Lim 2014
Lim 2014
Lim 2014
ARTICLE IN PRESS
Catalysis Today xxx (2013) xxxxxx
Catalysis Today
journal homepage: www.elsevier.com/locate/cattod
a r t i c l e
i n f o
Article history:
Received 10 September 2013
Received in revised form 1 November 2013
Accepted 16 November 2013
Available online xxx
Keywords:
CO2 reduction
Electrochemistry
Fuel cells
Metal electrodes
Molecular catalysts
a b s t r a c t
In this review article, we report the development and utilisation of fuel cells, metal electrodes in aqueous
electrolyte and molecular catalysts in the electrochemical reduction of CO2 . Fuel cells are able to function
in both electrolyser and fuel cell mode and could potentially reduce CO2 and produce energy at the same
time. However, it requires considerably high temperatures for efcient operation. Direct reduction using
metal electrodes and molecular catalysts are possible at room temperatures but require an additional
applied potential and generally have low current densities. Density functional theory (DFT) studies have
been used and have begun to unveil possible mechanisms involved which could lead to improvements
and development of more efcient catalysts.
2013 Elsevier B.V. All rights reserved.
1. Introduction
Since the industrial revolution, atmospheric carbon dioxide
(CO2 ) level of about 278 ppm has continuously increased due to
extensive use of fossil fuels [1]. Recent report showed that atmospheric CO2 level has reached the 400 ppm mark [2] and it is
expected to continue to rise. The climate modelling study performed by Solomon et al. has shown that the effects of climate
change due to the increase in greenhouse gases would have lasting effects of up to 1000 years even when there is zero emission
of greenhouse gases [35]. Currently, the release of CO2 into the
environment is uncontrolled from various industrial and anthropogenic processes. The open loop nature of these processes which
releases CO2 into the environment is not only unsustainable; but
is also contrary to the natural carbon cycle which is a closed loop
process. Although the study by Solomon et al. [3,4] do not take
into account forced or articial sequestration of greenhouse gases
especially CO2 , it is still important that the release of CO2 into
the environment has to cease as soon as possible to minimise the
adverse impact of climate change. Instead of sequestrating atmospheric CO2 only, the captured CO2 should also be turned into fuels
which not only closes the open loop processes that we have but
also reduces our dependency on fossil fuels. Various other potential uses for CO2 in the industry has also been highlighted where the
utilisation of CO2 may reach an upper limit of 700 Mt/yr [6]. Being
able to efciently turn CO2 into fuels would have the largest benet
and impact in the effort to reduce CO2 emissions. This is because
80% of our energy is derived from fossil fuels [7].
Energy from carbon neutral sources such as renewables and/or
nuclear can be used to synthesise these fuels by storing the energy
in a chemical form. Besides, it would be possible to utilise existing
infrastructures to transport and store the fuels [8,9]. Energy from
renewables and nuclear is generally in the form of electricity, hence
electrochemistry potentially plays an important role in the production of fuels from CO2 . The FischerTropsch (FT) process is also an
important process in the endeavour to turn CO2 into fuels. The FT
process utilises carbon monoxide (CO) and hydrogen (H2 ) or synthesis gas (syngas) and produces hydrocarbons in the presence of
iron or cobalt catalyst (Eq. (1)) [10]. A side and competing reaction is the watergas shift (WGS) reaction where CO2 is produced
instead (Eq. (2)).
CO + 2.15H2 hydrocarbons + H2 O
(1)
CO + H2 O CO2 + H2
(2)
0920-5861/$ see front matter 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.cattod.2013.11.037
Please cite this article in press as: R.J. Lim, et al., A review on the electrochemical reduction of CO2 in fuel cells, metal electrodes and molecular
catalysts, Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.11.037
ARTICLE IN PRESS
G Model
CATTOD-8759; No. of Pages 12
main products, have good current densities and does not require an
applied potential [12,13]. The selectivity and production of CO in
the aqueous electrochemical reduction of CO2 at room temperature
is dependent on the electrode material [11]. The main drawback
of this process is that it requires an additional applied potential
and relatively low current density. However, it is possible to obtain
other products such as hydrocarbons, formate and alcohols from
this process which is not possible in the high temperature electrochemical reduction process [1418].
It is much more favourable thermodynamically to form stable molecules via proton-coupled multi-electron steps rather than
single electron steps. However, the formation kinetics is not
favourable as conrmed by other studies where higher applied
potentials are required [1418]. Briey, Eqs. (3)(8) are used to estimate the standard reduction potential (with respect to the standard
hydrogen electrode (SHE)) from formation energies in aqueous
media at pH 7, atmospheric pressure and 25 C [19].
CO2 + e CO2
(1.90 V)
(3)
Fig. 1. CO2 electrolysis current density vs applied cell potential for Pt based SOFC.
