Electrochemical hydrodehalogenation of chlorinated phenols in aqueous solutions I. Material aspects

Article abstract of DOI:10.1149/1.1531971

Pentachlorophenol (PCP) and 2,4-dichlorophenol (DCP) have been hydrodehalogenated by electrochemical reduction in aqueous solutions at ambient and elevated temperatures. Galvanostatic and/or potentiostatic hydrodehalogenation (HDH) was carried out in conventional H-cells or solid polymer electrolyte cells operated in batch or/and batch-recycle modes. The processes were monitored by both chloride ion analysis and high performance liquid chromotography product analysis. The effect of the cathode, separator, and cell type on the rate and efficiency of HDH of 1 mM DCP and 0.071 mM (saturated) PCP in 0.05 M Na₂SO₄/H₂SO₄ (pH 3) and/or pure water are reported. Several types of cathodes, e.g., Fe gauze and foil, Pd/Fe gauze, Pd/Fe foil, carbon cloth, and Pd/carbon cloth, were tested. Palladized cathodes showed high catalytic activity and long term stability for the HDH processes. Both cation and anion exchange membranes were employed to separate the divided H-cells and solid polymer electrolyte cells. Complete HDH of 1 mM DCP with a current efficiency of 70% and an energy consumption below 20 kWh/kg DCP, was realized in an H-cell with a Pd cathode at ambient temperature. Similar results were obtained for HDH of 0.071 mM PCP but with lower current efficiency, e.g., 16% and higher energy consumption, e.g., 80 kWh/kg PCP. The cell gave a current efficiency of 15% and an energy consumption of 11.6 kWh/kg DCP for complete HDH of 1 mM DCP in pure water solution. For complete HDH of 0.071 mM PCP in the SPE cell, current efficiency and energy consumption were 10% and 90 kWh/kg PCP.

Full text of DOI:10.1149/1.1531971

Journal of The Electrochemical Society, 150 ͑2͒ D17-D24 ͑2003͒
D17
0013-4651/2002/150͑2͒/D17/8/$7.00 © The Electrochemical Society, Inc.
Electrochemical Hydrodehalogenation of Chlorinated Phenols
in Aqueous Solutions
I. Material Aspects
,z
H. Cheng,a K. Scott,a, and P. A. Christensenb,
*
*
b
aDepartment of Chemical and Process Engineering and Department of Chemistry, University of Newcastle
upon Tyne, Newcastle upon Tyne, Northumberland NE1 7RU, United Kingdom
Pentachlorophenol ͑PCP͒ and 2,4-dichlorophenol ͑DCP͒ have been hydrodehalogenated by electrochemical reduction in aqueous
solutions at ambient and elevated temperatures. Galvanostatic and/or potentiostatic hydrodehalogenation ͑HDH͒ was carried out in
conventional H-cells or solid polymer electrolyte cells operated in batch or/and batch-recycle modes. The processes were moni-
tored by both chloride ion analysis and high performance liquid chromotography product analysis. The effect of the cathode,
separator, and cell type on the rate and efficiency of HDH of 1 mM DCP and 0.071 mM ͑saturated͒ PCP in 0.05 M
Na2SO4 /H2SO4 ͑pH 3͒ and/or pure water are reported. Several types of cathodes, e.g., Fe gauze and foil, Pd/Fe gauze, Pd/Fe foil,
carbon cloth, and Pd/carbon cloth, were tested. Palladized cathodes showed high catalytic activity and long term stability for the
HDH processes. Both cation and anion exchange membranes were employed to separate the divided H-cells and solid polymer
electrolyte cells. Complete HDH of 1 mM DCP with a current efficiency of 70% and an energy consumption below 20 kWh/kg
DCP, was realized in an H-cell with a Pd cathode at ambient temperature. Similar results were obtained for HDH of 0.071 mM
PCP but with lower current efficiency, e.g., 16% and higher energy consumption, e.g., 80 kWh/kg PCP. The cell gave a current
efficiency of 15% and an energy consumption of 11.6 kWh/kg DCP for complete HDH of 1 mM DCP in pure water solution. For
complete HDH of 0.071 mM PCP in the SPE cell, current efficiency and energy consumption were 10% and 90 kWh/kg PCP.
© 2002 The Electrochemical Society. ͓DOI: 10.1149/1.1531971͔ All rights reserved.
Manuscript received June 11, 2002. Available electronically December 23, 2002.
