Investigating the role of atomic hydrogen on chloroethene reactions with iron using Tafel analysis and electrochemical impedance spectroscopy

Article abstract of DOI:10.1021/es0264605

Metallic iron filings are commonly employed as reducing agents in permeable barriers used for remediating groundwater contaminated by chlorinated solvents. Reactions of trichloroethylene (TCE) and tetrachloroethylene (PCE) with zerovalent iron were investigated to determine the role of atomic hydrogen in their reductive dechlorination. Experiments simultaneously measuring dechlorination and iron corrosion rates were performed to determine the fractions of the total current going toward dechlorination and hydrogen evolution. Corrosion rates were determined using Tafel analysis, and dechlorination rates were determined from rates of byproduct generation. Electrochemical impedance spectroscopy (EIS) was used to determine the number of reactions that controlled the observed rates of chlorocarbon disappearance, as well as the role of atomic hydrogen in TCE and PCE reduction. Comparison of iron corrosion rates with those for TCE reaction showed that TCE reduction occurred almost exclusively via atomic hydrogen at low pH values and via atomic hydrogen and direct electron transfer at neutral pH values. In contrast, reduction of PCE occurred primarily via direct electron transfer at both low and neutral pH values. At low pH values and micromolar concentrations, TCE reaction rates were faster than those for PCE due to more rapid reduction of TCE by atomic hydrogen. At neutral pH values and millimolar concentrations, PCE reaction rates were faster than those for TCE. This shift in relative reaction rates was attributed to a decreasing contribution of the atomic hydrogen reaction mechanism with increasing halocarbon concentrations and pH values. The EIS data showed that all the rate limitations for TCE and PCE dechlorination occurred during the transfer of the first two electrons. Results from this study show that differences in relative reaction rates of TCE and PCE with iron are dependent on the significance of the reduction pathway involving atomic hydrogen.

Full text of DOI:10.1021/es0264605

Environ. Sci. Technol. 2003, 37, 3891-3896
and chloride ions (1, 2). Chlorocarbon reduction may
Investigating the Role of Atomic
Hydrogen on Chloroethene
Reactions with Iron Using Tafel
Analysis and Electrochemical
Impedance Spectroscopy
occur via direct electron transfer from the iron or via in
direct electron transfer from atomic hydrogen produced
from water reduction (2, 3). Although the chemical inter-
mediates between reactants and products are generally
known, the relative contributions of the direct and indirect
reduction mechanisms have not been investigated. Under-
standing the conditions favoring each reduction mechanism
is important for interpreting field and laboratory data,
especially where there is competition between oxidants for
reactive sites on the iron surfaces. This research investigated
the mechanisms involved in reductive dechlorination of
trichloroethylene (TCE) and tetrachloroethylene (PCE) using
Tafel analysis and electrochemical impedance spectroscopy
J I A N K A N G W A N G A N D J A M E S F A R R E L L *
Department of Chemical and Environmental Engineering,
University of Arizona, Tucson, Arizona 85721
(
EIS).
The most commonly observed products of TCE and PCE
Metallic iron filings are commonly employed as reducing
agents in permeable barriers used for remediating
groundwater contaminated by chlorinated solvents.
Reactions of trichloroethylene (TCE) and tetrachloroethylene
reactions with zerovalent iron are ethene and ethane, and
only trace levels of chlorinated products have been observed
by most investigators (4-8). A primary pathway that results
in complete chloroethene dechlorination contrasts with the
sequential hydrogenolysis pathway for chloroalkane reduc-
tion in which slower reacting products containing one fewer
chlorine atom are produced (2). Orth and Gillham postulated
that the general absence of chlorinated intermediates for
chloroethene reactions may be attributed to a chemisorption
mechanism that retains the compound at a reactive site
sufficiently long to allow for complete dechlorination (8).
Other investigators have proposed that TCE and PCE form
di-sigma bonds between their carbon atoms and the iron
surface (5). This proposed mechanism is similar to that
involved in catalytic hydrogenation of ethene, which is known
to occur via formation of di-sigma bonds with iron and other
transition-metal catalysts (9).
The breadth of evidence indicates that PCE reaction
rates with zerovalent iron are 2-5 times faster than those
for TCE (10, 11). However, several studies have found that
TCE reaction rates are faster than those for PCE (1, 5, 6).