Roman numerals indicate low (I), constant (II) and high (III) current density regions.
Reproduced from Ref. [31] by permission from Elsevier.
CO2 + H2 O + 2e CO + 2OH
(0.52 V)
(0.25 V)
(5)
(0.34 V)
(4)
(0.33 V)
(6)
(7)
2.1. Platinum
(0.32 V)
(8)
(9)
Please cite this article in press as: R.J. Lim, et al., A review on the electrochemical reduction of CO2 in fuel cells, metal electrodes and molecular
catalysts, Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.11.037
G Model
CATTOD-8759; No. of Pages 12
ARTICLE IN PRESS
R.J. Lim et al. / Catalysis Today xxx (2013) xxxxxx
electrode was not responsible for the formation of O2 and (c) the
rate determining step is the charge transfer process.
CO2,ad + 3H2 CH4 + H2 O
(10)
(11)
(12)
Fig. 3. Voltagecurrent and powercurrent prole for different CO2 feed concentrations at a Cu based SOFC at 800 C.
Reproduced from Ref. [12] by permission from Elsevier.
2.4. Copper
Huang et al. proposed and demonstrated the use of copper in SOFCs as it was frequently used in the electrochemical
reduction of CO2 in aqueous solutions [12]. The mixed oxide
of lanthanum-strontium-chromium-iron (LSCF) with gadoliniadoped ceria (GDC) impregnated with copper was used as the
electrode. Pure CO2 or CO2 /Argon mixture was used as the
fuel which generated CO as product. Highest power density of
2.67 mW/cm2 was achieved for pure CO2 feed (see Fig. 3) and the
production of CO increases with temperature and CO2 feed concentration with no coking observed. It was postulated that in order
to produce CO, the dissociation of CO2 had to occur (Eq. (14)).
The reported activation energy for the reduction to occur was
2.72 kcal/mol which is much lower than values reported in heterogeneous catalysis [34].
CO2 CO + O
(14)
Recently, Zhan et al. used a similar SOFC to Kim et al. and demonstrated that it is possible to use the cell reversibly as an electrolyser
and a fuel cell [13]. The fuel source is a mixture of 25% H2 75% CO2
and managed to achieve a current density of 1.0 A/cm2 at 1.30 V. It
was postulated that steam undergoes reduction to H2 (Eq. (13)) and
CO production proceeds through the reverse WGS reaction (Eq. (2))
when in electrolyser mode.
H2 O + 2e O2 + H2
2.5. Other
(13)
Xie et al. used a solid oxide electrolyser based with iron supported on iron oxide as a cathode, a solid electrolyte oxide mixture
of barium, zinc, ceria, zirconia and yttria (BCZYZ) and Ni/BCZYZ as
the anode [36]. This electrolyte behaves differently as compared
to YSZ as it functions like a proton exchange membrane instead of
oxygen ions. 3% H2 O/H2 and pure CO2 was fed to the anode and
cathode, respectively. Major product from the process is CO while
the minor product was methane. Current efciencies for CO and
methane were not very high at only 29.5% and 2.4%, respectively
at very high applied potentials of 2.6 V and current density of
1.5 A/cm2 .
3. Metallic electrodes in aqueous solution
Early studies into aqueous reduction on CO2 using metallic electrodes began in the late 20th century and have continued until
today. In spite of the extensive literature, this method has yet to be
Please cite this article in press as: R.J. Lim, et al., A review on the electrochemical reduction of CO2 in fuel cells, metal electrodes and molecular
catalysts, Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.11.037
G Model
CATTOD-8759; No. of Pages 12
ARTICLE IN PRESS
R.J. Lim et al. / Catalysis Today xxx (2013) xxxxxx
commercialised, citing various drawbacks such as high overpotentials and production of mixed product [15]. Much of the focus had
been on copper electrodes instead of other metals that have been
tested. The reason behind this is the ability of copper to directly
produce hydrocarbons and other products at a high current efciency as compared to other metals that tend to be very selective
towards a single product; namely CO (Au, Ag, Zn, Ga, Pd), formate
(Pb, Hg, In, Sn, Cd, Tl) or hydrogen only (Ni, Fe, pt, Ti) [11].
3.1. Copper
In 19851986, Hori et al. reported the ability of copper to produce hydrocarbons via electrochemical reduction of CO2 [37,38].
A current efciency of 65% for methane formation was achieved
at 0 C but dropped rapidly with the rise in temperature while the
current efciency for ethylene steadily increased with temperature
to about 20% at 40 C. Doubling of the current density from 5 to
10 mA/cm2 did not affect the current efciencies for both methane
and ethylene [38]. The efciencies of ethylene formation is higher
than methane at low current densities (1 mA/cm2 ) and at high pH
[39]. Kim et al. showed that methane and ethylene begins to form
at 1.50 to 1.60 V vs SHE and methane is formed at a much lower
rate from CO than from CO2 . The rate limiting step was suggested
to be the electron transfer to adsorbed CO [40]. A similar study
discovered that the electrode was poisoned by graphitic carbon
which could be due to the decomposition of formate and a possible
mechanism was suggested (Eq. (15)) [18]. Cook et al. proposed that
formaldehyde, formic acid and acetaldehyde are possible intermediates in the formation of hydrocarbons and reported the formation
of carbon deposits on the electrodes in the reduction of formic acid
[16].
