Methods of

All* Sampling
a n d Analysis
THIRD EDITION
Methods of
A i r Sampling
a n d Analysis
THIRD EDITION

JAMES P. LODGE, JR., EDITOR

Taylor & Francis

Taylor & Francis Group
Boca Raton London N e w York

CRC is a n i m p r i n t o f t h e T a y l o r & Francis G r o u p ,

a n i n f o r m a business
 L i b r a r y of Congress Cataloging-in-Publication Data

Methods of air sampling and analysis.-3rd ed. / James P. Lodge, Jr., editor
p. cm.
Bibliography: p.
Includes index.
ISBN 0-87371-141-6
1. Air-Pollution-Measurement-Handbooks, manuals, etc. 2. A i r -
Analysis-Handbooks, manuals, etc. I . Lodge, James P. I I . Intersociety Committee.
TD890.M488 1988
628.5'3—dcl9 8 8-11462
CIP

This book contains information obtained from authentic and highly regarded sources. Reprinted material
is quoted with permission, and sources are indicated. A wide variety of references are listed. Reasonable
efforts have been made to publish reliable data and information, but the author and the publisher cannot
assume responsibility for the validity of all materials or for the consequences of their use.

Neither this book nor any part may be reproduced or transmitted in any form or by any means, electronic
or mechanical, including photocopying, microfilming, and recording, or by any information storage or
retrieval system, without prior permission in writing from the publisher.

The consent of CRC Press L L C does not extend to copying for general distribution, for promotion, for
creating new works, or for resale. Specific permission must be obtained in writing from CRC Press LLC
for such copying.

Direct all inquiries to CRC Press L L C , 2000 N.W. Corporate Blvd., Boca Raton, Florida 33431.

Trademark Notice: Product or corporate names may be trademarks or registered trademarks, and are
used only for identification and explanation, without intent to infringe.

Visit the C R C Press Web site at www.crcpress.com

© 1988 by CRC Press

No claim to original U.S. Government works

International Standard Book Number 0-87371-141-6
Library of Congress Card Number 88-11462
16 17 18 19 20
 PREFACE
This is the third edition of the manual of
methods adopted by the Intersociety Com­
mittee for a Manual on Methods o f A i r
Sampling and Analysis according to its es­
tablished procedures. The Intersociety
Committee, which was founded in 1960,
with its subcommittees consists o f several
scores o f experts currently representing
eight major national professional societies
concerned with environmental measure­
ments. These experts have volunteered their
time and services to evaluate, codify, and
publish hundreds o f test methods for mea­
surement of air contaminants affecting the
health and welfare of industrial workers, as
well as of the general public.
While many methods have been pub­
lished by various agencies, for hundreds o f
materials agreed-upon methods o f ade­
quate accuracy remain to be established.
The Committee and subcommittees have
served as a workshop and forum for inter­
action and coordination of professionals
from different governmental agencies, aca-
demia, and industry, each with knowledge
of the methods i n use by his organization.
In the decade that has passed since the sec­
ond edition, many changes have occurred
in the status o f the Committee, its proce­
dures, and the field o f atmospheric
chemistry.
Financial support from the Environmen­
tal Protection Agency and the National I n ­
stitute for Occupational Safety and Health
has ceased, and with it the ability of the
Committee to meet frequently, and to con­
vene the subcommittees. Consequently,
most o f the procedures previously carried
out face-to-face have been consigned to
mail and, on occasion, telephone. There
have been changes in personnel, as noted
below. The grand plan for a major program
of interlaboratory testing o f methods has
been replaced by occasional round-robin
testing where it was specifically needed,
driven by economic concern for EPA ac­
ceptance o f commercial instruments, or
conducted to resolve dispute. New, more
powerful methods have emerged and be­
come widely accepted, and previous con­
cerns have disappeared. Plans to prepublish
all methods i n "Health Laboratory Science"
for comment have not been implemented
since the second edition.
While the American Public Health Asso­
ciation had previously given valuable sup­
port to the Committee by serving as the
contracting and administrating society, we
are now an independent organization. Nev­
ertheless, there still appears to be a great
need for a manual such as the present one.
While instruments for routine air monitor­
ing are today uniquely qualified or disqual­
ified by the Environmental Protection
Agency, many local agencies are justifiably
reluctant to buy expensive monitoring
equipment prior to preliminary tests to as­
sess the nature of their problems. Such tests
are generally made with available equip­
ment and manual methods o f measure­
ment. For a large number of chemical spe­
cies there are as yet no approved methods;
nevertheless, it is o f considerable impor­
tance to characterize their concentration in
the environment. I n the workplace, moni­
toring instruments are still less common
than individual chemical tests on integrated
samples spanning a significant portion o f
the workday. The likelihood that all these
methods will be subjected to rigorous inter­
laboratory comparison is probably declin­
ing rather than increasing.
Accordingly, a number o f changes have
been made i n the present manual. The sta­
tus o f methods as "tentative" or "recom­
mended" has been removed throughout.
Every method carries statements of preci­
sion, accuracy, and interferences from
which the reader can quickly learn how
much is known about the dependability o f
vi METHODS O F AIR SAMPLING
that method. I n the absence o f sponsor­
ship, the requirement to list method num­
bers according to the several agencies has
also ceased. The given methods are ad­
vanced by the subcommittees and, where
they are knowledgeable, by members o f the
parent Committee. They are among the
best available, whatever their formal status,
and those needing EPA and/or N I O S H
method numbers are undoubtedly in posi­
tion to ascertain them.
For further simplification, certain ubiq­
uitous references have been omitted from
the bibliographies. For example, virtually
every method requires the use of reagent
water o f some type or other, as that is de­
fined by the American Society for Testing
and Materials, and the use o f reagent chem­
icals, as these are defined by the American
Chemical Society. Few analysts are un­
aware of these requirements, and it seems
redundant to enter the appropriate refer­
ences in every bibliography.
We have kept the previously assigned
method numbers within the book, and as­
signed new numbers to added methods.
Thus, some numbers are missing due to the
judgment of subcommittees that the old
methods were no longer in use, or should
not be in use. The previous text, where it is
retained, has been carefully edited to re­
move grammatical and typographical er­
rors. Even in cases where old methods are
largely retained, they have been carefully
checked to be certain that they were clear
and easy to follow; the overall result is that
few pages o f the second edition have sur­
vived without at least minor change. A
number o f new sections have been added to
Part I , on general techniques, with the
emergence of new laboratory tools.
In a number o f situations it has been nec­
essary to go outside the primary member­
ship to reach truly knowledgeable scientists
in the use o f particular techniques. To give
appropriate credit in these cases, a new con­
vention has been adopted. The actual au­
thors o f the technique are listed first, fol­
lowed by the principal reviewers, i f any,
and then the balance of the subcommittee is
listed in alphabetical order. Where changes
are minor, the listing o f the previous sub­
committee, presumably responsible for the
original writing, has been preserved.
The original 12 subcommittees have been
reduced in number to 9, in effect, by the
merger of Subcommittees 4 and 5, and the
termination of Committees 7 and 12. The
founding chairman of the Intersociety
Committee was Dr. Leonard Greenburg.
He was succeeded by Dr. E. R. Hendrick-
son, Professor A . C. Stern, Dr. B . E. Saltz-
man, M r . R. F. Toro, Dr. W. T. Ingram,
Dr. R. J. Thompson, and Dr. R. S. Saltz-
man, the present chairman. Successive edi­
tors have been Dr. M . D . Thomas, Dr.
Morris Katz, and Dr. James P. Lodge, Jr.,
the editor for this edition. Over the entire
history o f the Committee, Dr. George J.
Kupchik has been Executive Secretary. A l l
correspondence concerning these methods
should be directed to him at the School of
Health Sciences, BC 612, Hunter College,
425 East 25th Street, New York, N Y
10010.
The Intersociety Committee gratefully
acknowledges the past support of the U.S.
Public H e a l t h Service under Grant
AP-00256; o f the Environmental Protec­
t i o n Agency under Contract Grant
68-02-0004, and of the National Institute
for Occupational Safety and Health under
Contract number H S M 99-73-89. Ac­
knowledgments of permission to reprint,
with editorial modifications, methods from
other societies and agencies are duly noted
below.
Robert S. Saltzman, ISA
Chairman
Willard A . Crandall, A S M E
Andrew P. H u l l , HPS
William T. Ingram, A P W A
Theodore J. Kneip, A P C A
Bernard E. Saltzman, A O A C
Richard J. Thompson, ACS
Richard F. Toro, A I C h E
 A KEY TO A CLEANER ENVIRONMENT
Other books have been written about air
sampling and analysis, but there are none
like this one.
To do any job well, you must use the
proper tools, and this book, as with its two
predecessor editions, gives you the know-
how you need to sample and analyze air.
Updated, with new methods added and i n a
streamlined form for ease o f reading and
understanding, this new Third Edition is in
itself a tool for today's use.
Though full o f changes from the edition
of 11 years ago, this all-new work leans
heavily on the two previous editions. W i t h ­
out the farsightedness of the leaders who
gave us the first and second books through
the Intersociety Committee for a Manual
on Methods o f A i r Sampling and Analysis
(beginning in 1960) this new work could
have been created only by starting from
"scratch." Even so, it was no mean task for
the scores o f committees and their editor to
come forth with this matchless volume
that will guide today's professionals who
work daily to understand and solve the
problems o f both indoor and outdoor air
pollution.
Our hat is o f f to the experts who labored
long and hard —with intelligence and
integrity—to make this book a reality.
(These pioneers are noted i n the front pages
of this book and i n the committee rosters
appended to each method.) We are ex­
tremely pleased to be associated with the
people who created this worthwhile and
useful venture.
Methods of Air Sampling and Analysis,
Third Edition is both a timely volume and a
book destined for a long life. I t describes
and evaluates new methods along with the
classic. I t is a current and up-to-date refer­
ence to be regarded as a principal tool i n the
continuing, expanding battle to help solve
the never-ending problem o f maintaining
clean air and keeping our environment
workably pure.
- T h e Publisher
 INTERSOCIETY COMMITTEE
METHODS OF AIR SAMPLING
AND ANALYSIS
Representatives of Participating Societies

Organization* Representative Dates o f Service

ACS Henry Freiser 1971-80
Richard J. Thompson 1982-
AIChE Robert Coughlin 1972-74
Richard F. Toro 1974-
APWA James V. Fitzpatrick 1972-79
William T. Ingram 1981-
ASME Arthur C. Stern 1963-74
Wesley C. L . Hemeon 1976-79
Willard A . Crandall 1982-
AOAC Bernard E. Saltzman 1966-
AWMA E. R. Hendrickson 1963-86
Theo J. Kneip 1987-
HPS Lysle C. Schwendiman 1972-83
Andrew H u l l 1984-
ISA Robert L . Chapman 1972-74
Robert S. Saltzman 1974-

*Participating Organizations
ACS —American Chemical Society
A I C h E —American Institute o f Chemical Engineers
APWA—American Public Works Association
A S M E —American Society o f Mechanical Engineers
A O A C —Association o f Official Analytical Chemists
A W M A — A i r and Waste Management Association
H P S - H e a l t h Physics Society
ISA— Instrument Society o f America
x INTERSOCIETY COMMITTEE

ISC C H A I R M E N A N D S T A F F

Name Dates o f Service

Chairman Leonard Greenburg 1963-66
E. R. Hendrickson 1966-69
Arthur C. Stern 1969-72
Bernard E. Saltzman 1972-76
Richard F. Toro 1976-80
William T. Ingram 1980-83
Richard J. Thompson 1983-88
Robert S. Saltzman 1988-
Editor Moyer Thomas 1963-69
Morris Katz 1969-83
James P. Lodge 1983-
Executive George J. Kupchik 1963-
Secretary
SUBCOMMITTEE CHAIRMEN

Name Dates of Service

#1 (Sulfur Compounds)
Donald F. Adams 1966-83
Purnendu K . Dasgupta 1983-

#2 (Halogen Compounds)
Lester V. Cralley 1966-70
C. Ray Thompson 1970-81
Richard A . Mandl 1981-

#3 (Oxidants and Nitrogen Compounds)

Paul M . Giever 1966-67
Bernard E. Saltzman 1967-71
Evaldo L . Kothny 1971-74, 1976-79
Basil Dimitriades 1974-75
Dario Levaggi 1980-

#4 (Carbon Compounds)*
Ralph Smith 1966-79
M i l t o n Feldstein 1979-80

#5 (Hydrocarbon Compounds)*
Eugene Sawicki 1966-79

#4/5 (Carbon and Hydrocarbon Compounds)

M i l t o n Feldstein 1981-

#6 (Metals I)
William G. Fredrick 1966-70
Theo J. Kneip 1971-79
Richard Thompson 1980-82
Michael Kleinman 1983-

* Terminated 1980
 INTERSOCIETY COMMITTEE xi

#7 (Metals II)*
Elbert C. Tabor 1966-70
Robert E. Kupel 1970-72

#8 (Radioactive Compounds)
Bernard Shleien 1966-72
Andrew P. H u l l 1972-83, 1987-
Donald C. Bogen 1984-87

#9 (General Precautions)
Morris Katz 1967-69
A . L . Linch 1969-79
A l v i n L . Vander Kolk 1980-85
John N . Harman I I I 1986-

#10 (Particulate Matter)

Robert S. Sholtes 1968-71
Robert A . Herrick 1971 -74
Edward Stein 1975-79
Dale A . Lundgren 1980-83
John Watson 1984-

#11 (Stationary Source Sampling)

John S. Nader 1971-73
M . Dean High 1974-80
Knowlton Caplan 1981-83
Walter S. Smith 1984-

#12 (Standardization Coordination)**

William D . Kelley 1971-72
Robert W. Garber 1972-74
Robert Spinas 1975-80

* Terminated 1972
** Terminated 1980
 PERMISSIONS
The Intersociety Committee expresses its
appreciation for permission to use data as
indicated from the following:
Academic Press, O r l a n d o , F l o r i d a
32887, for use in Part I , Section 14, o f data
from "Detection of Volatile Organic Com­
pounds and Toxic Gases i n Humans by
Rapid Infrared Techniques," by R. D .
Stewart and D . S. Erley, i n Progress in
Chemical Toxicology, A . Stolman, E d . ,
Volume 2, 1965, pages 183-200.
The American Conference o f Govern­
mental Industrial Hygienists ( A C G I H ) ,
6500 Glenway Avenue, Cincinnati, Ohio
45211 for use i n Part I , Section 27, of text
from "Detector Tubes, Direct Reading Pas­
sive Badges and Dosimeter Tubes," by B. E.
Saltzman and P. E. Caplan; and for use i n
Part I , Section 29, of text from "Filter Me­
dia for A i r Sampling," by M o r t o n Lipp-
mann in Air Sampling Instruments for
Evaluation of Atmosphere Contaminants,
7th ed., American Conference o f Govern­
mental Industrial Hygienists, Cincinnati,
1988, in press.
The American Industrial Hygiene Asso­
ciation, 475 W o l f Ledges Parkway, A k r o n ,
Ohio 44311-1087 for data i n Part I , Section
13, from " U - V Absorptivity Curves"; and
of material from A I H A Analytical A b ­
stracts, 1965 (UV) for use i n Part I , Section
14, in the American Industrial Hygiene As­
sociation Journal, Volume 32, 412, June
1971.
The American Public Health Associa­
tion, 1015 15th Street, N . W . , Washington,
D . C . 20005 for use in Part I , Section 19, o f
text from Part 11, Section 104A o f Stan­
dard Methods for the Examination of Wa­
ter and Waste Water, 14th ed., 1976.
The American Society for Testing and
Materials, 1916 Race Street, Philadelphia,
Pennsylvania 19103 for use i n Part I , Sec­
tion 11, o f text from E:131; for use i n Part
I , Section 12, o f text A S T M E60; and for
use on Part V o f a table from A S T M E380.
Preston Publications, a Division o f Pres­
ton Industries, Inc., 7800 Merrimar Ave­
nue, Niles, Illinois 60048 for use i n Part I I ,
Method 101, o f "Adsorption Chromatogra­
phy on P L O T Columns: A New Look at
the Future of Capillary GC," by J. de
Zeeuw, R. C. M . de Nijs and L . T.
Henrich, in Journal of Chromatographic
Science, Volume 25, 76, 1987, Figure 6.
John Wiley & Sons, Inc., 605 Third Ave­
nue, New York, New York 10158, for per­
mission to use material i n Part I , Section
20, from "Safety i n the Analytical Labora­
tory," by R. F. Stalzer, J. R. M a r t i n , and
W. E. Railing, in I . M . Kolthoff, P. J. Elv-
ing, and E. B. Sandell, eds., Treatise on
Analytical Chemistry, Part I , Volume 11,
pp. 6743-6752, 1975.
James P. Lodge, Jr. attended the University
of Illinois, with his undergraduate training
there interrupted by two years o f service i n
the U.S. Navy during World War I I . After
graduation, he studied at the University o f
Rochester with Professor V. Boeckelheide,
receiving his doctorate i n 1951. He was
Assistant Professor o f Chemistry at Keuka
College i n New York from 1950 to 1952.
During the summer o f 1952, he joined the
Meteorology Department at the University
of Chicago, and has worked i n the field o f
Atmospheric Chemistry ever since. From
1955 to 1961, he was Section Head i n
Chemistry i n the Community A i r Pollution
Program, U.S. Public Health Service, i n
Cincinnati; this was one o f the predecessors
of the present U.S. Environmental Protec­
tion Agency.
From 1961 to 1974, Dr. Lodge was a Pro­
gram Scientist at the National Center for
Atmospheric Research i n Boulder, Colo­
rado, and since then, has been a Consultant
in Atmospheric Chemistry. He was chair­
man o f the section on air i n the American
Chemical Society report, "Cleaning Our
Environment: The Chemical Basis for
Action," for which the Board o f Directors
of ACS awarded him a Special Certificate.
He held the same position for the revision,
"Cleaning Our Environment, A Chemical
Perspective." Dr. Lodge has also received
the Distinguished Service Award o f the D i ­
vision of Environmental Chemistry, ACS,
and the Frank A . Chambers Award o f the
A i r Pollution Control Association.
As Chairman o f the Colorado State A i r
Pollution Control Commission for six
years, Dr. Lodge received several local
awards. He has contributed to the fields o f
air pollution analysis, atmospheric kinetics
and photochemistry, atmospheric electric­
ity, and theory o f filtration, publishing over
200 papers. I n addition to serving as West­
ern Hemisphere Editor o f Atmospheric En­
vironment, he is its principal book re­
viewer. Dr. Lodge has also had several
papers published in the field o f his hobby,
the study o f theology.
 TABLE OF CONTENTS
Page
Preface v
Publisher's Preface vii
Committees ix
Permissions , xii

PART I GENERAL TECHNIQUES

1. Physical Precautions 3
2. Calibration Procedures 15
3. Dynamic Calibration o f A i r Analysis Systems with Permeation Tubes . . . 18
4. Sampling and Storage o f Particles 27
5. Sampling and Storage o f Gases and Vapors 37
6. Use and Care of Volumetric Glassware 42
7. Reagent Water 47
8. Common Acid, A l k a l i , and Other Standard Solutions 48
9. Recovery and Internal Standard Procedures 55
10. Interferences 56
11. Terms and Symbols Relating to Molecular Spectroscopy 59
12. Photometric Methods for Chemical Analysis 67
13. Ultraviolet Absorption Spectroscopy 76
14. Infrared Absorption Spectroscopy 78
15. Atomic Absorption Spectroscopy 83
16. Inductively Coupled Plasma Emission Spectroscopy 89
17. Gas Chromatography 93
18. Radioactivity Analysis 103
19. Precision and Accuracy 106
20. General Safety Practices 110
21. A i r Purification 120
22. Liquid Chromatography 127
23. Thin-Layer Chromatography 131
24. Instrumental Neutron Activation Analysis o f Atmospheric Particulate
Matter 143
25. Use of Selective Ion Electrodes to Determine A i r Pollutant Species
(Ambient and Source Level) 150
26. Quality Control for Sampling and Analysis 156
27. Direct Reading Colorimetric Indicators 171
28. Fluorescence Spectrophotometry 187
29. Sampling Aerosols by Filtration 190
30. X-Ray Powder Diffraction 214
31. X-Ray Fluorescence Analysis 218
32. Chemiluminescent Analyzers 223
33. Anodic Stripping Voltammetry (ASV) 226
34. Ion Chromatography 230
xvi TABLE OF CONTENTS

PART II METHODS FOR AMBIENT AIR SAMPLING

AND ANALYSIS

100 Carbon Compounds

A . Hydrocarbons
101 Determination o f C Through C Atmospheric Hydrocarbons . .
x 5 243
102 Separation and Determination o f Polynuclear Aromatic
Hydrocarbons and Benzo[a]pyrene 249
102A Extraction and Cleanup Procedures for Polynuclear Aromatic
Hydrocarbons i n Atmospheric Particulate Matter 249
102B Separation and Microanalysis of Airborne Particulate Matter
for Benzo[a]pyrene Using Thin Layer Chromatography and
Spectrofluorimetry 251
102C Measurement o f Benzo[a]pyrene and Benzo[k]fluoranthene by
Spectrofluorimetry 253
102D Measurement o f Polynuclear Aromatic Hydrocarbons Using
Liquid Chromatography with Fluorescence Detection 257
108 Continuous Determination o f Total Hydrocarbons i n the
Atmosphere (Flame Ionization Method) 262
109 Flame Ionization Detector 265

B. Related Organic Compounds

114 Determination o f Acrolein Content o f the Atmosphere
(Colorimetric) 271
116 Determination o f Formaldehyde Content o f the Atmosphere
(Colorimetric Method) 274
117 Determination of Formaldehyde Content of the Atmosphere
( M B T H Colorimetric Method —Application to Other
Aldehydes) 279
118 Determination o f Mercaptan Content o f the Atmosphere 285
121 Determination o f Phenols i n the Atmosphere (Gas
Chromatographic Method) 290
122 Determination o f C -C Aldehydes i n Ambient A i r and Source
x 5

Emissions as 2,4-Dinitrophenylhydrazones by H P L C 293

C. Carbon Monoxide, Gas Analysis

128 Determination of Continuous Carbon Monoxide Content o f the
Atmosphere (Nondispersive Infrared Method) 296
133 Determination o f 0 , N , CO, C 0 , and C H (Gas
2 2 2 4

Chromatographic Method) 303

134 Constant Pressure Volumetric Gas Analysis for Oxygen,
Carbon Dioxide, Carbon Monoxide, and Nitrogen (Orsat) 307
135 Determination o f Volatile Organic Compounds i n Architectural
and Industrial Surface Coatings 308
 TABLE OF CONTENTS xvii

200 Halogen and Halogen Compounds

201 Chloride Content o f the Atmosphere 312
202 Determination o f Free Chlorine Content o f the Atmosphere
(Methyl Orange Method) 313
203 Determination o f Fluoride Content o f the Atmosphere and
Plant Tissues (Manual Methods) 316
204 Determination o f Fluoride Content o f the Atmosphere and
Plant Tissues (Semiautomated Method) 332
205 Determination o f Fluoride Content o f Plant Tissues
(Potentiometric Method) 344
206 Determination o f Gaseous and Particulate Fluorides i n the
Atmosphere (Separation and Collection with Sodium
Bicarbonate Coated Glass Tube and Particulate Filter) 347
207 Determination o f Gaseous and Particulate Fluorides i n the
Atmosphere (Separation and Collection with a Double Paper
Tape Sampler) 352

300 Metals
301 Determination of Particulate Antimony Content of the
Atmosphere 357
302 Determination o f Arsenic Content of Atmospheric Particulate
Matter 361
303A General Method for the Preparation o f Non-Tissue
Environmental Samples for Trace Metal Analysis 365
311-322 Determination o f a Number of Metals i n Atmospheric
Particulate Matter 370
317 Determination o f Elemental Mercury i n Ambient and
Workroom A i r by Collection on Silver Wool and Atomic
Absorption Spectroscopy 371
319 Determination o f Molybdenum Content o f Atmospheric
Particulate Matter by Atomic Absorption Spectrophotometry . . 376

400 Inorganic Nitrogen Compounds and Oxidants

401 Determination o f Ammonia i n the Atmosphere (Indophenol
Method) 379
404 Determination o f Nitrate i n Atmospheric Particulate Matter
(Brucine Method) 382
405 Determination o f Nitric Oxide Content o f the Atmosphere 385
406 Determination o f Nitrogen Dioxide Content o f the Atmosphere
(Griess-Saltzman Reaction) 389
407 Determination o f Total Nitrogen Oxides as Nitrate
(Phenoldisulphonic Acid Method) 395
408 Analysis for Atmospheric Nitrogen Dioxide (24-H Average) 399
411 Determination o f Oxidizing Substances i n the Atmosphere 403
413 Determination o f Ozone i n the Atmosphere by Gas-Phase
Chemiluminescence Instruments 407
415 Determination o f Atmospheric Nitric Acid 412
416 Continuous Monitoring o f Atmospheric Nitric Oxide and
Nitrogen Dioxide by Chemiluminescence 415
417 Continuous Monitoring of Ozone i n the Atmosphere by
Ultraviolet Photometric Instruments 422

500 Particulate Matter

501 High Volume Measurement of Size Classified Particulate
Matter 427
xviii TABLE OF CONTENTS

502 Particle Fallout Container Measurement o f Dustfall from the

Atmosphere 440
503 Continuous Tape Sampling o f Coefficient of Haze 446
507 Integrating Nephelometer Measurement o f Scattering
Coefficient and Fine Particle Concentrations 450

600 Radioactivity
601 Determination o f Gross Alpha Radioactivity Content o f the
Atmosphere 458
602 Determination o f Gross Beta Radioactivity Content o f the
Atmosphere 461
603 Determination o f the Iodine-131 Content o f the Atmosphere
(Particulate Filter Charcoal Gamma) 464
606 Determination of Radon-222 Content of the Atmosphere 469
609A Determination o f Tritium Content o f Water Vapor in the
Atmosphere 479
609B Determination of Elemental Tritium Content of the
Atmosphere 483

700 Sulfur Compounds

701 Determination o f Hydrogen Sulfide Content o f the Atmosphere 486
704A Determination o f Sulfur Dioxide Content of the Atmosphere
(Tetrachloromercurate Absorber/Pararosaniline Method) 493
704B Determination o f Sulfur Dioxide Content of the Atmosphere
(Formaldehyde Absorber/Pararosaniline Method) 499
704C Determination of Sulfur Dioxide Content of the Atmosphere
(Hydrogen Peroxide Method) 503
707 Continuous Monitoring o f Atmospheric Sulfur Dioxide with
Amperometric Instruments 506
708 Determination o f Mercaptan Content o f the Atmosphere 511
709 Determination o f Sulfur-Containing Gases i n the Atmosphere
(Continuous Method with Flame Photometric Detector) 512
709A Determination o f Sulfur-Containing Gases in the Atmosphere
(Following Chromatographic Separation, with the FPD) 512
709B Determination o f Sulfur-Containing Gases in the Atmosphere
(Total Gaseous Sulfur with the FPD) 512
711 Determination of Gaseous Sulfuric Acid and Sulfur Dioxide i n
Stack Gases 523
713 Semi-Continuous Determination o f Atmospheric Particulate
Sulfur, Sulfuric Acid and Ammonium Sulfates 529
714 Determination o f Sulfur Dioxide Emissions i n Stack Gases by
Pulsed Fluorescence 533
720A Suppressed A n i o n Chromatography 538
720B Nonsuppressed Anion Chromatography 543
720C Flow Injection Determination o f Aqueous Sulfate
(Methylthymol Blue Method) 548
720D Barium Sulfate Turbidimetry 556
720E Barium Perchlorate Microtitration 559
720F Barium Chloranilate Spectrophotometry 563
730 Determination o f Sulfur i n Particulate Material by X-Ray
Fluorescence 566
 TABLE OF CONTENTS xix