Different relative reaction rates under different experimental
conditions using the same iron reactants have also been
observed for 1,2-dichloroethylene (DCE) and TCE (12). The
effect of experimental conditions on the relative reac-
tion rates of chlorinated ethenes suggests that there is
more than one reaction mechanism for these compounds
on iron surfaces. Two reaction mechanisms that have
been proposed involve direct electron transfer and indirect
reduction via atomic hydrogen (2, 6, 13). In palladium-
catalyzed systems employing hydrogen as the reducing
agent, chloroethene reaction rates increase with decreasing
extent of chlorination (14). This suggests that the mech-
anism responsible for faster TCE versus PCE dechlorination
in iron systems may involve atomic hydrogen as the reducing
agent.
(PCE) with zerovalent iron were investigated to determine
the role of atomic hydrogen in their reductive dechlorination.
Experiments simultaneously measuring dechlorination and
iron corrosion rates were performed to determine the
fractions of the total current going toward dechlorination
and hydrogen evolution. Corrosion rates were determined
using Tafel analysis, and dechlorination rates were
determined from rates of byproduct generation. Electro-
chemical impedance spectroscopy (EIS) was used to
determine the number of reactions that controlled the
observed rates of chlorocarbon disappearance, as well
as the role of atomic hydrogen in TCE and PCE reduction.
Comparison of iron corrosion rates with those for TCE
reaction showed that TCE reduction occurred almost
exclusively via atomic hydrogen at low pH values and via
atomic hydrogen and direct electron transfer at neutral
pH values. In contrast, reduction of PCE occurred primarily
via direct electron transfer at both low and neutral pH
values. At low pH values and micromolar concentrations,
TCE reaction rates were faster than those for PCE due
to more rapid reduction of TCE by atomic hydrogen. At
neutral pH values and millimolar concentrations, PCE reaction
rates were faster than those for TCE. This shift in relative
reaction rates was attributed to a decreasing contribution
of the atomic hydrogen reaction mechanism with increasing
halocarbon concentrations and pH values. The EIS data
showed that all the rate limitations for TCE and PCE
dechlorination occurred during the transfer of the first
two electrons. Results from this study show that differences
in relative reaction rates of TCE and PCE with iron are
dependent on the significance of the reduction pathway
involving atomic hydrogen.
EIS Analyses. The overall rate of an electrochemical
reaction may be controlled by the rate of electron transfer,
by mass transfer limitations, or by the rates of slow chemical
reactions that precede or follow an electron-transfer step
(
15). In terms of their effect on the flow of electrons, these
three mechanisms are analogous to resistance, capacitance,
and inductance in an electrical circuit (16). This analogy
allows the contribution of these three rate limitations on the
observed rates of electrochemical reactions to be probed
using EIS techniques.
Reactions on corroding iron have been modeled with the
circuit shown in Figure 1 (17). This circuit is consistent with
the three well-known hydrogen evolution reactions (HER)
on iron surfaces (15)
Introduction
Zerovalent iron is commonly used as a reactive medium to
remove chlorinated solvents from contaminated ground-
waters. The iron serves as a reducing agent to transform the
chlorinated compounds to their nonchlorinated analogues
*
Corresponding author phone: 520-621-2465; fax: 520-621-6048;
e-mail: farrellj@engr.arizona.edu.
1
0.1021/es0264605 CCC: $25.00

2003 Am erican Chem ical Society
VOL. 37, NO. 17, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 8 9 1
Published on Web 08/02/2003

FIGURE 1. Equivalent circuit for reactions on corroding iron surfaces.
+
-
•
Fe + H + e f Fe-H
(1)
(2)
•
2
Fe-H f 2Fe + H2
•
+
-
FIGURE 2. Sample Tafel profiles for four different iron wires at a
pH value of 3 in potassium phosphate electrolyte solutions for TCE
concentrations of 0.5, 1.0, 1.5, and 2.0 mM.
Fe-H + H + e f Fe + H2
(3)
•
where H is an atomic hydrogen radical. The equivalent circuit
in Figure 1 consists of a solution resistance (R ) and three
circuit elements, corresponding to reactions 1-3. Circuit
element I consists of the charge-transfer resistance (R ) for
reaction 1 coupled with the double-layer capacitance (Cdl).
Circuit element II accounts for the resistance and capaci-
tance associated with reaction 2, and element III accounts
for the inductance and resistance associated with reac-
tion 3.
The primary pathway and the rate-limiting step for
hydrogen evolution depends on the potential and solution
pH value. At neutral pH values hydrogen evolution on freely
corroding iron proceeds primarily via reactions 1 and 2, with
reaction 1 being the rate-limiting step (18). Reaction 3
becomes increasingly important with increasing surface
coverage of adsorbed atomic hydrogen (θ ), which increases
H
with decreasing pH values and decreasing potentials (19).
