Electrochemical reduction of CO2 in a mixed supercritical fluid

Article abstract of DOI:10.1021/jp9932867

The electroreduction of CO₂ in the liquid and supercritical states was studied on both Pt and Pb electrode surfaces using a mixture of 1,1,1,2-tetrafluoroethane (HFC 134a) and CO₂ solvent. A decrease in the reduction potential on both electrode materials was observed in the supercritical state compared with the liquid state. Platinum was far better than lead as a cathode material on which to reduce CO₂ in supercritical CO₂/HFC 134a mixture by both voltametric and bulk electrolysis experiments. The use of the supercritical mixture, rather than an aprotic liquid as electrochemical solvent, significantly improved the faradaic efficiency of oxalate formation at a Pt electrode, which can be due to the high CO₂ concentration at the electrode surface in the mixture.

Full text of DOI:10.1021/jp9932867

J. Phys. Chem. B 2000, 104, 775-779
775
Electrochemical Reduction of CO2 in a Mixed Supercritical Fluid
Andrew P. Abbott* and Christopher A. Eardley
Chemistry Department, UniVersity of Leicester, Leicester LE1 7RH, United Kingdom
ReceiVed: September 15, 1999; In Final Form: NoVember 22, 1999
A mixture of 1,1,1,2-tetrafluoroethane (HFC 134a) and carbon dioxide is used as a medium for electrochemical
study. The reduction of CO2 in the liquid and supercritical states was investigated on both Pt and Pb electrode
surfaces. It was shown that there is a decrease in the reduction potential on both electrode materials in the
supercritical state compared to the liquid. The reduction of CO2 on a Pt electrode is shown to have a high
•-
faradaic efficiency for oxalate production. This is discussed in terms of the adsorption of the CO2 anion
radical on the electrode surface.
Introduction
for formate production at a mercury electrode in aqueous
solution, but this is of limited value.12,13
Following the work of Royer in 1870,1 the electroreduction
of carbon dioxide has been studied at a variety of cathode
materials in both aqueous and nonaqueous electrolyte solutions.2-5
A variety of organic compounds have been produced accord-
ingly. The most abundant reactions, however, are
More recent investigations have concentrated on the elec-
troreduction of CO2 in aqueous solutions at a variety of
nonmetallic cathodes, such as oxide and semiconductor
electrodes.14-19 Bandi demonstrated the electrosynthesis of
methanol with a faradaic efficiency of 50-70% by employing
an electrode prepared from mixed oxides of ruthenium and
titanium.15 Taniguchi and co-workers also produced methanol
with reasonable faradaic efficiency (44%) on a p-type GaP
cathode in the presence of a crown ether.17 A number of studies
have also investigated the electroreduction of CO2 in aqueous
media containing complex catalysts.20-22
CO2 + H+ + 2e- f HCOO-
CO2 + 2H+ + 2e- f CO + H2O
CO2 + 6H+ + 6e- f CH3OH + H2O
CO2 + 8H+ + 8e- f CH4 + 2H2O
2CO2 + 2e- f (COO-)2
(1)
(2)
(3)
(4)
(5)
(6)
The problems associated with hydrogen evolution in the
electrochemical reduction of CO2 can be completely irradicated
by using aprotic solvents. Reactions 1-4 can then be neglected.
Hence, the electroreduction of CO2 in nonaqueous media
promises high yields of oxalate anions. In the first such study
in 1967, Haynes and Sawyer investigated CO2 reduction in
dimethyl sulfoxide at gold and silver electrodes via a variety
of electroanalytical techniques.23 On both metals, somewhat
unexpectedly, CO and formate were the main reaction products.
The electrosynthesis of formate ions on gold, however, was
shown to be strongly dependent on the water content of the
solvent.
-
2CO2 + e- f CO + CO3
Electroreduction of CO2, therefore, commonly produces a
mixture of organic compounds. Large overpotentials are required
to activate the CO2 molecule because of its stability.6 Thus, in
aqueous solutions hydrogen evolution always accompanies the
reduction of CO2.7,8 The faradaic yield of each product is
therefore a critical parameter used to assess the reaction
selectivity.
