Published 1982

Chapter 3
Aqueous Pyrite Oxidation and the
Consequent Formation of
Secondary Iron Minerals
DARRELL KIRK NORDSTROMl

ABSTRACT
The oxidation of pyrite in aqueous systems is a complex biogeochemical pro-
cess involving several redox reactions and microbial catalysis. This paper reviews
the kinetic data on pyrite oxidation, compares available data on the inorganic vs.
microbial oxidative mechanisms and describes the occurrence of mineral products
resulting from pyrite oxidation. Although oxygen is the overall oxidant, kinetic
data suggests that ferric iron is the direct oxidant in acid systems and that tempera-
ture, pH, surface area, and the presence of iron and sulfur-oxidizing bacteria can
greatly affect the rate of reaction. The vast amount of literature on the microbial
and geochemical investigations on this subject have limited usefulness for under-
standing natural systems. Additional research is needed on the hydrologic, geo-
logic and microbiologic characteristics of field sites where oxidation occurs. The
acid water resulting from pyrite oxidation may precipitate a large suite of soluble
and insoluble iron minerals depending on pH, degree of oxidation, moisture con-
tent, and solution composition.

INTRODUCTION

The oxidation of common sulfide minerals such as pyrite plays a key

role in (a) the supergene alteration of ore deposits, (b) the formation of
acid mine waters, (c) the formation of acid sulfate soils, (d) the source and
distribution of dissolved sulfate in natural waters, (e) the source and dis-
tribution of heavy metals in the aquatic environment, and (f) the in situ
solution mining and dump leaching of heavy metals for economic re-
covery. Pyrite oxidation is a complicated process which includes several
types of oxidation-reduction reactions, hydrolysis and complex ion forma-
tion, solubility controls and kinetic effects. To interpret these reactions in

1 Department of Environmental Sciences, Clark Hall, Univ. of Virginia, Charlottesville,

VA 22903. Current address: Water Resources Division, U.S. Geological Survey, 345 Middle-
field Road, Menlo Park, CA 94025.

Copyright © 1982 Soil Science Society of America, 677 S. Segoe Rd., Madison, WI 53711.
Acid Sulfate Weathering.
37
38 NORDSTROM

natural aquatic systems requires the additional knowledge of microbial

ecology, hydrology, and geology. It is not surprising that important
aspects of this process are not clearly understood and many questions
remain unresolved.
The overall process describing pyrite oxidation is commonly given by
the following incongruent reaction:

in which pyrite and water, in the presence of oxygen, form insoluble fer-
ric hydroxide and sulfuric acid. Although the ultimate driving force is at-
mospheric oxygen, the fundamental mechanism and the major rate-
determining step(s) may not involve oxygen at all. Furthermore, many
other iron minerals may form in addition to or instead of ferric hydroxide.
In natural environments the relative importance of physical, chemical
and microbiological factors may vary widely. This paper (1) reviews the
known rates and suggested mechanisms for pyrite oxidation with some
discussion of the effects of geology, hydrology, and climate and (2) de-
scribes the occurrence of hydrated iron sulfate and iron oxide minerals
which result from pyrite oxidation. Illustrative examples are given from
acid mine drainage studies because more research has been done in this
area than any other. The conclusions, however, are just as applicable to
acid sulfate soils.
Investigations on pyrite oxidation usually center on the question of
whether an inorganic or microbiological mechanism is more important in
controlling the rate of the oxidative process. The relative importance of
these two mechanisms will dictate the preferred approach for "at-source"
control of acid mine drainage and acid sulfate soil formation.

INORGANIC OXIDATION MECHANISMS

The oxidation and hydrolysis shown by reaction 1 involves the loss of

1 electron by iron, 14 electrons by sulfur, and the gain of 7Y2 electrons by
oxygen per mole of pyrite. Also, 1 mole of iron is hydrolyzed and precipi-
tated. Clearly, all of these changes cannot take place in one step. Kinetic
studies and quantum mechanical calculations indicate that one-electron-
transfer reactions are most likely to occur, two-electron reactions much
less likely and more than two are highly unlikely (Basolo and Pearson,
1967). Furthermore, the oxidative half-cell reaction does not have to take
place at the same rate as the reductive half-cell. A maximum of 22 elec-
tron transfer reactions are possible, and there may be the same number of
possible rate-determining steps. More steps are possible if other oxidizing
agents are considered. Experimental measurements on this heterogeneous
system indicate that two electrons are frequently transferred and that all
22 steps do not need to be individually considered. The following discus-
sion will begin with pyrite in aqueous solution and sequentially consider
the loss of electrons from pyrite as well as the rates and mechanisms of
oxidation whenever data are available.
AQUEOUS PYRITE OXIDATION 39

Several investigators have shown that the first step in the breakdown
of a sulfide in an aqueous solution is the dissolution of the metal and the
oxidation of sulfide to neutral sulfur. High temperature ( -100 to 200 C)
metallurgical reactions consistently have the general stoichiometry:

[2]

for a divalent metal, M (Burkin, 1966; Wadsworth, 1973). Using pyrite as

an electrode, Sato (1960) made potentiometric measurements at 25 C
which indicated the initial step for pH ::5 3.0 was:

[3]

and this reaction was also found as a limiting condition in the experiments
of Garrels and Thompson (1960). Current-potential measurements on
pyrite by Biegler and Swift (1979) at 25 C and pH ::5 3.0 were interpreted
as a combination of

[4]

and

[5]

where reaction 5 is not an intermediate leading to reaction 4. In order to

speed up the reaction rate, Biegler and Swift (1979) had to make these
measurements at rather high potentials (0.95 to 1.4 V) which oxidized
dissolved iron to the ferric state. Lower potentials were achieved by
Bailey and Peters (1976) by increasing the temperature to 110 C, and in
these experiments there were significant amounts of ferrous iron. All of
these measurements were done in the absence of oxygen. In neutral to
basic solutions Sato (1960) proposed the reaction:

FeS2(S) + 2H 20(I) - Fe(OH)a(s) + SO + 3Htaq) + 3e- [6]

based on potentiometric measurements. The neutral sulfur produced in

reactions 5 and 6 would, at first, simply remain at the pyrite surface as
part of the structure while ferrous ions leached out from the surrounding
lattice sites. This sulfur-rich surface would eventually become unstable
and do one of two things: (1) disrupt and reorganize into elemental sulfur
(such as S8 rings) or (2) disrupt into solution as dimers which would be
highly unstable and would form sulfate rapidly in the presence of a strong
oxidizing agent. Electrochemical measurements support these hypotheses
(Peters, 1977; Biegler and Swift, 1979). Metal ions leach readily from sul-
fide minerals and for some the sulfur will remain behind in the original
structure. This residual sulfur behaves like elemental sulfur in that it dis-
solves in carbon disulfide (Sullivan, 1930). There is evidence for both the
presence and absence of elemental sulfur resulting from pyrite oxidation.
Clark (1966) has cited several instances of pyrite-derived sulfur from coal
40 NORDSTROM
seams. Presumably the organic matter retards the further oxidation of
sulfur. Smith et al. (1968), and Stokes (1901) found that elemental sulfur
was produced in the initial stages of oxidation using freshly-prepared
pyrite, whereas samples conditioned by prior oxidation did not produce
sulfur in the oxidation experiments. In highly oxidizing sulfide mineral
deposits little or no sulfur is found (Kinkel et al., 1956; Nordstrom, un-
published data). However, Bergholm (1955) has documented the produc-
tion of elemental sulfur from pyrite at low pH and in the presence of ferric
iron. In Bergholm's experiments most of the sulfur coated the pyrite as
elemental sulfur and retarded the oxidation. Brock et al. (1976) found
that elemental sulfur did not reduce ferric iron at a pH of 1.6 and
temperatures of 60 to 90 C. Decreasing temperature and decreasing pH
both led to increased sulfur production atlOO to 130 C in the experiments
of McKay and Halpern (1958). Also, McKay and Halpern (1958) have hy-
pothesized two independent reactions of the same stoichiometry as Biegler
and Swift (1979) which gives additional credance to the dual reaction
paths, one leading to the production of sulfur and the other to sulfate.
These observations suggest that the oxidation rate of free sulfur to sulfate
relative to the oxidation rate of sulfide-sulfur to free sulfur determines
whether free sulfur will occur or not. In turn, these rates will depend
upon the transport rate of the oxidizing agents to the pyrite surface, the
pH, and the temperature. The evidence cited indicates that these two
rates are somewhat comparable but that sulfide to sulfur is faster.
The oxidizing agent in waters of neutral pH is oxygen since the only
other likely oxidant, Fe3+, is highly insoluble. The reduction of oxygen on
pyrite proceeds through the following steps:

[7]

[8]

At pH values above 4, reaction 8 appears to be the rate-limiting step,

whereas at lower pH values the rate is independent of pH and the limiting
step is given by reaction 7 (Biegler et aI., 1977). This conclusion is also
verified by the rate data of Smith et al. (1968). The initial oxidation-re-
duction reaction of pyrite in water and oxygen results from combining re-
actions 3, 7, and 8:

[9]

Elemental sulfur formation is probably faster than peroxide formation for

a system containing pure pyrite in solutions of neutral pH where dif-
fusional processes are not rate-limiting. Therefore pyrite oxidation
kinetics and sulfur formation depends upon the availability and rate of
oxygen reduction. In soils and bedrock, the rate is probably related to the
rate of oxygen infiltration.
The three products of reaction 9 are unstable and rapidly oxidized or
decomposed. At neutral pH values ferrous iron oxidizes very rapidly ac-
cording to the expression:
AQUEOUS PYRITE OXIDATION 41

d[Fe(II)]1 dt = k[Fe(II)][ 02]/[H+)2 [10]

(Ghosh, 1974; Singer and Stumm, 1968; Stumm and Lee, 1961). Brackets
denote concentrations, k is the rate constant and t is the time. Equation
[10] holds for pH values above 4.5 and for poorly buffered solutions.
Ghosh (1974) pointed ot:.t that the oxidation rate increases with increasing
carbonate alkalinity and that highly buffered waters may differ by more
than an order of magnitude in their ferrous oxidation rates. In poorly
buffered perchlorate solutions and at a pH of 7.0, the rate constant is
about 102 day-I by extrapolating the data of Singer and Stumm (1968).
When sulfate is used instead of perchlorate the rate is different; Singer
and Stumm (1968) found that the rate decreased, whereas Huffman and
Davidson (1956) found the rate increased with increasing sulfate concen-
tration. The discrepancy must be resolved by additional measurements
over a wider range of concentrations.
A review of pyrite oxidation kinetics by Shumate et al. (1971) pOinted
out that the rate dependence on oxygen concentration differs somewhat
between investigators. They reported a rate expression based on an ad-
sorption equilibrium hypothesis: -

[11]

where r is the rate of pyrite oxidation in micromoles pyrite per hour per
gram of sample, k is the rate constant, KJ is the adsorption equilibrium
constant for oxygen on pyrite, K2 is the adsorption constant for an inert
gas, I, on pyrite. Equation [11] fits the data well but does not provide any
insight into the electrochemical nature of the reaction mechanism.
Once the iron is oxidized it will hydrolyze and precipitate as ferric
hydroxide:

or, if the hydrogen peroxide produced by reaction 7 is the oxidizing agent:

The presence of iron hydroxide on the surface of oxidizing pyrite has been
observed by Baker (1972) using Mossbauer spectroscopy.
It is interesting to note that if the sulfur produced in reaction 3 did
not further oxidize there would be no change in the pH of the solution
since the protons consumed in reaction 9 exactly balance the protons pro-
duced in reaction 12 or 13:

FeS2(S) + (3/4) O2 + (3/2) H20 - Fe(OH)a(s) + S~ [14]

When the sulfur is oxidized to sulfate, the pH drops:

[15]
42 NORDSTROM

Relatively little has been studied on the rates and mechanisms of reaction
15 with regard to pyrite oxidation, and this reaction could be an im-
portant rate-determining step in the initial weathering of pyrite to
produce acid conditions from neutral solutions. Six electrons are trans-
ferred per sulfur atom, and the formation of several intermediate sulfur
species such as thiosulfate, sulfite and polythionates are possible
complicating factors. Sulfite, SOl-, is not stable and rapidly oxidizes to
sulfate in the presence of oxygen or any other oxidizing agent. Thiosul-
fate, S203 2 -, is more stable at high pH than low pH and readily decom-
poses to elemental sulfur and sulfite (and hence to sulfate) in acid solu-
tions. Polythionates are more stable than thiosulfate or sulfite, especially
in acid media, but they too tend to breakdown by disproportionation to
simpler species such as sulfur and sulfate. Thiosulfate can be oxidized by
ferric iron to tetrathionate and the indication of thiosulfate formation on
air-oxidized sulfides by Steger and Desjardins (1978) leads to the hy-
pothesis that polythionates are important intermediates during aqueous
pyrite oxidation in neutral to acidic solutions. This hypothesis is verified
by the study of Nor and Tabatabai (1977) who reported that nearly all the
dissolved sulfur present during the oxidation of elemental sulfur in soils
was sulfate at a pH of 7.8, whereas tetrathionate was dominant at pH
values of 5 to 6. Goldhaber (1977, personal communication) found tetra-
thionate and sulfate to be present in equal proportions from pyrite oxida-
tion at a pH of 6 whereas thiosulfate was dominant at pH values of 8 and
9.
Reaction 15 is the initial acid-producing reaction which can reduce
the pH to about 4.5 where the rate of ferrous iron oxidation slows down
significantly (Singer and Stumm, 1968) and ferric hydroxide is more solu-
ble. Ferrous iron oxidation becomes independent of pH below a pH of 3
with a rate constant of 10-3 . 5 day-I. For example, in a 9 X 10-4 M ferrous
iron solution it takes about 5V2 days to oxidize 1 X 10-5 moles at a pH of 2,
a Po, of 0.2 atm and a temperature of 25 C.
As the ferric concentrations increase with the increased acidity, the
role of ferric iron becomes more important as an oxidizing agent. Meas-
urements by Garrels and Thompson (1960) and by Smith et al. (1968)
have shown that pyrite is rapidly oxidized by ferric iron in the absence of
oxygen and at low pH values according to the stoichiometry:

FeS2(S) + 14Fe3 + + 8H20 - 15Fe2+ + 2S0/- + 16W [16]

No residual sulfur was noted in these experiments which runs contrary to

the work of earlier investigators such as Bergholm (1955). From the data
of Garrels and Thompson (1960), 50% of a 2 X 10-3 m solution of ferric
sulfate was reduced by pyrite in 5 hours. Singer and Stumm (1969) carred
out similar experiments over a broader range of conditions and showed
the first order dependence on both Fe3 + and FeS2:

[17]

where [FeS2] represents the molar concentration of pyrite. Singer and

AQUEOUS PYRITE OXIDATION 43

Stumm (1969) calculated rate constants which ranged from 0.389 day-I to
17.4 day-I depending on the proportions of total Fe 3+ and FeS2 to the solu-
tion volume.
Figure 1 compares the oxidation rates of ferrous ion to ferric ion by
oxygen (III), pyrite to acid ferrous sulfate solution by oxygen (II), and
pyrite by ferric ion (I) as a function of pH. At low pH values (s 3.0) ferric
iron oxidizes pyrite much more rapidly than oxygen and more rapidly
than dissolved ferrous can be oxidized by oxygen. At neutral to alkaline
pH values the rate of ferrous oxidation rises rapidly, but the dissolved fer-
ric concentration also decreases greatly due to the precipitation of ferric
hydroxide. The results of these studies support the contention that pyrite
is initially oxidized by oxygen, and the pH consequently decreases
depending on the rate of oxidation of sulfur to sulfate. When the pH de-

