Selective reduction of CO2 to formate through bicarbonate reduction on metal electrodes: New insights gained from SG/TC mode of SECM

Article abstract of DOI:10.1039/c4cc03099k

We discovered using SECM of the electro-reduction of CO₂ on a Au substrate in CO₂-saturated KHCO₃ solutions that (i) formate comes solely from the direct reduction of bicarbonate; and (ii) CO forms only from CO₂ reduction (under low pH conditions) and at higher applied potentials. The results point to the possibility of the selective reduction of CO₂ to the formate product.

Full text of DOI:10.1039/c4cc03099k

ChemComm
View Article Online
View Journal | View Issue
COMMUNICATION
Selective reduction of CO2 to formate through
bicarbonate reduction on metal electrodes: new
insights gained from SG/TC mode of SECM†
Cite this: Chem. Commun., 2014,
50, 11143
Received 26th April 2014,
Accepted 26th July 2014
Narayanaru Sreekanth and Kanala Lakshminarasimha Phani*
DOI: 10.1039/c4cc03099k
www.rsc.org/chemcomm
We discovered using SECM of the electro-reduction of CO2 on a Au to their ability to convert carbon dioxide to carbon monoxide
substrate in CO2-saturated KHCO3 solutions that (i) formate comes selectively in aqueous solutions at room temperature.8
solely from the direct reduction of bicarbonate; and (ii) CO forms
According to Hori and later workers, CO2 electro-reduction
only from CO2 reduction (under low pH conditions) and at higher on Au and Ag in 0.1 M KHCO3 solution produce CO as the
applied potentials. The results point to the possibility of the selec- major product with a small percentage of formate ions.8,9
tive reduction of CO2 to the formate product.
The proximity of the potentials at which both the CO2 reduction
and hydrogen evolution reaction (HER) take place, precludes
Converting carbon dioxide into useful chemicals in a selective the possibility of observing current–potential features in aqueous
and efficient manner remains a major challenge in renewable voltammetry. This is made more difficult as the presence of CO2
and sustainable energy research.1 Among the several approaches for moves the solution pH into the acidic region, and thus the HER
carbon dioxide fixation,2 its cathodic reduction is one of the more becomes a significant interference. As the current–potential plots
promising steps in the total process of carbon dioxide conversion.1,3 are invariably featureless, preliminary identification of the nature
While the molecular level understanding of the reduction process of the products poses a challenge. Hence, it is necessary to resort
has paved the way for the discovery of different molecular cata- to other ways of detecting the reduction products online. In bulk
lysts,4–6 the electrochemical reduction of CO2 at metal electrodes7 in electrolysis, the reduction is usually carried out at a fixed potential
aqueous solutions leads to products like CO, HCOOÀ, CH4, C2H4 and the products are analyzed by titration methods,14 GC-MS,15
and alcohols in addition to H2.8–10 Recently, Compton et al.11 NMR16 and differential electrochemical mass spectrometry.17 How-
reviewed the work on the use of room temperature ionic liquids ever, simpler and faster methods of preliminary screening of
for the electro-reduction of CO2 and in one of their communica- catalytic surfaces prior to undertaking bulk electrolysis are preferred.
tions12 formic acid electro-synthesis was studied in acidified ethyl- It is advantageous that formate/formic acid, alcohols and CO,
methylimidazolium bis(trifluoromethylsulfonyl)imide. However, invariably found in the analysis of CO2 reduction products are
there is a need to evolve electrocatalysts that are effective in aqueous electroactive and hence amenable to electrochemical detection.
