Corrinoid-mediated reduction of tetrachloroethene, trichloroethene, and trichlorofluoroethene in homogeneous aqueous solution: Reaction kinetics and reaction mechanisms

Article abstract of DOI:10.1021/es9603867

It is shown that in homogeneous aqueous solution containing titanium(III) citrate or titanium(III)-NTA as bulk electron donor, cobalamin, cobinamide, and cobamide are effective electron transfer mediators for the reduction of tetrachloroethene (PCE), trichloroethene (TCE), and trichlorofluoroethene (TCFE). For a given chlorinated ethene, the reaction rate varied only slightly with pH and type of corrinoid present and was about 5 and 50 times faster for PCE as compared to TCFE and TCE, respectively. Evidence is presented that the first and rate-limiting step of the reduction of PCE, TCE, and TCFE by super-reduced corrinoids is a dissociative one- electron transfer yielding the corresponding vinyl radicals. Furthermore, the elimination of a chloride radical from the 1,1-dichlorovinyl radical yielding chloroacetylene and subsequently acetylene is proposed to account for the direct formation of acetylene out of TCE. Finally, it is demonstrated that at higher reduction potentials the corrinoid mediators may be blocked by the formation of addition products.

Full text of DOI:10.1021/es9603867

Environ. Sci. Technol. 1997, 31, 253-260
PCE reductase, the enzyme catalyzing this reduction, probably
contains a corrinoid (16, 17).
Corrinoid-Mediated Reduction of
Tetrachloroethene, Trichloroethene,
and Trichlorofluoroethene in
Homogeneous Aqueous Solution:
Reaction Kinetics and Reaction
Mechanisms
In a variety of studies, it has been shown that reduced
corrinoids are potent reductants for chlorinated compounds
in homogeneous aqueous solution (18-25). Several groups
have demonstrated that such reduction reactions may also
be carried out with immobilized cobalt corrins (26-28) or in
bicontinuous microemulsions containing Co(I) macrocycles
(29). However, rather little is known about the mechanism-
(s) and rate-limiting step(s) of the reaction between reduced
corrinoids and halogenated compounds, particularly with
halogenated ethenes. Such knowledge is important from both
a fundamental scientific and a practical point of view, because
corrinoids could be very useful electron transfer mediators
for the abiotic treatment of halogenated solvent wastes (28).
G U Y G L O D , † W E R N E R A N G S T , ‡
C H R I S T O F H O L L I G E R , † A N D
R E N EÄ P . S C H W A R Z E N B A C H * , ‡
Swiss Federal Institute for Environmental Science and
Technology (EAWAG) and Swiss Federal Institute of
Technology (ETH), CH-8600 Du¨ bendorf, Switzerland, and
Limnological Research Center, EAWAG,
Super-reduced corrinoids are very strong nucleophiles (30)
that may react by a nucleophilic substitution, addition, or
electron transfer mechanism with halogenated compounds
(31). Any remediation or waste treatment technique using
corrinoids as electron mediators should guarantee the rapid
regeneration of the reactive corrinoid species without sig-
nificant loss of activity over long time periods. Furthermore,
it has to be assured that no toxic products accumulate. It is
therefore necessary to understand the influence of important
factors such as redox potential, pH, and structure of the
corrinoid on the reaction kinetics as well as their effect on
the product distribution. The reduction of PCE and TCE by
corrinoids has been reported to occur via sequential hydro-
genolysis yielding vinyl chloride as an intermediate (22). This
compound is of great concern as a drinking water contaminant
because of its high toxicity (32). In addition, it has been
postulated that radicals may be formed as intermediates in
dehalogenation reactions (24, 25, 33) that may lead to the
formation of a variety of products, because such radicals may
react with any nucleophile present (24).
In the study presented in this paper, we have investigated
the effect of redox potential and pH on the cobalamin-
mediated reduction of PCE, TCE, and trichlorofluoroethene
(TCFE) in homogeneous aqueous solution using titani-
um(III) citrate or titanium(III)-NTA (Ti(III)-NTA) as bulk
electron donors. In addition, experiments were conducted
with cobinamide and cobamide as mediators. The major
goals of this study were to evaluate the effect of the redox
potential on reduction rates and product distributions and
to get hints on the reaction mechanism(s).
CH-6047 Kastanienbaum, Switzerland
It is shown that in homogeneous aqueous solution containing
titanium(III) citrate or titanium(III)-NTA as bulk electron
donor, cobalamin, cobinamide, and cobamide are effective
electron transfer mediators for the reduction of tetrachlo-
roethene (PCE), trichloroethene (TCE), and
trichlorofluoroethene (TCFE). For a given chlorinated ethene,
the reaction rate varied only slightly with pH and type
of corrinoid present and was about 5 and 50 times faster
for PCE as compared to TCFE and TCE, respectively.
Evidence is presented that the first and rate-limiting step
of the reduction of PCE, TCE, and TCFE by super-reduced
corrinoids is a dissociative one-electron transfer yielding the
corresponding vinyl radicals. Furthermore, the elimination
of a chloride radical from the 1,1-dichlorovinyl radical
yielding chloroacetylene and subsequently acetylene is
proposed to account for the direct formation of acetylene
out of TCE. Finally, it is demonstrated that at higher
reduction potentials the corrinoid mediators may be blocked
by the formation of addition products.
Introduction
Experimental Section
Reductive dehalogenation reactions are presently of great
interest, primarily because of their potential use in the
treatment of halogenated solvent wastes as well as in
remediation approaches to removing such chemicals from
contaminated soils and aquifers (1-3).
