cone penetrometer resistance values were found to be on the order of 30 to 80 blows per
0.3 m (12 in.). This represents an increase on the order of four to six.
Conclusions
Based upon the data obtained from the dynamic compaction of this 4049-m 2 (58-acre) site,
it is concluded that:
1. Densification of the nuclear waste was successfully accomplished by the dynamic com-
paction procedure.
// F////A. \ BACKFIll
,w, -10 - I~ ' ---" "-- SOIL / WASTE

I.iJ ~6E OF " ~GE OF "',. MATRIX LO
; PRE COMPACTION \ POST COMPAC110N "-.
Z !9...CPTESIS ] \ ~ ,,, OFfTBTS \ Z
~_ -2o ".. :, \
\
\ 6 :z
I--
13-
W Ld
123
IN SITU SOIL s
-30
I I I I I I I I I I I I I I I I I I I
10 20 30 40 50 60 70 B0 90 100
BLOWS/FOOT 1ft. =.304m
Downloaded/printed by FIG. 3--Cornparison o f cone penetrometer test results.
2 The craters formed by the impact of the drop weight averaged 1 7 m (5 67 ft), which
resulted in an average ground compression of the nuclear waste of approximately
12.9%
3 Cone penetrometer tests taken before and after dynamic compaction indicated a slgmf-
1cant increase m the penetraUon resistance, thereby confirming the h~gh degree of den-
Slficatlon within the nuclear waste deposit
4 Safety was maintained at all times by limiting the crater depth to less than 1 8 m (6 0 It)
Measurements were taken w~th air monitors around each compactor, monitoring of cra-
ters before backfill, and wipe tests of the weights d u n n g the enUre dynamic compachon
construction period These confirmed that radloacUve debris was not discharged
References
[1] Charles, J A, Burford, D, and Watts, K S, "F~eld Studies of the Effectwenessof Dynamic Consol-
ldaUon," Tenth International Conference on Sod Mechantcs and Foundatton Engmeermg, Stock-
holm, Vol 3, 1981, pp 617-622
[2] Leonards, G A, Cutter, W A, and Holtz, R D, "Dynamic Compaction of Granular Softs," Journal
of the GeotechnlcalEngtneerlngDtvlston, ASCE, Vol 106, GT4, Jan 1980, pp 35-44
[3] Lukas, R G , "Denslficatlon of Loose Deposits by Pounding," Journal of the Geotechntcal Engmeer-
mgDtvtslon, ASCE, Vol 106, April 1980, pp 435-446
[4] Lukas, R G , "Denslficatlon of a Decomposed Landfill Deposit," Eleventh Internattonal Conference
on Sod Mechamcs and Foundatton Engmeermg, San Franosco, CA, Vol 3, 1985, pp 1725-1728
[5] Lukas, R G , "Dynamic Compaction for Highway Construction, Vol 1, Design and Construction
Gmdehnes," Report No FHWA/RD-86/133, Federal Highway Administration, Washington, DC,
July 1986
[6] Mayne, P W, Jones, J S, and Dumas, J C, "Ground Response to Dynamic Compaction," Journal
of the GeotechnlcalEngtneerlngDtvtslon, ASCE, Vol 110, No 6, pp 757-774
[ 7] Menard, L and Brolse, Y, "Theoretmal and Practical Aspects of Dynamic Compactxon," Geotech-
tuque, March 1975
Robert D. Rogers, 1 John W. McConnell, Jr., 1 Julie D. Jastrow, 2 and
Diana S. Wickliff 3
Contributions of Lysimeter Data to the

Development of Site Specific Performance
Assessment Plans
REFERENCE: Rogers, R D , McConnell, J W, Jr, Jastrow, J D , and Wmkhff, D S, "Con-
tributlons of Lysimeter Data to the Development of Site Specific Performance Assessment
Plans," Stabdtzatzon and Sohdtficatton of Hazardous, Radioactive, and Mlxed Wastes 2nd Vol-
ume, A S T M S T P 1123, T M Gdham and C C Wdes, Eds, Amencan Sooety for Testing and
Materials, Phlladelphm, 1992, pp 448-465
ABSTRACT" Accurate data on the long-term performance of ra&oactwe wastes m a dxsposal

system are becoming a necessity m part because of restrictive federal and state regulaUons that
are being promulgated Data on the performance of buried ra&oactlve waste forms can be
obtained from lyslmeter arrays Lys~meters are Ideal instruments for the acqmsmon of actual
field test data When properly designed and operated, lys~meters can be used to isolate and then
study sod/waste systems under actual environmental condmons The complexity of interactions
occurring under field condltmns can never be completely duphcated by standard laboratory test-
mg This paper provides data from two instrumented, operational lyslmeter arrays containing
waste forms fabricated with highly loaded exchange resin materials The lyslmeters have been m
operation for five years and have been provtdlng data for both meteorological events and ra&o-
nuchde content of percolating water for the past four years Thin paper provides data obtained
from the lys~meters, a dlscussmn of the data, and a manlpulat~on of the database for calculating
some code parameters that can be used as input into codes used to verify performance of buried
radmacUve waste forms
KEY WORDS: lyslmeters, field testing, performance assessment, waste form, leaching, cesium,
cobalt, strontium
Intense p u b h c and soentlfic concern over the pracUces assocmted w~th the &sposal o f r a & o -
actwe wastes has resulted in a very real need to obtain accurate data on the long-term perfor-
m a n c e o f these wastes in disposal e n v i r o n m e n t s Regulations have been enacted by the U S
Nuclear Regulatory C o m m i s s i o n ( N R C ) (10 C F R Part 61) [ 1] and m o r e recently by the U S
D e p a r t m e n t o f Energy (DOE) ( D O E 5820 2A) [2] that hnk low-level ra&oacUve waste (LLW)
acceptance c n t e n a to long-term satisfactory performance o f the waste U n d e r 10 C F R Part 61,
c o m m e r c i a l l y generated ra&oactlve waste is classified in one of three classes. Wastes classified
as either Class B or Class C must be stablhzed with the stabd~ty being ensured for a m i n i m u m
o f 300 years The D O E order is intended to place responslbdlty on D O E and its contractors to
m a i n t a i n similar control over D O E - p r o d u c e d ra&oactlve wastes
Scientific specialist and senior engmeenng speclahst, respectively, Blotechnology and Waste Manage-
ment Programs, Idaho National Engmeenng Laboratory, P O Box 1625, Idaho Falls, ID 83415
2 Scientist, Environmental Research Division, Argonne National Laboratory, Argonne, IL 60439
3 Assistant soentlst, Environmental Sciences Division, Oak Ridge National Laboratory, Oak Ridge,
TN 37831
448

ROGERS ET AL ON LYSIMETER DATA 449
To verify the 300-year stablhty of waste forms, NRC originally specified the use of short-
term (accelerated) standardized tests with the thought that such tests in pnnclple could provide
information relevant to near-surface bunal performance objectives These tests, which were
published as the NRC Branch Technical Position (TP) on Waste Form [3], are currently
undergoing critical review to determine their apphcabihty to the 300-year stability
requirements
The DOE order is much more broad in its apphcatlon A central requirement outlined in
Section 3 of DOE 5820 2A is the need for a detailed understanding of the waste form m the
context of a burial site source and the effect of this source on site performance Among other
items, the order specifies that a site performance assessment (PA) be developed and Imple-
mented It is intended that the PA will predict whether a proposed disposal site will meet the
performance objectives as stated by the DOE order Correct assumptions regarding the per-
formance of the buried waste have a direct beanng on the outcome of the PA analysis. For
example, overly conservative assumptions regarding rates of radlonuchde release from buried
waste forms will result in costly overdeslgn in containment, on the other hand, ignoring the
release of special mobile constituents, such as organo-metalhc complexes, will overpredlct
containment performance
An appealing method that will provide "hard" data for both waste form stability determi-
nations and site PA development is the use of lyslmeters, which have been a standard in agri-
cultural and environmental research for years Uses have included studies of soil moisture bal-
ance, vahdatlon of environmental models, chemical movement and degradation studies, and,
not surprisingly, studies Involving the environmental fate of radionuchdes Lysimeters are
1deal instruments for obtaining actual field test data because when properly designed and oper-
ated they can be used to isolate and then manipulate soil/waste systems under actual environ-
mental conditions. The intricacies of Interactions under such conditions can never be com-
pletely duplicated by standard laboratory testing It is possible, however, to alter the design
and function of a lyslmeter so as to obtain the type and quantity of data required. Lyslmeter
design can range from small volumes of soil isolated from surrounding areas by impervious
dividers to concrete or metal caissons that are buried in the field so that the contents will be
fully exposed to the environment Such confinements lend themselves to instrumentation for
the acquisition of environmentally related data as well as the collection and storage of migrat-
ing precipitation
Several parameters associated with waste form stability as well as the more important overall
question of developing site-specific PA (that incorporates waste form stability data) can be
addressed with lysimeter data Applicable parameters that are best obtained with lyslmeters
include relevant monitoring and model validation data In terms of relevant monitoring data,
Information from the near-field (that area comprised of the waste form and the surrounding
soil) includes mass balance of released constituents, radionuchde speclation, species solubility
in geochemical systems, and retardation or dispersion of released constituents dunng trans-
port to the far-field Data m the form of soil pore water chemistry (inorganic and radioactive
constituents), soil mineralogy, soil water/mineral mass ratio, net Infiltration rate, soil profile
moisture and temperature, porosity, hydraulic conductivity, and dlsperslvity can be obtained
from instrumented lyslmeters These data are invaluable for the validation of process-level and
performance assessment codes used in PA models since the data represent a field data set that
contains complete Information characterizing environmental, hydrogeolog~cal, geochemical,
and waste form effects The relationship between code parameters and lyslmeter data is com-
pared in Table 1
The purpose of this paper is to discuss data obtained from two Instrumented, operational
lysimeter arrays that have been in operation for five years and demonstrate how these data can
serveby as
Copyright preliminary
ASTM inputMon
Int'l (all rights reserved); forFebcode validation
24 23:07:03 EST 2020 The study reported here was developed to per-
TABLE 1--Relationship between performance assessment code parameters and lystmeter data
Code Parameters Data Collected from Lysimeters
Q = inventory Known inventory is introduced by experimental design

P = annual percolation Amount of rainfall on lysimeter
Amount of evapotransp~raUon
S = fraction of saturation Soil moisture content
vv = water velocity Mass or volume of effluent water per umt Ume
R = retardauon factor Mass or volume of effluent solute per umt time relative to V.
ds = soil bulk density From experimental design of lysxmeter
Ps = effective soft porosity Can be estimated for saturated condmons from mass of
effluent water, volume of soil, sod bulk density
L = inventory released Radlonuchde concentrations m soil pore water and in effluent
Vw = trench volume From experimental design of lyslmeter
Cw = rad~onuchde concentration Radmnuchde concentration in effluent
M, = molahty Effluent concentrations
M I N = minerals dxssolved or From mineralogical characterization of sod at end of
preop~tated expenment
form field testing o f waste forms c o m p o s e d o f solidified ion exchange resin matenals from
E P I C O R - I I prefilters used in the cleanup of the Three Mile Island Nuclear Power Station
Waste used in the study is s~gnlficant because o f its high loading of radlonuchdes as well as
being comprised of actual ion exchange wastes of the type used by the nuclear industry

Waste form composition and lyslmeter design are discussed in detail elsewhere [4] and are
only briefly discussed here Wastes used in the experiment Include a mixture of synthetic
organic ion exchange resins as well as a mixture of organic exchange resins and an inorganic
zeolite Solidification agents e m p l o y e d to produce the 4 8 by 7.6-cm cyhndncal waste forms
used m the study were portland Type I-II c e m e n t and D o w vinyl ester-styrene (VES) Seven
of these waste forms (either c e m e n t or VES) were stacked end to end and inserted into each
lyslmeter to provide a 1-L v o l u m e of waste The inventory and approximate radlonuchde con-
tent o f waste forms used in each lyslmeter is found in Table 2 The p r e d o m i n a n t e nuclide is
Cs- 137 (71 to 94%) followed by Sr-90 ( 1 to 25%) There are also significant a m o u n t s ofCs- 134
as well as trace a m o u n t s o f Co-60 and Sb-125
There are ten lyslmeters, five at Oak Rxdge National Laboratory ( O R N L ) and five at
Argonne National Laboratory (ANL) Lyslmeters used in this study were designed to be self-
contained units that will be disposed o f at the termination o f the 20-year study Each is a 0 91-
by 3 12-m right circular cylinder divided into an upper c o m p a r t m e n t , which contains fill
material, waste forms, and instrumentation, and an e m p t y lower c o m p a r t m e n t for collecting
leachate (Fig 1) F o u r lyslmeters at each site are filled with soil, while a fifth (used as a control)
is filled with inert silica oxide sand Lyslmeters at A N L contain soil indigenous to that site,
whde the O R N L lyslmeters contain soil obtained from Savannah River Laboratory, SC Per-
tinent soil chemical and physical data on the fill materials are found in Table 3 Instrumen-
tation in each lysimeter includes porous cup soil-water samplers and soil moisture/tempera-
ture probes
The probes are connected to an on-site data acquisition and storage system (DAS) that also
collects data from a field meteorological station located at each site This makes it possible for
Copyright
the by ASTM Int'l (all rights
continuous reserved); Mon
recording o f Feb
soil24 23:07:03
m o i s t uEST
r e2020
/ t e m p e r a t u r e at three elevations m each lyslmeter
TABLE 2--Lyslmeter waste form inventory
Number and Number and

Type of Waste Total Curie Type of Waste Total Curie
Lys~meter Forms Content a LysImeter Forms Content a
ANL1 2CIA b 3 5 X 10 l ORNL 1 2 C1

