Reductive capacity of natural reductants

Article abstract of DOI:10.1021/es025830m

The reductive capacities of soil minerals and Silawa soil for Cr(VI) and chlorinated ethylenes were determined and characterized to understand in situ treatment using these natural reductions. The reductive capacity of soil minerals for Cr(VI) was 3-16 times greater than that for tetrachloroethylene (PCE), indicating that Cr(VI) is more susceptible to the reduction by soil minerals than PCE. Green rust (GR_SO4) showed the greatest reductive capacity for Cr(VI) and PCE followed by magnetite, pyrite, biotite, montmorillonite, and vermiculite. The major transformation product in pyrite and GR_SO4 suspensions was acetylene rather than dichloroethylene (DCE) and vinyl chloride (VC). For VC degradation, ethylene was the main transformation product with a low concentration of ethane observed. Fe(II) content in soil minerals was directly proportional to the reductive capacity of soil minerals for Cr(VI) and PCE, suggesting that Fe(II) content is an important factor that significantly affects reductive transformations of target contaminants in natural systems.

Full text of DOI:10.1021/es025830m

Environ. Sci. Technol. 2003, 37, 535-541
have different transport characteristics. For example, hexa-
Reductive Capacity of Natural
Reductants
valent chromium (Cr(VI)) can be converted to the less mobile
and less toxic trivalent form (Cr(III)) by suboxic aquifer
materials (19). PCE, one of the most common chlorinated
solvents found in contaminated environments, can be
reductively transformed by microbial processes to more toxic
daughter compounds such as TCE, dichloroethylenes (DCEs),
VC, or to harmless C2 hydrocarbons (28). Many effective and
cost-efficient remedial technologies have been developed
that are based on redox chemistry. Natural attenuation and
in-situ redox manipulation (28, 29) are examples of attractive
remediation techniques because they transform contami-
nants using natural reductants and indigenous microorgan-
isms. The natural reductants can cause abiotic reductive
degradation without extensive changes at the contaminated
site. Some of the most abundant natural reductants are soil
minerals that contain reduced forms of iron and sulfur. Such
soil minerals can significantly increase the transformation
rate of chlorinated organics in heterogeneous suspension
with HS- and Fe(II) and can also reductively transform
organic and inorganic contaminants by themselves (7, 24,
30-33).
The term “reductive capacity” has been used in different
ways to describe systems in which redox reactions occur.
One approach is to define the reductive capacity analogously
to the definition of buffer capacity, i.e., the derivative of the
concentration of reductant with respect to Eh (34). As such,
this definition measures the extent to which the redox
potential is buffered in a system. An alternative is to define
reductive capacity as the amount of an oxidant that can be
reduced when sufficient time is given so that the reaction
proceeds to its maximum extent. This definition is similar to
that of reductive strength (35, 36). However, this definition
describes the potential amount of reduction achievable in
a system and is attractive for use in remediation activities,
so it is the definition that will be used in this paper.
W O O J I N L E E * , † A N D B I L L B A T C H E L O R ‡
Environmental Science Research Center,
School of Public and Environmental Affairs,
Indiana University, Bloomington, Indiana 47405, and
Department of Civil Engineering, Texas A&M University,
College Station, Texas 77843
Reductive capacities of soil minerals and soil for Cr(VI)
and chlorinated ethylenes were measured and characterized
to provide basic knowledge for in-situ and ex-situ treatment
using these natural reductants. The reductive capacities
of iron-bearing sulfide (pyrite), hydroxide (green rust; GRSO4),
and oxide (magnetite) minerals for Cr(VI) and tetrachloro-
ethylene (PCE) were 1-3 orders of magnitude greater
than those of iron-bearing phyllosilicates (biotite, vermiculite,
and montmorillonite). The reductive capacities of surface
soil collected from the plains of central Texas were similar
and slightly greater than those of iron-bearing phyllosilicates.
The reductive capacity of iron-bearing soil minerals for
Cr(VI) was roughly 3-16 times greater than that for PCE,
implying that Cr(VI) is more susceptible to being reduced by
soil minerals than is PCE. GRSO4 has the greatest reductive
capacity for both Cr(VI) and PCE followed by magnetite,
pyrite, biotite, montmorillonite, and vermiculite. This order
was the same for both target compounds, which indicates
that the relative reductive capacities of soil minerals are
consistent. The reductive capacities of pyrite and GRSO4 for
chlorinated ethylenes decreased in the order: trichloro-
ethylene (TCE) > PCE > cis-dichloroethylene (c-DCE) > vinyl
chloride (VC). Fe(II) content in soil minerals was directly
proportional to the reductive capacity of soil minerals
for Cr(VI) and PCE, suggesting that Fe(II) content is an
important factor that significantly affects reductive trans-
formations of target contaminants in natural systems.
Barcelona and Holm suggested that the oxidative and
reductive capacity of soils could be related to the concentra-
tions of labile chemicals in the soil (37). Heron et al.
introduced Ti3+-ethylenediaminetetraacetic acid extraction
method for the determination of the oxidation capacity of
aquifer sediments (38). Kozuh and Schara developed a
method for rapid characterization of soil reductive capacity
using electron paramagnetic resonance (EPR) spectroscopy
(39). Fruchter et al. used the capability of reduced soil to
consume oxygen as a measure of its reductive capacity with
respect to chromium (29). Lee et al. developed a standard
procedure to test the reductive capacity of soil for Cr(VI) by
measuring the amount of Cr(VI) reduced per unit mass of
soil (40). The reductive capacity of soil for Cr(VI) may provide
an estimate of the reductive capacity of soil for other target
contaminants. Knowing the reductive capacity of soil miner-
als for a variety of target contaminants would be useful for
developing remedial technologies based on redox reactions
by providing a means to estimate the potential for contami-
nant transformation. It could also improve understanding
of transformation mechanisms, particularly identifying biotic
and abiotic processes. However, no significant efforts have
been made to measure and characterize the reductive
capacity of soil minerals for various target contaminants.
