Electrochemical reduction and oxidation of DDT

Article abstract of DOI:10.1139/v99-113

Electrolysis has been studied as a possible method to treat DDT wastes. In methanol, the major process was dehydrochlorination to DDE followed by further reduction. In an aqueous emulsion containing 1% heptane and 0.1% Triton SP-175, DDT was reduced at a deposited lead electrode with sodium sulphate as the supporting electrolyte by sequential hydrodechlorination of the aliphatic chlorine atoms. An excellent material balance was achieved, but the current efficiency was poor, even at low current densities. Electrooxidation of DDT was also investigated; in aqueous solutions or emulsion, little oxidation occurred because of competing oxidation of water at the highly positive potentials needed to oxidize DDT. In acetonitrile, electrooxidation occurred with high current efficiency by way of 'electrochemical combustion' of DDT and its intermediate oxidation products to CO₂. We conclude that development of an electrolytic technology for destroying DDT wastes is unlikely.

Full text of DOI:10.1139/v99-113

1
281
Electrochemical reduction and oxidation of DDT
Simona G. Merica, Wojceich J“dral, Susan Lait, Peter Keech,
and Nigel J. Bunce
Abstract: Electrolysis has been studied as a possible method to treat DDT wastes. In methanol, the major process was
dehydrochlorination to DDE followed by further reduction. In an aqueous emulsion containing 1% heptane and 0.1%
Triton SP-175®, DDT was reduced at a deposited lead electrode with sodium sulphate as the supporting electrolyte by
sequential hydrodechlorination of the aliphatic chlorine atoms. An excellent material balance was achieved, but the
current efficiency was poor, even at low current densities. Electrooxidation of DDT was also investigated; in aqueous
solutions or emulsion, little oxidation occurred because of competing oxidation of water at the highly positive
potentials needed to oxidize DDT. In acetonitrile, electrooxidation occurred with high current efficiency by way of
“
electrochemical combustion” of DDT and its intermediate oxidation products to CO2. We conclude that development
of an electrolytic technology for destroying DDT wastes is unlikely.
Key words: electroreduction, electrooxidation, voltammetry, surfactant media.
Résumé : On a évalué l’utilité potentielle de l’électrolyse comme méthode de se débarrasser de résidus de DDT. Dans
le méthanol, le processus majeur est une déhydrochloruration en DDE qui est suivie d’une réduction ultérieure. Dans
une émulsion aqueuse contenant 1% d’heptane et 0,1% de Triton SP-175®, en présence de sulfate de sodium comme
électrolyte de support, le DDT peut être réduit au niveau d’une électrode de plomb, par hydrodéchlorurations
successives des atomes de chlore aliphatiques. On a pu établir une excellente balance des produits, mais l’efficacité de
courant est mauvaise, même à de faibles densités de courant. On a aussi étudié l’électrooxydation du DDT; en solution
aqueuse ou à l’état d’émulsion, il se produit peu d’oxydation à cause d’une oxydation compétitive de l’eau aux
potentiels hautement positifs requis pour oxyder le DDT. Dans l’acétonitrile, l’électrooxydation se produit avec de très
bonnes efficacités de courant par le biais d’une “combustion électrochimique” du DDT et de son produit d’oxydation
intermédiaire en CO . On conclut que le développement d’une technologie électrolytique d’élimination des résidus du
2
DDT est peu probable.
Mots clés : électroréduction, électrooxydation, voltampérométrie, milieu d’agents de surface.
[
Traduit par la Rédaction] Merica et al. 1287
Introduction
Reddy (3) showed the reduction of DDT to DDD to be an ir-
reversible two-electron reaction. They proposed, following
the work of Lambert and Ingall (4), that two electrons were
added to DDT, to produce Cl + DDD which then
protonated. (Unlike aryl chlorides (5), the electroreduction
of alkyl chlorides does not afford radical anions as discrete
intermediates).
Rosenthal and Lacoste (6) observed three two-electron re-
ductions in 90% ethanol:10% 1.0 M aqueous methyltributyl-
ammonium chloride. These were attributed to successive
hydrodechlorination of DDT’s three aliphatic chlorines, on
the evidence that DDD showed reduction waves at the same
potentials as the second and third reduction waves for DDT.
Tandon et al. (7), who worked in 4% DMF:50% etha-
nol:46% aqueous tetramethylammonium bromide, noted
three well-defined reduction waves at –0.370 to –0.430 V,
–0.795 to –0.955 V, and –1.110 to –1.225 V vs. SCE, and
observed the three stepwise reductions, ultimately to DDO,
Although the use of DDT (1,1-bis(p-chlorophenyl)-2,2,2-
trichloroethane) was discontinued in North America in the
early 1970s, it remains a significant environmental problem,
both because of leaching from abandoned waste dumps and
because it is still used for mosquito control elsewhere. DDT
is persistent in the environment due to its resistance to
biodegradation (1). Along with previous authors, we have
been studying electrolysis as a possible method for convert-
ing DDT to more biodegradable compounds. Figure 1 shows
the structures and common abbreviations of DDT and its de-
rivatives.