Adsorption of amorphous carbon was also reported to occur
within 1 h of electrolysis which can be prevented by modulating the
potential which causes the structure of the copper surface to change
[41]. This method of potential modulation or pulsed potential was
developed by Nogami et al. and was shown to perform much better
than conventional electroreduction [42,43]. Faradaic efciencies
for methane and ethylene were relatively constant throughout the
process. Peak efciency of 55% and 10% for methane and ethylene,
respectively, was achieved at 10 C with a high applied cathodic
potential of 2.60 V vs SHE. This process is illustrated in Fig. 4.
Applying this method to the CuAg (Cu/Ag = 28/72) alloy electrode
seems to suggest that it is possible to control the product selectivity especially towards C2 products such as ethylene, ethanol and
acetaldehyde [44]. Total efciency for the C2 products was approximately 50% and was suggested that an oxide layer was formed
and caused hydrogen atoms to desorb. In addition, the difference
in adsorption on Cu and Ag may play a role in the improved C2
efciency.
An attempt to overcome the poisoning by carbon adsorption
was by electroplating the copper electrodes with gold. However,
electroplating of the electrodes resulted in the reduction in current efciencies for hydrocarbons and alcohols but an increase for
(15)
It is evident from these early studies of aqueous electrochemical reduction of CO2 with copper electrodes that methanol is not
formed or detected in any of the studies. However, in the industry,
one of the main catalysts for methanol production is copper supported on zinc oxide (Cu/ZnO) with additional promoters [48]. The
reason behind this is still unclear at the time. To further understand
the process, an in-depth study was performed by Hori et al. [14].
The study conducted at ambient temperature (19 C), ethanol and
propanol in addition to methane and ethylene were detected as
products. It was proposed that CO2 was reduced to CO and HCOO
at potentials more positive than 1.10 V vs SHE. The adsorbed CO
inhibits the evolution of H2 and undergoes further reduction to
hydrocarbons and alcohols when higher potentials are applied. The
formation of ethylene and alcohols is also much more favourable
when inorganic salts are used as electrolytes. Methane formation
is much more favourable when more hydrogen atoms are available
on the electrode surface. Finally, it was concluded that adsorbed
CO binds weakly to the surface of the electrode. The overall proposed mechanism is illustrated in Fig. 5 which is similar to the
FischerTropsch process.
Hara et al. demonstrated that it was possible to control the selectivity of the reduction products by manipulating the pressure of
CO2 , current density and stirring. The increase in CO2 pressure, current density and absence of stirring improved the current efciency
for methane and ethylene formation while decreased the efciency
for formic acid and CO formation and vice versa [49]. Hori et al.
was the rst to report on the formation of acetic acid when using
single crystal electrodes [5052]. Acetic acid and other C2+ hydrocarbons were detected when the Cu(1 1 0) surface is the dominant
surface. The Cu(1 1 1) dominant surface is much more selective in
Fig. 5. Possible mechanism for the formation of methane, ethylene, ethanol and propanol with copper as [14].
Please cite this article in press as: R.J. Lim, et al., A review on the electrochemical reduction of CO2 in fuel cells, metal electrodes and molecular
catalysts, Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.11.037
G Model
CATTOD-8759; No. of Pages 12
ARTICLE IN PRESS
R.J. Lim et al. / Catalysis Today xxx (2013) xxxxxx
Fig. 6. Current efciencies for formation of (a) methane, (b) ethylene, (c) formate and other C2+ products and (d) applied potential for various crystal series electrodes. Roman
numerals (I), (II) and (III) indicate Cu(1 0 0), Cu(1 1 1) and Cu(1 1 0) dominant surfaces, respectively [53].
Please cite this article in press as: R.J. Lim, et al., A review on the electrochemical reduction of CO2 in fuel cells, metal electrodes and molecular
catalysts, Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.11.037
G Model
CATTOD-8759; No. of Pages 12
ARTICLE IN PRESS
R.J. Lim et al. / Catalysis Today xxx (2013) xxxxxx
(16)
Route3 : CO + CO enediol/enediolate/
oxametallacycle ethylene
(18)
The selectivity of copper electrodes towards ethylene and current density was enhanced greatly and hydrogen evolution was
greatly suppressed when halide ions were present in the electrolyte
[5860]. Higher concentrations of halide ions improve the current
efciency of ethylene formation and suppress hydrogen evolution
further. A current efciency for ethylene formation of up to 80% is
achievable when copper(1) halides are conned within the copper
mesh electrodes [58]. Ethylene is proposed to be formed through
the coupling of adsorbed COOH [59]. Stronger bonding of the halide
ions with the electrode (Cl < Br < I ) may have contributed to the
improved current efciency of up to 13.2 mA/cm2 by increasing the
stability of adsorbed CO2 thus allowing it to be protonated [60].
