Springer-Verlag Berlin Heidelberg GMBH

Springer-Verlag Berlin Heidelberg GmbH
Douglas E. Rawlings (Ed.)
Biomining:
Theory, Microbes and Industrial Processes
i Springer
Douglas E. Rawlings
Department of Microbiology
University of Cape Town
Rondebosch, 7700
South Africa
ISBN 978-3-662-06113-8
LIbrary of Congress Cataloging-in-Publication Data

Biomining: theory, microbes, and industrial processes I [edited by] Douglas E. Rawlings.
p.em. - (Biotechnology intelligence unit)
Includes bibliographical references and index.
ISBN 978-3-662-06113-8 ISBN 978-3-662-06111-4 (eBook)
DOI 10.1007/978-3-662-06111-4
1. Minerals-Biotechnology. 2. Bacterial leaching.
3. Bioremediation. 4. Extreme environments-Microbiology.
5. Thermophilic bacteria I. Rawlings, Douglas E. II. Series.
TN 688.3·B33B55 1997
622'.7-dC21 97-23624 ClP
This work is subject to copyright. All rights are reserved, whether the whole or part of the mate-
rial is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recita-
tion, broadcasting, reproduction on microfilm or in any other way, and storage in data banks_
Duplication of this publication or parts thereof is permitted only under the provisions of the
German Copyright Law of September 9, 1965, in its current version, and permission for use
must always be obtained from Springer-Verlag Berlin Heidelberg GmbH.
Violations are liable for prosecution under the German Copyright Law.
© Springer-Verlag Berlin Heidelberg 1997
Originally published by Springer-Verlag Berlin Heidelberg New York in 1997
Softcover reprint of the hardcover 1st edition 1997
The use of general descriptive names, registered names, trademarks, etc_ in this publication
does not imply, even in the absence of a specific statement, that such names are exempt from
the relevant protective laws and regulations and therefore free for general use.
Product liability: The publisher cannot guarantee the accuracy of any information about dosage
and application thereof contained in this book. In every individual case the user must check
such information by consulting the relevant literature.
Typesetting: Landes Bioscience Georgetown, TX, U.S.A.
SPIN:10634500 31/31ll - 5 4 3 2 1 0 -Printed on acid-free paper
=====PREFACE=====
T wo excellent books entitled Biohydrometallurgy (written by Giovanni
Rossi) and Microbial Mineral Recovery (edited by Henry Ehrlich and
Carole Brierley) were published in 1990 and this book has been written
to build on those. During the past decade there has been much re-
newed interest in biomining. Several new leaching/oxidation processes
have been developed and the number of sites at which such processes
operate has risen substantially. Relatively low-rate dump or heap leach-
ing processes have been supplemented by processes that employ high-
rate stirred tank reactors. In addition, the more traditional dump and
heap-leaching processes have been applied to ores and concentrate-
coated supports in ways not previously used. The stirred tank reactor-
based process built at the Ashanti goldfields (Ghana) almost certainly
represents the largest fermeter-based biotechnology process on earth.
The size and number of operating bioleaching/oxidation plants and the
broadening of application indicates that biomining has become part of
the main stream of biotechnology.
This book deals with the theory and application of bioleaching
and biooxidation technology and has been written by a combination of
people employed in industry and academia. It has been compiled to
provide a state-of-the-art description of several industrial bioleaching
processes, the theory that underpins those processes and a description
of the biology of the microorganisms involved. A major aim of the
book is an attempt to provide the interested industrialist and engineer
a starting point from which to further investigate bioleaching/
biooxidation technology. Presentations of the Gencor process using
mesophilic bacteria and the BacTech process using moderate thermo-
philic bacteria in stirred tank reactors illustrate some of the high-rate
technology. Up-to-date applications of heap leaching for copper (e.g.,
the Quebrada Blanca mine, Chile) and gold (Newmont Gold's Quarry
mine, Nevada) extraction, as well as the Geobiotics process for heap
leaching concentrate coated support rock are presented to illustrate
somewhat lower-rate and less expensive treatment processes.
Biomining technology has been built upon the sustained and out-
standing work of a number of individuals of whom there are too many
for each to have contributed a chapter to this book. The choice of chap-
ter authors does not reflect the contribution of individual people to
this field. Due to space constraints, the book provides a largely applied
overview of the subject and no attempt has been made to give an in-
depth coverage of engineering, bioleaching chemistry or the molecular
biology and biochemistry of leaching microorganisms.
For academics, the writing of book chapters is often viewed as a
nuisance because it takes time from writing the journal articles which
are the real "credentials" of scientific scholars. For the industrialists,
putting "pen to paper" can be an unwelcome distraction and frustrat-
ing to those who are out of practice. I wish to gratefully acknowledge
with thanks the sacrifice that the contribution of each of the authors
represents. I especially wish to thank Nikki Campbell for redrawing
some of the figures in such a professional way. My thanks also to many
of the staff at Landes Bioscience for their assistance and particularly to
Maureen Jablinske via whom the invitation to produce this book was
received and for her help in the early stages.
Douglas E. Rawlings
r;::::::=:==== CONTENTS ====::::::;-]
Section I - Overview
1. Mining Biotechnology:
Research to Commercial Development and Beyond ................ 3
Corale L. Brierley
Introduction ................................................................................. 3
Industrial Applications of BioleachinglBiooxidation .............. 3
Stirred-Tank Biooxidation ......................................................... 4
Bioheaps and Copper Dump Leaching ..................................... 8
In Situ Bioleaching ...................................................................... 9
Applications of Bioremediation in the Mining
Environment .......................................................................... 10
R&D Leading to Today's Commercial Operations ................. 11
New Developments .................................................................... 13
Summary .................................................................................... 16
Section II - Industrial Processes
2.. Bioleaching of Copper ................................................................ 2.1

Henry A. Schnell
Introduction ............................................................................... 2.1
Leaching Definitions ................................................................. 22
Copper Mineralogy .................................................................... 22
Physico-Chemical Leaching Variables .................................... 23
Bacterial Leaching Variables .................................................... 26
Nutrients .................................................................................... 27
Heap Operating Variables ........................................................ 29
Leach Solution Processing ........................................................ 32
Environmental Considerations ................................................ 36
Commercial Installations ......................................................... 36
Conclusions ................................................................................ 42
3. The BIOXe Process for Biooxidation of Gold-Bearing

Ores or Concentrates .................................................................. 45
David W. Dew, Ellen N. Lawson and Jennifer L. Broadhurst
Introduction ............................................................................... 45
The BIOX® Process Flowsheet and Plant
Operating Practice ................................................................ 46
The BIOXCI> Bacterial Culture ................................................... 54
Engineering Design and Process Requirements ..................... 58
Treatment and Disposal ofBIOX® Effluent Streams ............ 70
Conclusion ................................................................................. 77
4. The Design and Operating Practice of Bacterial Oxidation
Plant Using Moderate Thermophiles
(The BacTech Process) ................................................................ 81
Paul C. Miller
Introduction ............................................................................... 81
Current Reasons for Considering Bacterial Leaching ........... 82
Oxidation of Refractory Gold Concentrates-
Process Flowsheet and Economics ....................................... 83
Aspects of Reactor Design and Operation .............................. 87
Integration of Bacterial Oxidation with
Other Unit Operations .......................................................... 97
Trouble Shooting and Philosophy of Operation ................... 99
Future Applications and Conclusions ................................... 100
5. Heap Leaching of Gold-Bearing Deposits:

Theory and Operational Description ...................................... 103
James A. Brierley
Introduction ............................................................................. 103
Laboratory Scale Testing and Development ......................... 104
Concepts and Theory of the Biooxidation-Heap Process .... 104
Microbiology ............................................................................ 106
Applications ............................................................................. 107
Summary ................................................................................... 112
6. Biooxidation of Refractory Gold Ores

(The Geobiotics Process) ........................................................... 117
James L. Whitlock
Introduction .............................................................................. 117
Biooxidation of Refractory Gold Ores .................................... 117
The Geobiotics Process ............................................................ 119
Process Design Overview ......................................................... 119
Laboratory and Pilot Test Results .......................................... 122
Costs ......................................................................................... 126
Conclusions .............................................................................. 126
7. Technical Potential for Bioleaching and Biobeneficiation

of Ores to Recover Base Metals (Other than Iron
or Copper), Platinum-Group Metals and Silver .................... 129
Henry L. Ehrlich
Introduction ............................................................................. 129
Leaching with Autotrophic and Mixotrophic Bacteria ........ 130
Leaching with Heterotrophic Bacteria ................................... 138
Conclusion ............................................................................... 145
Section III - Process Fundamentals
8. Recent Developments in Modeling the Kinetics

of Bioleaching ............................................................................. 153
Geoffrey S. Hansford
Introduction .............................................................................. 153
Bacterial Ferrous Iron Oxidation ............................................ 153
Kinetics Models for Bioleaching ............................................ 160
Conclusions .............................................................................. 170
9. Physical Chemistry of Bacterial Leaching .............................. 177

Frank K. Crundwell
Introduction .............................................................................. 177
The Surface Chemistry of Bacterial Attachment
to Mineral Surfaces .............................................................. 178
Electrochemistry of Mineral Dissolution and Bacterial
Leaching ................................................................................ 186
Applications of Electrochemical Mechanism
of Leaching ........................................................................... 190
Incorporation of the Electrochemical Kinetics
into a Model of Bacterial Leaching of Sphalerite .............. 195
Conclusions .............................................................................. 198
10. Optimization ofBiooxidation Heaps ...................................... 201

A. Ian M. Ritchie
Introduction ............................................................................. 201
Basics ........................................................................................ 202
The Intrinsic Oxidation Rate ................................................. 206
Mathematical Simulation of Heap Performance .................. 210
Optimizing the Performance of an Open Based Heap ......... 212
Monitoring the Performance of a Biooxidation Heap ......... 219
Outstanding Issues .................................................................. 222
Section IV - Leaching Microorganisms
11. Mesophilic, Autotrophic Bioleaching Bacteria:

Description, Physiology and Role ........................................... 229
Douglas E. Rawlings
Introduction to Microorganisms ........................................... 229
Taxonomic Description of the Mesophilic
Bioleaching Bacteria ............................................................ 230
Phylogeny .................................................................................. 231
Nutrition and Energy .............................................................. 232
Mineral Oxidizing Ability ....................................................... 237
The Role of the Mesophilic Obligate Acidophiles
in Mineral Bioleaching ........................................................ 239
Other Physiological Characteristics ....................................... 239
Summary .................................................................................. 241
12. Thermophiles and Bioleaching ............................................... 247
Paul R. Norris
Introduction ............................................................................. 247
The Microorganisms ............................................................... 248
Metabolism and Molecular Biology of Thermophiles ......... 252
Mineral Sulfide Oxidation by Thermophiles ........................ 253
13. Heterotrophic Acidophiles and Their Roles

in the Bioleaching of Sulfide Minerals ................................... 259
D. Barrie Johnson and Francisco F. Roberto
Introduction ............................................................................. 259
Biodiversity of Acidophilic Heterotrophic Prokaryotes ...... 259
Ecology of Acidophilic Heterotrophic Prokaryotes ............ 264
Molecular Biology of Acidophilic
Heterotrophic Bacteria ........................................................ 270
Role of Acidophilic Heterotrophs in Mineral Dissolution .. 271
Conclusions .............................................................................. 275
14. Molecular Methods for the Identification

and Enumeration of Bioleaching Microorganisms .............. 281
Carlos A. Jerez
Introduction ............................................................................. 281
Bioleaching Microorganisms .................................................. 282
Some of the Classical Methods for the Determination
of Biomining Microorganisms ........................................... 282
Adherence of Microorganisms and Biofilm Formation ...... 282
Immunological Methods ......................................................... 284
Methods Involving Nucleic Acids .......................................... 289
Determination of Microorganisms In Situ ............................ 293
Conclusions .............................................................................. 294
Index ..................................................................................................... 299

r.======EDITOR======;-]
Douglas E. Rawlings
Department of Microbiology
University of Cape Town
Rondebosch, South Africa
Chapter 11
1=1============ CONTRIBUTORS =============J:j

Corale L. Brierley Geoffrey S. Hansford
Brierley Consultancy LLC Gold Fields Minerals
Hoghlands Ranch, Colorado, U.S.A. Bioprocessing Laboratory
Chapter 1 Department of Chemical
Engineering
James A. Brierley University of Cape Town
Newmont Metallurgical Services Rondebosch, South Africa
Englewood, Colorado, U.S.A. Chapter 8
Chapters
Carlos A. Jerez
Jennifer L. Broadhurst Departamento de Bioquimica
Gencor Process Research Facultad de Medicina
Laboratories Universidad de Chile
Randburg, South Africa Santiago, Chile
Chapter 3 Chapter 14
Frank K. Crundwell D. Barrie Johnson

School of Process and Materials School of Biological Sciences
Engineering University of Wales
University of the Witwatersrand Bangor, United Kingdom
Johannesburg, South Africa Chapter 13
Chapter 9
Ellen N. Lawson
DavidW.Dew Department of Microbiology
Gencor Process Research University of the Witwatersrand
Laboratories Johannesburg, South Africa
Randburg, South Africa Chapter 3
Chapter 3
Paul C. Miller
Henry L. Ehrlich BacTech (Australia) Ltd.
Department of Biology Belmont, Western Australia
Rensselaer Polytechnic Institute Chapter 4
Troy, New York, U.S.A.
Chapter 7
Paul R. Norris Henry A. Schnell
Department of Biological Sciences Compania Minera Quebrada Blanca
University of Warwick S.A.
Coventry, United Kingdom Santiago, Chile
Chapter 12 Chapter 2
A. Ian M. Ritchie James 1. Whitlock

Australian Nuclear Science and Geobiotics, Inc.
Technology Organization Spearfish, South Dakota, U.S.A.
Menai, New South Wales, Chapter 6
Australia
Chapter 10
Francisco F. Roberto
Idaho National Engineering
Laboratory
Idaho Falls, Idaho, U.S.A.
Chapter 13
SECTION I
Overview
CHAPTER 1
Mining Biotechnology:
Research to Commercial
Development and Beyond
Corale 1. Brierley
Introduction
T oday large-scale, bioleachinglbiooxidation facilities extract copper and enhance

gold recovery from ores and concentrates and commercially applied environ-
mental biotechnologies remediate metal-contaminated waters and degrade cya-
nide. These commercial operations convincingly demonstrate the qualities-ro-
bust nature, operational simplicity, health, safety and environmental benefits,
capital and operating cost advantages and improved performance-that have made
bioleachinglbiooxidation and bioremediation viable process options for the min-
ing industry. Commercial reality is a momentous milestone for scientists and en-
gineers who have labored some 30 years and extolled the virtues of biotechnology
for mining. It also represents a major shift in the industry's approach to mineral
processing and environmental management. With commercial success the mining
industry now anticipates better efficiency from existing biotechnical processes,
improvements in applications and novel processes to further enhance the utility
and extent of application. Bioprocessing innovations in various stages of research
and development are expected to fulfill this expectation.
This introductory chapter strives to:
Show the extent and scale of current commercial biotechnology applica-
tions in mineral processing and environmental management;
Outline industry's reasons for using bioprocessing;
• Overview research and development, to present day commercial applica-
tions of biotechnology in mining; and
Point to new developments that promise to significantly influence future
technical and economical applications of biotechnology in mining.
Industrial Applications of BioleachinglBiooxidation

Although the terms bioleaching and biooxidation are often used interchange-
ably, there are distinct technical differences between these process technologies.
Bioleaching refers to the use of bacteria, principally Thiobacillus ferrooxidans,
Biomining: Theory, Microbes and Industrial Processes,

edited by Douglas E. Rawlings. © Springer - Verlag and Landes Bioscience 1997.
4 Biomining: Theory, Microbes and Industrial Processes
Leptospirillum ferrooxidans and thermophilic species of Sulfobacillus, Acidianus

and Sulfolobus, to leach a metal of value such as copper, zinc, uranium, nickel and
cobalt from a sulfide mineral. Bioleaching places the metal values of interest in the
solution phase during oxidation. These solutions are handled for maximum metal
recovery and the solid residue is discarded. Minerals biooxidation refers to a pre-
treatment process that uses the same bacteria as bioleaching to catalyze the degra-
dation of mineral sulfides, usually pyrite or arsenopyrite, which host or occlude
gold, silver or both. Biooxidation leaves the metal values in the solid phase and the
solution is discarded.
Since 1986,11 commercial bioleaching/biooxidation plants have been commis-
sioned with nine in continuous operation today (Table 1.1). Not listed are copper
dump leaching facilities, which rely on bioleaching to extract copper from mil-
lions of tons of submarginal ore. Biooxidationlbioleaching applications can be cat-
egorized as aerated stirred-tanks, bioheaps, dump bioleaching, and in situ
bioleaching.
Stirred-Tank Biooxidation
Aerated, stirred-tank bioreactors, typically reserved for mineral concentrate
feeds because of the capital and operating costs of the systems, involve three or
more stages in series. The first stage has several tanks in parallel to allow longer
retention of the feed. Subsequent stages are usually single tanks in series (Fig. 1.1).
Tanks are rubber-lined or constructed of high -grade stainless steel because of the
corrosiveness of acidic ferric sulfate. Considerable heat is generated by the oxida-
tion of mineral sulfides. Therefore, tanks are equipped with cooling coils or water
jackets to maintain the tank contents at optimum temperature for the bacteria
(35°-45°C for ThiobacilluslLeptospirillum species and 45°-55°Cfor the moderately-
thermophilic bacteria). The leach process requires large volumes of air, supplied
by blowers. An agitator in each tank promotes uniform solids suspension and al-
lows for oxygen mass transfer.
For refractory, sulfidic gold concentrates, the discharge from the final stage is
subjected to water washing and solidlliquid separation in thickeners. The aqueous
discharge is treated with limestone, lime or both to stabilize arsenic and iron. The
biooxidized residue is water-washed, removing acid and soluble metals that con-
sume lime and cyanide. The washed residue is neutralized and leached in a cya-
nide circuit to recover gold.
All of the commissioned, aerated, stirred-tank reactor commercial plants
(Table 1.1) are technically biooxidation facilities as they operate with refractory-
sulfidic gold, flotation concentrates as feedstocks. However, there is no reason why
stirred-tank reactors cannot be applied to bioleach base metal concentrates. The
principal advantages cited by mine operators for selecting stirred-tank biooxidation
over the more conventional roasting and pressure autoclave technologies, are:
Lower capital and operating costs;
Greater gold recovery;
Shorter construction period;
• Robust process;
• Less onerous environmental requirements;
• Arsenic stabilize;
Operation simple, requiring less skilled labor; and
Plants safer and healthier.
Mining Biotechnology: Research to Commercial Development and Beyond 5
Table 1.1. Commercial-scale bioleachlbiooxidation plants commissioned

since 1986
Project
Type & Size Technology History & Status
Fairview, South Africa

35 mt/day gold Aerated, stirred tanks; Commissioned 1986;
flotation concentrate Genmin' process In operation
Sao Bento, Brazil
150 mt/day gold Single-stage, aerated Commissioned 1990;
flotation concentrate stirred tank preceding In operation
pressure autoclave;
Genmin' process
Harbour Lights, Western Australia
40 mt/day gold Aerated stirred tanks; Commissioned 1992;
flotation concentrate Genmin' process Shut-down 1994
for lack of feed
Wiluna Mine, Western Australia
flotation concentrate Genmin' process In operation
Sansu, Ghana
1,000 mt/day gold Aerated stirred-tanks; Commissioned 1994;
flotation concentrate Genmin' process Expanded in 1995;
In operation
Youanmi, Western Australia
flotation concentrate BacTech' process In operation
Lo Aguirre, Chile
3,500 mt/day Bio-heap with SX/EW; Commissioned 1980;
chalcocite ore grading SMp3 process Terminated 1996
1.4% CUi Production for lack of feed
14,000-15,000 mt Cu/yr
Cerro Colorado, Chile
16,000 mt/day Bio-heap with SX/EW; Commissioned 1993;
chalcocite ore grading SMp3 process Expanded in 1995;
114% CUi Production In operation
60,000 mt Cu/yr
Quebrada Blanca, Chile
17,300 mt/day Bio-heap with SX/EW; Commissioned 19941
chalcocite ore grading SMp3 process In operation
1.3% CUi Production
75,000 mt Cu/yr
continued ...
Table 1.1. Commercial-scale bioleachlbiooxidation plants commissioned

since 1986 (continued)
Project
Type & Size Technology History & Status
Ivan-Zar, Chile
1,500 mtJday Bio-heap with SX/EW; Commissioned 1994;
chalcocite ore grading SMp3 process In operation
2.5% Cu; Production
10,000-12,000 mt Cu/yr
Mt. Leyshon, Queensland, Australia
1,370 mtJday Thin-layer bio-heap Commissioned 1992;
chalcocite/gold ore to leach Cu & heap In closure
grading 1,750 g Cu/mt cyanidation to leach Au;
& 1.73 g Au/mt Process developed at
Mt. Leyshon
Girilambone, New South Wales, Australia
2,000 mtJday Bio-heap with SX/EW; Commissioned 1993;
chalcocite ore grading Process developed at In operation
2.5% Cu; Production Girilambone
14,000 mt Cu/yr
Newmont-Carlin, Nevada, USA
10,000 mtJday sulfidic Bio-heap & heap Commissioned 1995;
gold ore grading cyanidationlheap In operation
-1 g1mt gold thiosulfate; Process
developed by Newmont
Gunpowder's Mammoth Mine, Queensland, Australia
-1.2 million mt In situ bioieach; Commissioned 1991;
chalcocitelbornite Process developed by In operation
grading !2.2% Cu; mine owners
Design capacity of
13,000 mt Cu/yr
1 Developed by Genmin Process Research GENCOR, Johannesburg, South Africa, the process
uses Thiobacillus ferrooxidans and Leptospirillum ferrooxidans and operates at about 40°C.
2 Developed by BacTech (Australia) Limited, Perth, Australia, the process uses moderately-
thermophilic bacteria and operates at around 50°C.

3 Developed by Sociedad Minera Pudahuel.
Fig. 1.1. At Youanmi Mine, Western Australia, 120 mt/day of refractory-sulfidic gold
concentrate are biooxidized by moderately-thermophilic bacteria to enhance gold
recovery.
Capital and operating costs, highly discernible elements in process selection

by mining companies, are based on the base/precious metal grade in the ore and
concentrate, reagent consumption for precious metal recovery, and the predomi-
nate determinant-the amount of S'--S (sulfide-sulfur) requiring oxidation to
achieve satisfactory metal recovery. The power demand by the blowers and agita-
tors of aerated, stirred-tank reactors is directly related to the amount of 5'--5 re-
quiring oxidation to achieve satisfactory metal recovery. Capital costs of aerated,
stirred-tank bioreactors are site-specific and related to which mineral sulfide is
being oxidized and the amount of S'--S requiring oxidation. Important consider-
ations are:
Residence time of the feed in the tanks, which dictates the number and size
of the tanks and number of stages in the circuit;
Cost of the blowers;
The size and cost of the cooling circuit, which is related to the type and
amount of mineral sulfide oxidized as each mineral has a specific heat of
reaction; and
• The neutralization circuit for refractory gold plants.
The principal disadvantages of aerated, stirred-tank reactors compared to
bioheaps are the capital and operating costs. The technology of these processes is
reviewed in detail in chapters 3 and 4.
Bioheaps and Copper Dump Leaching

Bioheap reactors are in commercial use today to pretreat low-grade, refrac-
tory-sulfidic gold ores and to leach copper from chalcocite ores (Fig. 1.2). Mined
ore is crushed, acidified with sulfuric acid to condition the ore for the bacteria,
agglomerated in some cases to bind fme materials to the coarser ore particles, and
stacked on lined pads on which aeration piping may be placed. The stack height of
the ore varies from about two to ten meters. Maximum heap height depends on
heat generation and dissipation, acid balance throughout the heap, and air/water
permeability. During the bacterial leach cycle, heaps can be aerated by forcing air
through the piping on the pad using low-pressure fans.
In chalcocite heap leaching the stacked ore is irrigated with effluent from the
solvent extraction/electrowinning (SX/EW) plant. The solution percolates through
the heap where bacteria catalyze the release of copper. Soluble copper is recovered
by SX/EW (see chapter 2).
For refractory-sulfidic gold ores the crushed ore is agglomerated with acid
and/or an acidic-ferric sulfate solution containing bacteria (see chapters 5 and 6).
The stacked ore is usually initially irrigated with an acidic, ferric solution contain-
ing bacteria and later with recycled bioheap leach effluent. After oxidizing S~- -S to
levels predetermined from testwork, the biooxidized ore is water washed, remov-
ing residual acid and "cyanicides" (constituents that consume cyanide). The heap
is dismantled and the biooxidized ore is agglomerated with lime or cement, re-
stacked on lined pads and leached with a dilute cyanide solution to extract gold. If
the ore is preg-robbing, that is refractory due to the presence of carbonaceous
material which re-adsorbs the gold-cyanide complex, ammonium thiosulfate can
be used as an alternative lixiviant.
Larger ore particle sizes are used in bioheap leaching, therefore overall metal
recovery is lower than is achieved by alternative process methods including stirred-
tank bioreactors. However, the reasons mining companies select bioheap leaching
over alternative processes are compelling and differ somewhat between base metal
and gold projects. Among the reasons bioheap leaching is selected for copper sul-
fide deposits are:
• The small size of the project can't sustain the construction cost of a smelter;
Smelter charges, a penalty imposed by a toll smelter because of contami-
nants in the concentrate feed, are too high;
• The remoteness of the mine precludes preparing and shipping a concen-
trate to another smelter because of high transportation costs; and
• There is no need to construct a tailings pond or an effluent treatment plant
as there is no aqueous discharge.
For refractory gold projects the reasons for using bioheap leaching involve the
following:
• The gold grade is so low that when the ore is concentrated the grade is still
too low to support the cost of a roaster, pressure autoclave or even a stirred-
tank bioreactor;
• The material containing small amounts of gold is mined with higher grade
ore and is considered waste, but it can be bioheap leached because of the
low capital and operating costs of the process;
The material is over-burden and recovering the low gold values will off-set
the cost of stripping, making overall project economics more favorable;
Fig. 1.2. Minera Quebrada Blanca, at 4.400 m above sea level in the Andes of northern
Chile, is the largest stand-alone bioheap leach facility in the world, processing 17,300
mt/day of chalcocite ore and producing 75,000 mt/yr of cathode copper.
The mineable reserves may be too low to support the capital costs of a roaster
or pressure autoclave, but the project economics are very favorable for
bioheap leaching even if the gold recoveries are somewhat less;
• The ore can't be concentrated because of mineralogy; and
• Bioheap leaching reduces high reagent consumptions caused by some con-
stituents (for example, copper).
Copper dump leaching was first initiated in the late 1960s and is still used to-
day to scavenge copper from run-of-mine (material not subjected to comminu-
tion), submarginal ore that can't be economically processed by any other means.
Copper dump bioleaching, employed by nearly every company mining a porphyry
deposit, is less technically sophisticated than stirred-tank, heap and in situ plants.
The submarginal ore is transported to a hydrologically isolated area and piled to
depths of up to 350 meters. The giant ore piles, containing billions of tonnes of
material, are acidified and the leaching bacteria facilitate the extraction of copper,
which is recovered by SX/EW. Copper dump leaching remains an economically
vital process to the copper industry and the technology has been consequential in
the genesis of the bioheap leaching technology commercially applied today.
In Situ Bioleaching
In situ bioleaching has been commercially used as a scavenger technology for
nearly 30 years to extract uranium and copper from depleted underground mines.
When conventional mining is completed, the underground workings are blasted
to fragment the ore and over-burden material establishing permeability. Shafts are
left intact to allow for aeration of the fragmented ore and to recover the metal-
bearing solutions from sumps. Acidified leach solutions, applied to the top surface
of the entire rubblized ore zone, percolate through the fragmented ore. The leach-
ing bacteria become established and facilitate metal extraction. Metal-rich solu-
tions, recovered in sumps, are pumped to the surface for metal recovery. Ultimate
metal recovery from in situ operations is dependent on the degree of ore fragmen-
tation and uniform irrigation of the fragmented ore zone. Poor metal recoveries
can often be traced to leach solutions following preferential flow paths and not
contacting the ore uniformly.
Applications of Bioremediation in the Mining Environment

Commercial applications ofbioremediation in the mining environment involve
the immobilization and recovery of soluble metals from aqueous wastes and the
microbial degradation of cyanide species.
The largest commercial bacterial sulfate reduction project to remove metal
contaminants from water is at a zinc refinery in Budel-Dorplein, The Netherlands.
This plant, commissioned in 1992, is designed to treat 300 m3/hr of effluent con-
taining 100 mg/L zinc, 1 mg/L cadmium, and 1,000 mg/L sulfate. The microbial
treatment process, commercially developed by Thiopaq Sulfur Systems B. V. (Balk,
The Netherlands) and Paques, Inc. (Exton, Pennsylvania, USA), achieves an efflu-
ent discharge of <0.3 mg/L zinc, <0.01 mglL cadmium and <200 mg/L sulfate. The
metal sulfides, precipitated by the H2 S, and elemental sulfur, produced from mi-
crobial oxidation of excess H2 S, are returned to the smelter where the metals are
reclaimed and the sulfur is converted to sulfuric acid in the acid plant. Total costs
(chemicals, energy and depreciation) of a microbial sulfate and metals removal
facility depend on waste stream characteristics and discharge criteria. However, as
an example the estimated cost for a plant reducing 10,000 million tons (mt) per
year sulfate would be about U.S.$330/mt sulfate.
Homestake Mining Company, Lead, South Dakota, USA, developed and com-
missioned in 1989 a microbial cyanide oxidation system (Fig. 1.3) to treat some
850 m3/hr of tailings containing 62 mg/L SCW, 4.1 mglL total CW, 0.56 mglL CuH
and 5.6 mg/L NH4 +. The plant consists of two stages of rotating biological
contactors-the first stage with Pseudomonas species, the second stage with nitri-
fying bacteria-resulting in an effluent with <0.05 mg/L SCW, 0.07 mg/L total
CW, and <0.50 mglL NH4+. A discharge of <0.07 mglL CuH is achieved when the
metal, released from the cyanide complex, is bound to the biomass on the rotating
biological contactors. The operating cost of the Homestake cyanide oxidation fa-
cility is about U.S.$0.1l/m3 of effluent treated.
Microbial cyanide oxidation to degrade cyanide during the decommissioning
of precious metal cyanide heap leach operations is reportedly used on a commer-
cial scale. However, the process has not been subjected to rigorous analysis to con-
firm performance and cost.
Wetlands are employed throughout the hard-rock and coal mining industries
to treat acid rock drainage. Anaerobic wetlands use waste organics, such as mush-
room composite, manure and straw, as an energy source for sulfate-reducing bac-
teria that produce H2 S to precipitate metals as sulfides. The carbonate produced
by the bacteria neutralizes the acid waters.
There have been no compelling reasons for the mining industry to consider
bioremediation, as conventional treatment methods are generally efficient and cost-
effective. However, because the strictest discharge standards for some constitu-
Fig. 1.3. At Homestake Mining Company in South Dakota (USA) 850 m3/day of tailings
are treated cost effectively with bacteria on rotating biological contactors to achieve
strict discharge standards for cyanide species and copper.
ents (e.g., N03- and Se) are very difficult to achieve, the new generation of miners
is taking a closer look at bioprocesses for clean-up and mine closure remediation
activities.
R&D Leading to Today's Commercial Operations

BioleachinglBiooxidation R6-D
The discovery in 1947 that bacteria are associated with acid rock drainage and
the characterization and naming of Thiobacillus ferrooxidans in 1951 spurred re-
search on the role of these organisms in oxidizing mineral sulfides. Research pub-
lished in the 1950S led to the 1958 patent that preceded the industrial scale applica-
tion of copper dump leaching by Kennecott Copper Corporation in the early 1960s.
Research and development between 1960 and 1980 yielded important informa-
tion:
Metabolic pathways for sulfur and iron oxidation in thiobacilli were de-
scribed;
Morphological and structural characteristics of the leaching bacteria iden-
tified;
• Genetics of metal-microbe interactions introduced;
Moderate and extreme thermophiles that oxidize reduced sulfur and iron
compounds and mineral sulfides discovered and partially characterized;
• Microbial-mineral interactions and mineral metabolism described;
Metal, iron concentration, nutrient, light, pressure, aeration, and tempera-
ture, affects on bioleaching quantified;
• Extent of microbial leaching of mineral sulfide ores and concentrates (for

example, pyrite in coal, arsenic sulfides, chalcopyrite, and complex mineral
sulfides) defined;
Ecology of copper dump leach operations studied;
Heterotrophic microbial processes of mineral's leaching surveyed;
• Bioleaching of mineral sulfide concentrates in continuous, stirred-tank re-
actors perfected;
Large-scale test facilities to evaluate copper bioleaching employed.
The results of these studies were chronicled in the International Biohydro-
metallurgy Symposia of the 1970S,t-4 an excellent book5 on thermophilic microor-
ganisms, and hundreds of journal publications.
Important technological advancements made in bioleaching during the 1980s
are summarized in Table 1.2 and detailed in the four proceedings of the Interna-
tional Biohydrometallurgy Symposia. 6 -9
The 1990S has been the decade of commercial awakening of bioleachingl
biooxidation. Twelve of the world's fourteen commercially-operating facilities were
commissioned in the first five years of this decade (Table 1.1). This commercial
activity sparked notable fmdings reported in an increasing number of journal pub-
lications, a book/o and symposiaU- 18 devoted to the discipline. Research and de-
velopment areas in Table 1.2 were amplified with consequential contributions made
in the biochemistry, genetics and molecular aspects of the leaching organisms
and -at the other end of the spectrum - testwork requirements, engineering con-
siderations and economics of tank and heap bioleaching/biooxidation scale-up.
As we approach the end of another decade it is somehow paradoxical that, despite
the considerable advances that have been made over 30 years, a fundamental as-
pect-the relative importance and contribution of direct and indirect leaching-
eludes scientists and practitioners of the technology.
Minerals Bioremediation R&D

Minerals bioremediation research and development have been ongoing since
the 1960s with the discovery of microbial cyanide oxidation and reports of metal-
binding by microorganisms. Research and development efforts since 1960 have
largely focused on these microbial mechanisms of metals immobilization and cya-
nide degradation:
• Precipitation of metals by bacterially-generated substances, such as phos-
phate and H 2 S;
• Binding of metals onto microbial cell surfaces by physical-chemical pro-
cesses and intracellular metal's accumulation;
Microbial oxidation or reduction of a metal or cyanide species, immobiliz-
ing metals and attenuating cyanide. Metals released from metal-cyanide com-
plexes by oxidation are bound to biomass.
Studies on sulfate-reducing bacteria (SRBs) have gone in two directions:
(1) use of SRBs in anaerobic wetlands for immobilization of metals as sulfides and
neutralization of acid rock drainage, and (2) use of SRBs in specially-built
bioreactors to treat large volumes of water, such as contaminated groundwater,
containing relatively dilute concentrations of metals and sulfate.
During the 1980s extensive research and development (R&D) was carried out
on understanding the mechanisms and extent of metal-binding to algae, fungi and
bacteria. Several technologies emerged using nonliving, immobilized micro organ-
isms for the accumulation of metals from contaminated streams. These metal-bind-
ing agents could be stripped of metals and re-used in a manner similar to ion
exchange resins. Despite the considerable effort in development of these
bioremediation processes, none of these technologies has been successfully com-
mercialized.
Cyanide degradation research has focused on both aerobic and anaerobic at-
tenuation, with the former in commercial application. In the last decade research
attention has turned to using microorganisms for attenuating cyanide in decom-
missioning cyanide heap leach facilities. Washing the heaps to remove residual
cyanide requires considerable time and fresh water. Bacterial degradation of WAD
(weak-acid dissociable) and free cyanide reduces both the time and amount of
water required.
R&D efforts in minerals bioremediation have been recorded extensively in the
proceedings of the International Biohydrometallurgy Symposia,1-4.6 -9.11.U.14,18
BIOMINE symposia15.16 in several bookslO.19-21 and numerous papers published in
biological and mining journals.
New Developments
Commercial applications of aerated, stirred tanks for refractory-sulfidic gold
concentrates and bioheaps for chalcocite and refractory-sulfidic gold ores have
shifted the focus of R&D somewhat to topics that will improve these processes,
diminish capital and operating costs and extend the process applications. New
developments that are promising in this context are:
• Improved aeration systems for stirred-tanks that will improve utilization
efficiency, and decrease operating costs;
• Aerated, stirred-tank and bacterial ferric generation systems for chalcopy-
rite concentrate bioleaching-developments that promise significant cost
reductions over smelting and could open new mines in remote locations;
• Aerated, stirred tanks for Co, Zn and other mineral sulfide bioleaching and
for recovery of less-common metals (e.g., In);
• Bioheap leaching of concentrates, which would reduce capital and operat-
ing costs for processing concentrates;
Techniques to decrease cyanide consumption of biooxidized residues, re-
ducing operating costs of gold plants;
Innovative bioheap technologies for diverse climates (e.g., high rainfall ar-
eas, tropics, Arctic regions), disparate ore types (e.g., clayey ores), and en-
hanced aeration;
• Effective application of extremely thermophilic bacteria in stirred-tank re-
actors and bioheaps; and
• Better understanding of solution chemistry equilibria in bioheaps to im-
prove kinetics and minimize leach solution treatment costs.
Biooxidationlbioleaching research that promises to make a difference from a
fundamentals perspective encompasses:
• Bioleaching of chalcopyrite ore to allow low-cost heap and in situ leaching
of deposits;
• Genetically altered bacteria which improve biooxidationlbioleaching and
can effectively compete with native bacteria in commercial plants;
• Defmitive understanding of the direct versus indirect attack and the relative
importance to stirred tank and bioheap reactors;
Table 1.2. Technological advancements in bioleachinglbiooxidation in the

1980s
Research & Development Area

- Advancements
Genetics
- Genetic system established
- Plasmids characterized
- Molecular probe construction and use initiated
- DNA sequencing of T. ferrooxidans started
- Immunological differentiation of T ferrooxidans strains noted
- Problems and opportunities of T. ferrooxidans genetic manipulation
identified
Basic microbiology
- Thiobacilli enzymology studied
- More thiobacilli species identified and characterized and phylogeny
established
- Leptospirillum ferrooxidans characterized
- Electron transport of T. ferrooxidans described
- Biochemistry and structure of moderate and extreme thermophiles
investigated
Fundamentals of bioleaching
- Importance of mixed cultures recognized
- Enhanced moderate thermophile bioleaching noted with CO. addition
- Toxicity of As(III) observed
- Role of attached and unattached thiobacilli debated
- Performance and rates of T. ferrooxidans and extreme thermophiles
compared
- Issue of high solids density in tank leaching of concentrates with extreme
thermophiles reported
Base metal bioleaching
- Bioleaching kinetic differences noted with different chalcopyrite concentrates
- Hypotheses developed to explain slow kinetics of chalcopyrite bioleaching
- Elemental sulfur formation during chalcopyrite concentrate bioleaching
reported
- Electrochemical coupling with bioleaching described
- Silver-catalyzed bioleaching of chalcopyrite reported
- Improved chalcopyrite leaching with extreme thermophiles observed
- Assessment of other base metal sulfide bioleaching continued
Refractory precious metal leaching
- Biooxidation with T. ferrooxidans, 1. ferrooxidans, and moderate and extreme
thermophiles of pyrite and arsenopyrite to expose occluded gold evaluated
- Process mineralogy used to extensively evaluate biooxidation
continued...
Table 1.2. (continued)
Research & Development Area

- Advancements
Modeling
- Power and agitation requirements of tank bioleaching first modeled and
issues of O2 transfer reported
- Kinetics ofbioleaching modeled
- Modeling of ore bioleaching in columns performed
- Kinetics ofbioleaching in copper dumps modeled
Scale-up/commercial activity
- Aerated, stirred-tank biooxidation of refractory precious metal concentrate
piloted outside lab with commissioning of first refractory gold biooxidation
plant still in operation
Commercial development initiative on chalcocite leaching commenced,
resulting in first bio-heap facility
Aerated, stirred-tank bioreactor developments progressed
In situ copper, zinc and uranium bioleaching piloted and first in situ uranium
facility commissioned
Alternative leaching
- Anaerobic leaching of mineral sulfides by T. ferrooxidans reported
- Lateritic nickel leaching evaluated
- Manganese, rare earth elements, silicate and bauxite leaching examined
Coal desulfurization
- Partial oxidation of pyrite by T. ferrooxidans tested for pyrite and other
mineral sulfide rejection with oil-agglomeration
- Bioleaching and froth flotation coupled for pyrite removal
- Thermophilic bioleaching of pyritic coals examined
- Degradation of organic sulfur by leaching bacteria studied
• Better overall understanding of mineral-microbial interactions in relation

to direct versus indirect attack and iron deposition during leaching;
• Assessment of how mixed cultures perform under extreme solution condi-
tions (i.e., extremely high acidity and iron) that develop in the bioheap leach-
ing of high sulfide ores;
• Explanations and resolutions for extended lag times that bacteria exhibit
with certain ores;
• Better understanding of microbial metabolism of mesophilic and thermo-
philic bacteria;
• Comprehensive description of the heterotrophic and autotrophic microbial
populations in bioheaps and their relevance to biooxidationlbioleaching;
and
• Cost-effective, easy and accurate methods of monitoring bacterial activity
in heaps.
New developments in bioremediation that promise to assist the mining indus-

try include:
Methodical assessment of the mechanisms, performance and cost of micro-
bial cyanide degradation in decommissioning cyanide heap leach facilities
and subsequent nitrate removal;
Integration of cost-effective bioremediation technologies into bioheap leach-
ing facilities (e.g., bioremediation for iron recovery, elemental sulfur pro-
duction and base metal recovery); and
Improved and cost-effective denitrification processes for treating nitrate in
aqueous discharges from mines (e.g., cold water and limited nutrients) and
decommissioned heaps following cyanide oxidation.
Summary
Since research on bioleaching was initiated in the late 1950S, bioleaching and
biooxidation have emerged as accepted commercial practices that offer the min-
ing industry cost-effective, simple, robust, high performance and environmentally
friendly alternatives to conventional mineral processing methods. Aerated, stirred-
tank biooxidation and heap biooxidation/bioleaching are used throughout the
world to process refractory gold ores and concentrates and chalcocite ores. Miner-
als bioremediation processes, also enjoying commercial success, have emerged as
alternatives to more expensive and cumbersome chemical/physical processes. New
fundamental and applied developments in bioleaching/biooxidation and minerals
bioremediation promise to extend applications of the technology, reduce costs
and enhance performance. In the next few years, we can look forward to greater
commercial application of mineral bioprocesses, incorporating innovations that
will assist the industry in remaining competitive, and mining in an environmen-
tally responsible way.
References
1. Schwartz W, ed. Conference Bacterial Leaching 1977-GBF Monograph NO.4.
Weinheim: Verlag Chemie, 1977.
2. Murr LE, Torma AE, Brierley JA, eds. Metallurgical Applications of Bacterial Leach-
ing and Related Microbiological Phenomena. New York: Academic Press, 1978.
3. Trudinger PA, Walter MR, Ralph BJ, eds. Biogeochemistry of Ancient and Mod-
ern Environments. Canberra: Australian Academy of Science, 1980.
4. Proceedings of the International Conference on the Use of Microorganisms in
Hydrometallurgy. Pecs: Hungarian Academy of Sciences, 1980.
5. Brock TD. Thermophilic Microorganisms and Life at High Temperatures. New
York: Springer-Verlag, 1978.
6. Rossi G, Torma AE, eds. Recent Progress in Biohydrometallurgy. Iglesias, Italy:
Associazione Mineraria Sarda, 1983.
7. Lawrence RW, Branion RMR, Ebner HG. Fundamental and Applied Biohydro-
metallurgy. Amsterdam: Elsevier, 1986.
8. Norris PR, Kelly DP, eds. BioHydroMetallurgy. Kew Surry: Science and Technol-
ogy Letters, 1988.
9. Salley J, McCready RGL, Wichlacz PL, eds. Biohydrometallurgy-CANMET
SP89-1o. Ottawa: Canada Centre for Mineral and Energy Technology, 1989.
10. Ehrlich HL, Brierley CL, eds. Microbial Mineral Recovery. New York: McGraw-
Hill Publishing Company, 1990.
11. Karavaiko GI, Rossi G, Avakyan ZA, eds. Biohydrometallurgy-Proceedings of

an International Seminar on Dump and Underground Bacterial Leaching of Met-
als from Ores. Moscow: Centre for International Projects, USSR State Committee
for Environment Protection, 1990.
12. Duarte JC, Lawrence RW, eds. IX International Symposium-Biohydrometallurgy
'91. Troia: Forbitec Editions, 1991.
13. Torma AE, Wey JE, Lakshmanan VI, eds. Biohydrometallurgical Technologies, Vol.
I. Bioleaching Processes. Warrendale PA: The Minerals, Metals and Materials So-
ciety, 1993.
14. Torma AE, Apel ML, Brierley CL, eds. Biohydrometallurgical Technologies, Vol.
II. Fossil Energy Materials, Bioremediation, Microbial Physiology. Warrendale PA:
The Minerals, Metals and Materials Society, 1993.
15. BIOMINE '93. Adelaide: Australian Mineral Foundation, 1993.
16. BIOMINE '94. Adelaide: Australian Mineral Foundation, 1994.
17. Vargas T, Jerez CA, Wiertz JV, Toledo H. Biohydrometallurgical Processing, Vol.
I. Santiago: University of Chile, 1995.
18. Jerez CA, Vargas T, Toledo H, Wiertz JV. Biohydrometallurgical Processing, Vol.
II. Santiago: University of Chile, 1995.
19. Westbroek P, De Jong EW, eds. Biomineralization and Biological Metal Accumu-
lation. Dordrecht: D. Reidel Publishing Company, 1982.
20. Eccles H, Hunt S, eds. Immobilization of Ions by Bio-Sorption. Chichester: Ellis
Horwood Limited, 1986.
21. Volesky B, ed. Biosorption of Heavy Metals. Boca Raton: CRC Press, 1990.
SECTION II
Industrial Processes
CHAPTER 2
Bioleaching of Copper
Henry A. Schnell
Introduction
N atural bioleaching has been taking place for almost as long as the history of the
world, but it is only in the last few decades that we have realized that bioleaching
is responsible for acid production in some mining wastes, and that this bacterial
activity can be used to liberate some metals. The application of the bioleaching
reaction for copper has been exploited and used to develop suitable methods to
recover copper from copper-bearing solutions.
The heap leaching of copper has been practiced for several decades, mostly
with oxide ores. Probably bacteria aided some of these oxide operations, but this
was without serious study. More recently, mines such as LoAquirre, Cerro Colo-
rado, Giralambone and Quebrada Blanca have made a concerted effort to optimize
bacterial activity and have chosen to rely on bacteria for their copper production.
Bioleaching of copper is of major importance in metals production and this chap-
ter will look at recent operations and the innovations that are making economic
recovery of secondary sulfide copper possible.
Before examining the application of bioleaching to copper recovery, I wish to
review the basic mineralogy and some recent ideas which are providing a new
viewpoint on the chemistry of leaching. This chapter also includes a short review
of copper leaching chemistry and the contribution of bacteria to these reactions.
We lack a complete understanding of the various mechanisms and are dealing with
living organisms which have somewhat unpredictable behavior. Included is a short
review of copper recovery from bioleach solutions and the production of the final
product in an economically viable process.
This chapter is intended as a review of copper bioleaching, its application, re-
cent bioleaching experience and some examples of existing operating plants. No
doubt the advance of learning and its application will create more economically
viable and efficient projects. A geologist friend left me with an interesting com-
ment some years ago: "Remember that geologists find mines and metallurgists
make them into waste!" Bioleaching technology is letting us make some waste into
mines.

Leaching Definitions
One of the most confusing aspects in reviewing copper leaching is a confusion
of terms used for each type of leach operation. Therefore, for the purposes of this
chapter the following terminology will be used:
Dump leaching is leaching from existing run-of-mine leach dumps which
were previously considered waste and includes run -of-mine ore placed spe-
cifically for leaching. This type ofleaching does not use size reduction equip-
ment and relies on mine blasting for size reduction.
• In situ leaching is the leaching of ore in place without removal from the ore
body, using adits or drill hole solution systems.
Heap leaching refers to crushed ores placed on prepared pads in layers for
leaching.
Permanent pad refers to a prepared leach pad that is used continuously with-
out removal of the leached ore prior to stacking of fresh ore.
Dynamic or ON/OFF pad refers to prepared pads from which the ore is re-
moved prior to placing fresh ore for leaching.
The main confusion occurs with reference to the terms "dump" and "heap"
leaching. Dump leaching is carried out on untreated uncrushed run-of-mine ma-
terial whereas heap leaching involves crushed, pretreated ore.
Copper Mineralogy
This short glance at the mineralogy of copper minerals is not intended to be
complete as there are over 350 copper minerals. What will be discussed are the
most common copper sulfide species considered suitable for bioleaching. Also in-
cluded is the oxidation of pyrite which forms a fundamental part of copper
bioleaching.
Pyrite
A short review of pyrite oxidation is appropriate since these reactions are the
driving force behind what happens mineralogically inside heap and dump leach-
ing operations. 1a The oxidation of iron disulfide to the sulfate ion involves the trans-
fer of as many as 16 electrons with the potential formation of complex intermedi-
ate species. The overall reaction takes place at pH <4 in heaps and dumps and
releases about 1500 KJ of heat in the following reaction:
FeS.(s) + 14 Fe 3+(aq) + SHoO --t 15Fe2+(aq) + 2S0 4'-(aq) + 16H+
The reaction rate is fast, but limited in the absence of assistance from bacteria
such as Thiobacillus ferrooxidans which catalyzes the reoxidation of ferrous to fer-
ric iron as shown below:
Secondary Sulfides
The most common secondary copper mineral considered in heap bioleaching
are chalcocite and covellite. There is generally a local participation of ferric iron in
the process derived from dissolution of either pyrite or chalcopyrite. The earliest
work by Sullivan1b in 1930 indicated that chalcocite dissolved in two stages accord-
ing to the reactions:
Bioleaching of Copper 23
Cu 2S + 2Fe3+ ~ Cu2+ + 2Fe2+ + CuS (not the mineral covellite)

CuS + 2Fe3+ ~ Cu2+ + 2Fe2+ + SO
In 1967, Thomas lC et alleached both synthetic chalcocite and digenite and found
that chalcocite was converted to digenite and thereafter both samples were con-
verted into CUl.lS, at that time considered to be one "mineral" -blaubleibender
covellite. The proposed system from this discovery was identical to the reaction
given by Sullivan,lb but with quite a different reaction path.
More recently, Gobleld,le published a study of the leaching of chalcocite with
Fe3+(aq.) and found that an entire series of nonstoichiometric copper sulfides were
produced. From this work a variety of possible chalcocite-series leaching sequences
are possible. (Recent work in leaching of chalcocite in bacterial and sterile condi-
tions has shown that the bacteria do increase the rate of reaction and can be in-
volved in these reactions). Possible chalcocite leaching sequences are as follows:2,3
1t 1t l
Cu2 S ~ CUI •97S ~ CUl.8S ~ CUI .75S ~ CUl.6S ~ Cul.4S ~ Cul.US ~ CuS
chalcocite dijurleite digenite anilite geerite spionkopite yarrowite covellite
~ .u. * 1t *
Cul.7sS ~ CUI •74S ~ Cul.US => common reactions
roxbyite "digenite" yarrowite ~ rare reaction
Primary Sulfides
The major copper primary sulfide considered economically important is chal-
copyrite. Chalcopyrite is bioleachable, but is currently considered to be nonviable
economically due to the long leach times required. Some operations have used
waste dumps, irrigated over a four to six year period to recover copper from chal-
copyrite. Reported recoveries are only about 15%. Considerable effort is being ex-
pended to improve recovery of primary sulfides through bioleaching in heaps. Most
work has been on the bioleaching of concentrates rather than mined ore. Much of
the concentrate leaching uses stirred tank reactors; this has not been reviewed in
this chapter since these processes have not yet been applied commercially.
Physico-Chemical Leaching Variables

Most of the variables that effect leaching rates and recoveries are common to
heap, dump, or in situ leaching systems. These variables can be divided into those
effecting the chemical part of leaching and those that effect the bacterial compo-
nent. The primary variable is the mineralogy of the ore and it is an assumed vari-
able. The remaining variables are discussed in the following three sections.
Surface Area
The rock size used for leaching has a direct effect on the leach rate. 4 Whether
produced by pit blasting or by crushing, rock size is specific to each ore deposit
and depends on the mineral occurrence and whether it is disseminated or in bound-
ary layers within the rock. The effect of particle size must be tested for each ore
type. In general, the finer the particle size the better the final recovery. It is inter-
esting to note that frequently, initial recovery is faster with ore of a larger size due
to the reduced surface area, although the final recovery is adversely affected. In
situ leaching is primarily affected by the mineral formation and fissure sizes. s
The chemical leach component is affected by the ability of the solution to con-
tact the copper and to transport the dissolved copper out of the rock fissures. Work
on the use of a wetting agent to reduce solution surface tension to allow better
fissure penetration has taken place, especially with oxide ores. 6 Little work of this
type has been carried out on bacterial leaching of sulfide minerals. In general it is
felt that wetting agents or surfactants are detrimental to the bacteria because of
the potential of these agents to rupture bacterial cells.
Acid Levels
In dump leaching, a preconditioning step with high acid levels is frequently
employed.? The benefit of this increased acid level is specific to each ore type. Too
much acid will increase the overall acid usage due to acid consumption by the
gangue. In heap leach operations, agglomeration is commonly practiced. This ag-
glomeration step releases some chemically available copper. The acid addition rate
during the agglomeration step is ore-specific. As with dump leaching pretreat-
ment, the initial acid level for agglomeration must be balanced against gangue
mineral consumption and the acid required to optimize bacterial activity.
Oxidants
Ferric ions are considered to be the primary oxidant in the dissolution of cop-
per and it is the production of ferric ions that is the main benefit of bioleaching. A
possible alternative to bioleaching would be the physical addition of ferric iron.
However, this is an expensive alternative unless there is a means of reoxidizing the
ferrous iron. There are several patents on the regeneration of ferric iron, but none
of these alternatives are as cost -effective as bacterial oxidation. Copper grades suffi-
ciently high to carry the cost of copper recovery by chemical leaching are not con-
sidered in this chapter. Other potential oxidants cannot be used in conjunction
with bacterial leaching since they have a detrimental effect on the bacteria.
Agglomeration
A common innovation in heap leach operations is to mix the ore with acid and
water to form an agglomerate. 8 - 10 This agglomerate differs from traditional gold
or concentrate agglomerates in that a true 'pellet' is not formed. The purpose of
agglomeration is to prevent the segregation of fine and coarse material during
stacking. The moisture level within the agglomerate also partially determines the
permeability of the stacked material.
Agglomeration takes place either in rotating drums (Fig. 2.1), reversing con-
veyor belts or conveyor belts with a series of plows to mix the ore and water/acid.
Most newer operations have included rotating drums during initial design, but in
some cases these have been fitted later to improve agglomerate quality. Typical
agglomerate discharge moisture is about 10% by weight. A larger crushed ore size
will decrease the required moisture levels to form a suitable agglomerate. Hot wa-
ter can be used for agglomeration to add heat to the leach pile and start more rapid
leaching. ll
Fig. 2.1. Agglomeration drums at the Quebrada Blanca Operation, Chile.
An interesting phenomenon associated with agglomeration is that some cop-

per is liberated quickly as a result of this process. Leaching an ore in the laboratory
with acid does not liberate the same quantity of copper as does the agglomeration
process. The use of a binder to improve permeability of the agglomerate may be
considered in the future, but there are no reports of the application of binders in
copper heap leaching.
Curing Time
The time required for the acid and moisture to act on the minerals must be
considered. In heap leach operations using agglomeration, the time after agglom-
erating, prior to irrigation is considered important. At the LoAguirre copper op-
eration three days is specified, at Cerro Colorado several weeks is allowed and at
the Quebrada Blanca operation the curing time is not considered an important
variable. Tests must be carried out on each ore type to determine the appropriate
curing time.
Permeability
The permeability of a heap or dump helps to determine the solution distribu-
tion and ingress of oxygen required for bacterial activity. Solution ponding on top
of the heap must be avoided since this will inhibit oxygen penetration. The per-
meability will decrease as a dump or heap ages and solution irrigation rates may
have to be reduced. Asystem of ON/OFF irrigation may also be employed in older
leach areas. Good agglomeration greatly improves permeability and consistency
on the pile and prevents solution channeling.
26 Biomining: Theory. Microbes and Industrial Processes
Bacterial Leaching Variables

Many of the variables discussed above also have a bearing on bacterial activity.
The available surface area effects bacterial growth and solution transport, as does
the level of acid and other oxidants. The heap permeability has a direct effect on
oxygen availability for bacterial growth. The more important variables affecting
bacterial activity are discussed below.
Acidity
The amount of acid added initially effects the chemical leaching of copper, but
suitable pH levels for bacterial growth must be maintained. Desirable pH levels
are 1.8 to 2.2. 12 Typical solution acid concentrations are 6.0-8.0 giL sulfuric acid. In
heap or dump leaching, consideration must be given to the change in acid concen-
trations from the top to the bottom of a leach pile (i.e., the pH rises as acid is
consumed while the solution drains through the pile). This fact makes acid control
difficult at times and influences the height of heap construction. Adaptation of the
bacteria helps this situation, and if there is sufficient oxygen, there will be a bacte-
rial front moving through the pile.
Oxygen
A theory of bacterial growth in a heap holds that the major area of bacterial
growth is in the top 1.5 meters of a leach pile due to the limits of oxygen diffusion
into the surface of a heap.13,14 Measurements on nonaerated heaps have shown that
oxygen levels drop after the first 1.5 meters from the top of the leach pile to below
5% oxygen at the bottom. To improve natural diffusion of oxygen, cyclical irriga-
tion has been used. The theory of cyclical irrigation is that the draining of the
irrigation solution will result in oxygen being drawn into a leach pile. No data is
available to demonstrate this theory.
In dump leach operations the natural segregation of coarse and fine ore as the
dump is constructed allows air to ingress into the bottom of a leach dump. Air
ingress has been further improved by the "finger dump" design.ls The idea of this
design is to optimize the "chimney effect" within the dump. The effect of drilled
air holes within dumps and the addition of air by use of fans have been also been
tested. 16 Finger dump construction is the norm for leaching operations and forced
air addition has been limited to test work.
The situation for heap leaching is somewhat different. The finer crushed ore
does not allow for the same amount of coarse and fine material segregation and
this segregation is not desirable for permeability reasons. In heap leaching, air
diffusion theory can be accepted for heap operation, but improved leach and re-
covery rates are possible. Recent developments in bacterial heap leach operations
have seen the addition of air injection systems to reduce leach cycle times. Initial
air addition tests used the bottom drain systems to inject air. In at least one opera-
tion a separate air injection system that uniformly distributes air through the heap
has been installed. The benefits of aeration are demonstrated in Figure 2.2, where
the reduced leach cycle times achieved with air injection on a large scale operation
are clearly seen. The effects on overall copper recovery are not clearly defined since
this plant achieved high copper recovery without aeration, although with longer
leach cycle times.
eu gain to pregnant leach solution

Sector At3 vs A25
23 28 33 38 43 48
Time of Irrigation, days
Fig. 2.2. Effect of heap aeration on copper extraction. Sector A13 without
aeration is compared with sector A25 with aeration.
The addition of air to a heap leach operation is accompanied by several changes.

When air is first added, iron levels within the leach solution will markedly dimin-
ish with the precipitation of jarosite or hydroxides of iron. A further observation
(still under investigation) is that the amount of ferric iron in the discharge solu-
tion tends to decrease although the leach rate increases. This effect may suggest
that bacteria have a direct mechanism in the leaching of copper sulfides and that
their role is not limited to the production of ferric iron.
Nutrients
Leaching bacteria require ammonium nitrogen and phosphate, supplied as
(NH 4)zS04 and either KH zP0 4 or H3P0 4. Dosages of these nutrients should be
10-20 mglL NH4- and 30-40 mglL POr for bioleaching. Care must be taken dur-
ing addition of the ammonium since jarasite formation which results in the re-
moval of iron from the leach cycle is favored and may cause heap permeability
problems. To minimize these problems the heap should be at a pH <2. Analysis of
the ore and solutions is required to determine the level of nutrients and the pres-
ence of detrimental ions. Generally there is sufficient phosphate from mining ac-
tivities and only ammonium addition may be required. ' ?
Heat
It is generally agreed that T. ferrooxidans and T. thiooxidans grow best at
20°-35°C,although activity is evident outside these ranges. Temperature is impor-
tant and a general rule of thumb is that bacterial activity halves for every 7° tem-
perature drop.
The heat within a leach pile is generally determined by the following major
variables-local climate (ambient temperature, solar radiation, wind velocity),
evaporation rates, heat of reaction, placement temperatures, irrigation tempera-
ture, irrigation strategy and irrigation rates. These variables must be addressed
during design and operation of a leach pile. A favorable local climate mayelimi-
nate the effects of some of these variables, but all variables must be considered
when operating in severe climatic regions. Heat modeling can help greatly in iden-
tifying the effects of variables and addressing them during design. I8
Evaporation is an important (sometimes forgotten) heat consideration. It is
effected by factors including irrigation rates, irrigation strategy, heap and solution
temperatures and local climatic conditions. Historically, copper operations have
used who bier type irrigation systems, but high evaporation rates associated with
this system result in high temperature losses. Several newer operations (Quebrada
Blanca, Cerro Colorado and El Abra) use drip-type irrigation to avoid low tem-
peratures in the heaps, even in warm climates. In the drip irrigation Quebrada
Blanca operation, waste heat from a power plant is added to the irrigation solution
and the heap is covered with a permeable "shade" cloth to reduce the cooling effect
of evaporation. In this operation the ore is heated prior to agglomeration and near-
boiling water is added to assist in agglomeration and to maximize the agglomerate
placement temperature. In this way a solution temperature of over 20°C is main-
tained throughout the year, even in a severe climate were the annual temperature
averages only 5°C.
Production of heat due to the exothermic nature of the leaching reaction is
only significant if high quantities of sulfides are oxidized over relatively short pe-
riods of time. Heat is an integral part of bacterial leaching and must be considered
in any heap design.
Mineralogy
The mineralogy of the ore and gangue constituents affect bacterial activity as
described above. pilot testing is required on each ore to determine its suitability
for bacterial leaching. The quantity and access of sulfides to bacterial leaching are
both very important. The copper sulfide minerals are not alone in determining
bacterial leach rates, but association with other minerals such as pyrite (which
helps activity) or atacamite (which reduces activity) are important factors. This
chapter is too short to review all aspects of mineralogy as related to bacterial leach-
ing, however, involvement of a process mineralogist from project initiation is
important.
Bacterial Inoculation
In many new operations the inoculation of bacteria into the pile to help start or
to sustain bacterial populations is considered. In other operations the right envi-
ronment for bacterial growth is present and deliberate inoculation is not carried
out. No reports on the benefits of bacteria inoculation in copper bioleaching have
been found. Consideration must be given to the ability of the bacteria to adapt to
the change from synthetic media to leach media. Bacteria in a leach pile also adapt
and performance improves with time. Growth of bacteria in synthetic media does
not allow for this adaptation process. An important variable when starting a new
heap leach pile is to maintain a suitable acid level with very high-grade chemicals
since the soluble copper sulfate produces the equivalent in acid during solvent
extraction.
Iron
Bacterial leaching of copper is primarily concerned with the conversion of fer-
rous to ferric iron. There has been much discussion on the iron levels required for
efficient copper leaching. It is easy to calculate the amount of iron required to leach
a given quantity of copper if only a once-through ferrous to ferric conversion is
assumed and if only ferric sulfate leaches the copper. What complicates this calcu-
lation is that ferrous to ferric iron conversion can occur multiple times within a
pile depending on bacterial activity and the availability of oxygen. In addition, if
copper conversion is also a direct result of bacterial activity, this calculation is
unreliable. There are plants which operate successfully with iron levels in excess of
10 giL to levels below 2 giL. Iron is definitely necessary for the leaching reaction,
but the absolute levels required are dependent on the factors that affect bacterial
activity. Iron has been introduced in several start-up operations whereas natural
iron build up has been allowed in others.
Heap Operating Variables

Irrigation Distribution
Earlier reference was made to sprinkler (Whopler) and drip irrigation systems.
Most copper bacterial heap leach operations have converted to drip-line irrigation
in order to reduce the effects of evaporation. Various makes of drippers are used
and most operations have ongoing programs to test different dripper types for
drip-rate consistency and resistance to plugging and other failures. Drippers can
cause a "dome" effect that may require moving the drippers from time to time. The
drippers may be plowed into the pile in areas where freezing is common. 19 Special
equipment has been designed (Fig. 2.3) to reduce labor for drip-line installation.
Most operations maintain extensive systems for managing solution distribution,
especially when hundreds of kilometers of drip lines spread over a large heap area
are involved, each at different point in the leach cycle (Fig. 2.4).
Solution Stacking
Solution stacking refers to the practice of recycling the pile effluent back to
either the same pile or another pile at a different stage of leaching. This has been
most commonly practiced in dump leach operations of low-grade ores to increase
solution concentrations or to increase leach area without an increase in solution
flow. The use of solution stacking is accompanied by increased evaporation losses,
pond usage and pumping requirements and a decrease in solution temperatures.
Fig. 2.3. Installation of drip lines above and below the heap surface at the Quebrada
Blanca Operation, Chile.
Fig. 2.4. Typical drip irrigation distribution system at the Quebrada Blanca Operation,
Chile.
Solution stacking has generally been restricted to dump leach operations, although
several heap leach operations now use this technique to improve solution discharge
grades or to recycle higher grade ferric solutions to poorly leaching areas. The
ability to recycle some solution is worth considering during plant design.
Solution Collection
As described earlier, dump leach operations generally rely on the natural seg-
regation of coarse and fine ore to provide natural drainage below the pile. Many of
these piles were former waste dumps that are later leached, or are run-of-mine ore
which is delivered to a prepared pad or a hillside. Environmental regulations usu-
ally require installation of membrane liner below the pad before ore placement.
These pads usually have a minimum of drainage piping installation below the ore.
In heap leach operations, solution collection systems are vital to reduce the
phreatic heads and stabilize stacked leach ore. A high density polyethylene (HDPE)
liner is installed below the pad and then HDPE "Drainaflex" tubing is laid down in
various configurations to collect the leach solution. The understanding of
geotechnical issues has improved to such an extent that the phreatic head can be
calculated during design and the correct drainage piping installed. 21 During de-
sign of a solution collection system consideration must be given to the slopes of
the pad, earthquake potential, material permeability, environmental protection
and related factors.
The phreatic head should be maintained as low as economically possible in
order to allow air penetration with good pile stability. Expense should not be the
overriding consideration where drainage is concerned, as failure to collect the valu-
able product from a heap can be catastrophic.
Pad Stacking/Configuration
A wide variety of stacking equipment is available. The most common are the
cascade or grasshopper conveyors,20 a continuous crawler type,21 or a stacker fed
directly from a truck. 22 Stacking of the ore for heap leaching is critical to maintain
permeability in the heap. The cascade or truck type of stacker have generally been
favored to reduce traffic on the heap (Fig. 2.5).
There are basically two types of heap operation-the permanent heap and the
ON/OFF or dynamic heap. In a permanent heap, fresh ore is stacked on top of
leached ore, whereas in the ON/OFF system ore is removed from the prepared pad
before placing fresh ore.
In the permanent pad it is usual to prepare the top surface of the pad before
placing another level of fresh ore in order to provide a good surface for equipment,
reduce solution inventory and provide a solution bleed to reduce build up of ions.
Preparation of the various lifts is done either by rolling and compactingl1 or by
installation of a plastic liner between lifts.23 In some operations an interlift imper-
meable liner is not provided in order to allow for further recovery of copper from
the lower lifts.
The ON/OFF pad offers the advantage of a reduced preparation area, but at the
expense of removing the leach ore from the leach area. Another problem associ-
ated with the ON/OFF pad are the problems encountered when the leach time is
longer than expected. Secondary leaching of removed ore can be introduced if
suitable space is available.
Fig. 2.5. Agglomerate stacking at the Quebrada Blanca Operation, Chile.
Leach Solution Processing

The solution discharged from a leach pile ( usually called Pregnant Leach Solu-
tion or PLS) requires further processing to achieve a marketable product. This PLS
solution usually contains 1.5-6.0 gIL copper, up to 20 giL iron and a variety of other
ions (calcium, magnesium, sodium, manganese, potassium, chlorides, etc.)-all
in a high sulfate solution. Another aspect which should be considered is that dur-
ing processing, the soluble ions as well as the resulting acidity when copper is re-
moved leave the sulfate to form sulfuric acid.
Copper Cementation
The most traditional way of recovering copper from acid solutions has been by
cementation with iron. The copper is allowed to make contact with iron filings or
other fine iron products at a ratio of three tonnes of iron per tonne of copper to
form a cemented copper precipitate. The precipitate is washed, dried and sold to a
refinery for further processing. The solution is recycled to leaching. l4 This process
is simple, but is affected negatively by the cost of iron, the low return on the sale of
copper and the high manpower requirements. One of the few remaining opera-
tions using this method is the La Cascada operation in northern Chile.
Direct Electrowinning
An early alternative copper cementation, introduced by Inspiration Consoli-
dated Copper Co. in 1929, was direct electrowinning (EW) of leach solutions. This
method has also been practiced in Zaire and Zambia. The copper solution was fed
directly to the electrowinning plant to produce an impure cathode containing cop-
per, iron, zinc and other impurities. This product was sold as a subgrade metal. l5
Solvent Extraction
Copper solvent extraction (SX) in its modern form began in the mid 1960s
when General Mills (now Henkel) produced its first selective solvent reagents. The
first commercial operation was installed at the Bluebird mine, Arizona, USA in
1968. The reagents have been under continuous development since that time. The
newest reagents are highly selective against iron and can tolerate a much wider
range of copper and acid conditions. These developments have led to the produc-
tion of very pure copper solutions, with subsequent production of a highly mar-
ketable copper product from the electrowinning plants. This high quality copper
cathode is usually sold directly for the production of copper bar or wire, brass, or
other finished copper products without the need for additional refining. By the
year 2000 it is expected that 20% of western copper production will be by use of
the SX/EW process. 26
The basic copper solvent extraction circuit consists of three closed solution
loops. In the first loop, the PLS solution from leaching is fed to the extraction
portion of the SX circuit and contacted with an organic mixture of the selective
extractant. The copper is extracted from the PLS solution to the organic solution.
The two solutions being immiscible, separate and the barren solution (raffinate) is
then reused for further leaching.
The second loop, within the SX plant proper, contacts the now copper laden
(loaded organic) solution with a highly acid spent electrolyte to strip the copper
from the organic. The now barren organic flows back around the loop to the ex-
traction circuit.
Finally, in the third loop, the strong electrolyte is sent to the electrowinning
plant. High purity copper is deposited by electrolysis and the spent electrolyte is
returned back to the SX stripping section (Figs. 2.6and 2.7).
Solvent extraction provides several benefits. The acid is regenerated and only
the acid consumed by the gangue is required for leaching. The organic is very se-
lective, purifying the copper solution to EW to allow for production of a very high
quality copper product. In the SX circuit there is some carryover of iron to EW.
This iron is usually returned to SX using a small bleed stream of electrolyte. A
problem is that fine solids from leaching can form a solids/organic crud that must
be periodically removed from the SX circuit.
Fig. 2.6. Recycle loops in a Leach/SX/EW circuit.

extractant
PLS fee d
pond & pump
iron breed
from EW-..I...-~
to crud t.ank . . . -
to EW _ -..r-'L
Fig. 2.7. Typical SX flowsheet.
Electrowinning
Electrowinning is carried out in a number of acid resident cells through which
electrolyte is circulated (Fig. 2.8). The cells contain a lead alloy anode and either a
stainless steel cathode or a copper starter sheet cathode. The cathodes with their
copper deposit are typically harvested in a seven-day cycle. Copper is removed
from the stainless steel blanks either manually or with an automatic or semiauto-
matic stripping machine. The final cathode product is typically one meter square
and weighs 40-55 kg/sheet. Asmall amount of cobalt sulfate is added to the elec-
trolyte to help stabilize the anodes and a smoothing agent such as Gaur Floc is
added to reduce cathode nodularization. Electrowinning is usually carried out at
40 o-4SoC.
rr.----------------.-rr-------~
.- -.
Tankhouse Crane
r~8~o~ile-r-.
Package
& Heat
Exchanger
Commercial Cells
Cathode
Washing
Stripping
& Waxing
Fork
Lift
Sampling
Weighing
&8anding
Water
Spent Electrolyte
Cathode
Electrolyte Circulation
Product
Tank & Pump Tank & Pump
ToSX
FromSX
Fig. 2.8. Typical EW flowsheet.
Most larger modern plants use polymer concrete EW cells with stainless steel
permanent cathodes. The plant may also contain fully automatic cranes. The elec-
trolyte is distributed within the cell through a manifold located at the bottom of
the cell. 27 An automatic stripping machine is used for cathode harvesting. The elec-
trolyte is heated with a weak-to-strong electrolyte heat exchanger and also from
an external boiler. During the plating cycle, oxygen is generated at the anodes which
liberates some acid mist that must be controlled within the cell house. Acid mist
can be controlled with a layer of balls, beads or a mixture of each placed on top of
the cells. A plant may have a cross-flow forced air ventilation system. Some plants
employ a surfactant (FCllOO) to reduce the solution surface tension.'s Surfactants
must be tested to determine any potential detrimental effects on SX operation or
potential negative effects on leaching.
Biomining: Theory, Microbes and Industrial Processes
Environmental Considerations
A key feature of copper bioleaching, followed by solvent extraction and elec-
trowinning is that it is an environmentally friendly process. It produces finished
high quality copper without the need for a conventional concentrator or smelter.
There are minimal process emissions or active tailings. All process solutions are
recycled. The only environmental concerns are dust control in mining and crush-
ing, acid mist control in electrowinning and control of the overall water balance.
Dust emission in the mining operation is usually controlled by watering of all
travel ways. In crushing circuits, generally dry bag house dust collection systems
are used. Wet dust collection is generally not used because disposal or reintroduc-
tion of dust laden water is difficult in a heap leach circuit. The Zaldivar circuit,
where wet crushing is used with separation of the -100 mesh dust, is an excep-
tion. 29 Also, dust losses may represent important copper losses so a separate treat-
ment circuit (either leaching or flotation) may be required.
Acid mist in electrowinning is primarily a worker exposure hazard that is
handled by passive ventilation or a forced air ventilation system. The required
acid mist levels are being reduced by regulatory authorities and design levels of
<0.5 mglm3 should be employed. There is environmental pressure to capture and
scrub the air of acid from cell house emissions and this will certainly be one of the
next developments in SX-EW plants.
Water is a requirement of all leaching operations. The water consumption is
primarily the result of the high evaporation caused by the large heap leach areas
under irrigation. Most operations are located in dry climates and therefore have a
net negative water balance. However, several operations do experience periodic
heavy rain which may require overflow containment ponds. Water can generally
be disposed of by evaporation during dry weather periods, either from the heap or
aided by use of sprinklers on the heap. If water does need to be discharged, neu-
tralization along with removal of any SX organic compounds is required.
During the planning of modern operations a conscious decision is made at the
outset of a project to build and operate the facility to at least North American envi-
ronmental standards. Background baseline studies are undertaken and the plant
is constructed with environmentally sound design and operating features. The leach
pads use welded HDPE liners, areas are contoured so that spillage will be con-
tained and ponds and tanks have emergency containment facilities. Extensive en-
vironmental monitoring is part of normal operations.3D
Commercial Installations
This section presents three examples of operations which demonstrate the three
main types of bioleaching used: in situ leaching, dump leaching and heap leach-
ing. Little work has been carried out in the bioleaching aspects of in situ and dump
leaching, although bacteria are known to be present. Examples given are intended
to provide an overview ofleaching technology as applied commercially to the pro-
duction of copper.
In Situ, San Manuel, BHP Copperl J,32

In situ leaching represents leaching within an undisturbed ore body where no
mining activity is required. There is limited evidence that bacteria playa role in in
situ mining. On the other hand, existing secondary deposits are the result of bac-
terial action over an extended period of time. In the interest of completeness, the
in situ operation at the San Manuel operation of BHP Copper is presented to illus-
trate most of the principles of in situ leaching.
San Manuel is located in the southwestern United States about 60 km north-
east of Tucson, Ariwna. Underground mining of sulfide ore since 1955 has resulted
in substantial surface subsidence. In 1985 Magma Copper began an open pit op-
eration to recover oxide ore by acid leaching, followed by an SX/EW circuit to re-
trieve the copper from the leach solution. Owing to pit geometry, mining econom-
ics and the irregular distribution of the remaining open pit ore, an in situ alternative
was considered to recover the remaining ore using the existing SX/EW facilities.
Initial in situ mining began in 1988 using an array of wells to inject acidified
leach solution into the mineralized ground. The leaching solution was gravity-
driven through the ore, collected in the abandoned underground workings and
pumped 725 meters to the surface. It was difficult to coordinate injection sites be-
cause these depended on open-mining activity, fluid flow paths and underground
collection areas which were already in place. Further development was undertaken
to improve control of this in situ leach operation. A subsequent well-to-well in situ
leaching operation that used pumping wells to collect injected leach fluids was
implemented in 1989. After open-pit mining ceased in 1995, production continued
at a rate of 20,000 tonnes of copper cathodes per year as a result of this in situ leach
program.
Drilling was undertaken to define the distribution of copper grade. These data
were used to defme wnes favorable for in situ leaching and to design the pumping
system. The basic well pattern employed is a seven-spot cell consisting of six injec-
tion wells distributed around a central production well. Individual cells are linked
by shared injection wells along the central axis of a linear well pattern. The depth
and screening of wells is based on subsurface grade and structure. The wells were
drilled by reverse circulation with a 3.8 cm casing diameter and the production
wells are 15.25 cm in diameter. Wells were drilled to 100-150 meter depth and cased
with PVC. Fluid flow rates in injection and production wells are logged hourly and
the system shows a 13.5% fluid loss.
The leach solutions are a combination of raffinates from the ongoing heap op-
eration and the in situ operation. During just over a year the acid content of the
injection fluid was maintained at 26 gIL. Copper concentration in the production
fluid started at about 2 gIL and then plateaued at 1.1 gIL for a leach time of over two
years. Overall recovery is almost negligible at starting grades of <0.2% copper, but
rises to above 80% for grades of over 1.5% copper. Overall recovery of a test section
averages about 60%. This depends also on the ratio of oxide to sulfide minerals
present. A significant effect is formation of gypsum and jarosites which is caused
by the mineralization of the gangue materials combined with the high dissolved
solids content of the leach solution.
This in situ operation is producing cathode copper from a low-grade irregular
ore body that would have become waste. The low operating costs combined with
an existing facility continue to provide an economic process. The future will allow
for development of in situ leaching for new low-grade deposits with minimal dis-
turbance of the surrounding areas.
Dump Leaching-Baja Ley Plant at the Chuquicamata Division

of Codelco, Chile'S
Most traditional dump leach operations have involved the leaching of existing
waste piles to extract low-grade ore that was previously considered waste. Several
operations for leaching low-grade run-of-mine ore dumps have been designed. A
good example of an operation that has a "designed" dump leach rather than an
"after the fact" waste dump, is the Baja Ley operation of the Chuquicamata Divi-
sion of Co delco, Chile.
The first studies for this facility started in 1970, with conceptual and basic en-
gineering design commencing in 1983. Construction began in 1991 with plant start-
up in 1993. The plant was designed for 15,000 tonnes of cathode production from a
0.35% copper run-of-mine finger dump operation at a cost of U.S.$40 million.
The ore is placed with mine trucks in 35 meter high 'fingers' 150 meters wide.
Leach irrigation modules of 70 x 35 meters, using drip irrigation are established on
top of these fingers (Fig. 2.9). The leach cycle consists of preconditioning, leach-
ing, rest, conditioning, and wash cycles over a 78-week period. The solution used
is the raffinate from the adjoining SX/EW plant. Acid consumption is 9.5 kglkg of
copper produced. The resulting PLS solution is collected at the bottom of a valley
in a 34,000 m 3 PLS pond. This solution is then transferred to a modern SX/EW
plant employing stainless steel cathodes and a semiautomatic stripping machine.
The plant employs 27 workers and produces copper at less than U.S.$0.40 per
pound. Overall recovery is estimated at 20%. This plant is an example of a success-
ful design run-of-mine dump leach operation on a low-grade ore, producing fm-
ished copper at low cost.
Heap Leaching-The Quebrada Blanca Operationu ,Jo,J3

The Quebrada Blanca operation is located in northern Chile in the Alti plano
desert at 4,400 meters above sea level. It is a stand-alone crushing, bacterial leach,
solvent extraction and electrowinning facility. The operation includes its own power
plant and obtains water from a Salar 38 km from the mine. This facility processes
17,300 tonnes/day of sulfide ore to produce 206 tonnes of London Metal Exchange
(LME) grade cathode copper per day (Fig. 2.10).
The property was explored in the late 1970s, drilled and an underground ex-
ploration adit was developed which provided valuable samples for pilot testing.
Conventional processing was considered to be uneconomical. In 1988 the property
was offered in an international tender and pilot studies were initiated with SMP
Tecnologia S.A. After feasibility studies, a decision to proceed with production
was made late in 1991. Plant start-up was in January 1994, with the first cathode
produced in August 1994. The Quebrada Blanca operation is based on a supergene
enriched porphyry copper (primarily chalcocite) deposit with a reserve of 91 mil-
lion tonnes at 1.3% copper.
Sulfide ore from the mine is crushed in three stages to 100% minus 9 mm. The
crushed ore is fed to a 1500 tonne surge bin that includes a steam-to-ore heat ex-
changer to increase the temperature of the ore. The ore is then agglomerated in
rotating drums with 5-7 kglt of sulfuric acid and 85°C hot water to obtain about
10% moisture in the final agglomerate. The ore is discharged from the agglomera-
tion drums at 22°-24°Cand is conveyed to the stacking circuit. The agglomerate is
moved from a tripper conveyor to a series of shiftable conveyors which transport
the ore to a shuttle conveyor and stacker. The stacker distributes the ore uniformly
in an arc to form a 6-6.5 m high continuous pile or sector, 85 m in width.
Fig. 2.9. Dump leach operation at the Baja Ley plant.
Fig. 2.10. The Quebrada Blanca open pit mine and leach/SX/EW operation.
Fig. 2.11. Heap leach operation at the Quebrada Blanca mine site, Northern Chile.
The leach pile is lined with 60 mm HDPE welded liner. A network of 4" dia.
Drainaflex pipe is placed at 2 m intervals on the liner or on compacted leached ore
(after the bottom lift is completed). Immediately after the ore is stacked, two net-
works of drip irrigation lines are installed on top of the pile, one on the surface and
the other 20 cm below surface. The leach area is irrigated at 0.1-0.141/min/m 2
of 7.0 giL sulfuric acid raffinate solution from solvent extraction. The raffinate
solution is heated to 28°C in a set of heat exchangers prior to being fed to the drip
lines. Heating of the raffinate solution also serves to cool the generators in the
power plant.
The 3.5 gIL Cu PLS solution at 23°C is collected in the Drainaflex lines and flows
to a series of collection header lines that are buried within the leach pile. The leach
circuit also includes several innovations to help improve leaching under the severe
weather conditions found at this site. Aseparate series of air lines is installed be-
low the heap to distribute air from a series of air fans. Nutrients are added to the
leach solutions to maintain adequate ammonia and phosphate levels for bacterial
activity. The top of the pile is covered with shade cloth to reduce evaporative cool-
ing. This operation carries out extensive heat, oxygen, solids and liquid monitor-
ing to help optimize leaching. This monitoring includes on-site bacterial activity
measurements. Measurements from residue samples indicate that a recovery of
better 80% of total copper is obtained. The layout of the leaching operation is shown
in Figure 2.11.
Bialeaching of Copper 41
Fig. 2.12. Copper cathode harvesting at Quebrada Blanca.
Over 3,000 m 3/hr of PLS solution at over 20°C and at 3.0-3.5 giL copper is for-
warded to the solvent extraction facility which consists of three process trains in
parallel, each with two extraction stages followed by one stripping stage. The PLS
is contacted in a counter-current stream of organic solvent to extract the copper in
two stages with a recovery efficiency of 93%. The solvent consists of up to 13.5%
LIX 984 carried in a low aromatic kerosene (SX 12). The loaded organic solvent
reports to a loaded solvent surge tank, while the aqueous raffinate is returned to
leaching. The loaded solvent is advanced to the stripping stage where it is con-
tacted with 35 giL Cu lean electrolyte in two mixer stages to transfer the copper to
the aqueous electrolyte solution. The barren organic solvent is advanced from strip-
ping to extraction, while the 49 giL Cu rich electrolyte is discharged from the strip
stage to feed two flotation columns in series, followed by a garnet/anthracite pres-
sure fIlter for removal of entrained organic solvent.
The EW cell house contains 264 polymer concrete cells, each with 60 perma-
nent stainless steel cathodes and 61 lead/tin/calcium alloy anodes. The cell house
is operated at a nominal 260 A/m2 for a seven-day cathode growth cycle. One-sixth
of the cathodes are harvested daily using an automatic WENMEC stripping ma-
chine (Fig 2.12), and the stripped blanks are returned to the cells. A bottom wax is
applied to each cathode. Guar is added to reduce nodule formation and 100 ppm of
cobalt sulfate is used to reduce anode corrosion. All water used in the cell house is
treated in a reverse osmosis plant to reduce chloride levels. The stripped LME grade
cathodes, 45-50 kg each, are stacked in 2.5 tonne bundles and strapped for truck
haulage to Iquique for shipping to the final destinations.
This operation produces high quality LME Grade A copper at around U.S.$0.50
per pound from an initial capital investment of U.S.$360 million. It demonstrates
an advancement of the application bioleaching of copper at a large scale to a re-
gion with adverse climatic conditions.
Conclusions
This chapter has reviewed the many elements that must be considered in plan-
ning, designing and operating a successful copper bioleaching operation. Outwardly,
this technology appears very simple, but there are many factors that must be con-
trolled to keep the bacteria alive and active. Existing operations and research or-
ganizations are continuing to advance the knowledge base required to make this
technology more reliable. The few operations included in this chapter demonstrate
that ore previously considered to be waste can indeed be mined economically us-
ing bioleaching technology.
The future will see this technology expand to include more difficult-to-leach
ores such as chalcopyrite and the treatment of copper concentrates. Bioleaching is
a low cost technology that is environmentally friendly and it produces a high qual-
ity finished product. It is an alternative that must be considered whenever a new
ore body is considered for development.
References
la. Bruynesteyn A. Bacterial leaching; its potential impact upon the Canadian non-
ferrous metals industry. 86 th Annual General Meeting of the CIM. April 17, 1983.
lb. Sullivan JD. Chemistry of leaching chalcocite, TP-473. u.S. Bureau of Mines. 1930.
lC. Thomas G, Ingraham TR, MacDonald RJC. Kinetics of dissolution of synthetic
dignite and chalcocite in aqueous acidic ferric sulfate solutions. Canadian Metal-
lurgical Quarterly 1967; 6:281-291.
Id. Whiteside LS, Goble RJ. Structural and compositional changes in copper sulfides
during leaching and dissolution. Canadian Minerologist. 1986; 24.
Ie. Goble RJ. Copper sulfides from Alberta: Yarrowite CU9S8 and Spionkopite
CU39S28. Canadian Minerologist. 1980; 18:511-518.
2. Scott DJ. The mineralogy of copper leaching: concentrates and heaps. Copper '91,
Copper Hydrometallurgy Short Course. Ottawa, June 1991.
3. Scott DJ. The mineralogy of copper leaching: concentrates and heaps. Copper '95,
Copper Hydrometallurgy Short Course. Santiago, November 1995.
4. Bruynesteyn A, Duncan DW. Effect of particle size on the microbiological leach-
ing chalcopyrite bearing ore. Solution Mining Symposium 1974.
5. D'Andrea D, Chamberlain PG, Fletcher LR, Ground characterization for in situ
copper leaching. Proceedings of the Las Vegas Symposium on Leaching and Re-
covering Copper from As-Mined Materials, February 1980.
6. Farias L et al. Acid leaching of copper ores. Copper '95, Copper Hydrometallurgy
Short Course. Santiago, November 1995.
7. Woodcock JT. Copper waste dump leaching. Proceeding Australian Institute Min-
ing and Metallurgy. Dec 1967.
8. Domic EM. Resultados tecnico-economicos de la opera cion industrial del proceso
TL en chile, Simosio Internacional sobre la Actual Tecnologia del Cobre,
Bucaramanga, Colombia, November 1982.
9. Jo M, Bustos S, Espejo R et al. Bacterial thin layer leaching of copper sulfide
ores. Proceeding of Copper '91 Symposium. Ottawa, June 1991.
10. Montealegre R, Bustos S, Rauld J et al. Copper sulfide hydro metallurgy and the
thin layer bacterial technology of Sociedad Minera Pudahuel. Proceedings of Cop-
per '95 Symposium. Santiago, November 1995.
11. Schnell HA. The Quebrada Blanca operation, SME, March 1996.
12. Bryner LC, Beck JV et al. Microorganisms in leaching sulfide minerals, Industrial
and Engineering Chemistry, Vol 46, 1954.
13. Herrera MN, Wiertz JV et al. A phenomenological model of the bioleaching of
complex sulfide ores Hydrometallurgy Vol 22, Elsevier Science Publishers, 1989.
14. Bartlett RW. Simulation of ore heap leaching using deterministic models. Hydro-
metallurgy Vol 29. Elsevier Science Publishers, 1992.
15. Anon. Sulfuros de baja ley aporta 15 mil T/afio a Chuquicamata, Mineria Chilena,
July 1994.
16. Moodry RP. Compressed air injection into a sulfide leach dump, AS 116, August
1976.
17. Trivedi NC, Tsuchiya. Microbial mutualism inleaching of Cu-Ni sulfide concen-
trate. International Journal of Mineral Processing, Elsevier Scientific Publishing
Co., 1975.
18. Montealegre R, Bustos S et al. Application of the thin layer process to Quebrada
Blanca ores. Biohydrometallurgical Technologies, The Minerals, Metals and Ma-
terials Society, 1993.
19. Anon. Quebrada Blanca-the first sx/ew project at over 4,000 m altitude. E&MJ,
February 1995.
20. Clifford D. Stacking systems in heap leaching. Mining Magazine, August 1996.
21. Anon. The big heap. World Mining Equipment, November 1996.
22. Pino F. Division salvador of Codelco Chile introduces the Wsx/ew process as a
new line of production. Proceedings of Copper '95 Symposium. Santiago, Novem-
ber 1995.
23. Anon. Mina 10 aquirre sociedad minera pudahuel Itda. y cia. C.p.a., Minera
Chilena, July 1983.
24. Fletcher AW. Copper recovery from low-grade ore by bacterial leaching. In: Mi-
crobiological Aspects of Metallurgy, chapter 8, 1970.
25. Anon. Trends and implications of the continued developments of sx/ew copper
production. Pin cock, Allen and Holt, Inc., March 1990.
26. Anon. Predicted sx/ew copper production. Mining Journal, February 1996.
27. Jenkins JG, Eamon MA. Plant practices and innovations at Magma Copper Com-
pany, San Manuel. Proceedings of Copper '91 Symposium. Ottawa, June 1991.
28. Davies JA, Hopkins WR. Recent developments in electrometallurgical tankhouse
environmental control, CIM Bulletin, June 1994.
29. Anon. Minera zaldivar, Minera Chilena, August 1995.
30. Schnell HA. Quebrada Blanca and the environment. Proceedings of Copper '95
Symposium. Santiago, November 1995.
31. Beane R, Ramey D. In situ leaching at San Manuel porphyry copper deposit. Pro-
ceedings of Copper '95 Symposium. Santiago, November 1995.
32. Ramey D, Beane R. In situ project evaluation; magma copper's approach. Pro-
ceedings of Copper '95 Symposium. Santiago, November 1995.
33. Schnell HA. The Quebrada Blanca project, Copper '95, Copper Hydrometallurgy
Short Course. Santiago, November 1995.
34. Lynch AJ, Taylor A, Avendafio C. Solvent extraction boom in Latin America,
E&MJ, December 1994.
35. Anon. Title page, Revista Innovacion, University of Antofagasta, May 1995.
CHAPTER 3
The BIOX· Process

for Biooxidation of Gold-Bearing
Ores or Concentrates
David W. Dew, Ellen N. Lawson and Jennifer L. Broadhurst
Introduction
G ENCOR S.A. Ltd. has pioneered the commercialization of biooxidation of re-

fractory gold ores. Development of the BIOX® process started in the late 1970S
at GENCOR Process Research, in Johannesburg, South Africa. The early work was
championed by Eric Livesey-Goldblatt, the manager of GENCOR Process Research
who directed pioneering and innovative research into bacterial oxidation of re-
fractory gold ores prior to cyanidation. This work was driven by the need to re-
place Fairview's outmoded Edward's roasters, which at the time were seriously
contributing to pollution in the Barberton area.
In 1984, a pilot plant was commissioned at GENCOR Process Research to treat
Fairview flotation concentrate and was operated for some two years. Its success led
to the decision to construct an industrial plant at Fairview to treat 40% of the
mine's production. This plant was commissioned in October 1986. In 1991, the
Fairview BIOX® plant was extended to treat the total production of the mine,
amounting to 40 tonnes/day of flotation concentrate, containing up to 160 g/t of
gold. The Edward's roasters were finally shut down and demolished, removing the
threat of future environmental pollution.
The BIOX® process has been both a technical and economic success and is fast
becoming the norm in refractory gold ore processing. Biooxidation offers real eco-
nomic advantages over roasting and pressure oxidation, and the BIOX® process
has been clearly demonstrated to be environmentally acceptable and robust enough
to be operated in remote areas.
Within the GENCOR group, a second plant was built in 1986 at Sao Bento
Mineracao in Brazil, and in 1988 the company took the unusual step (for a South
African mining house) of entering the international technology licensing busi-
ness. This decision formed part of GENCOR's thrust to establish itself more strongly
internationally.

