ChemComm
View Article Online
COMMUNICATION
Published on 17 January 2018. Downloaded by Indian Institute of Technology Kanpur on 12/11/2023 7:34:43 PM.
Electrocatalytic CO2 reduction by a cobalt
bis(pyridylmonoimine) complex: effect of acid
Cite this: Chem. Commun., 2018,
54, 1579 concentration on catalyst activity and stability†
Received 6th November 2017,
Accepted 17th January 2018 Weixuan Nie and Charles C. L. McCrory
DOI: 10.1039/c7cc08546j
rsc.li/chemcomm
A Co complex with a redox-active bis(pyridylmonoimine) ligand has
been prepared and shows catalytic activity for electrochemical CO2
reduction in acetonitrile. Addition of a proton source such as water
or trifluoroethanol dramatically improves the activity and stability
of the molecular catalyst. The Co complex reduces CO2 to CO
selectively at 1.95 V vs. Fc+/0 in the presence of high concentrations
of water. The activity of the Co complex for CO2 reduction compares
favorably to other molecular Co-based catalysts in acetonitrile
solutions.
The electrochemical CO2 reduction reaction (CO2RR) is a promising
strategy for the conversion of waste CO2 into value added
products.1–3 State-of-the-art solid-state CO2 reduction catalysts
such as Cu reduce CO2 to highly-reduced products, but suffer
from poor selectivity and generate multiple gaseous and liquid
products.4,5 In contrast, several molecular electrocatalysts
reduce CO2 selectively to single products, making them an
important class of CO2RR catalysts.6,7 Moreover, the activity
and selectivity of molecular catalysts can be tuned by altering
their ligand structure.8,9
Redox-active ligands have been shown to play an important
role in CO2RR by molecular electrocatalysts.9–16 Redox-active
ligands are able to store electrons in the ligand structure,
facilitating the multi-electron reduction of CO2. Previous
reports have shown that cobalt complexes with pyridylimine-
based ligands efficiently reduce CO2 to CO in the presence of
H2O14,17–19 and propose that the redox-active properties of the
ligand contribute to the preferential reduction of CO2 over
competitive proton reduction. However, inactivation and
decomposition of the catalyst is also prevalent in these systems.
Department of Chemistry, University of Michigan, 930 North University Ave,
Ann Arbor, MI, 48109-1055, USA. E-mail: cmccrory@umich.edu
† Electronic supplementary information (ESI) available: Synthesis procedures of
the ligand and complexes; selected bond lengths and angles; NMR, IR and UV-vis
spectra; paramagnetic susceptibility measurements; selected electrochemical
measurements; SEM-EDS results; TOF calculation; ICP-MS results. CCDC
1571483. For ESI and crystallographic data in CIF or other electronic format
see DOI: 10.1039/c7cc08546j
This journal is © The Royal Society of Chemistry 2018
View Journal | View Issue
*
The decomposition mechanism may involve the formation of a
catalyst-CO2 adduct via CO2–ligand bonding with the redox
active ligands.9,20,21
In this work, a Co complex with a bis(pyridylmonoimine)
ligand [Co(L–L)Br2] has been prepared (Scheme 1) and shows
catalytic activity for CO2 reduction in acetonitrile (CH3CN). The
addition of the proton sources H2O and 2,2,2-trifluoroethanol
(TFE) accelerates the rate of electrocatalytic CO2 reduction and
also improves catalyst stability and Faradaic efficiency. In the
presence of high concentrations of H2O and TFE, the [Co(L–L)Br2]
catalyst reduces CO2 selectively to CO with high Faradaic efficiency
and high turnover frequency (TOF) compared to other reported Co
complexes with similar ligands.14,17–19
The ligand L–L was synthesized via a classic Schiff base
condensation reaction of ethylenediamine with 2-acetylpyridine.