PART III METHODS FOR CHEMICALS IN AIR OF THE

WORKPLACE AND IN BIOLOGICAL SAMPLES

800 Methods for Chemicals in Air of the Workplace and in Biological Samples
801 Determination o f Ammonia in A i r 571
804 As, Se, and Sb in Urine and A i r by Hydride Generation and
Atomic Absorption Spectrometry 572
805 Determination o f Chloride i n A i r 579
806 Determination o f Free Chlorine in A i r 581
807 Determination o f Chromic Acid Mist i n A i r 582
808 Determination o f Cyanide i n A i r 586
809 Determination o f Fluorides and Hydrogen Fluoride i n A i r 589
810 Determination o f Gaseous and Particulate Fluorides in A i r 590
811 Determination o f Fluoride in Urine 594
812 Determination o f Hydrogen Sulfide in A i r 595
815 Determination o f Mercury i n Urine 596
816 Determination o f Nitric Oxide i n A i r 599
817 Determination o f Nitrogen Dioxide in A i r 600
818 Determination o f Nitrogen Dioxide in A i r 601
819 Determination o f Ozone i n A i r 605
821 Determination o f Phosgene i n A i r 606
822 General Atomic Absorption Procedure for Trace Metals i n
Airborne Material Collected on Filters 608
822A General Method for Preparation o f Tissue Samples for
Analysis for Trace Metals 619
822B X-Ray Fluorescence Spectrometry for Multielement Analysis o f
Airborne Particulate and Biological Material 623
824 Determination o f Airborne Sulfates 639
825 Determination o f Acrolein in A i r 645
826 Determination of Acrolein i n A i r 646
827 Determination o f Aromatic Amines i n A i r 649
828 Determination o f Bis-(Chloromethyl) Ether (Bis-CME) i n A i r . . 654
829 Determination o f Chloromethyl Methyl Ether ( C M M E ) and
Bis-Chloromethyl Ether (Bis-CME) in A i r 658
830 Determination o f 3,3'-Dichloro-4,4'-Diaminodiphenylmethane
(MOCA) in A i r 662
831 Determination o f p,p-Diphenylmethane Diisocyanate ( M D I ) in
Air 665
832 Determination o f Nitroglycerin and Ethylene Glycol Dinitrate
(Nitroglycol) i n A i r 669
833 Determination o f N-Nitrosodimethylamine in Ambient A i r 673
834 Determination o f Organic Solvent Vapors in A i r 678
835 Determination of E P N , Malathion and Parathion i n A i r 686
836 Determination o f Total Particulate Aromatic Hydrocarbons
( T p A H ) i n A i r : Ultrasonic Extraction Method 690
837 Determination o f 2,4-Toluenediisocyanate ( T D I ) in A i r 697
xx TABLE OF CONTENTS

PART IV STATE-OF-THE-ART REVIEWS

1. The Measurement o f Strong Acids in Atmospheric Samples 703
2. Determination and Speciation of Ambient Particulate Sulfur
Compounds 715
3. Source Sampling for Regulated A i r Pollutants from Stationary Sources . 731

PART V CONVERSION FACTORS 739

INDEX 741
 PART I
GENERAL TECHNIQUES
 PART I
GENERAL TECHNIQUES
1. Physical Precautions
1.1 HOMOGENEITY OF THE SAMPLE. N O
analytical result regardless o f the accuracy
and precision o f the procedure can be any
better than the quality o f the sample sub­
mitted for analysis. Therefore, the primary
concern o f the investigator must be directed
to the collection o f representative samples
and the homogeneity o f the air mixtures
employed to calibrate both the collection
and analytical systems. Human sensory
perceptions cannot guide the uninitiated
with respect to variations o f contaminant
concentrations i n either space or time. A l ­
most without exception gaseous compo­
nents are either colorless or are present at
such low concentrations as to be effectively
colorless. The human sense o f smell is noto­
riously deceptive with respect to concentra­
tions of even highly odoriferous compo­
nents, and therefore cannot be relied upon
for estimating relative quality or uniformity
of the ambient atmosphere (2,4-Part
3,7,43,44).
The source o f the contaminant, air flow
direction and velocity whether due to wind
or thermal gradients, density of the con­
taminant, intensity o f sunlight, time o f day,
presence o f obstructions such as trees,
buildings, partitions, machinery, etc.,
which act as baffles to produce turbulence,
humidity, and half-life o f the contaminant
together determine the concentration at any
given location. That the concentration can
vary by several orders o f magnitude within
a relatively short radius from the point o f
reference has been amply confirmed by
many investigators. A i r may flow either in
a stream line or turbulent flow and this fac­
tor alone can determine the dispersal pat­
tern o f the pollutant (2,3,5,7,46).
The density o f the pollutant i n many
cases will counteract diffusion processes to
an extent which will establish stratification.
This phenomenon prevails i n the case o f
many highly toxic and irritant gases such as
phosgene, mustard gas, lewisite, chlorine,
etc. (5). Natural convective circulation and
diffusion in confined spaces such as silos,
coal mines, wells, caves, tanks, etc., is not
3
sufficient to maintain a normal atmosphere
containing approximately 2 1 % oxygen.
The location of sampling sites whether
on a temporary or "grab" sample basis or
on a fixed station basis for continuous
monitoring over long periods of time is of­
ten critical and will determine the validity
of the conclusions drawn from the results
(1,2,3,7,46). I n the vicinity of industrial at­
mospheres the variability may be suffi­
ciently extreme to vitiate the results from
fixed station monitors regardless o f the
care taken i n locating the samplers. I n an
alkyl lead manufacturing facility, for exam­
ple, no relationship could be established be­
tween exposure as indicated by urinary lead
excretion and numerous fixed station sam­
plers. I n this case, correlation was not es­
tablished until the monitors i n the form of a
small filter-microimpinger assembly were
actually worn during the entire 8-h shift by
the workmen assigned to these areas (1).
Several excellent treatises have been pub­
lished on planning the air pollution survey
and the effects of meteorological and geo­
graphical conditions on the validity o f the
results. This subject is outside o f the field
of General Techniques and Precautions;
therefore, the reader is advised to consult
one or more o f these publications for full
details (2,3,6,7,32,33,34,47).
On a smaller scale, even greater care is
required to establish uniform mixtures o f
pollutants i n air for use as calibration mix­
tures whether the system be dynamic or
static i n principle (9). Passage through
lengths o f small-diameter tubing to ensure
turbulent mixing is one relatively simple ex­
pedient for dynamic mixing (3). Other sys­
tems employ some form o f baffled cham­
ber inserted between the mixing ports and
the delivery point (10). Rotating paddles or
even loose pieces o f sheet metal or inert
plastic to be activated by shaking have been
inserted i n containers for static mixtures for
mixing by turbulence (11). Probably a more
effective alternative is to evacuate the con­
tainer, then mix the contaminant with the
air at a uniform rate as it is released into the
void (11). (Inject from a hypodermic sy­
ringe through a needle inserted through a
4 GENERAL TECHNIQUES
septum.) Diffusion alone cannot be relied
upon to produce a homogeneous mixture in
most cases.
Other effects such as container wall ab­
sorption, humidity, volumetric errors, etc.,
discussed elsewhere in this section must be
recognized and appropriate precautions
taken to avoid concentration errors derived
from these seemingly extraneous sources
when attempts are made to produce homo­
geneous, known concentrations for calibra­
tion purposes.
1.2 A B S O R P T I O N E F F E C T S O N C O N T A I N E R
W A L L S AND CONNECTING TUBES. Failure
to recognize absorption effects can lead to
serious errors both i n collection o f samples
and in preparation o f calibration mixtures,
especially i n static systems. Losses o f ana-
lyte can occur either by primary adsorp­
tion, which may be reversible i n response to
pressure or temperature changes, or by sec­
ondary adsorption, which may involve
chemical reaction with another material
previously adsorbed (2,4,7). Long-term sta­
bility and accuracy o f calibration blends
have been studied as a function o f the con­
tainer material by several investigators.
(48,49,50,51,52). The affinity o f glass for
water, nitrogen dioxide, benzene, aniline,
the isocyanates, etc., has been recognized
(8,12,13,31). The interior surface of a 5-
gallon (19-L) borosilicate glass carboy can
adsorb as much as 50% o f the benzene in
the 100 to 200 ppm range from a calibration
mixture in air within 30 min after prepara­
tion. These losses can be reduced to some
extent by preconditioning at a concentra­
tion higher than the working level but de-
sorption may contribute positive errors
(Table 1:1). Losses by adsorption of poly-
nuclear aromatic compounds on glass have
also been reported (14).
This adsorption effect can be minimized
in the case of collection of air samples i n
evacuated glass flasks (Shepherd flasks) or
sampling tubes by the introduction of a
suitable solvent for the vapor, such as iso-
octane for benzene and xylidine, directly
into the container with thorough rinsing o f
the inner surfaces (31). I n some cases, such
as the analysis for nitrogen dioxide, the sol­
vent or reagent is introduced before the
sample is taken; e.g., 100 m L sample drawn
into a syringe containing 10 m L o f Saltz-
man's reagent (15).
Nitrogen oxides are so tenaciously ad­
sorbed on glass surfaces that rinsing with
Saltzman's reagent has been found neces­
sary to obtain reliable results. Adsorption
of isocyanates on glass, metal or plastic sur­
faces occurs to such a degree as to preclude
preparation of static calibration mixtures
even under strictly anhydrous conditions
(8).
Positive errors can be included when
evacuated double entry gas sampling tubes
(fitted with 2 stopcocks) are used for sam­
ple collection (7,16). I n this case, after re­
lease o f the vacuum, additional sample is
drawn through the tube for 3 min before
closing the stopcocks. Recovery of xylidine
increased to 120% in the 50 to 70 ppm
range under these conditions. This error
was increased to 40% by adding glass beads
to the sampling flask to increase the surface
area (31). These conditions can be avoided
by collection o f the pollutant i n an i m -
pinger assembly (Greenburg-Smith, midget
or micro).
I n dynamic systems, prolonged equilibra­
tion is required before dependable calibra­
tion can be assured. Calibration o f toluene
diisocyanate ( T D I ) in the T L V range re­
quired passage o f the air-TDI vapor mix­
ture through a mixing chamber composed
of a 6' (1.8 m) length o f V4" (6 mm) copper
tubing under strictly anhydrous conditions
for 60 to 72 h to reach equilibrium. This
phenomenon was not entirely caused by the
film o f water normally adsorbed on the i n ­
ner walls o f the tubing, as the system had
been carefully heated under vacuum to re­
move volatile impurities before use (8).
Adsorption is a critical factor i n making
up standard mixtures for phosgene calibra­
tion at concentrations below 1 ppm even
after meticulous removal of the adsorbed
water film. The use o f metal cylinders espe­
cially must be avoided. I n the development
of a continuous monitor for phosgene con­
centrations below 5 ppm, decomposition i n
metal (steel, stainless steel, copper, alumi­
num, etc.) or glass tubing was found to be
so extensive that their use i n the sampling
system could not be considered. I n this case
polyethylene tubing was acceptable. Water
adsorbed to the contact surfaces was proba­
bly responsible for the losses as HC1 was
observed to be a by-product. Catalytic de-
 PHYSICAL PRECAUTIONS 5

Table 1:1 Static Calibration Mixture

Benzene Loss by Absorption on Container Walls
Borosilicate Glass Temperature 73 ± 5°F (24 ± 3°C)
Container
Time Conditioned Remarks
Elapsed Benzene PPM Percent Time Benzene Inside Surface
Hours Added Found Loss Hours PPM Approx. 800 Sq. In. (0.52 m2)
0.5 100 56 44 0 none
0.5 100 67 33 0 none
16 200 100 50 0 none
0.5 100 84 16 16 100
1 — 78 22 — —
0.5 100 87 13 16 200
0.5 100 107 -7 — — Swept, 2nd 100 ppm added
0.5 200 128 36 — — 3rd 100 ppm added
0.5 50 56 -12 Swept, 50 ppm added
- -
0.1 100 75 25 0 none Idle 30 days
0.25 — 72 28 — —
0.5 — 55 45 — —
1.0 — 55 45 — —
4.5 — 48 52 — —
24 — 42 58 — —
192 — 0 100 — —
Note: Five-gallon (19-L) borosilicate glass carboy jacketed with a canvas cover. Calculated quantity
of benzene was added and mixed by rotating a polytetrafluoroethylene (PTFE) paddle and sampled at
the indicated time after addition. In two cases indicated by 16 h (column 5), equilibrated by standing
overnight, then refilled.

composition on anhydrous surfaces cannot
be ruled out, however (17).
I n a study o f the efficiency o f midget and
micro-impingers by collection o f aniline
from air, loss by adsorption i n the glass
batch-type generator (5 to 15% i n the 100
fig range) was encountered unless the sys­
tem was heated to 60°C i n a hot water bath.
Even greater were losses due to absorption
by gum rubber tubing. Even 3-inch (8-cm)
connections introduced significant errors
(10% loss) but the losses were not entirely
proportioned to hose length. These losses
increased as the age and number o f uses
increased; e.g., recoveries decreased from
83 to 57% over a period of 8 days (18).
Failure to recognize the effect o f adsorp­
tion on container walls delayed the determi­
nation o f the half-life o f S 0 during a study
2 phasize the importance o f this source o f er­
of the fate o f S 0 i n the atmosphere at Syr­
2 rors. Heavy metal ions, especially lead,
acuse University. The half-life was found to
be unexpectedly short even after correction
for adsorption effects (19).
Tubing through which high concentra­
tions o f tetraethyl lead vapor have been
drawn will become a source o f contamina­
tion when "clean" air is passed through the
system. These effects are eliminated by use
of all-glass systems connected by standard
taper, ball joints, " O " ring seals or polytet-
rafluoroethylene (Teflon) compression
connectors. Chemical transformation oc­
curred when sampling oxides o f nitrogen
through sampling probes from a simulated
combustion source. The species inter con­
version or loss was a function o f tempera­
ture and sample probe material and was
found to be a function of the air-fuel ratio
of the input sample. Glass sampling probes
were found to be inert, while the stainless
steel probes were found to be reactive (53).
The adsorption o f traces o f analytes
from liquid solution onto container walls is
probably somewhat more widely recog­
nized but certainly requires review to reem-
mercury (22), silver (20,21), cadmium, and
antimony, are readily adsorbed by glass and
in some cases by plastic surfaces. Lead i n
urine specimens, for example, is adsorbed
to an extent sufficient to introduce signifi­
cant errors after storage for more than 3
days i n borosilicate glass specimen bottles
6 GENERAL TECHNIQUES
(23). This effect probably is produced by
bacterial fermentation which elevates the
p H by ammonia release, and coprecipitates
lead with calcium and magnesium phos­
phates on the glass. I n the early stages the
precipitate is invisible but nevertheless may
contain a significant fraction o f the total
lead i n the sample. I n many cases these
losses can be held within tolerable limits by
reducing the p H to 2 or less with nitric acid.
This adsorption loss can be a critical factor
in the storage o f standard reference solu­
tions and must be taken into account when
collaborative testing programs are set up.
Efforts to prepare stable aqueous standard­
izing solutions of metallic mercury met with
failure unless excess mercury was main­
tained i n contact with the liquid phase. Fur­
thermore, centrifugation is necessary to re­
move dispersed microdroplets (or HgO) i f
shaking is employed to accelerate satura­
tion, and a relatively large temperature co­
efficient o f solubility must be taken into
account. I n spite o f these limitations, this
saturated aqueous mercury solution can be
employed as a convenient source of mer­
cury for calibration by comparing recovery
from a collecting system with direct analy­
sis o f the standard (22).
Filtration of samples collected i n a liquid
medium should in all cases be avoided i f the
desired contaminant is to be recovered
quantitatively i n the liquid phase. N o t only
will the contact surfaces o f the filter funnel
and the filtrate receiver adsorb trace mate­
rials, but the filter medium (paper, mem­
brane, asbestos mat, glass fiber, micro me­
tallic, etc.) may be highly adsorbent (14).
Also, the insoluble fraction removed may
retain the analyte as i n the case when cal­
cium phosphate is separated from a mixture
containing traces o f soluble lead com­
pounds (24).
The problem of adsorption on glass sur­
faces previously discussed requires re-
emphasis i n the case o f porous glass used
either as a filter for particulate matter or as
a diffuser for collection i n liquid absorp­
tion systems. Filtration o f aqueous solu­
tions o f phenanthrene, naphthalene, pyrene
or anthracene through a porous glass plate
reduced the hydrocarbon concentration as
much as 40% (14). When ultraviolet (UV)
absorption is employed for analysis, errors
from background absorption produced
from UV-absorbing materials leached from
the filters and light scattering from sus­
pended particles must be taken into account
(23). A base-line technique for correction is
often required as the solute often alters the
light scattering and sedimentation proper­
ties with the result that the reference blank
and the sample are no longer exactly com­
parable. I f removal o f sediment is neces­
sary, filtration through polytetrafluoroeth-
ylene (PTFE) mat or decantation after
centrifugation will minimize losses. I n any
case, potential losses at each step must be
recognized and determined.
Rust, scale and other metallic corrosion
products can introduce serious errors
through adsorption o f air pollution compo­
nents and should be meticulously scoured
out o f systems used for air sampling, stor­
age or analysis. Many o f these oxidation
and corrosion products are highly catalytic,
as illustrated by the rate at which allkyl lead
compounds are oxidized by hydrated iron
and bismuth oxides (25). Pipe dope used on
threaded connections and oils used i n cut­
ting threads or lubricating moving parts
also can present significant reversible ab­
sorption effect. Polytetrafluoroethylene
(PTFE) ribbon thread sealant, which has a
very low absorption coefficient, is recom­
mended for threaded joints that require l u ­
brication. Moving parts machined from
PTFE are self-lubricating and can be sealed
with modified PTFE " O " rings.
The reverse effect has been encountered
also. The diffusion o f H F i n Conway dishes
is enhanced ten-fold when silicone grease is
used as a sealant for the dual concentric
chambered dish cover. Fluoride can be re­
moved quantitatively from aqueous H C 1
by gentle boiling i n the presence o f silicone
fluid (26).
Adsorption of a secondary layer on con­
tainer walls may completely alter the charac­
ter o f the surface presented to the collection
and analytical system. Soaps and synthetic
detergents are especially troublesome i n this
respect. The film o f soap on glass is not re­
moved by 5% trisodium phosphate, chromic
acid cleaning mixture, or concentrated nitric
acid. Only by baking out in an annealing
furnace can "chemical sterility" be achieved.
In the colorimetric determination of aro­
matic nitro and amino compounds in air,
this soap f i l m reacts with the color-

generating reagents. These adsorbed layers
appear to retain nitrous acid or nitrogen ox­
ides in a state that does not react with starch-
iodide indicator or sulfamic acid but does
react with the coupling agent (1-naphthyl
ethylenediamine or Chicago Acid) to pro­
duce a deep orange color. This spurious
color reaction produces high results, which
may be sufficiently uniform to produce false
calibration curves (zero intercept at some
light transmittance value below 100%). To
avoid this nitrite-absorbing contamination
and remove residual azo dyes which may be
adsorbed on the coupling flasks and spectro­
photometer cuettes, soak all glassware in 5%
aqueous trisodium phosphate for several
hours and rinse thoroughly with distilled wa­
ter. Soak Corex cuvettes only in 37% nitric
acid wash (27).
The nonabsorption o f heavy metal ions
on hydrophobic surfaces o f PTFE and
polyethylene (PE) may be reversed by the
presence o f certain organic anions. I n trace
quantities, cesium is adsorbed on glass sur­
faces by interaction with the layer o f ad­
sorbed hydroxyl ions that account for the
hydrophilic character o f glass surfaces. A d ­
dition o f a salt o f high ionic activity such as
sodium nitrate reduces this adsorption ef­
fect. The ion exchange capacity o f the glass
surface can also become a significant factor
at very low concentrations. The overall ef­
fect is orders o f magnitude less (adsorption
coefficient increases in the order PTFE-PE-
glass) for the nonionic hydrophobic plastic
surfaces. However, the absorption coeffi­
cients reverse when the tetraphenyl borate
(TPB) ion is introduced. This reversal is
due to adsorption o f TPB ions by van der
Waals forces on the hydrophobic PTFE
and PE which then sequester cesium ions.
On the hydrophilic glass surface the ad­
sorption is ionic i n character, which reduces
the binding sites for the cesium ions (28).
Lubricants i n stopcocks and ground glass
joints must be avoided as these materials
are a source o f contamination and/or ab­
sorb those constituents o f interest either di­
rectly from the air sample or from the liq­
uid reagent system after collection. Teflon
plug cocks for stopcocks and sleeves for
joints provide seals that require no lubrica­
tion or liquid-phase sealant for air-tight
joints.
Fungus growth within an atmospheric
PHYSICAL PRECAUTIONS 7
analyzer can be a particularly vexing prob­
lem. Fungus within an S 0 analyzer that
2
uses liquid reagents results in an indicated
concentration lower than that actually
present. Fungus effect can be minimized i n
an analyzer by: (a) Keeping the instrument
running continually, (b) cleaning periodi­
cally, and (c) using a small amount o f fun­
gicide i n the reagent. A fungus problem will
usually be found on dynamic calibration.
1.3 SOLID ADSORBENT C O L L E C T I N G SYS­
TEMS. Extended surface adsorbents such as
activated silica gels, alumina gels, charcoals
and carbons, glass beads, and spheres offer
a very attractive solution to the collection
and concentration o f gas phase pollutants
from atmospheric samples. However, the
tenacity o f the forces that provide quantita­
tive extraction o f the components sought
may prove to be self-defeating when quan­
titative removal o f these same components
for analysis is attempted (54). Failure to
recognize and equate these factors may,
and has, produced serious errors. Mercury,
either as metallic vapor or as volatile or­
ganic derivatives, is not removed quantita­
tively either by heating or elution with liq­
uid reagents from silica gel. The same
deficiency has been encountered in at­
tempts to use activated charcoal for collec­
tion o f alkyl lead compounds.
Since the polarity o f the adsorbed ele­
ment or compound determines the binding
strength on silica and alumina gels, compo­
nents o f high polarity will displace compo­
nents o f lesser polarity —the principle o f
chromatography. Therefore, i n attempting
to collect relatively nonpolar compounds
such as benzene, the presence o f coexisting
polar compounds such as phenol, acetic
acid or ammonium chloride must be recog­
nized as a source of interference i n the use
of solid adsorbents for sample collection.
Under high humidity conditions the ab­
sorbent may be deactivated by saturation
with water vapor (23). Activated charcoal
exhibits this effect only to a minor degree.
Aromatic hydrocarbons are not removed
quantitatively from charcoal (32).
Selective displacement may be taken ad­
vantage o f in eluting the sample from the
adsorbent. A n inert carrier gas with or
without heat, for example, may be used to
separate a mixture o f aromatic hydrocar­
bons from a silica gel column. Proper
8 GENERAL TECHNIQUES
choice o f eluting solvent may well serve to
separate components by liquid extraction
also. A n example would be separation o f
nitrobenzene from aromatic hydrocarbons
for analysis by U V spectrophotometry.
Liquid eluents must be chosen and used
with care however as decrepitation o f the
gel may occur. The finely dispersed frag­
ments can produce difficulties i n the deter­
mination o f optical density in the final step
of the analysis. This problem is especially
acute in U V spectroscopy. Usually this ef­
fect is caused by the heat generated when a
polar solvent such as ethanol is introduced.
Dilution with a nonpolar solvent such as
isooctane will usually avoid this difficulty
(23). Elution from charcoal is dependent on
molecular size rather than polarity (CS ,
C C 1 , etc.)
4 rials, which include natural rubber, neo-
Change in air flow resistance due to ei­
ther swelling, shrinking or channeling o f
the adsorbent is sometimes encountered.
Collection o f alkyl lead vapor in crystals o f
iodine was plagued with channeling and
shrinkage due to iodine sublimation. A d d i ­
tion o f a head o f fine inert sea sand elimi­
nated the problem by providing enough
pressure to keep the iodine crystals com­
pacted without appreciably altering the
pressure drop across the column (29).
Many dusts are highly adsorbent and of­
ten carry a significant gaseous component
present in trace amounts such as S 0 at
concentrations below 0.5 ppm. I f a prefil-
ter is installed in the collection o f trace con­
taminants, particles collected on the filter
must be analyzed also i f a quantitative ac­
counting for the total contaminant present
is desired (30). I n some cases such as in the
analysis for lead i n air, the filter would be
analyzed separately from the fraction col­
lected in the liquid reagent to provide a dif­
ferential analysis for organic and inorganic
lead. This is especially important in cases
where the TLVs for the two forms are dif­
ferent and coefficients must be used for
hazard evaluation (1).
The use o f pretreatment or guard systems
such as solid dehydrating agents to remove
moisture, soda lime to remove C 0 , or sul­
2 calibration mixtures o f certain trace com­
furic acid on silica gel to remove N H inter­3
ference, and liquid scrubbers such as an ox­
idizing solution to remove reducing
interferences such as S 0 and H S before
2 2
adsorption o f the desired pollutant can i n ­
troduce serious errors (23,35). Silica based
and anhydrous CaS0 granules may parti­
4
ally or completely adsorb the subject ele­
ment or compound as well as eliminate the
interference in the air sample. Careful cali­
bration o f the collection system under use
conditions is critical when interferences
must be eliminated from the air sample by
pretreatment before collection o f the de­
sired component.
1.4 DIFFUSION EFFECTS OF PLASTICS.
Diffusion outward from the air sample,
into, or through plastic containers or tub­
ing probably is one o f the most frequent
sources o f error encountered in air analysis.
These irreversible losses are in addition to
the reversible surface adsorption effects
2 noted under Section 1.2. Elastomeric mate­
prene and plasticized polyvinyl chloride
(PVC) are especially troublesome. (See sec­
tion 3.1 on Permeation Tubes — Dynamic
Calibration.)
The plasticizers employed in the manu­
facture o f flexible tubing from PVC type
polymers may act as cosolvents to remove
organic components from air samples by
absorption. This mechanism is often tem­
perature sensitive and to some extent may
be reversible in a fashion resembling gas
chromatography column packings. I n a
study o f the recovery o f aniline vapor from
2 air, losses as high as 1.7%/cm o f exposed
gum rubber tubing were encountered (18).
Similar experience has been obtained with
alkyl lead and mercury compounds, aro­
matic hydrocarbons (benzene, toluene,
etc.), chlorinated hydrocarbons, and aro­
matic nitro compounds. A n all-glass system
with joints and connections fitted with
PTFE gaskets and the shortest possible
path between air sample intake and the col­
lecting medium to a great extent eliminates
these losses. However, adsorption effects
must not be neglected even in all-glass sys­
tems.
The permeability o f even the inert plas­
tics exemplified by PTFE has been em­
ployed for the constant rate generation o f
ponents such as S 0 , H F , C O C l , aromatic
2 2
hydrocarbons, etc. Although this property
would be undesirable in the collection o f air
samples, the magnitude o f loss by diffusion
through short inert plastic sampling lines