It has been shown that chlorinated ethenes including
tetrachloroethene (perchloroethene, PCE) and trichloro-
ethene (TCE) can be reduced microbially under methano-
genic, acetogenic, and sulfate-reducing conditions to dichlo-
roethenes (mainly cis-1,2-dichloroethene), vinyl chloride,
ethene, and ethane (4-7). Enzymes containing tetrapyrrole
cofactors like iron porphyrins (8, 9), cobalamins (10), and
factor F430 (11, 12) are thought to be involved in the
dehalogenation reactions. Recently, it has been demonstrated
that certain bacteria can even use PCE and TCE as electron
acceptors by reducing it to cis-dichloroethene (13-15). The
Chem icals and Stock Solutions. Tetrachloroethene (per-
chloroethene, PCE) (IR spectroscopy grade), trichloroethene
(TCE) (puriss), cis-dichloroethene (cis-DCE) (purum), ethanol
(puriss), trisodium citrate dihydrate (MicroSelect), nitrilo-
triacetic acid (NTA) (puriss), hexane (Burdick & Jackson),
pentane (Burdick & Jackson), vitamin B12 (cyanocobalamin)
(>98%), piperazine-N,N′-bis(2-ethanesulfonic acid) (PIPES)
for pH 7 solutions and glycine for pH 9 solutions were from
Fluka AG (Buchs, Switzerland). Trichlorofluoroethene (TCFE)
(Fluorochem LTD, United Kingdom) was obtained from
Chemie Brunschwig AG (Basel, Switzerland). 1-Chloro-3-
fluorobenzene (99%) and trans-dichloroethene (trans-DCE)
were purchased at Aldrich Chemie, Switzerland. 2-Amino-
2-(hydroxymethyl)-1,3-propanediol (Tris) for pH 8 solution,
TiCl3 (15% in 10% HCl), and Na2CO3 were from Merck AG
(Dietikon, Switzerland). A gas mixture containing methane,
acetylene, ethene, ethane, carbon monoxide, and carbon
dioxide (1% in N2) (Scott Specialty Gases) was purchased from
Supelco. d7-Isopropyl alcohol (99.5%) and D2O were from
Armag AG (Do¨ ttingen, Switzerland). Cobinamide dicyanide
* Corresponding author e-mail address: schwaba@eawag.ch; tele-
phone: 41 1 8235109; fax: 41 1 8235471.
† EAWAG, Kastanienbaum.
‡ EAWAG and ETH, Du¨ bendorf.
9
S0013-936X(96)00386-0 CCC: $14.00
1996 Am erican Chem ical Society
VOL. 31, NO. 1, 1997 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 2 5 3

(approximately 95%) was from Sigma Chemie, Switzerland.
Deuterochloroform (CDCl3) was from Dr. Glaser AG (Basel,
Switzerland). p-Cresolyl cobamide was isolated from Sporo-
musa ovata by a procedure similar to the one described in
ref 34. All chemicals were used as received.
taken when TCE or TCFE were the reactants and analyzed by
GC/ MS.
Experiments in D2O. The 100 mM stock solutions of
titanium(III) citrate (pH 9) and the 100 mM glycine buffer
solution were lyophylized and then resuspended in D2O.
Cobalamin stock solutions were prepared in D2O. The
experimental setup was the same as for the experiments with
d7-isopropyl alcohol. A total of 20 µL of an anaerobic ethanolic
stock solution of PCE, TCE, or TCFE, respectively, was added
to start the reaction. After 3 half-lives, the products were
analyzed by GC/ MS.
Stock solutions of the halogenated ethenes (0.1 M) were
prepared in ethanol and were found to be stable for at least
10 days. Stock solutions of the corrinoids were prepared in
water at a concentration of about 5 mM and stored at 4 °C
in the dark. The exact concentration of the corrinoids was
determined spectrophotometrically as dicyano complexes
using the ꢀ-value of 30 500 M-1 cm-1 (31).
Analytical Procedures. The hexane and pentane extracts
were analyzed by GC on a Carlo Erba HRGC 5160 equipped
with an electron capture detector (Carlo Erba ECD 400 with
a Ni-63 source) or on a Carlo Erba MFC Mega Series GC
equipped with a QMD 1000 mass spectrometer. The column
used was a 30-m fused-silica DB-624 from J&W Scientific. As
mentioned above, pentane or hexane contained 1-chloro-
3-fluorobenzene or cis-DCE as standards, and concentrations
were calculated by comparison with a calibration curve.
Solutions of 100 mM buffer were titrated using HCl or
NaOH to the desired pH at 35 °C. Stock solutions of 100 mM
titanium(III) citrate were prepared in an anaerobic glovebox
containing 96% N2 and 4% H2 according to the procedure
described in ref 35. The final concentration of citrate was
300 mM. Ti(III)-NTA stock solutions of 100 mM were
prepared as described in ref 36. The pH was adjusted with
Na2CO3. Stock solutions of Ti(III) were kept in the glovebox
at room temperature.
Head space samples were analyzed by GC on a Carlo Erba
HRGC 5160 equipped with a flame ionization detector (FID-
40, Carlo Erba) or on a Fisons Instruments GC 8165 equipped
with a Fisons MD 800 mass spectrometer. The column was
a 30-m megabore GS-Q from J&W Scientific. Quantification
was done with pentane as the standard and by comparing
with a calibration curve established at 35 °C.
Experim ental Procedures. The kinetic experiments were
conducted in 59-mL serum flasks sealed with a Viton stopper
(Maagtechnik, Du¨ bendorf, Switzerland) and an aluminum
crimp cap. All experiments were done in triplicate unless
otherwise stated or as indicated by omission of standard
deviations in the figures. The reaction mixtures were prepared
in the glovebox. The pH buffer concentration was 90 mM,
and the titanium(III) citrate or Ti(III)-NTA concentrations
were 10 mM. The concentrations of corrinoids were ajusted
by adding the appropriate aliquot from the stock solution.
The concentration of the corrinoid was in the range of 0.4-
8.0 µM for the experiments with PCE, 20-75 µM for TCE, and
2-3 µM for TCFE and was always smaller than the concen-
tration of the halogenated compound. The headspace of the
reaction mixture was less than 2 mL.
The relative amounts of the deuterated versus non-
deuterated products were calculated from the intensity ratios
of the respective molecular ions, i.e., m/z 131 and 130 for
TCE, m/ z 97 and 96 for cis- and trans-DCE, and m/z 115 and
114 for the products of TCFE reduction.
For the quantification of the relative amounts of the
dehalogenation products of TCFE, the reaction solution was
extracted with CCl3D, and 1H-NMR spectra were recorded on
an ASX-400 NMR spectrometer from Bruker at 400 MHz
because no standards of the products were available.