[Co(L–L)Br2] was prepared by allowing L–L to react with
1 equivalent of CoBr2 in THF. Aerobic oxidation of [Co(L–L)Br2]
in the presence of 1 equiv. HBr(aq) for 18 hours afforded green
[Co(L–L)Br2]Br (Scheme 1), which was characterized by 1H-NMR
(Fig. S2, ESI†) and elemental analysis. The X-ray single crystal
structure of [Co(L–L)Br2]Br indicates that the central cobalt metal
adopts a bipyramidal six-coordinated configuration. The tetra-
dentate ligand L–L coordinates to Co with four N atoms in a
slightly distorted square planar mode. Two Br atoms occupy the
axial coordination sites and a third Br atom exists outside the
coordination sphere as a counter anion (Fig. 1). Note that
[Co(L–L)Br2] (the Co2+ complex) was used for all electrochemical
measurements and [Co(L–L)Br2]Br (the Co3+ complex) was used
only for characterization of the complex (see ESI†).
Scheme 1 Synthesis of the ligand L–L, complex catalyst [Co(L–L)Br2], and
[Co(L–L)Br2]Br.
Chem. Commun., 2018, 54, 1579--1582 | 1579

Communication
Published on 17 January 2018. Downloaded by Indian Institute of Technology Kanpur on 12/11/2023 7:34:43 PM.
Fig. 1 X-Ray crystal structure of [Co(L–L)Br2]Br with ellipsoids shown at
50% probability (H atoms are omitted for clarity).
Fig. 2 CV of 1.2 mM [Co(L–L)Br2] in CH3CN with 0.1 M nBu4NPF6 under
N2 (black: under N2; red: under CO2; conditions: scan rate: 50 mV s 1;
working electrode: glassy carbon working electrode; reference electrode:
Ag/AgNO3 (1 mM); counter electrode: carbon rod).
Cyclic voltammograms (CV) of complex [Co(L–L)Br2] were
recorded in a nitrogen-saturated CH3CN solution with 0.1 M
nBu4NPF6 (Fig. 2). Note that all CVs have been IR compensated.
Two reversible peaks are observed at 0.42 V vs. Fc+/0 and
1.00 V vs. Fc+/0 and are assigned as the Co3+/2+ and Co2+/1+
couples, respectively. The broad quasi-reversible redox feature
observed at B 1.6 V vs. Fc+/0 is assigned to a ligand-based
redox process. This assignment is consistent with that of metal
complexes with other redox-active pyridylmonoimine9 and
bipyridyl8 ligands. The most-negative reversible peak at 2.16 V
vs. Fc+/0 is assigned to the Co1+/0 couple.
The cyclic voltammogram (CV) of complex [Co(L–L)Br2] in a
CO2-saturated CH3CN solution shows a large increase in the
reductive current consistent with an electrocatalytic process
(Fig. 2). The onset potential of the electrocatalytic current is
B 1.65 V vs. Fc+/0, the approximate potential of the ligand
reduction. Adding the proton source H2O to the electrolyte
solution increases the magnitude of the catalytic current
(Fig. 3a for 0.82 M and 11.0 M H2O, other concentrations
shown in Fig. S9, ESI†). Increasing the concentration of H2O
also slightly shifts the catalytic peak potential, from 2.0 V vs.
Fc+/0 at 0.82 M H2O to 1.95 V vs. Fc+/0 at 11.0 M. A similar
phenomenon was observed when adding TFE as a proton
source (Fig. 3b and Fig. S10, ESI†). Since TFE is more acidic
than H2O in CH3CN,22–24 a lower concentration of TFE is
necessary to reach a comparable catalytic current under CO2.