The first licensed plant was commissioned at the Harbour Lights mine, West-
ern Australia in 1992. Asarco Australia also decided to use the BIOX® process at its
Wiluna operation in Western Australia. This plant, with a capacity of ll5 tonnes of
concentrate per day, was successfully commissioned in 1993. The plant was to be
expanded in 1996 to treat 155 tonnes/day.
The BIOX® process took a quantum leap with the construction of a 720 tonne
per day BIOX® plant at the Ashanti Goldfields Company's Obuasi mine, at Obuasi
in Ghana. The plant design consisted of three independent BIOX® modules. Con-
struction was completed in January 1994 and the plant was successfully commis-
sioned one month later. A fourth BIOX® module was constructed and commis-
sioned in 1995, to increase the plant capacity to 960 tonnes/day.
Laboratory and pilot scale testwork to evaluate the amenability of ores to
biooxidation remains a major activity at GENCOR Process Research. To date, over
150 ore samples have been subjected to amenability tests at laboratory scale and
over 12 integrated pilot runs have been completed, to provide information for com-
mercial plant designs and project feasibility studies.
Process development is ongoing in order to improve the performance of
biooxidation with respect to reducing operating costs and improving process per-
formance and operability. A major development in recent years has been the use of
the LIGHTNIN A315 fluid foil impeller instead of the conventional radial flow tur-
bines for gas dispersion and solid mixing, resulting in a 30% power saving for the
operation of large scale bioreactors. The development of rate models describing
biooxidation and models describing gas liquid mass transfer in bioreactors has
been undertaken, to allow improved optimization of the process design
specification.
New development work currently in progress at GENCOR Process Research
will develop bioleach processes for the treatment of nickel and copper ores for
metals recovery. Recently, the trade name BioNIC has been registered as the pro-
cess name describing the bioleach process for nickel recovery from sulfide
concentrates.
The BIOX- Process Flowsheet and Plant Operating Practice

BIOX- Process Flowsheet
A typical GENCOR BIOX® flowsheet for treatment of a refractory gold ore con-
centrate is shown in Figure 3-1. The flowsheet consists of concentrate feed to the
biooxidation plant, biooxidation product to solid/liquid separation by counter-
current decantation washing, washed thickener under flow to cyanidation, thick-
ener overflow to neutralization, and the neutralized product to disposal on a tail-
ings dam.
The feed concentrate to biooxidation would typically be milled to a d so of
75 11m, with a solids content in the slurry feed of 20% by mass.
Generally, the pulp residence time in the biooxidation plant will be 4 days de-
pending on the oxidation rate achieved, which is a function of the sulfide content
and mineralogical composition of the concentrate. The biooxidation plant is
configured to include a primary and secondary oxidation stage, as shown in
Figure 3-1. Depending on the sulfide oxidation rates achieved, the pulp residence
time will be 2 days per stage. Both stages will normally consist of three equally
sized tanks, with the three tanks operating in parallel in the primary stage and in
Fig. 3.1. A typical ;!
BIOX® process flow- Water
'"b:J
sheet. '1(H20}~ Cooling Tower
ltitl '"..~ ~.
"1::1
• ~Nutrient
<3
ro MakeUp
~ ~
Stock Tank '"
Feed
~ '0'
...
b:J
o·
o
Secondary Oxidation Tanks ~
it
l:>
g.
:::!
~
~
o
is:
~
...'"l:>
~.
CCD
Wash Water
o
a
...o
l)
o
:::!
~
Lime ~
i:l
~
Neutralization
~
series in the secondary stage. This configuration allows a slurry residence time of
2 days per tank in the primary stage, compared to 0.67 days per tank in the sec-
ondary stage. A longer residence time in the primary stage is necessary in order to
allow a stable bacterial population to be established and to prevent bacterial wash-
out. Once the bacterial population has been created and bacterial attachment to
the concentrate has occurred, a shorter residence time per tank can be tolerated in
the secondary tanks, where sulfide oxidation is completed.
The operating slurry solids content is determined primarily by the sulfide con-
tent of the concentrate feed and the rate of oxygen transfer necessary to maintain
the required rate of sulfide oxidation as described by Hansford and Bailey.l,2 A
solids content of 30% by mass may be tolerated, if the ore or concentrate being
treated contains 5% sulfide or less.
The BIOX® culture consists of a mixed culture of mesophilic bacteria, as de-
scribed later in this chapter. These bacteria operate over a temperature range from
25°-45°C.The pulp temperature of commercial plants is controlled at 40°-45°C which
allows maximum rates of sulfide oxidation to be obtained while minimizing cool-
ing requirements.
Nutrients nitrogen, phosphorous and potassium are supplied to promote bac-
terial growth, with standard addition rates being:
Nitrogen 2 kg/tonne concentrate
Phosphorous 0.3 kg/tonne concentrate
Potassium 0.9 kg/tonne concentrate
The dosages required may be reduced on commercial plants, depending on the
composition of the ore or concentrate being treated.
The biooxidation product is washed in a three-stage counter-current decanta-
tion circuit. The washed product would normally contain less than 2 gIL total iron
with an acid pH of about pH 2-4. Iron removal by decantation washing is neces-
sary to promote gold recovery on cyanidation of the biooxidation product.
The final thickener overflow liquor is neutralized to pH 7.0-8.0 in a two-stage
neutralization plant, to produce a stable solid precipitate of iron and arsenic, for
safe disposal on a tailings dam. The neutralization process as well as the engineer-
ing design and process requirements for the BIOX® process are described in detail
in a later section of this chapter.
Plant Operating Practice
Description of Commercial Plants

The five commercial BIOX® plants mentioned in the introduction, have been
described in a number of papers presented at international conferences.3- 6 A short
summary of the five operations is presented below.
a. Fairview BIOX® plant

The Fairview demonstration BIOX® plant, commissioned in 1986, was designed
to treat 10 tonnes/day of concentrate with a liquid/solids (L/S) ratio of 8:1 for the
feed and a residence time of four days. As a result of ongoing development work,
the LIS ratio of the concentrates in the feed to the plant was decreased to 4:1 and
the feed rate increased to 17 tonnes/day.
The BIOX'" Process for Biooxidation of Gold-Bearing Ores or Concentrates 49
Table 3.1. Summary ofFairview BIOX- plant performance for the period 1988
to 1992
Yearly Average
1988 1989 1990 1991 1992
Concentrate feed (t/month) 26 3 265 350 712 853

Concentrate gold content (glt) 99 108 109 127 150
Concentrate sulfur content (% S) 27·4 24·5 23·1 22·9 19.1
Gold recovery (%) 93·0 93·2 92·5 93-4 94·5
Pulp feed LIS 6·5 4·8 4·0 4·0 4·0
Residence time (days) 4·0 4·0 4·0 4·0 4·0
Operating temperature (OC) 41·5 40·9 40.3 40.0 41.0
Plant availability (%) 99 98 98 98 97
Table 3.2. Typical ore and concentrate analyses for Sao Bento
Goldglt Sulfide % Arsenic %
Run-of-mine ore 7·5 4·5

Concentrate 30.0 18.0
The total concentrate production at Fairview amounts to 35 tonnes/day. With

the satisfactory performance of the BJOX'" plant, it was decided to expand the dem-
onstration plant to treat the full 35 tonneslday of concentrate. The expanded plant
was commissioned during the first half of 1991 and the Edwards roasters were de-
commissioned during June 1991.
After 10 years of operation, the BJOX'" plant at the GENCOR Fairview Mine can
be considered to be a commercial success. The key variables in the process have
been established to be air dispersion, LIS ratio of pulp feed, heat removal and the
stability of the arsenic-contaminated residues. The bacterial strain has proved to
be extremely hardy and can withstand the normal fluctuations encountered under
commercial plant operation. The plant can be operated by unskilled personnel,
and a very high availability is maintained.
b. The Sao Bento BIOX- Plant

The GENCOR Sao Bento Mine (SBM) in Brazil was commissioned in 1986 and
employs a pressure oxidation circuit to treat the refractory gold ore concentrate.
The concentrate typically consists of pyrite (16%), arsenopyrite (38%) and pyr-
rhotite (46%).
The Sao Bento pressure oxidation plant was originally designed to treat 240
tonnesl day of concentrate. The remainder of the circuit was designed to treat 480
tonnes/day of concentrate. With the decision to double treatment capacity of the
oxidation plant, it was decided to investigate the possibility of incorporating a
biooxidation plant in the circuit to preoxidize the sulfur, thus increasing the pres-
sure oxidation treatment capacity.
GENCOR Process Research were provided with samples of concentrate from

SBM on which specific process development work was undertaken during the late
1980s.AlI the bioresearch was performed in-house and included bacterial amena-
bility, toxicity, effluent neutralization, and pressure oxidation laboratory tests, as
well as a continuous biooxidation pilot plant campaign. This campaign confirmed
the reactivity of pyrrhotite and arsenopyrite to biooxidation, permitting a reten-
tion time in the biooxidation circuit that was significantly shorter than that at
Fairview.
Culminating from the successful pilot plant campaign, a decision was taken to
install a single stage BIOX® tank as a demonstration unit. This module was de-
signed to treat 150 tonnestday of concentrate containing 18.7% sulfide sulfur and
achieve an overall sulfide oxidation of 30%.
Hot commissioning commenced during December 1990, with continuous feed
achieved by February 1991. From October 1991 the plant has been in continuous
operation, after modifications to the cooling coil clamping arrangement and the
water circuit. Pressure oxidation performance was, however, not negatively influ-
enced by the BIOX® product; indications were, in fact, that the performance
improved.
Results from Table 3.3 indicate that although the planned feed rate of 150 tonnest
day of concentrate has not been achieved, the plant has performed well above de-
sign (144%) based on tonnes of sulfide oxidized per day. The pilot plant studies
predicted that the sulfide oxidation rate would decrease when the residence time
falls below 4 days. However, plant operation has indicated that sulfide oxidation
rates of above 70% can be achieved at a 1.5 day residence time. The installation of a
second tank to increase the Sao Bento plant oxidation capacity by 50% was com-
pleted in December 1994.
c. The Harbour Lights BIOX® Plant

The Harbour Lights mine is situated near the town of Leonora in Western Aus-
tralia. A refractory sulfidic orebody was mined with the gold associated mainly
with the arsenopyrite and pyrite. Conventional sulfide flotation was used to pro-
duce a concentrate containing 80 g1t Au, 18.0% sulfur and 8% arsenic. Flotation
concentrate was cyanided on site, with the residue, containing 40 g1t gold, stock-
piled. The GENCOR BIOX® technology was licensed to Harbour Lights for the treat-
ment of stockpiled concentrate. A detailed process design package was provided
to an Australian engineering company for the construction of the 40 tonne per day
plant.
The project commenced during June 1991 and hot commissioning started on
December 16,1991. Continuous operation commenced on February 1, 1992 and the
first gold was produced from the plant in early March 1992. Full production was
achieved by June 1992. A plant performance test run was completed during Octo-
ber 1992 to demonstrate that the guaranteed gold recovery of 92% could be achieved.
The successful commissioning of the Harbour Lights BIOX® plant again con-
firmed that the process is commercially viable. BIOX® plants are easy to operate
(only one person is needed per shift) and the GENCOR bacterial strain can adapt
to new environments within a relatively short period. The mining operation at
Harbour Lights was completed in 1994 and the plant equipment has been dismantled
and sold.
;i
Flotation Tailings <II
Mllhng Flotation
b:I
~.
~
~
~ ~ ~ ~ '"~
FlotatIOn Concentrate ~ RegnndMl1i 'C'
...
Vent Tailings Dam
b:I
o·
c
~
i:t
~
o·
;::
~
C'l
c
NeutralizatIOn l:;:
b:,
<II
;:,
....
~.
o
a
c
....
Q
;::
BIOX®
~
::t
i:l
~
Acid Liquor
Fig. 3.2. Sao Bento flowsheet incorporating BIOX®. 'C

Table 3.3. Summary of BIOX® plant performance at Sao Bento
Operating Performance
Concentrate feed (tpd) 27 42 51 75 105 120

Residence time (days) 3·8 2·4 1.9 1.4 1.0 0·9
Feed Sulfide (%) 22.1 23·7 24·0 22·3 18.2 17.6
Sulfide oxidation (%) 81.4 79·4 73·6 72.5 71.9 68.8
Design parameters: Concentrate feed per day, tonne (tpd) - 150

Residence time, days 0.6
Sulfide, % in feed
Sulfur oxidized per day, tonne
Sulfide oxidation, % - 30.0
Table 3.4. Summary of BIOX® performance test results at Harbour Lights
Day Gold Dissolution %

Plant Laboratory
91.83 94·43
2 92.58 92.82
3 92·96 92. 22
4 91.27 91.49
5 9 2.45 9 2. 8 4
6 92.76 91.98
7 93-37 92.62
Average 92·46 92.63
Target 92.00
Tonnes treated: target/week - 280 (40 tons per day), actual - 286.
d. The Wiluna BIOX® plant

After a five year program of metallurgical testwork, Asarco Australia decided
early in 1992 to use the BIOX® process for their refractory gold plant at Wiluna,
Western Australia. They considered various other process options, including whole-
ore-roasting, two-stage concentrate roasting and pressure oxidation, but finally
selected BIOX® on the grounds of better gold recoveries achieved, lower capital
and operating costs, a shorter permitting and construction period, and environ-
mental compatibility. In the Wiluna sulfide ore, approximately 95% of the gold is
locked within the sulfide lattice, mainly in arsenopyrite and to a lesser extent in
pyrite and stibnite.
The Wiluna BIOX® plant was designed to treat 115 tonnes/day of flotation con-
centrate with an average grade of 24% sulfide and 10% arsenic. Each of the six
equidimensional reactors has a live volume of 468 m 3, giving an overall retention
time of 5 days at the design feed rate. The plant was successfully commissioned in
early 1993. Due to insufficient concentrate tonnage, the performance guarantee
The BIOX e Process for Biooxidation of Gold-Bearing Ores or Concentrates 53
Table 3.5. Capital and operating cost comparison for a refractory gold
project in Canada
SO Tonne/Day Concentrate Plant

Operating Costs Capital
C$/Tonne C$ Million
Pressure oxidation 176 17·2

Biooxidation 97 6.0
The major elements of capital and operating costs are determined by the following:
- Sulfide content of ore or concentrate determines aeration and agitation requirements
(power input) and heat removal surface area.
- Required sulfide oxidation for optimum gold recovery.
Table 3.6. Cost comparisons for oxidation plants
Capital Expenditure Operating Costs
BIOX® 1.00
Roasting 1.11
Pressure oxidation 1.14
test could only be performed in December 1993. During the seven-day test, an av-
erage sulfide sulfur oxidation of 96.5% was achieved, substantially exceeding the
guarantee of 93.6%. The plant has operated within the design expectations and an
expansion of the plant began in 1995 to allow treatment of up to 155 tonnes/day.
e. Ashanti's Sansu BIOX® Plant

The Sansu sulfide treatment at Obuasi in Ghana formed a major part of Ashanti
Goldfields Company's expansion program and increased the mine's gold produc-
tion to over 1 million ounces per annum in 1995. A major breakthrough came for
BIOX® when, after more than two years of process studies, it was selected as the
optimum process for the project. The 720 tonnes/day plant constituted a signifi-
cant scale-up from the 115 tonnes/dayWiluna plant.
The Sansu BIOX® plant consists of three biooxidation modules, with six 900
m 3 reactors per module. At design capacity, each reactor module will treat 10 tonnes
of concentrate per hour with an average grade of 11.4% sulfide and 7.7% arsenic.
The plant has been designed with sufficient flexibility to treat concentrates from
both semioxidized and primary sulfide ores, having widely different mineralogy
and oxidation characteristics.
The Sansu BlOxe plant was successfully commissioned during February 1994.
Guaranteed performance was met from May 1994. As a result of the performance,
no guarantee test run was required and the plant was accepted on August 12, 1994.
Sulfide oxidation rates in excess of 94% are being achieved, against the guaranteed
91%. The plant has operated exceptionally well since commissioning and in Sep-
tember 1995, a fourth module was commissioned, increasing the plant tonnage to
960 tonnes/day.
Cost Comparison
Costs are central to any decision on choice of process, but to quote generalized
comparisons can be misleading as cost figures are site-specific. Obviously, given
that GENCOR is operating second generation plants and is spending a significant
amount of research funding on bacterial oxidation, cost comparisons have been
done.
These figures tend to compare well with other studies. The capital cost of a
BIOX® plant is significantly less than pressure oxidation at this scale and operating
costs are also lower. Recent cost estimates completed for a Canadian refractory
gold plant indicated costs as shown in Table 3.5. The sulfur content of the concen-
trate is 24%.
A study was conducted by GENCOR to estimate capital and operating costs for
a typical flotation concentrate considering the three conventional oxidation pro-
cesses: pressure oxidation, biooxidation and roasting. For the roasting option, it
was assumed that gas emitted will be scrubbed to recover arsenic and produce
sulfuric acid. Sales revenue from sulfuric acid was taken into account, but no ar-
senic sales were considered.
The concentrate used in the exercise had the following analysis: Au, 100 gttonne;
S, 15%; As, 4%. A general conclusion from the exercise is shown in Table 3.6.
The comparison indicated that biooxidation, using BIOX® technology, offered
considerable capital and operating cost savings over roasting and pressure
oxidation.
Operating Plant Status and Future Operations

The concentrates treated in the plants consist mainly of pyrite, arsenopyrite
and pyrrhotite. Sulfide sulfur content of the concentrates vary between 12% and
24%, with the arsenic content varying from 4-14%.
The evaluation and pilot scale testing of biooxidation for implementation on
commercial projects remains a major activity at GENCOR Process Research. Sev-
eral projects are currently at the feasibility stage or await finalization of financial
backing. It is expected that within the next five years the number of commercially
operating BIOX® plants will have increased to seven or more. The continuing de-
velopment work supported by GENCOR will also result in the application of
biooxidation for recovery of base metals. The BioNIC process is currently being
evaluated in a 250 kg/day demonstration plant at GENCOR Process Research. Com-
mercial implementation of the BioNIC process is expected by 1999.
The BIOX® Bacterial Culture

The heart of the BIOX® process is the bacteria; they are a mixed population of
Thiobacillus ferrooxidans, Thiobacillus thiooxidans and Leptospirillum ferrooxidans.
They are all chemolithotrophic Gram -negative acidophiles and are motile by means
of a single polar flagellum.? T. ferrooxidans and L. ferrooxidans oxidize reduced
iron compounds for energy, whereas T. thiooxidans obtains energy from reduced
sulfur compounds, as can T. ferrooxidans also. The thiobacilli are straight rods
The BIOX· Process for Biooxidation of Gold-Bearing Ores or Concentrates 55
Table 3.7. A summary of the commercial operating BIOX® plants, at the date
ofpublication
Mine Country Plant Reactor Date of

Tonnes Capacity Tank Size Commissioning
Concentrate per Day m3
Fairview South Africa 40 90 1991

Sao Bento Brazil 150 550 1990
Harbour Lights >4- Australia 40 160 1991
Wiluna Australia 155 480 1993
Ashanti Ghana 960 900 1994
>4- Mining operations were completed in 1994 and the plant equipment has been sold.
Table 3.8. Typical distribution of sulfide minerals in concentrates
Concentrate Pyrite % Arsenopyrite % Pyrrhotite %
Fairview 70 30
Sao Bento 16 38 46
Harbour Lights 73 27
Wiluna 57 43
Ashanti 18 46 36
with a diameter of 0.3-0.6 )lm and a length of 1-3.5 )lm. Although Leptospirillum
has similar dimensions, it is vibroid when young and spiral when 0lder.8 It is also
highly motile. Leptospirillum is not related to the thiobacilli and neither are T.
ferrooxidans and T. thiooxidans related to each other as indicated by DNA-DNA
similarity values.
The bacteria collaborate in oxidizing metal sulfides such as arsenopyrite and
pyrite and facilitate this by becoming attached to the ore. When bacterial attach-
ment was examined by the DAPI staining technique,9 it was found that they tend
to attach specifically to the metal sulfide. This has been confirmed by other inves-
tigations. 10- 12 There has been disagreement about the relative importance, and
mechanism of direct and indirect leaching, where indirect leaching involves the
production of ferric sulfate by the bacteria, which causes chemical breakdown of
the metal sulfide. Attachment tends to support the idea of direct bacterial involve-
ment. Recently, it has been claimed that all leaching is indirect,13 The specific rate
of oxygen utilization in the presence and absence of pyrite was found to be the
same and to follow the same kinetics. From this it was concluded that pyrite is
oxidized by means of an indirect mechanism, in which it is leached chemically by
ferric iron and the role of the bacteria is to generate ferric iron and maintain a high
redox potential in the system.
These conclusions were based on an investigation where the bacteria consisted

mainly of L. ferrooxidans with less then 5% being T. ferrooxidans. Since
Leptospirillum can only oxidize ferrous iron, it is likely that pyrite breakdown by
this bacterium does occur by the indirect leaching mechanism. However, this is
not necessarily true for T. ferrooxidans which can also oxidize sulfides.
Since the composition of the bacterial population can be influenced by various
factors such as high temperatures and low pH which enhance Leptospirillum num-
bers, it is important to determine the nature of the mixed population. This, in
turn, can have a profound effect on the quality of the leach. The dot-blot immuno
binding assar4 as well as the peR technique '5 have been used to analyze bacterial
populations.
The BIOX® population has been investigated with the microscopic immuno-
fluorescence technique. '6 This was made quantitative by counting the total num-
ber of bacteria in the same field of view using phase contrast after the fluorescent
bacteria (which may be Leptospirillum, T. ferrooxidans or T. thiooxidans depend-
ing on the fluorescent antibodies used) had been counted under UV light. A num-
ber of microscopic fields were counted for anyone species.
The Sao Bento BIOX® tank was found to contain 48% L. ferrooxidans, 34% T.
thiooxidans and 10% T. ferrooxidans, as free unattached bacteria. A similar pro-
portion of bacteria was determined in the Fairview multistage continuous plant
where in the primary tank (equivalent to the Sao Bento tank) Leptospirillum made
up 57%, T. thiooxidans 30% and T. ferrooxidans 13% of the population offree bac-
teria. The composition of the bacteria attached to the ore was obtained by releas-
ing them into acid water (pH 1.8) using Triton X100 and vortexing. '7 Of this, 57%
was again Leptospirillum, with 26% T. thiooxidans and 17% T. ferrooxidans. Since
counting errors do arise, it is possible that no real difference exists in the propor-
tions of free bacteria and attached bacteria.
However, in progression towards the final tank, the proportion of T. thiooxidans,
particularly those attached to the ore, was found to increase while less Leptospirillum
remained attached. This can be explained by a decrease in oxidizable iron and an
increase in sulfur intermediates which provide an energy source for T. thiooxidans.
Over the cascade of tanks, the amount of free T. ferrooxidans dropped to a low 5%,
while the amount attached increased slightly and then remained relatively con-
stant. That L. ferrooxidans plays a more dominant role in the continuous process
than T.ferrooxidans has been confirmed by other studies. 18,'9 It has also been indi-
cated that a predominance of Leptospirillum-like bacteria increases the iron-leach-
ing rate by a factor of approximately 24 '8
In ores rich in arsenopyrite, as is the case for many gold-bearing refractory
ores, the first step in leaching is growth of bacteria attached to the ore. This en-
hances direct bacterial oxidation, causing the appearance of the first corrosion
patterns on the mineral and solubilization of Fe H , AS3+ and S2-. Subsequently, the
presence of high concentrations of FeH iron in solution enhances the growth of
free bacteria and Fe H and AS3+ become oxidized to Fe3+ and AS5+. The Fe3+ iron
generated plays a role in arsenic and sulfide oxidation and in precipitation of fer-
ric arsenate, thereby coating the attached bacteria. 20
Bacteria growing on arsenopyrite may be inhibited by the arsenic released. The
BIOX® bacteria are tolerant to arsenic (V) concentrations of 15-20 gIL. However,
they are less tolerant to arsenic (III) and become inhibited above arsenic (III) con-
centrations of 6 g/L. 21 Just as arsenic can be toxic to the bacteria, so can other
The BIOX" Process for Biooxidation of Gold-Bearing Ores or Concentrates 57
constituents of the ores, such as antimony and mercury. Tolerance limits of the
BIOXIt bacteria towards these two metals is 10% by mass for antimony and at least
0.12% for mercury. An unusual feature of the acidophiles is their high sensitivity
toward chloride ions, the reason for chloride toxicity appears to be membrane
damage.:U
Other environmental factors that impact on the BIOX'" bacteria are pH, tem-
perature, carbon dioxide and nutrients. When shake-flask tests were carried out
on BIOXIt T. ferrooxidans inoculated into 9K medium23 with pH adjusted to 1,1.6,
1.8, 2, 3 and 4, it was found that oxidative activity was inhibited at a pH less than
1.6. The optimum was in the range of pH 2-3. Both 1. ferrooxidans and T. thiooxidans
were still active at pH 1, although T. thiooxidans, grown in a sulfur medium, showed
little growth between pH 0.5-1.0. Thus, bacterial leaches at very low pH would
have few T. ferrooxidans and more of the other bacteria.
Using a Chemap fermentor and varying temperatures between 25°C and 50°C
with other parameters constant, the optimum operating temperature of the BIOXIt
bacteria was found to be 40°C, although their oxidative potential was only slightly
lower at 35°C and 45°C. The bacteria were not killed at 50°C, but their oxidation
rate slowed down considerably, with the time required for complete conversion of
ferrous to ferric iron increasing from approximately one day at 40°C to 3 weeks or
more at 50°C.24 Growth of Leptospirillum will be favored compared to T.ferrooxidans
in an environment with alowpH « pH 1.5) and a temperature of 40°C or higher. T.
ferrooxidans will often be prevalent in batch cultures operating at high ferrous
concentrations, and controlled pH and temperature of pH 1.4-1.6 and 35°-40°C,
respectively.
The effect of added carbon dioxide was assessed by the rate of iron oxidation
using air (0.03% CO 2) and air supplemented with CO 2to give concentrations of 0.3
and 1%. Tests were carried out in a Chemap fermentor at two different tempera-
tures, 30°C and 40°C. It was found that the different concentrations of CO 2 had no
effect at 30°C, while at 40°C added CO 2 markedly shortened the time for complete
oxidation of ferrous iron. At 30°C, temperature was probably more limiting on
oxidative activity than was carbon dioxide concentration.
The nutrient formulation used to grow T. ferrooxidans and Leptospirillum in
the absence of metal sulfides is 9K. This contains basic salts and 50 giL ferrous
sulfate. In the presence of metal sulfides, oK, which lacks iron but is otherwise
identical, is supplied. This formulation has been described as providing an excess
of nutrients.25-'·7 Our laboratory investigations proved that this is the case and a
modified 9K medium containing 50 giL FeS04.7H20, 1 giL (NH4)2S04' 0.4 gIL
K2HP0 4 and 0.1 giL MgS04.7H20 did not diminish bacterial oxidation rates.
Ammonium sulfate is included in the nutrient medium because there is no
definite proof that all bacteria in the BIOXIt population can use gaseous nitrogen;
i.e., they have the capacity to fix nitrogen. Diazotrophy (ability to fix nitrogen) has
been described for Leptospirillum and T. ferrooxidans but not for T. thiooxidans. 28
However, a restriction enzyme analysis using the nifHDK probe showed that not
only the BIOXIt Leptospirillum and T. ferrooxidans bacteria had nitrogen fixing
genes, but these genes were also present in Thiobacillus thiooxidans. That these
genes are operative has yet to be proven. In addition, the high aeration required
for oxidation of the metal sulfides would probably inhibit nitrogen fixation, since
the nitrogenase enzymes are sensitive to oxygen.
Engineering Design and Process Requirements
Chemical Reactions and the Influence of Ore Mineralogy

The BIOX® process comprises contacting the refractory sulfide ore or concen-
trate with a preparation of the GENCOR BIOX® mixed bacterial culture for a suit-
able treatment period while maintaining an optimum operating environment. The
bacteria cause accelerated oxidation of the sulfide minerals thereby liberating the
occluded gold for subsequent recovery via cyanidation. The principal sulfide min-
erals associated with refractory hydrothermal gold ores are arsenopyrite, pyrite
and pyrrhotite.
The bacterial oxidation reactions of the sulfide minerals pyrite, arsenopyrite
and pyrrhotite, to achieve gold liberation, may be summarized as follows:
4FeS2 + 1502 + 2H20 -t 2Fe2(S04)3 + 2H2S0 4 3-1
2FeAsS + 702 + H2S0 4+ 2H20 -t 2H3As04 + Fe2(S04)3 3.2
4FeS + 902 + 2H2S0 4 -t 2Fe2(S04)3 + 2H20 3.3
These reactions clearly indicate the high oxygen demand of sulfide oxidation.
Also evident is the acid-consuming nature of pyrrhotite and arsenopyrite oxida-
tion with pyrite oxidation being the sole acid-producing reaction. The oxidation
reactions are also highly exothermic. In addition to the direct oxidation of sulfide
minerals, several indirect chemical and bacterially assisted reactions occur as
follows:
FeS + Fe 2(S04)3 -t3FeS04 + So 3·4
FeS 2 + Fe 2(S04)3 -t 3FeS04 + 2So 3·5
2FeS + 2H 2S0 4 + O2-t 2FeS0 4+ 2So + 2H 20 3.6
4FeS04 + 2H2S04 + O2 -t 2Fe2(S04)3 + 2H20 3-7
2So + 302 + 2H 2 0 -t 2H 2 S0 4 3·8
Important secondary reactions include precipitation of ferric arsenate (FeAs0 4),
acid dissolution of carbonates and precipitation of jarosite, according to the fol-
lowing reactions:
2H3As04 + Fe2(S04)3 -t 2FeAs0 4+ 3H2S04 3.9
CaMg(C03)2 + 2H 2S0 4 -t CaS0 4 + MgS04 + 2CO, + 2H 20 3.10
3Fe2(S04)3 + 12H 20 + M2S0 4-t 2MFe3(S04)2 (OH)6 + 6H 2S0 4

where M+ = K+, Na+, NH4 +, H30+ 3.11
The relative proportions of each mineral dictate various process requirements

such as cooling, acid consumption/production, oxygen demand, degree of pre-
cipitation and neutralization. The heat of reaction, acid demand and oxygen de-
mand for oxidation and chemical leaching of principal refractory gold ore miner-
als is presented in Table 3.9. The actual values, for treatment of a particular
concentrate, will be dictated by the relative proportions of the major minerals.
Typically the overall heat of reaction is about 30 MJ/kg sulfide with an oxygen
demand of 2.2 kg/kg sulfide oxidized.
Examples of the effects of the major minerals upon the operation ofbiooxidation
and the BIOX® process follow:
The BIOX'" Process for Biooxidation of Gold-Bearing Ores or Concentrates 59
Table 3.9. Process data for sulfide mineral oxidation
Mineral Pyritic Heat of Reaction Oxygen H.S04

Sulfur % Demand Demand
kg O./kg S'- kg/kg
Mineral
Mineral Sulfide
kJ/kg kJlkg S'-
Pyrrhotite FeS 36.4 -11,373 -31,245 2.25 0·557

Arsenopyrite FeAsS 19.6 -9,415 -48,036 3.51 0·301
Pyrite FeS, 53.3 -12,884 -24,173 1.88 -0·408
Ankerite Ca(Fe,Mg)(C03), - 219.2 0·979
Siderite FeC03 -326.7 (0.069 kg/kg 1.267
mineral)*
* Oxidation ofiron (II) to iron (III)
1.Pyrite
Bacterial oxidation of pyrite is highly acid-producing. Therefore, treatment of
a concentrate with a high pyrite content will be acid-generating and maintenance
of the pH within the required operating range requires addition oflime or limestone.
2. Pyrrhotite/Pyrite
Due to the acid-consuming nature of pyrrhotite, the relative proportion of py-
rite to pyrrhotite is an important factor affecting the overall lime and/or acid re-
quirements, and one which also influences solution redox potential. The acid dis-
solution of pyrrhotite releases ferrous iron and elemental sulfur. Although the
formation of elemental sulfur by this means is reversed by the bacteria present in
the culture, excessive elemental sulfur formation, due to an abnormally high pyr-
rhotite content, cannot be accommodated in the plant and may lead to an increase
in cyanide requirements and lower gold recovery.
The higher ferrous level in solution is beneficial in that it promotes a large
bacterial population in the liquor phase of the primary reactors, which in turn
reduces the possibility of bacterial washout occurring. However, the higher fer-
rous concentration lowers the redox potential, which can alter the oxidation chem-
istry of the process. The most serious effect of low redox potentials, combined
with a low iron to arsenic ratio in solution, is the possibility of arsenic (III) forma-
tion. Arsenic (III) may precipitate as the less stable calcium arsenite compound;
hence formation of arsenic (III) should be minimized as far as possible. In addi-
tion, arsenic (III) has a greater toxicity effect upon the bacteria than arsenic (V).
3. Arsenopyrite
The ratio of arsenopyrite to pyrite also influences acid consumption, but to a
lesser extent than that of the pyrrhotite to pyrite ratio. More critical is the ratio of
arsenopyrite to pyrrhotite and pyrite as given by the iron to arsenic ratio.
The iron to arsenic ratio is critical as it dictates the stability of ferric arsenate
precipitates formed on neutralization of the BIOX® waste liquor. The molar ratio
of iron to arsenic in a concentrate is generally required to be greater than 3 to
achieve stable effluent products, with respect to arsenic solubilization.
4. Carbonate Minerals
Carbonate content has two major effects on the BIOX® operation. Firstly, a
minimum content is required to ensure production of sufficient CO 2 to promote
bacterial cell production. If no carbonate is present, limestone must be added to
the primary vessels, or the carbon dioxide content of the air injected must be fur-
ther enriched with carbon dioxide.
The second effect is that of carbonate dissolution on pH. At a low sulfide to
carbonate ratio, the primary stage becomes acid-consuming. The degree of pre-
cipitation increases and results in coating of the sulfide surfaces. Formation of
coatings may result in lower oxidation rates, which in turn reduces liberation of
gold for dissolution on cyanidation. Presence of carbonate at a high sulfide to car-
bonate ratio is beneficial, not only for CO 2 production, but also in reducing lime
requirements for pH control during biooxidation.
Effect of Temperature and Cooling Requirements
Extent of Sulfide Oxidation

The BIOX® bacterial culture is an adapted mixed culture of mesophilic bacte-
ria, as described earlier in this chapter. Continuous pilot testwork has shown that
although this culture operates best at 40°C, operation at a temperature of 45°C is
possible in the primary reactors (where initial bacterial growth occurs) and at 50°C
in the final secondary reactors!9
The effect of operating at 45°C for biooxidation of the Fairview and an Austra-
lian concentrate is shown in Figure 3.3. It is noted that no significant decrease in
the final sulfide oxidation occurs, compared to operation at an "optimum" tem-
perature of 40°C.
Cooling Requirements
The oxidation of sulfide minerals is extremely exothermic as shown in Table
3.9. Biooxidation with mesophiles requires operation at a temperature in the range
of 30°-45°C.At a high rate of sulfide mineral oxidation, cooling of the bioreactors
is necessary in order to stay within this temperature range. The possible heat sinks
and heat loads for biooxidation, excluding the relatively small radiative and con-
vective heat losses, are as follows:
1. Heat of reaction of sulfide mineral oxidation.
2. Heat generation by absorption of agitation power.
3. Sensible heat loss or gain from inlet air on adjusting to slurry temperature.
4. Heat loss to heating the incoming slurry to the vessel operating tempera-
ture.
5. Evaporative cooling provided by the sparged air at slurry temperature.
6. Heat loss due to air expansion.
Typical heat loads for biooxidation plants are 30 MJ/kg sulfide oxidized. The
most cost-effective method for cooling is by circulation of cooling water through
internal coils in the reactor and removal of the heat from the water in an evapora-
10
.0
l: 70
g ao
IV
50
. . 40
o
-8 .0 _ _ Fa_ .. dog. C
:E " 20 .-...o6-At.tAhn eanc.ntnl" 40 0.;_ C
J! .......... ~tr..." canc.ntraMo 45 drtg . C
o
o 20 .0 00 .0 100 120
Rete ntio n Time (houri)
Fig. 3.3. Effect of temperature on sulfide mineral oxidation.
tive cooling tower. The efficiency of evaporative cooling is dependent on the local
climatic conditions, principally the wet bulb temperature. The efficiency is poor
in equatorial regions with high ambient temperatures and high humidity.
Operating biooxidation at a temperature of 45°C compared to 40°C may reduce
the cooling water flow rate required by 37% and the cooling tower cross sectional
area by 25%. However, although this saving is significant it would only represent a
reduction of about 5% in the total capital cost of a biooxidation plant and a negli-
gible savings in operating costs. Nevertheless, operating biooxidation at 40°C gives
a safety margin whereby the slurry temperature may increase by 5°C or more be-
fore bacterial activity is adversely affected. A temperature rise may occur on oper-
ating plants due to failure of cooling water supply, or scaling of the internal cooling
coils if correct maintenance is not carried out.
pH Control
Optimum rates of sulfide oxidation and gold recovery will require control of
the slurry pH to maintain an active bacterial culture. A typical operating pH is in
the range pH 1.2-2.0. A low pH reduces the extent of sulfide oxidation. Ahigh pH
may also reduce the extent of oxidation and can decrease gold recovery due to
metal salt precipitation resulting in occlusion of gold particles. Limellimestone or
acid addition will be required for pH control, depending on whether the sulfide
oxidation is net acid-producing or consuming. As indicated in Table 3.9, concen-
trates with a low carbonate and high pyrite content are acid-producing, concen-
trates with a high pyrrhotite and carbonate content are acid-consuming.
Operating costs may be minimized for biooxidation of acid-generating ores by
using a locally sourced limestone for acid neutralization and pH control. In the
case of acid-consuming ores, recycle of acidic biooxidation product slurry or solu-
tion may be incorporated in the process design to reduce consumption of fresh
acid.
In the case of treatment of refractory gold ore concentrates, upstream flotation
offers the opportunity to produce a concentrate where acid-consuming minerals
such as carbonates or pyrrhotite, are rejected to the flotation tailings. In designing
a biooxidation plant, considerable attention must be given to the upstream con-

centration or preparation process to ensure that the optimum grade and grind of
ore (or concentrate) is produced for treatment by biooxidation in order to ensure
maximum gold recovery and thus, financial return.
Oxygen Supply
The supply of air to biooxidation represents the largest consumer of power in
the process and hence the major contributor to the overall operating cost. Typi-
cally, sulfide mineral oxidation requires about 2.2 kg oxygen per kg sulfide oxi-
dized.
For a biooxidation plant treating 240 tonnes/day of concentrate, at 12% sulfide
and 90% sulfide oxidation, the sulfide oxidation duty is 1,080 kg/h. The equivalent
oxygen demand is 2,376 kg/h or 7,912 Nm 3/h of air at 100% oxygen utilization. In
practice, the oxygen utilization will be much lower than 100%. Assuming a utiliza-
tion of 30%, the air demand for biooxidation increases to 26,373 Nm3/h.
In order to attain acceptable oxygen transfer rates and oxygen utilization in the
bioreactor, the air supplied must be well dispersed. Air dispersion is achieved by
mechanical agitation. The power required for supplying oxygen, air delivery and
dispersion, represents 30-40% of the total power requirements for biooxidation of
metal sulfide concentrates. The power for air supply can be minimized by design-
ing reactors with a low height-to-diameter aspect ratio, so that the compressor or
blower pressure head required is as low as possible. A low pressure head permits
the use of blowers instead of compressors, which consume less power per m 3 of air
delivered.
Agitator Selection
The criteria for the agitator specification can be summarized as follows:
1. Sufficient power must be provided to the impeller to prevent flooding; flood-
ing occurs when air passes through the impeller and is not dispersed by
fluid flow.
2. The oxygen mass transfer coefficient (kl.a) for the agitator/aeration system
must meet or exceed the required kl.a, determined by the oxygen demand
of the process.
3. The impeller pumping rate must be sufficient to achieve uniform solids sus-
pension (as defined by Gates et al30 ).
The air supply must be adequate to meet the process oxygen demand deter-
mined by the rate of sulfide oxidation and to maintain a dissolved oxygen concen-
tration in solution of not less than 1.5 ppm in order to allow an adequate oxygen
supply to the bacteria.
The input power necessary to attain the required oxygen uptake rates, for
biooxidation of metal sulfide concentrates, is normally sufficient to be the control-
ling factor in the agitator specification. Radial flow impellers such as the Rushton
turbine are the traditional impellers used for high gas dispersion rates. However,
recently developed axial flow fluidfoil impellers, such as the LIGHTNIN A315 im-
peller, offer improved efficiency, giving equivalent oxygen transfer rates at reduced
power consumption. The fluid flow developed by these impellers is also much higher
than radial flow turbines, per unit power input, allowing solids suspension at re-
duced shear rates and lower power levels.
The BIOX" Process for Biooxidation of Go/d-Bearing Ores or Concentrates 63
•
...•o
I. Total - -Agitator -+-Btower I
Fig. 3.4. Minimization of total power vs. air flow supplied (Units are not indicated as
this is a generic graph and units will depend on the size ofbioreactor and the process
duty).
0.08
0.08 -+-0 % SolIdo - - 2 3.1 % Solids
0.07
-'-40.3 % Solids ~53.6 % SolIdo
0.08
i 0.05
!! 0.04
:00:
0.03
0.02
0.01
0
0 0 .1 0 .2 0.3 0 .4 0.5 0 .6
Power per Unit Volume (Wlm')
Fig. 3.5. Effect of solids concentration on oxygen mass transfer.
The design of agitators for bioreactors has been well described by Fraser.3'•3'
The agitator must be selected to match the required reactor duty so that the overall
power consumption for oxygen supply is minimized. as illustrated in Figure. 3.4.
The solids content of the process slurry has a direct influence on the oxygen
mass transfer rate, as shown in Figure 3.5, after Mills et al.33 Bailey and Hansford'"
have demonstrated that oxygen mass transfer is the limiting factor controlling the
solids content of process slurries in biooxidation. For typical sulfide concentrates
at 20-30% sulfide, the required oxygen transfer rates limit the solids content and
hence sulfide content per unit volume, to about 20% by mass. For low-grade con-
centrates or ores at 2-5% sulfide, solids concentrations of over 30% by mass may be
tolerated.
The optimum operating slurry density must be determined from pilot testwork,
the process results providing the required agitator specifications for the commer-
cial process.
The relationship between the rate of oxygen demand and the oxygen mass trans-
fer coefficient (kl.a) may be expressed as:
R = kl.a(C* - Cd
where: R Rate of oxygen demand, (mg/L/s)
C' Saturated dissolved oxygen concentration (mg/L)
CL Dissolved oxygen concentration in the bulk fluid (mg/L)
kl.a Oxygen mass transfer coefficient (lIs)
The oxygen mass transfer coefficient for air dispersion in a mechanically stirred
tank can be correlated with agitator power and superficial gas velocity by the fol-
lowing empirical correlation.
kl.a = c (Pg/V)a (us)b
where: Pg aerated impeller power draw (W)
V unaerated fluid volume (m3)
Us superficial gas velocity (m/s)
c,a,b empirical constants
Typically the constant c is in the range 0.01-0.04 and the constants a and bare
in the range of 0.3-0.6.
Process Modeling and the Effect of Bioreactor Configuration
Process Operating Curves

GENCOR Process Research has developed operating curves to describe
biooxidation performance, by applying the logistic model, as described by Miller,34
to continuous pilot plant data. These curves relate plant retention, concentrate feed
rate, sulfide grade and mass of sulfide oxidized. The operating curves provide valu-
able criteria whereby biooxidation plant activity may be assessed and target oper-
ating conditions set, in order to achieve the required sulfide oxidation and con-
comitant gold recovery. In addition, these curves allow prediction of the effect of
variation in feed grade and mineralogy, as well as operating conditions, on sulfide
oxidation performance.
Operating curves for the Fairview biooxidation plant, with a concentrate feed
grade of 24% sulfide, are shown in Figure 3.6. Two curves are shown, sulfide oxi-
dized (tonnesl day) against concentrate feed rate (tonnesl day) and the correspond-
ing curve for recoverable gold (%).
The operating curve defines the upper limit of mass of sulfide oxidized at any
given concentrate grade and feed rate under optimum biooxidation performance.
The Fairview plant operating curve indicates a linear increase in mass of sulfide
oxidized up to 6.0 tonnes/day,as the concentrate feed rate is increased to 2]tonnesl
day (6.0 day retention). Thereafter, the rate of increase in sulfide oxidation de-
creases until a maximum of 9.0 tonnes/day sulfide is oxidized at a concentrate feed
rate of 54 tonnes/day (3.0 day retention). Further increase in concentrate feed rate,
with the concomitant decrease in plant retention, results in a decrease in the total
mass of sulfide oxidized as high bacterial numbers can no longer be sustained.
- 100
x xxxxx X)OO( x x x
x 90
• Sufflde Oxidized
14 • 60% Oxidation 80
>: ~
i 12
-
•
X
85% Oxidation
Recoverabte Oold (%)
Poly. (SuHlde Oxidized)
70
'V
'0
j 10
60
0
50 .!
Ji! 6 .a
S 40 ••
~
-8 6 >
:= 30 0
ri 4 20
•
0
a:
2 10
O ~------------------------------------------~ 0
o 10 20 30 40 50 60 70 80 90
Concentrate Feedrate (tld ay)
Fig. 3.6. Fairview process operating curve.
The straight lines shown in Figure 3.6 radiating from the origin are curves for
determining extent of oxidation. These have been given for 60%, and 85% sulfide
oxidation. The plant operating point should generally be to the left of the perfor-
mance curve peak. Operating to the right results in a drop in sulfide oxidized and
a decrease in the recoverable gold. The expanded Fairview BIOXII> plant was de-
signed to treat 40 tonnes/day concentrate (4 day retention), which corresponds to
a sulfide oxidation of 85% and a sulfide oxidation rate of about 8.2 tonnes/day. The
recoverable gold at this operating point is 95%.
Bioreactor Configuration
The optimum configuration of reactors for a biooxidation plant is related to
the rate of sulfide mineral oxidation and the corresponding rate of bacterial growth
achieved. Sufficient residence time must be allowed in the primary bioreactors to
allow a stable bacterial population to develop. If the residence time is too short,
bacterial washout will occur and the rate of sulfide oxidation will decrease. In the
case of biooxidation of the Sao Bento concentrate, a high specific rate of sulfide
oxidation was achieved (15 kglm3/day). By modeling the results of continuous pilot
testwork, it was found that optimum process performance may be achieved by
operation of four bioreactors in series to give an overall design retention time of
1.8 days, or 0.45 days per reactor, for a plant treating 150 tonnes/day of concentrate
with the primary tank retention being 25% of the total plant retention.
The effect of tank configuration on process performance is illustrated in Fig-
ure 3.7, for biooxidation of the Fairview concentrate. The graph clearly shows that
a low primary tank volume will result in a significantly lower sulfide oxidation for
a given plant retention time, thus limiting the feed tonnage to the plant.
The optimum reactor configuration will be dependent on the ore mineralogy,
and the rate of sulfide mineral oxidation and bacterial growth rate. Ores contain-
ing minerals which oxidize rapidly will allow biooxidation operation with a short
retention time. The influence of the ore mineralogy on the rate of biooxidation is
further discussed below.
100
. 90 I
....., 80 I
~
- 70 !
C
0
I
!;;
.
60
50 I
_ _ Primary SO %
0 40 I
~ 30 ,
!E.
::J 20
I/) - + - Primary 17 %
10
0 4 8 8 10 12
Retention Ti me (dey.)
Fig. 3.7. The influence of primary tank volume as a percentage of total plant volume on
sulfide oxidation.
100 I • Pyrrhotite
90 .
• Ar.enopyrUa
60 i
.. Pyrite
I --Poly. (Pyrrhotite)
- -Poly, (Ar•• nopyrlle)
- 70 I
t I
--Poly. (Pyrite)
60
c
~ 50 I
I
i
'a 40 I
M
o 30 i
20 I
o 10 15 20 25 30 35 40 45
Retention Time (hour.)
Fig. 3.8. Sao Bento sulfide mineral oxidation vs. retention time.
Rate of Sulfide Mineral Oxidation and Gold Dissolution

Pyrrhotite is readily oxidized during biooxidation by ferric leaching, accord-
ing to the reaction given by Equation 3-4- Sulfur produced will be partially oxi-
dized by bacterial oxidation to produce sulfuric acid. Treatment of the Sao Bento
and Fairview concentrates has shown that arsenopyrite is oxidized prior to
pyrite and at a faster rate. The difference in rate of leaching for the sulfide miner-
als pyrrhotite, arsenopyrite and pyrite in the Sao Bento concentrate, is shown in
Figure 3.8.
In sulfide mineral concentrates such as the Fairview concentrate, biooxidation
will achieve complete release oflocked gold with only partial oxidation of the sul-
fide minerals present. This effect is caused by preferential oxidation of arsenopy-
rite in which the bulk of the gold is "chemically bound." Research has also demon-
strated that bacterial oxidation will occur preferentially along grain boundaries
100
80
"0_ 10
~! 70
0 10
-c ~A.ra.nopyrit. Oxidation
:!;:
• 0 50
>C ~
°0 .0
e.!
.0
30
c
i 20 - ....- Gold Dluoludoo
10
0
0 I I 10 12 U 11 11 20
Batch Treatment Period (day.)
Fig. 3.9. Gold dissolution vs. mineral oxidation for Fairview concentrate.
100 I
90
80
l 70
C
.2 60
';
'0
•• 50
Q .0
'a
'0 30
<:I
20
10
0
0 20 40 60 80 100
Sulfide Oxidation (%)
Fig. 3.10. Gold dissolution vs. sulfide oxidation for an Australian concentrate.
and in arsenic-rich zones, resulting in liberation of gold which is concentrated in

these areas. 35 This effect is shown in the biooxidation of Fairview concentrate
(Fig. 3.9), in which preferential oxidation of arsenopyrite results in the early lib-
eration of locked gold.
Biooxidation of refractory pyrite concentrates containing gold often show a
linear dependence of gold dissolution on sulfide oxidation. This is illustrated in
the biooxidation treatment of an Australian concentrate (Fig. ).1o).
General Process Requirements and Conditions for Operations
Grind Size
In general, the optimum grind size of 80% passing 75 mm and 95% pass-
ing 150 11m is stipulated for BIOX® plant operation. An increase in grind size
reduces the sulfide oxidation rate and would result in a lower overall sulfide
oxidation in a fixed volume plant. Production of a large proportion of fines (mi-

nus 25 )lm) may enhance the sulfide oxidation rate. However, down-stream thick-
ening and filtration may become increasingly problematic and viscosity of CIL
slurries will increase.
Nutrients
The nutrients potassium, nitrogen and phosphate are required for bacterial
cell growth. Nutrient addition used originally in 1987, for the Fairview plant, was
as follows; in kg/tonne of concentrate: 9.62 NH4+, 13.05 K+ and 1.5 POl+. This was
reduced over a period of two years to 8.40 NH/, 1.42 K+ and 1.56 P0 43+.
The minimum amount of nutrients required will be specific to the ore being
treated. Often potassium occurs in silicate minerals present in the ore which are
leached during biooxidation so that the required addition of potassium may be
eliminated or reduced to a level of less than 1 kg per tonne of concentrate.
Toxins
Biooxidation relies on a living culture of bacteria to promote sulfide mineral
oxidation. Certain chemical reagents and naturally occurring elements are toxic
to the bacteria and must be prevented from entering or building up to toxic con-
centrations, in the biooxidation plant. Reagents used upstream of the biooxidation
plant must be cleared by toxicity testing at the relevant dosages. These reagents
include flotation reagents, flocculants and defoamers.
Reagents which are particularly toxic include:
- All cyanide and thiocyanate species above specified operating levels
- Bactericides and fungicides
- Descalants and corrosion inhibitors
Maximum thickening of the flotation concentrate prior to use as BIOX® feed is
very important in order to minimize ingress of flotation reagents into BIOX®, and
prevent the concomitant froth formation and possible inhibitory effects on bacte-
rial oxidation of sulfide minerals.
Toxic or inhibitive effects of soluble inorganic species are discussed in the sec-
tions on the BIOX® bacterial culture (earlier) and the effects of chloride and ar-
senic (to follow).
Material Selection
It is normal practice to construct the BIOX® reactors and agitator impellers
from stainless steel (304L, 316L or SAF2205), or from mild steel with rubber lining,
for protection against corrosion and wear. Stainless steel is used for construction
of cooling coils, with series 304L or 316L normally being adequate grades for cor-
rosion resistance under BIOX® operating conditions. However, higher grades such
as SAF2205 or 904L are required where dilution water contains high chloride con-
centrations (above 1,000 ppm CI) as experienced in certain areas of Western
Australia.
The Effects of Chloride and Arsenic on the BIOX® Process
Effect of Chloride
High concentrations of chloride in solution have been shown to cause damage
to the membranes of the bacteria. 22 Toxicity tests in which the rate of ferrous iron
oxidation by the BIOX® bacterial culture was determined at different solution con-
The BIOX· Process for Biooxidation of Gold-Bearing Ores or Concentrates 69
centrations of chloride, have shown that little inhibition occurs at concentrations

in the range 0-5 giL chloride. At concentrations of 7 gIL chloride, the rate of fer-
rous iron oxidation decreases, such that only 55% oxidation occurs after 24 hours,
compared to 80% for the control tests (0.1 giL chloride) and complete oxidation is
achieved within two days. Above 19 giL chloride, complete inhibition of bacterial
activity occurs (GENCOR, unpublished).
The bacteria in the BIOXCD bacterial culture may operate at chloride concen-
trates in solution of up to 5 giL and maintain adequate sulfide mineral oxidation
rates. However, the presence of high chloride concentrations in solution during
biooxidation will promote precipitation of jarosite. This may become excessive
and inhibit gold recovery on subsequent cyanidation of the washed biooxidation
product, due to the coating of liberated gold. The precipitation of jarosite may also
inhibit sulfide mineral oxidation by coating the sulfide mineral surfaces, thus pre-
venting direct bacterial attachment. Continuous biooxidation of a refractory gold
ore concentrate may only give satisfactory results at a relatively low chloride con-
centration in solution. Experience with continuous biooxidation of an Australian
concentrate showed that best results in terms of sulfide oxidation and gold recov-
ery were achieved at a chloride level of 1.3-1.5 giL.
Operation at high chloride concentrations will also require the use of more
expensive stainless steel for internal cooling coils and the protection of pipes and
pump internals.
The Effect of Dissolved Arsenic

The BIOXCD bacterial culture is tolerant to arsenic (V) concentrations of
15-20 giL for continuous biooxidation of sulfide mineral concentrates. However,
the culture is less tolerant to arsenic (III) concentrations, as discussed earlier. This
difference may be explained by the toxic effect of arsenic (III), which is believed to
inactivate enzymes which playa key role in cell metabolism. Arsenic (V) is less
toxic than arsenic (III) but its action interferes with oxidative phosphorylation,
leading to energy loss in the bacteria.36•37
Biooxidation of the Fairview concentrate results in an arsenic (V) concentra-
tion in solution of 12 giL. However, even in the primary bioreactors, the arsenic
(III) concentration remains below 500 ppm during continuous operation. It is be-
lieved that arsenic (III) is rapidly oxidized to arsenic (V) by ferric ions, at pyrite
surfaces. Fairview concentrate has a low pyrrhotite content and consequently,
biooxidation operates at a high redox potential (550-600 m V vs. Standard Calomel
Electrode), which may promote oxidation of arsenic (III) to arsenic (V).3 8
In the case of biooxidation of the Sao Bento concentrate, high concentrations
of 3-6 giL arsenic (III) occur in solution. It appears that the pyrite content of the
concentrate (10.1%) does not promote the oxidation of arsenic (III) to arsenic (V).
This is believed to be due to the high pyrrhotite content and/or low pyrite arse-
nopyrite mass ratios in the Sao Bento concentrate.38 The mechanism of arsenic
(III) formation and oxidation is currently being researched, to verify the interac-
tion of solution potential and available catalytic surfaces, such as pyrite, on the
rate of oxidation.
The presence of arsenic (III) in the biooxidation product liquors may also ad-
versely affect the subsequent neutralization process, particularly if the arsenic (III)
concentration is greater than 3 gIL and the iron/arsenic mole ratio is less than
6/1. 39 Oxidation of neutralization liquors by a strong oxidant such a hydrogen per-

oxide may be require, in order to convert arsenic (III) to arsenic (V) and produce
stable solid residues. 38
Summary of Process Requirements for Biooxidation

The following is a summary of the process requirements for optimum opera-
tion of a biooxidation plant:
1. Adequate nutrient supply.
2. An adequate oxygen supply to meet the process oxygen demand, determined
by the rate of sulfide oxidation, and to maintain a dissolved oxygen concen-
tration in solution of not less than 1.5 ppm.
3. Sufficient supply of carbon dioxide for supply of carbon to build up bio-
mass. The carbon dioxide may be sourced from the air supplied for oxygen-
ation, from dissolution of carbonate minerals in the ore or, in the case of
acid-generating ores, by dissolution of limestone added for pH control.
4. Maintenance of an optimum operating temperature in the range 30°-45°C
for a mesophile bacterial culture such as the BIOX® bacterial culture.
5. Maintenance of an optimum operating pH, normally in the range pH 1.2-1.8.
Operation at a low pH « 1.2) will result in a decrease in bacterial activity.
Operation at a high pH (> pH 2) will result in the increased precipitation of
metal salts such as jarosite which may coat mineral sulfides, retarding bac-
terial oxidation and possibly resulting in occlusion of gold particles and
reduction in gold recovery on cyanidation of the biooxidation product.
6. Grind of ore or concentrate to be nominally 80% passing 75 }lm with 100%
passing 150 }lm.
7. The elimination of toxins and the control of soluble elements toxic to the
bacteria.
Treatment and Disposal of BIOX· Effluent Streams
Background
The bacterial oxidation of conventional pyrite/arsenopyrite ores or concen-
trates produces a liquor phase containing arsenic (V) and ferric sulfate according
to the following general equations, as presented in the section on engineering
design.
2 FeAsS + 70. + H.S0 4 + 2H.O ~ 2H3As04 + Fe.(S04)3
4FeS. + 150. + 2H.O ~ 2Fe.(S04)3 + 2H.S0 4
As arsenic compounds are toxic to all life, strict regulations for the control of
arsenic-bearing wastes exist throughout the world. These regulations apply to liq-
uid wastes as well as to any leachate which may be formed by the dissolution of
solid wastes on dumps, due to exposure to air and/or water.
The conventional method for the fixation of arsenic from solution is by lime
neutralization, preferably in the presence of excess iron, to produce ferric arsen-
ate, FeAs0 4 x H2 0, precipitates. The United States Environmental Protection Agency
(EPA) considers chemical precipitation of ferric arsenate be the Best Demonstrated
Available Technology (BDAT) for the treatment of wastewater forms (which in-
clude mining effluents) of arsenic-bearing wastes. 40
The BIOXe Process for Biooxidation of Gold-Bearing Ores or Concentrates 71
The As(V)-bearing BIOX® liquors are neutralized in a two-stage process em-

ploying limestone and/or lime. In the first stage, As(V) is precipitated as stable
ferric arsenate by adjusting the pH to between 4 and 5. The pH of the slurry is
subsequently adjusted to an environmentally acceptable level (between 6 and 8) in
the second neutralization stage. The use of limestone allows for better pH control,
particularly in the first neutralization stage, and is generally more economic than
lime. The overall chemistry of the two-stage BIOX® neutralization process is rep-
resented by the following equations:
Stage 1: Neutralization to pH 4-5

Fe~(S04h + H3As0 4 + CaC0 3+ 2H~O ~
Fe(OHh(s) + CaS04(s) + FeAs0 4(s) + 2H~S04 + CO~
Stage 2: Neutralization to pH 6-8

H~S04 + CaC03 ~ CaS0 4(s) + CO~ + H~O
or
H~S04 + Ca(OH)~ ~ CaS0 4 (s) + 2H~O
Development of the Two-Stage BIOX® Neutralization Process

Although it is generally accepted that coprecipitation of As(V) with ferric hy-
droxide is the most viable process for the removal of arsenic from process solu-
tions, a literature review indicates that there is still much controversy regarding
optimum precipitation conditions, the structures and the long-term stability of
these precipitates. Numerous investigations have thus been carried out at GENCOR
Process Research in order to establish optimum conditions for the neutralization
of BIOX® liquors, such that the process produces environmentally acceptable ef-
fluent liquors and stable ferric arsenate precipitates.
Process Parameters
The effects of neutralization operating parameters have been investigated on a
batch laboratory-scale, using synthetic liquors, and on a continuous pilot-scale,
using liquors derived from the continuous bacterial oxidation of two arsenopy-
rite/pyrite concentrates.39.41 The continuous neutralization pilot-plant was config-
ured in two stages, employing pure limestone to a final pH of 5 in stage 1 and pure
lime to a final pH of between 8 and 11 in stage 2. The stability of the pilot-plant
residues was determined by the standard Toxicity Characteristic Leaching Proce-
dure (TCLP) of the u.s. EPA4:>"43 as well as a modified stability test procedure which
utilizes dilute H~S04 or NaOH solutions to adjust the leach pH.
The results of the continuous pilot-plant testwork (Table 3.10) indicate that con-
tinuous neutralization of BIOX® liquors, with Fe/ As mole ratios ~3/1, results in
environmentally acceptable effluent arsenic concentrations (~0.04 ppm), and pro-
duces stable ferric arsenate precipitables ([As]TCLP«5 ppm), over a relatively
wide range of operating pH values (pH 5-11).
The neutralization pilot-plant results in Table 3-10 also indicate that the stabil-
ity of the arsenate precipitates decrease as the precipitation and leach pH values
increase above 5. These findings are consistent with those of numerous other in-
vestigators, who have reported that a pH range of between 3 and 6 is required for
Table 3.10. Results of the neutralization pilot-plant testwork

Concentrate Neutralization Ooerations Preciottate Stal lillY
Description Stage Final As in Solution Fe/As Mole Storage Stability As Value
pH (Dim) Ratio in Period Test in Extract
Inttlal Final Precioitate Extract loom}
Harbour Lights: 1 5.4 6400 0.12 - - - -
FeS2 =22%
FeAsS=8% 2 10.8 - 0.04 4.4 Fresh TClP 1.25
1 week TClP 0.26
8 months TClP <0.02
pH3 0.07
pH5 0.04
pH 7 0.07
oH 10 <0.02
Wiluna: 1 5.0 2685 <0.02 3.4 3 months TClP 0.04
FeS2=28% pH3 <0.02
FeAsS=21% pH5 <0.02
DH7 0.05
2 8.3 - <0.02 3.0 3 months TClP 0.13
pH3 0.13
pH5 0.08
pH7 0.14
oH 10 0.28
Wiluna: 1 5.5 1903 <0.02 - - - -
FeS2=28% 2 9.7 - <0.02 3.0 3 months TClP 0.24
eAsS=21% pH3 3.14
pH5 0.16
pH7 0.09
oH 10 0.20
optimum As(V) removal44-48 and ferric arsenate stability.45-4M9-51 Fe/As mole ra-
tios are, however, reported to have an even greater effect on the stability of ferric
arsenate precipitates than pH, with the stability increasing as the Fe/As mole ratio
increases from 1 to 16.44-46.48.5 0 It appears to be generally accepted that ferric arsen-
ates with Fe/As mole ratios ~3-4 are sufficiently stable for land disposal.44.48-5 1 The
coprecipitation of gypsum, formed when neutralizing sulfate-bearing liquors with
lime or limestone, and base-metals (Zn, Cu, Cd), is also reported to have a stabiliz-
ing effect on the ferric arsenate precipitates. 48 -51 Infra-red characterization has in-
dicated that the relatively low stability, particularly at pH 3, of the Wiluna neutral-
ization precipitates formed at a pH of 9.7 may be attributed to the coprecipitation
of arsenic as amorphous calcium arsenate, Ca3(As04)2..39.41 Calcium arsenates are
more soluble than ferric arsenate precipitates and are generally not accepted to be
sufficiently stable for long-term disposal, as they decompose to form carbonates
under the influence of C02. in air.52. No calcium arsenate precipitates could be de-
tected in any of the other pilot-plant neutralization residues. 3M1 Further infra-red
characterization testwork on the precipitates formed by neutralizing synthetic
Na2.HAs0iFe2.(S04h solutions with lime indicated that coprecipitation of calcium
arsenates only occurs at high pH (>10), and then only in conjunction with rela-
tively low Fe/ As mole ratios «3.4/1).39.41According to Barrett et al,53 neutralization
of As(V)/Fe(III)/sulfate solutions in the pH range 2-7 produces ferric arsenate pre-
cipitates only, even at a Fe/As mole ratio as low as 1.7/1.
The BIOXe Process for Biooxidation of Gold-Bearing Ores or Concentrates 73
Long-Term Stability of Ferric Arsenate Precipitates

The results of the continuous neutralization testwork at GENCOR Process Re-
search (see Table 3-10) indicate that the mobility of the arsenic in the residues de-
creases on aging (in the absence of air). These findings are consistent with those of
Therdiattikul and Dahlstrom,47 who attribute the increase in stability of ferric ar-
senates on aging to structural changes to a more ordered and insoluble form, and
with Krause and Ettel,49,5 0 who reported that leaching of ferric arsenate precipi-
tates at pH 5 for up to 3.7 years resulted in < 0.2 ppm arsenic in solution.
Although a number of investigators 45,46,49,5 o have reported that the precipitates
formed during the ambient temperature lime neutralization of solutions contain-
ing As(V) and Fe(lII) comprise an amorphous chemical compound of the type
FeAs0 4 x Fe(OHh, termed basic ferric arsenate, Robins 44>54>55 and Swash et al56
found that the precipitates consisted mainly of amorphous ferric hydroxide or
ferrihydrate (Fe 20 3 x H 20) onto which As(V) was absorbed. It was suggested that
these precipitates were not suitable for long-term disposal, as they would slowly
decompose to form goethite and soluble arsenic at pH ~2,44>54>55 particularly if
dehydrated. 56
Testwork has been carried out at GENCOR Process Research to investigate the
effects of dehydration (vacuum filtration and oven drying at 60°C) and rehydration
(repulping in tap water at L/S of 2/1) on the stability of the precipitates formed
during the continuous two-stage neutralization of BIOX® liquor with a Fe/As mole
ratio of 2311. The results in Table 3.11 indicate that the dehydration/rehydration
process, which simulates seasonal weathering on a tailings dam, results in a de-
crease in the mobility of arsenic in the neutralization residue.
Extent and Effect of As(III) Formation

A number of investigators57-6o have reported the formation of As(III) to at least
some extent during the bacterial oxidation of arsenopyrite concentrates, although
there appears to be little consensus regarding the mechanism, kinetics or condi-
tions for As(III) and/or As(V) formation. The relatively high solubility of metal
arsenites is well known,61 and the removal of As(III) from process solutions gener-
ally relies on the preoxidation of As(III) to As(V), with strong oxidants such as
hydrogen peroxide, and subsequent precipitation of As(V) as stable ferric
arsenates. 62
Testwork at GENCOR Process Research has indicated that although As(III) is
formed as a primary product during the bacterial oxidation of all arsenopyrite
bearing concentrates, it is rapidly and effectively oxidized to As(V) during bacte-
rial oxidation of conventional arsenopyrite/pyrite concentrates. The rapid oxida-
tion of As(III) in the presence of Fe(lII) and a moderately themophilic mixed cul-
ture growing on pyrite or chalcopyrite was also observed by Barrett et al. 63,64 Barrett
et al concluded that sulfide minerals catalyze the oxidation of As(III) by Fe(III),
with the bacterial action enhancing this catalytic effect through the regeneration
of clean surfaces and reoxidation of Fe(I1) to Fe(III). An assessment of 19 arseno-
pyrite-bearing concentrates processed at GENCOR Process Research has indicated
that As(III) is only present in the final BIOX® liquors (in the range 0.6-6.4 giL)
arising from concentrates containing significant levels of pyrrhotite (~16%) or
having low pyrite/arsenopyrite mass ratios «1/1). Laboratory testwork has also
indicated that As(III) only has a significant effect on the performance of the neu-
tralization process ([As] in effluent liquor ~0.5 ppm and [As] in TCLP extract ~5
Table 3.11. Dehydration/rehydration testwork results
As Concentration
(ppm)
Filtrate after first dehydration/rehydration <0.02

Filtrate after eighth dehydration/rehydration <0.02
TCLP extract before dehydration/rehydration 0.06

TCLP extract after eighth dehydration/rehydration 0.03
ppm) under conditions where the BIOX® feed liquor has an As(III) content of
>3 giL and a Fe(T)/As(T) mole ratio of ~6/1. In such cases, the adverse effect of
As(III) on the BIOX® neutralization process can be avoided by either adjusting the
concentration levels and/or composition of the feed liquor, or by adding H 2 0 2 be-
fore or during neutralization.
BIOX@ Neutralization Process Design and Performance
Continuous Mini-Plant Investigations

A continuous neutralization circuit is incorporated in the standard BIOX® pi-
lot-plant testwork programs in order to obtain plant design parameters, as well as
to ensure that environmental legislative limits will be met. A basic flowsheet of the
continuous neutralization mini-plant, employed at GENCOR Process Research since
1993, is shown in Figure 3.11.
The mini-plant is generally configured as seven tanks in series, each having an
aerated working volume of 2.2 liters. The tanks are agitated with pitch blade im-
pellers at approximately 350 rpm. Air is sparged into each tank in order to ensure
complete oxidation of any ferrous iron, present in the feed liquor, as well as to
facilitate solids suspension. Dissolved oxygen concentrations remain at saturated
levels (6-8 ppm). Limestone and, if required, lime slurries (80 gIL) are agitated in
make-up vessels, and both are circulated via a ringmain using a peristaltic pump.
The pH is measured and controlled in the range 4.85-5.5 in Tank 1 by limestone
addition from the ringmain via a solenoid valve. If required, either limestone or
lime is added to Tank 4 in order to ensure that the pH of the effluent liquor
(Tank 6 overflow) is ~ 6. The mini-plant also has facilities for recycling neutral-
ized slurry from the third to the first tank, via a peristaltic pump. Recycling is
employed in order to improve the settling characteristics of the slurry by provid-
ing seed for precipitation to promote growth of large particles. 65 Recycle flow to
feed flow ratios of between 0 and 5/1 are generally employed. Feed liquor is pumped
to the first tank (Tank 0) from a feed reservoir via a dosing pump at the required
feed rate. The retention for each vessel is approximately 1 hour, giving a total plant
retention of 7 hours. A summary of the results obtained for the continuous neu-
tralization of BIOX® liquors, derived from the continuous bacterial oxidation of
various sulfide concentrates over the past three years, is presented in Table 3.12.
;i
'"
b:l
Limestone Ringman
~
~
<'\
~
'"
"C'
Lime Ringmain
..
b:l
c·
Q
l-!
~
c·
;:!
~
C')
1 Q
$:i:
~
!:>
. e ..'"
~.
o
Recycle ~
'"
Q
.
(j
Q
;:!
<'\
lM'~
t
+ '"~
; .. : ~~~~ ~
'~' lJJi'l ~.:,~'" ... ~
AIR·················· .. · .
Fig. 3.11. Flowsheet of the continuous neutralization mini-plant. 0l

Table 3.12. Summary of the continuous neutralization testwork results
Description of Neutralization Feed Liquor Arsenic Concentration (ppm)

Fe As Fe/As Mole pH In Effluent In TCLP
(giL) (giL) Ratio Liquor Extract from
Neutralization
Residue
8·7-17·4 2·4-4·8 3·8-4.8 1.2-1·3 <0.25 0·50-1.00

10.7-15.7 0·4-0·7 25·4-50·3 1·4-1·5 <0.25 <0.02-0.84
21.9 0·4 70.6 1·5 <0.02-0.03 <0.02-0.06
25·6 1.2 28.0 1.1 0.03-0.04 0.07
Table 3.13. Stability of arsenic-bearing residues from the commercial plants
Plant Description of Waste Streams As in

Fe/As TCLP
Source Mole Extract
Ratio (ppm)
Ashanti Fresh tailings from the BIOX· neutralization circuit 3·0 2.00
Fairview Fresh tailings from the BIOX" neutralization circuit 3·4 2.62
Composite sample from the BIOX· neutralization
tailings slime dam 4.6 0.86
Fresh tailings from the BIOX" CIP circuit 6.0 1.92
Sao Bento Fresh neutralization thickener underflow 51.0 0.04
Fresh sample of blended neutralization and CIP tailings 8·7 0.13
Note: The neutralization circuit at Sao Bento treats effluent liquor from the BIOX· and pressure
oxidation circuits, as well as flotation tailings (fines fraction). The neutralization thickener
underflow is subsequently blended with the elP tailings, from both the sulfide and oxide ore
circuits, prior to disposal on the slimes dam.
Continuous neutralization resulted in environmentally acceptable effluent li-

quors ([Asl<0.5 ppm) and stable neutralization residues ([AslTCLP«5 ppm) in
all cases.
Commercial Applications
The standard two-stage neutralization process has been employed at all five
commercial BIOX® plants. Effluent arsenic concentrations are monitored on a regu-
lar basis to ensure compliance with local legislation. Standard TCLP tests have also
been carried out on selected arsenic-bearing residues from the Ashanti, Fairview
and Sao Bento commercial plants. The results, summarized in Table 3.13, indicate
that none of the residues tested exhibit the characteristics of u.S. EPA toxicity with
respect to arsenic ([As lTCLP>5 ppm), and are thus sufficiently stable for land dis-
posal, according to u.s. EPA standards. 66
Conclusion
The BIOX® process has been established as a commercially viable process for
the treatment of refractory gold ores, offering lower capital and operating costs
compared to alternative process routes such as roasting and pressure oxidation.
The neutralized effluent produced from the BIOX® process may be safely dis-
posed of to a tailings dam and conforms to the U.S. Environmental Protection
Agency specifications regarding solubility of arsenic.
The demand for BIOX® technology continues to grow and several new com-
mercial plants are expected to be constructed within the next five years.
References
1. Hansford GS, Bailey AD. A fluidised bed reactor as a tool for the investigation of
oxygen availability on the biooxidation rate of sulfide minerals at high solid con-
centrations. Minerals Engineering 1993; 6:387-396.
2. Hansford GS, Bailey AD. Factors affecting biooxidation of sulfide minerals at high
concentrations of solids. Biotechnol Bioeng 1993; 42:1164-1174.
3. Dew DW, Miller DM, van Aswegen PC. GENMIN'S Commercialization of the Bac-
terial Oxidation Process for the Treatment of Refractory Gold Concentrates. Randol
Gold Forum Beaver Creek '93, Sept. 7-9. 1993:229-237.
4. van Aswegen PC. Biooxidation of Refractory Gold Ores-The GENMIN Experi-
ence, Biomine '93 Conference, Adelaide, March 1993.
5. van Aswegen PC. Commissioning and operation of biooxidation plants for the
treatment of refractory gold ores, Milton E Wadsforth (IV) Symposium on Hy-
drometallurgy, Salt Lake City, August 1993.
6. van Aswegen PC, Marais HJ. Overview of Biooxidation Treatment of Refractory
Gold Ores. Reference not available from GENCOR Process Research, Randburg,
South Africa.
7. Acuna J, Rojas J, AM Amano AM et al. Chemotaxis of Leptospirillum ferrooxidans
and other acidophilic chemolithotrophs: comparison with the Escherichia coli
chemosensory system. FEMS Micro Lett 1992; 96:37-42.
8. Sand W, Rohda K, Sobotke B et al. Evaluation of Leptospirillum ferrooxidans for
leaching. Appl Environ Micro 1992; 58:85-92.
9. Huber H, Huber G, Stetter KO. A modified DAPI Fluorescence staining proce-
dure suitable for the visualization of lithotrophic bacteria. System Appl Microbiol
1985; 6:105-106.
10. Shrihari J, Modek JM, Kuner R et al. Dissolution of particles of pyrite mineral by
direct attachment of Thiobacillus ferrooxidans. Hydrometallurgy 1995; 38:175-187.
11. Malatt K, Ralph D. Fundamental aspects of arsenopyrite leaching. Biomine'94 Con-
ference Proceedings. Perth, W Australia 1994.
12. Ohmura N, Kitamura K, Saiki H. Selective adhesion of Thiobacillus ferrooxidans
to pyrite. Appl Environ Micro 1993; 59:4044-4050.
13. Boon M, Hansford GS, Heijnen JJ. The role of bacterial ferrous iron oxidation in
the biooxidation of pyrite. In: Vargas T, Jerez CA, Wiertz JV, Toledo H. eds,
Biohydrometallurgical Processing Vol 1. University of Chile Press 1995:153-163.
14. Jerez CA, Arredondo R. A sensitive immunological method to enumerate
Leptospirillum ferrooxidans in the presence of Thiobacillus ferrooxidans. FEMS
Micro Lett 1991; 78:99-102.
15. Rawlings DE. Restriction enzyme analysis of 165 DNA genes for the rapid identi-
fication of Thiobacillus ferrooxidans, Thiobacillus thiooxidans and Leptospirillum
ferrooxidans strains in leaching environments. In: Vargas T, Jerez CA, Wiertz JV,
Toledo H, eds, Biohydrometallurgical Processing Vol 2. University of Chile Press
1995:9-17.
16. Schloter M, Abmus B, Hartman A. The use of immunological methods to detect

and identify bacteria in the environment. Biotechnology Advances 1995; 13:75-90.
17. Garcia A, Jerez CA. Changes of the solid-adhered populations of Thiobacillus
ferrooxidans, Leptospirillum ferrooxidans and Thiobacillus thiooxidans in leach-
ing ores as determined by immnological analysis. In: Vargas T, Jerez CA, Wiertz
JV, Toledo H. eds, Biohydrometallurgical Processing Vol 2. University of Chile
Press 1995:19-30.
18. Helle U, Onken U. Continuous microbial leaching of a pyrite concentrate by
Leptospirillum-like bacteria. Appl Microbial Biotechnol 1988; 28:553-558.
19. Goebel BM, Stackebrandt E. Cultural and phylogenetic analysis of mixed micro-
bial populations found in natural and commercial bioleaching environments. Appl
Environ Micro 1994; 60:1614-1621.
20. Monnoy Fernandez MG, Mustin C, de Donato P et al. Occurrences at mineral-
bacteria interface during oxidation of arsenopyrite by Thiobacillus ferrooxidans.
Biotech Bioeng 1995; 46:13-21.
21. Dew DW. Comparison of performance for continuous biooxidation of refractory
gold ore flotation concentrates. In: Vargas T, Jerez CA, Wiertz JV, Toledo H. eds,
Biohydrometallurgical Processing. Vol 1. University of Chile Press 1995:239-251.
22. Lawson EN, Nicholas CS, Pellat H. The toxic effect of chloride ions on Thiobacillus
ferrooxidans. In: Vargas T, Jerez CA, Wiertz JV, Toledo H. eds, Biohydrometal-
lurgical Processing Vol 1. University of Chile Press 1995:165-174.
23. Silver MP, Lundgren DG. Studies on the chemoautotrophic iron bacteria
Ferrobacillus ferrooxidans, II Manometric Studies. J Bacteriol 1959; 78:326-323.
24. Lawson EN. Environmental influences on BIOX·. Bacterial Oxidation Colloquium
SAIMM 1991.
25. Tuovinen OH, Niemela SI, Gyllenberg HG. Effect of mineral nutrients and or-
ganic substances on the development of Thiobacillus ferrooxidans. Biotech Bioeng
1971; 13:517-527.
26. Denisov GV, Kovrov BG, Tnibachev IN et al. Composition of a growth medium
for continuous cultivation of Thiobacillus ferrooxidans. Microbiolgiya 1980;
49:341-341.
27. McCready RGL, Wadden D, Marchbank A. Nutrient requirements for the in place
leaching of uranium by Thiobacillus ferrooxidans. Hydromet 1986; 17:61-71.
28. Norris PR, Murrell JC, Hinson D. The potential for diazotrophy in iron and sul-
fur-oxidizing acidophilic bacteria. Arch Microbiol 1995; 164:294-300.
29. Miller DM. Effect of Temperature on BIOX· Operations, SAIMM Bacterial Oxida-
tion Colloquium, Johannesburg, June 1991.
30. Gates LE, Morton JR, Fondy PL. Selecting agitator systems to suspend solids in
liquids, Chern Eng 24 May 1976.
31. Fraser GM. Mixing and oxygen transfer on mineral bioleaching. Biomine '93 Con-
ference Adelaide 22-23 March 1993=16.1-16.11
32. Fraser GM, Kubera PM, Schutte MD et al. Process mechanical design aspects for
LIGHTNIN A315 agitators in minerals oxidation. Randol Gold Forum, Beaver Creek
'93 7-9 September 1993:247-253.
33. Mills DB, Bar R, Kirwaa. Effect of solids on oxygen transfer in agitated three-
phase systems. AIChE Journal September 1987:Vol. 33 NO.9.
34. Miller DM. Use of the logistic model for the interpretation of bacterial sulfide
oxidation. Report No. PR91/51F (Note for the Record) Internal Report GENCOR
Process Research 28 March 1991.
35. Classen R. The effect of mineralogy on the bacterial oxidation of a refractory
gold-bearing sulfide from a Barberton deposit. Colloquium on Bacterial Oxida-
tion SAIMM Johannesburg June 18, 1991.
The BroX'" Process for Biooxidation of Gold-Bearing Ores or Concentrates 79
36. Lawson EN. Private communication with the author (GENCOR Process Research,
South Africa), 1995.
37. Silver S, Budd K, Leahy K M et al. Inducible plasmid determined resistance to
arsenate, arsenite and antimony (III) in Escherichia coli and Staphylococcus aureus.
J Bacteriol 1981; 146:983-996.
38. Broadhurst JL. Arsenite build-up during BIOX· operations: A review and assess-
ment. GENCOR Process Research, Internal Report No. PR95/15 24, February 1995.
39. Broadhurst J L. Neutralization of arsenic bearing BIOX· liquors. Minerals Eng
1994; 7:1029-1038.
40. Rosengrant L, Fargo L. Final best demonstrated available technology (BOAT) back-
ground document for K031, K084, K101, K102, characteristic arsenic wastes (0004),
characteristic selenium wastes (0010), and P and U wastes containing arsenic and
selenium listing constituents. United States Environmental Protection Agency,
Report No. PB90-234014, 1990.
41. Broadhurst JL. The nature and stability of arsenic residues from the BIOX· pro-
cess. Biomine '93 Conference proceedings, Adelaide, 1993.
42. U.S. EPA toxicity characteristic leaching procedure. In: US Environmental Pro-
tection Agency Register 1991; 40 CFR, chapter I, Appendix II of part 261:66-81.
43. U.S. EPA Identification and listing of hazardous waste. In: US Environmental
Protection Agency Register 1991; 40 CFR, chapter I, part 261:27-261.
44. Robins RG, Huang JC, Nishimura T et al. The absorption of arsenate ion by fer-
ric hydroxide. In: Reddy RG, Hendrix JL, Queneau PB, eds. Arsenic Metallurgy
Fundamentals and Applications. AIME Publication, 1988:99-112.
45. Papassiopi N, Stefanakis M, Kontopoulus A. Removal of arsenic from solutions
by precipitation as ferric arsenates. In: Reddy RG, Hendrix JL, Queneau PB, eds.
Arsenic Metallurgy Fundamentals and Applications. AIME Publication, 1988:
321-334·
46. Stefanakis M, Kantopoulos A. Production of environmentally acceptable arseni-
tes-arsenates from solid arsenic trioxide. In: Reddy RG, Hendrix JL, Queneau PB,
eds. Arsenic Metallurgy Fundamentals and Applications. AIME Publication,
1988:287-303·
47. Therdkiattikul S, Dahlstrom DA. Safe disposal of liquor containing arsenic and
heavy metals from bacterial leaching of refractory gold concentrates. In: Randol
Gold Forum Proceedings. Beaver Creek, 1993:373-377.
48. Vircikova E, Molnar L, Lech P et al. Solubilities of amorphous Fe-As precipitates.
Hydrometallurgy 1995; 38:111-123.
49. Krause E, Ettel VA. Solubilities and stabilities of ferric arsenate compounds. Hy-
drometallurgy 1989; 22:311-337.
50. Krause E, Ettel VA. Ferric arsenate compounds-are they environmentally safe?
In: Impurity Control and Disposal. The 15th Annual CIM Hydrometallurgy Meet-
ing Proceedings. Vancouver, 1985:5-20.
51. Harris GB, Monette S. The stability of arsenic bearing residues. In: Reddy RG,
Hendrix JL, Queneau PB, eds. Arsenic Metallurgy Fundamentals and Applications.
AIME Publication, 1988:469-488.
52. Nishimura T, Tozawa K. Removal of arsenic from waste water by addition of cal-
cium hydroxide and stabilization of arsenic-bearing precipitates by calcination.
In: Impurity Control and Disposal. The 15th Annual CIM Hydrometallurgy Meet-
ing Proceedings. Vancouver, 1985:3-1 to 3-18.
53. Barrett J, Hughes MN, Simons C. The nature and stability of precipitated solids
from arsenopyrite biooxidation effluents. In: Randol Gold Forum Proceedings.
Cairns, 1991:179-183.
54. Robins RG. The stability and solubility of ferric arsenate: an update. In: Gaskell
PR, ed. EPD Congress Proceedings. The Minerals, Metals and Materials Society,
1990:93-104·
55. Robins RG, Wong PLM, Nishimura, T et al. Basic ferric arsenates-nonexistant.
In: Randol Gold Forum Proceedings. Cairns, 1991:197-2.00.
56. Swash PM, Monhemius AJ. Hydrothermal precipitation from aqueous solutions
containing iron (III), arsenate and sulfate. Hydrometallurgy '94 Proceedings.
Chapman & Hall; 1994:177-190.
57. Lawrence RW, Marchant PB. Biochemical pre-treatment in arsenical gold ore pro-
cessing. In: Reddy RG, Hendrix JL, Queneau PB, eds. Arsenic Metallurgy Funda-
mentals and Applications. AIME Publication, 1988:199-2.11.
58. Morin D, Olliver P, Toromanoff I et a1. Bioleaching of gold refractory arsenopy-
rite rich sulfide concentrate: laboratory and pilot case studies. In: Gaskell PR, ed.
EPD Congress Proceedings. The Minerals, Metals and Materials Society,
1991:577-589.
59. Adam K, Komnitsas C, Papassiopi NL et al. Stability of arsenical bacterial oxida-
tion products. In: Hydrometallurgy '94 Proceedings. Chapman & Hall, 1994:2.91-311.
60. Longhans D, Lord A, Lampshire D et a1. Biooxidation of an arsenic-bearing re-
fractory gold ore. Minerals Engineering 1995; 8 (112.): 147-158.
61. Nishimura T, Itoh CT, Tozawa K. Stabilities and solubilities of metal arsenites
and arsenates in water and effect of sulfate and carbonate ions on their
solubilitites. In: Reddy RG, Hendrix JL, Queneau PB, eds. Arsenic Metallurgy Fun-
damentals and Applications. AIME Publication. 1988:77-98.
62.. Teixera LA, Monteiro AG, Kohler HM. The detoxification of effluents containing
arsenic with iron sulfate and hydrogen peroxide In: Gaskell PR, ed. EPD Con-
gress Proceedings. The Minerals, Metals and Materials Society, 1990:189-2.01.
63. Barrett J, Ewart DK, Hughs MN et a1. The oxidation of arsenic in arsenopyrite:
the toxicity of As(III) to a moderately thermophilic mixed culture. In: Biohydro-
metallurgy Conference Proceedings, 1989:49-57.
64. Barrett J, Ewart DK, Hughs MN et al. Chemical and biological pathways in the
bacterial oxidation of arsenopyrite and arsenic (III). Paper presented at the
Biohydrometallurgy Symposium, Portugal, 1991.
65. Kust RN. Improved Technology for precipitating metal hydroxides. In: Reddy RG,
Imrie WP, Queneau PB, eds. Residues and Effluents-Processing and Environ-
mental Considerations, Warrendale, PA. The Minerals, Metals and Materials So-
ciety, 1991:793-800.
66. U.S. EPA. Land disposal restrictions. In: U.S. Environmental Protection Agency
Register 1991; 40 CFR, chapter 1, part 2.68:744-846.
CHAPTER 4
The Desip and Operating Practice

of Bacterial Oxidation Plant
Using Moderate Thermophiles
(The BacTech Process)
Paul C. Miller
Introduction
I t has been most satisfying to see the emergence of bacterial leaching technology
from laboratory experimentation through to commercial acceptance. The use of
moderate thermophiles for commercial bacterial leaching processes is an evolving
area which offers a number of possible advantages over the use of more traditional
mesophilic (lower temperature) organisms.
BacTech has been the first company to develop and commercialize a tank
bioleach process with moderate thermophiles (BACOX), which is currently in use
for the oxidation of a refractory gold concentrate at the Youanmi mine in Western
Australia. This process has been successfully operating for over two years and the
technology is now being developed and adapted for the potential treatment of base
metal sulfides. In particular, this includes process technology for the leaching of
copper from chalcopyrite and the leaching of nickel and cobalt from complex
polymetallic concentrates.
The objective of this chapter is to discuss the process application of moderate
thermophiles, which have optimum temperatures for growth of between 45°C and
55"C. This chapter draws upon the experience gained from the Youanmi operation
and examines the possible choices available to the engineer in plant design and
operation. Many of the issues discussed are also likely to be applicable to bacterial
leaching processes using mesophilic bacteria. The development of the BacTech tech-
nology for the treatment of base metal sulfides is at an advanced stage and there is
confidence that it will shortly progress through to commercialization. For this rea-
son, while the chapter is biased towards the treatment of refractory gold materials,
it also aims to translate the experience gained from current plant practice into the
requirements for the treatment of base metal sulfides.