Decreasing potentials exponentially increase the rates of the
electron-transfer reactions and also increase θ . Therefore,
H
at sufficiently low potentials, reaction 3 may become the
rate-limiting step for hydrogen evolution (18).
Reaction 1 is known to be a concerted electron-transfer
reaction in which electron transfer and bond formation occur
simultaneously (15, 20). In contrast, reaction 3 involves a
slow chemical reaction that precedes an electron-transfer
step. This slow chemical reaction is the diffusion of atomic
hydrogen to a vacant cathodic site (21). This distinction is
important because it affects the influence that an oscillating
voltage has on the rate of each reaction. The overall rate of
a multistep reaction may not respond to an alternating voltage
signal if a slow chemical reaction precedes the electron
transfer step. Slow chemical reactions are therefore analogous
to inductance in an electrical circuit because inductors
contribute to increasing impedance with increasing oscil-
lation frequency of the voltage source.
S
platinum wire encased in a Nafion (Dupont) sheath and a
Hg/ Hg SO reference electrode (EG&G, Oak Ridge, TN). TCE
2
4
1
or PCE dechlorination was monitored by chloride analysis
and gas chromatography. The experiments were conducted
at nearly constant halocarbon concentrations given that the
loss of TCE or PCE in each ca. 2 h experiment was on the
order of 5%.
Electrochem ical Analyses. All direct current voltam-
metry experiments were carried out with an EG&G model
273A potentiostat and EG&G Powersuite software. Tafel scans
were performed at the end of each batch experiment by
polarizing the iron wire (100 mV with respect to its open
circuit potential at a scan rate of 5 mV/ s. EIS experiments
were conducted by coupling the 273A potentiostat with an
EG&G 5210 impedance phase analyzer. A sinusoidal ampli-
tude modulation of (10 mV was used over a frequency range
from 1 mHz to 100 kHz. The impedance data were fitted
using EG&G ZsimpWin software to a model circuit for iron
corrosion (17).
Chronoamperometry (CA) experiments were performed
to determine the amount of atomic hydrogen on the iron
surface as a function of the electrode potential and pH value.
These experiments were performed using an iron rotating
disk electrode (RDE) at 100 rpm in nitrogen-purged electrolyte
solutions. The method of Elhamid et al. (19) was used to
calculate the atomic hydrogen surface coverages from the
measured currents as a function of potential. All potentials
are reported with respect to the standard hydrogen electrode
(SHE).
Chrom atography Analyses. Halocarbon concentrations
in the electrochemical cell were measured with a Hewlett-
Packard 5890 Series II gas chromatograph (GC) equipped
with an electron capture detector (ECD) and autosampler.
Samples were prepared by injecting 100 µL aqueous samples
into 1 g of pentane. Reaction product concentrations were
determined via headspace analysis using a Hewlett-Packard
Materials and Methods
5
890 GC equipped with a flame ionization detector. The
Batch Experim ents Dechlorination rates of TCE and PCE
were measured in a 25 mL sealed glass cell in 20 mL of
background electrolyte solution. Each experiment utilized a
different 2.5 cm long by 1.2 mm diameter iron wire of 99.9%
purity (Aesar, Ward Hill, MA) as the reactant. The iron wires
were polished with fine sandpaper and rinsed thoroughly
with ultrapure water prior to use. Duplicate experiments were
performed in potassium phosphate and potassium sulfate
electrolytes at a total ionic strength of 0.30 M. Electrolyte pH
values were adjusted by adding small quantities of sulfuric
acid. The electrolyte solutions were purged with 100 mL/
min of nitrogen gas for 15 min prior to injecting the reactants
through a rubber septum. The solutions were stirred at 200
rpm for 30 min after adding 0.5-3.6 µL of neat TCE or PCE
to the reaction cell. The iron wire was then inserted through
the rubber septum. The reaction cell also contained a counter
electrode consisting of a 3 cm long by 0.3 mm diameter
standard error for the GC analyses was 4.9%. Chloride ion
concentrations were determined using ion chromatography
(
IC) by triplicate analyses of 5 mL samples using a Dionex
DX 500 ion chromatograph. The standard error of the IC
analyses was 3.5%, and the detection limit for chloride was
ca. 10 µg/ L.
Results and Discussion
Iron wire corrosion rates were determined using Tafel analysis
of polarization diagrams such as those shown in Figure 2.