The objectives of this research were to measure and compare
reductive capacities of natural reductants for different types
of chemical compounds and to identify factors that signifi-
cantly affect them. Six soil minerals (pyrite, GRSO4, magnetite,
biotite, vermiculite, and montmorillonite) and soil (Silawa
soil) were chosen for study because of their importance in
Introduction
A number of anthropogenic chemicals that have contami-
nated soil and water resources are relatively susceptible to
reductive transformation in natural environments (1-3). It
has been shown that chlorinated solvents (4-8), organo-
chlorine and carbamate pesticides (3, 9-14), and nitro-
aromatic compounds (15-18) are reductively degraded in
natural and artificially induced reducing environments.
Inorganic contaminants such as chromium (19-21), arsenic
(22, 23), and nitrate (24, 25) and nuclear wastes such as
uranium (26) and plutonium (27) have also been reported
to undergo reductive transformations.
Redox reactions play a crucial role in the mobility,
transport, and fate of both inorganic and organic contami-
nants in natural and engineered systems because they can
convert contaminants among different oxidation states that
* Corresponding author phone: (812)855-8486; fax: (812)855-1881;
e-mail: woojlee@indiana.edu.
† Indiana University.
‡ Texas A&M University.
9
10.1021/es025830m CCC: $25.00
Published on Web 12/19/2002
2003 Am erican Chem ical Society
VOL. 37, NO. 3, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 5 3 5

TABLE 1. Characteristics of Soil Minerals
iron contenta
(Fe(II)/Fe(III), (mg/g))
surface areab
(m2/g)
particle size
soil minerals
pyrite
green rust (GRSO4
m agnetite
biotite
verm iculite
formula
pH
pzc
(µm)
FeS2
94.8/120
464/212
206/516
114/3.1
14.2/42.5
1.2/6.3
27.8
86.3
57.2
1.9
26.7
488
1.2-2.5c
nad
63-250
<63
)
FeII4FeIII2(OH)12SO4‚yH2O
Fe3O4
6.4-6.9e
na
<63
K(Mg,Fe2+)3(Si3Al)O10(OH)2
Mg0.33(Mg,Al,Fe3+)3(Si3Al)O10(OH)2
Nax[(Al2-xMgx)Si4O10(OH)2]
63-250
63-250
<63
na
m ontm orillonite
2.5f
a
b
The extraction m ethod is explained in analytical procedures. Surface area m easured by ethylene glycol m onoethyl ether (EGME) m ethod.
Ref 4. na, not available. Ref 55. f Ref 56.
c
d
e
TABLE 2. Characteristics of Silawa Soil
Section A: Particle Size Distribution
sand
total
silt
total
clay
total
sand
silt
clay
size (m m )
content (%) 0.3
2.0-1.0 1.0-0.5 0.5-0.25 0.25-0.10 0.10-0.05 2.0-0.05 0.02-0.002 0.05-0.002 <0.0002 <0.002
0.6
15.5
46.0
18.7
81.1
5.5
12.2
3.7
6.7
Section B: Characteristics
b
surface area (m2/g)a
11.67
organic carbon (%)
Fe(II) (mg/g)
Fe(III) (mg/g)
pH
0.69
1.2
5.06
6.1
a
b
Surface area m easured by EGME m ethod. Ref 57.
natural environments. Cr(VI) and a series of chlorinated
ethylenes (PCE, TCE, c-DCE, and VC) were chosen as oxidants
for this research.
were ground with a grinder in the anaerobic chamber. Pyrite
(30 g) was then washed with 30 mL of 1 M HCl twice for 2
min and rinsed with 60 mL of ddw several times until the pH
of the supernatant was greater than 6.7 in order to remove
oxidized material from its surface. Each 30-g portion of clay
mineral (biotite, vermiculite, and montmorillonite) was
pretreated with 200 mL of 0.1 M dithionite solution for 4
days and rinsed several times with 200 mL of ddw to remove
residual dithionite. The pretreatment with dithionite provided
a method of converting oxidized components of the minerals
to their reduced forms so that the full reductive capacity of
the soil minerals could be measured (40). GRSO4 was prepared
by a method modified by Koch and Hansen (25). The dark
blue-green precipitates synthesized were washed several
times with ddw until no Fe(II) was detected in the super-
natant. Magnetite was synthesized by modifying the method
developed by Taylor et al. (42). The black precipitates formed
were washed several times with ddw to remove Fe(II). The
soil minerals were then freeze-dried, dry-sieved, and stored
in the anaerobic chamber, except GRSO4 which was dried in
the chamber. Two size fractions (63-250 and <63 µm) were
used for this research.
Experimental Section
Anaerobic Environm ents. To simulate anaerobic conditions
in natural reducing environments, all reagents and samples
were prepared in an anaerobic chamber (Coy Laboratory
Products Inc.) containing a mixed gas atmosphere (95%
nitrogen and 5% hydrogen). A colorimetric redox indicator
solution (resazurin, 89%, Aldrich) was used to ensure that
anaerobic conditions were maintained (8). The solution is
colorless when the redox potential is less than ∼218 mV at
pH 9 but turns pink as the redox potential (oxygen level)
increases (41). Aqueous solutions and chemicals sensitive to
redox reaction were deoxygenated in an airlock (Coy
Laboratory Products Inc.) and kept in the anaerobic chamber.
