Professional Documents
Culture Documents
Springer-Verlag Berlin Heidelberg GMBH
Springer-Verlag Berlin Heidelberg GMBH
Springer-Verlag Berlin Heidelberg GMBH
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
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
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
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
Project
Type & Size Technology History & Status
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-
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.
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
10 Biomining: Theory, Microbes and Industrial Processes
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.
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.
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;
14 Biomining: Theory, Microbes and Industrial Processes
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...
Mining Biotechnology: Research to Commercial Development and Beyond 15
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
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.
Mining Biotechnology: Research to Commercial Development and Beyond 17
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
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.
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
24 Biomining: Theory, Microbes and Industrial Processes
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
Bioleaching of Copper 25
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
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.
Bioleaching of Copper 27
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.
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
28 Biomining: Theory, Microbes and Industrial Processes
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.
Bioleaching of Copper 29
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.
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.
30 Biomining: Theory, Microbes and Industrial Processes
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.
Bioleaching of Copper 31
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.
32 Biomining: Theory, Microbes and Industrial Processes
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
Bioleaching of Copper 33
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.
extractant
PLS fee d
pond & pump
iron breed
from EW-..I...-~
to crud t.ank . . . -
to EW _ -..r-'L
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.
Bioleaching of Copper 35
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
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.
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.
Biomining: Theory, Microbes and Industrial Processes
Fig. 2.10. The Quebrada Blanca open pit mine and leach/SX/EW operation.
40 Biomining: Theory, Microbes and Industrial Processes
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
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.
42 Biomining: Theory, Microbes and Industrial Processes
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.
Bioleaching of Copper 43
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
Introduction
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.
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
~
Biomining: Theory, Microbes and Industrial Processes
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.
Table 3.1. Summary ofFairview BIOX- plant performance for the period 1988
to 1992
Yearly Average
1988 1989 1990 1991 1992
Table 3.2. Typical ore and concentrate analyses for Sao Bento
~.
~
~
~ ~ ~ ~ '"~
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
Operating Performance
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.
Table 3.5. Capital and operating cost comparison for a refractory gold
project in Canada
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.
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.
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.
Table 3.7. A summary of the commercial operating BIOX® plants, at the date
ofpublication
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.
Biomining: Theory, Microbes and Industrial Processes
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.
58 Biomining: Theory, Microbes and Industrial Processes
Mineral Sulfide
kJ/kg kJlkg S'-
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.
60 Biomining: Theory, Microbes and Industrial Processes
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.
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-
The BIOX" Process for Biooxidation of Gold-Bearing Ores or Concentrates 61
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
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
62 Biomining: Theory, Microbes and Industrial Processes
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
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
0.07
-'-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')
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.
Biomining: Theory, Microbes and Industrial Processes
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.
- 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)
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.
66 Biomining: Theory, Microbes and Industrial Processes
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.
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.
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
68 Biomining: Theory, Microbes and Industrial Processes
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.
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
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
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
72 Biomining: Theory, Microbes and Industrial Processes
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
As Concentration
(ppm)
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.
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·················· .. · .
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.
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
The BIOX" Process for Biooxidation of Gold-Bearing Ores or Concentrates 77
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.
Biomining: Theory, Microbes and Industrial Processes
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.
80 Biomining: Theory, Microbes and Industrial Processes
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
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.
REGEl'ERATED
II F GROUN D
CA LCRET
~19 reo"~ "',,' r
LOADED CARBON TO
~~~'", '"~11 ill=rl ~
GOLD ROOM
NEUTRALIZATION
~
86 Biomining: Theory, Microbes and Industrial Processes
Table 4.1. Comparison ofdesign criteria (left column) and current operation
for BacTech biooxidation plant at Youanmi mine, Western Australia (Nov.
1996)
continued...
Bacterial Oxidation Using Moderate Thermophiles 87
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.
88 Biomining: Theory, Microbes and Industrial Processes
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.
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.
Bacterial Oxidation Using Moderate Thermophiles 91
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.