Fukami and Nakajima in 1953 (2) electrolysed DDT at a
dropping mercury electrode at –1.6 V vs. a saturated calomel
electrode (SCE) and obtained DDD in 80% yield together
with a small amount of yellow resinous substance (this may
have been DDMS, which is a pale yellow oil). Reddy and
–
–
Received January 7, 1999.
1
S.G. Merica, W. J“dral, S. Lait, P. Keech, and N.J. Bunce. Department of Chemistry and Biochemistry, University of Guelph,
Guelph, ON N1G 2W1, Canada.
1Author to whom correspondence may be addressed. Telephone: (519) 824-4120, ext. 3962. Fax: (519) 766-1499.
e-mail: bunce@chembio.uoguelph.ca
Can. J. Chem. 77: 1281–1287 (1999)
© 1999 NRC Canada

1
282
Can. J. Chem. Vol. 77, 1999
Fig. 1. Names and structures of DDT and derivatives.
Cl
CH
CCl
Cl
Cl
CH
CHCl
Cl
3
2
1,1'-(2,2,2-trichloroethylidene)bis(4-chlorobenzene) 1,1'-(2,2-dichloroethylidene)bis(4-chlorobenzene)
("DichloroDiphenylTrichloroethane" = DDT)
("DichloroDiphenylDichloroethane"= DDD)
Cl
CH
CH Cl
Cl
Cl
CH
CH
Cl
2
3
1
,1'-(2-chloroethylidene)bis(4-chlorobenzene)
1,1'-ethylidenebis(4-chlorobenzene)
("DichloroDiphenylethane, 0 aliphatic chlorines" = DDO)
("DichloroDiphenylMonochloroethane,
Saturated" = DDMS)
Cl
C
Cl
Cl
C
Cl
CCl
CHCl
2
1,1'-(2-chloroethenylidene)bis(4-chlorobenzene)
1,1'-(2,2-dichloroethenylidene)bis(4-chlorobenzene)
("DichloroDiphenylMonochloroethylene,
Unsaturated"= DDMU)
("DichloroDiphenyldichloroEtylene" = DDE)
O
Cl
C
Cl
Cl
4
C
Cl
CH
2
,4'-dichlorobenzophenone (DBP)
1
,1'-(ethenylidene)bis(4-chlorobenzene)
("DichloroDiphenylethylene, No aliphatic chlorines,
Unsaturated"= DDNU)
at these respective potentials. They proposed a mechanism
in which each dechlorination step involves a fast electron
transfer and proton addition to give the alkyl radical,
followed by a second, rate-limiting, electron transfer and
proton addition to complete the replacement of -Cl by -H.
This mechanism is consistent with the localization of 98%
of the electron density in the LUMO of DDT in aliphatic C-
Cl σ* antibonding orbitals (8). Four irreversible reduction
steps were also observed by cyclic voltammetry in carefully
prising given that its oxidation peak is only 0.02 V more
positive than that of DDT. Cyclic voltammetry indicated that
all of the oxidation steps were irreversible; the peak poten-
tials for DDT were at +2.3 and +2.6 V vs. SCE.
The ability to remove all three aliphatic chlorines from
DDT by electrolysis suggested that it might be possible to
develop a technology in which DDT could be transformed
electrochemically into less toxic, more biodegradable prod-
ucts, provided that the reaction could be achieved with high
chemical and electrical efficiency under “environmentally
friendly” conditions. An ideal situation would be a sequence
in which “side chain” reduction of DDT to DDO at the cath-
ode would be followed by anodic oxidation of DDO to prod-
ucts that might be compatible with a subsequent biological
treatment.
dried
acetonitrile
(ACN)
and
DMF
using
tetraethylammonium phosphate as the supporting electrolyte
(
9), also suggesting consecutive ECE mechanisms (10).
Electrochemical oxidation of DDT has been studied less
extensively than reduction. Keenan and Stuart (11) oxidized
DDT at +2.10 V vs. SCE in 5% aq 0.1 M tetraethylammo-
nium perchlorate:95% ACN solution at a Pt anode, and
obtained 4,4′-dichlorobenzophenone (DBP), 1,1,1-trichloro-
Experimental
2-(4-chlorophenyl)-2-(4-chloro-2,5-benzoquinone)ethane and
a small amount of 2,2,2-trichloro-1,1-bis-(4-chlorophenyl)
ethanol (Kelthane). The low yield of Kelthane was not sur-
DDT, DDD, DDE, 4,4′-dichlorobenzophenone (DBP), and
tetraethylammonium chloride (TEACl) were obtained from
©
1999 NRC Canada

Merica et al.