(PTFE, polyethylene, polypropylene) can
be neglected. However, when samples are
stored i n plastic bottles or bags, diffusion
may become a controlling factor. Indica­
tion o f contamination from fluoropolymer
bags used i n air pollution studies has also
been noted (45). Plastic containers must be
used to facilitate collection or transporta­
tion; then the diffusion rate factor must be
determined for each component sought
over the concentration and temperature
range to be encountered and a correction
factor applied to correct for the time inter­
val between collection and analysis. A l u m i ­
num foil-lined polyester bags (Mylar),
which materially reduce diffusion losses,
are available for C 0 , CO, water vapor and
2 would be a reasonable precaution to avoid
some other gases but are not recommended
for the highly polar gases ( S 0 , N 0 and
2 2
O3) (36). Polyester film in 2-mil (50 fim)
thickness is considered the best all-around
material for sample-bag fabrication. How­
ever, polyvinylidene chloride (Saran) film
(1-mil [25 fim] thickness) which has been
pre-aged to reduce outgassing o f com­
pounds from the plastic into the contents o f
the bag has been widely accepted on the
basis o f inertness to many common gases
and vapors, typical high tensile strength
and heat scalability. Data relative to gas
transmission is summarized i n the follow­
ing table.
Gas Transmission: c m V m fim d atm at
2
23°C
0 32-43
2
C0 150-236
2
N 4.8-6.3
2
Air 8.3-17.3
Water Vapor Transmission: g / m d 2
at 2 4 ° C - 0 . 3 1
at 3 8 ° C - 3 . 0 5
For ultimate chemical inertness fluoroplas-
tic film bags are available. Laminated
plastics — polyethylene on polyester, poly­
ethylene on cellulose acetate and polypro­
pylene on either base —used i n packaging
foods offer interesting alternatives for gas-
impervious containers and a choice o f sur­
face for contact with the air sample. Semi-
flexible 10-L polyethylene bottles with
greater wall thickness also offer a solution
PHYSICAL PRECAUTIONS 9
to the diffusion problem and are conven­
ient for preparation o f calibration mix­
tures. (See Static Calibration, section 2.1.)
Inward diffusion also may be encoun­
tered and may create problems i n shipment
or i n prolonged storage. Samples taken i n a
low humidity environment may be altered
by inward diffusion o f water vapor when
the plastic container is stored i n a high-
humidity climate. Solvent vapors, gasoline
vapors, and gasoline and jet engine exhaust
components may contaminate the sample i n
shipment and show up as unexplained
peaks i n gas chromatographic analysis, or
introduce interferences i n wet chemical pro­
cedures. Shipment i n air-tight containers
contamination from this source.
Although obvious, the testing o f all plas­
tic containers, especially thin film bags, for
gross leaks before use must be emphasized.
The absence o f pin holes, gaps i n the seams,
and leakage around the sampling port and
in the retaining valve or septum assembly
must be verified before each use. This ex­
amination may consist o f full inflation un­
der slight positive pressure (2 to 3" water,
0.5-0.8 kPa) under water or i n a rigid
sealed container with the air line introduced
through a bulkhead fitting and connected
to a water manometer. Soap solution stabi­
lized with glycerol may be painted or
sprayed on for leak detection to pinpoint
"micro" leaks.
1.5 M E C H A N I C A L D E F E C T S O F S A M P L I N G
E Q U I P M E N T . A S well recognized as this
source o f error has become, leaks and
faulty seals remain one o f the major con­
tributors to unreliable results. A n y system
used to collect air samples, to remove con­
tamination components by absorption, or
to calibrate such systems must be checked
for leakage, either inward or outward de­
pending upon whether the air stream is
pushed or pulled. Perhaps the most expedi­
ent procedure involves either an increase or
decrease o f the pressure i n the system
within limits o f the maximum operational
range o f the system, by closing o f f tightly
and measuring the rate o f loss o f either
pressure or vacuum with a water manome­
ter. Pinholes i n glass seals, especially ring
seals; plastic or metal tubing; joints that do
not seal perfectly or are insufficiently gas-
keted; " O " rings that may be cracked, loose
10 GENERAL TECHNIQUES
or hardened or lack a sealing lubricant;
loose glass-to-rubber or plastic tubing con­
nections; dirt i n glass-to-glass joints; glass
plugcocks that do not perfectly mate with
the stopcock barrels; poor valve stem pack­
ing or insufficient sealing pressure; and
poorly sealed threaded fittings provide the
most frequent leak sources. Pinholes i n
stainless steel valves, reducing valve dia­
phragms, and two stage regulators also
have contributed their share o f difficulties
(10). A review o f the requirements for gases
to be used with gas chromatographic sys­
tems and the overall gas sampling require­
ments is valuable i n determining how pollu­
tant standards should be handled (55).
Fouling o f rate meters and total volume
meters by particulate matter or corrosion
products often is not so obvious and re­
quires eternal vigilance to minimize. Orifice
type and positive displacement meters and
rotameters should be protected by an effi­
cient filter to prevent reduction i n orifice
opening, erosion o f moving parts, or com­
plete plugging. Rotameter calibration can
be shifted significantly by dust accumula­
tion, and plastic tube models require pro­
tection from solvents which soften or ren­
der the bore tacky. Operating
characteristics o f flow meters were dis­
cussed (56). Metal floats are attacked by
corrosive gases and some reagents such as
iodine may in time render the tube body
opaque. Corrosive gases and aerosols must
be rigidly excluded from total volume and
positive displacement meters to maintain
calibration validity and integrity of moving
vanes or pistons. Liquid aerosols may be
especially troublesome by forming a liquid
film that by surface tension and capillary
effects reduces air flow i n small orifice
throats, interferes with the movement of
rotameter floats, and increases the resist­
ance to movement o f critical components i n
positive displacement meters. Galvanic cor­
rosion from airborne electrolytes under
high-humidity conditions can be a problem
with metal components.
Flow meters based on thermal conductiv­
ity (hot filament type) also are affected by
dust and reactive gases, but i n a different
fashion. Corrosion will alter the electrical
conductivity and leads ultimately to failure
of the element, and coating by inert mate­
rial will reduce heat transfer and ultimately
reduce the response o f the instrument.
Chlorides and lead are especially destruc­
tive. The volumetric meter and the pump or
aspirator sub-assembly o f any air sampling
system should be adequately protected by
appropriate traps that are renewed fre­
quently. Activated charcoal granules (5 to
10 mesh) probably offer the most efficient
protection for all but a few destructive
gases and vapors. Cellulose acetate mem­
brane filters (0.8 /mi pore size) will effec­
tively remove over 99% o f the dust particles
that might affect instrument performance
without introducing prohibitive pressure
drop (37).
The orifice i n impinger-type collecting
devices, and the interstices i n porous glass
diffusers are also vulnerable to partial or
complete closure by particulate matter and
should be protected by a filter when the
solid fraction is not an integral part o f the
sample. Porous glass diff users may accu­
mulate particles that are not removed by
conventional cleaning procedures after re­
peated use (10). Also irreversible retention
of analyte sometimes occurs. I n the collec­
tion o f alkyl lead vapors from 5 to 10% o f
known quantities added to calibration mix­
tures were retained i n a pencil-type diff user
substituted for the impinger tube i n the
standard midget impinger sampler (31).
The increase i n pressure drop produced by
the decrease i n free cross-sectional area will
contribute to volumetric sampling errors,
and may significantly reduce the collection
efficiency. I n dust collection by impinge­
ment, this factor becomes critical.
Pressure drop must be taken into account
when a sampling system is calibrated volu-
metrically. Rotameters and orifice type me­
ters should be calibrated with the pressure
reducing or elevating components i n place
or volume correction made to compensate
for the deviations from ambient conditions
(38). I n systems based on filter collection,
the pressure drop across the system may
steadily increase as the filter medium loads
up with retained particles. Under these con­
ditions an averaged correction may be
within tolerable limits o f volumetric error.
Changes i n barometric pressure usually in­
troduce only marginal error i f ignored. (De­
viation o f 2 from 101.3 kPa will introduce a
2% volumetric error).
In the use of cold traps to collect contami-

nated air components (freeze-out technique),
consideration must be given to the possible
accumulation of ice crystals in the delivery
tube. This possibility can be minimized by
introducing the sample through a side arm
into the chamber and exhausting through the
inner tube, i.e., reverse flow (11).
1.6 PARTIAL VAPOR PRESSURE EFFECTS.
In collection systems that depend exclu­
sively upon solubility o f the component
sought in a nonreactive solvent rather than
retention by chemical reaction to a nonvol­
atile derivative, the amount absorbable is
determined by the partial pressure of ab-
sorbate. The ratio of the air sample volume
to the volume of absorbant is a critical fac­
tor in collection efficiency and there is a
limiting value of the ratio for any system
that cannot be exceeded i f a given degree of
efficiency is to be attained. I n the case o f
acetone in water the limit would be approx­
imately 5 L of air through 20 m L of water
aliquoted between two bubblers (39). I n the
case of mesityl oxide, which has poor solu­
bility in water to begin with, absorption ef­
ficiency would be very poor although rela­
tive volatility is low. This is especially
critical where a low T L V is involved, as
large volumes of air would be required. The
formation of lower-boiling binary and ter­
nary mixtures also can limit absorption.
Failure to collect benzene vapor in ethanol
even when chilled undoubtedly can be at­
tributed to revaporization of the higher
vapor-pressure ternary alcohol-benzene-
water mixture.
In general, the vapor pressure of the vol­
atile analyte should be reduced to the low­
est practical limit by refrigeration to pro­
mote efficient collection. I n many cases a
freeze-out trap, without solvent, in solid
carbon dioxide, liquid air, or a mixture o f
liquid air and ethanol proves to be a more
effective procedure (11).
Sampling in the region at or close to the
vapor saturation limit can produce serious
errors by condensation in sampling lines.
This condition is most likely to be encoun­
tered when collecting samples of air at tem­
peratures above ambient and may involve
condensation of water vapor which in turn
will entrain or dissolve a significant portion
of the components present in the sample
under high humidity conditions.
PHYSICAL PRECAUTIONS 11
1.7 SOLUBILITY EFFECTS. Limitation
may occur in cases where chemical reaction
should but actually does not occur readily.
The classical example is the failure for S 0 3
to be absorbed in pure water. The collection
of aniline vapor in dilute aqueous H C 1 is
not quantitative; therefore, chilled alcohol
is the medium of choice (13). Until the reac­
tion is initiated by a catalyst, fluorine gas
will bubble through distilled water without
reacting (to produce F 0 and H F ) although
2
once initiated, the reaction may be violent.
Other examples that can be cited are failure
to obtain absorption of alkyl mercury va­
por in acid permanganate, and the less-
than-quantitative absorption of alkyl mer­
cury and lead compounds in aqueous iodine
reagent (22). The assumption that a given
contaminant will be collected based on its
reactivity with the reagents can be disas­
trous i f calibration o f the system under use
conditions is omitted.
I f the product from the reaction of the
analyte with the collecting reagent has a
very low solubility product, losses are to be
expected unless care is exercised to avoid
the saturation region. The precipitate may
adsorb to the container walls, orifice tube,
or diffuser surfaces or fail to react in subse­
quent color development. I f the solid phase
is finely dispersed and has a refractive in­
dex approximately the same as the liquid
system, the losses may not be observed;
e.g., silica in chloroform.
Conversely, i f the collecting train does
not include a prefilter, particulate fractions
may go into solution and produce false pos­
itive results or interferences; e.g., lead dust
or fume in the collection of tetra-alkyl lead
vapor (30).
1.8 T E M P E R A T U R E E F F E C T S . The physi­
cal effects of heat on the collection system
under normal operating conditions is usu­
ally ignored without introducing apprecia­
ble difficulties. However, at the design and
development stages, attention must be
given to the upper and lower temperature
limits to be anticipated. A t low tempera­
tures components of the liquid reagent may
crystallize out or the solvent may freeze.
Frequently, a change of solvent will solve
this problem; e.g., aqueous K I versus me­ 3
thanols iodine reagent (22,29). Close atten­
tion must be given to possible formation o f
ice or reagent crystals in orifices and po-
12 GENERAL TECHNIQUES
rous glass diffusion beds when operating
near the freezing range.
On the other end o f the temperature
scale, excessive evaporation may limit the
choice o f reagents, or sampling time may be
reduced below practical limits (again alco­
holic versus aqueous iodine reagents). Re­
frigeration usually can be applied i n fixed
station or mobile systems t o alleviate this
problem. A portable assembly that derives
its sampling suction from the expansion o f
difluorodichloromethane vapor through an
aspirator also provides useful refrigeration
by forming the vaporizer coil into a cavity
which will receive a midget impinger (11).
In some cases a colorimetric system can be
adapted to a nonvolatile solvent, as illus­
trated by the use o f diethyl phthalate as a
solvent for the detection o f phosgene i n
trace ranges over extended periods o f
t i m e - 4 to 6 h (40).
As noted i n section 1.6 (Partial Vapor
Pressure Effects) the temperature coeffi­
cient may be quite large and preclude col­
lection by solution i n a solvent above a crit­
ical temperature, above which no collection
will be obtained, and which may be only
slightly above room temperature (39). I n
general, the adsorption capacity o f solid
adsorbents is decreased exponentially with
rising temperature, which may limit the
thermal operating range. Temperature co­
efficients should be determined before sam­
pling is attempted (23).
Excessive evaporation at elevated tem­
perature aggravates the scavenging problem
for the metering devices and pump down­
stream from the collecting unit. Larger acti­
vated charcoal cartridges and more fre­
quent replacement must be provided.
Generation o f volatile components from
the reagent will create an even greater prob­
lems, e.g., H C 1 from aqueous iodine
monochloride solution (1).
A word o f caution with respect to liquid-
air cooled freeze-out traps —liquid oxygen
from the air sample may condense and cre­
ate an explosion hazard that may not arise
until the trap is removed from the cold
bath. I n other subambient systems, mois­
ture may be a mixed blessing. Condensa­
tion o f water vapor often assists with the
retention o f contaminants, but i n the end
interferes with the final analysis. Analytical
systems that require anhydrous conditions
cannot be employed for air analysis unless
dehydration can be employed before the
sample enters the collecting system without
loss o f the analyte; e.g., CO by infrared.
Although temperature extremes suffi­
cient to alter the mechanical performance
of collecting systems are not often encoun­
tered, this factor cannot be ignored when
arctic or tropical desert conditions are en­
countered. Batteries for power supplies
must be chosen with these conditions i n
mind. Piston-type pumps may bind or blow
by; check valves may become inoperable;
volumetric factors will require recalibra-
tion; metering valve and plug cock toler­
ances may shift significantly; elastomeric
components lose elasticity beyond tolerable
limits (tubing, "0"-ring seals, check valves,
gaskets, etc.); and diffusion through plastic
containers, especially sample collection and
storage bags may become excessive.
Instruments that require electrical cir­
cuitry or electronic components for gener­
ating analytical data will be affected by
temperature extremes and must either be
provided with thermal insulation or com­
pensated for altered performance. Extrapo­
lation o f response obtained i n the range o f
normal ambient temperature is not a reli­
able substitute for actual testing under use
conditions.
1.9 V O L U M E T R I C E R R O R S . Rate meters
of all types require frequent calibration to
detect and correct drift or shifts i n refer­
ence points, (see Section 1.5). Portable i n ­
struments should be checked every 40 to
50 h o f operation with at least a secondary
standard (wet or dry test meter or glass ro­
tameter) which i n turn is compared fre­
quently with a primary standard o f the pos­
itive displacement total volume type
(spirometer).
In volumetric calibration procedures, the
pressure drop across the instrument being
calibrated and across the system i n which it
will operate must be determined and appro­
priate corrections applied (see Section 1.5
for effects o f fouling). When filter systems
are used, the increasing pressure drop and
decreasing sampling rate that develop as the
filter loads must also receive consideration.
Unnecessary restrictions to free sample
flow also produce undesirable pressure
drop increments that may limit the range of
the sampling system.

Temperature effects do alter volumetric
calibration significantly i f the system is op­
erated under ambient conditions more than
± 2 0 ° C from the original calibration tem­
perature conditions. A high degree o f volu­
metric accuracy requires determination o f
the instrumental temperature coefficient as
well as correction o f the sample volume to
standard temperature and pressure (25 °C
and 101.3 kPa).
Variability i n certain types o f pumping
systems, both between units o f the same
model or within any given unit, may pre­
vent constant rate sampling within tolerable
limits over extended time intervals. Direct
current motors are especially vulnerable to
rate fluctuations unless effective control
circuits are incorporated i n the design.
Manually operated squeeze-bulb devices
seldom deliver reproducible volumes and
should be avoided i f better than semi­
quantitative accuracy is required. Positive
displacement syringe-type hand pumps are
capable o f a high degree o f volumetric ac­
curacy and are to be preferred (41).
1.10 O B S E R V A T I O N A L E R R O R S . The
cal judgment o f color shade and optical
density is one, i f not the most, variable fac­
tor to be considered i n air analysis. Reason­
ably accurate results can be attained i f the
individual using a color comparator is will­
ing to calibrate the instrument for his own
personal use, but he must take into account
the quality and intensity o f the light avail­
able for illumination o f the comparator,
day-to-day variations i n his own optical
acuity, shade changes introduced by inter­
ferences, and stability o f the color stan­
dards. Field kits based on optical color
standards provide a definite, valuable ser­
vice not available by any other means, but
the unavoidable sacrifice o f accuracy and
precision must be carefully weighed when
planning and evaluating investigations
based on such equipment (17,29,42).
Optical acuity and judgment also enters as
a factor in the detection o f air contaminants
with the length-of-stain type o f detector
tubes. Diffusion, migration and trailing may
so obscure the stain front as to render accu­
rate reading very questionable. Again indi­
vidual calibration with knowns can be quite
helpful in providing a basis for judgment.
Errors i n the application o f volumetric
glassware are not unique to air sampling
PHYSICAL PRECAUTIONS 13
and are recognized by those familiar with
analytical laboratory techniques. Such
items as "to deliver" versus "to contain,"
delivery time for pipets, whether or not to
"touch o f f " or "blow out" the drop retained
in the pipet tip, capillary effects i n miniscus
reading, error introduced by a water-
repellant film on the glass, and chipped bu­
ret and pipet tips should be too well recog­
nized to require further elaboration. The
relative accuracy of the original glassware
calibration may be overlooked. Most
laboratory-ware manufacturers adhere to
two standards: precision-grade (within NBS
tolerances) and laboratory grade (twice
precision-grade tolerances). The choice o f
buret, volumetric or Mohr pipet or gradu­
ated cylinder will be determined by the rela­
tive accuracy required in the measurement.
That is, volumetric pipets or a buret would
be used to measure out standardizing solu­
tions whereas a graduated cylinder is suit­
able for making up most reagents.
Reading instruments that indicate by me­
chanical movement, such as a rotameter
opti­
float, voltmeter needles, etc., involves criti­
cal judgment also. Whether or not to read
the top, bottom, or center o f a rotameter
float, parallax effects i n observing meter
needles (some o f the better precision meters
provide a mirror background to eliminate
this problem), capillary effects on the me­
niscus o f liquid-filled manometers, and
scale units, illustrate several o f the more
common observational errors. The increas­
ing relative error inherent i n any instrument
as either the minimum or maximum range
limits are approached may introduce seri­
ous observational error. The upper and
lower 10% o f any instrument's range
should be avoided, and i f possible the ca­
pacity should be so chosen as to permit
operating only i n the center 50% o f the
overall range.
1.11 OPTIMUM SAMPLING RATE. Cali­
bration o f any sample collection device
should include a sufficient range o f rates to
delineate the optimum range consistent
with the efficiency demanded o f the system
and to establish the penalty incurred when
the rate exceeds or falls below these limits.
Failure to include this factor has contrib­
uted serious errors (dust collection by i m -
pinger) which are inexcusable. Collection
14 GENERAL TECHNIQUES
efficiency cannot be assumed; it must be
determined by calibration.
The design o f the collector in liquid re­
agent collection systems for gases and va­
pors is, in most cases, not critical i f in addi­
tion to acceptable collection efficiency the
design provides retention o f entrainment
(baffles), sufficient freeboard to retain
bubbles and foam, and interchangeability
of component parts, and the device is easily
assembled, dismantled and cleaned (29).
Only i n those cases that involve surface re­
actions and adsorption will a "fritted" or
porous glass diffusing element perform
more efficiently than a standard single-
orifice impinger (10,12). Frequently these
d i f f users create problems such as retention
of analyte, excessive frothing and foaming,
variable pressure drop, fragility, and parti­
cle retention that are not encountered with
the use o f impinger-type absorbers.
REFERENCES
1. L I N C H , A . L . , E . G . WIEST, AND M.D. C A R T E R ,
1970. Evaluation of Tetraalkyl Lead Exposure by
Personnel Monitor Surveys, Amer. Ind. Hyg. As­
soc. J . 31:170.
2. A M E R I C A N S O C I E T Y FOR T E S T I N G A N D M A T E R I ­
ALS. 1982. Annual Book of ASTM Standards, Vol.
11.03, A S T M Designation: D1357-82, 1982,
A S T M , Philadelphia.
3. STERN, A.C., 1968. Air Pollution and Its Effects,
Vol. LPart II (2nd Edition) Academic Press, New
York.
4. STERN, A.C., 1976. Air Pollution, Vol. 1 "Air Pollu­
tants, Their Transformation & Transport," Vol. 2 "The
Effects of Air Pollution," Vol. 3 "Measuring, Monitor­
ing and Surveillance of Air Pollution," Vol. 4 "Engi­
neering Control of Air Pollution," Vol. 5 "Air Quality
Management": Supplementary volumes published in
1986, Vol. 6 "Air Pollutants, Transformation, Trans­
port and Effects," Vol. 7 "Measurement, Monitoring,
Surveillance and Engineering Control of Air Pollu­
tion," and Vol. 8 "Management of Air Quality," Aca­
demic Press, New York.
5. C R A L L E Y , L . V . , L . J . C R A L L E Y , A N D G . D . C L A Y ­
25.
TON, 1968. Industrial Hygiene Highlights 1:297. In­
dustrial Hygiene Foundation of America, Inc.,
Pittsburgh, PA.
6. WEISBURD, M.I. (ed.), 1962. Air Pollution Control
Field Operations Manual, U.S. Department of Health, 26.
Education and Welfare, Public Health Service.
7. A M E R I C A N I N D U S T R I A L H Y G I E N E ASSOCIATION, 27.
I960. Air Pollution Manual, Part I-Evaluation,
Amer. Ind. Hyg. Assoc.
8. M A R C A L I , KALLMAN, 1957. Microdetermination
of Toluenedisocyamates in Atmosphere, Anal.
Chem. 29:552.
9. COTABISH, H.N., P.W. M C C O N N A U G H E Y , A N D
H.C. MESSER, 1961. Making Known Concentra­
tions for Instrument Calibration. Amer. Ind. Hyg.
Assoc. J. 22:392.
10. SALTZMAN, B.E., 1961. Preparation and Analysis
of Calibrated Low Concentrations of Sixteen Toxic
Gases. Anal. Chem. 33:1100.
11. LINCH, A . L . , R.C. CHARSHA, 1960. Development
of a Freeze-Out Technique and Constant Sampling
Rate for the Portable Uni-Jet Air Sampler. Amer.
Ind. Hyg. Assoc. J. 21:325.
12. S A L T Z M A N , B . E . , A N D A.F. W A R T B U R G , J R . ,
1965. A Precision Flow Dilution System for Stan­
dard Low Concentrations of Nitrogen Dioxide.
Anal. Chem. 37:1261.
13. LINCH, A . L . , AND M. CORN, 1965. The Standard
Midget Impinger-Design Improvement and Minia­
turization. Amer. Ind. Hyg. Assoc. J . 26:601.
14. INSCOE, M.N., 1966. Losses Due to Adsorption
During Filtration of Aqueous Solutions of Polycy-
clic Aromatic Hydrocarbons. Nature 211:1083.
15. SALTZMAN, B.E., 1954. Colorimetric Microdeter­
mination of Nitrogen Dioxide in the Atmosphere.
Anal. Chem., 26:1949-1955.
16. YAFFE. C D . , D.H. BYERS, AND A.D. HOSEY,
1956. Encyclopedia of Instrumentation for Indus­
trial Hygiene. Ann Arbor, Michigan.
17. LINCH, A . L . , S.S. LORD, K.A. KUBITZ, A N D
M.R. DE BRUNNER, 1965. Phosgene in Air-
Development of Improved Detection Procedures.
Amer. Ind. Hyg. Assoc. J . 26:465.
18. LINCH, A.L., AND M. CORN, 1965. The Standard
Midget Impinger-Design Improvement and Minia­
turization. Amer. Ind. Hyg. Assoc. J 26:601.
19. KATZ, M., 1970. Photochemical Reactions of At­
mospheric Pollutants. Canad. J. of Chem. Eng.
48:3-11.
20. W E S T , F.K., P.W. W E S T , A N D F.A. IDDINGS,
1966. Adsorption of Traces of Silver on Container
Surfaces. Anal. Chem. 38:1566.
21. DYCK, W. 1968. Adsorption of Silver on Borosilic-
ate Glass. Anal. Chem. 40:454.
22. LINCH, A . L . , R.F. STALZER, AND D.T. LAF-
FERTS, 1968. Methyl and Ethyl Mercury
Compounds-Recovery from Air and Analysis.
Amer. Ind. Hyg. Assoc. J . 29:79.
23. KEENAN, R.G., 1963. Determination of Lead in
Air and in Biological Materials, Manual of Analyti­
cal Methods, American Conference in Governmen­
tal Industrial Hygienists.
24. WOESSNER, W.W., AND J . CHOLAK, 1953. Im­
provements in the Rapid Screening Method for
Lead in Urine. A.M.A. Arch. Ind. Hyg. & Occup.
Med. 7:249.
DOWNING, F.B., AND A.L. LINCH, 1946. Process
for Stabilizing of Deactivating Sludges, Precipitates
and Residues Occurring or Used in the Manufacture
of Tetraalkyl Leads. U.S. Patent 2,407,261.
TARES, D.R., 1968. Effect of Silicone Grease on
Diffusion of Fluoride. Anal. Chem. 40:204.
KONIECKI, W.B., AND A.L. LINCH, 1958. Deter­
mination of Aromatic Nitro-Compounds. Anal.
Chem. 30:1134.
28. SKULSKII, I.A., AND V.V. GLASMO, 1966. Absorp­
tion of Caesium Tetraphenylborate from Aqueous
Solutions on Glass, Polyethylene and Polytetraflu-
oroethylene. Nature 211:631.
29. LINCH, A . L . , R.B. DAVIS, R.F. STALZER, AND
W.F. ANZILOTTI, 1964. Studies of Analytical
Methods for Lead-in-Air Determination and Use