Chem icals. Hexavalent chromium and chlorinated or-
ganics used as target compounds were ACS or higher grades.
These included K2CrO4 (99%, Sigma), PCE (99.9%, Sigma),
TCE (99.6%, Sigma), c-DCE, (97.0%, Sigma), trans-dichlo-
roethylene (t-DCE, 98%, Sigma), 1,1-dichloroethylene (1,1-
DCE, 99.0%, Sigma), and VC (20,000 mg/ L, Sigma). Ethane,
ethylene, and acetylene (99.0%, 1.0%, and 1000 mg/ L,
respectively) were purchased from Scott Specialty Gases and
used as standard gases for the analysis of non-chlorinated
daughter products. Stock solutions (1 M) of PCE, TCE, c-DCE,
and VC were prepared by diluting them in methanol (99.8%,
HPLC grade, EM). Deaerated deionized water (ddw) was
prepared by deoxygenating 18 MΩ‚cm deionized water with
99.99% nitrogen for 2 h and then by deoxygenating with the
mixed gases in the anaerobic chamber for 12 h. A stock
solution of dithionite (0.1 M) was prepared from Na2S2O4
(88%, Sigma) and used within 2 days. A buffer solution (10
mM) was prepared from NaHCO3 (100.3%, Sigma).
Samples of a Silawa soil were collected from a ranch at
Texas A&M University, College Station, TX. The soil is an
alluvium of the Brazos River and is classified as siliceous,
semi-active, thermic, and fine loamy sand (Alfisols) (43). The
samples were dried in the air for 14 days and screened with
a 0.425-mm sieve. The soil samples (120 g) were equilibrated
with the anaerobic atmosphere for 2 days and reduced by
250 mL of 0.1 M dithionite solutions for 4 days (40). The
pretreated soil samples were washed several times with ddw
to remove residual dithionite (40) and were dried and stored
in the anaerobic chamber. All soil minerals and soil prepared
were used for experiments no later than 7 days after
pretreatment and synthesis to preclude an aging effect (4)
on reductive degradation of target compounds. Some
characteristics of soil minerals and soil used for this research
are summarized in Tables 1 and 2.
Preparation of Soil Minerals and Soil. Pyrite (Zacatecas,
Mexico), biotite (Bancroft, Canada), and vermiculite (Trans-
vaal, Africa) were purchased from Ward’s (Rochester, NY),
and montmorillonite (Gonzales, TX) was obtained from the
Clay Minerals Society (Columbia, MO). Pyrite was ground
with ceramic mortar and pestle, and biotite and vermiculite
Experim ental Procedures. Reductive capacities for Cr(VI)
of soil minerals and soil were measured by modifying the
basic procedure for measuring soil reductive capacity that
9
5 3 6 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 3, 2003

was developed by Lee et al. (40). The effect of Cr(VI) contact
time was investigated for soil minerals and soil used for this
research. Amber borosilicate glass vials (nominally 20 mL,
Kimble) with an open-top screw cap and three-layered
septum system (2 mil PTFE film (Norton Performance Plastics
Co.), lead foil (3M), and PTFE film lined rubber septum
(Kimble)) were used as batch reactors for the measurement
of reductive capacity. The reactor was originally designed to
prevent oxidation of the sample by intrusion of oxygen from
air and to minimize loss of volatile organic compounds by
volatilization. It has been reported to keep anaerobic
conditions for over 2 months (43). The recoveries of target
compounds in controls in this research exceeded 98% at
each sampling time. The vials were introduced into the
anaerobic chamber and equilibrated with anaerobic atmo-
sphere for 2 days to remove oxygen sorbed on their surfaces.
Soil minerals and soil (1 g) were weighed and transferred to
vials. An aliquot (10 mL) of Cr(VI) solution with 10 mM
NaHCO3 was then added to each vial while in the chamber.
The mass ratio of solid to water was 0.1 (g/ g). The concen-
tration of Cr(VI) solution was approximately twice the
expected reductive capacity of each soil mineral and soil. A
51.5 mM Cr(VI) solution was used to measure the reductive
capacities of pyrite, green rust, and magnetite, and a 2.57
mM Cr(VI) solution was used to measure the reductive
capacities of iron-bearing phyllosilicates and soil. The pH
values of the soil mineral and soil suspensions were adjusted
to 7 by adding H2SO4 and NaOH solutions (1.0 M). Control
samples were prepared as the same manner without soil
minerals and soil. The batch reactors were tightly capped,
taken out of the anaerobic chamber, mounted on a tumbler,
and then completely mixed at 7 rpm at room temperature
(22 ( 0.5 °C). After contact with Cr(VI), the vials were
introduced into the anaerobic chamber, and 0.142 g of
Na2SO4 was added to result in 0.1 M sulfate. The vials
were taken out of the chamber and completely mixed for
1 day to promote desorption of Cr(VI) from soil mineral and
soil surfaces (19, 40) and centrifuged at 2800g for 10 min.