Any adventitious water present in the system acts a proton
donor, facilitating reaction 1 and thus increasing the faradaic
efficiency of formate production at the expense of that of oxalate
and CO.9 Subsequent studies in nonaqueous solvents have
shown that high yields of oxalate ions can be obtained if the
water content of the system is minimized and an appropriate
electrode material employed.9,24-26 In 1972, for example, Baizer
et al. employed a mercury electrode in dimethylformamide and
acquired a faradaic efficiency of approximately 90% for oxalate
production in the absence of water.26 The variation in product
selectivity achieved with metallic cathodes in aprotic solutions
prompted Ito and co-workers to introduce a scheme of electrode
classification.9
In aqueous solutions, the majority of organic compounds
formed possess a single carbon atom.2,9 The nature of the
cathode and electrolyte solution largely determines the faradaic
yields and composition of the electroreduction products. The
metallic cathodes can be generally grouped in accordance with
the primary reaction products.2 Of all the organic products listed
above, formic acid is the least desirable. It is difficult to separate
from solution, and on oxidation to carbon dioxide its energy
content is extremely low.
Many studies in aqueous solutions have focused on metal
electrodes that require high overpotentials for the cathodic
evolution of hydrogen. By employing cathode materials such
as lead, mercury, tin, and indium,7,9-12 hydrogen evolution can
be minimized at low overpotentials. Teeter and van Ryssel-
berghe, for example, observed faradaic yields close to 100%
The low solubility of CO2 in both protic and aprotic liquid
media commonly limits the rate of the reduction process. The
concentration of CO2 in solution can be raised by increasing
the CO2 pressure. At 25 °C, for example, CO2 has a solubility
in water of only 0.033 mol dm-3, a concentration that increases
* Corresponding author. Fax: UK
andrew.abbott@le.ac.uk.
+ 116 2523789. E-mail:
10.1021/jp9932867 CCC: $19.00 © 2000 American Chemical Society
Published on Web 01/04/2000

776 J. Phys. Chem. B, Vol. 104, No. 4, 2000
Abbott and Eardley
to 1.17 mol dm-3 at a CO2 pressure of 60 atm.27 Ito and co-
workers investigated the effects of increasing pressure on the
electroreduction of CO2 in aqueous solution at Zn, Pb, In, and
Sn electrodes.11,28 Increasing the CO2 pressure to approximately
20 atm led to a gain in both the faradaic efficiency and current
density of the reductive process. Further investigations have
shown that CO2 can be reduced with relatively high faradaic
efficiencies (16-62%) under high CO2 pressures (50-60 atm)
on Co, Fe, Pt, Pd, and Ni electrodes.29 The electroreduction of
CO2 was negligible, however, when employing the same
electrode materials and a CO2 pressure of 1 atm.
Bandi and co-workers suggested that liquid CO2, containing
a tetraalkylammonium electrolyte, could be employed as the
electrochemical solvent in the reduction of CO2.30,31 The
problems associated with CO2 solubility and hydrogen evolution
would, therefore, be irradicated. Although CO2 can be made
conducting through the introduction of bulky hydrophobic
electrolytes,32 the resistivity of the medium is still too high as
to make the reduction reaction impractical.
Figure 1. Schematic of the high-pressure apparatus.
Supercritical media are miscible with gases in all proportions,
which facilitates very high CO2 concentrations. Mass transport
of an electroactive species to or from the electrode surface is
also rapid in such low-viscosity media. Furthermore, CO2 itself
has conveniently low critical constants (Tc ) 31.4 °C; pc )
72.3 bar). As a result, supercritical CO2 has been reduced under
nonelectrochemical conditions using catalysts such as ruthe-
nium(II) complexes.33
Abbott and Harper were the first to investigate the electro-
chemical reduction of CO2 in the supercritical state.34 The
reductive process was studied at platinum and lead electrodes
in a supercritical CO2/H2O solvent mixture. The aqueous
component of the mixture acts as a polar modifier, permitting
the dissolution of a TBABF4 electrolyte and hence substantially
increasing the conductivity of the medium. At platinum
electrodes formate was the only significant product detected,
with faradaic efficiencies below 30%. Oxalate was shown to
be the major organic product formed at a lead electrode with a
faradaic efficiency of approximately 16%. The low faradaic
efficiencies of the organic products were attributed to competi-
tion from the hydrogen evolution process
Figure 2. High-pressure cell.