The use of thin copper oxide layer on the electrodes had
also attracted some attention due to its ability to reduce CO2 to
(19)
(20)
Please cite this article in press as: R.J. Lim, et al., A review on the electrochemical reduction of CO2 in fuel cells, metal electrodes and molecular
catalysts, Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.11.037
G Model
CATTOD-8759; No. of Pages 12
ARTICLE IN PRESS
R.J. Lim et al. / Catalysis Today xxx (2013) xxxxxx
3.2. Platinum
Extensive studies have been performed to study the adsorption
of CO and CO2 on platinum electrodes but with differing opinions.
However, most of the studies agreed that some form of reduced
CO2 is present on the electrode as a product [76]. Studies on single
crystal electrodes showed that at platinum surfaces (i.e., Pt(1 0 0)
and Pt(1 1 1)) have the lowest activity for CO2 reduction to CO
while high kink densities and stepped surfaces have the highest
activity [77,78]. Platinum electrodes modied with a composite of
RuO2 /TiO2 in the form of nanoparticles and nanotubes were able to
produce methanol with high efciencies of up to 60% at 0.8 V vs
SHE [79]. Electrodes modied with the nanotube composite were
much more selective (60.5%) in the production of methanol as compared to the nanoparticle counterpart (40.5%). This could be due
to the better dispersion of RuO2 on the TiO2 nanotubes as compared to the nanoparticles thus providing more surface area for the
electro-reduction to occur.
3.3. Palladium
A series of studies was carried out by Ohkawa et al. using hydrogenated palladium electrodes with applied potentials of 1.60 to
1.80 V vs SHE [8083]. Hydrogenation and addition of copper to
the palladium electrode increased the current efciency for the
production of CO and formic acid as compared to the unmodied electrodes [80,81]. The increased in efciency is due to the
participation of adsorbed hydrogen. Surface morphology from the
modication with copper had no effect on the efciency when compared to unhydrogenated electrodes. Methanol was reported to
have formed with very low Faradaic efciency (1%). Methane and
ethylene were also reported with efciencies of 35% and 5%, respectively. The remaining products of CO and formic acid were formed
with efciencies of about 20% and 60%, respectively. The copper
deposition may have suppressed the formation of H2 which allows
the hydrogenation of the adsorbed intermediates thus allowing the
formation of reduction products [82].
Kolbe et al. showed that the reduction of CO2 to CO or formic
acid on palladium electrodes begins at about 1 V vs SHE accompanied by hydrogen evolution. At 1.40 V volatile CO is formed thus
freeing the surface for further reduction of CO2 . At 0.50 V oxidation
of adsorbates (CO or formic acid) occurs [84]. The effect of different
crystal structures on the rate of CO2 reduction to CO was studied by
Hori et al. and found that the Pd(1 1 1) surface has the highest activity which is contrary to other Pt group metals (Pt, Rh, Ir) at 0.50 V
vs SHE [85]. Moreover, the Pd electrodes were prone to poisoning
the adsorbates [86].
3.4. Others
Oxide derived gold nanoparticles were shown to be able
to reduce CO2 much more efciently than polycrystalline gold
nanoparticles [87]. Oxide derived gold nanoparticles begin to
reduce CO2 to CO at 0.25 V vs SHE with a current efciency of 65%
compared to the polycrystalline alternative which only begin at
above 0.35 V vs SHE. At higher applied potentials of up to 0.50 V
vs SHE, the efciency increases to more than 96% for oxide derived
gold and only about 80% for polycrystalline gold. Current density is
an order of magnitude higher for oxide derived gold (24 mA/cm2
compared to 0.30.5 mA/cm2 ). In terms of stability, oxide derived
gold nanoparticles are much more stable and is able to maintain
activity although sintering occurred within 15 min of reduction.
The electron transfer to adsorbed CO2 was proposed to be the rate
determining step on polycrystalline gold while hydrogen transfer
by HCO3 to adsorbed CO2 radical which indicates that adsorbed
CO2 radical is much more stable on oxide derived gold nanoparticles.
Lead (Pb) granules in a xed bed reactor were used in conjunction with high pressures (up to 500 kPa) and temperatures
(up to 80 C) to electrochemically reduce CO2 to formic acid [88].