Excellent reviews and publications have been produced by both Brierley' and
Norris 2 concerning the characterization and microbial aspects of thermophiles and
such topics will not be discussed here. After a brief introduction to the current
reasons for considering bacterial leaching as a process option, an overview of the
process flowsheet used at the Youanmi mine is given together with a summary of
process economics.
Most of the chapter is dedicated to a discussion concerning the appropriate
plant design and operation to meet the reaction criteria. A short discussion is in-
cluded concerning the methods of integration of bacterial leaching into the
flowsheet in order to satisfy the upstream and downstream processing operations.
As a biological process, the philosophy of plant operation is quite different than
that of other on site operations and the chapter highlights these areas of differ-
ences through experience gained by staff in commissioning and troubleshooting.
The chapter concludes by giving a brief summary of the potential future use of
moderate thermophiles in bacterial leaching plant and reference is made to areas
of development which will broaden the commercial acceptance of this new tech-
nology in the mining industry.
Current Reasons for Considering Bacterial Leaching

A large number of articles have made reference to the potential advantages of
using bacterial leaching in comparison to traditional process routes. Early refer-
ences sometimes made an incorrect deduction that as a simpler concept, bacterial
leaching was inevitably a lower cost alternative. Pressure oxidation and high tem-
perature technologies had proven track records of both technical and economic
viability and were preferred alternatives. This led to the restriction of bacterial
leaching to heap leaching or in situ leaching projects on low-grade materials. The
presence of a number of commercial bioleach tank operations has now helped to
redress this balance in demonstrating the technical and economic viability of the
bacterial process. Further to this, environmental restrictions are now eliminating
smelting or roasting options on many projects, thereby giving a narrower field of
process choice.
The current practical reasons for considering bacterial leaching for refractory
gold or base metal projects as given by potential operators may be listed as follows:
1. Environmentally more stable and desirable end products are produced. For
the treatment of arsenical refractory gold ores the arsenic is stabilized
through precipitation instead of an impure arsenic byproduct from roast-
ing, which is difficult to dispose of, expensive to purify and usually
unsaleable.
2. By adopting bacterial leaching technology the time taken to obtain environ-
mental permits may be shorter, resulting in a reduction in lead times for
exploitation of an orebody.
3. The cost and time for monitoring and reporting of waste products is often
reduced with bacterial leaching processes.
4. Both the capital and operating costs may be lower. However, the process is
sensitive to the cost of power for the provision of air, and in the case of base
metals, the cost of power for downstream processing if solvent extraction
and electrowinning are to be considered.
5. Bacterial leaching can have a dual function by releasing valuable base met-
als into solution and for recovery and exposing gold values in the residue
Bacterial Oxidation Using Moderate Thermophiles
which can be recovered by cyanidation. This is of particular interest for the

leaching of gold-bearing chalcopyrite concentrates.
6. The equipment requirements in terms of agitation, tankage and aeration,
together with the control systems are relatively unsophisticated. This makes
the technology appropriate to environments where there is little support in
terms of skilled personnel and spares for maintenance.
7. Construction time is sometimes reduced by using bacterial leaching tech-
nology, thereby reducing the lead time to exploit an orebody.
8. The moderate thermophile-based process has demonstrated a robustness
in which the inertia of the system results in a 'forgiving' process which is
able to cope with a wide range of process fluctuations.
9. The operator has greater onsite control over final metal production as op-
posed to producing a concentrate for transport to an offsite contract smelter.
The financial pipeline may also be shortened and be more predictable with
bacterial leaching.
10. Bacterial leaching is versatile enough to accept a variety of head grades in
terms of sulfide values and mineralogical variations over a period of time.
This makes it more flexible for the exploitation of an orebody in which varia-
tions in sulfide mineralogy may occur.
11. Bacterial leaching may accept a lower sulfide grade of concentrate than com-
peting technologies, which often leads to a higher recovery of a few percent
in concentrate production.
12. Conceptually, the final acidic bacterial liquor produced may be a useful
lixiviant for the heap leaching of low-grade sulfide or oxide material. This
may also offset the cost of neutralization of the final liquor which is required
to produce a ferric precipitate suitable for disposal.
13. When using thermophilic organisms, the potential exists to take advantage
of the catalytic effect of temperature on certain reaction rates and a lower
heat load has to be removed from the exothermic system to maintain a set
temperature above ambient conditions.
14. For the treatment of refractory gold ores, a greater control is available to
minimize the oxidation necessary to expose gold values for recovery, thereby
reducing the oxygen requirements for oxidation. Also, when using thermo-
philic organisms to extract base metals, a potential may exist for the forma-
tion of sulfur intermediates rather than sulfuric acid, thereby reducing the
aeration requirements and neutralization costs.
Oxidation of Refractory- Gold Concentrates-

Process Flowsheet and Economics
It is current practice to produce a concentrate for treatment by tank leaching
rather to treat the whole ore. The cost of beneficiation to produce a concentrate is
often minimal in comparison to the cost factors associated with a bacterial oxida-
tion tank operation for handling large tonnages of ore. Waste gangue rejection at
an early stage also minimizes reagent costs such as acid requirements to satisfy
carbonate values in the ore during bacterial oxidation. Whole ore treatment is usu-
ally confined to heap leaching oflow-grade materials although heap leaching tech-
nology is itself an area of promise for new developments. 3
In the case of bacterial oxidation of refractory gold concentrates the economic