a model 10-600 pump (Hydraulic Engineering Corp., Los
Angeles, CA), driven by compressed air. The temperature of
the high-pressure cell was measured using an iron/constantan
thermocouple, the tip of which was in contact with the solvent
close to the cell center, and retained at a given value ((0.5 K)
using a CAL 9900-controlled heater. The pressure was moni-
tored ((0.2 MPa) using a UCC type PGE 1001.600 manometer.
Figure 2 shows the cell used in the high-pressure experiments.
The cell was constructed from 316 stainless steel and was rated
to 1.5 kbar. The cell, lined with a layer of Teflon 1 mm thick,
had a volume of approximately 15 cm3. A copper ring was used
to provide a high-pressure seal between the head and base of
the cell. The electrical feedthroughs consisted of microwave
cable (RS Components Ltd.) sealed by Swagelok fittings.
Voltammetric studies were carried out on a 10 µm diameter
electrode using a Ag/Ag+ pseudoreference electrode. A PG-
STAT 20 potentiostat (Ecochemie, Utrecht, Holland), computer
controlled via GPES software was used to collect the voltam-
metric data.
A concentric cylindrical two-electrode configuration, with an
electrode separation of 2 mm, was used in all bulk electrolyses.
The outer stainless steel cylinder formed the anode and was
platinized by the standard technique.37 The inner cathodic
cylinder was formed from either platinum black on stainless
steel or lead. The electrode areas of the anode and cathode were
45.4 and 38.6 cm2, respectively. Each electrolysis was performed
under potentiometric control, employing an EG&G Princeton
Applied Research potentiostat.
H2O + e- f 1/2H2 + OH-
(7)
In this work, 1,1,1,2-tetrafluoroethane (HFC 134a) rather than
water will be employed as a polar modifier in the supercritical
mixtures. Mixtures of CO2 and HFC 134a were recently shown
to be useful fluids because of their low critical constant, lack
of toxicity, and relatively high polarizability per unit volume.35
Significant solubility of quaternary ammonium electrolytes can
be obtained, which permits electrochemical processes to be
studied. Reactions 7 and 1-4 can then be completely neglected
at negative overpotentials because HFC 134a is aprotic. As a
result, the electroreduction of CO2 should be more efficient and
the formate anion should not form. HFC 134a also has a
considerably lower critical temperature than water, leading to
relatively mild critical conditions for the solvent mixture.36 The
current work reports the product distributions and efficiencies
associated with the electroreduction of CO2 at Pt and Pb
electrodes in supercritical CO2/HFC 134a solvent mixtures.
Experimental Section
The 30.6 mol % 1,1,1,2-tetrafluoroethane (HFC 134a) in CO2
mixture (ICI Klea group, 99.99%) and the electrolyte, tetrabu-
tylammonium tetrafluoroborate (Fluka, electrochemical grade),
were used as received.
A schematic of the high-pressure apparatus is shown in Figure
1. Prior to each experiment the reaction vessel was purged with
the HFC 134a/CO2 gas mixture. Pressure was then applied using

Electrochemical Reduction of CO2
J. Phys. Chem. B, Vol. 104, No. 4, 2000 777
Figure 3. Voltammograms recorded at a 10 µm diameter Pt disk
microelectrode in CO2/HFC 134a (xHFC 134a ) 0.304) with 20 mM
TBABF4; A ) 30 °C and 150 bar (liquid), B ) 60 °C and 150 bar
Figure 4. Voltammograms recorded at a 10 µm diameter Pt disk
microelectrode in supercritical CO2/HFC 134a (xHFC 134a ) 0.304) at
60 °C with 20 mM TBABF4; A ) 200 bar, B ) 150 bar. Scan rate )
(supercritical). Scan rate ) 20 mV s-1
.
20 mV s-1
.
Following electrolysis, all gaseous samples were transferred
from the electrochemical cell into a high-pressure sample vessel.