Peak current efciency using this setup was 94% at 1.80 V vs
SHE. This result is similar to what has been reported by Hori
et al. where Pb electrodes have high selectivity towards formate at
ambient conditions [11]. Although the electrochemical reduction
using Pb-granules in a xed bed reactor does exhibit higher current efciencies at higher temperatures and pressures compared to
ambient conditions, such efciencies were achievable under ambient conditions using Pb electrodes alone [11].
Electrodeposited tin (Sn) on carbon paper was used to create a
gas diffusion electrode [89]. The formation of formic acid was only
reported when operated at 1.60 V vs SHE and slightly elevated
temperatures of 40 C. Current efciency was very low at only 18%
compared to the regular Sn electrode used by Hori et al. which
reported an efciency of 88% at only 1.48 V [11].
Please cite this article in press as: R.J. Lim, et al., A review on the electrochemical reduction of CO2 in fuel cells, metal electrodes and molecular
catalysts, Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.11.037
G Model
CATTOD-8759; No. of Pages 12
ARTICLE IN PRESS
R.J. Lim et al. / Catalysis Today xxx (2013) xxxxxx
Molybdenum (Mo) exhibits the unique ability to electrochemically reduce CO2 directly to methanol [90]. When pre-treated with
HCl together with voltage cycling between 1.20 and +0.20 V, the
electroreduction at 0.80 V vs SHE is able to achieve an efciency
of 370%. The disadvantage of Mo electrodes is that it suffers from
corrosion during open circuit conditions.
Recently, a DFT study was performed on metal-functionalised
porphyrin-like graphene [91]. Various types of metals such as late
transition metals (Cu, Ag, Au, Ni, Pd, Pt, Co, Rh, Ir, Fe, Ru, Os), B, Al,
Ga and Mg had been used as the active site. The best catalyst for the
system was calculated to be Rh and Fe for production of methanol
and methane, respectively, due to the low voltage requirements
(0.19 V and 0.93 V respectively). Despite being predicted to be
able to reduce CO2 at such low voltages, the system also suffers
from a competing reaction where H* competes for the active site. It
was suggested that CO to be used instead of CO2 for the reduction
due to the competing reaction.
4. Molecular electrocatalysts
Various metal complexes such as copper [9294], palladium
[9599], nickel [100108] and cobalt [21,109116] have been
reported to be able to catalyse the electrochemical reduction of CO2 .
Benson et al. suggested that specic catalysts should be utilised in
specic steps of the CO2 electrochemical reduction process as metal
complexes can provide high selectivity towards specic products
such as methanol [20].
4.1. Copper complexes
Copper complexes have been shown to be able to xate CO2
directly from air when the pH is above 7 [92,94]. CO2 xated to the
complexes will form carbamates when there is only one copper
centre present while two copper centres will cause the formation
of carbonates instead [94]. The reduction of CO2 at the complex
was shown to occur in between 1.00 and 1.25 V vs SHE [92]. A
dinuclear copper complex synthesised by Angamuthu et al. spontaneously captured and reduced CO2 with applied potentials of only
0.03 V vs SHE. Two CO2 molecules formed bridges between the
copper complexes forming oxalate which is then extracted using
lithium salt [93].
4.2. Palladium complexes
Electrochemical reduction under anhydrous conditions begins
at about 1.30 V when monodentate palladium complexes are
used. Presence of water in the electrolyte produced hydrogen and
formic acid but no CO was detected in both conditions [95]. Presence of pyridine and pyrazole ligands in the complexes could
function as electron acceptors at suitable potentials. However, the
increase in donor strength of the ligands enhanced the amount of
CO and decreases the formic acid production [96].
DuBois et al. synthesised mono- and bipalladium phosphine
complexes with palladium as reaction centres. The complex consists of a weakly binding solvent portion (i.e., acetonitrile) that
will dissociate during catalysis revealing a vacant site for binding
of water and assisting in the cleavage of C O bond [9799]. The
reaction scheme for their catalyst is as shown in Fig. 7. The rate
determining step is dependent on the concentration of acid present
in the electrolyte. At high concentrations, the binding of CO2 to the
palladium site is the rate determining step (step 3 4) as H2 production is much more favourable. On the other hand, the cleavage
of the C O bond (step 8 9) is rate determining and will require
much higher potentials.
Fig. 7. Proposed mechanism for electrochemical reduction of CO2 using Pd complexes synthesised by DuBois et al.
Reproduced from Ref. [99] by permission from the American Chemical Society.
Please cite this article in press as: R.J. Lim, et al., A review on the electrochemical reduction of CO2 in fuel cells, metal electrodes and molecular
catalysts, Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.11.037
G Model
CATTOD-8759; No. of Pages 12
ARTICLE IN PRESS
R.J. Lim et al. / Catalysis Today xxx (2013) xxxxxx
Fig. 8. Possible mechanism for the regeneration of the pyridinium catalyst [121].
the nitrogen atom and the electron density reduced after reduction. The intermediate, pyridinium-CO2 , a carbamate-like species
was proposed to be the rate determining step for the process [123].