viability can be dependent upon the gold to sulfide ratio of the concentrate. This
gold value refers to the percentage increase recovery obtained as a result ofbacte-
rial oxidation pretreatment while the sulfide value refers to the minimum quantity
of sulfide oxidation necessary to achieve this. It is essential therefore that there is a
clear understanding of the relationship between sulfide oxidation and gold recov-
ery at an early stage in the project. When this is known it is usual to conduct a
costing exercise at ±30% or even ±50% sulfide oxidation to demonstrate the over-
all economic feasibility of a project.
For example, the feed to the thermophilic bacterial oxidation process at Youanmi
is a cleaner flotation concentrate upgraded from the ore at a 5:1 ratio, giving a
concentrate feed averaging 50-60 glt gold containing between 20-30% sulfur. Only
50% of this gold would be directly recoverable by cyanidation without oxidative
pretreatment. The majority of the remaining gold values can be exposed byoxi-
dizing only 30% of the sulfide values. Further to this, the majority of these gold
values are associated with the arsenopyrite fraction as opposed to pyrite and the
arsenopyrite is oxidized preferentially due to its less refractory nature. These are
beneficial features which are not inherent to all projects.
The Youanmi project has a concentrate throughput rate of about 120 t/dayand
experiences current operating costs through bacterial oxidation of less than U.S.$30
per tonne of concentrate. In effect this equates to 300 g of gold cost equivalent to
oxidize 11 t/day of sulfur, liberating about 3,500 glday of gold values solely due to
bacterial oxidation. For this material there is a good relationship between the quan-
tity of sulfide oxidized and the gold liberation. For some projects the gold to sul-
fide ratio is such that the tonnage quantity of sulfide to be oxidized leads to high
capital and operating costs by virtue of longer residence time and higher air de-
mands for oxidation. This cannot always be justified if relatively low gold values
are present.
The grade of concentrate produced is important as it is not always possible to
produce a high-grade concentrate with a good recovery due to the mineralogy of
some complex sulfides. Bacterial oxidation then becomes a favored option as a
lower grade of sulfide feed can often be acceptable and it may be possible therefore
to increase the overall recovery in concentrate production. This aspect is discussed
in greater detail in the section on upstream considerations for bacterial oxidation.
A process flow diagram of the Youanmi plant is given in Figure 4.1 and has
been discussed previously by Budden and Bunyard. 4 Many of the unit operations
are common to other commercial bacterial oxidation operations in terms of: crush-
ing; grinding; flotation concentrate production; bacterial oxidation; CCD thicken-
ing and washing; neutralization; and cyanidation. The flotation tails at Youanmi
contain gold values in oxide components and therefore these report to the
cyanidation circuit for gold recovery.
Brierley and Winby5 have previously detailed the design criteria used for the
Youanmi plant together with the performance and cost for the BacTech plant. An
update of these criteria is presented as a summary in Tables 4.1 and 4.2. These
tables indicate the improved performance and lower operating costs which have
been realized over the past year by comparing more recent cost data with that
previously presented.
Fig. 4.1. Youanmi b:I
COOUNGTOWERS I:>
plant flowsheet. ::;.
...
'"§.:
IIh\1 ==
1,,(1L~""Elt
CRUSHED
ORE
~o ~
I I
I
'--_---1..__ ,
~
.....
o·
;:s
!) :~X
REGRIND MILL J BLOWER
s:
~.
("'~ """",
~
l:>...
19 '"~
'---1-;:"-).,:..."-..,, ~',~ ~
I ~ ~ Ir ;i
SLAKEDUME
WASH WATER BACTERIA LOXIDATION[ J ; J
CYANIDE 3'"
~
CCDTHICKENER [
-----;-l ~
TA LUNGSTANK
REGEl'ERATED
II F GROUN D
CA LCRET
~19 reo"~ "',,' r
LOADED CARBON TO
~~~'", '"~11 ill=rl ~
GOLD ROOM
NEUTRALIZATION
~
Table 4.1. Comparison ofdesign criteria (left column) and current operation
for BacTech biooxidation plant at Youanmi mine, Western Australia (Nov.
1996)
Ore Treatment Rate (mtpa) 200,000 204,000

Feed Grades (Ore)
Au (glmt) 15
Ag (glmt) 2
S (%) 7·5
As(%) 1.0
Sb(%) 0.1
Flotation Recoveries
Mass (%) 20 21·5
Au(%) 80-85 88-90
S (%) 92-97
As (%) 82-85
Fe (%) 68-72
Flot Cons Treatment Rate
mtpa 40,000 43,860
mtpd 120 120
Flotation Tails
Au (glmt) 2.8
Flotation Concentrate Grades
Au (glmt) 60 50
S (%) 28 28
As(%) 4·3 2.8
Fe (%) 26 16
Bacterial Oxidation
Retention Time (Days) 5 3·8
Slurry Density (% Solids) 15 18
Number of Stages 6 4
Primary 3/4 3/4
Secondary 3/2
Operating pH 1.2 1.5
Operating Temperature (0C) 45-55 49-50
Arsenopyrite Oxidation (%) 85/95 96
Pyrite Oxidation (%) 28 30
Total Sulfur Oxidation (%) 32 34
Sulfur Oxidation (mt/day) 10.8 11
CCD/Neutralization
Wash Stages 3 3
Area (m1/mt/day) 0.65/0.24
Total Flocculant Usage (glt) 250 300
Neutralization Stages - Liquor 4 4
- Solids 1
Residence Times - Liquor (h/stage) 1·5 1·5
- Solids (hlstage) 2 2
Solution Neutralization pH 3.5/5.5 3.2/5.6
continued...
Bacterial Oxidation Using Moderate Thermophiles 87
Cyanidation (Oxidized Cons)

Gold Dissolution (%) 90-95 90-95
Silver Dissolution (%) 50
Leach Slurry Density (% Solids) 35 43
Leach Time (Hrs) 32 38
Adsorption Time (Hrs) 19 19
Carbon Loading (g/t - Au + Ag) 8,000 2,500
Reagent Consumptions
NaCN: Cons Leach (kg/mt cons) 7·5 7·5
Flot Tails Leach (kg/mt tails) 0·5 0·5
Total (kg/mt Ore) 1·9 1·9
Quicklime: Cons Leach (kg/mt cons) 25·0 22.0
Flot Tails Leach (kg/mt tails) 2.0 0.8
Total (kg/mt ore) 16.6 5·0
Calcrete Solution Neut. (kg/mt cons) 190
Aspects of Reactor Design and Operation

Conceptual Process Requirements
The translation of the reaction requirements for bacterial leaching into a prac-
tical full-scale reactor system is an area under constant review, with improvements
being made on existing plant and research being undertaken to develop new de-
signs. 6,7 It is well recognized that a critical aspect of the process is the provision of
aeration and mixing of the pulp to ensure that sufficient dissolved oxygen is avail-
able for bacterial growth while satisfying the requirements of the oxidation reac-
tions. 8 The requirement for cooling of reactors to remove the excess heat load is
also a critical area and the use of thermophiles may offer potential advantages in
reducing this requirement. 9,lo The following section discusses these issues in more
detail and aims to offer constructive criticism and reasoning on aspects of plant
design and operation.
Materials of Construction
The aggressive nature of the reaction environment, using highly oxidizing acidic
slurries at a minimum temperature of 50°C, creates many technical challenges,
particularly in the choice of materials for plant construction. Careful consider-
ation is required for the materials of construction in both the bacterial oxidation
reactors and downstream areas. The choice is either to use rubber lined mild steel
or stainless steels, both of which have advantages and disadvantages. Rubber lined
mild steel is sometimes favored for its slighdy lower cost and continuity with other
areas of the plant, while stainless steels are favored for their corrosion resistance
and heat transfer properties, giving greater choices in the methods of heat removal
from the reactors. Further reference is made to this in the discussion on methods
of heat removal. Failure of the rubber lining on mild steel tanks would result in
catastrophic failure and as a relatively high power input is required to the reactor
for mixing and aeration, the slightest imperfection in the lining or adhesion is
rapidly propagated into a major weakness. Failure of the rubber lining on agita-
tors results in disabling of the agitator due to corrosion in hours rather than days.
Table 4.2. Performance and cost for Youanmi's BacTech plant
(Dec. 9s-Feb. 96)

Monthly Average Nov. 96
Concentrate Treated - mt/month 2820 36 45

Average %As 2·46 2.8
Average % Fe 17·53 16
Average mt/hr to each primary tank 1.29 1.67
Bacterial Oxidation Consumable Costs

Contractor's Service 1582 1400
Equipment (Leased & Minor) 272 1900
Flocculant 1896 6210
Limestone (Note 1) 3000 6420
Nutrients 1859 4890
Sulfuric acid (Note 2) 27565 5580
Total 36174 26400
Bacterial Oxidation, Maintenance, and Material Costs

Contractor's Maintenance 145 400
Iron and Steel 95 0
Oil and Gas 10 0
Pipes and Valves 5476 2800
Maintenance, Mechanical 419 930
Maintenance, Agitators 524 0
Electrical and Instrument Spares 2469 1400
Maintenance, CCD Circuit 3360 3891
Maintenance, Neutralization 125 120
Maintenance, Compressor 698 720
Pump Spares 3682 4560
Miscellaneous 3 345
Total 17428 15166
Power Costs (Note 3 & 5) 33750 475 60

Labor Costs (Note 4) 17500 17500
Grand Total 104852 106626

U.S.$/MT of Concentrate Treated 37 29
Note 1: Limestone is cost of mining local calcrete. Mining is done in short campaigns.
Note 2: Acid consumption is high, because of the high throughput rate and the fact that little
pyrite oxidation occurs in the primary tanks.
Note 3: Power is not separately metered. Installed power is: three primary tanks at 75 kWeach;
one secondary tank at 55 kW; one blower at 270 kW. The total power is 600 kw at 90% full
load and 100% running time at 8.6 U.S. cents/kW-hr.
Note 4: One operator/shift @ U.S.$45,000Iyrlperson + one-half tradesman.
Note 5: Power requirement has increased not only due to extra tonnage but due to an increase
in ancillary equipment on site.
Bacterial Oxidation Using Moderate Thermophiles
When considering the use of stainless steels, an assessment of the most suit-
able type is required. Also, long lead times may be necessary on ordering, and a
very detailed quality control and inspection program is required during construc-
tion to ensure no damage or flaws occur. The chloride levels likely to be experi-
enced in solution must be considered when selecting stainless steels to ensure that
they are below the threshold of corrosion. In this sense a flrm understanding of
the water quality to be used through the life of the project is required not only to
ensure that it is below bacterial toxicity threshold with respect to total dissolved
solids, but also that the material of construction is suitable.
An area of major concern is poor welding which would lead to corrosion propa-
gation, and considerable care is required during tank fabrication to prevent slight
damage which may propagate to a major flaw during operation. This is relevant
both on the outside and inside of the reactors.
The material of construction must be nontoxic to the bacteria, and in the case
of rubber lining it is necessary to have a more comprehensive flushing program
during commissioning to leach out any organic toxins from the lining before plant
startup. It is also necessary to ensure that the adhesive used for attaching the lin-
ing is non-sulfur-based as the bacteria will destroy the adhesive.
Consideration must also be given to the availability of skilled labor for fabrica-
tion of the chosen material and the ability to carry out repairs according to a regu-
lar maintenance schedule.
Galvanized materials cannot be used in any associated areas of construction
for projects in which arsenic is present in the leach slurry due to the potential for
producing highly toxic arsenine gas. All of these considerations led to the use
of stainless steel reactors using SAF2205 for the Youanmi project. l l Stainless
steels reported to be used on other plants using mesophilic bacteria, which have
been reported include 304L for reactors where chloride levels are lower and SAF2304
for CCD thickeners.t~13
The use of stainless steel at Youanmi has given a relatively trouble-free mainte-
nance schedule and stainless steel may remain the preferred material for future
projects. Corrosion inspections are a regular feature of the maintenance schedule,
particularly with respect to internal welds due to the relatively high energy input
to the reactor system. It is believed that the maximum potential for corrosion in
the tankage occurs in the free board area at the pulp air interface at the top of the
tanks and consideration of an epoxy lining may be appropriate here even when
using specialized steels.
Aeration and Agitation

The provision of dissolved oxygen by sparged aeration represents a large com-
ponent of the capital and operating cost in bacterial oxidation. Agitation has the
combined duty of pulp suspension and of shearing slugs of air for dispersion
through the pulp. The agitation system at Youanmi creates a relatively low shear
environment with the use of A315 rubber sheathed impellers as opposed to tradi-
tional energy intensive turbines normally associated with slurry mixing. The use
of axial flow impellers for bacterial leaching applications has been discussed by
Fraser. 8 Controversy remains concerning the quantiflable effect of shear on bacte-
rial activity, and development work on impeller types will undoubtedly continue
to be an area of interest. Seventy-flve kW motors are used on the three primary
reactors at Youanmi sized at 480 m3 with 55 kW motors on the three secondary
reactors which are of similar volume; at this later stage of reaction, it was believed
that the pulp would have a lower specific gravity requiring less power for mixing.
Also, a lower air input would be required therefore less power for air dispersion
was needed. The use of smaller sized agitators resulted in only a small savings in
cost, and for the sake of reactor uniformity, it is likely that all agitators on future
plant would be of similar size. The cost of agitation is a much smaller cost compo-
nent in comparison to the cost of air addition and leads to speculation that there is
benefit in examining marginally higher power agitation systems which could make
more effective use of the air supplied.
A potential disadvantage to using thermophilic organisms relates to a decrease
in the solubility of oxygen in the pulp at an increased temperature.14 However, in
the commercial application at Youanmi no evidence has been obtained to suggest
that this has an effect at the current operating temperature of soCC. This could
become an important issue if thermophilic processes are developed to operate at
much higher temperatures.
Air supply is the most critical process service required for bacterial leaching. It
is believed that failure of supply may give rise to serious consequences in terms of
loss of bacterial activity for a considerable time even after the air supply has been
resumed. To date no major failure has occurred in the air supply on the Youanmi
site to allow such consequences to be fully understood. However, failures over a
short time period of a few hours have not resulted in any perturbation of the pro-
cess when the air supply has resumed. In the event of a power failure the order of
priority is aeration; temperature control; agitation and then nutrient addition. Two
blowers are available with one operating while the other is on standby, and they
are provided with an independent power supply and control systems. Each blower
is rated at a provision of 7,200 m 3 free air per hour.
The stoichiometric demand for the provision of oxygen to solution to satisfy
the oxidation reactions is very large, such that a tonne of oxygen is required while
for arsenopyrite 0.7 tonnes of oxygen is required for every tonne of pyrite oxi-
dized. The requirement at Youanmi for only 30% oxidation of the sulfide values, a
portion of which is mineralized as arsenopyrite, benefits the operation.
It is critical that the oxygen mass transfer rate from gaseous air into pulp solu-
tion be greater than the demand, and at the Youanmi site the residual dissolved
oxygen level in the pulp is maintained above 2 ppm to ensure this. The blower
capacity for the provision of air is exaggerated beyond that required for the sto-
ichiometric demand, and the oxygen transfer efficiencies used for design purposes
are typically as low as 20-25%. These factors are believed to be a common design
feature of other bacterial oxidation plants and Youanmi is not an exception. How-
ever, the actual efficiencies of air utilization are believed to be higher than 2S%,
but conservative design parameters are commonly used by manufacturers to en-
sure warranty performance on equipment. Technically, the low efficiencies are due
to a number of factors such as:
1. Poor dissolution properties of oxygen into water suggest saturation values
of between only 5.S ppm to 9 ppm using air.
2. The potential for gas transfer may be decreased further by operating in a
slurry environment rather than water.
3. Neither the depth of reactors used, nor the mixing characteristics allow for
the best transfer of oxygen from the air before disengagement.
In summary, the overall power requirements for the provision of oxygen for
bacterial oxidation is generally recognized to be an order of magnitude higher
than that associated with cyanidation practices in gold extraction. The design of
reactor mixing and aeration systems for the relatively unique environment of bac-
terial oxidation will undoubtedly be an expanding area of interest for research
development and innovation.
Provision for Heating and Cooling

Although heating of reactors is generally required at process startup, (in order
to facilitate initial bacterial growth and oxidation), under normal operating con-
ditions considerable quantities of heat have to be removed to maintain the reac-
tors at the temperature most conducive for bacterial activity in the case of the
moderate thermophile culture used at Youanmi. This temperature is 50°C. Devia-
tions from the optimum temperature may have serious consequences, although
no system failure has occurred at Youanmi mine for this effect to be studied in
detail. At a lower temperature the oxidation reactions will become slower due to a
decrease in bacterial activity, while cell death is likely to occur at temperatures
higher than 60°C.
The quantity of heat generated is a function of the mineralogy and the degree
of sulfide oxidation performed. The heat produced by oxidation of arsenopyrite is
9415 kJ/kg while that for pyrite is 37% higher at 12,884 kJ/kg. As the principal min-
eral of oxidation at Youanmi mine is arsenopyrite, less heat is generated than if
complete pyrite oxidation was required. The above values still require the removal
of a considerable heat load. The energy input by agitation is also a factor which is
accounted for in the design for heat removal, but is trivial in comparison to that
generated by sulfide oxidation.
The heat losses that occur in the system are due principally to four factors:
convective heat losses; heat absorption by incoming feed and loss in exit slurry
product; heat loss due to moisture in the air leaving the system; and evaporative
losses. Heat losses are also controlled by the ambient operating conditions of a
particular site with respect to temperature variations and wind velocity. The abil-
ity to operate at a higher temperature by using moderate thermophiles in com-
parison to mesophiles has an advantage in creating a higher temperature driving
force, thus increasing heat losses further and reducing the heat load for removal
by a cooling system. This gives a considerable process advantage, particularly in
operating environments with even modestly high ambient daytime temperatures.
It is evident that the quantity of heat to be removed is process-specific and the
magnitude of heat generated may be of the order of 3 MW to 12 MW for a reason-
ably small plant tonnage throughput of 100 tpd concentrate. As the heat load is a
direct reflection of the quantity of sulfide oxidation, this may vary through the life
of a project as a progression is made through transition ores to higher grade sul-
fides. The process design has to be robust enough to cater to these likely variations.
The use of internal reactor coils or reactor jackets are currently the two favored
design methods for control of the reaction temperature. If rubber lined mild steel
tanks are considered then the cooling options may be limited to the use of internal
coils due to the insulating quality of the rubber. This is unless a heat exchanger is
used external to the reactor through which the oxidizing slurry is passed and then
returned to the reactor. This method is less likely to be favored but may provide an
alternative to be considered in the future.
Specialized stainless steel reactors were used at Youanmi due to the simplicity
of being able to use water jackets for both a heating and cooling duty. The Youanmi
jackets are in effect 'water curtains' located as an external shell towards the base of
the reactor. The water jackets operate under atmospheric conditions thereby sim-
plifying the construction and design. However, it is recognized that if complete
oxidation of pyrite were required then the jackets may not be adequate to remove
the extra heat load generated.
The use of internal reactor coils for heating and cooling has a number of disad-
vantages. Firstly, translation of the heat duty required for a system results in calcu-
lated values which suggest that a few kilometers of stainless steel piping is required.
Due to the relatively high energy input by aeration and agitation, all reactor internals
must be extremely well secured and methods of mounting the stainless steel coils
inside the reactor require careful consideration. The presence of the coils also has
the potential for disturbing the geom~try of pulp flow and air bubble residence
time and occupy valuable internal reactor space. The coils must be designed for
ease of removal and replacement in the event of failure, and the antiscalant used to
maintain high efficiencies in the cooling circuit must be non toxic to the bacteria.
Failure of the coils could result in catastrophic consequences to the process with
considerable delay in being able to replace or repair coils and bring the reactor
back on line.
Irrespective of the method of cooling used, it is believed that failure to provide
cooling water may result in a 'run away' reaction with the temperature increasing
to levels well above bacterial tolerance. Although no such event has occurred at
Youanmi, it is believed that depending on the severity of overheating, it may take
many days or possibly weeks for a system to fully recover. This is the major reason
for considering the simplest method possible for the removal of the excess heat
generated from the process. The use of stainless steel jacketed vessels at Youanmi
has provided a trouble-free process for temperature control.
The hot return water from process cooling reports to a cooling water tower
before being returned to the circuit (see later section on process control). The lo-
cation of this tower needs to be sited away from high dust areas or upwind of the
plant to minimize cooling water contamination with airborne particles.
Provision of heating is likely to be required during the startup of reactors. For
the use of moderate thermophlles, hot water is supplied to the jackets to maintain
a temperature of about 46"C in the bacterial oxidation pulp. The hot water can be
sourced from waste heat generators or from a specific purpose heater. When the
temperature of the pulp increases to 48°C (indicating that oxidation has initiated),
cooling water is supplied to the water jackets and a normal operating temperature
of 50"C is maintained by removing the excess heat generated.
This protocol generally applies at commissioning for the startup of each reac-
tor in a sequential manner such that each reactor is operated in batch mode before
starting a continuous feed. When a reactor is brought back into service after main-
tenance, initial heating to start the reaction is not always necessary provided that a
slightly greater time can be allowed for a reactor to be integrated back into the
normal operation.
Feed Control, Nutrient and Reagent Addition

Issues relating to the upstream process operations of concentrate production
including flotation are discussed later. The first stage of feed control to bacterial
oxidation makes use of a feed stock tank and dilution tank prior to bacterial oxi-
dation. The flotation concentrate is fed into the stock tank as a 50% w/w thickener
underflow. This stock tank is essential in order to act as a feed storage buffer to
allow for any disruption in milling and concentrate production. The dilution tank
downstream of the stock tank provides the tankage into which dilution water is
added to adjust the pulp density to between 15-20% before feeding to bacterial
oxidation.
At Youanmi, the pulp feed from this dilution tank is circulated around a ring
main located above the primary reactors and discharged into each reactor indi-
vidually. The feed rate to each reactor can be regulated by means of a valve on the
discharge line to a particular reactor. This is a better alternative to the feed enter-
ing a splitter box arrangement where it could only be diverted in equal amounts to
each of the primary reactors.
Nutrients are prepared in a holding tank of two-day capacity from dry solids
with the addition of clean water ofless than 10,000 mglL TDS. The exact nutrient
composition is considered to be proprietary information, but consists of low cost
fertilizer type ingredients for the provision of nitrogen, phosphates, potassium
and magnesium. The nutrient solution is transferred into a holding tank and me-
tered from the holding tank to the slurry stock tank. A reasonable quality of dilu-
tion water is also required for pulp dilution to avoid potential contamination of
the bacterial oxidation process circuit.
The concentrate used in the Youanmi operation contains some carbonate val-
ues which give rise to an occasional need for acid addition. As only 30% oxidation
is required to expose gold values, a relatively small amount of acid is produced
during sulfide oxidation. Therefore when reagent additions are required to con-
trol the pH, they are made manually. The pH is generally regulated such that its
value is between 1 to 2 but considerable short-term deviations can occur with no
apparent adverse effect on the process.
Residence Time, Pulp Density, Reactor Configuration,

Pulp Transfer and Discharge
The most common reason cited against the acceptance of bacterial oxidation
processes is the longer residence time required for oxidation in comparison to
competing technologies. A critical issue is the extent of oxidation necessary in terms
of tonnage of sulfide values requiring oxidation. Within limits, it is possible to
consider the effectiveness of a bacterial oxidation plant in terms of the kilograms
of sulfur oxidized per cubic meter of reactor volume per day. This feature has been
noted by other workers 15 and it appears that a larger tonnage oflower grade sulfide
material can be treated over the same time period as a smaller tonnage of high-
grade sulfide material or that at a fixed pulp density the residence time can be
varied for different sulfide concentrations. There are clear limits to this in that if
the grade of material is too low, the residence time required in the reactor is too
short to enable the bacteria to grow and divide within the system and washout of
the bacteria will occur. If efforts are made to increase the pulp density and extend
the residence time then high pulp densities may be detrimental to bacterial activ-
ity. Conversely, if the grade of a material is too high requiring a longer residence
time, the ability to accommodate the products of reaction into solution may be-
come too difficult and thereby influence the overall rate of oxidation. This is un-
less a very low operating pulp density is considered, which is often unreasonably
dismissed by process engineers. The situation is more complex for the treatment
of base metals as the requirement of liquor grades for solvent extraction may im-
pose constraints on parameters such as pulp density for single pass bacterial leach -
ing systems.
Currently the Youanmi process operates with a mean residence time of just
under four days, although the original design criteria was for five days. A resi-
dence time of about three days is possibly the minimum practical residence time
to allow for bacterial growth and doubling to maintain an active population. As
this feature of bacterial growth represents a process constraint, a linear relation-
ship between residence time and sulfide oxidation is unlikely to exist when very
low levels of oxidation are required. Testwork has suggested that complete oxida-
tion of the Youanmi material may be achieved within a six to seven day residence
time. This suggests that there are practical limits to comparing plant efficiencies
on a tonnage of sulfide oxidized basis. The pulp density at Youanmi is 180/0 w/w and
this is similar to other bacterial oxidation plants.
With respect to reactor configuration, design theory suggests that a four-stage
reactor system is necessary in order to reduce short circuiting of feed material
from exiting the reactor system before being oxidized. Common design practice is
for the volume of the first stage to be at least double or triple the size of the later
bacterial oxidation stages. This is usually achieved by the use of multiple tanks in
the primary stage. This increases the residence time of the first stage to allow for
bacterial growth and division before material is carried through to later stages of
oxidation. It also provides a greater buffering capability against the adverse effects
of the fresh incoming feed. For example, if the feed has acid-consuming carbonate
values it may increase the pH value of the liquor in the first stage and therefore
require acid addition. However, a higher residence time in the first reactor pro-
motes oxidation of pyrite values thereby generating sulfuric acid to counteract
this acid consumption. The initial dissolution of arsenopyrite is also acid-consum-
ing with acid generation only being produced on precipitation of ferric arsenate
which may occur in the later stages of oxidation.
Although the volume of the primary stage is larger than subsequent stages, it is
usual to use reactor vessels of equal size. This uniformity is important as it pro-
motes flexibility in a circuit by being able to change the use of reactor vessels from
first-stage to secondary or later stage reactors and allows an easier maintenance
schedule. The versatility in having a greater number of reactors may also be im-
portant as grades and tonnage throughput may change through the life of a project,
giving redundant capacity. In some cases it is also conceivable that bacterial oxi-
dation reactors have a dual purpose over the life of a project-using as cyanidation
tanks in earlier stages of a project for the treatment of oxide material and using a
bacterial oxidation process for sulfide zones mined at greater depths within an
orebody.
The Youanmi project uses six 480 m3 reactors, three of which are classified as
primary reactors and the remainder as secondary. In normal operation the incom-
ing feed is distributed equally between the three primaries but can be varied as
required. All reactors are at the same level on concrete pads and have pulp inlet
and outlet configurations which allow reactors to be used as either first-stage pri-
mary or later stage reactors. This is achieved by having a common outlet manifold
to which pulp can report from the primary reactor outlets before being fed to the
secondary reactors. Air lifts are used to transfer pulp out of reactors. The layout of
the six reactors is such that there are two lines of three reactors and the air and
cooling water services and pulp outlet mains are positioned between the two lines.
The secondary reactors are designed to form a cascade of stages with the pulp
outlet from the final reactor reporting to the counter-current decantation circuit.
This type of circuit, using three primary reactors in parallel followed by three sec-
ondary reactors in series has been reported in use at other bacterial oxidation
plants. IS However, a number of alternative configurations have been tested at
Youanmi.
Control and Monitoring

A PLC-based (Proportional Logic Control) control system is used, allowing op-
erator interrogation with control instrumentation scanning and continuous data
acquisition. However, the control systems are relatively simple and the control loops
for the reactors are limited to the temperature and airflow for each reactor using
set points. As dissolved oxygen probes which can withstand the rigors of operating
in the aggressive pulp environment are now available, it is possible that the air
supply to reactors will be controlled from feedback measurement of residual dis-
solved oxygen concentration in the pulp. Measurement of the depletion of oxygen
and carbon dioxide in the exit air stream is also likely to form a useful tool for
measurement of reactor performance. It is current operator practice to aim at main-
taining a dissolved oxygen concentration of about 2 ppm and adjust the volume of
air supplied accordingly. As the provision of air is a major operating cost, it is
sensible to examine critically its most effective utilization to ensure that the best
available use is made of the blower capacity. When considering complete oxida-
tion of the material, the secondary reactors in which oxidation is nearing comple-
tion may receive too much air and the supply may be reduced. Differences in oxy-
gen uptake rates between primary and secondary reactors, due to variations in
reaction rates have been noted by other workers. ls
The temperature of the reactors is controlled by varying the pumping rate of
the return cooling water according to the set point of the temperature of the cool-
ing water leaving the reactors. A small forced draught cooling water tower is ad-
equate in cooling the water before its return. As the temperature fluctuations dur-
ing operation of the reactors are reasonably small and occur fairly slowly, the system
responds well and is effective to within an accuracy of one degree centigrade or
more. This accuracy applies even when a new reactor is being brought on line.
The measurement and control of pH is conducted manually by addition of sul-
furic acid or calcrete or lime by the plant operator. For high acid-consuming feed
material such as carbonates, it would be expected that pH adjustment and control
would be required in the slurry stock tank prior to feeding to the dilution tank and
primary bacterial oxidation reactors. Conversely, for materials in which a high
acid production beyond the pH tolerance of bacteria is anticipated, neutralization
may be required to maintain high reaction rates. The moderate thermophiles have
demonstrated a robustness in operation at low pH values of between 0.5 and 0.8,
although the final pH value of slurry in the Youanmi operation is usually between
1.2 and 1.6.
Adequate pH control by neutralization with alkali reagents can be a difficulty.

This is due to the problem of uniformly adding lime and/or calcrete to the reactors
and preventing concentrated areas of high alkalinity while mixing occurs. A fur-
ther consideration is ferric iron precipitation as a result of neutralization and the
formation of quantities of inert gypsum which may impact on the operating pulp
density of the system. Precipitation of ferric iron onto unreacted material, thereby
preventing leaching. Another possibility is no observations that ferric precipita-
tion inhibits leaching have been made, however. As indicated above, control of pH
is not considered to be a critical process parameter at the Youanmi site due to the
good tolerance by the culture fluctuations in pH value.
Samples are taken from all of the reactors at Youanmi together with the feed on
a twice daily basis. The dissolved metal species are quantified by atomic absorp-
tion spectrophotometry and titration. One of the most valued indicators of perfor-
mance is the amount of ferrous iron present in solution. Higher concentrations of
ferrous iron are encountered in the primary reactors with a decrease in level oc-
curring through the reactor stages. The quantities experienced under normal op-
erating conditions are a balance of the quantity of acid-labile iron species in the
mineral, combined with the reduction of ferric iron lixiviant during arsenopyrite
oxidation. This is balanced against the bacterial ability to convert ferrous to ferric
iron. In theory, an increase in solution ferrous iron concentration occurring under
steady state conditions suggests a decrease in bacterial activity. This may be in-
dicative of a contaminant entering the system, a lower air supply to the reactors or
a rise in temperature above the optimum value, due to cooling water failure. No
major failure has occurred in the process at Youanmi to allow these aspects to be
studied in detail with the exception of a short period of contamination with thio-
cyanate which was readily rectified (see section on troubleshooting). In practice,
changes in ferrous iron concentration are often a reflection on a change in feed
rate to the plant, a change in the mineralization of the feed or a change in feed
particle size. Changes in the other dissolved metal species reporting into solution
may also occur simultaneously with the changes in ferrous iron concentration;
this provides supportive evidence for some variation in feed exerting an influence
on the equilibrium of the system. Changes in the appearance of the color of the
pulp and the type of foam present on the surface of the pulp can all be indicative of
changes in performance. Sometimes a decrease in cooling water demand may be
noticeable due to a lower reaction rate. Other process variables which are logged
and controlled are the concentrate feed and dilution control in the stock tank to-
gether with nutrient addition rates. Downstream monitoring and control includes
counter-current decantation discharge rates and the rate neutralization agent
addition.
It is important to note that when using plant measurements as a tool to assess
performance, the response time for the measurement of parameters will vary. This
is particularly important when monitoring the alterations which occur when a
change is imposed on the system, such as an increase in feed rate. In order of re-
sponse, the parameters of measurement can be noted as follows:
Parameter Response Time

Changes in foam/froth appearance and quantity Within minutes
Dissolved oxygen Within minutes
Ferrous iron concentration Within 1 day
Change in cooling requirement Within 1 day
Change in iron arsenic solution values Within 1 day
Change in pH value Within a few days
Bacterial counts Within a few days
This clearly illustrates the importance of considering dissolved oxygen mea-

surement or methods of measuring oxygen uptake rates on line as the first process
variable to signal a change in the system. It is currently the most effective method
available for directly measuring bacterial oxidation activity with an increase in
dissolved oxygen value or decrease in uptake rate suggesting a decrease in reac-
tion rate. The development of monitoring and control systems for a bacterial oxi-
dation plant is still an area for research and development. More effective systems
will allow bacterial oxidation plants to be more cost-effective and minimize the
response time needed to take remedial action to avoid lengthy downtimes if prob-
lems are not resolved quickly enough.
Integration of Bacterial Oxidation with Other Unit Operations

In common with the introduction of any new technology, bacterial leaching
processes have often been developed as a separate unit operation isolated from the
upstream and downstream process areas and other site activities. This may lead to
difficulties in obtaining the 'best fit' for bacterial leaching into a process in which
the upstream and downstream process areas have been decided on previously and
committed in detail. This means that the bacterial leaching process can make few
demands on the most appropriate range of feeds which should be produced. The
best possible integration of bacterial leaching into the flowsheet must be consid-
ered at the earliest possible stage to enhance the viability of future bacterial leach-
ing projects. This has the potential for offering a wider scope of advantages for
using bacterial leaching than is currently seen, giving increased momentum to its
commercial acceptance. This is particularly true for base metal projects in which
very careful consideration has to be given to iron rejection and downstream se-
quential metal recovery. On some projects, consideration is being given to using
excess acid produced during bacterial oxidation for heap leach projects established
at the same site. The acid balance, with respect to consumption by gangue materi-
als and production by sulfide oxidation, together with the water balance are fun-
damental criteria for the development a successful flowsheet. These issues in up-
stream and downstream processing are discussed briefly below so as to highlight
some of the more important considerations.
Upstream Process Considerations

It is important to have a detailed estimate of the ore grades anticipated over the
life of a project, particularly the oxide and sulfide grade values together with re-
spective metal contents. Testing ore variability in laboratory studies is essential at
an early stage not only to characterize the behavior of the dissolution of metal
species from different sections of the ore body, but also to understand the behav-
ior of gangue minerals in terms of: acid consumption, organic carbon content in
the preg robbing phenomenon of gold, sliming and settling qualities and also, pos-
sible toxicity problems. This may also include potential contaminants in the source
of calcrete needed for neutralization.
Many mining operations will not only produce different ore types and grades,
but the quantity will also be variable. Unlike all of the other unit operations, bacte-
rial oxidation cannot be shut down and started up on an intermittent basis to cope
with these variations. In most operations it is usual to stockpile ore, but the size of
the stockpile may have to be increased when bacterial oxidation is to be used. It is
less common to stockpile concentrate due to greater storage difficulties of pulp or
the need for reclamation and repulping of a stored concentrate product. It is not
possible to make up lost production in the event of a restriction in the feed to
bacterial oxidation merely by increasing the tonnage throughput when more feed
becomes available. This type of action will actually result in losses in production
due to the shock load having an adverse effect on the bacterial process and is likely
to take a significant time to revert back to a normal operation.
Much emphasis has been placed on the need for a fine particle size feed to
bacterial oxidation and at Youanmi a regrind mill is available for grinding oversize
material from flotation before being used as feed to bacterial oxidation. However,
the regrind mill has never been used and the feed produced from milling and flo-
tation averages a PSo value of 75 J1m with a significant coarse fraction component.
If there was a requirement to oxidize a greater amount of pyrite as well as arse-
nopyrite, then facilities for a regrind may be more appropriate. For even more
refractory materials such as chalcopyrite associated with base metal leaching, se-
rious consideration is being given to the use of ultrafine grinding to produce PSo
values of 15 J1m or less. The benefits in metal release rates, reduced residence times
and a higher total recovery, often far exceed the cost of fine grinding.
Flexibility in the bacterial oxidation reactor usage should also be considered at
the design stage to enable reactors to be taken out of operation during long-term
reductions in sulfide material production. This is a further reason for having uni-
formity in reactor design such that a reactor is not necessarily dedicated to being
used as a primary or secondary reactor, but circuit configurations of reactors can
be altered to match long-term variations in feed. The production of concentrate by
flotation is a favored route in which either a rougher or cleaner concentrate is made.
A potential advantage of bacterial oxidation which is gaining prominence is the
ability of bacterial oxidation to treat a lower grade concentrate than competing
processes. This often enables an increase in recovery of a few percent to be made
during concentrate production. The possible toxicity of the flotation reagent suite
used needs consideration due to reagent carryover to bacterial oxidation. These
have so far not presented a problem but laboratory testing is required to gain a
firm understanding of the process sensitivity to all upstream reagents.
Downstream Processing
Solid liquid separation and washing by counter-current decantation is tradi-
tionally the first step after bacterial oxidation. However, the possible benefits of
recycling portions of either solid or liquid, or both as a slurry, to the primary reac-
tors of bacterial oxidation or to the feed stock tank is being assessed. This may be
for the recycling of unreacted coarse sulfide material to oxidize it further, in pref-
erence to extending the bacterial oxidation reactor residence time or in preference
to regrinding any oversize fraction of the original feed. A reason for recycling final
liquor on its own may be to provide a source of acidic ferric iron for neutralizing
the acid-consuming components of fresh sulfide feed entering the primary reac-
tors. This latter scheme would also reduce the downstream neutralization require-
ments of the final liquor needed for iron rejection. Recycling portions of the final
product is also a method of biomass recycle for improving the stability of the bio-
mass population. The overall impact of recycling in bacterial leaching systems may
be to reduce residence time requirements together with reagent costs while im-
proving the stability of the leaching system. It is possible that such schemes will
become an integrated feature of bacterial oxidation flowsheets, thereby improving
the process effectiveness in competing with other oxidation technologies.
Filtration of feed and residues is avoided where possible, due to its high cost
and greater mechanical complexity, so it is usual to consider handling streams of
thickener underflow at 50% w/w. The CCD (Counter Current Decantation)
underflow containing the gold values is carried forward to neutralization and
cyanidation. The detrimental effects of bacterial debris are often noticeable in the
cyanidation process. This is believed to be due to cell lysis, resulting in a stable
mousse-type foam sometimes giving difficulties in effective cyanidation. Antifoam
agents can be useful in disrupting the foam, while residual sulfides in the residue
can also give high cyanide consumption and a slight decrease in gold recovery
which a passivating agent such as lead nitrate may help to alleviate. For the
bioleaching of base metals in which solvent extraction is to be used for the recov-
ery of the metals, consideration has to be given to the potential for bacterial debris
resulting in crud formation.
The acidic ferric waste liquor from bacterial oxidation in refractory gold pro-
cessing is neutralized in two stages by the addition of calcrete and lime or in single
stage using calcrete alone. A stable precipitate is formed over a 6-hour period in
which the pH is adjusted to a value of between 5 and 6. If arsenic is present then a
stable precipitate is formed if the ferric iron to arsenic ratio exceeds 3 in the solu-
tion and often a ratio just greater than 2 is sufficient. The majority of projects can
usually meet the required iron to arsenic ratio with ease. Very stable residue prod-
ucts are produced from the bacterial process, giving substantial environmental
benefits when compared to competing technologies. This is a major benefit to us-
ing a bacterial leaching process.
Trouble Shooting and Philosophy of Operation

The operation of bacterial oxidation plant requires a different philosophy when
compared to the traditional physical and chemical operations of beneficiation and
extraction. Reference has already been made to the need for small incremental
adjustments over relatively long time periods when efforts are being made to in-
crease process efficiency. If the process deviates from the operating norm then it
must be bought back in small incremental changes. Shock loads to the system are
not well tolerated but over a period of time the process can adapt to a remarkably
wide range of changes in operating conditions over the life of a project. The com-
mercial system using moderate thermophiles at Youanmi has demonstrated a re-
markable robustness and an inertia which absorbs perturbations in feed condi-
tions which would not be tolerated at laboratory scale. A major problem area
is the potential for introducing contaminants into the system through the water
makeup. Every effort should be made at the design stage to ensure that a reason-
able water quality is supplied to bacterial oxidation. For the moderate thermo-
philes TDS values of about 8,000 mglL can be tolerated which is possibly higher
than other bacterial oxidation plants. Any recycle lines must be critically evalu-
ated for potential increases in elements which would exceed toxicity thresholds
over time and would remain unnoticed if a once-through system were employed.
The inability to secure a good supply of reasonable quality water should not limit
the use of bacterial processes for the majority of projects.
For the treatment of refractory gold materials, a high potential exists for con-
tamination with cyanide waters and the need for separate process water circuits is
critical. Contamination of the bacterial oxidation circuit with thiocyanate and loss
of bacterial activity represents the only major crisis which has ever occurred at the
Youanmi operation. This was the result of an operator error in which a connection
to a hosing point outside the bacterial oxidation circuit was made for cleaning the
reactors, thereby introducing water containing thiocyanate into the reactor area.
The matter was quickly rectified with a process downtime limited to a few days. A
similar problem of contamination by thiocyanates has been reported during the
commissioning of another bacterial oxidation plant.15
Another potential problem is foaming at the top of the reactors which can oc-
cupy valuable reactor volume. This can require the reactor design to incorporate
significant free board to accommodate the foam. The type of foam experienced
also changes with the degree of oxidation, suggesting that more than one mecha-
nism of foam formation may be responsible. Initially the foam resembles that ex-
perienced in flotation production and is more likely to be due to the carryover of
flotation reagents combined with the natural flotation of sulfide minerals. The
coarse foam bubble size changes to a much finer size as the pulp progresses through
the bacterial oxidation system and this suggest that the more stable fine bubble
foam is associated with microbial activity, possibly a result of extracellular prod-
ucts. This type of foam is a well recognized phenomenon in fermentation indus-
tries. Possible methods for resolving this problem include mechanical foam break-
ers or draught tubes fitted around agitator shafts which draw the foam back into
the pulp.
Future Applications and Conclusions

There is no reason to doubt that the number of commercial bacterial leaching
plants will increase substantially in the future. For many mining ventures the
need to exploit sulfide reserves which are often complex, combined with a smaller
choice of process routes, provides the focus to examine the option of using bacte-
rial leaching.
This chapter has cited a number of potential advantages for considering the
use of moderate thermophiles in bacterial leaching, as opposed to more traditional
lower temperature organisms. It is believed that the use of moderate thermophiles
in leaching will increase not only for the treatment of refractory gold materials,
but will also expand to encompass the leaching of base metals from complex sul-
fides. This is particularly with respect to the leaching of copper from chalcopyrite
and also the leaching of cobalt and nickel from complex polymetallic sulfides. The
potential advantages which the use of thermophilic organisms offer is likely to
lead to an acceleration in the development of processes which would be able to
operate at even higher temperatures, than the current use of moderate thermo-
philes operating at 50°C. Such processes would use both thermophiles and extreme
thermophiles and this will lead to an emphasis on the development of better reac-
tor systems which create environments more conducive to bacterial activity in terms
of the provision of oxygen and nutrients at higher temperatures in low shear
environments.
In addition to pursuing the more obvious development areas for improving the
acceptance of bacterial leaching processes, there is a constant need to address the
abilities and limitations of downstream operations. It will be essential to match
the effort in the development of bacterial leaching systems with an equal amount
of effort to understand current developments in downstream processing for the
selective removal of metal values from solution. The expansion of bacterial leach-
ing as a commercial process is likely to be dependent upon its skillful integration
with other nonbiological unit operations, which will require a higher level of inno-
vation in flowsheet development than is currently tolerated.
During the preparation of this chapter, it was evident that little published in-
formation exists to give a practical guide to either the design and operation of
bacterial leaching plant or its integration into upstream/downstream processing.
It not possible to aim to redress this balance with a single chapter such as this, but
it is hoped that the discussions have highlighted some of the principle issues. In
order for this technology to become more widely accepted, it is essential that the
existing base of knowledge be used wisely to promote the commercial growth of
bacterial leaching processes. The challenges presented for using thermophilic or-
ganisms are by no means insurmountable and will undoubtedly provide a rich
area for future research development and innovation.
References
1. Brierley CL. Practical role of thermophilic bacteria in bioleaching and bioxidation.
Biomine 93, Applications of Biotechnology to the Minerals Industry, International
Conference and Workshop Adelaide, South Australia, March 22-23, 1993. Glenside
South Australia, Australian Mineral Foundation.
2. Norris PRo Acidophilic bacteria and their activity in mineral sulfide oxidation.
In: Ehrlich HL, Brierley CL, eds. Microbial Mineral Recovery. New York: McGraw
Hill, 1990:3-27.
3. Shield JW, Crowell RM. Heap bioxidation of sulfidic gold concentrates. Randol
96. Squaw Creek, California. April 21-24 1996.
4. Budden JR, Bunyard M. Pilot plant testwork and engineering design for the
BacTech bacterial oxidation plant at the Youanmi mine. Biomine 94. Applica-
tions of Biotechnology to the Minerals Industry, International Conference and
Workshop Perth Western Australia, September 19-20, 1994. Glenside, SA Austra-
lian Mineral Foundation.
5. Brierley CL, Winby R. BacTech's thermophilic bioxidation plant at Youanmi mine:
An update on performance and cost. Randol 96. Squaw Creek, California. April
21-24 1996.
6. Hansford GS. Gold biohydrometallurgy: Current design and operation of bio oxi-
dation plants, new research tools and challenges. Proc 1995 Eng Found Conf Min
Process 11 1995, Snowbird, ur, USA.
7. Griffin EA, Luinstra L. Bioreactor scale up, practical considerations for biologi-
cally assisted gold recovery. Biohydrometallurgy 89. Proceedings of International
Symposium. Jackson Hole, Wyoming, August 13-18 1989.
8. Fraser GM. Mixing and oxygen transfer in mineral bioleaching. Biomine 93 Ap-
plications of Biotechnology to the Minerals Industry, International Conference
and Workshop. Perth Western Australia, Glenside, SA Australian Mineral Foun-

dation. September 19-20, 1994.
9. Spencer P, Budden J. Effect of bacterial processes on the heat balance in bacterial
oxidation. Randol Innovations in Gold and Silver Recovery Phase IV Volume 7
1992. Golden, Colorado. August 1, 1992.
10. Budden J, Spencer P. The effect of temperature and water quality on bacterial
oxidation: The advantages of a moderately thermophilic culture over conventional
Thiobacillus cultures. Applications of Biotechnology to the Minerals Industry,
International Conference and Workshop. Perth, Western Australia, Glenside, SA
Australian Mineral Foundation. September 19-20, 1994.
11. Brierley CL, Brans R. Selection of BacTech's thermophilic bioxidation process for
Youanmi mine. Biomine 94. Applications of Biotechnology to the Minerals In-
dustry, International Conference and Workshop. Perth, Western Australia,
Glenside, SA Australian Mineral Foundation. September 19-20, 1994.
12. Nicholson HM et al. Design and commissioning of Ashanti's Sanu biox plant.
Biomine 94. Applications of Biotechnology to the Minerals Industry, International
Conference and Workshop. Perth, Western Australia, Glenside, SA Australian Min-
eral Foundation. September 19-20, 1994.
13. Van Aswegen PC. Commissioning and operation ofbioxidation plants for the treat-
ment of refractory gold ores. Hydrometallurgy fundamentals, technology and in-
novations. P.roceedings of Milton E Wadsworth (IV) International Symposium
Hydrometallurgy, AIME, Salt Lake City Utah 1-5 August 1993.
14. Batty JD. Bioxidation of refractory gold ore concentrates: Don't be misled. Randol
96. Applications of Biotechnology to the Minerals Industry, International Confer-
ence and Workshop. Perth, Western Australia, Glenside, SA Australian Mineral
Foundation. September 19-20, 1994.
15. Van Aswegen PC. Biooxidation of refractory gold ores-the Genmin experience.
Biomine 93. Applications of Biotechnology to the Minerals Industry, International
Conference and Workshop. Perth, Western Australia, Glenside, SA Australian Min-
eral Foundation. September 19-20, 1994.
CHAPTER 5
Heap Leaching of Gold-Bearing

Deposits: Theory
and Operational Description
James A. Brierley
Introduction
B iooxidation pretreatment of refractory precious metal ores and concentrates is

now a recognized, full-scale production process which may serve as an alterna-
tive to roasting and pressure oxidation. The Gencor (BIOXO) biooxidation pro-
cess'-3 for pretreatment of refractory gold-ore concentrates has been commercially
applied at five mine sites: Fairview, South Africa; Sao Bento, Brazil; Harbour Lights,
Australia; Wiluna, Australia; and Ashanti, Ghana (see chapter 3). A second com-
mercial biooxidation process, which employs BacTech technology, using thermo-
philic iron-oxidizing bacteria (see chapter 4), is in operation at the Youanmi Mine,
Australia:M Both the BIOX· and BacTech processes are carried out in highly aer-
ated stirred tank reactors.
The use of a biooxidation heap is an alternative to the use of stirred tank reac-
tors for the biooxidation of an ore. Heap "reactors" are particularly suitable for
pretreating refractory, lower-grade (e.g., 1.0-2.4 g/tonne) whole ores which may
not be amenable to flotation -concentration process treatment. Because of low gold
grades and other mineralogical characteristics, whole ores are usually not ame-
nable to processing in stirred tank reactors. In addition, biooxidation-pretreat-
ment or bioleaching processes using heaps can be an economically favorable alter-
native processes for use in developing countries. 6
E. Livesey-Goldblattl considered an alternative to stirred tank reactors for the
biooxidation pretreatment of refractory gold ore in mill tailings. Testing demon-
strated that slimes, loosened into a granular mass and inoculated with iron- and
sulfide-oxidizing bacteria, could be effectively biooxidized to facilitate subsequent
gold leaching. This pioneering work was extended by Lawson et al, 8 who confirmed
that bacterial cultures which included the bacterium Thiobacillus ferrooxidans of-
fer an alternative to flotation-roasting for gold recovery from slimes dams.

Bioleaching of copper is another process that makes use of the heap "reactor"
concept for mineral biooxidation and bioleaching. Heaps, specifically designed
for copper bioleaching, have been reported and the "bacterial thin layer" heap pro-
cess (see chapter 2) is currently used in Chile for extraction of copper from sulfidic
ores. 9 -11
An early indication of potential for the use of heaps for biooxidation pretreat-
ment was the phenomenon of acid drainage that is associated with sulfidic mine
waste. These acid drainages are attributable to bacterial activity within the waste
rock piles!2 Information gleaned from study of the acid-generating waste rock
systems can provide leads for control and development of an engineered heap in
which bacterial activity is promoted.
In this chapter the recent development and applications of the biooxidation-
heap technology for pretreatment of refractory gold ores is reviewed, the concepts
and theory of the process are considered and commercial application is emphasized.
Laboratory Scale Testing and Development

Conceptual evaluation of the heap as a refractory gold ore pretreatment
bioreactor was first reported in 1990. Column test studies, simulating the
biooxidation of refractory, sulfidic, gold ore in heaps were conducted and reported
by Greene,13 Burbank et al,14 Harrington et al,15 and Brierley and Luinstra.16 These
studies showed that after several months of biooxidation, gold recovery from re-
fractory ores was substantially increased. The effect of particle size on pretreat-
ment time, the extent of sulfide oxidation and gold extraction was also considered
in the preliminary work. Column tests to evaluate particle size effects and other
important process parameters are critical to the development of an economic
bioheap process.
Successful development of bioheaps expands mineable reserves and increases
mine life. This process may also be used to treat materials which might otherwise
be considered waste and could lead to acid rock drainage problems if not pro-
cessed to prevent further sulfide oxidation.
Concepts and Theory

of the Biooxidation-Heap Process
The promising results oflaboratory testing of the biooxidation pretreatment of
ore have lead to critical evaluation of the variables controlling the process and a
consideration of heap design. Mathematical models in conjunction with labora-
tory studies have been used by Genmin in a theoretical evaluation of the bacterial
heap leach process for pretreatment of lower grade refractory ores.17 Genmin's
model predicted the rate and extent of bacterial oxidation of pyritic ores, but de-
termination of maximum extent of heap leaching still requires experimentation
given the unique reactive characteristics of each specific ore. From an economic
perspective, the application of bacterial heap leaching depends on several ore char-
acteristics. Heap biooxidation becomes viable when the flotation recovery of gold
is below 90%. Also, acid-consuming properties of the gangue material are critical.
Occurrence of carbonate in the ore as calcite and or dolomite will consume acid
and prevent the establishment of conditions suitable for biooxidation. If condi-
tions are suitable, bioheap leaching can be effective for pretreatment of smaller
low-grade refractory ore deposits.
Heap Leaching of Gold-Bearing Deposits: Theory and Operational Description 105
Extensive research and modeling of acid-generating waste dumps and

biooxidation pretreatment heaps conducted by the Australian Nuclear Science and
Technology Organization, have led to a definition of important controlling param-
eters for the biooxidation process. Modeling of pyrite oxidation in waste dumps
included the variables of oxygen transport, heat transport, thermal convection,
heap geometry and depletion of oxygen by oxidation ofpyrite. '8 The model pre-
dicted that oxygen penetrates the interior of a heap from the sides as a thin reac-
tion zone, a condition which would limit bacterial activity in the internal portions
of a heap. Within the heap the highest temperatures are localized in this reaction
zone. When the temperature reaches lethal limits for the bacteria (e.g., >40°C for
mesophiles), the biooxidation of sulfides is halted and there is an increase in oxy-
gen concentration which can then penetrate further into the heap.
The waste dump models identified key parameters for the biooxidation heap
pretreatment process. '9 Computer simulations indicated that oxygen transport
through the heap will be governed by diffusion unless the air permeability of the
heap is quite high. Diffusion-driven oxygen transfer would be too low to achieve
sufficient oxidation of the sulfide minerals within a reasonable time for commer-
cial operation. Forcing air in at the base of a biooxidation heap would most likely
be required to achieve sufficiently rapid rates ofbiooxidation. The model also in-
dicated that irrigation of the heap would influence biooxidation. Infiltrating water
can remove heat from the heap-although this may be insufficient to prevent re-
gions of the heap from exceeding temperatures for bacterial survival. Too much
water can retard convective aeration by filling pore spaces.
Modeling the biooxidation heap pretreatment process was performed to evalu-
ate the constraints imposed by the upper temperature lim it'o for growth of
mesophilic iron and mineral oxidizing bacteria. This limit could be higher (> 40°C)
if thermophilic bacteria were present in a heap. The model indicated a sufficient
rate of biooxidation of the sulfide could be achieved with mesophilic bacteria
in conditions of high air permeability with solution infiltration for temperature
control.
The overall oxidation rate in a biooxidation heap is also sensitive to the intrin-
sic oxidation rate of the particular sulfidic material (e.g., pyrite, arsenopyrite,
arsenianpyrite) at the process site. 21 This aspect of the model considers the unique
characteristics of the mineralogy of the sulfidic ore for a given biooxidation pro-
cess. In the modeling process, the intrinsic oxidation rate is coupled with rate of
oxygen transport in a heap to determine the overall oxidation rate. This informa-
tion, which is determined empirically, can be used to design an operation to achieve
a high oxidation rate.
A process of "aeration curing" over a long time period for wetted heaps has
been suggested as a biooxidation pretreatment process!2 The conditions required
for this type of pretreatment process are: the ore must be kept wet but drained; the
ore should be crushed to minus 2 cm to adequately oxidize internal sulfide miner-
als; up to three years may be required; forced air ventilation is likely to be needed
and thorough laboratory testing should precede processing in order to define the
oxidation kinetics and potential extent of metal extraction. ''Aeration curing" could
be a very low-cost method for the pretreatment of refractory ores, but the time
required might discourage operators from implementing this process. However,
the operating costs for a biooxidation heap do not increase proportionately with
pretreatment time and rate. 23
Particle Size
The importance of particle size on the rate ofbiooxidation of sulfidic minerals
was stressed by Bartlett in a comprehensive review of the biooxidation-heap pro-
cess. 24 It was concluded that refractory ore should be crushed to at least -19 mm or
even to -13 mm for effective biooxidation pretreatment. Rate limiting conditions
for the biooxidation heap process also include the heap energy balance and tem-
perature. Mesophilic iron-oxidizing bacteria will be subject to thermal death at
temperatures exceeding 40°-45"<:. Conversely, temperatures below 20"<: will slow
the rate ofbiooxidation (refer to the work of Ahonen and Tuovinen25-28 for the low
temperature effect).
The importance of mineral particle size has also been stressed in a consider-
ation of the use of heaps as bioreactors for the bacterial leaching of metals from
low-grade ores and the biodesulfurization of coal. 29•30 Larger particles will limit
the rate of biooxidation as the sulfide minerals will be present as inaccessible in-
clusions. In contrast, smaller particles limit oxygen and CO2 diffusion into the
heap, thereby limiting bacterial activity. The optimum particle size for a heap will
be effected by mineralogy. Clayey ores should not be crushed to small particles,
whereas hard siliceous ores can withstand crushing to smaller particles without
generation of fines which can plug a heap. A low-cost treatment process such as
that provided by heap reactors is especially important for these low value, low-
grade ores.
The design of a biooxidation pretreatment heap may involve the removal of
problematic fines and clays from the oreY·32 Fines can be segregated from the ore
and the gold-bearing sulfidic fmes concentrated from the rest by conventional flo-
tation or gravity separation. These sulfidic fmes can be processed for gold recov-
ery either in stirred tank reactors or returned to the ore and included in the heap
for biooxidation pretreatment by agglomerating the fines onto larger ore particles.
Residual, nonrefractory gold in the nonsulfide portion of the fines can be recov-
ered by conventional metallurgical procedures. Separation of the problematic fines
from the ore enhances the potential for heap ventilation and reduces solution per-
colation problems.
Microbiology
A comprehensive study of the microbiology of a biooxidation heap pretreat-
ment processes has yet to be reported. The population of acidophilic iron-oxidiz-
ing bacteria present on ore from an experimental biooxidation-heap was estimated
using a most-probable-number (end point dilution) procedure with -10 mesh frac-
tions of ore samples.33 The initial population density, following inoculation, was
determined to be about 5.3 X 105 bacterialg ore. The bacterial population increased
to about 3.5 X 107/g ore by day 30 ofbiooxidation pretreatment. Bacterial numbers
were maintained at a level of about 1.1-1.3 X 107/g over 98 days of pretreatment. The
bacterial population included bacteria of the T. ferrooxidans and Leptospirillum
ferrooxidans type. Unnamed moderately-thermophilic iron-oxidizing bacteria have
also been cultured from biooxidation heap samples.
It is unlikely that bacterial populations in biooxidation heaps for pretreatment
of refractory gold ores would differ greatly from those found in metal bioleaching
heaps and other acid-generating waste rock piles. However, specific characteris-
tics of heaps, such as the amount of sulfide present, could create conditions of high
acidity andlor high temperature which would select for particular groups of min-
Heap Leaching of Go ld-Bea ring Deposits: Theory and Operational Description 107
eral oxidizing bacteria. The pioneering work of Ehrlich34 indicated the diversity of
the micro flora present in the bioleaching of copper from mine waste. The variety
of microorganisms present in the acidic and high metal content environment in-
cluded autotrophic and heterotrophic bacteria, fungi and protozoa. How these di-
verse microorganisms interact and contribute to the biooxidation of the mineral
remains an open question.
The microbial diversity in uranium mine waste heaps has been reported and
thiobacilli dominated the microflora. 3s T. ferrooxidans was consistently detected
in samples, but L. ferrooxidans rarely occurred. The distribution of the T.
ferrooxidans population in the waste heap appeared to be influenced by the avail-
ability of oxygen. Up to 99% of these bacteria were restricted to the top 1.5-2 m of
the heap, presumably a zone with sufficient available oxygen. A study of a copper
mine waste leaching operation gave similar results. 36 The population of acidophilic
iron-oxidizing bacteria was apparently limited to the surface layer of the dump as
determined from two drill holes. Chemical analysis indicated an increasing ratio
of unoxidized ferrous iron to ferric iron with increasing depth in the dump pro-
viding further evidence for restriction of microbial activity to locations near the
surface.
Natural inoculation of a heap from the surface can potentially limit the distri-
bution of bacteria to near-surface locations. A sulfidic ore body was inoculated
with a mixed culture of T. ferrooxidans, T. thiooxidans, and L. ferrooxidans for in
situ stope leaching for extraction of copper and zinC. 37 The ore was inoculated to
contain 107 cells/g. The inoculation was successful as bioleaching became appar-
ent after eight weeks. However, the distribution of bacteria was limited to the solu-
tion flow path in the rubble-ized stope and effective bioleaching was limited to this
flow path.
Applications
The idea of using a biooxidation heap for pretreatment of refractory gold ore
has gone beyond the conceptual and laboratory testing stage. Numerous pilot plant
scale tests have been conducted at mine sites with favorable results and a better
understanding of the operational requirements of the process. The process has
also been tested at a demonstration scale to provide the information required for
commercialization. This section presents a description and the results of scale-up
of the biooxidation heap.
Pilot Scale Testing
Newmont Gold Company

A process for inoculation/agglomeration of refractory ore for biooxidation heap
pretreatment was developed and tested by Newmont Gold Company.38.39 T.
ferrooxidans that has been grown on ferrous iron is added to the crushed ore con-
current with formation of the heap. This process assures distribution of an active
population of bacteria throughout the entire mass of the heap for rapid initiation
of the biooxidation process and facilitates homogenous biooxidation of the sul-
fides. Addition of the acidic bacterial culture to the ore also benefits the process by
facilitating a degree of agglomeration of the ore fine particles. The inoculation/
agglomeration process was first field tested with six pilot-scale biooxidation
heaps.40 .41 The gold grade in the pilot test heaps varied from 1.62 glAu/tonne to
8.62 g Au/tonne. With one exception, 8.62 g Au/tonne, the respective ore grades for
testing were consistent with the range of low-grade ore, 1.0-2.4 g Au/tonne, for
which the biooxidation-heap pretreatment process is considered to offer a practi-
cal alternative to either pressure-oxidation or roasting. In order to evaluate a pos-
sible limiting concentration of oxidizable sulfide, one of the test ores had an un-
usually low sulfide-S content (0.2-0.4%).
A nutrient solution containing ammonium sulfate and potassium phosphate
was applied to each heap with a system of drip-emitter tubing at a rate of about
10-12 L/m2 /h. The effluent from each heap was collected in either tanks or a pond
for recirculation to the heap. Routine monitoring of the progress of the
biooxidation-pretreatment was accomplished by measuring the pH, redox poten-
tial, ferrous-, ferric- and total-iron concentrations in samples of the effluent solu-
tion from each heap. The biooxidation-process increased the total dissolved iron
from the test heaps. Total-iron was almost entirely in the ferric form and the solu-
tion Eh range was +495 to +770 mV (SeE), indicating effective bacterial activity.
Initiation of iron solubilization was rapid, occurring in less than 10-20 days for
most of the test heaps. The short lag for commencement of the biooxidation pro-
cess demonstrated the utility of the inoculation/agglomeration process for dis-
tributing an active culture of the bacteria throughout the heap. The population of
the acidophilic, iron-oxidizing bacteria increased to 3.5-8.7 X 107/g in the -10 mesh
fraction of ore samples after 30 days of biooxidation, indicating suitable condi-
tions within the test heaps for growth of the bacteria.
Low sulfide content (0.2-0.4%) of ore did not prevent biooxidation as indicated
by the solubilization of iron. Even with the low pyrite content, biooxidation re-
sulted in rapid solubilization of iron up to about 4 giL. At this sulfide-mineral con-
cent ration the refractory nature of the ore could be attributed to factors other than
sulfide locking of gold such as the presence preg-robbing carbon which binds the
gold-cyanide complex to the ore, preventing the use of cyanide to leach the gold.
Although the biooxidation-heap process has a low (=0.2-0.4%) sulfide-S concen-
tration requirement for support of bacterial activity, this low sulfide ore should
not contain acid-consuming carbonates unless an inexpensive source of sulfuric
acid was available. Otherwise more acid would be required for biooxidation than
the ore could generate.
Gold extraction results for the biooxidized ores from the respective test heaps
are presented in Table 5.1. 41
Although test heap 1 had a high degree of sulfide oxidation (50-75%), gold ex-
traction with cyanide was low (30%). This heap contained some preg-robbing car-
bonaceous-sulfidic ore in addition to the cyanide-amenable sulfidic ore. However,
the biooxidation-heap process was beneficial for enhancing gold extraction. Test
heap 3 was leached using cyanide because the ore was sulfidic and not preg-rob-
bing. The gold extraction of 50% (a 41% increase above an unoxidized ore sample),
would have been greater with further oxidation of the sulfide. Iron sulfide oxida-
tion was still taking place at the termination of the biooxidation period, indicating
the potential for further sulfide oxidation with time. The preponderance of par-
ticles of a small size and clays in this heap affected the solution flow by decreasing
porosity. Biooxidation was deleteriously effected by the poor solution flow and
ventilation of the heap. Subsequent cyanide leaching was most likely also affected
by the hydrological characteristics of this heap due to the small particle size of the
ore. Heap 3 demonstrated the importance of proper heap design, ore particle size
and operation in order to enhance the aeration required to support biooxidation.
Table 5-1. Gold extraction from ore biooxidized using a heap pretreatment
process
Test Untreated Sulfide Leach Extraction

Heap CN Extraction % Oxidation % Process %
14 50-75 cyanide 30
2 nil 35-40 thiourea 18
3 4 45 cyanide 50
4 nil 42 thiosulfate 51
5 41 48 thiosulfate 61
6 nil 47 thiosulfate 65
Test heap 2 was composed entirely of preg-robbing carbonaceous-sulfidic ore

which cannot be leached with cyanide following biooxidation as previously re-
ported.33 Thiourea was tested as a lixiviant for leaching the gold from this heapY
Thiourea leaching was attractive for use with biooxidation pretreatment since thio-
urea functions in acidic conditions thus eliminating the need to neutralize the acidic
ore. However, very low extraction (18%, Table 5.1) was achieved even though labo-
ratory-column tests had provided more encouraging extraction results. Thiourea
leaching yielded poor gold extraction kinetics because of the large particle size of
the ore (29.3% >2.54 cm) and cold temperatures during the leach process. In addi-
tion, thiourea consumption was relatively high and the reagent was found to un-
dergo oxidation with unfavorable end-products, such as elemental sulfur. This made
thiourea leaching uneconomic for use with biooxidation-heap pretreatment.
Given the preg-robbing nature of much of Newmont Gold Company's refrac-
tory ore, an alternative to cyanide and thiourea was required for leaching the
biooxidized ore. Ammonium thiosulfate was found to effectively leach gold from
biooxidized ore with preg-robbing carbon and a process patent was granted. 43 Five-
tonne composite samples from test heaps 5 and 6 were leached using thiosulfate
(Table 5-1). Test heaps 4 and 6 contained actively preg-robbing ores, and test heap
5 contained sulfidic-carbonaceous ore which had low preg-rob activity. Leach re-
sults from ore samples of test heaps 5 and 6 indicated gold extraction of 61 and
65%, respectively, which is sufficiently high for an economically favorable com-
mercial process. The gold extraction of 51% obtained for test heap 4 is also consid-
ered adequate for economic recovery.
Brohm Mining
Biooxidation heap pretreatment was also tested at the pilot scale by Brohm
Mining and Geobiotics (see chapter 6) at the Gilt Edge Mine. 44 The biooxidation
heap contained 4,300 tonnes of ore crushed to 9 mm and stacked to an initial height
of 7.6 m. A T. ferrooxidans culture was added to the ore during the stacking pro-
cess. An agglomeration aid, Nalco 7534 polymer (50-100 glt), was used to amelio-
rate the effect of ore fines. Biooxidation was conducted for a period of 11 months.
The solution application rate to the heap ranged from 0.04-0.16 L/min/m2. During
the biooxidation process, a portion of the circulating solution was treated with
limestone for neutralization to pH of 5.5-6.0. The purpose of neutralization was

reportedly to remove inhibitory components from the circulating solution. How-
ever, the nature of the inhibitory components was not revealed.
Progress and effect ofbiooxidation pretreatment were determined using com-
posite samples collected from the heap by auger-drilling every 30 days. Cyanide
leach of the biooxidized composite samples was conducted in columns and com-
pared with a baseline extraction of 56.2% for an unoxidized control. Biooxidation
improved gold extraction to 71.6%. Sulfide oxidation was 21%, equivalent to a rate
of 0.06% sulfide/day.
The cost of the biooxidation heap process was estimated at $1.12/t ore. The total
cost, including mining, crushing, leaching, biooxidation pretreatment, reclama-
tion and administration was about $7.95/t. The capital cost for the biooxidation
pretreatment facility was estimated at U.S. $11.6 million. These costs indicate the
biooxidation process to be suitable for pretreatment and gold recovery from lower
grade refractory ore deposits.
ZlataMine
Biooxidation pretreatment has also been tested at the pilot scale with sulfidic
refractory ore at the Zlata Mine in Bulgaria.45 The heap contained 1,200 tonnes of
ore stacked to 2 m height. The heap was inoculated with a mixed culture contain-
ing T. ferrooxidans, L. ferrooxidans, T. thiooxidans, T. acidophilus and Acidiphilium
sp. The culture was added to the heap after the ore was placed. Inoculation of the
ore was achieved by percolation of the culture through the heap, a procedure which
was not likely to pose a problem as the heap was relatively shallow. The concentra-
tion of the iron-oxidizing bacteria on the ore was from 10 7 to 10 8 cells/g. Biooxidation
pretreatment was conducted for a period of six months. The pretreatment resulted
in sulfide-sulfur oxidation of 47% and a decrease in the initial sulfide content of
0.80-0.42%. The most rapid oxidation of the sulfide occurred over the first four
months and continued at slower rate to the termination of pretreatment.
Gold leaching was conducted with a solution containing a protein hydrolysate
from Saccharomyces [actis, ammonium thiosulfate and copper sulfate. Following
biooxidation, 68.4% gold extraction was achieved compared to 19.6% for unoxidized
ore. The gold grade was relatively high (3.2 g Au/t) but within a range appropriate
for biooxidation heap pretreatment. This pilot test provided more evidence dem-
onstrating the utility of a heap as a reactor for biooxidation pretreatment.
Plant Operations
Mt. Leyshon Gold Mine

A 500,000 tonne/annum two stage biooxidation-heap plant to treat a super-
gene copper/gold ore was operated at the Mt. Leyshon Gold Mine in Australia. 46
The gold in the ore was not considered refractory, however, the presence of copper
caused an uneconomically high cyanide consumption during gold recovery. In this
case, the biooxidation-heap was used as a pretreatment for copper leaching in or-
der to remove the cyanocidal copper prior to gold leaching with cyanide. Success-
ful, economic recovery of gold was achieved following the bioleaching of copper.
This process was anticipated to be in operation during 1995 when the supergene
ore supply became depleted.
Fig. 5.1. Aerial view of the Newmont Gold Company biooxidation pretreatment
heap. The dark ore is carbonaceous-sulfidic refractory and the gray ore is sulfidic
refractory. Photo by B. Staver, courtesy of Newmont Gold Company.
Newmont Gold Company

A biooxidation heap pretreatment demonstration plant has been constructed
at Newmont Gold Company's Gold Quarry mine in Carlin, Nevada, USA.4' The
facility was designed to process 900,000 tonnes oflow-grade refractory gold-bear-
ing ore per annum. The biooxidation heap pad is oblong in shape with the long
axis approximately 457 m; the width varies but is typically 152 m, with a heap height
of 8.5-10.7 m (Fig. 5.1). Ore is placed by a series of ten 0.9 by 32 m-Iong field "grass-
hopper" conveyors constructed to withstand the acidic conditions of service. The
conveyors were provided with piping and sprays to both inoculate and acidify the
ore depending on the initial carbonate concentration of the ore. The system can
add about 378.5 L/min of acidified water and 757 L/min of bacterial culture which
is equal to between 4% and 6% surface moisture or about 27.5-34.3 L/metric ton.
The conveyor system is capable of transporting 1089 tonnes of inoculated ore per
hour. A radial stacker with a stinger is used for placement of ore on the heap.
Tanks were used for preparation of the mixed T. ferrooxidans and L. ferrooxidans
ore inoculum. A small mix tank was used for preparation of the ferrous sulfate
nutrient solution which contained 161 mM ferrous iron. Sulfuric acid was used to
adjust the solution to pH 1.8. The original design criteria specified a production
rate of 254 L/min of bacterial inoculum, using two trains of three insulated tanks
(6.4 m diameter, 7.3 m height, 189 m 3 capacity) with a three-day retention (Fig.
5.2). Current experience indicates that the flow rate could be increased to at least
double the design flow rate. Low pressure air was injected into the six tanks at a
rate of 84.6 Llsec at 82.7 kPa. The bacterial inoculum was stored for use in two
ponds of 946 and 2422 m 3 capacity. A pump with a capability to deliver 2271 L/min
Fig. 5.2. Tank system for production of T. ferrooxidans culture for inoculation/agglomera-
tion of ore for biooxidation pretreatment. Photo by T. Logan, courtesy of Newmont Gold
Company.
was used to provide the culture to the inoculation/agglomeration system and the
heap. The biooxidation activity in the heap was monitored using thermistors and
gas samplers. The O2 and CO 2 concentrations were determined for the gas phase
within the heap. Microbiological sampling was performed periodically.
The biooxidation-heap pretreatment process, incorporating inoculation with
an iron-grown bacterial culture of T. ferrooxidans and L. ferrooxidans as the crushed
ore is stacked on a pad, has resulted in the rapid biooxidation of the sulfide miner-
als occluding the gold. Ores need not remain on the pads for long time periods
(perhaps one year) to accomplish acceptable levels of sulfide oxidation and subse-
quent gold recovery. As a result, much lower grade ores can be processed than
when using alternate processes. Sulfidic material with as little as 1.0 g Au/tonne
can be profitably biooxidized and leached with 60-7°% recovery of gold. Cost of
the biooxidation heap pretreatment and gold extraction is in the range of
U.S.$4-6/ton ore processed. This technology allowed Newmont Gold Company to
boost reserves by 37 X 10 6 g gold by being able to process low-grade ore which
would otherwise have been considered as waste.
Summary
The concept of biooxidation for the pretreatment of sulfidic refractory ore in
heaps is rapidly developing to a commercial-scale process. Through inoculation/
agglomeration of ore with the acidophilic iron-oxidizing bacteria it is possible to
initiate biooxidation of sulfide ores with minimum lag time. The biooxidation heaps
require some control for insuring adequate aeration (to optimize bacterial activ-
ity) and solution application (for possible temperature control and maintaining
conditions for bacterial growth). The biooxidation pretreatment will most likely
require a period of 270-360 days. Biooxidation pretreatment can be followed with
conventional metallurgical gold extraction by cyanide or the newly developed
ammonium thiosulfate leach for the biooxidized refractory preg-robbing carbon-
aceous sulfidic ores. A comprehensive study of the microbial populations active in
the biooxidation heap process is needed. We do not yet have a thorough under-
standing of the relationships, activities and responses of the diverse microflora
within a biooxidation heap in order to further development and a better under-
standing of the bioreactor-heap process.
References
1. van Aswegen PC. Bio-oxidation of refractory gold ores, the Genmin experience.
In: Biomine '93. Glenside: Australian Mineral Foundation, Inc., 1993: chapter 15.
2. van Aswegen PC. Commissioning and operation of bio-oxidation plants for the
treatment of refractory gold ores. In: Hiskey JB, Warren GW, eds. Hydrometal-
lurgical Fundamentals, Technology and Innovations. Littleton: Society for Min-
ing, Metallurgy and Exploration, Inc., 1993:709-725.
3. Odd PAR, Craven B., Irvine W. Bioleaching: a feasible process for Wiluna refrac-
tory gold ores. In: Biomine '93. Glenside: Australian Mineral Foundation, Inc.,
1993: chapter 18.
4. Budden JR, Bunyard MJ. Pilot plant testwork and engineering design for the
BacTech bacterial oxidation plant at the Youanmi gold mine. In: Biomine '94.
Glenside: Australian Mineral Foundation, Inc., 1993: chapter 4.
5. Brierley CL, Brans R. Selection of BacTech's thermophilic bio-oxidation process
for Youanmi mine. In: Biomine '94. Glenside: Australian Mineral Foundation, Inc.,
1993: chapter 5.
6. Acevado F, Gentina JC. Bioleaching of minerals-a valid alternative for develop-
ing countries. J Biotechnol1993; 31:115-123.
7. Livesey-Goldblatt E. Bacterial leaching of gold, uranium, pyrite bearing compacted
mine tailing slime. In: Lawrence RW, Branion RMR, Ebner HG, eds. Fundamen-
tal and Applied Biohydrometallurgy. New York: Elsevier, 1986:89-96.
8. Lawson EN, Taylor JL, Hulse GA. Biological pre-treatment for the recovery of
gold from slime dams. Journal of South African Institute of Mining and Metal-
lurgy, 1990; 90:45-49.
9. Montealegre R, Bustos S. Industrial application of the bacterial thin layer process
(BTL). In: Badilla R, Vargas T, Herrera L, eds. Bioleaching: From Molecular Biol-
ogy to Industrial Applications. Santiago: University of Santiago, 1990:95-106.
10. Bustos S, Castro S, Montealegre R. The Sociedad Mineral Pudahuel bacterial thin-
layer leaching process at Lo Aguirre. FEMS Microbiol Rev 11; 1993:231-236.
11. Montealegre R, Bustos S, Rojas et al. Application of the bacterial thin layer leach-
ing process to Quebrada Blanca ores. In: Torma AE, Wey JE, Lakshmanan VI,
eds. Biohydrometallurgical Technologies, Bioleaching Processes, vol. 1. Warrendale
PA: The Minerals, Metals & Materials Society, 1993:1-14.
12. Harries JR, Ritchie AIM. The microenvironment within waste rock dumps under-
going pyrite oxidation. In: Rossi G, Torma AE, eds. Recent Progress in
Biohydrometallurgy. Sarda:Association Mineralogy, 1983:377-392.
13. Greene JW. 1990. Microbial column leaching of a refractory, carbonaceous gold
ore. In: Randol Gold Forum '90. Golden: Randol International, 1990:89-92.
14. Burbank A, Choi N, Prisbrey K. Biooxidation of refractory gold ores in heaps. In:
Fuerstenau MC, Hendrix JL, eds. Advances in Gold and Silver Processing, Littleton:
Society for Mining, Metallurgy and Exploration, Inc., 1990:151-159.
15. Harrington JG, Bartlett RW, Prisbrey K. Kinetics of biooxidation of coarse refrac-
tory gold ores. In: Hiskey JB, Warren GW, eds. Hydrometallurgical Fundamen-
tals, Technology and Innovations. Littleton: Society for Mining, Metallurgy and
Exploration, Inc., 1993:691-708.
16. Brierley JA, Luinstra 1. Biooxidation-heap concept for pretreatment of refractory
gold ore. In: Torma AE, Wey JE, Lakshmanan VI, eds. Biohydrometallurgical Tech-
nologies, Bioleaching Processes, vol. 1. Warrendale PA: The Minerals, Metals &
Materials Society, 1993:437-448.
17. Holtum DA, Murray DM. Bacterial heap leaching of refractory gold/sulfide ores.
Min Eng 1994; 7:619-631.
18. Bennett JW, Harries JR, Pantelis G et al. Limitations on pyrite oxidation rates in
dumps set by air transport mechanisms. In: Salley J, McCready RGL, Wichlacz
PL, eds. Biohydrometallurgy. Canada Centre for Mineral and Energy Technology,
1989:551-561.
19. Ritchie AIM, Pantelis G. Optimization of oxidation rates in dump oxidation of
pyrite gold ores. In: Biomine '93. Glenside: Australian Mineral Foundation, Inc.,
1993: chapter 9.
20. Pantelis G, Ritchie AIM. Rate controls on the oxidation of bacterially catalyzed
process. FEMS Microbiol Rev 11; 1993:183-189.
21. Ritchie AIM. Optimizing the performance of a biooxidation heap: the importance
of the intrinsic oxidation rate. In: Biomine '94. Glenside: Australian Mineral Foun-
dation, Inc., 1993: chapter 15.
22. Bartlett RW. Aeration pretreatment of low-grade refractory gold ores. Minerals &
Metall Proc 7; 1990:22-29.
23. Bartlett RW. Optimum duration for bioheap pretreatment of refractory sulfidic
gold ores. Proceedings of the IMPC San Francisco, 1995:197-201.
24. Bartlett RW. Biooxidation heap pretreatment of sulfide refractory gold ore. Min-
eral Processing and Metallurgy Review (In press).
25. Ahonen L, Tuovinen OH. Microbiological oxidation of ferrous iron at low tem-
peratures. Appl Environ Microbiol1989; 55:312-316.
26. Ahonen L, Tuovinen OH. Kinetics of sulfur oxidation at suboptimal temperatures.
Appl Environ Microbioll990; 56:560-562.
27. Ahonen L, Tuovinen OH. Temperature effects on bacterial leaching of sulfide
minerals in shake flak experiments. Appl Environ Microbiol1991; 57:138-145.
28. Ahonen L, Tuovinen OH. Bacterial oxidation of sulfide minerals in column leach-
ing experiments at suboptimal temperatures. Appl Environ Microbiol 1992;
58:600-606.
29. Andrews G. Large-scale bioprocessing of solids. Biotechnol Prog 1990; 6:225-230.
30. Andrews GF, Stevens q, Leeper SA. Heaps as bioreactors for coal bioprocessing.
Appl Biochem Biotech 1993; 39/40:427-443.
31. Kohr WJ. Method for rendering refractory sulfide ores more susceptible to
biooxidation. US Patent 5,431,717, 1995.
32. Kohr WJ, Johansson CE, Shield JW. Improved sulfide gold ore biooxidation. In:
McClelland GE, Scheiner BJ, Muhtadi 0 et aI, eds. Practical Aspects of Interna-
tional Management and Processing. Littleton: Society for Mining, Metallurgy, and
Exploration, Inc., 1996:101-105.
33. Brierley JA, Wan RY, Luinstra 1. Gold recovery from refractory sulfidic-carbon-
aceous ore. Part I: Biooxidation-heap pretreatment. In: Warren GW, ed. EPD
Congress 1995. Warrendale PA: The Minerals, Metals & Materials Society (TMS),
1995=155-163.
34. Ehrlich HL. Microorganisms in acid drainage from a copper mine. J Bacteriol
1964; 86:350-352.
35. Schippers A, Hallmann R, Wentzien, S et al. Microbial diversity in uranium mine
waste heaps. Appl Environ Microbiol 1995; 61:2930-2935.
36. Bhappu R, Johnson P, Brierley J et al. Theoretical and practical studies on dump
leaching. AIME Transactions 1969; 244:307-320.
37. Sand W, Hallmann R, Rohde K et al. Controlled microbiological in situ stope
leaching of a sulfidic ore. Appl Microbiol Biotechnol 40; 1993:421-426.
38. Brierley JA, Hill DL. Biooxidation process for recovery of gold from heaps of
low-grade sulfidic and carbonaceous sulfidic ore materials. US Patent 5,246,486,
1993·
39. Brierley JA, Hill DL. Biooxidation process for recovery of metal values from sul-
fur-containing ore materials. US Patent 5,332,559, 1994.
40. Brierley JA. Biooxidation-heap technology for pretreatment of refractory sulfidic
gold ore. In: Biomine '94. Glenside: Australian Mineral Foundation, Inc., 1994:
chapter 10.
41. Brierley JA, Wan RY, Hill DL et al. Biooxidation-heap pretreatment technology
for processing lower grade refractory gold ores. In: Vargas T, Jerez CA, Wiertz
JV et al, eds. Biohydrometallurgical Processing. Vol I. Santiago: University of Chile,
1995:253-262.
42. Wan RY, Luinstra L, Brierley JA. Gold recovery from refractory sulfidic-carbon-
aceous ore. Part II: Thiourea leaching following biooxidation-heap pretreatment.
Warrendale PA: The Minerals, Metals & Materials Society (TMS), 1995:165-173.
43. Wan RY, LeVier KM, Clayton RB. Hydrometallurgical process for the recovery of
precious metal values from precious metal ores with thiosulfate lixiviant. US Patent
5,345,359, 1994·
44. Shield JW, Whitlock JL, Johansson CE et al. Sulfide biooxidation: pilot heap at
Gilt Edge. Mining Engineering, 1996; March:48-54.
45. Groudev SN, Sasova II, Groudeva VI et al. Pilot scale microbial heap leaching of
gold from a refractory ore at the Zlata Mine, Bulgaria. In: Vargas T, Jerez CA,
Wiertz JV et al, eds. Biohydrometallurgical Processing. Vol I. Santiago: Univer-
sity of Chile, 1995:425-435.
46. Ellis SJ. Bacterial copper heap leach followed by heap leach recovery of gold at
Mt. Leyshon Gold Mine. In: Biomine '94. Glenside: Australian Mineral Founda-
tion, Inc., 1994: chapter 8.
CHAPTER 6
Biooxidation of Refractory
Gold Ores
(The Geobiotics Process)
James 1. Whitlock
Introduction
I nnovative biooxidation technologies may well hold the key to reduced process
costs for mining in the future. Given a relatively stable gold price, mining enti-
ties must strive for continual cost reduction to remain competitive. As free-milling
oxide deposits are depleted on a world-wide scale, the mining companies capable
of efficiently mining and processing both low- and high-grade refractory sulfide
deposits will become industry leaders. This chapter addresses a new process for
the biooxidation of refractory gold ores developed by GeoBiotics, Inc.
At present, three major categories of refractory gold ore processing are being
utilized on a commercial scale. These processes are pressure oxidation or auto-
daving, roasting and biooxidation. Refractory sulfide gold ores contain gold in
association with the sulfides and do not lend themselves to significant gold recov-
ery following cyanidation. In many cases even fine grinding and subsequent
cyanidation will not liberate the gold. Thus, a sulfide oxidizing pretreatment is
required. Pressure oxidation and roasting have gained acceptance as oxidative pro-
cesses in the industry. Both processes involve substantial capital and operating
costs. Therefore these processes are limited to high-grade refractory gold depos-
its. In addition, these deposits must be large tonnage deposits to justify the capital
required for construction of the oxidation plants. In the biooxidation pretreat-
ment process, the oxidation of the sulfides renders the gold particles available to
cyanidation processes, thus greatly improving the gold recovery on refractory ores.
The economics of the biooxidation process over a broad range of sulfide ore types
and grades needs to be demonstrated at full-scale for biooxidation to gain full
fledged acceptance in the mining industry.
Biooxidation of Refractory Gold Ores

Stated simply, biooxidation is the utilization of naturally occurring biological
and chemical processes to oxidize sulfide minerals. These processes occur slowly
in nature as conditions for biological oxidations are rarely optimum. Temperature,

pH, moisture, oxygen availability, nutrient requirements and removal of reaction

end products are normally rate limiting. Application of process engineering to the
general procedure for the biooxidation of refractory sulfide ores has resulted in a
speeding up of the process ores by orders of magnitude."7 Process design is di-
rected at supplying near optimum conditions for the biooxidation process and
overcoming rate limiting conditions found in nature.
Biooxidation offers some important advantages when compared with pressure
oxidation and roasting. Lower capital and operational costs can be designed into
these biological systems and therefore lower grade refractory ores can be treated.
Simple design offers more rapid mobilization once the decision is made to pre-
treat refractory ores. Both pressure oxidation and roasting require long equip-
ment production lead times. Biooxidation operates at near ambient temperatures
which is much lower than the alternate processes. Simplicity of design for
biooxidation also reduces the required skill levels of operators as well as safety
concerns when compared to pressure oxidation and roasting.
Stirred tank biooxidation of refractory gold concentrates or in a few examples,
stirred tank whole ore and whole ore heap biooxidation of refractory gold ores are
the presently available biooxidation technologies (see chapters 3-6). Stirred tank
biooxidation of refractory gold concentrates offers high percentage gold recovery
in a process with a short retention time of less than 4-6 days. Capital costs for the
acid-resistant tankage and mixing mechanisms that are required are significant.
Concentrate formation, agitation, aeration are the major operating costs.
Pulp densities in stirred tank biooxidation of refractory gold concentrates are
normally kept in the range of 10-20%. Agitation and oxygen transfer problems
arise when pulp densities are increased. Stirred tank biooxidation is best suited
for higher grade concentrates of 18 g/t or greater and faster biooxidizing sulfides
such as arsenopyrite. Pyrite and marcasite type refractory sulfide ores require longer
periods of biooxidation and this increased retention time greatly affects the gold
recovery and/or process economics. Whole ore stirred tank biooxidation systems
usually result in very marginal economics even for higher grade ores. When high-
grade ores are biooxidized in tanks, excessive acid consumption often becomes
problematic.
Whole ore biooxidation in a heap configuration has drawn attention for treat-
ment of lower grade refractory sulfide ores although the process may also be ap-
plicable to higher grade ores. Typically, whole ore is crushed and stacked in a heap
configuration, with or without agglomeration, and inoculated with bacteria. Of-
ten the crushing process produces fine material which requires agglomeration or
removal to allow for solution and air flow through the heap, as described by Kohr
et al.B.'o
Agglomeration requires an acid stable agglomeration aid. Conventional cement
agglomeration as used in cyanide heap leaching will not withstand acidic bioheap
solutions. Although capital and operational costs are low with whole ore heap
biooxidation, a period of 9-13 months is required. Following biooxidation pretreat-
ment, gold recoveries utilizing conventional cyanidation processing are in the
50-70% range. Adequate neutralization of the acidic biooxidized ore prior to
cyanidation normally requires restacking the product while mixing with lime or
other neutralizing agents. Useful overviews of these biooxidation processes are
available."7
Biooxidation of Refractory Gold Ores (The Geobiotics Process) 119
The Geobiotics Process

GeoBiotics Inc. has focused its efforts on an innovative and economic process
for the biooxidation of refractory sulfide gold ores. This new process is designed
to incorporate the advantages of the proven biooxidation technologies of stirred
tank systems as well as whole ore heap systems. At the same time this new system
seeks to avoid many of the disadvantages of both processes. The newly developed
process takes advantage of the more rapid biooxidation rates associated with con-
centrates as utilized in stirred tank systems and the subsequent high gold recover-
ies of 80-95%. The GeoBiotics process is economic with lower grade concentrates
and with sulfides such as pyrite which are slower to biooxidize. This new process
includes the biooxidation heap concept with its associated low capital and operat-
ing cost. By combining high gold recoveries in short time periods with the lower
cost of heap biooxidation, the GeoBiotics process promises to improve economics
when compared to either of the two accepted biooxidation technologies.
The GeoBiotics process is based on the coating of refractory sulfide gold con-
centrates onto a support material with subsequent biooxidation pretreatment tak-
ing place in a stacked heap configuration. A period of 30-90 days is normally re-
quired to oxidize the concentrate coating. The extent of biooxidation required is
normally 50-70% to yield gold recoveries of 80-95%. Biooxidation of the refrac-
tory sulfides renders the gold concentrate amenable to efficient cyanide gold dis-
solution and recovery using conventional processes. The biooxidized concentrate
coating may be recovered by washing and then treating by carbon in leach (CIL) or
other means of cyanidation. It is often advantageous to utilize refractory crushed
whole ore as the support rock. The high concentration of ferric iron liberated from
the concentrate coating rapidly oxidizes this type of support rock. The extent of
biooxidation of the support rock is greater per unit time than would occur in a
whole ore heap process. The support rock may be neutralized following
biooxidation pretreatment and conventionally treated with cyanide for gold re-
covery. Alternatively, sizing of the support after each biooxidation cycle removes
ore support rock that has been reduced in size by attrition or gradual degradation.
This thoroughly oxidized material can then be forwarded to cyanide leaching and
gold recovery.
Process Design Overview

The refractory gold concentrate may be formed by flotation or gravity separa-
tion techniques. In some cases even high-grade tailings may be used as the coat-
ing. Experiments indicate that a mesh size of 65-400 is preferred. Concentrate with
gold grades as low as 15 glt may be economically biooxidized. The amount of con-
centrate which can be coated on support rock varies, however, in general it will be
about 20% of the weight of the support rock. The thickness of the concentrate
coating is then about 0.05 inches, as shown in Figures 6.1 and 6.2. The concentrate
coating is applied at high solids density in slurry form via spraying or tumbling.
Success has also been achieved with coating dry concentrate onto wet support
material. The hydrophobic nature of sulfide material creates a thin fllm on the
support rock. Acid-stable binding agents have been tested but are not usually re-
quired. Numerous studies have confirmed the integrity of the concentrate coating
to remain in place on the support material under heap biooxidation conditions.
Recovery of the concentrate postbiooxidation by rinsing on screens or in a trommel
system is almost totally complete.
Suppor' Rock Sulfide Cooling

Vat or Heap 1/2 - , Inch d la . <0 .05 Inch thick
Fig. 6.1. Diagram indicating the coating of support rock with refractory gold concen-
trate.
Fig. 6.2. Support rock coated with concentrate in preparation for biooxidation.
The support material can be almost any acid-stable medium. In most cases it
would be refractory ore crushed to 3/8 x 1 inch. Fine material from crushing can be
removed and treated separately or agglomerated and treated separately. Choosing
the most economical support rock size may be a complex issue. The size of the
support rock affects the surface area per unit volume and if we assume that there
exists an optimum concentrate coating thickness and that it needs to be uniform,
then the size of the support material affects the entire economics of the heap. A
larger size of support rock would decrease the weight ratio of concentrate to sup-
port rock for a given concentrate coating thickness. This would require an in-
crease in the pad size and/or height for a given volume of concentrate to be treated.
The fluid flow rate would increase at a flxed application rate and oxygen and heat
transfer may be improved. A smaller sized support rock would reduce the required
pad size and may be preferred if air and solution flow is adequate.
GeoBiotics'
Biooxidation Process
Hypothetical Flowsheet
..... ~
Cons ~
t Recycled Support Rock
Oversize I
•
Fig. 6.3. Hypothetical flowsheet for a typical GeoBiotics' biooxidation process.
A major design option is the selection of the type of support rock. Chemical
characterization and acid stability are considerations. For example, it may be ad-
vantageous to select a support on the basis of its carbonate or pyrite sulfur content
to facilitate process regulation. If the support rock is refractory sulfide ore it may
be processed for gold recovery after biooxidation. In one scenario, the biooxidized
concentrate coating is collected for processing through rinsing and the support
rock is re-coated and recycled to the pad, as in the flow sheet shown in Figure 6.3.
Bacteria are well established in the retained solutions. This established bacterial
culture shortens the lag time normally found during start-up and shortens the
biooxidation period. It may also be advantageous to recycle the support rock when
it is a refractory sulfide ore. Further oxidation after recycling can render the final
product very low in its potential to produce acid rock drainage and thereby allow
the rock to be disposed of in an environmentally safe manner (after gold recov-
ery). This complete level ofbiooxidation can not be economically achieved in whole
ore biooxidation.
The GeoBiotics process utilizes fermentation to provide a biooxidation inocu-
lum with a very high bacterial cell count. Components in the initial rinses of these
biooxidation pads may frequently be a source of toxicity to the microbes or at least
can be rate limiting. Therefore it is prudent to inoculate after acidification and
rinsing of the pad. Column pilot testing can determine whether toxic or inhibitory
components are present in the initial drain and subsequent rinse waters. This in-
formation is used to determine the percentage of rinse water that must be removed
from the system or neutralized and recycled. The water management plan also
comes into play in determination of these strategies. Addition of nutrients to sat-
isfy the growth requirements of the bacteria are made to the feed solution system.
The potential exists for these additions to be modified to regulate formation of
undesirable chemical products within the biooxidation pad solutions.
GeoBiotics heap biooxidation of concentrate-coated support rock utilizes a
reusable heap-leach type lined containment. Once the liner (typically HPDE)
is correctly sized for the application and in place, a drain and protection layer of
material is added. The characteristics of the material for this drain layer may be
selected to facilitate the biooxidation process. The drain layer can be selected to
provide some neutralization capacity as well as serve as a source of carbon diox-
ide. Low pressure air is forced into the heap using blower( s) and a perforated pipe
grid system. Location of the pipe and perforations is based on the pad height and
configuration to maximize the air utilization. Thermocouples are installed in pilot
columns and on the full-scale pad to provide temperature data and profiles. This
information is utilized in heat regulation in the pad. Solution application rates and
the level of air addition is then controlled to fit this strategy.
In general terms, feed solution at a pH of 1.1-1.9 is applied to the heap through
emitters or wobblers at rates of 0.0015-0.003 gpm/ft2. Monitoring of the feed solu-
tions to the pad and effiuent solutions from the pad is performed on a daily basis.
Parameters of general interest are pH, Eh, acidity, sulfate, temperature, dissolved
oxygen and metals. The composition of the feed solution and the percentage of
effiuent recycled may be adjusted to maintain process specifications. Also, mea-
surements of effiuent solution microbial cell counts and measurement of residual
nutrients are performed periodically.
The extent of biooxidation can be approximated by the levels of solubilized
iron, solubilized sulfur and ratios of solubilized metals such as iron and arsenic.
Also, the pad is relatively easy to core bore and sulfur determinations or gold re-
coveries can be used as measures of the extent of biooxidation. One of the advan-
tages of biooxidation is that only partial oxidation of the sulfide content may be
required to gain very high gold recoveries. When the extent ofbiooxidation reaches
predetermined economic and or strategic levels, the biooxidized material can be
off-loaded from the pad. The biooxidized concentrate is then trommeled and rinsed
from the support rock and collected. Conventional cyanidation of the pretreated
product followed by gold adsorption (e.g., in a CIL circuit) is performed. The sup-
port rock can then be recycled or treated for conventional gold recovery if it is a
biooxidized ore.
All processes for oxidizing sulfide ores produce acidic effiuents and acidic resi-
dues. Gold dissolution lixiviants may be selected for use in alkaline or acidic me-
dia. Regardless of the process chosen, neutralization is likely to come into play at
some point. As neutralization costs can defeat the favorable economics of any oxi-
dation process, a tailored and cost-effective neutralization plan should be prede-
termined.
Laboratory and Pilot Test Results

The development of the GeoBiotics concentrate coated on support rock
biooxidation process has evolved from the results of extensive testing of more than
30 different ore types. A wide variety of different support rock types have been
tested and most performed well. In many cases barren support rock was utilized
so as not to contribute gold, iron or sulfur values to the metallurgical balance.
However, when refractory sulfide ore was utilized as the support rock, the support
was biooxidized at a much faster rate as compared to biooxidation of the same
Fig. 6-4- The 4-foot diameter biooxidation test column.
support rock material in an uncoated whole ore scenario. This rapid biooxidation
of the support rock is believed to be a function of the high ferric iron environment
created by biooxidation of the concentrate coating.
A shake flask system has been developed for determination of biooxidation
potential for refractory gold ores. Concentrates and ores are placed in shake flasks
at approximately 20% pulp densities, nutrients are added and the pH adjusted
prior to inoculation with nonadapted bacteria. Solutions are replaced on a fre-
quently to prevent the accumulation of rate limiting end products. The extent of
biooxidation is determined by the amount of iron solubilized. Previous testing has
shown a strong correlation between solubilized iron and sulfide oxidized in these
flask systems.
More than 30 column biooxidations utilizing columns of 3-6 inch diameter and
up to 9 feet high have been performed as suggested by Shield et al. ll Normal col-
umn biooxidation times are 30-90 days, however, shorter and longer periods have
also been tested. Columns are also utilized to test coated concentrate integrity in
relation to support rock types. Sloughing of the concentrate from the support rock
has been very limited in column and pilot studies. Gold recovery, before and after
biooxidation, is determined by cyanide roll bottle tests.
Two large column tests were performed utilizing a 4-foot diameter by 13-foot
high column. Both tests were completed outdoors, one in summer and one in se-
vere winter conditions on the same concentrate and support rock (Fig. 6.4). Smaller
6-inch diameter by 9-foot high columns were tested in parallel with the larger col-