The stainless steel collection vessel had a volume of 250 cm3
and a maximum working pressure of 30 bar. A measured volume
of the gas sample was then introduced into a Tedlar bag with
an internal volume of 1 dm3. A Chrompack CP9000 with flame
ionization detection was used to analyze the sample, where the
Poropack Q-packed column was 2 m long and had an internal
diameter of 4 mm. The oven temperature was held at 30 °C,
while nitrogen was employed as the carrier gas.
The formate and oxalate content of the bulk electrolysis
product was analyzed by high-performance liquid chromatog-
raphy. The recovered electrolyte was first dissolved in 0.1 M
H2SO4(aq) to ensure protonation of any formate and oxalate
anions. Following filtration through a 20 µm glass sinter, the
solutions were analyzed via a Hewlett-Packard 164 high-
performance liquid chromatograph. A Whatman Partisphere C18
reversed phase column was used to detect both formic and oxalic
acids. The mobile phase consisted of 10 mM H3PO4(aq), where
the flow rate was maintained at 1 mL min-1 and the sample
injection volume was 20 µL. The products were detected by
UV absorbance, employing a fixed wavelength of 210 nm.
Although analysis for carbonate was not carried out, it can be
assumed that the faradaic efficiency was the same as that for
CO.
corresponding to the onset of CO2 reduction is approximately
-0.19 V. Negative shifts in the onset potential of CO2 reduction
are commonly ascribed to the adsorption of intermediates, such
as CO, on the electrode surface.38-41 It is, therefore, implied
that adsorbed species such as CO exhibit a lower surface
coverage on the platinum electrode under the supercritical
conditions studied. Note that the observed positive shift in
potential cannot be attributed to an increase in CO2 concentra-
tion, as described by Ito and co-workers,28 because the density
of the specified supercritical mixture is lower than that of the
liquid mixture studied.
Figure 3 also shows that the cathodic current associated with
CO2 reduction at a platinum electrode is higher under super-
critical conditions. This result implies that the rate of CO2
reduction is limited by mass transport in the liquid phase at 30
°C and 150 bar. The lower viscosity of supercritical CO2/HFC
134a will allow any species involved in the electroreduction of
CO2 to diffuse to and from the electrode surface at higher rates
than in the liquid.
The potential for the onset of CO2 reduction is shown in
Figure 4 to undergo a slight negative shift to approximately
-0.25 V as the pressure of the supercritical fluid is increased
from 150 to 200 bar at 60 °C. The density of the fluid increases
with increasing pressure, suggesting that the negative potential
shift is caused by an increase in intermediate adsorption on the
electrode surface and not a CO2 concentration decrease. At the
lower pressure of 150 bar, there is also an increase in the
cathodic current. This increase can be assigned to the drop in
fluid viscosity and consequent increase in the diffusivity of any
species involved in the reductive process.
Figure 5 shows voltammograms recorded in liquid and
supercritical CO2/HFC 134a at a lead microelectrode. At 30 °C
and 165 bar the reduction of CO2 only occurs at potentials
negative of -2.15 V. This value is considerably more negative
than the corresponding reduction potential on platinum. Vassiliev
and co-workers also reported that the overpotential of CO2
reduction in liquid aprotic solvents is more negative on lead
than on platinum.42 The large difference in onset potentials
reported in this work suggests that there is a change in the major
reaction mechanism of CO2 reduction on replacement of the
platinum electrode with lead.
Results and Discussion
A CO2/HFC 134a mixture with a HFC 134a mole fraction,
xHFC 134a, of 0.306 was employed as the solvent in all of the
voltammetric experiments presented in the current work. This
composition exhibits relatively mild critical constants (Tc ) 57
°C; pc ) 72 bar),36 while simultaneously facilitating the
dissolution of a sufficient amount of TBABF4 electrolyte to
support electrochemistry without significant ohmic distortion.
All of the CO2/HFC 134a/TBABF4 systems employed in the
current work were present as a single phase, as confirmed by
visual observation using an optical cell.
Figure 3 shows cathodic voltammograms recorded at a
platinum microelectrode in both liquid and supercritical CO2/
HFC 134a mixtures with TBABF4 concentrations of 20 mM.