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.
Bacterial Oxidation Using Moderate Thermophiles 93
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-
Bacterial Oxidation Using Moderate Thermophiles 95
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.
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
Bacterial Oxidation Using Moderate Thermophiles 99
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.
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.
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
102 Biomining: Theory, Microbes and Industrial Processes
Introduction
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.
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.
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.
Heap Leaching of Gold-Bearing Deposits: Theory and Operational Description 109
Table 5-1. Gold extraction from ore biooxidized using a heap pretreatment
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
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
110 Biomining: Theory, Microbes and Industrial Processes
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
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.
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
Heap Leaching of Gold-Bearing Deposits: Theory and Operational Description 113
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.
114 Biomining: Theory, Microbes and Industrial Processes
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.
Heap Leaching of Gold-Bearing Deposits: Theory and Operational Description 115
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.
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.
Biooxidation of Refractory Gold Ores (The Geobiotics Process) 121
GeoBiotics'
Biooxidation Process
Hypothetical Flowsheet
..... ~
Cons ~
t Recycled Support Rock
Oversize I
•
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-
122 Biomining: Theory, Microbes and Industrial Processes
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.
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
124 Biomining: Theory, Microbes and Industrial Processes
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 ~
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--
~
126 Biomining: Theory, Microbes and Industrial Processes
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
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.
7. Ritchie AIM, Pantelis G. Optimization of oxidation rates in dump oxidation of
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
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
Biomining: Theory, Microbes and Industrial Processes,
edited by Douglas E. Rawlings. © Springer - Verlag and Landes Bioscience 1997.
130 Biomining: Theory, Microbes and Industrial Processes
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
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
132 Biomining: Theory, Microbes and Industrial Processes
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.
Biobeneficiation for Base and Platinum Group Metals 133
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.
134 Biomining: Theory, Microbes and Industrial Processes
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
136 Biomining: Theory, Microbes and Industrial Processes
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
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.
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
138 Biomining: Theory, Microbes and Industrial Processes
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.
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.
Biobeneficiation for Base and Platinum Group Metals 141
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.
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)
Biomining: Theory, Microbes and Industrial Processes
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.
Biobeneficiation for Base and Platinum Group Metals 147
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
Materials Society, 1993:47-56.
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.
Biomining: Theory, Microbes and Industrial Processes
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.
Warrendale, PA: The Minerals, Metals, and Materials Society, 1991:207-218.
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.
Warrendale, PA: The Minerals, Metals, and Materials Society, 1993:293-301.
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
Introduction
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+ ]
=
Ks
I + _...."...-_"'::'-..".-__
[Fe 2+] - [Fe 2 + ]thres
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
= 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:
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
Recent Developments in Modeling the Kinetics of Bioleaching 157
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:
8.12
In this way the rate of bacterial growth, rx,is equal to the rate of carbon dioxide
utilization, -reo l :
= 8.13
-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
158 Biomining: Theory, Microbes and Industrial Processes
Equations 8.12 and 8.13 can be combined into a single balanced equation for
bacterial growth on pyrite as substrate:
= 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
-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.
Recent Developments in Modeling the Kinetics of Bioleaching 159
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.
= 8.25
= 8.26
so that:
~ ma, = q';"~' - ms Ys~ 8.29
similarly
~ = qo , Y~' -mo Y~ 8·30
and:
~ma, = q~~ -mo Yo":' 8.3 1
160 Biomining: Theory, Microbes and Industrial Processes
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.
2.5 r---------------------------...,
••
~ 1.5
-0
E
;;;:
Q
!
8 •
<T
0.5
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.
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.
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
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.
Biomining: Theory, Microbes and Industrial Processes
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]
1 + ~+ K, [As(VW
[Fe 2+] Kj [Fe 2+]
and
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+ ]
166 Biomining: Theory, Microbes and Industrial Processes
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-________________________________________~
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.