1283
Aldrich; a standard solution of DDMU in methanol was ob-
tained from AccuStandard. DDNU, DDO, and DDMS were
synthesized as described below. All compounds were as-
sayed by HPLC and (or) GC–MS. Sodium sulfate supporting
electrolyte and HPLC grade solvents were obtained from
Fisher. Triton SP-175® was generously supplied by Union
Carbide Canada.
ture at 2 mL/min was employed. All solvents were HPLC
grade and filtered using 0.45 µm Nylon 66 membranes.
GC–MS
The instrument was a Hewlett Packard 5971 Series mass
selective detector interfaced to a Hewlett Packard 5890 Se-
ries II gas chromatograph equipped with a J&W DB5 capil-
lary column (0.25 mm i.d. × 30 m, 0.25 µm thick film) using
helium as the carrier gas. The temperature program used for
identification of electroreduction products was 100°C (3 min);
100–180°C at 12°/min; 180–300°C at 7°/min, 300°C (5 min).
In the case of electrooxidation, the following program was
employed: 70°C (3 min); 70–200°C at 20°/min; 200–300°C
at 15°/min.
DDNU
The method employed was part of a longer synthesis by
Grummitt et al. (12). 2,2,2-Trichloro-1,1-bis-(4-chlorophenyl)
ethanol (1.0474 g, 3.92 mmol) was heated under reflux in an
oil bath at 210–220°C for 15 min. After cooling to 80°C and
addition of 1.0 mL of 20% aq H SO , the mixture was stirred
2
4
at reflux at 140–150°C for 1 h. Cooling and two recry-
stallizations from ethanol (10 mL) gave pale yellow needles
Electrolyses
of DDNU (0.4884 g, 1.96 mmol, 50.0%, mp 83–84°C, lit.
The solvents used in exhaustive electrolyses were HPLC
grade methanol, HPLC grade acetonitrile, reagent grade t-
BuOH (which was purified by successive recrystallization),
and Triton SP-175 emulsion (surfactant:heptane:water =
0.1:1:98.9 v/v). The latter contained 0.05 M Na SO as sup-
1
(
12) 84–86°C); H NMR (CDCl ) δ (ppm): 5.45 (s), 7.3 (m).
3
DDO
DDNU (0.3639 g, 1.461 mmol), AlCl (10 mg), 5% Pd–C
3
2
4
(
77.1 mg), and cyclohexene (4.0 mL) were stirred for 5 days
porting electrolyte; it was prepared by sonication and was
indefinitely stable at neutral pH.
under dry argon. Filtration and removal of solvent gave a
yellow oil, which was chromatographed over silica gel (60–
2
00 mesh, 22 g) and eluted with hexane (5 mL fractions).
Continuous flow electroreductions
These experiments were performed using a sandwich-type
flow-through cell previously described (13). Current-
Fractions were checked for composition by TLC, and the
presence of DDO was verified by GC–MS. The DDO-rich
fractions were pooled, evaporated, and crystallized from
2
controlled electroreduction at a Hg pool cathode (A = 6 cm )
2
9
5% ethanol (1 mL) to give white fluffy needles of DDO
was performed using a Ti/IrO anode (A = 20 cm ) and satu-
2
(
(
40.6 mg, 0.162 mmol, 11.1%, mp 44–45°C); 1H NMR
rated Ag/AgCl reference electrode. Prior to electrolysis, the
catholyte was purged with argon for 20 min, and a slow argon
flow was maintained throughout the experiment. A 25.0 mL
sample of 1 mM, DDT 0.05 M TEACl in methanol was circu-
lated through the cathodic compartment at 2.5 mL/min using
a peristaltic pump, and 25 mL of TEACl–methanol was
pumped through the anodic compartment. Samples (0.1 mL)
were taken at intervals during the electrolysis. Extraction of
the electrolyzed solution with CH Cl allowed for product
CDCl ) δ (ppm): 1.6 (d), 4.1 (q), 7.3 (m). The literature
3
melting point for DDO is 54–55°C (4); however, GC–MS
and HPLC analysis showed contamination by traces of
DDNU even after purification.