The supernatant was filtered through a 0.2-µm membrane
filter (Whatman), and Cr(VI) concentration of filtered super-
natant was measured colorimetrically. All soil mineral and
soil samples and their controls were prepared in duplicate
inside the anaerobic chamber.
spectrophotometer. Cr(VI) concentration in aqueous solution
was measured using the colorimetric reagent, diphenyl-
carbazide, at a wavelength of 542 nm (44). Iron contents of
soil minerals were measured after acid extraction with a
colorimetric procedure using 1,10-phenanthroline at a wave-
length of 562 nm (45, 46). A 10% solution of hydroxylamine
was added to the acid extractant to reduce Fe(III) to Fe(II)
for the measurement of total iron content. The ferrozine
method (47) was also used as an alternative method to
measure Fe(II) and total iron concentrations in aqueous
solution.
Target chlorinated ethylenes and their transformation
products were measured by gas chromatographic analysis.
PCE, TCE, DCEs, and VC were measured by a HP G1800A
GCD system with a DB-VRX column (60 m × 0.25 mm i.d.
× 1.8 µm film thickness, J&W Scientific) and an electron
ionization mass spectrometer detector. The temperature of
the injection port was 230 °C, and the temperature of detector
was 300 °C. The oven temperature was isothermal at 80 °C
for 8 min, increased to 160 °C at the rate of 20 °C/ min, and
then held for 2 min. Batch reactors were centrifuged to
separate aqueous and solid phases at 2800g for 30 min. A
4-mL sample of supernatant was rapidly withdrawn with a
10-mL gastight syringe (Hamilton) and transferred to an 8-mL
extraction vial (Kimble) with 2 mL of extractant (pentane
with 0.22 mM toluene). Supernatant in the reactors was
removed, and 10 mL of extractant was added to extract target
organics and transformation products sorbed on soil miner-
als. Both vials for extraction were mounted on an orbital
shaker and shaken for 30 min at 250 rpm. A 1-µL portion of
extractant was automatically introduced into split/ splitless
injection port at a split ratio of 30:1.
Non-chlorinated transformation products were analyzed
by a HP 6890 GC with a GS-alumina column (30 m × 0.53
mm i.d., J&W Scientific) and a flame ionization detector. The
temperatures of split/ splitless injection port and detector
were both 150 °C, and the temperature of oven was isothermal
at 100 °C. A 10-mL sample of supernatant was rapidly
transferred with a 10-mL gastight syringe to a 20-mL amber
vial. The vial was tightly capped after the addition of sample,
shaken for 1 h at 250 rpm, and allowed to stand for 2 h in
order to equilibrate the aqueous and gas phases. A gas-phase
sample of 50-100 µL was withdrawn from the headspace
with a 100-µL gastight syringe (Hamilton) and introduced
into the injection port at a split ratio of 5:1. The concentrations
of C2 hydrocarbons in aqueous solution were calculated using
dimensionless Henry’s law constants at room temperature
(20.4, ethane; 8.7, ethylene; 1.1, acetylene) (48, 49).
Experiments were conducted to measure reductive ca-
pacities of soil minerals and soil for chlorinated ethylenes.
Preliminary experiments were conducted to determine the
contact time required to ensure complete reaction with the
target organic. Batch experiments were then conducted to
determine the extent of degradation of target organics and
production of transformation products. All soil mineral and
soil samples and controls (without soil minerals and soil) for
these experiments were prepared in duplicate inside the
anaerobic chamber. After equilibration of 20-mL amber vials
with anaerobic atmosphere, weighed soil minerals and soil
were transferred to vials and 10 mM NaHCO3 solution was
added, generally resulting in a mass solid/ liquid ratio of 0.1.
A mass ratio of 0.007 was used for GRSO4 because its reductive
capacity was expected to be higher than those of other
materials. The pH values of the suspensions were adjusted
to 7 with H2SO4 and NaOH solutions (1.0 M). Batch
evaluations started by spiking stock solutions of target
organics to soil mineral and soil suspensions. Vials were
rapidly capped, taken out of the anaerobic chamber, and
mounted on the tumbler at 7 rpm for complete mixing. Target
organics and transformation products sorbed on the solid
phases and in aqueous solution were monitored at each
sampling time.
Results and Discussion
Developm ent of Test Methods. Optimal Cr(VI) contact time
with soil minerals and Silawa soil was determined by
measuring the reduction rate of Cr(VI) until it dropped to a
level where it could not be statistically distinguished from
zero. The optimal contact time for pyrite, GRSO4, and
magnetite to achieve maximum Cr(VI) reduction was 1 day
and that for iron-bearing phyllosilicates (biotite, vermiculite,
and montmorillonite) and Silawa soil was 4 days (see Figure
S-1, Supporting Information). A Cr(VI) contact time of 4 days
was chosen for all solids. Reductive capacity of soil minerals
and soil for Cr(VI) was determined by dividing the difference
between the equivalents of Cr(VI) originally added and that
remaining after extraction by the mass of solids present. This
approach assumes that all Cr(VI) that is removed was reduced
to Cr(III).
The amounts of chlorinated organics reduced as functions
of time were measured in preliminary experiments in which
the concentrations of target organics were monitored in both
aqueous and solid phases. Optimal contact times of target
organics were determined by applying the same technique
Analytical Procedures. Both chromium (Cr(VI)) and iron
(Fe(II) and total iron) were determined by spectroscopic
analysis using a Hewlett-Packard (HP) 8452A diode-array
9
VOL. 37, NO. 3, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 5 3 7

TABLE 3. Optimal Contact Times of Target Organics and
Distribution of Transformation Products Relative to Initial
Target Compound in Pyrite and Green Rust (GR )
SO4
Suspensions
contact
type of
target
timea
(day)
transformation distributionb
minerals
organics
products
(%)
pyrite
PCE
TCE
40.6
TCE
4.0
28.0
6.0
3.3
40.0
2.2
12.0
5.0
31.7
5.9
5.6
0.7
8.5
1.0
1.1
7.1
1.6
24.7
1.2
C2H2
C2H4
c-DCE
C2H2
C2H4
C2H2
C2H4
C2H4
C2H6
C2H2
C2H4
C2H2
C2H4
C2H2
C2H4
C2H6
C2H4
C2H6
32.3
c-DCE
VC
39.4
32.4
40.5
31.6
40.8
green rust
(GRSO4
PCE
)
TCE
c-DCE
VC
40.8
a
All inference tests were conducted in the range of 95% confidence
b
level under t-distribution. Relative to initial target com pound and based
on the m ass recovery (as %) of transform ation products at optim al
contact tim es.