..."""
I

"'C
~
ca
/1l' -
i/ ___
____-1-------- 8
'-'-1

-2
~

bI- 3

- 0
-4

-5

-6

pH
FeS, + 14Fe3+ + 8H,O - 15Fe'+ + 2S0~- + 16H+
(data from Singer and Stumm, 1969)
II FeS, + 7/20, + H,O - Fe'+ + 2S0~- + 2H+
(data from Smith, et aI., 1969)
III Fe'+ + 1/40, + H+ - Fe3+ + V2H,O
(data from Singer and Stumm, 1968)
Fig. 1. Comparison of rate constants as a function of pH for I. The oxidation of pyrite by
ferric iron, II. The oxidation of pyrite by oxygen and III. The oxidation of ferrous iron by
oxygen.
44 NORDSTROM

creases to 4.5, ferric iron becomes more soluble and begins to act as an
oxidizing agent, and below a pH of 3.0 it is the only important oxidizer of
pyrite. The presence or absence of oxygen makes no difference to the oxi-
dation rate by ferric iron (Singer and Stumm, 1969). Since pyrite can re-
duce ferric ion to ferrous faster than ferrous can be regenerated into ferric
by oxygen, the pyrite will simply reduce all the ferric ions and then the re-
action will stop. Thus, the ferrous to ferric oxidation has been called the
rate-determining step in the production of acid mine drainage (Singer and
Stumm, 1970). One of the strongest catalysts of this reaction is the iron-
oxidizing bacterium, T. jerrooxidans, which is known to increase the fer-
rous oxidation rate by five to six orders of magnitude (Lacey and Lawson,
1970; Singer and Stumm, 1970; Nordstrom, 1976). This increase makes
the ferrous oxidation rate, in the presence of the bacterium, comparable
or slightly greater than the pyrite oxidation by ferric iron and thereby
makes pyrite oxidation a rapid self-perpetuating process.
Other factors which can affect the inorganic oxidation rate include
temperature, surface area, the presence of impurities such as trace metals,
and the presence of other minerals such as chalcopyrite, sphalerite,
calcite, etc. These factors have been summarized by Shumate et al. (1971)
and Clark (1966). Temperature causes the oxidation rate by oxygen to
double for every 10 C rise. Pyrite can vary significantly in grain size and
morphology. Pyrite in coal deposits has been found in at least six different
forms, and the most reactive form is framboidal pyrite having pyrite cry-
stals less than a micrometer in size (Caruccio, Geidell, and Sewell, 1976).
The occurrence of framboidal pyrite has been used to estimate the acid-
forming potential of coal mine refuse (Caruccio, 1975). Studies cited by
Clark (1966) and Shumate et al. (1971) described "sulfur ball" pyrite as
having surface areas about an order of magnitude greater than museum
grade pyrite of the same diameter. Sulfur ball pyrite is assumed to be
equivalent to "framboidal" pyrite.
The occurrence of trace metals appears to have no significant effect
on the oxidation rate, but the co-existence of other sulfides such as chalco-
pyrite and sphalerite tends to decrease the oxidation rate. Sveshnikov and
Dobychin (1956) pointed out that rates of metal release from different sul-
fides are related to their electrode potentials and that a mixture of sulfides
releases more metals into solution and decreases the pH more than mono-
mineralic samples. Pyrite greatly increases metal release from other
sulfides while its own dissolution is reduced by galvanic protection. These
phenomena have been used to explain oxidation and supergene enrich-
ment in sulfide ore bodies by considering the mass of sulfides as a natural
galvanic cell (Sveshnikov and Ryss, 1964; Thornber, 1975). The presence
of relatively inert conductors such as graphite has also been proposed to
speed up the oxidation rate by increasing the electron flow between the
anodic and cathodic portions of an ore body (Cameron, 1979).
These additional factors mayor may not be important depending
upon the particular geological and climatological circumstances of the
oxidizing pyrite. The surface area effect seems to be very important in the
initial stages of acid sulfate production, but once the pH has decreased to
AQUEOUS PYRITE OXIDATION 45

3 or less it would seem to be less important. On the other hand, bacterial

oxidation could be the single most important factor over the whole range
of pH of natural waters.

MICROBIAL OXIDATION MECHANISMS

There are at least two well-established facts of microbial ecology: (1)

microbes are ubiquitous on this planet, existing under a wide range of
natural conditions (Kushner, 1978) and (2) microbes catalyze many reac-
tions including those related to the formation and oxidation of sulfide ore
deposits (Kuznetsov et aI., 1963; Zajic, 1969). Very few microorganisms
have been studied as much as the sulfur-oxidizing genus, Thiobacillus.
This bacterium utilizes some form of sulfur as an energy source and re-
duces CO 2 for a carbon source and is therefore termed chemoautotrophic.
Three species of Thiobacillus have been isolated from acid mine wastes:
T. jerrooxidans which oxidizes ferrous iron and pyrite as well as sulfur, T.
thiooxidans which oxidizes only sulfur and pyrite, and T. acidophilus
which is a facultative autotroph (grows on either inorganic or organic
substrates, Guay and Silver, 1975) oxidizing sulfur but not ferrous iron.
Thiobacillus acidophilus cannot oxidize pyrite unless it is in a mixed cul-
ture. These bacteria are acidophilic with optimal growth conditions
around a pH of 2 to 3, although they can survive up to pH values as high
as 6 and 7.
The importance of Thiobacillis in pyrite oxidation and the mechan-
ism of microbial attack has long been a controversial issue. Thiobacillus
jerrooxidans has been implicated as an essential catalyst in the production
of acid mine waters because (1) it can usually be isolated from acid mine
waters; (2) it actively accelerates pyrite oxidation in lab experiments; and
(3) it can speed up the oxidation of dissolved ferrous ions by five to six
orders of magnitude over the inorganic rate, thereby providing a way of
rapidly regenerating ferric ions in acid solutions. Direct evidence for
microbial growth in the deeper portions of the unsaturated zone, how-
ever, has not been demonstrated. Growth may occur in the surface layer
of a soil but some skepticism has been raised about the possibility of
growth deeper in the subsurface such as in underground mines or in coal
refuse piles. Kleinmann (1979) and Kleinman and Crerar (1979) have re-
futed these criticisms in an oxidation study of pyritic coals and over-
burden where they simulated the well-oxygenated surface layer, the un-
saturated zone above the water table and the saturated zone below the
water table. They found that T. jerrooxidans catalyzed pyrite oxidation
in the unsaturated zone for a few days following rainfall infiltrations as
well as in the surface layer.
There has been considerable discussion about the mechanism of
microbial pyrite oxidation. Thiobacillus jerrooxidans can rapidly gener-
ate ferric ions from ferrous ions in acidic media; pyrite is then directly and
inorganically oxidized by ferric iron. This process has been named the
"indirect contact" mechanism. Some investigators (Bryner and Jameson,
46 NORDSTROM