To confirm that the catalytic peak is due to CO2 reduction and
not competitive H2 evolution, CVs were also measured in the
absence of CO2 (Fig. 3c for H2O, Fig. S11, ESI† for TFE). The
onset of H2 evolution occurs only at much more negative
1580 | Chem. Commun., 2018, 54, 1579--1582
View Article Online
ChemComm
Fig. 3 CV of 0.3 mM [Co(L–L)Br2] in CH3CN with 0.1 M nBu4NPF6 (a)
under CO2 in the presence of H2O of different concentrations; (b) under
CO2 in the presence of TFE of different concentrations; (c) under N2 and
CO2 with H2O.
potentials than the CO2 reduction peak. Note that the catalytic
peak shape in the presence of proton source resembles an
‘‘inverted peak’’ (the inverted return wave when potential is
scanned back).25 We attribute this peak shape to the overlap of the
catalytic current response for CO2 reduction with the more negative,
reversible Co1+/0 redox feature. A similar behavior has been observed
in the case of CO2 reduction by [Mn(mesbpy)(CO)3(MeCN)]+.24
To further confirm the catalytic ability of [Co(L–L)Br2] for
CO2 reduction, we conducted a series of 30 minute controlled
potential electrolysis (CPE) experiments in CH3CN (Table 1 and
Table S5, ESI†). The CPE experiments were conducted with no
IR compensation for solution resistance, and the reported
electrolysis potentials are the actual applied potentials. Note
that the electrolyte solutions as prepared contain up to B0.04 M
H2O even when no additional proton source is added.26 Gas and
liquid phases were analyzed by GC and HPLC, respectively, and
the Faradaic efficiency of each product (FEproduct) was deter-
mined. Turnover frequencies (TOFs) were calculated using two
methods: (1) by dividing the total amount of CO generated by
the catalyst amount in solution and (2) calculated from CPE
data with the equations described by Savéant et al.27,28 (see the
ESI† for further discussion of TOF). The electrolysis potentials
were chosen based on the peak potential of CO2 reduction in
Fig. 3. After each electrolysis, the un-rinsed electrode surface
was investigated with both scanning electron microscopy-energy
dispersion spectroscopy (SEM-EDS) and inductively coupled
plasma-mass spectrometry (ICP-MS) to detect the formation of
This journal is © The Royal Society of Chemistry 2018
 View Article Online

ChemComm Communication

Table 1 Conditions and product analysis of the controlled potential electrolysis for CO2 reduction

Faradaic efficiency/%
[Proton Co weight% on the
E/V vs. Fc+/0 a source]/M Charge/C TOFA/s 1b
TOFB/s 1c
CO H2 HCOOH electrode surface from EDSd
5
1.95 — 2.2  0.3 (2.9  2.1)  10 0.053 2.8  2.6 0 6.3  2.6 0.52%
5
2.15 — 4.3  1.4 (8.1  2.5)  10 0.453 4.2  1.7 0 2.7  0.9 0.52%
Published on 17 January 2018. Downloaded by Indian Institute of Technology Kanpur on 12/11/2023 7:34:43 PM.

4
1.95 1.1 M H2O 3.4  0.4 (2.7  0.9)  10 4.26 16.3  5.1 0 4.5  1.1 —e
3
1.95 5.5 M H2O 7.9  0.7 (2.3  0.5)  10 393 67.4  11.0 0.3  0.1 1.9  0.5 —e
3
1.95 11.0 M H2O 8.3  0.9 (3.2  0.4)  10 620 80.5  4.7 1.1  0.5 0.7  0.2 0.35%
3
2.15 11.0 M H2O 16.2  2.5 (8.1  1.3)  10 3961 104.3  5.5 0.6  0.4 0 0.18%
3
1.85 5.5 M TFE 10.1  2.3 (4.2  0.5)  10 1089 87.7  8.2 1.7  0.7 0 0.01%
3
2.05 5.5 M TFE 16.8  2.5 (8.2  0.9)  10 4131 102.7  3.10 1.0  0.4 0 0.007%
a
Applied potential without IR compensation. b TOFA is calculated based on total concentration of catalyst in solution, and therefore is a
significant underestimate of catalytic activity (see ESI). c TOFB is derived from CPE data as described by Savéant et al. (see ESI).27,28 d ‘‘Co weight%’’
means the weight% of Co in the total surface atoms. e Not measured.

solid-state deposits (Fig. S15–S22 and Table S8, ESI†).29 Control
experiments consisting of 30 min CPE with the glassy carbon
electrode and clean electrolyte, only CoBr2 in solution, and only
the ligand L–L in solution all showed no activity for CO2
reduction (Table S6, ESI†).