5C1 5 C1A 35 • 10 -I
ANL 2 7 C2B 15 3 )< 10 t ORNL 2 7 C2B 15 3 • 10 -I
ANL 3 7D1A c 5 2 >( 10 -j ORNL 3 7 DIA 52 • 10 -1
ANL4 7D2 2 0 6 • 10 t ORNL 4 7 D2 20 6 )< 10 1
ANL 5 4 CIA ORNL 5 7 C2B 15 3 • 10 -~
3 CI 35 • 10 -1
a Waste forms have the following average curies contents
134C8 137C8 9~ Total
DI and D I A 4 38 • 10 3 66 22 • 10 -3 3 92 • 10 -3 74 52 • 10 3
D2 and D2A 1822 X 10 .3 27545 )< 10 .3 064 • 10 .3 29431 • 10 .3
C1 and CIA 2 95 X 10 .3 44 58 • 10 -3 2 64 • 10 .3 50 17 • 10 .3
C2A and C2B 13 53 • 10 -3 204 59 • 10 3 0 47 • 10 -3 218 59 • 10 -3
b Portland Type II Cement

c Dow vinyl ester-syrene
a l o n g w i t h t h e c o m p l e t e w e a t h e r h i s t o r y E a c h m o n t h , d a t a stored o n a cassette t a p e are

retrieved f r o m t h e D A S a n d t r a n s l a t e d i n t o a P C - c o m p a t i b l e disk file T h e s e d a t a are subse-
q u e n t l y r e d u c e d i n t o t a b u l a r a n d g r a p h i c d i s p l a y s D u r i n g e a c h q u a r t e r w a t e r is d r a w n f r o m
t h e p o r o u s c u p s o d - w a t e r s a m p l e r s a n d t h e l e a c h a t e collection c o m p a r t m e n t o f e a c h l y s l m e t e r
T h e t r a c k i n g o f n u c l i d e m i g r a t i o n is a c c o m p l i s h e d b y h a v i n g selected water s a m p l e s a n a l y z e d
for beta- a n d g a m m a - p r o d u c i n g n u c l i d e s
TABLE 3--Phystcal and chemtcal charactensttcs of sotls used at

ANL-E and ORNL
Soil
Charactenstic ANL ORNL ~
Soil bulk density, g/cm 3 1 74 NA b

Texture, %
Sand 29 58
Silt 29 2
Clay 42 39
Clay mineralogy, %
Vermiculite b 10
Kaohmte b 80
Percent carbon 4 20 0 07
Cation exchange capaoty, meq/100 g 84 49
pH (1 1 paste method) 83 62
Percent moisture-holding capacity 40 6 44 55
" Sod used at ORNL was obtained from the Savannah River
Laboratory, SC, because ItS characteristics approximate those of
the soils of the low level radioactive waste disposal site at Barn-
well, SC
b Not available
452 HAZARDOUSWASTES
10"
CT-O - Route moisture-cup

tubing as required--
5 places
2r0".~
J. % oLj_L~.
3 ~
-Route detector
cables as required--
180" 3 places
--'~dQ N~m
waste forms
Access tube
Moisture cup
5 places -
312.1
, e/ ~ ..-'I ~1"3 cm
"'~'7" -7"~'// I "Moisture/temper

91.4 cm t probe--3 places
38.1 cm I . .
89
] 8-339S
FIG. I--Detail of lysimeter vessel with component locations.

The term performance assessment (PA) is almost all-inclusive since it appears to cover all
those steps taken to develop an understanding of the risk associated with disposal of nuclear
waste [5]. What is evident, however, is that there is an underlying need for a PA to be based
on "hard" data in order to predict reasonably the overall performance of solidified waste. Such
data should reflect waste form stability in the context of a burial site source and indicate how
that source relates to the surrounding environment. With the above in mind, data obtained
from the ANL and O R N L lysimeter networks is discussed, and an example of its use as a data-
base for code input purposes is demonstrated. The sources of data are those obtained from the
site DAS and the radiochemical analysis of water samples from each lysimeter. Examples of
DAS output are provided to acquaint the reader with how monitoring data were obtained, and
together with-results of radiochemical analysis, they are presented as reduced data in figures
or tables. Data from the radiochemical analyses for the preceding 48 months are presented in
a cumulative format.
Weather and Sod Data

Precipitation, mr temperature, wind speed, and relative humidity are recorded by the A N L
and O R N L DAS d u n n g each twelve-month reporting penod Patterns of precipitation for the
sites can be obtained from the overall cumulative preopltatlon (Fig 2) Over the course of the
.study A N L had a cumulative average of 97% of the normal preopltatlon O R N L cumulatwe
precipitation has averaged 85% of total It ~s normal for A N L to expenence periods of freezing
temperatures from mid-November until near mid-March (Fig 3) Oak Ridge has had only
occasional days where the air temperature is as low as 0*C However, the sod at neither site has
been frozen at the depth of the waste forms Figure 4 provides an example of a typical yearly
cycle for soft temperature
Output from the moisture probes in the soil horizons of the lys~meters at both A N L and
O R N L has been slowly deteriorating dunng the past 24 months of operation However, data
recorded over the penod indicate that the lysimeter soil columns at both sites have remained
moist The actual momture of the soft column m each lyslmeter IS determined gravlmetncally
once each year. Soil moisture (as gravlmetrlcally determined) at each sampling depth has
remained uniformly consistent between mtraslte lyslmeters during the past several years (Figs.
5 and 6).
The umformlty of soil moisture in the A N L lyslmeters (Fig 5) is somewhat surpnsmg gwen
the long-term decrease m water infiltratmn m A N L 1 and 2 and now 4 (Fig 7) Action to
improve drainage of these two lysimeters was begun in July 1987, however, what has now
become obvious is that the inmal rate of drainage cannot be restored This is due to the present
structureless condition of the soils Loss of structural definmon is c o m m o n m disturbed softs
The problems assocmted with structure might be indicative of those that could occur ff a dm-
posal trench were constructed in this soft.
500
450 9 ANL-E /
400
350
E
v 300
250
200
150
100
50
I I I I I I I I I I I I I I I
0 200 400 600 800 1000 1200 1400 1600

DAYS SINCE SUMMER 1985
FIG 2--Cumu&ttveamountsofprectpltatton atANL and ORNLlyslmetersUes
55
50
25
20
15
10
Od
Ld 5
(3-
1
I/v
Ld
0
-5
-10
-15
-20 i i i i i i i i i
JUL AUG SEP OCT NOV DEC JAN FEB MAR APR MAY JUN
MONTHS 1 9 8 8 - 8 9
FIG. 3--Example of recorded data for site air temperature.
24
..,,,
22 f-,,,, ,
20
18
16
14
12
rY
Ld
O- \ I //
10
p_
8
6 ',.;, \. f
0 i i i i i i i i [ i
JUL AUG SEP OCT NOV DEC JAN FEB MAR APR MAY JUN
MONTHS 1 9 8 8 - 8 9
Copyright by ASTM Int'l (all FIG. 4--Example
rights reserved); of23:07:03
Mon Feb 24 recorded data for site lysimeter soil temperature.
EST 2020
ROGERS ET AL. ON LYSIMETER DATA 455
[ l l l r 0
I] ] ] ] ] ~ ] ] ] ] ] ] ] -0
L l l l f e l e l f f f f f , LID
--p..
I
I g W g I g ~ * ' I L ~ I I I 0
--LID
I] ] ] ] ]~ ~ ~ ] ] ] ~ ] ~
I 1 1 1 1 1 1 1 1 1 1 1 1
I ] ] ] ] ] ] ] ] ] ] ] ]
l l l l l l l l l l l l l l ,
I I I * I I I I # I # I # # --LID
i
'" ] ] ] ] ] ] ~ ] ] ~ ] ]
o ~
--LID .~
'~ ] ] ] ] ] ] ] ] ] ] ] v "~
w ~
k~
0 ~
--U~ 0 ~
[~1111111111~ii
Go
~o
Z ~
m ~
r / ,~ ~, _~ 0 ~
'i4{{444{{{{{i W ~
IN. \1~
_o E "~
c~
i I l l l l l f l l l l l l l
[ I g g g f f f g g f f W f g f
t ~ \ \ \ \ x \ \ x \ ~ \ \ \ l
d2
l l l l z # * # * z l l l l r
I I
0 LID
3~DISIOM 710S %
i / / / / ~ / / ~ / / / / / / I o
_ix3
" i ? ? ? 5 ? ? ? ? ~ ? ? ? ?1
l i l / i i l / / / I / I / / ix)
c'~l
I I ] ] ] ~ ) ] ] ] ] ] ] ] ] ~
I g i / / / l / l l l l l l l l l o
_0
i 3 5 5 ? 5 5 ? 5 3 " " \1
i ~ l l l l l l l l l l l l l
i i i i i i / / I / / / / / I o ~
'lh h ? ? ~ ? ? ] ] . . . . I -u3 .~
I I/ / / / / / / / / / !
L \ \ \ \ \ \ ~ \ \ \ \ J
-J
c~
o~
I ~ l / l l l l l l l l i l l
I I I I I I I I I I I I I I I I
i~ h 5 5 "5 5 ? 5 5 5 5 5 3 5 ?
l i i i l l t l / ' l l l l
7 ~
I I I I I 1 ~ 1 1 / 1 1 1 1
I ] ? ] ] ] ] ] ] ? . . . .
r l / l l / l l / l l / i / / / /
I~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ ~ 7
I
o~
r,.. &
I i / f f f f ( f ( f f f / / / ,
I h ~ h h h h h h h h h ~ h
1 1 3 5 ) ) h ) 3 h 3 h ) ) 3
i 1 ( / ( ( ( ( ( ( ( / / / /
I h h h h h h h h h h h h
I / l i d
<~ , ,q
~ ( ( ( ( Z / / z / /
m J
I I
O O
3EIr'llSIO~ qlos %
ROGERS ET AL. ON LYSIMETER DATA 457
2.8
2.6
2.4 9 L1
+ L2
2.2
o L3
2
,', L4
1.8
x L5
1.6
w~
d O 1.4
#_
v
1.2
0.8
0.6
0.4
0.2
0 I
0 2 0t0 , 4 0,0 , 6 0t0 , 8 0,0 , 1 0 0~ 0 ' 12'00 ~ 1 4 0' 0 1600

DAYS SINCE 8/I/85
FIG. 7--Cumulative volume of leachate obtainedfrom ANL lysimeters.
As is apparent from data presented in Figs. 5 and 6, after initial wetting the vertical moisture
content within the lysimeter soil columns at each of the sites appears to have remained rela-
tively uniform. At the time of the last sampling, the average soil moisture of ANL-E soils was
55.7% of the soil moisture-holding capacity (MHC), while at O R N L this value was 35.9%.
By using the cumulative rainfall data from each site since the time the lysimeters were placed
in operation (Fig. 4), it is possible to calculate the volume of water that has been received by
the exposed lysimeter surfaces (6489.5 cm2). The cumulative volume of precipitation received
by each ANL lysimeter was 2111.7 L; at ORNL, this value was 2991.6 L. It appears that the
throughput is dependent on site conditions (period of time soil surface was frozen, amount of
precipitation received as snow, and so on) and lysimeter fill material. At ANL an average of
998.9 + 255.6 L, or 47.3% of total precipitation, passed through the soil lysimeters; for the
control lysimeter, this value was 2080 L, or 98.5% of available precipitation (Fig. 7). For
ORNL, the values were 2558 _+ 28.5 L (85.5%) for the soil-filled lysimeters and 2991 L
(100%) for the control lysimeter (Fig. 8). Not only does precipitation have more of an oppor-
tunity to move into the O R N L lysimeters (an observation made by comparing the control
lysimeters at each site), but the O R N L soils are more permeable than the ANL soils (an obser-
vation made by comparing the control lysimeter at each site with that site's soil lysimeters).
The total volumes of precipitation that have moved through the lysimeters represent an aver-
age of 1.40 pore volumes for the ANL soil lysimeters and 3.60 pore volumes for soil lysimeters
at ORNL, while the control lysimeters at ANL and O R N L were 3.57 and 5.34 pore volumes,
respectively.
Radionuclide Analysis
Four times over a twelve-month period water samples are taken for beta and gamma anal-
ysis. The cumulative
Downloaded/printed by amounts of nuclides as determined in water samples obtained from
mois-
2.8
2.6
2.4 9 L1
+ L2
2.2
o L3
2
,6 L4
~ 1.8
X
n'o 1.6
~o t.4
~ 1.2
0.8
0.6
0.4
0.2
0 I I I I I 1 I I I I I I [ I I
0 200 400 600 800 1000 1200 1400 1600

DAYS SINCE 7 / 1 / / 8 5
FIG. 8--Cumulative volume of leachate obtained from ORNL lysimeters.
ture cups numbered 3 and leachate collectors for all sampling periods are displayed graphically
in Figs. 9 through 16.
What is apparent from these data is that not all nuclides are appearing consistently in the
water obtained from the moisture cups and leachate collectors. The nuclide that appears with
the most regularity at both sites is Sr-90. Consistent significant occurrences, however, have
been observed in all the No. 3 cups at ANL, the No. 3 cups at ORNL, except 4-3 (though it
was found in this cup during the last two samplings), and the No. 5 leachate collectors at both
sites (Figs. 9 through 12). There continue to be standout amounts of Sr-90 retrieved from mois-
ture cup samples at both sites. Those include 270 090 pCi from Cup 3-3 at A N L (Fig. 9) and
17 285 pCi from Cup 1-3 at O R N L (Fig. 10). The amount of Sr-90 leached from the A N L Cup
3-3 represents about 0.001% of the waste form inventory.
At ANL, Sr-90 retrieved from Cup 3 of the soil lysimeter ranges from 74 to 3200% of that
found in the leachate collectors, while at O R N L these values range from 0.1 to 70%. These
percent values have been increasing from year to year and could be a result of both an
increased quantity of St-90 moving into the area near the cups and a decrease in movement
of that radionuclide through the entire soil profile into the leachate collectors. This is evident
in that during the most recent twelve months, only the leachate water from the control
lysimeters and none from the soil lysimeters at each site contained significant amounts of Sr-
90. This is comparable to the trend seen in previous years and is in sharp contrast to the No.
3 cups data, which continue to demonstrate that substantial amounts of Sr-90 are still being
released from the waste forms. The percent of total Sr-90 being measured in the leachate col-
lectors and the No. 3 cups is somewhat inconsistent between the two sites. This could indicate
differences in waste form performance at the two sites. These data are interesting because the
waste forms at each site have been experiencing similar exposure to moisture and temperature.
ROGERS E T AL. O N L Y S I M E T E R DATA 459
280
26O
24O
moisture cups
22O
9 1.3
2OO + 2.5
180 o 3.3
16O a 4.3
I c
~g 140
x 5.3
~o
120
100
8O
6O
4O
2O
0 I 9 I i
0 2O0 400 600 800 1000 1200 1400 1600
DAYS SINCE 8 / 1 / 8 5
FIG. 9 - - A N L cumulative Sr-90 collected in moisture Cup 3.
18
17
16
15 moisture cups
14 9 1.3
13 + 2.5
12 o 3.3
11 A 4.3
o%"
10 x 5.3
I c
9
'-- o
cD.c: 8
7
6
5
I
I
) 200 400 600 800 100 1200 1400 1600
FIG. I O - - O R N L cumulative Sr-9O collected in moisture Cup 3.
340
320
300
leachate c o l l e c t o r
280
9 1
260
+ 2
240
o 3
220
,x 4
o'~" 200
O7"O x 5
Lg 180
00 160
140
120
100
80
60
4O
2O
0 , ,- , , =-r-~ "--'~ I
200 400 600 800 1000 1200 1400 1600