1958; Beck and Brown, 1968) have questioned this mechanism and have
proposed that the bacterium makes direct contact with the pyrite crystals
and oxidizes them through enzymatic pathways. The bacterial population
on the pyrite surface is much greater than in solution (Tuovinen and
Kelly, 1972), and it seems that T. thiooxidans oxidizes elemental sulfur by
direct contact. Thus, a direct contact mechanism has been suggested for
the microbial oxidation of pyrite. This mechanism is also indicated by
studies on the oxidation of non-ferrous sulfides such as sphalerite, chalco-
cite, and covellite in the presence of T. ferrooxidans (Silver and Torma,
1974; Duncan and Walden, 1972; Nielson and Beck, 1972) where iron is
not present in the system. Silverman (1967) attempted to determine which
mechanism was more important, and he concluded that both mechanisms
were operating concurrently. However, in every experiment the bacteria
were allowed to make direct contact with the pyrite grains, and in the ex-
periments without bacteria he failed to measure and compare the reaction
rates. A series of critical experiments were carried by Arkesteyn (1979) in
which he not only demonstrated the direct contact mechanism but also
showed that the rate of pyrite and elemental sulfur oxidation at a pH of
5.0 was faster than the oxidation rate of ferrous iron, all in the presence of
T. ferrooxidans. In addition, when Arkesteyn (1979) separated the
bacteria from pyrite by a dialysis bag the oxidation rate decreased. These
results confirmed other sets of experiments in which he individually in-
hibited the oxidation of ferrous iron or sulfur by the organism without
stopping the oxidation process.
A further point that tends to confuse the whole microbial picture is
the ability of T. thiooxidans and Sulfolobus acidocaldarius (a thermophil-
ic sulfur and iron-oxidizing bacterium) to reduce ferric iron during the
aerobic oxidation of elemental sulfur (Brock and Gustafson, 1976). Thio-
baccillus ferrooxidans also reduces ferric iron during sulfur oxidation
under anaerobic conditions while utilizing the oxidized iron as an energy
source. These data have very important implications concerning our
understanding of sulfide mineral oxidation in subsurface environments
such as in tailings piles, water-logged soils, and sulfide mineralization lo-
cated below the water table. First, it must be recognized that these iron
and sulfur-oxidizing bacteria are facultative with respect to their oxygen
requirements. Clearly the presence of a suitable electron acceptor such as
ferric ion will do just as well as oxygen. Thus, we can expect pyrite oxida-
tion to continue by microbial activity under low or even undetectable
oxygen concentrations. No one, however, has determined if the microor-
ganisms catalyze the anaerobic oxidation process. If they do, then clearly
mine sealing and flooding is an inappropriate method of preventing acid
mine drainage from abandoned mines.
Another source of controversy is the possible role of microorganisms
in the initial stages of pyrite oxidation. Thiobaccillus ferrooxidans can
play an active role once the pH has decreased to 4 to 4.5. How does a
neutral soil or ground water initially become acid from pyrite oxidation?
One possibility is the inorganic oxidation by molecular oxygen, but this
mechanism is quite slow. It is more likely that the initial step is microbial-
ly catalyzed by some of the heterotrophs and/or autotrophs which occur
AQUEOUS PYRITE OXIDATION 47

in the soil horizon and are known to metabolize inorganic sulfur com-
pounds (Roy and Trudinger, 1970). Walsh and Mitchell (1972a, 1972b)
have proposed that Metallogenium, an acid-tolerant iron-oxidizing bac-
terium, initiates a pH-dependent succession of bacteria. Metallogenium
optimally oxidizes iron between pH values of 3.5 and 5. Once Metallogen-
ium brings the pH down to about 4 then T. jerrooxidans takes over and
reduces the pH to below 4. In his studies on simulated coal refuse environ-
ments, Kleinman (1979) found that (1) inoculation with and without
Metallogenium made no difference in the rate of initial acidification and
(2) T. jerrooxidans not only survives at pH values up to 7, but it also initi-
ates pyrite oxidation to provide the acidity and ferrous ions. It should be
kept in mind that although the bulk solution phase may have a neutral
pH, the pH right at the surface of a pyrite grain may be considerably less
and may be the site location of viable Thiobacilli. He suggested that in the
initial stage of oxidation the direct contact mechanism was probably more
important since ferric ions are too low in concentration. Several other
species of Thiobacillus are also capable of performing the same function
as initiator (Sokolova and Karavaiko, 1968; Zajic, 1969) provided that
free sulfur is available from the pyrite. Arkesteyn (1980) found that the
presence of T. thiooxidans, thioparus, intermedius, and perametabolis
does not increase the pyrite oxidation rate over sterile blanks when start-
ing at pH values of 6.0. There seems to be a slight enhancement of the rate
when Arkesteyn (1980) used T. jerrooxidans but not by an amount greater
than the experimental uncertainty. These data seem contrary to his ex-
periments carried out at pH 5.0 in which microbial catalysis by direct
contact was evident. In addition, new evidence summarized by Kellyet
al. (1979) indicates that several other bacteria are important in pyrite
oxidation and that mixed cultures can be more effective in oxidizing
pyrite than single species cultures. Arkesteyn (1980) was also unable to
isolate Metallogenium from some acid sulfate soils. The conclusion from
all these investigations is that the initiation of pyrite oxidation in soils or
sulfide refuse piles probably occurs more rapidly by microbial catalysis
than by a purely inorganic mechanism and that T. jerrooxidans and
possibly other microorganisms which normally inhabitat the soil environ-
ment may catalyze these reactions. The potential role of Thiobacilli
species in initiating pyrite oxidation is another area of research worthy of
further investigation. The importance of the microbial vs. the inorganic
mechanisms for the initiation step is still subject to controversy.
If the rate-determining steps are controlled primarily by T. jerrooxi-
dans, then the oxidation rate depends upon the factors controlling the
bacterial growth kinetics such as the availability of oxidizing agents, car-
bon dioxide, and nutrients. Periodic rainwater infiltration provides the
oxygen and carbon dioxide requirements in the unsaturated subsurface
(Tuovinen and Kelly, 1972; Kleinmann, 1979). Regular flushing not only
provides the needed aeration, but it transports oxidation products away
from the reaction zone so that fresh pyrite surfaces are exposed. Nutrient
requirements (such as nitrate and phosphate) for T. jerrooxidans are not
limiting to growth because extremely small concentrations are quite suf-
ficient for growth. Thiobaccillus jerrooxidans grows in the absence of any
48 NORDSTROM

added N compound, and it has been suggested that it may fix atmospheric
N2 (Tuovinen and Kelly, 1972). Nitrogen fixation has recently been
demonstrated by Mackintosh (1978). Trace quantities of phosphate and
Mg are sufficient for growth. Sulfate is a requirement for iron oxidation
(Lazaroff, 1963), but it does not limit growth because it is always avail-
able as an oxidation product. Since the only important limiting factors are
oxygen and carbon dioxide, it follows that no growth occurs below the
oxygenated water table. Kleinman (1979) confirmed that there is no sig-
nificant growth in saturated environments, and field observations have
shown that when deep coal mines are flooded by mine sealing techniques
there is usually a significant, albeit incomplete, reduction of acidity
(Foreman, 1972). Complete inhibition of the oxidation process may never
be possible because of the availability of ferric iron as an alternate oxidiz-
ing agent.
Microbial catalysis of pyrite oxidation is an established fact, and the
reaction rate may be primarily determined by the growth kinetics of T.
jerrooxidans. However, recent data from Silver (1978) demonstrates that
the DNA base composition of T. jerrooxidans can vary depending on the
substrate used for growth. Thiobaccillus jerrooxidans may be a hetero-
geneous culture and other bacteria, as yet unidentified, may be more
crucial to the oxidation process. Alternatively, mixed cultures may have
symbiotic relationships which could produce different oxidation rates
than those obtained from pure cultures adapted to specific media. In
addition to oxygen or ferric iron concentrations or pH, rates should also
be expressed in terms of environmental factors such as hydrologic flow
path, residence time, pyrite surface area, temperature, climatic patterns,
and soil properties if they are to be applied to actual field situations.