FIGURE 1. Reductive capacities of soil minerals and soil for target
inorganic (Cr(VI)) and organic (PCE). Error bars represent the ranges
of observed reductive capacities. Some error bars are not shown
because of their small values.
described above. Batch experiments were then conducted
to measure the concentrations of target organics and
transformation products sorbed on solids and in aqueous
solution at each optimal contact time, which were used to
calculate the reductive capacity for target organics. Reductive
capacity of soil minerals and soil for target organic was
determined as the difference between the initial concentra-
tion of target organic and the sum of concentrations of target
organic and all transformation products measured in aqueous
solution and on solids at the optimal contact time, all
expressed as electron equivalents relative to ethane (PCE, 10
equiv/ mol; TCE, 8 equiv/ mol; DCEs, 6 equiv/ mol; VC, 4 equiv/
mol; ethylene, 2 equiv/ mol). The difference was divided by
the mass of solids in batch reactors.
Table 3 summarizes the optimal contact times of target
organics and transformation product distribution in pyrite
and GRSO4 suspensions. The major transformation product
was acetylene rather than DCEs and VC, which are more
often found as products of biotic dechlorination. This suggests
that the major transformation pathway may be reductive
elimination. This result is consistent with the results of other
abiotic reductive dechlorination studies by Sivavec and
Horney (30) and Butler (7). TCE and c-DCE were observed
as minor transformation products when PCE and TCE were
degraded by pyrite, but they were not observed in experi-
ments with green rust suspensions. Accumulation of DCEs
and VC was not observed in either mineral suspension. The
distribution of transformation products of c-DCE was dif-
ferent in each mineral suspension. More c-DCE was trans-
formed to acetylene than ethylene in pyrite suspension (see
Figure S-2, Supporting Information). However acetylene,
ethylene, and ethane were evenly produced in green rust
suspension. In the case of VC degradation, ethylene was the
main transformation product with a low concentration of
ethane observed.
than others. Duplicate analyses for reductive capacity of soil
minerals and Silawa soil for Cr(VI) are fairly consistent and
reproducible as shown in Figure 1. The relative standard
deviations (RSD) of the reductive capacities are less than
4.7%, except for pyrite (9.0%). This implies that the hetero-
geneity of soil minerals and soil does not significantly affect
measurement of the reductive capacity for Cr(VI). Reductive
capacities for Cr(VI) of pyrite, GRSO4, and magnetite are
roughly 1-3 orders of magnitude greater than those of iron-
bearing phyllosilicates. This result may be due to the lower
levels of iron in the phyllosilicates, which had 1-3 orders of
magnitude less iron than iron-bearing sulfides, hydroxides,
and oxides. Reductive capacity of pyrite for Cr(VI) was
surprisingly low as compared to that of GRSO4 and magnetite.
This is not compatible with other studies that have empha-
sized the reactivity of pyrite (31, 50). This may be explained
by the low content of Fe(II) and Fe(III) (94.8 and 120.3 mg/ g,
respectively) in the pyrite used for this experiment. The
measured iron content of pyrite was less than 50% of the
iron that should be present, which may be due to the
incomplete dissolution of iron during the acid extraction.
This suggests that there could be some iron in the pyrite that
is not dissolved by acid and is unavailable for the reduction
of Cr(VI). Although Fe(II) content in Silawa soil was 2 orders
of magnitude lower than that in biotite, its reductive capacity
(6.7 µequiv/ g) was 1.9 times greater than that of biotite. This
may be caused by high reactivity of reduced natural organic
matter (NOM) in Silawa soil pretreated with dithionite. It
has been reported that NOM acts as a redox catalyst in
electron-transfer reaction and accelerates the reductive
transformation of target contaminants in reducing environ-
ments (15, 16, 51).
Measurem ent of Reductive Capacity. The measured
reductive capacities of soil minerals and soil for Cr(VI) and
PCE are represented in Figure 1. The reductive capacities of
GRSO4 and magnetite are shown with a different scale (Figure
1b) because their reductive capacities are substantially higher
The duplicate analyses for reductive capacity of soil
minerals and Silawa soil for PCE were reasonably consistent
and reproducible as shown in Figure 1. RSD of the measured
reductive capacity for PCE was less than 8.8%, which is similar
9
5 3 8 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 3, 2003

FIGURE 2. Reductive capacities of pyrite and GRSO4 for a series of
chlorinated ethylenes. Error bars represent the ranges of observed
reductive capacities.
FIGURE 3. Reductive capacities for Cr(VI) for iron-bearing phyllo-
silicates and Silawa soil with and without dithionite pretreatment.
Error bars represent the ranges of observed reductive capacities.