umns using the same concentrate and support rock under very similar conditions.
The large winter column was insulated and heated solution was recirculated to
mimic the laboratory temperatures. Recovery data showed a very high degree of
correlation between the large and small columns in biooxidation rates and
Fig. 6.5. The 5,000 ton whole ore biooxidation heap built by GeoBiotics in South Da-
kota.
subsequent gold recoveries. The larger column allowed for thermocouple and oxy-
gen measurements that were difficult to gain from small columns. This initial level
of confidence in scale-up was required to evaluate techniques and process devel-
opment from the smaller columns. At present a full-scale, coated concentrate on
support rock, in-heap configuration, test is being evaluated in parallel with a 4-foot
diameter column and 6-inch diameter columns.
GeoBiotics gained experience with heap configurations when a 5,000 ton whole
ore biooxidation test heap (Fig. 6.5) was constructed and evaluated over a one-
year period at the Brohm mine of the Dakota Mining Company in the Black Hills
of South Dakota as reported by Shield et al.12 This test was to evaluate the improve-
ment in recovery by whole ore biooxidation of refractory sulfide ore as well as to
collect design data on full-scale fermentation and the heap biooxidation concept.
The overall gold recovery of the biooxidized whole ore was improved from 55-74%.
A severe winter period was managed during the test and provided valuable infor-
mation on engineering design. Additional air was forced into the heap which was
equipped with a thermocouple monitoring system. Air delivery systems were tested
and evaluated. Monthly core bore samples of the pad were used to evaluate the
extent of biooxidation, gold recovery improvement and to provide data on solu-
tion distribution.
Obviously, each ore tested presents a different set of conditions. In some cases
toxic rinse waters were noted at column test initiation. Some ore types form higher
grade concentrates than others, with corresponding sulfide-sulfur concentrations.
It is well known that different sulfide mineralogical types are biooxidized at different
O:l
Table 6.1. Results of column testing of coated biooxidation technology Ig'
>t
i:i.:
I:>
Ore Sample Concentrate Coating Support Before Extent of Biooxidation After g.
Location Source Grade Ratio Rock Recovery Biooxidation Time Recovery ;:s
~
South America Drill core 1.38 opt Au Cinder ~
5:2 55.2% Au 57% 46 days 98.0% Au
~
Float@GBI ~
Western US Existing pyrite 0.26 opt Au 5:1 Rounded 46% Au 80% 91 days 86% Au "0...
flotation circuit 0.85 0ptAg Gravel 34% Ag 81% Ag ~
C)
South America Leached Mill Tails 0-41 opt Au 4:1 Barren < 20% Au 64% 71 days 77% Au j 0
Float@GBI Rock is:
0
WesternUS Whole ore 1.11 opt Au 5:1 Ore 54.9% Au 44% 68 days 84.3% Au ....
Float@GBI '"
Africa Existing pyrite 1.35 opt Au 7:1 Barren 37.8% Au 97% 57 days 97.9% Au
---'":;i
flotation circuit Rock '"C)
0
'"
Africa Existing pyrite 2.81 opt Au 7:1 Barren 53.6% Au 94% 55 days 94.8% Au ~
flotation circuit Rock §."

Mexico Existing pyrite 0.95 opt Au 7:1 Barren 53% Au 4?'Yo 22 days 94% Au 5'
'1:l
flotation circuit 4.59 0ptAg Rock 42% Ag 72% Ag ....
0
Nevada Gravity concentrate 2.3 opt Au 5:1 Cinder 79.5% Au 82% 47 days 98.4% Au "'"'"
P70 100 mesh ~
Nevada Pilot Scale pyrite 1.04 opt Au 5:1 SAG Mill 51.2% Au 56% 56 days 89.9% Au
flotation circuit reject
Chile Whole ore 0.35 opt Au 5:1 Barren 28.6% Au 80% 60 days 92.4% Au
Float@GBI Rock
Nevada Drill core 1.17 opt Au 5:1 DrillCore 50% Au 60% 57 days 93.4% Au
Float@GBI
* Recovery from refractory flotation concentrate formed from elL tailings 1--
~
Table 6.2. Coated concentrate biooxidation cost estimates
Mill Feed Rate Capital Cost Operating Cost

(st/day) (U.S. $1,000 ) (U.S. $/st mill feed)
2,500 1.58
10,000 0.71
rates, therefore the sulfide mineralogy may have an enormous effect on the rate of
biooxidation. Particle size of the sulfide to be biooxidized also may have a pro-
nounced effect.
Given the complexity of the concentrates used in these evaluations and the dif-
ferences in experimental variables, we can still draw some very important general-
ized conclusions as presented in Table 6.1, and by Shield et al.'3 Biooxidation of
concentrate-coated support rock treated in a heap configuration typically results
in approximately 60% refractory sulfide oxidation in 60 days. Post biooxidation
gold recovery utilizing the GeoBiotics coated concentrate technology averaged 92%
when the extent ofbiooxidation was about 60%. Process strategies can reduce cya-
nide consumption to very favorable levels as well as produce environmentallyac-
ceptable end products.
Costs
The capital costs for equipment associated with the GeoBiotics coated concen-
trate technology are typical of conventional heap leaching utilized in the gold min-
ing industry. These costs are largely the lined pad and collection system, convey-
ance, stacking and loading. As with refractory sulfide ore processing, concentrate
formation and neutralization costs are also similar.
Kvaerner-Davy (San Ramon, CA) was commissioned to generate a prefeasibility
study of the biooxidation portion of the GeoBiotics process as presented in
Table 6.2, and in Kearns et al.'3 This study represents a hypothetical case in a mod-
erate climate with mill and recovery systems in place. Sulfide refractory gold ore
was assumed to be pyritic with a 10:1 ore to concentrate formed ratio. Extent of
biooxidation was projected at 50% with a 60-day biooxidation period. Mill feed
rates for the two conceptual ore feeds were 2,500 st/day and 10,000 st/day. This
would correlate to 250 st/day and 1,000 st/day of refractory gold concentrate.
Conclusions
Refractory sulfide gold ores offer enormous opportunities to mining compa-
nies. This refractory sulfide gold ore resource will become increasingly important
as the world's oxide gold ore reserves are depleted. Oxidation technologies will be
paramount in achieving economic gold recovery from this resource, and commer-
cialized biooxidation technologies will be extremely strong competitors for be-
coming the pretreatment technology of choice. Biooxidation, with lower capital
and operating costs and the ability to operate at low temperatures and pressures,
can offer numerous advantages when compared to pressure oxidation and roasting.
Biooxidation of Refractory Gold Ores (The Geobiotics Process) 12 7
The newly developed GeoBiotics proprietary concentrate-coated support rock

heap biooxidation is believed to combine the best qualities of the two major com-
mercial applications ofbiooxidation of refractory sulfide ores, namely concentrate
stirred tank and whole ore heap leaching. The concentrate-coated technology has
the advantages of rapid biooxidation rates and high gold recoveries of stirred tank
concentrate biooxidation combined with the low costs associated with biooxidation
in a heap configuration. This combination results in greater potential for
biooxidation of refractory sulfide ores where the type of sulfide is more resistant
to biooxidation, such as pyrite.
The GeoBiotics process has been extensively researched, engineered and care-
fully scaled up to its present level of small-scale commercial application. Due to
the high-grade concentrate coating, the pad size required is much smaller and
material retention times are much less than whole ore heap biooxidation. Research
efforts are now focused on a number of improvements which will shrink the pad
size, optimize biooxidation rates, enhance gold recovery and further improve the
overall process economics.
References
1. Brierley JA. Biooxidation heap technology for pretreatment of refractory sulfidic
gold ore. In: Biomine '94. Adelaide, Australia: Australian Mineral Foundation,
1994·
2. Bruynesteyn A. Biological treatment of refractory gold ores. In: Biomine '93.
Adelaide, Australia: Australian Mineral Foundation, 1994.
3. Lawrence RW, Poulin R. Evaluation of the potential for biotechnology in the Ca-
nadian mining industry. CANMET Report 95-029R, 2nd ed. 1996:61.
4. Bartlett RW. Aeration requirements for heap biooxidation of refractory gold ores.
Olympic Valley, CA: Randol Gold Forum, April 1996:273-276.
5. Hansford GS, Bailey AD. Oxygen transfer limitations of biooxidation at high sol-
ids concentrations. Proceedings of International Biohydrometallurgical Sympo-
sium. Jackson WY. August 1993:469-478.
6. Gormely LS, Branion RMR. Engineering design of microbiological leaching reac-
tors. Proceedings ofInternational Biohydrometallurgical Symposium. Jackson WY.
August 1989:499-518.
pyrite gold ores. Proceedings of International Biohydrometallurgical Symposium.
Jackson WY. 1993:731-738.
8. Kohr WJ, Johansson CE, Shield JW. Improved sulfide gold ore biooxidation. In:
McClelland GE, Scheiner BJ, Muhtadi O. Practical aspects of international man-
agement and processing. 1996: 48-54
9. Kohr WJ. Method for rendering refractory sulfide oresmore susceptible to
biooxidation. U.S. Patent No. 5.431717. 1995.
10. Kohr WJ. Method for rendering refractory sulfide ores more susceptible to
biooxidation. U.S. Patent No. 5,573,575. 1996.
11. Shield JW, Crowell RM. Heap biooxidation of sulfidic gold concentrates. Randol
Gold Forum. Olympic Valley, CA. 1996:277-280.
12. Shield JW, Whitlock JL, Johansson CE et al. Sulfide bioxidation pilot heap at Gilt
Edge. Mining Engineering 1996; 48(3): 48-54.
13. Kearns DP, Shield JW. Lowering the threshold for refractory gold deposits with
new biooxidation technologies. Mine-EXPO International '96. Las Vegas NY. 1996.
CHAPTER 7
Technical Potential for Bioleaching

and Biobeneficiation of Ores
to Recover Base Metals (Other than
Iron or Copper), Platinum-Group
Metals and Silver
Henry 1. Ehrlich
Introduction
T he extraction of copper from sulfidic copper ores through the agency of acido-
philic iron-oxidizing bacteria, e.g., Thiobacillus ferrooxidans, has become an
industrially accepted technology in mining and mineral processing, especially for
low-grade ore, and is now also receiving serious consideration for high-grade ores
and ore concentrates (see chapters 2-6 and also Rossi,' Ehrlich and Brierlef>. Other
commercially successful applications of bioleaching have been in the in situ ex-
traction of uranium from some uranium ores, and in the biobeneficiation of pre-
cious metal ores, especially sulfidic gold ores. A potential exists for the microbial
extraction of base metals besides iron and copper from sulfidic and other types of
ores, but none of these processes have so far found commercial application. Among
the reasons for lack of commercial exploitation of these processes have been a
preconception that some may be technically difficult to run on an industrial scale
and/or that their economics are unfavorable. Other reasons include considerations
of limited size of pertinent mineral reserves and future demand for a particular
metal. 3 As stricter environmental laws dealing with control of atmospheric, water,
and soil pollution that results from conventional extraction by pyro- and hydro-
metallurgy are enacted, many processes of base metal extraction by bioleaching
currently viewed as unattractive economically are likely to become competitive
with the conventional processes, especially if the efficiency of these bioleaching
processes can be further improved.
In this chapter, recent advances in our knowledge of bioleaching of a series of
base metals other than copper will be reviewed. These processes include some that
are based on metal extraction by autotrophic bacteria, in particular iron-oxidizing
acidophiles, and others that are based on extraction by heterotrophs, including
aerobic and anaerobic bacteria, and fungi, which are generally aerobes. Leaching
by acidophilic autotrophs has the advantage of requiring supplementation of the
culture medium with only small amounts (concentrations in a gram per liter or
less) of one or a few inexpensive, inorganic nutrients, especially in reactor leach-
ing. In in situ-, heap- or dump-leaching of sulfidic ores by acidophiles, nutrient
supplementation is mostly unnecessary because all needed nutrients are present
in the environment in sufficient quantities.
Heterotrophic leaching, on the other hand, requires the addition of significant
quantities (concentrations in the range of 10-100 giL or more) of an organic en-
ergy/carbon-source and an inorganic and/or organic nitrogen source in the lixiviant
to support the growth and activity of the leaching microorganisms. The energy/
carbon source may be carbohydrate such as sugars or other polysaccharides, pro-
teins, alcohols, organic acids, aliphatic or aromatic hydrocarbons, heterocyclic
compounds, or others. The choice is governed by the ability of an organism to
metabolize a given energy/carbon source, its ready availability and cost, and its
potential for exerting a selective effect for the organism( s) that are the active agents
in a leaching process. Unlike leaching processes with acidophilic autotrophs, in
which the very acidic growth conditions (pH range from <1.5-2.5) generated by the
organisms exert a highly selective effect, leaching with heterotrophs may require
different controls to suppress growth of undesirable organisms. The undesirables,
may consume the energy/carbon source without extraction of base metal from an
ore mineral. One kind of selective control may involve the use of a specialized
energy/carbon source that is unavailable or toxic to undesired organisms but readily
used by the leaching organism(s), e.g., phenol. Another control would be to limit
the access of oxygen if the leaching organisms were microaerophilic or anaerobic.
Temperature and pH controls are two other alternatives for maintaining selective
growth conditions. Elimination of interference by undesirable organisms may also
be achieved with two reactors running in tandem. This approach is applicable when
direct contact between the leaching organism and the ore to be leached is not re-
quired because the leaching organism produces one or more water-soluble com-
pounds when metabolizing the energy/carbon source and these compounds do
the leaching. In a two-reactor system, the organism generates the lixiviant axeni-
cally in one reactor, and the spent growth medium is then allowed to leach the ore
in a second reactor or in a heap or dump without special precautions against con-
tamination. The lixiviant can thus be generated without interference from con-
taminants. Since the spent medium containing the lixiviant may not support growth
of interfering organisms, aseptic conditions become unnecessary for running the
second reactor. For more detailed discussion of this subject, see Ehrlich. 4
Leaching with Autotrophic and Mixotrophic Bacteria

Besides copper and iron, other base metals, among them Co, Ni, Zn, Pb, Mo,
and Ga, may occur in industrially significant quantities in sulfidic ores. Silver and
platinum-group metals (platinum, rhodium, ruthenium, palladium, osmium, and
iridium) may also be associated with sulfidic ores. In some instances, these metals
may be the chief ore constituent, whereas in other instances they occur in
polymetallic ores. Laboratory studies have shown that sulfidic ores containing these
metals can be leached by acidophilic iron bacteria such as Thiobacillus ferrooxidans,
either by an indirect or a direct mechanism, or both. In an indirect mechanism,
the organisms generate an acidic ferric lixiviant which oxidizes the metal sulfide(s),
Biobeneficiation for Base and Platinum Group Metals 131
whereas in a direct mechanism, the organisms oxidize the metal sulfide in the ore
directly while in contact with the mineral surface and interacting enzymatically
with its crystal lattice (for a detailed discussion, see Ehrlich,s,6 Rojas et al7). In
extracting polymetallic ores, galvanic effects can sometimes be exploited for the
preferential leaching of a desired metal constituent. 8
The chemistry underlying the leaching of the base metals considered in this
chapter is the oxidation of the sulfide moiety to sulfate. In chemical oxidation of
sulfide by ferric iron, the sulfide is converted to elemental sulfur (So). This sulfur
may accumulate on the mineral surface and prevent further oxidation by blocking
further access of ferric iron to the sulfide mineral. However, organisms like T.
ferrooxidans and T. thiooxidans readily oxidize So to sulfate and thereby sustain
chemical metal sulfide oxidation. With the exception of PbSO 4 and jarosite, a basic
ferric sulfate, all base metal sulfates are soluble in the acid lixiviant and thus do not
precipitate on the surface of the sulfide mineral from which they derive. The
bioleaching of galena (PbS), requires special conditions to prevent sparingly wa-
ter-soluble PbS0 4 build-up on the galena surface. In the laboratory, continual agi-
tation on a shaker has been found to give satisfactory results. 9
Studies prior to 1989 on the leaching of sulfidic ores containing Co, Ni, Zn, Pb,
Ga, Mo, and Ag were reviewed by Rossi' and Rossi and Ehrlich.'o The following is a
discussion of studies published since 1989.
Cobalt
Several recent studies on Co extraction by acidophiles were chiefly concerned
with testing the amenability of different cobalt-containing raw materials to leach-
ing. Thus, Torma et al" investigated bioextraction of cobalt from cobaltite (CoAsS)
concentrate from the Blackbird Mine in Idaho (USA) by T. ferrooxidans. The chief
mineral constituents of the ore were pyrite, arsenopyrite, cobaltite and silica. The
ore concentrate contained 44.3% total S, 32.7% Fe, 9.9% As, 5.7% Co, and 1.0% Si.
Leaching was performed in batch experiments in 250 mL Erlenmeyer flasks con-
taining iron-free mineral salts solution plus the ore concentrate at a desired pulp
density (PD). Experimental flasks were inoculated with iron-grown T. ferrooxidans.
Ore concentrate at PD of 3-5% gave best results. In 42 days, 98% of the Co was
extracted at 3% PD, and 78% at 5% PD. At higher pulp densities, total Co extrac-
tion in the same length of time was markedly smaller; e.g., at 10% PD, it amounted
to only 13.2%, and at 30% PD to just 4.5% Leached Co was recoverable from the
pregnant solution by solvent extraction with 15% (v/v) cyanex 272 or cyanex 302,
and 85% (v/v) Exxsol D-80 after precipitation of Fe and As as basic ferric jarosite
and ferric arsenate, respectively. Co was stripped from the extractants with dilute
sulfuric acid.
In a comparison of 29 strains of T. ferrooxidans in leaching of Blackbird cobal-
tite ore and concentrate, and synthetic cobalt sulfide, Thompson et all2 found 5
strains to give the best extraction. Optimal leaching without added iron occurred
at an initial pH of 2.0. However, in lixiviant containing ferrous iron, optimal leach-
ing occurred at an initial pH of 1.8. More than 95% of the cobalt in ore concentrate
was extracted in 28 days by wild-type strain Fel in the initial presence of Fez+ of
1500 mg at a PD 1%. Leaching was found to occur by direct and by indirect attack
(with Fe3+) of the ore or its concentrate. Most of the nitrogen and phosphorus
requirements of the bacteria were met by the ore. When oxidation of synthetic CoS
by strain T5 was studied, So% of the cobalt became dissolved in 200 h while only
S% was dissolved in sterile controls. The mechanism of action in this case ap-
peared to be mostly by the direct mode.
Morin et al13,14 performed bench-scale leaching experiments on a cobaltiferous
pyrite concentrate with a mixed culture of T. ferrooxidans, T. thiooxidans and
Leptospirillum ferrooxidans. The cobalt was finely disseminated in the pyrite ma-
trix and amounted to 1.4% of the total mineral. Leaching was performed in a modi-
fied 9K medium. 15 In batch tests in airlift reactors, optimal leaching occurred when
the pH was maintained in a range of 1.1-2.0. A build-up of ferric iron in excess of 35
gIL was found to be inhibitory to pyrite oxidation. Addition of CO 2 at 1% of the air
supply (source of oxygen) stimulated bacterial activity. Optimal Co extraction was
observed at a particle size of <20 JlIl1 and a pulp density of 10%. Up to So% of the
pyrite was oxidized.14>16 When the same ore was leached in a reactor system con-
sisting of a set of four 20-L leach tanks connected in series that were continuously
fed with a medium similar to that in the batch experiments, and in which the spent
medium was recycled, a cobalt concentration of 5 giL in the pregnant solution was
achieved, with So% of the Co extracted from the ore. Prolonged operation of the
reactor system with a gradual increase in pulp density of the ore from 10-20%
permitted sustained Co extraction. The pH during reactor operation was main-
tained between 1.3 and 1.7 with addition of limestone pulp. Not surprisingly in
view of the low pH, the authors reported that L. ferrooxidans displaced T.
ferrooxidans and T. thiooxidans in the second and third tanks of a four-tank series.
L. ferrooxidans replaced the other two organisms essentially completely in the fourth
tank, in which the pyrite oxidation rate was very slow.13
Baldi et al17 reported that small amounts of calcitic gangue (0.01-1.01%) in py-
rite ore can interfere in its oxidation by T. ferrooxidans and in the accompanying
release of cobalt and zinc that may be present in the ore. The reason for the inter-
ference is the neutralization by the calcite of acid required for growth of T.
ferrooxidans. Continual addition of small amounts of acid (5 mM H2 S04) to ore in
percolator setups removed the calcite. T. ferrooxidans began to grow and oxidize
the ore with release of Fe, Co and Zn once the calcite was removed. The lag phase
depended on calcite content of the ore and the time required for its removal. Al-
though leaching rates were greater in percolator setups than in shake flasks, the
duration of the lag phase was shorter in shake flasks than in percolators.
Nakazawa and Sat018 took the unusual approach of using T. ferrooxidans in
leaching Co from cobalt-rich ferromanganese crusts, which have no sulfidic con-
stituents. The leaching of the crusts was accomplished in 9K medium'5 with el-
emental sulfur or pyrite (-100 mesh) replacing ferrous sulfate as the energy source
for the bacteria. The crust contained 20.S% Mn, 13.S% Fe, 0.S2% Co, 0.5% Ni, and
0.12% Cu. Leaching was performed at 25°C in shaken flasks with crustal material
(-100 mesh) at a pulp density of 0.1%. Although acidification with H 2 S0 4 to
pH 1 in the absence of bacteria released nearly all Cu and some of the Ni and Fe in
IS days, no Co or Mn were released under these conditions. With the addition of
1% elemental S to Fe-free 9K medium, the bacteria leached all Cu, nearly all Ni, Co
and Fe, and more than 60% of the Mn in IS days. Raising the sulfur concentration
in the medium to 3% speeded up Co leaching significantly, but the full extent of
leaching remained nearly the same (between So and 90%) as with 1% sulfur.
Use of a copper-tolerant strain of T. ferrooxidans improved the kinetics of Co and
Ni leaching.
Olson et aP9 reported successful leaching of cobalt from smelter wastes (sulfidic
dross furnace matte containing Co and Ni) by T. ferrooxidans in batch experiments.
In six weeks, they were able to solubilize two-thirds of the 8.5% cobalt in the matte
at a pulp density of 4% and a mesh size of -200 (<74 JllIl) in the basal salts solution
of 9K medium. At a mesh size of -50 +too mesh (150-300 Jlm), only 43% of the Co
was leached. Unlike Co, the Ni in the matte was leached by acid alone. Addition of
pyrite did not increase the rate or extent of Co leaching.
Nickel
Some studies on nickel extraction from nickel sulfide and sulfidic nickel ores
and ore tailings by acidophiles have appeared since 1990. One such study by
Amaratunga et al~o explored the bioextraction potential of ore tailings for heap
and vat leaching. Pyrrhotitic, nickel-containing (0.7-0.8%) tailings material from
Falconbridge, Ontario, Canada was subjected to leaching by a strain of T.
ferrooxidans in 9K salts medium in shake flasks at a pulp density of 8-10%. As
much as 50% of the Ni was extracted in 28 days in the presence of T. ferrooxidans.
It was suggested that agglomerating the tailings with gypsum hemihydrate (flue
gas desulfurization gypsum) prior to leaching might minimize segregation of fine
clays in the tailings, which could cause channeling in leach heaps and reduce solu-
tion contact with ore particles, but no test results were presented.
Kai et al~l studied the effect of enhancement of nickel tolerance to 1 giL in a
strain of T. ferrooxidans on nickel solubilization from NiS. The original strain
was isolated from acid mine water of the Yahara Mine in Japan. Tests were run in
batch experiments in 500 mL Erlenmeyer flasks containing 300 mL of a ferrous
iron (8 giL) solution at pH 2.0 and 0.1 g of reagent-grade NiS and shaken at 30OC.
About 40% nickel was extracted by the adapted strain in 150 h compared to 30% by
the unadapted strain; however, about 25% of the Ni in the NiS was solubilized in a
sterile control. No difference in the rate of FeH oxidation was noted between the
Ni-adapted and unadapted strains. From this the authors inferred that NiS oxida-
tion by the unadapted strain was indirect (due to oxidation by Fe3+ produced by
the bacteria) whereas the additional NiS oxidized by the adapted strain was the
result of direct attack. The authors reported that 30-50% of the cells of T.
ferrooxidans were adsorbed to the NiS.
Zinc
Zinc sulfide (sphalerite) has been shown by various investigators to be oxidiz-
able by iron-oxidizing acidophiles like T. ferrooxidans. Among studies since 1989,
one by Krafft and HallbergU assessed the possibility of in situ mining of zinc sul-
fide ores from two different Swedish mines, Saxberget and Kristineberg. Ore from
Saxberget mine contained 15% Zn and 1.5% Cu, whereas that from the Kristineberg
mine contained 8% Zn and 2.5% Cu. Leaching studies were performed in columns
and flasks using mixed cultures from the Falu copper mine in Sweden and the Rio
Tinto mine in Spain. The dominant species in these cultures was T. ferrooxidans.
The culture media were various modifications of 9K medium!5 In column leach-
ing, only 0.84-1.02% of the Zn and 1.8-3.2% of the Cu in the Saxberget ore were
leached in 87 days, and only 0.2-1.2% of the Zn and 0-1.2% of the Cu in Kristineberg
ore. Highest leaching rates with respect to Zn were obtained with particles in the
16-32 mm size-range. In batch-leaching experiments 80% of the Zn in either ore
was leached in 150 days.
In an attempt to enhance the kinetics of selectively leaching the Zn from cop-

per-zinc sulfide concentrate, Carranza et al23 employed a two-phase system in which
they generated ferric lixiviant with T. ferrooxidans in a first phase and then leached
the ore concentrate with the lixiviant in one or more reactors in series in the sec-
ond phase. The lixiviant-generating phase was operated at 31°C and the leaching
phase at 70°C. Galvanic effects due to the presence of chalcopyrite with the zinc
sulfide caused preferential leaching of Zn. Maximum Zn extraction of 80% at a
pulp density of 12% was obtained in a continuous process. The authors predicted
that maximum Zn extraction can be achieved with mean residence times ranging
from 8-22 h, depending on the number of reactors in series employed for the sec-
ond phase. The zinc in the pregnant solution from this process was recoverable by
solvent extraction and electrowinning.
Different leaching conditions can affect the rates and extents of Zn extraction
from Zn ores. Ballester et al24 compared three strains of T.ferrooxidans adapted to
FeS0 4, CuFeS 2 and ZnS, respectively, for their ability to oxidize ZnS in 9K me-
dium.'5 The two strains adapted to the sulfide ores solubilized more Zn from an
ore concentrate and without an initial lag than did the FeS0 4-adapted strain. When
three zinc-ore concentrates differing in the amounts and kinds of iron impurities
were leached with the ZnS-adapted strain of T.ferrooxidans, best results were ob-
tained with a concentrate in which ore impurities were dissolved in the sphalerite
(ZnS). Next best results were obtained with a concentrate comprised mainly of
sphalerite and pyrite. Least but nevertheless significant amounts of Zn, were dis-
solved from a concentrate comprising sphalerite, pyrite and galena. These differ-
ences may be at least in part explainable in terms of galvanic effects. Ballester et
al24 also suggested that FeH when dissolved in ZnS enhances the conductivity of
ZnS and its chemical leachability. In comparing leaching rates at 5, 10 and 15% PD
of ore concentrate, fastest leaching occurred at the 5% PD and slowest at 15% PD.
Ballester et al 24 suggested that the lower pulp density facilitated O2 and CO 2trans-
fer in the reactor system. They also found that progressively more Zn was dis-
solved as the reactor temperature was increased from 25°-30° and then to 35OC,
although initial rates at 30° and 35°C were similar. In bioleaching, the sphalerite
was preferentially attacked at the site of cracks and structural defects.
Sukapun et al 25 were able to extract Zn from zinc silicate ore from Thailand
using acid ferric sulfate lixiviant generated by T. ferrooxidans. The action was dem-
onstrated in shake flasks and columns. In shake flasks at a pulp density of 10%,
about 60% of the Zn in the ore was extracted in 20 days, during which the bacteria
generated ferric iron from ferrous iron. Ferrous iron leached as much as 40% of
the zinc in the ore, presumably by displacing it. At a pulp density of 2-5%, the
bacteria mobilized almost 75% of the Zn in the ore. In column experiments less
than 20% of the Zn in the ore was mobilized. In these experiments the medium
containing the culture was continually recycled. Control of pH was critical to main-
taining bacterial activity in the columns.
Complex Sulfide Ore

In leaching of a complex sulfide ore that included pyrite, pyrrhotite, pentland-
ite, chalcopyrite and sphalerite in a column with an acidophilic culture enrich-
ment from mine water, Ahonen and Tuovinen 26 found that new solid phases of
covellite (CuS) appeared on the surface of pyrrhotite inclusions, and that jarosite,
especially in the form of NaFe 3(S04MOH)6, and Fe(III) hydroxide formed on
leached mineral surfaces. The source of the sulfide for covellite formation was as-
sumed to have been nonoxidative dissolution of pyrrhotite. The authors noted
some elemental S formation from the partial oxidation of pyrrhotite. As the dis-
solved ferric iron concentration in columns increased, the leaching rates of Co, Cu
and Zn increased. When pH and redox potential in the columns exhibited low val-
ues, acid consumption by the system was greatest. However, Ni leaching did not
correlate with the level of dissolved ferric iron. In trickle leaching, in which the ore
in the column was not saturated with lixiviant, more acid was produced and the
redox potential was higher than in flood leaching. Variations in pH and Eh af-
fected the leaching rates of Co, Cu, Ni, and Zn from the ore differently. The active
bacteria in the mine water enrichment culture were not identified.
Gallium
A novel bacterium, an archaeon related to the Acidothermus genus of the
Sulfolobaceae, has been shown by Bowers-Irons et al27 to be able to leach gallium
from several ores and from waste semiconductor wafers. The organism is an aci-
dophilic (growth range pH 1.5 and 4.5), and thermophilic (growth range 85°-95"C),
facultative aerobe. It was isolated from a thermal spring in a mine in the state of
Utah (USA) and is culturable in 9K medium15 at an initial pH of 1.8 with and with-
out FeS0 4• When FeS0 4 is omitted, an alternate energy source such as a suitable
ore or GaAs is needed for growth. Iron was reported to interfere with gallium oxi-
dation. In the case of GaAs, Ga and As appear to become oxidized to Ga2 0 3 and to
arsenite and arsenate, respectively.
Molybdenum
Tedesco et al28 made the interesting observation in preliminary experiments,
that T. ferrooxidans attacked a low-grade sulfidic ore from EI Nevado de Famatina
(Argentina) containing Cu and Mo by selectively leaching the Cu. The test was
carried out in shake cultures in 9K medium of Silverman and Lundgren15 at 30"C.
In 60 days, 60% of the copper but only 0.34% of the molybdenum were leached
with the bacteria, whereas in uninoculated controls, just 9% of the Cu and no mea-
surable Mo were leached. The authors seem to have assumed that leaching in this
system was by the direct mode and did not consider the possibility that ferric iron
generated by the bacteria in the 9K medium was the primary oxidant. Further-
more, they also did not comment on the possibility that oxidized iron produced by
the bacteria might have precipitated solubilized molybdenum as ferric molybdate,
accounting for the apparent selective leaching of Cu. Molybdate precipitation would
also have warded off molybdate toxicity.
Donati et al29 reported, as previously mentioned by Tedesco et al28 that the
seemingly limited attack of MoS 2 in their experiments, was because their T.
ferrooxidans strain did not attach readily to the surface of molybdenite whereas
the cells readily attached to other metal sulfides tested. Pistaccio et al30 subsequently
showed that 5000 ppm Tween 80 in the absence of FeH in the medium increased
attachment of T. ferrooxidans to molybdenite from 17.9-87% of the total cell popu-
lation. In a bioleaching experiments with molybdenite in iron-free 9K medium
and T. ferrooxidans, they found very low extraction of Mo in the absence of added
Tween 80, but in the presence of 5000 ppm Tween 80, they found -16 ppm (mg/L)
molybdenum in solution after 28 days at 30"C. Growth and oxidation of molyb-
denite presumably ceased when toxic concentrations of molybdate in solution had
been reached. The authors showed that with their culture of T.ferrooxidans in 9K
iron medium,15 a molybdate concentration of 1 mM was partially inhibitory to iron
oxidation and 2 mM was fully inhibitory. Since in the molybdenite oxidation ex-
periments, iron-free 9K medium in which reagent grade MoS 2 was the energy
source, was used, the authors reasoned that the mode of bacterial attack of the
molybdenite was by direct oxidation. Thus they confirmed that molybdenite oxi-
dation by T. ferrooxidans is limited by molybdate toxicity. It was previously re-
ported that Acidianus brierleyi can oxidize molybdenite readily without the toxic-
ity effects experienced by T. ferrooxidans. 31
Silver and Manganese

No major studies on bacterial leaching of silver from sulfidic ores appear to
have been undertaken since the work of Ehrlich.3:>'33 A probable reason is the an-
ticipated toxicity of Ag+ for T. ferrooxidans and other acidophiles that are active in
leaching. The toxicity problem can be avoided if the soluble silver ion is complexed,
for instance by the excess chloride in a lixiviant like 9K medium.34
Porro and Tedesc035 reported the use of T. thiooxidans strain Thth 04 in
beneficiating an Argentinean manganiferous silver ore from Farall6n Negro (16.5%
Mn02; 100 ppm Ag; 10 ppm Au) by selective extraction of the manganese. Experi-
ments were performed in shake flasks and columns. Inocula were grown in a So-
mineral salts medium at an initial pH 2. Since the ore itself did not contain an
energy source for T. thiooxidans, and since leaching in the So-medium by T.
thiooxidans was poor, the effect of using FeS as an energy source was tested. In
shake flasks, 2 g FeS per 400 mL resulted in optimal Mn solubilization (-85%) in
70 days at 25"C. Intermediate products of oxidation of the sulfide in FeS by T.
thiooxidans were held responsible for the reductive dissolution of Mn(IV) oxide in
the ore. When the ore residue after Mn extraction was treated chemically with
NaCN reagent, 53% of the silver was recovered, which is an improvement over a
30%-silver recovery from an ore pretreated chemically with S02 to remove Mn.
Mn extraction in column experiments was less successful. Less than 20% of the
Mn was extracted in 60 days from 200-mesh ore in the presence of FeS and nearly
none in the absence of FeS. It seems as if a heterotrophic leaching approach as
(discussed below) might have yielded industrially more promising results.
Some anaerobic, facultatively mixotrophic, hydrogen-oxidizing bacteria can
reduce manganese(IV) in manganese ore to Mn(II) which may appear in solution
or as MnC03 (rhodochrosite). Although not directly investigated, these bacteria
probably can also reduce manganese in ores that occurs predominantly in the +3
oxidation state. In these instances, Mn(IV) or Mn(lII) serve as terminal electron
acceptors in the oxidation of hydrogen. Examples of organisms capable of using
H2 as electron donor in Mn(IV) reduction include Shewanella putrefaciens36 and
Geobacter sulfurreducens.37 None of these organisms have so far been tested to
assess the feasibility of their commercial exploitation in extracting Mn from man-
ganese oxide ores mixotrophically.
Two green phototrophic sulfur bacteria were recently shown to be able to use
freshly precipitated MnS and FeS as electron donors in CO 2 assimilation in the
light but not in the dark. 38 NiS, CuS, ZnS, CdS or PbS could not substitute for MnS
or FeS, probably because of their lower solubility. The organisms were Chlorobium
limicola and C. paeobacteroides. Manganese and iron from the corresponding sul-
fides passed into solution as MnH and FeH as a result of the oxidation. The sulfide
of the metal sulfides served as the source of reducing power in the photosynthetic
CO 2 assimilation. The sulfide in MnS was oxidized to sulfate with an intermediate
accumulation of elemental sulfur, whereas the sulfide in FeS was only oxidized to
elemental sulfur. Although the authors have speculated that these organisms may
oxidize mackinawite (FeS) in nature, no evidence exists that they can oxidize py-
rite (FeS 2) or the rare mineral hauerite (MnS 2), and therefore the industrial poten-
tial of these microbial capabilities remains in question.
Platinum Group Metals

Attempts have been made to enhance the extraction of encapsulated platinum
group metals (platinum, rhodium, palladium) by a biobeneficiation process. In
such a process, sulfidic ore components that encapsulate the platinum group met-
als are removed by bacterial oxidation prior to extraction of the platinum-group
metals by cyanidation. T. ferrooxidans was found to oxidize base metal sulfides in
Stillwater Complex (Montana, USA) flotation concentrate (in g/mt: Pd 1,142; Pt
328; Rh 11; and Au 19) at a temperature of 30°-35°Cin a stirred bioreactor contain-
ing mineral salts solution at pH 2.5 with the ore concentrate at 10% pulp density.39,4o
The organism oxidized 94% of the sulfide present in 35 days. In subsequent cya-
nide treatment of the ore residue at 80°C,76% of the palladium, 94% of the rhodium,
and 34% of the platinum were recovered.
Uranium
Although bioleaching of uranium has been extensively studied in the past (see
reviews by McCready and Gould4I and Ehrlich 6), a few recent studies deserve men-
tion. A bioleaching study of a Spanish uranium ore revealed differences in leach-
ing efficiency with different microbial populations, in this instance a natural mix-
ture of T. ferrooxidans, T. thiooxidans and L. ferrooxidans, moderate thermophiles
and heterotrophs, individual cultures of T. ferrooxidans, T. thiooxidans and L.
ferrooxidans, and by different mixtures of the pure cultures. 42 The ore used in these
investigations came from the Fe mine and contained about 4% pyrite. Its uranium
was mainly in the form of UO, with some U0 3, which was an alteration product.
Experiments were done in shake flasks and columns using dilute iron-free mineral
salts solution. At 35° and 40°C, the natural mixture of organisms was most effec-
tive. In artificial mixtures at these two temperatures, best results were obtained
when both T. thiooxidans and L. ferrooxidans were present. At 25°C in shake cul-
ture, T. ferrooxidans was as effective or better than the natural mixture, but in
columns, the natural mixed culture was more efficient in leaching uranium than
the pure cultures or mixtures of them. Altered ore (air-oxidized prior to bacterial
treatment) was leached more effectively than unaltered ore. With appropriate in-
oculum, about 70% of the uranium was mobilized from altered ore in the first two
weeks.
A Brazilian study by Garcia Junior43 in which uranium was bioleached from a
low-grade ore is an excellent example of a commercial feasibility assessment. The
ore in this study was from the Figueira mine in Paran. The assessment was based
on results from laboratory, semipilot and pilot-scale experiments. The bulk com-
position of the ore included quartz, silicates, kaolinite, gypsum and pyrite. On an
elemental basis, the ore contained 0.08% Ups, 6.91% Fe 20 3, 3.84% sulfide-sulfur,
and 3.84% total sulfur. The organism used in the leaching study was a strain of T.
ferrooxidans isolated from Figueira mine waters and cultured in 9K medium. 's
Bench-scale tests including suitable controls were carried out in shake flasks con-
taining 30 g of ore (-14 mesh) in water or iron-free 9K medium, and in columns
with 500 g of ore (particle sizes <0.25 inches) with water as leaching medium.
In both types of experiments, pH was stabilized at 2.8 by additions of 1 N H2S0 4
before inoculation. A semipilot test was run in a concrete column, 3 m-high by
1 m-diameter, filled with -3 tons of ore (particle sizes <10 inches) and 500 L of
water inoculated with the T. ferrooxidans strain after pH stabilization and con-
tinually recycled through the ore. For a pilot-scale test, heap-leaching was done
with 850 tons (particle size <10 inches) on a 2500-m2 impermeable pad connected
to three 200-m3 ponds by PVC pipelines. The ore was watered by a system of sprin-
klers. No pH adjustment of the ore was needed prior to initiation of the leaching
process. The pilot test apparently did not require inoculation but utilized the bac-
teria naturally associated with the ore. In shake flasks, about 58% of the uranium
was extracted in 60 days with bacterial inoculum in acidified water. About 60% of
the uranium was extracted from the ore in 40 days with bacterial inoculum in
iron-free 9K medium and 40% without inoculum. A lower pH (1.3) was generated
in the iron-free 9K medium than in the water (1.8). In the laboratory columns, 50%
of the uranium in the ore was leached in 45 days with bacteria and 30% without. In
the semipilot test, 48.5% of the uranium was mobilized in 90 days after stepwise
acidification of the ore with a total of 28 kg sulfuric acid per ton of ore over 45 days
of bacterial action. The acidification solubilized the ferric iron previously gener-
ated by the bacteria and needed for further uranium mobilization. In the heap-
leach test, 51.3% of the uranium was extracted from the ore in 83 days after stepwise
acidification with 32.4 kg sulfuric acid per ton of ore over 35 days of bacterial ac-
tion. The author concluded that heap-leaching is a feasible approach for uranium
recovery from the Figueira ore because the acid consumption was considerably
lower than for conventional acid leaching in a stirred tank. Capital outlays and
operational costs were estimated to be low.
A bioleaching study with Pakistani-sandstone-uranium-ore (0.049% uranium)
revealed that slag waste from a sulfuric acid plant, which contained iron and el-
emental sulfur, was a cheap and plentifully available energy source for T.
ferrooxidans and T. thiooxidans in leaching uranium in shake flask experiments
efficiently.44 The slag waste was used to replace FeS04 or elemental sulfur in other
experiments. Although iron as energy source was somewhat more effective than
sulfur in mobilizing maximum amounts of uranium, very significant amounts of
uranium were mobilized in the presence of sulfur, presumably through U(IV) oxi-
dation by reduced sulfur intermediates formed during bacterial sulfur oxidation
to sulfuric acid.
Leaching with Heterotrophic Bacteria

The following discussion of heterotrophic leaching reviews publications that
have appeared since the reviews by Rossi' and Ehrlich and by Brierley. 2
Metal Mobilization by Microbially Generated AcidsILigands
Cobalt and Nickel

A possible approach to extracting Co as well as other base metals from lateritic
ore and nonsulfidic metalliferous waste is heterotrophic leaching (for a general
overview, see Burgstaller and Schinner45 ). A number of investigators have taken
this approach in their studies of Co and Ni extraction from laterite. They used
fungal acids as lixiviant.
Alibhai et al46 explored Ni extraction from six different Greek lateritic ores,
using bioacids either of analytical grade or as generated in culture medium by
selected strains of Aspergillus and Penicillium. Leaching with analytical grade
bioacids was done in shaken-flask as well as in fIxed-bed trickle columns. In the
latter mode, the lixiviant was sprayed on the top of a column at a specifIed rate
with limited recycling. Leaching with microbially produced bioacids was done in
single- and two-phase systems. In the two-phase mode, the bioacids were gener-
ated in one reactor, and the resultant spent medium was then applied to ore in a
second reactor. In the single-phase mode, the bioacids were produced by the fungi
in the presence of the ore. In leaching with purifIed bioacids in the shaken mode,
up to 40% of the Ni in low-grade ore was extracted in 12 days, and 55% of the Ni in
high-grade ore in 9-15 days. In the column mode, >30% of the Ni in low-grade ore
was extracted with citric acid after 100 days. Sulfuric acid was somewhat more
effective than citric acid but both acids were more selective for Ni than for Fe.
Oxalic acid extracted Fe in preference to Ni. Its presence is thus not desirable among
fungally produced bioacids used as lixiviants for nickeliferous laterites. Leaching
with bioacids generated in flask cultures of Aspergillus and Penicillium strains was
performed in one-phase and two-phase systems. The authors reported that 68% of
the Ni in low-grade Litharakia ore (0.73% Ni; 0.043% Co) was leached in 51 days in
a two-phase system, using an Aspergillus strain in a sucrose-containing medium.
This result was attributed to optimal citrate production. Tests in single-phase mode
were expected to give somewhat better results than tests in two-phase mode, but
were not reported in detail by these authors. Detailed results from such tests with
a different ore were reported by Kar et al47 (see below).
A more detailed study of the action of analytical-grade organic acids was un-
dertaken by Tzeferis and Agatzini-Leonardou. 48 The action of these organic acids
on high-grade Kastoria laterite (4.23% Ni; 0.028% Co) and low-grade Litharakia
ore (0.73% Ni; 0.043% Co) singly and in combination with and without added
H 2 S0 4 was compared to that of H 2 S0 4 alone. The organic acids tested included
citric, lactic, formic, oxalic and salicylic. Leaching tests at temperatures up to 50"C
were carried out in stirred conical flasks in a water bath, whereas tests from 50°-95°C
were carried out in spherical, stirred glass reactors heated by a thermostatically
controlled thermal mantle. Lixiviant concentrations, pulp densities, temperatures,
and reaction durations were varied for each of the different lixiviants. Of all the
organic acids tested, citric was the most effective (max. extraction -60%), but
equimolar concentrations of citric and sulfuric acid were in some instances slightly
more effective. Reaction kinetics for citric acid were slower than for sulfuric acid.
Citric acid lixiviant whose acidity was adjusted with sulfuric acid to a free acid
concentration of 0.5 giL was more effective than an equivalent concentration of
citric acid whose free acid concentration was not adjusted, or sulfuric acid alone.
This effect was explained in terms of the chelation of Ni by citric acid. Oxalic acid
was found to extract iron preferentially. It liberated very little Ni in the tests. The
extent of maximum Ni extraction from different ores under identical conditions
varied.
Experimental evidence for the greater effectiveness of the single-phase mode
versus the two-phase mode in extracting Ni from lateritic ore was presented by
Kar et al.47 They found that more cobalt was extracted from lateritic nickel ore
from India when the fungus Rhizopus arrhizus produced lixiviants in the presence
of the ore than when using lixiviant consisting of culture medium in which the
fungus had grown in the absence of the ore and from which the fungus had been
removed by fIltration after its growth. They attributed this difference in effect to
higher local concentration of the bioacids formed in the presence of the fungus
than in fungus-free spent medium. Their culture medium was potato-dextrose
broth. The experiments were performed in 250 mL conical flasks on a shaker at
370C and 120 rev/min. After 20 days, 73% of the cobalt and 4.5% of the Ni were
extracted from -350 mesh laterite containing 0.039% Co and 1.1% Ni at a pulp den-
sity of 5%. The acids produced by R. arrhizus in this study were not identified but
can be assumed to have functioned as protonation agents in the displacement of
cobalt and nickel from the ore and/or as ligands of these metals. On extended in-
cubation, the fungal mycelium probably bound some of the solubilized metal.
Alibhai et al49 reported on Ni extraction from high-grade laterite ore (Kastoria
deposit, Greece) separately with acetic, formic, lactic, oxalic, and citric acids. Each
of these can be microbiologically formed as end products of energy metabolism.
Of these acids, 1.5 M citric acid gave the best results with 40% Ni extraction in 20
days at 5% pulp density. By contrast, 1.5 M H"SO 4 extracted 60% of the Ni under
the same conditions. Some fungal strains of Penicillium and Aspergillus extracted
50-60% of the Co and Ni from low-grade Litharakia ore, which contained 0.92 NiO
and 0.060% CoO in sucrose-mineral salts medium. From 2.5-3.5% of the Ni was
adsorbed by the fungus mycelium. When molasses replaced sucrose in the me-
dium, 20% less Ni was extracted from the ore.
Tzeferis et alSo performed similar bioleaching experiments with the low-grade
nickel ore from Litharakia. They also used strains of the Penicillium and Aspergil-
lus in dextrose- and sucrose-mineral salts media at an incubation temperature of
30OC. The ore had a mesh size of -150 «106 11m). The ore sample used in this study
had a nickel content of 0.73% but the Co content was not reported. Cultures were
shaken at 400 rev/min. They found 60% of the Ni and 50% of the Co in the ore to
be extracted in sucrose medium in 48 days, but much less in dextrose medium. A
portion of the leached Ni became bound to the fungal mycelium. The lixiviant
produced by the fungi contained citric and oxalic acids. Citric acid production
was much better with sucrose as carbon/energy source than with dextrose. Tarasova
et alS1 reported that direct ultrasonic treatment at a frequency of 18-20 kHz and an
intensity of 1-9 W/cm-" as a 30- or 120-sec pulse at the beginning of a leaching run
with Aspergillus or Penicillium sp. improved Ni extraction from low-grade laterite
ore from Greece from 40-50%. The authors suggested that the ultrasonic energy
uncovered occluded minerals in the ore and/or opened channels that facilitated
access of the lixiviant to the valuable minerals. This treatment was, however, seen
as too costly to apply commercially in the bioleaching of low-grade nickeliferous
laterite.
Tzeferis 52 studied the bioleaching of a low-grade laterite ore (-150 mesh or

<106 11m particle size) from LARCO S.A. (Larymna, Greece) in a two-phase sys-
tem using a molasses-based medium. The ore contained 0.73% Ni and 0.043% Co.
He generated lixiviant (bioacids, mainly citric and some oxalic) in a molasses-
based medium with citric acid-producing Penicillium and Aspergillus strains in
air-lift fermenters over eight days. The molasses was pretreated with potassium
ferrocyanide to control the concentrations of Fe, Cu, Ca, and Mn, which affect
citric acid production by the molds. The ore was then leached with the spent me-
dium containing the bioacids in spherical stirred-glass reactors at 95OC. Maximum
Co recovery from the ore ranged from 44-49% and maximum Ni recovery from
54-62%. Since molasses is cheaper than sucrose (cane or beet sugar) or dextrose,
its use in leaching Co and Ni from laterite on an industrial scale is more economic.
However, its pretreatment for control of Fe, Cu, Ca, and Mn levels in the fungal
culture medium to maximize organic acid production is essential.
Sukla and Panchanadikar53 bioleached a nickeliferous laterite from Sukinda,
India with an Aspergillus niger strain isolated from the ore. Leaching was carried
out on 2 g of the ore (particle size of -150 mesh) in 250 mL Erlenmeyer flasks con-
taining 100 mL of a medium that included dextrose (2-10%) and 20% potato ex-
tract (single-phase system) on a rotary shaker at 37OC. The ore contained 1.1% Ni,
and 0.03% Co. Unlike experiments by some other investigators who found sucrose
more effective than dextrose for the production ofbioacids, Sukla et al found that
with 10% dextrose, 92% of the nickel and 34% of the cobalt were extracted in 20
days at an initial pH of 6.5, which fell to 1.75-2.0 after 10 days.
Sukla et al54 reported that in leaching Ni from nickeliferous laterite from
Sukinda, India, with nickel-tolerant, autotrophic T. ferrooxidans, and heterotrophic
Bacillus circulans, Bacillus licheniformis and A. niger, T. ferrooxidans was least ef-
fective and B. circulans most effective among the three bacterial cultures. How-
ever, A. niger was even more effective than B. circulans. Thus, in 20 days at a pulp
density of 5% in appropriate growth medium for each of the organisms in shaken
flasks, T. ferrooxidans leached 6% of the Ni in the ore, B. licheniformis 35%, B.
circulans 85.4% and A. niger 92%. However, A. niger exhibited a lag of about 10
days before the rate of leaching became significant whereas B. circulans exhibited
no such lag.
Zinc and Lead

Zinc can be leached from some non sulfidic ores and waste materials. 45
Tungkaviveshkul et al55 demonstrated Zn bioextraction from Zn silicate residue
left after conventional processing of high-grade Zn silicate ore. They used Aspergil-
lus niger ATCC 11414 and a strain of Penicillium isolated from the residue in their
study. For Zn leaching, the cultures were grown in the presence of the ore at 10%
pulp density in a sucrose-mineral salts medium. Both strains leached Zn from the
ore. More than 60% Zn was extracted in 49 days, most of that in the first 14 days. A.
niger produced citric, oxalic and tartaric acids, whereas Penicillium produced
mainly citric and tartaric. The authors showed that this combination of acids was
more effective in extracting the Zn than the individual acids. The fungal mycelium
was found to adsorb some of the mobilized zinc. Some mobilized Zn was precipi-
tated as Zn phosphate hydrazine in the culture medium. To maximize Zn recovery
in solution, the culture conditions require further optimization.
The yeast Brettanomyces lambicus was shown to be able to leach 65% of Zn

from filter dust residue of a copper refinery in a medium containing glucose and
yeast-malt extract, 11.2% of the lead in a yeast extract-peptone medium, and 16.5%
of the copper in a glucose-mineral salts medium. 56 Twenty other yeast strains ex-
hibited different abilities to leach Zn, Pb, and Cu from the filter dust. Best results
were obtained at a pulp density of 0.5%. The leaching agents produced by the yeasts
and implicated by the authors included lactic and acetic acids and possibly amino
acids. The authors also reported that more metal was extracted by 17 of the 21
strains when they grew in the presence of filter dust than by their respective spent
culture medium. It is noteworthy that Brettanomyces lambicus produced acid only
in the presence of filter dust.
Gallium
The ability of five of six alkalophilic bacteria to produce siderophores that can
react with gallium has been reportedF They included two gram-negative rods,
two Gram-positive rods, one of which was a sporeformer and the other a coryne-
form, and a gram-positive coccus. The complexation of the Ga by the siderophores
was measured in terms of uptake by the respective bacteria. The range of uptake
was from 4-89%, depending on the culture. Iron in the medium interfered with
gallium uptake when both were supplied in the medium at 10 J1M concentration.
Previous experimentation had shown that 10 JlM Fe lowered siderophore produc-
tion, by as much as 10-fold. Ga uptake by one of the organisms was followed in a
time-course experiment, which showed that Ga removal from solution occurred
during the exponential and stationary growth phases and was correlated with
siderophore production. The structure of the siderophores was not established in
this study. These findings suggest that it may be possible to extract Ga from ores
with siderophores. For optimal exploitation, bacterial strains capable of produc-
ing siderophores that can complex gallium might be genetically engineered so that
they will produce the siderophore while being rendered incapable of accumulat-
ingGa.
Manganese and Silver

A number of heterotrophic bacteria have been shown to be able to reduce
Mn(IV) and probably Mn(III) aerobically and/or anaerobically (for reviews, see
Ehrlich,6 Ghiorse,5 8 Lovley,59 Nealson and Saffarini6o ). Such bacteria have poten-
tial usefulness in extracting manganese from low-grade manganese-oxide ores.
A comparison of three different modes of leaching of a wad ore (-16% Mn,
-1% Fe) has been reported by Nobel et al61 and Baglin et al. 62 Tests were carried
out in shaken flasks, columns, and heaps. The column tests, which were run at
22°C, employed wad ore of -6.3 and -19 mm particle size with molasses solution fed
to the top of the columns and continually recycled for 4- to 6-week periods, after
which it was completely replaced with fresh medium. Some leach tests in shaken
flasks were done in a glucose-mineral salts medium with and without phosphate,
and others in a molasses solution (3-5% w/v) at an initial pH of 5.4-5.7. Omission of
phosphate from the glucose-mineral salts medium suppressed development of fungi
but not bacteria and also prevented the formation of Mn3(P04)2.3H20 precipitate.
In most tests, microorganisms naturally associated with the ore were employed as
the microbial leaching agents; they included bacteria and fungi. In one of the shake-
flask tests, as much as 47% of the manganese in a wad ore (15.6% Mn) was solubi-
lized in six weeks by a mixed bacterial culture in glucose-mineral salts medium,

whereas in factory-grade molasses, 97% of the Mn was leached. In a column test,
as much as 99% of the Mn in the wad ore was leached in 51 days by microorgan-
isms in molasses solution. On extended leaching, little difference in Mn solubiliza-
tion was seen between columns inoculated with a mixed bacterial culture from a
flask test and uninoculated columns. In a heap-leach test, only 10% of the Mn in
the wad ore was leached in 57 days.
Toro et al63 evaluated the effect of different culture conditions on bacterial leach-
ing of an Italian manganese ore (17.4% Mn, 1.7% Fe; -200 mesh size) in shake-
flasks. They observed faster leaching with molasses than with sucrose in a mineral
salts medium that contained a trace of yeast extract, although the same amount of
Mn was finally solubilized with either carbon/energy source. Sucrose was con-
sumed at a faster rate and more completely in the presence of the ore than in its
absence. Sucrose concentration as well as pulp density affected the extent of Mn
leaching. At a pulp density of 15%, an initial sucrose concentration of 25 g/L al-
lowed for the most efficient extraction of Mn from the ore.
An attempt has been made to relate bioleaching rates of individual manganese
minerals and binary mixtures of manganese minerals to electrochemical poten-
tials. 64 The experiments were performed at 28°C using Achromobacter delicatulus
(Enterobacter cloacae) 182-A in columns containing one or a mixture of two test
ores that were fed with a glucose-mineral salts medium. According to the authors,
A. delicatulus 182-A produced citric and malic acids from glucose, which mobi-
lized the Mn in the minerals tested. The medium was not recycled but replaced
with fresh medium every 3-4 days. The organism lowered the pH of the medium,
with the extent of the pH drop from -7.7, depending on the ore that was being
leached. The electrochemical potential of individual manganese minerals was found
to be inversely related to the extent to which A. delicatulus 182A leached them.
With binary mixtures of manganese minerals in which one of the components was
either manganite or todorokite, the Mn leaching rate varied directly with the mag-
nitude of the electropotential difference between the two minerals. The synergis-
tic effect due to potential difference between component minerals was greater with
todorokite mixtures than with manganite mixtures.
The possibility of Mn recovery by heterotrophic bioleaching of manganese di-
oxide tailings has been explored. 65 Tests were run with Groote Eylandt Mining
Company (Australia) tailings (17% Mn, 8% Fe, 28% SiO a, and 22% Ala0 3), equiva-
lent to a low-grade ore, in a 10% (w/v) nonsterile molasses medium at initial pH
5.9 in stationary flasks and in continuously stirred reactors. The medium in flask
experiments was replaced with fresh medium every three days. In such experi-
ments at a pulp density of 5%, the best leaching rate, 3 g MnH /LI day (equivalent to
20 g Mna+Ikg/day) before the first medium change, was obtained with a mixed
culture at 50°C. This was 2.8 times the rate at 70°C. In a stirred reactor at 32°C, at
5% pulp density and 15% molasses concentration, the leaching rate was 1.5 g MnH /
L/day. The author suggested that under optimized conditions, this process of Mn
extraction could be economical if linked to formation of a valuable product like
high-grade manganese dioxide from the leached manganese.
Reductive dissolution of Mn(IV) oxides may be exploited in the beneficiation
of manganiferous silver ores in which the silver is encapsulated in manganese
oxides, which makes it inaccessible to cyanide lixiviant. Although such reductive
dissolution can be achieved nonbiologically by treating the ore with SO..,66,67 this
approach is presently deemed uneconomical on an industrial scale. Reductive dis-

solution of Mn(IV) oxides can also be achieved by heterotrophs, using one of two
possible approaches, namely indirect and direct attack of the Mn(IV) oxide. An
indirect approach was taken by Gupta and Ehrlich,68 who used a strain of Penicil-
lium isolated from a manganiferous silver ore, which produced a lixiviant in a
medium containing glucose and peptone dissolved in dilute seawater or 3% NaC!.
The lixiviant solubilized the Mn reductively by a nonenzymatic mechanism. At
low chloride concentration, Mn was solubilized selectively, whereas at elevated
chloride concentrations, silver was also solubilized due to its complexation by ex-
cess chloride.
An industrially more attractive approach is one involving enzymatic reduction
of Mn(IV) and was taken by Rusin et al,69 who selectively leached manganese from
manganiferous silver ore with Bacillus polymyxa D1 and B. circulans MBX1 in sepa-
rate experiments by reductive dissolution of Mn(IV) oxide. In a subsequent step,
they extracted 70 to more than 90% of the silver from the residual ore with cya-
nide (1 g NaCN per liter of ore slurry at pH 11). In a modification of the procedure
that involved addition of an organic collector for the metals (composition propri-
etary) to the growth medium, Rusin et al70 demonstrated simultaneous solubiliza-
tion of 99.8% Mn and 86% of the Ag from an ore by B. circulans MBX 1 at 15% pulp
density. An additional 8.5% of the residual silver in the ore residue was subse-
quently extracted by cyanidation. The same organism was also able to solubilize
molybdenum from manganiferous molybdenum ores. The reader is referred to
Rusin and Ehrlich for additional discussion of these processes.71
Lithium
The possibility of extracting lithium from spodumene (LiAISi2 0 6) through ac-
celeration of a microbial weathering process has been explored using bacteria and
fungi. The possibility of microbial participation in the mobilization of Li in spo-
dumene through weathering was suggested by Karavaiko et al.72 Such microbial
activity was subsequently demonstrated in laboratory experiments.73•74 The fungi
Penicillium notatum and A. niger were shown to liberate Li, AI and Si from spo-
dumene in a mineral salts medium containing 5% sucrose in shaken flasks.73 In the
same study, T. thiooxidans was shown to mobilize Ii in Waksman's sulfur medium,
and "silicate" bacteria, later identified as Bacillus mucilaginosus and B. circulans,
in a sucrose medium in shaken flasks. Of the three types of organisms, the activi-
ties in decreasing order were fungi> > silicate bacteria>T. thiooxidans. Other bac-
teria naturally associated with the spodumene were only weakly active. Slime
formed by the silicate bacteria was by itself able to mobilize Ii from spodumene.73
Avakyan et al74 explored the action of the fungal mixture of Trichoderma lignorum
and P. notatum in mobilizing Li from pegmatite rock from a Li deposit in a 20%
sucrose-mineral salts medium in percolators at 24-26"C. After one month, the fungi
had leached 0.2% of the Li from a mixture of 50 g spodumene, 30 g pegmatite, and
20 g shale, and after 7 months they had leached 0.38%.
Ilgar et al75 examined lithium extraction from spodumene by the fungus A.
niger. This organism produces oxalic and citric acid from sucrose in a mineral
salts solution. These acids extract the Ii from ~-spodumene through acidolysis
and complexation reactions. Most efficient Li extraction by spent culture medium
from the fungus culture was achieved at 90"C and at a pulp density of 20%.
Biobenejiciation for Base and Platinum Group Metals 145
Bauxite
Bauxite is the ore from which aluminum is obtained. An ideal bauxite for alu-
minum production should contain >40-50% A120 3 (gibbsite, diaspore, boehmite),
<20%Fe20 3, and 3-5% combined silica (silicate (kaolinite) + Si02).76 Some poten-
tially economic deposits may contain Fe203 and/or combined silica in excess of
these concentrations. Excess combined silica is undesirable because in commer-
cial chemical extraction of the aluminum from bauxite by the Bayer process, solu-
bilized silicate from the combined Si02will combine with some of the solubilized
aluminum and form aluminosilicate. This sequestered aluminum is lost in the sub-
sequent electrolytic recovery of aluminum metal. Removal of excess Fe and/or Si
by microbial means has been considered. Ogurtsova et al77 compared the activity
of various microorganisms in shake-cultures on a kaolinite-hematite-boehmite-
containing bauxite. They found that strains of the mycelial fungi A. niger and Peni-
cillium chrysogenum mobilized the AI, Fe and Si in the ore to varying degrees
whereas several yeasts exhibited only weak solubilizing activity. Among bacteria,
Pseudomonas spp. were more active than the yeasts but much less active than the
mycelial fungi. Strains ofB. mucilaginosus, B. circulans, and B. polymyxa, so-called
silicate bacteria, were inactive. T. thiooxidans solubilized some aluminum but none
of the Si or Fe. Groudev7s and Groudeva and Groudey79 had previously reported
success in partially desilicifying some other bauxite ores with B. circulans and B.
mucilaginosus. The desilicifying activity was attributed to the exopolysaccharide
produced by the bacteria. Karavaiko et also found that in the case of hematite-
kaolinite-gibbsite and hematite-goethite bauxites, desilicifying activity by B.
mucilaginosus was due to adsorption of silica- and silicate-containing ore fines by
the bacterial exopolysaccharide. These authors also reported that citric acid pro-
duced by A. niger and Aureobasidium pollulans could release aluminum from
aluminian goethite, which could then be recovered for subsequent processing to
aluminum metal.
Ehrlich et al S1 reported extensive iron solubilization from a pisolitic bauxite
under anaerobic conditions by a mixed bacterial flora associated with the ore. The
solubilized iron was mainly in the ferrous form.
Conclusion
While the foregoing discussion clearly shows that microbes have a potential for
assisting in extraction of base metals and some other metals from ores, either
through their direct mobilization (bioleaching) or through removal of interfering
ore constituents (biobeneficiation), economic exploitation of these activities on a
commercial scale remains in most cases to be demonstrated. In many of the in-
stances discussed here, the microbial activities need to be optimized and incorpo-
rated into industrial process designs before a meaningful economic analysis is
possible. In most instances, rates of reaction at mesophilic or thermophilic tem-
perature ranges need to be accelerated significantly over those observed in initial
laboratory experiments, irrespective of whether autotrophs or heterotrophs are
the biological agents. In the case of heterotrophs, an economically viable process
also requires cheap organic substrates and selective culture conditions that can be
easily established and maintained on a large scale at minimum cost. Microbiologi-
cal bioleaching or biobeneficiation processes of the ores discussed in this chapter
will be competitive with conventional ore processing (pyro- or hydrometallurgy)
if plant investments, costs of production, and control of environmental pollution

are favorable, considering the extent and quality of the ore resource to be pro-
cessed and the value of the product.
References
1. Rossi G. Biohydrometallurgy. Hamburg: McGraw-Hill, 1990.
2. Ehrlich HL, Brierley CL, eds. Microbial Mineral Recovery. New York: McGraw-
Hill, 1990.
3. Lawrence RW, Poulin R. The demand for biotechnology in mining in the 21st
century. In: Vargas T, Jerez CA, Wiertz JV et al. Biohydrometallurgical Process-
ing. Vol 1. Santiago: University of Chile, 1995:185-195.
4. Ehrlich HL. Microbes for biohydrometallurgy. In: Smith RW, Misra M, eds. Min-
eral Bioprocessing. Warrendale, PA: The Minerals, Metals and Materials Society,
1991:27-41.
5. Ehrlich Hi. Metal extraction and ore discovery. In: Lederberg J, ed. Encyclopedia
of Microbiology. Vol 3. San Diego, CA: Academic Press, 1992:75-80.
6. Ehrlich HL. Geomicrobiology. 3rd edition. New York: Marcel Dekker, 1995.
7. Rojas J, Giersig M, Tributsch H. Sulfur colloids as temporary energy reservoirs
for Thiobacillus ferrooxidans during pyrite oxidation. Arch Microbiol 1995;
163:352-356.
8. Natarajan KA. Electrochemical aspects of bioleaching of base-metal sulfides. In:
Ehrlich HL, Brierley CL, eds. Microbial Mineral Recovery. New York: McGraw-
Hill, 1990:79-106.
9. Silver M, Torma AE. Oxidation of metal sulfides by Thiobacillus ferrooxidans
grown on different substrates. Can J Microbiol1974; 20:141-147.
10. Rossi G, Ehrlich HL. Other bioleaching processes. In: Ehrlich HL, Brierley CL,
eds. Microbial Mineral Recovery. New York: McGraw-Hill, 1990:149-170.
11. Torma AE, Wey JE, Pesic B. Bioextraction of cobalt from cobaltite. Metallwissen-
schaften und Technik (Berlin) 1993; 47:648-651.
12. Thompson DL, Noah KS, Wichlaz PL et al. Bioextraction of cobalt from complex
metal sulfides. In: Torma AE, Wey JE, Lakshmanan VL, eds. Biohydrometallurgical
Technologies. Vol 1, Bioleaching Processes. Warrendale, PA: The Minerals, Met-
als, and Materials Society, 1993:653-664.
13. Morin D, Battaglia F, D'Hugues P. Bioleaching of cobaltiferous pyrite. In: Vargas
T, Jerez CA, Wiertz JV et aI, eds. Biohydrometallurgical Processing. Vol 1. Santiago:
The University of Chile, 1995:471-482.
14. Morin D, F. Battaglia F, Ollivier P. Study of the bioleaching of cobaltiferous py-
ritic concentrate. In: Torma AE, Wey JE, Lakshmanan VI, eds. Biohydrometallurgy.
Vol 1, Bioleaching processes. Warrendale, PA: The Minerals, Metals, and Materi-
als Society, 1993=147-156.
15. Silverman MP., Lundgren DG. Studies on the chemo-autotrophic iron bacterium
F. ferrooxidans: 1. An improved medium and harvesting procedure for securing
high cell yields. J Bacteriol1959; 77=642-647.
16. Battaglia F, Morin D, Ollivier P. Dissolution of cobaltiferous pyrite by Thiobacillus
ferrooxidans and Thiobacillus thiooxidans: factors influencing bacterial leaching
efficiency. J Biotechnol 1994; 32:11-16.
17. Baldi F., Bralia A, Riccobono F et al. Bioleaching of cobalt and zinc from pyrite
ore in relation to calcitic gangue content. World J Microbiol Biotechnol 1991;
7:298-308.
18. Nakazawa H, Sato H. Bacterial leaching of cobalt-rich ferromanganese crusts. Int
J Miner Process 1995:43:255-265.
19. Olson GJ, Sakai CK, Parks EJ et al. Bioleaching of cobalt from smelter wastes by
Thiobacillus ferrooxidans. J Ind Microbiol 1990; 6:49-52.
20. Amaratunga L, Tackaberry P, Lakshmanan VI et al. Extraction potential using
biological techniques on agglomerated material in a heap or vat leach process.
In: Torma AE, Wey JE, Lakshmanan VI, eds. Biohydrometallurgical Technolo-
gies. Vol 1. Bioleaching Processes. Warrendale, PA: The Minerals Metals and
21. Kai T, Nishi M, Takahashi T. Adaptation of Thiobacillus ferrooxidans to nickel
ion and bacterial oxidation of nickel sulfide. Biotechnol Lett 1995; 17:229-232.
22. Krafft C, Hallberg RO. Bacterial leaching of two Swedish zinc sulfide ores. FEMS
Microbiol Rev 1993; 11:121-128.
23. Carranza F, Iglesias N, Romero R et al. Kinetics improvement of high-grade sul-
fides bioleaching by effects separation. FEMS Microbiol Rev 1993; 11:129-138.
24. Ballester A, BI zquez ML, Gonzalez F et al. The influence of different variables on
the bioleaching of sphalerite. Biorecovery 1989; 1:127-144.
25. Sukapun J, Thiravetyan P, Tanticharoen M. Bioleaching of zinc from zinc silicate
residue. In: Vargas T, Jerez CA, Wiertz JV et aI, eds. Biohydrometallurgical Pro-
cessing. Vol 1. Santiago: University of Chile, 1995:351-358.
26. Ahonen L, Tuovinen OH. Bacterial leaching of complex sulfide ore samples in
bench-scale column reactors. Hydrometallurgy 1995; 37:1-21.
27. Bowers-Irons G, Pryor R, Bowers-Irons T et al. The bio-liberation of gallium and
associated metals. In: Torma AE, Wey JE, Lakshmanan VI, eds. Biohydrometal-
lurgical Technologies. Vol 1. Bioleaching Processes. Warrendale, PA:The Minerals
Metals and Materials, 1993:335-342.
28. Tedesco PH, deCordo LB, Piro A. Lixiviaci6n selectiva de molibdeno y de cobre a
partir de un mineral de baja ley (Selective leaching of molybdenum and copper
from a low-grade ore). Rev Metal (Madrid) 1990; 26:151-156.
29. Donati E, Curutchet G, Porro S et al. Bioleaching of metallic sulfides with
Thiobacillus ferrooxidans in the absence of iron(II). World J Microbiol Biotechnol
1992; 8:305-308.
30. Pistaccio L, Curutchet G, Donati E et al. Analysis of molybdenite bioleaching by
Thiobacillus ferrooxidans in the absence of iron. Biotechnol Lett 1994:16:189-194.
31. Brierley CL, Murr LE. Leaching: use of a thermophilic chemoautotrophic microbe.
Science 1973; 179:488-490.
32. Ehrlich HL. Bacterial leaching of silver from a silver-containing mixed sulfide ore
by a continuous process. In: Lawrence RW, Branion RMR, Ebner H, eds. Funda-
mental and Applied Biohydrometallurgy. Amsterdam: Elsevier, 1986:77-88.
33. Ehrlich HL. Bioleaching of silver from a mixed sulfide ore in a stirred reactor. In:
Norris PR, Kelly DP eds. Kew Surrey, UK: Biohydrometallurgy. Science and Tech-
nology Letters, 1988:223-231.
34. Ehrlich HL. Inorganic hazardous waste amenable to biological transformation.
In: Stoner DL, ed. Biotechnology for the Treatment of Hazardous Waste. Boca
Raton, FL: Lewis Publishers, 1994:27-44.
35. Porro S, Tedesco PH. Biolixiviaci6n de un mineral de manganeso y plata (Bio-
lixiviation of a mineral containing manganese and silver). Rev Metal (Madrid)
1990; 26:36-40.
36. Lovley DR, Phillips EJP, Lonergan DJ. Hydrogen and formate oxidation coupled
to dissimilatory reduction of iron and manganese by Alteromonas putrefaciens.
Appl Environ Microbiol 1989; 55:700-706.
37. Caccavo Jr F, Lonergan DJ, Lovley DR et al. Geobacter sulfurreducens sp. nov., a
new hydrogen- and acetate-oxidizing dissimilatory metal-reducing microorgan-
ism. Appl Environ Microbiol 1994; 60:3752-3759.
38. Borrego C, GarCja-Gil J. Photosynthetic oxidation of MnS and FeS by Chlorobium

sp. Microbiologia SEM (Madrid) 1995; 11:351-358.
39. Yopps D, Baglin EG. Extracting platinum-group metals from Stillwater complex
flotation concentrate. In: Smith RW, Misra M, eds. Mineral Bioprocessing.
Warrendale, PA: The Minerals, Metals and Materials Society, 1991:247-260.
40. Langhans DL, Baglin EG. Biological oxidation of a platinum-group metal flota-
tion concentrate and converter matte. In: Torma AE, Wey JE, Lakshmanan VI,
eds. Biohydrometallurgical Technologies. Vol I. Bioleaching Processes. Warrendale,
PA: Minerals, Metals, and Materials Society, 1993:315-324.
41. McCready RG., Gould WD. Bioleaching of uranium. In: Ehrlich HL, Brierley CL,
eds. Microbial Mineral Recovery. New York: McGraw-Hill, 1990:107-125.
42. Munoz JA, Gonzalez F, Ballester A et al. Bioleaching of a Spanish uranium ore.
FEMS Microbiol Rev 1993; 11:109-120.
43. Garcia Junior O. Bacterial leaching of uranium ore from Figueira-PR, Brazil, at
laboratory and pilot scale. FEMS Microbiol Rev 1993; 11:237-242.
44. Khalid AM, Anwar MA, Shemsi AM et al. Biohydrometallurgy of low-grade, car-
bonate bearing sandstone uranium ore. In: Torma AE, Wey JE, Lakshmanan VI,
eds. Biohydrometallurgical Technologies. Vol I. Bioleaching Processes. Warrendale,
PA: The Minerals, Metals, and Materials Society, 1993:285-292.
45. Burgstaller W, Schinner F. Leaching metals with fungi. J Biotechnol1993; 27:91-116.
46. Alibhai K, Leak DJ, Dudeney AWL et al. Microbial leaching of nickel from low-
grade Greek laterites. In: Smith, RW, Misra, M, eds. Mineral Bioprocessing.
Warrendale, PA: The Minerals, Metals, and Materials Society, 1991:191-205.
47. Kar RN, Sukla LB, Panchanadikar VV et al. Extraction of cobalt and nickel from
lateritic nickel ore using Rhizopus arrhizus. In: Vargas T, Jerez CA, Wiertz JV et
aI, eds. Biohydrometallurgical Processing. Vol 1. Santiago: University of Chile,
1995:417-424.
48. Tzeferis PG, Agatzini-Leonardou S. Leaching of nickel and iron from Greek non-
sulfide nickeliferous ores by organic acids. Hydrometallurgy 1994; 36:345-360.
49. Alibhai KAK, Dudeney, AWL, Leak DJ et al. Bioleaching and bioprecipitation of
nickel and iron from laterites. FEMS Microbiol Rev 1993; 11:87-96.
50. Tzeferis PG, Agatzini S, Neratzis ET. Mineral leaching of non-sulfide nickel ores
using heterotrophic micro-organisms. Lett Appl Microbiol 1994; 18:209-213.
51. Tarasova, II, Khavski NN, Dudeney AWL. The effects of ultrasonics on the
bioleaching of laterites. In: Torma AE, Wey JE, Lakshmanan VI, eds. Biohydro-
metallurgical Technologies. Vol 1. Bioleaching Processes. Warrendale, PA:The
Minerals, Metals, and Materials Society, 1993:357-361.
52. Tzeferis PG. Mineral leaching of low-grade laterite ores using "bioacids" by mo-
lasses fungal metabolism. Trans Soc Min Metall Explor 1994; 296:125-129.
53. Sukla LB, Panchanadikar VV. Bioleaching of lateritic nickel ore using an indig-
enous microflora. In: Torma AE, Wey JE, Lakshmanan VI, eds. Biohydrometal-
lurgical Technologies. Vol 1. Bioleaching Processes. Warrendale, PA: The Miner-
als, Metals, and Materials Society, 1993:373-380.
54. Sukla LB, Panchanadikar VV, Kar RN. Microbial leaching of lateritic nickel ore.
World J Microbiol Biotechnol1993; 9:255-257.
55. Tungkaviveshkul T, Thiravetyan P, Tanticharoen M. Mechanism for bioleaching
of zinc silicate residue by organic acid-producing microorganisms. In: Vargas T,
Jerez CA, Wiertz JV et al, eds. Biohydrometallurgical Processing. Vol 1. Santiago:
University of Chile, 1995:385-393.
56. Wenzl R, Burgstaller W, Schinner F. Extraction of zinc, copper, and lead from
filter dust by yeasts. Biorecovery 1990; 2:1-13.
57. Gascoyne DJ, Connor JA, Bull TA. Capacity of siderophore-producing alkalophilic
bacteria to accumulate iron, gallium and aluminum. Appl Microbiol Biotechnol
1991; 36:136-141.
58. Ghiorse WC. Microbial reduction of manganese and iron. In: Zehnder, AJ, ed.
Biology of Anaerobic Microorganisms. New York: Wiley, 1988:305-331.
59. Lovley DR. Dissimilatory metal reduction. Annu Rev Microbiol1993; 47:263-291.
60. Nealson KH, Saffarini D. Iron and manganese in anaerobic respiration: Environ-
mental significance, physiology and regulation. Annu Rev Microbiol 48:311-343.
61. Noble EG, Baglin EG, Lampshire DL et al. Bioleaching of manganese from ores
using heterotrophic microorganisms. In: Smith RW, Misra M, eds. Mineral
Bioprocessing. Warrrendale, PA:The Minerals, Metals, and Materials Society,
1991:233-245.
62. Baglin EG, Noble EG, Lampshire DL et al. Solubilization of manganese from ores
by heterotrophic micro-organisms. Hydrometallurgy 1992; 29:131-144.
63. Toro L, Veglio F, Terreri M et al. Manganese bioleaching from pyrolusite: Bacte-
rial properties reliable for the process. FEMS Microbiol Rev 1993; 11:103-108.
64. Serebrjanaja M, Yakhontova L, Petrova, 1. Bacterial transformation of manga-
nese oxide. In: Torma AE, Wey JE, Lakshmanan VI, eds. Biohydrometallurgy. Vol
I, Bioleaching processes. Warrendale, PA: The Minerals, Metals, and Materials
Society, 1993:277-284.
65. Madgwick Je. Bio-Ieaching of manganese dioxide tailings. In: Torma AE, Wey JE,
Lakshmanan VI, eds. Biohydrometallurgical Technologies. Vol 1, Bioleaching pro-
cesses. Warrendale, PA: The Minerals, Metals, and Materials Society, 1993:343-355.
66. Pahlmann, JE, Khalafalla SE. Leaching of domestic manganese ores with dissolved
S02' U.S. Bureau of Mines, Rept Inv 1988: RI 9150.
67. Pahlmann, JE, Rhoades CA, Chamberlain PG. Dual leaching method for recover-
ing silver and manganese from domestic manganiferous silver deposits. U.S. Bu-
reau of Mines, Rept Inv 1988: RI 9126.
68. Gupta A, Ehrlich H1. Selective and non-selective bioleaching of manganese from
a manganese-containing silver ore. J Biotechnol 1989; 9:287-304.
69. Rusin PA, Sharp JE, Arnold RG et al. Enhanced recovery of manganese and silver
from refractory ore. In: Smith RW, Misra M, eds. Mineral Bioprocessing.
70. Rusin P, Sharp J, Arnold R et al. Enhanced extraction of silver and other metals
from refractory oxide ores through bioreduction. Mining Eng 1992; 45:1467-1471.
71. Rusin P, Ehrlich H. 1995. Developments in microbial leaching-Mechanisms of
manganese solubilization. Adv Biochem Eng/Biotechnol 1995; 52:2-26.
72. Karavaiko GI, Avakyan ZA, Krutsko VS et al. Microbiological investigations on a
spodumene deposit. Mikrobiologiya 1979; 48:502-508 (Engl. transl. pp. 393-398).
73. Karavaiko GI, Krutsko VS, Mel'nikova EO et al. Role of microorganisms in spo-
dumene degradation. Mikrobiologiya 1980; 49:547-551 (Engl. transl. pp. 402-406).
74. Avakyan ZA, Karavaiko GI, Mel'nikova EO et al. Role of microscopic fungi in
weathering of rocks and minerals from a pegmatite deposit. Mikrobiologiya 1981;
50:156-162 (Engl. trans!' pp. 115-120).
75. Ilgar E, Guay R, Torma AE. Kinetics of lithium extraction from spodumene by
organic acids produced by Aspergillus niger. In: Torma AE, Wey JE, Lakshmanan
VI, eds. Biohydrometallurgical Technologies. Vol I. Bioleaching Processes.
76. Valeton, I. Bauxites. Amsterdam: Elsevier Publishing Company, 1972.
15 0 Biomining: Theory, Microbes and Industrial Processes
77. Ogurtsova LV, Karavaiko GI, Avakyan ZA et a!. Activity of various microorgan-
isms in extracting elements from bauxite. Mikrobiologiya 1989; 58:956-962 (Eng!.
trans!' pp. 774-780.
78. Groudev SN. Use of heterotrophic microorganisms in mineral biotechnology. Acta
Biotechnol 1987; 7:299-306.
79. Groudeva VI, Groudev, SN. Dressing of bauxite ores by means of microorgan-
isms. XVI International Mineral Processing Congress, June 5-10, 1988, Stockholm,
Sweden.
80. Karavaiko GI, Avakyan ZA, Ogurtsova LV et a!. Microbiological processing of
bauxite. In: Salley J., McCready RGL, Wichlacz PL, eds., Biohydrometallurgy. Ot-
tawa, Canada: CANMET SP89-10, 1989:93-102.
81. Ehrlich HL, Wickert LM, Noteboom D et al. Weathering of pisolitic bauxite by
heterotrophic bacteria. In: Vargas T, Jerez CA, Wiertz JV et aI, eds. Biohydro-
metallurgical Processing. Vol 1. Santiago, University of Chile, 1995:395-403.
SECTION III
Process Fundamentals
CHAPTER 8
Recent Developments in Modeling

the Kinetics of Bioleaching
Geoffrey S. Hansford
Introduction
A s the bioleaching of sulfide minerals becomes increasingly applied by the min-

ing industry for the pretreatment of refractory gold ores 1,2 and for the
bioleaching of base metals, there is a growing need for mechanistically-based ki-
netic models for the design, operation, optimization and control of bioleaching
processes. This chapter will review kinetic models for bioleaching as a whole, as
well as the important subprocesses of chemical ferric leaching of sulfide minerals,
bacterial ferrous iron oxidation and the bacterial oxidation of the sulfur moiety.
As models published prior to 1990 have been reviewed elsewhere,3 this chapter
will concentrate on more recent developments in the elucidation of bioleaching
mechanisms and modeling the kinetics.
Two mechanisms have been proposed for the bioleaching of sulfide minerals,
viz., the direct and the indirect mechanisms. In the direct mechanism the bacteria
are assumed to attach to the sulfide mineral surface and directly leach the mineral
by means of some enzymatic processes. The details of these have not been eluci-
dated. In the indirect mechanism the primary attack of the sulfide mineral is as-
sumed to be a chemical ferric leach with the role of the bacteria, whether attached
or not, being to oxidize ferrous iron to the ferric form and maintain a high redox
potential and also to oxidize the sulfur products of the ferric leach to sulfate. Most
recently, Sand and coworkers 4 and Boon and Heijnen5 have reviewed the experi-
mental evidence for these mechanisms. It is generally accepted that whether at-
tached or planktonic microorganisms are involved, that bioleaching occurs via a
primary chemical ferric leach with the role of the microorganisms being to oxi-
dize the iron and sulfur products. The process is regarded as involving a number
of chemical and microbial subprocesses. 406
In addition to models based on the direct or indirect mechanism, some work-
ers have successfully used purely empirical models to predict bioleaching kinetics.
Bacterial Ferrous Iron Oxidation

It has been shown that the bacterial oxidation of ferrous iron is an important
subprocess in the bioleaching of pyrite7,8 and it is believed to be rate-controlling in
the bioleaching of all sulfide minerals. 9

The kinetics of bacterial ferrous iron oxidation were first measured by Lacey
and Lawson'o at several total initial iron concentrations in batch culture and fitted
to the Monod equation. They observed that the specific growth rate decreased at
increasing iron concentrations and that the specific growth rate was optimum at
31 cC. Using a rate equation of the Monod form:
= fx = Il max e s
Il K + 8.1
ex s es
to describe bacterial growth where bacterial concentration was determined by plate
and microscope counts. Substrate utilization was considered to be directly related
to growth by a yield constant, neglecting maintenance:
- rs
= = 8.2
ex
MacDonald and Clark" also used Monod kinetics to describe bacterial ferrous
oxidation but concluded that the bacteria appeared to be more active in continu-
ous than batch culture. They pointed out the inaccuracies introduced if wall growth
was not eliminated. In continuous cultures the value of Ks was lower while Jlmax
was higher than in batch.
The influence of high total iron concentrations on the kinetics were reported
by Guay et al." They did not consider either maintenance or ferric iron inhibition,
but report lower values of Jlmax with increasing total iron concentrations.
In an extensive study by Kelly and Jones'3 in batch culture and using respiro-
metric measurements, they found different kinetic constants for different ranges
of iron concentration, and because of the assumption of nongrowing cells in
respirometry experiments, the kinetic data were not related to the batch data. Jones
and Kelly,'4 presented evidence for both competitive inhibition by ferric iron de-
pending on the operating conditions, with different values for the yield y sx max and
a maintenance coefficient, ms
1 + ~+~[Fe3+]
8·3
[Fe 2+ ] Kj [Fe 2+ ]
Braddock et al'5 introduced the concept of a threshold ferrous iron concentra-

tion below which growth did not occur.
=
Ks
I + _...."...-_"'::'-..".-__
[Fe 2+] - [Fe 2 + ]thres
Several other studies have been published on bacterial ferrous oxidation by

Thiobacillus ferrooxidans and Leptospirillum ferrooxidans.,6-2 3 These describe the
kinetics in terms of product inhibition and include the effect of pH, total iron,and
bacterial concentration on the kinetics as well as including bacterial decay terms.
There have been no reports which quantified the effect of maintenance according
to the Pirt equation!4 A more comprehensive review of these together with a com-
parison of the rate equations and kinetic constants has been presented by Boon. 6
Recent Developments in Modeling the Kinetics of Bioleaching 155
With the exception of Jones and Kelly4 and Liu et al,t9 simultaneous measure-
ments of iron oxidation rates, oxygen consumption rates and bacterial growth rates
have not been reported.
Recent doctoral theses have reported attempts to model ferrous iron kinetics
using mechanistically-based models and extend the range of operating conditions
studied so as to reconcile many of the discrepancies previously reported. Huberts 25
has used the chemiosmotic theory of Ingled~6 together with electrochemical
theory and Monod kinetics to develop a kinetic model for the bacterial oxidation
of ferrous iron which is based on redox potential. Assuming the electron transport
mechanism proposed by Ingledew, Huberts developed an equivalent electrical cir-
cuit of the biochemical cell accounting for the flow of electrons and protons from
the bulk medium through the cell membrane into the cytoplasm. The formation of
NADPH necessary for carbon dioxide fixation is included. The potentials are set
up due to the following six steps required for the formation of ATP, viz.; the oxida-
tion of ferrous iron outside the cell, the transfer of electrons via the electron trans-
port chain into the cytoplasm, the transport of protons through the membrane
with the accompanying production of ATP, the diffusion of oxygen into the cell,
the reaction of oxygen with protons to form water and the excretion of water from
the cell.
The theoretical free energy changes during these steps was then determined
for the formation of one ATP from the number of ferrous ions required. He pro-
posed ten linked steps which are required for the formation of NADPH. It was
found from his experimental measurements, that the free energy changes are lower
than those calculated by Ingledew using the highest possible changes for ATP and
NADPH formation.
The steps which make up the bacterial oxidation of ferrous iron involve the
transfer of charged species across potential differences. Huberts used standard
electrochemical theory to derive kinetic expressions for these steps, assuming each
of the steps could be rate-limiting in turn. The expressions were tested against
experimental data and a reasonable fit was obtained over the Eh range 800 to 900
m V. These tests showed that it is possible to model the kinetic data using redox
potential as a measure of ferric and ferrous iron concentrations and dissolved oxy-
gen concentration as the major variables and that pH also has an effect. However,
the rate expressions used are rather complex. The approach, however, has the po-
tential of being extended to develop a kinetic model incorporating the fundamen-
tal microbial processes involved in ferrous iron metabolism and based on electro-
chemical and chemiosmotic theory.
Huberts also used Michaelis-Menten kinetics to model the rate incorporating
.ffect of bO~ w.'(Olv<: oxw)n[ODd fi:mc/:~UB-iron "']tiO'

rFe ,· a 1 kB +o~o. Fe 2+ + K (1+ Fe 3+) 8.5
Fe'· K'
with constants as follows:
at = llFeH ions/Jls/cell
Kb = 0.022 atm
KPe2+ = 0.0012 giL
K' 0.57 giL
This model can be written in the form:
= q:;+ [°2 ]
fpe 2 +
1 + ~ + Ks [Fe 3+] K02 + [0 2 ] 8.6
Fe 2+ Ki [Fe 2+]
Nemati and Webb 27 have developed a kinetic model for the biological oxida-
tion of ferrous iron by T. ferrooxidans using the initial rate method and measuring
ferrous iron concentration by changes in redox potential. They have used an ap-
proach based on Michaelis-Menten kinetics and included the effects of tempera-
ture, ferrous iron and bacterial concentration as well as substrate and cell inhibi-
tion. The model is of the form:
- r 2+
= K c [Fe 2+]
0 X
E
exp( _ a)
Fe ~ (1 + ~: J+ [Fe 2+] + (1 - c~X)( [Fe~+ F J RT 8.7
The effect of ferric iron on the kinetics has not been investigated. The inhibi-
tion by high concentrations of bacteria is similar to that reported by Lizama and
SuzukV3 and explained in terms of inhibition of the active sites on the primary
iron-oxidizing enzyme of bacteria. The possible effect on the kinetics by carbon
dioxide transfer at high bacterial concentrations has not been considered. Bacte-
rial concentrations are reported in terms of cell number.
The coefficients of the overall kinetic equation for bacterial ferrous oxidation
are given as:
Coefficient Value Value

K" 114.96 kmo1!m 3/h/(cells/mL) 6438 kg/m3/h/( cells/mL)
Ea 68.4 kJlmol
Km' 1.2 x 10-3kmol/m 3 .0672kg/ m3
K/ 2.68 x 107 cell/mL
p 7.8 x 10 8 cell/mL
a 0.466 kmol/m 3 26.1 kg/m3
This equation and the data gave the same shape curve of qFe2+max versus redox
potential as that of Boon5 for T. ferrooxidans and of van Scherpenzeel, Hansford,
Boon et al, (unpublished data, 1997) for a Leptospirillum-like culture.
Boon6 has investigated the bacterial oxidation of ferrous iron in batch and con-
tinuous cultures with a pure strain of T. ferrooxidans (Boon, Thone, Heijnen et aI,
unpublished data, 1997) and in continuous culture with Leptospirillum-like bacte-
ria (van Scherpenzee1, Hansford, Boon et aI, unpublished data, 1997). The growth,
concentration and activity of the bacteria were determined using off-gas analysis
to measure the uptake of carbon dioxide and oxygen. They determined the kinet-
ics in terms of a specific oxygen utilization rate, q02 and included both growth and
maintenance in terms of the Pirt equation. 24
8.8
And they described the specific oxygen utilization kinetics using a rate equation
incorporating both ferric iron inhibition from Kelly and Jones 14 and a threshold
value of ferrous iron from Braddock et aps
q';,'
+ Ks [Fe 3+]
K; [Fe 2+]- [Fe 2+ ]thre.
and an equation for the specific growth rate:
m +m o ymax
/..l = K
max
+K.
ox
[Fe 3+]
1+ s 8.10
[Fe 2+ ] - [Fe 2+]thres K; [Fe 2+] - [Fe 2+] thres
Using off-gas analysis for the measurement of oxygen and carbon dioxide uti-
lization, the kinetics offerrous iron utilization by T. ferrooxidans and Leptospirillum-
like bacteria have been determined in continuous culture at 30°C and pH 1.6.6•a8
The approach used assumes that the oxidation of ferrous iron takes place accord-
ing to the following stoichiometry:
4Fe 2+ + O 2 + 4H+ = 8.11
This is accompanied by the growth of bacteria which are assumed to have a

stoichiometric formula, CH1.80 0.sNo,a and use COl and NH4+ as sole carbon and
nitrogen sources, respectively. The stoichiometric formula is the general one sug-
gested by Roels 19 and agrees with that of CH1.7800.S3NO.17 measured for T.
ferrooxidans by Jones and Kelly.14
8.12
In this way the rate of bacterial growth, rx,is equal to the rate of carbon dioxide
utilization, -reo l :
= 8.13
By performing an elemental and charge balance, or a "degree-of-reduction"

balance over Equations 8.12 and 8.13 it is possible to relate the rate of ferrous iron
oxidation, -rFes2 , to the rates of oxygen, -ro 2 , and carbon dioxide, utilization, -reOl'
as:
-r
Fe
2. = - 4r 0 2
- 4.2r co
2
8.14
Introducing a yield, Y sx> defmed as the moles of biomass produced per mole of
substrate, in this case ferrous iron, utilized:
=
8.15
Equations 8.12 and 8.13 can be combined into a single balanced equation for
bacterial growth on pyrite as substrate:
CO 2 + O.2NH; + [I-42Ysxx ]0 2+ _I_Fe 2+ + [_I_- O.2 ]H+

4Ysx Ysx Ysx
=
8.16
It is frequently convenient to make use of a yield based on oxygen, Yox, defined

as the moles of biomass produced per mole of oxygen utilized:
= 8.17
This can be related to the yield on substrate by means of Equation 8.4 to give:
I = 1- 42Ysx
Yox 4Ysx 8.18
Details of these have been given elsewhere.6

The rates of oxygen and carbon dioxide utilization are measured by means of
the off-gas analyzers. In continuous culture at steady state, a biomass balance gives:
D.c x = rx - rC02 8.19
so that;
- rco z
cx = 8.20
--0-
It is convenient to consider the substrate utilization providing energy for both
biomass synthesis and maintenance. This can be expressed in terms of the Pirt
equation:
rx r
- rs = = _x_+ cxffi
Ysx ymax s 8.21
sx
or
_ rs = I = I + ms
rx Ysx Y;;X Ii1 8.22
similarly for oxygen utilization:
-r = rx = -rx- + cxmo
Oz y max 8.23
Yox ox
and
_ roz = I = 1 + mo
~ Yox Y:," Ii1 8.24
Equations 8.21 and 8.23 can be used to determine the maximum yields and
maintenance coefficients from the rates of ferrous iron, oxygen and carbon diox-
ide utilization, as shown in Figure 8.1.
180
~ 170
0
'0 160
!
::::.
150 =
+ 140
('oj
:. 130
o 120
E
110
>-= 100
90
80
0 20 40 60 80 100 120
110 (h)
I • Data Equation 22 1
Fig. 8.1. Evaluation of yield and maintenance coefficient from Equation 8.22.
The kinetic equations are written in terms of biomass and substrate-specific

rates defined as the biomass specific substrate utilization rate, q.. and the specific
oxygen consumption rate, q02'
= 8.25
= 8.26
The specific growth rate is then measured by:

= rx = -rco,
~ ~
ex e x
and related to the specific substrate and oxygen utilization rates by Equations 8.23
and 8.25 as:
~ = qs~X -msY;:' 8.28
so that:
~ ma, = q';"~' - ms Ys~ 8.29
similarly
~ = qo , Y~' -mo Y~ 8·30
and:
~ma, = q~~ -mo Yo":' 8.3 1
The rate equation for the specific rate of oxygen utilization can be approxi-
mated by a simpler form by neglecting the ferrous iron saturation term and the
threshold ferrous iron concentration:
This simplified form has been found to give a satisfactory fit to bacterial fer-
rous oxidation data9•28 and assumes that the kinetics are determined by the ferric/
ferrous-iron ratio or redox potential according to the chemiosmotic theory pro-
posed by Inglede~6 for the growth of T. ferrooxidans on ferrous iron. The kinetics
may also be written in terms of the specific ferrous iron utilization rate:
q~,
1+K[Fe 3+]
[Fe 2+]
From the results of Boon6 and van ScherpenzeeJ28 shown in Figure 8.2, the fol-
lowing values of the kinetic parameters have been obtained, as shown in Table
8.1.
The models presented by Huberts25 and Nemati and Webb27 are in terms of
numbers of bacteria whereas the model presented by Boon6 is in terms of biomass
carbon. Without further information regarding the carbon content per bacterial
cell under the conditions used, it is therefore not possible to compare the actual
values of the parameters used. However, all three models give reverse sigmoid curves
when the specific rate is plotted against ferric/ferrous iron-ratio or redox poten-
tial. This dependence of the bacterial ferrous iron oxidation kinetics on redox po-
tential is in agreement with the predictions of the chemiosmotic theory proposed
by Ingled~6 and will be used later in this chapter in the development of a mecha-
nistic model for sulfide mineral bioleaching.
Kinetics Models for Bioleaching
Empirical Models Based on the Logistic Equation

Successful modeling of bioleach kinetics has been achieved by the use of an
empirical approach with kinetics based on the logistic equation. Pinches et apo
showed that the logistic equation could be used to predict the rate of bioleaching
of a refractory gold-bearing pyrite-arsenopyrite concentrate in continuous
bioreactors. Hansford and coworkers31-35 used the logistic equation to model the
kinetics of both batch and continuous bioleaching of pyrite and pyrite-arsenopy-
rite concentrates. Miller and Hansford35 showed that the logistic equation could be
used to model the kinetics of both iron and arsenopyrite bioleaching from an arse-
nopyrite concentrate and that parameters obtained in batch culture were in rea-
sonable agreement with those from continuous culture. Furthermore, they were
able to model conversions obtained in the bioreactors of a two-stage pilot plant.
Recent Developments in Modeling the Kinetics of Bio/eaching 161
2.5 r---------------------------...,
••
~ 1.5
-0
E
;;;:
Q
!
8 •
<T
0.5
•• .00 1000 10000
IFo3+I'IFa2+]
• Thiobaclilus ferrooxidans- EqL8t1on8.32 • Leptospirillum ferrooxidllns- Equatione.32
Fig. 8.2. Specific oxygen utilization rate for T. ferrooxidans and L. ferrooxidans as a
function of the ferric/ferrous iron ratio.
Table 8.1. Kinetic parameters for ferrous iron oxidation by Thiobacillus

ferrooxidans and Leptospirillum ferrooxidans
Thiobacillus Leptospirillum Units

ferrooxidans ferrooxidans
q max molO 2/molCIh

02 2.2 1.7
qFe Hmax 8.8 6.8 moIFe 2+/moIC/h
K02 0.05 0.0005
KFe H 0.05 0.0005
y max 0.051 0.045 moIC/moIO.
ox
y max molC/molFe H
sx 0.0114 0.0099
ffio 0.1 0.026 molO 2/molCIh
ms 0.25 0 ·45 molFe H /molCIh
The logistic equation is written for the rate of conversion of sulfide mineral,
where conversion is defined as the fraction of feed sulfide mineral leached. In a
batch bioleach system this is:
dX = k X(I-
dt m
~)
X 8·34
max
For batch bioleaching, the fraction bioleached with time is given by integrating
Equation 8.34 to give:
X(t) =
X(O)
1- X-(I- exp( kmt)) 8·35
max
which can be written in logarithmic form more convenient for parameter estima-
tion.
k mt + In( XmaxX(O)- X(O) ) 8·36
The fraction leached for a single stage bioleach reactor at steady state is given
by:
8·37
More complicated expressions can be derived for the fraction leached in a se-
ries ofbioreactors. 33
Hansford and Chapman 31 found that the logistic equation could be used to suc-
cessfully fit batch pyrite bioleach data, as shown in Figure 8.3 but found different
values for the parameters for different size fractions. Miller and Hansford3' found
that the logistic equation gave a good fit to both iron and arsenic dissolution in
batch bioleaching of an arsenopyrite-pyrite concentrate (Fig. 8.4). Similarly, they
found that Equation 8.37 gave a good fit to arsenopyrite-pyrite bioleach data ob-
tained from a pilot-scale four bioreactor cascade (Fig. 8.5).
Crundwe1l36 has used a population balance37 to account for the size distribu-
tion of the mineral slurry and changes occurring due to leaching via a shrinking
particle process. This has given an improved fit to the data of Hansford and
Chapman31 and Miller and Hansford33, particularly with regard to the values of the
parameters for the different size fractions. This approach has proved successful
using a shrinking particle mechanism in spite of the fact that bioleaching of pyrite
occurs by the formation of pores in the mineral particles rather than uniform
shrinkage of the surface as shown by Hansford and Drossou.38
Models Based on Attached Microorganisms

The direct mechanism has been used by several workers to develop models for
the bioleaching of pyrite. 39>40 In a very similar approach, Konishi and coworkers 42-44
have developed a model assuming that the attached microorganisms directly leach
the sulfide mineral. They have shown that the equations developed successfully fit
data for the batch bioleaching of pyrite 4•• 43 and zinc sulfide concentrate. 44 Their
Recent Developments in Modeling the Kinetics ofBioleaching
0 .9
•
0 .8
0 .7
i
'0 0 .6
;;
o
~
:E 0 .5
'c:"
II)
o
tl 0.4
...
~
0 .3
0 .2
0. 1
10 15 20 25 30
Tlm o (doyo)
• +53·75 microns-logis tic fit • +3&-53 microns - logistic • -38 ml(:(ons - Iog r'tlc III
Fig. 8.3. Batch bioleaching of three different size fractions of a pyrite concentrate.
model assumes a rapid adsorption of the bacteria to the sulfide mineral surface
according to a Langmuir adsorption isotherm. For the case of pyrite bioleaching
they use the erroneous assumption without giving evidence that no chemical fer-
ric leaching of pyrite takes place and develop a model for bioleaching due to the
growth of attached bacteria. The kinetics of bacterial growth are assumed to be
proportional to the concentration of attached bacteria and the area of unoccupied
mineral surface area. For the case of zinc concentrate bioleaching they include the
ferric leaching step according to first order kinetics. No use is made of redox po-
tential kinetics nor ferrous iron concentration kinetics for bacterial growth. How-
ever, their model was able to fit their batch experimental data for both pyrite and
zinc concentrate.
Nagpal et al45 have studied the bioleaching of an arsenopyrite-pyrite concen-
trate in a continuous bioreactor in which the oxygen and carbon dioxide utiliza-
tion rates were measured. They assumed that the pyrite and arsenopyrite leaching
occurred solely by chemical ferric iron oxidation of the minerals with rate de-
pendent on the difference between the redox potential of the leach solution and
the rest potentials of the minerals. 46 Their model assumed that bacterial growth
rates were directly proportional to ferrous oxidation rates and they used Monod
kinetics modified to include inhibition by arsenic in solution. They assumed that
both attached and nonattached bacteria contributed to the oxidation of ferrous
iron. They did not include the bacterial oxidation of sulfur species in their model.
0 .9
'a 0 .8
•
.c
:= 0 . 7
•
'0
m0 .6
; 0 .5
c
i 0 .4
c
o
~ 0 .3
I!!
I&. 0 .2
0 .1
oL-~-=~~~ __ ~ ____ ~ ________ ~ __-J
o 2 4 8 8 10 12 14
Time (deys)
- --
• Arsenopyrite _ Pyr ite FeAsS logistic iii - - FeS2 logistic flI]
Fig. 8.4. Batch bioleaching of an arsenopyrite-pyrite concentrate.
Bacterial growth is directly proportional to ferrous iron oxidation:
with the rate of ferrous iron oxidation as:
1 + ~+ K, [As(VW
[Fe 2+] Kj [Fe 2+]
and the rates of ferric leaching of pyrite and arsenopyrite:
and
-fpeA,s = ~~,s «PeA,s [FeAsS ] (Eh - EpeA,s )
respectively.
This model was not a particularly good prediction of the total dissolved iron
concentrations but gave good prediction of dissolved arsenic concentration. The
trends of dissolved iron and arsenic against dilution rate in a continuous bioleach
Recent Developments in Modeling the Kinetics of Bioleaching
I::
0 .9 •
0 .8
.,
"U
&.
0 .7
.,'<>"
"0 0 .6
iii
GI 0 .5
II
"U
:::::0
III
0 .4
c:
0
.~ 0 .3
!!
II..
0.2
0.1
0
0 2 4 6 8 10 12 14
Residence Time (days)
• Stage One • Stage Two Stage Three Stage Four - - Log istic fit
Fig. 8.5. The performance of a four-stage continuous bioleach reactor cascade treating
an arsenopyrite-pyrite concentrate.
reactor are predicted quite well. The model also gave good predictions to transient
data obtained by removal of carbon dioxide supplementation and pulse additions
of arsenic and ferrous sulfate.
Boon6 used off-gas analysis of carbon dioxide and oxygen to measure the rate
of bacterial growth and pyrite oxidation in batch cultures of an enrichment cul-
ture of Leptospirillum-like bacteria to which pyrite was added every four hours.
The redox potential dropped with each addition and the rate of oxygen utilization
rose almost immediately to a new level. Using the rates of oxygen and carbon diox-
ide and a degree of reduction balance, it was possible to calculate the rate of pyrite
leaching and therefore the bacterial and pyrite concentrations with time. Total
carbon analysis confirmed the bacterial concentration reported as moles carbon.
From these the specific oxygen rate was determined with respect to bacteria and
pyrite and plotted against redox potential (or ferric/ferrous-iron ratio). Details of
the methodology and results are reported elsewhere. 47•48 Boon6 used Equation 8.8
and the following equation to describe the bacterial and pyrite-specific oxygen
utilization rates, respectively:
-r 0 , nmax
n = =
0,
0,
[peS 2] 1+B [Fe 2+] 8.42
[Fe 3+ ]
0 .8 . . . - - - - - - - - - - - - - - - - - - - - ------,
0 .7
0 .6
:2
205
o
E
N
Q 0 .4
o
S ....
S 0 .3
C"" •
0 .2
0.1
o L-________________________________________~
1 00 1000 1 0000 100 000 1000000

[Fe3 + )/ [Fe2 +)
• Ferrous Iron .. Pyrite -Equation 32
Fig. 8.6. Specific oxygen utilization rates of Leptospirillum-like bacteria growing on

ferrous iron and pyrite, respectively.
The form of Equation 8.42 was chosen because it was found by Boon6 to fit the
data for pyrite specific oxygen consumption rate over the range of experimental
redox potentials obtained.
Taking bacteria from the pyrite batch, Boon 6 measured the specific oxygen uti-
lization rate in a respirometer when these bacteria were oxidizing ferrous iron in
the absence of any pyrite. Both oxygen utilization rate and redox potential were
measured simultaneously. It was found that the specific rate of oxygen utilization,
qo, was the same at the same redox potential as shown in Figure 8.6. From this
Boon et al7 concluded that the primary substrate used by the bacteria was ferrous
and proposed a two-step mechanism for the bioleaching of pyrite by Leptospirillum-
like bacteria.
A Two Subprocess Mechanism for the Bioleaching of Pyrite

In this mechanism the two subprocesses of bacterial ferrous iron oxidation
and chemical ferric leaching of the sulfide mineral are linked at pseudo steady
state by equating the rate of ferrous iron production from the chemical ferric leach
reaction to the rate of consumption of ferrous iron consumption by the bacteria.9
In order to do this the kinetics of the two subprocesses must be rewritten for fer-
rous iron production and utilization using the stoichiometry of Equation 8.12 and
the overall bioleaching of pyrite, Equation 8.43 below:
4Fe 3+ + 8S0~' + 4W
Equations 8.42 and 8.34 become:

-rFe 2+
[Fe _
+B _ 2+]
[Fe 3+]
where the quantity ~ is introduced to base the kinetics on the surface area of the
sulfide mineral. In this way the concentration, size and surface roughness of the
mineral can be taken into account and:
-rFe 2+ q::"
[Fe 3+]
1 +K _ _
[Fe 2+]
so that at a particular pyrite and bacterial concentration the pseudo-steady state

will be defined by:
s;:', (X. [FeS 21 q max

Fe2+
C
X
fPe ,chern -f 2+
[Fe 2+] 2+ Fe ,bact [Fe 3+]
l+B-- l+K--
[Fe 3+] [Fe 2+]
where it is possible to express the ferric/ferrous-iron ratio in terms of the redox

potential using the Nernst equation as:
~[Fe 2+]
= ex
P
rEh _EO]
RT
nF
Figure 8.7 shows the rates of ferrous iron generation by ferric leaching of pyrite
and ferrous consumption by Leptospirillum ferrooxidans and Thiobacillus
ferrooxidans. The curves are plotted for a concentration of 10 gIL of +53-75 Jlm
pyrite and for bacterial concentrations of 150 mgC/L at a total iron concentration
of 12 gIL. The point of intersection of the curves represents the pseudo-steady state
giving the rate of ferrous iron turnover and the redox potential. It can be seen that
for the bioleaching of pyrite the ferric leach curve intersects the bacterial ferrous
oxidation curve of L. ferrooxidans at a higher rate of ferrous turnover than T.
ferrooxidans and therefore be L. ferrooxidans will be the dominant species. This
has been confirmed by Rawlings 49 who has found that L. ferrooxidans predomi-
nates in the bioreactors of the GENCOR BIOX· process treating an arsenopyrite-
pyrite concentrate. Similar observations have been made by Brierley (personal com-
munication, 1995) for the Newmont Bioheap leaching operations in Nevada.
The point of intersection of the chemical and bacterial curves, defines the
pseudo-steady state redox potential and the rate of ferrous iron turnover, which
can be related stoichiometrically to the rate of pyrite bioleaching. The intersection
point depends on both the concentration of bacteria and active surface area con-
centration of the pyrite. In this way the model presented here can be related to
those which are based on a bacteria-to-mineral ratio, Cx /[FeS 2 1. As the bacterial
concentration and/or the surface area concentration change, the redox potential
and overall pyrite bioleaching rate will change accordingly.
L.pto.pirillum
fefrooxld.".
~------~~-------
N
~
:::IE
!.
('11--------------
+
.50 700 750 000 .50 0.0 . .0 1000
Eh [mV(SHEll
Fig. 8.7. Selection of L. ferrooxidans over T. ferrooxidans based on the two subprocess
mechanism with pyrite as mineral substrate.
In the bioleaching of sulfide minerals which have lower rest potentials, the fer-
ric leach curve will intersect the bacterial curves at a lower redox potential where
the ferrous iron oxidation rate of T. ferrooxidans may be higher than that of L.
ferrooxidans and then it will be the dominant organism, (Fig. 8.8), which is what is
reported for copper bioleaching from chalcocite (Rossi MC, personal communica-
tion,1995 and Garcia C, personal communication, 1995).
According to this two-step mechanism for sulfide mineral bioleaching it is pos-
sible to determine the kinetics of the chemical and bacterial subprocesses inde-
pendently and then use the kinetic constants so derived to predict both the steady
state and dynamic performance of bioleach systems. Recent work on the purely
chemical ferric leaching of pyrite by Ralph et al50 has shown that the values for
~Fe2+ max and B agree with those obtained from the data of Boon et al,l for the
bioleaching of pyrite. The dependence of the bacterial kinetics on redox potential
is consistent with the chemiosmotic theory of Ingledew,26 while the dependence of
the ferric leach kinetics on redox potential is in accord with electrochemical theory.
May, Ralph and Hansford (unpublished data, 1997) have shown that the experi-
mentally determined ferric leach rate of pyrite does not level off as predicted by
Equation 8.42, but appears to increase with increasing redox potential. This is what
would be expected if ferric leaching is assumed to be an electrochemical process
predicted by the Butler-Volmer equation.
= 1. ( exp
o
((i-a) Fll) (-a Fll))
RT
- exp -----
RT
Where the corrosion current can be related to the leach rate of the sulfide min-
eral. May (unpublished work, 1997) has shown that an equation of the form:
relating pyrite leach rate with redox potential gives a reasonable fit to experimen-
tally measured ferric leach rates of pyrite, Figure 8.9.
Ch a lcoc i te
o 650 700 750 800 850 900 950 1 000
Eh (mY)
Fig. 8.8. Selection of T. ferrooxidans in the bioleaching of copper from chalcocite.
0 .000014
0 .0000 12
0 .00001
-;r
::.
0 0 .000008
.E
'0
E
~ 0 .00 0 008
VJ
If
*;' 0 .000004
~
'!:
0 .000002
0
620 630 640 650 660 670 680
E (mY V8 Ag/AgCI)
• Experimental dala Buller-Volmer model I
Fig. 8.9. Prediction of the ferric leaching of pyrite based on the Butler-Volmer equation.
This two-step mechanism appears to be able to explain why L. ferrooxidans is

the dominant bacterial species in the bioleaching of pyrite and arsenopyrite, as
reported by Boon et al9 and Rawlings,49 and why arsenopyrite is preferentially
bioleached ahead of pyrite, as reported by Miller and HansfordY It also could be
the reason for T. ferrooxidans being the predominant bacteria in the bioleaching
of copper from chalcocite.9 This hypothesis and approach, while yet to be exten-
sively tested and rigorously proved, shows great potential of becoming a very use-
ful way of modeling the kinetics of bioleaching with engineering and industrial
applications.