with an Improved Self-Powered Portable Sampler.
Amer. Ind. Hyg. Assoc. J. 25:81.
30. PILAT, M.J., 1968. Application of Gas-Aerosol
Adsorption Data to the Solution of Air Quality
Standards. J. Air Poll. Control Assoc., 18:751.
31. ANDREWS, M.L., A N D D.C. P E T E R S O N , 1947. A
Study of the Efficiency of Methods for Obtaining
Vapor Samples in Air. J. Ind. Hyg. Toxic. 29:403.
32. C A M P B E L L , E . E . , AND H.M. I D E , 1966. Air Sam­
pling and Analysis with Microcolumns of Silica
Gel. Amer. Ind. Hyg. Assoc. J. 27:323.
33. SALTZMAN, B.E., 1970. Factors in Air Monitoring
Network Design. Proceedings of the Eleventh Con­
ference on Methods in Air Pollution and Industrial
Hygiene Studies, California Air and Industrial Hy­
giene Laboratory. Berkeley, California. March 30.
34. SALTZMAN, B.E., 1970. Significance of Sampling
Time in Air Monitoring. J . Air Poll. Control As­
soc., 10:660.
35. N I C H O L S , R., AND A. TOPPING, 1966. Absorption
of Ethylene by Self-Indicating Soda-Lime. Nature
211:217.
36. C A L I B R A T E D INSTRUMENTS, I N C . , 729 Sawmill
River Road, Ardsley, NY 10502.
37. M I L L I P O R E F I L T E R C O R P . , 80 Ashby Road, Be­
dford, Mass. 01730 Type Aerosol (MAW G037A0).
38. V E I L L O N , C , AND J.Y. PARK, 1970. Correct Pro­
cedures for Calibration and Use of Rotameter-Type
Gas Flow Measuring Devices. Anal. Chem., 42:
684.
39. E L K I N S , H.B., 1959. The Chemistry of Industrial
Toxicology, (2nd Ed.) pp. 284-288. John Wiley &
Sons, New York
40. NOWEIR, M.H., A N D E . A . P F I T Z E R , 1971. An Im­
proved Method for Determination of Phosgene in
Air. Amer. Ind. Hyg. Assoc. J . 52:163.
41. KITAGAWA, T., 1960. The Rapid Measurement of
Toxic Gases and Vapors. Proceedings of the 13th
International Congress on Occupational Health.
New York.
42. GRIMM, K . E . , AND A . L . LlNCH, 1964. Recent
Isocyanate-In-Air Analysis Studies. Amer. Ind.
Hyg. Assoc. J . 25:285.
43. TURK, A. A N D J.W. JOHNSTON, J R . , 1974. Hu­
man Responses to Environmental Odors. Academic
Press, New York.
44. F O X , D . L . , AND H . L . J E F F E R I E S , 1983. Air Pollu­
tion. Anal. Chem., 55:233R-245R.
45. LONNEMAN, W.A., A N D J . J . B U F F A L I N I , 1981.
Contamination from Fluorocarbon Films. Environ.
Sci. &Technol. 15:99-103.
46. M E Y E R , B., 1983. Indoor Air Quality. Addison
Wesley Publishing Co., New York, p. 56ff.
47. L U D W I G , F.A., 1978. Air Monitoring. Environ.
Sci. & Technol. 72:774-778.
48. BENNETT, B.I., 1979. Stability Evaluation of Am­
bient Concentrations of Sulfur Dioxide, Nitric Ox­
ide, and Nitrogen Dioxide Contained in Com­
pressed Gas Cylinders. EPA/600/4-79/006, U.S.
NTIS PB-292749.
49. K E B B E K U S , E . A N D F. C O R N A V A C C A , 1977. Fac­
tors in the Selection of Calibration Gas Standards.
American Laboratory, 9(7):51-57.
50. HUGHES, E . E . , AND W.D. DROKO, 1977. Long
Term Investigation of the Stability of Gaseous
Standard Reference Materials, NBS Spec. Publ.
464:535-539.
51. C A D O F F , B.C., 1977. Standard Reference Materials
CALIBRATION P R O C E D U R E S 15
for the Analysis of Organic Vapors in the Air. NBS
Spec. Publ. 464:541-543.
52. W R I G H T , R.S., W.C. E A T O N , C . E . D E C K E R , A N D
D . J . V O N L E H M D E N , 1983. The Performance Au­
dit Program for Gaseous Certified Reference Mate­
rials. Proc. Annual Meeting, Air Pollution Control
Assoc. 76th (Vol.2), pp. 83-98.
53. S A M U E L S E N , G.S., AND J.N. H A R M A N , III, 1977.
Chemical Transformation of Nitrogen Oxides
While Sampling Combustion Products. J . Air Poll.
Control Assoc. 27:648-655.
54. C R E C E L I U S , H . J . , AND W. FORWEG, 1975. Staub
Reinhalt Luft, 55:330, in German. APA 576017.
55. SCOTT R E S E A R C H L A B O R A T O R I E S , 1985. "Gas
Chromatography Newsletter," Fall 1985. Scott Re­
search Laboratories, San Bernadino, CA.
56. B A R K E R , W . C , A N D J.F. PONDROT, 1983. The
Measurement of Gas Flow, Part I. J. Air Poll. Con­
trol Assoc. 33:66-72.
Approved with modifications from
2nd edition
Subcommittee 9
J. N. H A R M A N , Chairman
2. Calibration Procedures
2.1 PREPARATION O F STATIC CALIBRA­
TION MIXTURES.
2.1.1 Introduction. Static calibration
mixtures are prepared by introducing a
known weight or volume o f contaminant
into a given volume o f air (1). Generally the
mixture is held i n a container o f fixed d i ­
mensions, but flexible chambers may be
used.
2.1.2 Measuring the Contaminants.
L i q u i d contaminants are commonly intro­
duced into the calibration system with a m i -
crosyringe or micropipet. Gaseous
materials are handled with a gas-tight sy­
ringe.
A pure gas sample may be obtained from
a lecture bottle by either o f the methods
shown i n Figure 2 : 1 . The first technique i n ­
volves attaching a lecture bottle septum d i ­
rectly to a regulated cylinder (2). Just
enough gas to purge the system is bled
through the pressure relief valve. The sam­
pling chamber can then be maintained at
about two atmospheres with no additional
loss o f gas to the atmosphere. Contaminant
gas is then obtained through the septum us­
ing a gas-tight syringe.
In the second method, the pure gas is
passed slowly through a T with a septum
over one arm. The gas then flows from a
dry midget impinger into one containing 10
i n L o f water. The pure gas is then with-
 16 GENERAL TECHNIQUES
Figure 2:1 —Two methods for removing small
amounts of contaminant gas from a lecture
bottle.
drawn through the septum after the gas
lines have been purged.
2.1.3 Rigid Chambers. Generally glass
bottles and flasks make the best containers.
However, plastic and metal chambers may
be used i f wall interactions with the con­
taminant gases are negligible.
The volume o f the container is obtained
by direct measurement o f the chamber
boundaries or volumetrically by filling with
water. The usual size is on the order o f 20 to
40 L , which allows the removal o f enough
useful gas without causing excessive dilu­
tion by the replacement gas.
A typical static calibration system is
shown i n Figure 2:2. The vessel is first
purged with fresh air to remove any resid­
ual contamination, then the contaminant is
"CONTAMINANT INLET
STIRRER
•RUBBER SEPTUM
VALVE
Figure 2:2 —Diagram of a typical state calibra­
tion system.
injected directly into the vessel through the
rubber septum. The gases and vapors are
evaporated and mixed by an externally
driven stirrer or magnetic stirrer or by agi­
tating with internally placed aluminum or
Teflon strips. The gas mixture is then with­
drawn after opening both valves. Place­
ment o f two or three o f these vessels i n se­
ries as shown i n Figure 2:3 greatly reduces
the dilution effect o f the air as the sample
gas mixture is removed (1).
Sometimes it is desirable to sample from
a rigid container without suffering makeup
gas dilution or a significant internal pres­
sure decrease. This can be done by attach­
ing a deflated plastic bag to the dilution-gas
inlet as shown i n Figure 2:4. As the mixture
is sampled, dilution air fills the plastic bag
rather than diluting the mixture. A bag
which does not absorb appreciable quanti­
ties o f the contaminant gas or vapor must
be selected for this purpose.
2.1.4 Piston Type Chambers. Fig­
ure 2:5 illustrates a piston type container
(3). A calibration mixture is prepared by
introducing the contaminant through the
inlet as the piston is raised. The inlet is then
closed and 15 min are allowed for mixing.
DILUENT
GAS rNONREACTIVE, NONABSORBING TUBING
4
' 4
'-^ t>TEST GAS
=
/ f ^ /-fS r f S
Figure 2:3 —Rigid chambers connected in series.
BEFORE REMOVAL
OF SAMPLE
Figure 2:4 —Removing a test gas from a rigid
chamber without causing a sample dilution.
 CALIBRATION P R O C E D U R E S 17

2.1.5 Nonrigid Chambers. Mixtures

may be prepared i n plastic bags, generally
made from Teflon, Mylar and aluminized
Mylar. The bag is alternately filled and de­
flated with air to remove residual contami­
nation and final purge is accomplished by
applying a vacuum to the system. The con­
tainer is connected to a flow meter or wet
or dry gas meter. As the bag fills, the con­
taminant is introduced by one o f the meth­
ods shown i n Figure 2:6.
The test mixture should be used as soon
as possible because the initial concentration
often decays with time. Such decay can be
lessened i f the container is first precondi­
tioned to the test substance however.
Figure 2:6—Four methods of introducing a sam­
2.2 C A L C U L A T I O N S . The concentration
ple into a nonrigid chamber.
by volume o f a contaminant gas or vapor
can be calculated either i n per cent or i n
parts per million from the following equa­ where V and V are the contaminant and
c

tions. system volumes respectively. I f a liquid is

For gas mixtures: added to a closed system, the resulting con­
centration is

C p p m =2 4 . 4 5 x l 0 ^ ( g i ) ( ^ ) ( ^ ) (3)

where p is liquid density ( g / m L ) , V is the L

volume o f liquid (mL), T is the temperature

(K), P is the pressure (kPa), M is the molec­
ular weight (g/mol) and V is the system vol­
ume (L).
2.3 SOURCES O F E R R O R . Lubricant
greases and sealants should be avoided
since they tend to absorb the trace contami­
nants. The use o f rubber and plastic tubing,
other than Teflon, should be avoided i f
possible. The compressed air should be o f
the highest quality available and a purifica­
tion train should be used to remove un­
wanted contaminants.
For additional sources o f error consult
principles stated in the chapter on "Physical
Precautions" which precedes this section.

REFERENCES
1. NELSON, G . O . 1971. Controlled Test
Atmospheres—Principles and Techniques. Ann Ar­
bor Science Publishers. Inc., Ann Arbor, Mich.
2. HAMILTON C O . , P.O. Box 10030, Reno, Nevada
89520
3. HOUSTON A T L A S C O . , P.O. Box 40052, Houston,
Texas 77041.
Approved with Modifications
from 2nd edition
Subcommittee 9
Figure 2:5 —A piston-type container. J.N. H A R M AN, Chairman
18 GENERAL TECHNIQUES
3. Dynamic Calibration of Air
Analysis Systems With
Permeation Tubes
3.1 P E R M E A T I O N T U B E S . The use o f per­
meation tubes as primary standards for
trace gas analysis was first documented by
O'Keeffe and Ortman i n 1966 (1). The prin­
ciple o f this device is based on diffusion o f
gas or vapor through a plastic membrane at
very slow rates. A liquefied gas or volatile
liquid sealed i n a section o f Teflon FEP
tubing placed i n a metered air stream can be
used as a dynamic calibration standard.
The diffusion rate is a nonlinear function
of temperature; therefore, constant temper­
ature conditions must be maintained for the
permeation tube during gravimetric stan­
dardization and use as a calibration source.
Also, the diffusion rate is influenced by the
molecular weight o f the carrier gas em­
ployed for dilution. The difference between
air and nitrogen is negligible; but i f a lighter
gas such as helium or a heavier gas such as
argon is used, corrections or recalibration
would be required.
Certain compounds react with oxygen or
water vapor at the surface o f the plastic
tubing; e.g., sulfur dioxide and hydrogen
sulfide. However, such compounds can be
used i f air is excluded from contact with the
loaded tube by storage and standardization
in a dry nitrogen atmosphere. A relatively
high concentration i n dry nitrogen carrier
gas then is diluted stepwise with air to pro­
vide calibration mixtures.
In general, Teflon fluorocarbon resins
are chemically inert. However, certain qual­
ifications that govern their application to
ambient air analysis problems must be con­
sidered. A clear distinction between chemi­
cal compatibility and purely physical prop­
erties will best serve to delineate the
application limits. Very few chemicals pro­
duce chemical degradation o f fluorocarbon
resins under normal conditions. Reaction
will occur with molten alkali metals, fluo­
rine, strong fluorinating agents, and so­
dium hydroxide above 300°C. I f ignition is
provided finely divided fluorocarbon resins
also will react with 100% oxygen, alumi­
num or magnesium dust. Otherwise, Teflon
fluorocarbons may be considered nonreac-
tive. Examples o f chemicals which Teflon
resists are listed i n Table 3:1 (2). I n most
cases the tests included the boiling range.
Nearly all plastics absorb small quantities
of certain materials on contact. Since no
chemical reaction or solubility occurs i n
Teflon, the absorption detectable by weight
increase is a result o f the material filling
submicroscopic voids between the polymer
molecules (Table 3:11). The amount is pro­
portional to contact time, pressure and
temperature and the effect is reversible. I f
temperature or pressure cycling includes the
boiling range, liquid inclusions within the
submicroscopic pores will vaporize and
thereby create internal pressure which may
enlarge the pores and eventually mechani­
cally damage the plastic. Surface blisters
containing liquid provide visible evidence
of this phenomenon. The absorption o f
aqueous solutions is minimal due i n part to
the low degree o f wettability exhibited by
fluorocarbon resins (3). Teflon is wetted by
most organic solvents.
Gases and vapors permeate through fluo­
rocarbon resins at a considerably lower rate
than through most other plastics. I n gen­
eral, permeation increases with tempera­
ture, pressure and surface contact area, and
decreases with increased film thickness and
polymer density (Figure 3:1, 2, 3). The per­
meability coefficient P is expressed as:
p _ Wt x Th
~ A x Tc x AP
Where:
Wt = Weight o f permeant.
Th = Film thickness.
A = Area o f film.
Tc = Contact time.
AP = Pressure difference.
Film thickness and AP can be combined
into pressure gradient, G = A P / T h = at­
mospheres per m i l (0.001 in.)
Gases and liquids permeate plastics by
molecular vibrations and motion between
the plastic molecules. Therefore, an in­
crease i n plastic density for a well-molded
resin will reduce the diffusion process due
to reduced intermolecular spacing (Fig­
ure 3:1). Since Teflon FEP resins are proc­
essed by melt extrusion, voids larger than
 DYNAMIC CALIBRATION 19

Table 3:1 Typical Chemicals with Which Teflon Resins are Compatible
Abietic acid Cyclohexanone Hydrogen peroxide Phthalic acid
Acetic acid Dibutyl phthalate Lead Pinene
Acetic anhydride Dibutyl sebacate Magnesium chloride Piperidine
Acetone Diethyl carbonate Mercury Polyacrylonitrile
Acetophenone Diethyl ether Methyl ethyl ketone Potassium acetate
Acrylic anhydride Dimethyl formamide Methacrylic acid Potassium hydroxide
Allyl acetate Di-isobutyl adipate Methanol Potassium permanganate
Allyl methacrylate Dimethylformamide Methyl methacrylate Pyridine
Aluminum chloride Dimethyl hydrazine Naphthalene Soap and detergents
Ammonia, liquid unsymmetrical Naphthols Sodium hydroxide
Ammonium chloride Dioxane Nitric acid Sodium hypochlorite
Aniline Ethyl acetate Nitrobenzene Sodium peroxide
Benzonitrile Ethyl alcohol 2-Nitro-butanol Solvents, aliphatic and
Benzoyl chloride Ethyl ether Nitromethane aromatic*
Benzyl alcohol Ethyl hexanoate Nitrogen tetroxide Stannous chloride
Borax Ethylene bromide 2-nitro-2-methyl Sulfur
Boric acid Ethylene glycol propanol Sulfuric acid
Bromine Ferric chloride n-Octadecyl alcohol Tetrabromoethane
n-Butylamine Ferric phosphate Oils animal and Tetrachloroethylene
Butyl acetate Fluoronaphthalene vegetable Trichloroacetic acid
Butyl methacrylate Fluoronitrobenzene Ozone Trichloroethylene
Calcium chloride Formaldehyde Pentachlorobenzamide Tricresyl phosphate
Carbon disulfide Formic acid Perfluoroxylene Triethanolamine
Cetane Furan Phenol Vinyl methacrylate
Chlorine Gasoline Phosphoric acid Water
Chloroform Hexachloroethane Phosphorus Xylene
Chlorosulfonic acid Hexane pentachloride Zinc chloride
Chromic acid Hydrazine
Cyclohexane Hydrochloric acid
Based on experiments conducted up to the boiling points of the liquids listed. Teflon resins have
normal service temperatures up to 500°F, (260°C) for TFE resins, 400°F (205°C) for FEP resins.
*Some halogenated solvents may cause moderate swelling.

intermolecular are not normally present
and permeation is basically molecular dif­
fusion only (2). See Figure 3:2 and Tab­
le 3:111.
Surface attractive forces between per-
meant and plastic barrier influence both
absorption quantity and permeability coef­
ficient P. Gases or vapors chemically re­
lated to the plastic normally show a higher
permeation rate than dissimilar materials.
High solubility o f a material i n a plastic is
indicative o f high permeation rate (molecu­
lar size and vapor pressure are also influen­
tial). Swelling produced by liquid absorp­
tion which increases molecular spacing o f
the plastic, and permeation is not a prob­
lem with fluorocarbon resins (2).
Increased temperature produces higher
molecular activity o f the diffusing material
and hence permeation (Figure 3:2). This i n ­
crease i n not only due to increased solvent
molecule activity but also due to the i n ­
creased vapor pressure. Since the perme­
ation rate is temperature dependent (as
much as 10% per degree C) precise temper­
ature control is critical.
Permeation tubes are available from sev­
eral commercial sources and a wide range
of permeant gas choices is available. Tab­
le 3 :IV illustrates a representative selection
of the currently available permeation tubes.
The vendors o f permeation tubes also offer
custom-built permeation tubes, with the
material designated by the customer, as spe­
cial items. Also available are alternative
permeation-based sources such as those em­
ploying a wafer o f a diffusion-limiting plas­
tic which may be connected to a lecture bot­
tle o f the desired permeant to afford a very
long-life source.
Perhaps the most important parameter i n
the standardization o f permeation tubes is
the time factor required for diffusion equi­
librium to be reached. Two to 5 days should
20 GENERAL TECHNIQUES

Table 3:11 Absorption of Common Solvents in Teflon Resins

Exposure Temp* Exposure Weight
Solvent °C (°F) Time Increase °/o
Acetone 25 (77) 12 mo 0.3
50 (122) 12 mo 0.4
70 (158) 2 wk 0
Benzene 78 (172) 96 hr 0.5
100 (212) 8 hr 0.6
200 (392) 8 hr 1.0
Carbon 25 (77) 12 mo 0.6
tetrachloride 50 (122) 12 mo 1.6
70 (158) 2 wk 1.9
100 (212) 8 hr 2.5
200 (392) 8 hr 3.7
Ethyl 25 (77) 12 mo 0
alcohol (95%) 50 (122) 12 mo 0
70 (158) 2 wk 0
100 (212) 8 hr 0.1
200 (392) 8 hr 0.3
Ethylacetate 25 (77) 12 mo 0.5
50 (122) 12 mo 0.70
70 (158) 2 wk 0.7
Toluene 25 (77) 12 mo 0.3
50 (122) 12 mo 0.6
70 (158) 2 wk 0.6
•Exposure at over the boiling point o f the solvent are at its vapor pressure.
Note: Values are averages only and not for specification purposes.

be allowed in a constant temperature oven

(30°C ± 0.02°C) under a flowing dry air
stream for accurate standardization. The
time is dependent upon diffusion rate and
accuracy of weighing. I n use, constant tem­
perature should be maintained to ± 0.1 °C
for precision within ± 1% ( 3 ) . Gas chro­
matography ovens are ideal for this pur­
pose and internal standards for calibration
of field gas chromatography apparatus can
be conveniently installed inside the calibra­
tion assembly (4).
Care must be taken to assure tight seals
at the tube ends; this is most important with
low-boiling compounds. It is doubtful i f
tubes can be fabricated using materials that
exert pressures of greater than 50 psig at
normal temperatures. The method used is
that described by O'Keeffe and Ortman (1).
Others are:
3.2 CALCULATION OF LENGTH R E ­
Q U I R E D . To a very good approximation,
the output of a permeation tube is propor­
tional to its active length (the length o f tube
in contact with the contained fluid). I f the
output of any given tube is known, then a
rate per unit length can be calculated.
Table 3:V shows rates for a particular size
and material. To achieve any desired out­
put, the length is then
1) Pr
R
where L = length (cm) of permeation
tube required.
Pr = output required (ng/min).
R = permeation rate in ng/min
cm.
Thus, i f an output of 4 5 0 0 ng/min o f S 0 is
2

needed, using the tubes considered in

a. Swagelok-type plugs at each end (may Table 3 : V , then the required length is 4 5 0 0 /
be cumbersome). 290 = 15.5 cm.
b. Tight fitting caps sealed with Apiezon 3 . 3 C O N V E R S I O N O F N G / M I N TO PPMV A T
W—for low pressure applications. 3 0 ° C . Because many instruments are cali-
 DYNAMIC CALIBRATION 21

0 2 4 6 8 10 12
2.14 2.15 2.16 2.17 2.18 2.19 THICKNESS mils
TFl DENSITY grams/ml

Notes: V a l u e s are a v e r a g e s only a n d not for specification Notes: V a l u e s are a v e r a g e s only and not for specification
purposes. To convert the permeation v a l u e s for 100 s q . in. to purposes. To convert the permeation v a l u e s for 100 s q . in. to
those for 1 s q . c m . multiply by 0 . 0 0 1 5 5 . O n e mil = 25.4 those for 1 s q . c m . multiply by 0.00155. O n e mil = 25.4 n.

Figure 3:1-Effect of density of "Teflon" TFE Figure 3:3 —Water vapor transmission rate of
resins on their permeability to gases at 30°C "Teflon" FEP resins at 40°C (104°F).
(86°F).

brated i n ppmv and outputs are given in

n g / m i n , a conversion formula is required.
To convert from n g / m i n to ppm the re­
quired formula is:

2) Pr • K
C =

where C concentration i n ppm (vol),

Pr output in n g / m i n .
F flow rate i n c m V m i n .
K ratio o f molar volume ( L /
mol) to molecular weight
(Table 3:VI).

Example: A permeation tube containing

propane and having an output o f 950 n g /
min is to be used at 30°C and a flow rate o f
500 c m V m i n . What is the concentration o f
propane (in ppm) i n the gas mixture?

0 20 40 60 80
TEMPERATURE °C (Reciprocal Absolute S a l e )
Notes: V a l u e s a r e a v e r a g e s only a n d not for specification
purposes. To convert the permeation v a l u e s for 100 s q . in. to
those for 1 s q . c m . multiply by 0 . 0 0 1 5 5 . O n e mil = 2 5 . 4 n.
Figure 3:2-Permeability of "Teflon" FEP resins
to gases, at various temperatures.
950 x 0.565
= 1.07
500
3.4 Permeation tubes that contain lique­
fied gases or volatile liquids are calibrated
gravimetrically and used to prepare stan­
dard concentrations of pollutants i n air as
follows (1,3,4,5). Analyses o f these known
concentrations give calibration curves
22 GENERAL TECHNIQUES

Table 3:111 Permeability of Teflon Fluorocarbon Resins to Vapors (g/100 in? 24 h)*
TFE FEP
23°C. 30°C. 23°C. 35°C. 50°C.
(73 °F.) (86°F.) (73 °F.) (95 °F.) (122°F.)
Acetic acid — — — 0.42 —
Acetone — — 0.13 0.95 3.29
Acetophenone 0.56 0.47
Benzene 0.36
-
0.80 0.15
-
0.64 —
-
n-Butyl ether — — 0.08 — 0.65
Carbon
tetrachloride 0.06 — 0.11 0.31 —
Decane — — 0.72 — 1.03
Dipentene 0.17 — 0.35
Ethylacetate
- - 0.06 0.77 2.9
Ethyl alcohol
-
0.13
-— 0.11 0.69 —
Hexane — — — 0.57 —
HC1. 20% <0.01 — <0.01 — —
Methanol — — — — 5.61
Piperidine 0.07 0.04 — —
"Skydrol"
-
hydraulic fluid 0.06 — 0.05 — —
NaOH. 50% 5 x 10 -5
— 4 X 10 — -5
—
H S 0 . 98%
2 4 1.8 x 10" — 5
8 x 10 — —
-6

Toluene — — 0.37 — 2.93

Water 0.35 0.09 0.45 0.89
-
•Test Method: ASTM E-96-53T (at vapor pressure; for 0.001" film thickness; one mil = 0.001" =
25.4 fim).
Notes: Values are averages only and not for specification purposes. To convert the permeation
values for 100 sq cm, multiply by 0.00155.

which simulate all o f the operational condi­
tions performed during the sampling and
chemical procedure. The calibration curves
include the important corrections for col­
lection efficiency at various concentrations
of pollutants.
Prepare or obtain (3,8) a Teflon FEP
permeation tube that emits gas or vapor at
a rate consistent with the desired concentra­
tion level. A permeation tube with an effec­
tive length o f 1 to 2 cm and a wall thickness
of 0.76 m m . (0.030") will usually yield the
desired permeation rate i f held at a constant
temperature o f 20°C for S 0 . 2 pass a stream o f nitrogen or air over the
Permeation tubes are calibrated under a
constant-temperature water bath main­
tained at the desired temperature; cylinder
containing pure, dry air with appropriate
pressure regulators; needle valves and flow
meters for the nitrogen, i f necessary; and
dry air, diluent gas streams. The diluent
gases are brought to temperature by pas­
sage through a 2-m-long copper coil i m ­
mersed i n the water bath. Insert a cali­
brated permeation tube (7) into the central
tube o f the condenser (maintained at the
desired temperature by circulating water
from the constant-temperature bath) and
tube at a fixed rate o f approximately 50 c m 3

stream o f dry nitrogen when reactions oc­
cur with atmospheric oxygen or moisture
on or within the walls o f the permeation
tube ( H S , S 0 , etc.).
2 2
3.5 To prepare standard concentrations
of pollutant, select either a system designed
for laboratory use or one for field use. (Fig­
ure 3:4 and 3:5, respectively). Assemble the
apparatus as shown i n one o f these systems,
consisting o f a water-cooled condenser;
min. Dilute this gas stream to the desired
concentration by varying the flow rate o f
the clean, dry air. Clean, dry air may also
be prepared by passing ambient air from a
relatively uncontaminated outside source
first through absorption tubes packed with
activated carbon and soda-lime, then
through an efficient fiber glass filter in se­
ries. This flow rate can normally be varied
from 0.2 to 15 L / m i n u t e . The flow rate o f
 DYNAMIC C A L I B R A T I O N 23

Table 3:1 V Commercially Available Permeation Tubes

Chemical Permeant Chemical Permeant
Material Gas Formula Material Gas Formula

Acetaldehyde CH CHO 3 Heptane (n) 7 16

C H

Acetic acid CH3COOH Hexane (n) 6 14

C H

Acetone CH3COCH3 Hydrazine NH NH -H 0 2 2 2

Acetonitrile CH3CN Hydrogen chloride HC1

Acrylonitrile CH CHCN 2 Hydrogen cyanide HCN
Ammonia NH 3 Hydrogen fluoride HF
Benzaldehyde C H CHO 6 5
Hydrogen sulfide HS 2

Benzene C H 6 6
Iodine h
Bromine Br 2
Isobutylene (CH ) CCH 3 2 2

Butane C H 4 1 0 Isopropyl alcohol (CH ) CHOH 3 2

Butanol C H OH 4 9 Isopropylbenzene C
9 12 H

Butene (1) C H 4 8 Isopropyl mercaptan (CH ) CHSH 3 2

Butyl acetate CH COOC H 3 4 9 Mercury Hg

Butyl mercaptan (n) C Ho,SH
4 Methanol CH3OH
Butyl mercaptan (Tert) C H SH 4 9
Methyl acetate CH C0 CH 3 2 3

Carbon disulfide cs 2
Methyl amine CH NH 3 2

Carbon tetrachloride CC1 4 Methyl bromide CH Br 3

Carbonyl sulfide COS Methyl ethyl ketone CH COC H 3 2 5

Chlorine Cl 2 Methyl ethyl sulfide CH3SC H^ 2

Chlorobenzene C H C1
6 5 Methyl iodide CH I 3

Chloroethane C H C1
2 5 Methyl isocyanate CH3NCO
Chloroform CHCI3 Methyl isopropyl sulfide CH SCH(CH ) 3 3 2

Chloromethane CH3CI Methyl mercaptan CH SH 3

Chlorotoluene (o) CH C H C1 3 6 4 Napthalene C

10 8 H

Cyclohexane C
6 12 H
Nitric acid HNO3
Cyclopropane CH CH CH 2 2 2
Nitrobenzene C H N0 6 5 2

Decane C
10 22 H Nitrogen dioxide N0 2

Dichlorobenzene (p) C H C1
6 4 2 Nonane (n) C
9 20 H

Dichloroethane (1,1) C1 CHCH 2 3 Octane (n) C

8 18H

Dichloroethane (1,2) C1CH CH C1 2 2 Octanethiol (n) C H SH

8 1 7

Dichloromethane CH C1 2 2 Pentane (n) C

5 12 H

Diethyl disulfide (C H ) S 2 5 2 2 Phenol C H OH

6 5

Diethyl sulfide (C H ) S 2 5 2 Phosgene COCl 2

Dimethyl amine (CH ) NH 3 2 Propane C H 3 8

Dimethyl disulfide (CH ) S 3 2 2 Propyl mercaptan (n) C H SH

3 7

Dimethyl ether CH OCH 3 3 Propylene C H 3 6

Dimethyl mercury Hg(CH ) 3 2 Propylene oxide C H 0 3 6

Dimethyl sulfide (CH ) S 3 2 Pyridine C H N

5 5

Dimethyl sulfoxide (CH ) SO 3 2 Pyrrole C H N

4 5

Dinitrotoluene (2,4) C H (N0 )

7 6 2 2 Sulfur dioxide so 2

Dipropyl sulfide (C H ) S 3 7 2 Sulfur hexafluoride SF 6

Epichlorohydrin C H OCl
3 5 Tetrachloroethylene C C12 4

Ethanol C H OH
2 5 Thionyl chloride SOCl 2

Ethyl acetate CH C0 C H 3 2 2 5 Thiophane C H S

4 8

Ethyl amine C H NH
2 5 2 Toluene C H CH
6 5 3

Ethyl benzene C
6 5 2 5 H C H Toluene diisocyanate H CC H (NCO)
3 6 3 2

Ethyl mercaptan C H SH
2 5
Trichloroethane (1,1,1) CH CC1 3 3

Ethylene dibromide C H Br
2 4 2
Trichloroethylene CHC1CC1 2

Ethylene oxide (CH ) 0 2 2 Trimethylamine (CH ) N 3 3

Formaldehyde CH 0 2 Trioxane (CH 0) 2 3

Formic acid HCOOH Vinyl acetate CH COOC H 3 2 3

Freon-11 CCI3F Water H 0 2

Freon-113 C C1 F
2 3 3 Xylene (m) C H (CH )
6 4 3 2

Freon-12 CC1 F 2 2 Xylene ( 0 ) QH (CH3) 4 2

Freon-21 CHC1 F 2 Xylene (p) C H (CH )

6 4 3 2

Freon-22 CHC1F 2
24 GENERAL TECHNIQUES

Table 3:V Typical Permeation Rates sampling time. This fixing of sampling time
Permeation Rate is desirable also from a practical stand­
Chemical (ng/min cm) at 3 0 ° C * point. I n this case, select a sampling time o f
290
30 min. Then to obtain a 30-L sample o f air
so
requires a flow rate o f 1 L / m i n . A 22-
2

N0 2 1200
H S
2 250 gauge hypodermic needle operating as a
Propane 80 critical orifice will control air flow at this
n-Butane 2 approximate desired rate (10). The concen­
Chlorine 1500 tration in air is computed as follows:
Ammonia 170
Methyl Mercaptan 30 Pr x K
C =
• M i n i m u m Tube Length is 2.5 cm. ri + r,
Where C = Concentration in ppm.
the sampling system determines the lower Pr = Output of tube in /xg/min.
limit for the flow rate of the diluent gases. K = Constant from Table 3 : V I .
The flow rates of the nitrogen and the di­ v { = Flow rate of diluent air,
luent air must be measured to an accuracy liter/min.
of 1 to 2 % . W i t h a tube emitting S 0 at a 2
r 2 = Flow rate of diluent nitro­
rate of 0.26 /-ig/min, the range o f concen­ gen, liter/min.
tration will be between 0.007 to 0.5 ppm (17
Data for a typical calibration curve are
to 1300 fig m ) , a generally satisfactory
3

listed in Table 3 : V I I .
range for ambient air conditions. When
A plot of the concentration of sulfur di­
higher concentrations are desired, calibrate
oxide in ppm (x-axis) against absorbance of
and use longer permeation tubes.
the final solution (y-axis) i U yield aw