environments. Though the metals with high hydrogen over- Earlier work on the use of a rotating ring-disk electrode set-up for
potentials gave the highest faradaic efficiencies, only those with analyzing the products formed an interesting part of the electro-
low hydrogen overpotentials offered higher energy efficiency. chemical studies on CO2 reduction.18 In comparing SECM and
Hence, obtaining a balance of faradaic vs. energy efficiency with RRDE,19 the former is preferred as its collection efficiency is higher
minimum hydrogen evolution requires better understanding of than that of RRDEs (improving the precision of the measurement)
`
CO2 reduction on various surfaces vis-a-vis product distribution. and is more convenient, since it allows variation of the substrate–tip
Nanoparticle variants of many metals13 have also been the subject distance (d) (compared to the fixed ring-disk spacing) and simple
of exciting recent research activity in the area of carbon dioxide replacement of the substrate (compared to replacement of the
conversion. Gold and silver are interesting electrocatalysts owing disk material). In this work, we employ the substrate-generation/
tip-collection (SG-TC) mode of scanning electrochemical micro-
À
Nanoscale Electrocatalysis Group, Electrodics & Electrocatalysis Division,
CSIR-Central Electrochemical Research Institute, Karaikudi, 630006, India.
E-mail: kanalaphani@yahoo.com, klnphani@cecri.res.in;
Fax: +91-4565-227779, 91-4565-227713
scopy (SECM) to sense the products of the reduction of CO2/HCO3
at a Au substrate (and Ag) using a platinum ultramicroelectrode
(tip). In the SG-TC experiments the tip travels within a thick
diffusion layer produced by the large substrate.19
† Electronic supplementary information (ESI) available: Chemicals, experimental
procedures, electrochemical measurements and NMR analysis. See DOI: 10.1039/
c4cc03099k
In the course of our investigations on CO2 reduction at Au
electrodes in 0.1 M KHCO3 solutions, we observed featureless
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 11143--11146 | 11143

View Article Online
Communication
ChemComm
To ensure the approach of the tip to the substrate in the
SG-TC mode of SECM, a standard probe-approach curve
[Fig. 2a and b] was constructed using a ferrocenemethanol
probe (1 mM in 0.2 M KCl), and the probe was brought E20 mm
(d/a = 4) (where ‘a’ is the tip radius) close to the substrate (as it
is appropriate to maintain d/a c 1 in order to avoid problems
due to interference between diffusion layers of the substrate
and the tip20) before each series of experiments on bicarbonate
or CO2 reduction. A set of cyclic voltammograms at a Pt UME tip
(10 mm) for the oxidation of products of the bicarbonate
reduction at the Au substrate in 0.1 M KHCO3 solution are
shown in Fig. 2d. The responses were acquired at the Pt UME
tip kept at a distance (d) of approx. 20 mm above the Au substrate
that is biased at various cathodic potentials. As can be seen from
the CV panel in Fig. 2d, at a potential of À0.85 V, features of the
oxidation of small organic molecules21 start emerging to assume
full-fledged characteristics of formic acid oxidation [Figure 1 of
ref. 22b] under lower alkaline conditions (the current character-
istic of the Pt UME tip in the region of À0.3 to À0.6 V (Fig. 2) is
likely to arise from the oxidation of molecular hydrogen22). 13C
NMR spectroscopy confirms the formation of formic acid as the
main product, see the ESI,† Fig. S11.
Fig. 1 Linear sweep voltammetry of CO2 reduction on the Au substrate in
0.1 M KHCO3 solution at a scan rate of 0.05 V sÀ1
.
voltammograms only marked by an infinite current increase
at the extreme cathodic potentials both in the absence and
presence of CO2 [Fig. 1]. However, interestingly when a Pt
ultramicroelectrode (Pt-UME probe) tip is placed close to the
Au substrate (cathode) in the SG-TC mode of SECM [Fig. 2c],
voltammetric features typical of formic acid oxidation on Pt
appeared in the tip voltammograms. In addition, voltammetric
features displaying the oxidation of both formic acid and CO
appear when KHCO3 is saturated with CO2, to a varying extent
depending on the pH of the electrolyte and the applied sub-
strate potential. We present in this communication (a) the
possibility of generation of formate by direct reduction of
bicarbonate alone; and (b) highlight the fact that CO forms
only under low pH conditions (subsequent to the formation of
formate) taking advantage of the SG-TC mode for determining
the exact potentials at which the onset of product generation
(here, formate and CO) occurs.