FIG. l 1 - - A N L cumulative Sr-90 collected in lysimeter leachate collectors.
80
7O
leachate c o l l e c t o r
60 9 1
+ 2
50 o 3
z~ 4
I c
x 5
4O
oo
0.~--
30
20 ~ c o ~..11..,--0 o o
10 x, ~ .~ .?. r A
I I I I I I I I I I I I I I I
200 400 600 800 1000 1200 1400 1600

FIG.
Copyright by ASTM Int'l (all rights1reserved);
2 - - O R Mon cumulative
N L Feb Sr-90
24 23:07:03 EST 2020 collected in lysimeter leachate collectors.
ROGERS ET A L ON LYSIMETER DATA 461
1.8
1.7
1.6
1.5 moisture c
1.4 9 2.3
1..3 + 5.3
1.2
1.1
~ 0.9
~5-c 0.8
0.7
0.6
0.5
0.4
0.3
0.2
0.1
0 I I 9 i ~ i ' i ' i ' '1 1 i i 1 i i i i
0 200 400 600 800 1000 1200 1400 1600
FIG. 13--ANL cumulative Cs-137 collected in moisture Cup 3.
110
100
90
leachafe collector
9
+
1
2
/
80 o 3
A 4
70
x 5
T~, 60
(j.-~
_ o 50
40
30
20
10
0 il -- I I I I I I ~ ~ [I I = P !
4'oo 200 6'0o 8'0 1000 1200 400 600

Copyright by ASTM Int'lFIG. 1 4reserved);
(all rights - - O R NMon
L cumulative
Feb 24 23:07:03 Cs-137
EST 2020 collected in lysimeter leachate collectors.
8O
70
leachate collector
60 9 5
5O
T~
S o 40
3O
/
m
20-
10-
0 I "i ~ i i i i i i i i l i i i
0 200 400 600 800 1000 1200 1400 1600

FIG. 1 5 - - O R N L cumulative Sb-125 collected in lysimeter leachate collectors.
280 m . . - . .
260
240 moisture
220 9 5.5
200
db
180
7 160
140
120
I00
80
60
40
20 i t i [ i r ~ i I i I p i I
200 400 600 800 1000 1200 1400 1600

Copyright by ASTM Int'l (all rightsFIG. 1 6 Mon
reserved); - - OFeb cumulative
R N24L23:07:03 EST 2020Sb-125 collected in moisture Cup 3.
Gamma-producing nuchdes have occurred with regularity at ANL but appear sporadically
at ORNL Cup 2-3 at ANL, below a cement waste form containing large amounts of Cs-137,
continually receives Cs-137 with a significant increase m quantities of this radlonuchde reap-
peanng after peaking m the February 1987 sample (Fig 13) Because there continues to be no
sustained occurrence ofCs- 137 m any of the ANL leachate water these data are not shown In
sharp contrast to past years, Cs- 137 is now being consistently detected in leach water from the
ORNL lysimeters (Fig 14) Breakthrough of Cs-137 into the ORNL No 5 leachate collector
occurred in November 1988, some seven months after its occurrence in the moisture cup
ORNL 5-3, and thus far a total of 100 096 pCi have passed through the lyslmeter In addition,
both ORNL 3 and 4 have had Cs- 137 occur in leachate collector water (Fig 14)
Although Sb-125 has not been found in water samples from ORNL Lyslmeter 5 since Octo-
ber 1987, it appeared again m ORNL 5 leachate collector samples beginning with the Novem-
ber 1988 sampling (Fig 15) It is now calculated that approximately 0 004% of the Sb-125
inventory from the ORNL 5 waste form has been recovered with 0 003% of that occurring
during the last year The nuclide no longer appears m the Cup 3 sample (Fig 16)
On an mtrasite comparison (Figs 11 and 12), the conclusion that the VES waste forms
(Lyslmeters 3 and 4) have released quantities of Sr-90 comparable to those lyslmeters contain-
ing cement waste forms (Lyslmeters 1 and 2) appears to be a valid conclusion since any move-
ment of Sr-90, or lack thereof, does not appear to correlate to either type of waste form Data
from the No 3 cups tend to support the evidence that VES is no better at retaining Sr-90 than
cement (Figs 9 and 10) Based on percent of total inventory release to No 3 cups, ANL-1 and
-2 (cement) have received 16 to 46 • 10-6%, respectively, of the Sr-90, while ANL-3 and -4
(VES) have received 16 to 800 )< 10-6%, respectively Comparable data at ORNL for the
cement are 94 to 96 E-6%, while those for VES are 1 to 17 E-6%
The data for Cs- 137 found m ANL Cups 2-3 and 5-3, and ORNL 5-3 as well as Sb- 125 from
ORNL-5, are of interest However, lack of Sb- 125 occurrence in other lyslmeters with the same
type waste forms makes it difficult to draw conclusions The continued appearance of Cs- 137
m ANL 2-3 and ORNL 5, as well as reappearance m ANL 5-3, would indicate that this occur-
rence is not an artifact These data, as well as those for Sr-90, continue to demonstrate the need
for long-term field testing of the present waste forms
Use of Lystmeter Data for Performance Assessment

It is becoming apparent through operational experience and cumulative data provided by
the NRC lysimeter array d u n n g the past four years that lysimeters are a valuable resource for
data applicable to calculating parameters used m developing site-specific performance assess-
ments The operational lyslmeters are providing continuous data from the near-field (that area
comprxsed of waste form and surrounding soil) m a number of areas that directly relate to
waste form stability
The relationship between Input parameters for codes and data derived from lyslmeter oper-
aUon were discussed m the introduction of the paper Code parameters listed m Table 1 are
those that initially appear to be the ones that can be either calculated from lyslmeter data or
derived from a knowledge of the lysImeter design They are common input parameters for
such codes as PATHRAE [6], PRESTO [ 7], and others that can be used to predict waste form
stability
Code parameters calculated from the current lyslmeter data are shown m Table 4 These
values were calculated for each of the lysimeters at ANL and ORNL The parameters for
annual percolation, vertical water velocity, inventory release, ra&onuclide concentration, and
retardation factor were calculated from cumulative data collected during the first 48 months
of lyslmeter
Copyright by ASTM Int'loperation as reported
(all rights reserved); m this
Mon Feb 24 23:07:03 ESTpaper
2020 The other parameters were calculated from
o~
0 ~ c-~ onr~ -- 0 c~ on ,~
oo I-~ ,~"
xo
~0
~n o~
I ~o CD
U~
< o~
h~
H
O
E
II
O
z
data on lyslmeter design and installation [8]. These data have not as yet been used for code
input, though it is apparent that they do provide a condensed summary of the lysimeter oper-
ation Another interesting point is that there is a natural varlabihty in the performance (annual
percolation, vertical water velocity, soil bulk density, etc ) of site soil lyslmeters even though
they are replicates of each other Such differences are not readily detected in unrephcated tests
or accelerated laboratory studies and could have important long-term effects on nuclide
movement
Other parameters that will require data from subsequent years of operation include calcu-
lating rates of waste form degradation, mobility, and dissolved or precipitated minerals
Acknowledgment
Appreciation is expressed to S A Rawson for her assistance m demonstrating how lysimeter
data can be used for code parameters
Notice
This report was prepared as an account of work sponsored by an agency of the United States
Government Neither the United States Government nor any agency thereof, or any of their
employees, makes any warranty, expressed or implied, or assumes any legal liability or respon-
sibihty for any third party's use, or the results of such use, of any information, apparatus, prod-
uct or process disclosed in this report, or represents that its use by such third party would not
infnnge pnvately owned nghts The views expressed in this report are not necessanly those of
the U S. Nuclear Regulatory Commission
References
[1] ~~ CFR 6 ~~``L1censmg Requ1rements ~f Land D1sp~sa~ ~f Radl~act1ve Waste~ C~de of Federal Reg-
ulatlons, Office of Federal Register, Washington, D C
[2] "Radioactive Waste Management," Order 5820 2A, United States Department of Energy, Washing-
ton, DC, 26 Sept 1988
[3] "Technical Position on Waste Form," United States Nuclear Regulatory Commission, Washington,
DC, 11 May 1983
[4] Rogers, R D, McConnell, J W, Jr, Davis, E C, and Fmdlay, M W, "Field TestmgofWaste Forms
Containing EPICOR-II Ion Exchange Resins Using Lysimeters," NUREG/CR-4498, EGG-2438,
Nuclear Regulatory Commission, Washington, DC, June 1986
[5] Case, M J and Otis, M D, "Guidelines for Radiologlcal Performance Assessment of DOE Low-
Level Radioactive Waste Disposal Sites," DOE-LLW-62T, Department of Energy, Washington, DC,
July 1988
[6] PATHRAE-EPA "A Low-Level Radioactive Waste EnvironmentalTransport and Risk Assessment
Code," EPA 520/1-87-028, Environmental Protection Agency, Washington, DC, Dec 1987, devel-
oped by V Rogers and C Hang
[7] PRESTO-EPA-POP "A Low-Level Ra&oactlve Waste EnvironmentalTransport and Risk Assess-
ment Code," EPA 520/1-87-024-1, December 1987, EnvironmentalProtection Agency,Washington,
DC, developed by D E Fields, C A Little, F Parrage, V Rogers, and C Hang
[8] Rogers, R D, McConnell, J W, Jr, Davis, E C, and Flndlay, M W, "Field Testing of Waste Forms
Containing EPICOR-II Ion Exchange Resins Using Lyslmeters," NUREG/CR-4498, EGG-2438,
Nuclear Regulatory Commission, Washington, DC, June 1986
Jeffrey L. Means, z Jeffery C. Heath, 2 and William J. McLaughlin 3
Laboratory Treatment and Field

Demonstration Lessons Learned from a
Waste Stabilization Project
REFERENCE: Means, J L, Heath, J C, and McLaughhn, W J , "Laboratory Treatment and
Field Demonstration Lessons Learned from a Waste Stabihzation Project," Stabthzatlon and
Sohd~catzon of Hazardous, Radloacttve, and Mtxed Wastes, 2nd Volume, ASTM STP 1123, T
M Gdham and C C Wiles, Eds, American Society for Testing and Materials, Phlladelphm,
1992, pp 466-478
ABSTRACT: For a ten-year period, 1976 to 1986, the U S Navy leased most of~ts naval statmn,
Treasure Island, Hunters Point Annex (HPA), to Triple A Machine Shop, which operated the
station as a commercial ship repair facility Triple A generated waste sandblasting grit contam-
inated wath toxic heavy metals (1 e, lead and copper) The grit accumulated in a 2300-m 3 (3000-
yd 3) pile This three-phase study evaluated the effectiveness of a sulfide-based chemical fixation
process m rendenng the gnt nonhazardous under U S Environmental Protection Agency and
State of Cahfornla hazardous waste rules and regulatmns
Two s~gnlficant lessons resulted from the study First, the unsatisfactory degree ofstablhzatlon
of the lead and copper during the field demonstration was a direct result of their encapsulation
in orgamc coatings (antffouhng agents, p~gments, etc ) Understanding the phys~cochemical
form of the contaminant is as ~mportant as knowing the type and amount of the contaminant
present Second, sulfide-based formulations used in the laboratory enhanced the degree of sta-
blhzation over that which occurred m the field Dunng laboratory treatability stu&es, it is impor-
tant to cure the waste m a setting that s~mulates the conditions that wdl be encountered in the
field
Sulfide binders, as shown in this study, have a role m stabdlzatmn/soh&ficatlon where con-
ventmnal technologxes are unable to meet performance criteria
KEY WORDS: stabxhzatlon, soll&ficatlon, lead, copper, sandblasting grit, leaching data, sam-
phng, statistical analys~s, treatablhty
F r o m 1976 to 1986, the U S N a v y leased most of its naval station, Treasure Island, Hunters
Point A n n e x (HPA), to Triple A Machine Shop, which operated it as a c o m m e r c i a l ship repair
facility Triple A's shipyard corrosion control operations generated waste sandblasting grit,
which was deposited in the industrial landfill area o f the shipyard An approximately 2300-m 3
(3000-yd 3) pile of waste sandblasting grit present at the facility is the subject o f this study
The overall objective o f this project was to evaluate the effectiveness of a sulfide-based chem-
ical fixation process in rendering the sandblasting g n t c o n t a m i n a t e d with toxic heavy metals
nonhazardous under U S E n v i r o n m e n t a l Protection Agency (EPA) and State o f California
hazardous waste rules and regulations The project was divided into three phases Phase I was
a sampling and analysis activity to identify the target c o n t a m i n a n t s Phase II was a laboratory
Department manager, Enwronmental Technology, Battelle Columbus Operations, Columbus, OH