Using DFT and continuum solvation model, Keith et al. showed
that the pyridinyl radical (Fig. 8(I)) is not the active intermediate responsible for CO2 reduction and should not form during the
process as it would require an applied potential of 1.53 V [124].
Further calculations revealed that a more probable intermediate is
the dihydropyridine (DHP) as the calculated reduction potential of
0.72 V is much closer to experimental values of 0.58 V vs SHE
[125]. DFT study for the formation of formic acid revealed that the
pyridinium cation was rst reduced by adsorbing the hydrogen to
the electrode surface which is susceptible to electrophilic attack
by CO2 (Eq. (21)). The reaction then proceeds with a two electron
proton coupled hydride transfer thus forming formic acid (Eq. (22))
[126].
PyrH+ + Pt + e Pyr + Pt-H
(21)
(22)
Please cite this article in press as: R.J. Lim, et al., A review on the electrochemical reduction of CO2 in fuel cells, metal electrodes and molecular
catalysts, Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.11.037
ARTICLE IN PRESS
G Model
CATTOD-8759; No. of Pages 12
10
Table 1
The electrochemical reduction products of CO2 over various catalysts.
Catalysts
Products
CO
CO, methane
CO
CO
Molecular electrocatalysts
Copper complexes
Palladium complexes
Nickel complexes
Cobalt complexes
Pyridinium
Oxalate
CO, formic acid
CO, oxalate, formic acid, formaldehyde
CO, formic acid, alcohols
Methanol, formaldehyde, formic acid
In an attempt to overcome the input requirement in the electrochemical reduction of CO2 , Zhao et al. combined a microbial
fuel cell (MFC) with a microbial electrolysis cell (MEC) utilising a
graphite electrode modied with Co-TaPc with carbon nanotubes
[138]. The main product for this setup was formic acid at a voltage of 0.50 V with a peak current efciency of 77%. The system
showed no degradation even after 10 four hour cycles of reduction
was performed.
Recently, a current density of up to 3 A/cm2 was achieved using
stainless steel electrodes together with Geobacter sulfurreducens
[139]. Only 0.6 V vs SHE was applied and glycerol was the only
product. The possible mechanism of the process is still unknown
but was proposed to be the consumption of succinate due to the
reduced CO2 (Eq. (23)).
1/2C4 H6 O4 + HCO3 + 8H+ + 7e C3 H8 O3 + 2H2 O
(23)
are dependent on the type of the electrodes used. Copper and its
oxide Cu2 O has gained a lot of interest as it is able to produce
a wide range of hydrocarbons and oxygenates as products with
high efciency. Nanostructured copper such as nanoparticles and
nanoarrays may provide suitable crystal facets or surfaces tuned
to produce specic product for efcient electroreduction of CO2 .
The actual mechanism is still a work in progress though recent
studies have begun to unveil the possible process that is occurring.
Despite having such unique properties to electrochemically reduce
CO2 , it is still prone to poisoning from trace elements present in
the electrolyte. Surface modication, alloying and voltage modulation could potentially improve the performance of Cu electrodes by
delaying the onset of deactivation but a more permanent solution
is needed.
Various complexes have been proposed having extremely high
activity towards CO2 capture and reduction, in particular cobalt
and nickel complexes. Required applied potentials are much lower
than that of metal electrodes while reported products are generally CO and formic acid. Some complexes exhibit good stability and
activity while there are some that degrade through extensive use.
The mechanism involved is still a work in progress as there are
many types of complexes synthesised but there is a general understanding of possible mechanism that is taking place. Pyridinium
shows good potential in the electrochemical reduction process with
reduction occurring at potentials of <1.00 V vs SHE. However, most
experimental and modelling studies are based on platinum and
palladium electrodes. For possibility of commercialisation a much
cheaper electrode with comparable performance would be necessary.
The stability of the possible intermediates formed during the
reduction process is important as it will inuence the energy or
voltage requirement of the system. Mechanistic studies will help
to uncover the possible intermediates thus allowing for better
electrocatalyst design. DFT studies have uncovered possible mechanisms that closely matched some experimental observations but
there still remain some disagreements on what could actually
be going on. Possible methods in stabilising intermediates in the
case for methane formation in aqueous electrolyte are alloying
of electrodes, stabilisation using ligand, tethering and additional
promoters on the electrode [140]. Increased stability of key intermediates should require much lower applied voltages and more
favourable protonation of the intermediate.
We acknowledge nancial support from the Ministry of Educations academic research fund AcRF tier 1 (RG8/12 and RG73/10),
Singapore. R. J. Lim is grateful to NTU for a scholarship. This
publication is made possible by the Singapore National Research
Foundation under its Campus for Research Excellence and Technological Enterprise (CREATE) programme.