DDMS
2,2-Bis-(4-chlorophenyl)ethanol (201.6 mg, 0.755 mmol)
and SOCl (0.500 mL, 7.15 mmol) were refluxed for 2 h with
2
2
2
stirring under dry air. The reflux condenser was removed, al-
analysis by GC–MS.
lowing most of the SOCl to evaporate while the solution
boiled for a further 15 min, then the residue was stirred for
Current-controlled electroreductions were also performed
2
2
using a cathode of deposited lead (A = 40 cm ) and an anode
2
1
h and the remaining SOCl allowed to evaporate overnight.
of Ti/IrO (A = 20 cm ). Deposition on a lead wire frame
2
2
The gummy yellow product (121.4 mg, 0.425 mmol, 56.3%)
was dissolved in hexane (1 mL), chromatographed over silica
gel (60–200 mesh, 20 g), and eluted with hexane (10 mL frac-
tions). The composition of fractions was checked by TLC and
GC–MS. The fractions containing DDMS with no significant
was accomplished by circulating 50 mL of 10 mM Pb(NO3)2
+ 125 mM disodium citrate through the cathodic compart-
ment while applying a current of 20 mA for 45 min. A
25.0 mL sample of 0.1 mM, DDT in Triton SP-175 emulsion
was circulated through the cathodic compartment at
5 mL/min, and 25 mL emulsion was circulated through the
anodic compartment. The reservoir of the cathodic emulsion
was stirred magnetically, and heptane was added periodi-
cally to maintain the emulsion. Samples (0.5 mL) taken at
intervals during electrolysis were treated with 5 M H SO
(50 µL) and heptane (0.25 mL) to destroy the emulsion and
to extract DDT and reduction products. Reaction rates were
corrected for the consequent change in the volume of the so-
lution.
contaminants were pooled to give a very pale yellow oil,
1
DDMS (27.4 mg, 0.0959 mmol, 12.7%); H NMR (CDCl ) δ
3
(ppm): 4.0 (d), 4.3 (t), 7.3 (m).
HPLC
The system comprised a Waters U6K injector, Waters
00E system controller, and Model 486 tuneable absorbance
detector set to λ = 240 nm (electroreduction) and 227 nm
electrooxidation). The system was connected to a PC con-
2
4
6
(
trolled by Millennium 2010 chromatographic software. Ana-
lytical separations were performed on a Waters µBondapack
C , 3.9 × 300 mm column. The column was eluted with
Current-controlled electroreductions were also performed
using a cathode of deposited nickel (from a 1:1 NiSO :NiCl
solution, A = 40 cm ) on a nickel wire frame. Experimental
conditions were comparable as for current-controlled
electroreduction on deposited lead.
4
2
2
1
8
5
5/45 ACN/water at 2.5 mL/min, for electroreduction exper-
iments, while for electrooxidation a 90/10 MeOH/water mix-
©
1999 NRC Canada

1
284
Can. J. Chem. Vol. 77, 1999
Single-flow electroreductions
A peristaltic pump passed 0.1 mM DDT in emulsion from
a 500 mL container through the cathodic compartment at
bration time of 10 min was allowed before samples were
collected.
Carbon dioxide was quantitated using the single-flow pro-
cedure, by directing the output from the cell to a closed con-
tainer connected to a methanol trap (3 mL) followed by a
0
.4 mL/min and circulated 25 mL emulsion through the an-
odic compartment. Unlike the continuous flow electrolyses,
the output from the cathodic compartment was directed to
waste. Samples were taken from the cathodic compartment
output and treated with H SO and heptane as in continuous
Ba(OH) trap (25 mL of 3 mM). During the sample collec-
2
tion period (10 min), Ar was flushed through the sample.
The Ba(OH) solution was allowed to stand for few hours to
2
4
2
flow experiments. To achieve a steady state, current was ap-
plied for 30 min before sampling.
clear, and the residual hydroxide ion was then titrated
against 10 mM HNO3.
Anodic voltammetry
The electrochemical cell was a glass container with a Tef-
lon lid which had four openings for electrodes and an argon
Results and discussion
2
Electroreductions
purging tube. The anode was a Pt wire (0.20 cm ); an exter-
We first performed continuous flow electrolysis of 1 mM
DDT in methanol with TEACl as the supporting electrolyte
using an Hg pool cathode, to allow product identification
corresponding to the reduction waves observed polarogra-
phically. Methanol was chosen because it is inexpensive,
miscible with water, and not toxic to microorganisms. Al-
though TEACl is too expensive an electrolyte for use on an
industrial scale, tetraalkylammonium salts have been used in
most of the published voltammetric studies involving DDT
nal reference Ag/AgCl electrode was used, with a bridge
filled with two solutions. The part that was in contact with
the Ag/AgCl electrode contained saturated aqueous KCl so-
lution, while the part that was in contact with the analyzed
solution was filled with 0.1 M LiClO – acetonitrile. The
4
2
counter electrode was a Pt foil with area of 5 cm . The solu-
tions were deaerated prior to electrolysis. The cell was con-
trolled by a EG&G, model 273, potentiostat–galvanostat
interfaced with a PC and operated by M270 electrochemical
software. The following parameters were used: pulse height,
(
1, 5–7, 10, 13).