The scenario totally changes when the bicarbonate solution
is saturated with CO2, with the associated change in pH from
8.3 to 6.8. Depending on the substrate potential, a sharp
voltammetric peak at 0.6 V vs. Ag/AgCl emerges to increase in
magnitude at the expense of the response of the formic acid
oxidation [Fig. 3a]. It is easy to understand the origin of this
Fig. 3 (a) Cyclic voltammetric responses of Pt UME (10 mm)-tip probe to
the products (HCOOÀ and CO) generated at Au substrate in 0.1 M KHCO3
solution saturated with CO2; tip scan rate: 0.05 V sÀ1. (b) Responses of Pt
UME (10 mm)-tip probe to the oxidation of CO in CO bubbled 0.1 M KHCO3
(red) and CO generated from Au substrate in 0.1 M KHCO3 solution
saturated with CO2 (black); (c) 0.1 M KHCO3 + PBS solution of pH = 6.8.
Fig. 2 (a) Positive and (b) negative approach curves of SECM using a Pt
10 mm UME; (c) schematic of the SG-TC mode for bicarbonate reduction–
formate oxidation. (d) Cyclic voltammetric responses of the Pt UME
(10 mm)-tip probe to the products (HCOOÀ) generated at the Au substrate
in 0.1 M KHCO3. Tip scan rate: 0.05 V sÀ1
.
11144 | Chem. Commun., 2014, 50, 11143--11146
This journal is ©The Royal Society of Chemistry 2014

View Article Online
ChemComm
Communication
comparison with our results with a Au substrate. In accordance
with the results reported in ref. 25, we noticed current peaks
corresponding to the direct reduction of bicarbonate in our
studies [ESI,† Fig. S3 and S4] and identified that the peaks are
due to the formation of formate ions, with the aid of the SG-TC
mode. Based on the magnitude of the Pt UME tip currents
recorded, it may be stated that the Au surface is more favorable
than Pd in producing formate ions from bicarbonate.
We examined the substrate surfaces belonging to the class
of metals (Hg, Bi and Sn) that are classified to yield only formic
acid/formate as the product of CO2 reduction,8,9 viz., Hg and Bi
as well as Pd are known to produce formate from the reduction
of bicarbonate.26 We collected the tip responses for the pro-
ducts of reduction at the Hg-, Bi- and Pd-substrates. In the case
of the Hg substrate, potentials as high as À2.0 V vs. Ag/AgCl were
required to initiate the generation of formate from bicarbonate,
Fig. 4 Cyclic voltammetric responses of Pt UME (10 mm)-tip probe to the
product (HCOOÀ) generated at the Au substrate in 0.1 M carbonate buffer
solution pH = 9 (black) and (ii) 0.5 M KOH solution saturated with CO2 pH = 8
(red); tip scan rate: 0.05 V sÀ1; substrate potential ES fixed at À1.2 V.
sharp feature, by recalling the conventional knowledge of while the potential is shifted 0.5 V less cathodic when the
CO-oxidation on a Pt surface.23 To prove that this is due to bicarbonate solution is CO2-saturated to produce only formate
the oxidation of CO, the following control experiments were [ESI,† Fig. S5 and S6]. Bismuth-coated glassy carbon substrates
carried out. Tip responses were acquired for the reduction of also show the same trend but with reduction occurring at much
species produced at the Au substrate from (a) the bicarbonate lower cathodic potentials than on Hg (ESI,† Fig. S7 and S8).
solution acidified using a phosphate-buffer of pH 6.8 (equiva- These results suggest that (a) formate/formic acid generation
lent to CO2-saturated bicarbonate solution) in the SG-TC mode; at these electrodes from CO2 is easier than from bicarbonate;
and (b) CO-oxidation at the Pt UME in bicarbonate solutions and (b) only formate/formic acid are generated until the potential
bubbled with carbon monoxide. The similarities in the shape limit of the hydrogen evolution reaction.