ENVIRONMENTAL INFLUENCES

Pyrite oxidation investigators have emphasized the rate-determining

step in a very complex heterogeneous system. The results indicate that the
governing factor is the growth kinetics of iron and sulfur-oxidizing bac-
teria, chiefly T. jerrooxidans. Laboratory studies on batch or continuous
cultures (chemostat cultures), however, are not appropriate simulations
of natural systems. They are useful in obtaining rate data, comparing
rates and deriving mechanisms, but our understanding of natural systems
also requires knowledge of cyclic processes such as seasonal, diurnal, and
rainstorm influences. These processes strongly influence reactions by
regulating temperature, pH, oxygen, water and nutrient availability.
Additional information on microbial ecology and the environmental fac-
tors which influence growth are needed to determine the rate at which
pyrite oxidizes under natural conditions. Unfortunately, very little work
has been done in this area, and few field studies contain quantitative data
on the microbiology, mineralogy, geology, hydrology, and climate.
Reviews on the microbiology of acid mine waters (Lundgren et aI.,
1972; Nordstrom, 1977) cite many papers containing species
identification but none describing the relative proportions of species.
Dugan et al. (1970a, 1970b) and Nordstrom (1977) have described acid
AQUEOUS PYRITE OXIDATION 49

slime streamers which contain an abundance of both motile and non-

motile bacilli (see color Plate 1A). The slime-excreting bacterium has
not been isolated and its function in acid mine waters is not known.
Natural acid sulfate waters contain mixed cultures of many species and
several families of microorganisms whose growth rates may be quite dif-
ferent from lab growth rates with single species. An example of the inter-
dependency of mixed cultures is a study by Arkesteyn (1980b)2 on the as-
sociation of T. jerrooxidans and T. acidophilus. Apparently T. jerrooxi-
dans cannot grow in the presence of many common organic compounds
without the presence of T. acidophilus which can utilize those compounds
as an energy source. It is also interesting to note that T. acidophilus was
present in approximately equal numbers as T. jerrooxidans in all the cul-
tures which Arkesteyn (1980b)2 examined.
A study has been made on the dissolved ferrous ion oxidation rate in a
mountainous stream containing acid mine wastes (Nordstrom, 1976,
1977). The rate was found to be nearly identical to that determined for
optimal growth of T. jerrooxidans in 9K culture media. The data indi-
cates that the bacteria are growing at optimal rates in this dynamic
aqueous environment because iron oxidation can be directly related to cell
population growth (Silverman and Lundgren, 1959). After a rainstorm,
the oxidation rates in the stream decreases significantly due to dilution
and to the flushing effect of high flow. This study is at least one example
of the role which climate and hydrology can play.
The geologic structure and composition of the rock strata are very
important to the development of acid mine waters. Hollyday and
McKenzie (1973) have shown how the positioning of carbonate strata
with respect to oxidizing pyritic coals and shales can reduce the produc-
tion of acid waters. Parizek and Tarr (1972) give an excellent account of
several hydrogeologic techniques which can be used to reduce or prevent
acid mine drainage. These techniques include diverting water flow away
from the pyritic layers by pumping and diverting water through carbon-
ate strata so that they develop high alkalinity before infiltrating pyritic
strata. Geidel and Caruccio (1977) have further investigated the rate of
alkalinity accumulation compared to acid accumulation by water
infiltration through limestones, sandstones, and shales. Their results
demonstrate that more frequent infiltration tends to reduce acid develop-
ment. When longer dry periods are permitted there is a continual buildup
of pyrite oxidation products (soluble sulfate salts) which cause higher
acidity in the leachates when the next rainfall occurs. Alkalinity produc-
tion, however, tends to level off with time. These conclusions are contrary
to the work of Kleinman (1979) and Kleinman and Crerar (1979) who
showed that the activity of T. jerrooxidans and the acidity production de-
crease if the dry intervals between infiltrations are increased. The pro-
cesses occurring during infiltration are extremely important, and more re-
search on short-term and seasonal variability is needed.

2Arkesteyn, G. J. M. W. 1980b. Thiobacillus acidophilus: a study of its presence in Thio-

bacillus jerrooxidans cultures. Unpublished manuscript.
50 NORDSTROM

OXIDATION PRODUCTS: THE FORMATION OF HYDRATED

IRON MINERALS

A large number of iron minerals may be found in acid, unsaturated

pyrite-rich soils, on sulfide mine dumps, and on the surfaces of exposed
sulfide ores and pyritic shales. These minerals have a wide range of
properties from very insoluble iron hydroxides to very soluble iron sulfate
hydrates. The more soluble sulfates are most commonly formed during
dry periods as evaporation promotes the rise of subsurface waters to the
uppermost soil surfaces by capillary action. As these waters reach the
upper portions of the soil, they become progressively more concentrated
and finally precipitate various salts in an efflorescence. This phenomenon
is quite comparable to the buildup of efflorescent crusts associated with
closed basin lakes in arid environments (Eugster and Jones, 1979). The
formation of these efflorescent iron sulfates is an important intermediate
step preceding the precipitation of the more common insoluble iron
minerals such as goethite and jarosite.
Hydrated ferrous sulfate minerals frequently occur on the surface of
weathering pyrite where moisture is present. When conditions become
sufficiently dry, the dissolved ferrous and sulfate ions produced by reac-
tion 14 first reach saturation with respect to melanterite, FeS0 4 7H2 0.
0

Under continued dryness, melanterite dehydrates to either rozenite,

FeS04 4H 2 0, or szomolnokite, FeS04 o H2 0. The mineral ferrohexahy-
0

drite, FeS0 4 6H 2 0, is much less comm<m and has a very limited stability.
0

The pentahydrate, siderotil, requires copper in its structure to be stable.

If these minerals are still in contact with soil water or humid air and
warm temperatures, they oxidize to copiapite, Fe2 +Fel+(S04)s(OHh o
20 H2 0 (see color Plate 4C) which can be stable for long periods of time
when protected from rainfall or flowing water. Copiapite may contain
several other divalent and trivalent metals substituted into the ferrous and
ferric sites. It also has a very yellow color making it difficult to distinguish
from sulfur. From the author's experience this mineral is one of the most
abundant minerals found on sulfide mine tailings and on oxidizing sulfide
ore minerals.
The transformation to copiapite may proceed according to the reac-
tion:

5FeS04 7H 2 0
0 + O 2 + H 2S04 - Fe 2+Fel+(S04)6(OH) 2 20H2 0
0

+ 15H20 [18]

which indicates that lower pH values as well as partial oxidation is re-

quired. Copiapite may precipitate directly from acid sulfate waters, but
no field evidence for this has yet been found. Copiapite overgrowths on
melanterite, however, have been observed (Nordstrom and Dagenhart,
1978) and a specimen from Shasta County, California (Nordstrom, 1977)
has been reproduced in color Plate lB. The bulk of the specimen is
cuprian melanterite (blue). Small amounts of rozenite occur as a white
AQUEOUS PYRITE OXIDATION 51

Overall Stoichiometry

ACID MEDIA (pH~3) SLIGHTLY ACID TO BASIC MEDIA (pH~4)

Fe5 Z --- Fe Z+ + 5 Z + Ze- Fe5 Z + 3HZO - . Fe(OH)3 + 5 Z + 3H+ + 3e-

initiation { {
phase 0z + e- -- 0Z- 0z + Ze- + ZH+ - HZOZ

acid-generating phase: 5 Z + 30 Z + 2H 20 _ 2s0t + 4H+

catalytic phase: FeS2 + 14Fe 3+ + 8H 20 - 15Fe 2+ + 2S0~- + 16H+

Fig. 2. The major steps during the course of pyrite oxidation.