The experiment was started by adding 57 µL of an
anaerobic ethanolic stock solution of the halogenated ethene
with a gas-tight Hamilton syringe through the Viton stopper.
The flasks were incubated at 35 °C in the dark. This
temperature was chosen because it is the optimal temperature
for many bacteria. At appropriate time intervals, 100-µL
samples were withdrawn with a gas-tight syringe and extracted
with 1 mL of hexane containing 10-4 M 1-chloro-3-fluo-
robenzene as the standard. The samples containing TCFE
were extracted with 0.8 mL of pentane containing cis-DCE
as the standard.
Spectroscopic Measurem ents. Spectroscopic measure-
ments were conducted to study the change of the oxidation
state of the cobalt-center of the corrinoid during the reaction.
EPR Measurements. Two separate solutions were pre-
pared, one containing 1 mM PCE, 1 mM titanium(III) citrate,
and 95 mM glycine buffer (pH 9), and the other containing
1 mM cobalamin, 2.5 mM titanium(III) citrate, and 95 mM
glycine (pH 9). An argon-purged 500-µL syringe was first
filled with 200 µL of the PCE solution followed by 200 µL of
the cobalamin solution. The solution was then transferred
to an anaerobic EPR tube where mixing occurred and was
frozen immediately in liquid nitrogen. Continous-wave EPR
spectra were obtained with a Bruker ESP 300 spectrometer
at 77 K (microwave frequency, 9.4225 GHz; modulation
frequency, 100 kHz; modulation amplitude, 10 G; microwave
power, 12.5 mW).
The product studies of the reduction of TCE and TCFE
were carried out in 8-mL sealed serum flasks containing 2 mL
of reaction mixture. The preparation of the reaction mixture
was identical to the kinetic experiments. The concentration
of cobalamin was 20 or 100 µM for TCE and 10 or 100 µM for
TCFE experiments. The reaction was started by adding 10
µL of the ethanolic stock solution of the compound to the
mixture. The stock solution contained pentane as the internal
standard. The vials were incubated at 35 °C in the dark; 200-
µL samples of the headspace were taken at appropriate times
with a gas-tight syringe during more than 3 half-lives.
D• abstraction experiments were conducted in duplicate
in 8-mL sealed serum flasks containing 4 mL of the reaction
mixture. Experiments with PCE and TCE were carried out at
pH 9.0, those with TCFE were carried out at pH 8.0.
Cobalamin was the mediator used for all abstraction experi-
ments. Different volumes of d7-isopropyl alcohol (0.1, 0.5, or
2% of the total volume of the solution) containing the
appropriate amount of PCE, TCE, or TCFE to yield the same
final concentration of the halogenated ethene in all bottles
were added to the reaction mixture. After about 3 half-lives,
0.1 mL of the PCE reaction mixture was extracted with hexane
and analyzed by GC/ MS. Head space samples (200 µL) were
Spectrophotometric Studies. A total of 10 µL of a 0.01 M
oxygen-free ethanol solution of PCE was added to a 1-cm
quartz cuvette with a gas-tight syringe. The cuvette was sealed
with a butyl rubber stopper and filled with 1 mL of a solution
containing 35 µM cobalamin, 90 µM titanium(III) citrate, and
100 mM glycine buffer at pH 9. Spectra were recorded at
room temperature with a Hitachi U 2000 spectrophotometer.
The influence of TCE and acetylene on cobalamin spectra
was studied in 1-cm quartz cuvettes sealed with a Teflon or
butyl rubber stopper, respectively, containing solutions made
of 1.6 mL of 25 µM cobalamin, 10 mM titanium(III) citrate,
and 90 mM PIPES buffer (pH 7). In one experiment, 10 µL
of a 0.03 M oxygen-free ethanol solution of TCE was added
to the cuvette in a glovebox. In another experiment, 30 µL
of pure acetylene was added to the cuvette through the stopper
9
2 5 4 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 31, NO. 1, 1997

FIGURE 2. Pseudo-first-order rate constants k′obs (O) obtained from
eq 3 as a function of initial cobalamin concentration; [PCE]0 ) 10
µM; [titanium(III) citrate] ) 10 mM; pH 9.0; 35 °C.
where k′obs is the pseudo-first-order rate constant for the
disappearance of the CE due to the reaction with the mediator.
The fact that, for a given mediator concentration, k′obs was
independent of the CE concentration even at very high CE
to mediator concentration ratios (e.g., 200:1, data not shown)
indicates that the reactive corrinoid was always present at a
constant steady-state concentration. Furthermore, as is
illustrated by Figure 2, k′obs was linearly dependent on the
total mediator concentration:
FIGURE 1. Reduction of 9.5 µM tetrachloroethene (PCE) (0)in 1µM
cobalamin and 10 mM titanium(III) citrate at pH 9.0 and 35 °C. (a)
Mass balance (2): The only product observed was trichloroethene
k′obs ) k2[mediator]tot
(4)
(TCE) (O). (b) Plot of ln ([PCE] /[PCE]0) (]) versus time.
t
where k2 is the apparent second-order rate constant of a given
CE for the reaction with the reduced corrinoid. Based on the
UV/ VIS spectra, it can be assumed that more than 90% of the
corrinoid was always present in the super-reduced form [i.e.,
as cobalt(I) corrinoid] before the addition of the CE to the
reaction solution. Thus, k2 values were determined for the
various compounds and solution conditions using eq 4. They
are summarized in Table 1. Considering the reaction with
cobalamin as mediator, it can be seen that, in general, PCE
reacted about 3-4 times faster than TCFE and about 50-80
times faster than TCE. The latter result is consistent with the
findings of Gantzer and Wackett (22). Furthermore, with
titanium(III) citrate as the bulk reductant, somewhat smaller
k2 values were obtained at pH 7 as compared to pH 8 and
pH 9. Finally, at a given pH, the k2 values were smaller if
Ti(III)-NTA was used as the bulk reductant instead of
titanium(III) citrate.