43201-2693
2 Project officer, Naval Clwl Englneeilng Laboratory, Port Hueneme, CA 93043-5003
3 Owner, ToxCo, Inc
466

MEANS ET AL ON WASTE STABILIZATION 467
demonstration activity to develop and verify the effectiveness and smtablhty of a chemical
stablhzatlon process p n o r to treating the 2300 m 3 of grit Phase III was the field treatment
demonstratmn of the bench-scale-tested treatment technology on the 2300-m 3 waste pile.
After treatment, the treated grit was to be disposed of in a municipal landfill as nonhazardous
waste
Phase I: Sampling and Analysis

Sampling
The field site consists of an accumulated pile of sandblasting grit situated on a cleared soil
area The pile is approximately 18 m (20 yd) wide by 41 m (45 yd) long, m the approximate
shape of a kidney bean, and is about 3 m (9 ft) high with a relatively flat top The pile has been
covered with a tarpaulin to reduce dust emissions and the infiltration of precipitation through
the pile
Because of possible variations in metal concentrations in different parts of the pile, samples
were collected from random locations and as a function of depth In November 1988 the pile
was gridded into equal surface areas by marking a coordinate every 2 m (6 ft) This resulted in
208 grids having surface areas of approximately 3 m 2 (36 ft 2) each (Fig 1) The grad areas were
numbered consecutively so that sample locations could be referenced Then sample grid num-
bers were selected from a random number table for each sampling location Twenty-four dif-
ferent samples were collected along with two bhnd rephcates Eight samples were collected at
each of three depth intervals (a) 0 to 1 m (0 to 3 ft, avoiding the top 8 cm or 3 m ), (b) 1 to 2
m (3 to 6 ft), and (c) 2 to 3 m (6 to 9 ft)
Samples were collected using a stainless steel shovel or sand auger, depending on depth A
portion of each sample was archlved for possible future use Samples were split in the field
using a riffle-type sphtter, placed in precleaned polyethylene (I-Chem Corp ) bottles, and
immediately shipped under chain of custody to the appropriate laboratory for analysis or stor-
age The splitter, sampling devices, and mixing tray were cleaned between each sample A field
quality control (QC) blank was collected using clean sand to verify that cross-contamination
was not occurring Permanent field notebooks were maintained w~th proper documentation
Analysts of Untreated Samples

As mentioned above, 24 different samples (plus two bhnd replicates) from three different
depth horizons were collected. Portions of selected untreated samples were analyzed for total
Cahforma Assessment Manual (CAM) and soluble (Waste Extraction Test leaching test) met-
als uslngState of California protocols Selected samples were also analyzed using the EP Tox
(U S EPA) leaching methodology All California tests were conducted by a California-certified
analytical laboratory
The results of the total metal analyses are provided in Tables 1 and 2 and lead to the follow-
lng conclusions" (a) with one exception, only copper and lead were present m concentraUons
theoretically suificient to exceed their soluble threshold limit concentrations (STLCs), and (b)
there was no statistically significant difference in the concentrations ofe~ther copper or lead as
a function of depth Note that c h r o m m m analyses of two samples suggest that chromium (Cr)
is present mostly (>90%) in the tnvalent form (Table 1) Therefore, Cr (VI) concentrations
are below STLC The exception mentioned above was nickel in one sample, but when WET-
tested, the nickel concentration was well below its STLC In addition, three samples were ana-
lyzed for p H and reactive sulfide (Claussen methodology) Average pH of the untreated grit
was by
Copyright 8 ASTM
6, and
Int'lreactive sulfide
(all rights reserved); Monwas
Feb 24<23:07:03
3 ppm EST 2020
46B HAZARDOUS
WASTES
, 3 9 s 6 7 ~ 1
l
•
17 18 ~ 20 21 22 23 24 "~ 1
iiiii~i ~ N
if 25 26 27 28 29 30 :i:i~1"1"i:i:i 32
3 34 35 36 37 38 39 40 41 SurfaceSamples
43 45 ~ ~ 48 49 50 51 Mid-Level Samples
$2 . ~ ss iii:~:::i!ii s? . s~ Lower-LevelSamples
62 63 64 65 66 67 68 69 70 71
72 73 74 75 76 77 78 79 80 \ 81~ 82
i \\\\'~
e3 84 es e6 87 as 89 90 91 92
'~9s
95 96
96 97 9~ 99 100 ~01 I0:, ~03 I04 I0s 106 \
\
107 108 109 110 111 112 113 114 115 116 117
~::::::::::::::::! \
~ 118 119 :~'1"20'~ 121 122 123 '~124 ~ 125 126 127 128 129/
.~.-. :..: ~ \ \ \ -~ ~ \ \ ' ~ ...... ~..~
130 131 ~132~ 133 134 ~13.5~136\ 137 138 139 140 ~141~
142 143 144 145 146 147 148 149 150 ~1 152
\\\\~
154 155 156 157 158 159 160 161 162 163
;:;:;:::.;:::i:i~
164 ::i~1:6"5::!::i 16G 167 168 169 170 ~ 172 173
:: 175 176 177 178 179 180 181 182
184 ::~:11"85::~i 186

187 188 189 190 "~:lti t
192 193 194 195 196 197 198
FIG. 1--Sampling locations on the waste pile,
oo~o~ o~ oo~
;>
VV VV VV VV
<
VVVV VVVVVVVV
VVVVVVVVVVVVV
0
vv~v V~vVv
VVVVVVVVVVVVV
- - ~ ~ ~R
VVVVVVVVVVVVV
I
II
2
<
vv~ ~vv~vv~
E
II
TABLE 2--Analytical data on untreated Hunters Pomt samples,

Nos 13 to 27
Metals, mg/kg
Sample
Identification Cu Pb
HP-13 1900 240

HP-14 1900 170
HP-14 Dup 1600 180
HP-15 1600 330
HP-16 1700 170
HP-17 1800 200
HP-18 1600 150
HP-19 2000 250
HP-20 1600 86
HP-21 2100 180
HP-22 2000 200
HP-23 1600 180
HP-24 1700 210
HP-25 2600 190
HP-26 1700 170
HP-27 12 <5
a Samples codes are as follows

1 Samples 1 to 8 and 25 collected from the 0- to l-m (0- t o
3-ft) level [after ehmmatlng the top 8 cm (3 in )] Sample
25 is a blind rephcate of Sample 1
2 Samples 9 to 16 and 26 collected from the 1- to 2-m (3-
to 6-ft) level Sample 26 is a blind replicate of Sample 10
3 Samples 17 to 24 collected from the 2- to 3-m (6- to 9-ft)
level
4 Sample 27 is a blank [reagent slhcon dioxide (SIO2) sand]
Average copper, lead, nickel, a n d zinc c o n c e n t r a t i o n s plus s t a n d a r d d e v i a t i o n s are s h o w n

in T a b l e 3, c o m p a r e d w i t h t h e i r total t h r e s h o l d l i m i t c o n c e n t r a t i o n ( T T L C ) values
T h e 14 s a m p l e s with t h e highest c o p p e r a n d lead c o n c e n t r a t i o n s were t h e n extracted accord-
ing to C a l i f o r n i a W E T m e t h o d o l o g y , a n d t h e extracts were a n a l y z e d for c o p p e r a n d lead
T w e l v e s a m p l e s were also a n a l y z e d using t h e E P T o x test Average W E T a n d E P Tox c o p p e r
(Cu) a n d lead (Pb) c o n c e n t r a t i o n s a n d s t a n d a r d d e v i a t i o n s are c o m p a r e d with t h e i r t h r e s h o l d
values m T a b l e 4 N o t e t h a t c o p p e r IS n o t a n E P T o x m e t a l
T h e c o n c e n t r a t i o n s o f b o t h m e t a l s exceed t h e i r STLCs, a n d the grit is therefore h a z a r d o u s
TABLE 3--Average total metal contents of Hunters Pomt

sandblasting grit samples
Average, Standard Dewatlon, TTLC,

mg/kg mg/kg mg/kg a
Copper 1830 • 392 2500

Lead 202 • 63 1000
Nickel 82 • 70 2000
Zmc 1017 • 761 5000
a TTLC = total threshold limit concentration

TABLE 4--A verage soluble metal contents of Hunters Point

sandblastmg grit samples
Standard
Average, Deviation,
mg/L mg/L STLC, mg/L a
WET copper 147 + 27 25

WET lead 19 + 6 5
EP Tox copper 22 + 4 not apphcable
EP Tox lead 06 + 04 5 (EPA standard)
a STLC = soluble threshold limit concentraUon
in C a h f o r n l a . H o w e v e r , this grit w o u l d n o t be consxdered h a z a r d o u s by the U S. E P A based

o n t h e results o f t h e E P T o x test U n a v e r a g e d d a t a f r o m b o t h the W E T a n d E P T o x tests are
p r o v i d e d m T a b l e 5 A n a l y t i c a l m e t h o d s used to collect t h e d a t a are listed In T a b l e 6 C A M
total a n d soluble m e t a l limits are c o m p i l e d m T a b l e 7
Phase II: Bench-Scale Treatability Study

T h i s section reports t h e results o f t h r e e different c h e m i c a l fixation p r o c e d u r e s tested at
b e n c h - s c a l e o n selected s a n d b l a s t i n g grit s a m p l e s f r o m H P A T h e bench-scale t r e a t a b l h t y
study was c o n d u c t e d by T o x C o , I n c , a C a h f o r m a - b a s e d stablhzat~on c o n t r a c t o r , with b o t h
N a v y a n d Battelle witnesses Battelle p r o v i d e d c h a i n o f custody for samples b o t h to a n d f r o m
t h e p r e m i s e s o f t h e stabd~zat~on c o n t r a c t o r .
TABLE 5--Results of leachmg tests on selected untreated

Hunters Pomt samples
WET, mg/L EP Tox, mg/L
Sample Cu Pb Cu Pb
Hp-I 150 18 25 2 0 53
HP-I Dup 25 2 0 52
HP-2 150 18 29 6 0 52
HP-3 130 35 19 9 0 22
HP-9 160 19 24 3 0 74
HP-10 100 10 15 1 0 19
HP-II 160 17 22 8 0 51
HP-12 200 17 25 1 0 41
HP-14 120 19 12 3 0 17
HP-14 Dup 120 21 12 4 0 21
HP-15 130 24 19 2 1 57
HI~-~ 180 21 22 3 0 71
HP-22 160 15 23 7 0 54
HP-24 150 16 23 3 0 58
HP-25 150 16 NA a NA a
HP-25 Dup 170 21
HP-26 100 11 NA a NA a
HP-26 Dup 120 32
a NA = not analyzed
TABLE 6--Methods used for chemtcal analyses
TTLC--Cahfornla Title 22 CCR 66699

STLC--Callforma Title 22 CCR 66699
EP Tox--EPA 1310
M E P - - E P A 1320
p H - - E P A 9045
Reactwe Sulfide--SW-846, Part If, Section 7 3
METALS, TTLC, STLC, EP Tox, AND MEP
Barium EPA 7080
Antimony EPA 7040
Arsemc EPA 7061
Barium EPA 7080
Berylhum EPA 7090
Cadmmm EPA 7130
Chromium, Total EPA 7190
Chromium (VI) EPA 7196
Cobalt EPA 7200
Copper EPA 7210
Lead EPA 7420
Mercury EPA 7471
Molybdenum EPA 7480
Nickel EPA 7520
Selemum EPA 7741
Sdver EPA 7760
Thalhum EPA 7840
Vana&um EPA 7910
Zinc EPA 7950
TABLE 7--Cahfornta Assessment Manual metal hmlts,

Cahfornta Admmtstrattve Code, Tttle 22, Section 66699
Element STLC, mg/L a TTLC, mg/kgb
Antimony 15 500
Arsenic 50 500
Ban um 100 10 000
Berylhum 0 75 75
Cadmium 10 100
Chromium, Total 560 2500
Chromium (VI) 50 500
Cobalt 80 8000
Copper 25 2500
Fluoride 180 18 000
Lead 50 1000
Mercury 02 20
Molybdenum 350 3500
Nickel 20 2000
Selenium 10 100
Silver 50 500
Thalhum 70 700
Vanadmm 24 2400
Zinc 250 5000
NOTE pH should not exceed 12 5 or be lower than 2 0, and

reactive sulfide should not exceed 500 mg/kg
a STLC = soluble threshold limit concentraUon
b TTLC = total threshold limit concentration
The performance criteria were as follows
1 WET Pb < 5 mg/L