4Fe 3+ + 8S0~' + 4W
Recent Developments in Modeling the Kinetics ofBioleaching
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+]
~[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.
168 Biomining: Theory, Microbes and Industrial Processes
L.pto.pirillum
fefrooxld.".
~------~~-------
N
~
:::IE
!.
('11--------------
+
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.
Recent Developments in Modeling the Kinetics ofBioleaching
Ch a lcoc i te
Eh (mY)
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.
170 Biomining: Theory, Microbes and Industrial Processes
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
Recent Developments in Modeling the Kinetics of Bioleaching 171
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
172 Biomining: Theory, Microbes and Industrial Processes
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).
Recent Developments in Modeling the Kinetics of Bioleaching 173
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.
174 Biomining: Theory, Microbes and Industrial Processes
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.
Recent Developments in Modeling the Kinetics of Bioleaching 175
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
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.
(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.
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-
Physical Chemistry of Bacterial Leaching 181
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
182 Biomining: Theory, Microbes and Industrial Processes
o
o
Distance
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
Biomining: Theory, Microbes and Industrial Processes
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.
Physical Chemistry of Bacterial Leaching 185
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
186 Biomining: Theory, Microbes and Industrial Processes
(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.
188 Biomining: Theory, Microbes and Industrial Processes
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.
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
Physical Chemistry of Bacterial Leaching
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
[Fe 3l-]US
rd'
lSS
=k--:-~
[H+ ]0.5 9. 21
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:
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.
-
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.
Physical Chemistry of Bacterial Leaching 193
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
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
194 Biomining: Theory, Microbes and Industrial Processes
... 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
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.
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
\: a k c )0.5 Nd\\Fe
fr[ 3+ (aq)]+[FeHS0 2+
4 ]
)0.5
Physical Chemistry of Bacterial Leaching 195
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
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.
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.
Physical Chemistry of Bacterial Leaching 197
[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
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
Biomining: Theory, Microbes and Industrial Processes
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.
Physical Chemistry of Bacterial Leaching 199
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·
200 Biomining: Theory, Microbes and Industrial Processes
Optimization
of Biooxidation Heaps
A. Ian M. Ritchie
Introduction
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
x
qw[Fe 3t
=--=--...;;;;...
1
F e S F
where
qW = specific discharge rate for the liquor (the irrigation rate);
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
10.6
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
Optimization of Biooxidation Heaps 205
t.>
~
208 Biomining: Theory, Microbes and Industrial Processes
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.
1.0
monod
0.8
a: 0.6
Q
.,>
~
a;
a: 0.4
linear
0.2
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.
210 Biomining: Theory, Microbes and Industrial Processes
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)
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
212 Biomining: Theory, Microbes and Industrial Processes
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.
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
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
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 .
214 Biomining: Theory, Microbes and Industrial Processes
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)
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
Optimization of Biooxidation Heaps 215
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.
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.
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.
Optimization of Biooxidation Heaps 217
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
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.
6
E
ECl 5
·S
::c 4
O~~~~~~~~~~~~
5.Ox10-7 1.0x10-8
7 month 6
6
E
E 5
.2>
Q)
::c 4
5.0x10-7 1.0x10-e
Table 10.3. The impact of varying Tsick and Tkill on the fraction of sulfide
sulfur oxidized
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.
thermistor strinq
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;
222 Biomining: Theory, Microbes and Industrial Processes
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
Optimization of Biooxidation Heaps 223
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.
224 Biomining: Theory, Microbes and Industrial Processes
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 &
Materials Society, 1993:437-448.
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,
Optimization of Biooxidation Heaps 225
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.
36. Ritchie AIM, Pantelis G. Optimization of oxidation rates in dump oxidation of
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
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
232 Biomining: Theory, Microbes and Industrial Processes
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
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
Mesophilic, Autotrophic Bioleaching Bacteria: Description, Physiology and Role 233
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.
234 Biomining: Theory, Microbes and Industrial Processes
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!
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~-
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.
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
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.