–2
–
1
After 20 min of amperostatic electrolysis at 2 mA cm ,
1
00 mV; pulse width, 50 ms; scan rate, 2 mV s . The solu-
most of the DDT had reacted. DDE was the major product;
DDD, DDMU, and a small amount of DDMS were minor
products. After 60 min, DDE was still the major product; the
concentration of DDT was not greatly altered, but the relative
amounts of DDMU and DDMS had increased, and DDNU
and DDO became detectable. Similar products were obtained
by Schweizer et al. (1) who electrolyzed 10 mM DDT in a
bicontinuous emulsion of dodecane, didodecyldimethyl-
ammonium bromide (DDAB), and water. They observed two
reduction waves at –1.4 and –2.0 V vs. Ag/AgBr.
Potentiostatic electrolysis at –1.4 V under deoxygenated con-
ditions gave DDE (31%) and DDD (10%) at 51% conversion
of DDT, with minor amounts of DDO, DDMU, and DDMS.
Electrolysis at –2.0 V to 100% conversion gave more exten-
sive reduction to DDO (55%), DDMS (27%), and traces of
tions were purged with argon for 15 min before taking mea-
surements. All potential data were reported using SHE as
common point of reference, and the anodic current was re-
ported as positive.
Electrooxidations
These experiments used the same type of cells as used for
electroreductions. For the electrolyses performed in aceto-
nitrile and in 1:1 t-BuOH:H O, the cell was made of high-
2
density polypropylene, while for the experiments performed
in emulsion, the cell was made of Plexiglas. Care was taken
to select appropriate solvent-resistant tubing for the peristal-
2
tic pump. The working electrode had an area of 25 cm and
was either a Pt foil or a Ti/IrO plate. The counter electrode
2
was a glassy carbon electrode. The cell was controlled amp-
erostatically by means of a EG&G, model 173, potentiostat–
galvanostat. In continuous flow experiments, a 25.0 mL
1,1-diphenylethane. Under aerated conditions, the yield of
1,1-diphenylethane increased to 34%, and 1-(4-chlorophenyl)-
1-phenylethane (22%) and DDO (37%) were also formed.
The formation of DDE and its further reduction products
sample of DDT solution (1 mM DDT – 0.1 M LiClO –
4
acetonitrile; 0.2 mM DDT in 50/50 t-BuOH/H O – 0.05 M
2
DDMU and DDNU is not compatible with DDT → DDD →
DDMS → DDO as the proposed pathway for electroreduc-
tion (2, 6, 7, 10). One possible mechanism is an initial re-
Na SO ; or 0.1 mM DDT – emulsion – 0.05 M Na SO ) was
2
4
2
4
continuously recirculated through the anodic compartment
–
1
using a peristaltic pump (flow rate = 5 mL min ), and
5.0 mL of electrolyte was pumped through the cathodic
·–
–
·
duction to DDT and loss of Cl , giving Ar CH-CCl
2
2
2
radicals, which can disproportionate into DDD and DDE.
Another is base-catalyzed elimination of HCl from DDT, in
which the base is formed in situ at the cathode as the by-
product of reduction of other DDT molecules.
compartment. The electrolysis was monitored by HPLC by
removing samples from the bulk solution; for product identi-
fication, the electrolyzed solution was extracted in dichloro-
methane and analyzed by GC–MS. In single-flow
experiments, a solution of 0.1 mM DDT in 0.1 M LiClO –
–
–
–
4
DDT + 2e + MeOH → DDD + Cl + MeO
acetonitrile was pumped from a 0.25 L container through the
anodic compartment at 1 mL/min, and the output from the
anode was directed to waste. Approximately 25 mL of elec-
trolyte solution was recirculated through the cathodic com-
partment, as in continuous flow experiments. Both solutions
were purged with argon for -30 min before electrolysis to
Experimental evidence in favour of the second option was
–
3
–1
obtained by allowing DDT (10 mol L in methanol) to re-
act with different concentrations of methanolic sodium
methoxide for time periods up to 140 min. After 50 min, for
example, the fraction of DDT converted to DDE was 0.0022
–
4
–1
–3
–1
remove dissolved CO . After current was applied, an equili-
(10 mol L NaOMe), 0.015 (10 mol L NaOMe), and
2
©
1999 NRC Canada

Merica et al.