2 W E T C u < 25 mg/L
3 p H < 125
4 Reactive sulfide < 500 mg/kg
Several &fferent bench-scale treatabd~ty tests were conducted, each w~th unsatisfactory
results The problem was traced to the physlcochem~cal form of the Pb and Cu m the grit Both
were found to be encapsulated m orgamc polymeric coatings, Pb most hkely m paint p~gment
compounds and C u m antffouhng compounds When the complex phys~cochemlcal forms of
Pb and Cu were understood, devising a successful bench-scale treatabd~ty study was then pos-
sible Three processes were tested All three processes used a sulfide-fixing agent in combin-
ation with a Sdlcate-settmg agent In Process No 1, the setting agent was a combmatmn
of fly ash and cement Process No 2 used fly ash only, and Process No 3 used bentomte
only
Soon after treatment, the waste has a rather moist texture and resembles a wscous slurry
Several days after stabd~zatlon and cunng, the treated sandblasting grit has a texture and con-
s~stency very much resembhng that of the untreated grit
The treated samples were allowed to stabd~ze and cure for two to four days The samples
were then analyzed for pH, reacUve sulfide, and WET Pb and Cu by a Cahfornla-certlfied ana-
lytical laboratory Averaged results of the chemical analyses (except for reactxve sulfide, which
is shown as a range) on the treated samples are shown m Table 8 by process Unaveraged chem-
ical data and 95% confidence intervals are presented in Table 9
All samples tested using all three processes passed the WET test for Cu and Pb, had a pH
below 12 5, and tested below the reactive sulfide guldehne of 500 mg/kg Therefore, any of the
three processes were consadered to be saUsfactory for the field demonstration Processes Nos
2 and 3 were preferred over Process No 1 because of the need to fix fly ash and cement (thus
entmllng the use of an addmonal hopper) m the latter Process No 2 was used in the field
Also, to assess the long-term stablhty of the treated waste, a multiple extraction procedure
(MEP) was conducted on three samples of th~s waste treated by a similar process The results
of this test showed no significant release of either lead or copper (Table 10) dunng the ten-day
testing penod Addmonal stu&es, both theoretical and experimental, are being undertaken to
understand better the long-term stablhty of metal sulfides and the potentml applicability of
sulfide-based stabdlzatlon/sohdlficatlon (S/S) processes in remedlatlon
TABLE 8--A veragedperformance datafrom bench-scaletreatabthty studtes
WET Pb, mg/L WET Cu, mg/L a Lab pH ReacUve Sulfide, mg/kg
PROCESS NO l
<0 4 <0 1 12 1 < 3 to 20
PROCESS NO 2
<0 4 <0 1 12 0 < 3 to 65
PROCESS NO 3
<04 <0 1 11 2 < 3 to 300
T A B L E 9 - - R e s u ~ o f l e a c h m g t e s t s o n b e n ~ scale-treatedHuntersPomtsamples
Sample WET Cu, mg/L WET Pb, mg/L pH Reactive Sulfide, mg/kg
PROCESS NO 1
HP-7 <01 <04 126 20
HP-13 <01 <04 120 <3
HP-17 <01 <04 117 <3
HP-19 <01 <04 120 <3
104 10
PROCESS NO 2a
HP-7 <01 <04 123 <3
HP-13 <01 <04 123 <3
HP-15 <01 <04 116 <3
HP-19 <01 <04 119 <3
HP-21 <01 <04 117 65
HP-22 <01 <04 121 <3
02 21
PROCESS NO 3a
HP-17 <0 1 <0 4 10 9 33

HP-23 <0 1 <0 4 10 8 <3
HP-24 <0 1 <0 4 11 8 300
06 186
a 95% Confidence interval
Phase III: Field Treatment Demonstration Field Activities

T h e full-scale d e m o n s t r a t i o n o f the t r e a t m e n t t e c h n o l o g y was c o n d u c t e d m D e c e m b e r 1989
after t h e a p p r o v a l o f a detailed w o r k p l a n by the U S N a v y a n d c o g n i z a n t regulatory agencies
I m p l e m e n t a t i o n o f t h e full-scale process was p r e c e d e d b y t h r e e pilot-scale tests o n approxi-
m a t e l y 15-m 3 (20-yd 3) b a t c h e s o f s a n d b l a s t i n g grit T h e m i x i n g device was a large double-screw
pugmlll, w i t h a 150 m 3 (200 yd 3) p e r h o u r capacxty W e o p e r a t e d t h e field d e m o n s t r a t i o n at
a p p r o x i m a t e l y 40 m 3 (50 yd 3) p e r h o u r t h r o u g h p u t T h e p r o p o r t i o n s o f the ingredients were
TABLE l O--Analyttcal results M E P test--Pb and Cu concentrations In sulfide-treated Hunters Point

sandblastmg grit
Posttreatment, Multiple Extraction Procedure

EP Tox,
Sample mg/L Dayl Day2 Day3 Day4 Day5 Day6 Day7 Day8 Day9
PB C O N C E N T R A T I O N S
HP-9 <005 <05 <005 <005 <005 <005 <005 <005 <005 <005
HP-I 1 <005 <05 <005 <005 <005 <005 <005 <005 <005 <005
HP-19 <005 <05 <005 <005 <005 <005 <005 <005 <005 <005
HP-19 < 0 05 < 0 05
CU CONCENTRATIONS
HP-9 032 006 003 003 002 001 001 001 002 002
HP-11 0 69 0 08 0 06 0 06 006 0 06 005 0 05 007 0 16
HP-19 0 17 007 005 004 005 003 003 004 004 007
HP-19 Dup 0 O4 0 03 0 04
Copyright
NOTE by ASTM
TheInt'lEP(all Tox
rights reserved);
test IS theMon first
Feb 24step
23:07:03
of EST
the 2020
multiple extraction procedure
varied shghtly between these three tests with the intent of identifying the most satisfactory mix
The results from these tests on samples collected immediately m the field and express-analyzed
by a local analytical laboratory showed that two of the three mixes passed all of the apphcable
performance standards Therefore, the decision was made to proceed with full-scale
demonstration.
After approximately one and a half days of field treatment, approximately 500 m ~(600 yd 3)
of sandblasting grit had been treated. However, field actlwtles were suspended because of inter-
nal matters between the Navy and the Cahfornm Department of Health Services Several
weeks after treatment, samples were collected from the treated waste pile and analyzed for the
four principal performance parameters The following sectmns briefly summarize the results
of these analyses
500 Cubic Meter Treated Pile, December 1989

Several samples of treated waste from the approximately 500-m 3 (600-yd 3) pile were chem-
ically analyzed (Table 11) The samples were collected as a function of depth The pile varied
m depth up to approximately 1 m (3 ft) The results of the first three samphngs (six samples
total) showed that Cu failed the WET test in two of six samples and Pb faded m five of s~x
samples The late January samples (two samples) were collected shghtly deeper m the treated
pde In these samples, Cu passed the STLC m both cases while Pb passed in one case The
February samples showed that the deeper samples were, in general, better stablhzed than the
shallow material
Clearly, some degree of Cu and Pb stabdlzatlon ~s evxdent in the field-treated samples The
TABLE 11--Data on the 500-m3 treated ptle
STLC, Cu, STLC, Pb, Reactive Sulfide,

Sample mg/L mg/L pH mg/kg
SAMPLES COLLECTED WEEK OF 18 DEC 1989

8 c m ( 3 m Depth) 93 44 13a <l
30 cm (12 in Depth) 10 11 13~ <1
SAMPLES COLLECTED WEEK OF 2 5 DEC 1989
8 cm (3 m Depth) 50 2 10 6 12 36 25 3
30 cm (12 m Depth) 42 0 10 3 12 37 175
SAMPLES COLLECTED 16 JAN 1990b
Sample 1 22 5 11 8 12 l <0 005
Sample 2 18 5 15 3 12 0 <0 005
SAMPLES COLLECTED 29 JAN 1990c
1320-3 13 4 30 12 2
1320-4 15 0 7 17 12 4
SAMPLES COLLECTED 2 0 FEB 1 9 9 0 d
I-3 65 8I 124 27
II-3 48 60 12 2 13
III-3 77 78 12 4 166
III-4 86 6l 12 l 19
apH rounded to nearest whole unit
b Collected from a depth of approximately 30 cm (12 in )
c Collected from a depth of approximately 30 to 61 cm (12 to 24 m )
a Collected
Copyright by ASTM Int'lfrom a depth
(all rights reserved);of approxlmately
Mon 612020
Feb 24 23:07:03 EST to 91 cm (24 to 36 In )
soluble Cu and Pb levels are lower than what can be accounted for by the small amount of
dilution that occurs from the stabilization chemicals and water However, the degree of sta-
bilization is not sufficient for the grit to be classified as nonhazardous in the state of Cahfornia,
especially in the upper parts of the treated waste pile
Laboratory Treata&hty Studies

Several laboratory treatabd~ty studies were conducted to help understand the cause of the
apparent failure, judging from the field data The first study compared the effect of the fly ash
that was used in the bench-scale test of June 1989 (the basis for the formulation used in the
field demonstration) with that of the fly ash that was used in the field demonstration The sta-
bdlzation contractor was unable to secure for the field demonstration the exact fly ash that was
used in the bench-scale treatabdity study We wanted to determine if using the different fly ash
could account for the lower degree of stabihzatlon observed in the field
The result (Table 12) was that there was no difference m the effect of the two fly ashes Both
resulted m satisfactory STLC Cu and Pb levels, pH, and reacuve sulfide as observed m the
bench-scale test of June 1989 Inodentally, both samples were cured m a j a r The pHs of the
two fly ashes were measured and were found to be within several tenths of a pH unit of each
other
Another treatabdity test evaluated the effect o f c u n n g the test samples on a gentle mchne m
the open air to better simulate field condmons Some excess stabdizatlon reagent was observed
draining off the grit pile, and the treated samples had a lighter color than the samples that cured
m the jar The results yielded WET Cu and Pb levels that were over a factor often higher than
the jar-cured samples for both metals (Table 12) These results strongly suggested that incom-
plete c u n n g was the cause of the lower degree of stabilization observed in the field
A final treatability test consisted of a similar series of tests as described above for the fly ash,
except that bentonite was used instead of fly ash Typical results (Table 12) shows that the
bentonite was more effective compared w~th fly ash m stabilizing the grit when cured m the
open air on an mchne However, the degree of stabilization was still not as great as when
the samples were cured in a j a r The bentomte appears clearly preferable to the fly ash in these
tests, but it is not clear whether the CAM STLC hmits would consistently be saUsfied in the
field application of this technology
Summary
The hypothesis for the higher-than-expected STLC-soluble Cu and Pb contents m some of
the field-treated grit samples is that the reaction between sulfide and metal ions was prevented
TABLE 12--Data from lab treata&hty studies conducted since 15 Dec" 1989
STLC, Cu, STLC, Pb, Reactive Sulfide,

rng/La mg/La pH mg/kg
COMPARISON OF 2 FLY ASHES CURED IN A CLOSED JAR
Lab Treatability Fly Ash 0 11 0 28 11 5

Field Treatability Fly Ash 0 13 0 13 12 0
CURED IN THE OPEN AIR ON A SLIGHT INCLINE
Lab Treatablhty Fly Ash 33 7 3 72 11 2

Field Demonstration Fly Ash 23 5 2 83 11 9
Bentonite 36
2 13 11 8
a STLC
Universidad = deSoluble
Industrial Santanderthreshold limit concentration
(Universidad Industrial de Santander) pursuant to License Agreement. No further reproductions authorized.
from going to completion because various environmental conditions caused the treated grit to
cure d~fferently in the field than in ajar One of the prlnopal factors appears to be the drainage
of the dense sulfide reagent away from the upper parts of the treated pile This effectively
stopped the stablhzatlon reaction before completion
In the laboratory, the samples were cured m the restricted environment of the sample jar,
and a higher degree of chemical reaction resulted An important complicating factor Is that
Cu and Pb in this grit are encapsulated in orgamc polymeric antlfouhng compounds and pig-
ments, which must be penetrated before the reaction between lomc metal speoes and sulfide
can occur From the bench-scale data, the caustic sulfide setting reagent appears to be capable
of slowly degrading these polymers, but the reaction stops when the reagent drains away
Note also that the soluble silicate binder was incapable of stablhzmg the grit in bench-scale
studies This is a commonly used binder, but was not applicable to Hunters Point grit because
of the grlt'S chemical complexity.
Lessons Learned
This paper addresses two of the more significant of the lessons learned
1 Comphcatlons due to physwochemlcalform of the target contaminant The unsatisfac-

tory degree of stabilization of the Cu and Pb d u n n g the field demonstration was a direct
result of their encapsulation in organic coatings of various types (antlfouhng com-
pounds, p~gments, etc ) Persons conducting stabilization/solidificationtreatability tests
frequently measure the type and a m o u n t of contaminant present But, In complex waste
forms such as the Hunters Point sandblasting grit, type and amount of contaminant are
not sufficient information Also ~mportant IS understanding the physlcochemlcal form
of the contammant
2 Jar effect during curing The jar environment promotes good contact between the binder
and waste form and clearly led to an enhanced degree of stabilization over that which
occurred m the field During laboratory treatablhty studies, c u n n g the waste m a setting
that simulates as closely as possible the condmons that will be encountered in the field is
important
Postscript
The data contained in this report might lead to the erroneous conclusion that the sulfide-
based stabilization technologies described hereto are ineffective To the contrary, there are
numerous instances where such technologies have been used successfully in the field, resulting
in products having stability xn an appropriate disposal environment Once the metal sulfide
salts are formed, they appear to be very stable for long periods of time unless exposed to acid
and oxidizing conditions simultaneously Free sulfide (unreacted with metals) left over after
completion of the S/S reactions appears to oxidize fairly readily upon aeration We are study-
ing the long-term stability of metal sulfides and the S/S conditions necessary to form metal
sulfides that have long-term stablhty m the environment
The sulfide-based formulations described in this report were selected after a number of other
conventional binders were clearly shown to be unsatisfactory for this waste form, which, as
indicated above, was difficult to stabilize because of the organic polymer coatings on the met-
als The sulfide binders have a role in stabilization technology where other conventional tech-
nologies are unable to meet the necessary performance criteria
Finally,
Copyright recent
by ASTM Int'l developments
(all rights in23:07:03
reserved); Mon Feb 24 the California
EST 2020 Department of Health Services policy appear
to make it possible to recycle the Hunters Pointtosandblasting
Universidad Industrial de Santander (Universidad Industrial de Santander) pursuant License Agreement. grit into
No further asphaltauthorized.
reproductions or other corn-
478 HAZARDOUSWASTES
posltes This policy, which is expected to be made official m 1991, would only apply to Cali-
fornia (not RCRA) wastes (a) that do not exceed the TTLCs, (b) whose WET-soluble metal
contents will be brought below the STLCs by dilution in the asphalting process, and (c) that
have a beneficial contrabutlon to the physical integrity of the composite Pending approval of
this new DOHS policy, the 2300 m s (3000 yd s) of sandblasting grit at Hunters Point will be
legally recycled into asphalt
J. C. N o m i n O ~ a n d A. B d l o n 2
Performance Appraisal of Full-Size