Halogenated liquid wastes are routinely produced in industrial
processes, e.g., approximately 1 million tonnes of such wastes are
generated annually in the U.K.1 Until very recently, the disposal
practice for these wastes was landfill for short-chain chlorinated
organics, incineration for resistant and intractable compounds such
as polychlorobiphenyls ͑PCBs͒ and pesticides, and use as a fuel
supplement in cement kilns where a high calorific value of the or-
ganic made this a practical proposition. Disposal to landfill is now
virtually precluded by the Environment Agency. Incineration has a
number of serious drawbacks, such as high capital costs of plant,
processing, and transport, production of harmful substances, e.g.,
dioxin and adverse public reaction. Therefore, other routes have
been explored, such as bioremediation,2 chemical and electrochemi-
cal dehalogenation.2-4
Bioremediation has been applied to dehalogenate a wide var-
iety of halogenated compounds using the metabolism of
microorganisms.2 Bioremediation greatly depends on the ability of
microorganisms to survive in an environment containing haloge-
nated compounds. A more challenging issue is that the products of
bioremediation are often toxic and, in some cases, may be more
harmful to human health than the parent compounds.4 Microorgan-
isms can evolve relatively quickly to develop biochemical traits but
in some cases, long-term operation is necessary, e.g., months for the
bioremediation of PCBs.5
Hydrodehalogenation ͑HDH͒ was considered as a low-waste
technology for detoxifying organic halogenated waste and regenera-
tion of the initial raw materials.6 Chemical dehalogenation has been
developed as an alternative to incineration for disposal of haloge-
nated organic compounds, which can proceed several ways, e.g.,
catalytic and reactive procedures. The reactive dehalogenation in-
volves the use of relatively expensive chemicals, such as LiAlH4 or
NaBH4 , as hydrogen donors, and is therefore considered only for
preparative synthesis.7 Catalytic HDH of organic halogenated com-
pounds is accepted as a practical choice at the moment and can be
carried out in the gas8 or liquid phase.9 As there are harsh conditions
in catalytic HDH, e.g., high temperature ͑above 400°C in most
cases͒,10 there are requirements for thermal, mechanical, and chemi-
cal stability of catalytic reactor components. More severely, catalytic
HDH often takes place at a gradually decreasing rate through pro-
gressive poisoning of the catalyst in some cases. This is accompa-
nied by a rapid deactivation of the catalyst. Low-temperature cata-
lytic HDH of bromobenzene was performed, e.g., at 40°C, but the
results and other reaction conditions are unacceptable for industry.7
More competitive ways of chemical HDH have been developed
based on zero-valent metals such as iron, known as dissolving metal
reductions.2,3 Over the last few years, it has been shown that these
metals can effect the HDH of a range of chlorinated organic com-
pounds in contaminated groundwater.2,3,11,12 Chemical HDH using
iron alone suffered from slow reaction rates, particularly for aro-
matic halogenated compounds, under ambient temperature and
pressure.5 Partial HDH of PCBs by iron was only achieved under
high-temperature and high-pressures, e.g., 250°C, 10 MPa.13
Interestingly, the deposition of small amounts, ca. 0.05 wt %, of
Ni or Pd onto the iron has been found to significantly enhance the
rate of HDH, e.g., one to two orders of magnitude for trichloroeth-
ylene ͑TCE͒, compared to the rate at Fe alone which extends the
range of halogenated organic compounds amenable to treatment
from TCE to polychlorinated biphenyls.14,15
Recently, the electrochemical HDH of chlorinated organic com-
pounds has been explored as another means to replace incineration
for disposal of halogenated organic wastes.16-21 The cathode mate-
rial has been found to have a major effect on the efficiency of the
electrochemical HDH of organic halogenated compounds. A typical
example is the HDH of 1,2,3,5-tetrachlorobenzene ͑TCB͒ and chlo-
robenzene ͑CB͒ in methanol or dimethylsulfoxide and acetonitrile
͑with 0.25 M tetraethylammonium bromide͒ at a cathode potential
of Ϫ3.3 V vs. Ag/AgCl.17 More than 95% conversion of 12 mM CB
was achieved with a current efficiency of 15-20% on the carbon
cloth or Pb cathodes. On the other hand, Pt, Ti, and Ni cathodes only
gave current efficiencies of ca. 5% and lower conversions.
The electrochemical dechlorination of 153 ppm 4-chlorophenol
to phenol in 0.05 M sodium acetate-acetic acid solution has been
reported. A conversion, up to 100% dechlorination, of 153 ppm
4-chlorophenol was achieved by using palladized carbon cloth
or graphite cathodes after 15 h electrolysis.18
The majority of work examining the effect of cathode materials,
even using Fe as a cathode, concentrated on mechanistic analysis
rather than practical applications. Moreover, environmentally unac-
* Electrochemical Society Active Member.
z E-mail: k.scott@ncl.ac.uk
Downloaded on 2015-03-18 to IP 134.71.135.75 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

D18
Journal of The Electrochemical Society, 150 ͑2͒ D17-D24 ͑2003͒
ceptable materials, such as Hg and Pb, have been used as cathode
materials,17,19 which makes such techniques unattractive for
industry.
Based on the above work on the HDH of halogenated organic
compounds, the aim of the work reported in this paper is to find a
technical way to effect the electrochemical HDH of halogenated
organic compounds in aqueous solutions. The first stage focused on
the HDH of 2,4-dichlorophenol ͑DCP͒ and pentachlorophenol ͑PCP͒
because of their importance in industrial waste-water treatment. In
Europe, the acceptable level of PCP in wastewater is in the range of
0.1-1 ppm, and that of its reduction product, DCP, is 4 ppm.20
In drinking water, the acceptable level is extremely low, i.e., only 1
ppb.21 In common with many halogenated organic compounds, it is
not possible to treat PCP and DCP routinely by common bioreme-
diation techniques, due to their toxicity to the bacteria employed.5
One accepted approach to eliminate PCP from water is absorp-
tion using granulated activated carbon, which causes some technical
and economic problems in the disposal of PCP-containing carbon
materials.21 Anodic dehalogenation of PCP in aqueous solutions has
been examined as an alternative to adsorption, via electrochemically
initiated condensation reactions at high-surface-area carbon felt an-
odes. In the best case, a current efficiency of 100% for the electro-
chemical condensation process was achieved in the controlled po-
tential electrolysis of 0.35 mM PCP in 1 M acetate buffer solutions.
However, at potentials less than 0.79 V vs. RHE, PCP was only
partially dehalogenated, and at higher potentials, e.g., 1.9 V vs.
RHE, the products were chlorinated compounds, e.g., 2,3,4,5,6-
pentachloro-4-pentachlorophenoxy-2,5-cyclohexadienone and were
essentially insoluble and caused anode passivation.21
Figure 1. Flow circuit for the electrochemical reductive HDH using a solid
polymer electrolyte reactor in the batch recycle mode. 1. Power supply. 2.
Catholyte reservoir. 3. Cell. 4. Thermostatic bath. 5. Catholyte. 6. Magnetic
stirrer. 7. Pump. 8. Cathode. 9. Membrane. 10. Anode. 11. Anolyte.