frosting and the small yellow crystals are copiapite. The sulfate hydrate
coquimbite, Fe2(S04)a·9H20, is intimately associated with copiapite in
oxidizing sulfide deposits. Melanterite, rozenite, szomolnokite, and copia-
pite are probably the most abundant efflorescences associated with oxi-
dizing coal deposits (Nuhfer, 1967).3 All of these sulfates are highly solu-
ble and may be at least partially responsible for the increased acidity and
dissolved solids load in receiving streams during rainstorm events (Nord-
strom and Dagenhart, 1978).
As the iron becomes fully oxidized in acid mine waters it eventually
reaches saturation with respect to either ferrihydrite 4 or jarosite,
KFe3(S04h(OH)6' Jarosite is stable at lower pH values than ferric hy-
droxide (Brown, 1971). Both of these minerals have been observed as pre-
cipitates in acid mine drainage (Nordstrom et aI., 1979) but are not stable
for more than a season. Color Plate 2A shows a bright yellow deposit of
jarosite precipitation from acid mine water. Fresh precipitates of iron in
acid mine drainage which produce the so-called "yellow boy" may in fact
be jarosite. As jarosite weathers or is exposed to dilute waters with higher
pH, it will gradually decompose to ferrihydrite or goethite. Mine tailings
may often be in equilibrium with both jarosite and ferrihydrite thereby
providing a buffer system (Miller, 1979). Miller (1979) has estimated the
pH of this buffer as 3.19 (± 0.17). Studies on acid mine drainage environ-
ments suggest that the soluble hydrated sulfates form during periods of
dry weather near the vicinity of oxidizing pyrite in unsaturated soil hori-
zons. Ferrihydrite, goethite, and jarosite are spatially distributed further
away from the pyrite and commonly form by precipitation from aqueous
systems such as the saturated zone or in receiving streams as in Plate 2A.

3Nuhfer, E. B. 1967. Efflorescent minerals associated with coal. M.S. Thesis. West Vir-
ginia University, Morgantown. 74 p.
• I have adopted the mineral name "ferrihydrite" here in place of ferric hydroxide, whether
amorphous or otherwise, following the recommendation of Schwertman and Taylor (1977)
and Schwertman (1979).
52 NORDSTROM

O2 ((71" Fe")

~ ~~ ":;/uQl,(LCt{fi

Fig. 3. The overall sequence of mineral reactions for pyrite oxidation showing the relation-
ships between oxidizing agents, catalysts, and mineral products.

SUMMARY

The oxidation of pyrite involves many electron transfer reactions

with the consequent reduction of oxygen in an aqueous environment and
ultimately produces ferric iron, sulfate, and high acidity. The key reac-
tions are summarized in Fig. 2 where the inorganic rate-limiting steps are
pH dependent for values greater than 4.0. Below a pH of 3.0 the oxidation
rates are independent of pH . The transition zone of pH = 3 to 4 includes
both steps. The first step is the initiation phase in which elemental sulfur
and ferrous iron are produced . At the higher pH values ferrous iron is
oxidized to ferric hydroxide. The acid-generating phase in which sulfuric
acid is produced from elemental sulfur follows next. Actual accumulation
of sulfur may depend upon the concentration and availability of oxidizing
agents such as sulfur-oxidizing bacteria. Once sufficient acid has been
generated (pH < 3.0), ferric iron rapidly continues the oxidation of pyrite
(catalytic phase). The most important catalyst is the iron-oxidizing bac-
terium, T. jerrooxidans, which greatly speeds up the oxidation of ferrous
to ferric thereby regenerating the supply of ferric irons in solution . The
scheme presented here is similar to the reaction sequence suggested by
AQUEOUS PYRITE OXIDATION 53

Kleinmann (1979). The oxidation products include several possible iron

sulfate, iron oxide, and iron hydroxide minerals whose genesis depends
upon water content, degree of oxidation, and time. The grand sequence
of reactants, products, and catalysts for pyrite oxidation is shown in the
schematic illustration of Fig. 3. This picture attempts to tie together the
roles played by (1) the oxidizing agents ferric iron and oxygen, (2) the
catalyzing agent T. ferrooxidans which utilizes oxygen for respiration,
and (3) the mineral products.

ACKNOWLEDGMENTS

I am grateful to Bob Kleinman and Dave Crerar for their helpful dis-
cussions on the chemistry of acid mine waters; to Don Thorstenson and
Martin Goldhaber for their reviews of the manuscript; and to Katherine
Baker, Carl Moses, and Aaron Mills for their stimulating ideas and sug-
gestions. However, the interpretations presented are solely my responsi-
bility. This manuscript was completed with the support of the U.S.
Geological Survey, Reston, the Virginia State Water Control Board and
NSF Grant EAR-7911144. I particularly thank Nico van Breemen for
bringing Arkesteyn's work to my attention and to both Delvin Fanning
and Nico for inviting me to rethink this whole problem through again.