References
[1] C. Keeling, S. Piper, R. Bacastow, M. Wahlen, T. Whorf, M. Heimann, H. Meijer, Atmospheric CO2 and 13CO2 exchange with the terrestrial biosphere and
oceans from 1978 to 2000: observations and carbon cycle implications, in: I.T.
Baldwin, M.M. Caldwell, G. Heldmaier, R. Jackson, O.L. Lange, H.A. Mooney,
E.D. Schulze, U. Sommer, J. Ehleringer, M. Denise Dearing, T. Cerling (Eds.), A
History of Atmospheric CO2 and its Effects on Plants, Animals, and Ecosystems,
Springer, New York, 2005, pp. 83113.
[2] J. Ewald, Carbon dioxide at NOAAs Mauna Loa Observatory Reaches New
Milestone: Tops 400 ppm, National Oceanic and Atmospheric Administration,
2013.
[3] S. Solomon, J.S. Daniel, T.J. Sanford, D.M. Murphy, G.-K. Plattner, R. Knutti, P.
Friedlingstein, Proceedings of the National Academy of Sciences of the United
States of America 107 (2010) 1835418359.
Please cite this article in press as: R.J. Lim, et al., A review on the electrochemical reduction of CO2 in fuel cells, metal electrodes and molecular
catalysts, Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.11.037
G Model
CATTOD-8759; No. of Pages 12
ARTICLE IN PRESS
R.J. Lim et al. / Catalysis Today xxx (2013) xxxxxx
11
[54] W. Tang, A.A. Peterson, A.S. Varela, Z.P. Jovanov, L. Bech, W.J. Durand, S. Dahl,
J.K. Norskov, I. Chorkendorff, Physical Chemistry Chemical Physics 14 (2012)
7681.
[55] R. Reske, M. Duca, M. Oezaslan, K.J.P. Schouten, M.T.M. Koper, P. Strasser, The
Journal of Physical Chemistry Letters 4 (2013) 24102413.
[56] K.J.P. Schouten, Y. Kwon, C.J.M. van der Ham, Z. Qin, M.T.M. Koper, Chemical
Science 2 (2011) 19021909.
[57] K.J.P. Schouten, Z. Qin, E.P. Gallent, M.T.M. Koper, Journal of the American
Chemical Society 134 (2012) 98649867.
[58] H. Yano, T. Tanaka, M. Nakayama, K. Ogura, Journal of Electroanalytical Chemistry 565 (2004) 287293.
[59] K. Ogura, R. Oohara, Y. Kudo, Journal of the Electrochemical Society 152 (2005)
D213D219.
[60] K. Ogura, J.R. Ferrell Iii, A.V. Cugini, E.S. Smotkin, M.D. Salazar-Villalpando,
Electrochimica Acta 56 (2010) 381386.
[61] K.W. Frese, Journal of the Electrochemical Society 138 (1991) 33383344.
[62] M. Le, M. Ren, Z. Zhang, P.T. Sprunger, R.L. Kurtz, J.C. Flake, Journal of the
Electrochemical Society 158 (2011) E45E49.
[63] Y. Terunuma, A. Saitoh, Y. Momose, Journal of Electroanalytical Chemistry 434
(1997) 6975.
[64] C.W. Li, M.W. Kanan, Journal of the American Chemical Society 134 (2012)
72317234.
[65] Y.P. Peng, Y.T. Yeh, P.Y. Wang, C.P. Huang, Separation and Purication Technology 117 (2013) 311.
[66] A.A. Peterson, F. Abild-Pedersen, F. Studt, J. Rossmeisl, J.K. Norskov, Energy
and Environmental Science 3 (2010) 13111315.
[67] W.J. Durand, A.A. Peterson, F. Studt, F. Abild-Pedersen, J.K. Nrskov, Surface
Science 605 (2011) 13541359.
[68] X. Nie, M.R. Esopi, M.J. Janik, A. Asthagiri, Angewandte Chemie International
Edition 52 (2013) 24592462.
[69] J.H. Montoya, A.A. Peterson, J.K. Nrskov, ChemCatChem 5 (2013) 737742.
[70] H. Wu, N. Zhang, H. Wang, S. Hong, Chemical Physics Letters 568569 (2013)
8489.
[71] P. Hirunsit, The Journal of Physical Chemistry C 117 (2013) 82628268.
[72] M. Watanabe, M. Shibata, A. Kato, M. Azuma, T. Sakata, Journal of the Electrochemical Society 138 (1991) 33823389.
[73] M. Watanabe, M. Shibata, A. Katoh, T. Sakata, M. Azuma, Journal of Electroanalytical Chemistry and Interfacial Electrochemistry 305 (1991) 319328.