1285
Fig. 3. Electroreduction of 0.1 mM DDT in 0.5 M Na SO –
Table 1. Pseudo-first-order rate constants for electrochemical
reduction of 0.1 mM DDT in Triton SP-175 emulsion – 0.05 M
Na SO at deposited Pb and Ni cathodes.
2
4
2
emulsion at a Pb cathode (A = 40 cm , single flow, F =
–
1
0.4 mL min ).
2
4
–5
–1
Cathode
Pb
Current density/mA cm–2
(k ± SE) × 10 /s
0.5
1
3
1.7 ± 0.3
8.5 ± 0.3
7.3 ± 0.8
14 ± 0.7
18 ± 0.5
2.2 ± 0.5
5.8 ± 0.5
5.5 ± 0.7
5
1
0
Ni
0.5
3
5
Fig. 2. Electroreduction of 0.1 mM DDT in 0.5 M Na SO –
2
4
2
–2
emulsion at a Pb cathode (A = 40 cm , i = 5 mA cm ,
–
1
continuous flow, F = 5 mL min ).
DDE was not formed, consistent with the concentration of
–
base (OH in this case) being low in the emulsion phase.
Figure 2 shows the progress of the reaction at Pb using a
–
2
current density of 5 mA cm . No more than traces of DDD
were present at any stage of the reaction, indicating that this
intermediate was immediately reduced to DDMS before re-
entering the bulk solution. Mass balances at the Pb cathode
–
2
were at least 97% at current densities ≤3 mA cm ; at higher
current densities, loss of heptane caused by H evolution led
2
to deposition of products from the solution.
The electroreductions at Ni also followed pseudo-first-
order kinetics, but gave poorer mass balances (typically -80%
after 3 h electrolysis). HPLC analysis showed an unidenti-
fied peak with a retention time more than twice as long (less
polar) than any of the standard DDT derivatives. The identi-
fied reduction products were also counter-intuitive. At 0.5
0
.14 (10–2 mol L–1 NaOMe). Detectable amounts of DDE
were even obtained at time zero (immediate quench of the
solution). Direct comparison with the electrolytic experiments
is not possible because the local concentration of MeO in
–
2
mA cm , the major product was DDO, corresponding to
complete dechlorination of the side chain; DDMS was the
–
–
2
the vicinity of the cathode is unknown.
major product at 3 mA cm and was formed exclusively at
–
2
The bicontinuous emulsion used by Schweizer et al.
5 mA cm . We anticipated that increasing the current den-
sity, and hence the polarization of the cathode, would in-
crease the extent of dechlorination, not the reverse. A possible
explanation is competing reduction by chemisorbed H2,
given that nickel is well known as a hydrogenation catalyst.
(
1) employs large volumes of surfactant and organic solvent
and is only 22% water. We showed previously (14) that
aqueous emulsions of 1% heptane, 0.1% Triton SP-175 were
stable for hours of electrolysis when stirred. We therefore in-
vestigated the use of this medium for the electroreduction of
–
2
Single-flow electrolyses (i = 0.1–10 mA cm ) were per-
formed at Pb in order to determine the current efficiency in
the emulsion medium. There was no significant disappear-
0
.1 mM DDT. Lead was selected as a cathode based on our
previous work (14), and nickel was selected based on the
work of Liu et al. (15) involving electroreduction of carbon
tetrachloride. Both metals were freshly deposited prior to
electrolysis of DDT. Small amounts of heptane had to be
added periodically when the current density was greater than
–
2
ance of DDT at current densities <1 mA cm ; at these low
current densities, the polarization of the cathode was insuffi-
cient for efficient electron transfer. The yield of DDMS was
–
2
nearly constant at i ≥ 1 mA cm , while the yields of DDD
–
2
–2
3
mA cm to replace heptane lost through ohmic heating of
and DDO increased with current (Fig. 3). At 10 mA cm ,
the solution and through entrainment by the H evolved. So-
84% of the DDT was reduced to DDD (24%), DDMS
(19%), and DDO (41%). The current efficiency was greatest
at i = 1 mA cm but, even so, was only 0.2%.
2
dium sulphate was selected as an inexpensive, nontoxic elec-
trolyte. The electrolyses were performed in the same way as
those in methanol except that a two-electrode system was
used. The loss of DDT followed first-order kinetics in all
cases: see Table 1.
In the Triton SP-175 emulsion, the products were consis-
tent with the pathway DDT → DDD → DDMS → DDO.
–
2
Schweizer et al. (1) also studied “mediated” (16) electro-
chemical reduction in the bicontinuous emulsions, using
2
+
Co(bipy)3 as a mediator. Significant electrolysis occurred
at the relatively positive potential of –1.0 V, consistent with
+
2+
Co(bipy) , formed by reduction of Co(bipy) , acting as
3
3
©
1999 NRC Canada

1
286
Can. J. Chem. Vol. 77, 1999
Table 2. Electrooxidation of 0.1 mM DDT – 0.1 M LiClO at Pt and Ti/IrO anodes.