None of the above models have considered the sulfur moiety, which may in-
deed be critical in many cases. Sand and coworkers 4,5 1,53,54 have studied the mecha-
nism of pyrite bioleaching by T. ferrooxidans and L. ferrooxidans and by mixtures
of these with strains of Acidiphilium sp. They have also considered the fate of sul-
fur in bioleaching and presented a critical evaluation of the mechanism by which
bioleaching takes place. Sand et al51 have shown that L.ferrooxidans was as abun-
dant as T. ferrooxidans in a heap leaching operation. In agreement with the find-
ings of Norris,52 they have shown that while the maximum ferrous iron oxidizing
activity of L. ferrooxidans is about half that of T. ferrooxidans, it retains this activ-
ity at high ferric iron levels where T.ferrooxidans shows significant product inhi-
bition.53 L. ferrooxidans attaches much more rapidly and strongly to mineral sur-
faces and this attachment is facilitated by exopolymers with high ferric iron
content.54 By investigating the sulfur chemistry of bacterial leaching of pyrite, they
have shown that in the case of the bioleaching by L. ferrooxidans, an organism
lacking sulfur-oxidizing capability, sulfur, tetra- and pentathionate accumulated
at similar levels to those measured for sterile ferric leaching. However, only trace
amounts of elemental sulfur were detected in the case of T. ferrooxidans. They
have proposed a leaching mechanism based on the hypothesis that ferric ions
complexed in the exopolymer layer of the bacteria chemically leach the mineral.
This ferric-rich layer has been shown by Blake et a155 to lead to strong attachment
of T. ferrooxidans to pyrite. The exopolymer layer provides a reaction compart-
ment for the ferric leaching and bacterial oxidation of the sulfur compounds, which
occurs in the vicinity of the mineral surface. They argue that because of its depen-
dence on ferrous iron solely as a source of energy, L. ferrooxidans attaches more
strongly than T. ferrooxidans. The hypothesis is consistent with the findings of
Luther56 and Moses et a15 7 for the chemical ferric oxidation of pyrite and includes
the evidence of Sugio et a15 8 of the existence of a sulfite oxidase in T. ferrooxidans
which can utilize ferric iron. This proposed hypothesis has provided valuable evi-
dence for the mechanism by which the bioleaching of sulfide minerals by T.
ferrooxidans and 1. ferrooxidans takes place. 4It remains to develop kinetic models
based on these postulates. The fate of the sulfur moiety in the bioleaching of sul-
fide minerals has not been included in any published kinetic models even though
it is clearly an important factor in the case of base metal bioleaching. It may be that
sulfur products are the cause of passivation of the mineral surface and the bacteria
have been claimed to assist in removing this passivation layer.
Conclusions
Research has led to a kinetic model for sulfide mineral bioleaching based on
the turnover rates of iron between the chemical ferric leaching of the mineral and
the bacterial oxidation of the ferrous iron to the ferric form. The balance between
these two rates defines the rate and redox potential at which a bioleaching system
will operate and is dependent on the concentration of bacteria and the available
surface area of the mineral. Both these processes have been shown to be strongly
dependent on the redox potential of the leach solution.
There are a number of kinetic models available for the bacterial oxidation of
ferrous iron as a function of redox potential as a measure of ferric/ferrous iron-
ratio, total iron concentration and temperature. However, there is still no compre-
hensive model which defines the kinetics fully in terms of chemiosmotic and elec-
trochemical theory.
Similarly, for the chemical ferric leaching of sulfide minerals, in spite of much
published research not reviewed here, there is still no comprehensive kinetic model
based on electrochemical theory which takes into account all the factors affecting
such kinetics. With the absence of such models it is not surprising that there is still
a lack of reliable kinetic parameters for process design and performance predic-
tion. There remains a need to refine some of the potentially attractive methods for
determining these parameters. Also, the role of the sulfur moiety in determining
bioleach kinetics is still not completely understood.
List of Symbols
a, bacterial ferrous iron oxidation rate FeH ions/cell/Jls
[As(V)] concentration of As(V) mol/L
B kinetic constant in pyrite oxidation dimensionless
Cs concentration of growth limiting substrate mol/L
Cx concentration of bacteria mol/L
D dilution rate l/h
Ea activation energy kJlmol
EFeAsS rest potential of arsenopyrite mV
EFeS• rest potential of pyrite mV
Eh redox potential - standard hydrogen electrode mV
F Faraday constant coulomb/mol
[Fe H ] concentration of FeH mol/L
[FeH)thres threshold concentration of Fe H mollL
[Fe 3+) concentration of Fe3+ mol/L
[FeAsS) concentration of FeAsS mol/L
[FeS.) concentration of FeS. mol/L
corrosion current amps
corrosion current amps
kinetic constant dimensionless
K' ferric iron inhibition constant mol/L
K'i bacterial inhibition constant cells/mL
K'm saturation constant mol/m 3
kB saturation constant for oxygen atm
KFeH ferrous iron saturation constant mol/L
Ki ferric iron inhibition constant mol/L
km maximum rate constant in logistic equation l/h
Ko maximum rate constant (mol/m/h)/( cells/mL)
Ko. saturation constant for oxygen molO./L
Ks substrate saturation constant mol/L
mo maintenance coefficient on oxygen (molO.)/(molC)/h
ms maintenance coefficient on substrate (molS)/(molC)/h
n number of electrons involved in reaction

[0 2 1 oxygen concentration moltL
P0 2 oxygen partial pressure atm
qPe2+ bacterial specific ferrous iron oxidation rate molFe2+ l(moIC)/h
qPe2+max maximum bacterial specific ferrous iron
oxidation rate (moIFe2+)/(moIC)/h
q0 2 bacterial specific oxygen utilization rate (moI0 2)/(moIC)/h
qO2max maximum bacterial specific oxygen utilization rate (moI0 2)/(molC)/h
q. bacterial specific substrate utilization rate (moIS)/(moIC)/h
R universal gas constant kJ/moltK
re0 2 carbon dioxide production rate molC/L/h
rPe2+ ferrous iron production rate molFe 2+IL/h
rPeH.bact bacterial ferrous iron production rate molFe 2+IL/h
rpe2.+Jchem chemical ferrous iron production rate molFe2+ IL/h
r0 2 oxygen production rate moIO,!L/h
r. substrate production rate molS/L/h
r" bacterial production rate molC/L/h
T temperature K
t time h
X fraction of pyrite bioleached dimensionless
X(O) fraction of pyrite bioleached at zero time dimensionless
X(t) fraction of pyrite bioleached at time, t dimensionless
Xmax maximum fraction of pyrite bioleached dimensionless
Yox bacterial yield on oxygen (moIC)/(moI0 2)
yoxmax maximum bacterial yield on oxygen (molC)/(moI0 2)
y." bacterial yield on substrate (molC)/(moIS)
Y.xmax maximum bacterial yield on substrate (moIC)/(moIS)
IX ferrous iron concentration constant in Equation 8.48 moltL
OtPeAsS specific surface area of arsenopyrite m 2/mol
OtPeS2 specific surface area of pyrite m 2/mol
~ bacterial concentration constant in Equation 8.49 cells/mL
'1 over potential mV
Jl specific growth rate l/h
Jlmax maximum of specific growth rate 1/h
V0 2 pyrite specific oxygen utilization rate (moI0 2)/(moIFeS2)/h
V02.max maximum pyrite specific oxygen utilization rate (moI0 2)/(moIFeS.)/h
~F.2+ area specific ferrous iron production rate (moIFe H )/m 2/h
~peHmax maximum area specific ferrous iron utilization rate (moIFeH)/m'/h
References
1. Oew ow. Comparison of performance for continuous bio-oxidation of refractory
gold ore flotation concentrates. In: Vargas T, Jerez CA, Wiertz JV, Toledo H, eds.
Biohydrometailurgical Processing. Vol I. Santiago: University of Chile, 1995:239-251.
2. Brierley JA, Wan RY, Hill OL, Logan T. Biooxidation-heap pretreatment technol-
ogy for processing lower grade refractory gold ores. In: Vargas T, Jerez CA, Wiertz
IV, Toledo H, eds. Biohydrometailurgical Processing. Vol I. Santiago: University
of Chile, 1995:253-262.
3. Rossi G. Biohydrometallurgy. 1st ed. Hamburg:McGraw-Hill, 1990.
4. Schippers A, Josza P-G, Sand W. Sulfur chemistry in bacterial leaching of pyrite.
Appl Environ Microbiol 1996; (in press).
5. Boon M, Heijnen JJ. Mechanisms and rate limiting steps in bioleaching of sphaler-
ite, chalcopyrite and pyrite with Thiobacillus ferrooxidans. In: Torma AE, Wey
JE, Lakshmanan VI, eds. Biohydrometallurgical Technologies. Vol I Bioleaching
Processes. Warrendale PA: The Minerals, Metals and Materials Society,
1993:217-236.
6. Boon M. Theoretical and Experimental Methods in the Modelling of Bio-oxida-
tion Kinetics of Sulfide Minerals. PhD Thesis, Department of Biochemical Engi-
neering, Kluyver Laboratory for Biotechnology, Delft University of Technology
The Netherlands. 1996.
7. Boon M, Hansford GS, Heijnen JJ. The role of bacterial ferrous iron oxidation in
the bio-oxidation of pyrite. In: Vargas T, Jerez CA, Wiertz JV, Toledo H, eds.
Biohydrometallurgical Processing. Vol I. Santiago: University of Chile, 1995:153-181.
8. Sanheuza A, Escobar B, Vargas T. The leaching of pyrite in the presence and
absence of ferric iron and thiobacilli. Paper presented at the International
Biohydrometallurgy Symposium, Vitia del Mar, Chile, November 1995.
9. Boon M, Heijnen JJ, Hansford GS. Mechanism and kinetics of bioleaching sulfide
minerals. Appl Metall Rev 1996; in press.
10. Lacey DT, Lawson F. Kinetics of the liquid-phase oxidation of acid ferrous sul-
fate by the bacterium Thiobacillus ferrooxidans. Biotechnol Bioeng 1970;
12(1):29-50.
11. MacDonald DG, Clark RH. The oxidation of aqueous ferrous sulfate by
Thiobacillus ferrooxidans. Can J Chern Eng 1970; 48:669-676.
12. Guay R, Silver M, Torma AE. Ferrous iron oxidation and uranium extraction by
Thiobacillus ferrooxidans. Biotechnol Bioeng 1977; 19:727-740.
13. Kelly DP, Jones CA. Factors affecting metabolism and ferrous iron oxidation in
suspensions and batch cultures of Thiobacillus ferrooxidans: relevance to ferric
iron leach solution regeneration. In: Murr LE, Torma AE, Brierley JA, eds. Metal-
lurgical Applications of Bacterial Leaching and Related Microbiological Phenom-
ena. New York: Academic Press, 1978:19-44.
14. Jones CA, Kelly DP. Growth of Thiobacillus ferrooxidans on ferrous iron in
chemostat culture: influence of product and substrate inhibition. J Chern Tech
Biotechnol1983; 33B(4):241-261.
15. Braddock JF, Luong HV, Brown EJ. Growth kinetics of Thiobacillus ferrooxidans
isolated from arsenic mine drainage. Appl Environ Microbiol 1984; 48:48-55.
16. Smith JR, Luthy RG, Middleton AC. Microbial ferrous iron oxidation in acidic
solution. J water Pollut Control Fed 1988; 60(4):518-530.
17. Shrihari SRB, Kumar R, Gandhi KS Modelling of FeH oxidation by Thiobacillus
ferrooxidans. Appl Microbiol Biotechnol 1990; 33:524-528.
18. Norris PR, Barr DW, Hinson D. Iron and mineral oxidation by acidophilic bacte-
ria: affinities for iron and attachment to pyrite. In: Norris PR, Kelly DP eds.
Biohydrometallurgy, Proceedings of International Symposium. Kew UK: Science
and Technology Letters, 1988:43-59.
19. Liu MS, Branion RMR, Duncan DW. The effects of ferrous iron, dissolved oxygen
and inerts solids on the growth of Thiobacillus ferrooxidans. Can J Chern Eng
1988; 66:445-451.
20. Schnaitman CS, Korczynski MS, Lundgren DG. Kinetic studies of iron oxidation
by whole cells of Ferrobacillus ferrooxidans. J BacterioI1969:99(8):552-557.
21. Pesic B, Oliver DJ, Wichlacz. An electrochemical method for measuring the oxi-
dation rate of ferrous to ferric iron in the presence of Thiobacillus ferrooxidans.
Biotechnol Bioeng 1989;33(4):428-439.
22. Suzuki I, Lizama HM, Tackberry PD. Competitive inhibition of ferrous iron oxi-
dation by Thiobacillus ferrooxidans by increasing concentration of cells. Appl
Environ Microbiol1989; 55:1117-1121.
23. Lizama HM, Suzuki I. Synergistic competitive inhibition of ferrous iron oxida-
tion by Thiobacillus ferrooxidans by increasing concentrations of ferric iron and
cells. Appl Environ Microbiol 1989; 32:2588-2591.
24. Pirt SJ. Maintenance energy: a general model for energy limited and energy suffi-
cient growth. Arch Microbiol 1982; 133:300-302.
25. Huberts R. Modelling of Ferrous Sulfate Oxidation by Iron Oxidizing Bacteria-a
Chemiosomotic and Electrochemical Approach. PhD Thesis, Department of Chemi-
cal Engineering, University of the Witwatersrand, Johannesburg, South Africa 1994.
26. Ingledew WJ. Thiobacillus ferrooxidans-the bioenergetics of an acidophilic
chemolithotroph. Biochim Biophys Acta 1982; 683:89-117.
27. Nemati M, Webb C. A kinetic model for biological oxidation of ferrous iron by
Thiobacillus ferrooxidans. Biotechnol Bioeng; 1997: (In press).
28. van Scherpenzeel DA. The Kinetics of Ferrous Iron Oxidation by Leptospirillum-
like Bacteria. MSc (Eng) Thesis, Department of Chemical Engineering, University
of Cape 1997.
29. Roels JA. Energetics and Kinetics in Biotechnology. Amsterdam: Elsevier Biomedi-
cal Press, 1983.
30. Pinches A, Chapman JT, te Rie1e WAM et al. The performance of bacterial leach
reactors for the preoxidation of refractory gold-bearing sulfide concentrates. In:
Norris PR, Kelly DP eds. Biohydrometallurgy, Proceedings of International Sym-
posium. Kew UK: Science and Technology Letters, 1988:329-344.
31. Hansford GS, Chapman JT. Batch and continuous bio-oxidation kinetics of re-
fractory gold-bearing pyrite concentrate. Minerals Eng 1992; 5:597-612.
32. Miller DM, Hansford GS. Batch bio-oxidation of a gold-bearing pyrite-arsenopy-
rite concentrate. Minerals Eng 1992; 5(6):613-629.
33. Miller DM, Hansford GS. The use of the logistic equation for modelling the per-
formance of a bio-oxidation pilot plant treating a gold-bearing arsenopyrite-py-
rite concentrate. Minerals Eng 1992;5(7):737-749.
34. Hansford GS, Bailey AD. The logistic equation for modelling bacterial oxidation
kinetics. Minerals Eng 1992; 5:1355-1364.
35. Hansford GS, Miller DM. Bio-oxidation of a gold-bearing pyrite-arsenopyrite con-
centrate. FEMS Microbiol Rev 1993; 11:175-182.
36. Crundwell FK. The prediction of the bioleaching of refractory gold ores in a con-
tinuous plant from the batch data. In: Holmes DS, Smith RW, eds. Minerals
Bioprocessing II. Warrendale PA: The Minerals, Metals and Materials Society,
1995:17-39·
37. Crundwell FK, Bryson AW. The modelling of particulate leaching reactors-the
population balance approach. Hydrometall 1992; 29:275-295.
38. Hansford GS, Drossou M. A propagating pore model for the batch bioleach ki-
netics of a gold-bearing pyrite. In: Norris PR, Kelly DP eds. Biohydrometallurgy,
Proceedings of International Symposium. Kew UK: Science and Technology Let-
ters, 1988:345-358.
39. Chang YC, Myerson AS. Growth models of the continuous bacterial leaching of
iron pyrite by Thiobacillus ferrooxidans. Biotechnol Bioeng 1982; 24:889-902.
40. Bagdigian RM, Myerson AS. The adsorption of Thiobacillus ferrooxidans on coal
surfaces. Biotechnol Bioeng 1986; 28:467-479.
41. Kargi F, Weissman JG. A dynamic mathematical model for microbial removal of
pyritic sulfur from coal. Biotechnol Bioeng 1984; 26:604-612.
42. Konishi Y, Asai S, Katoh H. Bacterial dissolution of pyrite by Thiobacillus
ferrooxidans. Bioproc Eng 1990; 5:231-237.
43. Asai S, Konishi Y, Yoshida K. Kinetic model for batch bacterial dissolution of
pyrite particles by Thiobacillus ferrooxidans. Chem Eng Sci 1992; 47(1):133-139.
44. Konishi Y, Kubo H, Asai S. Bioleaching of zinc sulfide concentrate by Thiobacillus

ferrooxidans. Biotechnol Bioeng 1992; 39(1):66-74.
45. Nagpal S, Dahlstrom 0, Oolman T. A mathematical model for the bacterial oxi-
dation of a sulfide ore concentrate. Biotechnol Bioeng 1994; 43(5):357-384.
46. Wadsworth ME. Advances in the leaching of sulfide minerals. Mineral Sci Eng
1972; 4(4):36-47.
47. Boon M, Heijnen JJ, Hansford GS. Recent developments in modelling bio-oxida-
tion kinetics, Part I: Measurement methods. In: Holmes OS, Smith RW, eds. Min-
erals Bioprocessing II. Warrendale PA: The Minerals, Metals and Materials Soci-
ety, 1995:41-61.
48. Boon M, Hansford GS, Heijnen JJ. Recent developments in modelling bio-oxida-
tion kinetics, Part II: Kinetic modelling of the bio-oxidation of sulfide minerals
in terms of the critical sub-processes involved. In: Holmes OS, Smith RW, eds.
Minerals Bioprocessing II. Warrendale PA: The Minerals, Metals and Materials
Society, 1995:63-82.
49. Rawlings DE. Restriction enzyme analysis of 16S rONA genes for the rapid iden-
tification of Thiobacillus ferrooxidans, Thiobacillus thiooxidans and Leptospirillum
ferrooxidans in leaching ores environments In: Vargas T, Jerez CA, Wiertz JV,
Toledo H, eds. Biohydrometallurgical Processing. Vol I. Santiago: University of
Chile, 1995:9-18.
50. Ralph DE, May N, Hansford GS. The electrochemistry of bioleaching. Presented
at International Biohydrometallurgy Symposium Sydney Australia, August 1997.
51. Sand W, Rohde K, Sobotke B et al. Evaluation of Leptospirillum ferrooxidans for
leaching. Appl Environ Microbiol 1992; 58:85-92.
52. Norris PRo Iron and mineral oxidation with Leptospirillum-like bacteria. In: Rossi
G, Torma AE eds. Recent Progress in Biohydrometallurgy. Iglesias Italy:
Associazione Mineraria Sarda, 1983:83-96.
53. Hallmann R, Friedrich A, Koops H-P et al. Physiological characteristics of
Thiobacillus ferrooxidans and Leptospirillum ferrooxidans and physicochemical
factors influence microbial metal leaching. Geomicrobiol J 1993; 10:193-206.
54. Sand W, Gehrke T, Hallmann R et al. Sulfur chemistry, biofilm and the (in)direct
attack mechanism-a critical evaluation of bacterial leaching. Appl Environ
Microbiol 1995; 43:961-966.
55. Blake RC, Shute EA, Howard GT. Solubilization of minerals by bacteria: electro-
phoretic mobility of Thiobacillus ferrooxidans in the presence of iron, pyrite and
sulfur. Appl Environ Microbiol 1994; 60:3349-3357.
56. Luther GW. Pyrite oxidation and reduction: molecular orbital theory consider-
ations. Geochim Cosmochim Acta 1987; 51:3193-3199.
57. Moses CO, Nordstrom JS, Herman JS et al. Aqueous pyrite oxidation by dissolved
oxygen and by ferric iron. Geochim Cosmochim Acta 1987;51:1561-1571.
58. Sugio T, White KJ, Shute E et al. Existence of hydrogen sulfide:ferric iron-oxi-
doreductase in iron-oxidizing bacteria. Appl Environ Microbiol1992; 58:431-433.
CHAPTER 9
Physical Chemistry
of Bacterial Leaching
Frank K. Crundwell
Introduction
F errous ions and sulfide minerals may be oxidized by chemical or bacterial leach-
ing. The rate of bacterial oxidation is higher than that of chemical oxidation
under the same conditions.1These enhanced rates of reaction have been exploited
in the extraction of gold, copper and uranium and in the bioremediation of sulfidic
tailings. 1,2 An understanding of the mechanisms and physical chemistry of bacte-
rial leaching could be profitably used in the design and optimization of
bioremediation and metal extraction processes. 3
Two broad mechanisms of bacterial leaching have been proposed: 1 (1) the'di-
rect' mechanism, in which bacteria attach to the surface of the mineral, and attack
the mineral at the point of attachment, and (2) the 'indirect' mechanism, in which
the role of the bacteria is to oxidize ferrous ions to ferric ions, and the ferric ions
dissolve the sulfide mineral.
The direct mechanism of bacterial leaching of mineral sulfide, for example,
sphalerite, can be represented by the overall reaction:1
9·1
whereas the indirect mechanism requires the presence of iron in solution:'
ZnS +2Fe 3+ ~ Zn2 + + S+ 2Fe 2+
9·3
In both cases, bacteria are able to utilize sulfur as a substrate. This reaction is
given by:1
2S + 302 + 2H 20 bacteria) 2H 2S0 4 9.4
Both mechanisms of bacterial leaching are surface processes. The attachment

of bacteria to the mineral surface is affected by hydrophobic and electrostatic pa-
rameters and can be treated by the methods of surface chemistry. All dissolution
reactions in aqueous solutions include an electrochemical step at the surface, and
this step often controls the rate of reaction.

The aim of this chapter is to discuss the physical chemistry of bacterial leach -
ing. Two topics are examined here in particular: the surface chemistry of bacterial
attachment and the electrochemistry of chemical and bacterial leaching. The ex-
perimental results and a theoretical interpretation of the surface phenomena are
outlined for each topic. The surface chemistry of bacterial attachment is first dis-
cussed, and then the electrochemical mechanisms of leaching presented.
The Surface Chemistry of Bacterial Attachment

to Mineral Surfaces
The attachment of bacteria to a mineral surface is a prerequisite for the leach-
ing of minerals by the direct mechanism. Indeed, the prime evidence of the direct
mechanism is, at this stage, that Thiobacillus ferrooxidans attaches to mineral sur-
faces. On its own this is unconvincing evidence because bacterial cells generally
adhere readily to many types of surfaces. There is little consensus about the sig-
nificance of bacterial attachment in microbial leaching. For example, Espejo and
Ruiz4 determined that the total activity associated with the direct mechanism to
be between 1-10%, while Boogerd et a15 determined that 83-92% of the pyrite was
oxidized by the direct mechanism. The clarification of the role of bacterial attach-
ment to mineral surfaces would provide a significant step towards the effective
design and operation of bacterial leaching processes.
The colonization of a surface by bacteria is visualized as occurring in four dis-
tinct steps. These steps are illustrated in Figure 9.1. In the first step bacteria are
transported to the surface by diffusion, convection or active movement. Once in
the vicinity of the surface, the next step is the initial attachment or adhesion of the
bacteria to the surface. This process is thought to be dominated by physicochemi-
cal forces, such as the electrostatic and hydrophobic forces between particles and
surfaces. The terms initial attachment and adhesion are used interchangeably to
describe this second step. The attachment of bacteria to the surface is the third
step in the process. Bacteria that adhere to the surface may develop special struc-
tures such as fibrils, or secrete polymers such as polysaccharides to become more
firmly attached to the surface. The fourth and final process is the colonization of
the surface by the formation of micro colonies and biofilms.
The study of bacterial attachment to and biofilm formation on minerals is a
developing field and the methods of surface and colloid science are just beginning
to throw new light on the processes of bacterial leaching. Physicochemical aspects
of the adhesion of bacteria to the surface of minerals, the formation of biofllms on
pyrite and the formation of colloids during bacterial leaching are discussed below.
Adhesion and Attachment of Bacteria to Mineral Surfaces

The attachment of bacteria to a mineral can be divided into two steps:6 (1) the
initial attachment, which is governed by surface properties of the mineral and the
bacteria; and (2) the specific attachment, which is the growth of appendages or
proteins that anchor the bacterium to the surface. The initial attachment can be
reversible or irreversible. Reversible attachment is the weaker of the two, and the
bacterium can move back into the solution. A dynamic equilibrium exists between
the reversibly adsorbed bacteria on the surface and those in the solution, and most
mathematical models of bacterial leaching include terms describing this equilib-
rium/ A cell that is irreversibly attached does not move along the surface nor can
it be removed by moderate shear forces. s
Physical Chemistry of Bacterial Leaching 179
(a) (b)
~ Transportation
- - -. . 0) Secondary
adhesion
Primary
adhesion
(c)
Polymer Fibril
attachment attachment
Fig. 9.1. The four stages in the colonization of a surface by bacteria. (a) transport of
bacteria to the surface, (b) adhesion or initial attachment of bacteria to the surface at
either the primary or the secondary minima, (c) attachment of bacteria to the surface
by the excretion of lipopolysaccharides or by the growth of fibrils, and (d) the forma-
tion of micro-colonies and biofllms of bacteria.
Van Loosdrecht et al9 -12 have shown that hydrophobic and electrostatic forces
determine the initial attachment of many different types of bacteria to surfaces.
They examined the adhesion of more than 20 types and strains of bacteria to poly-
styrene and found that the surface coverage by the bacteria was more strongly
correlated with the contact angle of the bacteria than with the electrophoretic
mobility. This suggested that the hydrophobic interaction was more important than
the electrostatic force.
Van Loosdrecht and Zehnder8 summarized the previously published results as
follows: (1) adhesion increases with increasing hydrophobicity ofthe bacterium or
solid; (2) adhesion decreases with increasing electrostatic repulsion; (3) usually
hydrophobicity is more important than electrostatic interactions; (4) adhesion is
generally reversible; (5) irreversible adhesion occurs when the electrostatic inter-
action is weak or when the hydrophobicity is strong; and (6) there is evidence of a
layer of water between the bacterium and the surface.
The mechanism of adhesion of T. ferrooxidans to minerals has not been thor-
oughly investigated. Blake et aP found that the cells grown at pH 2.0 on pyrite or
ferrous ions were negatively charged, while those grown on sulfur were close to
their isoelectric point. Both pyrite and sulfur are negatively charged at pH 2.0.
In spite of the electrostatic repulsion, mixtures of cells and pyrite or cells and sul-
fur aggregated to form a colloid with properties of an intermediate of the two start-
ing materials. Blake et al3 did not examine the hydrophobic interactions. Solari et
al' 4 found that T. ferro oxida ns is slightly hydrophobic, and that the contact angle
180 Biomining: Theory. Microbes and Industrial Processes
increases when the pH decreases from 3.0 to 2.0. They showed that the bacteria
adhere preferentially to hydrophobic surfaces such as mineral sulfides. This adhe-
sion is weak, representing reversible adsorption of the bacteria to the mineral
surface.l4
The second step in the process of bacterial attachment is the development of a
means of specific attachment to minerals. Arrendondo et al l5 removed part of the
polysaccharide material that surrounds the T. ferrooxidans cell wall and showed
that this resulted in an increased number of cells that adhere to sulfide minerals.
They concluded that this was not only due to the increased hydrophobicity of the
cell wall, but that proteins in the cell wall play an important part in the mechanism
of adhesion. This view is supported by Devasia et all6 who concluded that an ap-
pendage made of protein is present on the surface of the cell and is responsible for
adhesion of T. ferrooxidans to sulfide minerals. The structures or proteins that
would allow for such specific interactions in T. ferrooxidans are yet to be identified.
Theory of Bacterial Attachment to Surfaces

Bacteria in solution have been frequently described as colloidal suspensions
and the adhesion of bacteria has been treated in terms of the theory of colloid
chemistry.S-u The physical chemistry of colloids treats the long-range interactions
of a colloidal particle and a surface as the combination of the van der Waals force
and the electrostatic force.1],lS
The van der Waals force arises when the oscillating charges of the positive
nucleus and the negative electrons in an atom (or molecule) cause an instanta-
neous dipole moment and induce an instantaneous dipole moment in another atom
(or molecule).t7.ls This force is also responsible for the long-range attractive forces
between macro-bodies, such as that between a colloid and a surface. The presence
of a medium such as water alters the magnitude, but not the dependence on dis-
tance of the van der Waals force. l7 Hydrophobicity arises when the magnitude of
the van der Waals interactions between water molecules is greater than the inter-
actions of the water molecules with the colloid and with the surface. l7 The hydro-
phobic interaction is effectively the solvent expelling the colloid to the surface. l7
The interaction energy, VA, arising from the van der Waals force between a
spherical colloid particle of radius, r, separated from a planar surface by a dis-
tance, H, is given by:ls
VA =_ A132r[1+_H_+ Hln(H+2r)]
6H H +2r r H 9·5
where Al32 is the Hamaker constant of the system. The Hamaker constant is depen-
dent on the polarizability and the number density of the atoms of each of the com-
ponents of the system. The Hamaker constant of the system is given by:l],lS
where Au is the Hamaker constant for the surface-colloid interaction, A33 for the
water-water interaction, Al3 for the surface-water interaction, and A23 for the col-
loid-water interaction. Thus the presence of the water will increase the colloid-
surface interaction if A'3' is larger than A'l> that is, if A33 is larger than A'3 + A'3" '7 In
other words, if the water-water interactions are large, then the presence of water
increases the Hamaker constant, and the effect is that the colloid is expelled from
the water towards the surface,'? This is the hydrophobic interaction.
The hydrophobicity of the bacterial cell wall has been determined using a num-
ber of techniques, the most popular of which are the measurement of the contact
angle of a drop ofliquid on a layer of cells, the liquid-liquid partitioning of cells in
an aqueous-hydrocarbon two-phase system and hydrophobic interaction chroma-
tography. 8 The measurement of the contact angle has been the most common
method of evaluating the hydrophobicity of the bacterial cell in studies related to
bacterial leaching. The contact angle is measured directly using a goniometer. Cells
are placed on a substrate, such as a glass slide or a ffiter membrane, then a drop of
solution is placed on top of the layer of cells. The angle between the meniscus of a
drop of solution and a layer of bacterial cells is then measured.
However, the relationship between the hydrophobic interaction and measur-
able quantities such as the contact angle is not straightforward. The Hamaker con-
stants can be calculated from the contact angle by a method outlined by Fowkes.'9
He described a method for calculating the contribution of the dispersion forces to
the surface free-energy of a substance from the contact angle and showed that this
quantity can be used to calculate the Hamaker constants. In contrast, van Oss et
al'o obtained an expression for the contact angle and the Hamaker constant in
terms of the Lifshitz-van der Waals component of the surface free-energy.
The electrostatic force, which is generally repulsive in bacteria-surface interac-
tions,8-" arises because both the surface and the colloid are charged. The develop-
ment of a surface charge (and as a result, a potential difference between a solid and
the solution) is one of the principle properties of surfaces.•1-2. The surface charge
originates from four different phenomena: (1) the termination of bonds of the solid
lattice at the surface may result in uncompensated charge; (2) the degree of ioniza-
tion of function groups such as -SH at the surface and thus the surface potential
may be dependent on the pH of the solution; (3) the polarization of the surface as
a result of charge-transfer reactions; and (4) the adsorption of ions or hydropho-
bic species on the surface.
The solid-solution phase boundary consists of a double-layer structure in which
the charge on the solid surface is balanced by the charge in the solution. Three
double layers exist at the solid-solution boundary!'-'3 These are the space-charge
layer which forms on the mineral side of the boundary, the compact Stern layer
which forms between the solid surface and the distance of closest approach of
nonadsorbed ions, and the diffuse Gouy layer in the solution. 2l -'3 Figure 9.2 illus-
trates the formation of the compact Stern layer and the diffuse Gouy layer. The
compact Stern layer is represented by the layer between the surface and the outer
Helmholtz plane, which is assumed to be sufficiently close to the plane of shear for
them to be regarded as occurring at the same position.
The interaction energy, VE, arising from the electrostatic force between a spheri-
cal colloid of radius r separated from a flat surface by a distance H is given by:'4
VE = 1tE~r~", 13 + '" 23)2 In{l + exp( -ldI)}+ ('" 13 - "'23)2 In{l- exp(-1d:I)}]
9·7
o
o
Distance
IHP Shear Diffuse Gouy Layer

Plane
Fig. 9.2. Schematic representation of a possible distribution of potential at an interface.

IHP indicates the inner Helmholtz plane, located at the surface of a layer of adsorbed
ions. Anions are shown here as unhydrated, and cations as hydrated. The plane of shear
is beyond, but sufficiently close to the next layer of adsorbed ions (referred to as the
outer Helmholtz plane) for the two to be regarded as occurring at the same position.
where is the potential difference between the surface (1) and the solution (3), \(123
is the potential difference between the colloid (2) and the solution, 1( is the inverse
of the thickness of the diffuse Gouy layer, E is the dielectric constant and Eo is the
permitivity of free space.
The potential difference of the double layer cannot be directly measured,23in-
stead different techniques are used in different fields of study. The potential differ-
ence can be determined with respect to a reference electrode,21 such as the hydro-
gen electrode or calomel electrode, or it can be estimated from the surface charge
by calculation using the Gouy-Chapman theory of the diffuse double-Iayer. 21,23
The measurement of the zeta potential is often used in surface and colloid sci-
ences to determine the surface potential. The zeta potential represents the poten-
tial between the plane of shear and the bulk solution!3 Mineral particles are inert
in these studies and the surface potential, and hence the zeta potential, is affected
by the adsorption of ions including H+ and OH-, and surfactants. Typical mea-
surements show the effect of an adsorbed species, such as a surfactant, as a func-
tion of pH. 23 However, it is difficult to determine accurately the distance between
the plane of shear and the surface of the particle, and as a result it is difficult to
determine the surface potential accurately. Consequently, the surface potentials in
Physical Chemistry of Bacterial Leaching
>-
e>
Q)
c:
W
c:
o
n~
c
Q)
Secondary adhesion
VA Primary adhesion
Fig. 9.3. Interaction energy for the long range forces between a spherical colloidal par-
ticle and a planar surface as a function of the distance of separation. The curve of total
interaction energy indicates the possibility of two minima at different distances from
the surface. 8"7>18
Equation 9.7 are often replaced by the values of the zeta potential. U The zeta po-
tential and the electrophoretic mobility are related to each other by the
Smoluchowski equation.l7>l8
The total interaction energy, VTot' is the sum of the van der Waals energy and
the electrostatic energy, and is given by:'7.l8
The total interaction energy is shown in Figure 9.3 as a function of the distance
of separation, H. This figure indicates that there are two minima in the energy
diagram: there is a primary minimum close to the surface representing irrevers-
ible adsorption and a secondary minimum at a distance of between 5-20 nm from
the surface representing reversible adsorption.
Solari et al l4 compared the contribution of the hydrophobic interaction and the
electrostatic interaction to the adhesion of T. ferrooxidans on various mineral sur-
faces. They determined the hydrophobicity of the bacteria by measuring the con-
tact angle of a layer of cells deposited on a surface and by liquid-liquid partition-
ing. The electrostatic interaction was determined by the measurement of the
electrophoretic mobility and it was reported that the water contact angle of the
bacterial cell wall was 20.3° on a fresh layer of cells and 23.3° on a dry layer of cells.
Between 14- 22% of the cells were extracted into hexadecane during liquid-liquid
partitioning tests. These results were interpreted as indicating that the bacterial
cell wall is slightly hydrophobic. Of the minerals tested by liquid-liquid partition-
ing, only quartz was considered hydrophilic since all of the quartz remained in the
aqueous phase. Chalcopyrite and pyrite were concentrated at the interface between
the aqueous and the hydrocarbon phases, indicating that these minerals are hy-
drophobic even though their contact angles are about 60°. The isoelectric point of
the bacterial cells occurred at a pH of about 3.0 and that of most of the sulfide
minerals between pH 2.0-3.0. Therefore both the sulfide mineral and the bac-
terium have the same charge at pH 2.0 and below and the electrostatic force is
repulsive.
Solari et al14 determined that the bacterial adhesion is reversible and corre-
sponds to adsorption at the secondary minimum. They concluded that the hydro-
phobic interactions dominate the adsorption of T. ferrooxidans to mineral sur-
faces. In addition to these measurements, Solari et al14 also determined the kinetics
and equilibrium of bacterial adhesion to various minerals. They showed that bac-
terial adhesion was complete after about 4-6 hrs, which was relatively slow com-
pared to work of others.14 The isotherm of the adhesion of T. ferrooxidans to quartz
was close to linear. This isotherm indicated that about 30% of the bacteria adhere
to the quartz, irrespective of the concentration that is initially present. The cover-
age of the surface of the minerals by T. ferrooxidans was 23.0, 20.0 and 9.8% on
chalcopyrite, pyrite and quartz, respectively.14 It was concluded that the bacteria
attached preferentially to the hydrophobic sulfide minerals rather than to the hy-
drophilic quartz.14
Ohmura et al25 compared the attachment of T. ferrooxidans and Escherichia
coli to different minerals. The contact angles of T. ferrooxidans and E. coli were
approximately 23 °and 31°,respectively, and these values agreed with those reported
previously. However, while 60% of the E. coli were extracted in hexadecane in aque-
ous-hydrocarbon partition tests, only 25% of the T. ferrooxidans were extracted in
the hexadecane. Both the measurements of the contact angle and the liquid-liquid
partitioning indicated that T. ferrooxidans cells were more hydrophilic than E. coli
cells. Ohmura et al 25 also measured the contact angles of minerals tested. The val-
ues of the contact angle with pH 2.0 solution were 28.4°,68.9°,83.4°and 80.9° for
quartz, pyrite, chalcopyrite and galena, respectively. The adhesion studies indi-
cated that 9.9, 18.9, 26.8 and 30.8% of the E. coli cells adhered to quartz, pyrite,
chalcopyrite and galena, respectively.25These results suggested a relationship be-
tween the contact angle of the bacterial cells and the contact angle of the mineral.
However, results for the adhesion of T. ferrooxidans onto these minerals were more
variable. Ohmura et al 25 determined that 4.7,24.0, 14.0 and 1.4% of the T.
ferrooxidans cells adhered to quartz, pyrite, chalcopyrite and galena, respectively.
They concluded that there was a special interaction between T. ferrooxidans and
the reduced iron in the minerals, that is T. ferrooxidans recognized the reduced
iron in pyrite and chalcopyrite and adhered selectively to these minerals but not to
galena, by a strong interaction. They believed that this interaction was not physi-
cochemical in nature.
In order to test their hypothesis Ohmura et al25 added ferrous ions to the solu-
tion which reduced the number of cells that adhered to pyrite. Superficially this
confirmed their reasoning, however, this result could be explained as follows. The
zeta potential for both pyrite and T. ferrooxidans is negative under the conditions
of their experiments 25 and the addition of iron to the solution would reduce the
potential of the mineral surface (see the mixed potential model, Equations 9.18 or
9.31, discussed later in this chapter). Thus, with the addition of ferrous ions to the
solution the surface of the mineral would become more negative and the electro-
static repulsion between the cells and the mineral would increase thereby reduc-
ing the number of attached cells.
Devasia et al16 determined the adhesion of T. ferrooxidans that had been grown
on different substrates on various minerals. It was observed that about 20% of the
cells grown in sulfur, pyrite and chalcopyrite were extracted into the hexadecane
phase in liquid-liquid partitioning experiments, while negligible numbers of cells
grown on ferrous ions were removed from the aqueous phase. They reported a
similar trend for the results from hydrophobic interaction chromatography using
octyl-Sepharose as the extractant. About 17% of the cells grown on ferrous ions
were bound to the octyl-Sepharose, while about 50% of the cells grown on sulfur,
pyrite, and chalcopyrite were bound to the octyl-Sepharose. The isoelectric point
of sulfur, pyrite and chalcopyrite moved to higher pH values (approximately from
2.5-4.5) when bacterial cells were present on the surface. It was concluded that
both electrostatic and hydrophobic interactions are important in the adhesion of
T. ferrooxidans to minerals.16
Devasia et al16 also reported the growth of a surface component when cells
were grown on sulfur, pyrite and chalcopyrite. This component was not present
when the cells were grown on ferrous ions. This result was obtained by raising
antibodies to cells grown on sulfur. Devasia et al16 argued that the FTIR difference
spectrum indicated that this component was a protein. However, the quantity of
lipopolysaccharide on the cell surface is expected to increase with attachment to
and growth on a solid substrate such as sulfur and mineral sulfides and Devasia et
al16 did not compare the spectrum with that of the lipopolysaccharide produced by
attached cells.
Sam et al2.6 measured the hydrophobicity of cell suspensions by liquid-liquid
partition in aqueous and organic (m-xylene) phases. They measured the electro-
static interactions by determining the zeta potential of the suspensions. They
showed that cells grown on sulfur, pyrite, and chalcopyrite exhibit higher hydro-
phobicity than cells grown on ferrous ions.
From this presentation of work reported in the literature, it is clear that there is
agreement on some key parameters. Measurements of the zeta potential of cells of
T. ferrooxidans and the particles of minerals indicate that their surfaces are nega-
tively charged and that the electrostatic interaction between cells and minerals is
therefore repulsive. The measurements of the contact angle of T. ferrooxidans con-
sistently yield values of 20°-23°C,and the liquid-liquid partitioning with hexadecane
as the hydrocarbon results in extraction of 20-25% of the cells into the organic
phase. The adhesion isotherm appears to be linear. However, the surface coverage
by the cells is low ranging between 1-25% and it does not appear to be directly
related to the contact angle of the mineral.
There is little consensus about the mechanisms affecting bacterial attachment
to minerals. Hydrophobic interactions are clearly important but they are difficult
to quantify and no clear pattern has emerged.2.5 A significant group of workers
believe that the attachment of cells to the mineral involves a specific biochemical
interaction rather than a physical interaction and they postulate that components
of the cell wall are able to determine the source of energy.16,25
Biofilm Formation
Once bacteria adhere to the surface, colonization of the surface begins. Bac-
teria may develop special cell structures such as fibrils and secrete polysaccha-
rides to attach more securely to the surface.2.7-29 Other responses, such as a loss of
flagella may occur. The bacteria grow and form micro-colonies, creating a local
micro-environment in which the concentrations of ions and nutrients at the sur-

face are different from those in the bulk. Stable growth on the surface results in the
formation of biofilms. 27- 29
The formation of biofilms of T. ferrooxidans has not received much attention.
Crundwell30 examined the colonization of a polished section of pyrite by a mixed
culture of iron-oxidizing bacteria using a confocal microscope. Clusters of micro-
colonies were seen within 3-4 days. After a period of 13 days, the surface was com-
pletely covered by a biofilm of 20-50 Jl11l thick. A layer of ferric hydroxide formed
below the biofilm. Vertical sections of the biofilm showed that the bacteria were
concentrated on the solution side of the biofilm. Crundwell30 concluded that the
bacteria did not require direct attachment to unreacted pyrite in order to utilize it
as an energy source. Rather, he speculated that iron was cycled between the bacte-
ria and the pyrite within the biofilm.
Sand et al31 proposed a similar model except that iron was cycled in the li-
popolysaccharide layer between a single bacterium and the pyrite surface, rather
than in a colony of bacteria. The models of Sand et al31 and Crundwe1l30 are comple-
mentary and are shown in Figure 9.4.
The models of bacterial leaching discussed here fall between the classical mod-
els of direct and indirect leaching. These newer models suggest that it may be pos-
sible to increase the rate of bacterial leaching by increasing the number of bacteria
that are attached to the surface and by maintaining a high redox potential in
solution.
Colloid Formation on Bacterial Cells

Two sets of workers have reported the formation of colloids on the surface of T.
ferrooxidans. Rojas-Chapana et alP detected colloidal sulfur particles of between
4-70 nm in size in the organic capsule around the cell wall. No doubt these sulfur
colloids were intermediates in the dissolution process, and were precipitated and
stabilized by the lipopolysaccharides surrounding the cell. Rojas-Chapana et alP
speculated that these colloids may serve as an energy reserve. However, Pooley
and Shrestha33 detected colloidal particles of silver sulfide surrounding the bacte-
rial cell wall when silver was added to the leaching solution. The silver sulfide was
probably precipitated in the lipopolysaccharide surrounding the cell. However, it
is difficult to explain why this silver sulfide was not dissolved by the leaching solu-
tion. 33 The detection of colloids of silver sulfide surrounding the bacterial cells
seems to suggest that the presence of colloids of sulfur may not be a special feature
of the survival strategy of T. ferrooxidans; instead they simply form as a result of
the reduced solubility of sulfur compounds in the lipopolysaccharide and the cell
periplasm.
Electrochemistry of Mineral Dissolution

and Bacterial Leaching
Dissolution is the process in which material moves from the solid phase, across
the charged phase boundary, and into the liquid phase. Chemical bonds are bro-
ken in the solid and are formed in solution. The rate of dissolution is affected by
the bonding in the solid, the movement of ions and electrons across the phase
boundary and the formation of bonds in solution. Dissolution reactions may oc-
cur in the presence or absence of a redox couple. Those dissolution reactions that
occur in the absence of a redox couple involve only the transfer of ions. There is no
(a)
Ferric hydroxide
Pyrite
(b)
Fig. 9.4. More recent models of the mechanism of bacterial leaching. (a) The biofilm
model in which iron is cycled between the bacteria in the biofllm and the surface of the
mineral. Corrosion products have been observed between the mineral and the biofllm.
(b) A model similar to that of the biofllm model except that the iron in the lipopolysac-
charide that surrounds a single cell is cycled between the cell and the mineral surface.
change in oxidation state and they are referred to as nonoxidative reactions. Dis-
solution reactions in the presence of a redox couple involve the transfer of ions
and electrons and are referred to as oxidative reactions when the mineral is oxi-
dized and as reductive reactions when the mineral is reduced. All of these types of
reactions are common in leaching chemistry.
For example, the leaching of sphalerite in the presence of ferric ions occurs by
oxidative dissolution:
ZnS + 2Fe 3+ ~ Zn2+ + 2Fe 2+ + S 9·9
while the leaching of sphalerite in an acidic solution in which no oxidant is present
occurs by nonoxidative dissolution:
ZnS + 2H + ~ Zn2+ + 2H2 S 9.10
Both the oxidative and nonoxidative dissolution of minerals are electrochemi-
cal processes, since charged species move across the phase boundary. In the next
section, the electrochemical model of leaching is discussed, the kinetic equations
of Nicol et al34 derived and their application in both leaching and bacterial leach-
ing studies demonstrated.
Electrochemical Mechanism of Oxidative and Reductive Reactions

Dissolution reactions that result in the change of oxidation state of the mineral
can be separated into anodic and cathodic half-reactions. 21,34 For example, during
the dissolution of a metal, M, in an aqueous solution containing the oxidant, BH,
the dissolution reaction is:
M(s) + B2+ (aq) ~ M+ (s) + B+ (aq)
This reaction can be separated into the anodic dissolution of metal M, and the
cathodic reduction of oxidant BH. The anodic half-reaction is: 21
M(s) ~ M+(aq) +e- (s) 9. 12
and the cathodic half-reaction is given by:21
B2+ (aq) +e- (s) -4 B+ (aq)
Note that the separation of the reaction into anodic and cathodic half-reac-
tions is a formal separation and that it does not mean that the reaction necessarily
occurs at different locations on the mineral surface. 21 The anodic and cathodic
half-reactions occur at all sites on the particle surface, unless the mineral particle
is clearly heterogeneous as in the case of galvanic corrosion.
Since the rate of production of electrons is equal to the rate of consumption of
electrons, the general requirement for dissolution is that the sum of the rates of the
anodic reactions must be equal to the sum of the rates of the cathodic reactions. 21,34
This condition is given by:
where ra and rc are the rates of the anodic and cathodic half-reactions, respec-
tively. This condition is used to derive an expression for the potential difference at
the mineral surface and the rate of dissolution.
The rate of an electrochemical reaction is dependent on the potential differ-
ence across the electrode-solution interface, E, the concentration of reacting spe-
cies and the temperature, T. The potential, E, is that presented earlier as 1jJ ' 3' but is
measured with respect to a reference electrode, that is, E = 1jJ 13 _ 1jJ .3. 21
If the anodic half-reaction is considered irreversible, then the rate of the an-
odic half-reaction is given by:21
fa =ka exp(aaFE/RT) 9.15
where ka is a rate constant, (Xa is the charge-transfer coefficient, which has a value
between 0.4-0.6, F is Faraday's constant and R the gas constant. Similarly, the rate
of the cathodic half-reaction is given by:21
rc =kJB 2+]exp{-(1-a c)FE / RT} - k~[B+]exp(acFE/ RT) 9.16
where kc and kc1are rate constants and (Xc is the charge-transfer coefficient for the
cathodic reaction which also has a value between 0.4-0.6.
Substituting the expressions for ra and rc (Eqs. 9.15 and 9.16), into Equation
9.14 gives the expression:
k. exp(<x.FE IRT) = k c[B 2+]exp{-(1- <x.) FE I RT} - k![B+]exp(<XcFE I RT)
9·17
If it is assumed that aa = (xc> then Equation 9.17 may be rearranged to give an
expression for the potential difference, E:34
This potential is called the mixed potential and as a result the electrochemical
mechanism of leaching is sometimes referred to as the mixed-potential modePl
Substituting this expression for E back into Equation 9.14 gives the following equa-
tion for the rate of dissolution:34
where aa = (Xc = (X and rdiss represents the rate of dissolution, which is equal to the
rate of the anodic or the cathodic half-reactions, i.e., r diss =r a =r cby Equation 9.14.
The value of (X is expected to be close to 0.5, so this model indicates that the rate of
dissolution is expected to be one-half order in the concentration of the oxidant,
BH. Many dissolution reactions display a one-half order dependence on the con-
centration of the oxidant in solution.
The derivation of Equation 9.19 represented a major advance in the understand-
ing of leaching chemistry. It was first presented by Nicol et al34 for the leaching of
UO z in ferric sulfate solutions. The kinetics of reductive leaching, such as the re-
duction of pyrolusite (Mn02) by sulfur dioxide can be described by a similar mecha-
nism. It is easy to show that the rate of dissolution for a reductive leaching reaction
is given by:
=k (
kc[B+] ),-a
rd"
ISS c kc +k! [B2+] 9.20
The value of a is expected to be close to 0.5, so that order of reaction with

respect to the reductant is expected to be close to 0.5. A large number of leaching
studies have been published and of particular significance is the result that the
order of reaction is close to 0.5 for many oxidative and reductive dissolution reac-
tions.35 This indicates that the electrochemical mechanism ofleaching is generally
applicable.
Influence of the Electronic Structure of the Mineral

on the Rate ofDissolutiow6
The rates of leaching of a mineral from different sources are different. For ex-
ample, the rate of dissolution of sphalerite from different ores varies depending
mainly on the concentration of iron in the sphalerite lattice. This phenomenon is
explained by the influence that the electronic structure of the mineral exerts on
the electrochemistry of dissolution. 36,37 In order to dissolve a mineral, electrons
need to be removed from bonding orbitals. In the case of sphalerite it is energeti-
cally more favorable to remove electrons from the bonding orbitals of iron in the
lattice than from zinc in the lattice. 37
Different minerals dissolve at different rates. For example, it is well known that
pyrite dissolves slowly in comparison to other iron sulfides, such as pyrrhotite.
These differences can also be explained by the semiconductor-electrochemical
model of leaching. 36 The electronic structure of pyrite is such that the upper va-
lence band, which interacts with common oxidizing agents, does not contribute to
the bonding of the minera1. 36 Therefore, removal of electrons from these
nonbonding orbitals will not result in the breaking of bonds, and as a result disso-
lution is slow. 36 Stronger oxidizing agents interact with the lower valence band
which is composed of bonding orbitals and therefore dissolution will occur. 36
In the next sections, the electrochemical model will be decribed in detail for
three different dissolution reactions.
Applications of Electrochemical Mechanism of Leaching

Chemical Leaching of Pyrite by Ferric Ions
Pyrite is a common mineral that is found in many ore deposits and tailing sites.
The primary cause of acid-mine drainage is the bacterial leaching of the pyrite in
tailing sites. Gold that is encapsulated by pyrite and arsenopyrite cannot be treated
by the conventional cyanidation process. Pretreatment of these ores by bacterial
leaching liberates this gold. Thus, pyrite plays an important role in the study of the
mechanisms of bacterial leaching and the study of the electrochemical mecha-
nism of dissolution of pyrite will assist in the process of understanding bacterial
leaching.
McKibben and Barnes38 found that the rate of leaching of pyrite in ferric sul-
fate solutions is given by:
[Fe 3l-]US
rd'
lSS
=k--:-~
[H+ ]0.5 9. 21
These results may be explained in terms of the electrochemical mechanism of

dissolution. If it is assumed that the initial step is the formation of a hydroxide
layer on the surface of the pyrite and the rate-determining step is the electrochemical
reaction of this layer, then we can write the first two steps in the anodic dissolution
of pyrite as follows:
The chemical and electrochemical reactions that follow these two initial reac-
tions are assumed to be much faster than the rate-determining reaction and there-
fore do not influence the rate of reaction. The rate of formation of the adsorbed
species FeS~OH- ads by Equation 9.22 is given by:
where q is the fraction of surface covered by FeS~OH-ads' Since this reaction is fast
compared to the subsequent rate-determining step and there is no accumulation
of the adsorbed layer of FeS~OH- ads, then r 1 '" o. Assuming that the surface cover-
age 8 is small (8 « 1), we obtain the following expression for 8:
9= kl
k_I[H+] 9·25
The rate of the rate-determining electrochemical step (Eq. 9.23) is given by:
fa = kagexp(a.aFE/RT)
The substitution of Equation 9.25 into Equation 9.26 gives:
f = kakl exp(a.aFE I RT)

--=.-.:....--=;..;....::...,,--~
9·27
a k_I[H+]
The cathodic reaction is the reduction of ferric ions at the pyrite surface, given
by:
The rate of the cathodic reaction is given by:
Following the same method of derivation described in the section on the elec-
trochemical model of leaching (i.e., using Eq. 9.14 to get an expression for E, and
substituting this expression into the rate equation for the anodic reaction), the
following expression for the dissolution of pyrite is obtained:
I-ex
k.kl k
(.[Fe])
( ) 3+ ex
f diss = k_l[H+]
where a =aa =a c•
Since aa and a c are expected to have values close to 0.5, it is clear that this rate
equation is the same as that obtained by McKibben and Barnes.38 This means that
the electrochemical model describes the kinetics of dissolution of pyrite.
Effect of Bacterial Action on the Mixed Potential of Pyrite

In the derivation of the electrochemical mechanism of leaching, we obtained
an expression for the mixed potential, given by Equation 9.18. The mixed potential
of a dissolving mineral can be measured with respect to a reference electrode. This
-
W 440
() Slopes
C/)
;>
-E
co
:;::;
400
-
c:
Q) 360
0
Q.
"0
Q) 320
.~
~ + Biofilm
• Sterile
280
2 4 6 8 2 4 6 8 2
10.2 10-1
[Fe3+], [Fe2+] (mol/L)
Fig. 9.5. The mixed potential of pyrite in the presence and absence of bacteria at vari-
ous concentrations of ferrous and ferric ions. 39
is an easy and useful method of determining the effect that changes in the solution
will have on the leaching reaction. Crundwell et al39 measured the mixed potential
of a pyrite electrode in the presence and absence of bacteria to determine the in-
fluence of biofilm formation on the kinetics of bacterial leaching. The results of
this mixed potential study are shown in Figure 9.5.
The experiments shown in Figure 9.5 were conducted in two steps.39 Initially,
the pyrite electrode was pretreated in a leaching solution in either the presence or
absence of bacteria for 10 days. After this process the electrode was removed from
the solution and the mixed potential was determined at different concentrations
of ferrous and ferric sulfate in a sterile solution containing 0.4 M sodium sulfate
and 0.2 M sulfuric acid. A slight blackening of the surface was observed in experi-
ments in the presence of bacteria. This was thought to be caused by the formation
of a ferric hydroxide layer on the pyrite surface. In addition, micro-colonies of
bacteria or biofilms, were clearly visible by the end of the pretreatment process.
The results shown in Figure 9.5 indicated that there was an increase in the mixed
potential of about 10 m V when bacteria were present in the solution. This slight
increase in the mixed potential may be explained by a slight increase in the con-
centration of ferric ions at the surface as a result of the presence of the bacterial
biofilm. However, from an examination of Equation 9.15, the increase in the leach-
ing rate as a result of an increase in the mixed potential of 10 m V can be no greater
than 22% at 25°C (i.e., exp{a 0.01 F/RT} = 1.22. Such a small increase in the rate of
leaching as a result of bacterial activity suggests that the contribution of bacterial
attachment to the overall rate ofleaching is small.
These results are explained by the electrochemical mechanism ofleaching.39 If

it is assumed that the partial anodic current due to the dissolution of pyrite is
small compared to the partial currents due to the oxidation and reduction of the
ferric and ferrous ions in solution, that is ka «k/[FeH ], then the mixed potential
Equation 9.18 may be written as: 39
RT ([Fe 3
E = E pyr + -In --2-+-
+])
F [Fe]
where Epyr is RT/Fln(kJkc1).
The slopes presented in Figure 9.5 indicate that the slope of the mixed potential
with respect to In[Fe3+] and In[FeH ] are close to 26.5 mV,i.e., the value ofRT/F at
25°C. These results indicate that the effect of changing the concentrations of fer-
rous and ferric ions on mixed potential in the presence and absence of bacteria is
described by Equation 9.31. This suggests that the most likely reason for the slight
increase in the mixed potential in the presence of bacteria is a slightly higher con-
centration of ferric ions at the pyrite surface.39
The Kinetics of the Oxidative Dissolution of Sphalerite

The dissolution of sphalerite in ferric sulfate or ferric chloride solutions occurs
according to the reaction: 40•41
ZnS+ 2Fe3+ ~ Zn2+ +2Fe 2+ +S 9·32
Iron occurs as an impurity in all natural sphalerite. This iron has a pronounced
effect on the rate of dissolution in ferric solutions. Crundwelp7 determined that
the rate of dissolution is proportional to the iron content in the sphalerite. Values
of Lltrxn, which is directly proportional to the rate of dissolution, are shown in
Figure 9.6 for four samples of southern African sphalerite. (L is the initial particle
size, and trxn is the time for complete conversion of particles initially of size L).
The anodic half-reaction for the dissolution of sphalerite is written as:37.40.41
9·33
and the cathodic half reaction is given by:37.4 0.41
2Fe3+ + 2e - ~ 2Fe 2 +
9·34
The presence of iron that substitutes for zinc atoms in the sphalerite lattice
results in a d-orbital band within the band-gap of sphalerite.37 The iron d-orbitals
of this band are of bonding character. This means that the removal of an electron
from this band also results in dissolution of the solid.37 The iron impurity and its
associated d-orbital affects the dissolution of sphalerite by presenting a narrow
localized band with which the transfer of electrons is energetically more favorable
than with the valence band. 37
This means that the rate of the anodic reaction is given by:3 6.37
9·35
... Wards
2.0
• Zincor 850C
(0
0
T""
•
l>. Rosh Pinah
Gamsberg
•
1.6
X
c:
'E
....... 1.2
E
650C
C: 0.8
~
:)
0.4 450C
0.0
o 2 4 6 8 10
Iron content in (Zn,Fe)S, %
Fig. 9.6. The effect of iron content in the lattice of the sphalerite on the rate of dissolu-
tion at various temperatures. 37 LIt is proportional to the rate of leaching.
and the cathodic reaction is given by:3 6,37
where Nd is the relative concentration of iron in the sphalerite (mols Felmols Zn).
Since ra = rc by Equation 9.14, E can be eliminated between Equations 9.14,9.35
and 9.36, giving the expression for the rate of dissolution: 37
rdiss = (k akc r N d[Fe 3+] a

9·37
where a = (Xa = (xc> and (X has a value close to one-half.

This equation indicates that the rate of reaction is first order in the concentra-
tion of iron in the sphalerite and is one-half order in the concentration of ferric
ions in solution. Experimental results shown in Figure 9.6 demonstrate that the
rate of reaction is a linear function of the concentration of iron in the lattice. In a
detailed analysis of the ferric sulfate complexes in solution using the stability con-
stants calculated by Dry,42 CrundwelPMO,41 argued that of the ferric sulfate species
in solution only FeHSO/+ and Fe3+(aq) were active at the sphalerite surface.
The rate of reaction is then:
rdiss= "k
\: a k c )0.5 Nd\\Fe
fr[ 3+ (aq)]+[FeHS0 2+
4 ]
)0.5
10-1 9
8
7
6
5
4
3
c:: 2
'E
.......
,... 10-2 9
C 8
7
~ 6
-t
,... 5
Slope = 0.45
4
10-3
2 3456789 2 3 4 5 6
10-2
[Fe 3+] + [FeHS04 2+], mollL
Fig. 9.7. The effect of solution speciation on the rate ofleaching.4' lit is proportional to
the rate ofleaching.
The rate of reaction, rdiss, is proportional to Ihrxn, where trxn is the time for
complete reaction of the particle. Values of Ihrxn are plotted as a function of the
concentration of the active ferric species in solution in Figure 9.7. Since the slope
of the regression line of data is 0.45, which is close to 0.5, this figure indicates that
Equation 9.38 is in agreement with the experimental results for the kinetics of the
oxidative dissolution of sphalerite.3MO,41
The discussion indicates that the electrochemical mechanism is able to describe
the two most important features of the leaching of sphalerite: that the rate of reac-
tion is a linear function of the concentration of iron in the lattice and the rate is
one-half order in the concentration of oxidant in solution.
Incorporation of the Electrochemical Kinetics

into a Model of Bacterial Leaching of Sphalerite
The kinetics of bacterial leaching reactions have not been generally interpreted
in terms of the electrochemical mechanism of leaching. Part of the reason for this
is that in typical bacterial leaching studies the concentrations of ferric and ferrous
ions vary during the course of the experiment. This makes it difficult to analyze
the data to obtain the orders of reaction that are required to identify the reaction
as being controlled by the electrochemical step. In order to undertake an analysis
of batch data of this sort, it is necessary to present the electrochemical mechanism
in the form of a mass balance describing the leaching reaction in a batch reactor.
As an example, the leaching of sphalerite by ferric sulfate is examined. This
derivation can be easily applied to the leaching of other minerals. The mass bal-
ance for the leaching of sphalerite in a batch reactor is given by:43
dN ZnS
d i = - rdiS A 9·39
where NZns represents the number of moles of sphalerite in the batch reactor, r diss
represents the rate of consumption of sphalerite by the leaching reaction (units of
mollm 2 s), and A represents the surface area (m2 ) of sphalerite that is available for
reaction.
To develop the mathematical model, expressions for the rate of reaction, rdiss,
and for the available surface area, A, are required. These two expressions are de-
rived from the electrochemical mechanism of leaching and from the shrinking-
particle model.
An expression for the rate of reaction has been presented in Equation 9.38.
Although Crundwe1l41 showed that not all of the ferric species are active at the
sphalerite surface, it is assumed that all the ferric species are active at the surface
in order simplify this presentation. Thus the rate of dissolution is given by Equa-
tion 9.19, and if it can be assumed that ka << kc'[FeH ], then:
9·40
The change in the number of moles of sphalerite is related to the change in size
of the particle. For particles that are approximately spherical with radius I (or which
maintain their geometric proportion during leaching), the following relationships
are obtained: 43
dN Zns = 41tA2p Zns dA 9.41
A = 41tA.2
X=l-(~J 9·43
where rznS is the molar density of sphalerite, X is the reaction conversion and L is
the initial particle size.
Substituting Equations 9.40-9.43 into the mass balance, Equation 9.39, and af-
ter some rearrangement, the mass balance for the leaching of sphalerite in a batch
reactor can be written as: 44
dX = ~(l_xf([Fe3i-])a
dt tun [Fe2+ ] 9·44
where t rxn• the time for complete reaction, is given by t rxn = P zns L(k! ) a
k. kc
The physical significance of trxn is that it represents the time that a particle of
size L will take to be completely dissolved if the concentration of ferric and ferrous
ions remains at the initial value throughout the course of the reaction.
The effect of temperature is accounted for by the Arrhenius equation,
t rxn = t;:n exp(E A I RT) . The initial condition is X = 0 at t = o.
[Fe3+].
O.938M
0.8 O.466M
C O.172M
0 0.6
.~
O.093M
~
C
0 0.4
U
0.018 M
0.2
0.0
o 10 20 30 40 50 60 70
Time, min
Fig. 9.8. Effect of different concentrations of ferric ions on the rate ofleaching of sphaler-
ite at 65°C, 0.038 M FeH , 0.1 M HzS0 4,1 giL solids. 44
Equation 9.44 is solved in conjunction with the solution mass balances, given
by:44
[Fe 3+]=[Fe 3+]0 - 2FZnSX -2FFeS X 9·45

[Fe 2+] = [Fe2+]o + [Fe 3+]0 - [Fe 3+] + FFeSX
where Fzns and Fpes represent the initial molar concentrations of zinc sulfide and
iron sulfide added to the batch reactor.
The fit of the model parameters to the leaching data presented by Verbaan and
Crundwe1l40 is shown in Figures 9.8 and 9.9. 44 The sphalerite contained 50.9% zinc,
and 9.1% iron and the initial particle size was 22 JlDl. The model is represented by
the lines and the data by the points on these figures.
The close correspondence between the model (Eqs. 9.44-9.46) and the data
suggests that the model is a good representation of the data and supports the as-
sumption that, for this data, all ferric species are active in the reaction. 41 All the
lines in both these figures have been obtained with the same set of parameters,
that is, with 3!t~rxn = 1.88 X 105 min-t, a = 0.39, and Ea = 48 kJ/mol. 44 These param-
eters are similar to those obtained from other studies of chemicalleaching.3•37>41
The value of 0.39 obtained for a is close to the value predicted from the electro-
chemical mechanism. Therefore, it can be concluded that the leaching of sphaler-
ite is described by the electrochemical mechanism.
The parameters obtained from batch leaching studies can be used to predict
the operation of a continuous leaching plant. 3 The piloting of a new continu-
ous operation is a slow and expensive task. In contrast, small-scale batch leaching
Temp
0.8
c: 0.6
0
...
'iii
Q) D
>
c:
0 0.4 45°C
()
0.2
25°C
0.0
0 10 20 30 40 50 60 70
Time, min
Fig. 9.9. Effect of temperature on the rate of leaching of sphalerite with 0.45 M Fe 3+,
0.04 M Fe 2 +, 0.1 M H2 S0 4, 1 giL solids. 44
experiments are easy to perform. Therefore, a thorough analysis of the batch leach-
ing data in the manner shown in this chapter and the use of the kinetic parameters
obtained from this analysis to predict the performance of a continuous plant could
save a significant amount of time and money in feasibility and piloting studies. In
addition, these kinetic parameters can be used in models to simulate and optimize
the operation of the full-scale continuous plant.
Conclusions
The physical chemistry relating to bacterial attachment to minerals and bacte-
rial leaching of minerals has been discussed. Results indicating that the initial at-
tachment of T. ferrooxidans to minerals is governed primarily by hydrophobicity
have been presented and the theory of colloid chemistry that is used to interpret
these results outlined. The electrochemistry of bacterial leaching revolves around
the electrochemical model of dissolution, and its derivation has been presented in
detail. The electrochemical model has been shown to hold for three different dis-
solution reactions and shows much promise for the interpretation of the kinetics
of bacterial leaching.
References
1. Rossi G. Biohydrometallurgy. New York: McGraw-Hill, 1990.
2. Ehrlich HL, Brierley CL. Microbial Mineral Recovery. New York: McGraw Hill,
199 0 .
3. Crundwell FK. Mathematical modelling of batch and continuous bacterial leach-
ing, Chern Engin Biochem Engin J, 1994; 54:207-220.
4. Espejo RT, Ruiz P. Growth of free and attached Thiobacillus ferrooxidans in ore
suspension. Biotech Bioeng 1987; 30:586-592.
5. Boogerd FC, van den Beemd C, Stoelwinder T et al. Relative contributions of bio-
logical and chemical reactions to the overall rate of pyrite oxidation at tempera-
tures between 30 D C and 70 D C. Biotech Bioeng 1991; 38:109-115.
6. van Loosdrecht MCM, Lyklema J, Norde W et al. Influence of interfaces on mi-
crobial activity. Microbiol Rev 1990; 54:75-87.
7. Haddadin J, Dagot C, Fick M. Models of bacterial leaching. Enzyme and Micro-
bial Technology 1995; 17:290-305.
8. van Loosdrecht MCM, Zehnder AJB. Energetics of bacterial adhesion. Experientia
1990; 46:817-822.
9. van Loosdrecht MCM, Norde W, Lyklema J et al. Hydrostatic and electrostatic
parameters in bacterial adhesion. Aquatic Sciences 1990; 52:103-114.
10. van Loosdrecht MCM, Lyklema J, Norde W et al. Electrophoretic mobility and
hydrophobicity as a measure to predict the initial steps of bacterial adhesion.
Appl Environ Microbiol1987; 53:1898-1901.
11. van Loosdrecht MCM, Lyklema J, Norde W et al. The role of bacterial cell wall
hydrophobicity in adhesion. Appl Environ Microbiol 1987; 53:1893-1897.
12. Lyklema J, Norde W, van Loosdrecht MCM et al. Adhesion of bacteria to polysty-
rene surfaces. Colloids and Surfaces 1989; 39:175-187.
13. Blake RC II, Shute EA, Howard GT. Solubilization of minerals by bacteria: Elec-
trophoretic mobility of Thiobacillus ferrooxidans in the presence of iron, pyrite
and sulfur. Appl Environ Microbiol1994; 60:3349-3357.
14. Solari JA, Huerta G, Escobar B et al. Interfacial phenomena affecting the adhe-
sion of Thiobacillus ferrooxidans to sulfide mineral surfaces. Colloids and Sur-
faces 1992; 69:159-166.
15. Arredondo R, Garcia A, Jerez CA. Partial removal of lipopolysaccharide from
Thiobacillus ferrooxidans affects its adhesion to solids. Appl Environ Microbiol
1994; 60:2846-2851.
16. Devasia P, Natarajan KA, Sathyanarayana DN et al. Surface chemistry of
Thiobacillus ferrooxidans relevant to adhesion on mineral surfaces. Appl Environ
Microbiol1993; 59:4051-4055.
17. Adamson AW. Physical Chemistry of Surfaces. New York: John Wiley and Sons,
1986:258-290,379-420.
18. Hunter RJ. Foundations of Colloid Science. Vol. 1. Oxford: Oxford University Press,
1987.
19. Fowkes FM. Attractive forces at interfaces. Industrial and Engineering Chemistry
1964; 56(12):40-52.
20. van Oss q, Chaudhury MK, Good RJ. Interfacial Lifshitz-van der Waals and po-
lar interactions in macroscopic systems. Chemical Reviews 1988; 88:927-941.
21. Bockris JOM, Reddy AKN. Modern Electrochemistry. New York: Plenum Press,
1970.
22. Stumm W. Chemistry of the solid-water interface. Processes at the mineral-water
and particle-water interface in natural systems. New York: John Wiley and Sons,
1992.
23. Hunter RJ. Zeta potential in colloid science. London: Academic Press, 1988.
24. Hogg R, Healy TW, Fuerstenau DW. Trans Faraday Soc 1966; 62:1638-1651.
25. Ohmura N, Kitamura K, Saiki H. Selective adhesion of Thiobacillus ferrooxidans
to pyrite. Appl Environ Microbiol1993; 59:4044-4050.
26. Sam L, Rema V, Devasia P et al. Surface properties of Thiobacillus ferrooxidans
and its adhesion to mineral surfaces. Current Sci 1993; 65:974-978.
27. Costerton JW, Lappin-Scott HM. Behaviour of bacteria in bioflims. ASM News.
1989; 55:650-654·
28. Blenkinsopp SA, Costerton JW. Understanding bacterial biofilms. Trends in

Biotech 1991; 9:138-143.
29. Costerton JW, Lewandowski Z, de Beer D et al. Biofilms, the customised
microniche. J. Bacteriology 1994; 176:2137-1242.
30. Crundwell FK. Formation of biofilms of iron-oxidizing bacteria on pyrite. Miner-
als Engineering 1996; 9:1081-1089.
31. Sand W, Gerke T, Hallmann R et al. Sulfur chemistry, biofilm and (in)direct at-
tack mechanism-a critical evaluation of bacterial leaching. Appl Microbiol
Biotechnol 1995; 43:961-966.
32. Rojas-Chapana JA, Giesig M, Tributsch H. The path of sulfur during the bio-oxi-
dation of pyrite by Thiobacillus ferrooxidans. Fuel 1996; 75:923-930.
33. Pooley FD, Shrestha GN. The distribution and influence of silver in pyrite bacte-
rial leaching systems. Minerals Engineering 1996; 9:825-836.
34. Nicol MJ, Needes CRS, Finkelstein NP. Electrochemical model for the leaching of
uranium dioxide: 1. Acidic media. In: Burkin AR ed Leaching and Reduction in
Hydrometallurgy. IMM, Camelot Press. 1975:1-19.
35. Nicol MJ. The role of electrochemistry in hydrometallurgy. In: Hiskey JB, Warren
GW (eds) Hydrometallurgy Fundamentals, Technology and Innovations. Society
for Mining, Metallurgy and Exploration. 1993.
36. Crundwell FK. The influence of the electronic structure of solids on the anodic
dissolution and leaching of semiconducting sulfide minerals. Hydrometallurgy
1988; 21:155-190.
37. Crundwell FK. Effect of iron impurity in zinc sulfide concentrates on the rate of
dissolution. AIChE Journal 1988; 34:1128-1134.
38. McKibben MA, Barnes HL. Oxidation of pyrite in low temperature acid solutions:
rate laws and surface textures. Geochimica et Cosmochimica Acta 1986;
50:1509-1520.
39. Crundwell FK, Holmes P, Harvey P. Electrochemical aspects of bacterial leaching.
In: Doyle FM, ed. Electrochemistry in Mineral and Metal Processing IV. The Elec-
trochemical Society, 1996.
40. Verbaan B, Crundwell FK. An electrochemical model for the leaching of a sphaler-
ite concentrate. Hydrometallurgy 1986; 16:345-359.
41. Crundwell FK. Kinetics and mechanism of the oxidative dissolution of a zinc sul-
fide concentrate in ferric sulfate solutions. Hydrometallurgy 1987; 19:227-242.
42. Dry MJ. Kinetics of leaching a low-grade matte in ferric sulfate solution. Ph.D.
Thesis, University of the Witwatersrand, Johannesburg 1984.
43. Levenspiel O. Chemical Reaction Engineering. New York: J. Wiley and Sons, 1972.
44. Crundwell FK. The role of iron in bacterial leaching. Minerals Processing and
Extractive Metallurgy Rev 1996: submitted.
45. Konishi Y, Kubo H, Asai S. Bioleaching of zinc sulfide concentrate by Thiobacillus
ferrooxidans. Biotech Bioeng 1992; 39:66-74.
46. Garcia 0, Bigham JM, Tuovinen OH. Sphalerite oxidation by Thiobacillus
ferrooxidans and Thiobacillus thioxidans. Can J Microbiol 1995; 41:578-584.
CHAPTER 10
Optimization
of Biooxidation Heaps
A. Ian M. Ritchie
Introduction
L arge scale opencut mining is leading to a rapid reduction in reserves of oxide

gold ores at many locations around the world. At a number of these locations
the underlying primary sulfidic ores are refractory and often only a small fraction
of the contained gold can be won by traditional cyanide leaching. As oxidation of
some fraction (typically 50%) of the sulfide leads to much higher fractions of gold
recovered in subsequent cyanide leaching, a number of techniques have been de-
veloped to oxidize sulfidic gold ores. These techniques include roasting and pres-
sure oxidation with the tank biooxidation process, BIOX·, pioneered by Gencor
and being applied at an increasing number of mine sites.l Heap biooxidation is
similar in concept to tank leaching in that bacterially catalyzed oxidation of pyrite
forms the basis of the technique, but the ore is crushed rather than ground and the
heap rather than a set of stirred tanks is the reactor. An attraction of the technique
is lower costs with recent estimatesl - 3 quoting these in the range $1-6/t. Lower ore
grades can be treated economically with a consequent increase in ore reserves. For
example, access to heap leaching technology has allowed the Newmont Gold Com-
pany to boost its gold ore reserves by 37,000 kg. l A further benefit in the use of
heap biooxidation is a reduction in the quantity of sulfidic mine wastes with a
consequent reduction in the environmental impact of polluted drainage from the
wastes. These pollutants are generated within the wastes by the same sulfide oxi-
dation process which forms the basis of the heap biooxidation technique. As the
biooxidation heap technique requires about 50% of the sulfide sulfur to be oxi-
dized and since the most readily oxidized fraction is likely to be oxidized first, the
material that eventually goes to waste will not only contain less sulfide sulfur but
should be material which oxidizes more slowly. A reduction in oxidation rate does
in many cases lead to a reduction in the overall pollutant production rate in the
waste pile. 4
Biooxidation heap technology is currently at a much earlier stage of develop-
ment than is stirred tank biooxidation. A program of research and development
conducted by the Newmont Gold Company has progressed from laboratory based
measurements, through pilot scale heaps ranging in size from a few hundred tonnes
to about 30,000 tonnes, to a commercial sized heap of about 900,000 tonnes at
their Carlin mine site in Nevada, USA.t,5,6 Shield et aF describe a similar approach
conducted by Brohm Mining and Geobiotics culminating in a 4,300-tonne test heap
built at the Gilt Edge mine in South Dakota, USA. The 500,000 tonne per year heap
leach plant commissioned at the Mt. Leyshon gold mine in Queensland, Australia
in 19927 can claim to be the first commercial sized biooxidation heap for pretreat-
ment of a gold ore. While this ore was not a refractory sulfidic ore it was a super-
gene ore containing about 0.16% copper as secondary sulfides. These sulfides caused
unacceptably high cyanide consumption and their removal in the biooxidation
heap substantially improved the efficiency of gold recovery in subsequent cyanide
heap leaching.
The economics of heap biooxidation (see for example, ref. 3) require that about
50% of the sulfide sulfur in the heap has to be oxidized within a year or less. As
commercial sized heaps are likely to contain about a million tonnes or more of
material the heap has to be as high as possible to avoid profligate use of land area
at the mine site. The consequence of these constraints on properties required of
the heap material, on conditions within the heap and on heap operation are dis-
cussed in broad terms in the next section and in more detail later.
In the last few years a number of texts have appeared 8-lD which cover in detail
various aspects of biohydrometallurgy, in particular those processes that control
the rate of sulfide sulfur oxidation. Similarly recent publications l ,3 cover such as-
pects of heap biooxidation as the appropriate size for crushed ore and ways to
construct a heap that ensure that a high proportion of appropriate bacteria de-
velop in the heap in a short period of time. These aspects will therefore not be
covered in detail elsewhere in this chapter (see chapter 5). The contents of this
chapter are intended to provide a link between the science-based studies of
biohydrometallurgy and the needs of a mining engineer to construct a high per-
formance biooxidation heap. The third section contains a discussion on the oxida-
tion rate of sulfide sulfur and how this rate depends on parameters over which the
operator has some control in the construction of the heap and in the choice of
material comprising the heap. This is followed by a section which contains infor-
mation on mathematical modeling of a heap, how such modeling can be used to
identify key parameters and limitations on the range of values of these parameters
to ensure operational goals are achieved. Such modeling also shows that the heap
must have a certain geometry and be operated in a certain way to achieve opera-
tional goals. The fifth section contains information on how to optimize heap per-
formance. Parameters that can be monitored to ensure that conditions in the heap
are optimal and to quantify the fraction of sulfide sulfur oxidized are discussed in
the sixth section. Some important questions that are not adequately answered at
the current stage of development of biooxidation heaps are discussed in the last
section.
Basics
The purpose of a biooxidation heap to treat refractory gold ore is to oxidize the
sulfide sulfur of the material comprising the heap as rapidly as possible. The quan-
tity of the effluent and its chemical composition are of no real interest except in
that they provide information on the performance of the heap. In this sense a
biooxidation heap is very different from a copper leach heap where the quantity of
the effluent and its copper concentration are crucial in ensuring the cost-effective
operation of the heap and its associated copper extraction system.
Optimization of Biooxidation Heaps 203
As indicated above, a typical target is to oxidize about 50% of the sulfide sulfur
in a heap in about six months. Assuming that all of the material in the heap is
oxidizing at the same rate, then the time to complete oxidation is:
t = P,m~
, S 10.1
where;
ts =the timescale to oxidize the sulfide sulfur (say six months = 1.6 X 107
seconds);
ps = the bulk density of the heap material (say 1500 kglm3);
(ilsS = the mass fraction of sulfide sulfur in the heap material (say 1%);
S = the sulfide sulfur oxidation rate.
With ts set at six months and using the typical values above for the other pa-
rameters it follows from Equation 10.1 that S needs to be about 1 x 10- 6 kg(S)/
m 3/s or 6.7 x 10-10 kg(S)/kg(heap)/s. If the sulfide sulfur oxidation rate ofthe bulk
material is much lower than about 1 x 10-6 kg(S)/m3/s, then the economics of pre-
treatment of refractory gold ores in a biooxidation heap may no longer be attrac-
tive. It is useful to call the oxidation rate of the bulk material the intrinsic oxida-
tion rate (lOR) and to treat it as a property of the bulk material.
The oxidation of sulfide sulfur is frequently described by the following reac-
tions. 3
10.2
with the ferric ion being regenerated in the bacterial catalyzed reaction;
1 1
Fe 2+ + -0 + H+ ~ Fe 3t +-H 0 10.3
4 2 2 2
In a biooxidation heap the reaction described by Equation 10.3 is very rapid

with the oxidation rate of pyrite being controlled primarily by the rate of reaction
10.2. The ratio of the concentration of ferric ion to ferrous ion depends on the rate
constants of both reactions. If the pH is maintained close to some optimal value
throughout the heap and the oxygen concentration is uniformly high then the ra-
tio of ferric to ferrous ion will be approximately constant throughout the heap
with the actual concentration being determined by the total iron in solution in the
liquor infiltrating the heap.
It is, however, instructive to consider the situation where there is no regenera-
tion of ferric ion within the heap and by what distance into the heap all the ferric
ion in the liquor irrigating the heap is used up. An expression for this distance is:
x
qw[Fe 3t
=--=--...;;;;...
1
F e S F
where
qW = specific discharge rate for the liquor (the irrigation rate);
[Fe3+ 1= the concentration of ferric ion in the irration liquor;

EF = the ratio of the mass of ferric ion used to sulfide sulfur oxidized;
If the irrigation rate is 2.7 x 10- 6 mls (about 10 L/m 2 /hr) and contains about
20 gIL of ferric ion then Xp is about 4.4 m. It is clear therefore that in a biooxidation
heap higher than about 4 m, with the high bulk sulfide sulfur oxidation rate re-
quired it is essential that ferric ion is regenerated within the heap. It can be seen
from Equation 10.3 that this regeneration process requires access to oxygen.
The amount of oxygen required is given by the following equations which are
derived directly from Equations 10.2 and 10.3.
7
FeS2 +-02 +H20~ FeS04 + H 2S04 1O·sa
2
10.Sb
The appropriate combination of these two depends on the ratio of ferric to

ferrous ion.
It is instructive to estimate the flux of oxygen required to oxidize the sulfide
sulfur at the rate required. The expression required is similar to that of Equation
10.4 but what is of interest is qg the gas specific discharge required to provide oxy-
gen at the rate required in say a 10 m high heap.
10.6
Yo = the height of the heap;

£0 = ratio of the mass of oxygen used to sulfide sulfur oxidized;
JM = the intrinsic density of gas in the heap;
ro~ =mass fraction of oxygen in the gas phase.
For typical values of these parameters the gas specific discharge is about 7 x
10-sm/s. This value assumes that the oxygen is all used in the 10 m high dump. As
the oxidation rate is expected to decrease in some way with decreasing oxygen
concentration (see following section), it is reasonable to assume a value of about 1
x 10-4m/s as the gas specific discharge required to oxidize the sulfide sulfur in the
heap at the rate required.
The gas specific discharge is related to other parameters by the expression
K dP
q =--- 10·7
g /-lg dx
where;
K = the gas permeability in the heap;
Jl8 = the viscosity of the gas in the heap;
dP
= the pressure gradient in the heap
ax
Measurements in waste rock dumps l l have shown that the gas permeability
ranges from a high of about 1 x 10-9 m 2 to 1 x 10-13 m 2 or less. Bartlett3 quotes a
range of 10-12 to 10-10 m2 for crushed material which has subsequently been ag-
glomerated. Using expression 10.7 it follows that at the highest permeability a flow
rate of 1 x 10-4 mls can be maintained across 10 m with a pressure difference of
about 20 Pa. The pressure difference required is closer to 100 Pa as described in

the section on optimizing heap performance. Note that while pressure gradients
needed to achieve the required gas discharge rates are quite small, the magnitude
of the area of the heap dictates that the overall gas flow rate to the heap is large. If
gas is to be supplied to the heap then the technology is one which requires fans of
high throughput but low pressure.
The reaction described by Equation 10.5 is exothermic and about 1410 kJ/mole
is released. It follows that, at the high oxidation rates required in a biooxidation
heap, the heat production rate within the heap is also high. Making the simplify-
ing assumption that the lateral dimensions of the heap are sufficiently larger than
the height so that we can consider the heap to be one-dimensional, that the heat
generation rate is the same throughout the heap and that the top and bottom of the
heap are maintained at the same (ambient) temperature then it follows that the
maximum temperature is in the middle of the heap (if we neglect any cooling by
fluid flow through the heap). The temperature rise /).T above ambient as quoted by
Ritchiell is;
l1T= L2~S
10.8
41C th
L = height of heap (saY10m)
6 = heat released per mass of sulfide sulfur oxidized (2.2 x 107 J/kg(S»
1Cth = thermal conductivity of heap (about 2.0 WIm/°e)
Again making the assumption that the thermal conductivity of a biooxidation
heap will not be too different from that of a waste rock dump,ll the temperature
rise is well above 100°C at the oxidation rates required. It is clear therefore that the
heap must be cooled if the temperature within the heap is to be maintained within
the operating range of the bacteria which catalyze the oxidation process.
It is worthwhile understanding what physical processes are responsible for heat
removal from the heap. Two heat transport mechanisms come into play, thermal
conduction and advective heat transport by fluid flow through the pore space. There
are two fluids involved in advective transport, the liquor infiltrating the heap and
gas flowing through the heap. In both cases the advective heat transport is propor-
tional to the quantity af where;
where f indicates fluid.

Using appropriate values of the parameters for water and for gas;
af = 13 J/m'/s/oC for water
af = 0.13 J/m'/s/oC for gas.
It can be seen that water appears more effective in transporting heat in the
system than is gas but if the infiltration rate is decreased and the gas discharge rate
increased, both of which are likely in practice, then the two values can become
similar. Note that if the fluid flow rates are sufficiently low, the fluids take on the
temperature at the boundaries and then neither fluid actually carries any heat out
of the system. Rather the process is one where the temperature profile is shifted
towards a boundary by advective processes and heat loss is then by conduction
along a greater temperature gradient.
The Intrinsic Oxidation Rate

It is apparent from the discussion above that the oxidation rate of the sulfide
sulfur is a basic parameter needed in quantifying the performance of a biooxidation
heap. It can be defined either as the rate of sulfide sulfur oxidation or as the rate
that oxygen is used up in sulfide sulfur oxidation. These are conceptually the same
and related by a scale factor defined by the stoichiometry of an equation such as
10.5. The most convenient units for this parameter in mathematical modeling of a
biooxidation heap is mass/unit volume/unit time but in some cases, particularly
where the parameter is being measured in laboratory experiments the units mass/
per mass of material/unit time are more appropriate. In the rest of this chapter the
units used will be kg(S)/m 3/s, for sulfide sulfur oxidized and kg(O)/m3/s for oxy-
gen used. Unless otherwise stated the scale factor relating to these two is 1.75 and it
comes from Equation 10.5a.
The use of such a parameter assumes some averaging over a "representative
volume" and is, in this sense, a bulk property of the material comprising the heap.
In principle it should be related to the oxidation rate of sulfide at the microscopic
level and evaluated by modeling the oxidation rate at the particle level and inte-
grating over the particle size distribution. While this is a useful way to see what the
dependence other such parameters as the pore gas oxygen concentration and the
remaining sulfide sulfue2 may be, the results achieved do not always give good
agreement with measured rates for the bulk material (see for example refs.13-14).
A direct measurement of the bulk parameter is therefore the preferred route at this
stage of our knowledge.
This property of the bulk material is called the intrinsic oxidation rate (lOR).
In some cases, particularly in modeling heap performance, the value quoted is the
maximum value for the parameter when other parameters are at the value where
the sulfide sulfur oxidation rate is a maximum. Table 10.1 lists a number of (maxi-
mum) lOR values and indicates what the consequence of these is in a large, well
aerated heap.
The magnitude of the lOR depends on a number of parameters which define
the chemical, physical and microbiological conditions within the heap. For ex-
ample, it has been reported 8 (and chapter 5) that the rate depends on such chemi-
cal conditions as pH, Eh, the ratio of ferric to ferrous ion concentrations, the par-
tial pressure of oxygen in the pore space and the chemical form of the sulfide, on
such physical conditions as temperature, particle size distribution, porosity of the
particles, and crystalline form of the sulfides and on the type and numbers of mi-
croorganisms in the heap.
It must be emphasized that if mathematical modeling is to be useful in opti-
mizing the performance of a biooxidation heap then the model must describe the
way that each parameter evolves in space and time if that parameter appears in the
functional dependence of the lOR. For example, if it is believed that the oxidation
rate is a function of the type of microorganisms present and the number of each
type of organism then an equation describing the evolution of each species must
appear in the model. However,it has been shown 15 that the oxidation rate of a 1 g
sample of pyrite ground to -58+91 )lm and added to 50 mL of 1/1O-strength 9 K
salts in a 250 mL reactor did not increase if bacterial numbers were greater than
about 106/mL when other conditions were optimized. Moreover, assuming the dou-
bling time of the microorganisms to be about 1 day, the time for the population to
increase by a factor of a 1000 is just ten days. It follows that if conditions in the
~§.
~.
I::.
Table 10.1. The significance of the magnitude of the IOR Ig·;:;
~
lOR Relevance of Rate Time to Use Time to Use Time to Use Approx Sulfate Copper b:l
c·
c
(kglm/s) Up Pyrite (y) Up Carbonate Up Initial Pore Temp Rise Cone. Cone. ~
Space Oxygen (OC) (mg/L) (mg/L) 5..:
I::.
Note 1 Note 2 Note 3 Note 4 Notes .....
c·
;:;
10-5 Biotic rate seen in laboratory 0.167 2·4d

experiments
I~~
'"
10- 6 Rate needed in biooxidation heap 1.67 3·4w 0.9d
Rate for chemical oxidation
10-7 Effectively infmitely high rate in 16·7 0.66y 1.3 W
a waste rock dump
10-8 Typical rate found in waste 167 6.57Y 13.1 W -10 21,600 430
rock dumps
10-9 Typical low rate found in waste 1670 6S·7Y 2·52 Y -1 2160 43
rock dumps
10-10 Marginal environmental problem 16,700 657Y 25· 2 Y 216 4·3
Note 1 Bulk density 1500kglm3; sulfur density as pyrite 30
kglm3; oxidation rate maintained throughout dump
Note 2 Carbonate density 3.45 kglm3 (0.23%)
Note 3 Gas filled pore space 0.3; oxygen concentration in air 0.265 kglm3
Note 4 Sulfate assumed conserved; dump height 20 m; infiltration 1.59 x 10-8m/s (0.5 mly)
Note 5 Sulfate to copper concentration in drainage 50:1
t.>
~
heap are close to those ideal for bacterial growth and provided the heap is "spiked"
with suitable microorganism on construction, the microbial population will rise
to its maximum in a comparatively short period of time and stay there unless some
parameter which effects their viability moves out of the near optimal range.
In the modeling described here the lOR is assumed to be a function only of the
sulfide sulfur concentration in the heap material, the temperature and the pore gas
oxygen concentration.
Dependence on Sulfide Sulfur Concentration

A large number of mathematical models have been developed to describe the
dissolution of a solid reactant disseminated throughout a porous particle. Many of
these models were developed to describe the heap leaching of copper ore with those
of Harris '6 in 1969 and Braun et a1'7 in 1974 being typical of the earlier models. A
common theme has been the concept of a "shrinking core" of reactant in a spheri-
cal particle. Box and Prosser 'S derived a more general model for leaching of a num-
ber of minerals involving more than one reagent. The advent of fast computers has
allowed more detail to be included in such models (see for example ref. 19).
It can be demonstrated 2o that replacing the rather complex dependence of the
lOR on sulfide sulfur density in the shrinking core model with a simple linear
dependence on sulfide sulfur density leads to little change in the evolution of heap
performance.
It has also been demonstrated 21 that while details in oxygen profiles changed
with changing complexity of the model used to describe the IOR, the overall oxi-
dation rate in the heap was very insensitive to the details of the IOR model. Since
in the present study it is the overall performance of the heap that is of interest, the
model for the IOR described in this chapter has a simple functional dependence
on the sulfide sulfur density. The lOR is assumed either to decrease linearly as the
sulfide sulfur decreases or to be described by Monod kinetics. It is believed that
the actual dependence for any given material falls somewhere between these two.
If it is further assumed that the maximum lOR is known for the material, the de-
pendence of the IOR on decreasing values of the sulfide sulfur density is as shown
in Figure 10.1. Note that as an objective of a biooxidation heap is to have a similar
oxidation rate over as large a volume of the heap as possible and as a goal is to
oxidize about 50% of the sulfide sulfur, it is likely that the functional dependence
of the IOR on the sulfide sulfur density at very low densities does not need to be
well known.
Dependence on the Pore Gas Oxygen Concentration

Since water and gas transport rates are expected to be small within a
biooxidation heap (about 3 x 10-5 mls and 3 x 10-3mIs, respectively) it is reasonable
to expect that the dissolved oxygen concentration in the thin film of water in water
filled pore space of the heap is in equilibrium with the oxygen concentration in the
pore space. If the rate of oxidation of the sulfide sulfur material is proportional to
the dissolved oxygen concentration, as it is in the shrinking core model,22 then the
rate is also proportional to the pore gas oxygen concentration. Hammack and
WatzlaF3 indicated that the rate in pyritic mine waste is proportional to the pore
gas oxygen concentration when the reaction is largely abiotic but can be described
by Monod kinetics if bacterial catalysis is important. The dependence of the lOR
on the pore gas oxygen concentration in the modeling described here takes either
1.0
monod
0.8
a: 0.6
Q
.,>
~
a;
a: 0.4
linear
0.2
0.2 0.4 0.6 0.8 1.0
Relative mass fraction (w!fiJ.... )
Fig. 10.1. Two forms for the functional dependence of the lOR on the mass fraction of
sulfide sulfur remaining and of oxygen in the pore space gas. Reprinted with permis-
sion from Ritchie AIM, In: BIOMINE '94. International Conference and Workshop, Perth
1994, Adelaide: The Australian Mineral Foundation, 1994:15-1-15.19.
of the two forms shown in Figure 10.1. Again, it is likely that the details of the
dependence on oxygen concentration are not required at low oxygen concentra-
tions. As will be shown below it is practical in a heap with a well chosen geometry
to ensure that there is a plentiful supply of oxygen throughout the heap.
Dependence on Temperature
The temperature dependence has two main components. The first reflects the
properties of the microbial ecology of the system. It has been known for some time
that the mesophilic bacteria which catalyze sulfide sulfur oxidation have an opti-
mum at about 35°C and lose viability at temperatures much above 40°C,24 while
moderate thermophiles have an optimum at about 50°C and lose viability much
above 55°C.15 The relative numbers of mesophiles and moderate thermophiles de-
pends on the temperature profile in the heap and the oxidation rate catalyzed by
the two different species. As details of these rates are not readily available and as
the overall heap performance is almost certainly sensitive to gross features of the
temperature dependence rather than the detailed dependence, the simple depen-
dence used first by Cathles 26 is employed in the modeling reported here. This as-
sumes a constant rate of oxidation up to a temperature T sick at which the rate starts
to decrease, reaching zero at a temperature Tkill. The temperatures Tsick and Tkill
are chosen to reflect the expected microbial population in the heap.
With Tsick and Tkill set at say 40°C and 60°C this simple form of the temperature
dependence assumes that the rate of oxidation catalyzed by mesophiles and mod-
erate thermophiles are not very different. If there is evidence to suggest that the
rates are substantially different then it is comparatively easy to generate a func-
tional form reflecting this difference but still using the concept of a rick and :till to
indicate the changeover in the microbial population from one species to the other.
The second component reflects the activation energy associated with the sul-
fide sulfur oxidation reaction. It has been shown 27 that where pyrite is oxidized
under well controlled conditions the temperature dependence of the reaction fol-
lows the Arrhenius formulation with an activation energy of about 70 kJ/mole. It is
debatable that this formulation will describe the temperature dependence of the
lOR since other mechanisms such as diffusion through a water film or through to
a reaction front in a particle may obscure the temperature dependence of the sul-
fide oxidation reaction. If this is so then the activation energy Ea will be smaller
than the 70 kJ/mole of the pyrite oxidation reaction.
The temperature dependence of the lOR used in some simulations reported
below is shown in Figure 10.2. The overall dependence of the lOR on pore gas
oxygen concentration, sulfide sulfur concentration in the heap material and heap
temperature takes following forms;
s = (J'
1 1 (J' 2
g
ats a T ex _ [ E a
+ OJ~ (J' 3 + OJ~ ( ) P R(B + T)
OJ 0 1 10.10
Sz = (J' _ g
OJ 0 OJSS a(T)exp- [E]
_ 0 10.11
gOJ00 OJSSo R(B + T)
where;
au C1 = the lOR at maximum oxygen concentration in the pore gas and at
the initial sulfide sulfur concentration;
1
0,0
2
=parameters in Monod kinetics defining where the rate has dropped
to half of its maximum rate;
rot = mass fraction of oxygen in the gas phase;
roto = mass fraction of oxygen in air;
ro~ = mass fraction of sulfide sulfur
ro~o = initial mass fraction of sulfide sulfur
T = the temperature in DC;
a(T) =a function which is unity at T =T'ick and decreases smoothly to zero
at T=TkiI\;
e = 273°C;
E. = an activation energy (kJlmole);
R = the gas constant (8.315 J/mole)
Mathematical Simulation of Heap Performance

Mathematical simulation of polluted drainage from sulfidic waste, pregnant
liquors from leach heaps and sulfide oxidation in biooxidation heaps all have in
common the need to quantify the oxidation of sulfide in a large mass of material.
Most authors have followed the approach pioneered by Cathles and his cowork-
ers. 28 -30 Pantelis and Ritchie31-38 developed a two-dimensional computer code which
takes account of heat transport, gas transport, water transport and sulfide con-
20
Ea=64kJ
15 -
~
0
N
.,
~ 10
c:
0
~
:2
)(
0 Ea = 40 kJ
5
Ea= 20 kJ
O ~~~~~L-~~~~L-~~~~L-~~~L-L-L-~~~
10 20 30 50 60
To T""
Temperatu re ( · C)
Fig. 10.2. The effect of varying the magnitude of the activation energy on the tempera-
ture dependence of the lOR. Reprinted with permission from Ritchie AIM, In: BIOMINE
'94. International Conference and Workshop, Perth 1994,Adelaide: The Australian Min-
eral Foundation, 1994:15-1-15.19.
sumption and has been applied to modeling pyritic waste dumps, biooxidation
heaps and to leach heaps. Casas et al39 have developed a similar model where the
primary goal is to model copper leach heaps. Jaynes et al,4 0 .41 Guo and Parizek41
and Scharer et al 43 have developed models where the primary goal is to model
deposits of pyritic mine wastes.
Cathles and Apps's showed that convection driven by thermal gradients in a
heap generated by the heat produced in sulfide sulfur oxidation was an important
gas transport mechanism. Pantelis and Ritchie 31 further showed that convection
added significantly to the overall oxidation rate (the global oxidation rate or GOR)
only if the gas permeability was greater than about 10-9 m' and that convection
was a process which started at the edge of a heap and works its way into the heap
interior. Casas et al 39 have shown similar results. If a heap has been constructed in
such a way that a no flow boundary condition is applicable at the base then Pantelis
and Ritchie33 have shown that 50% of the sulfide sulfur can be oxidized in about six
months only if the gas permeability is very high. Moreover, the oxidation rate
throughout the heap is very nonuniform. Based on data from measurements in
waste rock dumpsll the very high values of gas permeability required are not likely
to be achievable and the nonuniform oxidation rate through the heap is not attrac-
tive from an operational point of view.
It is convenient to describe a heap with a no flow condition at the base as a
closed-based heap. An open-based heap is one where the base is designed to allow
gas to flow through and, in particular, to allow a forcing pressure to be applied at
the base. Pantelis and Ritchie 35 have shown that in such a heap gas flow is substan-
tially one-dimensional and that a one-dimensional model suffices to simulate con-
ditions in such a heap. Of considerable practical importance is the fact that an
acceptable fraction of sulfide sulfur can be oxidized in about six months in a 10 m
high heap with a gas permeability of 10-10 m 2 and a forcing pressure of only 1000
Pa. This gas permeability is likely to be achievable particularly in a biooxidation
heap where there is some scope to select the particle size distribution and to stack
the material in such a way so as to avoid compaction.
Figure 10.3 shows the space and time dependence of some of the parameters of
importance in assessing heap performance for a heap with the properties given in
Table 10.2. Of note is the relatively uniform sulfide sulfur oxidation rate and oxy-
gen concentration profiles through the heap. This means that the material is oxi-
dizing at close to the maximum rate possible throughout the heap. Figure 10.4
shows the fraction of sulfide sulfur oxidized (GO) and the overall rate of sulfide
sulfur oxidation in the heap (GOR) as a function of time. The decrease in the over-
all oxidation rate with time is a consequence of sulfide sulfur being consumed.
Optimizing the Performance of an Open-based Heap

The conceptual model of a biooxidation heap described in earlier sections as-
sumes that parameters important in controlling the sulfide sulfur oxidation rate
are the oxygen availability, sulfide sulfur availability and temperature. It has been
assumed that other parameters that might effect the oxidation rate such as pH,
total iron concentration, nutrient availability and water availability are maintained
in a range where the oxidation rate is close to optimal. Optimizing the heap per-
formance then becomes a question of maintaining the pore gas oxygen concentra-
tion and temperature in a range where the oxidation rate is close to maximum.
The mathematical model quantifies the interaction between the various mecha-
nisms which control the oxygen concentration and temperature and hence it is
possible to quantify the sensitivity of the biooxidation heap performance to the
physical parameters which effect these mechanisms. The primary impact of these
parameters is either on the pore gas oxygen concentration profile or on the tem-
perature profile. The impact of these parameters on heap performance has there-
fore been grouped below under the headings of impact on oxygen profile and im-
pact on temperature profile.
Impact on Oxygen Concentration Profile
Gas Permeability and Forcing Pressure

Gas transport in an open-based heap is dominated by advection. There is there-
fore no need to discuss the impact of the gas diffusion coefficient or how accu-
rately it needs to be known. The physical parameters which determine the rate of
gas discharge are the gas permeability and the forcing pressure applied at the base
of the heap. As discussed above, there is reason to believe on the basis of measure-
ments in waste rock dumpsll that it may be difficult to achieve gas permeabilities
much above 10-9 m 2 and that 10-10 m 2 may be an appropriate figure to use in initial
feasibility design. Figure 10.5 shows the GO as a function of forcing pressure in a
heap with the properties shown in Table 10.2. Figure 10.3 shows that at a forcing
pressure of 1000 Pa both the oxygen concentration and the sulfide sulfur oxidation
rate are reasonably uniform throughout the heap. As the forcing pressure decreases,
10 10
7 7 month 6
~6 ~6
-
E E
E 5 .c 5
Cl Cl
·iii ·iii
J: 4 J: 4
3 3
2 2
Temperature ( ·C) Sulfur oxidation rate ( kg/m3/s )
10
10
9
9-
8 month 6
8 month 6
7
7
~6
~6 E
E
E 5
E 5 Cl
·iii
Cl
·iD J: 4
J: 4
3
3
2
2
0.002 0.004 0.006 0.008 0.010

0.05 0.10 0.15 0.20
Sulfur mass fraction (kglkg)
Oxygen mass fraction (kg/kg)
Fig. 10.3. The space and time dependence of the temperature, sulfide sulfur oxidation
rate, oxygen mass fraction in the pore space and remaining sulfide sulfur in a
biooxidation heap with the properties shown in Table lO. 2 .
1.0x l 0 '" 1.0
0.9
- - - GO
GOR 0.8
- 0.7
~ ..-/- 0.6
~::> 5.0xl 0'" // - 0.5 0
(!l
'"
Oi /
/
"'"
[(
0.4
o
(!l 0 .3
Fig. 10.4. The fraction of
sulfide sulfur oxidized 0.2
(GO) and the overall rate
of sulfide sulfur oxidation 0 .1
(GOR) as a function of
time in a biooxidation 0.0
heap with the properties o 2 3 4 5 6
shown in Table 10.2. time (months)
Table 10•.2. Physical and material properties of a typical biooxidation heap
Property Description or Value Units
Height 10 m
Bulk density of heap 1500 kg/m3
Initial sulfide sulfur fraction 0.01 %
Form for IOR linear form
0 1 X 10-6 kg(S)/m3/S
Activation energy 0 kJ/mole
Forcing pressure at heap base 1000 Pa
Ambient temperature (temperature
applied at boundaries) 0 °C
Irrigation rate 2.7 x 10- 6 mls
0.7
0.6
U>
£;
c:
o
E
<0
o(!)
0.1
0.00L...l....................
200L...l.................4...0....0..................6-0....0..................S
- 00
'-'-...............1-'
000 Fig. 10.5. Variation of the frac-
tion of sulfide sulfur oxidized
Po ( Pa) (GO) with forcing pressure.
the flux of gas through the heap decreases and the oxygen concentration towards
the top of the heap is lowered due to consumption in lower parts of the heap. The
overall oxidation rate is decreased and hence the fraction of sulfide sulfur oxidized
(GO) at six months is lower.
Simulations at gas permeabilities of 10-11 and 10-9 and forcing pressures of 10,000
and 100 Pa, respectively produce results very similar to those with a gas perme-
ability of 10-10 m' and 1000 Pa.At a gas permeability of 10-8 m ' the indications are
that no forcing pressure is required as convection provides adequate advective gas
flux. From a practical point of view the apparent ability to compensate changes in
gas permeability with changes in forcing pressure implies that achieving a par-
ticular gas permeability in a heap is not as crucial as was the case with a closed-
based heap.
Monod and Linear Kinetics

The use of Monod kinetics to describe the lOR means that oxidation is main-
tained at a high level over a large range of oxygen concentration. This further means
that if a heap is well aerated the oxidation rate is uniform throughout the heap and
the overall oxidation rate is higher than if linear kinetics is used. This is demon-
strated in Figure 10. 6 which shows the GO in a heap as a function of time for the
two forms for the lOR and a high and low forcing pressure. This heap has similar
but not exactly the same properties as the one specified in Table 10.2. It can be seen
that the use of Monod rather than linear kinetics leads to a predicted increase of
0.6
a) linear p= 1000 Pa
0.5 b) linear p= 300 Pa
"C
CD
N
'5 c) monod p= 1000 Pa
';(
0 0.4
d) monod p= 300 Pa
,..~
:::J
'" 0.3
'0
c:
0
~
II..
0.2
0.1
0.0
0 2 3 4 5 6
Time (months)
Fig. 10.6. The effect on the performance of a biooxidation heap using linear and Monod
kinetics for the rate dependence on oxygen concentration. Reprinted with permission
from Ritchie AIM, In: BIOMINE '94. International Conference and Workshop, Perth
1994, Adelaide: The Australian Mineral Foundation, 1994:15.1-15.19.
about 40% in the GO at six months. Figure 10.7 shows the oxidation rate profile
and demonstrates that while the GO at 300 Pa and assuming Monod kinetics ap-
pears to give a good result, the heap is not adequately aerated. This demonstrates
the interaction between the different mechanisms and the usefulness of simula-
tions in identifying a potential shortfall in heap performance.
Impact on Temperature Profile
Irrigation Rate
When the irrigation rate in a heap with the properties indicated in Table 10.2 is
reduced by a factor of 6.7, the GO at six months is reduced from about 62% to 50%.
The impact on the temperature profile and hence on the sulfide sulfur oxidation
rate profile is very marked, as can be seen by comparing Figures 1O.8a and 10.3.
The reason for the decrease in the oxidation rate towards the center of the heap is
that temperatures in that region are in the range between Tsick and Tkill. In this
simulation the ambient temperature is assumed to be zero. This ambient tempera-
ture is lower than is the case in many parts of the world. As an increase in ambient
temperature just adds to the temperature in the heap, it is clear that with a higher
ambient temperature an even larger volume of the under cooled heap would be in
a temperature regime where oxidation rates are reduced. Irrigation has therefore
the very important function of keeping temperatures in the heap within an appro-
priate range.
10 ~--------~~-----------r~~----------------mr
E 6
a) linear p= 1000 Pa
~
'"
'iii
:I:
b) linear p= 300 Pa
4 c) monod p= 1000 Pa
d) monod p= 300 Pa
Sulphur oxidation rate ( ~glm3fs )
Fig. 10.7. The effect on the sulfur oxidation profile of variable forcing pressure when
using the two forms for the rate dependence on oxygen concentration. Reprinted with
permission from Ritchie AIM, In: BIOMINE '94. International Conference and Work-
shop, Perth 1994, Adelaide: The Australian Mineral Foundation, 1994:15-1-15-19.
High rates of irrigation can, however, be counterproductive. The gas perme-