The linear portions of the anodic and cathodic Tafel regions
were extrapolated to their point of intersection to determine
the iron corrosion rate as a function of the TCE or PCE
concentration (15). Figure 3 shows the currents associated
with TCE, PCE, and water reduction in potassium phosphate
electrolyte solutions at a pH value of 3. Similar results to
3
8 9 2
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 17, 2003

FIGURE 3. Corrosion current densities (i) and current densities for
hydrogen evolution, TCE (a) reduction and PCE (b) reduction at a
pH value of 3 in potassium phosphate electrolyte solutions. Error
bars on the currents for TCE and PCE reduction represent 95%
confidence intervals.
FIGURE 4. Corrosion current densities (i) and current densities for
hydrogen evolution, TCE (a) reduction and PCE (b) reduction at a
pH value of 7 in potassium sulfate electrolyte solutions. Error bars
on the currents for TCE and PCE reduction represent 95% confidence
intervals.
those shown in Figure 3 were observed in the potassium
sulfate electrolyte and are presented in the Supporting
Information. The currents for TCE and PCE reduction were
determined from the rates of chloride ion generation and
the dechlorination products. Reaction products for TCE and
PCE were 75 ( 6% ethene and 25 ( 6% ethane at both pH
values. Mass balance differences between the GC and IC
analyses were less than 10%. Changes in chloride ion
concentrations were used to determine the number of moles
of TCE or PCE that reacted. The product distribution indicates
that each mole of TCE or PCE that reacted released 3 or 4
mol of chloride ions, respectively. TCE reduction to ethene
and ethane requires 6 and 8 mol, respectively, of reducing
equivalents; while PCE reduction to ethene and ethane
requires 8 and 10 mol, respectively, of reducing equivalents.
The currents associated with hydrogen evolution were
determined by the difference between the total current and
the halocarbon reduction currents. The corrosion current
density for hydrogen evolution in the blank electrolyte in
PCE reduction was 16 µA/ cm . This indicates that the primary
2
mechanism for PCE dechlorination involved direct electron
transfer. The 33% greater rate of PCE dechlorination com-
pared to the increase in corrosion current suggests that PCE
may also react with atomic hydrogen produced from water
reduction.
Data in Figure 4a suggest that there was also indirect TCE
reduction at a pH value of 7. Adding TCE to the potassium
sulfate electrolyte resulted in increasing corrosion currents
with increasing TCE concentrations. Similar results to those
shown in Figure 4 were observed in the potassium phosphate
electrolyte and are presented in the Supporting Information.
The increasing corrosion currents with increasing TCE
concentration indicate that TCE was able to directly oxidize
the iron. However, the increase in corrosion current associ-
ated with direct TCE reduction was small compared to the
overall dechlorination current. For example, TCE at a
concentration of 1 mM increased the corrosion rate by 18%
over that in the blank electrolyte. However, 65% of the total
corrosion current went toward TCE dechlorination at this
concentration. This suggests that there was also indirect
reduction of TCE by atomic hydrogen at neutral pH.
The data in Figure 4b show that PCE had a more dramatic
effect on iron corrosion rates at neutral pH than at a pH
value of 3. Adding PCE to the solution at a concentration of
1 mM increased the corrosion rate by 117% over that in the
blank electrolyte. This shows that direct PCE reduction was
the major reaction contributing to iron oxidation. In fact,
the current efficiency for PCE reduction reached 69% at a
concentration of 1 mM. The data in Figure 4b also suggest
that there may be indirect PCE reduction at neutral pH. The
current going toward PCE reduction at a concentration of 1
mM was 27% greater than the increase in total current
associated with adding PCE to the solution. This is slightly
less than the 33% increase observed at a pH value of 3.
Dechlorination reactions involving atomic hydrogen
would be more important at lower pH values due to increasing
2
Figure 3a was 27 µA/ cm . Upon adding TCE to this solution
there was little change in the overall rate of iron corrosion.
However, the product generation rate showed that a sig-
nificant fraction of the current went toward TCE dechlori-
nation. At a TCE concentration of 2 mM, 90% of the corrosion
current went toward TCE dechlorination while only 10% went
toward hydrogen evolution. The high current efficiency for
TCE dechlorination, despite little change in the overall
corrosion rate from the blank solution, suggests that the
primary pathway for TCE dechlorination was reduction by
atomic hydrogen produced from reaction 1.