Another option is electrochemical HDH of DCP and PCP
Cl2C6H3OH ϩ 4eϪ ϩ 2Hϩ → C6H5OH ϩ 2ClϪ
Cl5C6OH ϩ 10eϪ ϩ 5Hϩ → C6H5OH ϩ 5ClϪ
͓1͔
͓2͔
The only reported work on the electrochemical HDH of PCP was
using a flow-through cell with carbon fiber cathodes operated in a
batch-type recycle mode.20 The HDH was performed in 0.1 M
Na2SO4/0.1 M NaOH solution at constant current densities. At 500
mA cmϪ2, complete dehalogenation of the PCP to phenol and chlo-
ride ions was achieved. However, the current efficiency was very
low, about 1%, and the energy consumption was high, about 400
kWh/kg PCP for 90% conversion. At lower current densities, PCP
was only partially dehalogenated to tetrachlorophenol, dichlorophe-
nol, or monochlorophenol, respectively.20
The overall aim of the work reported in this paper was to develop
a technical process for the dehalogenation of aqueous solutions con-
taining chlorophenols, including PCP and DCP, based upon electro-
chemical HDH. This paper deals with the effect of cell materials,
e.g., cathode, anode, and cell separator, on the efficiency of the
HDH of PCP and DCP. The following paper examines the influence
of operational parameters.
(NH4)2SO4 ͑99%, Aldrich͒, NiSO4 ͑98%, Aldrich͒, H3BO3 ͑99%,
BDH͒, NaOH ͑AnalaR, BDH͒, HCl ͑37%, BDH͒, H2SO4 ͑98%,
AnalaR, BDH͒, Nafion solution ͑5%, Aldrich͒, CH3OH ͑HPLC-
grade, Fisons͒, and acetic acid ͑99.9%, Aldrich͒.
All solutions were prepared using water with a resistance of 18.2
M⍀ cm obtained from a Millipore-Q system. Due to the very low
solubility of PCP in acidic and neutral aqueous solutions, the con-
centrated PCP solutions, above the saturated concentration, were
prepared by dissolving the desired amount of PCP in methanol first
and then diluting that to the required concentration. PCP solutions
with less than 1% methanol, above the saturated concentration, were
in the form of emulsions.
Two types of membranes were used in this work: Nafion 117
membrane ͑DuPont͒ and FuMATech FT-FKE-S membrane ͑FuMAT-
ech͒. Pretreatment of the Nafion 117 membranes is detailed
elsewhere.22 The FuMATech FT-FKE-S membranes were used after
immersion in water for at least 2 h.
Experimental
Cells and apparatus.—Two cells were used, an H-cell and a
solid polymer electrolyte zero gap flow cell. The H-cell, used for
voltammetric measurements and the electrochemical HDH of PCP
and DCP, consisted of two compartments, each with a volume of 80
cm3. Nafion 117 membrane, FuMATech FT-FKE-S membrane, or a
glass frit was employed as the separator between the cell compart-
ments. Nitrogen was bubbled through the catholyte of the H-cell at
the start of the voltammetric experiments. The solid polymer elec-
trolyte cell, operated in a batch recirculation mode for the electro-
chemical HDH of PCP and DCP, was made from either stainless
steel or graphite blocks with machined flow channels. The flow cir-
cuit of the solid polymer electrolyte cell, as shown in Fig. 1, con-
sisted of a laboratory scale two-electrode cell, two pumps ͑H. R.
Flow Inducer, England͒, two reservoirs ͑1 dm3 in volume each͒ for
anolyte and catholyte, respectively, and thermostatic baths ͑B-480
Waterbath, Buchi, Switzerland͒ for temperature control.
Materials and chemicals.—The following materials were all
used as received: Iron gauze ͑99%, BDH͒, iron wire ͑diameter 0.25
mm, 99.9%, Aldrich͒, iron foil ͑99.5%, 2.5 ϫ 2.5 mm, Aldrich͒,
mild steel mesh ͑wire diameter 0.15 mm, open area 35%͒, stainless
steel mesh ͑AISI 304, wire diameter 0.25 mm, open area 37%,
Goodfellow͒ or foil ͑AISI 316, thickness 0.25 mm, Goodfellow͒, Ti
mesh ͑99.6%, open area 37% wire diameter 0.2 mm͒, carbon cloth
͑GC-14, E-TEK Inc.͒, iron powder ͑99.0%, maximum particle size
60 ␮m, Goodfellow͒, palladium-charcoal ͑5% or 10% Pd, BDH͒,
and palladium activated carbon powder ͑30% Pd, Aldrich͒. Reagents
were K2PdCl6 ͑99%, Aldrich͒, PdCl2 ͑99%, Aldrich͒, pentachlo-
rophenol ͑PCP, 98%, Aldrich͒, 2,4-dichlorophenol ͑DCP, 99%, Ald-
rich͒, 4-chlorophenol ͑CP, 99%, Lancaster Synthesis͒, phenol
͑99.9%, Aldrich͒, Na2SO4 ͑97%, Aldrich͒, KCl ͑99%, Aldrich͒, and
NaCl ͑99%, Aldrich͒, NaBH4 ͑99%, BDH͒, FeSO4 ͑99%, Aldrich͒,
Downloaded on 2015-03-18 to IP 134.71.135.75 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 150 ͑2͒ D17-D24 ͑2003͒
In operation, the catholyte and anolyte, each with a volume of
D19
100 cm3, were pumped through the cell and then returned to the
reservoirs for recycling. All electrolyses were carried out under gal-
vanostatic control, ranging from 5 to 100 mA cmϪ2, for a period
between 30 min and 20 h. The concentrations of chloride ion were
measured with an ion analyser and the concentrations of reactant,
intermediates, and products were monitored using HPLC. The data
was used to determine the process efficiency, as a function of time.
Various materials, i.e., iron gauze, iron wire, iron foil, mild steel
mesh or sheet, stainless steel mesh or plate, Ni foam, nickel, palla-
dium, and nickel-palladium on Ti mesh, Pd/carbon cloth, Pd-Ni/
carbon cloth, and Pd-Ni/stainless steel mesh, with geometric areas
between 2 and 10 cm2, were tested as cathodes. The anode was a
platinum mesh ͑about 10 cm2͒. Commercial saturated calomel elec-
trode ͑Russell͒ or Ag/AgCl electrode ͑Radiometer model K801͒ was
used as the reference, although all electrode potentials are reported
on the reversible hydrogen electrode ͑RHE͒ scale. For the measure-
ments using three electrodes, a Luggin probe was used to measure
the electrode potential.