For electrolyte solutions containing [Co(L–L)Br2] in the
absence of any added proton source, the solution became
turbid after 30 min of CPE at 1.95 V vs. Fc+/0 and 2.15 V
vs. Fc+/0. Only a small amount of CO (FECO = 2.8%@ 1.95 V,
4.2%@ 2.15 V) and formic acid HCOOH (FEHCOOH =
6.3%@ 1.95 V, 2.7%@ 2.15 V) were detected after the electro-
lyses (Table 1 and Table S7, ESI†). SEM-EDS measurements of
the electrodes post electrolysis show Co-based particles deposited
on the surface with a weight percent of 0.52% of total atoms
measured (Fig. S15, ESI†). ICP-MS measurements were used to
better quantify the amount of Co deposited and indicate the
amount of Co on the electrode surface corresponds to B8.5% of
initial Co complex loading in the electrolysis solution (Table S8,
ESI†). These measurements suggest that the catalyst decomposes
during electrolysis in the absence of added proton source.
Upon addition of a proton source, significantly more charge
was passed in 30 min CPE and the Faradaic efficiency for CO
increased while the Faradaic efficiency for HCOOH decreased.
A summary of the product distribution after 30 min CPE at
1.95 V vs. Fc+/0 for selected concentrations of H2O is shown in
Table 1 (and all concentrations in Table S5, ESI†). At the
highest concentration of H2O investigated (11.0 M), CO2 was
reduced to CO with 80.5% efficiency at 1.95 V vs. Fc+/0.
Similarly, CPE conducted in 5.50 M TFE showed selective
CO2 reduction to CO with 87.7% Faradaic efficiency at 1.85 V
vs. Fc+/0. In both cases, the TOF values for CO increase dramatically.
During the electrolysis, the solution remained transparent and
the UV-vis spectra of the solution before and after 30 min CPE
remained nearly identical (Fig. S30, ESI†). Both SEM-EDS and
ICP-MS measurements of the electrode surfaces post-electrolysis
showed less catalyst decomposition to adsorbed solid-state
materials (Table 1 and Fig. S17, S18, Table S8, ESI†). When
the CPE experiments were conducted at 0.2 V more negative
applied potentials (to manually account for IR drop due to
B60 O solution resistance in our electrolysis cell), higher
Faradaic efficiencies and TOF values were observed for CO
production and even less decomposition to surface Co species
was observed by SEM-EDS (Table 1). Compared to other Co-based
CO2RR catalysts, [Co(L–L)Br2] reduces CO2 to CO with high Faradaic
efficiency and fast TOF at less negative applied potentials
(Table S11, ESI†).
Note that when CPE experiments were conducted in 11.0 M
H2O at potentials which are more positive ( 1.75 V vs. Fc+/0)
and more negative ( 2.35 V vs. Fc+/0) than the catalytic peak
potential in Fig. 3, Faradaic efficiency for CO production
decreased and an increase in Co decomposition/deposition
was observed (Table S7, ESI†). Longer time CPE measurements
were also conducted in 11.0 M H2O at 1.95 V vs. Fc+/0, and
similar catalyst activity and Faradaic efficiencies for CO were
observed after 60 min and 90 min electrolysis when compared
to the results from the 30 min electrolyses (Table S10, ESI†).
However, significant loss of CO2 activity and increased catalyst
decomposition was observed after 120 min CPE in both 11.0 M
H2O and 5.5 M TFE (Table S10 and Fig. S28, S29, ESI†). These
results indicate that the stability of the catalyst under CO2
reduction conditions is both potential and time dependent.
The results of our CPE experiments suggest that adding a
proton source not only accelerates the catalysis but also
increases the catalyst’s stability. We propose a mechanism to
explain our results in Scheme 2. Here, [Co(L–L)]2+ is reduced
sequentially by 2 e to form [Co(L–L)], the active species for CO2
Scheme 2 Proposed catalytic mechanism of CO2 reduction with [Co(L–L)].