LITERATURE CITED

1. Arkesteyn, C. J. M. W. 1979. Pyrite oxidation by Thiobacillusjerrooxidans with special

reference to the sulfur moiety of the mineral. Antonie van Leeuwenhoek J. Microbiol.
Serol. 45:423-435.
2. - - - - . 1980. Microbial processes in the formation of acid sulfate soils. Plant and Soil
(In press).
3. Bailey, L. K., and E. Peters. 1976. Decomposition of pyrite in acids by pressure leaching
and anodization: the case for an electrochemical mechanism. Can. Metall. Q. 15:333-
334.
4. Baker, R. A. 1972. Evaluation of pyritic oxidation by Mossbauer spectrometry. Water
Res. 6:9-17.
5. Basolo, F., and R. C. Pearson. 1967. Mechanisms of inorganic reactions: A study of
metal complexes in solution. 2nd ed. John Wiley and Sons, Inc., N.Y. 701 p.
6. Beck, J. V., and C. D. Brown. 1968. Direct sulfide oxidation in the solubilization of
sulfide ores by Thiobacillus jerrooxidans. J. Bacteriol. 96: 1433-1434.
7. Bergholm, A. 1955. Oxidation of pyrite. Jernkontorets Ann. 139:531-549.
8. Biegler, T., D. A. J. Rand, and R. Woods. 1977. Oxygen reduction on sulphide miner-
als. p. 291-302. In J. O·M. Bockris, D. A. J. Rand, and B. J. Welch (ed.) Trends in
electrochemistry. Plenum Press, N.Y.
9. - - - - , and D. A. Swift. 1979. Anodic behavior of pyrite in acid solutions. Electro-
chim. Acta 24:415-420.
10. Brock, T. D., S. Cook, S. Petersen, and J. L. Mosser. 1976. Biogeochemistry and bac-
teriology of ferrous iron oxidation in geothermal habitats. Ceochim. Cosmochim. Acta
40:493-500.
11. - - - - , and J. Gustafson. 1976. Ferric iron reduction by sulfur- and iron-oxidizing
bacteria. App. Environ. Microbial 32:567-571.
54 NORDSTROM
12. Brown, J. B. 1971. Jarosite-goethite stabilities at 25°C, 1 atm, Mineral. Deposita, 6:
245-252.
13. Bryner, L. C., and A. K. Jameson. 1958. Microorganisms in leaching sulfide minerals.
Appl. Microbiol. 6:281-287.
14. Burkin, A. R. 1966. The chemistry of hydrometallurgical processes. Van Nostrand,
Inc., N.J. 157p.
15. Cameron, E. M. 1979. Effect of graphite on the enhancement of surficial geochemical
anomalies originating from the oxidation of sulphides. J. Geochem. Expl. 12:35-43.
16. Caruccio, F. T. 1975. Estimating the acid potential of coal mine refuse. p. 197-205. In
M. J. Chadwick and G. T. Goodman (ed.) The ecology of resource degradation and
renewal. Blackwell Sci. Pub., London.
17. - - - - , G. Geidell, and J. M. Sewell. 1976. The character of drainage as a function
of the occurrence of framboidal pyrite and ground water quality in eastern Kentucky.
p. 1-16. In 6th Symp. Coal Mine Drainage Res., Louisville, Ky.
18. Clark, C. S. 1966. Oxidation of coal mine pyrite. J. Sanit. Eng. Div. Am. Soc. Civ.
Eng., 92 Proc. Paper 4802. p. 127-145.
19. Duncan, D. W., and C. C. Walden. 1972. Microbiological leaching in the presence of
ferric iron. Dev. Ind. MicrobioI.13:66-75.
20. Dugan, P. R., C. B. MacMillan, and R. M. Pfister. 1970a. Aerobic heterotrophic bac-
teria indigenous to pH 2.8 acid mine water: Microscopic examination of acid streamers.
J. Bacteriol. 101 :972-981.
21. - - - - , - - - - , and - - - - . 1970b. Aerobic heterotrophic bacteria indigenous
to pH 2.8 acid mine water: Predominant slime-producing bacteria in acid streamers.
J. Bacteriol. 101:982-988.
22. Eugster, H. P., and B. F. Jones. 1979. Behavior of major solutes during closed-basin
brine evolution. Am. J. Sci. 279:609-631.
23. Foreman, J. W. 1972. Evaluation of mine sealing in Butler County, Pennsylvania.
4th Symp. Coal Mine Drainage Res., Louisville, Ky. p. 83-95.
24. Garrels, R. M., and M. E. Thompson. 1960. Oxidation of pyrite in ferric sulfate solu-
tion. Am. J. Sci. 258:57-67.
25. Ghosh, M. M. 1974. Oxygenation of ferrous iron (II) in highly buffered waters. p. 193-
217. In A. J. Rubin (ed.) Aqueous-environmental chemistry of metals. Ann Arbor Sci-
encePub. Inc., Mich.
26. Geidel, G., and F. T. Caruccio. 1977. Time as a factor in acid mine drainage pollution.
7th Symp. Coal Mine Drainage Res., Louisville, Ky. p. 41-50.
27. Guay, R., and M. Silver. 1975. Thiobacillus acidophilus sp. nov.; isolation and some
physiological characteristics. Can. J. Microbiol. 21 :281-288.
28. Hollyday, E. F., and S. W. MacKenzie. 1973. Hydrogeology of the formation and
neutralization of acid waters draining from underground coal mines of western
Maryland. Dep. Nat. Res., Maryland Geol. Survey Rep. Invest. 20. 50 p.
29. Huffman, R. E., and N. Davidson. 1956. Kinetics of the ferrous iron-oxygen reaction in
sulfuric acid solution. J. Am. Chern. Soc. 78:4836.
30. Kelly, D. P., P. R. Norris, and C. L. Brierley. 1979. Microbiological methods for the ex-
traction and recovery of metals. p. 263-308. In Bull, Ellwood and Ratledge (ed.)
Microbial technology. Soc. Gen. Microbiol.
31. Kinkel, A. R., Jr., W. E. Hall, and J. P. Albers. 1956. Geology and base-metal deposits
of West Shasta copper-zinc district Shasta County, California. U.S. Geol. Survey Prof.
Paper 285. 156 p.
32. Kleinmann, R. L. P. 1979. The biogeochemistry of acid mine drainage and a method to
control acid formation. Ph.D. Thesis. Princeton Univ., Princeton. 104 p. (Diss. Abstr.
7919776).
33. - - - - , and D. A. Crerar. 1979. Thiobacillus ferrooxidans and the formation of
acidity in simulated coal mine environments. Geomicrobial. J. 1:373-388.
34. Kushner, D. J. (ed.). 1978. Microbial life in extreme environments. Academic Press,
N.Y. 465p.
AQUEOUS PYRITE OXIDATION 55