Some error bars are not shown because of their small values.
to that for Cr(VI). Pyrite, GRSO4, and magnetite have 1 or 2
orders of magnitude higher reductive capacity for PCE than
iron-bearing phyllosilicates. Reductive capacity of pyrite for
PCE was still lower than that of GRSO4 and magnetite. The
reductive capacity of Silawa soil for PCE (0.80 µequiv/ g) was
similar to that of the iron-bearing phyllosilicates. The trends
in these results are similar to those observed with Cr(VI)
as target except that the reductive capacity of Silawa soil
for PCE was about the same as that of iron-bearing phyllo-
silicates.
Discussion. Reductive capacities of soil minerals and soil
for inorganic and organic contaminants can be compared in
Figure 1 for the two targets. The reductive capacity of soil
minerals for Cr(VI) was 3-16 times greater than that for PCE,
which indicates that Cr(VI) is more susceptible to the
reduction by soil minerals than PCE. GRSO4 has the greatest
reductive capacity for Cr(VI) and PCE followed by magnetite,
pyrite, biotite, montmorillonite, and vermiculite. A consistent
relationship was observed for the relative reductive capacities
of soil minerals for Cr(VI) and PCE. This suggests that the
reductive capacities of soil minerals for either Cr(VI) or PCE
could be used as representative values for representing the
reducing power of soil minerals. The relationship between
reductive capacities of soil minerals for Cr(VI) and PCE was
examined so that one measurement could predict the other:
treatment are shown in Figure 3. Duplicate analyses for the
reductive capacity without pretreatment were fairly repro-
ducible with a RSD less than 4.8%. The reductive capacity
of pretreated phyllosilicates and Silawa soil for Cr(VI) was
greater than the reductive capacity of the same samples
without pretreatment, but the extent of the difference varied
among the solids. Pretreatment with dithionite increased
the content of Fe(II) in biotite, vermiculite, montmorillonite,
and Silawa soil (1.8, 12.7, 350, and 140% increase in Fe(II)
content, respectively). However, it also resulted in slight
dissolution (<10%) of structural Fe(II) in biotite, vermiculite,
and Silawa soil. The relatively slight increase of Fe(II) content
combined with the loss of Fe(II) could be a possible cause
for the relatively small increase of reductive capacity of biotite
and vermiculite after pretreatment. In contrast, the reductive
capacity of pretreated montmorillonite and Silawa soil for
Cr(VI) was 15.8 and 17.7 times greater than that of mont-
morillonite and Silawa soil without pretreatment, respec-
tively. The content of Fe(II) in montmorillonite was quad-
rupled after dithionite pretreatment, and no dissolution of
structural Fe(II) was observed. This may explain the robust
increase of reductive capacity of montmorillonite after the
pretreatment. Pretreated Silawa soil showed the greatest
increase in reductive capacity, although the increase of Fe(II)
content after pretreatment was smaller than that observed
in montmorillonite and a slight loss of Fe(II) was observed.
The large increase in reductive capacity after pretreatment
may be due to the additional effect of reduced NOM. Except
in biotite, most of the structural iron in phyllosilicates and
soil was present as Fe(III), and this form was still dominant
after the pretreatment. Less than 24% of initial Fe(III) was
reduced to Fe(II) by the pretreatment, and the Fe(II) content
stayed between 0.7 and 1.6 mg/ g.
RCPCE ) 0.0622 × RCCr(VI) (r2 ) 0.993)
(1)
where RCPCE and RCCr(VI) represent reductive capacities of
soil minerals for PCE and Cr(VI) in µequiv/ g.
The reductive capacities of pyrite and GRSO4 for a series
of chlorinated ethylenes (PCE, TCE, c-DCE, and VC) are
compared in Figure 2. RSD of reductive capacities of duplicate
pyrite and GRSO4 samples for target organics was less than
8.8%. The reductive capacity of GRSO4 for target organics was
approximately 1 order of magnitude greater than that of
pyrite. This may be due to the higher content of reactive
compound (Fe(II)) in GRSO4. Fe(II) content in GRSO4 was 5
times greater than that in natural pyrite. Pyrite and GRSO4
show a trend in the order of magnitude of reductive capacities
for target organics. In both cases, the order of reductive
capacities for target organics was TCE > PCE > c-DCE > VC.
The reductive capacities of pyrite and GRSO4 for TCE were
greater than those for PCE. This is very interesting because
PCE is more highly oxidized than TCE; therefore, it would
be more susceptible to reductive dechlorination than TCE
(52). The similar trends in the higher susceptibility of TCE
have been also observed in the reductive dechlorination
kinetics of PCE and TCE by zerovalent metals (53) and iron
sulfide (54).
Figure 4 shows that the reductive capacities for PCE and
Cr(VI) for different groups of soil minerals are directly
proportional to Fe(II) contents, regardless of the soil mineral
type. The correlation factors (r) of linear regression lines are
significant in 95% confidence limit, indicating that Fe(II)
content significantly affects the reductive capacity of soil
minerals. The linear relationships between reductive capacity
of all soil minerals and Fe(II) content are well-described by
the following regression equations:
RCCr(VI) ) 160.33 × {Fe(II)} (r2 ) 0.936)
RCPCE ) 10.00 × {Fe(II)} (r2 ) 0.9332)
(2)
(3)
where {Fe(II)} is Fe(II) content in soil minerals in mequiv/ g.
However, the slopes indicate that only 16% of the Fe(II) is
able to reduce Cr(VI) and that only 1% of the Fe(II) is able
Reductive capacities for Cr(VI) of iron-bearing phyllo-
silicates and Silawa soil with and without dithionite pre-
9
VOL. 37, NO. 3, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 5 3 9

(9) Grittini, C.; Malcomson, M.; Fernando, Q.; Korte, N. Environ.