and position of the anodic peak pattern in Fig. 3b and c confirm
It is now clear that at Au and Ag surfaces (a) bicarbonate
that the product of the reaction mixtures of (a) and (b) is indeed undergoes direct reduction; (b) in CO2-saturated bicarbonate
carbon monoxide, arising from CO2 reduction on the Au surface. solutions (pH 6.8), bicarbonate undergoes reduction first to
Whereas, the solutions of (i) CO2 + KOH, and (ii) carbonate produce formate and at higher applied potentials, CO2 is reduced
buffer (1: 1 mol% K2CO3 : KHCO3) yield formate as the product to CO; (c) in alkaline solutions (pH 8.5 to 10), CO2-saturated
[Fig. 4]. That is, when electroreduction is performed in CO2 + solutions yield only formate as the main product; and (d) even in
0.5 M KOH solutions (pH = 8.3), only formate ions are formed. slightly acidic solutions (pH 6.5), CO appears to be the major
However, when the pH was changed to 6.5, CO is obtained as a product. More importantly, it is demonstrated here that one can
major product with minor quantities of formate. The following have control over the nature of the product species by exactly
simple equilibria are of relevance to the situation that we are fixing the applied potential. This would suggest the possibility of
describing:
obtaining selectivity between formate and CO based on the
applied potentials. It will be much more advantageous if CO2 is
converted to bicarbonate which in turn undergoes reduction at
lower applied potentials. The present results in juxtaposition with
the suggestion of Koper et al.26 assume significance when other
À
CO2 + OHÀ $ HCO3
(1)
(2)
CO2 + H2O $ HCO3 + H+
À
Similar results were obtained in the case of Ag with inherent more catalytic metals or alloys (for example, copper) are used.
differences in the position of the potentials. In spite of the Incidentally, the local pH measurements made [ESI,† see the
reported similarities in catalytic behavior of Ag and Au, accord- notes after Fig. S10] during various time intervals of electro-
ing to Hori’s classification of metals with CO as the major lysis show a pH shift towards more alkalinity, pointing to a
product, the (substrate) potentials are significantly more cathodic plausible reduction mechanism involving the following reac-
than those observed with Au (ESI,† Fig. S1 and S2). The SG-TC tions [eqn (3) and (4)].
mode is also unique in that it enables identification of the exact
HCO3 + H2O + 2eÀ - HCO2 + 2OHÀ
(3)
(4)
À
À
substrate potential [e.g., Fig. 2 and 3] at which the useful products
are formed (formate/formic acid, and CO) with total control over
the potential needed for the avoidance of hydrogen evolution.
A few control experiments involving other metal surfaces that
CO2 + H2O + 2eÀ - CO + 2OHÀ
The present results will eventually lead to the classification
are electrocatalytic for the reduction of (i) bicarbonate to formate; of metal surfaces for bicarbonate reduction, as has been done
and (ii) CO2 to formate, are required here. Direct reduction of for CO2 reduction (ca. the work of Hori et al.). With a thesis of
bicarbonate to formate at a palladium electrode was established by recent observations on the bicarbonate reduction at copper
Andre and Wrighton,24 and Spichiger-Ulmann and Augustynski.25 electrodes26 and other studies reported on nanostructured
Thus, Pd can serve as a good benchmark for quantitative electrocatalysts13,27 it will be possible to design surfaces for
This journal is ©The Royal Society of Chemistry 2014
Chem. Commun., 2014, 50, 11143--11146 | 11145

View Article Online
Communication
ChemComm
10 M. Jitaru, D. A. Lowy, M. Toma, B. C. Toma and L. Oniciu, J. Appl.
Electrochem., 1997, 27, 875–889.