35. Kuznetsov, S. I., M. V. Ivanov, and N. N. Lyalikova. 1963. C. H. Oppenheimer (ed.)

Introduction to geological microbiology. McGraw-Hill Book Co., N. Y. 252 p.
36. Lacey, D. T., and F. Lawson. 1970. Kinetics of the liquid-phase oxidation of acid fer-
rous sulfate by the bacterium Thiobacillus ferrooxidans. Biotech. Bioeng. 12:29-50.
37. Lazaroff, N. 1963. Sulfate requirement for iron oxidation by Thiobacillus ferrooxidans.
J. Bacterio!' 85:78-83.
38. Lundgren, D. G., J. R. Vestal, and F. R. Tabita. 1972. The microbiology of mine drain-
age pollution. In R. Mitchell (ed.) Water pollution microbiology. Wiley-Interscience,
N.Y. 147p.
39. Mackintosh, M. E. 1978. Nitrogen fixation by Thiobacillus ferrooxidans. J. Gen.
Microbio!. 105:215-218.
40. McKay, D. R., and J. Halpern. 1958. A kinetic study of the oxidation of pyrite in
aqueous suspension. Trans. Metal!. Soc. AIME 212:301-309.
41. Miller. S. D. 1979. Chemistry of a pyritic strip-mine spoi!. Ph.D. Dissertation. Yale
University. New Haven. 189 p. (Diss. Abstr. 7926660).
42. Nielson, A. M., and J. V. Beck. 1972. Chalcocite oxidation and coupled carbon dioxide
fixation by Thiobacillus ferrooxidans. Science 175: 1124-1126.
43. Nor, Y. M., and M. A. Tabatabai. 1977. Oxidation of elemental sulfur in soils. Soil Sci.
Soc. Am. J. 41:736-741.
44. Nordstrom, D. K. 1976. Kinetic and equilibrium aspects of ferrous iron oxidation in acid
mine waters. Abstract. Geo!. Soc. Am. Ann. Mtg., Denver, Colo.
45. - - - - . 1977. Hydrogeochemical and microbiological factors affecting the heavy
metal chemistry of an acid mine drainage system. Ph.D. Dissertation. Stanford Univ-
versity. Stanford. 210 p. (Diss. Abstr. 7718232).
46. - - - - , and T. V. Dagenhart. 1978. Hydrate iron sulfate minerals associated with
pyrite oxidation: Field relations and thermodynamic properties. Abstract. Geo!. Soc.
Am. Ann. Mtg., Toronto, Canada.
47. - - - - , E. A. Jenne, and J. W. Bal!. 1979. Redox equilibria of iron in acid mine
waters. In E. A. Jenne (ed.) Chemical modeling in aqueous systems: Speciation, sorp-
tion, solubility and kinetics. Am. Chern. Soc. Symp. Series 93:51-79.
48. Parizek, R. R., and E. G. Tarr. 1972. Mine drainage pollution prevention and abate-
ment using hydrogeological and geochemical systems. p. 56-82. In 4th Symp. Coal Mine
Drainage Res., Pittsburgh. Pa.
49. Peters,E. 1977. The electrochemistry of sulphide minerals. p. 267-290. In J. O'M.
Bockris, D. A. J. Rand, and B. J. Welch (ed.) Trends in electrochemistry. Plenum Press,
N.Y.
50. Roy, A. B., and P. A. Trudinger. 1970. The biochemistry of inorganic compounds of
sulphur. Cambridge University Press. 400 p.
51. Sato, M. 1960. Oxidation of sulfide ore bodies, II. Oxidation mechanisms of sulfide
minerals at 25°C. Econ. Geo!. 55:1202-1231.
52. Schwertman. U. 1979. Is there amorpholls iron oxide in soils? Abstract. Ann. Mtg. Soil
Sci. Soc. Am .. Ft. Collins. Colo.
5.3. - - - - . and R. M. Taylor. 1977. Iron oxides. p. 145-180. III J. B. Dixon and S. B.
Weed (ed.) Minerals in soil environments. Soil Sci. Soc. Am., Madison. Wis.
54. Shumate, K. S., E. E. Smith, P. R. Dugan, R. A. Brant, and C. I. Randles. 1971. Acid
mine drainage formation and abatement. E.P.A. Rep. 14010 FPR 04171.
55. Silver, M. 1978. Metabolic mechanisms of iron-oxidizing Thiobacilli. p. 3-17. In Matal-
Murr, A. E. Torma, and J. A. Brierley (ed.) Metallurgical applications of bacterial
leaching and related microbiological phenomena. Academic Press, N. Y.
56. Silver, M., and A. E. Torma. 1974. Oxidation of metal sulfides by Thiobacillu~ferroxi­
dam grown on different substrates. Can. J. Microbio!. 20:141-147.
57. Silverman, M. P. 1967. Mechanism of bacterial pyrite oxidation. J. Bacterio!' 94:1046-
1051.
56 NORDSTROM
58. Silverman, M. P., and D. G. Lundgren. 1959. Studies on the chemoautotrophic iron
bacterium Ferrobacillusjerrooxidans. J. Bacterio!. 77:642-647.
59. Singer, P. C., and W. Stumm. 1968. Kinetics of the oxidation of ferrous iron. Second
Symp. Coal Mine Drainage Res., Pittsburgh, Pa. 12-34.
60. - - - - , and ----.1969. Oxygenation of ferrous iron. FWQA Rep. 14010-06/69.
61. - - - - , and - - - - . 1970. Acid mine drainage: the rate determining step. Science
167:1121-1123.
62. Smith, E. E., K. Svanks, and K. S. Shumate. 1968. Sulfide to sulfate reaction studies.
2nd Symp. Coal Mine Drainage Res., Pittsburgh, Pa. p. 1-11.
63. Sokolova, G. A., and G. I. Karavaiko. 1968. Physiology and geochemical activity of
Thiobacilli (Russian trans.), Israel Prog. for Scientific Trans!. 283 p.
64. Steger, H. F., and L. E. Desjardins. 1978. Oxidation of sulfide minerals, 4. Pyrite, chal-
copyrite and pyrrhotite. Chern. Geo!. 23:225-237.
65. Stokes, H. N. 1901. On pyrite and marcasite. U.S. Geo!. Survey Bull. 186.
66. Stumm, W., and G. F. Lee. 1961. Oxygenation of ferrous iron. Ind. Eng. Chern. 53:
143.
67. Sullivan, J. D. 1930. Chemistry of leaching chalcocite. U.S. Bur. Mines. Tech. Paper
473.24 p.
68. Sveshnikov, G. B., and S. L. Dobychin. 1956. Electrochemical solution of sulfides and
dispersion aureoles of heavy metals. Geokhimiya 1956. No.4. p. 413-419.
69. - - - - , and Yu. S. Ryss. 1964. Electrochemical processes in sulfide deposits and their
geochemical significance. Geochem. Int!. p. 198-204.
70. Thornber, M. R. 1975. Supergene alteration of sulphides. I. A chemical model based on
massive nickel sulphide deposits at Kambalda, Western Australia. Chern. Geo!. 15: 1-14.
71. Tuovinen, O. H., and D. P. Kelly. 1972. Biology of Thiobacillus jerrooxidans in relation
to the microbiological leaching of sulphide ores. Z. Allg. Mikrobio!. 12:311-346.
72. Wadsworth, M. E. 1973. Kinetics of heterogeneous systems. Ann. Rev. Phys. Chern. 23:
355-384.
73. Walsh, F., and R. Mitchel!. 1972a. An acid-tolerant iron-oxidizing Metallogenium.
J. Gen. Microbio!. 72:369-376.
74. - - - - , and - - - - . 1972b. A pH-dependent succession of iron bacteria. Environ.
Sci. Techno!. 6:809-812.
75. Zajic, J. E. 1969. Microbial biogeochemistry. Academic Press, N.Y. 345 p.
Plate 1. A. Slime-excreting bacteria growing along the stream bottom of a creek containing
acid mine drainage in Shasta Co., California. These "acid slime streamers" contain very
dense populations of bacilli enmeshed in a fibrillar polymer network similar to those found
by Dugan, MacMillan and Pfister (1970a). Distance lengthwise across area shown by
photo is about 0.8 m (See Chapter 3) . Plate 1. B. Specimen of efflorescent growth contain-
ing copiapite (yellow) overgrowths on rozenite (white) which has dehydrated from cuprian
melanterite (blue) . This specimen measures about 1 cm across. (See Chapter 3) .
Plate 2. A. Jarosite (yellow) precipitation from an acid mine water (pH 2) in Shasta Co., Cali-
fornia . Grey color on stream bed is a coating of bacterial slime. Scale is provided by a ring
of keys in center foreground. (See Chapter 3) . Plate 2. B. Face of lignite surface mine
showing oxidized (about upper 115 of exposed soil-geolOgiC column) and reduced zones of
the pre-mining column. (See Chapter 10).
Plate 3. A. Soil profile of a Hapludult in Maryland with jarosite (pale yellow) and iron
"oxide" mottles in its C horizons, which commence at a depth of about 60 cm. Depth incre-
ments on tape are in decimeters. The soil was on a slope of about 30 % where the profile oc-
curred. The area is presently forested, but it apparently was once a severely eroded pasture.
(See Chapter 7). Plate 3. B. Closer view of jarosite and iron "oxide" mottles at depth of
from 70 to 120 cm in C horizons of profile shown in Plate 3A. Matrix materials are
glauconitic "greensands". (See Chapter 7). Plate 3. C. Soil profile (Profile 1 of Wagner
et al.) with a sulfuric horizon (0-30 cm) at its surface, formed on a scalped (by man) land
surface in a highway cloverleaf at Central Avenue and the Washington Beltway, Prince
Georges County, MD. Depth increments on tape are in decimeters. Jarosite and iron
"oxide" mottles are evident in the sulfuric horizon. Dark, sulfide bearing glauconitic ma-
terial below 30 cm contains (white) calcareous fossil shell fragments. (See Chapter 7). Plate
3. D. Close up of barite as nodular, white crystalline masses in soil matrix at a depth of
about 1 m in an Albaqualf near College Station, TX. Finger gives scale. (See Chapter 8).
Plate 4. A. Jarosite (yellow) and "free iron oxides" (yellowish brown) on plate faces of hand
specimen of oxidized zone of Eocene shale collected from scalped land surface (cut) at Col-
lege Station, TX . (See Chapters 8 and 10). Plate 4. B. Hand specimen gypsum collected
from surface of spoil at text pit for San Miquellignite mine in Yequa formation in Atascosa
County, TX . (See Chapters 8 and 10) . Plate 4. C. Copiapite (lemon yellow) efflorescence
on hand specimen of underclay of coal seam. Specimen collected beneath overhanging rock
in roadcut on Interstate 64 near Cannonsburg, KY. (See Chapters 3, 7, and 10). Plate 4. D.
Rozenite (white) occurring on coal hand specimen. Specimen collected a few cm above
specimen shown in C. (See Chapter 10) . Plate 4. E. Hand specimen of siderite (pale yellow
zone shown in broken face) weathered to a crust of goethite on surfaces. Specimen col-
lected from a larger mass exposed in a scalped land surface (cut) in Eocene shale at College
Station, TX. (See Chapter 10) . Plate 4. F . Rozenite (white efflorescence) at surface of
lignite mine spoil in Texas . (See Chapter 10).