3.6 P R O C E D U R E F O R P R E P A R I N G S I M U ­
straight line, the slope o f which is the factor
L A T E D C A L I B R A T I O N C U R V E S . Obviously
for conversion o f absorbance to ppm. This
one can prepare a multitude of curves by
factor includes the correction for collection
selecting different combinations of sam­
efficiency. A n y deviation from linearity at
pling rate and sampling time. The following
the lower concentration range indicates a
description for S 0 represents a typical pro­
change in collection efficiency of the sam­
2

cedure for ambient air sampling of short

pling system. Actually, the standard con­
duration with a brief mention of a modifi­
centration o f 0.01 ppm and below o f sulfur
cation for 24-h sampling. The system is de­
dioxide is slightly below the dynamic range
signed to provide an accurate measure in
of the method. I f this is the range o f inter­
the 0.01 to 0.5 ppm range. It can be easily
est, the total volume of air collected should
modified to meet special needs.
be increased to obtain sufficient color
The dynamic range o f the colorimetric
within the dynamic range of the colorimet­
procedure fixes the total volume of the
ric procedure. Also, once the calibration
sample at 30 L ; then, to obtain linearity be­
factor has been established under simulated
tween the absorbance o f the solution and
conditions the conditions can be modified
the concentration in ppm, select a constant
so that the concentration of S 0 is a simple
2

multiple of the absorbance of the colored

solution.
Table 3:VI K Values. For long-term sampling of 24-h duration
*K Values at 3 0 ° C the conditions can be fixed to collect 300 L
of sample in a larger volume o f tetrachloro-
so 0.389
2
mercurate. For example, for 24 h at 0.2 L /
N0 0.541
min, approximately 288 L of air are col­
2

H S
2
0.732
Propane 0.565 lected. A n aliquot representing 0.1 of the
N-Butane 0.429 entire amount o f sample is taken for the
Chlorine 0.351 analysis.
Ammonia 1.463 3.7 E Q U I L I B R I U M . Each tube's perme­
Methyl Mercaptan 0.518 ation rate changes with its temperature.
•Diverse density, L / g . Hence, time must always be allowed for the
 DYNAMIC CALIBRATION 25

PERMEATION TUBE SCHEMATIC FOR LABORATORY USE

FLOW METER
OR NEEDLE VALVE
DRY T E S T
. METER CLEAN DRY AIR

FLOW METER
OR
CRITICAL ORIFICE

Figure 3:4 —Gas dilution system for preparation of standard concentrations of sulfur dioxide for
laboratory use by the permeation tube method.

tube to reach an equilibrium condition. by a permeation rate vs. air temperature

This is particularly true for tubes stored in a calculation (9).
cool place and then inserted i n a warm 3.8 V E N T I L A T I O N . Each tube permeates
chamber for gas generation control. In such continuously. Thus i f a tube is enclosed in a
cases the tube will accurately provide a gas chamber without ventilation, concentration
source based on its actual temperature, buildup will occur. This is particularly true
rather than the temperature o f the warm air for tubes stored i n a warm, closed chamber
flowing through the chamber. Output con­ environment. I n such cases the actual con­
centration will be lower than that predicted centration coming out o f the chamber upon
the first flush o f ventilation air will be
Table 3:VII Typical Calibration Data higher than that predicted by a permeation
rate vs air temperature calculation. Often,
Amount of S 0 2
considerable time must be allowed to "de­
Concentrations in fiL for Absorbance
gas" a saturated system (9). A n attractive
of S 0 ppm 30 L of sample
2
alternative is to maintain a slow purge on
0.005 0.15 0.01 the tube during periods o f storage so that
0.001 0.30 0.02
evolution o f a high concentration plug o f
0.05 1.50 0.117
3.00 0.234
sample is avoided when the permeation de­
0.10
0.20 6.00 0.468 vice is first connected.
0.30 9.00 0.703 3.9 CONDITIONING. When working with
0.40 12.00 0.937 low concentrations o f gas (i.e., concentra-
26 GENERAL TECHNIQUES
PERMEATION TUBE
Figure 3:5 —Gas dilution system for preparation of standard concentrations of sulfur dioxide for field
use by the permeation tube method.
tions near nominal ambient levels) it is i m ­
portant to condition the system by allowing
the gas to pass through the system for a
period o f day(s) before conducting any ex­
periments. Conditioning is necessary even
i f nominally inactive materials such as glass
or stainless steel are used. Active materials
such as rubber should, o f course, be
avoided (9).
3.10 T E S T I N G VS C A L I B R A T I O N .
permeation tubes can be used for testing
atmospheric analyzers without temperature
control. However, the existing temperature
needs to be measured and taken into ac­
count i n calculating the approximate per­
meation rate. Testing is considered to be an
operation to establish whether an analyzer
is operating "about right." I t should be per­
formed frequently. Testing, on the other
hand, is not a substitute for less frequent,
periodic calibrations i n which known con­
centrations o f gas are delivered to the ana­
lyzer with precision (9).
Acknowledgment. M r . Arthur Johnston as­
sisted with this section as Chairman o f the
PERMEATION TUBE SCHEMATIC FOR FIELD USE
NEEDLE VALVE
ffi 4 CLEAN AIR
DRIER
^..CYLINDER
^ AIR
OR
NITROGEN
WATER BATH
IMPINGER
Analytical Chemistry Committee Standard­
ization Subcommittee.
REFERENCES
1. O'KEEFFE, A . E . , AND G.C. ORTMAN, 1966. Pri­
mary Standards for Trace Gas Analysis. Anal.
Chem. 38:760.
2. DIGEL, W.A., (ed.), 1970. The Journal of
"Teflon," Reprint No. 41. E . I . Dupont de Nemours
Company.
Dynacal 3. ANALYTICAL INSTRUMENT DEVELOPMENT,
INC., 250 S. Franklin Street, West Chester, Pa.,
19380.
4. PATTY, F.A., 1963. Industrial Hygiene and Toxi­
cology. 2nd Edition, Vol. II. Interscience Pub­
lishers. New York.
5. SCARINGELLI, F.P., A.A. FREY, AND B . E . SALTZ­
MAN, 1967. Evaluation of Teflon Permeation
Tubes for Use with Sulfur Dioxide. Amer. Ind.
Hyg. Assoc. J . 28:260.
6. THOMAS, M.D., AND R.E. AMTOWER, 1966. Gas
Dilution Apparatus for Preparing Reproducible
Dynamic Gas Mixtures in any Desired Concentra­
tion and Complexity. J . Air Pollut. Contr. Assoc.
16:618.
7. SCARINGELLI, F.P., A . E . O'KEEFFE, E . ROSEN­
BERG, AND J.P. BELL, 1970. Preparation of
Known Concentrations of Gases and Vapors with
Permeation Devices Calibrated Gravimetrically.
Anal. Chem. 42:871.

8. VICI METRONICS, 2991 Corvin Drive, Santa
Clara, Calif. 95061.
9. M E T R O N I C S A S S O C I A T E S , 3201 Porter Drive,
Stanford Industrial Park, Palo Alto, California.
10. L O D G E , J.P., J R . , J . B . P A T E , B.E. AMMONS, A N D
G.A. SWANSON, 1966. The Use of Hypodermic
Needles as Critical Orifices in Air Sampling. J . Air
Pollut. Control Assoc. 16:197.
Approved with modifications
from the 2nd edition
Subcommittee 9
J . N. H A R M A N , Chairman
4. Sampling and Storage of
Particles
4.1 I N T R O D U C T I O N . The collection o f a
particle sample that is not too different
from the population o f interest is often dif­
ficult and requires attention to details. This
section deals with general information con­
cerning the collection of such samples. Spe­
cific methods o f analysis are covered i n
other sections.
Particles are sampled for many reasons
such as:
a. to determine whether there are hazard­
ous concentrations o f pollutants i n the at­
mosphere or i f ambient air standards have
been exceeded;
b. to determine the effectiveness o f con­
trol programs in reducing ambient concen­
trations o f pollutants;
c. to determine the emission levels from a
source;
d. to determine the effectiveness o f con­
trol equipment;
e. to determine the sources contributing
to pollution at a receptor; or
f. to identify pollutants i n the atmos­
phere.
Aerosols are defined here to mean dis­
persions of any material in the solid or liq­
uid phase i n a gas stream or the atmos­
phere. Dusts, smoke, soot, mist, fumes,
and fog are terms used on occasion to de­
scribe certain types o f aerosols. Some o f the
more useful terms may be defined as fol­
lows:
a. Settleable Particles—larger particles
that settle out of the air fairly fast, such as
those caught i n an open jar. They are usu­
ally larger than 30 fim diameter, but in still
AIRBORNE PARTICLES 27
air some particles 10 fim or smaller may
settle.
b. Suspended Particles—particles that
tend to remain suspended i n the atmos­
phere for long periods o f time. These are
generally smaller than 30 fim i n diameter.
Commonly, the material collected by a
High Volume Sampler (Hi-Vol) is limited
by the shelter design and filter to particles
less than 40 fim.
c. Condensation Nuclei— sometimes
called Aitken nuclei, are small particles that
act as condensation sites for super­
saturated vapors i n the atmosphere. They
are predominantly 0.01 to 0.1 fim diam­
eter.
d. Agglomerates—particles that are com­
posed o f several smaller particles that are
attracted to a large particle or to each other
and travel together i n the atmosphere as a
single particle.
e. Modes—the distribution o f particle
sizes i n the atmosphere is usually bimodal
or trimodal. The size range defined above
as condensation nuclei are now termed the
nuclei mode. Particles 0.1-2.5 fim are called
the accumulation mode, being most persist­
ent i n the atmosphere. These two classes are
collectively called fine particles. Particles
larger than 2.5 fim are called coarse parti­
cles.
It should be kept i n mind that there are
no sharp cutoffs between classifications
and the particle sizes i n each class overlap
the neighboring classes. This may be due to
several factors that tend to make the proper
collection and classification of particles dif­
ficult and somewhat inexact. These include
the shape o f the particles (which affects
their aerodynamic properties), their density
and velocity (which affects their inertia),
and electrical charge. These will be dis­
cussed in more detail later in this section.
There is a great variety of methods avail­
able for the collection o f particles, and the
method selected will often depend on the
purpose for which the sample is being
taken. For instance, i f information on indi­
vidual particles is desired, a few hundred
particles may be collected for microscopic
examination. This can determine particle
shape and size distribution but does not tell
all about aerodynamic properties, weight,
or chemical composition. A larger amount
28 GENERAL TECHNIQUES
may be collected i n an inertial collector in
which the particle size distribution can be
determined. This does not provide separate
particles that can be examined nor does it
usually provide enough samples for chemi­
cal analysis. For chemical analysis, a H i -
Vol sampler is usually employed. The aver­
age composition o f all o f the particles can
be determined, but this does not provide
information on the composition o f individ­
ual particles or their sizes or shapes, or on
the compounds present. Sometimes optical
properties such as visibility reductions are
of interest and none o f the above methods
is entirely suitable.
4.2 P R E C A U T I O N S I N C O L L E C T I N G A E R O ­
SOL S A M P L E S .
4.2.1 The Problem. The most bother­
some aspect of aerosol sampling is a result
of the momentum o f the particles, which is
a product o f its mass times its velocity
(MV). Thus, this problem becomes more
severe with larger particles or fast moving
streams. Since particles are much larger
than gas molecules, each time the flow di­
rection o f the gas stream changes, as at a
bend i n a pipe or i n flow around an object,
the larger particles tend to continue on their
original line and are displaced somewhat
from the original part o f the gas stream
they were with, as illustrated i n Figure 4 : 1 .
This often leads to deposition on a nearby
surface or unevenly dispersed streams
where more particles will be found in one
side of a pipe than another. Different sized
particles will be displaced by different
amounts. This makes it necessary for all
parts o f the stream to be sampled and prop­
erly weighted to be representative of the
whole stream.
4.2.2 Isokinetic sampling refers to
taking a sample under such conditions that
there is no change i n momentum, so that
the sample will be representative o f the
gases and aerosols i n that portion o f the
stream being sampled. This is accomplished
by using a thin-walled tube aligned with the
stream flow and drawing sample into it at
the same linear velocity as the stream flow
at that point. Even when these precautions
are taken, there can be several reasons why
the sample is not truly representative:
a. The probe always has a finite wall
thickness that disturbs the flow;
b. the point sampled may not be repre­
sentative o f the whole stream;
c. i f the flow is turbulent at the point o f
sampling there is no way to get a truly iso­
kinetic sample; the sample will contain
smaller portions o f the larger particles; or
d. sample may be lost by deposition or
changed by agglomeration or deagglomera-
tion i n the sample line after it enters the
probe and before it enters the main sam­
pling device.
Fortunately, it is not always necessary to
use isokinetic sampling to obtain a reasona­
bly good sample. Figure 4:2 shows the col­
lection efficiency for particles o f various
sizes as a function o f stream velocity. Fig­
ure 4:3 shows the effect o f not aligning the
sample tube in the same direction as the
stream flow.
These data show that:
a. Sample velocity and alignment are
both important conditions o f isokinetic
sampling;
b. under isokinetic conditions all particle
sizes are collected efficiently;
c. small particles (under about 3
diameter) do not require isokinetic sam­
pling for efficient collection—their small
mass minimizes inertial effects;
d. stagnant gases, such as calm ambient
air, do not require isokinetic conditions for
efficient sampling o f all particle sizes be­
cause slow moving particles have little or no
momentum.
4.2.3 Gravitational Effect. I n the case
of quiescent air masses, as discussed imme­
diately above, there will be a small effect
because o f gravitational settling of the par­
ticles. I f the probe points up, the concentra­
tion of particles collected will be increased
by a factor o f (1 + V / U ) where V is the
s s
settling velocity o f a particle and U is the
sampling velocity. I f the probe aims down
(such as the shed opening o f a Hi-Vol), the
sample will contain less particles by the fac­
tor ( l - V / U ) . Here again, small particles
are not much o f a problem. The correction
factor would predict a reduction o f about
1 °7o for 11 fim particles entering a H i - V o l
(assuming 645 c m of opening and 1.42 m V
2
min) but for 36 fim particles there would be
10% less, for 85/i particles about 50% less,
and no particles larger than 135fim would
enter. O f course, turbulent winds could
 AIRBORNE PARTICLES 29

S T R E A M LINES S T R E A M LINES

Concentration
of ajj[sizes
the same here
as here

O C) () ()

(a) I L L U S T R A T I O N OF I S O K I N E T I C (b) S A M P L E - T U B E N O T I N L I N E
SAMPLING NON-ISOKINETIC
T o o f e w large particles are c o l l e c t e d

S T R E A M LINES S T R E A M LINES

O CI o
o c o
S t r e a m lines
b e n t at
sample p o i n t

oo
oo

° o 0 ,
° b <
V o; o; °
o
o o
o o
o o

(c) SAMPLE V E L O C I T Y T O O LOW (d) S A M P L E V E L O C I T Y T O O H I G H

NON-ISOKINETIC NON-ISOKINETIC
T o o m a n y large particles c o l l e c t e d T o o f e w large particles c o l l e c t e d

Figure 4:1 —Schematic illustration of isokinetic and non-isokinetic sampling.

help a few larger particles to reach the f i l ­ lectors such as cascade impactors, Ander­
ter. (Obviously in the real system other fac­ son samplers, Rotorods, and jars with
tors enter, since experimentally the Hi-Vol sticky paper all are dependent on the Stokes
sampler collects 50% of particles of diam­ number. For impactors,
eters 30-45 fim depending on wind speed.)
4.2.4 Sampling Rate. The linear sam­
pling rate can have an effect on several
other types of aerosol collectors. For in­ R 9fiR
stance, the efficiency o f impingement col­
30 GENERAL TECHNIQUES

where I is the stopping distance, R is the

{

radius o f the impingement orifice, r the

radius o f the particle, V the velocity o f
0

the stream as it leaves the orifice, p the

density o f the particle, and the viscos­
ity o f the medium.

The mass o f the particle, its linear velocity

and the size o f the jet or object are impor­
tant factors.
Other types o f collectors such as electro­
static precipitators and filters also depend
on velocity, but to a much smaller extent.
4.2.5 Miscellaneous Factors. Previous
mention was made o f particles being lost
from a sample by impingement on surfaces
or bends i n the sampling line, which can
cause errors even after a good sample has
been taken at the probe. I n addition to this,
STREAM LINES
Large particles can settle out i n tubes or other
components o f the sampler or can be lost
by electrostatic attraction to various sur­
faces i n the sampler—especially noncon­
ducting surfaces such as glass or plastics.
Figure 4:l(e)-Illustration of particle segregation Particles can also be lost due to adhesion on
at a pipe bend. wet, oily, or sticky surfaces. As shown i n
Figure 4:4 and Table 4:1 this can vary with
relative humidity and surface roughness.

Ratio of wind speed to inlet air speed (\J <>/ U )

Figure 4:2—Diagram showing change in ratio of observed concentration to true concentration with
departure from isokinetic conditions.
(Watson, H . 1954, Amer. Ind. Hyg. Assoc. Quart. 15:21.)
 AIRBORNE PARTICLES 31

ISOKINETIC FLOW

x 4 M Spores
& 32>x Spores

30° 60° 90° 180°

Angle between inlet tube and stream direction
Figure 4:3 — Variation of the ratio of observed concentration to true concentration with angle of yaw
of inlet tube for 4,12 and 37 micrometer mass median diameter diethyl phthalate droplets and for 4
and 32 micrometer spores.
(Watson. H . 1954. Amer. Ind. Hyg. Assoc. Quart. 15:21)

Several factors can result i n changes i n
size o f particles, which may change the col­
lection efficiency as discussed above. Dur­
ing periods o f high relative humidity some
particles such as sulfuric acid can absorb
water vapor and get larger as illustrated i n
Figure 4:5. Most particles will adsorb water
or other vapors; therefore, collected sam­
ples are generally equilibrated to a standard
relative humidity, such as 40%, before
weighing. Water adsorption is generally
minimal up to about 50% R . H . but can
cause considerable variation and error
above this value.
There is generally a tendency for small
particles to agglomerate to larger particles.
This can occur i n the sampling system but is
not fast enough to be o f concern for parti­
cles larger than about 0.2^m. Even for this
size, particles only agglomerate at a rate o f
about 1%/h. I n impingement collection
methods, particles that are normally ag­
glomerated i n the atmosphere may be frac­
tured to smaller particles when they strike
the collector. Some o f these smaller parti­
cles may then be redispersed and either col­
lected as smaller particles or lost from the
collector.
The following precautions should be
mentioned as possibilities but seldom create
a serious problem. Collected particles may
react after collection so that the material
analyzed may not represent the composi­
tion o f the original aerosol. For instance,
sulfuric acid particles may react with ce­
ment kiln dust (CaO) to form calcium sul­
fate. Particles may also react with the col­
lection media such as Millipore filters, or i n
wet collectors they may react with the sol­
vent or partially dissolve i n i t . Gases may
adsorb on solid or liquid aerosols. Col­
lected aerosols may also sublime or evapo­
rate after collection. I t is not uncommon
for the concentration o f pollutants to vary
throughout the day or for the temperature
to vary so that a pollutant collected at night
32 GENERAL TECHNIQUES

Saturation (G/G)
O 0.001 0.002 0.003 0.004 0.005
100 I l i ' 1

90
Water vapor /
adsorption curve /
80 for quartz /

70 7 4 1 m m Hg < P < 7 5 6 m m Hg / o( _
24°C < T < 27° C J /
• - 2 1 fim particle / uL
60
O - 5 1 /im particle / j ,r
A - 9 0 /*m particle 1/
50
40 - -
30 -
20 -
10 -
t i l l I I 1 |
0 10 20 30 40 50 60 70 80 90 100
Full adhesion (%)
Figure 4:4—Adhesion variation with relative humidity (quartz particles to pyrex flat). (Corn, M.
1961. J . Air Poll. Control Assoc. 11:566.)

or in the morning may be lost from the sam­ pressures and evaporative loss is small. For
pler after a warm afternoon. Although it is many purposes, particulate samples tend to
common for aerosol sampling to extend for keep well for long periods of time.
24 h or more, most solids have low vapor 4.3 F A C T O R S A F F E C T I N G S E L E C T I O N O F
SITES FOR A E R O S O L SAMPING OF T H E ATMO­
S P H E R E . Site selection is important i n any
Table 4:1 Effect of Surface Irregularities on
the Adhesion of Quartz Particles air sampling but especially so for particles
to Pyrex Glass because they are much less uniformly dis­
persed i n the ambient air as well as i n pro­
Root-mean- cess equipment. (Factors involved i n select­
Range of square
profilometer surface Adhe- ing source sampling locations are discussed
reading irregularity sion in the next section, 4.4). Special consider­
Surface (nm) (nm) (%)* ation must be given to sources, obstacles,
Pyrex flat 1 203.2-228.6 215.9±12.7 100 meteorology, and time. Particles o f all sizes
Pyrex flat 2 203.2-381.0 292.1 ±12.7 67 are being continually emitted into the atmo­
Pyrex sphere. The larger particles fall out rapidly
microscope and moderate sized ones more slowly, as
slide 304.8-381.0 342.9±12.7 59 listed i n Table 4:11. A t the same time the
Pyrex flat 3 457.2-508.0 482.6 ± 12.7 45 very small particles (Aitken nuclei) tend to
*As percentage of adhesion to pyrex flat 1. become attached to larger particles. This
(Corn, M. 1961. J . Air Poll. Control Assoc. process is fastest for the very small particles
11:566) (80%/min. for 0.02 fim particles) becoming
 AIRBORNE PARTICLES 33

1x 10"

1x 10 5

0.2 0.4 0.6 0.8 1.0 1.2 1.4 1.6 1.8 2.0
Particle Size,

Figure 4:5 —Calculated number of droplets per m3 in various size intervals at different relative
humidities in a sulfuric acid mist sample having a concentration of 39 [ig/m*.
(National Air Pollution Control Administration, 1969. U.S. Dept. Health, Education and Welfare,
Air Quality Criteria for Sulfur Oxides. AP-50, pp. 1-22.)

Table 4:11 Settling Velocity of Unit Density tides) becoming attached to larger parti­
Spheres in Still Air cles, and much slower for larger particles
Settling Velocity, V ,
s
(14%/day for 0.2^m size). These two pro­
Particle Diameter, fim cm/s cesses then tend to produce an equilibrium
particle size distribution in the atmosphere,
1000 385
500 200 but because of the variable speeds of these
220 76 processes there will be significant gradients
100 25.1 near major sources, especially i f they emit
50 7.2 large particles. The height of the source, the
40 4.8 wind velocity and turbulence, and particle
30 2.7 size distribution will determine how fast the
10 0.30 particles settle out.
1 0.0035
Point sources do not generally distribute
34 GENERAL TECHNIQUES
their particles evenly i n all directions. The
particles travel with the wind which is quite
variable. A sampler located directly i n the
path o f a plume may show 10 or 100 times
as much aerosol as one located a few meters
away. I n the lee o f buildings, trees, hills or
other obstacles, the air is often quiescent
and will contain fewer particles. Strong gra­
dients usually exist i n the vertical direction
also. Because most sources are near the
ground and vertical mixing is frequently
limited, particle loading usually decreases
with height.
Time is also an important variable. I n
most cities the mixing height varies greatly
throughout the day. Often there is an over­
night inversion which keeps all o f the pollu­
tion close to the ground and produces rela­
tively high concentrations. When the sun
comes up the inversion gradually lifts and
generally breaks i n the morning or early
afternoon, at which time there is good dis­
persion. Particle concentrations at a given
location can easily vary by a factor o f 10 or
more during a 24-h period. Thus, a sample
taken i n the afternoon may give very little
information on the air pollution problem.
On the other hand, a 24-h sample might
average a few hours o f critically severe pol­
lution over a 24-h period and cause little
alarm although a serious problem may ex­
ist. I n spite o f this, 24-h samples have be­
come a standard practice, and samplers are
generally started at midnight by an auto­
matic timer and run until the following
midnight. From the point o f view o f
record-keeping this keeps the entire sample
on the same date, but from an air pollution
point o f view it produces a sample that has
one day plus parts o f two different nights.
It would be preferable to include a single
night by starting 24-h samples at 6.00 P . M .
or 6.00 A . M . or to sample for a shorter
period, or for the critical period only.
To summarize:
a. A sample should be taken at the point
of major interest, i f possible, but care
should be exercised i n extrapolating its rele­
vance to the ambient air o f a whole city or
even a nearby part o f i t ; and
b. i f it is desired to get a general measure
of atmospheric aerosol as it affects man,
plants and animals, one should:
(1) not be directly downwind from a ma­
j o r point source;
(2) locate the sampler about 1.5 m above
ground level;
(3) locate downwind from major obsta­
cles a distance o f about 10 times their
height;
(4) take several such samples at different
locations i n the area o f interest;
(5) sample the time o f day o f greatest
interest but do not consider this representa­
tive o f all day or o f any small part o f the
sampling period.
4.4 FACTORS INVOLVED IN S O U R C E SAM­
PLING.
4.4.1 General Considerations. Source
sampling refers to the collection o f pollu­
tants at their source as they are being re­
leased into, and before they are diluted by
the ambient atmosphere, or for design pur­
poses at a point where controls may be i n ­
stalled. The objective is to collect a sample
that is representative o f the material that is
emitted or to be controlled. The material
may not always be confined i n a stack or
duct, and the composition and amount may
vary with times, as i n batch processes. I t
may be coming out o f vents, windows, etc.
Collection o f such samples is usually d i f f i ­
cult and costly and requires advanced plan­
ning. Because o f this, source testing is sel­
dom done unless there is reason to suspect
pollution or to determine the efficiency o f
control equipment or to get data for design
purposes.
4.4.2 Planning and Preparation for the
Test. The specific points o f sampling are
generally determined by examination o f
drawings or flow diagrams, and discussions
with plant engineers or others who under­
stand the process or source o f emissions
and its variation with time. A site visit is
generally required to make the final selec­
tion. Criteria for selection o f the sampling
location are:
a. A t the point o f greatest interest (such
as stack outlet);
b. i n a straight section o f pipe 5 or more
diameters downstream and 3 or more diam­
eters upstream from any bends or flow dis­
turbances;
c. accessible to sampling personnel and
equipment;
 AIRBORNE PARTICLES 35

d. utilities such as electricity, water, and

air available i f needed; and
e. the point o f sampling is a safe location
for testing.
Often an ideal site cannot be found and a
compromise must be reached.
8 HOLES-0.04 DIAMETER E Q U A L L Y
Generally the selected site must then be SPACED F R E E FROM BURRS

prepared for testing by making access holes

(confined source) or a collection system
(unconfined source). A hole to which a 3"
(8 cm) pipe coupling can be welded is gen­
erally about the right size but the particular
sampling probes that are to be inserted
must be considered before this decision is
reached. A scaffold accessible to these
holes may have to be constructed and utili­
ties supplied. Some estimate o f flow rate,
temperature, pressure, dew point, particle
size and gas composition must be made i n
advance to plan the sample collection. A f ­
ter the detailed procedure for collection has
been worked out, the sampling and analysis Figure 4:6 —Standard pitot tube.
equipment must be assembled and prepared
for transportation to the site. The process
P = absolute pressure o f gas i n duct,
variables must be considered i n determining
inches o f H g .
the proper time or times for sampling. The
T = absolute temperature of gas in
management or persons responsible for the
duct, ° R.
source must be contacted and arrangements
made to have access at the desired times. I f measurements are made i n SI units (ve­
The sampling crew must be organized and locity i n m/s, pressures in kPa, temperature
briefed and the analysts alerted. in K ) , the constant i n the above equation
4.4.3 Sampling. The equipment and becomes 23.9. The specific gravity is the
crew are transported to the site. Source ratio o f mean molecular weight o f the gas
flow conditions must first be measured. A l ­ in the stream to that o f air, 28.95. From this
though a number o f methods are available, point onward, this discussion will use En­
pitot tubes are generally preferred. I n es­ glish units; operation in SI units is reasona­
sence a pitot tube consists o f two concentric bly obvious.
tubes (Figure 4:6) one o f which has an A number o f measurements must be
opening facing upstream and the other has made at various points in the stream. The
openings perpendicular to the flow. The cross-section should be divided into equal
pressure difference between these is called areas o f about one square foot. I n a rectan­
the velocity head and may be used to deter­ gular duct, a measurement should be made
mine the flow rate from handbook tables or in the center o f such rectangle with a mini­
the following equation: mum o f nine measurements.