Electrode preparation.—Several catalyzed electrodes, i.e., Fe/Ti
mesh, Ni/Ti mesh, Pd/Ti mesh, Pd-Ni/Ti mesh, Pd/stainless steel
mesh, Pd-Ni/stainless steel mesh, Pd/Fe gauze, and Pd/carbon cloth,
with geometric areas between 2 and 10 cm2, were prepared by
electrodeposition.17,22
Figure 2. Linear sweep voltammograms for electrochemical HDH of PCP
and DCP on palladized cathodes. Cell: H-cell divided by a Nafion 117 mem-
brane. Cathode: as shown in the figure ͑2 mg Pd/cm2, 4 cm2͒. Anode: Pt
mesh ͑10 cm2͒. Catholyte: 0.05 M Na2SO4 ͑pH 3͒ solution without ͑blank͒
or with saturated ͑0.071 mM͒ PCP and saturated ͑20 mM͒ DCP. Anolyte:
0.05 M Na2SO4 ͑pH 3͒ solution. Scan rate: 5 mV sϪ1. Temperature: 21.5
Ϯ 0.5°C.
Deposition of Pd onto iron ͑gauze, wire, or foil͒ and mild steel
from K2PdCl6 was carried out spontaneously via the reaction5
2Fe ϩ PdCl26Ϫ ϭ 2Fe2ϩ ϩ 6ClϪ ϩ Pd
͓3͔
by exposing to 20 mL of a 2.5 mM solution of K2PdCl6 ͑20 mg
K2PdCl6 in 20 mL water͒ for 5 to 15 min. The amount of palladium
deposited on the iron surface was assumed to be the result of a
100% completion of Reaction 3, which was determined by compar-
ing the substrate weights before and after deposition and was be-
tween 2 and 5 mg/cm2 geometric area.
Preparation of metal/membrane cathodes used the ion exchange
method.23 The sandwiched membrane electrode assemblies were
prepared by the hot press method from Nafion-bonded iron powder
or Pd/carbon powders.22
products of the dechlorination during the course of the electrolysis.
The HPLC apparatus consisted of a PU 4010 pump ͑Pye Unicam,
Ltd., England͒, a Discovery™ C18 column ͑5 ␮m particle size and
25 ϫ 0.46 cm, Supelco͒, PU 4020 UV detector ͑Pye Unicam, Ltd.,
England͒, and a LY 16100-11 X-Y recorder. The wavelengths used
in HPLC measurements were determined using UV-vis spectroscopy
͑UV-160A UV-visible recording spectrophotometer, Shimadzu, Ja-
pan͒. Normally, the PU 4020 UV detector was set to 270 nm for
phenol, 280 nm for DCP and CP, and 300 nm for PCP. The mobile
phase was a 0.1 M acetic acid aqueous solution/methanol mixture
͑60/40 by volume͒ with flow rate of 1.0 mL/min. The peaks for
phenol ͑retention time tr ϭ 10.0 min), CP ͑retention time,
tr ϭ 23.9 min), DCP ͑retention time, tr ϭ 32.0 min), and PCP ͑re-
tention time, tr ϭ 42.3 min) were characterized by using standard
solutions. Quantification of phenol production and distributions of
the chlorophenols was accomplished by the use of calibration curves
with standards. A sample volume of 20 ␮L was generally employed.
The detection limit of this method was 0.1 ppm for phenol, CP, and
DCP and 0.5 ppm for PCP.
A number of electrodes were deposited under identical condi-
tions to check reproducibility.
Voltammetric measurements.—All voltammetric measurements
were performed in an H-cell using a Ministat precision potentiostat
͑Thompson Electrochemical, Ltd.͒ with a PCI-100 MK3 computer
interface ͑Sycopel Scientific, Limited͒ controlled by Sycopel Scien-
tific electrochemistry software ͑Sycopel Scientific, Limited͒. All of
the solutions studied were thoroughly degassed using oxygen-free
nitrogen ͑BOC, Ltd͒. To obtain stable and reproducible voltammo-
grams, it was necessary to treat the cathode electrochemically before
collecting data, e.g., three scans between 0.4 and Ϫ1.2 V vs. RHE at
a scan rate of 50 mV/s.
Parameter definitions.—Energy consumption was calculated
according to the following equation24,25
Energy consumption ECN͒ ϭ nFECell /␸M
͓4͔
͑
Batch electrochemical HDH.—Batch electrolyses were per-
formed in the H-cell and the solid polymer electrolyte cell to deter-
mine the HDH efficiency. Galvanostatic electrolysis was generally
used, and the constant currents were provided by a Farnell LS60-5
power supply. Potentiostatic electrolyses were also performed, and
the total charges passed in the experiments were monitored. The
catholytes in the H-cells were magnetically stirred.
where n is the number of electrons passed during the reaction, F is
the Faraday constant, ECell is the cell voltage, ␸ is the current effi-
ciency, and M is the molar mass.
The current efficiency of the HDH reaction was calculated as the
part of current ͑or charge͒ passed that was used to convert the start-
ing PCP or DCP to the products according to Eq. 1 and 2.
Product analysis.—Chloride concentrations were determined us-
ing a pH/ion meter ͑Corning model 135, Corning Glass Works or
Orion model 920A, Orion Research, Inc.͒ and a combination chlo-
ride electrode ͑model 96-17B, Orion Research, Inc.͒. Calibration
curves were obtained from standard solutions, 5 ϫ 10Ϫ6 to 0.1 M
NaCl in pure water or in 0.05 M aqueous Na2SO4 solution, before
and after each experiment.
Results and Discussion
Voltammetric data.—Voltammetric data were collected on a
range of cathodes. Typical linear sweep voltammograms ͑LSV͒ ͑Fig.