The cathodic current can be assigned to the reduction of CO2
because both HFC 134a and TBABF4 are electroinactive over
the potential range investigated. In the liquid phase at 30 °C
and 150 bar, CO2 is electroreduced at potentials negative of
approximately -0.6 V vs Ag/Ag+. On increasing the temper-
ature to 60 °C the reduction wave is shifted to more positive
potentials. Under these supercritical conditions, the potential
In supercritical solutions at 60 °C and 240 bar the reduction
wave reported in the liquid phase shifts to more positive
potentials by approximately 1.2 V. This shift is larger than that
observed on platinum and suggests that intermediates such as
CO are adsorbed to a greater extent on lead under the liquid-
phase conditions specified. The overpotential of CO2 reduction

778 J. Phys. Chem. B, Vol. 104, No. 4, 2000
Abbott and Eardley
Figure 5. Voltammograms recorded at a 25 µm diameter Pb disk
microelectrode in CO2/HFC 134a (xHFC 134a ) 0.304) with 20 mM
TBABF4; A ) 30 °C and 165 bar (liquid), B ) 60 °C and 240 bar
Figure 6. Cyclic voltammogram recorded at a 10 µm diameter
platinum disk microelectrode in CO2/HFC 134a (xHFC 134a ) 0.304) at
60 °C and 260 bar following the completion of a bulk electrolysis
(supercritical). Scan rate ) 20 mV s-1
.
experiment. Scan rate ) 20 mV s-1
.
TABLE 1: Electrochemical Reduction Products of CO2 at
Platinum and Lead Electrodes in CO2/HFC 134a (xHFC 134A
0.304) with 0.1 M TBABF4 (Charge Passed ) 100 C)
•-
desorption of the CO2 anions at higher temperature. At 80
)
•-
°C there will, therefore, be a lower surface coverage of CO2
radical anions on the electrode surface and a reduced tendency
to form oxalate ions.
faradaic efficiency/%
electrode
Pt
T/°C
p/bar
(COOH)2
CO
HCOOH
It is evident from the results in Table 1 that the combined
faradaic efficiencies of oxalate and CO formation fall some way
short of 100%. This deficit can probably be attributed to design
of the electrochemical cell, where the anode and cathode are
not separated. Voltammetry was performed at a platinum
microelectrode in the electrochemical cell after the bulk
electrolysis of CO2 at 60 °C and 260 bar, as shown in Figure 6.
An anodic peak was observed at approximately +0.55 V, which
can probably be ascribed to the oxidation of dissolved oxalate.
In an attempt to prove this idea the cathode and anode were
separated using a medium porosity glass frit. The 3 mm thick
frit was placed centrally between two rectangular platinum
electrodes (26 × 10 × 0.5 mm), separated by a distance of 8
mm. The anode was then enclosed in Teflon so that the anodic
compartment was only separated from solution by the glass frit.
A bulk electrolysis experiment was then performed at 80 °C
and 260 bar in supercritical CO2/HFC 134a with 0.1 M TBABF4
at -4 V. After the passage of 50 C of charge, the faradaic
efficiency of oxalate production was 43.0%. This value is higher
than that of 37.3% observed in the nonseparated electochemical
cell, which tends to confirm the idea that the faradaic efficiencies
listed in Table 1 are affected by the reoxidation of the oxalate
ions at the anode. The oxidation of oxalate may be inhibited
more efficiently by the employment of an ion exchange
membrane. Considering that the experimental conditions in the
present study were not optimized, supercritical CO2/HFC
mixtures appear to be extremely promising media for the
efficient electrosynthesis of oxalate.
60
80
60
80
260
260
260
260
41.6
37.3
17.5
13.7
14.6
16.4
42.0
42.9
0
0
0
0
Pb
is, however, still significantly more negative than that on
platinum. Electroreduction of CO2 on platinum is, therefore, a
more attractive proposition because it is a more energy efficient
process.
Bulk electrolysis experiments were performed in supercritical
CO2/HFC 134a mixtures at both platinum and lead electrodes.
In each case, a constant potential of -4 V was applied to the
working electrode until 100 C of charge were passed. The
faradaic efficiencies of the detected products are given in Table
1.