In contrast to the effect of TCE, the corrosion current
increased when PCE was added to the blank electrolyte
solution, as shown in Figure 3b. This indicates that PCE was
able to directly oxidize the iron. At a PCE concentration of
1
2
mM, the corrosion current density in the blank solution of
4 µA/ cm increased by 12 µA/ cm by the presence of PCE.
2
2
The product generation rate indicated that the current for
VOL. 37, NO. 17, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 8 9 3

The effect of TCE and PCE concentration on their relative
reaction rates at a pH value of 7 is shown in Table 1. The
first-order rate constants in Table 1 show that the TCE
reaction rate was 2.0 times faster than that for PCE when
each was present at a concentration of 6 µM and was 1.5
times faster at a concentration of 30 µM. However, at a
concentration of 1 mM the PCE reaction rate was 45% faster
than that for TCE at a pH value of 7. This shift in relative
reaction rates with concentration supports the hypothesis
that TCE reacts faster with atomic hydrogen than PCE. This
result is similar to that of Arnold and Roberts (5), who reported
that TCE reaction rates were up to 13 times faster than those
for PCE at micromolar concentrations on acid-washed iron
filings.
EIS Analyses. The data in Table 1 and Figures 3 and 4
suggest that there are different primary pathways for
dechlorination of TCE and PCE. The proposed reaction
sequence for both compounds begins with a chemisorption
step in which the halocarbon forms di-sigma bonds with
iron atoms, as given by (5)
FIGURE 5. Fractions of the electroactive iron surface covered by
adsorbed atomic hydrogen at pH values of 3 and 7.
TABLE 1. Apparent First-Order Rate Constants for TCE and PCE
in Batch Reactors Along with Their 95% Confidence Intervals
concn
k1(TCE)/
k1(PCE)
(µM) pH
TCE k1 (min-1)
PCE k1 (min-1
)
6
0
000
000
7
7
7
3
4.84 ( 0.35 E-05 2.41 ( 0.17 E-05
3.23 ( 0.15 E-05 2.12 ( 0.16 E-05
4.37 ( 0.29 E-06 6.31 ( 0.42 E-06
2.00 ( 0.39 E-04 1.14 ( 0.18 E-04
2.01
1.52
0.69
1.75
3
1
1
H
θ values with decreasing pH, as shown in Figure 5 for pH
where R ) Cl for PCE or H for TCE. The next reaction step
involves reduction by an electron or atomic hydrogen
according to
values of 3 and 7. Therefore, the effect of pH on the relative
current efficiencies and reaction rates for TCE versus PCE
indicates that reaction of TCE with atomic hydrogen was
faster than that for PCE. At a pH value of 3 and a concentration
of 1 mM, the current efficiency for TCE dechlorination was
8
1% while that for PCE was only 45%. However, at a pH value
of 7 and a concentration of 1 mM, the current efficiency for
PCE of 69% was similar to the current efficiency of 65%
observed for TCE. Comparison of the rates of TCE reduction
in Figures 3 and 4 show that TCE dechlorination rates were
The adsorbed radical is then reduced by an electron or
atomic hydrogen according to
4
7
5-50 times faster at a pH value of 3 than at a pH value of
. This effect of pH on TCE reaction rates was three times
that observed for PCE, for which the dechlorination rates
increased by factors of 15-18 upon lowering the pH value
H
from 7 to 3. Greater θ values at lower pH should lead to
faster reaction rates via an indirect mechanism involving
atomic hydrogen. Therefore, the greater effect of pH on TCE
reaction rates suggests that reduction of TCE by atomic
hydrogen was more facile than that for PCE. In fact, this
difference was significant enough to make the rate of TCE
dechlorination 75% faster than that for PCE at a pH value of
The adsorbed species produced by reaction 6 may then
desorb to yield mono- or dichloroacetylene, as observed by
Arnold and Roberts (5), or undergo further reduction to
ethene or ethane. The fact that most studies have not
observed chloroacetylenes suggests that the predominant
pathway after reaction 6 involves further reduction of the
adsorbed species rather than desorption. Although trace
amounts of chlorinated acetylenes were observed in solution
by Arnold and Roberts, no intermediate products between
mono- or dichloroacetylene and ethene were observed (5).
This suggests that the rate limitations to chloroethene
reduction were associated with reaction steps occurring prior
to formation of mono- or dichloroacetylene. Under this
scenario, all rate limitations to TCE and PCE dechlorination
are associated with the transfer of the first two electrons,
either directly or via atomic hydrogen.