4
2
Electrode
Pt
IrO /Ti
2
i/mA cm–2
DDT left/%
DBP/%
0.5
58.4
—
37.9
96
1
42
8.8
42.9
93
3
21.4
5.4
65.7
92
5
7.8
2
85.7
95
0.5
81.6
1.9
14.3
97
1
3
14
24.5
57.1
94
5
66.8
6.2
21.4
94
9.4
14.2
71.4
94
CO /%
2
Mass balance
Current efficiency/%
31.7
18.2
9.2
7.2
12.1
9.2
8.2
6.1
Fig. 4. Differential pulse voltammograms of 1 mM DDT, DBP,
Fig. 5. Electrooxidation of 1 mM DDT in 0.1 M LiClO –
4
acetonitrile at Pt and IrO /Ti anodes (i = 5 mA cm ).
2
–
2
and HQ in 0.1 M LiClO – acetonitrile at a Pt anode.
4
the reducing agent for DDT. After 3 h, the major products
were DDE (24%), DDMU (17%), DDD (17%), DDMS
are resistant to oxidation, because it is impossible to avoid
concurrent, parasitic oxidation of water.
Using Triton SP-175 emulsion as the reaction medium, no
(
12%) and DDO (24%). Further electrolysis increased the
yield of DDMU at the expense of DDD and DDMS.
loss of DDT was observed at either Pt or Ti/IrO , using vari-
2
–
2
ous current densities 0.2–20 mA cm . In both acetonitrile
and 50/50 t-BuOH/H2O, the starting material was observed
to disappear but the mass balance of organic products was
very poor. Using a Pt anode and current density i = 5
Electrooxidations
Voltammetry of DDT at a Pt anode in ACN – 0.1 M
LiClO revealed half-wave potentials at +2.29 and +2.58 V
4
vs. SHE (cf. ref. 11). Oxidation at +2.29 V involves conver-
sion of DDT to 4,4′-dichlorobenzophenone (DBP), demon-
strated by showing that the second oxidation corresponded
to the oxidation potential of DBP. Differential pulse volt-
ammograms of 1 mM DDT, DBP, and hydroquinone (HQ,
–
2
mA cm , DDT disappeared more than 10 times faster in
–
4
–1
0.1 M LiClO4 – ACN (k = 6.5 × 10 s ) than in 50/50
–
5
–1
t-BuOH/H2O (k = 5.3 × 10 s ). In the former case, the
only product detected was DBP (Fig. 5), but in t-BuOH/H2O,
DBP was accompanied by traces of 2,2,2-trichloro-1-(4-
chlorophenyl)ethanone. In 0.1 M LiClO4 – acetonitrile, the
rate of disappearance of DDT at a Ti/IrO2 anode (i = 5
–
which is known to undergo a 2e oxidation (17)) in 0.1 M
LiClO – ACN at a Pt anode (Fig. 4) were obtained. The rel-
4
ative heights of these voltammetric peaks show that both
–
–2
–4 –1
DDT and DBP undergo 4e oxidations.
mA cm ) was slower (k = 5.8 × 10 s ) than at Pt; DBP
was again the major organic product, along with traces of
chlorophenol and 2,2,2-trichloro-1-(4-chlorophenyl)ethanone.
A rationalization of these products involves proton loss from
Continuous flow electrolyses were performed in various
solvents: 0.1 M LiClO – acetonitrile, 0.05 M Na SO in
4
2
4
5
0/50 t-BuOH/H O, and 0.05 M Na SO – Triton SP-175
2 2 4
·+
·
emulsion. The disappearance of starting material and appear-
ance of products were monitored by removing samples from
bulk solution and analyzing them by HPLC. Pt and Ti/IrO2
DDT to give Ar2C -CCl3, followed by either attack of O2
(from the atmosphere) at the radical centre or a second oxi-
dation to Ar2C -CCl3 and nucleophilic attack by water.
+
were used as anodes; Ti/IrO is an example of a “dimen-
We considered that the poor mass balance of organic
products might result from “electrochemical combustion”
(18) of DDT and intermediate products at the anode. We
2
sionally stable anode,” the mechanism of action of which
has been proposed to involve surface-generated hydroxyl
radicals as the reactive oxidant. This type of anode was of
interest in that the hydroxyl radicals might be able to attack
DDT (or its reduction products) at potentials less positive
than those needed for direct oxidation of the substrate. This
is an important consideration for substances like DDT that
therefore modified the apparatus to trap any CO formed
2
during electrooxidation in ACN – 0.1 M LiClO in Ba(OH)
solution. The residual amount of HO was determined by ti-
tration. Experiments were performed at both Pt and Ti/IrO2
4
2
–
using various current densities. Once CO was included,
2
©
1999 NRC Canada

Merica et al.
