unreactive. Habeck and Sublette (8) investigated the
reduction of PCE in batch and column systems using B12
immobilized by adsorption onto Duolite S-761 resin and
titanium(III) citrate as the bulk reductant. Significant
sorption of PCE onto the resin occurred, creating an
additional process to account for. Reaction kinetics and
carbon mass recoveries were not reported.
The reduction of chlorinated methanes by immobilized
cobalamins (immobilized by adsorption onto talc) using
titanium(III) citrate as the reductant in batch systems was
investigated by Matheson (7). Reuse of the talc-im-
mobilized B12 resulted in successive decreases in the
observed reaction rate. Loss ofthe cobalamin bydesorption
or deactivation of the cobalamin may have contributed to
the reduction in rate.
In this study, the reduction of PCE and TCE catalyzed
by vitamin B12 in batch homogeneous and heterogeneous
systems is examined. Vitamin B12 was immobilized by
covalent bond formation to agarose, a solid support that
does not adsorb either chlorinated substrate. Particular
attention is paid to volatile organic reaction products and
intermediates and to carbon mass recoveries. The product/
intermediate formation and reduction kinetics of the two
batch systems were compared to assess the effect of
immobilization of vitamin B12 on catalytic activity. The
reusage potential of the immobilized vitamin B12 was also
examined. Evidence will be presented in support of a
reductive â-elimination reaction mechanism, a previously
unreported reaction pathway in the reduction of PCE and
TCE by vitamin B12.
present study. A 5 mM vitamin B12 solution served as a
stock reagent for the homogeneous system. For the
heterogeneous system, the required volume of gel was
added to a graduated pipet and washed with 66 mM Tris.
The final vitamin B12 concentrations were 10 and 9.7 µM
for the homogeneous and heterogeneous systems, respec-
tively (heterogeneous system B12 concentrations are re-
ported in terms of µmol of B12/ L suspension; 0.97 ( 0.05
µmol of vitamin B12 immobilized on agarose was added to
each serum vial). The stated masses of TCE or PCE and
400 µg of pentane were spiked into each vial and allowed
to equilibrate between the aqueous and vapor phases for
several days prior to commencing the reactions by placing
vials on a roller drum and rotating (g8 rpm) vertically as
the vial axis remained horizontal at 20 °C. Methanol was
used as a co-solvent in some cases and did not exceed 2
µL per vial. Care was taken to ensure that the septa were
not in contact with the vial headspace in order to minimize
losses. Reduction reactions were initiated by spiking with
the titanium(III) citrate stock solution to yield a final
concentration of 15 mM Ti[III], and the reactions were
allowed to continue on the roller drum. Positive controls
omitted vitamin B12. Negative controls consisted of the
respective chloroethylene in water (no titanium(III) citrate
or B12). In controls using only the vitamin B12 agarose
(results not shown), it was determined that sorption of PCE
and TCE to agarose was negligible. At selected intervals,
200-µL headspace samples were taken from the reaction
vials for analysis.
In heterogeneous system studies in which the im-
mobilized B12 was reused, after an 8-h time course, the
reaction vials were opened and allowed to sit overnight in
a fume hood. The vitamin B12 agarose was then washed
seven times with 75 mL of water, allowing the agarose to
settle by gravity for 45 min between washes to ensure
minimal loss of agarose. After the final wash, the system
was prepared for reaction as described previously. The
vitamin B12 immobilized on agarose was used to reduce
PCE four times in this fashion. Sampling was conducted
as described above. A loss of less than 5% of the initial B12
on agarose was confirmed volumetrically at the end of this
immobilized B12 reusage experiment.
Analytical Methods. Calibration standards containing
each organohalide and hydrocarbon gas component were
prepared using the 160-mL crimp-top serum vials with 100
mL of water and 60 mL of headspace volume. Quantifica-
tion using headspace samples was by the internal standard
method (using pentane as the internal standard). The
concentrations could be determined as mass per vial, since
the solution/ headspace ratio was the same as that used in
the reaction systems thus accounting for vapor/ water
partitioning. Headspace samples (200 µL) were analyzed
by a dual-column, column sequence reversal gas chro-
matographic method (15) using a Hewlett-Packard 5890
GC. The chlorinated ethenes and C2 hydrocarbon gases
were separated on 1% SP-1000 (60/ 80 mesh Carbopack B,
8 ft × 1/ 8 in. stainless steel (ss), Supleco) and Carboxen
1000 (60/ 80 mesh, 4 ft ×1/ 8 in. ss, Supelco) packed columns,
respectively. Initial column sequence was the SP-1000
followed by the Carboxen 1000. Samples were injected
splitless at 200 °C. Carrier gas was He at 30 mL/ min. Oven
temperature program was 60 °C for 1 min, ramp 18 °C/ min
to 210 °C, and hold for 3 min. The column switching valve
was rotated at 1 min runtime. Flame ionization detector
temperature was 240 °C. Due to co-elution ofcis- and trans-
Materials and Methods
Chem icals. PCE was obtained from J. T. Baker. TCE was
supplied by Fisher Scientific. cis-1,2-Dichloroethylene (cis-
DCE), trans-1,2-dichloroethylene (trans-DCE), 1,1-dichlo-
roethylene (1,1-DCE), and sodium citrate (citric acid,
trisodium salt) were obtained from Aldrich. Vitamin B12
(cyanocobalamin), sodium hydroxide, and Trizma base
were obtained from Sigma. Vitamin B12 immobilized on
4% beaded agarose (1.2 mg/ mL packed gel) was also
supplied by Sigma. In this product, vitamin B12 was
attached at the corrin ring to the agarose by binding
monocarboxyl derivatives of B12 to cyanogen bromide-
activated agarose via hexanediamine linkages (i.e., a C8
spacer) (13). Pentane was supplied by Burdick and Jackson.
Vinyl chloride (VC) in nitrogen and a gas mixture containing
acetylene, ethene, ethane, methane, carbon dioxide, and
carbon monoxide were obtained from Alltech Associates.
Titanium trichloride (13% in 20% HCl) was obtained from
Fluka. Stock solutions of 250 mM titanium(III) citrate in
660 mM Tris buffer (pH 8.2) were prepared in serum vials
as described in Smith and Woods (14). Working stock vials
were stored in the anaerobic chamber (10% H2 in N2) prior
to use. Milli-Q (Millipore) distilled deionized water, argon-
sparged (using an in-line O2 trap) for at least 1 h, was used
to prepare all solutions in this study.
Reaction System s. The batch systems (foil-wrapped
160-mL serum vials, crimp-sealed with Teflon-lined septa,
with final volumes of 100 mL of aqueous solution and 60
mL of headspace) were prepared, in duplicate, in the
anaerobic chamber. Vapor/ liquid exchange in these sys-
tems is rapid (90% of equilibrium for chloroethenes is
reached within 5 min). Air-water partitioning may safely
be assumed to be at pseudo-equilibrium for compounds
whose reactions require many hours or more, as in the
9
3 0 4 8 ENVIRONMENTAL SCIENCE & TECHNOLOGY / VOL. 30, NO. 10, 1996