The most likely cause for the differences found in the k2
values of a given compound are the different redox conditions
prevailing in the various experiments. Although, as men-
tioned above, the cobalamin was initially completely reduced
in all cases, its regeneration rate may have become a limiting
factor at higher reduction potentials. In the pH range of the
experiments, the reduction potential of the Ti(IV)/ Ti(III)
couple is strongly pH-dependent [i.e., ∆E/ ∆pH ) -60 mV at
25 °C(37)]. It is therefore possible that in the case of titani-
um(III) citrate at pH 7, in contrast to pH 8 and pH 9, the
steady-state concentration of the reactive cob(I)alamin during
the reaction was smaller than the total cobalamin concentra-
tion, thus yielding a smaller apparent k2 value. The reduction
potential of the cobalamin itself is not pH-dependent in this
pH range (38). The same arguments can be used for the
experiments with Ti(III)-NTA, which has a higher reduction
potential than titanium(III) citrate. In fact, at pH 7, as
indicated by the UV/ VIS spectrum, the cobalamin was not
completely reduced in 10 mM Ti(III)-NTA. Furthermore,
the above explanation is supported by the findings that the
reduction of cob(II)alamin and cob(III)alamin occured at
slower rates at pH 7 than pH 8 and pH 9 in titanium(III)
with a gas-tight Hamilton syringe. Spectra were recorded on
a Kontron Uvikon 810 spectrophotometer at 12 °C.
Results and Discussion
Reaction Kinetics. With cobalamin present as a mediator at
various concentrations, the reduction kinetics of PCE, TCFE,
and TCE were studied at pH 7, pH 8, and pH 9 in 10 mM
aqueous titanium(III) citrate and at pH 8 and pH 9 in 10 mM
aqueous Ti(III)-NTA. In addition, experiments were carried
out with cobinamide or cobamide at pH 9 in 10 mM aqueous
titanium(III) citrate. Except for the reaction of TCE at pH 8
in aqueous Ti(III)-NTA, the rate of disappearance of a given
chlorinated ethene (CE) could be described by a pseudo-
first-order rate law:
d[CE]
dt
-
) kobs[CE]
(1)
and, thus
[CE]t
[CE]0
ln
) -kobst
(2)
where [CE]t and [CE]0 are the concentrations of the chlori-
nated ethene at time t and time zero, respectively. This is
illustrated in Figure 1 for the reaction of PCE in 10 mM
titanium(III) citrate at pH 9.0 in the presence of 1 µM
cobalamin. Note that the pseudo-first-order rate constant,
k
obs, derived from this type of experiment is a composite of
the reaction of the CE with the mediator and with the bulk
reductant, i.e., titanium(III) citrate or Ti(III)-NTA, respec-
tively. For all compounds and conditions investigated, the
contribution of the reaction with the Ti(III) bulk reductant
to the overall reaction rate was found to be less than 10% and
was accounted for by eq 3:
k′obs ) kobs - kobs,bulk
(3)
9
VOL. 31, NO. 1, 1997 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 2 5 5

TABLE 1. Second-Order Rate Constants k2 (M-1 s-1)a of PCE, TCFE, and TCE for Reaction with Cobalamin
titanium(III) citrate
pH 8.05
Ti(III)-NTA
b
pH 7.05
pH 9.0
pH 8.05
pH 9.0
PCE
TCFE
TCE
125 ( 7
28 ( 1
179 ( 10
155 ( 21
(200 ( 21)c
(154 ( 5)d
49 ( 5
102 ( 12
120 ( 15
41 ( 1
28 ( 1
37 ( 2
(98 ( 1)c
(nd)d
2.4 ( 0.2
3 ( 0.1
2.7 ( 0.2
(5.4 ( 0.2)c
(nd)d
(1.2 ( 0.1)e
1.5 ( 0.1
a
b
c
The intervals given are standard deviations. The error range of the pH is 0.05 in all the reactions. Rate constants for the reaction with
d
e
cobinam ide. Rate constants for the reaction with cobam ide. Initial reaction rate constant. nd, not determ ined.
citrate. Similarly, at a given pH (i.e., pH 8 and pH 9), the
reduction was slower in Ti(III)-NTA than in titanium(III)
citrate (data not shown). Finally, similar effects of the redox
conditions on the cobalamin-mediated reduction of halo-
genated compounds have been observed in other studies
(20, 23, 33).
To evaluate the influence of the coordinating base on the
reactivity of the corrinoid, the reaction of PCE with cobina-
mide and cobamide was studied at pH 9 in 10 mM titani-
um(III) citrate. Cobinamides can be obtained from cobal-
amins by hydrolytic cleavage of the nucleotide side chain at
the phosphate group. Thus, their side chain has lost the
coordinating ability toward the cobalt atom. p-Cresolyl
cobamide is a corrinoid that has been isolated from Sporo-
musa ovata, a homoacetogenic bacterium, which has been
shown to reduce PCE cometabolically (39). In this corrinoid
the base 5,6-dimethylbenzimidazole of cobalamin is replaced
by p-cresol that cannot coordinate the cobalt. The results
summarized in Table 1 show that only minor differences were
found between the k2 values obtained for the three mediators,
FIGURE 3. Reduction of 220 nmol of trichloroethene (TCE) (O) in 20
which indicates that the effect of the axial ligand on the
reactivity of the corrinoid is not very pronounced. This is in
accordance with the results of Lewis (24), who also observed
only a slightly faster reduction rate of CCl4 with cobinamide
as compared to cobalamin.
µM cobalamin and 10 mM titanium(III) citrate at pH 9.0 at 35 °C.
Time courses of the masses of TCE (O) and its transformation products.
The products observed were cis-dichloroethene (cis-DCE) (9), trans-
dichloroethene (trans-DCE) (1), acetylene (0), and ethene ([); mass
balance (2). The lines are the best fits using the model described
by eq 5.
Initial Product Form ation. As illustrated in Figure 1a, in
all experiments summarized in Table 1, initially PCE was
converted quantitatively (i.e., g 95%) to TCE. No other
products could be detected after 3 reaction half-lives with
the analytical techniques used. These findings are consistent
with the results of Gantzer and Wackett (22). Particularly, no
evidence was obtained that PCE also reacted byâ-elimination
to yield dichloroacetylene and subsequently acetylene, as has
been proposed for its reduction by zero-valent metals (3).