1287
mass balances improved to ≥92%; current efficiencies
ranged up to >30% (Table 2), with the highest current effi-
ciencies being achieved by operating at the smallest practi-
cal current densities. The first step in the oxidation yields
DBP; however, at the potential required to oxidize DDT,
DBP is further oxidized at the electrode surface.
References
1
. S. Schweizer, J.F. Rusling, and Q. Huang. Chemosphere, 28,
61 (1994).
9
2
. H. Fukami and M. Nakajima. Botyu-Kagaku, 18, 6 (1953),
through Chem. Abstr. 47, 5815g (1953).
3
4
5
. A.S. Reddy and S.J. Reddy. Indian J. Chem. 23A, 619 (1984).
. F.L. Lambert and G.B. Ingall. Tetrahedron Lett. 3231 (1974).
. N.J. Bunce, S.G. Merica, and J. Lipkowski. Chemosphere, 35,
Conclusions
2
719 (1997).
At a Pb cathode, DDT is sequentially reduced to less chlo-
rinated derivatives, particularly DDMS and DDO. No loss of
chlorine is observed from the phenyl rings, consistent with
our earlier experiments with polychlorinated benzenes (13,
6. I. Rosenthal and R.J. Lacoste. J. Am. Chem. Soc. 81, 3268
(1959).
7. R. Tandon, I. Bala, and M. Singh. Z. Phys. Chem. Leipzig,
267, 340 (1986).
8. F.A. Beland, S.O. Farwell, P.R. Callis, and R.D. Geer. J.
Electroanal. Chem. 78, 145 (1997).
1
4). Reduction takes place in an almost entirely aqueous me-
dium, based on the new surfactant Triton SP-175, using so-
dium sulfate as an environmentally friendly supporting
electrolyte. Mass balances are excellent at all stages of the
reaction, but the current efficiency of the process (0.2%) is
completely unacceptable for any practical technology to be
contemplated.
9
. G.I. Udovich. An electrochemical investigation of the reduc-
tion of DDT in ethanol, acetonitrile and N,N-
dimethylformamide and the development of a method for the
differential pulse polarographic determination of DDT. Ph.D.
thesis. University of Wyoming, Wyoming. 1979.
1
0. R.S. Nicholson and I. Shain. Anal. Chem. 37, 178 (1965).
1. R.R. Keenan and J.D. Stuart. J. Electrochem. Soc. 123, 1637
1976).
2. O. Grummitt, A.C. Buck, and E.I. Becker. J. Am. Chem. Soc.
7, 2265 (1945).
Conversely, in the case of electrooxidations, high current
efficiencies (>30%) were obtained, but only in a nonaqueous
solvent; this time, it is the solvent that is technologically im-
practical. The potential required to oxidize DDT is suffi-
ciently positive that neither an aqueous cosolvent (t-BuOH)
nor an aqueous surfactant allowed oxidation of DDT to com-
pete successfully with oxidation of water. In acetonitrile, the
highly positive potential required to oxidize DDT leads to
concurrent oxidation of all reaction intermediates, and “elec-
1
1
1
1
1
(
6
3. S.G. Merica, N.J. Bunce, W. Jedral, and J. Lipkowski. J. Appl.
Electrochem. 28, 645 (1998).
4. S.G. Merica, C.E. Banceu, W. Jedral, J. Lipkowski, and N.J.
Bunce. Environ. Sci. Technol. 32, 1509 (1998).
5. Z. Liu, R.G. Arnold, and E.A. Betterton. In Physical, chemi-
cal, and thermal technologies remediation of chlorinated and
recalcitrant compounds. Edited by G.B. Wickramanayake and
R.E. Hinchee. Battelle Press, Columbus. 1998. p. 473.
6. J.W. Sease and R.C. Reed. Tetrahedron Lett. 393 (1975).
trochemical combustion” gives CO as a major product.
2
Acknowledgements
1
We thank the Natural Sciences and Engineering Research
Council of Canada for financial support, Union Carbide for
providing Triton SP-175 surfactant, and a referee for sug-
gesting base-catalyzed elimination from DDT.
17. M. Gattrell and D.W. Kirk. J. Electrochem. Soc. 140, 1534
(1993).
1
8. C. Comninellis and C. Pulgarin. J. Appl. Electrochem. 21, 703
1991).
(
©
1999 NRC Canada