2͒ using Pd/Fe gauze as the working electrode showed that the ad-
dition of PCP and DCP to the electrolyte caused an increase in
cathodic current densities, although all curves showed flat profiles at
potentials less negative than Ϫ1.0 V vs. RHE. The rapid increase in
current density with potential more negative than Ϫ1.25 V was ac-
High-performance liquid chromatography ͑HPLC͒ was used to
determine concentrations of starting material, intermediates, and
Downloaded on 2015-03-18 to IP 134.71.135.75 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

D20
Journal of The Electrochemical Society, 150 ͑2͒ D17-D24 ͑2003͒
Figure 3. Comparison between the chemical and electrochemical HDH of
DCP and PCP in terms of chloride ion release. Cathode: Pd/Fe gauze ͑2 mg
Pd/cm2, 4 cm2͒. Reaction conditions: ‘‘Chemical HDH’’ at the open circuit;
‘‘Electrochemical HDH’’ at a potential of Ϫ1.20 V vs. RHE. Electrolyte
volume: 60 mL. Other conditions: see Fig. 2.
Figure 4. Effect of catalyst on chloride ion release during electrochemical
HDH of 1 mM DCP. Cathode: shown in the figure ͑5 mg catalyst cmϪ2, 6
cm2͒. Controlled cathode potential: Ϫ1.0 V vs. RHE. Other conditions: see
Fig. 2.
Ni/Ti mesh and Ti mesh alone. The best performance was obtained
using the Pd/Ti mesh, which dehalogenated 1 mM DCP completely
within 75 min. No improvement in HDH efficiency was observed
with a Fe/Ti mesh ͑not shown in the figure for clarity͒, due to the
rapid corrosion of the Fe deposit. The inactivity of Ni with respect to
the HDH process was something of a surprise given the increased
activity of Fe metal with Ni, and Ni itself towards chemical HDH
reported in the literature for liquid phase14,15 and gas phase26 pro-
cesses, respectively. Moreover, the deposition of Ni onto Pd/Ti mesh
cathode produced the same HDH characteristics as for the Pd/Ti
mesh cathode ͑not shown in the figures for clarity͒.
It was found that hydrogen plays a key role in the HDH and that
both platinum and palladium are excellent materials for electro-
chemical evolution of hydrogen. More important, palladium has the
ability to absorb hydrogen into its lattice and to maintain a high
surface concentration of hydrogen, and platinum lacks this ability.
Consequently, palladium gave higher HDH rates than platinum.18,27
Our data can be explained on these observations, although further
research is necessary to understand the data thoroughly.
companied by rapid evolution of gas ͑hydrogen͒ bubbles at the cath-
ode. The striking feature of the voltammetric responses using Pd/Ti
mesh is that higher cathodic current densities were observed from
more positive potentials than those at the Pd/Fe gauze in both PCP
and DCP solutions ͑Fig. 2͒. Under conditions used in this work, both
HDH and hydrogen evolution reactions occurred simultaneously
even at less negative potentials, so it was impossible to distinguish
the characteristics of these processes from the voltammetric mea-
surements only. Moreover, the voltammetric data could not com-
pletely show the effectiveness of cathode materials for releasing
chloride ion. This was demonstrated by the electrolysis data shown
later. For example, stainless steel mesh and plate cathodes showed
different cyclic and linear sweep voltammograms in the blank or in
the PCP and DCP solutions. But when these cathodes were used to
attempt HDH of PCP and DCP by preparative electrolysis, no chlo-
ride ions were detected. Regardless of this, the preliminary voltam-
metric data suggested that the Fe, Pd/Fe, Pd/carbon cloth, and Pd/Ti
mesh were potential cathodes for the HDH process. The program of
research thus focused on quantitative preparative electrolysis.
The effectiveness of HDH greatly depends on the cathode sub-
strate, as shown in Fig. 5 as a comparison of different cathodes
during the electrochemical HDH of 1 mM DCP in an H-cell with a
Nafion membrane. As expected, uncatalyzed substrates, e.g., Ti, re-
leased very few chloride ions, e.g., below 7 ppm within 2 h elec-
trolysis ͑Fig. 5͒. Fe gauze gave a slightly greater chloride ion re-
lease, but still less than 10 ppm. The Pd/Fe gauze cathode ͑5 mg
Pd/cm2͒ gave more than a twofold increase in chloride ion release.
Higher activities were obtained using the Pd/carbon cloth cathode ͑5
mg Pd/cm2͒, e.g., about 50 ppm at 120 min. This value was even
better than that achieved at a 30% Pd/activated carbon powder cath-
ode, although the latter had higher Pd loading ͑15 mg Pd/cm2͒. The
use of stainless steel as an alternative metal support for Pd produced
a performance similar to that of Pd/Ti mesh, although the time for
complete HDH was increased by some 25%. Both the metal mesh
supported cathodes gave significantly better performances than
those of the carbon cloth electrodes, as shown in Fig. 5.
Comparison of chemical and electrochemical dechlor-
ination.—The rates of HDH of DCP and PCP using Fe or Pd/Fe
electrodes at open circuit were very small, and significantly less than
those observed during electrolysis, see Fig. 3. For example, after 60
min, 16.6 and 5.9 ppm chloride ions were detected from the elec-
trochemical HDH of 1 mM DCP and PCP, respectively. Only 2.4
and 0.97 ppm chloride ions were released from the chemical dechlo-
rination of 1 mM DCP and PCP, respectively ͑Fig. 3͒.
Chloride ion release.—The principal analytical method used to
measure performance was chloride ion release as this provided a
reliable and instantaneous measurement of the HDH process.