11 N. V. Rees and R. G. Compton, Energy Environ. Sci., 2011, 4, 403–408.
12 B. C. M. Martindale and R. G. Compton, Chem. Commun., 2012, 48,
6487–6489.
selective reduction. We are currently pursuing approaches
for electrocatalyst selection that in turn need deployment of
product-specific ultramicroelectrode tips for in situ process
monitoring28 and for mechanistic elucidation that is more
involved than the simplistic picture presented in the reactions
in eqn (3) and (4).
NS thanks India’s UGC for a research fellowship. KLNP
acknowledges support from India’s DST for a grant under
SR/NM/NS-1036/2011 and CSIR funding under MULTIFUN:
CSC-0101.
13 (a) Y. Chen, C. W. Li and M. W. Kanan, J. Am. Chem. Soc., 2012, 134,
19969–19972; (b) Q. Lu, J. Rosen, Y. Zhou, G. S. Hutchings,
Y. C. Kimmel, J. G. Chen and F. Jiao, Nat. Commun., 2014, 5, 3242;
(c) B. A. Rosen, A. Salehi-Khojin, M. R. Thorson, W. Zhu, D. T.
Whipple, P. J. A. Kenis and R. I. Masel, Science, 2011, 334, 643–644;
¨
(d) W. Zhu, R. Michalsky, O. Metin, H. Lv, S. Guo, C. J. Wright, X. Sun,
A. A. Peterson and S. Sun, J. Am. Chem. Soc., 2013, 135, 16833–16836.
14 Y. Hori and S. Suzuki, J. Electrochem. Soc., 1983, 130, 2387–2390.
15 A. S. Bandarenka, E. Ventosa, A. Maljusch, J. Masa and
W. Schuhmann, Analyst, 2014, 139, 1274.
16 K. P. Kuhl, E. R. Cave, D. N. Abram and T. F. Jaramillo, Energy
Environ. Sci., 2012, 5, 7050–7059.
Notes and references
1 (a) E. E. Benson, C. P. Kubiak, A. J. Sathrum and J. M. Smieja, Chem. 17 (a) K. Jambunathan and A. C. Hillier, J. Electrochem. Soc., 2003, 150,
Soc. Rev., 2009, 38, 89–99; (b) D. T. Whipple and P. J. A. Kenis, J. Phys.
Chem. Lett., 2010, 1, 3451–3458; (c) M. Jacoby, The Hidden Value of
Carbon Dioxide, C&EN, 2013, vol. 91, pp. 21–22.
2 (a) O. Vechorkin, N. Hirt and X. Hu, Org. Lett., 2010, 12, 3567–3569;
(b) P. G. Jessop, I. Takao and N. Ryoji, Chem. Rev., 1995, 95, 259–272;
312–320; (b) K. Jambunathan, S. Jayaraman and A. C. Hillier, Lang-
muir, 2004, 20, 1856–1863.
18 (a) J. Zhang, W. J. Pietro and A. B. P. Lever, J. Electroanal. Chem.,
1996, 403, 93–100; (b) A. Aoki and G. Nogami, J. Electrochem. Soc.,
1995, 142, 423–427.
(c) X. Zhang, F. Han, B. Shi, S. Farsinezhad, G. P. Dechaine and 19 (a) C. M. Sanchez Sanchez, J. Rodriguez Lopez and A. J. Bard, Anal.
K. Shankar, Angew. Chem., Int. Ed., 2012, 51, 12732–12735; (d) Y. Oh
and X. Hu, Chem. Soc. Rev., 2013, 42, 2253–2261.
3 J. Qiao, Y. Liu, F. Hong and J. Zhang, Chem. Soc. Rev., 2014, 43,
631–675.
Chem., 2008, 80, 3254–3260; (b) P. Sun, O. Laforge and M. V Mirkin,
Phys. Chem. Chem. Phys., 2007, 9, 802–823; (c) S. Cannan, J. Cervera,
R. J. Steliaros, E. Bitziou, A. L. Whitworth and P. R. Unwin, Phys.
Chem. Chem. Phys., 2011, 13, 5403–5412.