Radioactive Waste Packages
REFERENCE. Nomln6, J C and Blllon, A, "Performance Appraisal of Full-Size Radioactive
Waste Packages," Stabdtzatton and Sohdtficatlon of Hazardous, Radtoacttve, and Mixed
Wastes, 2nd Volume, ASTMSTP 1123, T M Gdham and C C Wiles, Eds, American Sooety
for Testing and Materials, Phlladelphm, 1992, pp 479-494
ABSTRACI" Before final storage m a subsurface s~te, each kind of encapsulated or lmmoblhzed
ra&oactlve waste has to be recogmzed as acceptable for this situation To reach this goal, French
radmacUve waste pohcy recommends an exhaustive charactenzatmn of radioactive packages
Complying w~th th~s regulation, each processing step, from the waste producUon to ultimate stor-
age 0dentlficatmn, sorting out, conditioning, transporting, storage, and so on) necessitates a
knowledge of specific properties Among them are physical and mechanical charactensUcs, con-
finement characteristics, and behavior m relatmn to the staNhty and the aging of the embedded
wastes To give an overview, examples of results w~th comments will be presented and compared
KEY WORDS" radioactive waste, charactenzatmn, apprmsal
Intermediate and low-level radioactive wastes are currently Immobilized in different matri-
ces as varied as hydraulic binders, bitumen, or polymers These materials, before their accep-
tance for final storage in a subsurface site, must be shown to be suitable for this type of disposal
For this, all forms of heterogeneous and homogeneous intermediate and low-level wastes (ILW
and LLW) are subjected to technical inspections and basic safety rules (R F S --R6gles Fon-
damentales de Sfiret6--"Fundamental Safety Rules") These specifications give general and
specific conditions for determining acceptance of the wastes for shallow land burial Thus, a
data bank of the properties of the waste products is required If the results are satisfactory,
A N D R A (the French National Agency for Radioactive Waste Management) can store the
packages in shallow land burial
With this in mind, measurements concern either the encapsulated waste or the whole pack-
age, including the camsters An inventory of the required properties can, to a large extent, be
established as follows
For the encapsulated wastes
Specific characteristacs, c o m m o n to all kinds of radioactive waste, include

a temperatures for changes of state (only for polymer and bitumen matnces)
b homogeneity of the processed waste
c mechanical properties
Research and development engineer, Atomic Energy Commission, Nuclear Research Center of
SACLAY, 91191 Gff-sur-Yvette, France
2 Research and development engineer, Atomic Energy Commission, Nuclear Research Center of
FONTENAY-AUX-ROSES, BP6, 92265 Fontenay-aux-Roses, Cedex, France
479

480 HAZARDOUSWASTES
d water exuded under compression

e compressive and bending strengths
Retention properties
a leaching, weathering and radiation resistance
b tests, c o m m o n to all matrices
c sensitivity to water contact
As a second step, additional measurements may be necessary Among these we can note
1 Permeability to air and water

2 Density, porosity, weight loss
3 Shrinkage
4 Tensile strength
For quahficatlon or appraisal of the whole package, other properties have to be taken into
account the filling percentage of the drums or other containers, the weight, the resistance
under load, fall tests, integrity, fire resistance, dose rate, surface contamination, and so on
This extensive characterization is required by French policy The authorities in charge of
this policy are the Ministry of Industry and ANDRA
To reach this goal, full-size packages, generally with actual radioactive waste, are selected at
the producer's facihty and sent to otficml laboratories where the required tests are carried out.
The aim of this presentation is to give several representative and pragmatic results that can be
considered as the most important for qualification, that is, determining if the materials spec-
ifications meet acceptance standards, or the appraisal of the packages
The packages come from the following producers
1 ElectrlClt6 de France, with its electronuclear power plants

2 Nuclear research centers from the Commissariat fi l'Energie Atomlque (Saclay, Cadar-
ache, Grenoble, and so on)
3 COGEMA (Compagnle g6n6rale des matl6res nucl6alres), for reprocesslng facilities,
power plants, hospitals, and university laboratories
Homogeneity of the Package

Homogeneity is measured for the whole package, one method is to take cores to obtain sam-
ples for which density maps can be determined and destructive radiochem~cal analyses earned
out Another method is continuous scanning gamma-ray spectrometry Figures 1 and 2 show,
respectively, a conng operation and a scanmng gamma-ray spectrometer With the latter, ~t is
possible to locate the hot points In the structure (Fig 3)
Mechanical Properties
It 1s also important to know the mechanical properties that govern the general behavior of
the processed wastes, one can think of product hardness and lntergmy The main properties
measured are compressive, bending, and breaking strengths They are obtained either by using
samples at the producer's facility or by taking cores and machining with packages submitted
for appraisal Figure 4 shows a hot cell wlth master-slave manipulators where these tests are
earned out, and an example of mechanical properhes of cores measured in this cell is given in
Table
Copyright 1 Int'l (all rights reserved); Mon Feb 24 23:07:03 EST 2020
by ASTM
NOMINE AND BILLON ON PERFORMANCE APPRAISAL 481
FIG. 1--Removal o f a core from a cement matrix.
FIG. 2 - - The scanning gamma-ray spectrometer used for measuring activites in cores.
482 HAZARDOUS
WASTES
Relative variation
60Co (20 mm) ~l" , ~. _ ~L"~,.~ /~\
', / . / i
o-'-~/v ,:o~o~,,
///
"-..M~,o,,,ooo,
4 . o 4 " ~ (values parenthers are the
__..o.....- - ' ~ width of collimation)
__External surface Depth (ram)

0 (~ 0 h i I 50-- I i ~ , 100 -"-
FIG. 3--An example of the results obtained in measuring the homogeneity of a core with a continuous
gamma-ray spectrometer.
FIG. Int'l
Copyright by ASTM 4 - -(all e hot
T hrights cell used
reserved); in 24
Mon Feb measuring
23:07:03 EST the
2020properties of various radioactive waste packages.
TABLE 1--Examples of the values for mechantcal propertles

measuredfor cores taken from full-stze packages
Age of Flexural Compressure
Samples Stnnkage Strength, Strength,
m Days Density in #m/m MPa MPa
7 2 25 52 3 16
2 20 48 35 17
2 10 50 3 16
2 20 51 4 17 5
2 01 50 36 16
221 51 3 16
14 1 90 105 4 26
2 01 110 5 27
1 85 110 4 5 27
1 97 112 5 28
200 111 5 27
1 98 105 4 5 28
28 1 85 260 5 31
1 90 260 6 32
1 93 265 65 33
1 90 265 5 30
1 85 260 6 32
1 98 260 6 33
Full-Scale Leaching Tests
Leachablhty can be considered as one of the most important characteristics, this atttrlbute
allows us to know precisely the ability of the embedding matenal to retain the radlonuchdes
The French approach is to carry out leaching tests on actual full-size waste units The value of
these tests hes in the fact that they are made with packages coming directly from the producer's
facility and chosen at random The tests are carried out following a stnct procedure In general,
the test is static, made using a standard synthetic leach water (simulating the shallow land bur-
lal groundwater table), simulated water has been fabricated with delonlzed water containing.
NaC1 (117 mg/L), Ca(HCO3)2 (2 3 rag/L), K2SO4 (89 mg/L), Ng(NO3)2 (40 mg/L)], CaSO4
2H20 (63 mg/L), NgSO4 7H20 (27 mg/L) The leach water is kept at 23 _+ 3~ and is pen-
odically renewed (cycles of 2 X 15 days, 2 X 30 days, and 4 X 90 days) At the end of each
cycle samples are taken for
Spectrometer measurements The main radionuchdes followed are 239pu, 241Am, 137Cs,
6~ and 9~
Chemical and physicochemlcal analyses, which allow us to show the extent of ion trans-
fer between the radioactive waste packages and the environment The ions that we look
for depend on the nature of the m a t n x and the waste Thus, for example, In the case of
a cement-nitrate block we would prefer to look for nitrate and sodium What is impor-
tant to note is that the evaluation of mtrate leachability is of great interest In particular,
the evaluation makes it possible to calculate the increase In leachability due to the
increase in porosity resulting from the attack by water
Leaching tests are carried out in a special faclhty where blocks of different volumes (from a
few htres to 2 m 3) can be studied A partial view of this faclhty is given in Fig 5 The first
example
Copyright of Int'l
by ASTM the(allresults obtained
rights reserved); Mon Feb is
24 for 200-L
23:07:03 drums containing a bitumen-concentrate matnx
EST 2020
484 HAZARDOUSWASTES
FIG. 5--Leachingfacilityfor full-size radioactive waste packages.
[mainly sodium nitrate (NaNO3) with a, t , and 3' emitters], which were subjected to leaching
for 455 days. The most important results obtained can be summarized as follows.
1. The annual fractions leached, covering the total constant leaching period, are in accor-
dance with the authorized limits specified by French policy. 3
2. Chemical analyses and visual observations at the end of the test show that a significant
amount of salts was leached and that there was swelling of the packages (Table 2 and Fig.
6).
The following examples illustrate acceptable and nonacceptable encapsulated wastes for
storage in a final respository.
Type A packages can be considered as acceptable for all the radionuclides present except for
the 6~ which is very easily leached. This kind of embedded waste also shows a high NaNO3
solubility and a swelling effect (see Fig. 6); for these reasons Type A packages have to be
improved before being accepted for final storage.
On the other hand, Type B packages are in conformity with French policy requirements; no
excessive release of any radionuclide nor of any chemical components was detected; the
annual fractions leached are below the maximum permissible limits.
The next example is for cement containing waste products (mainly NaNO3 containing ~3VCs,
9~ 6~ and traces of a-emitters).
The similar reasons, Type C packages are not satisfactory and Type D packages are accept-
able (Table 3). For the former, the release of ~37C0is above the limit, and after the leaching test
3 The authorized leaching limits (regulated by French policy) are the following: (1) for a specific ~a
activityless than 1 Ci/T, the speed will be less than 5 X 10- 3cm X Ci; (2) for a specific fla activity between
1 andby l0
Copyright ASTM Ci/T, this
Int'l (all value
rights willMon
reserved); be Feb
less24than l EST
23:07:03 • 2020
l0 -3 cm X Ci; (3) with a specific fla activity higher than
l0 Ci, the speed
Downloaded/printed by is limited to 2 X 10 6 cm X Ci-L
NOMINE AND BILLON ON PERFORMANCEAPPRAISAL 485
TABLE 2--Leaching results for full-size packages" of bitumen concentrates.
Package A Package B
Radionuclides Annual Fractions Leached Radionuclides Annual Fractions Leached
137Cs 2.0 X 10 5
9~ 1.7 X 10 -4 9~ < 1 . 5 X 10 -6
6~ 3.5 X 10 3
239pu 3.6 X 10 5 239pu 1.8 X 10 -4
241Am 9.5 X 10 -s 2 4JAm 1.2 X 10 -5
CHEMICAL RELEASE
Dryextract 7.0 X 10 2 Dry matter undetectable
sodium 20 X 10 -2
nitrates 3.2 X 10 2
sulfates 1.5 X 10 -2
FIG. 6--Appearance of a 200-L drum containing a bitumen-concentrate matrix after a leaching test.
TABLE 3--Cement-waste systems: results of the leaching tests.
Package C Package D
Radionuclides Annual Fractions Leached Radionuclides Annual Fractions Leached
137Cs 0.3 ~37Cs Not present

6~ <2.8 X 10 -3 6~ Not present
9~ 2.0 X 10 4 239pu 1.5 X 10 -4
239pu < 1.8 X 10 -s 241Am 5.3 X 10 -4
24~Am 3.2 X 10 6
CHEMICAL COMPONENT
Dry extract 3 X 10 -2 Dry extract 3.3 X 10 2
Nitrates
Copyright 9 XFeb10
by ASTM Int'l (all rights reserved); Mon 2
24 23:07:03 EST 2020 Nitrates 7X10 2
FIG. 7--1nitial appearance o f a 200-L cement block containing t3 and gamma-ray emitting nuclides
with traces o f a radioactivity.
the cement-waste system has lost its integrity. Figures 7 and 8 show several significant defects
in this type of package. For Type D packages the annual fractions leached are very low, and
thus these packages can be stored.
The last example of the tests carried out on full-size packages is for a polymer-ion exchange
resin where the main radionuclides are 137Csand 6~ The 137Csleaching rate as a function of
time is shown in Fig. 9, and the final appearance of the drum is given in Fig. 10.
Weathering Tests
These, also, are among the most interesting tests for determining the acceptance for storage
of embedded ILM and LLW. Climatic changes can influence the long-term integrity of the
packages, and consequently weathering tests are required. These tests consist of submitting the
materials to various temperature cycles, generally, with - 20~ as the lowest temperature and
Downloaded/printed by FIG. 8--Appearance o f the block in Fig. 7 after a leaching test.
NOMINI c AND BILLON ON P E R F O R M A N C E APPRAISAL 487
~
~rt'
~ m
~m ~TllN
mlL~lll
r,J lug
,\!
IJ, I I.~
!--- v 4
,.~
mr- C,~
X ""
r,~
Z
H
"r
-.J ~ ,||IIiHt ~i||
0 t00 200 300 400 50~

Time in Days
+_____+2: VL
x xS: VL 04]:R13
POLYMER-ION EXCHANGERESIN PACKAGE

FIG 9--Leachtng of137Cs as a functton of ttme
40~ as the highest An intermediate step is spnnkhng with water at 5~ These tests are fol-
lowed by mechamcal measurements to determine any damage induced m the structure Tests
carried out on a 200-L drum of bltumen-r gamma waste on samples representative of a poly-
mer-ion exchange resin, and on a cement-concentrate product are shown in Figs 1 l, 12, and
13, respecuvely Clearly, In the case of this cement-concentrate material, the effect of the freez-
ing cycle demonstrates the poor quality of the product
Permeability
Permeabdity is another very important and interesting property, its measurement allows us
to make
Copyright correlations
by ASTM w~th the
Int'l (all rights reserved); leachability
Mon Feb 24 23:07:03 EST Generally,
2020 the test is carried out using air or water
as the Industrial
Universidad permeating
de Santanderfluid A schematic
(Universidad drawing
Industrial de Santander) and
pursuant a photograph
to License of reproductions
Agreement. No further the apparatus
authorized.used are
FIG. 1O--Appearance of a 200-L drum containing a polymer-ion exchange resin matrix after a leaching
lest.
FIG. 1 1 - - T w o IO0-L drums of bitumen containing fl and gamma-ray emitting waste in a weathering
chamber.
LNOMINE AND BILLON ON PERFORMANCE APPRAISAL 489
FIG. 12--Samples of a polymer-ion exchange resin matrix from a 200-L drum after a weathering test.
shown in Figs. 14 and 15, respectively. Measurements are possible only on hard materials
made with hydraulic binders or polymers (Table 4).
The following results are for blocks of cement-a-activity waste concentrate. The perme-
ability values in comparison with values obtained with conventional mortars show that the
L L W block considered is satisfactory; the permeability measurements are not an obstacle to
acceptance of these materials for shallow land burial.
Appraisal of Old Bitumen-Concentrates Packages