Mesophilic, Autotrophic Bioleaching Bacteria: Description, Physiology and Role 239
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
240 Biomining: Theory, Microbes and Industrial Processes
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
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.
5. Norris PRo Acidophilic bacteria and their activity in mineral sulfide oxidation.
In: Ehrlich HL, Brierley CL, eds. Microbial Mineral Recovery. New York: McGraw-
Hill, 1990:3-2.7.
6. Hallberg KB, Lindstrom EB. Characterization of Thiobacillus caldus sp nov., a
moderately thermophilic acidophile. Microbiology 1994; 140:3451-3456.
7. Goebel BM, Stackebrandt E. Cultural and phylogenetic analysis of mixed micro-
bial populations found in natural and commercial bioleaching environments. Appl
Environ Microbiol1994; 60:1614-1621.
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.
Biomining: Theory, Microbes and Industrial Processes
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.
14. Sand W, Rohde K, Sobotke B et al. Evaluation of Leptospirillum ferrooxidans for
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
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.
Appl Environ Microbiol 1992; 58:2227-2230.
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-
tification of Thiobacillus ferrooxidans, Thiobacillus thiooxidans and Leptospirillum
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
ferrooxidans, Leptospirillum ferrooxidans and Thiobacillus thiooxidans in leach-
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
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
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
250 Biomining: Theory, Microbes and Industrial Processes
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~.
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
252 Biomining: Theory, Microbes and Industrial Processes
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).
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.
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
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
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).
Thermophiles and Bioleaching 255
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.
Appl Environ Microbiol 1978; 36:523-525.
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.
Biomining: Theory, Microbes and Industrial Processes
13. 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.
14. Rawlings DE. Restriction enzyme analysis of 16S rRNA genes for the rapid iden-
tification of Thiobacillus ferrooxidans, Thiobacillus thiooxidans and Leptospirillum
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
Thermophiles and Bioleaching 257
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
Introduction
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
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
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).
Biomining: Theory, Microbes and Industrial Processes
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
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.
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.
266 Biomining: Theory, Microbes and Industrial Processes
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).
Heterotrophic Acidophiles in the Bioleaching of Sulfide Minerals
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
Heterotrophic Acidophiles in the Bioleaching of Sulfide Minerals 273
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,
274 Biomining: Theory, Microbes and Industrial Processes
~ /F,0,.
Other heterotrophic acidophiles ~ \ORG-A-N-I-C-C--',
~sulfide minerals
(MS,)
~ 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.
Heterotrophic Acidophiles in the Bioleaching of Sulfide Minerals 275
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.
Biomining: Theory, Microbes and Industrial Processes
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.
Heterotrophic Acidophiles in the Bioleaching of Sulfide Minerals 279
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
Introduction
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?
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).
Biomining: Theory, Microbes and Industrial Processes
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
Methods for Identification and Enumeration of Bioleaching Microorganisms 285
~ .
•• • ~
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
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).
288 Biomining: Theory, Microbes and Industrial Processes
----10 um
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.
-
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
Methods for Identification and Enumeration of Bioleaching Microorganisms 291
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.
Methods for Identification and Enumeration of Bioleaching Microorganisms 293
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.
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.
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.
Methods for Identification and Enumeration of Bioleaching Microorganisms 295
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.
Appl Environ Microbiol 1992; 58:2227-2230.
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
Environ Microbiol 1994; 60:3470-3473.
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
ferrooxidans, Leptospirillum ferrooxidans and Thiobacillus thiooxidans in leach-
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.
Methods for Identification and Enumeration of Bioleaching Microorganisms 297
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-
and iron-oxidizing eubacteria. J Bacteriol1992; 174:269-278.
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
Environ Microbiol 1993; 59:945-952.
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.
Appl Environ Microbiol 1990; 56:1919-1925.
52. Seeger M, Jerez CA. Phosphate limitation affects global gene expression in
Thiobacillus ferrooxidans. Geomicrobiol J 1992; 10:227-237.
Index
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