4 (a) I. Bhugun, D. Lexa, J. Save and D. Diderot, J. Am. Chem. Soc., 1996, 20 M. V. Mirkin, in Scanning Electrochemical Microscopy, ed. A. J. Bard
118, 1769–1776; (b) K. Leung, I. M. B. Nielsen, N. Sai, C. Medforth and and M. V. Mirkin, Marcel Dekker, 2001, ch. 5, p. 145.
J. A. Shelnutt, J. Phys. Chem. A, 2010, 114, 10174–10184; (c) K. Ogura 21 (a) H. Wang, L. Alden, F. J. DiSalvo and H. D. Abruna, Phys. Chem.
and I. Yoshida, J. Mol. Catal., 1988, 47, 51–57.
5 (a) Z. F. Chen, C. C. Chen, D. R. Weinberg, P. Kang, J. J. Concepcion,
D. P. Harrison, M. S. Brookhart and T. J. Meyer, Chem. Commun.,
Chem. Phys., 2008, 10, 3739–3751; (b) F. J. E. Scheijen, G. L. Beltramo,
S. Hoeppener, T. H. M. Housmans and M. T. M. Koper, J. Solid State
Electrochem., 2008, 12, 483–495.
2011, 47, 12607–12609; (b) M. R. Dubois and D. L. Dubois, Acc. Chem. 22 (a) J. John, H. Wang, E. D. Rus and H. D. Abruna, J. Phys. Chem. C,
Res., 2009, 42, 1974–1982.
6 (a) E. B. Cole, P. S. Lakkaraju, D. M. Rampulla, A. J. Morris, E. Abelev
2012, 116, 5810–5820; (b) J. Joo, T. Uchida, A. Cuesta, M. T. M. Koper
and M. Osawa, J. Am. Chem. Soc., 2013, 135, 9991–9994.
and A. B. Bocarsly, J. Am. Chem. Soc., 2010, 132, 11539–11551; (b) M. Z. 23 G. Gonzalo and M. T. M. Koper, Phys. Chem. Chem. Phys., 2008, 10,
Ertem, S. J. Konezny, C. M. Araujo and V. S. Batista, J. Phys. Chem. Lett.,
2013, 4, 745–748.
3802–3811.
24 J. F. Andre and M. S. Wrighton, Inorg. Chem., 1985, 24, 4288–4292.
7 (a) J. L. Dimeglio and J. Rosenthal, J. Am. Chem. Soc., 2013, 135, 25 M. Spichiger-Ulmann and J. Augustynski, J. Chem. Soc., Faraday
8798–8801; (b) K. J. P. Schouten, Y. Kwon, C. J. M. van der Ham,
Z. Qin and M. T. M. Koper, Chem. Sci., 2011, 2, 1902–1909.
8 Y. Hori, in Modern Aspects of Electrochemistry, ed. C. G. Vayenas,
Trans. 1, 1985, 81, 713–716.
26 R. Kortlever, K. H. Tan, Y. Kwon and M. T. M. Koper, J. Solid State
Electrochem., 2013, 17, 1843–1849.
R. E. White and M. E. Gamboa-Aldeco, Springer, 2008, vol. 42, pp. 89–189. 27 S. Zhang, P. Kang and T. J. Meyer, J. Am. Chem. Soc., 2014, 136, 1734–1737.
9 W. Li, in Advances in CO2 conversion and utilization, ed. Y. Hu, 28 E. D. Rus, H. Wang, A. E. Legard, N. L. Ritzert, R. B. Van Dover and
ACS Symp. Ser., 2010, ch. 5, pp. 55–76.
H. D. Abruna, Rev. Sci. Instrum., 2013, 84(2), 024101.
11146 | Chem. Commun., 2014, 50, 11143--11146
This journal is ©The Royal Society of Chemistry 2014