Figure 3 shows the time course of the initial product
formation when reducing 220 nmol of TCE by cobalamin at
pH 9 in titanium(III) citrate. As is indicated by the mass
balance, under these conditions, all TCE initially added could
be recovered in the detected products after 30 h: cis-DCE
(63%), trans-DCE (5%), acetylene (7%), and ethene (25%).
The time courses of the concentrations of the various
compounds shown in Figure 3 can be modeled by using the
following simple reaction scheme:
The indicated pseudo-first-order rate constants were deter-
mined by using the simulation tool AQUASIM (49). Figure
3 shows that the experimental data could be modeled quite
well (solid lines) with this reaction scheme. Note that in this
scheme (eq 5), it is assumed that all of the cis- or trans-DCE
is converted to acetylene. However, as is indicated by the
rate constants, even under these assumptions, no significant
fraction of the acetylene was formed by the reduction of cis-
or trans-DCE, thus suggesting that acetylene was produced
by another pathway. Furthermore, this is supported by the
finding that cis-DCE and trans-DCE are indeed reduced very
slowly by cobalamin under the given conditions (data not
shown). The data also suggest that ethene was primarily
formed by the reduction of acetylene, which is in agreement
with the finding that acetylene was readily reduced to ethene
under the same reaction conditions (data not shown).
Figure 4 shows that with decreasing pH and thus with
increasing redox potential, the amount of acetylene, and
particularly of ethene produced, decreased and that the mass
balance became less complete. A similar product distribution
as found for pH 7 and titanium(III) citrate (Figure 4) was
obtained at pH 8 using Ti(III)-NTA as the bulk reductant.
Obviously, with increasing redox potential, part of the
reduction products of TCE was present as nonvolatile product-
(s) as no other volatile products could be detected with the
analytical setup used. With cobinamide as the mediator at
pH 9 in 10 mM titanium(III) citrate, the amount of DCEs
k
TCE
k
cis-DCE
k3 = 0.034 h–1
trans-DCE
k
(5)
k
acetylene
k4 = 0.09 h–1
ethene
9
2 5 6 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 31, NO. 1, 1997

FIGURE 4. Product distribution of the reduction of trichloroethene (TCE) at pH 7, pH 8, and pH 9 in 10 mM titanium(III) citrate and 20 µM
cobalamin.
FIGURE 5. Product distribution of the reduction of trichlorofluoroethene (TCFE) at pH 7, pH 8, and pH 9 in 10 mM titanium(III) citrate and
3 µM cobalamin.
produced did not change while the recovery of TCE trans-
formation products was just about 90% and less acetylene
and ethene were detected (data not shown).
For TCFE, after 3 half-lives, only the three hydrogenolysis
products could be detected with cis-dichlorofluoroethene
(DCFE) being the major product formed (see Figure 5). At
all pH values, the same product distribution was found.
Reaction Mechanism s and Reaction Pathways. Several
different reaction mechanisms have been postulated for the
reaction of Co(I) corrinoids with halogenated compounds.
For many halogenated alkanes, nucleophilic substitution is
generally thought to be the initial reaction step (20, 25, 30),
while addition to double bonds has been proposed for the
reaction with halogenated ethenes (40). However, in many
cases it is difficult to distinguish clearly between these two
mechanisms and a pure electron transfer mechanism that
and a new maximum at 470 nm characteristic of cob(II)alamin
may also occur (41, 31).
FIGURE 6. UV/VIS spectra of cobalamin (35 µM) in 90 µM titani-
um(III) citrate solution at pH 9.0 at room temperature before and
after the addition of 10 µL of 10 mM tetrachloroethene(PCE). The
maximum at 390 nm characteristic of cob(I)alamin is disappearing,
appears.
In order to gain insight into the mechanism(s) of the
reactions of PCE, TCFE, and TCE with the reduced cobalamin,
the reactions were followed by UV/ VIS spectroscopy under
conditions where re-reduction of the cobalamin was not
possible. Figure 6 shows that upon the addition of 100 µM
PCE from an anaerobic stock solution to 35 µM cobalamin,
which had been reduced with 90 µM titanium(III) citrate in
100 mM glycine (pH 9), the large peak at 390 nm (characteristic
of cob(I)alamin), disappeared and maxima around 310 and
470 nm (typical of cob(II)alamin) appeared. The same
observations were made at pH 7 and pH 8 and after the
addition of TCE or TCFE instead of PCE. When a pure
anaerobic solution of ethanol was added to the cobalamin as
a control, cob(I)alamin did not react and the spectrum
remained constant. Note, however, that similar spectra as
the ones shown in Figure 6 are also obtained for methyl-
aquocobinamide (42) and other alkylcorrinoids (20). This
led Gantzer and Wackett (22) to postulate from similar results
that a carbon-cobalt bond formed between cobinamide and
tetrachloroethene.
As cobalamin is paramagnetic in the cob(II)alamin state,
EPR measurements were carried out to verify whether really
cob(II)alamin was formed. EPR spectra were recorded upon
9
VOL. 31, NO. 1, 1997 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 2 5 7

explanation for the observed difference is that, because of
the low stability of the 1,1-dichlorovinyl-radical, the elimina-
tion of a chloride radical yielding chloroacetylene and
subsequently acetylene is much faster than the abstraction
of a D•:
Cl
•
–Cl
•
C
C
HC
CCl
HC
CH (7)
Cl
H
The proportion of acetylene and ethene formed relative to
the DCEs was unaffected when increasing the amount of
d7-isopropyl alcohol in solution, suggesting that the formation
of acetylene was unaffected by the presence of a good
D•-donor. Furthermore, when 100 µM cobalamin was used
instead of 20 µM, the product distribution of TCE did not
change. A recombination of the radicals with cobalamin to
form a Co-C bond, a process that has been postulated for
many Co(I) substitutions of alkyl halides (31), can therefore
be excluded. The cleavage of the resulting Co-C bond would
lead to acetylene, and thus, an increase of the amount of
acetylene formed would be expected with increasing con-
centration of cobalamin.