In the presence of an added proton source, the mechanism proceeds via
path (a) forming CO as the final C-containing product. In the absence of
added proton source, the mechanism proceeds via path (b) to catalyst
decomposition with stoichiometric HCOO production.

This journal is © The Royal Society of Chemistry 2018 Chem. Commun., 2018, 54, 1579--1582 | 1581

Communication
Scheme 3 Proposed mechanism of the catalyst decomposition.
Published on 17 January 2018. Downloaded by Indian Institute of Technology Kanpur on 12/11/2023 7:34:43 PM.
reduction. [Co(L–L)] coordinates CO2 to form a [Co(L–L)CO2]
adduct. At high proton concentration, [Co(L–L)CO2] undergoes
a 1 e reduction and 2 H+ protonation by the proton source to
generate CO and H2O and regenerate the [Co(L–L)]+ complex
(Scheme 2, path (a)). However, at low proton concentration, the
[Co(L–L)CO2] abstracts a proton from the ligand to generate
HCOO , leading to decomposition of the complex (Scheme 2,
path (b) and Scheme 3). This latter step is supported by recent
reports of catalyst inactivation and decomposition of Re, Mo and
Mn complexes with pyridylimine ligands due to the formation of
catalyst-CO2 adducts involving CO2–ligand bonding.9,20,21,30 Note
that other mechanisms might explain the observed catalyst
decomposition at low proton concentration including a purely
electrochemical degradation pathway at the electrode surface.
In conclusion, we have prepared a Co complex with a redox-
active bis(pyridylmonoimine) ligand L–L and studied its CO2
reduction activity in CH3CN. We have shown that adding a
proton source not only accelerates the electrocatalytic rate of CO2
reduction by [Co(L–L)Br2], but also improves the stability of the
catalyst by promoting intermolecular protonation of the CO2-catalyst
intermediate and preventing intramolecular H+-abstraction. The
[Co(L–L)Br2] system reduces CO2 selectively to CO in the presence
of added proton source with high TOF at less negative potentials
compared to previously reported Co-based catalysts.
This work was supported by the University of Michigan and
NSF grant CHE-0840456 for X-ray instrumentation. We thank
Dr James Windak for ICP-MS assistance, Dr Jeff Kampf for
crystallographic assistance, and Prof. Songsong Bao for assistance
with interpreting crystallographic data.
Conflicts of interest
There are no conflicts to declare.
Notes and references
1 A. M. Appel, J. E. Bercaw, A. B. Bocarsly, H. Dobbek, D. L. DuBois,
M. Dupuis, J. G. Ferry, E. Fujita, R. Hille, P. J. A. Kenis, C. A. Kerfeld,
1582 | Chem. Commun., 2018, 54, 1579--1582
View Article Online
ChemComm
R. H. Morris, C. H. F. Peden, A. R. Portis, S. W. Ragsdale,
T. B. Rauchfuss, J. N. H. Reek, L. C. Seefeldt, R. K. Thauer and
G. L. Waldrop, Chem. Rev., 2013, 113, 6621–6658.
2 S. Berardi, S. Drouet, L. Francas, C. Gimbert-Surinach, M. Guttentag,
C. Richmond, T. Stoll and A. Llobet, Chem. Soc. Rev., 2014, 43, 7501–7519.
3 J. Qiao, Y. Liu, F. Hong and J. Zhang, Chem. Soc. Rev., 2014, 43,
631–675.
4 Y. Hori, H. Wakebe, T. Tsukamoto and O. Koga, Electrochim. Acta,
1994, 39, 1833–1839.
5 K. P. Kuhl, E. R. Cave, D. N. Abram and T. F. Jaramillo, Energy
Environ. Sci., 2012, 5, 7050–7059.
6 E. E. Benson, C. P. Kubiak, A. J. Sathrum and J. M. Smieja, Chem.
Soc. Rev., 2009, 38, 89–99.