Fast subsequent reactions after two electrons have been
transferred is consistent with the Bode phase plots shown
in Figure 6. Figure 6 shows a phase plot for reduction of
water, TCE, and PCE at a pH value of 7. Each peak in the
profile represents a unique reaction that affects the overall
impedance to current flow (22). In the blank electrolyte
solution there was only one peak corresponding to reaction
1. The absence of peaks for reactions 2 and 3 in the blank
solution shows that the rates of these reactions do not
3
and a concentration of 1 mM, as shown in Table 1. In
contrast, at a pH value of 7 the dechlorination rate for PCE
was 45% faster than that for TCE at a concentration of 1 mM.
The faster reaction rates for PCE versus TCE that have
been observed in most investigations (10, 11) can likely be
attributed to a small contribution of the atomic hydrogen
reduction mechanism under the conditions used in those
investigations. In the experiments presented in Figure 4, the
free corrosion potential of the iron wires ranged from -512
to -528 mV. The data in Figure 5 show that in this potential
range there was a very low concentration of atomic hydrogen
H
adsorbed on the iron surface. The low θ values at neutral
pH suggest that the atomic hydrogen reduction mechanism
will be saturated at low TCE concentrations. Therefore, at
neutral pH values this mechanism could be the dominant
TCE reduction pathway only at very low TCE concentrations.
2
The finite amount of H evolution in Figure 4a does not
preclude saturation of the atomic hydrogen reduction
mechanism for TCE, since not all the atomic hydrogen may
be accessible by TCE. Reaction sites covered by porous oxides
penetrable to water but not TCE would result in H
2
evolution,
despite a shortage of atomic hydrogen for TCE reduction.
3
8 9 4
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 17, 2003

FIGURE 7. Absolute value of the phase shift between the applied
potential and resulting current at a pH value of 3 for an iron wire
immersed in a blank electrolyte solution and in electrolyte solutions
containing TCE or PCE at a concentration of 1 mM.
FIGURE 6. Absolute value of the phase shift between the applied
potential and resulting current at a pH value of 7 for an iron wire
immersed in a blank electrolyte solution and in electrolyte solutions
containing TCE or PCE at a concentration of 1 mM.
measurably affect the rate of hydrogen evolution. This is
consistent with past observations that reaction 1 is the rate-
limiting reaction for hydrogen evolution at neutral pH values
(
18, 19, 23). In the solutions containing TCE or PCE, the two
additional peaks indicate that there were two additional
reactions whose rates affected the overall flow of current.
Because ethene production from TCE and PCE requires six
and eight electrons, respectively, the absence of peaks in the
Bode plots for these additional reactions indicates that their
rates do not affect the observed rates of TCE or PCE reduction.
The close agreement of the currents associated with PCE
reduction with the increase in total current resulting from
adding PCE to the solutions indicates that PCE reactions 5
and 6 primarily involve reduction by electrons. In contrast,
the primary dechlorination pathway for TCE depends on the
relative concentrations of TCE and atomic hydrogen on the
iron surface. Where the ratio of the atomic hydrogen to TCE
concentration is high, TCE reactions 5 and 6 primarily involve
atomic hydrogen. This is the case at a pH value of 3 where
adding TCE to the solution did not measurably increase the
total corrosion current over that in the blank, but reduction
of TCE accounted for up to 90% of the total current. Faster
reduction of TCE by atomic hydrogen than by electrons can
explain why reaction rates for TCE are more pH and
concentration dependent than those for PCE. Lower TCE
concentrations allow a greater contribution of atomic
hydrogen to the overall rate of TCE reduction. Faster
reduction by electrons of chemisorbed PCE versus TCE is
consistent with linear free energy relationships involving
standard reduction potentials (4) and the energies of the
lowest unoccupied molecular orbitals (E-LUMO) (24).
FIGURE 8. Resistance values determined from fitting the equivalent
circuit depicted in Figure 1 to the EIS data for TCE shown in Figure
6. To account for surface roughness, the double-layer capacitance
(Cdl) was replaced by a constant phase element (30) when fitting
the data. The error bars represent the 95% confidence intervals.
reaction 3 for TCE and PCE. These reactions may be of the
form
Evidence that the TCE and PCE reactions involving
induction also involve electron transfer from the electrode
is shown for TCE reduction in Figure 8. The resistance
3
associated with the inductive circuit element (R ) showed a
The phase angle for the third peaks in the Bode phase
plots for TCE and PCE in Figure 6 were negative and thereby
indicate the presence of induction in electron-transfer
reactions involving TCE and PCE. Induction arises from the
presence of electron-transfer reactions that involve surface-
adsorbed intermediates, such as atomic hydrogen. Induction
in the hydrogen evolution reaction was seen in the third
peak in the Bode plot at a pH value of 3, shown in Figure 7.