Using a similar method, Chiu and Reinhard (23) showed
that the reduction of CCl4 proceeded via the formation of
radicals. It should be pointed out that the formation of
radicals would also be expected if cob(I)alamin added to the
double bond and the resulting Co-C bond was cleaved
homolytically. However, other studies have shown (43, 44)
that in protic solvents the heterolytic cleavage of the Co-C
bond is favored, leading to cob(III)alamin and the anion. The
reduction of Co(III) to Co(II) in the experiment shown in
Figure 6 would proceed slowly as the reductant Ti(III) is
limiting. Thus, the rapid formation of cob(II)alamin and the
formation of radical intermediates support the postulated
electron transfer mechanism.
Table 2 shows that the less halogenated radicals, i.e., the
radicals formed from the reduction of TCE abstract a D• much
more efficiently from d7-isopropyl alcohol than the radicals
formed from PCE and TCFE. For the radicals, an alternative
reaction pathway is reduction by the bulk reductant (that is
present in large excess) to the anion with subsequent
protonation. This pathway was verified by conducting the
reaction in deuterated water. For PCE and TCFE, over 90%
of the reaction products were deuterated, while TCE only
yielded about 60% deuterated cis- and trans-DCE, which is
consistent with the results of the D• abstraction experiments
(Table 2). Note that in neither case was there any indication
of the elimination of a Cl- from the anionic intermediate,
which would have yielded the corresponding acetylene. This
is in agreement with observations made by Houser (45), who
found that the elimination of chloride from the trichloro-
ethene anion is a slow process. These findings support the
hypothesis that the formation of acetylene and ethene from
the reduction of TCE (Figure 4) occurred primarily through
the 1,1-dichlorovinyl radical (eq 7). The preferential formation
of cis isomers was found by Kopchik and Kampmeier for
other ethenes, and they concluded that the cis/ trans ratio
was kinetically controlled (48).
FIGURE 7. EPR spectra at 77 K of 0.5 mM cobalamin in 1.7 mM
titanium(III) citrate at pH 9.0 upon the addition of an anaerobic solution
of ethanol (thick line) and upon the addition of an anaerobic solution
of 1 mM tetrachloroethene (PCE) (thin line).
addition of PCE to a titanium(III) citrate-reduced cobalamin
solution. As shown in Figure 7, cobalamin exhibited a EPR
signal attributable to cob(II)alamin as indicated by the g values
(gxy is approximately 2.23 and gz is 2.00). The decrease of the
signal of Ti(III) (gxy is approximately 1.94 and gz is 1.98) upon
addition of PCE showed that titanium(III) was oxidized when
PCE was added to the solution. Extraction of the reaction
mixture afterwards revealed the presence of TCE, meaning
that PCE was reduced to TCE with concurrent formation of
cob(II)alamin. The addition of pure ethanol instead of PCE
gave no Co(II) signal (Figure 7), indicating that oxygen
contamination was not an inherent problem. These obser-
vations suggest that the first step of the reduction of PCE,
TCFE, and TCE is a one-electron transfer of cob(I)alamin to
the halogenated ethene with formation of cob(II)alamin.
It is likely that, after the proposed one-electron transfer,
Cl- is eliminated, thus forming one (when X ) Cl) or three
(when X ) F, H) possible radical species:
–
•
Cl
Cl
X = Cl, F, H
C
C
X
Cl
Cl–
Cl–
(6)
Cl–
Cl
Cl
Cl
Cl
X
Cl
X
•
•
C
•
C
C
C
C
C
Cl
X
Unfortunately, spin-trapping or spin-labeling agents could
not be used to demonstrate the formation of these radicals,
because these agents reacted readily with cob(I)alamin and
titanium(III) citrate. Nevertheless, good indications for the
formation of the above postulated radicals were obtained
from experiments carried out in the presence of different
amounts of d7-isopropyl alcohol, which is a good D•-donor.
Table 2 shows that with increasing amounts of d7-isopropyl
alcohol, the amount of the deuterated products of PCE, TCE,
and TCFE increased. In the case of TCFE, the three deuterated
products indicated that all three expected radicals were
formed, while evidence for only two radicals, namely, the
two 1,2-dichlorovinyl radicals, was found for TCE because
no 1,1-dichloroethene was detected. The relative abundances
of the deuterated isomers and of the non-deuterated isomers
formed were identical.
Figure 5 shows that with decreasing pH both the amount
of acetylene and ethene formed from TCE and the recovery
of TCE in volatile products decreased. A spectrophotometric
analysis of the reaction mixtures of the TCE experiments at
pH 7 and pH 8, where the mass balance was incomplete,
revealed that the cobalamin was not completely present as
cob(I)alamin. The mixture showed
a weak absorption
maximum around 520 nm (spectra not shown), which is
typical for alkylcobalamins (23). When adding 300 µM TCE
to 25 µM cobalamin that had been reduced with 10 mM
titanium(III) citrate at pH 7 at 12 °C, the maximum at 390 nm
disappeared slowly and a maximum appeared at 520 nm,
In analogy to TCFE, the formation of all three radicals
could have been expected for TCE as well. A possible
9
2 5 8 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 31, NO. 1, 1997

TABLE 2. Ratio of Deuterated and Non-Deuterated Products of PCE, TCE, and TCFE as a Function of the Amount of d7-Isopropyl
Alcohol Present in Solution
amount of d7-isopropyl alcohol
(in % of total vol)
% of deuterated TCE,
the products of PCE
% of deuterated DCEs,
the products of TCE
% of deuterated DCFEs,
the products of TCFE
2 (1 for PCE)
0.5
0.1
10.4 ( 0.9
8.5 ( 0.6
6.5 ( 0.5
31 ( 3
14.5 ( 2
7.5 ( 1
5.2 ( 0.3
4.6 ( 0.02
4.4 ( 0.01
FIGURE 9. Possible reaction pathways of chloroacetylene to ethene
in the presence of cob(I)alamin.
γ-bands, it cannot be concluded from our spectra which of
the alkylcobalamins were present in the solution. Currently
work is in progress to determine the structure of the
alkylcobalamin(s) formed from the reduction of TCE.