Sci. Technol. 1995, 29, 2898-2900.
(10) Chuang, F.; Larson, R. A.; Wessman, M. S. Environ. Sci. Technol.
1995, 29, 2460-2463.
(11) Baxter, R. M. Chemosphere 1990, 21, 451-458.
(12) Sokol, R. C.; Kwon, O.; Bethoney, C. M.; Rhee, G. Environ. Sci.
Technol. 1994, 28, 2054-2056.
(13) Perkins, P. S.; Komisar, S. J.; Puhakka, J. A.; Ferguson, J. F. Water
Res. 1994, 28, 2101-2107.
(14) Strathmann, T. J.; Stone, A. T. Abiotic Reduction of Oxime
Carbamate Pesticides by Fe(II): Catalytic Role of Mineral
Surfaces. In Specialty Chemicals in the Environment; Stone,
A. T., Ed.; 219th ACS National Meeting, American Chemical
Society, San Francisco, CA, 2000; American Chemical Society:
Washington, DC, 2000; pp 141-144.
(15) Schwarzenbach, R. P.; Stierli, R.; Lanz, K.; Zeyer, J. Environ. Sci.
Technol. 1990, 24, 1566-1574.
(16) Dunnivant, F. M.; Schwarzenbach, R. P.; Macalady, D. L. Environ.
Sci. Technol. 1992, 26, 2133-2141.
(17) Heijman, C. G.; Grieder, E.; Holliger, C.; Schwarzenbach, R. P.
Environ. Sci. Technol. 1995, 29, 775-783.
(18) Klausen, J.; Tro¨ ber, S. P.; Haderlein, S. B.; Schwarzenbach, R.
P. Environ. Sci. Technol. 1995, 29, 2396-2404.
FIGURE 4. Linear relationship between the reductive capacity of
iron-bearing soil minerals and Fe(II) content. Some error bars are
not shown because of their small values.
(19) Anderson, L. D.; Kent, D. B.; Davis, J. A. Environ. Sci. Technol.
1994, 28, 178-185.
(20) Eary, L. E.; Rai, D. Am. J. Sci. 1989, 289, 180-213.
(21) James, B. R.; Bartlett, R. J. J. Environ. Qual. 1983, 12, 173-176.
(22) Brewster, M. D.; Passmore, R. J. Environ. Prog. 1994, 13, 143-
148.
(23) McGeehan, S. L. J. Environ. Sci. Health, Part A: Environ. Sci.
Eng. Toxic Hazard. Subst. Control 1996, A31, 2319-2336.
(24) Hansen, H. C. B.; Koch, C. B.; Nancke-Krogh, H.; Borggaard, O.
K.; Sørensen, J. Environ. Sci. Technol. 1996, 30, 2053-2056.
(25) Koch, B. C.; Hansen, H. C. B. Adv. GeoEcol. 1997, 30, 373-393.
(26) Charlet, L.; Liger, E.; Gerasimo, P. J. Environ. Eng. 1998, 124,
25-30.
(27) Rusin, P. A.; Quintana, L.; Brainard, J. R.; Strietelmeier, B. A.;
Tait, C. D. Environ. Sci. Technol. 1994, 28, 1686-1690.
(28) Gossett, J. M. In Microbiological Aspects Relevant to Natural
Attenuation of Chlorinated Ethenes; Ward, C. H., Ed.; Symposium
on Natural Attenuation of Chlorinated Organics in Ground
Water; U.S. EPA: Dallas, TX, 1996; pp 10-13.
(29) Fruchter, J. S.; Amonette, J. E.; Cole, C. R. In Situ Redox
Manipulation Field Injection Test Report-Hanford 100-H Area;
DE97050892-XAM; U.S. Department of Energy: Washington,
DC, 1996.
(30) Sivavec, T. M.; Horney, D. P. Preprints of Papers Presented at
the 213th ACS National Meeting, April 13-17, 1997, San
Francisco, CA; American Chemical Society: Washington, DC,
1997; Vol 37 (1), pp 115-117.
(31) Kriegman-King, M. R. Environ. Sci. Technol. 1994, 28, 692-700.
(32) Kriegman-King, M. R.; Reinhard, M. Reduction of Hexa-
chloroethane and Carbon Tetrachloride at Surface of Biotite,
Vermiculite, Pyrite, and Marcasite. In Organic Substances and
Sediments in Water; Baker, R. A., Ed.; Lewis Publishers Inc.:
Chelsea, MI, 1991; Chapter 16, pp 349-364.
to reduce PCE. This result suggests that all Fe(II) in the soil
minerals cannot be used for the reduction of target com-
pounds and that the reactivities of Fe(II) sites are different
and specific to the target compounds. The reductive capaci-
ties of soil minerals predicted by these regression equations
are in good accordance with the measured reductive capaci-
ties of soil minerals except for those of biotite and vermiculite.
The measured reductive capacities of biotite and vermiculite
were overestimated by at least 1 order of magnitude by the
regression equations. This indicates that Fe(II) in these
minerals may not be as reactive as that in other soil minerals
tested. The correlation between the reductive capacity of
soil mineral and Fe(II) content observed in this research
suggests that Fe(II) is an important and reactive compound
in soil minerals for the reductive transformation of target
compounds. It would provide an approximate estimation to
predict and evaluate the potentials of natural reductants that
could be applied to the remedial technologies such as natural
attenuation and in-situ redox manipulation.