Induction in the blank electrolyte indicates that reaction 3
significantly contributed to hydrogen evolution. Similarities
in the Figure 7 Bode plots for TCE, PCE, and the blank
electrolyte suggest that there may be reactions analogous to
strong potential dependence and decreased by 13 orders of
magnitude between a potential of -0.5 and -0.9 V. This is
consistent with the exponential relationship between the
charge-transfer resistance and potential dictated by the
Butler-Volmer equation for electron-transfer reactions (15).
In contrast, the resistance associated with the capacitive
2
circuit element (R ) showed no potential dependence. This
is consistent with a reaction that does not involve the transfer
of an electron from the electrode, such as the reaction of
TCE with atomic hydrogen. For reaction 1, which involves
the transfer of an electron from the electrode to a solution
1
species, R showed the expected strong potential dependence.
VOL. 37, NO. 17, 2003 / ENVIRONMENTAL SCIENCE & TECHNOLOGY
9
3 8 9 5

Development (R-825223-01-0) and the National Institutes
for Environmental Health Sciences of the National Institutes
for Health (2P42ES04940-11). Although the research described
in this paper has been funded by these agencies, it has not
been subject to the agencies’ peer and policy review and
therefore does not necessarily reflect the views of the agencies,
and no official endorsement should be inferred.
Supporting Information Available
The corrosion experimental data at a pH value of 3 in
potassium sulfate electrolyte solutions and at a pH value of
7
in potassium phosphate electrolyte solutions. This material
is available free of charge via the Internet at http:/ /
pubs.acs.org.
FIGURE 9. Inductance values determined from fitting the equivalent
circuit depicted in Figure 1 to the EIS data shown in Figures 6 and
7
. To account for surface roughness, the double-layer capacitance
Literature Cited
(1) Gillham, R. W.; O’Hannesin, S. F. Ground Water 1994, 32, 958.
(Cdl) was replaced by a constant phase element (30) when fitting
the data. The error bars represent the 95% confidence intervals.
(
2) Matheson, L. J., Tratnyek, P. G Environ. Sci. Technol. 1994, 28,
045.
2
Figure 9 shows the inductance values measured in the
blank, TCE, and PCE solutions. At a pH value of 7, there was
very little induction in the blank due to the insignificant
contribution of reaction 3 to the overall rate of hydrogen
evolution. However, there was considerable induction in both
the PCE and TCE EIS spectra. This indicates that electron-
transfer reactions involving chemisorbed intermediates were
involved in their reduction. The greater induction for TCE
versus PCE is consistent with the more important role of
atomic hydrogen in its reduction. Slow diffusion of adsorbed
atomic hydrogen to a cathodic site for electron transfer to
TCE may be responsible for the induction. The lower
inductances for TCE and PCE at a pH of 3 as compared to
(
(
3) Brewster, J. H. J. Am. Chem. Soc. 1954, 76, 6361.
4) Roberts, A. L.; Totten, L. A.; Arnold, W. A.; Burris, D. R.; Campbell,
T. J. Environ. Sci. Technol. 1996, 30, 2654.
(5) Arnold, W. A.; Roberts, A. L. Environ. Sci. Technol. 2000, 34,
1794.
(
(
(
6) Farrell, J.; Melitas, N.; Kason, M.; Li, T. Environ. Sci. Technol.
000, 34, 2549.
2
7) Burris, D. R.; Campbell, T. J.; Manoranjan, V. S. Environ. Sci.
Technol. 1995, 29, 2850.
8) Orth, W. S.; Gillham, R. W. Environ Sci. Technol. 1996, 30, 66.
(9) Masel, R. I. Principles of Adsorption and Reaction on Solid
Surfaces; John Wiley and Sons: New York, 1996.
10) Tratnyek, P. G.; Johnson, T. L.; Scherer, M. M.; Eykolt, G. R.
Ground Water Monit. Rem. 1997, Fall, 108.
11) Johnson, T. L.; Scherer, M. M.; Tratnyek, P. G. Envrion. Sci.
Technol. 1996, 30, 2634.
12) Wust, W. F.; Kober, R.; Schlicker, O.; Dahmke, A. Environ. Sci.
Technol. 1999, 33, 4304.
(13) Balko, B.; Tratnyek, P. G. J. Phys. Chem. B 1998, 102, 1459.