FIGURE 8. UV/VIS spectra of cobalamin in 10 mM titanium(III) citrate
solution at pH 7.05 at 12 °C. (a) Spectra before and after the addition
of 300 µM trichloroethene (TCE). (b) Spectra before and after the
addition of acetylene.
Zhou et al. (44) showed that electron-withdrawing groups
at the Co-bound C-atom shift the reduction potential of the
corrinoid to more positive values. So, it is likely that adducts
like vinyl- and ethynylcobalamin are more difficult to reduce
(44), thus slowing down or even blocking the regeneration of
the reactive Co(I) species. In fact, Figure 10 shows that, at
pH 8 with Ti(III)-NTA as the reductant, the reaction of 100
µM TCE with 3 µM cobalamin as a mediator did not follow
pseudo-first-order kinetics; it proceeded more slowly than
anticipated after a few hours. The regeneration of the reactive
cobalamin became probably rate-limiting. Hence, if cor-
rinoids are to be used for remediation purposes, the redox
potential of the reaction medium has to be kept low enough
to allow quick regeneration of the reactive Co(I) species.
indicating the formation of an alkylcobalamin (Figure 8a).
After the disappearance of cob(I)alamin, the reaction mixture
was extracted, and about 70 µM cis-DCE could be detected.
Thus, significantly more cis-DCE was formed than cob(I)-
alamin had disappeared. As shown before, the formation of
cis-DCE did not proceed via an alkylcobalamin. However,
another fraction of the TCE reacted to give an alkylcobalamin
and blocked the regeneration of the reactive cobalamin
species. Part of the unrecovered TCE at low pHs or with
Ti(III)-NTA as bulk electron donor may, therefore, have been
present as alkylcobalamins. These were most likely the
products of the reaction of cobalamin with chloroacetylene
(Figure 9) that was formed from 1,1-dichlorovinyl radical (eq
7). The reaction of acetylene in 25 µM cobalamin that had
been reduced with 10 mM titanium(III) citrate at pH 7 at 12
°C gave a series of spectra very similar to the one found with
TCE (Figure 8b). It is known that acetylene and bromoacety-
lene (40) form addition products with cobalamin. Johnson
(46) demonstrated that acetylene reacts with cobalamin to
form vinylcobalamin while bromoacetylene yields ethynyl-
cobalamin and bromovinylcobalamin. The spectra of vin-
ylcobalamin and bromovinylcobalamin were identical, and
the spectra of ethynyl-, vinyl-, and bromovinylcobalamin differ
only by their most intense bands (γ-bands) in the 350-370-
nm region. If one assumes that bromovinylcobalamin and
chlorovinylcobalamin exhibit the same spectrum, and be-
cause titanium(III) citrate itself absorbs in the region of the
In conclusion, the results of this study suggest that the
first and rate-limiting step of the reduction of PCE, TCE, and
TCFE by super-reduced corrinoids is a dissociative one-
electron transfer yielding the corresponding vinyl radicals.
No evidence for a reductive â-elimination mechanism leading
to the direct formation of chloroacetylenes and finally
acetylene, as has been proposed for the reduction of
chlorinated ethenes by zero-valent metals (3), has been found.
The significant amounts of acetylene and ethene formed from
the reduction of TCE can be attributed to a dissociation
reaction of the 1,1-dichlorovinyl radical yielding chloro-
acetylene. As compared to the two 1,2-dichlorovinyl radicals,
the 1,1-dichlorovinyl radical seems to be very unstable in
9
VOL. 31, NO. 1, 1997 / ENVIRONMENTAL SCIENCE & TECHNOLOGY 2 5 9

(8) Li, S.; Wackett, L. P. Biochemistry 1993, 32, 9355-9361.
(9) Castro, C. E.; Wade, R. S.; Belser, N. O. Biochemistry 1985, 24,
204-210.
(10) Stromeyer, S. A.; Stumpf, K.; Cook, A. M.; Leisinger, T. Biodeg-
radation 1992, 3, 113-123.
(11) Krone, U. E.; Laufer, K.; Thauer, R. K.; Hogenkamp, H. P. C.
Biochemistry 1989, 28, 10061-10065.
(12) Holliger, C.; Schraa, G.; Stupperich, E.; Stams, J. M.; Zehnder,
A. J. B. J. Bacteriol. 1992, 174, 4427-4434.
(13) Holliger, C.; Schraa, G.; Stams, A. J. M.; Zehnder, A. J. B. Appl.
Environ. Microbiol. 1993, 59, 2991-2997.
(14) Holliger, C.; Schumacher, W. Antonie van Leeuwenhoek 1994,
66, 239-246.
(15) Neumann, A.; Scholz-Muramatsu, H.; Diekert, G. Arch. Microbiol.
1994, 162, 295-301.
(16) Neumann, A.; Wohlfarth, G.; Diekert, G. Arch. Microbiol. 1995,
163, 276-281.
(17) Schumacher, W.; Holliger, C. J. Bacteriol. 1996, 178, 2328-2333.
(18) Zanette, D.; Nome, F. J. Org. Chem. 1979, 44, 2308-2309.
(19) Rusling, J. F.; Miaw, C. L.; Couture, E. C. Inorg. Chem. 1990, 29,
2025-2027.
(20) Krone, U. E.; Thauer, R. K.; Hogenkamp, H. P. C. Biochemistry
1989, 28, 4908-4914.
(21) Krone, U. E.; Thauer, R. K.; Hogenkamp, H. P. C.; Steinbach, K.
Biochemistry 1991, 30, 2713-2719.
(22) Gantzer, C. J.; Wackett, L. P. Environ. Sci. Technol. 1991, 25,
715-722.
(23) Chiu, P. C.; Reinhard, M. Environ. Sci. Technol. 1995, 29, 595-
603.
FIGURE 10. Reduction of trichloroethene (TCE) (O)in 3µM cobalamin
and 10 mM Ti(III)-NTA at pH 8.05 at 35 °C. Inset: Plot of ln ([TCE]t /
[TCE]0) (O) versus time.
(24) Lewis, T. A.; Morra, M. J.; Brown, P. D. Environ. Sci. Technol.
1996, 30, 292-300.