Several cathode substrates and catalysts were evaluated in terms
of chloride ion release from the HDH of DCP and PCP, and the latter
was found to have the most significant effect upon the efficiency of
the HDH process. Figure 4 shows chloride ion release data obtained
during the electrolysis of 1 mM DCP in the H-cell using a Nafion
membrane as the separator and three catalyzed cathodes. In terms of
the release of chloride ions, the Ni/Ti mesh gave virtually the same
result as the Ti mesh substrate, whereas the Pt/Ti mesh showed an
almost fourfold increase in chloride ion release compared to the both
A similar cathode effect was observed for the HDH of 0.071 mM
PCP, as shown in Fig. 6. The poor performance at the iron cathodes
mainly resulted from material instability. It was observed that the Fe
gauze and the Pd/Fe gauze cathodes formed surface oxides ͑rusted͒
after 15 min electrolysis and rusting was severe after 2 h. In con-
Downloaded on 2015-03-18 to IP 134.71.135.75 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 150 ͑2͒ D17-D24 ͑2003͒
D21
Figure 5. Effect of cathode substrate on chloride ion release during electro-
chemical HDH of 1 mM DCP. Cathode: shown in the figure ͑5 mg catalyst
cmϪ2, 6 cm2, except for the ‘‘Pd/C powder’’ with 15 mg Pd cmϪ2͒. Con-
trolled current density: 10 mA cmϪ2. Other conditions: see Fig. 2.
Figure 7. Effect of cell and separator on chloride ion release during electro-
chemical HDH of 1 mM DCP. Cell and separator: shown in the figure.
Cathode: Pd/carbon cloth ͑5 mg Pd/cm2, 9 cm2͒. Anode: Pt mesh ͑10 cm2͒.
Controlled current: 200 mA. Catholyte solvent and anolyte: 100 ml 0.05 M
Na2SO4 ͑initial pH 3͒ solution. Flow rate for the solid polymer electrolyte
cell: 150 mL/min. Temperature: 20.5 Ϯ 0.5°C.
trast, the Pd/carbon cloth, Pd/stainless steel mesh, and Pd/Ti mesh
cathodes showed great stability under the HDH conditions. Particu-
larly, there was no visual damage of the Pd/Ti mesh cathode after
more than 50 h of use. The palladized mesh cathodes have much
more open areas, which led to better mass transport between PCP
and the catalysts and thus higher HDH rates, compared to the pal-
ladized carbon cloth. In the case of the stainless steel-supported
cathodes, loss of Pd from the surface was observed after use. Thus
the stainless steel supported cathode would not seem to be a viable
cathode for long term HDH, at least not with the particular prepa-
ration procedure used.
Pd/carbon powder was made using bonded palladized carbon pow-
der. For the Pd/carbon powder, quite severe mass-transport limita-
tions could occur in the carbon cloth substrate and the bonded cata-
lyst region, especially when a large amount of hydrogen gas was
generated. The system also relied on good ionic contact between the
Pd and Nafion membrane. As a consequence of these factors, the
Pd/carbon powder cathode showed poorer HDH efficiency than the
Pd/carbon cloth cathode.
In this work, an objective was to explore for possible cell con-
figurations for electrochemical HDH. Two types of cells were com-
pared, an H-cell and solid polymer electrolyte zero-gap flow cell
͑solid polymer electrolyte cell͒. Figure 7 shows data obtained from
galvanostatic HDH ͑22 mA/cm2͒ of 1 mM DCP solutions in these
cells with Pd/carbon cloth cathodes ͑5 mg Pd/cm2͒. The solid poly-
mer electrolyte cell gave good performance for the HDH of 1 mM
DCP solution, although the best result was obtained using the
H-cell, e.g., in terms of total amount of released chloride ions, after
3 h of electrolysis, ca. 0.174 and 0.14 mmol of chloride ions were
released using the H-cell and the solid polymer electrolyte cell,
respectively.
In this work, the Pd/carbon cloth was formed by electrochemi-
cally depositing palladium onto the carbon cloth substrate and the
One aspect of the work was a preliminary evaluation of two solid
polymer membranes, a cation exchange, and an anion exchange. A
comparison of chloride ion in the cathode chamber of the H-cell,
with three types of separator, using the Pd/carbon cloth cathodes ͑5
mg Pd/cm2͒ under identical conditions for HDH of 1 mM DCP, is
shown in Fig. 7. After 3 h of electrolysis at 200 mA, the total
released chloride ions were 0.093, 0.132, and 0.174 mmol in the
cells with the glass frit, the FuMATech membrane and Nafion 117
membrane, respectively. The Nafion proton conductor effectively
excludes chloride ion transport and thus exhibits the highest chloride
content. The anion exchange ͑FuMATech͒ membrane, however, al-
lows a significant amount of chloride ion transport into the anode
chamber, i.e., lower chloride ion content. Thus with the anion ex-
change membrane the simultaneous dechlorination and separation of
phenol from chloride ion is possible. The use of a simple porous
glass frit enables a greater chloride ion transfer into the anode cham-
ber, but this is at the expense of some chlorophenol transfer.
Figure 6. Effect of cathode material on chloride ion release during electro-
chemical HDH of 0.071 mM PCP. Cathode: shown in the figure ͑5 mg
catalyst cmϪ2, 6 cm2͒. Controlled cathode potential: Ϫ1.0 V vs. RHE. Other
conditions: see Fig. 2.
Product distribution.—In order to confirm the effectiveness of
the electrochemical HDH approach as demonstrated by chloride ion
Downloaded on 2015-03-18 to IP 134.71.135.75 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

D22
Journal of The Electrochemical Society, 150 ͑2͒ D17-D24 ͑2003͒
Figure 9. Effect of cathode on current efficiency for the electrochemical
HDH of 1 mM DCP in 0.05 M Na2SO4 solution. Palladium loading for the
palladized cathodes: 5 mg cmϪ2. Other conditions: same as those in Fig. 7
for the H-cell.
Figure 8. HPLC concentration profiles of reactant and product during elec-
trochemical HDH of 1 mM DCP. Conditions: same as those in Fig. 7 for the
H-cell.
analysis, product distributions were determined by HPLC with UV
detection at 270 nm for phenol, 280 nm for CP and DCP, and 300
nm for PCP. Figure 8 shows product distributions, including DCP,
the intermediate 4-chlorophenol ͑CP͒ and phenol, obtained from
HPLC measurements during the electrochemical HDH of 1 mM
DCP in aqueous solution. As expected, a rapid decrease in the con-
centration of DCP and corresponding increase in the concentration
of phenol due to the effective dechlorination of DCP can be seen.