(14) Lowry, G. V.; Reinhard, M. Environ. Sci. Technol. 1999, 33, 1905.
15) Bockris, J. O’M.; Reddy, A. K. Modern Electrochemistry; Plenum
Press: New York, 1970; Vol. 2.
16) Christensen, P. A.; Hamnet, A. Techniques and Mechanisms in
Electrochemistry; Chapman & Hall: Oxford, 1994.
17) Mansfeld, F. Corrosion 1981, 36, 301.
(
(
(
H
a pH of 7 can be attributed higher θ values that make the
slow diffusion of atomic hydrogen to cathodic sites less of
a rate limitation. The appearance of induction in the blank
solution at a pH value of 3 is consistent with a significant
contribution of reaction 3 to hydrogen evolution.
The reactions proposed in this study require that TCE
and PCE have access to the metallic iron surface. This occurs
only at cracks in the oxide film and suggests that only a small
fraction of the iron surface is available for halocarbon
reactions involving chemisorption. This is consistent with
TCE reaction rates on cathodically protected iron that are
(
(
(
(18) Bockris, J. O’M.; Carbajal, J. L.; Scharifker, B. R.; Chandrasekaran,
K. J. Electrochem. Soc. 1987, 134, 1957.
(
19) Abd Elhamid, M. H.; Ateya, B. G.; Weil, K. G.; Pickering, H. W.,
J. Electrochem. Soc. 2000, 147, 2148.
2
-3 orders of magnitude faster than those on oxide-coated
iron (25, 26). On iron coated with electrically conductive
magnetite or semiconducting Fe(III) oxides, halocarbon
reduction may also occur via electron tunneling through the
oxides (13, 27). Electron tunneling to a physically adsorbed
chloroethene may be responsible for the sequential hydro-
genolysis pathway that produces stable products with one
fewer chlorine atom than the parent compound (4, 28).
A chloroethene reaction mechanism involving chemi-
sorption and atomic hydrogen may explain why the linear
free energy correlations between E° values and reaction rates
developed for chloroalkanes underpredict rates of chloro-
ethene reduction (29). Reaction mechanisms involving
atomic hydrogen may make laboratory determination of
steady-state reaction rates under field conditions more
difficult, since experiments must mimic the adsorbed
hydrogen concentrations found in the field. Therefore, sealed
batch experiments, which are inherently not steady state, or
accelerated flow-rate column testing may be of limited
usefulness for predicting chloroethene reaction rates under
field conditions.
(
(
20) Marcus, R. A. Annu. Rev. Phys. Chem. 1964, 15, 155.
21) Diard, J. P.; Landaud, P.; LeGorrec, B.; Montella, C. J. Electroanal.
Chem. 1988, 255, 1.
(22) Wang, S.; Kato, T.; Nagata, S.; Honda, T.; Kaneko, T.; Iwasita,
N.; Dokiya, M. Solid State Ionics 2002, 146, 203.
(
(
23) Li, T.; Farrell, J. Environ. Sci. Technol. 2001, 35, 3560.
24) Scherer, M. M.; Balko, B. A.; Gallagher, D. A.; Tratnyek, P. G.
Environ. Sci. Technol. 1998, 32, 3026.
25) Li, T.; Farrell, J. Environ. Sci. Technol. 2000, 34, 173.
26) Farrell, J.; Kason, M.; Melitas, N.; Li, T. Environ. Sci. Technol.
2000, 34, 514.
(
(
(27) Scherer, M. M.; Balko, B. B.; Tratnyek, P. G. The Role of Oxides
in Reduction Reactions at the Metal-Water Interface. In Mineral-
Water Interfacial Reactions: Kinetics and Mechanisms; Sparks,
D. L., Grundi, T. J., Eds.; ACS Symposium Series 715; American
Chemical Society: Washington, DC, 1998; pp 301-332.
(
28) McCarty, P.; Criddle, C. S.; Vogel, T. M. Environ. Sci. Technol.
1987, 21, 722.
(29) Liu, Z.; Betterton, E. A.; Arnold, R. G. Environ. Sci. Technol.
2000, 34, 804.
(
30) Macdonald, J. R. Impedance Spectroscopy; John Wiley & Sons:
New York, 1987.
Received for review December 20, 2002. Revised manuscript
received June 16, 2003. Accepted June 19, 2003.
Acknowledgments
This research has been supported by the United States
Environmental Protection Agency Office of Research and
ES0264605
3
8 9 6
9
ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 37, NO. 17, 2003