(25) Wood, J. M.; Kennedy, F. S.; Wolfe, R. S. Biochemistry 1968, 7,
1707-1713.
(26) Ukrainczyk, L.; Chibwe, M.; Pinnavaia, T. J.; Boyd, S. A. Environ.
Sci. Technol. 1995, 29, 439-445.
(27) Marks, T. S.; Maule, A. Appl. Microbiol. Biotechnol. 1992, 38,
413-416.
aqueous solution, which may explain why 1,1-dichloroethene
has never been found in abiotic or microbially mediated
reductions of TCE (4, 13, 15, 22). To our knowledge this is
the first report demonstrating a direct pathway for acetylene
and ethene formation from the reduction of TCE by super-
reduced corrinoids.
From a practical point of view, the results of this study are
important for several reasons. First, we have shown that at
a low enough redox potential, independent of pH, several
corrinoids are effective electron transfer mediators that could
be used for the reductive dehalogenation of PCE, TCE, and
TCFE. Furthermore, we have demonstrated that TCE may
be transformed by a pathway that does not lead to the
formation of vinyl chloride. For waste treatment or reme-
diation schemes, a further evaluation of this pathway should
be of great interest. Finally, we have also shown that, at higher
reduction potentials, the regeneration of the super-reduced
corrinoid species may become rate-limiting and that these
species may even be blocked by addition products. The
formation of such addition products, particularly during the
reduction of dichloroethenes and vinyl chloride, is presently
under investigation in our laboratory.
(28) Habeck, B. D.; Sublette, K. L. Appl. Biochem. Biotechnol. 1995,
51/52, 747-759.
(29) Zhou, D. L.; Gao, J. X.; Rusling, J. F. J. Am. Chem. Soc. 1995, 117,
1127-1134.
(30) Schrauzer, G. N.; Deutsch, E. J. Am. Chem. Soc. 1968, 91, 3341-
3350.
(31) Dolphin, D. B12; John Wiley & Sons: New York, 1982; Vol. 1.
(32) Pontius, F. W. J. Am. Water Works Assoc. 1995, 87, 48-58.
(33) Assaf-Anid, N.; Hayes, K. F.; Vogel, T. M. Environ. Sci. Technol.
1994, 28, 246-252.
(34) Stupperich, E.; Steiner, I.; Ru¨ hlemann, M. Anal. Biochem. 1986,
155, 365-370.
(35) Zehnder, A. J. B.; Wuhrmann, K. Science 1976, 194, 1165-1166.
(36) Moench, T. T.; Zeikus, J. G. J. Microbiol. Methods 1983, 1, 199-
202.
(37) Zehnder, A. J. B. Ph.D. Dissertation, ETH Zurich, No. 5716, 1976.
(38) Lexa, D.; Save´ant, J. M.; Zickler, J. J. Am. Chem. Soc. 1977, 99,
2786-2790.
(39) Terzenbach, D. P.; Blaut, M. FEMS Microbiol. Lett. 1994, 123,
213-218.
(40) Pratt, J. M. Inorganic chemistry of vitamin B12; Academic Press:
London and New York, 1972.
(41) Zhou, D.-L.; Walder, P.; Scheffold, R.; Walder, L. Helv. Chim.
Acta 1992, 75, 995-1011.
(42) Friedrich, W. Vitamin B12 und verwandte Corrinoide; Georg
Thieme Verlag: Stuttgart, 1975; pp 47-55.
(43) Tinembart, O.; Walder, L.; Scheffold, R. Ber. Bunsen-Ges. Phys.
Chem. 1988, 92, 1225-1231.
(44) Zhou, D.-L.; Tinembart, O.; Scheffold, R.; Walder, L. Helv. Chim.
Acta 1990, 73, 2225-2241.
Acknowledgments
We thank Thomas Bucheli for help with GC/ MS measure-
ments, Roland Hany for the NMR measurements, Paul
Schosseler for the EPR measurements, and Francisco Vazquez
for technical assistance. We are indepted to Jo¨ rg Klausen,
who performed the simulations shown in Figure 3. Johanna
Buschmann, Rick Devlin, Jo¨ rg Klausen, Klaus Pecher, Anja
Sinke, and Wolfram Schumacher are kindly acknowledged
for reviewing the manuscript.
(45) Houser, T. J.; Bernstein, R. B.; Miekka, R. G.; Angus, J. C. J. Am.
Chem. Soc. 1955, 77, 6201-6203.
(46) Johnson, A. W.; Mervyn, L.; Shaw, N.; Smith, E. L. J. Chem. Soc.
1963, 4146-4156.
Literature Cited
(1) Holliger, C. Curr. Opin. Biotechnol. 1995, 6, 347-351.
(2) Matheson, L. J.; Tratnyek, P. G. Environ. Sci. Technol. 1994, 28,
2045-2053.
(3) Roberts, A. L.; Totten, L. A.; Arnold, W. A.; Burris, D. R.; Campell,
T. J. Environ. Sci. Technol. 1996, 30, 2654-2659.
(4) Freedman, D. L.; Gossett, J. G. Appl. Environ. Microbiol. 1989,
55, 2144-2151.
(5) Fathepure, B. Z.; Boyd, S. A. FEMS Microbiol. Lett. 1988, 49,
149-156.
(47) Belay, N.; Daniels, L. Appl. Environ. Microbiol. 1987, 53, 1604-
1610.
(48) Kopchik, R. M.; Kampmeier, J. A. J. Am. Chem. Soc. 1968, 90,
6733-6741.
(49) Reichert, P. Schr. EAWAG, Du¨ bendorf, Switzerland 1994, No. 7.
Received for review May 1, 1996. Revised manuscript received
August 27, 1996. Accepted August 28, 1996.X
(6) Bagley, D. M.; Gossett, J. M. Appl. Environ. Microbiol. 1990, 56,
2511-2516.
ES9603867
(7) Tandol, V.; Distefano, T. D.; Bowser, P. A.; Gossett, J. M.; Zinder,
S. H. Environ. Sci. Technol. 1994, 28, 973-979.
X Abstract published in Advance ACS Abstracts, November 1, 1996.
9
2 6 0 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 31, NO. 1, 1997