X X X

X X X

where V = flow velocity i n feet per min­ X X X

ute at measured conditions.
P = velocity head i n inches o f water.
v I n the case o f a circular duct, annular rings
S = specific gravity o f the gas stream of equal area (about 1 f t . ) should be visu­
2

compared to air. alized and measurements should be made at

36 GENERAL TECHNIQUES
the center o f each such ring on each o f two
diameters which are 90° from each other
(with a minimum o f 8) as shown below i n
Figure 4:7.
After calculating velocities, each velocity
is multiplied by the area it represents and
these flow rates may be added to obtain the
total flow rate. Samples should be taken at
each o f the flow measuring points and the
amounts collected and weighted together i n
proportion to the individual flow rates
found at the corresponding points.
Particulate sampling should be carried
out with probes inserted i n the duct at each
of the points o f flow measurement. The
probes must be pointed directly upstream
and the sampling rate adjusted to provide
isokinetic conditions at each point.
It is often more convenient to make flow
measurements at the same time as the sam­
ple is collected by using a combined probe.
This is especially useful i n cases where the
flow rate is varying and the sampling rate
R E G U L A R DUCT
CIRCULAR DUCT
Figure 4:7 —Location of pitot traverse points.
Note: • Indicates points of locations of Pitot
Tube.
can be correspondingly varied to maintain
isokinetic conditions.
Lack o f equipment and personnel gener­
ally dictates that the points are sampled
successively rather than simultaneously.
However, i n determining the efficiency o f
control equipment, it is highly desirable to
take simultaneous samples before and after
the control. Greater variability ( < ± 50%)
for flow rate or sample concentration from
point to point or time to time make it advis­
able to increase the number o f samples rec­
ommended above and place more emphasis
on simultaneous sampling, whereas a some­
what smaller number can be taken i f the
stream is more consistent ( < ± 20%).
Sample concentrations determined at vari­
ous cross-sectional points at a given loca­
tion should be weighted together i n propor­
tion to the flow rates found at each point
and the areas represented as follows:
M = 60(C V A + C V A + . . . C V A )
1 1 1 2 2 2 n n n
Where M = the mass o f pollutant i n
pounds per hour,
Ci» C ,
2
C n = the concentration o f the
pollutant found at sam­
ple points 1, 2, . . . n,
respectively. I n pounds
per cubic foot at mea­
sured conditions,
v„ v , . . . v
2 n the flow velocity at sam­
ple points 1, 2, . . . n
respectively i n feet per
minute at the same con­
ditions used for values
of C.
Aj, A ,
2 the square feet o f cross-
sectional areas repre­
sented by sample points
1,2, . . . n respectively.
Separate calculations will need to be made
for each pollutant analyzed.
Diffuse emissions such as wafts out
doors or windows, or dust from an open
quarry, or construction site, generally re­
quire individual ingenuity i n sampling. A
useful technique applicable to some o f
these is referred to as variable dilution sam­
pling. I n this method a blower is operated
at the site and ducted so that all o f the emis­
sions from the location o f interest enter this
duct. Although the emissions rate may vary
considerably, the fan mass-flow rate is kept

constant at a value exceeding the emission
rate. A tight seal is not made so that when
the effluent rate is below the blower rate,
ambient air can enter freely to make up the
difference so the total flow remains con­
stant. This approach prevents creation of
any appreciable unnatural suction at the
source that might cause excessive emis­
sions. A t the output o f the blower or some
other point where the emissions and
makeup air are thoroughly mixed, a sample
can be taken for analysis. The sample con­
centration (C) times the total fan rate (V)
will give the emission rate. I f a constant fan
mass-flow rate isused, a sample can be col­
lected at any smaller constant mass-flow-
rate and combined over any desired period
of time and the combined sample concen­
tration will be a true measure o f the
weighted pollution concentration over that
period. This can be multiplied by the total
blower flow over that period to obtain the
mass o f emissions for the period.
Care must be exercised to make sure that
particles do not settle out i n the duct prior
to the sample point and that the sample
point is operated isokinetically, i f neces­
sary. Another precaution of importance is
to separate the particulate sample at the
temperature of interest. I f the temperature
is too hot, liquid aerosols o f interest may be
vaporized and not be collected, whereas i f
the temperature is too low, water or other
condensable vapors may form mists or
aerosols which will collect with the solids
and plug up the filter or give unrealistic
results. The variable dilution technique
tends to minimize condensation o f vapors,
which may be beneficial or undesirable de­
pending on the particular case.
While the sample is being taken, the fol­
lowing data should generally be taken si­
multaneously:
Time and Date
Sample Location
Sample Flow Rate
Sample Pressure
Sample Temperature
Dew Point
Plant Operating Conditions
Flow Rates and Variability
Production Rates and Variability
Pressures
Temperatures
G A S E S AND V A P O R S 37
Persons Involved i n Sampling
Approved with modifications from
2nd edition
Subcommittee 9
J . N . H A R M A N , Chairman
5. Sampling and Storage of Gases
and Vapors
Gases and vapors follow the normal laws
of diffusion and mix freely with the sur­
rounding atmosphere. They are not af­
fected by inertial or electrostatic forces
which may disturb particles. These charac­
teristics make gases and vapors easier to
sample than dusts or fumes.
5.1 G E N E R A L CONSIDERATIONS. In the
selection o f an appropriate sampling proce­
dure, some important factors should be
considered:
a. The integrity o f the stored gases or
vapors collected; is it maintained prior to
analysis, i.e. will degradation or reactions
take place in the sample probes or collec­
tion media?
b. The sampling procedure; is it optimal
for the analytical procedure to be used?
c. The stability o f the gas or vapor o f
interest; is it capable o f being stored with­
out wall losses i n the container chosen for
sampling?
d. That all precautions to prevent leaks
and sample contamination have been ad­
dressed by appropriate quality control
measures.
In general, one should always attempt to
reduce time intervals between sample col­
lection and analysis —good technique calls
for protection o f collected samples against
sunlight and temperature extremes during
transport.
5.2 G A S P U M P S ( A I R M O V E R S ) . Appropri­
ate gas pumps are available in several dif­
ferent types and sizes. They should have
adequate capacity and perform uniformly.
Mechanical pumps, nearly always driven
with electric motors, are necessary for pro­
longed periods o f operation. Induction mo­
tors are preferred, since this type operates
uniformly in spite o f variations in line
load.
38 GENERAL TECHNIQUES
Pumps that are Teflon* or stainless steel
lined are available from many commercial
sources. These pumps with inert surfaces
are necessary when i t is required to have a
sample stream pass through the pump prior
to the sample collector. This type o f pump
is essential when sampling for reactive gases
such as ozone, or collecting trace quantities
of hydrocarbons and halocarbons.
Other air movers such as hand pumps,
air aspirators or siphons are rarely, i f ever,
used. Use o f these devices as air movers is
not recommended, as they are not able to
control flow rates accurately or be used for
prolonged sampling periods.
5 . 3 G A S M E T E R I N G D E V I C E S . I t is usually
necessary to measure the total volume o f air
sampled i n an experiment. For this purpose
two types o f meters are available, volume
meters and rate meters. Volume meters re­
cord the total volume o f sample which is
passed through the sampling train. Rate
meters indicate the flow rate, and this rate
multiplied by the sampling time gives the
total volume sampled.
Rate meters consist o f rotameters, elec­
tronic flow controllers, and critical orifices.
Rotameters are made to cover a large range
of flow rate, from cubic centimeters per
minute to hundred o f liters per minute, and
are made o f glass or plastic. Glass rotame­
ters are preferred when sampling for reac­
tive gases or vapors.
Electronic mass flow meters are a recent
development that are read out electroni­
cally. These units are considered more accu­
rate than rotameters, as dependable, but
are more expensive.
Use o f critical orifices offers a conven­
ient and inexpensive way to produce con­
stant flow rates i n sampling trains. Use o f
hypodermic needles is particularly advanta­
geous ( 1 ) . To maintain constant flow i t is
necessary that the pressure drop across the
critical orifice be equal to or greater than
0.55 times the upstream pressure. A sam­
pling pump o f sufficient capability to main­
tain this pressure drop must be used. The
critical orifice must also be kept free o f
dust to prevent plugging. A typical sam­
pling train containing a critical orifice for
flow control is shown i n Figure 5:1.
•Teflon is a registered trademark of E.I. DuPont
de Nemours Co., Inc., Wilmington, Delaware
Figure 5:1 — Use of critical orifices for sample
collections
5.4 S A M P L E P R O B E S . I t is essential that
an air sample stream be able to reach the
collection device i n an unaltered state. Sam­
pling probes used for sample transport are
sometimes o f substantial length; e.g., sam­
pling the outside air from within a building.
To insure no losses o f gases or vapors the
sampling probe must be constructed o f i n ­
ert materials such as glass or Teflon ( 2 ) .
Use o f materials such as rubber or nylon is
not advised even when sampling for rela­
tively inert gases such as carbon monoxide
or nitric oxide. Sampling probes should be
kept to the shortest possible length, and
never allowed to accumulate particulate
matter or moisture.
5.5 SAMPLING TECHNIQUES FOR C O L L E C ­
TION O F T H E W H O L E SAMPLE WITHOUT
CONCENTRATION O F GASES AND VAPORS.
These techniques include the use o f plastic
bags, glass containers and metal containers.
For many low-boiling compounds such as
methane, ethane, propane, and the fixed
gases ( N , 0 and C 0 ) , this type o f sam­
2 2 2
pling is the preferred and/or sole method o f
collection.
5.5.1 Plastic Bags. The typical materi­
als used i n plastic bags are Aluminized
Scotch-pak*; Scotch-pak, Mylar**, Sa-
ran***, and Tedlar****. Entry into the
bags is by a fitting seated i n and connected
to the bag to form an integral part o f the
*Scotch-pak is a registered trademark of Minnesota
Minning and Manufacturing Co., St. Paul, Minne­
sota
**Mylar is a registered trademark of E.I. DuPont
de Nemours Co. Inc., Wilmingon, Delaware
***Saran is a registered trademark of Dow Chemi­
cal, USA, Midland, Michigan
****Tedlar is a registered trademark of E.I.
DuPont de Nemours Co. Inc., Wilmingon, Dela­
ware

bag. Some bags have two fittings to facili­
tate sample taking by simply passing sam­
ple air through the bag. These plastics are
appropriate for inert gas collection and
storage. Bags are particularly well suited
for integrated air sampling. A n integrated
bag sampling train is depicted i n Figure 5:2.
Tedlar bags are especially resistant to wall
losses for many reactive gases, and have
been i n use i n a large toxics air monitoring
network (3).
Whether a substance can be sampled and
stored i n a plastic bag should be determined
in the laboratory prior to field use. Litera­
ture references may also be used to deter­
mine proper usage o f bag sampling. Bags
can be reused after purging with clean air
and checking for any residual gaseous com­
ponents. Even "inert" plastics display out-
gassing properties (4).
5.5.2 Glass Containers. Glass contain­
ers can be used both i n an evacuated or
pass-through displacement mode. I n the
displacement mode gas sample tubes such
as those i n Figure 5:3 are opened on both
ends. A pump is attached to one end and a
sampling probe is attached on the other
end, i f needed. The pump is activated and
approximately 10 times the sampling tube
volume is passed through the system. The
stopcocks are immediately shut and the
sample is then secured. Alternatively, the
tubes may be evacuated using a vacuum
pump, then the sample collected by simply
opening one o f the stopcocks. Glass con­
tainers, by nature o f their small volumes,
provide basically instantaneous "grab"
samples.
Glass containers are also convenient for
CONTAINER
Figure 5:2—Integrated bag sampling train
G A S E S AND V A P O R S 39
TYPE A
SCREW CLAMP AND TUBING
TYPE B
WITH STOPCOCKS
Figure 5:3—Gas sample tubes for collection of
sample by displacement or vacuum
collecting samples directly i n a chemical ab­
sorbent i n which they are later analyzed.
When used i n this manner, rather than sam­
ple tubes, 1-3 L flasks are better suited,
with only one stopcock used for both evac­
uation and later sample entry. This is a
proven and frequently used technique
(5,6).
I n general, glass containers are excellent
for inert gases such as oxygen, nitrogen,
methane, carbon monoxide and carbon d i ­
oxide. Reactive gases such as hydrogen sul­
fide, oxides o f nitrogen and sulfur dioxide
are not recommended for direct collection
and storage. The reactive gases are, how­
ever, amenable to the chemical absorbent
technique referred to earlier, which fixes
and stabilizes the reactive gas prior to
analysis.
5.5.3 Metal Containers. Stainless steel
containers are also widely used for the col­
lection o f the inert gases mentioned i n Sec­
tion 5.5.2. They are not suitable for a ma­
jority o f the reactive gases commonly
encountered. I n metal containers, wall
losses are extremely rapid for the reactive
gases. Metal containers come i n a variety o f
sizes from 1 to 34 L . A n advantage o f using
the large sizes is that a longer, more inte­
grated, sampling time is possible when
compared to the glass containers.
A n important improvement i n stainless
steel containers has recently taken place. A
treatment process for the containers has
been introduced which reduces the surface
adsorption o f gases by reduction o f the i n ­
ternal exposed metal surface area. This pro-
40 GENERAL TECHNIQUES

prietary process o f S U M M A * polishing is

available i n various container sizes up to 6
L from two manufacturers (7,8). Extensive
studies have shown excellent stability, i n the
order o f weeks, for hydrocarbons and vari­
ous halocarbons even i n the low and sub
parts per billion (9). Extensive studies on
inorganic reactive compounds have not
been performed to date. Figure 5.4 depicts
a typical sampling setup for long term (up
to 24 h) sampling for ambient hydrocar­
bons and halocarbons. This technique is
considered state-of-the-art for the sampling
of low level toxics i n air.
5.6 SAMPLING TECHNIQUES W H I C H E X ­ Figure 5:5A—Bubbler absorbers with diffusers
T R A C T A N D C O N C E N T R A T E G A S E S A N D VA­
PORS F R O M A I R S A M P L E S .
5.6.2 Dry Collection Systems. A n y gas
5.6.1 Wet Collection Systems. Sam­
or vapor w i l l , to some degree, adhere to any
pling techniques employing simple bubblers
solid surface at ordinary or low tempera­
or bubblers with diffusers are widely used,
and a large diversity o f applications may be tures. This phenomenon is called adsorp­
found. Bubblers, depending on design and tion. Porous solids expose not only their
use, may contain from 5 to 200 m L o f ab­ exterior surface, but their interior surfaces
sorbent. Bubblers with diffusers to provide as well, and some such solids indeed possess
more efficient air-to-absorbent contact are a vast network o f extremely minute chan­
especially needed for acceptable collection nels and submicroscopic pores within their
of gases that are difficult to absorb; e.g., bodies. Such materials have practical value
nitrogen dioxide (5). Figure 5:5 shows the as adsorbents. These solids include acti-
basic design differences among bubbler
types.
Specific use o f a bubbler sampling train
must be based on prior knowledge that a
bubbler-absorbent-flow rate combination
will result i n an acceptable collection effi­
ciency, usually 90% or greater. A typical
complete sampling train using bubblers is
shown i n Figure 5:6.

SAMPLE PROBE

Figure 5:5B —Simple bubble absorbers

Fl
FLOW CONTROLLER

J * M
I > = = D RUBBER TUBING

TER
Figure 5:4—Use of stainless steel canisters for
integrated air samples

GLASS WOOL

*SUMMA—Proprietary process of Molectrics, Figure 5:6 —Typical sampling train for bubbler
Inc., Carson, California absorption of gases
 G A S E S AND V A P O R S 41

vated carbon, silica gel, activated alumina, Table 5:1 Retention Volume Estimates for
and various active earths. Synthetic poly­ Compounds on Tenax
meric materials have also been used for ad­ Estimated Retention
sorption o f gases, particularly volatile or­ Volume at 100° F
ganic gases. Among the most commonly Compound (38° C), L / g .
utilized of the synthetics has been Tenax*. Benzene 19
A l l these differ widely in the number and Toluene 97
kinds o f substances they absorb, as well as Ethyl Benzene 200
in the amount of sorbed substances they Xylene(s) 200
will retain. Cumene 440
Silica gel, charcoal and Tenax are the n-Heptane 20
1-Heptene 40
most widely used adsorbents, and will be
individually addressed as to their sampling Chloroform 8
characteristics. These materials adsorb a Carbon Tetrachloride 8
1,2-Dichloroethane 10
variety of volatile organic gases, and have
1,1,1 -Trichloroethane 6
excellent storage properties prior to analy­ 80
Tetrachloroethylene
sis. Rarely are the materials used for the Trichloroethylene 20
quantitative collection and analysis o f inor­ 1,2-Dichloropropane 30
ganic gases. I n the use o f the materials it is 1,3-Dichloropropane 90
important after the sampling is completed Chlorobenzene 150
that the adsorbent tubes be immediately Bromoform 100
sealed and properly stored. Storage in a Ethylene Dibromide 60
sealed plastic or glass vial is most Bromobenzene 300
satisfactory.
Silica gel is a highly polar substance, and
therefore will preferentially adsorb water
more strongly than organic gases and va­ per billion gas concentrations. Charcoal has
pors. Caution must be exercised when sam­ been used for a wide spectrum of volatile or­
pling high relative humidity gas streams to ganic gases and vapors. Solvent extraction,
insure high collection efficiency for gases usually with carbon disulfide, prepares the
of interest. Silica gel is an excellent adsorb­ samples for GC or IR analysis. When using
ent for a variety o f organic substances in charcoal for specific applications, refer to the
the part per million concentration range. literature to ascertain the quantity of charcoal
The adsorbed gases must be solvent ex­ to use, mesh size, and the sampling parame­
tracted, for subsequent GC or IR analysis. ters; i.e., flow rate and volume. This is essen­
Extractants are usually polar substances tial to insure acceptable collection efficiencies
such as alcohols or dimethylsulfoxide or for a particular gas or vapor.
water in combination with carbon disulfide Tenax, a polymeric material (poly-2,6-
(10). Experimental data show excellent re­ diphenyl phenylene oxide) is used for air
coveries from standard gas streams for a sampling at part per billion and lower levels
large variety of gases (10). of volatile organic compounds (11). Tenax
Activated charcoal is the most widely is nonpolar, and water vapor does not af­
used adsorbent for the concentration and fect sampling in any way. Unlike silica gel
storage of organic vapors from air streams. and activated charcoal, the adsorbed gases
Many N I O S H / O S H A and A C G I H meth­ are not solvent extracted after collection,
ods are based on the use of activated char­ but are thermally desorbed. The thermal
coal. Since charcoal is nonpolar, sample desorption takes place over 10-15 seconds
streams of high relative humidity do not at a temperature of approximately 240° C.
present the difficulties mentioned for silica This rapid desorption is accomplished by
gel. Larger sample volumes may be taken, heating to 240° C while purging the absorb­
resulting in procedures which can provide ent with helium gas, allowing all the col­
sensitivity capabilities for analysis of subpart lected gases to be introduced into a gas
chromatograph as a single injection. This
T e n a x — Developed by A K Z O Research Labora­ accounts for the extreme sensitivities possi­
tories, marketed by Enka N . V . , the Netherlands. ble when using Tenax. Table 5:1 shows the
42 GENERAL TECHNIQUES
TENAX ABSORBERS CRITICAL ORIFICES
Figure 5:7 —Multiple and variable flow setup for
Tenax sampling
sampling limitations o f specific com­
pounds, expressed as retention volumes. It
is recommended that, prior to using Tenax
for sampling, Reference 10 be reviewed. As
with any methodology dealing with parts-
per-billion detections, stringent quality
control procedures must be followed. Fig­
ure 5:7 depicts a multiple and variable sam­
pling system recommended when using
Tenax absorbers (12). Flow rates normally
used are in the 5 to 5 0 L / m i n range. The
flow rate and time of sampling are deter­
mined primarily by the retention volume o f
the compound(s) o f interest (see Table 5:1).
Multiple parallel sampling using variable
flow rates allows specific indications o f
data quality. This sampling approach adds
an analytical burden, but provides impor­
tant data regarding adequate retention, ar­
tifact formation, and hence, applicability
to specific compounds.
REFERENCES
1. LODGE J.P., J.B. PATE, B.E. AMMONS AND G.A.
SWANSON. 1966. The Use of Hypodermic Needles
as Critical Orifices in Air Sampling. J . Air Pollu­
tion Control Assoc. 16:197-200
2. FEDERAL REGISTER. Vol. 44 No. 92, Thursday,
May 10, 1979 p 27.
3. OSLUND W.E. 1986. Implementing a Quality As­
surance Program for Sampling and Analysis of
Ambient Air Toxics Compounds. Proceedings of
the 1986 EPA/APCA Symposium on Measurement
of Toxic Air Pollutants, pp 364-374.
4. LONNEMAN W.A., J . J . BUFALINI, R.L. KUNTZ
AND S.A. MEEKS. 1981. Contamination from Fluo­
rocarbon Films. Anal. Chem. 15:99-103.
5. ASTM COMMITTEE D-22. 1977. Method of Test for
Oxides of Nitrogen in Gaseous Combustion Prod­
ucts (Phenoldisulfonic Acid Procedures) ASTM
Designation D 1608-77 (Reapproved 1985). Book
of ASTM Standards, Vol. 11.03.
6. SALTZMAN B . E . 1954. Colorimetric Determination
of Nitrogen Dioxide in the Atmosphere. Anal.
Chem. 26:1949-1955.
7. D E M A R Y S C I E N T I F I C INSTRUMENT L T D . , S.E.
1122 Latah Street, Pullman, Washington 99163.
8. B I O S P H E R I C RESEARCH CORP., 1121 N.W.
Donelson Road, Hillsboro, OR 97123.
9. O L I V E R K.D., J.D. P L E I L AND W.A. M C C L E N N Y .
1986. Sample Integrity of Trace Level Volatile Or­
ganic Compounds in Ambient Air Stored in
SUMMA Polished Cannisters. Atmos. Environ
20:1403-1411.
10. F E L D S T E I N M., S. BALESTRIERI A N D D.A.
L E V A G G I . 1967. The Use of Silica Gel in Source
Testing. Am. Ind. Hyg. Assoc. J. 28:381-385.
11. RIGGIN R.M. 1984. Compendium of Methods for
the determination of toxic organic compounds in
ambient air. EPA-600/4-84-041. April 1984,
USEPA, Research Triangle Park, N. C.
12. W A L L I N G J . F . 1984. The Utility of Distributed Air
Volume Sets When Sampling Ambient Air Using
Solid Absorbents. Atmos. Environ. 18:855-859.
Subcommittee 9
D.A. L E V A G G I
J.N. H A R M A N , Chairman
6. Use and Care of Volumetric
Glassware
6.1 G E N E R A L C O N S I D E R A T I O N S . The Na­
tional Bureau o f Standards has specified
certain minimum requirements which must
be adhered to in the manufacture o f volu­
metric glassware.
6.1.1 The instrument must bear, i n
legible characters, the capacity, the temper­
ature at which it is to be used, and whether
it is to contain or to deliver the specified
volume.
6.1.2 The instrument must bear the
name or trade-mark o f the manufacturer
and an identification number. I f the instru­
ment has detachable parts, such as stoppers
or stopcocks, such parts must also bear the
same numbers.
6.1.3 The inscription and capacity
mark must be etched or engraved but not
scratched on the instrument.
6.2 VOLUMETRIC FLASKS. Volumetric
flasks most generally employed are those
which are calibrated to "contain" a definite
volume and have only one mark etched on
the neck o f the flask. There are also flasks

available which have two etched marks.
The lower mark indicates the volume o f so­
lution the instrument "contains," while the
upper mark indicates the volume of solu­
tion the instrument will "deliver." The flask
may be provided with a ground glass or
plastic stopper to prevent any spillage when
inverted for mixing solutions.
The sizes o f volumetric flasks generally
employed in the laboratory "to contain"
(TC) are 1000-mL, 500-mL, 250-mL, 100¬
m L , 50-mL, 25-mL, and 10-mL. Flasks cal­
ibrated "to deliver" (TD) are not suffi­
ciently accurate since the amount o f liquid
adhering to the sides o f the flask after emp­
tying can only be approximated.
6.3 T H E U S E O F V O L U M E T R I C F L A S K S .
6.3.1 Preparation of a Solution of Def­
inite Volume. Transfer the solution, which
has been previously prepared i n a beaker,
by means o f a funnel, to the volumetric
flask. Pour the solution with the aid o f a
glass rod held across the lip o f the beaker.
Wash the beaker, when empty, thoroughly
into the flask by means o f a wash bottle,
holding the beaker with its lip in the funnel
and then directing a stream o f distilled wa­
ter from the wash bottle around the inside
walls of the beaker. Do not use an excessive
amount o f wash water, so that the capacity
of the flask is not exceeded. Rinse the fun­
nel into the flask and then remove. Stopper
the flask and mix the solution by shaking
and inverting the flask several times. Re­
turn the flask to the upright position and
add water until the level of the liquid is
slightly below the graduation mark. Repeat
the mixing procedure. Return again the
flask to the upright position. W i t h the eye
on the level o f the graduation mark, add
water slowly by means o f a pipet until the
level o f the liquid reaches exactly the gradu­
ation mark. Stopper and shake well by i n ­
verting many times, so that the contents are
completely mixed.
It should be kept in mind that the solu­
tion and the diluting water must both be at
room temperature.
A t times, it is necessary to use a portion
of an "aliquot" of a prepared solution of
definite volume. The "aliquot" portion,
taken from the volumetric flask, is mea­
sured out by means of a pipet, buret, or a
smaller volumetric flask calibrated to "de­
liver."
VOLUMETRIC GLASSWARE 43
Volumetric flasks calibrated to "deliver,"
should be emptied by inclining them gradu­
ally until they are almost in a vertical posi­
tion. Allow to drain in this position for
about 30 s and then bring the mouth o f the
flask in contact with the receiving vessel so
that the last adhering drop is removed from
the flask.
6.3.2 Calibration of Volumetric
Flasks. Calibrated volumetric flasks pur­
chased from reliable manufacturers are suf­
ficiently accurate. However, volumetric
flasks should be calibrated when the accu­
racy o f the capacity o f the flask is in doubt.
Circular N o . 602 o f the National Bureau
of Standards gives capacity tolerances for
flasks. Determine from the weight o f the
water contained the capacity o f the flask at
20°C using Table 6:1. For example, i f the
weight o f water contained in a 50-mL volu­
metric flask at 23°C is 49.8310 g, the ca­
pacity o f the flask calculated from the table
is 49.8310 x 1.0034 or 50.000 m L .
6.4 B U R E T S . The most common size bu­
ret employed i n the laboratory is the 50-mL
capacity buret, graduated to 0.2 m L in its
smallest subdivision. The buret has a capil­
lary tip o f definite size so that the free out­
flow time of a liquid does not exceed 3 min
nor is it less than 90 s. The optimum out­
flow time for a 50-mL buret is about 100 s.
Other sizes of burets are o f 10-, 25-, and
100-mL capacity.
The Friedman and La Mer weighing bu­
ret o f somewhat different construction al­
lows increased accuracy in titration. The
amount o f liquid necessary for a titration is
weighed rather than measured by volume.
The buret is short, of light weight, and
equipped with lugs so that it can be sus­
pended from hooks on the stirrups o f a bal­
ance. It is also provided with a glass stop­
per, and a cap that fits over the delivery tip
to prevent evaporation.
There are also microburets available
which have a capacity o f 5 m L or less, with
the error o f delivery not exceeding 0 . 1 % .
The 5-mL microburet is graduated to 0.01
m L in its smallest subdivision.
The microburet is provided with a deliv­
ery tip drawn out to a fine point. Thus, the
liquid is delivered in minute drops and the
amount adhering to the tip is negligible.
These burets are available with a variety of
detachable tips that are attached to the bu-
44 GENERAL TECHNIQUES
Table 6:1 Apparent Weights and Volumes of Water Weighed in Air for Calibration of Volumetric
Glass Apparatus, Coefficient of Cubical Expansion, 0.000025 per °C
Temperature Weight of Volume of
°C 1 ml, g lg, mL
15 0.9979 1.0021
16 0.9978 1.0022
17 0.9977 1.0023
18 0.9975 1.0025
19 0.9973 1.0027
20 0.9972 1.0028
21 0.9970 1.0030
22 0.9968 1.0032
ret by means o f an adapter and can be eas­
ily removed for cleaning.
There are available microburets which
discharge very small volumes o f liquids
with high precision and reproducibility.
The Rehberg microburet does not have a
stopcock but uses a mechanical device
which discharges very small volumes o f liq­
uid by means o f a micrometer screw push­
ing against a column of mercury which, i n
turn, is in contact with the solution to be
measured. The Rehberg microburet is o f
0.125- or 0.225-mL capacity and, when
used, the tip is dipped below the surface o f
the solution to be titrated.
6.4.1 Use of the Buret. The buret is
thoroughly cleaned before use. I f the buret
has a glass stopcock, it should be greased
with stopcock grease, not petrolatum, so
that it is well lubricated and the capillary tip
allows the liquid to flow freely. The tem­
perature o f the solution withdrawn from
the buret should be near 2 0 ° C .
When measuring liquids by means o f a
buret, the readings are taken at the lowest
part o f the meniscus since this area is most
clearly defined. Exceptions are made for
dark liquids, such as permanganate and io­
dine solutions, where the bottom o f the me­
niscus cannot be seen. I n this case, the top
of the meniscus is taken as the reading. I t is
essential that the point chosen for reading
the meniscus should be the same for all
readings.
I n observing the bottom o f the meniscus,
the eye should be i n the same plane as the
meniscus as otherwise there will be an error
due to parallax. After recording the buret
reading, allow about 30 s for drainage be­
fore taking the final reading.
Temperature Weight of Volume of
°C 1 mL, g lg, mL
23 0.9966 1.0034
24 0.9964 1.0036
25 0.9961 1.0039
26 0.9959 1.0041
27 0.9956 1.0044
28 0.9954 1.0046
29 0.9951 1.0049
30 0.9948 1.0052
6 . 4 . 2 Calibration of Burets. Burets are
generally purchased from reliable firms
that guarantee their burets to meet the tol­
erances specified by the National Bureau o f
Standards. Burets are also available certi­
fied for accuracy by the National Bureau o f
Standards, with a Certificate o f Accuracy.
Circular N o . 602 o f the National Bureau
of Standards gives capacity tolerances for
burets. The capacities of burets (even the
certified ones) may change after a period o f
use, especially i f caustic solutions are used.
Therefore, burets should be calibrated
when the need arises.
6 . 4 . 3 Titrating with a Buret. Titration
is a method by which the quantity o f a sam­
ple dissolved i n a liquid is determined by
adding a volume or weight o f a reagent so­
lution that just neutralizes or completely re­
acts with this sample.
A standard solution is one o f definite
and accurately known concentration, usu­
ally expressed i n terms o f normality.
The standard solution is prepared by dis­
solving a definite weight o f a pure sub­
stance (primary standard) i n a specific vol­
ume o f liquid. The standard solution may
also be prepared by dissolving a definite
weight of a known substance (secondary
standard) i n a specific volume o f liquid and
then standardizing i t , i.e., determining its
concentration by titrating against a primary
standard.
The stoichiometric point is defined as the
point at which the addition o f a definite
amount of substance is exactly equivalent
to the dissolved substance being titrated.
Actually, this is the theoretical end point
which is determined from the equation of
the reacting substances. However, the stoi-