Acknowledgments
This research has been funded entirely with funds from the
State of Texas as part of the program of the Texas Hazardous
Waste Research Center. The contents do not necessarily
reflect the views and policies of the sponsor, nor does the
mention of trade names or commercial products constitute
endorsement or recommendation for use.
(33) Erbs, M.; Hansen, H. C. B.; Olsen, C. E. Environ. Sci. Technol.
1999, 33, 307-311.
Supporting Information Available
(34) Nightingale, E. R., Jr. Anal. Chem. 1958, 30, 267-272.
(35) Scott, M. J.; Morgan, J. J. Energetic and conservative properties
of redox systems. In Chemical Modeling of Aqueous Systems II;
Melichior, D. C., Bassett, R. L., Eds.; American Chemical Society
Symposium Series; American Chemical Society: Washington,
DC, 1990; pp 368-378.
Two graphs showing the effect of Cr(VI) contact time on the
reduction of Cr(VI) and reductive transformation of c-DCE
by pyrite. This material is available free of charge via the
Internet at http:/ / pubs.acs.org.
(36) Langmuir, D. Aqueous Environmental Geochemistry; Prentice
Hall: New York, 1997.
(37) Barcelona, M. J.; Holm, T. R. Environ. Sci. Technol. 1991, 25,
1565-1572.
(38) Heron, G.; Christensen, T. H.; Tjell, J. C. Environ. Sci. Technol.
1994, 28, 153-158.
(39) Kozuh, N.; Schara, M. Ann. Chim. 1995, 85, 257-265.
(40) Lee, W.; Batchelor, B.; Schlautman, M. A. Environ. Technol. 2000,
21, 953-963.
(41) Clark, W. M. Oxidation-Reduction Potentials of Organic Systems;
The William & Wilkins Company: Baltimore, MD, 1960.
(42) Taylor, R. M.; Maher, B. A.; Self-P., G. Clay Miner. 1987, 22,
411-422.
(43) Hwang, I. Ph.D. Dissertation, Texas A&M University, College
Station, TX, 2000.
(44) Greenberg, A. E.; Clesceri, L. S.; Eaton, A. D., Eds. Standard
Methods for the Examination of Water and Wastewater, 18th
ed.; APHA, AWWA, WEF: Washington, DC, 1992.
Literature Cited
(1) Tsukano Y. Contam. Hydrol. 1986, 1, 47-63.
(2) Hill, D. W.; McCarty, P. L. J. Water Pollut. Control Fed. 1967, 39,
1259-1444.
(3) Sokol, R. C.; Bethoney, C. M.; Rhee, G. Y. Chemosphere 1994,
29, 1735-1742.
(4) Kriegman-King, M. R. Ph.D. Dissertation, Stanford University,
Stanford, CA, 1993.
(5) Kleopfer, R. D.; Easley, D. M.; Haas, B. B. J.; Deihl, T. G.; Jackson,
D. E.; Wurrey, C. J. Environ. Sci. Technol. 1985, 19, 277-280.
(6) Parsons, F.; Wood, P. R.; DeMarco, J. J. Am. Water Works Assoc.
1984, 76, 56-59.
(7) Butler, E. C. Ph.D. Dissertation, University of Michigan, Ann
Arbor, MI, 1998.
(8) Barbash, J. E. Ph.D. Dissertation, Stanford University, Stanford,
CA, 1993.
9
5 4 0 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 3, 2003

(45) Amonette, J. E.; Templeton, J. C. Clays Clay Miner. 1998, 46,
51-62.
(46) Stucki, J. W.; Anderson, W. L. Soil Sci. Soc. Am. J. 1981, 45,
633-637.
(47) Gibbs, C. R. Anal. Chem. 1976, 48, 1197-1201.
(48) Mackay, D.; Shiu, W. Y. J. Phys. Chem. Ref. Data 1981, 10, 1175-
1199.
(49) Arnold, W. A.; Roberts, A. L. Environ. Sci. Technol. 1998, 32,
3017-3025.
(50) Lee, W.; Batchelor, B. Abiotic Reductive Dechlorination of
Chlorinated Ethylenes by Iron Bearing Soil Minerals and
Potential Interactions with Biotic Processes. In Chemical-
Biological Interactions in Contaminant Fate; Tratnyek, P. G.,
Adriaens, P., Roden, E. E., Eds.; 220th ACS National Meeting;
American Chemical Society: Washington, DC, 2000; pp 338-
340.
(52) Vogel, T. M.; Criddle, C. S.; McCarty, P. L. Environ. Sci. Technol.
1987, 21, 722-736.
(53) Roberts, A. L.; Totten, L. A.; Arnold, W. A.; Burris, D. R.; Campbell,
T. J. Environ. Sci. Technol. 1996, 30, 2654-2659.
(54) Butler, E. C.; Hayes, K. F. Environ. Sci. Technol. 1999, 33, 2021-
2027.
(55) Regazzoni, A. E.; Blesa, M. A.; Maroto, A. J. G. J. Colloid Interface
Sci. 1983, 91, 560-570
(56) Stumm, W. Aquatic Surface Chemistry: Chemical Processes at
the Particle-Water Interface; Wiley: New York, 1987.
(57) SSSA. Methods of Soil Analysis, Part III. Chemical Methodss
Agronomy Monograph; Madison, WI, 1996.
Received for review May 29, 2002. Revised manuscript re-
ceived November 7, 2002. Accepted November 13, 2002.
(51) Curtis, G. P.; Reinhard, M. Environ. Sci. Technol. 1994, 28, 2393-
2401.
ES025830M
9
VOL. 37, NO. 3, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 5 4 1