The faradaic efficiency of oxalate formation is reasonably
high at a platinum electrode at the supercritical conditions listed.
In comparison, previous studies in liquid aprotic media have
reported faradaic efficiencies no higher than 5%.9 In these liquid
media CO was the main product with faradaic efficiencies higher
•-
than 60%. It was thus suggested that the CO2 radical anions
easily desorb from the Pt surface and are then attacked by CO2
molecules in the bulk to form CO. In the supercritical CO2/
HFC 134a mixture, however, the CO2 concentration will be
much higher at the electrode surface than in the aforementioned
liquid bulk electrolysis experiments. There will, therefore, be a
•-
higher surface coverage of CO2 radical anions, facilitating
dimerization to the oxalate anion, as shown in eq 5. Conse-
quently, a lower percentage of CO2•- radical anions will desorb
from the electrode surface to form CO, as exemplified by the
lower faradaic efficiencies of CO formation in Table 1. These
results, therefore, show that when very high concentrations of
CO2 are employed, the efficiency of oxalate formation is less
dependent on the energetics of CO2 adsorption at the electrode
surface.
Table 1 also shows that the faradaic efficiency of oxalate
formation undergoes a slight decrease when the temperature of
the supercritical system is raised from 60 to 80 °C. There is
also a slight concurrent increase in the faradaic efficiency of
CO formation. These changes could result from the lower
density of the solvent at 80 °C and the consequent decrease in
CO2 concentration at the electrode surface or the increased
Conclusion
It has been demonstrated for the first time that CO2 can be
reduced electrochemically in aprotic mixtures in the supercritical
phase. Voltammetry has shown that at a platinum electrode in
a CO2/HFC 134a mixture the reduction of CO2 is under mass
transport control. The kinetics of reduction are, therefore,
enhanced by employing a supercritical CO2/HFC 134a mixture.
The nature of the electrode material has been shown to affect
both the kinetics and products of CO2 electroreduction in
supercritical CO2/HFC 134a. Platinum has been shown to be
far superior than lead as a cathode material on which to reduce
CO2 in supercritical CO2/HFC 134a mixtures by both voltam-
metric and bulk electrolysis experiments. Employing a super-

Electrochemical Reduction of CO2
J. Phys. Chem. B, Vol. 104, No. 4, 2000 779
(20) Ogura, K.; Migita, C. N.; Nagoka, N. J. Mol. Catal. 1989, 56, 276.
(21) Fischer, B.; Eissenberg, R. J. Am. Chem. Soc. 1980, 102, 7361.
(22) Beley, M.; Collin, J.-P.; Rupert, R.; Sauvage, J.-P. J. Chem. Soc.,
Chem. Commun. 1984, 19, 1315.
(23) Haynes, L. V.; Sawyer, D. T. Anal. Chem. 1967, 39, 332.
(24) Kaiser, U.; Heitz, E. Ber. Bunsen-Ges. Phys. Chem. 1973, 77, 818.
critical CO2/HFC 134a mixture rather than an aprotic liquid as
the electrochemical solvent dramatically improves the faradaic
efficiency of oxalate formation at a platinum electrode. This
increase in efficiency can be mainly attributed to the high
concentration of CO2 at the electrode surface in supercritical
CO2/HFC 134a mixtures.
(25) Fischer, J.; Lehmann, Th.; Heitz, E. J. Appl. Electrochem 1981,
11, 743.
References and Notes
(26) Tysee, D. A.; Wagenknecht, J. H.; Baizer, M. M.; Chruma, J. L.
Tetrahedron Lett. 1972, 47, 4809.
(27) Kudo, A.; Nakagawa, S.; Tsuneto, A.; Sakata, T. J. Electrochem.
Soc. 1993, 140, 1541.
(28) Ito, K.; Ikeda, S.; Okabe, M. Denki Kagaku 1980, 48, 247.
(29) Nakagawa, S.; Kudo, A.; Azuma, M.; Sakata, T. J. Electroanal.
Chem. 1991, 308, 339.
(30) Maier, C. U.; Bandi, A. Abstracts 42nd ISE Meeting, Montreux,
1991; Abstr. 7-020.