M.L. Avramov-Ivic,
N.B. Vukovic,
Journal of Electroanalytical Chem[74] J.P. Popic,
istry 421 (1997) 105110.
[75] J. Li, G. Prentice, Journal of the Electrochemical Society 144 (1997) 42844288.
[76] J. Sobkowski, A. Czerwinski, Journal of Physical Chemistry 89 (1985) 365369.
[77] N. Hoshi, S. Kawatani, M. Kudo, Y. Hori, Journal of Electroanalytical Chemistry
467 (1999) 6773.
[78] N. Hoshi, Y. Hori, Electrochimica Acta 45 (2000) 42634270.
[79] J. Qu, X. Zhang, Y. Wang, C. Xie, Electrochimica Acta 50 (2005) 35763580.
[80] K. Ohkawa, K. Hashimoto, A. Fujishima, Y. Noguchi, S. Nakayama, Journal of
Electroanalytical Chemistry 345 (1993) 445456.
[81] K. Ohkawa, Y. Noguchi, S. Nakayama, K. Hashimoto, A. Fujishima, Journal of
Electroanalytical Chemistry 348 (1993) 459464.
[82] K. Ohkawa, Y. Noguchi, S. Nakayama, K. Hashimoto, A. Fujishima, Journal of
Electroanalytical Chemistry 367 (1994) 165173.
[83] K. Ohkawa, Y. Noguchi, S. Nakayama, K. Hashimoto, A. Fujishima, Journal of
Electroanalytical Chemistry 369 (1994) 247250.
[84] D. Kolbe, W. Vielstich, Electrochimica Acta 41 (1996) 24572460.
[85] N. Hoshi, M. Noma, T. Suzuki, Y. Hori, Journal of Electroanalytical Chemistry
421 (1997) 1518.
[86] C. Iwakura, S. Takezawa, H. Inoue, Journal of Electroanalytical Chemistry 459
(1998) 167169.
[87] Y. Chen, C.W. Li, M.W. Kanan, Journal of the American Chemical Society 134
(2012) 1996919972.
[88] F. Kleli, D. Balun, Applied Catalysis A: General 274 (2004) 237242.
[89] R.L. Machunda, H. Ju, J. Lee, Current Applied Physics 11 (2011) 986988.
[90] D.P. Summers, S. Leach, K.W. Frese Jr., Journal of Electroanalytical Chemistry
and Interfacial Electrochemistry 205 (1986) 219232.
[91] V. Tripkovic, M. Vanin, M. Karamad, M.E. Bjrketun, K.W. Jacobsen, K.S. Thygesen, J. Rossmeisl, The Journal of Physical Chemistry C 117 (2013) 91879195.
P. Gavina,
S. Tatay, A.
[92] B. Verdejo, S. Blasco, J. Gonzlez, E. Garca-Espana,
Domnech, M.T. Domnech-Carb, H.R. Jimnez, C. Soriano, European Journal
of Inorganic Chemistry 2008 (2008) 8497.
[93] R. Angamuthu, P. Byers, M. Lutz, A.L. Spek, E. Bouwman, Science 327 (2010)
313315.
P. Gavina,
J. Latorre, C. Soriano, B. Verdejo, Journal of the
[94] E. Garca-Espana,
American Chemical Society 126 (2004) 50825083.
[95] A.G.M.M. Hossain, T. Nagaoka, K. Ogura, Electrochimica Acta 41 (1996)
27732780.
[96] A.G.M.M. Hossain, T. Nagaoka, K. Ogura, Electrochimica Acta 42 (1997)
25772585.
[97] B.D. Steffey, C.J. Curtis, D.L. DuBois, Organometallics 14 (1995) 49374943.
[98] J.W. Raebiger, J.W. Turner, B.C. Noll, C.J. Curtis, A. Miedaner, B. Cox, D.L. DuBois,
Organometallics 25 (2006) 33453351.
[99] M. Rakowski Dubois, D.L. Dubois, Accounts of Chemical Research 42 (2009)
19741982.
[100] M. Beley, J.P. Collin, R. Ruppert, J.P. Sauvage, Journal of the American Chemical
Society 108 (1986) 74617467.
[101] J.P. Collin, A. Jouaiti, J.P. Sauvage, Inorganic Chemistry 27 (1988) 19861990.
Please cite this article in press as: R.J. Lim, et al., A review on the electrochemical reduction of CO2 in fuel cells, metal electrodes and molecular
catalysts, Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.11.037
G Model
CATTOD-8759; No. of Pages 12
12
ARTICLE IN PRESS
R.J. Lim et al. / Catalysis Today xxx (2013) xxxxxx
Please cite this article in press as: R.J. Lim, et al., A review on the electrochemical reduction of CO2 in fuel cells, metal electrodes and molecular
catalysts, Catal. Today (2013), http://dx.doi.org/10.1016/j.cattod.2013.11.037