Often, a knowledge of the changes occurring in packages with time (aging problems) is
needed to appraise old waste packages received either for intermediate or final storage.
The next example concerns bitumen-concentrates and sludges as wastes which are more
than 15 years old.
FIG. 13--Cement-concentrate samples fractured by weathering (freezing) effects in the presence of

water.by ASTM Int'l (all rights reserved); Mon Feb 24 23:07:03 EST 2020
Copyright
] ~ Gas
Valve
Pressure gauge
P
Eau
Water level
A o~..), ~ , r r 1 6 2
Sealing Ring
Sample
I 1 1 1
%
FIG. 14--Schematic drawing of permeability measurement system.
Different samples were taken and the following operations carried out.
1. Drums were sectioned to examine the internal structure. No major defects or failures, no
hot points (in the distribution of the radionuclides, mainly '37Cs and 6~ and no voids
were detected, thus, confirming the good behavior of the matrices (Fig. 16).
2. Metallographic examination (microscopy) of sections of the metal drums was made to
verify that there was no corrosion (Fig. 17).
3. Cores of the bitumen matrices for radiochemical and chemical analyses were taken to
demonstrate the homogeneity of the packages.
By means of these tests a large number of drums containing different matrices have been exam-
ined to ensure that the wastes sent for storage and disposal are consistently of an acceptable
quality.
Tests of the Compressive Strength of Drums

In general, in the final disposal, the drums are stacked on each other, and it is, thus, necessary
to know their compressive strength. This type of test is carried out by placing a calibrated load
on the top of the package (Fig. 18). In the example shown in this figure the test was made on
a 200-L metal drum containing high-activity ion exchange resins [a contact dose rate of ~ 5 0
mGy. h 1 (5 tad. h-')]. This explains the presence of the concrete container surrounding the
drum and acting as a biological radiation shield. Detectors placed on the drum made it possible
to measure
Copyright its(alldeformation
by ASTM Int'l during
rights reserved); Mon the 24-h
Feb 24 23:07:03 length of the test.
EST 2020
F I G . 15--The permeability measurement system.
T A B L E 4--Permeabilities o f mortar containing a liquid wastes.
Permeability
T o N2, m 2 To Water, m 2
S a m p l e No. 1 6.9 X 10 -17 1.3 X 10 - I s

3.2 X 10 17 1.3 X 10 -18
9.7 • 10 17 1.2 • 10 -18
S a m p l e No. 2 2.4 • 10 ~7 1.7 X 10 - i s
1.1 X 10 -17 1.6 X 10 18
1.4 X 10 -17 1.6 X 10 -18
FIG. 16--Sections of a 200-L drum of a bitumen matrix containing 13and gamma-ray emitters.
These, then, are the different examples intended to show the importance that we place in
France on knowledge of the characteristics, and their time dependence, of packages that may
be stored or those already in storage (old packages).
Only when these properties have been measured--nearly always for full-size actual radio-
active packages--can we decide which of two actions should be taken. The first is to draw up
a complete file for submission to the safety authorities to obtain acceptance for delivery to the
disposal site. The second is to ask that the package be retained by the producer and be modified
Copyright by ASTM Int'lFIG. 17--Sampling

(all rights reserved); Mon Febsections
24 23:07:03o EST
f a 200-L
2020 drum fi~r microscopic examinations.
FIG. 18--Loading a drum to determine its compressive strength (with 10 tons, using an overhead crane
for the placement of the calibrated load).
before further tests are requested. In both cases what is involved is quality assurance. This con-
sists of rigorous requirements which the producers must meet. Subsequent tests for quality are
always carried out.
Only by c o n t i n u i n g to make these efforts can we ensure the long-term integrity of the radio-
active waste packages and that they have n o harmful effects on the biosphere.
References
[ 1] Nominr, J. C., Bernard, A., and Farges, L., "Long Term Leaching Tests on Full Scale Blocks of Radio-
active Wastes," ORNL Conference on the Leachability of Radioactive Solids, Oak Ridge National
Lab., Oak Ridge, TN, 9-12 Dec. 1980.
[2] Marcaillou, J., Faure, J. C., Bernard, A., and Nominr, J. C., Pratique en matirre de conditionnement
des drchets solides radio-actifs en vue de leur stockage temporaire sur le CEN-Cadarache en attente
Copyright d'rvacuation.
by ASTM Int'l (all rights reserved);AIEA-CCE.
Colloque Mon Feb 24 23:07:03 EST 2020
UTRECHT--PAYS-BAS, 21-25 Juin 1982.
[3] Nougmer, H, Nomm6, J C, and Vaunols, P, "Example ofa Quahty Control Operation Performed
on a Nuclear Reactor Waste Package," Waste Management, March 1983
[4] Nomm6, J C and Vejmelka, P, "Expeilence with Full-Scale Leaching of Low and Medmm Level
Waste," CEC Seminar on Testing, Evaluatwn and Shallow Land Burtal of Low and Medtum Radto-
acttve Waste Forms, Commission of European CommumUes, Belgium, September 1983
[5] Pottier, P E and Glasser, F P, "Characterization of Low and Medium Level Radioactive Waste
Forms," Final Report EUR 10579, 2nd Programme 1980-1984, Commission of the European
Communities,
[6] NommS, J C and Blllon, A, "Experience Acqmred m the Field of Long-Term Leaching Tests on
Blocks of Radloactwe Sohds," Fourth Internattonal Symposmm on Waste, Envzronmental Aspect of
Stabdlzatwn/Sohdtficatwn of Hazardous and Radioactive Waste, Atlanta, GA, May 1987
[7] Nomm6 J C and Bdlon, A, "Leaching Scale Effect for Cement Waste Forms," 12th Internatwnal
Symposmm on tke Scientific Basts for Nuclear Waste Management, Berlin, October 1988
[8] Nomln~, J C and Bdlon, A, "Industnal Aspects of the Leaching Tests Camed Out on Full Size
Radioactive Blocks," IRPS, Dubrovnlk, Yugoslavia, Oct 1989
STP1123-EB/Jan. 1992
Author Index
A G
Aloy, A S, 86 Garvey, Wllham F, 242

Assmann, Dons, 90 Gdham, T Michael, ix, 61,394
Gong, Runzhl, 103
Gress, David L, 119
Bajpayee, T S, 378 H
BalzaI.o, Stephanaa, 182
Bangart, Richard L, 3 Han, Kyong-Won, 204
Banch, III, John J, 12 Heath, Jeffrey C, 466
Barth, Edwin F, 18, 409 Hjelmar, Ole, 320
Bentley, William, 73
Blllon, A, 479
Bishop, Paul L, 103
Bostlck, William D, 132 Isenburg, Jerry, 361
Brlcka, R Mark, 193
Brodda, Berthold-Gunther, 90 J
Brodersen, Knud, 320
Brown, Robert E, 43 Jastrow, Juhe D, 448
Buckley, Leo P, 204 Jlndal, Blpender S, 43
Bull, Michel, 227 Jones, Larry W, 193, 385
Bulzan, Joseph D, 43
K
C Keener, Tim C, 103
Kllley, R W Doug, 266
Collins, M Robin, 119
Klm, Huan-Young, 338
Culhnane, M John, 193, 385
Kam, Hwan-Young, 297
Klm, Jong-Oh, 283
D Klm, Joon-Hyung, 338
Klm, K1-Hong, 297
De Angehs, Glorglo, 182 Kroner, Nancy E, 119
de Groot, Gerard J, 149 Kleppe, James H, 426
Del Cul, Gulllermo D, 394 Kolycheva, T I, 86
Dollnar, George M, 266 Koo, Ja-Kong, 283
Donato, Aldo, 255 Kuznetsov, B S, 86
Durhng, Diane B, 119
Duslng, David C, 103 L
Lee, Jae-Owan, 204

LI, Cong-Le, 103
Finn, John T , 426 Lukas, Robert G , 440
Fontalne, Thomas A, 415 Lyons, W Berry, 119
495
Copyright9 1992byASTMInternational www.aslm.org

M Rico, Gmseppe A, 255

Rogers, Robert D, 448
Malnlero, Richard J, 378
Marchettl, Antoma, 182 S
Mason, B J, 12
McConnell, Jr, John W, 448 Schexnayder, Chff, 440
McIntosh, Thomas H, 73 Scocoa, Glancarlo, 171
McLaughhn, Wllham J, 466 Sell, Nancy J, 73
Means, Jeffrey L, 466 Serrlnl, Glorglo, 217
Med~o, Franco, 171 Shashukov, E A, 86
Merh, Carlo, 171 Shin, Eung-Bal, 283
Moore, Michael, 361 S h i n , Hang-Slk, 283
Morgan, Ivan L, 61, 132 Soundararajan, Rengarajan, 33
Mortensen, Svend, 320 Spalding, Brian P, 415
Myers, Tommy E, 304 Spence, Roger D, 61
Suh, In-Suk, 297, 338
N
T
Nomm6, J C, 479
Tokar, M~chael, 3
O
Oliver, Jean, 227 V

Osborne, Steven C, 61
Otten, Mark T, 426 van der Sloot, Hans A, 149
Vlshnevskn, A S, 86
P Volpe, Roberto, 171
Park, Hun-Hwee, 297,338 W

Perry, Klmberly J , 242
Phlhpose, Ken E, 266 Wlckhff, Diana S, 448
Prange, Ned E, 242 Wiles, Carlton C, 18
R Z
Revall, Mark A, 73 Zamoranl, Edmond, 217

Revertegat, Ellane, 227 Zappl, Mark E, 304
STP1123-EB/Jan. 1992
Subject Index
A generalized acid neturallzatlon capacity
test, 361
Acetone, wastes containing, 61 hazardous wastes, 255
Adsorption, permanence evaluation, 33 in-line grout meter, 394
Alkahnlty, solidified heavy metal sludges, low-level radioactive waste disposal, 3
283 wastes containing volatile organic
Aluminum phosphate hardener, hazardous compounds, 61
waste Incinerator residues, 90 Cementation, borate waste by adding slaked
Anion exchange resins, solidification, grout- lime, 338
based waste forms, 133 Cemented boiling water reactor, cesium
ANS 16 1 test, 204 from evaporator concentrates, 182
Apprmsal, full-size radioactive waste Cementing agents, hazardous organic
packages, 479 wastes, 43
Arsenic, metal-contaminated sods, 426 Cement matrices, toxic element release, 171
Cement waste forms
anion exchange resins, 133
leachability of Co-60, Sr-85, and Cs-137,
Benzene, wastes containing, 61 204
Binders, chemistry, 374 Ceramic-clay matrices, Incorporation of
Blogeochemistry, mumclpal solid waste radioactive wastes, 86
incinerator ash, 119 Ceramics, incorporation of radioactive
Bitumen, low-level radioactive waste wastes, 86
disposal, 3 Cesium
Boiler ash, hazardous waste lnclneratcr GEODUR additive and leachability, 320
residues, 90 leachability, 182
Borate waste lyslmeter data and performance
cementation by adding slaked lime, 338 assessment plans, 448
dried, solidification with polyester, 297 Cesium nitrate, immobilized, hydrating
water effect, 217
C Characterization, full-size radioactive waste
packages, 479
Cadmium Chemical bonding, permanence evaluation,
GEODUR additive and leachability, 320 33
mumclpal solid waste incinerator ash, 119 Chemical fixation, flue gas desulfurizatlon
Calcium hydroxide, cement attack of water sludges and fly ash, 103
or undersaturated lime solutions, 227 Chlonnated phenol contaminated soil,
Calcium oxide, solidification with polyester, solidification and stabilization, 73
297 Chlorobenzene, wastes containing, 61
Calcium silicate hydrates, cement attack of Chloroform, wastes containing, 61
water or undersaturated lime Chromium, metal-contaminated soils, 426
solutions, 227 Clays, incorporation of radioactive wastes, 86
Cement Chnoptxlolte, leachability of Co-60, Sr-85,
additives, 320 and Cs-137, 204
effect of water or undersaturated lime Cobalt, lyslmeter data and performance
solutions, 227 assessment plans, 448
497
498 HAZARDOUSWASTES
Compressive strength tests, metal- Fly ash

contaminated soils, 426 flue gas desulfunzation, 103
Cone index test, 385 generalized acid neturahzatlon capacity
Copper test, 361
generalized acid neturahzat~on capacity municipal solid waste incinerator, 119
test, 361 Founer transform infrared spectra,
GEODUR additive and leachability, 320 stablllZatlon/sohdification
municipal sohd waste incinerator ash, 119 evaluation, 33
toxic element release, 171
waste sandblasting grit, 466 G
D Gap test, explosive-contaminated sohd