(31) Bandi, A.; Maier, C. U.; Bloss, W. H. FRG Patent 4126349, C25B
3/04.
(32) Abbott, A. P.; Harper, J. C. J. Chem. Soc., Faraday Trans. 1996,
92, 3895.
(33) Jessop, P. G.; Ikariya, T. Noyori, R. Nature 1994, 368, 231.
(34) Harper, J. C. Ph.D. Thesis, Leicester University, 1998.
(35) Abbott, A. P.; Eardley, C. A.; Scheirer, J. E. J. Phys. Chem. B
1999, 103, 8790.
(36) Kordikowski, A.; Robertson, D. G.; AguiarRicardo, A. I.; Popov,
V. K.; Howdle, S. M.; Poliakoff, M. J. Phys. Chem. 1996, 100, 9522.
(37) Vogel, A. I. Textbook of Inorganic QuantitatiVe Analysis; Long-
man: London, 1978.
(1) Royer, E. C. R. Acad. Sci. 1870, 70, 371.
(2) Bagotzky, V. S.; Osetrova, N. V. Russ. J. Electrochem. 1995, 31,
409.
(3) Ziessel, R. NouV. J. Chim. 1983, 7, 613.
(4) Taniguchi, I. In Modern Aspects of Electrochemistry; Bockris, J.
O’M., White, R. E., Coway, B. E., Eds.; Plenum.: New York, 1989; Vol.
20, p 327.
(5) Electrochemical and Electrocatalytic Reactions of Carbon Dioxide;
Sullivan, B. P., Krist, K., Guard, H. E., Eds.; Elsevier: Amsterdam, 1993.
(6) Lamy, E.; Nadjo, L.; Saveant, J. M. J. Electroanal. Chem. 1977,
78, 403.
(7) Udupa, K. S.; Subramanian, G. S.; Udupa, H. V. K. Electrochim.
Acta 1971, 16, 1593.
(8) Hori, Y.; Kikuchi, K.; Suzuki, S. Chem. Lett. 1985, 1695.
(9) Ikeda, S.; Takagi, T.; Ito, K. Bull. Chem. Soc. Jpn. 1987, 60, 2517.
(10) Ito, K.; Murata, T.; Ikeda, S. Bull. Nagoya Inst. Tech. 1975, 27,
209.
(11) Ito, K.; Ikeda, S.; Iida, T.; Niwa, H. Denki Kagaku 1981, 49, 106.
(12) Teeter, T. E.; van Rysselberg, P. J. Chem. Phys. 1954, 22, 759.
(13) Teeter, T. E.; van Rysselberghe, P. Abstracts of Papers, 6th Meeting
of CITCE; Butterworth: London, 1955; p 538.
(14) Koudelka, M.; Monnier, A.; Augustinski, J. J. Electrochem. Soc.
1984, 131, 745.
(38) Hori, Y.; Murata, A.; Takahashi, R. J. Chem. Soc., Faraday Trans.
1989, 85, 2309.
(39) Hori, Y.; Murata, A. Electrochim. Acta 1990, 35, 1777.
(40) Hori, Y.; Murata, A.; Yoshinami, Y. J. Chem. Soc., Faraday Trans.
1991, 87, 125.
(41) Taguchi, S.; Aramata, A.; Enyo, M. J. Electroanal. Chem. 1994,
372, 161.
(15) Bandi, A. J. Electrochem. Soc. 1990, 137, 2157.
(16) Bandi, A.; Ku¨hne, H. M. J. Electrochem. Soc. 1992, 139, 1605.
(17) Taniguchi, I.; Yoneygama, H.; Tamura, H. Bull. Chem. Soc. Jpn.
1982, 55, 2305.
(18) Nogami, G.; Aikoh, Y.; Shiratsuchi, R. J. Electrochem. Soc. 1993,
140, 1037.
(19) Aurian-Blajeni, B.; Taniguchi, I.; Bockris, J. O’M. J. Electroanal.
Chem. 1983, 149, 291.
(42) Vassiliev, Yu. B.; Bagotzky, V. S.; Khazova, O. A.; Mayorova, N.
A. J. Electroanal. Chem. 1985, 189, 295.