Small amounts of 4-chlorophenol were also detected. As mentioned
previously, the electrochemical HDH of DCP involved the removal
of chloride ions and the addition of four or two electron͑s͒ to the
phenol ring, depending on the product, phenol or chlorophenol. The
above product distribution agreed well with the results obtained
from analysis of chloride ion concentrations ͑see Fig. 7͒. The mass
balances in the above HPLC measurements were close to 100%
when DCP, CP, and phenol were considered.
3.5% in the solid polymer electrolyte cell for the HDH of 1 mM
DCP and 1 mM PCP, respectively. However, no supporting electro-
lyte was used for the operation of the solid polymerelectrolyte cell.
This feature should be beneficial for practical applications of this
type of cell provided that operating conditions are optimized. The
preferred cell for HDH should be an solid polymer electrolyte cell.
Energy consumption.—A key parameter when designing an elec-
trochemical process is energy consumption.24,25 Not surprisingly,
energy consumption varied with cell types and cathode materials for
the electrochemical HDH of DCP and PCP. Figure 11 shows the data
for the HDH of 1 mM DCP in the H-cell using the Pd/Ti mesh
cathode or in the solid polymer electrolyte cell using different cath-
odes. The solid polymer electrolyte cell gave lower energy con-
Current efficiency.—The influence of the cathode material on the
current efficiency of the HDH process is shown in Fig. 9. For the
H-cell, the current efficiency of dechlorination of 1 mM DCP solu-
tion varied from 0.75 to 65%, depending on the materials and elec-
trolysis time. The best current efficiencies, 60-65%, were achieved
using the Pd/Ti mesh cathode. The high surface area deposits, dem-
onstrated by the SEM measurements ͑not shown here͒, and open
structure of the Pd/Ti mesh cathode gave very high efficiency. The
Pd/carbon cloth and Pd/Fe gauze cathodes gave lower current effi-
ciencies, around 10 and 8%, respectively. With other materials, cur-
rent efficiencies were very low, i.e., less than 3%. As stated above,
the poor performance of the iron cathodes was mainly a result of
their instability. The mass-transport limitations of the Pd/carbon
powder cathode led to the lower efficiency for the HDH process,
although it had three times the Pd loading as those of the Pd/Ti mesh
and Pd/carbon cloth cathodes. It must be stated that these electrode
structures were not optimized and represent our adapted method of
manufacture. For HDH of PCP solutions, lower efficiencies were
observed, e.g., below 10%, even using a Pd/Ti mesh cathode
͑Fig. 10͒.
Current efficiencies varied greatly in the different cells, as shown
in Fig. 10. Current efficiencies obtained in the solid polymer elec-
trolyte cell were lower than those obtained in the H-cell for the
HDH of PCP and DCP solutions. The current efficiencies were in a
range of 60-65% and 14-28% in the H-cell and 3.1-7.5% and 1.3-
Figure 10. Effect of cell type on the current efficiency of HDH of 1 mM
DCP and PCP solutions. Conditions: same as those in Fig. 7.
Downloaded on 2015-03-18 to IP 134.71.135.75 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

Journal of The Electrochemical Society, 150 ͑2͒ D17-D24 ͑2003͒
D23
Figure 11. Effect of cathode on energy consumption for the electrochemical
HDH of 1 mM DCP-pure water solution in the solid polymer electrolyte cell
with different cathodes. Cathode: shown in the figure ͑palladium loading: 5
mg/cm2͒. Other conditions: same as those in Fig. 7.
Figure 12. Change of catholyte pH values during the electrochemical HDH
of 1 mM DCP in 0.05 M Na2SO4 solution. Cathode: Pd/Fe gauze ͑5 mg
Pd/cm2, 6 cm2͒. Other conditions: same as those in Fig. 7.
sumption than the H-cell, e.g., 30 and 59 kWh/kg DCP after 3 h for
the solid polymer electrolyte cell and the H-cell, respectively.
In the solid polymer electrolyte cell, the Pd/Ti mesh and the
Pd/carbon cloth cathodes showed much lower energy consumptions
than other materials for the HDH process, which can be attributed to
their high effectiveness and stability during the HDH process. For
example, the energy consumption was 400, 350, 32, and 30 kWh/kg
for the Pd/Fe gauze, 30% Pd/activated carbon powder, Pd/carbon
cloth and Pd/Ti mesh cathodes, respectively. High energy consump-
tion observed at the Pd/Fe gauze, as mentioned above, mainly re-
sulted from its low current efficiencies. The Pd/activated carbon
powder gave lower energy consumption than the Pd/Fe gauze.
ever, the corrosion of these iron-based materials was severe, particu-
larly during long-term operation. As technically more acceptable
electrodes, carbon cloth, and titanium mesh supported Pd electrodes
showed high catalytic activity and stability for HDH process. Using
these cathodes good HDH performance of DCP and PCP were
achieved with dilute solutions. In particular, using a simple batch
H-cell with pH 3, 0.05 M sodium sulfate/H2SO4 supporting electro-
lyte, electrolysis resulted in virtually complete HDH of DCP and
PCP with current efficiencies up to 70 and 16%, respectively. For
the low concentrations of organics treated the current efficiencies
were quite respectable and energy consumptions were reasonable, 6
and 21 kWh kgϪ1 for DCP and PCP, respectively. Pd/stainless steel
also hydrodehalogenated PCP and DCP efficiently, although the de-
posit stability needs to be improved.
Although either cation or anion exchange membranes could be
used for HDH, the use of an anion exchange membrane may be
particularly useful when the simultaneous removal of chloride ion is
desirable.
To scale-up the electrochemical HDH to industrial application,
optimization of electrode fabrication, HDH cells, and operation pa-
rameters is necessary and is in progress in this laboratory.
Catholyte pH change.—An interesting observation during the
HDH of DCP and PCP was that the catholyte pH greatly depended
on cell type. Figure 12 shows such a typical change in pH during
HDH of 1 mM DCP in 0.05 M Na2SO4 with the Pd/Fe gauze cath-
ode ͑5 mg Pd/cm2͒. The pH increased due to the accumulation of
OHϪ ions from the electrolysis of water in the H-cell. In the solid
polymer electrolyte flow cell, protons formed at the anode crossed
the Nafion membrane into the catholyte, which partly cancelled the
effect of the hydrogen evolution and led to a small catholyte pH
change.