7 H. Takeda, C. Cometto, O. Ishitani and M. Robert, ACS Catal., 2017,
7, 70–88.
8 M. L. Clark, K. A. Grice, C. E. Moore, A. L. Rheingold and C. P.
Kubiak, Chem. Sci., 2014, 5, 1894–1900.
9 C. W. Machan, S. A. Chabolla and C. P. Kubiak, Organometallics,
2015, 34, 4678–4683.
10 E. E. Benson, M. D. Sampson, K. A. Grice, J. M. Smieja, J. D.
Froehlich, D. Friebel, J. A. Keith, E. A. Carter, A. Nilsson and
C. P. Kubiak, Angew. Chem., Int. Ed., 2013, 52, 4841–4844.
11 Z. Chen, C. Chen, D. R. Weinberg, P. Kang, J. J. Concepcion,
D. P. Harrison, M. S. Brookhart and T. J. Meyer, Chem. Commun.,
2011, 47, 12607–12609.
12 T. C. Simpson and R. R. Durand, Electrochim. Acta, 1988, 33, 581–583.
13 E. Fujita and J. T. Muckerman, Inorg. Chem., 2004, 43, 7636–7647.
14 D. C. Lacy, C. C. L. McCrory and J. C. Peters, Inorg. Chem., 2014, 53,
4980–4988.
15 J. M. Smieja and C. P. Kubiak, Inorg. Chem., 2010, 49, 9283–9289.
16 J. A. Keith, K. A. Grice, C. P. Kubiak and E. A. Carter, J. Am. Chem.
Soc., 2013, 135, 15823–15829.
17 C. M. Lieber and N. S. Lewis, J. Am. Chem. Soc., 1984, 106, 5033–5034.
18 A. H. A. Tinnemans, T. P. M. Koster, D. H. M. W. Thewissen and
A. Mackor, Recl. Trav. Chim. Pays-Bas, 1984, 103, 288–295.
19 L. Chen, Z. Guo, X.-G. Wei, C. Gallenkamp, J. Bonin, E. Anxolabéhère-
Mallart, K.-C. Lau, T.-C. Lau and M. Robert, J. Am. Chem. Soc., 2015, 137,
10918–10921.
20 D. Sieh, D. C. Lacy, J. C. Peters and C. P. Kubiak, Chem. – Eur. J.,
2015, 21, 8497–8503.
21 D. Sieh and C. P. Kubiak, Chem. – Eur. J., 2016, 22, 10638–10650.
22 K. T. Ngo, M. McKinnon, B. Mahanti, R. Narayanan, D. C. Grills,
M. Z. Ertem and J. Rochford, J. Am. Chem. Soc., 2017, 139, 2604–2618.
23 N. P. Liyanage, H. A. Dulaney, A. J. Huckaba, J. W. Jurss and
J. H. Delcamp, Inorg. Chem., 2016, 55, 6085–6094.
24 M. D. Sampson, A. D. Nguyen, K. A. Grice, C. E. Moore, A. L. Rheingold
and C. P. Kubiak, J. Am. Chem. Soc., 2014, 136, 5460–5471.
25 B. t. Limoges and J.-M. Savéant, J. Electroanal. Chem., 2004, 562,
43–52.
26 C. C. L. McCrory, N. K. Szymczak and J. C. Peters, Electrocatalysis,
2016, 7, 87–96.
27 C. Costentin, M. Robert and J.-M. Saveant, Chem. Soc. Rev., 2013, 42,
2423–2436.
28 C. Costentin, S. Drouet, M. Robert and J.-M. Savéant, Science, 2012,
338, 90–94.
29 N. Kaeffer, A. Morozan, J. Fize, E. Martinez, L. Guetaz and V. Artero,
ACS Catal., 2016, 6, 3727–3737.
30 R. Stichauer, A. Helmers, J. Bremer, M. Rohdenburg, A. Wark,
E. Lork and M. Vogt, Organometallics, 2017, 36, 839–848.
This journal is © The Royal Society of Chemistry 2018