wastes, 378
Degradation, modelling, cement attack, 227 Generahzed acid neutralization capacity
Detonation, explosive-contaminated solid test, 361
wastes, 378 GEODUR process, 320
1,2-Dlchloroethene, wastes containing, 61 Grout
Differential scanning calorimetry, in-line meter, 394
stabillzatlon/sohdlfiCatlon m SltU, radmactlve solid waste bunal
evaluation, 33 trenches, 415
Diffusion coefficient, immobilized cesium waste forms, solidification of anion
nitrate release, 217 exchange resins, 133
Diffusion models, environmental product wastes containing volatile organic
certification, 149 compounds, 6
Dilution, permanence evaluation, 33
Dolomite, generalized acid neturallzatlon H
capacity test, 361
Durablhty testing, anion exchange resin Hazardous waste
solidification, 133 environmental product certification, 149
Dynamic compaction, radioactive waste, long-term leaching performance, 242
440 metal-contaminated sods, 426
organic, effectiveness of stablhzatlon and
E sohdlfiCatlon, 43
solidified, disposal, 283
Electrical resistance, in-line grout meter, solidification and stabilization, 18
394 short-term penetration testing, 385
Environmental product certification, uncontrolled sites, 409
leaching charactenstlcs, 149 Hazardous waste incinerators, residue
Enwronmental Protection Agency, SITE fixation, 90
demonstrations, uncontrolled Hazardous waste sites, public and regulatory
hazardous waste sites, 409 acceptance of performance standards,
Explosive reactiwty, explosive- 12
contaminated solid wastes, 378 Heavy metals
Extraction procedure toxicity tests, metal- flue gas desulfurlzation sludges and fly
contaminated soils, 426 ash, 103
GEODUR additive and leachability,
320
hazardous waste incinerator residues,
Filter dust, hazardous waste Incinerator 90
residues, 90 sohdlfied sludge, disposal, 283
Flow-around leach tests, 204 stablhzatlon, 193
Flue ash, hazardous waste incinerator stablhzatlon/sohdxficatlon in sediment,
residues, 90 304
Flue gas, desulfurlzation sludges and fly ash, High-integrity containers, low-level
chemical fixation, 103
radioactive waste disposal, 3
SUBJECT INDEX 499
Hydrating water, effect on diffusion release Lead

of cesium nitrate, 217 generahzed acid neturallzatlon capacity
Hydraulic conductivity, in SltU grouting test, 36 l
with polyacrylamlde, 415 GEODUR additive and leachability, 320
metal-contaminated soils, 426
municipal solid waste incinerator ash, 119
SITE demonstrations, uncontrolled
Immoblhzatlon hazardous waste sites, 409
heavy metals in sediment, 304 waste sandblasting gnt, 466
proper application, 18 LIMB ash, 103
toxic element release, 171 Lime, generalized acid neturallzatlon
wastes containing volatile organic capacity test, 361
compounds, 61 Llme-pozzolan matrices, toxic element
Internal ignition test, explosive- release, 171
contaminated solid wastes, 378 Lime solutions, undersaturated, cement
Ion exchange resins, leaching behavior of attack, 227
Co-60, Sr-85, and Cs-137, 204 Lithium toxic element release, 171
Lyslmeter, site specific performance
K assessment plans, 448
Kiln dust, generahzed acid neturahzatlon M

capacity test, 361
Manne disposal
L GEODUR additive and leachabihty, 320
municipal solid waste incinerator ash, 119
Leachability Manne sediment, municipal solid waste
cesium, 182 incinerator ash, ocean disposal
Co-60, Sr-85, and Cs- 137 from cemented effects, 119
ion-exchange resm waste forms, 204 Metal-contaminated soils, stablhzation/
flue gas desulfurizatlon sludges and fly solidification, 426
ash, 103 Metal finishing waste treatment,
generalized acid neturahzatlon capacity leachability, 193
test, 361 Metals, SITE demonstrations, uncontrolled
GEODUR additive and, 320 hazardous waste sites, 409
hazardous wastes Methyl ethyl ketone, wastes containing, 61
Incinerator residues, 90 Municipal solid waste incinerator ash
radioactive waste solidification cement additives, 320
methods, 255 ocean disposal and marine sediments,
heavy metals in sediment, 304 119
hydrating water effect, 217 Munitions, explosive-contaminated solid
long-term, commercially stabilized waste, wastes, 378
242
lyslmeter data and performance N
assessment plans, 448
sandblasting grit, 466 Netherlands Energy Research Foundation,
solidified dried borate waste, 297 leach database, 149
sohdlfied heavy metal sludges, 283 New Bedford Harbor Superfund site,
stabilization/solidification of metal- stablhzatlon/sohdlfiCatlon of heavy
contaminated soils, 426 metals in sediment, 304
strontium, 182 Nuclear waste See Radioactive waste
waste constituents and, 193
Leaching mechanisms, environmental
product certification, 149
Leaching test, toxic element release from PCBs, SITE demonstranons, uncontrolled
Copyright bysolidified
ASTM Int'l (allwastes, 171Mon Feb 24 23:07:03 EST 2020
rights reserved);
hazardous waste sites, 409
500 HAZARDOUSWASTES
Penetration resistance, in s~tu grouting w~th Risk assessment, wastes m hazardous waste
polyacrylamide, 415 sRes, 12
Perchloroethene, wastes contalnmg, 61
Performance assessment S
full-size radioactive waste packages, 479
site specific plans, lys~meter data, 448 Salts, hazardous waste lncmerator residues,
Performance standards, wastes m hazardous 9O
waste sites, 12 Sandblasting grit, sohdlficatlon/
Phenol contaminated sod, sohdlfiCatlon and stablhzatlon, 466
stabilization, 73 Sediment
Polyacrylam~de, in situ grouting, radioactive mumcipal solid waste incinerator ash,
solid waste burial trenches, 415 ocean disposal effects, 119
Polyester radioactive contaminated, stabilization
dried borate waste sohdlfiCatlon, 297 and m sltu management, 266
resins, hazardous wastes, 255 stabdxzatlon/solldlfiCatlon of heavy
Polymer soli&ficatlon, hazardous wastes, metals, 304
255 Shallow land &sposal, hazardous waste
Porosity factor, ~mmobillzed cesmm mtrate incinerator residues, 90
release, 217 Slag, hazardous waste incinerator residues,
Port Hope Harbor, radioactive 90
contaminated sediments, Slaked lime, borate waste cementation, 338
stablhzatlon and in sItu management, Sohdlficatlon
266 anion exchange resins, 133
Pressurized water reactors, evaporator dried borate waste with polyester, 297
concentrate cementation, 338 effectiveness, hazardous organic wastes,
43
Q flue gas desulfurizatxon sludges and fly
ash, 103
Quality control, ln-hne grout meter, 394 hazardous waste, short-term penetration
Quaternary ammonium modified clays, testing, 385
hazardous organic wastes, 43 heavy metals xn sediment, 304
heavy metal sludges, 283
R in-hne grout meter, 394
leachability and waste constituents, 193
Radioactive ashes, incorporation into metal-contaminated soils, 426
ceramic-clay metrlces, 86 municipal sohd waste incinerator ash, 119
Radioactive contaminated sod, cement permanence evaluation, 33
additives, 320 phenol and chlorinated phenol
Radioactive waste contaminated soils, 73
cesium leaching, 182 proper application, 18
contaminated sediments, stabilization and radioactive waste methods, apphcatlon to
in sltu management, 266 hazardous wastes, 255
dynamic compaction, 440 SITE demonstrations, uncontrolled
incorporation into ceramic-clay metrlces, hazardous waste sites, 409
86 toxic element release, 171
in SltU grouting with polyacrylamlde, 415 waste sandblasting gnt, 466
low-level disposal, 3 wastes containing volatile organic
lyslmeter data and performance compounds, 61
assessment plans, 448 wastes m hazardous waste sites, 12
performance appraisal of full-size Sohd wastes, explosive-contaminated,
packages, 479 explosive reactivity, 378
sohdlfication methods, application to Stablhty, long-term, commercially stablhzed
hazardous wastes, 255 waste, 242
Residue fixation, hazardous waste Stablllzanon
incinerators,
Copyright by ASTM Int'l (all rights 90
cement additives, 320
SUBJECTINDEX 501
effectiveness, hazardous orgamc wastes, Toxicity Characteristic Leaching

43 Procedures, 73
flue gas desulfunzatlon sludges and fly commercially stabihzed waste, 242
ash, 103 SITE demonstrations, uncontrolled
hazardous waste, short-term penetration hazardous waste s~tes, 409
testing, 385 Toxic waste, soll&ficatlon/stabfllzatlon, 18,
heavy metals m sediment, 304 73
m-line grout meter, 394 Tnchloroethene, wastes containing, 61
in SltU ~routIng with polyacrylamlde,
415 U
leachabihty and waste constituents, 193
long-term leaching performance, 242 Unconfined compressive strength test, 385
low-level radioactive waste &sposal, 3 Urea-formaldehyde resins, hazardous
metal-contaminated softs, 426 wastes, 255
mumc~pal solid waste incinerator ash, U S Nuclear Regulatory Commission, low-
119, 320 level radioactive waste disposal, 3
permanence evaluation, 33
phenol and chlonnated phenol V
contaminated soils, 73
proper apphcatlon, 18 Vinyl ester styrene waste forms, anion
radioactive contaminated soft, 320 exchange resms, 133
sandblasting gnt, 466 Volatile organic compounds, wastes
SITE demonstrations, uncontrolled containing, stabilization/
hazardous waste sites, 409 solidification, 61
wastes containing volatile organic
compounds, 61 W
wastes in hazardous waste sites, 12
Static test, cesmm leaching, 182 Waste treatment, sohdlficatlon and
Strontmm stablhzatlon, 18
leaching, 182 Water, cement attack, 227
lysimeter data and performance Water glass, hazardous waste incinerator
assessment plans, 448 residues, 90
Sulfate wastes, cesmm leaching, 182 Wemng/drylng test, anion exchange resin
sohdlficatton, 133
T
Z
Tank leach test, 149
Thermogravlmetnc analysis, stablhzatlon/ Zeolon 900-Na, leachablhty of Co-60, Sr-85,
solidification evaluation, 33 and Cs-137,204
Toxic element release, solidification and, Zinc, G E O D U R ad&tlve and leachablhty,
171 320
ISBN 0 - 8 0 3 1 - 1 4 4 3 - 5

Stp1123-Eb 19568 PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Stp1123-Eb 19568 PDF

Uploaded by

Copyright:

Available Formats

Copyright by ASTM Int'l (all rights reserved); Mon Feb 24 23:07:03 EST 2020

Stabilization and Solidification of

T. Michael Gdliam and Carlton C. Wiles

ASTM Publication Code Number (PCN)

ASTM Pubhcat~on Code Number (PCN) 04-011230-56

Copyright 9 1992 AMERICANSOCIETYFOR TESTINGAND MATERIALS, Phdadelph~a, PA Pnor edition copy-

Peer Revmw Policy

This publication, Stabtllzatlon and Sohdtficatton of Hazardous, Radloactlve, and Mixed

REGULATORY AND TECHNICAL GUIDANCE

Low-Level Radioactive Waste Disposal in the United S t a t e s - -

Issues Related to the Public and Regulatory Acceptance of Performance Standards

Solidification/Stabilization:Is It Always Appropriate?--CARLTON C WILESAND

Guidelines for Evaluation of the Permanence of a Stabilization/Solidification

SPECIALTY WASTES ORGANICS, ASHES, AND RESINS

A Critical Review of the Effectiveness of Stabilization and Solidification of

Stabilization/Solidificationof Wastes Containing Volatile Organic Compounds in

Solidification and Stabilization of Phenol and Chlorinated Phenol Contaminated

Incorporation of Radioactive Ashes into Ceramic-Clay Matrices--A S ALOY,

Evaluation of the Impact of Ocean Disposal of Stabilized/Solidified Municipal

Performance Testing of Grout-Based Waste Forms for the Solidification of Anion

LABORATORY-SCALE LEACHABILITY STUDIES

Determination of Leaching Characteristics of Waste Materials Leading to

Release of Toxic Elements from Solidified Wastes: A Mathematical M o d e l - -

Leach Studies: Influence of Various Parameters on the Leachability of Cesium

Effects of Selected Waste Constituents on Solidified/Stabilized Waste

Effect of Hydrating Water on the Physical Characteristics and the Diffusion

A Model of the Attack of Pure Water or Undersaturated Lime Solutions on

Long-Term Leaching Performance for Commercially Stabilized W a s t e - -

Application of Radioactive Waste Solidification Methods to Hazardous Wastes in

Mixture Design Optimization Considering Three Disposal Options for Solidified

Solidification of Dried Borate Waste with PoIyester--HWAN-YOUNG KIM,

Laboratory Evaluation of Stabilization/Solidification Technology for Reducing the

Cementation of Borate Waste by Adding Slaked Lime--JOON-HYUNG KIM,

The Development of Cement Formulations for Radioactive Waste Encapsulation--

TEST METHOD DEVELOPMENT

Generalized Acid Neutralization Capacity Test--JERRY ISENBURGAND

Methods of Evaluating Explosive Reactivity of Explosive-Contaminated Solid

An Assessment of Short-Term Penetration Testing as an Indicator of Strength

Development of an In-Line Grout Meter for Improved Quality Control--

LARGE-SCALE EVALUATION OR DEMONSTRATION

Summary of Solidification/StabilizationSITE Demonstrations at Uncontrolled

Field Demonstration of In Situ Grouting of Radioactive Solid Waste Burial

Dynamic Compaction of Nuclear Waste Deposits--ROBERT G LUKASAND

Contributions of Lysimeter Data to the Development of Site Specific Performance

Laboratory Treatment and Field Demonstration Lessons Learned from a Waste

Performance Appraisal of Full-Size Radioactive Waste Packages--J c NOMINE

Inorganic Based Systems Organic Based Systems

portland cement b~tumen

Regulatory and Technical Guidance

Specialty Wastes: Organics, Ashes, and Resins

Laboratory-Scale Leachability Studies

Laboratory-Scale Process Development

Test Method Development

Large-Scale Evaluation or Demonstration

Low-Level Radioactive Waste Disposal in the

ABSTRACt. In this paper, we &scuss the U S Nuclear Regulatory Commlssmn's (NRC)

Copyright9 1992 by ASTM International www.astm.org

Waste Form Solidification and High-Integrity Containers

Toptcal Report Reviews

Vendor Docket No Type Disposition

Waste Chem WM-90 Sohdlfication (bitumen) Approved

H D P E containers Before the promulgation and Implementation of 10 CFR Part 61 (in

Alternattve Methods of Dtsposal