chiometric point will not always coincide
with the end point o f the titration.
The end point is the stage i n a titration
where the reaction is complete. This point is
determined by means o f indicators or other
devices and should coincide with or be as
close as possible to the stoichiometric
point.
6.5 V A R I A T I O N S O F V O L U M E S W I T H T E M ­
PERATURES. The National Bureau o f Stan­
dards has designated 20°C as the standard
temperature at which volumetric apparatus
will "contain" or "deliver" the stated vol­
ume.
Errors, although very slight, will occur i f
the volumetric apparatus is employed at
other temperatures. Glass will expand
slightly above 20°C and contract slightly at
lower temperatures. The average coeffi­
cient o f cubical expansion o f soft glass is
0.000025 and o f borosilicate glass,
0.00001/°C.
The variation i n volume, due to the ex­
pansion o f glass, over a small temperature
range from the standard 20°C is very slight
and may be disregarded for ordinary work.
For example, for a 50-mL buret, a variation
in volume due to a decrease or increase i n
temperature o f 5 ° C from the standard
20°C will be only plus or minus 0.007 m L .
For accurate work, however, the volume o f
the vessel may be corrected to 20°C by the
following formula:
V 2 0 = V [ l + a(20 - t)]
t
where V = volume at 20°C
2 0
V = recorded volume at ob­
T
served temperature
t = observed temperature
a = coefficient o f cubical ex­
pansion o f the material o f
the apparatus (for soft glass
0.000025/°C)
The volume o f a liquid is also affected by
temperature. Liquids generally expand at
higher temperatures. The more concen­
trated the solution, the greater will be the
variation. Fortunately, most standard solu­
tions are not too concentrated and usually
vary from 0.5 normal to less than 0.1 nor­
mal. Although the error involved i n the var­
iation i n the volume o f a 0.5 normal solu­
tion due to a 5 ° C change i n temperature is
small (about 0.05 m L for a 50-mL soft glass
VOLUMETRIC GLASSWARE 45
buret), nevertheless, it is important, when
employing volumetric apparatus or titrat­
ing, that the temperature o f the solutions
used is as close as possible to the standard
temperature o f 2 0 ° C .
6.6 P I P E T S . Pipets o f different types and
capacities are available. The transfer p i ­
pets, calibrated "to deliver" at 2 0 ° C , are
generally used to measure a single fixed vol­
ume.
The orifice o f the tip o f the pipet is o f
definite size so that the free outflow time
should not be more than 1 min nor less than
15 s for a 5-mL pipet; 20 s for a 10-mL
pipet; 30 s for a 50-mL pipet; 40 s for a
100-mL pipet and 50 s for a 200-mL pipet.
Pipets o f a capacity o f 0.1 m L or less are
calibrated to contain a definite volume.
When a liquid has been delivered from this
pipet into a receiving vessel, the pipet must
be rinsed with water at least three times.
These rinsings are added to the receiving
vessel.
Measuring or M o h r pipets are narrow
straight tubes which are graduated to de­
liver variable amounts o f liquids. They are
not ordinarily used for precise work.
6.6.1 Calibration of Transfer Pipets.
The accuracy o f a transfer pipet may be
checked when necessary. The maximum tol­
erances allowed for pipets by the National
Bureau o f Standards are listed i n Table
6:11.
The procedure used i n calibrating the p i ­
pet should be the same as that employed i n
the subsequent use o f the pipet. A glass-
stoppered weighing bottle o f a capacity o f
about five times the volume o f water to be
received is employed as the receiving vessel.
The bottle should be twice as tall as wide so
that the evaporation o f the water during
delivery will be negligible.
Table 6:11 Capacity Tolerances for Transfer
Pipets
Capacity (mL) Less
Than and Including Limit of Error (mL)
2 0.006
5 0.01
10 0.02
30 0.03
50 0.05
100 0.08
200 0.10
46 GENERAL TECHNIQUES
The bottle is weighed to the nearest milli­
gram by substitution i f a two-pan balance is
used.
6.7 G R A D U A T E D C Y L I N D E R S . Graduated
cylinders are calibrated "to deliver" or "to
contain" their stated volumes at 2 0 ° C .
These cylinders are not as accurate as bu­
rets, pipets, or volumetric flasks since they
are graduated to a tolerance o f 1 °7o.
Graduated cylinders are generally used to
deliver approximate volumes and where ex­
act volumes are not essential. Therefore,
for accurate measurements, use the other
volumetric apparatus described.
When using a graduated cylinder, select a
size which will be nearly filled by the vol­
ume of liquid to be measured.
6.8 C L E A N I N G G L A S S W A R E . Keeping lab­
oratory glassware clean is a necessary but
tedious chore. Much o f the glassware may
be cleaned by rinsing with distilled water i f
the rinsed glassware shows satisfactory
drainage of water without the appearance
of droplets adhering to its walls.
Glassware can be cleaned satisfactorily
by soaking directly after use i n a hot soapy
solution and then scrubbing with a brush.
The glassware is then rinsed thoroughly i n
hot tap water and, finally, in distilled water.
I f the glassware is greasy and dirty and
resists ordinary cleaning procedures, the
following general chemical methods may be
used.
6.8.1 Sulfuric Acid-Sodium Bichro­
mate Mixture. A solution of sodium d i -
chromate i n concentrated sulfuric acid
(cleaning solution) is widely used i n labora­
tories to clean volumetric glassware thor­
oughly. The oxidizing potential o f this mix­
ture, when hot, will cut most greases
adhering to glass walls and remove them
completely. The cleaning solution is very
corrosive and must be used with great care.
To prepare the solution, moisten about
2 0 g o f powdered sodium or potassium d i -
chrornate with water to form a thick paste
in a glass container. A d d about 500 m L o f
concentrated sulfuric acid to this paste and
stir. Store the mixture i n a glass-stoppered
bottle. Some o f the salt may be left undis­
solved in the solution. Do not remove this
excess salt but use the supernatant liquid.
The solution can be repeatedly used by
pouring back the unused portions into the
stock bottle. Continue using this red solu­
tion until it turns green, indicating that the
solution is spent. The color change is
caused by the chromium being reduced to
the chromic state.
Use this solution with the greatest care.
Never allow it to come into contact with
clothing or skin since sulfuric acid is very
corrosive and bad burns will result. I f some
of the solution does spill, wash o f f immedi­
ately with running water. Never store or use
this solution in metal or enameled contain­
ers. Finally, do not allow the solution to
spill on floors or desks.
6.8.2 The Use of the Cleaning Solu­
tion. Rinse the glass vessel with the clean­
ing solution or immerse it i n this solution.
The immersion time is dependent on the
stubbornness o f the dirt, and may vary
from a few minutes to 2 4 h. A hot cleaning
solution is more effective than a cold one.
After cleaning, rinse the glass vessel thor­
oughly with tap water until all traces o f the
solution are removed. Finally, rinse with
distilled water and allow to drain and dry.
A good test for cleanliness o f a glass vessel
is to fill it with water and then empty. A n
unbroken film o f water indicates that the
vessel is clean.
6.8.3 Alcoholic Sodium Hydroxide.
A n alcoholic solution o f sodium hydroxide
is very effective for removing greasy films
from glassware.
Prepare a solution by dissolving 120 g o f
sodium hydroxide i n 120 m L of water. A l ­
low to cool and add sufficient isopropyl al­
cohol to make 1 L .
Immerse the glassware in this solution
for not more than 3 0 min since the caustic
may attack the glass. Remove from the so­
lution and scrub with a stiff brush. Rinse
thoroughly with tap water and finally with
distilled water. Allow to drain and dry.
6.8.4 Synthetic Detergents. Anionic or
nonionic synthetic detergents are available
under proprietary names. Their cleaning
properties are due to their ability to reduce
surface tension and, at the same time, to
wet objects thoroughly. Synthetic deter­
gents are compounded with alkaline salts to
increase their detergency. They are efficient
in that they do not form scum on glassware
when used i n hard water as soaps do. Dilute
solutions o f about 0 . 5 % concentration are
sufficient to clean glassware.
Immerse the glassware i n a hot or cold

solution for about 30 min, remove from the
solution, and scrub with a stiff brush.
Rinse thoroughly with tap water and finally
with distilled water. A l l o w to drain and dry.
6.8.5 Cleaning Burets. Fill the buret
with the "cleaning solution" and allow to
remain overnight. Drain the solution into a
beaker and return to the stock bottles.
Rinse the buret thoroughly with tap water
and finally with distilled water. Grease the
stopcock with stopcock grease, i f necessary.
Invert the buret and allow to drain. A test
for the cleanliness o f a buret is to fill it with
water and then allow to drain. A n unbro­
ken film o f water, without droplets, should
remain on the walls o f the buret.
6.8.6 Cleaning Pipets.
a. Routine Procedure — Secure the pipet
in a clamp i n a vertical position with its tip
inserted into a beaker containing water or a
detergent. Attach, by means o f rubber tub­
ing, the mouth o f the pipet to a safety bot­
tle which, in turn, is connected to a vacuum
source. Draw up the solution by suction un­
til it almost fills the pipet. Release the suc­
tion and allow the solution to drain. Repeat
this procedure several times. Change the
beaker resting under the tip o f the pipet and
repeat the procedure with tap water and f i ­
nally with distilled water. Remove the pipet
from the clamp and allow to drain and dry.
b. Cleaning Greasy Pipets — Secure the
pipet i n a clamp i n a vertical position with
its tip inserted i n a beaker containing warm
sodium dichromate-sulfuric acid cleaning
solution. Draw up the warm solution until
it fills the pipet within an inch from the top,
using the procedure described earlier.
Pinch the rubber tubing with a clamp and
allow the solution to remain i n the pipet for
about 10 min. Then remove the pinch-
clamp and discharge the cleaning solution.
Rinse the pipet thoroughly with tap water
to remove any trace o f the solution and f i ­
nally rinse with distilled water. A l l o w the
pipet to drain and dry. To dry the pipet
quickly, rinse it several times with ethyl al­
cohol, after draining the distilled water,
and then rinse once with ether. Draw dry air
through the pipet until dry.
Approved with modifications from 2nd edition
Subcommittee 9
J . N . H A R M A N , Chairman
R E A G E N T WATER 47
7. Reagent Water
Wet chemical analysis implies that water
is used in the procedure. The water used i n
the reagents and for effecting solubiliza­
tion, dilution, and transfer is called reagent
water. Assumption that the reagent water
contributes insignificantly to the bias o f a
determination is warranted only when the
method o f test is not sensitive to the impu­
rity level i n the water.
7.1 P R E P A R A T I O N . Preferred practice is
to minimize the probability of error by pre­
paring and storing high purity water. The
preparation frequently begins with single
distillation from conventional apparatus.
Condensate from a conventional steam
boiler is suitable i f the conductivity o f this
condensed steam is no greater than 20 ptS/
cm. A condensate o f adequate quality can
almost always be obtained by partial con­
densation. The condensing train for this
purpose is arranged in steps, to (a) effect
partial condensation i n the steam pipe, (b)
dry the steam i n a separator (retains the
bulk o f the more soluble gases and the en­
trained boiler water), (c) effect additional
partial condensation o f the steam, and (d)
separate and store this condensate (the bulk
of the less soluble gases like carbon dioxide
remains i n the uncondensed steam).
7.2 D E M I N E R A L I Z A T I O N . Additional pro­
cessing o f the distillate or condensate is
now usually accomplished by demineraliza­
tion, the water being passed through a com­
mercially available mixture of anion and
cation exchange materials. Assumption o f
adequate upgrading by this process is un­
warranted, especially as ion exchange mate­
rials vary from lot to lot, and the effluent i n
each instance should be discarded until the
desired quality is obtained.
The successive steps o f distillation and
demineralization are both important. The
former removes nonvolatile electrolytes
and colloidal material, and the latter re­
moves ionizable volatiles such as carbon di­
oxide and ammonia (which together impart
the greater part o f the 20 /iS/cm conductiv­
ity o f the initial condensate).
7.3 F I L T R A T I O N . High purity water pre­
pared in this way is usually satisfactory for
the preparation o f reagents, for the labora­
tory preparation and dilution o f samples,
and for the final rinsing o f glassware and
48 GENERAL TECHNIQUES
the other apparatus. It may not be ade­
quate, however, for the determination of
trace concentrations o f some o f the materi­
als extracted from air samples. Organic
contaminants, in a range from 1 to 5 ppb,
are usually imparted to the water by its con­
tact with the ion-exchange materials. For
some determinations, such as tests for air­
borne micro-organisms, the "pure" water is
further purified by special low porosity f i l ­
tration. Examples of special purpose "puri­
fiers" are the sized Gelman or Millipore
membrane filters and the Barnstead Or­
ganic Removal Cartridge.
7.4 P R O T E C T I O N A G A I N S T CONTAMINA­
TION. Reagent water should at all times be
protected from atmospheric contamina­
tion. Replacement air should be passed
through a vent guard which removes unde-
sired species such as carbon dioxide and ox­
idant or reductant species from the makeup
air (for example, a drying tube filled with
equal parts o f 8- to 20-mesh soda lime, ox­
alic acid, and 4- to 8-mesh calcium chlo­
ride, each product being separated from the
other by a glass wool plug). Sample con­
tainers and tubing should be made o f mate­
rial that has been proven to be resistant to
even minor solvation by, or reaction with,
the water. Such materials usually are TFE-
f l u o r o c a r b o n , block t i n , quartz, or
polyethylene —in that order of preference.
It is general practice to filter the output o f a
reagent water source with a 10-/im filter to
remove particulate contamination. For spe­
cific grades o f reagent water, finer filters
are required.
The stored water is usually considered
suitable until it fails to pass the maximum
electrical conductivity test (0.1 /*S/cm at
25 ° C ) . I t may also be checked occasionally
for its consumption of potassium perman­
ganate, the requirement being that the per­
manganate color must persist for at least
one hour after 0.20 m L o f K M n 0 solution 4
(0.316 g / L ) is added to a mixture o f 500 m L
of the reagent water and 1 m L o f reagent
grade sulfuric acid (cone) in a stoppered
bottle of chemically resistant glass.
7.5 B L A N K D E T E R M I N A T I O N S . Despite all
of these preparations and precautions, min­
ute concentrations o f impurities in reagent
water may still affect the precision and ac­
curacy of some determinations. Pretesting
is a reasonable precaution. The usual prac­
tice is to run a "blank" to disclose whether
or not the water contains a detectable quan­
tity o f the material for which the analysis is
to be made. This "blank" is often designed
to include the reactive impurities in the re­
agents as well as those in the water.
Occasionally, the analyst will purposely
contaminate "pure" reagent water before it
is used. For example, water for the prepara­
tion of the reagents used to calibrate con­
ductivity cells is deliberately equilibrated
with the atmosphere in which the calibra­
tion will be made. This tends to stabilize the
content of dissolved gases, such as carbon
dioxide, that affect conductivity.
When the analyst is seeking especially
small concentrations of material in samples
of air, he must necessarily be especially cau­
tious in determining and correcting for any
bias introduced because the water is not
perfectly pure. He will be concerned, for
example, with the magnitude of the "blank"
in relation to the lower limit of detection o f
the method. He will determine the range o f
deviation i n the "blank" to be expected
from random error, and design his analysis
so that the "blank" is a proper correction on
the quantity he finds in the air sample.
Above all, he will start with reagent water
that is o f a purity consistent with the pro­
jected testing and with its practical utility.*
Approved with modifications from 2nd edition
Subcommittee 9
J . N . H A R M A N , Chairman
8. Common Acid, Alkali, and
Other Standard Solutions
8.1 I N T R O D U C T I O N . Reagent grade
chemicals shall be used in preparing and
standardizing all solutions. Reagents shall
conform to the current specifications of the
Committee on Analytical Reagents o f the
American Chemical Society, where such
specifications are available.** Other grades
may be used, provided it is first ascertained
that the reagent is of sufficiently high pu-
*For a detailed discussion of the various
grades o f reagent water, see A S T M D-1193.
**"Reagent Chemicals. American Chemical
Society Specifications," Amer. Chemical Soc,
Washington, D . C .

rity to permit its use without lessening the
accuracy o f the determination.
The National Bureau o f Standards offers
for sale certified standard samples o f arse­
nic trioxide, benzoic acid, potassium hy­
drogen phthalate, potassium dichrornate,
and sodium oxalate. These samples o f com­
mercially available primary standards are
to be used i n standardizing the volumetric
solutions.
Directions are given for the preparation
of the most commonly used concentrations
of the standard volumetric solutions.
Stronger or weaker solutions are prepared
and standardized i n the same general man­
ner as described, using proportionate
amounts o f the reagents. Similarly, i f quan­
tities larger than 1 liter are to be prepared,
proportionate amounts o f the reagents
should be used.
When quantities o f solution larger than 1
or 2 liters are prepared, special problems
are encountered i n being sure that they are
well mixed before being standardized.
While blade stirrers with glass or metal
shafts are suitable for many solutions, they
are not suitable i n every case. I n those cases
where contact of a glass or metal stirrer
with the solution would be undesirable it
may be possible to use a sealed polyolefin
coated stirrer. I n those cases where only
contact o f the solution with metal must be
avoided, the solution can be mixed by i n ­
serting a fritted glass gas dispersion tube to
the bottom o f the container and bubbling
nitrogen through the solution for 1 or 2 h .
I n order to make a solution o f exact nor­
mality from a chemical that cannot be mea­
sured as a primary standard, a relatively
concentrated stock solution may first be
prepared and standardized, and then an ex­
act dilution o f this may be made to the de­
sired strength. Another method is to make
a solution o f slightly higher concentration
than desired, standardize, and then make
suitable adjustments i n the concentration.
Alternatively, the solution may be used as
first standardized, with appropriate modi­
fication o f the factor used i n the calcula­
tion. This alternative procedure is espe­
cially useful i n the case o f a solution that
slowly changes strength—for example,
thiosulfate solution, which must be restan-
dardized at frequent intervals. Often, how­
ever, adjustment to the exact normality
COMMON STANDARD S O L U T I O N S 49
specified is desirable when a laboratory
runs a large number o f determinations with
one standard solution, since this simplified
the final calculation.
As long as the normality o f a standard
solution does not result i n a titration vol­
ume so small as to preclude accurate mea­
surement or so large as to cause abnormal
dilution o f the reaction mixture, and as
long as the solution is properly standard­
ized and the calculations are properly
made, the determinations can be considered
to be i n accord with the instructions i n this
manual.
I f a solution o f exact normality is to be
prepared by dissolving a weighed amount
of a primary standard or by dilution o f a
stronger solution, it is necessary that the
solution be brought to exact volume i n a
volumetric flask.
The stock and standard solutions pre­
scribed for the colorimetric determinations
in the chemical sections o f this manual
should also be accurately prepared i n volu­
metric flasks. Where the concentration
does not need to be exact, it is often easier
to mix the concentrated solution or the
solid with measured amounts o f water, us­
ing graduated cylinders for these measure­
ments. There is usually a significant change
of volume when strong solutions are mixed,
so that the total volume is less than the sum
of the volumes used. For approximate dilu­
tions, the volume changes are negligible
when concentrations o f 6 N or less are d i ­
luted.
Very thorough and complete mixing is es­
sential when making dilutions. One o f the
commonest sources o f error i n analyses us­
ing standard solutions diluted i n volumetric
flasks is failure to attain complete mixing.
Glass containers are suitable for the stor­
age o f most o f the standard solutions, al­
though the use o f polyolefin containers is
recommended for alkali solutions.
When large quantities o f solutions are
prepared and standardized, it is necessary
to provide protection against changes i n
normality due to absorption o f gases or wa­
ter vapor from the laboratory air. As vol­
umes o f solution are withdrawn from the
container, the replacement air should be
passed through a drying tube filled with
equal parts o f 8 to 20-mesh soda lime, ox­
alic acid, and 4 to 8-mesh anhydrous cal-
50 GENERAL TECHNIQUES

cium chloride, each product being sepa­ The weights o f dried K H C H 0 suitable
8 4 4

rated from the other by a glass wool plug. for other normalities o f N a O H solution are
8.2 S O D I U M H Y D R O X I D E , 0.02 T O 1.0 N . given in Table 8:11.
8.2.1 Preparation of 50% NaOH Solu­ 8.2.3 Calculation. Calculate the nor­
tion and of other Standard NaOH Solu­ mality o f the N a O H solution, as follows:
tions. Dissolve 162 g o f sodium hydroxide
(NaOH) in 150 m L o f carbon dioxide-free
water. Cool the solution to 25 °C and filter 0.20423 x C
through a Gooch crucible, hardened filter where A = normality o f the N a O H solu­
paper, or other suitable medium. Alterna­ tion,
tively, commercial 50% N a O H solution B = weight o f K H C H 0 used, g,
8 4 4

may be used. and,

To prepare a 0.1 N solution, dilute 5.45
m L o f the clear solution to 1 L with carbon
dioxide-free water, mix well, and store in a
tight polyolefin container.
For other normalities o f N a O H solution,
use the requirements given in Table 8:1.
8.2.2 Standardization. Crush 10 to
20 g o f primary standard potassium hydro­
gen phthalatet ( K H C H 0 ) to 100-mesh
8 4
C = volume o f N a O H solution
consumed, m L .
8.2.4 Stability. The use o f polyolefin
containers eliminates some o f the difficul­
ties attendant upon the use o f glass contain­
ers, and their use is recommended. Should
glass containers be used, the solution must
be standardized frequently i f there is evi­
4

fineness, and dry i n an open glass container dence o f action on the glass container, or i f
at 120°C for 2 h. Stopper the container and insoluble matter appears in the solution.
cool in a desiccator. 8.3 HYDROCHLORIC ACID. 0.02 TO
To standardize a 0.1 N solution, weigh 1.0 N .
accurately 0.95 ± 0.05 g o f the dried 8.3.1 Preparation. To prepare a 0.1 N
K H C H 0 , and transfer to a 500-mL coni­ solution, measure 8.3 m L of concentrated
8 4 4

cal flask. A d d 100 m L o f carbon dioxide-
free water, stir gently to dissolve the sam­
ple, add 3 drops o f a 1.0% solution o f
phenolphthalein in 95% ethanol and titrate
with N a O H solution to the appearance o f a
very faint pink color that persists.
Table 8:1 Sodium Hydroxide Dilution
Requirements
hydrochloric acid (HC1, sp gr 1.19) into a
graduated cylinder and transfer it to a 1-L
volumetric flask. Dilute to the mark with
water, mix well, and store in a tightly closed
glass container.
For other normalities o f HC1 solution,
use the requirements given in Table 8:111.
8.3.2 Standardization. Transfer 2 to
4 g o f anhydrous sodium carbonate
( N a C 0 ) to a platinum dish or crucible,
2 3

Volume o f 50% and dry at 250 C for 4 h. Cool in a desicca­

N a O H Solu-
tor.
Grams o f tion (25°C)
N a O H required required per L
To standardize a 0.1 N solution, weigh
Desired per L of Solu­ accurately 0.22 ± 0.01 g o f the dried
normality of solution tion, m L
0.02 0.8 1.09 Table 8:11 Weights of Dried Potassium
0.04 1.6 2.18 Hydrogen Phthalte
0.05 2.0 2.73
Weight o f dried
0.1 4.0 5.45
Normality KHC H 0
0.2 8.0 10.90 8 4 4

of Solution to be used, g*
0.25 10.0 13.63
0.5 20.0 27.25 0.02 0.19 ±0.005
1.0 40.0 54.54 0.04 0.38 ±0.005
0.05 0.47 ±0.005
fA primary standard grade o f potassium 0.1 0.95 ±0.005
hydrogen phthalate ( K H C H 0 ) is available 0.2 1.90 ±0.05
8 4 4

from the Office o f Standard Reference Materi­ 0.25 2.35 ±0.05

als, National Bureau o f Standards, Washington, 0.5 4.75 ±0.05
D . C . 20234. 1.0 9.00 ±0.05
 COMMON S T A N D A R D S O L U T I O N S 51

Table 8:111 Hydrochloric Acid Dilution C = volume of HC1 solution con­

Requirements sumed, m L .
Volume o f HC1
Desired to be diluted 8.3.4 Stability. Restandardize
normality to 1 L , m L monthly.
0.02 1.66 8.4 S U L F U R I C A C I D . 0.02 T O 1.0 N.
0.04 3.32 8.4.1 Preparation. To prepare a 0.1 N
0.1 8.3 solution, measure 3.0 m L o f concentrated
0.2 16.6 sulfuric acid ( H S 0 , sp gr 1.84) into a
2 4

0.5 41.5 graduated cylinder and slowly add it to one

1.0 83.0
N a C 0 , and transfer to a 500-mL conical
2 3
flask. A d d 50 m L o f water, swirl to dis­
solve the carbonate, and add 2 drops o f a
0 . 1 % solution o f methyl red i n 95% etha-
nol. Titrate with the HC1 solution to the
first appearance o f a red color, and boil the
solution carefully, to avoid loss, until the
color is discharged. Cool to room tempera­
ture, and continue the titration, alternating
the addition o f HC1 solution and the boil­
half the desired volume o f water in a 600¬
m L beaker. Rinse the cylinder into the
beaker with water. M i x the acid-water mix­
ture, allow it to cool, and transfer to a 1-L
volumetric flask. Dilute to the mark with
water, mix well, and store in a tightly closed
glass container.
For other normalities o f the H S 0 solu­ 2 4
t i o n , use the requirements given i n
Table 8:V.
8.4.2 Standardization. Transfer 2 to 4 g
of anhydrous sodium carbonate ( N a C 0 )
2 3

to a platinum dish or crucible, and dry at

ing and cooling to the first appearance o f a
faint red color that is not discharged on
further heating.
The weights o f dried N a C 0 suitable for
2 3
other normalities o f HC1 solution are given
in Table 8: IV.
250° C for 4 h . Cool in a desiccator.
For standardization o f a 0.1 N solution,
weigh accurately 0.22 ± 0.01 g of the dried
N a C 0 and transfer to a 500-mL conical
2 3
flask. A d d 50 m L o f water, swirl to dis­
solve the N a C 0 , and add 2 drops o f a
2 3

8.3.3 Calculation. Calculate the nor­ 0 . 1 % solution o f methyl red in 95% etha-
mality o f the HC1 solution, as follows: nol. Titrate with the H S 0 solution to the
2 4

first appearance o f a red color, and boil the

solution carefully, to avoid loss, until the
0.0530 x C
color is discharged. Cool to room tempera­
where A = normality o f the HC1 solu­ ture and continue the titration, alternating
tion, the addition o f H S 0 solution and the
2 4

B = weight o f N a C 0 used, g,
2 3
and
Table 8:IV Weights of Dried Sodium
boiling and cooling, to the first appearance
of a faint red color that is not discharged on
further heating.
The weights o f dried N a C 0 suitable for
2 3

Carbonate other normalities o f H S 0 solution are 2 4

given in Table 8:IV.