Acknowledgments
Conclusions
The authors thank the United Kingdom Engineering and Physical
Sciences Research Council ͑EPSRC͒ for a postdoctoral award dur-
ing this work.
Thanks also to FuMATech ͑Germany͒ for providing membranes
for this work.
1. The chemical HDH of DCP and PCP was not effective on
reducing metals, such as Fe and Pd. However, DCP and PCP were
effectively hydrodehalogenated by electrochemical reduction using
iron and palladium cathodes, which provides a new technology for
treatment of wastewater containing halogenated organic com-
pounds.
The University of Newcastle upon Tyne assisted in meeting the publica-
tion costs of this article.
2. HDH could be performed with H-cell or/and solid polymer
electrolyte cells and cathode material, including catalyst and sub-
strate, had a decisive influence on HDH regarding both reaction rate
and efficiency. Palladized cathodes were much more effective for
the HDH of DCP and PCP than iron and platinized cathodes. Elec-
trochemical HDH using iron cathodes was effective with almost
complete HDH possible at low concentrations of DCP and PCP. The
use of palladized iron electrodes in both the solid polymer electro-
lyte cell and the H-cell markedly improved the HDH efficiency
which, with 1.0 mM DCP and PCP solutions, gave current efficien-
cies up to 14 and 8%, respectively, at 94% chloride removal. How-
References
1. DOE, The Special Waste Regulations, HMSO, London ͑1996͒.
2. M. L. Hitchman, R. A. Spackman, N. C. Ross, and C. Agra, Chem. Soc. Rev., 1995,
423.
3. S. Zhang and J. F. Rusling, Environ. Sci. Technol., 27, 1375 ͑1993͒.
4. C. S. Criddle and P. L. McCarty, Environ. Sci. Technol., 25, 937 ͑1991͒.
5. C. Grittini, M. Macomson, Q. Fernando, and N. Korte, Environ. Sci. Technol., 29,
2898 ͑1995͒.
6. L. N. Zanaveskin, V. A. Averganov, and Y. A. Treger, Russ. Chem. Rev., 65, 617
͑1996͒.
7. M. A. Aramendia, V. Borau, I. M. Garcia, C. Jimenez, J. M. Marinas, and F. J.
Urbano, Appl. Catal., B, 20, 101 ͑1999͒.
Downloaded on 2015-03-18 to IP 134.71.135.75 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).

D24
Journal of The Electrochemical Society, 150 ͑2͒ D17-D24 ͑2003͒
8. S. T. Srinivas, L. T. Lakshmi, N. Lingaiah, P. S. S. Prasad, and P. K. Rao, Appl.
Catal., A, 135, L201 ͑1996͒.
9. E. N. Balko, E. Przybylski, and F. V. Trentini, Appl. Catal., B, 2, 1 ͑1993͒.
10. E. S. Creyghton, M. H. W. Burges, J. C. Jansen, and H. Van Bekkum, Appl. Catal.,
A, 128, 275 ͑1995͒.
18. I. F. Cheng, Q. Fernando, and N. Korte, Environ. Sci. Technol., 31, 1074 ͑1997͒.
19. S. Zhang and J. F. Rusling, Environ. Sci. Technol., 27, 1375 ͑1993͒; S. Zhang and
J. F. Rusling, Environ. Sci. Technol., 29, 1195 ͑1995͒.
20. D. Schmal, J. van Erkel, and P. J. van Duin, Inst. Chem. Eng., 98, 259 ͑1986͒.
21. M. Gattrell and B. MacDougall, J. Electrochem. Soc., 146, 3335 ͑1999͒.
22. H. Cheng, Ph.D. Thesis, University of Newcastle upon Tyne, Newcastle upon Tyne,
U.K. ͑1999͒.
11. G. W. Reynolds, J. T. Hoff, and R. W. Gillham, Environ. Sci. Technol., 24, 135
͑1990͒.
12. S. F. O’Hannesin, MA Thesis, University of Waterloo, Ontario, Canada ͑1993͒.
13. H. K. Yak, B. W. Wenclawiak, I. F. Cheng, J. G. Doyle, and C. M. Wai, Environ.
Sci. Technol., 33, 1307 ͑1999͒.
14. R. Muftikian, Q. Fernando, and N. Korte, Water Res., 29, 2434 ͑1995͒.
15. E. L. Appleton, Environ. Sci. Technol., 30, 536A ͑1996͒.
16. F. Bonfatti, S. Ferro, F. Lavezzo, M. Malacarne, G. Lodi, and A. De Battisti, J.
Electrochem. Soc., 146, 2175 ͑1999͒.
23. S.-A. Sheppard, S. A. Campbell, J. R. Smith, G. W. Lloyd, T. R. Ralph, and F. C.
Walsh, Analyst (Cambridge, U.K.), 123, 1923 ͑1998͒.
24. F. Goodridge and K. Scott, Electrochemical Process Engineering, pp. 15-191, Ple-
num Press, New York ͑1995͒.
25. D. Pletcher and F. C. Walsh, Industrial Electrochemistry, 2nd ed., pp. 74-75, Chap-
man and Hall, New York ͑1990͒.
26. E.-J. Shin and Mark A. Keane, Appl. Catal., B, 18, 241 ͑1998͒.
27. D. R. Lawson, M. J. Teirney, I. F. Cheng, L. S. Van Dyke, M. W. Espenscheid, and
C. R. Martin, Electrochim. Acta, 36, 1515 ͑1991͒.
17. S. M. Kulikov, V. P. Plekhanov, A. I. Tsyganov, C. Schlimm, and E. Heitz, Elec-
trochim. Acta, 41, 527 ͑1996͒.
Downloaded on 2015-03-18 to IP 134.71.135.75 address. Redistribution subject to ECS terms of use (see ecsdl.org/site/terms_use) unless CC License in place (see abstract).