ability is a very strong function of the gas fllied pore space of the heap material. As
the irrigation rate increases the fraction of water filled pore space increases and
consequently, the fraction of gas filled pore space decreases. This phenomenon
has been investigated33 for closed-based heaps of high intrinsic permeability. In
general there is an optimum irrigation rate with the optimum depending on the
magnitude of the intrinsic permeability and the value for Tkill.
The Impact of Tsick and Tkill

It follows from the discussion in the previous subsection, that if the irrigation
rate is low and temperatures within the heap are high, the lower Tsick and Tkill the
larger the volume of the heap with a lowered oxidation rate, the lower the overall
oxidation rate and the longer it takes to achieve a target GO. Table 10.3 shows the
impact of reducing Tsick and Tkill from 40° and 60°C respectively to 30° and 50°C
based on simulations for a heap with properties similar to those in Table 10.2. Fig-
ure 10.8b shows the impact of decreasing Tsick and Tkill under low irrigation
conditions.
Impact of the Activation Energy

A high activation energy means a rapid increase in the oxidation rate with in-
creasing temperature. With an activation energy of 70 kJ/mole the increase is about
a factor of two for every ten degree increase in temperature. The impact of increas-
ing Ea is initially one of positive feedback; as the temperature rises so does the
Fig.lO.8. The effect on the sul-

fur oxidation rate profile of a) 10
decreasing the irrigation rate
(lO.8a) and decreasing Tsick 9
and TkiU at low irrigation rate
(lO.8b). 8
6
E
ECl 5
·S
::c 4
O~~~~~~~~~~~~
5.Ox10-7 1.0x10-8
Sulfur oxidation rate ( kg/m3/s
7 month 6
6
E
E 5
.2>
Q)
::c 4
5.0x10-7 1.0x10-e
Sulfur oxidation rate ( kg/m3/s

Optimization ofBiooxidation Heaps 219
Table 10.3. The impact of varying Tsick and Tkill on the fraction of sulfide
sulfur oxidized
Irrigation Rate T.ick Tkill GO at 6 Months

(m/s) (0C) (0C) (%)
2.7 X10-6 40 60 62
4.2X- 7 40 60 50
4.2 X-7 30 50 44
oxidation rate, the heat output rises and the temperature rises even faster. If the
heap is not adequately cooled then the effect at later times is similar to that of a
decrease in Tsick and Tkill with a large volume of the heap oxidizing at less than
optimal rates.
The time evolution of temperature, oxygen and oxidation rate profiles also de-
pend on the boundary condition assumed to apply on the temperature at the base
of the heap. If the irrigation rate is high enough that the highest temperatures
occur towards the base of the heap but low enough that it is reasonable to assume
that the temperature at the base is ambient temperature (Dirichlet conditions), the
consequence is a large temperature gradient in the region of the base. This phe-
nomenon is exacerbated if Ea is large. Such a situation requires that the algorithms
used in the numerical solution of the equations are robust. The sulfide sulfur oxi-
dation can also evolve in an unusual way. If Ea is large, the highest oxidation rates
occur near the base of the heap early in the life of the heap, andthe sulfide sulfur
decreases rapidly in that region leading to only a small removal of oxygen in the
gas passing through the region. In some cases this gives rise to a situation where
oxidation of sulfide sulfur occurs mainly near the base early in the life of the heap
and progresses upwards at later times in heap's life. Such a situation is not the ideal
of obtaining a near uniform oxidation rate throughout the heap.
The use of boundary conditions (Neumann), where it is assumed that the tem-
perature on the boundary is set more by the water flow rate through the base,
should lead to less of a temperature gradient near the base. Such a move might
improve confidence in the numerical solutions, but will, if anything, lead to even
more nonuniform oxidation rate and oxygen concentration profiles.
Monitoring the Performance of a Biooxidation Heap

The parameters that should be monitored to assess the performance of the
biooxidation heap fall into two sets. The first set are parameters used to assess that
conditions within the heap are consistent with optimal sulfide sulfur oxidation
rates. The second set are to estimate the fraction of the initial sulfide sulfur that
has been oxidized.
Physical and Chemical Conditions Within the Heap

It is clear from the discussion above that it is important to ensure that the pore
gas oxygen concentration and the temperature throughout the heap are in a range
compatible with a high sulfide sulfur oxidation rate. These two parameters can be
monitored by installing a sufficient number of probe systems in the heap, each

probe system allowing the temperature to be measured as a function of depth and
pore gas samples to be collected from different depths. A number of systems have
been described,44-46 generally in the context of measuring pore gas concentration
and temperature profiles in tailings or waste rock dumps. Figure 10.9 depicts one
such system. The string of temperature sensors down the central pipe has the ad-
vantage that it can be placed or modified to cover regions where there is a pro-
nounced temperature gradient more closely. The vertical orientation is preferred
to other orientations as there is less of a tendency for the ends of the gas sampling
tubes to be clogged. The sand backfill also reduces the risk of clogging, particu-
larly if the heap material has a high clay content and infiltration rates are high. The
sand backfill will have the same temperature and pore gas properties as the sur-
rounding heap. The gas sampling points could also be used to monitor gas pres-
sure to note any change in the gas permeability of the heap material as it oxidizes.
Such measurements have not been reported but the required sensitivity should be
achievable with solid state pressure transducers currently available.
A further advantage of the central tube is that it could be used as access for
neutron moisture meters and gamma density probes if there is some need to as-
certain whether or not parts of the heap are water saturated. If it is planned to use
such probes then the central tube should not be made of PVC which has high neu-
tron absorption properties. Often the simplest indicator of water saturation is the
ease with which gas can be drawn through the gas sampling tubes. Their becom-
ing blocked from time to time is an indicator that the material in their vicinity may
be water saturated.
Monitoring the Extent of Sulfide Sulfur Oxidation

A knowledge of the dependence of the sulfide sulfur oxidation rate on param-
eters such as oxygen concentration, temperature and sulfide sulfur concentration
together with simulations from a mathematical model of the heap should provide
an estimate of the fraction of sulfide sulfur oxidized at different times after con-
struction of the heap. With our current state of understanding of the mechanisms
affecting sulfide sulfur oxidation rates in a heap the role of a model is, however,
more to ensure that conditions have been optimized as far as possible and to iden-
tify key parameters that affect heap performance than to predict the fraction of
sulfide sulfur oxidized.
It is usual to build the heap on a low water permeability pad in order to collect
heap drainage. There are a number of reasons for doing this, including minimiza-
tion of the environmental impact of drainage, allowing water recycle and so re-
ducing water demand at the mine site, using chemical analysis of the effluent to
provide information on heap conditions and controlling the chemical content of
the liquor used to irrigate the heap. The pad must be designed to avoid saturation
in as large a volume of the heap pore space as possible. It is also sensible to design
it in such a way that both the rate of flow and the chemical composition of drain-
age from different sections of the heap can be monitored.
Pyrite is usually the largest fraction of sulfide sulfur in a biooxidation heap, the
products of oxidation being iron and sulfate. Chemical analysis of iron and sulfate
concentration in drainage and measurement of the rate of flow of drainage pro-
vide data on the mass flux of iron and sulfate in drainage. Subtraction of the mass
flux of the iron and sulfate flux in irrigation water should provide the production
thermistor strinq
bundle of tubes -==SttW-w
gas sampling tubes and

thermistors at 1 meter
intervals
I Ion gas sampling
tube
thermistor
thermistor
string
sand backfill i
drilled hole
water-tight cap
PVC pipe
Fig. 10.9. Schematic diagram of a probe system to measure temperature and pore gas
oxygen concentration in a heap.
rate of these two species and hence the overall oxidation rate of sulfide sulfur in
the heap. Iron chemistry is complex and as some sulfates will be insoluble at the
concentrations expected in pore water, care has to be exercised in terms of the
mass flux of iron and sulfate in interpreting the overall oxidation rate. Water loss
both by evaporation at the heap surface and by water vapor carried out in gas flow
means that the irrigation rate needs to be monitored.
Effluent gas composition can in principle be used in a way similar to effluent
water composition to estimate the overall oxidation rate. The oxygen concentra-
tion in gas samples taken from locations near the top of the heap should provide lI;
measure of the oxygen concentration in gas effluent. The input to the gas reticula-
tion system, which maintains the required forcing pressure along the base of the
heap, is a measure of the total gas flow. An estimate of the rate of oxygen consump-
tion and hence the overall oxidation may then be calculated. Such an estimate as-
sumes no gas leakage in the gas reticulation system. For practical reasons this is
more difficult to achieve than ensuring that all the water infiltrating the base of the
heap reports to the effluent collection point. A more direct measure is to deter-
mine the sulfide sulfur remaining in samples of material taken from the heap. Such
a method suffers all the drawbacks of spot sampling techniques.
Summary
At this stage of development of the understanding of the operation and perfor-
mance ofbiooxidation heaps, it is likely that application of a number of the meth-
ods outlined above is required to form a reliable estimate of the fraction of sulfide
sulfur oxidized at different times after heap construction. A large number of probe
systems can be deployed over the heap as the cost of materials for the probe system
is small. It should also be emphasized that as biooxidation heap technology is at
an early stage of development good quality information on performance is crucial
if the technology is to advance. It is therefore vital to ensure that adequate re-
sources are made available to ensure collection of high quality data. Similarly, the
cost of analyzing a number of chemical species in water samples can be high but a
judicious choice of sampling period, a well thought out protocol for the chemical
species to be measured and use of data from within the heap to identify suitable
times to carry out a full suite of chemical analysis on effluent will provide the in-
formation at a reasonable cost.
Outstanding Issues
Improvement of Data on lOR
The lOR is a crucial parameter in modeling the performance of a biooxidation
heap. The information required is the functional dependence on the pore gas oxy-
gen concentration, on the temperature and on the sulfide sulfur content of the
material to be biooxidized over the range of variation expected for these param-
eters in a biooxidation heap. Moreover, the system under study must be one where
the oxidation rate of the bulk material is measured or where the rate for bulk ma-
terial can be quantified with confidence from the measurements. The fraction of
sulfide oxidized as a function of time has been the traditional way to assess the
amenability of material for heap leaching or to compare the efficacy of different
species of bacteria. Such a quantity is the integral of the parameter required and
must be manipulated to quantify the lOR. The nature of the manipulation intro-
duces uncertainties which could be avoided readily if the rate were measured
directly.
A direct measure of the oxygen consumption rate is better than a rate inferred
from sulfate or iron production rates. This is particularly so if the measurement is
made in bulk material unless the experimental protocol is such that adequate ac-
count can be taken of iron or sulfate retained in the mass of material. Quantifica-
tion of sulfide sulfur seems to be a continuing problem. As it is usual to infer the
quantity of sulfide sulfur left in a material from the difference between the initial
sulfide sulfur and the total quantity oxidized, at the time an oxidation rate mea-
surement is made, it follows that a systematic overestimate of the initial sulfide
sulfur follows through as a high value of the remaining sulfide sulfur.
The temperature dependence of the lOR of bulk material is also an issue which
needs to be clarified. It is possible that the dependence is material specific but
there is little information available in the literature on which to base an assess-
ment. As setting up a system to measure the temperature dependence is straight-
forward and as given the marked impact that even the modest temperature de-
pendence (a doubling in rate for each ten degrees increase) can have on the per-
formance of a biooxidation heap, work on temperature dependence would be
rewarding.
As discussed in the third section, the functional dependence of the lOR has
been limited to the effects of temperature, remaining sulfide sulfur concentration
and the pore gas oxygen concentration. As also discussed in that section, this limi-
tation can be removed if further terms are added to the mathematical model which
describes the evolution of sulfide sulfur oxidation in the heap. Including these terms
necessitates further data on the functional dependence of the lOR on the param-
eters included in the new description. An important question to be addressed is
the extent to which this more detailed model, carrying with it the need for more
data, leads to improved performance of a biooxidation heap.
Chemical Controls
Much of the discussion above has focused on physical processes which control
the overall rate of sulfide sulfur oxidation in a biooxidation heap. The primary
reason for this approach has been that a degree of control can be exercised on the
chemical conditions within the heap by controlling the chemistry of the recycled
liquor. Control of recycled water is generally well understood. Similarly, the bacte-
rial ecology is to a large extent self regulating if the chemical and physical condi-
tions are kept within appropriate ranges. There does seem to be scope for a better
understanding of chemical processes in the heap if the fraction of sulfide sulfur
oxidized is to be inferred with precision from the chemical composition of the
drainage effluent.
It must be emphasized that conditions in a biooxidation heap are controlled by
a set of rate processes. These are, for example, heat transport rates, gas transport
rates, water transport rate, sulfide sulfur oxidation rates and rates of interaction
between oxidation products and the minerals which make up the heap. It is these
last processes which relate the chemical composition of the drainage effluent with
the sulfide sulfur oxidation rates in the heap. It follows that if the chemical compo-
sition of the effluent is to be used as a guide to the fraction of sulfide sulfur oxi-
dized then data on these rates are also required. It is possible that some reactions
are fast enough compared to sulfide sulfur oxidation rates and to water transport
rates that equilibrium chemistry can be assumed. It is also possible that the degra-
dation of some of the gangue minerals can be ignored in the comparatively short
timescale associated with the operation of a biooxidation heap. It does seem likely
that reaction rates associated with iron and sulfate chemistry need to be better
known if iron and sulfate concentrations in effluent drainage are to be used to
estimate the fraction of sulfide sulfur oxidized.
Water Transport
The type of modeling discussed above indicates that the role of water is largely
to remove heat and keep temperatures in the heap within a range where oxidation
rates are high. As it is likely that the lOR decreases as sulfide sulfur is consumed
then the heat output should drop later in the life of the heap. It should therefore be
possible to reduce the irrigation rate as the heap ages. This may be advantageous
as degradation of the heap material may accompany pyrite oxidation and lead to a
reduction in water permeability. A reduction in average irrigation rate is often
achieved in practice by cycling irrigation applied to the top of the heap. This may
lead to ponding on the surface during the irrigation on period. The question then
arises as to the effect that this saturation of the top surface has on gas transport
rates. A further question is the nature of water transport in the heap and the extent
to which water and gas transport interact. Such a question may not be an issue for
heaps where there is little clay, but if the clay content is high and, particularly if
clay-like properties increase as the material oxidizes, the details of the water-gas
transport interaction may impact on heap performance.
References
1. Brierley lA, Wan RY, Hill DL, Cogan TC. Biooxidation-heap treatment technol-
ogy for processing lower grade refractory gold ores. In. Vargas T, Ierez CA, Wiertz
IV, Toledo H, eds. Biohydrometallurgical Processing. Vol!. Santiago: University
of Chile, 1995:253-261.
2. Shield JW, Whitlock JL, Christopher E. Mining Engineering. 1996; 48-54.
3. Bartlett RW. Biooxidation heap pretreatment of sulfide refractory gold ore. Min-
eral Processing and Extractive Metallurgy Review 1996 (In press).
4. Gibson DK, Pantelis G, Ritchie AIM The relevance of the intrinsic oxidation rate
to the evolution of polluted drainage from a pyritic waste rock dump. Interna-
tional Land Reclamation and Mine Drainage Conference and Third Conference
on the Abatement of Acidic Drainage. Vol 2. USA Department of the Interior
Bureau of Special Publications SP06A-94, 1994:258-264.
5. Brierley lA, Hill D1. Biooxidation process for recovery of gold from heaps of
low-grade sulfidic and carbonaceous sulfidic ore materials. U.S. Patent 5,246,486.
6. Brierley I, Luinstra 1. Biooxidation-heap concept for pretreatment of refractory
gold ore. In: Torma AE, Wey IE, Lakshmanan VI, eds. Biohydrometallurgical Tech-
nologies, Bioleaching Processes. Vol. 1 Warrendale, PA: The Minerals, Metals &
7. Ellis SI. Bacterial copper heap leach followed by heap leach recovery of gold at
Mt Leyshon Gold Mine. Biomine '94, International Conference and Workshop
Applications of Biotechnology to the Minerals Industry. Adelaide: The Australian
Mineral Foundation, 1994:8.1-8.7.
8. Rossi G. Biohydrometallurgy. New York: McGraw-Hill, 1990.
9. Evangelou VP. Pyrite Oxidation and its Control. New York: CRC Press, 1995.
10. Barrett I, Hughes MN, Karavaiko G1. Metal Extraction by Bacterial Oxidation of
Minerals. Chichester: Ellis Horwood. 1993.
11. Ritchie AIM. The environmental geochemistry of sulfide in mine-waste. In: Blowes
DW, Iambor IL eds. MAC Course Handbook on Environmental Geochemistry of
Sulfide Mine-wastes. Vol. 22. Nepean: Mineralogical Association of Canada.
1994:201-244.
12. Harrington IG, Prisbrey KA, Bartlett RW. Engineering aspects of heap biooxidation
of coarse-crushed refractory gold ores. In: Torma AE, Wey JE, Lakshmanan VI,
eds. Biohydrometallurgical Technologies, Bioleaching Processes. Vol 1, Warrendale,

PA:The Minerals, Metals & Materials Society, 1993:521-530.
13. Stromberg B, Banwart S. Kinetic modelling of geochemical processes at the Aitik
mining waste rock site in Northern Sweden. Appl Geochem 1994; 9:583-594.
14. White AF, Peterson ML. Role of reactive surface area characterization in geochemi-
cal kinetic models. In: Melchoir DC, Bassett RL, eds. Chemical Modelling of Aque-
ous Systems II. ACS Symposium Series 416. Washington: American Chemical So-
ciety 1990: 461-475.
15. Olson GJ. Rate of bioleaching by Thiobacillus ferrooxidans; results of an inter
laboratory comparison. Appl Environ Microbiol 1991; 57:642-644.
16. Harris JA. Development of a theoretical approach to the heap leaching of copper
sulfide ores. Proc Aust Inst Min Met 1969; 230:81-92.
17. Braun RL, Lewis AE, Wadsworth ME. In-place leaching of primary sulfide ores:
Laboratory leaching data kinetics model. Metallurgical Transactions 1974;
5:1717-1726.
18. Box JC, Prosser AP. A general model for the reaction of several reagents in heap
and dump leaching. Hydrometallurgy 1986; 16:77-92.
19. Dixon DG, Hendrix JL. A general model for leaching solid reactants from porous
ore particles: In: Hager JP ed. Laboratory and Leaching Data. EPD Congress 1993.
Warrendale, PA: The Minerals, Metals & Materials Society, 1993:21-43.
20. Pantelis G. FIDHELM: Description of model and users guide. Australian Nuclear
Science and Technology Organization Report, 1993; ANSTO/M123.
21. Davies GB, Ritchie AIM. A model of oxidation in pyritic mine wastes, III. Import
of particle size distribution. Appl Math Model 1987; 11:417-422.
22. Davies GB, Doherty G, Ritchie AIM. A model of oxidation in pyritic mine wastes,
II. Comparison of numerical and approximate solutions. Appl Math Model 1986;
10:323-330.
23. Hammack RW, Watzlaf GR. The effect of oxygen on pyrite oxidation. In: Pro-
ceedings of the Mining and Reclamation Conference. (Charleston, WV April 23-26)
1990:257-264.
24. Lawrence RW, Marchant PB. Comparison of mesophilic and thermophilic oxida-
tion systems for the treatment of refractory gold ores and concentrates. In: Norris
PR, Kelly DP, eds. Biohydrometallurgy. Kew: Science and Technology Letters
1988:359-374·
25. Karavaiko GI, Golovacheva RS, Pivovarova TA, Tzaplina lA, Vartanjan NS. Ther-
mophilic bacteria of the genus sulfobacillus. In: Norris PR, Kelly DP, eds.
Biohydrometallurgy. Kew Science and Technology Letters 1988:29-41.
26. Cathles LM. Predictive capabilities of a finite difference model of copper leaching
in low-grade industrial sulfide waste dumps. Math Geo11979; 11:175-191.
27. Nicholson RV. Iron-sulfide oxidation mechanisms: laboratory studies. In: Blowes
DW and Jambor JL eds. MAC Short Course Handbook on Environmental
Geochemistry of Sulfidic Mine-wastes. Vol 22. Nepean: Mineralogical Association
of Canada 1994:163-183.
28. Cathles LM, Apps JA. A model of the dump leaching process that incorporates
oxygen balance, heat-balance and air convection. Metallurgical Trans 1975;
6B:617-624.
29. Cathles LM, Schlitt WJ. A model of the dump leaching process that incorporates
oxygen balance, heat balance, and two-dimensional air convection. In: Schlitt WJ,
ed. Leaching and Recovering Copper from As-Mined Minerals. Soc Mining Eng
AIME, 1980:9-27.
30. Cathles LM, Murr LE. Evaluation of an experiment involving large scale column
leaching of low-grade copper sulfide waste: a critical test of the waste leaching
process. In: Schlitt WJ. ed. Leaching and Recovering Copper from As-Mined Min-
erals. Soc Mining Eng; AI ME 1980:29-48.
31. Bennett JW, Harries JR, Pantelis G, Ritchie AIM. Limitations on pyrite oxidation
rates in dumps set by air transport mechanisms. In: Salley J, McCready GL,
Wichlaz PL eds. Biohydrometallurgy. Montreal: CANMET SP89-10, 1989:551-561.
32. Pantelis G, Ritchie AIM. Macroscopic transport mechanisms as a rate-limiting
factor in dump leaching of pyritic ores. Appl Math Modelling 1991; 15:136-143.
33. Pantelis G, Ritchie AIM. Rate controls on the oxidation of pyritic material im-
posed by the upper temperature limits on the bacterially catalyzed process. In-
ternational Biohydrometallurgy Symposium, Troia, Portugal, September 1991.
FEMS Microbiological Reviews 1993; 11:183-190.
34. Pantellis G, Ritchie AIM. Rate limiting factors in dump leaching of pyritic ores.
Appl Math Modelling 1992; 16:553-560.
35. Pantelis G, Ritchie AIM. Optimizing oxidation rates in heaps of pyritic material.
In: Torma E, Wey JE, Lackshmanan, eds. Biohydrometallurgy Technologies,
Bioleaching Processes. Vol 1. Warrendale, PA. The Minerals, Metals & Materials
Society, 1993:731-738.
pyrite-gold ores. In: BIOMINE '93, Proceedings of a Conference on Applications
of Biotechnology to the Minerals Industry, Adelaide: The Australian Mineral Foun-
dation, 1993:9.1-9.8.
37. Ritchie AIM. Bio-oxidation heaps and AMD from waste rock dumps-the impor-
tance of the intrinsic oxidation rate. AusIMM Annual Conference, Darwin, Au-
gust 1994. Melbourne: AusIMM 1994:473-478.
38. Ritchie AIM. Optimizing the performance of a biooxidation heap: the importance
of the intrinsic oxidation rate. In: BIOMINE '94. International Conference and
Workshop, Perth 1994, Adelaide: The Australian Mineral Foundation, 1994:
15·1-15·19·
39. Casas JM, Martinez J, Moreno L, Vargas T. Two dimensional model of heat and
gas transport and mineral oxidation in copper bioleaching dump. In: Vargas T,
Jerez CA, Wiertz JV, Toledo H, eds. Biohydrometallurgical Processing. Vol 1.
Santiago: University of Chile 1995:447-457.
40. Jaynes DB, Rogowski AS, Pionke HB. Acid mine drainage from reclaimed coal
strip mines 1. Model description. Water Research 1984; 20:233-242.
41. Jaynes DB, Rogowski AS, Pionke HB. Acid mine drainage from reclaimed coal
strip mines 2. Simulation results of model. Water Research 1984; 20:243-250.
42. Guo W, Parizek RR. Field research on thermal anomolies indicating sulfide-oxi-
dation reactions in mine spoil. In: Alpers CN, Blowes DW, eds. Environmental
Geochemistry of Sulfide Oxidation. ACS Symposium Series 550 1994; 39:645-659.
43. Scharer JM, Byerley n, Kwong E, Nicholson RV. Role of biologically assisted pyr-
rhotite oxidation in acid mine drainage. In: Torma AE, Apel ML, Brierley CL,
eds. Biohydrometallurgy Technologies. Vol 2. Warrendale, PA: The Minerals, Met-
als & Materials Society, 1993:255-265.
44. Blowes DW, Reardon EJ, Jambor JL et al. The formation and potential impor-
tance of cemented layers in inactive sulfide mine tailings. Geochim Cosmochim
1991; 55:965-978.
45. Harries JR, Ritchie AIM. Pore gas composition in waste rock dumps undergoing
pyritic oxidation. Soil Science 1985; 140:143-152.
46. Bennett JW, Ritchie AIM. Measurement of the transport of oxygen into a reha-
bilitated waste rock heap. In: Proceedings of Second International Conference on
the Abatement of Acidic Drainage. Vol 3. Montreal: CANMET, 1991:289-298.
SECTION IV
Leaching Microorganisms
CHAPTER 11
Mesophilic, Autotrophic
Bioleaching Bacteria:
Description, Physiology and Role
Douglas E. Rawlings
Introduction to Microorganisms
A cidophilic bacteria capable of attacking metal sulfides are readily isolated from
sites of natural mineral oxidation. These bacteria have been divided according
to their preferred temperatures for growth into three groups: mesophiles, moder-
ate thermophiles and extreme thermophiles. The mesophiles are those bacteria
with optimum temperatures of between 25°-40°Cand are incapable of growth above
45°C. Mesophilic iron- or sulfur-oxidizing bacteria can be further subdivided into
those which are obligately autotrophic and those which are also capable of growth
on organic compounds.1 The moderate and extreme thermophiles are described
in chapter 12 and heterotrophic bacteria isolated from iron- and sulfur-rich envi-
ronments in chapter 13. This chapter deals primarily with the acidophilic, iron- or
sulfur-oxidizing obligately autotrophic bacteria. Good reviews on these bacteria
have been publishedz-4 and this chapter is intended to update and build on these.
The low pH, metal-rich, inorganic mineral environment in which bioleaching
reactions occur is populated by a group of bacteria which are highly adapted to
growth under these conditions. The bacteria most commonly isolated from inor-
ganic mining environments are Thiobacillus ferrooxidans, Thiobacillus thiooxidans
and Leptospirillum ferrooxidans. These bacteria are considered to be the most im-
portant in most industrial leaching processes. The recently described moderately
thermophilic bacterium, Thiobacillus caldus, 6 which grows optimally at 45°C also
grows well at between 30 and 40°C and may be readily isolated from bioleaching
processes which operate at this temperaturel,8 From a physiological viewpoint this
bacterium is the moderately thermophilic equivalent of Thiobacillus thiooxidans9
and the role of T. caldus in many industrial operations is almost certainly greater
than has been generally recognized. A number of species of heterotrophs which
grow in very close association with the obligate autotrophs have been found, most
of which belong to the genus Acidiphilium.10 Other acidophilic bacteria which have

been reported from leaching environments include the facultatively heterotrophic

bacteria Thiobacillus acidophilus!·l0 and Thiobacillus cuprinus. However, these fac-
ultative heterotrophs do not seem to playa major role in most leaching operations.
Taxonomic Description of the Mesophilic Bioleaching Bacteria

T. ferrooxidans, T. thiooxidans and 1. ferroxidans are all Gram-negative, obli-
gately autotrophic and obligately acidophilic bacteria. Both T. ferrooxidans and T.
thiooxidans are rod-shaped whereas 1. ferrooxidans has a distinct spiral-shape
(Fig.n.I). The major physiological difference between T. ferroxidans, T. thiooxidans
and 1. ferrooxidans is that T. ferrooxidans is capable of using ferrous iron or re-
duced sulfur compounds as an electron donor whereas T. thiooxidans is able to use
only reduced sulfur compounds and 1. ferrooxidans, only ferrous iron.! All three
bacteria use oxygen as a terminal electron acceptor, although T. ferrooxidans is
able to grow using ferric iron as an electron acceptor provided reduced sulfur com-
pounds are available to serve as an electron donor.
T. ferrooxidans has been described as a species of convenience.! Based upon
DNA homology studies, strains called T. ferrooxidans were placed by Harrison
into as many as seven subgroupsy o•1I Five T. ferrooxidans homology subgroups
had G+C contents of 56-59%, one subgroup was shown to be an L. ferrooxidans
isolate and another, represented by a single isolate (T. ferrooxidans m-I) had a
G+C content of 65% and has not been shown to oxidize sulfur. Therefore only five
of the subgroups can be considered to be T. ferrooxidans.
T. thiooxidans may also consist of more than one grouping with most strains
having a G+C content of 52-53%Y o An exception is strain DSM6I2 which has a
62% G+C content and may be more similar to Thiobacillus albertis. Another strain
has a G+C content of 58%, which falls within the range for T. ferrooxidans. A diffi-
culty in using the ability to oxidize iron or sulfur as a means of distinguishing T.
ferrooxidans from T. thiooxidans is that many T. ferrooxidans strains may exhibit a
delay in switching between iron and sulfur oxidation with the result that research-
ers may assume that the organism is incapable of oxidizing one or other energy
source.
The taxonomic status of the genus Leptospirillum remains to be fully resolved.
There is no valid description of the organism in Bergey's Manual of Determinative
Bacteriology, although reference is made to Leptospirillum in the section on the
genus Thiobacillus.! Acidiophilic, obligately autotrophic bacteria which grow on
ferrous iron and sulfide minerals and which have a helically curved rod-shaped
morphology are considered to be strains of Leptospirillum. The spiral forms can
be of varying length and generally have a long polar flagellum. The bacterium was
originally described by Markosyan from samples collected in Armenia but there is
evidence that this L.ferrooxidans LI5 strain may be atypical,l' Harrison and Norris
compared protein electrophoretic patterns, DNA-DNA homology and mol% GC
ratios for 1. ferrooxidans and several Leptospirillum-like bacteria.!3 They found
that the bacteria could be divided into two groups with a G+C content of either
51-52% or 55-56 mol%. Bacteria which fit the description of 1. ferrooxidans have
been isolated from numerous environments including water samples from ura-
nium mines in Canada and Mexico, copper mines in the USA/3 a mine in Bul-
garia,!4 coal spoil heaps in England,!5 mine samples in Australia7.!. and biooxidation
plants in South Africa.!6 L.ferrooxidans is as ubiquitous as T.ferrooxidans and T.
thiooxidans.
Mesophilic, Autotrophic Bioleaching Bacteria: Description, Physiology and Role 231
Fig. 11.1. A scanning electron microscope image of (a) L. ferrooxidans DSM2705 (mag-
nification "'18,000X) and (b) T.ferrooxidans ATCC33020 (magnification "'15,000X).
Phylogeny
The important leaching organisms grow in acidic, inorganic habitats and in
many instances cannot tolerate more than traces of organic matter. 17 These bacte-
ria could therefore be expected to have evolved in relative isolation from other
bacteria. Prior to the advent of DNA and RNA sequencing techniques it was not
possible to determine the evolutionary relationship of the acidophilic autotrophs
to the rest of the bacterial kingdom. Aconsiderable amount of molecular sequence
information has become available within the past 10 years including partial and
complete 5S rRNA and 16S rRNA sequences. 18,19 Based on these sequences T.
ferrooxidans and T. thiooxidans isolates are closely related and have been placed
within the proteobacteria at a point close to the division between the ~ and y sub-
groupS.19
The sequences of a number of other molecules such as the genes for the RecA
protein, glutamine synthetase and the ~ subunit of the FIFo ATP synthase have been
determined from a large number of bacteria. Phylogeny based on the amino acid
sequences of the RecA protein,2o,21 glutamine synthetasel6 and the 13 subunit of the
FIFo ATP synthase 22 have confirmed the classification of T. ferrooxidans as a ~
proteobacterium. An interesting exception is the product of the nifH gene. The T.
ferrooxidans nitrogenase iron protein is clearly most closely related to the equiva-
lent proteins of the genus Bradyrhizobium,16 which is an x proteobacterium, and
may be an example of lateral gene transfer.
Lane et al,19 analyzed partial 16S rRNA sequences of three isolates of
Leptospirillum-like bacteria. They reported that although the three isolates were
closely related to each other (ca. 94% similar), they were not specifically related to
any of the existing divisions of bacteria and suggested that the leptospirilli may
represent a new phylogenetic division. Near complete 16S rRNA sequence data of
two strains of 1. ferrooxidans are available in the data base and this has been used
in the construction of Figure 11.2. There is a discrepancy in the relationship of the
genus Leptospirillum to other bacteria if one compares the information in the
Ribosome Database Project (WWW) with that in the National Center for Biotech-
nology Information(NCBI) taxonomy base (WWW). In the NCBI data base the
leptospirilli have been placed within the group Nitrospira whereas the managers
of the Ribosome Database Project have not recognized a Nitrospira group. It is
Leptospirillum sp. (DSM 2391)
1--- Leptospirillum ferrooxidans str. Z2 (ATCC 29047)
r------- Magnetobaclerium bavaricum

Thermodesulfovibrio yellowslonii str. (ATCC 51303)
11"'---------- Synergistes jonesii str. (ATCC 49833)

Thiobacillus acidophilus (ATCC 27807)
Acidiphilium cryplum (ATCC 33463)
1------- Rhizobium meliloti str. (ATCC 9930)
_ - - - - - - Escherichia coli subsp. K-12
Pseudomonas aeruginosa sir. (ATCC 25330)
Thiobacillus caldus (DSM 8584)
Thiobacillus Ihiooxidans (DSM 612)
Thiobacillus thiooxidans str. (ATCC 19377)
Thiobacillus ferrooxidans (ATCC 19859)
Thiobacillus ferrooxidans (ATCC 23270)
Fig. 11.2. Dendrogram showing the phylogenetic relationship between the relatively
closely related bacteria T. ferrooxidans, T. thiooxidans and T. caldus and the distantly
related Leptospirillum species. A selection of other bacteria have been included as ref-
erence points. The dendrogram was constructed based on 165 rRNA sequences using
the Ribosome Database Project website (http://www.rdp.life.uiuc.edu).
interesting that one of the closest bacterial relatives to members of the genus
Leptospirillum that has been reported so far is the magnetotactic bacterium
Magnetobacterium bavaricum. 12,23
Nutrition and Energy

Bacteria used in bioleaching are remarkable in that they have very modest nu-
tritional requirements. Aeration of a sample of iron pyrite in acidified water is
sufficient to support the growth of T. ferrooxidans and L. ferrooxidans. Air pro-
vides the carbon, nitrogen and oxygen source, pyrite the energy source and trace
elements, and acidified water the growth environment. T. thiooxidans is not able
to oxidize ferrous iron to produce the ferric iron required to attack the mineral,
however, it readily grows on pyrite in combination with either T. ferrooxidans or L.
ferrooxidans.
Carbon Sources
T. ferrooxidans strains that have been confirmed as being pure are obligate
autotrophs. Some early studies appeared to show that after a period of adaptation
T. ferrooxidans was able to grow on organic substrates and that this was followed
by a permanent loss of the ability to oxidize iron. However, the G+C mole % ratio
of the cultures changed under these conditions and heterotrophic growth was al-
most certainly due to the inability of researchers to free their cultures from the
presence of the closely associated heterotrophic bacteria belonging to the genus
Acidiphilium.lO The iron-dependent, mixotrophic growth of one strain has been
reported but unfortunately that isolate has been lost. 24
Carbon dioxide fIXation in T. ferrooxidans takes place via the Calvin reductive
pentose phosphate cycle. One of the most important enzymes in this process,
ribulose l,5-biphosphate carboxylase (RuBPCase) has been characterized. 25 This
work also showed that growth on ferrous iron was reduced unless the concentra-
tion of CO 2 in the air was increased. This observation is in contrast to the work of
others26.27 in which it was found that the concentration of CO 2 in air was sufficient
to avoid limitation on growth on ferrous iron and mineral sulfide oxidation by T.
ferrooxidans. The bacterium responds to CO2 limitation by increasing the cellular
concentration of RuBPCase. Indeed, T. ferrooxidans strain Fel has two sets of the
structural genes for RuBPCase. 28 The two sets are separated by more than 5 kb and
the nucleotide sequence of the coding region of each set is identical although the
flanking regions varied substantially. The RuBPCase gene regulator, RbcR, has been
isolated and sequenced. 29
Very little work has been carried out on the enzymology or genetics of CO 2
fixation by either L. ferrooxidans or T. thiooxidans.
Nitrogen Sources
The study of the nitrogen requirements of bioleaching organisms is compli-
cated by the phenomenon that ammonia is highly soluble in acid solutions. Atmo-
spheric ammonia readily dissolves in leach solutions and may provide most, if not
all, of the nitrogen required for growth. As little as 0.2 mM ammonium has been
reported to be sufficient to satisfy the nitrogen requirement of T. ferrooxidans. 17
This value will be dependent on the amount of ferrous iron or mineral present in
the medium or leach liquor. High concentrations of inorganic or organic nitrogen
are inhibitory to iron oxidation. T. ferrooxidans is diazotrophic and is able to re-
duce atmospheric nitrogen to ammonia. This property was first reported by Mack-
intosh30 who demonstrated that T. ferrooxidans was able to incorporate 15N2 1abel
into cellular material. It has since been shown that alll5 isolates of T.ferrooxidans
tested contain the structural genes (nifHDK) for the nitrogen fixing enzyme, ni-
trogenase. '6 The ability to fix nitrogen is therefore almost certainly a general prop-
erty of T. ferrooxidans. The nifHDK genes from T. ferrooxidans ATCC 33020 have
been cloned and sequenced.31,32
There is evidence that L. ferrooxidans is also capable of fixing atmospheric
nitrogen. Genomic DNA from the L. ferrooxidans type strain was reported to give
a positive hybridization signal with a nifHDK gene probe from Klebsiella
pneumoniae. 33 L. ferrooxidans was also shown to reduce acetylene to ethylene and
oxidize ferrous iron to ferric iron at low oxygen concentrations. This ability was
repressed by added ammonium ions, behavior which is indicative of the ability to
fix nitrogen.
The ability of T. thiooxidans to fix nitrogen is uncertain. No hybridization sig-

nal was obtained when a nifHDK gene probe from Klebsiella pnuemoniae was used
against chromosomal DNA from T. thiooxidans ATCC 8085,33 but a positive signal
was obtained when a T. ferrooxidans nifHDK probe was hybridized to an unidenti-
fied T. thiooxidans isolate. 34
The role of nitrogen fixation in bioleaching operations is difficult to predict.
The dissolution of atmospheric ammonia in acid solutions could provide suffi-
cient ammonium to suppress nitrogen fixation. Furthermore, nitrogen fixation is
inhibited under fully aerobic conditions therefore might not occur in a well-aer-
ated leaching operation. In the highly-aerated, high oxidation rate, BIOX" tanks
used to pretreat gold-bearing arsenopyrite ores, addition of a small amount of
ammonia in the form of low-grade fertilizer is required to enhance mineral oxida-
tion (chapter 3).
Energy Sources
Iron Oxidation
As stated earlier, the energy requirements for growth of both T. ferrooxidans
and L. ferrooxidans are able to be met by the oxidation of ferrous to ferric iron
under aerobic conditions. Work by Blake and colleagues on the components of
iron oxidation in acidophilic bacteria has revealed that the ability to oxidize iron
appears to have evolved several times. At least four unique iron-oxidation mecha-
nisms exist. 35 -37 Two of these mechanisms are found in the mesophilic acidophiles.
The pathway for iron oxidation in T. ferrooxidans is characterized by the presence
oflarge amounts of the small copper protein, rusticyanin and c-type cytochromes.
Rusticyanin is not detectable in L. ferrooxidans or in any of the moderately or
extremely thermophilic iron-oxidizers. A novel red cytochrome (cytochrome 579)
which is clearly different from cytochrome a-, b- or c-type hemes and not found in
the other iron-oxidizers, dominates the electron transport chain of L.ferrooxidans.
This unique cytochrome was redox active with ferrous sulfate.37
The components of the iron-oxidation pathway in T. ferrooxidans have been
relatively well studied.38 These are a 92 kDa membrane porin,39 an Fe(II) oxidase,
cytochrome C552' rusticyanin and a cytochrome, oxidase of the aa 3-type. All the
above components have been isolated and characterized, the amino acid sequence
for rusticyanin has been determined 40 ,41 and gene for the Fe(II) oxidase have been
cloned and sequenced. 42 The exact order of the components and particularly, the
position of rusticyanin in the passage of the electrons is uncertain. 43 In a recent
review38 it has been postulated that the role of rusticyanin is to broaden the elec-
r
tron pathway from cytochrome '552 to the cytochrome oxidase as illustrated below.
rusticyanin!
Fe2+ ~ Fe(II) oxidase ~ cytochrome '552 ~ cytochrome c oxidase ~ O2

Oxygen electrode measurements have been used to determine the apparent
Michaelis constant (Km) of iron-oxidizers for ferrous iron.44 A Km value of 0.25
mM for L. ferrooxidans was considerably lower than the 1.34 mM obtained for T.
ferrooxidans cells. L. ferrooxidans is therefore able to grow more efficiently in an
environment with a low concentration of iron. Even more important was the find-
ing that ferrous iron oxidation by L. ferrooxidans was much less sensitive to end-
product inhibition by ferric iron (Kj38 mM) than T.ferrooxidans (Kj 2.5 mM). This
Mesophilic. Autotrophic Bioleaching Bacteria: Description. Physiology and Role 235
implies that in a continuous-flow leaching process (see chapter 3), where the quan-
tity of ferric iron in solution is high, L. ferrooxidans will have a distinct selective
advantage over T. ferrooxidans. Indeed, L. ferrooxidans has been reported to dis-
place T. ferrooxidans when mixed cultures were grown in chemostat cultures on
either ferrous sulfate- or pyrite-based media. 44.45
Sulfur Oxidation
Considerably more energy is available during the oxidation of reduced sulfur
compounds when compared with ferrous iron. For example, during the complete
oxidation of pyrite (FeS~), 1 electron is derived from the ferrous iron and 14 elec-
trons from the sulfur moiety. Although the oxidation of the iron component of a
mineral is probably the most important aspect of metal solubilization, the oxida-
tion of at least some of the sulfur component occurs. Evidence for this is that the
growth yield of T. ferrooxidans cells expressed as biomass produced per mole of
electrons is considerably higher on pyrite than it is on ferrous iron. 46
Attempts to investigate the pathways involved in sulfur oxidation by acidophilic
bacteria have proved difficult. This is partly because of the chemical reactivity and
hence lack of stability of many of the sulfur intermediates. What is clear is that
reduced sulfur compounds are oxidized to sulfate and that this results in a de-
crease in pH. Several enzymes involved in sulfur oxidation have been isolated and
this research has been reviewed. 46•47 Although the nature of some of the reactions
such as the conversion of elemental sulfur to sulfite and the nature of many of the
intermediate sulfur compounds is unknown, the steps in sulfur oxidation appear
to be as illustrated below.
S306~- ~ S~03~- ~ S406~- ~ S8 ~ S03~- ~ SO/-
i
S~-
Sugio and coworkers have identified a sulfite:ferric iron oxidoreductase in

T. ferrooxidans and a hydrogen sulfide:ferric ion oxidoreductase (SFORase) in a
number of strains identified as T.ferrooxidans and L.ferrooxidans. 48 •49 Both
T. ferrooxidans and L. ferrooxidans also possess a sulfide-binding protein.so The
presence of the hydrogen sulfide:ferric ion oxidoreductase and sulfide-binding pro-
teins in L. ferrooxidans was surprising as this organism is incapable of growth in
sulfur-based media. Even more surprising was the observation that three different
washed L. ferrooxidans cell preparations had sulfur-oxidizing activity with a pH
optimum proflle similar to other sulfur-oxidizing bacteria. These workers con-
firmed the results of previous researchers that L. ferrooxidans was not capable of
growth in sulfur-salts medium. The explanation of this paradox is still unknown.
A model has been proposed in which T. ferrooxidans is able to reduce elemen-
tal sulfur to sulfide using g1utathione.6~ This sulfide can be oxidized using Fe3+ as
an electron acceptor to produce sulfite and FeH • An NADH-dependent sulfite re-
ductase has been isolated from the periplasm of iron-grown T. ferrooxidans cells
which reduces sulfite to hydrogen sulfide and NADH+. The possible function of
sulfide cycling in T. ferrooxidans is uncertain, but it has been suggested that sul-
fide acts as a energy storage compound.s'
Alternate Carbon Sources, Electron Donors and Electron Acceptors

Although T. ferrooxidans is considered to be strictly autotrophic, it is capable
of growth on formic acid.5~ Small organic acids are toxic to acidophiles because at
low pH values the organic acids are present in their undissociated form and in this
form organic acids readily diffuse across the cytoplasmic membrane.53 Once in-
side the cell, organic acids dissociate because of the near neutral intercellular pH
which leads to a dissipation of the proton gradient. The secret of obtaining growth
of T. ferrooxidans on formate was to ensure that the formate was provided at low
concentration in a chemostat.5~ Using this approach, higher cell densities were
obtained using formate as an energy source than when using ferrous iron or re-
duced sulfur compounds. Utilization of formate was via the calvin cycle and six
out of seven T. ferrooxidans strains tested were able to oxidize formate with strain
ATCC 21834 being particularly active. None of the two T. thiooxidans strains tested
were able to grow on formate.
T. ferrooxidans has been shown to possess an inducible hydrogenase which
enabled the bacterium to grow aerobically using hydrogen as an energy source.54
The bacterium was less acidophilic when growing on hydrogen (pH range for
growth 2.5-6.0) and had a slightly longer doubling time (5.0 vs. 4.5 h) than when
grown with sulfur or ferrous iron. All three T. ferrooxidans strains but not the
single T. thiooxidans strain tested had the ability to utilize hydrogen.
Metal ions which exist in more than one oxidation state and have redox poten-
tials less than the O~/H20 couple, have the theoretical potential to serve as electron
donors for the growth of bioleaching bacteria. It has been reported that T.
ferrooxidans is able to directly oxidize U4+ to U6+ under aerobic conditions and
that this oxidation reaction is coupled to carbon dioxide fixation. 55 Similarly, the
direct oxidation of Cu+ to CuH has been coupled to CO 2 fixation. 56.57 However,
whenever iron is present, it is difficult to unequivocally demonstrate the direct
oxidation of the metal as opposed to the oxidation of ferrous iron to ferric which
may then oxidize the metal chemically. T. ferrooxidans has been reported to oxi-
dize M05+ to M0 6+ and the evidence for this is strong as a molybdenum oxidase
was purified from cell extracts.58 Bioleaching bacteria are able to oxidize arsenopy-
rite ores and the potential exists for the oxidation of As3+ to As5+ to serve as an
alternate electron donor. However, this property has not been conclusively dem-
onstrated. T. ferrooxidans may be more versatile with respect to its ability to oxi-
dize metals than is generally recognized. A similar metal-oxidizing capacity has
not been demonstrated for either L. ferrooxidans or T. thiooxidans, but much less
research has been carried out with these bacteria.
T. ferrooxidans is also versatile with respect to the compounds it is able to use
as electron acceptor. Typically T. ferrooxidans, T. thiooxidans and L. ferrooxidans
all respire using oxygen as a terminal electron acceptor. However, in the absence of
oxygen, T. ferrooxidans is able to use ferric iron as an electron acceptor provided
that a reduced form of sulfur, such as elemental, serves as the electron donor.48.59.60
This means that T. ferrooxidans is capable of growth in an anaerobic environment.
The practical implication of this is that metal solubilizing activity may take place
at the center of poorly aerated dump or heap using the ferric that was produced by
bacteria growing at the surface.61 T. thiooxidans strains do not appear to possess
this ferric iron-reducing ability and are therefore probably obligately aerobic. 59
Since the ferrous/ferric iron couple cannot be used as both electron donor and
acceptor, L. ferrooxidans is also likely to be an obligate aerobe.
Besides its ability to reduce ferric iron, T. ferrooxidans is also able to reduce
M0 6+, CuH and CoH when using elemental sulfur as electron donor.57,62 The abil-
ity to reduce M0 6+, CuH and CoH is a property of the hydrogen sulfide:ferric iron
oxidoreductase (SFORase), the same enzyme which couples the oxidation of el-
emental sulfur to the reduction of Fe3+. Interestingly, a SFORase has also been
found in a strain of L. ferrooxidans 50 but its role in this bacterium is uncertain.
Mineral Oxidizing Ability

Since T. ferrooxidans is able to oxidize both iron and sulfur, it is not surprising
that the bacterium is able to solubilize a wide variety of metals from ores in pure
culture. 63 Even though leptospirilli are able to use only iron as an energy source,
these bacteria are nevertheless able to degrade pyrite very efficiently. This is be-
cause both types of organism are able to oxidize ferrous iron to ferric and ferric
iron is required for mineral solubilization. At 30°C the growth rate of L.ferrooxidans
on iron in shaken batch cultures is about half that of T. ferrooxidans,5 however,
these two species degrade pyrite at similar rates. 44 In contrast to these bacteria, T.
thiooxidans is unable to degrade sulfidic minerals in pure culture. However, when
T. thiooxidans is grown in mixed culture with either T. ferrooxidans or 1.
ferrooxidans it may enhance the ability of either pure culture to degrade sulfidic
ores.
The relative roles of the mesophilic bacteria in bioleaching operations has been
the subject of much research and has been reviewed. 5 More recent work has reas-
sessed the relative roles of the mesophilic bacteria and suggested that T. ferrooxidans
may not be as dominant as previously thought. In percolation leaching experi-
ments using crushed pyrite or complex sulfidic ores and a temperature of 28°C,
L. ferrooxidans was at least as numerous as T. ferrooxidans and when in pure cul-
ture mobilized metals at least as efficiently.14Similar conclusions have been reached
when molecular DNA analysis techniques have been used to identify bacteria in
bioleaching environmentsp:z,64 An analysis of the spacer regions between the 16S
and 23S rRNA was used to identify the dominant bacteria during the column-leach-
ing of a complex copper inoculated with bacteria from the bottom of an industrial
heap-leaching operation at Lo Aguirre, Chile. 65 If ferrous iron was added to the
leach liquor, then T. ferrooxidans dominated the population. If no ferrous iron was
added, the ferrous iron concentration remained low and only L. ferrooxidans and
T. thiooxidans were detected. Under these conditions bioleaching efficiency re-
mained high with 90% of the copper being recovered. In another study, adherence
of T. ferrooxidans, L. ferrooxidans and T. thiooxidans to a complex copper sulfidic
ore during column-leaching using specific antibodies was measured. T. ferrooxidans
was dominant for the first 5 days but after 60 days of leaching, L. ferrooxidans and
T. thiooxidans outnumbered T. ferrooxidans by 1,000-fold. 66
An investigation of the bacteria present in the tanks of a continuously operat-
ing pilot plant treating gold-bearing arsenopyrite ores has revealed similar results.
The tanks had been inoculated with leach solution from an industrial plant and
were operating at 40°C and pH 1.6. Analysis of the 16S rRNA from these tanks
indicated that the population was dominated by 1. ferrooxidans and T. thiooxidans
(or Thiobacillus caldus) and that if present, the number of T. ferrooxidans cells was
too low to be detected. 64
T. ferrooxidans cell
iron impregnated
~S10"- exopolymer
'---------~~~--~------' layer
sulfidic mineral (eg FeSJ
rusticyanin
16 kDa acid stable
copper protein
inner membrane
pH 6.5
92 kDa
84 kDa cytochrome cSS2

Fen cytochrome c552 oxidase (aa 3 type)
oxidoreductase, 63 kDa sulfate requiring
8 - 10 subunits of a cytochrome c-552(s)
6 kDa high redox potential c-SS2(m)
iron-sulfur monomer c-sso(m)
Fig. 11.3. Diagram illustrating ferrous/ferric iron cycling in the exopolysaccharide layer
of a T. ferrooxidans cell attached to a mineral particle. Amodel of the path of electron
transport from ferrous iron into the cell is shown in the expanded region of cell envelope.
The Role of the Mesophilic Obligate Acidophiles

in Mineral Bioleaching
It has been elegantly argued67 that at low pH values (pH < 3.5) indirect solubili-
zation of the mineral by ferric iron hexahydrate is the major mechanism of min-
eral attack. Bacteria such as T. ferrooxidans and L. ferrooxidans adhere strongly to
the surface of an sulfidic ore 68,69 and are surrounded by an exopolymer layer which
is heavily impregnated with iron irons and polythionate granula (Fig. 11.3). These
metal-containing exopolymers give the cells a slightly positive zeta potential and
this was necessary for attachment as it overcomes the repulsion between the nega-
tive1y charged sulfide ore and negatively charged washed cells.7° The ferric iron
hexahydrate molecules at the surface attack the sulfidic ore indirectly (chemically)
and ferrous iron and thiosulfate are produced. 67
FeS z + Fe(H zO)63+ + 3HzO ~ FeH + SZ03z- + 6(HzO)63+ + 6H+
The ferrous iron is rapidly reoxidized to ferric by the bacterium and is recycled
within the exopolymer layer of the attached bacteria (Fig. 11.3). At low pH the un-
stable thiosulfate is converted to polythionates and elemental sulfur. Since elemental
sulfur is poorly soluble this is the form of sulfur most readily detected, often visu-
ally. In the case where T. ferrooxidans is the generator of ferric iron, thiosulfate
and polythionates may be may be oxidized to sulfate. Where L. ferrooxidans is the
ferric iron generator, sulfur-oxidizing bacteria such as T. thiooxidans, T. caldus or
even Thiobacillus acidophilus and other faculative sulfur oxidizers may be respon-
sible for the oxidation of thiosulfate and sulfur derivatives.
It has been pointed out that the stoichiometry and sulfur intermediate of this
equation is different from that in the equation most commonly referred to for py-
ritic ore attack.
Other Physiological Characteristics

In general, bioleaching bacteria are remarkably adaptable when faced with ad-
verse growth conditions. There appears to be a substantial variation in tolerance
to some growth conditions like pH, temperature, or metal ion tolerance between
strains or isolates which have the same genus and species designation. Adapted
populations may also differ substantially from the parental cultures. Furthermore,
there is some interdependence between variables, e.g., the pH of a medium may
affect temperature tolerance.
pH
T. ferrooxidans grows best within the pH range 1.8-2.5. There have been reports
of growth at a pH of 1.5 after selection in continuous culture.71 T. thiooxidans is
considerably more resistant to acid and is capable of growth at a pH of less than
0.8. 15 Leptospirillum is also more resistant to low pH than T. ferrooxidans and will
grow at a pH as low as 1.2.15
Temperature
The optimum temperature for the growth of both T. ferrooxidans and T.
thiooxidans is probably about 30°-35°C. However, some strains of T. ferrooxidans
are adapted to low temperatures. It has been reported that the growth rate halves
with each 6°C within the range 2So-2°C.72 Some strains are able to oxidize pyrite at
temperatures of as low as 10°C.s As may be expected, these cold-tolerant strains are
less tolerant of high temperatures than more typical mesophilic isolates. The up-
per limit for growth of T. ferrooxidans is probably close to 40°C. Cultures which
grow above 42°C are almost always dominated by mixtures of T. caldus and L. fer-
rooxidans rather than T.ferrooxidans. s
In general,L.ferrooxidans strains appear to be more tolerant of high tempera-
tures and less tolerant of low temperatures than T. ferrooxidans. Leptospirillum-
like bacteria have been reported to have an upper limit of about 4SoC73 and a lower
limit of about 20°C.74 A pilot scale (1 m 3) BIOX" plant operating at constant tem-
perature of 4SoC was found to be dominated by bacteria with an identical16S rRNA
restriction enzyme pattern to the L. ferrooxidans type strain.16.64 When growing
on iron at lower temperatures (lS0-20°C), L. ferrooxidans cells become embedded
in slime to form aggregates. s These macroscopic aggregates can take on a variety
of forms which include ribbons and almost spherical-like pellets. The observation
that L. ferrooxidans is more inhibited by low temperatures than either T. fer-
rooxidans or T. thiooxidans was confirmed by Sand and coworkers.14
Tolerance and Sensitivity to Metallic and Other Ions

An important characteristic of the mesophilic acidophilic chemolitho-
trophs is their tolerance of high concentrations of metallic and other ions. T. fer-
rooxidans and L. ferrooxidans are resistant to a large number of metal cat-
ions, although levels of resistance show considerable strain variation.
Adaptation to high levels of metal resistance on exposure to a metal is likely to be
responsible for much of the variation. T. ferrooxidans appears to be particu-
larly resistant. For example, the bacterium has been reported to grow in medium
containing Co H (30 giL), CuH (55 in giL), NiH (72 giL), ZnH (120 giL), Ups (12
gIL) and Fe H (160 glL).63
One L. ferrooxidans strain has been reported to be 4- to 5-fold more resistant to
uranium, silver and molybdenum than T. ferrooxidans, but sensitive to 100-fold
lower concentration of copper.73 In a comparative study of two T. ferrooxidans, two
L. ferrooxidans and a T. thiooxidans strain, it was found that T. ferrooxidans and L.
ferrooxidans were approximately equally resistant to CuH , ZnH, Al3+, NiH and
MnH, but that L. ferrooxidans was more sensitive «2 giL) than T. ferrooxidans to
CoH .74 T. thiooxidans was sensitive to less than 5 giL of all the cations used in the
comparative study with the exception of ZnH (10 giL). Levels of resistance were,
on average, lower than those reported previously.
T. ferrooxidans has been reported to be sensitive to the HgH, Ag+, As3+ and
M0 6+ cations and monovalent anions such as CI-, Br-, rand N03-.s3 Again, levels
of sensitivity were strain dependent. When ten T.ferrooxidans isolates were screened
for HgH resistance, three of the strains contained DNA which hybridized to a Tns01
mer gene probe. Bacteria carrying the resistance genes were, in general, 3-5 times
more resistant to Ht+ than strains which did not have mer genes. The mer genes
from T.ferrooxidans strain E-15 have been cloned and sequencedls-n
Although normally sensitive to As3+ and to a lesser extent Ass+, bioleachin§
bacteria can be adapted to high concentrations of As ions. For example the BIOX
process for the treatment of refractory gold-bearing arsenopyrite ores operates
with a total As concentration of >13 giL. Adaptation to As was achieved by expos-
ing the bacteria to increasing concentrations of arsenic in continuous flow reac-

tors,78.79 Recently, the As genes from T. caldus have been identified and the As re-
sistance genes from T. ferrooxidans ATCC 33020 have been cloned. 80
Sensitivity of bioleaching organisms to chloride ions is of considerable eco-
nomic importance. For example, in the dry parts of Western Australia or northern
Chile most of the available ground water is high in chloride and the use of this
water in leaching process make-up water is therefore limited. A mixture of thiobacilli
has been used to leach a sulfidic copper ore in the presence of 5 giL CI- but at-
tempts to adapt T. ferrooxidans to > 5 giL CI- have not been successful. 81.8~
Summary
The mesophilic iron- and sulfur-oxidizing autotrophic bacteria described in
this chapter play an essential role in many currently operating commercial
biooxidation processes. However, the number of types of bacteria that will be used
in future processes is likely to increase substantially as new processes are devel-
oped. Bacteria capable of oxidizing ores at higher temperatures are particularly
important because at higher temperatures the chemistry of microbially-assisted
leaching is much faster. This means that several processes which are uneconomic
at lower temperatures are likely to become financially viable. Each new organism
will have strengths and weaknesses which are likely to be different from the
mesophilic bacteria described in this chapter. Studies on moderately and extremely
thermophilic iron- and sulfur-oxidizing organisms are less advanced than those
with the mesophilic bacteria, but in the future there is likely to be a range of organ-
isms available, each with different advantages and disadvantages which will need
to be considered for use in the biooxidation of a given ore.
References
1. Kelly DP, Harrison AP. Genus Thiobacillus Beijerinck 1904b, 597. In: Staley JT,
Bryant MP, Pfenning N et al, eds. Bergey's Manual of Systematic Bacteriology.
Vol 3. Baltimore: Williams and Wilkins, 1989:1842-1858.
2. Brierley CL. Bacterial leaching. Crit Rev Microbiol 1978; 6:207-2.62..
3. Lundgren DG, Silver M. Ore leaching by bacteria. Ann Rev Microbiol 1980;
34:263-283.
4. Kelly DP, Norris PR, Brierley CL. Microbiolgical methods for the extraction and
recovery of metals. In: Bull AT, Ellwood DG, Ratledge C, eds. Microbial Technol-
ogy: Current State and Future Prospects. Cambridge, UK:Cambridge University
Press, 1979:263-308.
In: Ehrlich HL, Brierley CL, eds. Microbial Mineral Recovery. New York: McGraw-
Hill, 1990:3-2.7.
6. Hallberg KB, Lindstrom EB. Characterization of Thiobacillus caldus sp nov., a
moderately thermophilic acidophile. Microbiology 1994; 140:3451-3456.
8. Amaro AM, Hallberg K, Lindstrom EB et al. An immunological assay for detec-
tion and enumeration of thermophilic biomining microorganisms. Appl Environ
Microbiol 1994; 60:3470-3473.
9. Hallberg KB. PhD dissertation, 1995, University of Umea, Sweden.
10. Harrison AP. The acidophilic thiobacilli and other acidophilic bacteria that share
their habitat. Ann Rev Microbiol 1984; 38:265-292.
11. Harrison AP. Genomic and physiological diversity amongst strains of Thiobacillus
ferrooxidans and genomic comparison with Thiobacillus thiooxidans. Arch
Microbiol 1982; 131:68-76.
12. Goebel BM and Stackebrandt E. Molecular analysis of the microbial biodiversity
in a natural acidic environment. In: Jerez CA, Vargas T, Toledo H, Wiertz JV,
eds. Biohydrometallurgical Processing. Vol II. Santiago: University of Chile Press,
1995:43-51.
13. Harrison AP, Norris PRo Leptospiriilum ferrooxidans and similar bacteria: some
characteristics and genomic diversity. FEMS Microbiol Lett 1985; 30:99-102.
leaching. Appl Environ Microbiol1992; 58:85-92.
15. Norris PRo Iron and mineral oxidation with Leptospirillum-like bacteria. In: Rossi
G, Torma AE, eds. Recent Progress in Biohydrometallurgy. Iglesias: Associazione
Mineraria Sarda, 1983:83-96.
16. Rawlings DE. Unpublished observations.
17. Tuovinen OH, Niemela SI, Gyllenberg HG. Effect of mineral nutrients and or-
ganic substances on the development of Thiobacillus ferrooxidans. Biotechnol
Bioeng 1971; 13:517-527.
18. Lane DJ, Stahl DA, Olsen GJ et al. Phylogenetic analysis of the genera Thiobacillus
and Thiomicrospira by 5S rRNA sequences. J Bacteriol 1985; 163:75-81.
19. Lane DJ, Harrison AP, Stahl DA et al. Evolutionary relationships amoung sulfur-
and iron-oxidizing eubacteria. J Bacteriol1992; 174:269-278.
20. Karlin S, Weinstock GM, Brendel V. Bacterial classifications derived from recA
protein sequence comparisions. J Bacteriol1995; 177:6881-6893.
21. Karlin S, Brocchieri L. Evolutionary conservation of RecA genes in relation to
protein structure and function. J Bacteriol 1996; 178:1881-1894.
22. Brown LD and Rawlings DE. A comparison of the structure of the H+ -translocat-
ing ATP synthase from Thiobacullus ferrooxidans with those of other organisms.
In: Torma AE, Apel ML, Brierley CL, eds. Biohydrometallurgical Technologies.
Vol II. Warrendale, Pennsylvania: TMS Press, 1993:519-528.
23. Amann RI, Ludwig W, Schleifer K-H. Phylogenetic identification and in situ de-
tection of individual microbial cells without cultivation. Microbiol Rev 1995;
59:143-169.
24. Barros MEC, Rawlings DE, Woods DR. Mixotrophic growth of a Thiobacillus
ferrooxidans strain. Appl Environ Microbiol 1984; 593-595.
25. Holuigue L, Herrera L, Phillips OM et al. CO. fixation by mineral-leaching bacte-
ria: characteristics of the ribulose bisphosphate carboxylase-oxygenase of
Thiobacillus ferrooxidans. Biotechnol Appl Biochem 1987; 9:497-505.
26. Kelly DP, Jones CA. Factors affecting metabolism and ferrous iron oxidation in
suspensions and batch cultures of Thiobacillus ferrooxidans: relevance to ferric
iron leach solution regeneration. In: Murr LE, Brierley JA, Torma AE, eds. Metal-
lurgical Applications of Bacterial Leaching and Related Microbiological Phenom-
ena. New York:Academic Press. 1978:19-44.
27. Norris PRo Factors affecting bacterial mineral oxidation: the example of carbon
dioxide in the context of bacterial diversity. In: Salley J, McCready RGL, Wichlacz
PL, eds. Biohydrometallurgy-1989 Ottawa:CANMET 1989:1-14.
28. Kusano T, Takeshima T, Inoue, C et al. Evidence for two sets of structural genes
coding for ribulose biphosphate carboxylase in Thiobacillus ferrooxidans. J
Bacteriol 1991; 173:7313-7323.
Mesophilic, Autotrophic Bioleaching Bacteria; Description, Physiology and Role 243
29. Kusano T, Sugawara K. Specific binding of Thiobacillus ferrooxidans RbcR to the

intergenic sequence between the rbc operon and the rbcR gene. J Bacteriol 1993;
175:101 9-1025.
30. Mackintosh ME. Nitrogen fixation by Thiobacillus ferrooxidans. J Gen Microbiol
1978; 34:2.63-2.83.
31. Pretorius I-M, Rawlings DE, Woods DR. Indentification and cloning of Thiobacillus
ferrooxidans structural nifHDK genes in Escherichia coli. Gene 1985; 45:59-65.
32.. Pretorius I-M, Rawlings DE, O'Neill EG et al. Nucleotide sequence of the gene
encoding the nitrogenase protein of Thiobacillus ferrooxidans. J Bacteriol 1997;
169:367-370.
33. Norris PR, Murrell JC, Hinson D. The potential for diazotrophy in iron- and sul-
fur-oxidizing acidophilic bacteria. Arch Microbiol 1995; 164:2.94-300.
34. Lawson EN. Unpubublished observations.
35. Blake RC, Shute EA, Waskovsky J, Harrison AP. Respiratory components in aci-
dophilic bacteria that respire on iron. Geomicrobiol J 1992.; 10:173-192..
36. Blake RC, Shute EA, Greenwood MM et al. Enzymes of aerobic respiration on
iron. FEMS Microbiol Rev 1993; 11:9-18.
37. Blake RC, McGinness S. Electron-transfer proteins of bacteria that respire on iron.
In: Torma AE, Apel ML, Brierley CL, eds. Biohydrometallurgical Technologies.
Vol II. Warrendale, Pennsylvania: TMS Press, 1993; 616-62.8.
38. Yamanaka T, Fukumori Y. Molecular aspects of the electron transfer system which
participtes in the oxidayion of ferrous iron by Thiobacillus ferrooxidans. FEMS
Microbiol Rev 1995; 17:401-413.
39. Mjoli N, Kulpa CF. The identification of a unique outer membrane protein re-
quired for iron oxidation in Thiobacillus ferrooxidans. In: Biohydrometallurgy,
Norris PR, Kelly DP, eds. Kew Surrey: Science and Technology Letters, 1988:89-102..
40. Ronk M, Shively JE, Shute EA et al. Amino acid sequence of the blue copper
protein rusticyanin from Thiobacillus ferrooxidans. Biochemistry 1991;
30:9435-9442.·
41. Nunzi F, Woudstra M, Campese D et al. Amino acid sequence of rusticycanin
from Thiobacillus ferrooxidans and its comparison with other blue copper pro-
teins. Biochim Biophys Acta 1993; 1162:28-34.
42.. Kusano T, Takeshima C, Sugawara K et al' Molecular cloning of the gene encod-
ing Thiobacillus ferrooxidans Fe(I1) oxidase. J BioI Chern 1992; 2.67:112.42.-112.47.
43. Blake RC, Shute EA. Respiratory enzymes of Thiobacillus ferrooxidans. Kinetic
properties of an acid stable iron:rusticyanin oxidoreductase. Biochemistry 1994;
33:92.2.0-92.2.8.
44. Norris PR, Barr DW, Hinson D. Iron and mineral oxidation by acidophilic bacte-
ria: affinities for iron and attachment to pyrite. In: Biohydrometallurgy, Norris
PR, Kelly DP, eds. Kew Surrey: Science and Technology Letters, 1988:43-59.
45. Helle U, Onken U. Continuous bacterial leaching of a pyritic fiatation concen-
trate by mixed cultures. In: Biohydrometallurgy, Norris PR, Kelly DP, eds. Kew
Surrey: Science and Technology Letters, 1988:61-75.
46. Pronk JT, Meulenberg R, Hazeu W et al. Oxidation of reduced sulfur compounds
by acidophilic thiobacilli. FEMS Microbiol Rev 1990; 75:2.93-306.
47. Kuenen JG, Pronk JT, Hazeu W et al. A review of bioenergetics and enzymology
of sulfur compound oxidation by acidophilic thiobacilli. In: Torma AE, Apel ML,
Brierley CL, eds. Biohydrometallurgical Technologies. Vol II. Warrendale, Penn-
sylvania: TMS Press, 1993:487-505.
48. Sugio T, Domatsu C, Munakata 0 et a1. Role of a ferric ion-reducing system in
sulfur oxidation of Thiobacillus ferrooxidans. Appl Environ Microbiol 1985;
49:14°1-1406.
49. Sugio T, White KJ, Shute E et al. Existence of a hydrogen sulfide:ferric ion oxi-
doreductase in iron-oxidizing bacteria. Appl Environ Microbioll992; 58:431-433.
50. Suzuki H, Tanaka T, Tano T et al. Existence of sulfide binding protein in iron-
oxidizing bacteria. In: Torma AE, Apel ML, Brierley CL, eds. Biohydrometallurgical
Technologies. Vol II. Warrendale, Pennsylvania: TMS Press, 1993:423-431.
51. Sugio T, Tanaka K, Matsugi S et al. Purification and some properties of NADH-
dependent sulfite reductase from Thiobacillus ferrooxidans. In: Jerez CA, Vargas
T, Toledo H, Wiertz JV, eds. Biohydrometallurgical Processing. Vol II. Santiago:
University of Chile Press, 1995:109-117.
52. Pronk JT, Meijer WM, Hazeu W et al. Growth of Thiobacillus ferrooxidans on
formic acid. Appl Environ Microbiol 1991; 57:2057-2062.
53. Alexander B, Leach S, Ingledew WI. The relationship between chemiosmotic pa-
rameters and sensitivity to anions and organic acids in the acidophile Thiobacillus
ferrooxidans. J Gen Microbiol 1987; 133:1171-1179.
54. Drobner E, Huber H, Stetter KO. Thiobacillus ferrooxidans, a facultative hydro-
gen oxidizer. Appl Environ Microbiol 1990; 56:2922-2923.
55. DiSpirito AA, Tuovinen OH. Uranous ion oxidation and carbon dioxide fixation
by Thiobacillus ferrooxidans. Arch Microbiol 1982; 133:28-32.
56. Nielsen AM, Beck IV. Chalcocite oxidation and coupled carbon dioxide fixation
by Thiobacillus ferrooxidans. Science 1972; 175:1124-1126.
57. Sugio T, Tsujita Y, Inagaki K et al. Reduction of cupric ions with elemental sul-
fur by Thiobacillus ferrooxidans Appl Environ Microbiol 1990; 56:693-696.
58. Sugio T, Hirayama K, Inagaki K et al. Molybdenum oxidation by Thiobacillus
ferrooxidans. Appl Environ Microbioll992; 58:1768-1771.
59. Suzuki I, Takeuchi TL, Yuthasastrakosol TD et al. Ferrous iron and sulfur oxida-
tion and ferric iron reduction activities of Thiobacillus ferrooxidans are affected
by growth on ferrous iron, sulfur or a sulfide ore. Appl Environ Microbiol 1990;
56:1620-1626.
60. Pronk JT, de Bruyn JC, Bos P et al. Anaerobic growth of Thiobacillus ferrooxidans.
61. Goodman AE, Babij T, Ritchie AIM. Leaching of a sulfide ore by Thiobacillus
ferrooxidans under anaerobic conditions. In: Rossi G, Torma AE, eds. Recent
Progress in Biohydromeatallurgy. Iglesias: Associazione Mineraria Sarda,
1983:361-376.
62. Sugio T, Tsujita Y, Katagaki T et al. Reduction of M0 6+ with elemental sulfur by
Thiobacillus ferrooxidans. I Bacteriol 1988; 170:5956-5959.
63. Torma AP. The role of Thiobacillus ferrooxidans in hydrometallurgical processes.
Adv Biochem Eng 1977; 6:1-38.
64. Rawlings DE. Restriction enzyme analysis of 16S rRNA genes for the rapid iden-
ferrooxidans strains in leaching environments. In: Ierez CA, Vargas T, Toledo H,
Wiertz IV, eds. Biohydrometallurgical Processing. Vol II. Santiago: University of
Chile Press, 1995:9-17.
65. Pizarro I, Iedlicki E, Orellana 0 et al. Bacterial populations in samples of
bioleached copper ore as revealed by analysis of DNA obtained before and after
cultivation. Appl Environ Microbioll996; 62:1323-1328.
66. Garcia A, Jerez CA. Changes of the solid-adhered populations of Thiobacillus
ing ores as determined by immunological analysis. In: Jerez CA, Vargas T, To-
ledo H, Wiertz IV, eds. Biohydrometallurgical Processing. Vol II. Santiago: Uni-
versity of Chile Press, 1995:19-30.
Mesophilic, Autotrophic Bioleaching Bacteria; Description, Physiology and Role 245
67. Sand W, Gerke T, Hallmann R et al. Sulfur chemistry, biofllm, and the (in)direct
attack mechanism-a critical evaulation of bacterial leaching. Appl Microbiol
Biotechnol1995; 43:961-966.
68. Arredondo R, Garcia A, Jerez CA. Partial removal of lipopolysaccharide from
Thiobacillus ferroxidans affects its adhesion to solids. Appl Environ Microbiol
1994; 60:2846-2851.
69. Gerke T, Hallmann R, Sand W. Importance of exopolymers from Thiobacillus
ferrooxidans and Leptospirillum ferrooxidans for bioleaching. In: Jerez CA, Vargas
T, Toledo H, Wiertz JV, eds. Biohydrometallurgical Processing. Vol I. Santiago:
University of Chile Press, 1995:1-11.
70. Blake RC, Shute EA, Howard, GT. Solubilization of minerals by bacteria: Electro-
phoretic mobility of Thiobacillus ferrooxidans in the presence of iron, pyrite, and
sulfur. Appl Environ Microbiol1994; 60:3349-3357.
71. Vian M, Creo C, Dalmastri C et al. Tliiobacillus ferrooxidans selection in con-
tinuous culture. In Lawrence RW, Branion RMR, Ebner HG, eds. Fundamental
and Applied Biohydrometallurgy, Amsterdam:Elsevier Science Publishing. 1986;
395-406.
72. McCready RG. Progress in the bacterial leaching of metals in Canada. In:
Biohydrometallurgy, Norris PR, Kelly DP, eds. Kew Surrey: Science and Technol-
ogy Letters, 1988:177-195.
73. Norris PR, Parrot L, Marsh RM. Moderately thermophilic mineral-oxidizing bac-
teria. Biotech Bioeng Sym 1986; 16:253-262.
74. Sand W, Gerke T, Hallmann R et al. In situ bioleaching of metal sulfides: The
importance of Leptospirillum ferrooxidans. In: Torma AE, Wey JE, Lakshmanan
VI, eds. BiohydrometaUurgical Technologies. Vol I. Warrendale, Pennsylvania:
TMS Press 1993:15-27.
75. Shiratori T, Inoue C, Sugawara K et al. Cloning and expression of Thiobacillus
ferrooxidans mercury ion resistance genes in Escherichia coli. J Bacteriol 1989;
171:3458-3464·
76. Inoue C, Sugawara K, Shiratori T et al. Nucleo tide sequence of the Thiobacillus
ferrooxidans chromosomal gene encoding mercuric resistance. Gene 1989; 84:47-54.
77. Inoue C, Sugawara K, Kusano T. The merR regulatory gene in Thiobacillus
ferrooxidans is spaced apart from the mer structural genes. Mol Microbiol 1991;
5:2707-2718.
78. Rawlings DE, Woods DR. Development of improved biomining bacteria. In
Gaylarde CG, Videla HA, eds. Bioextraction and biodeterioration of metals. Cam-
bridge: Cambridge University Press. 1995; 63-84.
79. Rawlings DE, Silver S. Mining with microbes. Bio/Technology 1995; 13:773-778.
80. Rawlings DE, Deane SM, Butcher B. Unpublished observations.
81. Leong BJY, Dreisinger DB, Branion R et al. The microbiological leaching of a
sulfidic copper ore in a strongly saline medium (I): shakeflask and column stud-
ies. In: Torma AE, Wey JE, Lakshmanan VI, eds. Biohydrometallurgical Tech-
nologies. Vol I. Warrendale, Pennsylvania: TMS Press, 1993:117-126.
82. Lawson EN, Nicholas CJ, Pellat H. The toxic effects of chloride ions on Thiobacillus
ferrooxidans. In: Jerez CA, Vargas T, Toledo H, Wiertz JV, eds. Biohydrometal-
lurgical Processing. Vol I. Santiago: University of Chile Press, 1995:165-174.
CHAPTER 12
Thermophiles and Bioleaching

Paul R. Norris
Introduction
A cidophilic microorganisms that oxidize iron and sulfur can be exposed to high
temperatures in geothermal environments and in some heaps of ores and mine
wastes. A variety of types with different optimum temperatures for growth are
found across temperature gradients in the natural environments and may succeed
one another as exothermic oxidation reactions increase the temperature in the
industrial heaps. Most of the commercial, mineral-processing bioreactors are op-
erated with mesophilic bacteria at about 40°C1 (see chapter 3). Approaching this
temperature, however, activity of well-studied, mesophilic Proteobacteria such as
Thiobacillus ferrooxidans and Thiobacillus thiooxidans can be exceeded by that of
Thiobacillus caldus and Sulfobacillus species, which grow optimally at about 45°C.
At the extremes of their temperature ranges for growth, these moderate thermo-
philes can grow in mixed cultures with mesophiles or with extreme thermophiles.
One commercial bioreactor has been developed to utilize such organisms at 45°-50°C
for extraction of gold from a pyrite/arsenopyrite concentrate2 (see chapter 4). Be-
tween 50° and 55°C, their growth becomes progressively restricted whereas that of
the Sulfolobus-like archaea increases, with some strains active to at least 85°C. These
most thermophilic acidophiles are usually associated with sulfurous hot springs 3
but they have also been found in drainage of a copper mine 4 and in self-heating
heaps of waste from coal5 and uranium mining. 6
Thermophilic acidophiles may catalyze the solubilization of minerals in ore-
leaching heaps at temperatures that destroy mesophiles,7,8 The rapid dissolution
of finely ground mineral sulfide concentrates during autotrophic growth of mod-
erate9 and extreme thermophiles1o in agitated cultures has also been established
for some time but industrial application in stirred tanks is currently limited to a
single plant, as noted above. The efficient extraction of copper from chalcopyrite
concentrates, which cannot readily be achieved at low temperatures, is perhaps
the most notable potential application ofbioleaching at high temperatures. In this
chapter examples of the capacities of various thermophiles for metal extraction
from mineral sulfides follow an outline of the progress that has been made in char-
acterizing well-studied strains since previous reviews. n -13 The thermotolerant or

moderately thermophilic, Gram-positive bacterium Sulfobacillus thermosulfido-

oxidans and the extremely thermophilic, Sulfolobus-like archaea (of which several
genera have been described) have been the major subjects of work in this context.
The Microorganisms
Thermotolerance ofLeptospirillum Species
Leptospirillum ferrooxidans has appeared to be the major iron-oxidizing bac-
terium in laboratory bioreactors processing pyrite/arsenopyrite14 and zinc con-
centrates at 35°-40°C.15 Leptospirillum-like bacteria, rather than Thiobacillus
ferrooxidans, were previously shown to be the dominant iron -oxidizing acidophiles
at 40°C in pyrite-oxidizing mixed cultures from acidic drainage of mines, ore leach-
ing dumps and coal spoil sites.16 Only one out of four Leptospirillum-like strains
isolated from these cultures maintained some growth on pyrite at 45°C, however,
and this was beyond its optimum temperature. The precise temperature tolerance
of strains from cultures that appear to oxidize pyrite/arsenopyrite concentrates
efficiently at 45°C has not been reported.1A thermotolerant strain (proposed name
Leptospirillum thermoferrooxidans) with an optimum temperature for growth of
45°-50°C has been described.17 Further work is required to establish its mineral
sulfide-oxidizing capacities and its phylogenetic relationship to L. ferrooxidans.
Enrichment cultures of iron- and mineral sulfide-oxidizing organisms from vari-
ous locations generally comprise endospore-forming, Gram-positive Sulfobacillus
species at 45°-50°C rather than Leptospirillum-like bacteria.
Sulfobacillus Species
The first repore 8 of pyrite-oxidizing moderate thermophiles led to work with
" Thiobacillus-like" bacteria, such as strain THI (see ref. 12). This strain has since
been recognized as an isolate of Sulfobacillus thermosulfidooxidans.19.~O The phy-
logenetic relationships of moderately thermophilic Bacillus-like bacteria have been
established from analyses of their 16S rDNA sequences. S. thermosulfidooxidans is
most closely related~1.~~ to heterotrophic Alicydobacillus species~3 which share the
same acidic environments. However, the degree of relatedness is probably not as
close as suggested by one of the analyses 21 that appears to have assigned a sequence
from an Alicydobacillus species to S. thermosulfidooxidans. ~
Two species of Sulfobacillus, S. thermosulfidooxidans and S. acidophilus, have
been differentiated phenotypically as well as by 16S rDNA sequence analysis.
Chromosomal DNA from strains of S. thermosulfidooxidans and S. acidophi-
Ius has a guanine-cytosine content (mol% G + C) of 48-50 and 55-57, respectively.20
Iron-oxidizing cells of S. thermosulfidooxidans, but not those of S. acidophilus,
increase significantly in size when culture medium is supplemented with yeast
extract (Fig. 12.1a-d) and when growth is heterotrophic in the absence of iron. 20
The cells of both Sulfobacillus species can also appear elongated or in chains dur-
ing autotrophic growth on mineral sulfides when acidity develops to inhibitory
levels, and occasionally for unexplained reasons during growth on some concen-
trate samples. S. thermosulfidooxidans is the more active of the two species in oxi-
dation of iron and mineral sulfides in laboratory culture while S. acidophilus more
readily oxidizes sulfur, particularly in the absence of organic nutrients. 20 Strains
of S. thermosulfidooxidans and S. acidophilus were referred to as strains BCl and
ALV, respectively in earlier work that indicated a higher tolerance of ferric iron
Thermophiles and Bioleaching 249
•
(e
...
, (c) -
/
I
,
Fig. 12.1. Phase contrast microscopy of moderately thermophilic acidophiles grown on
ferrous iron. S.thermosulfidooxidans (a, b), S. acidophilus (c, d), A. ferrooxidans strain
ICP (e) and A. ferrooxidans strain TH3 (f) were grown autotrophically (a,c,e) or in the
presence of yeast extract (b, d, f). All to same scale; bar, 5 \lm.
could underlie the greater capacity of the former species for ferrous iron and py-
rite oxidation in batch culture!4 Both species appear to have a widespread distri-
bution in acidic environments, but their relative concentration or activity in ore
leaching heaps is unknown.
Several other moderate thermophiles with Suifobacillus-like morphologies and
similar capacities to grow autotrophic ally on ferrous iron and heterotrophically
on yeast extract have been isolated. Some can be distinguished from the named
Suifobacillus species on the basis of their mol% G + C content: 43 and 63 for strains
THWX and YTF1, respectivelt5 and 60 for strain LM2.'3 Unnamed Suifobacillus-
like strains have not yet been shown to have any advantages over S. thermosul-
fidooxidans for metal extraction from mineral sulfides. Examples of the character-
istics that may be worthy of further investigation, however, include the apparently
greater tolerance of acidity by strain }8 26 and a slightly greater temperature toler-
ance of a novel, yet-to-be-named species that grows at 62° but not 65°C (PR Norris,
unpublished data).
Acidimicrobium ferrooxidans
Autotrophic growth and associated oxidation of iron or pyrite by well-studied
strains of S. thermosulfidooxidans is slow when culture aeration is not supplemented
with carbon dioxide.'7 In contrast, some enrichment cultures oxidize iron and py-
rite rapidly at about 50°C under air!8 One of these enrichment cultures was found
to contain Suifobacillus-like organisms and a smaller bacterium of quite different
Fig. 12.2. Electron micrographs of thin sections of bacteria from a moderately ther-
mophilic, natural mixed-culture that oxidizes iron and pyrite rapidly under air
(a) and from cultures of S. thermosulfidooxidans (b) and A.ferrooxidans (c). All scale
bars; 0.5~.
morphology (Fig. 12.2), Acidimicrobium ferrooxidans. 29 The extent of iron oxida-

tion by A. ferrooxidans in batch culture is the same whether or not culture aeration
is supplemented with carbon dioxide, but it is less than that in carbon dioxide-
supplemented batch cultures of Sulfobacillus species that appear to have a greater
tolerance of ferric iron. When A. ferrooxidans is mixed with either S. thermo-
sulfidooxidans or S. acidophilus in culture without carbon dioxide supplementa-
tion, however, the extent of iron oxidation matches that by the Sulfobacillus spe-
cies under carbon dioxide-enriched air.29 The necessity for this mixture of the
bacteria to achieve extensive, growth-associated oxidation of ferrous iron under
air could be explained by efficient carbon dioxide uptake by A. ferrooxidans and
subsequent release of organic nutrients that are utilized by Sulfobacillus species.
Cells of A. ferrooxidans that are grown under air display a higher affinity for car-
bon dioxide uptake than cells grown under carbon dioxide-enriched air. This re-
sponse to a limiting carbon dioxide concentration is also shown by the mesophile
T.ferrooxidans but not by Sulfobacillus species (DA Clark and PR Norris, unpub-
lished data). A. ferrooxidans and T. ferrooxidans also share a high degree of simi-
larity in amino acid sequences of their key enzyme for fixation of accumulated
carbon dioxide, ribulose 1,5-bisphosphate carboxylase/oxygenase (RubisCO). This
similarity indicates that lateral transfer of the gene encoding the enzyme has oc-
curred between these organisms because on the basis of their 16S rRNA sequences,
they are not closely related phylogenetically (DA Clark and PR Norris, unpub-
lished data). In contrast, the phylogenetic distance between T. ferrooxidans and
Sulfobacillus species is clearly reflected in many more amino acid substitutions
outside of the conserved regions of their RubisCO sequences.30
A. ferrooxidans is not known to be as widely distributed as Sulfobacillus spe-

cies in acidic environments. However, it has been found in samples from a hot
spring29 (the type strain, ICP) and a copper leach dump7,31 (the strain formerly
referred to as TH3, which tends to grow more filamentously than strain ICP; Fig.
12.1). A moderate thermophile from acidic drainage of an Australian coal mine25
was probably another isolate of A. ferrooxidans, but was lost from culture before
direct comparison could be made with strains ICP and TH3. It also had a mol%
G+C content of 68 and, like strain TH3, tended to grow filamentously.
Thiobacillus caldus
Acidophilic thiobacilli that grow at higher temperatures than Thiobacillus
thiooxidans have been known for many years to inhabit hot springs.32,33 These bac-
teria do not oxidize ferrous iron. One strain with an optimum temperature for
growth on sulfur of 45°C and some capacity for growth at 55°C was referred to as
strain BC1334 before classification as Thiobacillus caldus. 35 Strain BC13 was used to
provide the sulfide/sulfur-oxidizing capacity that was required to complement iron
oxidation by L. ferrooxidans in a demonstration of mixed culture dissolution of a
chalcopyrite concentrate.13 The choice of the thermotolerant T. caldus rather than
the relatively heat -sensitive T. thiooxidans to partner L. ferrooxidans allowed fur-
ther illustration of the capacity of the iron-oxidizer to promote mineral sulfide
dissolution at temperatures that inhibited T. ferrooxidans. The leaching of zinc-
lead-iron concentrates in laboratory continuous reactors at 35°-40°C has appar-
ently involved a natural pairing of L. ferrooxidans and T. caldus as the dominant
iron - and sulfur-oxidizing bacteria, respectively.15 The rate of pyrite and chalcopy-
rite dissolution by S. thermosulfidooxidans is generally similar in the presence or
absence of T. caldus, but the pH is lower in the presence of the sulfur-oxidizing
thiobacilli, resulting in higher concentrations of iron in solution (PR Norris, un-
published work).
Sulfolobus-Like Archaea
The first report36 of mineral leaching at high temperature by a spherical, acido-
philic, iron- and sulfur-oxidizing organism closely followed the naming of a su-
perficially similar organism as Sulfolobus acidocaldarius.37 Both organisms were
from geothermal sites in Yellowstone National Park, USA. The leaching organism
was referred to as cferrolobus', and then known as Sulfolobus brierleyi before re-
classification as Acidianus brierleyi.38 Some confusion has surrounded Sulfolobus
acidocaldarius since cultures available from collections have been found by sev-
erallaboratories to comprise organisms that do not oxidize sulfur.10,39,40 It is pos-
sible that a non -sulfur-oxidizing, heterotrophic thermo acidophile became selected
from a mixed culture (yeast extract was usually included in culture media). The
exact nature of the thermophiles used under the name S. acidocaldarius in some
successful leaching studies (e.g., in coal desulfurization41 ) is therefore uncertain,
while the failure of some S. acidocaldarius cultures to oxidize mineral sulfides 42 is,
with hindsight, not surprising. Recent studies have further indicated that even sev-
eral of the species of Sulfolobus that do oxidize sulfur are not related closely enough
to be classified in the same genus. 43 Clearly, a variety of Sulfolobus-like thermo-
philes with some consequently different characteristics are available for evalua-
tion of their industrial potential for mineral sulfide-processing. Studies of mineral
sulfide oxidation at high temperature in several laboratories (e.g., refs. 44-47) have
utilized Sulfolobus strain Be. This strain was subsequently identified (PR Norris,
unpublished data) as an isolate of Sulfolobus metallicus,48 which is therefore now
recognized to inhabit hot springs and coal spoil heaps. Growth of S. metallicus is
inhibited at just over 70°C. Metallosphaera sedula has an optimum temperature of
about 75°C.39 Other Sulfolobus-like organisms that have been isolated from vari-
ous sulfur-rich, geothermal sites grow and oxidize mineral sulfides rapidly at
Boo-B5°C. These high temperature strains have yet to be classified and offer poten-
tially the most efficient leaching of chalcopyrite concentrates (see below).
Metabolism and Molecular Biology of Thermophiles

The following summary of studies that principally concern sulfur oxidation,
iron oxidation and the fixation of carbon dioxide by moderate and extreme ther-
mophiles indicates that most information concerning these key aspects of metabo-
lism for their growth on mineral sulfides is of a preliminary nature and has not yet
influenced attempts to optimize mineral leaching.
Sulfobacillus species are nutritionally versatile, being capable of autotrophic
growth using the Calvin cycle, mixotrophic growth with simultaneous utilization
of carbon dioxide and glucose49 and heterotrophic growth. 20 T. caldus probably
shares mechanisms of autotrophic growth with the closely related T. thiooxidans
and in addition, it appears to utilize glucose during growth on sulfur compounds.34.35
Sulfolobus metallicus has been described as obligately autotrophic 48 but can utilize
organic compounds during growth on sulfur compounds.50 The details of the path-
way of carbon dioxide fixation by the Sulfolobus-like archaea are not known. Acetyl
CoA carboxylation was suggested to be a key step in the pathway in S. metallicus.5'
When growth of this organism is carbon dioxide-limited, it increases synthesis of
a biotin carboxylase and biotin-carboxyl-carrier protein complex (NP Burton and
PR Norris, unpublished data). This complex is encoded by genes that are adjacent
to a gene encoding a putative propionyl CoA carboxyl transferase. These observa-
tions may be in accord with a suggestion that A. brierleyi has a modified
3-hydroxypropionate pathway for carbon dioxide fixation: acetyl and propionyl
CoA carboxylases were induced when this organism was switched from het-
erotrophic to autotrophic growth. 52
The kinetics of sulfur compound oxidation by T. caldus 53•54 and S. metallicus53
have been described. A sulfur oxygenase has been identified in Sulfolobus (now
Acidianus) brierleyi55 and in Desulfurolobus (now Acidianus) ambivalens and the
gene encoding the enzyme from the latter organism has been sequenced.56 Mecha-
nisms of iron oxidation in the thermophilic acidophiles are unknown. A novel cy-
tochrome is produced by S. metallicus, A. brierleyi and M. sedula only when they
are oxidizing ferrous iron 57.58 and is likely to be involved in electron transport from
this substrate, but it remains to be characterized. A protein that is greatly increased
in concentration in S. metallicus when this organism is grown on ferrous iron in-
stead of sulfur compounds is an anti-oxidant protein rather than a likely com-
ponent of an iron oxidation system.59 It might be involved in protection against
radical ions that could be formed intracellularly during 'iron overload' of cells.
Other proteins are involved in the thermal stress response of S. metallicus60 and
M.sedula. 61
A variety of plasmids (about 2-20 kb in size) have been detected in species of
Sulfobacillus and a 2.6 kb plasmid of S. thermosulfidooxidans has been sequenced.62
Gene transfer in the iron-oxidizing moderate thermophiles remains to be achieved,
Thermophi/es and Bioleaching 253
however. The development of gene transfer systems for the thermophilic, acido-
philic archaea has so far focused 63 on the heterotrophic strains of Sulfolobus (see
earlier) rather than on those involved in mineral sulfide oxidation.
Mineral Sulfide Oxidation by Thermophiles

Some of the early work assessing the potential application of high temperature
bioleaching involved low-grade ores in columns to simulate leaching heaps (see
ref. 12), but more recently the emphasis has been on oxidation of fmely ground
concentrates in reactors. Sulfolobus-like organisms produce the highest rates of
mineral sulfide dissolution when the mineral concentration is low, but the devel-
opment of high-temperature bioleaching may be restricted by inhibition of these
thermophiles at high concentrations of solids. These observations are illustrated
by the solubilization of iron from a pyrite/arsenopyrite concentrate by various
cultures in air-lift reactors gassed with 1% v/v CO 2 in air (Fig. 12.3). The principal
iron-oxidizing bacteria in the mixed cultures used at the lower temperatures were
the mesophiles L. ferrooxidans and T. ferrooxidans and the moderate thermophile
S. thermosulfidooxidans. Mesophiles, moderate thermophiles and Sulfolobus-like
organisms have also been used with a gold-bearing pyrite/arsenopyrite concen-
trate at 5% w/v solids to show progressively more efficient metal extraction as the
temperature was increased. 65 Most direct comparisons of this range of organisms,
however, have used lower concentrations of minerals (e.g., ref. 13). The effects of
the concentration of solids has not been studied as extensively with Sulfobacillus
species, A. ferrooxidans and T. caldus as with the mesophile T. ferrooxidans, but
there are no indications that the moderate thermophiles have a particularly greater
sensitivity than mesophiles to agitation at high mineral concentrations. The greater
rates of mineral oxidation that can be obtained at 48°C versus 30°C, at least with
10% w/v mineral for example (Fig. 12.3), could therefore be preserved at the higher
mineral concentrations preferred for industrial application. It remains to be shown
whether the potential benefit of using moderate thermophiles rather than
mesophiles is generally seen with continuous culture operation as well as in batch
leaching systems. One comparison of continuous leaching of a refractory, gold-
bearing pyrite/arsenopyrite concentrate by mesophiles (35°C) and moderate ther-
mophiles (45°C) has confirmed the potential advantages of operating at the higher
temperature with some concentrates66 and, as noted earlier, 2 one commercial pro-
cess already utilizes moderate thermophiles at 45°-50°C.
Maximum rates of mineral concentrate leaching by Sulfolobus metallicus have
been obtained with about 10% w/v pyrite (PR Norris, unpublished data) and 15%
w/v chalcopyrite45 concentrates in air-lift reactors and with 6-8% w/v pyrite in
stirred reactors. 67 The mineral concentration rather than the stirring speed was
critical in inhibition of the organism.67 There are indications that different strains
of Sulfolobus-like organisms have different tolerances of high concentrations of
solids 68 and more screening is required to find the most resilient strains. Gener-
ally, however, advantages of oxidizing pyrite or arsenopyrite with extreme ther-
mophiles rather than moderate thermophiles remain to be demonstrated at min-
eral concentrations that exceed about 5% w/v in stirred reactors.
A clear potential advantage of high temperature leaching has been demonstrated
with minerals, including chalcopyrite, for which the extent as well as the rate of
metal extraction is dependent on the temperature. The problem of incomplete ex-
traction of copper from chalcopyrite during low temperature bioleaching is well
Iron in solution A 15 Iron in solution

(gi l) (gil)
4
.8 ,
...
.~
I
I
I
3 10 I
I
I
~
2 f
•
I
..
I
5 I
·c
tJ
• Meso philes
A Moderate
thermophiles
30
48
... . . . ."
"." .".
• S . mQtaUicus 68
100 200 300 400 100 200 300 400 500

Time (hours)
Fig. 12.3. The solubilization of iron from (A) 2% and (B) 10% w/v pyrite/arsenopyrite
concentrate in cultures growing autotrophically at close to their optimum tempera-
tures in air-lift reactors. (Adapted from Clark DA and Norris PR. 64)
8
Copper in
solution
(g/l)
0"0
• 0 ..., ,. .
.....
4
·c
o 70
.d • 75
2
,.cv
• 80
..:I 84
0 " " _C .. ··o ' A 88
o 92
50 100 150 200

Time (hours)
Fig. 12-40 The effect of temperature on copper extraction from a chalcopyrite concen-
trate during autotrophic growth of a culture of Sulfolobus-like organisms (see text for
experimental details).
known. Figure 12.4 shows the dissolution of a refractory copper concentrate dur-
ing growth of a Sulfolobus-like organism at different temperatures in air-lift reac-
tors that were gassed with 1% v/v CO 2 in air. An initial mineral concentration of 1%
w/v was supplemented with a further 4% w/v mineral after 43 hours of incubation
(additional mineral was not added at 88° and 92°C because growth of the organ-
ism and copper extraction were clearly inhibited at these temperatures). The sig-
nificance of utilizing thermophiles that grow at higher temperatures than S.
metallicus was shown by fmal yields of 80% of the copper in solution at 80°-84°C
compared to 65% in solution at 70°C. Further work with a range of chalcopyrite
concentrates from various sources has shown solubilization of up to 95% of the
copper during growth of such cultures at about 80°C.
There is now considerable familiarity with the capacities of various microor-
ganisms to oxidize a range of mineral sulfides at elevated temperatures. There are
also novel strains already isolated but yet to be examined in this context. The greatest
potential advantages of utilizing thermophiles, however, may only be realized in-
dustriallywith modifications to standard bioreactor designs to facilitate high tem-
perature operation with the Sulfolobus-like organisms.
References
1. Dew OW. Comparison of performance for continuous bio-oxidation of refractory
gold ore flotation concentrates. In: Vargas T, Jerez CA, Wiertz JV et al, eds.
Biohydrometallurgical Processing. Vol I. Santiago: University of Chile, 1995:239-251.
2. Brierley CL, Brans R. Selection of Bactech's thermophilic biooxidation process
for Youanmi mine. In: Biomine '94. Glendale: Australian Mineral Foundation,
1994:5-1-5·7·
3. Brock TO. Thermophilic Microorganisms and Life at High Temperatures. New
York: Springer-Verlag, 1978.
4. G6mez E, L6pez AI, Marin I et a1. Isolation and characterization of novel
bioleaching microorganisms from Rio Tinto. In: Torma AE, Apel ML, Brierley
CL, eds. Biohydrometallurgical Technologies. Vol 2. Warrendale, PA: The Miner-
als, Metals and Materials Society, 1993:479-486.
5. Marsh RM, Norris PRo The isolation of some thermophilic, autotrophic, iron- and
sulfur-oxidizing bacteria. FEMS Microbiol Lett 1983; 17:311-315.
6. Fuchs T, Huber H, Teiner K et a1. Metallosphaera prunae, sp. nov., a novel metal-
mobilizing, thermo acidophilic Archaeum, isolated from a uranium mine in Ger-
many. System Appl Microbiol 1995; 18:560-566.
7. Brierley JA. Thermophilic iron-oxidizing bacteria found in copper leaching dumps.
8. Murr LE, Brierley JA. The use of large-scale test facilities in studies of the role of
microorganisms in commercial mineral leaching operations. In: Murr LE, Torma
AE, Brierley JA, eds. Metallurgical Applications of Bacterial Leaching and Related
Microbiological Phenomena. New York: Academic Press, 1983:491-520.
9. Marsh RM, Norris PRo Mineral sulfide oxidation by moderately thermophilic aci-
dophilic bacteria. Biotechnol Lett 1983; 5:585-590.
10. Marsh RM, Norris PR, Le Roux NW. Growth and mineral oxidation studies with
Sulfolobus. In: Rossi G, Torma AE, eds. Recent Progress in Biohydrometallurgy.
Iglesias: Associazione Mineraria Sarda, 1983:71-81.
11. Brierley CL. Bacterial leaching. Crit Rev Microbiol 1978; 6:207-262.
12. Brierley JA, Brierley CL. Microbial mining using thermophilic microorganisms.
In: Brock TO, ed. Thermophiles: General, Molecular and Applied Microbiology.
New York: Wiley, 1986:279-305.
In: Ehrlich HL, Brierley, CL, eds. Microbial Mineral Recovery. New York: McGraw-
Hill, 1990:3-27.
14. Rawlings DE. Restriction enzyme analysis of 16S rRNA genes for the rapid iden-
ferrooxidans strains in leaching environments. In: Vargas T, Jerez CA, Wiertz JV
et aI, eds. Biohydrometallurgical Processing. Vol II. Santiago: University of Chile,
1995:9-17.
15. Goebel BM, Stackebrandt E. Cultural and phylogenetical analysis of mixed mi-
crobial populations found in natural and commercial bioleaching environments.
Appl Environ Microbiol1994; 60:1614-1621.
16. Norris PRo Iron and mineral oxidation studies with Leptospirillum-like bacteria.
In: Rossi G, Torma AE, eds. Recent Progress in Biohydrometallurgy. Iglesias:
Associazione Mineraria Sarda, 1983:83-96.
17. Golovacheva RS, Golyshina OV, Karavaiko GI et al. A new iron-oxidizing bacte-
rium, Leptospirillum thermoferrooxidans sp. nov. Mikrobiologiya 1992; 61:744-750.
18. Le Roux NW, Wakerley DS, Hunt SD. Thermophilic Thiobacillus-type bacteria
from Icelandic thermal areas. J Gen Microbiol 1977; 100:197-201.
19. Golovacheva RS, Karavaiko GI. Sulfobacillus-a new genus of spore-forming ther-
mophilic bacteria. Microbiology (trans. Mikrobiologiya) 1979; 48:658-665.
20. Norris PR, Clark DA, Owen JP, Waterhouse S. Characteristics of Sulfobacillus
acidophilus sp. nov. and other moderately thermophilic mineral sulfide-oxidizing
bacteria. Microbiology 1996; 142:775-783.
21. Tourova TP, Poltoraus AB, Lebedeva IA et al. 16S ribosomal RNA (rDNA) se-
quence analysis and phylogentic position of Sulfobacillus thermosulfidooxidans.
System Appl Microbiol 1994; 17:509-512.
22. Durand P. Primary structure of the 16S rRNA gene of Sulfobacillus thermosulfido-
oxidans by direct sequencing of PCR amplified gene and its similarity with that
of other moderately thermophilic chemolithotrophic bacteria. System Appl
Microbiol 1996; 19:360-364.
23. Wisotzkey JD, Jurtshuk Jr P, Fox GE et al. Comparative sequence analyzes on the
16S rRNA (rDNA) of Bacillus acidocaldarius, Bacillus acidoterrestris, and Bacillus
cycloheptanicus and proposal for creation of a new genus, Alicyclobaci/lus gen.
nov. Int J Syst Bacteriol 1992; 42:263-269.
24. Norris PR, Barr, DW, Hinson, D. Iron and mineral oxidation by acidophilic bac-
teria: affinities for iron and attachment to pyrite. In: Norris PR, Kelly DP, eds.
Biohydro-metallurgy, Int Symp Proc. Kew: Science and Technology Letters,
1988:43-59.
25. Ghauri MA, Johnson DB. Physiological diversity amongst some moderately ther-
mophilic iron-oxidizing bacteria. FEMS Microbiol Ecol 1991; 85:327-334.
26. Hendy NA. Isolation of thermophilic iron-oxidizing bacteria from sulfidic waste
rock. J Ind Microbiol 1987; 1:389-392.
27. Norris PRo Factors affecting bacterial mineral oxidation: the example of carbon
dioxide in the context of bacterial diversity. In: Salley J, McCready RGL, Wichlacz
PL, eds. Biohydrometallurgy 1989. Ontario: CANMET, 1989:3-14.
28. Norris PR, Owen JP. Mineral sulfide oxidation by enrichment cultures of novel
thermo acidophilic bacteria. FEMS Microbiol Rev 1993; 11:51-56.
29. Clark DA, Norris PRo Acidimicrobium ferrooxidans gen. nov., sp. nov.: mixed-
culture ferrous iron oxidation with Sulfobacillus species. Microbiology 1996;
142:785-790.
30. Holden PJ, Brown RW. Amplification of ribulose bisphosphate carboxylase!
oxygenase large subunit (RuBisCO LSU) gene fragments from Thiobacillus
ferrooxidans and a moderate thermophile using polymerase chain reaction. FEMS

Microbiol Rev 1993; 11:19-30.
31. Norris PR, Barr OW. Growth and iron oxidation by acidophilic moderate ther-
mophiles. FEMS Microbiol Lett 1995; 28:221-224.
32. Schwartz A, Schwartz W. Geomikrobiologische Untersuchungen VII. Ober das
Vorkommen von Mikroorganismen in solfataren und heissen Quellen. Z Allg
Mikrobiol 1965; 5:395-405.
33. Fliermans CB, Brock TO. Ecology of sulfur-oxidizing bacteria in hot acid soils. J
Bacteriol 1972; 111:343-350.
34. Norris PR, Marsh RM, Lindstrom EB. Growth of mesophilic and thermophilic
acidophilic bacteria on sulfur and tetrathionate. Biotechnol Appl Biochem 1986;
8:318-329·
35. Hallberg KB, Lindstrom EB. Characterization of Thiobacillus caldus sp. nov., a
moderately thermophilic acidophile. Microbiology 1994; 140:3451-3456.
36. Brierley CL, Murr LE. Leaching: use of a thermophilic and chemoautotrophic
microbe. Science 1973; 179:488-490.
37. Brock TO, Brock KM, Belly RT, Weiss RL. Sulfolobus: a new genus of sulfur-oxi-
dizing bacteria living at low pH and high temperature. Arch Microbiol 1972;
84:54-68.
38. Segerer A, Neuner A, Kristjansson JK et al. Acidianus infernus gen. nov., sp. nov.,
and Acidianus brierleyi comb. nov.: facultatively aerobic, extremely acidophilic
thermophilic sulfur-metabolizing archae bacteria. Int J Syst Bacteriol 1986;
36:559-564·
39. Huber G, Spinnler C, Gambacorta A et al. Metallosphaera sedula gen. and sp.
nov. represents a new genus of aerobic, metal-mobilizing, thermo acidophilic
archaebacteria. System Appl Microbiol 1989; 12:38-47.
40. Zillig W, Kletzin A, Schleper C et al. Screening for Sulfobales, their plasmids and
their viruses. In: Pfeifer F, Palm P. Schleifer K-H, eds. Molecular Biology of
Archaea. Stuttgart: Fischer, 1994:109-128.
41. Kargi F, Robinson JM. Removal of sulfur compounds from coal by the thermo-
philic organisms Sulfolobus acidocaldarius. Appl Environ Microbiol 1982;
44:878-883·
42. Larsson L, Olsson G, Holst 0 et al. Pyrite oxidation by thermophilic archaebacteria.
43. Fuchs T, Huber H, Burggraf S et al. 16S rONA-based phylogeny of the archaeal
order Sulfolobales and reclassification of Desulfurolobus ambivalens as Acidianus
ambivalens comb. nov. System Appl Microbiol 1996; 19:56-60.
44. Norris PR, Parrott L. High temperature, mineral concentrate dissolution with
Slilfolobus. In: Lawrence RW, Branion RMR, Ebner HG, eds. Fundamental and
Applied Biohydrometallurgy. Amsterdam: Elsevier, 1986:355-365.
45. Le Roux NW, Wakerley OS. Leaching of chalcopyrite (CuFeS.) at 70 c C using
Sulfolobus. In: Norris PR, Kelly OP, eds. Biohydrometallurgy, Proc Int Symp. Kew:
Science and Technology Letters, 1988:305-317.
46. Liu X, Lindstrom EB, Petersson S. Bioleaching of refractory arsenical pyrite con-
centrates to enhance gold recovery. Scan dan avian J Metallurgy 1991; 20:346-350.
47. Barr OW, Jordan MA, Norris PR et al. An investigation into bacterial cell, ferrous
iron, pH and Eh interactions during thermophilic leaching of copper concentrates.
Minerals Engineering 1992; 5:557-567.
48. Huber G, Stetter KO. Sulfolobus metallicus, sp. nov., a novel strictly chemolitho-
trophic thermophilic archaeal species of metal-mobilizers. System Appl Microbiol
1991; 14:372-378.
49. Wood AP, Kelly DP. Growth and sugar metabolism of a thermoacidophilic iron-
oxidizing mixotrophic bacterium. Microbiology 1984; 130:1337-1349.
50. Wood AP, Kelly DP, Norris PR. Autotrophic growth of four Sulfolobus strains on
tetrathionate and the effect of organic nutrients. Arch Microbiol1986; 146:382-389.
51. Norris PR, Nixon A, Hart A. Acidophilic, mineral-oxidizing bacteria: the utiliza-
tion of carbon dioxide with particular reference to autotrophy in Sulfolobus. In:
de Costa MS, Duarte IC, Williams RAD, eds. Microbiology of Extreme Environ-
ments and Its Potential for Biotechnology. London: Elsevier, 1989:24-43.
52. Ishii M, Miyake T, Satoh T et al. Autotrophic carbon dioxide fixation in Acidianus
brierleyi. Arch Microbiol 1997; 166:368-371.
53. Nixon A, Norris PR. Autotrophic growth and inorganic sulfur compound oxida-
tion by Sulfolobus species in chemostat culture. Arch Microbiol1992; 157:155-160.
54. Hallberg KB, Dopson M, Lindstrom EB. Reduced sulfur compound oxidation by
Thiobacillus caldus. I Bacteriol 1996; 178:6-11.
55. Emmel T, Sand W, Konig WA et al. Evidence for the existence of a sulfur
oxygenase in Sulfolobus brierleyi. I Gen Microbiol1986; 132:3415-3420.
56. Kletzin A. Sulfur oxidation and reduction in archaea: sulfur oxygenase/reductase
and hydrogenase from the extremely thermophilic and facultatively anaerobic
archaeon Desulfurolobus ambivalens. In: Pfeifer F, Palm P. Schleifer K-H, eds.
Molecular Biology of Archaea. Stuttgart: Fischer, 1994:34-43.
57. Barr DW, Ingledew WI, Norris PR. Respiratory chain components of iron-oxidiz-
ing, acidophilic bacteria. FEMS Microbiol Lett 1990; 70:85-90.
58. Blake R, Shute EA, Waskovsky I et al. Respiratory components in acidophilic
bacteria that respire on iron. Geomicrobiol I 1992; 10:173-192.
59. Burton NP, Williams TD, Norris PR. A potential anti-oxidant protein in a ferrous
iron-oxidizing Sulfolobus species. FEMS Microbiol Lett 1995; 134:91-95.
60. Ierez CA. The heat shock response in meso- and thermo acidophilic chemolitho-
trophic bacteria. FEMS Microbiol Lett 1988; 56:289-294.
61. Peeples TL, Kelly RM. Bioenergetic response of the extreme thermoacidophile
Metallosphaera sedula to thermal and nutritional stress. Appl Environ Microbiol
1995; 61:2314-2321.
62. Burton NP, Gibson FE, Murrell IC et al. Development of genetic systems for mod-
erately thermophilic, mineral sulfide-oxidizing bacteria. In: Alberghina L, Frontali
L, Sensi P, eds. Proceedings of the 6th European Congress on Biotechnology.
Amsterdam: Elsevier 1994, 1169-1172.
63. Zillig W, Prangishvilli D, Schleper C et al. Viruses, plasm ids and other genetic
elements of thermophilic and hyperthermophilic Archaea. FEMS Microbiol Rev
1996; 18:225-236.
64. Clark DA, Norris PR. Oxidation of mineral sulfides by thermophilic microorgan-
isms. Minerals Engineering 1996; 9:1119-1125.
65. Hutchins SR, Brierley lA, Brierley CL. Microbial pretreatment of refractory sul-
fide and carbonaceous gold ores. In: Vassiliou AH, Hausen DM, Carson DIT, eds.
Process Mineralogy VII. Warrendale, PA: The Metallurgical Society, 1987:53-66.
66. Liu X, Petersson S, Sandstrom A. Mesophilic versus moderate thermophilic
bioleaching. In: Torma AE, Wey IE, Lakshmanan VI, eds. Biohydrometallurgical
Technologies. Vol 1. Warrendale, PA: The Minerals, Metals & Materials Society,
1993:29-38.
67. Lindstrom EB, Wold S, Kettanch-Wold N et al. Optimization of pyrite bioleaching
using Sulfolobus acidocaldarius. Appl Microbiol Biotechnol 1993; 38:702-707.
68. Norris PR, Owen IP. Strain selection for high temperature oxidation of mineral
sulfides in reactors. In: Ladisch MR, Bose A, eds. Harnessing Biotechnology for
the 21st Century. Washington: American Chemical Society, 1992:445-448.
CHAPTER 13
Heterotrophic Acidophiles and

Their Roles in the Bioleaching
of Sulfide Minerals
D. Barrie Johnson and Francisco F. Roberto
Introduction
T he most familiar and well-studied microorganisms indigenous to acidic min-

eralleaching environments are autotrophic sulfur- and iron-oxidizing bacteria
such as Thiobacillus ferrooxidans, Thiobacillus thiooxidans and Leptospirillum
ferrooxidans. Some photoautotrophs, such as the thermophilic rhodophyte
Cyanidium caldarium, may also be present in extremely acidic environments that
receive light. Other microorganisms which require pre-fixed (organic) carbon have
been isolated from mineral leach dumps and acid mine drainage (AMD) waters.
These heterotrophic microorganisms include eukaryotes, such as some fungi and
yeastsl and protozoa,2 as well as prokaryotic bacteria and archaea. It is somewhat
paradoxical, given that heterotrophy is the most widespread form of metabolism
among bacteria, that the first acidophilic heterotrophic bacterium which is indig-
enous and active in mineral leaching environments was isolated and character-
ized some 40 years after the iron/sulfur-oxidizing chemolithotroph T.ferrooxidans
and 70 years after the sulfur-oxidizing acidophile T. thiooxidans.
Biodiversity of Acidophilic Heterotrophic Prokaryotes

Mesophilic Bacteria
Acidophilic heterotrophic bacteria with temperature optima of 25°-37°C have
been isolated from AMD waters, mineral leach dumps and supposedly pure cul-