Weight o f dried
Normality N a C 0 to be
2 3

of Solution used, g
0.02 0.088 ± 0 . 0 0 1 *
0.04 0.176 ± 0 . 0 0 1 * Table 8:V Sulfuric Acid Dilution Requirements
0.1 0.22 ± 0 . 0 1 f
0.2 0.44 ± 0 . 0 1 f Volume o f H S 0 2 4

0.5 Desired to be diluted

1.10 ± 0 . 0 1 f
1.0 normality to 1 L , m L
2.20 ± 0 . 0 1 f
* A 100-mL buret should be used for this stan­ 0.02 0.60
dardization. 0.1 3.0
fThe listed weights are for use when a 50-mL 0.2 6.0
buret is used. I f a 100-mL buret is to be used, the 0.5 15.0
weights should be doubled. 1.0 30.0
52 GENERAL TECHNIQUES
8.4.3 Calculation. Calculate the nor­
mality o f the H S 0 solution, as follows:
2 4
0.0530 x C
where A = normality o f the H S 0 solu­
tion,
B = weight o f N a C 0 , g, and
2 3
C = volume of H S 0 solution
2 4
consumed, m L .
8.4.4 Stability. Restandardize
monthly.
8.5 I O D I N E (0.1 N).
8.5.1 Preparation. Transfer 12.7 g of
iodine and 60 g o f potassium iodide ( K I ) to
an 800-mL beaker, add 30 m L o f water,
and stir until solution is complete. Dilute
with water to 500 m L , and filter through a
sintered-glass filter. Wash the filter with
about 15 m L o f water, transfer the com­
bined filtrate and washing to a 1-L volu­
metric flask, dilute to the mark with water,
and mix. Store the solution in a glass-
stoppered, amber-glass bottle in a cool
place.
8.5.2 Standardization. Transfer 1 g of
arsenic trioxide ( A s 0 ) to a platinum dish,
2 3
and dry at 105°C for 1 h. Cool i n a desicca­
tor. Weigh accurately 0.20 ± 0.01 g o f the
dried A s 0 and transfer to a 500-mL coni­
2 3
cal flask. A d d 10 m L o f sodium hydroxide
solution ( N a O H , 40 g / L ) , and swirl to dis­
solve. When solution is complete, add 100
m L o f water and 10 m L o f sulfuric acid
( H S 0 , 1:35), and mix. Slowly add sodium
2 4
bicarbonate ( N a H C 0 ) until effervescence
3 B = weight o f K C r 0 used, g,
ceases, add 2 g o f N a H C 0 in excess, and
3 and
stir until dissolved. A d d 2 m L of starch so­
lution (10 g / L ) and titrate with the iodine
solution to the first permanent blue color.
8.5.3 Calculation. Calculate the nor­
mality o f the iodine solution, as follows:
0.040455 x C
where A = normality o f the iodine solu­
tion,
B = weight o f A s 0 used, g, and
2 3
C = volume o f iodine solution re­
quired for titration o f the so­
lution, m L .
8.5.4 Stability. Restandardize sealed bot­
tles monthly. Restandardize open bottles
weekly.
8.6 S O D I U M T H I O S U L F A T E (0.1 N).
8.6.1 Preparation. Dissolve 25 g o f
sodium thiosulfate pentahydrate ( N a S 0 • 2 2 3
5H 0) in 500 m L o f freshly boiled and
2
cooled water, and add 0.11 g of sodium
2 4
carbonate ( N a C 0 ) . Dilute to 1 L with
2 3
freshly boiled and cooled water, and let
stand for 24 h . Store the solution in a
tightly-closed glass bottle.
8.6.2 Standardization. Pulverize 2 g
of potassium dichromate ( K C r 0 ) , trans­
2 2 7
fer to a platinum dish, and dry at 120°C for
4 h . Cool i n a desiccator. Weigh accurately
0.21 ± 0.01 g o f the dried K C r 0 , and
2 2 7
transfer to a 500-mL glass-stoppered coni­
cal flask. A d d 100 m L o f water, swirl to
dissolve, remove the stopper, and quickly
add 3 g o f potassium iodide ( K I ) , 2 g o f
sodium bicarbonate ( N a H C 0 ) , and 5 m L
3
of hydrochloric acid (HC1). Stopper the
flask quickly, swirl to ensure mixing, and
let stand in the dark for 10 min. Rinse the
stopper and inner walls o f the flask with
water and titrate with the N a S 0 solution
2 2 3
until the solution has only a faint yellow
color. A d d 2 m L o f starch solution (10 g /
L ) , and continue the titration to the disap­
pearance o f the blue color.
8.6.3 Calculation. Calculate the nor­
mality o f the N a S 0 solution, as follows:
2 2 3
0.04904 x C
where A = normality of the N a S 0 so­ 2 2 3
lution,
2 2 7
C = volume of N a S 0 solution 2 2 3
required for titration o f the
solution, m L .
8.6.4 Stability. Restandardize weekly.
8.7 A M M O N I U M T H I O C Y A N A T E (0.1 N).
8.7.1 Preparation. Transfer 7.8 g o f
ammonium thiocyanate (NH SCN) to a 4
flask, add 100 m L o f water, and swirl to
dissolve the N H S C N . When solution is
4
complete, filter through a Cooch crucible,
hardened filter paper, or other suitable me­
dium. Dilute the clear filtrate to 1 L with
water and mix. Store the solution in a
tightly-stoppered glass bottle.
8.7.2 Standardization. Measure accu­
rately about 40 m L o f freshly standardized
0.1 N silver nitrate ( A g N 0 ) , and transfer
3

to a 250-mL conical flask. A d d 50 m L of
water, swirl to mix the solution, and add
2 m L o f nitric acid ( H N 0 ) and 1 m L of
ferric ammonium sulfate solution ( F e N H
( S 0 y i 2 H 0 , 80 g / L ) . Titrate the A g N 0
4 2
solution with the N H S C N solution until 4
the first permanent reddish-brown color
appears and persists after vigorous shaking
for 1 min.
8.7.3 Calculation. Calculate the nor­
mality of the N H S C N solution, as follows:
4
B x C
where A = normality o f the N H S C N so­
lution,
B = volume of A g N 0 solution
used, m L
C = normality o f the A g N 0 solu­
tion, and
D = volume of N H S C N solution
required for titration o f the
solution, m L .
8.7.4 Stability.
monthly.
8.8 P O T A S S I U M D I C H R O M A T E (0.1
8.8.1 Preparation. Transfer 6 g o f po­
tassium dichromate ( K C r 0 ) to a plati­
num dish and dry at 120°C for 4 h. Cool in
a desiccator. Place 4.9 g o f the dried
K C r 0 in a 1-L volumetric flask, and add
2 2 7 a volumetric pipet, 50 m L of the A g N 0
100 m L o f water. Swirl to dissolve and
when solution is complete, dilute to the
mark with water and mix. Store the solu­
tion in a glass-stoppered bottle.
8.8.2 Standardization. Place 40 m L of
water in a 250-mL glass-stoppered conical
flask, and add 40 m L , accurately mea­
sured, of the K C r 0 solution. Stopper the
2 2 7 A g N 0 with ammonium thiocyanate solu­
flask, swirl to mix, remove the stopper, and
add 3 g o f potassium iodide ( K I ) , 2 g o f
sodium bicarbonate ( N a H C 0 ) and 5 m L
of concentrated hydrochloric acid (HC1).
Stopper the flask quickly, swirl to ensure
mixing, and let stand in the dark for 10
min. Rinse the stopper and inner walls o f
the flask with water and titrate with freshly
standardized sodium thiosulfate solution
( N a S 0 ) until the solution has only a faint
2 2 3
yellow color. A d d 2 m L o f starch solution
(10 g / L ) , and continue the titration to the
disappearance o f the blue color.
8.8.3 Calculation. Calculate the nor­
mality o f the K C r 0 solution, as follows:
2 2 7 adequate ventilation.
COMMON S T A N D A R D S O L U T I O N S 53
B x C
3
A =
-^-
4
where A = normality of the K C r 0 so­ 2 2 7
3
lution,
B = volume of N a S 0 solution2 2 3
required for titration o f the
solution, m L
C = normality of the N a S 0 so­ 2 2 3
lution, and
D = volume o f K C r 0 solution
2 2 7
used, m L .
8.8.4 Stability. Restandardize
monthly.
4
8.9 S I L V E R N I T R A T E (0.1 N ) .
8.9.1 Preparation. Dry 17.5 g of silver
3
nitrate ( A g N 0 ) at 105°C for 1 h. Cool in a
3
desiccator. Transfer 16.99 g o f the dried
3
A g N 0 to a 1-L volumetric flask. A d d
3
500 m L o f water, swirl to dissolve the
4
A g N 0 , dilute to the mark with water, and
3
mix. Store the solution in a tightly-
stoppered amber-glass bottle.
8.9.2 Standardization. Transfer 0.3 g
Restandardize of sodium chloride (NaCl) to a platinum
dish and dry at 105°C for 2 h. Cool in a
N). desiccator. Weigh accurately 0.28 ± 0.01 g
of the dried NaCl and transfer to a 250-mL
2 2 7 glass-stoppered conical flask. A d d 25 m L
of water, swirl to dissolve the NaCl, and
add 2 m L o f nitric acid ( H N 0 ) . A d d , from
3
3
solution, mix, and add 1 m L o f ferric am­
monium sulfate solution ( F e N H ( S 0 ) 4 4 2
• 1 2 H 0 , 80 g / L ) and 5 m L of nitrobenzene
2
(CAUTIONX). Stopper the flask and shake
vigorously to coagulate the precipitate.
Rinse the stopper into the flask with a few
milliliters o f water and titrate the excess o f
3
tion ( N H S C N ) until the first permanent
4
reddish-brown color appears and persists
3 after vigorous shaking for 1 min. Designate
the volume o f N H S C N solution required
4
for the titration as Volume I .
J Caution —nitrobenzene, used in this section,
is extremely hazardous when absorbed through
the skin or when its vapor is inhaled and is cur­
rently considered a carcinogen. Such exposure
may cause cyanosis; prolonged exposure may
cause anemia. Do not get in eyes, on skin, or on
clothing. A v o i d breathing vapor. Use only with
54 GENERAL TECHNIQUES
Transfer 50 m L of the A g N 0 solution
to a clean, dry, 250-mL glass-stoppered
conical flask. A d d 25 m L o f water, 2 m L
of H N 0 , 1 m L o f F e N H ( S 0 ) . 1 2 H 0
3 4
solution (80g/L), stopper the flask, and
shake vigorously. Rinse the stopper into the
flask with a few milliliters o f water and t i ­
trate the A g N 0 solution with N H S C N so­
3 4
lution until the first permanent reddish-
brown color appears and persists after
vigorous shaking for 1 min. Designate the
volume o f N H S C N solution consumed as
4
Volume I I .
Measure accurately, from either a buret
or a volumetric pipet, 2.0 m L of the
A g N 0 solution, designate the exact vol­
3 8.10.2 Standardization. Transfer 2 g
ume as Volume I I I , and transfer to a 100¬
m L glass-stoppered conical flask. A d d
25 m L o f water, 2 m L o f N H 0 , 1 m L o f
F e N H ( S 0 ) .12 H 0 solution (80 g / L ) ,
4 4 2 2
and 5 m L o f nitrobenzene, stopper the
flask, and shake vigorously. Rinse the stop­
per into the flask with a few milliliters o f
water and titrate the A g N 0 solution with
N H S C N solution until the first permanent
4 solved. A d d 39 m L o f the K M n 0 solution
reddish-brown color appears and persists
after vigorous shaking for 1 m i n . Designate
the volume of N H S C N solution consumed
4 until the pink color disappears. Heat the
as Volume IV. §
8.9.3 Calculation. Calculate the nor­
mality of the A g N 0 solution, as follows:
3 pink color persists for 30 s. A d d the final
0.05845 x (C - D)
where A = normality o f the A g N 0 solu­
tion,
B = weight o f NaCl used,
C = volume o f A g N 0 solution
consumed by the total chlo­
ride, m L = 50 - (Volume I x
50/Volume I I ) , and
D = volume o f A g N 0 solution
consumed by any chloride
ion in the nitrobenzene, m L
= Volume I I I - (Volume I V
x 50/Volume I I )
8.9.4 Stability.
monthly.
§The ammonium thiocyanate titrant used i n
the three titrations must be from the same, well-
mixed solution. The nitrobenzene used in each
titration must also be from the same, well-mixed
container.
3 8.10 POTASSIUM PERMANGANATE (0.1
N).
8.10.1 Preparation. Dissolve 3.2 g o f
4 2 2
potassium permanganate ( K M n 0 ) i n 4
100 m L o f water, and dilute the solution
with water to 1 L . A l l o w the solution to
stand in the dark for two weeks and then
filter through a fine-porosity sintered-glass
crucible. Do not wash the filter. Store the
solution in glass-stoppered, amber-colored
glass bottles.
N O T E 1. Do not permit the filtered solu­
tion to come into contact with paper, rub­
ber, or other organic material.
of sodium oxalate ( N a C 0 ) to a platinum 2 2 4
dish and dry at 105°C for 1 h . Cool in a
3 desiccator. Weigh accurately 0.30 ± 0.01 g
of the dried N a C 0 and transfer to a 500-
2 2 4
m L glass container. A d d 250 m L of sulfuric
acid ( H S 0 , 1:19) that was previously
2 4
boiled for 10 to 15 min and then cooled to
3 27 ± 3 ° C , and stir until the sample is dis­
4
at a rate o f 30 ± 5 m L per min. while stir­
ring slowly, and let stand for about 45 s
solution to 6 0 ° C , and complete the titration
by adding K M n 0 solution until a faint
4
0.5 to 1.0 m L dropwise, and give the solu­
tion time to decolorize before adding the
next drop.
3
Carry out a blank determination on a
second 250-mL portion o f the H S 0 2 4
(1:19), and make sure that the pink color at
3
the end point matches that o f the standard­
ization solution. Correct the sample titra­
tion volume as shown to be necessary.
N O T E 2. The specified 0.30 g sample of
3
N a C 0 should consume about 44.8 m L o f
2 2 4
0.1 N K M n 0 . 4
NOTE 3. I f the pink color o f the solution
persists more than 45 s after the addition o f
the first 39 m L of K M n 0 solution is com­ 4
Restandardize plete, discard the solution and start over
with a fresh solution o f the N a C 0 , but 2 2 4
add less o f the K M n 0 solution. 4
N O T E 4. The blank correction usually
amounts to 0.03 to 0.05 m L .
8.10.3 Calculation. Calculate the nor­
mality of the K M n 0 solution as follows: 4
 R E C O V E R Y AND INTERNAL STANDARD P R O C E D U R E S
0.06701 x C
where A = normality o f the K M n 0 so­ 4 sought substance in the sample alone is de­
lution,
B = weight o f N a C 0 used, g,
2 2 4 from each o f the determinations consisting
and
C = volume o f K M n 0 solution 4
required for titration o f the
solution, m L .
8.10.4 Stability. Restandardize
weekly.
Approved with modifications from 2nd edition
Subcommittee 9
J . N . HARM AN, Chairman
9. Recovery and Internal Standard
Procedures
9.1 R E C O V E R Y P R O C E D U R E . A recovery
procedure does not enable the analyst to
apply any correction factor to the results o f
an analysis; however, it does give him some
basis for judging the applicability o f a par­
ticular method o f analysis to a particular
sample.
A recovery determination may be per­
formed at the same time as the determina­
tion itself. O f course, recoveries would not
be run on a routine basis with samples
whose general composition is known or
when using a method whose applicability to
the sample is well established. Recovery
methods are to be regarded as tools to re­
move doubt about the applicability o f a
method to a sample.
In brief, the recovery procedure involves
applying the analytical method to a reagent
blank, to a series o f known standards cov­
ering the expected range o f concentration
of the sample, to the sample itself, in at
least duplicate, and to the recovery sam­
ples, prepared by adding known quantities
of the substance sought to separate por­
tions o f the sample itself, each portion
equal to the size o f sample taken for the
run. The substance sought should be added
in sufficient quantity to overcome the limits
of error o f the analytical method, but with­
out causing the total in the sample to exceed
the range o f the known standards used.
The results are first corrected by sub­
tracting the reagent blank from each o f the
55
other determined values. The resulting
known standards are then graphically rep­
resented. From this graph the amount o f
termined. This value is then subtracted
of sample plus known added substance.
The resulting amount o f substance divided
by the known amount originally added and
multiplied by 100 gives the percentage re­
covery.
The procedure outlined above may be ap­
plied to a colorimetric, flame photometric,
or fluorimetric analysis. It may also be ap­
plied in a little more simple form to titri-
metric, gravimetric, and other types o f
analyses.
Rigid rules concerning the percentage re­
coveries required for acceptance o f results
of analyses for a given sample and method
cannot be stipulated. Recoveries o f sub­
stances in the range o f the sensitivity o f the
method may, o f course, be very high or
very low and approach a value nearer to
100 per cent recovery as the error o f the
method becomes small with respect to the
magnitude o f the amount o f substance
added. I n general, intricate and exacting
procedures for trace substances that have
inherent errors due to their complexity may
give recoveries that would be considered
very poor and yet, from the practical view­
point o f usefulness o f the result, may be
quite acceptable. Poor results may reflect
either interferences present in the sample or
real inadequacy o f the method o f analysis
in the range in which it is being used.
It must be stressed, however, that the j u ­
dicious use o f recovery determinations for
the evaluation o f analytical procedures and
their applicability to particular samples is
an invaluable aid to the analyst in both rou­
tine and research investigations.
9.2 INTERNAL STANDARD PROCEDURE.
The internal standard technique is used pri­
marily for emission spectrograph and po-
larographic procedures. This procedure
enables the analyst to compensate for elec­
tronic and mechanical fluctuations within
the instrument.
In brief, the internal standard method i n ­
volves the addition to the sample o f known
amounts o f a substance that will respond to
the instrument in a manner similar to the
contaminant being analyzed. The ratio o f
56 GENERAL TECHNIQUES
the measurement o f the internal standard to
the measurement o f the contaminant is the
value used to determine concentration o f
contaminant present in the sample. A n y
changes in conditions during analyses will
affect the internal standard and the con­
taminant the same and so will compensate
for such changes. The internal standard
should be o f similar chemical reactivity to
the contaminant, approximately the same
concentration as anticipated for the con­
taminant, and as pure a substance as possi­
ble.
Approved with modifications from 2nd edition
Subcommittee 9
J . N . H A R M AN, Chairman
10. Interferences
10.1 INTRODUCTION. The term interfer­
ence covers various effects o f dissolved or
suspended substances upon analytical pro­
cedures. The known interferences and re­
lated information have been included i n the
specific methods o f analysis. Many o f the
methods in this manual have been devel­
oped specifically to minimize common i n ­
terferences. The purpose o f this section is
to offer guidelines for detecting the pres­
ence o f interfering substances and over­
coming their effects when encountered.
In analytical chemistry, one must be pre­
pared to use an alternate procedure i f one
method would be less affected than another
due to the presence o f the interfering sub­
stance. The most expedient method, how­
ever, for overcoming a suspected interfer­
ence is often the use o f a smaller initial
aliquot. Thus, the effect o f the interfering
element is diminished or eliminated
through dilution o f the original sample.
The level o f the parameter being measured
is likewise reduced, so care must be taken to
insure an aliquot containing more than the
minimum detectable concentration. I f the
data display a consistently increasing or de­
creasing pattern with dilution, interference
is indicated. Dilution o f final reaction vol­
ume must never be used i n overcoming i n ­
terferences. Certain reagents are prepared
to react with a limited amount o f the test
element. I f this limit is exceeded, a portion
of the test element goes undetected. Then
the analyst must repeat the entire test pro­
cedure with a smaller initial aliquot.
10.2 INTERFERENCE REACTIONS. Essen­
tially, three reactions may occur to produce
an interference:
a. A n interfering substance may react
with the reagents i n the same manner as the
element being sought. (Positive interfer­
ence);
b. an interfering substance may react
with the element being sought to prevent
complete isolation. (Negative interference);
or
c. an interfering substance may combine
with the reagents to prevent further reac­
tion with the element being sought. (Nega­
tive interference).
These reactions will produce either high
or low results. A n estimate o f the magni­
tude o f an interference may be obtained by
the internal standard procedure. After es­
tablishing a calibration graph for the ap­
propriate range and performing an initial
analysis o f the sample, this procedure may
be employed.
10.3 I N T E R N A L S T A N D A R D P R O C E D U R E .
10.3.1 Select a standard which, when
added to the unknown sample, will produce
a final result approximately midrange o f
the standard calibration graph. Example:
Calibration graph: 0 to 100 ppb.
Sample analysis: 10 ppb.
Select a standard concentration, which
when added to 10 ppb, will produce a
total concentration o f 50 ppb; i.e., 40
ppb.
10.3.2 Analyze the prepared solution
(sample + standard) and read the result
from the graph.
10.3.3 Subtract from this result the
original sample analysis.
10.3.4 The difference divided by the
amount originally added and multiplied by
100 gives the percentage recovery. Exam­
ple:
Sample: 10 ppb
Sample + Standard:
10 ppb + 40 ppb = 50 ppb
Sample + Standard as determined:
56 ppb

Subtract original determination:
56 - 10 = 46 ppb
Recovery equals:
46
— x 100 = 115 percent
Thus, a positive interference is present. I f
the result is below 100%, a negative inter­
ference is indicated. By establishing the di­
rection of interference, the nature of the
reaction taking place may be determined
and appropriate steps taken for eliminating
the effect. A positive error indicates the in­
terference is reacting like the test element,
and the possibility of isolation by a p H ad­
justment should be considered. This proce­
dure should be applied i n triplicate to both
the sample and the sample plus standard
and must include a reagent blank.
10.4 R E M O V A L O F I N T E R F E R E N C E . I f fur­
ther treatment o f the sample for removal o f
the interfering substance is necessary, sev­
eral approaches are available to the analyst.
These may be physical or chemical or a
combination o f both, such as:
a. Distillation o f the sample leaving the
interference behind;
b. removal of the interference by ion ex­
change resins;
c. addition o f complexing agents;
d. extraction into organic solvents;
e. ashing;
f. p H adjustment;
g. different reaction rate;
h. different temperature.
These treatment methods serve as exam­
ples only, since a thorough review o f the
literature must be undertaken before pro­
ceeding to undertake a change i n an estab­
lished method.
10.5 SOURCES OF I N T E R F E R E N C E . In
preceding discussion, the analyst was given
instructions in detecting an interference i n
quantitative analysis. The question of the
source of the interference must be identi­
fied and corrected, i f possible. The origin
of an interference may be found in the fol­
low situations:
a. Present at the sampling site;
b. imparted during sample collection;
c. developed in sample storage; or
d. imparted or developed in laboratory
analysis.
INTERFERENCES 57
These four source situations may normally
be avoided in applying proper sampling and
laboratory techniques. These techniques
are discussed i n this section.
The sources may be a result of physical,
chemical, or biological phenomena. The
possibility of interaction (physical upon
chemical, etc.) should not be overlooked.
These three classifications o f sources o f er­
rors are further subdivided to assist the an­
alyst i n search of the action producing the
effect.
10.6 PHYSICAL SOURCES.
10.6.1 Heat.
a. Effect on chemical equilibria.
b. Promote side reactions at elevated
temperatures — decomposition.
c. Affect rate o f reaction — temperature
coefficient correction. Serious errors may
be encountered at low temperatures.
10.6.2 Light.
a. Effect o f visible spectrum — accelerate
photooxidation.
b. Effect o f U V - y e l l o w i n g o f K I
reagent — positive error.
c. Fading of detector reagent colors
(lead and mercury dithizonates) — negative
error.
10.6.3 Humidity.
a. Reaction of H 0 with contaminant —
2
toluene diisocyanate, phosgene.
b. Reaction of H 0 with the collecting
2
substrate.
c. Dilution o f the collecting substrate.
10.6.4 Time.
a. Deterioration rate after sample collec­
tion: half-life must be determined.
b. Fading of the developed color after
maximum density attainment.
the
c. Reagents often deteriorate with time.
10.6.5 Chemical Contamination.
a. Contamination o f sampling equipment
during use.
(1) Collection in high concentration
locations — tetraethyl lead.
(2) Extraneous dust and dirt.
(3) Failure to keep collection equip­
ment and reagents stoppered.
(4) Cross contamination in recharging
collector or changing filters, especially in
the field.
58 GENERAL TECHNIQUES

10.7 C H E M I C A L S O U R C E S . c. Failure to react in absence of catalyst.

10.7.1 pH Control in Aqueous Sys­
tems.
a. p H sensitive collection.
b. p H sensitive color responses.
c. Buffered versus unbuffered solutions.
10.7.2 Chemical Contamination.
a. Impure reagents — Sulfates in gum ara­
d. Surface reactions — porous glass bub­
blers.
e. Inhibition o f reactions.
10.7.6 Half-life in the Ambient Atmos­
phere.
a. Effect o f light, H 0 vapor, and coex­
2

isting contaminants.
ble used to stabilize B a S 0 colloid.4

b. Failure to run blanks and controls. (1) tetraethyl lead

(2) toluene diisocyanate
10.7.3 Interferences.
(3) C O C l , phosgene
2

a. Negative — N H in H g detector tube —

3 (4) Nitrogen Oxides
redox cancellations. (5) Carcinogenic Hydrocarbons*
b. Positive
(6) Polymerization
(1) Different shade or color
(7) Carbon Monoxide
produced — Dithizone + oxidizing agents.
(2) Reaction the same, increased by 10.8 B I O L O G I C A L S O U R C E S .
interference —Ozone —nitrogen oxides sys­
10.8.1 Algal Growth.
tem.
(3) Electrolytes in conductivity and p H 10.8.2 Insects.
meters. 10.8.3 Waste Products (Animal and
10.7.4 Sensitivity. Insects).
a. Concentration effects. 10.9 Procedures used herein are usually
(1) Adjust to obtain m a x i m u m well established and, when applied to the
effect —maximum color density. Reactant selected sample, give predictable results. I f
ratios often nonstoichiometric. a result is suspected to be in error, samples
(2) Beer's law may not be followed at of known composition should be analyzed
low concentrations. by the method in use and an alternate
(3) Background "noise" in electrolytic method. I f a method fails on the unknown
instruments. sample while giving correct results on a
known, an interference is indicated. The
(4) Effect of reagent concentrations on
best way to solve a problem o f this kind is
reaction rate.
by reviewing the literature. Many hours
(5) Effect of reagent concentrations on
may be saved through a few hours o f dili­
reaction equilibrium — C H 0 + fuchsin re­
gent library searching. Solving such prob­
2

agent.
lems depends on the skill and ingenuity of
(6) Attempting to operate outside the
the investigator.
optimum range for which the method was
developed.
Approved with modifications from 2nd edition
b. Color former structure versus extinc­
tion coefficient —Phenanthroline series for Subcommittee 9
iron detection. J . N . H A R M AN, Chairman
c. Length-of-stain tubes.
(1) Attempts to increase sensitivity by
repeated sample aliquots —benzene tube re­
quires recalibration.
(2) Change in sampling rate without
recalibration may produce serious errors.
10.7.5 Catalytic Effects.
a. Decomposition on contact with tubes
and containers — ozone on metals — H 0 + 2 2

heavy metal. •Anthracene and naphthalene (polynuclear

b. Promotion o f side reactions. aromatic hydrocarbons), benzene, nitrobenzene.