tures of chemolithotrophic acidophiles. Many of the first heterotrophic bacteria to
be isolated from AMD streams were found not to grow in synthetic media poised
at similar acidic pH values to the source materials, suggesting that obligatelyaci-
dophilic heterotrophs might not occur in such environments. For example, mac-
roscopic gelatinous growths colonizing AMD streams, known as 'acid streamers'
were considered by Dugan et al3 to be composed of primarily neutrophilic and
acid-tolerant heterotrophic bacteria that created and inhabited less acidic micro-
cosms within the slime matrix of the streamers. Wakao et al, 4 however, concluded
Biomining: Theory, Microbes and Industrial Processes.

that acid streamers from an abandoned sulfur/iron sulfide mine in Japan were
produced by a strain of the chemolithotroph T. ferrooxidans which synthesized
copious amounts of extracellular polymers, though a bacterium of this kind was
not actually isolated from the streamers. Acid streamers found in an abandoned
(70 years) pyrite mine (Cae Coch) in North Wales were reported by Johnson 5 to
contain a range of morphologically-distinct bacteria; isolates obtained by plating
streamer fragments onto selective solid media included a range of neutrophilic
heterotrophs (Bacillus spp. and others), acidophilic chemolithotrophs (T. ferro-
oxidans and L. ferrooxidans) and acidophilic heterotrophs (Acidiphilium spp.). More
recently, the same material has been the source of heterotrophic iron-oxidizing
acidophiles 6'7 and acid -tolerant sulfate-reducing bacteria,8 indicating that this most
obvious and dramatic manifestation of microbial life in acidic mine environments
(acid streamers within the Cae Coch mine have an estimated biovolume of over
100 m 3) can be highly complex in microbial diversity.
The first truly acidophilic heterotrophic bacteria were isolated and character-
ized in the early 1980s,9-11 often from supposedly pure cultures of T. ferrooxidans.
Chemolithotropic iron-oxidizing bacteria and heterotrophic acidophiles can form
highly stable mixed cultures; subculturing through 'inorganic' ferrous sulfate or
pyrite media may not eliminate the latter, which effectively scavenge the small
quantities of organic materials leaked by active and moribund T. ferrooxidans (or
L. ferrooxidans) cells, as well as those originating from other sources (contami-
nant materials present in 'inorganic' media, and atmospheric inputs). Such close
associations are probably why some strains of T. ferrooxidans had been reported
to grow heterotrophically on glucose and to lose their capacity for ferrous iron
oxidation when subcultured in organic media.12 Some years before the isolation of
the first obligately heterotrophic acidophilic bacterium, Thiobacillus acidophilus
had been isolated from a culture of T. ferrooxidans. 13 While T. acidophilus is highly
versatile in its metabolism (it can grow autotrophic ally or mixotrophically in the
presence of reduced sulfur compounds, or heterotrophically in their absence), its
presence in cultures of T. ferrooxidans grown in ferrous sulfate medium indicates
that it was living as a heterotrophic 'satellite' organism in these cultures (as
Acidiphilium isolates were later also shown to do). Interestingly, from a phyloge-
netic viewpoint, T. acidophilus is much more closely related to Acidiphilium spp.
than to other Thiobacillus spp.
The genus Acidiphilium was first introduced by Harrison1o to describe aerobic,
mesophilic rod-shaped bacteria that grow in lean organic media between pH 1.9
and 6.1. The type species, A. cryptum, was so-called to reflect its clandestine lifestyle
in laboratory cultures of T. ferrooxidans, though similar isolates were also obtained
from a variety of acidic mineral environments. Many other obligately acidophilic
heterotrophic mesophilic bacteria that have been isolated and characterized since,
also appear to be Acidiphilium spp. (Table 13-1). Differentiation of isolates from A.
cryptum has often been on the basis of physiological and nutritional characteris-
tics; chromosomal DNA base composition (G + C contents), DNA homology stud-
ies and lipid analysis. More recently, 16S rRNA gene sequencing has also been used
to confirm the novel status of Acidiphilium species.
Besides Acidiphilium, three other genera of mesophilic, heterotrophic acido-
philic bacteria are currently recognized (Table 13.1). Two previous Acidiphilium
isolates (A. facilis and A. aminolytica) were transferred to the genus Acidocella
following comparative 16S rDNA sequence analysis. 14 Acidimonas methanolica15 is
Heterotrophic Acidophiles in the Bioleaching of Sulfide Minerals 261
Table 13.1. Characteristics ofclassified species ofacidophilic heterotrophic

bacteria and archaea
Temperature Growth Rate

Organism G+C pH (OC) (max.)
(mol%) Opt. Range Opt. Range (t~h)
Mesophilic Bacteria
Acidiphilium spp.
A. cryptum 64-70 a 3 1·9-5·9 35-41 20-41 6.0
A. symbioticum 59-60 3-4 1·5-5 37 3.8
A. rubrum 63 2·5-6 14
A. angustum 67 2·5-6 11
A. organovorum 64 3 2-5·5 37 20-45 2·5
A. multivorum 66-68 -3·5 1.9-5.6 27-35 17-42 5·0
Acidocella spp.
Ad. facilis 65 2.5-6 25-37 3·3
Ad. aminolytica 59 3-6 20-37
Acidomonas
methanolica 63-65 2-5·5 <30-42 2.8
Acidobacterium
capsula tum 60 3-6 20-37
Thiobacillus acidophilusb 63-64 2·5-3 1·5-6·5 27-30 <25-37 9.0

Ferromicrobium
acidophilus 51-55 2-2·5 1·3-4·8 37 <20-40 5·5
Thermophilic Bacteria
Alicyclobacillus spp.
AI. acidocaldarius 61-62 3-4 2-6 60-65 45-70 0.6
AI. acidoterrestris 51-53 2.2-5·8 4 2-53 <35-55
AI. cydoheptanicus 54-57 3-5·5 48 40-53
Thermophilic Archaea
Thermoplasma spp.
Th. acidophilum 46 1-2 0·5-4 59 45- 63 5
Th. volcanium 38-40 2 1-4 59 33-67 5
Picrophilus spp.
P.oshimae 36 0·7 0-3·5 60 45-65 6
P. torridus 0·7 0-3·5 60 6
a 63-64% in recent reports; b also mixotrophic/autotrophic growth (see text).
a methylotrophic acidophile, and Acidobacterium capsulatum 16 a saccharolytic,

capsulated, and somewhat less acidophilic heterotroph. In common with all other
classified mesophilic acidophilic heterotrophs (and mixotrophic Thiobacillus spp.),
all of these bacteria are Gram-negative, nonsporulating rods.
Some mesophilic heterotrophic acidophiles also appear to be able to oxidize
ferrous iron. One such isolate, CCH7, was reported to be morphologically similar
to the iron oxidizing/precipitating neutrophilic bacteria Leptothrix and Sphaerotilus
spp.6 Isolate CCH7, which forms acid streamer-like growths in liquid media con-
taining ferrous sulfate and yeast extract, does not conserve the energy from iron
oxidation. In contrast, motile unicellular rod-shaped iron-oxidizing acidophiles
which require pre-fixed carbon (e.g., as yeast extract) and which do appear to use
iron oxidation as an energy-transducing reaction have been isolated from a vari-
ety of sourcesY7
Thermophilic Bacteria
Moderately thermophilic iron-oxidizing acidophilic bacteria differ from their
mesophilic counterparts in several fundamental ways, such as their highly versa-
tile nutritional capacities. IS Some isolates have been shown to grow heterotrophi-
cally on yeast extract, though growth rates are much slower than in media amended
with ferrous sulfate.19 These bacteria (Sulfobacillus and Acidimicrobium spp.) are
not primarily heterotrophic and will not be considered further in this chapter (see
chapter 12). Moderately thermophilic, obligately heterotrophic Bacillus-like
acidophiles have been isolated from acidic geothermal areas (e.g., hot springs, soils)
from various parts of the world. A common characteristic of three phenotypically-
related species (originally named B. acidocaldarius, B. acidoterrestris and B. cyclo-
heptanicus) is the possession of a unique type of lipid (w-alicyclic fatty acids) as
the major fatty acid component of their membranes. 2o Phylogenetic relatedness
(from 165 rDNA analysis) of these three acidophiles, and distinct differences be-
tween them and other Bacillus spp. has prompted their reclassification as a sepa-
rate genus, Alicyclobacillus. 20 Recently characterized isolates from two acidic geo-
thermal sites in Yellowstone National Park displayed some similar physiological
traits to Alicyclobacillus Spp.,21 and were also noted to reduce ferric iron when grown
under microaerophilic conditions.
Thermophilic Archaea
Besides the Alicyclobacillus-like bacteria described above, the only heterotrophic
microorganisms to have been isolated from acidic geothermal sites where the
ambient temperature exceeds 50°C are thermophilic archaea. Most of the
Sulfolobales (extremely thermophilic archaea, including Sulfolobus and Acidianus
spp.) can grow heterotrophically, but are more well-known for their sulfur-depen-
dent chemolithotrophic metabolisms and will not be considered in this section
(see chapter 12). Obligately heterotrophic archaea tend to be less thermotolerant
(temperature range generally 45°-65°C) than the Sulfolobales, though they do have
the distinction of being the most acidophilic of any microorganism to have been
isolated and characterized. The first of these, Thermoplasma acidophilum, was iso-
lated by Darland et al22 from a coal spoil heap that had undergone self-heating.
This archae on is considered to be widely distributed in thermal acidic environ-
ments, having been isolated from a variety of coal refuse piles in the USA23 and hot
springs in Japan. 24 Th. acidophilum is unique among archaea in lacking a cell en-
Heterotrophic Acidophiles in the Bioleaching of Sulfide Minerals
velope, and can grow either aerobically, or anaerobically by sulfur respiration. A

related species, Th. volcanium, has been isolated from various sites around the
world. 25 The recently characterized genus Picrophilus comprises two physiologi-
cally-similar isolates (P. oshimae and P. torridus) which differ from Thermoplasma
in being obligate aerobes and in possessing cellular surface layers, though of a
different structure to other archaea. 26 Probably the most intriguing facet of
Picrophilus spp. is, however, their requirement and tolerance of extreme acidity;
the pH optimum of isolates is 0.7, and growth occurs in media poised at 0.0.
Molecular Phylogeny ofAcidophilic Heterotrophic Bacteria

Unlike the wide phylogenetic distribution of the iron- and sulfur-oxidizing
phenotype in the eubacteria,27 the majority of acidophilic heterotrophic bacteria
fall into coherent, tightly associated groups of evolutionarily distinct lineages which
correlate well with their optimum growth temperatures. The report by Lane et al27
of evolutionary relationships between a variety of iron- and sulfur-oxidizing bac-
teria from diverse environments included organisms with some quite different
physiologies (only a subset were actually obligate acidophiles), which may explain
the seeming lack of phylogenetic coherence which was found.
From sequence analysis of 16S rRNA genes, it appears that known acidophilic
heterotrophs fall into a relatively small number (four) of widely-spaced bacterial
lineages (Fig. 13.1); a similar trend has been noted for other obligate acidophiles
with different temperature optima. Mesophilic isolates account for two of these
groupings. The best known of the acidophilic heterotrophs (Acidiphilium spp.) fall
within the Acetobacteriaceae as a major branch separated from Rhodopila globi-
formis, Acidomonas methanolica, Acetobacter and Thiobacillus acidophilus.As noted
earlier, members of the genus Acidiphilium have recently been transferred to the
closely related genus Acidicella14 based on differences in 16S rDNA sequence simi-
larity, lack of bacteriochlorophyll a, carotenoids, bisphosphatidylglycerol, and the
presence of 2-hydroxy fatty acids. Recent work has demonstrated the unique phy-
logenetic position of a group of mesophilic iron-oxidizing heterotrophs, for which
a novel genus and species label (Ferromicrobium acidophilus; type strain T-23,
DSM 1l138) has been proposed. 28 Data base searches have identified the most closely
related microorganism (by sequence analysis) to be Acidimicrobium ferrooxidans,
a Gram-positive, moderately thermophilic acidophilic high (G + C)% firmicute.
However, this relationship is quite distant in that the overall sequence similarity is
only 67%. Lane et al27 had suggested that Ac. ferrooxidans strainTH3 was evolu-
tionarily near the root of the Gram-positive eubacterial branch, and it may be that
F. acidophilus represents a Gram-negative organism also near the divergence of
the Gram-positive and Gram-negative lineages. It will be interesting to learn what
physiologies are associated with new 16S rDNA sequences that are more closely
related to these microorganisms in the future.
Most moderately thermophilic acidophilic bacteria reside phylogenetically
within the low (G + C)% Bacilli, and Genbank has placed both the heterotrophs
and the iron-oxidizing Sulfobacilli in the Alic),clobacillus group. The thermophilic
acidophiles Thermoplasma and Picrophilus spp. reside in separate branches of the
Thermoplasmales division of the Eurarchaeotal kingdom of the archaea. On the
other hand, sulfur-dependent extremely thermophilic acidophiles such as Sulfolobus
acidocaldarius and Acidianus brierle)'i are all found within the kingdom
Sulfolobus acidocaldarius
Picrophilus
oshimae
Thiobacillus
thiooxidans
Thiobacillus
ferrooxidans
F erromicrobium
acidophilus
Sulfobacillus
Sulfobacillus thermosulfidooxidans
acidophilus
0.10
Fig. 13.1. Phylogenetic tree showing the relative positions of acidophilic heterotrophic
bacteria and archaea, and other acidophilic microorganisms. The bar provides scale
reference for 0.1% divergence between species.
Crenarcheoata. 29 These disparate placements are interesting when contrasted with

the close association of autotrophic and heterotrophic moderately thermophilic
acidophilic bacteria in the Alicyclobacillus group.
Ecology of Acidophilic Heterotrophic Prokaryotes

Distribution in Mineral Leaching Environments
Quantitative data on the numbers and activities of acidophilic heterotrophic
prokaryotes in mineral leaching environments are relatively sparse. In addition,
since the majority of AMD and similar environments are oligotrophic (i.e., con-
centrations of dissolved organic carbon are very small) enumeration of acidophilic
heterotrophs using substrate- and nutrient-rich liquid and solid media probably
underestimates their abundance (parallel situations with oligotrophic neutrophiles
are well documented). Belly and Brock30 reported that acidophilic heterotrophic
bacteria,at 102_103/g, were far less numerous than acidophilic fungi (10 7_ 10 8/ g) in a
coal refuse pile; however, the medium used in their MPN analysis (0.1% yeast ex-
tract/o.1% glucose, pH 2.5) might inhibit rather than encourage growth of many
oligotrophic acidophiles. Counts of acidophilic heterotrophic bacteria from envi-
ronmental samples on solid media have been reported to be greatest using media
developed primarily for isolating and enumerating chemolithotrophic acidophiles
and which are necessarily depleted in organic materials.31 Once isolated, however,
many 'oligotrophic' acidophiles may be cultured successfully in more nutrient-rich
media. Numbers of heterotrophic acidophilic bacteria in streams draining indus-
trial-scale leach dumps in Bulgaria were found to be considerably less than those
of acidophilic chemolithotrophs (Table 13.2). A similar trend was noted by Sand et
al32 in a copperlzinc mine in Romania, though numbers of acidophilic heterotrophs
tended to exceed those of T.ferrooxidans (the dominant iron-oxidizer) in a urani-
um-containing waste heap in Germany, where a positive correlation between the
two groups of acidophilic bacteria was noted. 33 Effluents draining bioleached py-
ritic (20% FeS2) coal columns were found throughout 100 days of processing to be
dominated by chemolithotrophic rather than heterotrophic acidophiles (Table 13.2);
this was independent of whether bioleaching was carried out by indigenous acido-
philic populations or by an inoculated bacterial consortium containing T. fer-
rooxidans, L. ferrooxidans, T. thiooxidans and Acidiphilium spp.34 It was also noted
that acidophilic heterotrophs were significantly greater in leachate from the coal
columns processed with indigenous microflora over the first 60 days of the experi-
ment, though numbers were similar from days 60-100; this was possibly related to
the more rapid build-up of soluble ferric iron (which is relatively toxic to some
acidophilic heterotrophs; see later section) when the pyritic coal was leached by
the more efficient bacterial consortium. Relative numbers of iron-oxidizing
chemolithotrophs and acidophilic heterotrophs in a highly acidic (pH 2.2-2.9)
stream draining the derelict Parys copper mine in Anglesey, North Wales changed
with distance from its source (a mine adit); for the first 400 m iron -oxidizers domi-
nated, but from 500 m to 1 km downstream, acidophilic heterotrophs were more
numerous. 35 Three heterotrophic isolates, which could be differentiated from their
colony morphologies on yeast extract solid medium, were observed in Parys mine
AMD; the relative proportions of these heterotrophs also varied with distance from
the mine adit. Selective pressure in mineral leaching environments may produce
an apparent suppression or elimination of acidophilic heterotrophic bacteria.
Goebel and Stackebrandt36 isolated A. cryptum from batch cultures of zinc/lead
sulfide ore concentrate undergoing bioleaching, though when the bioreactor was
run in continuous culture mode heterotrophs were not recovered. More recently,
biomolecular approaches have been used to elucidate the microbial composition
ofbioleaching systems. Isolation of DNA, followed by gene (16S rRNA) amplifica-
tion, cloning and sequencing was used by Goebel and Stackebrandt37 to examine
the biodiversity in runoff water from a chalcopyrite overburden heap. Of the 120
clones obtained, 107 were affiliated to known mesophilic iron-oxidizers (princi-
pally L. ferrooxidans) and 3 were found to be identical to A. cryptum, though the
authors stressed the potential inaccuracy inherent in extrapolating data from such
analysis for estimating the relative abundance of acidophiles in situ. Pizarro et al38
used, as a molecular marker, the sizes of the spacer regions between the 16 and 23S
rRNA genes, comparing the distribution of those obtained from a copper heap
bioleaching system with those from known acidophilic bacteria. Although no di-
rect evidence for the presence of heterotrophic bacteria was presented, the authors
noted that the amplification products from the mine environment were highly com-
plex, and could have included materials from Acidiphilium-like isolates.
Table 13.2. Estimation of the relative number~ ofchemolithotrophic iron-

oxidizing bacteria and heterotrophic acidophiles in mineral leaching
environments
Site FeZ+-Oxidizing Heterotrophic Heterotrophic

Chemolithotrophs Acidophiles FeZ+-Oxidizers
Vlaikov Vrah Cu leach

dump (Bulgaria)40 10'-107 1-103 N.D.
Tsar Assen leach
dump (Bulgaria)4O 10'-105 <1-104 N.D.
Pyritic coal leach
effluent (U.K.)34 2 X105-4 X108 10'-106 N.D.
Parys Cu mine AMD
(Wales)35 1 x 103-7 X103 2 X10'-4 X104 N.D.
Denison & Stanleigh
U mines (Canada)4' 2.7 x 103-2.5 x 105 4 x 10'-9.7 X103 N.D.
Blackbird Co
mine AMD (USA)b 7XlO' 8X10' N.D.
Ronneberg U waste
heap (Germany)3' <1-2 x 103/g <1-103/g N.D.
Cae Coch pyrite
mine AMD (U.K.Y 9X104 1 X104 2X10 4
Mineral Park
well water (USAY 1X103 <10' 1X105
Sulfidic regolith
(U.S. Bureau of MinesY 1.5 x 10 6/g 2.2X10 6/g 2.7 X10 6/g
• as number/mL, unless stated otherwise
bD.B. Johnson, unpublished data
Rawlings 39 used restriction enzyme analysis of 168 rRNA genes to analyze the
microbial population of a bioleaching tank processing gold-bearing pyrite/arse-
nopyrite ore. While chemolithotrophs were identified, there was no evidence that
acidophilic heterotrophic bacteria were present, which may indicate that the more
extreme environments oflarge-scale biooxidation tanks select against Acidiphilium-
like isolates. However, the tolerance of some strains of the recently characterized
iron-oxidizing heterotroph Ferromicrobium acidophilus to extreme acidity and
soluble metals is similar to that of chemolithotrophic acidophiles (see later sec-
tion). It will be interesting to see whether heterotrophs of this type are found to be
widely-distributed in bioleaching systems. Preliminary data from acidic mineral
leaching sites indicate that Ferromicrobium-like bacteria may be more numeri-
cally abundant than both other heterotrophic acidophiles and chemolithotrophic
iron-oxidizers in some situations (Table 3.2).
Nutrition of Acidophilic Heterotrophic Bacteria

Acidophilic heterotrophic bacteria require organic materials as their source of
energy and cell carbon. A. cryptum has been reported to oxidize sulfur, though
only when provided with a suitable organic substrate,'0 and the presence of
bacteriochlorophyll-a in several Acidiphilium spp. facilitates limited light-enhanced
CO~ incorporation when growing under aerobic conditions. ~ Characterization of
the nutritional diversity of acidophilic heterotrophs has tended to focus on their
utilization (or otherwise) of a relatively narrow range of potential substrates, such
as simple sugars and small molecular weight alcohols and aliphatic acids. Citric
acid and glycerol are metabolized by the majority of characterized acidophilic het-
erotrophs. Growth factors (usually supplied in yeast extract) are required for the
growth of some acidophiles (e.g., A. rubrum and A. angus tum ) but not others (e.g.,
A. cryptum and A. organovorum). Substrate utilization has often been used as a
primary characteristic for distinguishing acidophilic heterotrophic isolates, e.g.,
Acidimonas spp. are methylotrophic (though A. multivorum also metabolizes
methanol), and Acidocella aminolytica can grow on a wide range of amino acids.
While some acidophilic heterotrophs are obligate oligotrophs, others (e.g., A.
organovorum and A. symbioticum) may grow in organic-rich media, producing
cell densities (10 9-10'0ImL) similar to copiotrophic neutrophiles. Utilization oflarge
molecular weight substrates appears to be somewhat limited, though
Acidobacterium capsulatum can grow on starch and cellobiose. Growth of acido-
philic heterotrophs on aromatic substrates and recalcitrant aliphatic materials has
received little or no attention; Wakao et al43 noted that p- and m-hydroxybenzoate
failed to support the growth of A. multivorum. A feature common to many acido-
philic heterotrophic bacteria is inhibition of growth by some small molecular weight
aliphatic acids such as acetic. In the low pH at which these bacteria grow, weak
acids of this type are primarily undissociated and membrane-permeable; on en-
tering the higher internal pH cellular environment they dissociate, producing
disequilibrium between internal and external acid concentrations, and continued
acid influx and potential severe acidification of acidophile cytoplasm. 44 Some spe-
cies (e.g., A. organovorum, inhibited by 12 mM acetic acid) appear less susceptible
than others (e.g., A. cryptum, which is inhibited by 1.2 mM acetic acid), and it is
possible that sublethal concentrations of these acids might actually support the
growth of acidophilic heterotrophs; this has been demonstrated for four
Acidiphilium spp., T. acidophilus and the 'chemolithotroph' T. ferrooxidans which
can all grow on formic acid when provided at 100 JlM.45
Some Alicyclobacillus spp. and other moderately thermophilic isolates also have
a requirement for one or more growth factors. Sugars and small molecular weight
organic acids and alcohols are widely used by these bacteria, and the ability to
metabolize polysaccharides appears to be more widespread than with mesophilic
acidophiles. Of the thermo acidophilic heterotrophic archaea, Thermoplasma
acidophilum has a requirement for yeast extract, and is often grown in complex
growth media (containing, e.g., casamino acids and yeast extract) though growth
may also be stimulated by some defined organic compounds such as glucose. Simi-
larly, Picrophilus spp. have a requirement for yeast extract, and addition of a vari-
ety of sugars increases cell yield; tryptone inhibits growth of this acidophile. ~6
Response to Environmental Variables

All of the bacteria and archaea described in this chapter are obligate acidophiles
rather than acid-tolerant microorganisms (Table 13.1). The most acid-tolerant of
the mesophilic heterotrophs appears to be Ferromicrobium acidophilus, whose pH
optimum and range is similar to that of iron-oxidizing chemolithotrophs. Of the
other acidophilic heterotrophic genera, Acidocella and Acidobacterium are less
acidophilic than most Acidiphilium and Acidomonas. Growth of Acidiphilium spp.
in media poised at relatively high pH values (4-6) may result in acidification of
cultures, presumably due to the production of organic acidsY Relatively few
mesophilic acidophilic heterotrophs grow well below pH 2, and this may limit their
activities in commercial bioleaching operations. The same is true of Alicyclobacillus,
the most acid-tolerant species of which appears to be AI. acidocaldarius, though
some unclassified isolates have been observed to grow well at pH 1.7.45 In contrast,
extreme acidophily appears to be a feature of thermophilic heterotrophic archaea,
and of Picrophilus spp. in particular.
Temperature optimum (and range) is a frequently-used parameter for differ-
entiating acidophilic microorganisms. Berthelot et al41 reported the isolation of
psychrotolerant strains of acidophilic heterotrophic bacteria from an uranium mine
(in situ temperature 13°-18"C) in Canada, though no psychrophilic isolates were
recovered. In low- to mid-temperature environments, acidophilic heterotrophic
bacteria compete for organic substrates with acid/metal-tolerant fungi and yeasts,
while at higher temperatures, obligate heterotrophs compete with other bacteria
(the metabolically-versatile iron- and sulfur-oxidizing moderately thermophilic
bacteria) or, at temperatures over ca. 55"C and above, with Sulfolobus-like archaea.
Over some temperature ranges there would be predicted overlap of heterotrophic
acidophiles, e.g., between Acidiphilium- and Alicyclobacillus-like bacteria at ca.
35°-40"C,and between Alicyclobacillus and ThermoplasmalPicrophilus at 50o-60"C,
though there is a dearth of data in this area.
All acidophilic heterotrophic bacteria (with the exception of F. acidophilus)
were described initially as obligate aerobes. Thermoplasma acidophilum is now
known to be a facultative anaerobe since, in the absence of oxygen, it can use sul-
fur as an electron sink. However, no acidophilic isolate has been shown to be ca-
pable of fermentation, or of anaerobic respiration using nitrate, sulfate, manga-
nese (IV) or carbon dioxide as electron sinks. Ferric iron tends to be highly
abundant in mineral leaching environments, and is present in soluble form at pH
ca. <2.3. It has been shown that Acidiphilium-like isolates and F. acidophilus can
use Fe3+ as an electron acceptor when grown under conditions of limiting or zero
oxygen, though growth tends to be restricted under strictly anoxic conditions. 28,46
Alicyclobacillus-like moderate thermophiles, isolated from Yellowstone National
Park, have also been shown to reduce ferric iron. 21 Whether acidophilic het-
erotrophic archaea are capable of ferric iron reduction is not known.
Another characteristic of acidophilic heterotrophic bacteria and archaea that
has a major influence on their distribution in mineral leaching environments is
their tolerance of heavy metals and metalloids. In general, heterotrophic acidophiles
are less tolerant to dissolved metals than are iron-oxidizing chemolithotrophs
(T. ferrooxidans in particular) though there are variations in their sensitivities
(Table 13.3). Tolerance to very high concentrations of ferrous iron, and to a lesser
extent of ferric iron, seems to be widespread in strains of F. acidophilus and
Acidiphilium spp., though other metals (such as copper) which are important in
Table 13.3. Tolerance of mesophilic heterotrophic acidophiles to some heavy

metals, as minimum inhibitory concentrations (mM)
Organism Metal Species

Fea+ Fe3+ CuH UO aa+ Mo042-
A. cryptum 47 >200 >200 50 2.0 0.2

Acidiphilium SJH47 >200 >200 20 0·5 0.1
T. acidophilus47 >200 100 2.0 0.2
F. acidophilu~ >400 >400 >150 5·0 1.0
Isolate CCH7 6 300 50 20 2.0 0.2
commercial bioleaching tend to be more toxic to heterotrophs other than F. acido-

philus. Increased tolerance may be achieved by subculturing in media containing
gradually increasing concentrations of heavy metals; Said47 found that Acidiphilium
SJH could be adapted to tolerate at least 20 mM copper (II) by this means, a con-
centration that inhibited the growth of nonexposed cultures.
Interactions with Autotrophic Acidophiles

The close association between strains of Acidiphilium spp.and T. ferrooxidans
has already been commented upon; similar associations have also been observed
with L. ferrooxidans. 48.49 F. acidophilus has also been shown to form stable mixed
cultures with both chemolithotrophic iron oxidizers and with the sulfur-oxidizer
T. thiooxidans (D.B. Johnson and P. Bacelar-Nicolau, unpublished data). Cellular
aggregates containing Acidiphilium spp. and L. ferrooxidans may appear as mac-
roscopic flocs in liquid cultures;49 in cultures containing fme-grain pyrite, the min-
eral may become intimately associated with these growths, causing particles to
become aggregated. Competition between acidophilic autotrophs and heterotro-
phs for resources such as nutrients (phosphate, etc.) and oxygen might be antici-
pated when supplies are limited. However, since most acidophilic heterotrophs
(with the exception of F. acidophilus) and iron-oxidizing mesophiles use totally
different substrates and sources of carbon, competition for these would be negli-
gible. At higher temperatures, this would not be the case since, as noted earlier, the
indigenous iron - and sulfur-oxidizing micro flora have more versatile metabolisms
and use organic as well as inorganic carbon. Carbon flow from active, senescent
and dead chemolithotrophs to acidophilic heterotrophic populations has been dem-
onstrated on numerous occasions.5•50 While the association between chemolitho-
trophs and heterotrophs might be commensal (beneficial only to one partner) in
some circumstances, in others it is mutualistic in that removal of accumulating
organic materials may 'detoxify' the environment for the chemolithotrophs. This
observation was the basis for the development of 'overlaid' Acidiphilium-contain-
ing solid media which facilitate the growth of iron- and sulfur-oxidizing mesophilic
(and thermophilic) acidophilesY Mixed cultures of iron-oxidizing and het-
erotrophic acidophiles are often more robust and show greater longevity than cor-
responding pure cultures of chemolithotrophs, especially strains of L. ferrooxidans
(D.B. Johnson, unpublished data). Acidophilic heterotrophs and autotrophs may
interact in other ways, for example the reduction of ferric iron to ferrous by some
Acidiphilium spp. 'regenerates' the substrate used by L. ferrooxidans and T. ferro-

oxidans, and cycling between the two ionic forms has been observed in mixed
cultures. 48 However, since T. ferrooxidans and T. thiooxidans can also use ferric
iron as an electron sink under conditions of oxygen limitation (using reduced sul-
fur as electron donor) competition with heterotrophic acidophiles for Fe3+ would
be anticipated in such circumstances. Interactions between the iron-oxidizing het-
erotroph P. acidophilus (which is also capable of iron reduction at low D0 2 values)
and autotrophic iron-oxidizers is particularly complex. The superior tolerance of
this heterotroph (compared with most others) both to low pH and high concentra-
tions of soluble metals means that such associations are likely to be particularly
stable in the more extreme conditions that can occur in controlled mineral leach-
ing experiments and commercial operations.
Molecular Biology of Acidophilic Heterotrophic Bacteria

Mesophilic Eubacteria
There has been a significant amount of research on the genetics and molecular
biological manipulation of acidophilic bacteria, though the focus of the bulk of
this has been the chemolithotroph T. ferrooxidans. The successes and failures with
T. ferrooxidans have been reviewed recently51 and those studies will be ignored
here, except to say that successful gene transfer into T. ferrooxidans has proven to
be difficult. On the other hand, genetic systems have been readily developed for
mesophilic heterotrophic heterotrophs, including Acidiphilium and Acidocella spp.,
though there has been a dearth of fundamental genetic studies on gene structure
and regulation in these bacteria. Investigations of the molecular biology of acido-
philic heterotrophs are aided by their relative ease of cultivation in both liquid and
solid media, compared with chemolithotrophic acidophiles. This has promoted
extensive studies in some areas, for example, a wide range of restriction endonu-
cleases have been purified and characterized from Acidiphilium Spp.5 2 and
Acidocella.53 Virtually all of these enzymes have proven to be isoschizomers of
known restriction endonucleases.
The recA gene of Ad. facilis was successfully cloned by Inagaki and cowork-
ers,54 and Ward et al55 described a bacteriophage which infects Acidiphilium and
Acidocella. However, these represent the only studies to date which relate to func-
tional genes or naturally occurring genetic systems in these organisms. The devel-
opment of artificial genetic systems, utilizing chimeric recombinant plasmids,
mobilizable, broad host range plasmids, conjugation and electroporation, has been
very successful in Acidiphilium and Acidocella spp. Conjugation of a variety of
broad host range plasmids from different incompatibility groups, as well as a re-
combinant, mobilizable T. ferrooxidans plasmid was reported by Roberto et al.56
Subsequently, Glenn et al 57 extended this work by employing electroporation as
the means for introducing exogenous plasmids, and included a series of recombi-
nant plasmids created by fusing antibiotic resistance genes to endogenous plas-
mids of Acidiphilium. Similar success employing electroporation has been reported
with Ad. facilis. 58
Conjugation has subsequently been used to introduce new biochemical fea-
tures into A. cryptum and Ad. facilis. A phenol-degrading strain of A. cryptum was
created by triparental mating of A. cryptum strain AC6 with E. coli harboring the
plasmid pPGHll.59 As discussed later, the presence of acidophilic heterotrophic
bacteria has been reported to improve rates of bioleaching of sulfide minerals,

though their sensitivities metals and metalloids may limit their activities. To im-
prove the arsenic tolerance of Acl.facilis (strain PWt), a broad range arsenic resis-
tance plasmid was constructed and mobilized through biparental mating into the
heterotroph. 60 The resistance of the organism to arsenite was increased, but the
expression of this foreign gene function appeared to affect the acid tolerance of
Acl. facilis. A novel aspect of this experiment was the broad host range nature of
the constructed arsenic resistance plasmid which was found to be capable of being
mobilized into other compatible bacteria in the same environment, including key
bioleaching organisms such as T. ferrooxidans and L. ferrooxidans.
Moderately Thermophilic Eubacteria

Very little molecular work has been done with these organisms, although their
evolutionary position relative to other low (G + C)% Gram-positive bacteria has
generated some discussion. The amylases of AI. acidocaldarius have been the sub-
ject of several studies on the acid-stability of enzymes. The genes for two such
enzymes have been cloned and sequenced61 and subsequent analysis of the puri-
fied enzyme62. suggested that the replacement of charged amino acids with neutral
polar residues at the surface of the protein contributes to its acid stability.
Thermophilic Acidophilic Archaea

A considerable number of studies have analyzed the genes, proteins and evolu-
tionary positioning of Th. acidophilum, perhaps reflecting the great interest within
the scientific community in probing the relationships between the archaea and the
equally distant eukarya and bacteria to define conceptually the earliest forms of
life on earth. Ribosomal gene organization and expression have been studied and
found to be unlike that in eubacteria, in that each ribosomal gene is expressed
independently.63 Ubiquitin-dependent proteasomes, common to eukaryotes but
absent in eubacteria,have been studied at the protein64 and DNA65 levels. The genes
for enzymes of intermediary metabolism, glucose dehydrogenase and cytoplas-
mic pyrophosphatase have been cloned and sequenced, and a novel restriction
endonuclease, ThaI, has been isolated. 66 The cellular machinery involved in pro-
tein synthesis has been an area of intense study in Th. acidophilum, and DNA-
dependent RNA polymerase,67 tRNAs,68 elongation factor EF_t 69 and transcrip-
tion and transiation70 studied in depth. These studies indicate that while some
features (notably transcription, translation and proteolysis) of the archaea are most
similar to those found in eukarya, others (e.g., pyrophosphatase) are similar to
those in bacteria, and the central role of the archaea in providing clues about the
divergence of the different domains from progenitor species is reinforced.
Role of Acidophilic Heterotrophs in Mineral Dissolution

General Aspects
The contribution of heterotrophic acidophiles to the bioleaching of sulfidic ores
is poorly understood. There have been relatively few investigations of the com-
parative effects on mineral oxidation by mixed cultures containing heterotrophic
acidophiles and pure cultures of chemolithotrophs. Contrasting evidence for the
occurrence of acidophilic heterotrophs in heap and tank leaching operations has
been described earlier, though the presence of these bacteria does not necessarily
infer that they affect rates of mineral dissolution. Acidophilic heterotrophic bacte-
ria may affect mineral bioleaching indirectly by their interactions (both positive
and negative) with iron-oxidizing chemolithotrophs. The reduction of ferric iron
by heterotrophic acidophiles could, in different circumstances, have potentially
opposing effects on mineral dissolution (discussed more fully in the next section).
Adverse effects may also result from the production of biofilms by heterotrophic
bacteria on mineral surfaces which could shield sulfide minerals from oxidative
attack; some heterotrophic acidophiles have been noted to produce copious
amounts of extracellular polymeric materials. 5•16 On the other hand, production of
vitamins, cofactors, chelating agents and surfactants by acidophilic heterotrophs
may enhance sulfide mineral leaching by chemolithotrophic acidophiles. 40
Laboratory experiments with mixed cultures of acidophilic heterotrophs and
chemolithotrophs on mineral leaching have yielded contrasting results. No en-
hancement of the depyritization of coal was found when mixed cultures contain-
ing Alicyclobacillus-like heterotrophs were compared with pure cultures of iron-
oxidizing moderate thermophiles. 34 Pyrite leaching by T. ferrooxidans was reported
not to be influenced by T. acidophilus71 and a similar result was found by Johnson
et al48 with Acidiphilium SJH. However, Wichlacz and Thompson7a found increased
solubilization of cobalt sulfide ores by T. ferrooxidans in the presence of a number
of acidophilic heterotrophic bacteria when leached for 7 and 28 days; this effect
was noted only when ferrous iron or glucose was added to cultures, and net in-
hibition of cobalt solubilization was found in the absence of added ferrous iron.
Leaching of pyrite by a mixed culture of an L. ferrooxidans isolate and Acidiphilium
SJH was found to be more rapid than that brought about either by pure cultures of
the iron-oxidizer or by pure and mixed (with Acidiphilium SJH) cultures of the
type strain of T. ferrooxidans. 48 Addition of 10 mM glucose completely sup-
pressed L. ferrooxidans grown in pure culture, but this inhibition was relieved when
Acidiphilium SJH was present; in contrast, 10 mM did not suppress T. ferrooxidans
and no net benefit in terms of mineral solubilization was observed in mixed cul-
tures with Acidiphilium SJH. Hallmann et al49 also reported that an Acidiphilium
spp. enhanced pyrite solubilization by L. ferrooxidans but not by T. ferrooxidans.
Two hypotheses were proposed to account for this: the removal of soluble organic
materials by the Acidiphilium spp. which was particularly beneficial to the more
sensitive L. ferrooxidans; and stimulation of exopolymer production by L. ferro-
oxidans in the presence of the Acidiphilium spp. which facilitated contact with and
enhanced bioleaching of pyrite crystals.
Iron-Reducing Heterotrophs
Direct and indirect mechanisms have been described for the bacterial oxida-
tion of sulfide minerals (reviewed by Sand et al73 ). In the indirect mechanism, fer-
ric iron generated by iron-oxidizing acidophiles oxidizes sulfide minerals abioti-
cally in a reaction which does not require molecular oxygen, though the subsequent
biological regeneration of the chemical oxidant (Fe3+) occurs only under aerobic
conditions. It follows that biological reduction of ferric iron to ferrous would be
deleterious to mineral oxidation by this mechanism. Some Acidiphilium spp. uti-
lize ferric iron as an electron sink even under aerobic conditions (e.g., in shake
flask cultures and aerobically-incubated agar plates;46 ); in controlled fermenter
cultures a dissolved oxygen concentration of <15% of maximum (under prevailing
culture conditions) was required for iron reduction by Acidiphilium SJH. 8 The
mixotrophic acidophile T. acidophilus can also reduce ferric iron when growing
heterotrophicallyJ4 The other requirement for heterotrophic ferric iron reduction
is provision of a suitable (organic) electron donor. Data illustrating negative ef-
fects on pyrite oxidation by mixed cultures of both T. ferrooxidans and L.
ferrooxidans with the acidophilic iron-reducer Acidiphilium SJH in cultures
amended with glucose have been published. 48 These reports contrast somewhat
with those of Wichlacz and Thompson72 described earlier. Two possible explana-
tions for this apparent anomaly are: (I) the heterotrophs used by Wichlacz and
Thompson could either not reduce ferric iron or else did so relatively slowly (aci-
dophilic heterotrophs vary widely in their propensities for iron reduction};46 (2)
since the cobalt ore leaching experiments were performed at relatively high pH
(cultures were poised initially at pH 3.0) there would have been little ferric iron
present in soluble form (and the indirect leaching mechanism of minor impor-
tance) and other factors such as production of chelating organic acids by the acido-
philic heterotrophs might have been more important to the net solubilization of
cobalt.
The ability of heterotrophic acidophiles to reduce not only soluble but also
solid-phase (amorphous and crystalline) ferric iron compounds could, however,
be advantageous to ore bioleaching in certain circumstances. Ferric iron formed
during bioprocessing of iron-containing sulfide minerals may precipitate in a va-
riety of mineralogical forms, such as jarosites, forming passivation layers on min-
eral surfaces and slowing down ore dissolution. Bridgel4showed that Acidiphilium
SJH was capable of the reductive solubilization of a wide range (all of those tested)
of ferric iron-containing minerals. Rates of dissolution varied with the crystallin-
ity of the mineral concerned; amorphous Fe( OHh was the most readily solubilized
and hematite the least of those tested. Though it is yet to be demonstrated in situ,
it is possible that, by controlling aeration in bioleaching operations involving aci-
dophilic consortia, passivation layers could be removed biologically, allowing more
rapid sulfide mineral dissolution to proceed; extraneous organic material might
also be required in order to promote rapid and effective ferric mineral removal by
acidophilic heterotrophs.
Iron-Oxidizing Heterotrophs
The indirect theory of sulfide mineral oxidation suggests that any biological
system that is capable of regenerating ferric iron in an acidic milieu should pro-
mote mineral dissolution. Heterotrophic iron-oxidizing bacteria should therefore
be capable of attacking pyrite and other sulfides, if provided with a suitable or-
ganic substrate. However, the filamentous iron-oxidizing bacterium CCH7 failed
to show any evidence of pyrite or chalcopyrite oxidation in cultures containing
ferrous iron and yeast extract6 though, in contrast, unicellular R acidophilus iso-
lates have been found to be oxidize pyrite when grown in media containing yeast
extract (D.B. Johnson and P. Bacelar-Nicolau, unpublished data). Ferrous iron is
also required, indicating that pyrite oxidation by iron-oxidizing heterotrophic
acidophiles is mediated only via the indirect mechanism (Fig. 13.2}. Rates of pyrite
oxidation by R acidophilus varied between strains, with one isolate (strain T-24)
being at least equal in this respect (in yeast extract-containing cultures) to the
type strain of T. ferrooxidans (Fig. 13.3}. One intriguing observation was that strains
of R acidophilus which were not able to oxidize pyrite in pure culture in the
absence of added organic carbon were able to do so in mixed cultures with either
T. thiooxidans or T. acidophilus. Since neither Thiobacillus is able to oxidize pyrite,
Extraneous sources Chemolithotrophs/(phototrophs)
~ /F,0,.
Other heterotrophic acidophiles ~ \ORG-A-N-I-C-C--',
~sulfide minerals
(MS,)
Heterotrophic iron-oxidizers Heterotrophic iron-reducers M2., S042-
~ sulfide
minerals
(MS,)
Fig. 13.2. Schematic representation of carbon flow and oxido-reduction of iron in acidic
environments.
3000
-
""'
....J
O"l
E
2000
CD
u..
1000
o L -_ _ _ _ _ _
~ ~ ____ ~ ___ ~ ___ ~
o 10 20 30 40 50
Time (days)
Fig. 13.3. Leaching of pyrite by pure and mixed cultures of R acidophilus and compari-
son with pure cultures of T. ferrooxidans and T. thiooxidans. Key: 'Y, R acidophilus
(strain T-21); ., R acidophilus (T-21)/T. thiooxidans mixed culture; &, R acidophilus
(strain T-24; culture amended with 0.02% (w/v) yeast extract); e, T.ferrooxidans;., T.
thiooxidans.
this observation is a novel example of bacterial synergy. Organic carbon to sustain

heterotrophic growth was considered to originate from the CO2 -fixing Thiobacillus
spp., which oxidize reduced sulfur compounds produced via ferric iron attack on
pyrite,73 As shown in Figure 13.3, pyrite leaching by the R acidophilus (strain T-21)1
T. thiooxidans consortium was similar to that by T. ferrooxidans though, in con-
trast to that by pure cultures of R acidophilus strain T-24, this was achieved in the
absence of added yeast extract.
In both commercial and environmental situations, R acidophilus is considered
to contribute to mineral leaching both by its generation of ferric iron, and by'detoxi-
fying' acidic milieu (from the perspective of chemolithotrophic acidophiles) by its
catabolism of dissolved organic carbon (Fig. 13.2). Although R acidophilus is also
capable of ferric iron reduction, this is not observed unless oxygen is very limiting
(D0 2 at 5-10% of maximum ambient) and its specific rate of iron reduction is far
lower than that of ferrous iron oxidation. 28
Conclusions
Acidophilic heterotrophic bacteria and archaea are an interesting and impor-
tant group of microorganisms in both pure and applied microbiology. Recent work
suggests that, far from comprising a limited number of genera and species, het-
erotrophic acidophiles which colonize mineral leaching environments include a
variety of phylogenetically-distinct microorganisms. These bacteria and archaea
may promote the oxidative dissolution of sulfide minerals either by their positive
interactions with chemolithotrophic metal-mobilizing acidophiles (especially L.
ferrooxidans) or, in the case of R acidophilus-like bacteria, by generating the chemi-
cal oxidant ferric iron. The isolation and characterization of organisms with dis-
tinct and sometimes novel physiological traits indicates that there may be biotech-
nologies other than bioleaching of sulfide ores in which heterotrophic acidophiles
could be utilized. In addition, there are some areas of acidophilic heterotrophic
microbiology (such as studies of anaerobic acidophiles) that have received little or
no attention; research in these areas will further expand our understanding of,
and possibly the exploitation of, acidophilic heterotrophic microorganisms.
References
1. Lopez-Archilla AI, Marin I, Amils R. Microbial ecology of an acidic river: bio-
technological applications. In: Vargas T, Jerez CA, Wiertz JV et aI, eds.
Biohydrometallurgical Processing II. Santiago: University of Chile, 1995:63-74.
2.. Johnson DB, Rang L. Effects of acidophilic protozoa on populations of metal-
mobilising bacteria during the leaching of pyritic coal. J Gen Microbiol 1993;
139:1417-142.3.
3. Dugan PR, Macmillan CB, Pfister RM. Aerobic heterotrophic bacteria indigenous
to pH 2..8 acid mine water: predominant slime-producing bacteria in acid stream-
ers. J Bacteriol 1970; 101:2.82.-2.88.
4. Wakao N, Tachibana H, Tanaka Y et al. Morphological and physiological charac-
teristics of streamers in acid mine drainage water from a pyritic mine. J Appl
Microbiol 1985; 31:17-2.8.
5. Johnson DB. Diversity of microbial life in highly acidic, mesophilic environments.
In: Berthelin J, ed. Diversity of Environmental Biogeochemistry. Amsterdam:
Elsevier, 1991:2.2.5-2.38.
6. Johnson DB, Ghauri MA, Said MF. Isolation and characterization of an acido-
philic hetero trophic bacterium capable of oxidizing ferrous iron. Appl Environ
Microbiol 1992.; 58:142.3-142.8.
7. Johnson DB, Bacelar-Nicolau P, Bruhn DF et al. Iron-oxidizing heterotrophic

acidophiles: ubiquitous novel bacteria in leaching environments. In: Vargas T,
Jerez CA, Wiertz JV et al, eds. Biohydrometallurgical Processing I. Santiago: Uni-
versity of Chile, 1995:47-46.
8. Johnson DB, McGinness S, Ghauri MA. Biogeochemical cycling of iron and sulfur
in leaching environments. FEMS Microbiol Rev 1993; 11:63-70.
9. Wichlacz PL, Unz RF. Acidophilic, heterotrophic bacteria of acidic mine waters.
Appl Environ Microbiol1981; 41:12.54-12.61.
10. Harrison AP Jr. Acidiphilium cryptum gen. nov., sp. nov., heterotrophic bacte-
rium from acidic mineral environments. Int J Syst Bacteriol 1981; 31:32.7-332..
11. Johnson DB, Kelso WI. Detection of heterotrophic contaminants in cultures of
Thiobacillus ferrooxidans and their elimination by subculturing in media con-
taining copper sulfate. J Gen Microbiol 1983; 12.3:2.969-2.972..
12.. Tabita R, Lundgren DG. Utilization of glucose and the effect of organic com-
pounds on the chemolithotroph Thibacillus ferrooxidans. J Bacteriol 1971;
108:32.8-333·
13. Guay R, Silver M. 1974. Thiobacillus acidophilus sp. nov.: isolation and some physi-
ological characteristics. Can J Microbiol1974; 2.1:2.81-2.88.
14. Kishimoto N, Kosako Y, Wakao N et al. Transfer of Acidiphilium facilis and
Acidiphilium aminolytica to the genus Acidocella gen. nov., and emendation of
the genus Acidiphilium. Syst Appl Microbiol 1995; 18:85-91.
15. Urakami T, Tamaoko J, Suzuki K et al. Acidomonas gen. nov., incorporating
Acetobacter methanolicus as Acidomonas methanolica comb. nov. Int J Syst
Bacteriol 1989; 39:50-55.
16. Kishimoto N, Kosako Y, Tano T. Acidobacterium capsulatum gen. nov., sp. nov.:
an acidophilic chemoorganotrophic bacterium containing menaquinone from
acidic mineral environment. Curr Microbiol1991; 2.2.:1-7.
17. Pronk JT, Johnson DB. Oxidation and reduction of iron by acidophilic bacteria.
Geomicrobiol J 1992.; 10:153-171.
In: Ehrlich HL, Brierley CL, eds. Microbial Mineral Recovery. New York: McGraw-
Hill, 1990:3-2.7.
19. Ghauri MA, Johnson DB. Physiological diversity amongst some moderately ther-
mophilic iron-oxidizing bacteria. FEMS Microbiol Ecol1991; 85:32.7-334.
2.0. Wisotzkey JD, Jurtshuk P Jr, Fox GE et al. Comparative sequence analysis on the
16S RNA (rDNA) of Bacillus acidocaldarius, Bacillus acidoterrestris and Bacillus
cycloheptanicus and proposal for creation of a new genus, Alicyclobacillus gen.
nov. Appl J Syst Bacteriol 1992.; 42.:2.63-2.69.
2.1. Johnson DB, Body DA, Bridge TAM et al. Biodiversity of acidophlic moderate
thermophiles isolated from two sites in Yellowstone National Park, and their roles
in the dissimilatory oxido-reduction of iron. In: Mancinelli R, Reysenbach A-L,
eds. Biodiversity, Ecology and Evolution of Thermophiles in Yellowstone National
Park. Washington D.C: American Society for Microbiology, 1996: (In press).
2.2.. Darland G, Brock TD, Samsonoff W et al. A thermophilic, acidophilic mycoplasma
isolated from a coal refuse pile. Science 1970; 170:1416-1418.
2.3. Bohlool BB, BrockTD. Immunofluorescence approach to the study of the ecology
of Thermoplasma acidophilum in coal refuse material. Appl Microbiol 1974;
2.8:11-16.
2.4. Yasuda M, Oyaizu H, Yamagishi A et al. Morphological variation of new
Thermoplasma acidophilum isolates from Japanese hot springs. Appl Environ
Microbiol 1995; 61:3482.-3485.
25. Segerer A, Langworthy TA, Stetter KO. Thermoplasma acidophilum and

Thermoplasma volcanium sp. nov. from solfatara fields. System Appl Microbiol
1988; 10:161-171.
26. Schleper C, Puehler G, Holz I et al. Picrophilus gen. nov., fam. nov.: a novel aero-
bic, heterotrophic, thermo acidophilic genus and family comprising archaea ca-
pable of growth around pH o. J Bacteriol 1995; 177:]050-7059.
27. Lane DJ, Harrison AP Jr, Stahl D et al. Evolutionary relationships among sulfur-
28. Johnson DB, Bacelar-Nicolau P, Bruhn D et al. Ferromicrobium acidophilus, gen.
nov., sp. nov.; obligately heterotrophic and acidophilic iron-oxidizing bacteria
isolated from acid mine drainage and related environments. Microbiol1997; (sub-
mitted).
29. Olsen GJ, Woese CR Ribosomal RNA: a key to phylogeny. FASEB J 1993; 7:113-123.
30. Belly RT, Brock TD. Ecology of iron-oxidizing bacteria inpyritic materials associ-
ated with coal. J Bacteriol1974; 117:726-732.
31. Johnson DB. Selective solid media for isolating and enumerating acidophilic bac-
teria. J Microbiol Meth 1995; 23:205-218.
leaching. App Environ Microbiol 1992; 58:85-92.
33. Schippers A, Hallmann R, Wentzien S et al. Microbial diversity in uranium mine
waste heaps. App Environ Microbiol 1995; 61:2930-2935.
34. Johnson DB. Biological desulfurization of coal using mixed populations of
mesophilic and moderately thermophilic acidophilic bacteria. In: Dugan PR,
Quigley DR, Attia YA, eds. Processing and Utilization of High-Sulfur Coals.
Amsterdam:Elsevier, 1991:567-580.
35. Walton KC, Johnson DB. Microbiological and chemical characteristics of an acidic
stream draining a disused copper mine. Environ Pollut 1992; 76:169-175.
37. Goebel BM, Stackebrandt E. Molecular analysis of the microbial biodiversity in a
natural acidic environment. In: Vargas T, Jerez CA, Wiertz JV et aI, eds.
Biohydrometallurgical Processing II. Santiago: University of Chile, 1995:43-52.
38. Pizarro J, Jedlicki E, Orellana 0 et al. Bacterial populations in samples of
bioleached copper ore as revealed by analysis of DNA before and after cultiva-
tion. Appl Environ Microbiol1996; 62:1323-1328.
39. Rawlings DE. Restriction enzyme analysis of 16S rDNA genes for the rapid iden-
ferrooxidans strains in leaching environments. In: Vargas T, Jerez CA, Wiertz JV
et aI, eds. Biohydrometallurgical Processing II. Santiago: University of Chile,
1995:9-18.
40. Tuovinen OH, Kelley BC, Groudev SN. 1991. Mixed cultures in biological leaching
processes and mineral biotechnology. In: Zeikus G, Johnson EA, eds. Mixed cul-
tures in Biotechnology. New York: McGraw-Hill, 1991:373-427.
41. Berthelot D, Leduc LG, Ferroni GD. Temperature studies of iron-oxidizing au-
totrophs and acidophilic heterotrophs isolated from uranium mines. Can J
Microbiol 1993; 39:384-388.
42. Wakao N, Nagasawa N, Matsuura T et al. Acidiphilium multivorum sp. nov., an
acidophilic chemoorganotrophic bacterium from pyritic acid mine drainage. J Gen
Appl Microbiol1994; 40:143-159.
43. Norris PR, Ingledew WJ. 1992. Acidophilic bacteria: adaptations and applications.
In: Herbert RA, Sharp RJ, eds. Molecular Biology and Biotechnology of Thermo-
philes. Glasgow: Blackie, 1992:115-142.
44. Pronk JT, Meijer WM, Hazeu W et al. Growth of Thiobacillus ferrooxidans on
formic acid. Appl Environ Microbiol 1991; 57:2057-2062.
45. Norris PR, Johnson DB. Acidophilic Microorganisms. In: Horikoshi K, Grant WD,
eds. Extremophiles:Microbial Life in Extreme Environments. New York: John
Wiley, 1996: (In press).
46. Johnson DB, McGinness S. Ferric iron reduction by acidophilic heterotrophic
bacteria. Appl Environ Microbiol 1991; 57:207-211.
47. Said MF. The tolerance of acidophilic bacteria to high concentrations of some
metals. Ph.D. Thesis, University of Wales, 1990.
48. Johnson DB, Said MF, Ghauri MF et al. Isolation of novel acidophiles and their
potential use in bioleaching operations. In: Salley J, McCready RGL, Wichlacz PL,
eds. Biohydrometallurgy 1989. Ottawa: Canada. 1990:403-414.
49. Hallmann R, Friedrich A, Koops H-P et al. Physiological characteristics of
Thiobacillus ferrooxidans and Leptospirillum ferrooxidans and physicochemical
factors influence microbial metal leaching. Geomicrobiol J 1992; 10:193-206.
50. Harrison AP Jr. The acidiphilic Thiobacilli and other acidophilic bacteria that
share their habitat. Ann Rev Microbiol 1984; 38:265-292.
51. Rawlings DE, Kusano T. Molecular genetics of Thiobacillus ferrooxidans. Microbiol
Rev 1994; 58:39-55.
52. Inagaki K, Hikita T, Yanagidani S et al. Restriction endonuclease Aon3HI from
Acidiphilium organovorum 13H, a new isoschizomer of BspMII: purification and
characterization. Biosci Biotechnol Biochem 1993; 57:1716-1721.
53. Dou K, Inagaki K, Oshima A et al. Restriction endonuclease AfaI from
Acidiphilium facilis, a new isoschizomer of RsaI: purification and properties.
Biochim Biophys Acta 1989; 1009:83-86.
54. Inagaki K, Tomono J, Kishimoto N et al. Cloning and sequencing of the recA
gene of Acidiphilium facilis. Nucl Acids Res 1993; 21:4149.
55. Ward TE, Bruhn DF, Shean ML et al. Characterization of a new bacteriophage
which infects baceria of the genus Acidiphilium. J Gen Viro11993; 74:2419-2425.
56. Roberto FF, Glenn AW, Bulmer D et al. Genetic transfer in acidophilic bacteria
which are potentially applicable in coal beneficiation. Fuel 1991; 70:595-598.
57. Glenn AW, Roberto FF, Ward TE. Transformation of Acidiphilium by electro-
poration and conjugation. Can J Bacteriol 1992; 38:387-393.
58. Inagaki K, Tomono J, Kishimoto N et al. Transformation of the acidophilic het-
erotroph Acidiphilium facilis by electroporation. Biosci Biotechnol Biochem 1993;
57:1770-1771.
59. Quentmeier A, Friedrich CG. Transfer and expression of degradative and antibi-
otic resistance plasmids in acidophilic bacteria. Appl Environ Microbiol 1994;
60:973-978.
60. Bruhn DF, Roberto FF. Maintenance and expression of enteric arsenic resistance
genes in Acidiphilium. In: Torma AE, Wey JE, Lakshmanan VI, eds. Biohydro-
metallurgical Technologies I. Warrendale PA: The Minerals, Metals and Materials
Society, 1993: 745-754.
61. Koivula TT, Hemila H, Pakkanen R et al. Cloning and sequencing of a gene en-
coding acidophilic amylase from Bacillus acidocaldarius. J Gen Microbiol 1993;
139:2399-2407.
62. Schwermann B, pfau K, Liliensiek B et al. Purification, properties and structural
aspects of a thermoacidophilic «-amylase from Alicyclobacillus acidocaldarius
ATCC 27009. Eur J Biochem 1994; 226:981-991.
63. Ree HK, Zimmermann RA. Organization and expression of the 16S, 23S and 5S
ribosomal RNA genes from the archaebacterium Thermoplasma acidophilum. Nucl
Acids Res 1990; 18:4471-4478.
64. Rivett AJ, Mason GG, Thomson S et al. Catalytic components of proteasomes and
the regulation of proteinase activity. Mol BioI Rep 1995; 21:35-41.
65. Zwicki P, Lottspeich F, Dahlmann B et al. Cloning and sequencing of the gene
encoding the large (a-) subunit of the proteasome from Thermplasma acidophilum.
FEBS Lett 1991; 278:217-221.
66. McConnell DJ, Searcy DG, Sutcliffe JG. A restriction enzyme ThaI from the ther-
mophilic mycoplasma Thermoplasma acidophilum. Nucl Acids Res 1978;
5:1729-1739.
67. Prangishvilli D, Zillig W, Gierl A et al. DNA-dependent RNA polymerase of the
thermo acidophilic archaebacteria. Eur J Biochem 1982; 122:471-477.
68. Kilpatrick MW, Walker RT. A nucleotide sequence of the tRNA Met from the
archaebacterium Thermoplasma acidophilum. Nucl Acids Res 1981; 9:4387-4390.
69. Tesch A, Klink F. Cloning and sequencing of thegene coding for the elongation
factor 1a from the archaebacterium Thermoplasma acidophilum. FEMS Microbiol
Lett 1990; 59:293-297.
70. Londei P, Altamura S, Cammarano P et al. Differential features of ribosomes and
poly (U)-programmed cell-free systems derived from sulfur-dependent archaebac-
terial systems. Eur J Biochem 1986; 157:455-462.
71. Norris PR, Kelly DP. The use of mixed microbial cultures in metal recovery. In:
Bull AT, Slater JH, eds. Microbial Interactions and Communities I. London: Aca-
demic Press, 1982:443-474.
72. Wichlacz PL, Thompson, DL. The effect of acidophilic heterotrophic bacteria on
the leaching of cobalt by Thiobacillus ferrooxidans. In; Norris PR, Kelly DP, eds.
Biohydrometallurgy: Proceedings of the International Symposium, Warwick 1987.
Kew (U.K.): Science and Technology Letters, 1982:77-88.
73- Sand W, Gehrke T, Hallmann R et al. Sulfur chemistry, biofilm, and the (in)direct
attack mechanism-a critical evaluation of bacterial leaching. Appl Microbial
Biotechnol 1995; 43:961-966.
74. Bridge TAM. Iron reduction by acidophilic bacteria. Ph.D. Thesis, University of
Wales, 1995.
CHAPTER 14
Molecular Methods for the

Identification and Enumeration
of Bioleaching Microorganisms
Carlos A. Jerez
Introduction
M icrobial ecology requires that the microorganisms of a given ecosystem be iden-

tified in situ, and that their spatial and temporal distribution be known. The
classical approach to enumerate the microorganisms in environmental samples
has been the plating technique combined with a simultaneous or subsequent dif-
ferentiation of the isolates based on physiological and biochemical properties. How-
ever, all techniques relying on cultivation are time-consuming. In addition, the
failure of many bacteria to form colonies is a widely acknowledged problem when
using plate counting procedures. Often the number of colony forming units is only
a minor fraction of the cell counts determined by direct microscopic procedures.
Therefore, only a small percentage (less than 20%) of the microorganisms within
autochtonous communities is known.1-5 Additional problems are found with the
determination of the population dynamics since most microbial communities in-
clude not only planktonic or free-living microorganisms, but also biofllms, sedi-
ments and particulates.1.4•5
A number of methodological approaches are now either under development or
available to detect and quantitate microorganisms and/or their activities in eco-
logical and environmental studies. These include adaptations of traditional mi-
croscopy and also scanning confocal laser microscopy. A spectrum of different
molecular approaches has also been developed and these are used in microbial
ecology and environmental microbiology.1-5 Other current approaches include
immunological methods, flow cytometry, image analysis technology, cellular bio-
chemical signatures, microcalorimetry and microelectrodes.1-5
The purpose of this chapter is to review some of the current molecular meth-
ods that have been employed for the detection and enumeration of bioleaching
microorganisms, with special emphasis in some of the immunological methods
developed in our laboratory for the quantitation of liquid suspended and ore-ad-
hered microorganisms.
Biomining: Theory. Microbes and Industrial Processes,

edited by Douglas E. Rawlings_ © Springer - Verlag and Landes Bioscience 1997.
Bioleaching Microorganisms
A complex population of chemolithotrophic microorganisms is involved in the
bioleaching of ores. Some of these, such as Thiobacillus ferrooxidans, Leptospirillum
ferrooxidans and Thiobacillus thiooxidans amongst others have been cultivated
and characterized (see chapters 11 and 12).6-8 Also, several iron-oxidizing het-
erotrophic acidophiles are present in leaching environments9 (see chapter 13). In
addition, it is possible that many other unknown nonculturable species exist in
the bioleaching environment. lo
Specific methods are required to study the role of each type of bioleaching
microorganism and their population dynamics during ore biodegradation. The
identification of specific microorganisms and their activity in bioleaching opera-
tions is important for their industrial control. For example, several stress condi-
tions for microorganisms are generated during operation of heap bioleaching when
effluents are recirculated. These liquids contain increasing salt concentrations (for
example, sulfate concentrations as high as 150 giL) and other inhibitors which will
affect the existing microbial population. This most likely causes a slowdown of the
bioleaching operation. A few of the many questions that can be asked regarding
the microbial populations and their changes in a bioleaching operation are: which
microorganisms remain under stress conditions; are they capable of oxidizing
minerals; and which bacteria are more sensitive to these operational conditions?
Some of the Classical Methods for the Determination

of Biomining Microorganisms
Some of the methods currently employed to monitor bioleaching microorgan-
isms involve determinations of most probable numbers or growth on solid me-
dia/· 8 The most probable number analysis, in addition to being a quantitative mea-
sure of viable cells, provides a sample which can later be used to identify subspecies.
This method has been widely used to enumerate T. ferrooxidans. According to
Southam and Beveridge,l1 the enumeration of this microorganism with this method
eliminates the problem of low plating efficiency and makes it the best available
technique to estimate the population of T. ferrooxidans in tailing environments
even though it provides underestimates of the actual population. However, the
most probable number method is not specific, and any iron-oxidizing microor-
ganism present in the minerals could also grow under these conditions.
Growth on solid media by plating is essential not only to isolate pure strains of
bioleaching microorganisms but also to genetically manipulate the cells. Several
solid media have been developed9•12-l4 but most of them are not selective, and in
general, the obligately chemoautolithotrophic bacteria that are thought to playa
major role in bioleaching are difficult to culture in solid media. This is more com-
plicated when thermophilic or moderate thermophilic microorganisms are to be
analyzed. Although a plating method has been described for Sulfolobus
acidocaldarius,15 quick and sensitive methods for the identification of thermophiles
are not general available. Also, as already mentioned, many of the microorganisms
present in environmental samples may not be culturable on laboratory media.
Adherence of Microorganisms and Biofilm Formation

Biofilms are special micro environments where microorganisms are firmly ad-
hered to surfaces and to each other by exopolymeric substances excreted by the
cells. In some cases, the exopolymeric compounds preclude the use of direct enu-
meration techniques such as epifluorescence microscopy, since the fluorescent
Methods for Identification and Enumeration of Bioleaching Microorganisms 283
IO,um
Fig. 14.1. Biofllm of T. ferrooxidans on the surface of a sulfur prill. Sulfur prills were
colonized by T. ferrooxidans for two weeks. After this time, the samples were processed
for scanning electron microscopy. The bar indicates 10 jlm.
antibodies are not able to react with the bacteria inside the biofilm. Most bioleaching
microorganisms exist in both planktonic and solid-adhered forms. It has been
described that depending on the conditions, at least 80% of the bioleaching micro-
organisms are adhered to ores, making it very difficult to count them, especially if
they grow within pores of the particles. 16 -2o T. ferrooxidans has been described to
form a monolayer or "biofilm of adsorbed cells:"1 An example obtained in our
laboratory (Levican and Jerez, unpublished results) of this kind of bacterial at-
tachment to a solid substrate is shown in Figure 14.1. The scanning electron mi-
croscopy (SEM) image shows an area from the surface of an elementary sulfur prill
colonized with T. ferrooxidans cells. Extensive microbial coverage of the valleys
and crevices present on the surface of the sulfur particle can be seen. The bacteria
are tightly packed in contact with each other and form what appears to be a fUm of
at least one layer of cells. In natural samples, biofUm formation will probably in-
clude several bioleaching microorganisms and some organic excretion exopolymers
in which they are embedded. 22 To separate the biofUm into single cells, mechanical
forces and or chemical treatments are needed (see below).
Enumeration ofMicroorganisms Adhered to Solid Surfaces

Most of the more classical procedures are useful for the direct enumeration of
planktonic microorganisms in effluents, but are seldom adaptable to the study of
bacterial adhesion to the ore or mineral surfaces. Since bacterial adhesion in the
solubilization of metals is a well-known phenomenon l6 -20 and methods for the
determination of the attached microbial populations are needed. Yeh et all7 have
employed epifluorescence microscopy using acridine orange for the direct counts
of T. ferrooxidans attached to solid particles. Given the proper sample prepara-
tion, direct cell counts can be performed both for free cells at densities below the
detection limits of most other techniques and for cells adhered to particulates.17
However, as this technique is based on a nucleic acid binding dye which fluoresces
under blue light, it will stain any bioleaching microorganism present in the sample
and is therefore nonspecific.
Detachment ofSolid-Adhered Microorganisms

The application of many of these techniques is hampered by the presence of
clays, metals and extreme pHs. For example, when employing antibodies, their
nonspecific adsorption to particles can occur. Consequently, some measurements
can only be performed after the separation of the cells from the solids. Such sepa-
ration also reduces problems with contaminating eukaryotic DNAs and particles
that can interfere with reactions such as PCR amplification and rRNA analysis (see
below).1,23- 25
We have studied in detail conditions for the release of ore-attached bioleaching
bacteria prior to their determination. To release adhered bacteria, ore samples were
exposed to different physical and chemical treatments, including stirring in the
presence of detergents. The bacteria released by this treatment (Fig. 14.2) can be
used for direct microscopic counts, for immunological analysis or to extract their
DNA for further analysis. The use of detergents such as Tween-20 has also been
reported for the detachment of T. ferrooxidans and 1. ferrooxidans from sulfur or
arsenopyrite for direct microscopic counting.18,19,26
Immunological Methods
The use of either polyclonal or monoclonal antibodies offers a potentially sen-
sitive and specific means of identifying environmentally important microorgan-
isms} The general principle involved in this immunological procedure is illus-
trated in Figure 14.3. We and others have previously described, that the main
antigens on the surface of a biomining microorganism such as T. ferrooxidans cor-
respond to lipopolysaccharides and/or outer membrane proteins. 27 The antibod-
ies can be reacted directly with whole cells or after separating the antigenic mol-
ecules by polyacrylamide gel electrophoresis followed by transfer of the separated
antigens into a membrane for Western blot analysis.
Immunofluorescence
Antibodies have been employed for the specific detection and enumeration of
biomining microorganisms. Apel et al,28 Baker and Mil1s 29 and Gates and Pham30
have reported the use of fluorescent antibodies for the determination of pyrite-
oxidizing microorganisms in acid mine drainage waters. We described the use of
specific antisera against whole T. ferrooxidans cells. 31 These antisera did not react
with L. ferrooxidans, one of the most common iron-oxidizers accompanying
Fig. 14.2. Detachment of ore-adhered

microorganisms. An ore-colonized
with bacteria (black. dots) is loaded
onto a column, and the ore washed
with a salts solution (C) or with the
same solution containing a given de-
tergent (A). The number of bacterial
cells liberated in each case was deter-
mined by direct microscopic count-
ing (bottom curves).
~ .
•• • ~
W
E 9
c A
o~
0''''
~ 0
o
o )(
6
~
u ~
"0
E OJ
..... If)
0
0
OJ 3
OJ OJ
.0
E
~
c
:l
Z
ou.~~--~------~--
o 5 10
Minutes af treatment
A ntlgen- Antibody Second Antibody Ant Igen-A nllbo dy

S econd-Anll body-E nzyme
Fig. 14.3. The general procedure of detection of microorganisms by using antibody-

antigen reactions. A primary antibody ( A) generated against whole cells of a given
bacterial species is reacted specifically with the antigens of the microorganism to be
identified. This reaction is recognized by a secondary antibody ( A)
specific to the
primary antibody. Since the secondary antibody can be conjugated with an enzyme,
the enzymatic reaction with a colored substrate reveals the interaction. This interac-
tion will be proportional to the amount of original antigens in the bacteria and will
depend on the number of bacteria present in the first reaction.
T. ferrooxidans in mining environments. Furthermore, the antisera were useful

not only to specifically detect T. ferrooxidans but also to distinguish two different
T. ferrooxidans strains based on the analysis of the immunoprecipitated radioac-
tively labeled components by SDS-PAGE.31 Muyzer et al32 reported the use of simi-
lar antibodies for a combined staining technique employing immunofluorescence
and DNA-fluorescence to assess the T. ferrooxidans abundance in coal samples
containing pyrite.
Dot Immunoassay (DIMA) to Determine Ore-Adhered Microorganisms

Later, we described the use of a convenient dot immunoassay (DIMA) which is
not only fast, but both sensitive and specific for the determination of T. ferrooxidans,
L. ferrooxidans and T. thiooxidans. 27,33 In this method the bacteria to be deter-
mined are applied to a nitrocellulose membrane by filtration in the form of dots.
After the bacteria contained in the dots are fixed to the membrane, the antigen-
antibody reaction is conducted and visualized by reaction with a secondary anti-
body (which specifically recognizes the primary antibody) conjugated to an en-
zyme. This enzyme will then react with a substrate, producing a colored precipitated
product whose intensity will be proportional to the original number of microor-
ganisms present in the dot and that were recognized by the primary antibody (see
Fig. 14.4). The number of cells can be easily quantitated by comparison with a
standard reaction obtained with known numbers of the corresponding type of
microorganism. The quantitation can be performed with an appropriate computer
image analysis program. We also adapted this immunological assay for the deter-
mination of the moderate thermophile T. caldus34 and for the thermophilic archaeon
Sulfolobus acidocaldarius. 34 In general, these methods are specific and very sensi-
tive, being able to detect as few as 103 cells per dot. 27,33,34 The presence of different
serotypes in biomining acidophiles such as T.ferrooxidans and T. caldus have been
detected.35,3 6 Serotype variation may cause an underestimation of bacterial num-
bers. However, as pointed out by Hallberg and Lindstrom,3 6 this potential draw-
back to the use of immunoassays can be avoided by the selection of an antiserum
which recognizes a common antigen shared among bacteria of the same species.
We have developed a convenient method to monitor the microorganisms ad-
hered to solid substrates. This involves the determination of ore-released bacteria
by using a dot immunoassay (DIMA).27,33>37 When sulfur prills colonized with T.
ferrooxidans (Fig. 14.5A) were subjected to vortexing for five or more minutes in
the presence of Triton X-100 to release the adhered cells, most of the microorgan-
isms were detached (Fig. 14.5B: Garcia, Levican and Jerez, unpublished results).
The remaining bacteria were more difficult to remove, apparently because they
are embedded or "stacked" in pit-like cavities on the sulfur surface. Alternatively,
the remaining adhered bacteria could be bound through the recently described
sulfur binding protein which allows a firm adherence of T. ferrooxidans to sulfur
particles.38 Association of T. ferrooxidans to similar cavities on the surface of py-
rite has been observed before by other groups.22,39 In general, stirring in the pres-
ence of 0.05% Triton X-lOO removes around 80% of the cells initially present in the
biofilm. Similar results were obtained when radioactively-labeled T. ferrooxidans
cells were employed to estimate the percentage removal by the treatment.37
With this method, it is possible to follow the changes of the individual bacterial
populations as bioleaching of a sulfidic ore progresses. When studying a sulfidic
ore subjected to bioleaching in a column, we found that by approximately 60 days,
Methods for Identification and Enumeration of Bioleaching Microorganisms 2B7
e-· • •
Fig. 14.4. The dot immunoassay (DIMA). Liquid samples containing the
suspended microorganisms are applied to a nitrocellulose membrane (A)
and subjected to vacuum fIltration to generate dots that contain the bac-
teria attached to the membrane (B). After treating the membrane with the
corresponding antibodies as described in Fig. 14.3, the number of micro-
organisms present in each dot is determined by comparison with known
numbers of the specific microorganism being analyzed and which has been
reacted in parallel under the same conditions (as shown at bottom of
figure).
----10 um
Fig. 14.5. Removal of the bacteria from a T. ferrooxidans mm covering a

sulfur prill. A. Biomm formed after growth and colonization of sulfur prills
by T. ferrooxidans for two weeks. B. The same kind of colonized samples
after treatment byvortexing during 10 min in the presence of 0.05% Triton
X-I00. Both samples were analyzed after processing for scanning electron
microscopy.
the ore-adhered T. ferrooxidans population greatly decreased, with T. thiooxidans

and L. ferrooxidans being the predominant species.37 This method is very useful to
specifically monitor the changing microbial population dynamics of some of the
main bacterial species present in bioleaching operations and is currently being
applied in several Chilean industrial bioleaching processes.
It is necessary to consider that these immunological methods and the ones
involving nucleic acid analysis (see next sections) measure both live and dead cells.
Therefore, to determine the activity of the microorganisms, another alternative
method should be applied in conjunction with these methods.
Fig. 14.6. The general principle of DNA-DNA hybridization. Total DNA extracted from
bacteria or the DNA present in single bacterial cells permeabilized to expose their DNA
will contain determined nucleotide sequences specific for the microorganism being
analyzed. A DNA probe is designed to have a nucleotide sequence which is comple-
mentary to only the microorganism of interest. A tag such as a fluorescent compound
( • ) is chemically bound to the probe. Under the appropriate conditions, a specific
hybridization will take place with only those microorganisms having the complemen-
tary DNA sequence.
Methods Involving Nucleic Acids

Dot-Blot and Southern Blot Hybridizations
Several detection and monitoring procedures based on DNA have been devel-
oped to study microbial population structure and dynamics. For DNA isolation
from solid samples such as soil or ores, direct lysis and cell extraction approaches
can be employed. Direct lysis involves treatment of a sample with hot sodium
dodecyl sulfate (SDS) and mechanical disruption of cells by shaking with inert
particles. Extracted DNA is then purified by standard procedures. DNA recovered
by this general method is sufficiently pure for dot blot DNA-DNA hybridization.
However, this DNA is usually sheared (average size less than 10 kb), excluding the
application of Southern blot analysis.
Alternatively, DNA extraction involves the separation of the cells from the ore
particles by physical means. Once the cells are separated, the DNA is extracted as
before and purified. The resulting DNA in this case is of adequate purity and size
for Southern blot analysis. 3 An illustration of the general principle of specific hy-
bridization is seen in Figure 14.6.
Yates et al40 were the first to suggest the use of genetic probes to identify differ-
ent species and strains of thiobacilli commonly isolated from biomining opera-
tions. These DNA probes can distinguish between strains of microorganisms as
well as between species and genera. The probes are very sensitive and can detect
less than 104 microorganisms. 4o Many applications of nucleic acid hybridization
using environmental samples involve probing immobilized DNA sequences fixed
to a nitrocellulose or nylon membrane. The probe can be double-stranded, and
usually comprises specific sequences of genomic or plasmid origin. When the se-
quence of a specific gene or part of it is known, the probes can be constructed in
vitro to detect the gene of interest. An example of the use of Southern blot analysis
in which we used a probe specific for a gene of T. ferrooxidans which is expressed
when this microorganism is grown in the presence of sulfur but not when grown
in the presence of ferrous iron is shown in Figure 14.7 (Toledo and Jerez, unpub-
lished results). The probe was synthesized as a mixture of oligonucleotides on the
basis of the N-terminal end sequence of the protein induced by growth in sulfur
and the frequency of codon usage of T. ferrooxidans. 41 The probe was able to react
Fig. 14.7. Southern blot analysis of

DNA from different microorgan-
isms. Different microorganisms
(A) are treated to extract their
DNA (B). DNA preparations are
\~/~ A
then subjected to hydrolysis by a ~ ~ ~ ~
~~tt~
restriction enzyme to generate
numerous fragments of DNA with B
different sizes. Fragments are
~ ~ ~ ~
then separated by electrophoresis
in an agarose gel as seen in (C.) c
After separation, the DNA frag-
ments are transferred to a nitro-
cellulose or nylon membrane by
Southern blotting. The trans-
ferred fragments are then reacted
with the labeled-probe which will
hybridize only with those frag-
ments of DNA having a nucle-
otide sequence complementary to
--
the probe (D). In this case, an oli-
-
gonucleotide probe to recognize
a gene of a protein induced when
T. ferrooxidans is grown in sulfur
was employed. Tf, T. ferrooxidans;
Lf, L. ferrooxidans; Tt, T. thio-
oxidans; Ec, E. coli.
Kb o
23.1--
9.4 - -
6.5 - -
-
4.3 - -
2.3--
2.0 - -
·1
1,
I
Tf Lf Tt Ec
specifically with a DNA fragment from T. ferrooxidans, a bacterium capable of

oxidizing ferrous iron and sulfur (Fig. 14.7D ). However, there was no reaction with
DNA from L. ferrooxidans. This was expected, since this microorganism is not
able to oxidize sulfur. When T. thiooxidans DNA was tested, no reaction was de-
tected in spite of its ability to oxidize sulfur and its derivatives. These results indi-
cate that it is possible to specifically differentiate these three different species un-
der the conditions employed. Therefore, total DNA extraction and subsequent
probing for a particular trait can be used to monitor the presence or absence of
that characteristic. With suitable probes, it is possible to detect specific nucleic
acid sequences not only in bacteria following culture but also directly from envi-
ronmental samples.l •M
Pulse Field Gel Electrophoresis (PFGE)

Using pure bacterial cultures, extracted DNAs can also be analyzed by pulse
field gel electrophoresis (PFGE). When total DNA ofbioleaching microorganisms
such as T. ferrooxidans and T. thiooxidans is cleaved by using certain restriction
enzymes which generate big fragments, it is possible to separate the entire frag-
mented genome of the microorganism by pulse field gel electrophoresis. Each type
of thiobacilli will give a different macrorestriction pattern, allowing its prelimi-
nary differentiation and possibly its identification by comparison with reference
patterns. 42
Polymerase Chain Reaction (PCR) Amplification of 16S rRNA and rDNA

If samples to be analyzed do not contain enough of the target microorganism
or its nucleic acid, it is possible to use the polymerase chain reaction (PCR) to
increase the relative concentration of the target following the extraction of total
DNA from the environmental sample. With this procedure, specific segments of
DNA can be amplified millions of fold when the appropriate primers are used to
copy the DNA fragment to be amplified (see Fig. 14.8).
Ribosomal RNA genes are essential for the survival of all organisms and are
highly conserved in the bacterial and other evolutionary domains. Therefore, the
characterization of the 16S rRNA gene is now a well-established standard method
for the identification of species, genera and families of microorganisms. l •43 The
rRNA offers advantages to be used as the target sequence, since it is naturally am-
plified in the cell and contains sequences that may be specific for the target organ-
ism. This procedure has revolutionized microbial ecology. The analysis employs
mainly the 16S rRNA or 16S rDNA from both pure and mixed cultures of bacte-
ria. 43 Oligonucleotide probes have been developed that are domain-, genus-, spe-
cies- or strain-specific. Partial16S rRNA sequences have already been obtained
from a number of bacteria commonly found in acidic biooxidation environ-
ments. 9•43-45
Rawlings 46 prepared genomic DNA from isolates of T. ferrooxidans, T. thio-
oxidans and L. ferrooxidans and the 16S rDNA from each strain was amplified us-
ing PCR, cloned and their restriction enzyme sites maps were obtained. From a
comparison of these restriction maps it was possible to choose restriction sites
allowing the rapid identification of each of the three types of bacteria. This meth-
odologywas applied to the population of bacteria in a biooxidation tank. As pointed
out by Rawlings,46 the 16S rRNA technique cannot be expected to be quantitative
I L
I I
292
Denatur;
+ ,
Synthesis o~ +
Add new DNA
Primers
1\ /\
IIII
1
Repeat several times
previous sleps
~
Amplified DNA
Fig. 14.8. The polymerase chain reaction (peR) to amplify DNA. The DNA fragment to
be amplified is denatured by heat to separate both strands and the specifically designed
oligonucleotide primers are added. After cooling, the primers hybridize as indicated.
New DNA is then synthesized by the DNA polymerase from the primers. After repeat-
ing the denaturation/renaturation and DNA synthesis steps in the presence of the same
primers 20-30 times, the original DNA can be fmally amplified millions of times.
in regard to being able to use the amount of peR product to estimate the number
of bacteria in a sample. It may, however, be quantitative in the sense that analysis
of the peR product reflects the distribution of bacterial species within a sample.
Although the techniques employing analysis of nucleic acids are very specific
and are the only way of identifying nonculturable microorganisms in environmental
samples, they also offer a number of problems that have to be considered. As an
example, it is important that the extraction of DNA from the different species in a
given sample be quantitative and that all species of bacteria are completely lysed.
Also, it is important to know the number of 16S rRNA genes for each of the species
under investigation, since it may vary substantially between different types of mi-
croorganisms (between 1 and 10 copies of these genes have been reported in differ-
ent bacteria),47
Another useful and promising technique to analyze microbial populations is
the use of denaturing gradient gel electrophoresis (DGGE). In DGGE, DNA frag-
ments of the same length are separated on the basis of their nucleotide sequence.
The fragments migrate through the denaturant gradient until they reach a concen-
tration at which the double strand opens and migration stops. The concentration
at which the fragment stops migrating depends on its melting behavior and there-
fore on its sequence. With this procedure, a high resolution separation of DNA
fragments is obtained and it can be used to separate peR-amplified 16S rDNA frag-
ments. 48 Although its use has not been reported for biomining microorganisms,
its application can be very useful on the characterization of these microorganisms.
Recently, Pizarro et al49 studied the bacterial population composition, the se-
lection occurring upon culturing and the potential species involved in the
bioleaching of copper ores. They used the method suggested by Jensen et al,5 0 which
consists of the amplification by peR of the spacer regions between the 16S rRNA
and 23S rRNA genes and subsequent comparison of the products by gel electro-
phoresis with those from the main species isolated from bioleaching systems. The
size of the amplified spacer region allows a preliminary identification of the differ-
ent bacteria providing the DNA.49.5 0 The pattern of the products obtained indi-
cated the presence of a complex mixture of bacteria which changed under differ-
ent leaching conditions. At low ferrous iron concentration, none of the products
corresponded to that obtained from T. ferrooxidans, indicating that the numbers
of this microorganism were practically undetectable. The two spacers observed in
larger abundance had similar electrophoretic migration to the one of the amplifi-
cation products ofDNAs from T. thiooxidans andL.ferrooxidans. The results were
confirmed by sequencing part of the 16S RNA genes adjacent to the spacer regions.
Rawlings, using the restriction enzyme analysis of 16S rRNA approach46 and
we,37 using an entirely different approach such as the immunological analysis, also
found that when analyzing the microorganisms present in a biooxidation tank and
in a copper sulfide mineral column, respectively, T. ferrooxidans was almost or
entirely absent, and T. thiooxidans and L. ferrooxidans were present in greater
amounts. It is interesting therefore, that three entirely different methods arrived at
similar conclusions regarding the changes in microbial populations of the same
microorganisms determined in different bioleaching experiments.
Determination of Microorganisms In Situ

Both antibodies and nucleic acid probes can be tluorescently labeled, thus al-
lowing detection and monitoring of microorganisms by employing a standard
epitluorescence or confocal microscope. The general principle is illustrated in Fig-
ure 14.9 (see also Fig. 14.6). This procedure requires that the probe employed be
specifically detected within morphologically intact cells (whole-cell DNA hybrid-
ization or immunodetection). In situ detection1 is more adequately employed for
the whole-cell detection in which microorganisms are detected in their natural
microhabitat. It is possible therefore to determine the cell morphology of an un-
cultured microorganism, its abundance and also its spatial distribution in situ.1
As mentioned in a previous section, the immunotluorescence approach gives
excellent spatial resolution and it can be instantaneously detected by epitluorescence
microscopy. Fluorescently labeled rRNA targeted probes have been demonstrated
to allow detection of individual cells.1.51 Like immunotluorescence, whole-cell hy-
bridization with tluorescently labeled rRNA-targeted oligonucleotides can be com-
bined with tlow cytometry for a high-resolution automated analysis of mixed mi-
crobial populationsY
We are currently developing antibodies against surface proteins of T. ferro-
oxidans which are specifically induced under certain growth conditions (for ex-
ample, the lack of the essential nutrient phosphate).52 With this kind of tluorescent
antibody, we expect to be able to monitor the in situ physiological state of
bioleaching microorganisms. This could in turn allow us to correct the environ-
mental conditions (for example, to add the detected missing nutrients) to improve
the bioleaching rates.
Fig. 14.9. The general principle of in situ detection of microorganisms by the use of
fluorescent probes. A fluorescent antibody ( ~ ) or fluorescent oligonucleotide ( ~ )
probe specific to recognize a given protein or nucleotide sequence in a microorganism
is used to react with a mixed population of microorganisms. Only the microorganisms
where the target protein or DNA is present will fluoresce under the epifluorescence
microscope, and can therefore be specifically distinguished from the rest of the bacteria.
In the near future, I expect an increase in use of molecular methods such as

those briefly reviewed here, especially fluorescent r RNA and immunological probes
for the in situ monitoring of bioleaching microorganisms in bioleaching opera-
tions.
Conclusions
There are several methods available for the detection, identification and enu-
meration of bioleaching microorganisms. Some of them require rather specialized
equipment and well trained personnel to handle them. However, several of these
can now be used in routine analysis in biomining operations, allowing changes in
the microbial population to be assessed. This information would indicate how to
avoid conditions that make the bioleaching process to slow down or stop. More
importantly, the incorporation of these new techniques will permit a definition in
greater detail of the entire bioleaching microbial population which has been rather
elusive until now.
Acknowledgments
Funding for the research conducted in our laboratory was provided by grants
from FONDECYT, UNIDO, SAREC and ICGEB.
References
1. Ammann RI, Ludwig W, Schleifer KH. Phylogenetic identification and in situ
detection of individual microbial cells without cultivation. Microbiol Rev 1995;
59: 143- 1 69.
2. McFeters GA, Yu FP, Pyle BH et al. Physiological assessment of bacteria using
fluorochromes. J Microbiol Meth 1995; 21:1-13.
3. Pickup RW. Development of molecular methods for the detection of specific bac-
teria in the environment. J Gen Microbiol1991; 137:1009-1019.
4. Busse, HJ, Denner, EBM, Lubitz W. Classification and identification of bacteria:

current approaches to an old problem. Overview of methods used in bacterial
systematics. J Biotechnol 1996; 47:3-38.
5. Gurtler V, Stanisich VA. New approaches to typing and identification of bacteria
using the 16S-23S rDNA spacer region. Microbiol 1996; 142:3-16.
6. Tuovinen OH, Kelly BC, Groudev SN. Mixed cultures in biological leaching pro-
cesses and mineral biotechnology. In: Zeikus JG, Johnson EA, eds. Mixed Cul-
tures in Biotechnology. New York: McGraw-Hill, 1991:373-427.
7. Harrison AP Jr. The acidophilic thiobacilli and other acidophilic bacteria that share
their habitat. Annu Rev Microbiol 1984; 38:265-292.
8. Hutchins SR, Davidson MS, Brierley JS et al. Microorganisms in reclamation of
metals. Annu Rev Microbiol1986; 40:311-336.
9. Johnson DB, Macvicar JHM, Rolfe S. A new solid medium for the isolation and
enumeration of Thiobacillus ferrooxidans and acidophilic heterotrophic bacteria.
J Microbiol Methods 1987; 7:9-18.9.
10. Goebel BM, Stackebrand E. Cultural and phylogenetic analysis of mixed micro-
Envir Microbiol1994; 60:1614-1621.
11. Southam G, Beveridge TJ. Enumeration of Thiobacilli within pH-neutral and acidic
mine tailings and their role in the development of secondary mineral soil. Appl
12. Manning HL. New medium for isolating iron oxidizing and heterotrophic acido-
philic bacteria from acid mine drainage. Appl Microbiol1975; 30:1010-1016.
13. Mishra AK, Roy P, Mahapatra SSR. Isolation of Thiobacillus ferrooxidans from
various habitats and their growth pattern on solid medium. Curr Microbiol 1983;
8:147-152.
14. Peng JB, Yan WM, Bao XZ. Solid medium for the genetic manipulation of
Thiobacillus ferrooxidans. J Gen Appl Microbiol 1994; 40:243-253.
15. Lindstrom EB, Sehlin HM. High efficiency of plating of the termophilic sulfur-
dependent archaebacterium Sulfolobus acidocaldarius. App1. Environ Microbiol
1989; 55:3020-3021.
16. Bagdigian RM, Myerson AS. The adsorption of Thiobacillus ferrooxidans on coal
surfaces. Biotechnol Bioeng 1986; 28:467-479.
17. Yeh TY, Godshalk JR, Olson GJ et a1. Use of epifluorescence microscopy for char-
acterizing the activity of Thiobacillus ferrooxidans on iron pyrite. Biotechnol
Bioeng 1987; 30:138-146.
leaching. Appl Environ Microbiol 1992; 58:85-92.
19. Monroy FM, Mustin C, de Donato P et al. Occurrences at mineral-bacteria inter-
face during oxidation of arsenopyrite by Thiobacillus ferrooxidans. Biotechnol
Bioengn 1995; 46:13-21.
20. Arredondo R, Garda A, Jerez CA. The partial removal of lipopolysaccharide from
Thiobacillus ferrooxidans affects its attachment to solids. Appl Environ Microbiol
1994; 60:2846-2851.
21. Karamanev DG. Model of the biofilm structure of Thiobacillus ferrooxidans. J
Biotechnol1991; 20:51-64.
22. Sand W, Gehrke T, Hallmann Ret a1. Sulfur chemistry, biofllm, and the (in)direct
attack mechanism-a critical evaluation of bacterial leaching. Appl Microbiol
Biotechnol 1995; 43:961-966.
23. Page S, Burns RG. Flow cytometry as a means of enumerating bacteria intro-
duced into soil. Soil BioI Biochem 1991; 23:1024-1028.
24. Tebbe CC, Vahjen W. Interference of humic acids and DNA extracted directly
from soil in detection and transformation of recombinant DNA from bacteria
and yeast. Appl Environ Microbiol 1993; 59:2657-2665.
25. Lindahl V, Bakken LR. Evaluation of methods for extraction of bacteria from soil.
FEMS Microb Ecol 1995; 16:135-142.
26. Pronk JT, deBruyn JC, Bos P et al. Anaerobic growth of Thiobacillus ferrooxidans.
27. Arredondo R, Jerez CA. A specific dot-immunobinding assay for detection and
enumeration of Thiobacillus ferrooxidans. Appl Environ Microbiol 1989;
55:2025-2029.
28. Apel WA, Dugan PR, Filppi JA et al. Detection of Thiobacillus ferrooxidans in
acid mine environments by indirect fluorescent antibody technique. Appl Environ
Microbiol 1976; 32:159-165.
29. Baker KH, Mills AL. Determination of the number of respiring Thiobacillus
ferrooxidans cells in water samples by using combined fluorescent antibody 2-(p-
iodophenyl)-3-) p-nitrophenyl)-5-phenyltetrazolium chloride staining. Appl
Environ Microbiol 1982; 43:338-344.
30. Gates, JE, Pham KD. An indirect fluorescent antibody staining technique for de-
termining population levels of Thiobacillus ferrooxidans in acid mine drainage
waters. Microb Ecol 1979; 5:121-127.
31. Jerez CA, Peirano I, Chamorro D et al. Immunological and electrophoretic differ-
entiation of Thiobacillus ferrooxidans strains. In: Lawrence RW, Branion RMR,
Ebner, HG, eds. Fundamental and Applied Biohydrometallurgy. Proceedings of
the sixth international symposium on biohydrometallurgy, Vancouver, British
Columbia, Canada. Elsevier, Amsterdam, 1986:443-456.
32. Muyzer G, DeBruyn AC, Schmedding DJM et al. A combined immunofluorescence-
DNA-fluorescence staining technique for enumeration of Thiobacillus ferrooxidans
in a population of acidophilic bacteria. Appl Environ Microbiol1987; 53:660-664.
33. Jerez CA, Arredondo R. A sensitive immunological method to enumerate
Leptospirillum ferrooxidans in the presence of Thiobacillus ferrooxidans. FEMS
Microbiol Lett 1991; 78:99-102.
34. Amaro AM, Hallberg KB, Lindstri)m EB et al. An immunological assay for the
detection and enumeration of thermophilic biomining microorganisms. Appl
35. Koppe B and Harms H. Antigenic determinants and specificity of antisera against
acidophilic bacteria. World J Microbiol1994; 10:154-158.
36. Hallberg KB, Lindstri)m EB. Multiple serotypes of the moderate thermophile
Thiobacillus caldus, a limitation of immunological assays for biomining microor-
ganisms. Appl Environ Microbiol 1996; 62:4243-4246.
37. Garda A, Jerez CA. Changes of the solid-adhered populations of Thiobacillus
ing ores as determined by immunological analysis. In: Jerez CA, Vargas T, To-
ledo H, Wiertz JV eds. Biohydrometallurgical Processing. Vol II.
Santiago:University of Chile, 1995:19-30.
38.0hmura N, Tsugita K, Koizumi JI et al. Sulfur-binding protein of flagella of
Thiobacillus ferrooxidans. J Bacteriol 1996; 178:5776-5780.
39. Mustin C, Donato P, Berthelin J. Quantification of the intragranular porosity
formed in bioleaching of pyrite by Thiobacillus ferrooxidans. Biotechnol Bioeng
1992; 39:1121-1127.
40. Yates JR, Lobos JH, Holmes DS. The use of genetic probes to detect micro-organ-
isms in biomining operations. J Int Microbiol 1986; 1:129-135.
41. Rawlings DE, Woods DR, Mjoli NP. The cloning and structure of genes from the
autotrophic biomining bacterium Thiobacillus ferrooxidans. In: Advances in Gene
Technology. Vol 2. London:JAI Press Ltd., 1991; 215-237.
42. Irazabal N, Moreira D, Amils R et al. Comparative genomic organization of
Thiobacilli using pulsed field gel electrophoresis. In: Jerez CA, Vargas T, Toledo
H, Wiertz JV eds. Biohydrometallurgical Processing. Vol II. Santiago:University
of Chile, 1995:31-42.
43. Pace NR, Stahl DA, Lane DJ et al. The analysis of natural microbial populations
by ribosomal RNA sequences. Adv Microbiol Ecol 1986; 9:1-55.
44. Lane DJ, Harrison AP, Stahl D et al. Evolutionary relationships among sulfur-
45. Woese CR. Bacterial evolution. Microbiol Rev 1987; 51:221-271.
46. Rawlings DE. Restriction enzyme analysis of 16S rDNA for the rapid identifica-
tion of Thiobacillus ferrooxidans, Thiobacillus thiooxidans and Leptospirillum
ferrooxidans strains in leaching environments. In: Jerez CA, Vargas T, Toledo H,
Wiertz JV eds. Biohydrometallurgical Processing. Vol II. Santiago:University of
Chile, 1995:9-17.
47. DeLong EF, Wickham GS Pace NR. Phylogenetic stains: ribosomal RNA-based
probes for the identification of single microbial cells. Science 1989; 243:1360-1363.
48. Muyzer G, Hottentrager S, Teske A et al. Denaturing gradient gel electrophoresis
of PCR-amplified 16S rRNA-a new molecular approach to analyze the genetic di-
versity of mixed microbial communities. In: Akkermans ADL, van Elsas JD, de
Bruijn FJ eds. Molecular Microbial Ecology Manual. Kluwer Academic Publish-
ers, Dordrecht, The Netherlands, 1996; 1-23.
49. Pizarro J, Jedlicki E, Orellana 0 et al. Bacterial populations in samples of
bioleached copper ore as revealed by analysis of DNA obtained before and after
cultivation. Appl Environ Microbiol1996; 62:1323-1328.
50. Jensen MA, Webster JA, Straus N. Rapid identification of bacteria on the basis of
polymerase chain reaction-amplified ribosomal DNA spacer polymorphisms. Appl
51. Amann RI, Binder BJ, Olson RJ et al. Combination of 16S rRNA-targeted oligo-
nucleotide probes with flow cytometry for analyzing mixed microbial populations.
52. Seeger M, Jerez CA. Phosphate limitation affects global gene expression in
Thiobacillus ferrooxidans. Geomicrobiol J 1992; 10:227-237.
Index
A Bioheap leaching, 9, 13, 15-16, 104,167

advantages, 8
Achromobacter delicatulus (Enterobacter Bioheap reactors, 8, 15
cloacae), 143 Bioleaching, 3-4, 103-104, 106-107, 110,
Acid levels, 24 129,131,134-135,137-138,140-141,143,
Acid rock drainage, 10-12, 104, 121 145,153-175,229-245,265
Acidianus brierleyi, 136, 251, 263 copper, 9, 21
Acidimicrobium ferrooxidans, 249-250, defmition, 3
263 in situ, 4, 9
Acidophilic heterotrophic bacteria, nutrients, 27
250-260,263-268,270,272,275 BioNIC, 46, 54
Acidothermus, 135 Biooxidation, 3-4, 11-13, 15-16, 45-46,
Agglomeration, 24-25, 28,38,107-109, 48-50,53-54,58,60-70, 103-113,
112,118 117,-119,121-124,126-127,201-206,208,
Agitator selection, 62, 89 210-212, 219-220, 222-223, 230, 241,
Alicyclobacillus, 248, 262-264, 266, 291, 293
267-268, 272 definition, 4
Arsenate precipitate(s), 60, 71-73, 99 nutrients, 48
Arsenate precipitation, 99 Biooxidation heap, 201-202
Arsenic Biooxidation heap theory, 203
toxic effect, 69 Bioreactor
inhibition, 56, 163 control and monitoring, 95
Arsenopyrite oxidation, 58 temperature control, 91
Aspergillus, 139-141 Bioreactor configuration, 64-65, 94
Aspergillus niger, 141 Bioreactor materials, 68, 87
Bioremediation, 3, 10, 12-13, 16, 177
B BIOX® bacterial culture, 54
BIOX® process, 240
Bacillus circulans, 141 Brettanomyces lambicus, 142
Bacillus mucilaginosus, 144 Brohm mining, 109, 202
Bacillus polymyxa, 144
BACOX,81
BacTech process, 81-103
c
Bacteria-surface Carbon dioxide fixation, 155, 233, 236,
electrostatic force, 181 250,252
Bacterial adhesion, 184, 284 Carbon dioxide utilization rate, 158
Bacterial attachment, 48, 55, 69, 178, 180, Carbonate content, 60-61
185, 192, 198, 283 Chalcocite, 8, 13, 16, 22-23, 38, 168, 170
Bacterial cell Chalcopyrite, 12-13, 22-23, 42, 73, 81, 98,
hydrophobicity, 181 100,134,183-185,247,251-253,255,
Bacterial inoculation, 29 265,273
Bacterial leaching Chloride
potential advantages, 82 effect of, 68
Bacterial mineral oxidizing ability, 237 sensitivity, 57
Bacterial numbers, 265 Citric acid, 139-141, 144-145, 267
Bauxite Cobalt, 4, 34, 41, 131-133, 139-141, 272-273
heterotrophic leaching, 145 heterotrophic leaching, 139
BiofIlm(s),281-282 Cold-tolerant strains, 240
formation of, 178, 185, 192 Colloidal sulfur, 186
3 00 Biomining: Theory, Microbes and Industrial Processes
Column biooxidations, 123 F

Concentrate coating, ll9-121, 123, 127
Concentrates Fairview Plant, 48
base metal(s), 4 Ferric iron inhibition, 154, 157
refractory, 4 Ferric/ferrous-iron ratio, 155,160,
Cooling requirements, 48, 60 165, 167
Copper, 3-4, 8-9,ll-12, 21-29, 31-34, Ferromicrobium acidophilus, 263,
36-38,40-42, 46, 81, 100, 104, 107, llO, 266,268
129-130,132-135,142,168,170,177,202, Finger dump construction, 26
208, 2ll, 230, 234, 237, 240-241, 247,
251, 253, 255, 265, 268-269, 293 G
cementation, 32
solvent extraction, 33 Gallium, 135, 142
Copper-tolerant, 132 heterotrophic leaching, 142
Covellite, 22-23,134-135 Gas permeability, 204, 2ll-212, 215,
Cyanicides, 8 217,220
Cyanidation, 99, 108 GENCOR, 45-46, 49-50, 54, 58, 64, 69,71,
Cyanide 73-74, 103, 167, 201
microbial degradation of, 10 Process Research, 45-46, 50, 54, 64, 71,
73-74
D Genmin,104
Geobacter sulfurreducens, 136
Detection of bacteria using DNA GeoBiotics, Inc., ll7
hybridization, 289 process, ll9, 121, 126-127
Detection of bacteria with Gold
antibodies, 284 refractory, 7-8, 16, 45-46, 49, 52,54, 58,
Direct mechanism, 27, 55, 130-131, 153, 61,69,77,81-84,99-100,103-104,
162, 177-178 106-107, ll7-ll9, 123, 126, 153,
Direct versus indirect attack, 13, 15 202-203, 240
Dot immunoassay, 286
Drip-line irrigation, 29 H
E HS
•precipitated, 10
Effect of chloride, 68 Harbour Lights Plant, 50
Effluent disposal, 70 Heap
Electrochemical mechanism ferric iron regeneration, 203
of dissolution, 190 gas flow, 212
Electrochemical mechanism of leaching, heat production, 205
189-191,193,195-196 irrigation rate, 216
Electrochemical model of leaching, 188, oxygen flux, 204
190-191 oxygen diffusion, 26
Electron acceptor, 136, 230, 235-236, 268 permeability, 25-27, 31, 105
Electron donors, 136, 236 pore gas measurement, 220
Electronic structure of a mineral, 190 sulfide sulfur oxidation, 206
Electrowinning, 8, 32-34, 36, 38, 82, 134 temperature measurement, 220
Enumeration ofbioleaching Heap biooxidation,16, 104, ll8-ll9,122,
microorganisms, 281, 294 124, 127, 201-202
Environmental considerations, 36 temperature dependence, 209
Epifluorescence microscopy, 282, Heap reactors, ll8
284,293 microbiology, 106
particle size, 106
Index 301
Heterotrophic iron -oxidizing Metal-binding, 12-13

bacteria, 273 Microbial cyanide oxidation, 10, 12
Heterotrophic leaching, 130, 136, 138-139 Moderate thermophiles, 81-82, 91-92, 95,
High temperature bioleaching, 253 99-100,137,209-210,229,247-249,
Homestake cyanide oxidation, 10 252-253, 268, 272
Molecular biology of acidophilic
I heterotrophs, 270
Molybdenum, 135, 144, 236, 240
In situ detection of microorganisms, 293 Monod kinetics, 154-155, 163, 208, 210,
Indirect mechanism, 55, 130, 153, 177, 215-216
272-273 Mt. Leyshon Gold Mine, 110, 202
Inoculum preparation, III
Intrinsic oxidation rate, 203,206 N
Iron oxidation, 11, 57, 68-69, 233-234,
250-252,260,262,275 Newmont Gold Company, 107, 109,
Iron oxidation rate, 160 111-112, 167, 201
Iron oxidation pathway, 234 Nickel, 4, 46, 81, 100, 133, 139-141
heterotrophic leaching, 139
Nitrogen flxation, 234
J Nutrients, 68, 93, 108
Jarosite, 27, 37, 58, 69-70, 131, 134, 273 Nutrition of acidophilic
heterotrophs, 267
K
o
Kinetic models, 153, 170-171
Off-gas analysis, 156-157, 165
Operating curves, 64
L
Optimizing heap performance, 212
Leaching Optimum grind size, 67
arsenopyrite, 50 Optimum temperature, 4, 81, 91, 229, 239,
direct and indirect, 12, 55, 186 247-248,251-252
dump operations, 4, 8-9, 11-12, 22, 24, Ore
26,29,31,36-38 preg-robbing,8
heap, 8, 21-22, 25-26, 31, 36, 38, 82-83, stacking, 31
103-104,118,126-127,170,201-202, submarginal, 4, 9
208,222 Ore mineralogy
heat, influence of, 58
effects of, 28 Organic carbon content, 97
in situ, 13, 22-24, 36-37, 82 Oxalic acid, 139-140
Leaching mechanism, 56, 170, 273 Oxidation reactions, 58, 91
Lead Oxygen supply, 62-63, 70, 90
heterotrophic leaching, 141 Oxygen utilization rate, 156-157, 159,
Leptospirillum ferrooxidans, 4, 54, 106, 165-166
132,154,167,229,248,259,282
Lithium p
Logistic equation, 160, 162 Penicillium, 139-141, 144-145
pH control, 60-61, 70-71, 96
Phylogenetic relationships, 263
M Picrophilus, 263, 267-268
Manganese, 32, 136, 142-144, 268 Platinum group metals, 137
heterotrophic leaching, 142 Polymerase chain reaction (PCR), 291
Mesophilic bacteria, 229 Pore gas oxygen concentration, 206, 208,
Metal tolerance, 240 210, 212, 219, 222-223
Power consumption, 62 T
Pulp density, 93-94, 96, 118
Pyrite oxidation, 22, 58, 91, 105, 132, 165,
210,224,249,273 Technological advancements, 12, 14
Pyrrhotite oxidation, 58 Thermophilic acidophiles, 247, 252, 263
Thermoplasma acidophilum, 262,
267-268
R
Thiobacillus ferrooxidans, 3, 11, 22,54,
Redox potential, 55, 59, 69, 108, 135, 153, 103,106-107,109-112,129-138,141,154,
155-156,160,163,165-168,171,186,236 156-157,160-161,167-168,170,178-180,
Refractory gold ore, 16, 45-46, 49, 58, 61, 183-186,198,229,247-248,259,282
69,77,83,103-104,106-107,117-118,123, adhesion Of,179
126,153,202-203 contact angles, 184
Release of adhered bacteria, 284 copper-tolerant, 132
Research and development, 3, 11-12, nickel tolerance, 133
97,201 zinc adapted, 134
Rhizopus arrhizus, 140 Thiobacillus acidophilus, 230, 239,
Rock size, 23, 120 260,263
16S rRNA, 231, 237, 240, 250, 260, 263, Thiobacillus caldus, 229, 237, 247, 251
265-266,291-293 Thiobacillus thiooxidans, 28, 54, 57, 107,
5S rRNA sequences, 231 110,131-132,137-138,144-145,229,247,
16S rRNA sequences, 231 251, 259, 282
Thiourea leaching, 109
s Toxic reagents, 68
Sulfobacillus acidophilus, 248-250 u

Sansu Plant, 53
Sao Bento Plant, 49 Uranium, 4,9,107,129,137-138,177,230,
Serotype variation, 286 240, 247, 265, 268
Shewanella putrefaciens, 136
Silver, 4, 129-130, 136, 142-144, 186, 240 v
Solution collection, 31 van der Waals force, 180
Solution stacking, 29, 31
Solvent extraction, 8, 29, 33, 36, 38,
40-41, 82, 94, 99, 131, 134
w
Specific growth rate, 154, 157, 159 Wiluna Plant, 52
Stability of ferric arsenate, 60,72-73
Stack height, 8 y
Stirred-tank bioreactors, 4,7-8
Stirred-tank reactors, 4, 7, 12-13 Youanmi mine, 81-82, 91, 103
disadvantages, 7
Sulfate-reducing bacteria, 10, 12, 260
Sulfobacillus, 4, 247-253, 262
z
Sulfobacillus thermosulfidooxidans, 248 Zeta potential, 182-185, 239
Sulfolobus acidocaldarius, 251, 263, Zinc, 4, 10,32,107,132-134,141,146-147,
282,286 162-163,190,193,197,248,251, 265
Sulfolobus metallicus, 252, 253 heterotrophic leaching, 141
Sulfur oxidation, 110, 138, 202-204, 206, Zlata Mine, 110
209-212,216,219-220,223,230,
235,252
Support rock, 119-124, 126-127

Springer-Verlag Berlin Heidelberg GMBH

Uploaded by

Copyright:

Available Formats

You might also like

Springer-Verlag Berlin Heidelberg GMBH

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Springer-Verlag Berlin Heidelberg GMBH

Uploaded by

Copyright:

Available Formats

Springer-Verlag Berlin Heidelberg GmbH

Douglas E. Rawlings (Ed.)

LIbrary of Congress Cataloging-in-Publication Data

Section II - Industrial Processes

2.. Bioleaching of Copper ................................................................ 2.1

3. The BIOXe Process for Biooxidation of Gold-Bearing

5. Heap Leaching of Gold-Bearing Deposits:

6. Biooxidation of Refractory Gold Ores

7. Technical Potential for Bioleaching and Biobeneficiation

8. Recent Developments in Modeling the Kinetics

9. Physical Chemistry of Bacterial Leaching .............................. 177

10. Optimization ofBiooxidation Heaps ...................................... 201

Section IV - Leaching Microorganisms

11. Mesophilic, Autotrophic Bioleaching Bacteria:

13. Heterotrophic Acidophiles and Their Roles

14. Molecular Methods for the Identification

Index ..................................................................................................... 299

1=1============ CONTRIBUTORS =============J:j

Frank K. Crundwell D. Barrie Johnson

A. Ian M. Ritchie James 1. Whitlock

T oday large-scale, bioleachinglbiooxidation facilities extract copper and enhance

Industrial Applications of BioleachinglBiooxidation

Biomining: Theory, Microbes and Industrial Processes,

Leptospirillum ferrooxidans and thermophilic species of Sulfobacillus, Acidianus

Table 1.1. Commercial-scale bioleachlbiooxidation plants commissioned

Fairview, South Africa

Table 1.1. Commercial-scale bioleachlbiooxidation plants commissioned

thermophilic bacteria and operates at around 50°C.

Capital and operating costs, highly discernible elements in process selection

Bioheaps and Copper Dump Leaching

Applications of Bioremediation in the Mining Environment

R&D Leading to Today's Commercial Operations

• Extent of microbial leaching of mineral sulfide ores and concentrates (for

Minerals Bioremediation R&D

Table 1.2. Technological advancements in bioleachinglbiooxidation in the

Research & Development Area

Table 1.2. (continued)

Research & Development Area

• Better overall understanding of mineral-microbial interactions in relation

New developments in bioremediation that promise to assist the mining indus-

11. Karavaiko GI, Rossi G, Avakyan ZA, eds. Biohydrometallurgy-Proceedings of

Biomining: Theory, Microbes and Industrial Processes,

Cu 2S + 2Fe3+ ~ Cu2+ + 2Fe2+ + CuS (not the mineral covellite)

Physico-Chemical Leaching Variables

Fig. 2.1. Agglomeration drums at the Quebrada Blanca Operation, Chile.

An interesting phenomenon associated with agglomeration is that some cop-

Bacterial Leaching Variables

eu gain to pregnant leach solution

The addition of air to a heap leach operation is accompanied by several changes.

Heap Operating Variables

Fig. 2.5. Agglomerate stacking at the Quebrada Blanca Operation, Chile.

Leach Solution Processing

Fig. 2.6. Recycle loops in a Leach/SX/EW circuit.

Fig. 2.7. Typical SX flowsheet.

Fig. 2.8. Typical EW flowsheet.

In Situ, San Manuel, BHP Copperl J,32

Dump Leaching-Baja Ley Plant at the Chuquicamata Division

Heap Leaching-The Quebrada Blanca Operationu ,Jo,J3