Photocatalytic CO2 Reduction under Visible-Light Irradiation by Ruthenium CNC Pincer Complexes

Article abstract of DOI:10.1002/chem.201905840

Photocatalytic CO₂ reduction using a ruthenium photosensitizer, a sacrificial reagent 1,3-dimethyl-2-(o-hydroxyphenyl)-2,3-dihydro-1H-benzo[d]imidazole (BI(OH)H), and a ruthenium catalyst were carried out. The catalysts contain a pincer ligand, 2,6-bis(alkylimidazol-2-ylidene)pyridine (CNC) and a bipyridine (bpy). The photocatalytic reaction system resulted in HCOOH as a main product (selectivity 70–80 %), with a small amount of CO, and H₂. Comparative experiments (a coordinated ligand (NCMe vs. CO) and substituents (tBu vs. Me) of the CNC ligand in the catalyst) were performed. The turnover number (TON_HCOOH) of carbonyl-ligated catalysts are higher than those of acetonitrile-ligated catalysts, and the carbonyl catalyst with the smaller substituents (Me) reached TON_HCOOH=5634 (24 h), which is the best performance among the experiments.

Full text of DOI:10.1002/chem.201905840

A Journal of
Accepted Article
Title: Photocatalytic CO2 reduction under visible-light irradiation by
ruthenium CNC pincer complexes
Authors: Yasuhiro Arikawa, Itoe Tabata, Yukari Miura, Hiroki Tajiri,
Yudai Seto, Shinnosuke Horiuchi, Eri Sakuda, and Keisuke
Umakoshi
This manuscript has been accepted after peer review and appears as an
Accepted Article online prior to editing, proofing, and formal publication
of the final Version of Record (VoR). This work is currently citable by
using the Digital Object Identifier (DOI) given below. The VoR will be
published online in Early View as soon as possible and may be different
to this Accepted Article as a result of editing. Readers should obtain
the VoR from the journal website shown below when it is published
to ensure accuracy of information. The authors are responsible for the
content of this Accepted Article.
To be cited as: Chem. Eur. J. 10.1002/chem.201905840
Link to VoR: http://dx.doi.org/10.1002/chem.201905840
Supported by

10.1002/chem.201905840
Chemistry - A European Journal
COMMUNICATION
Photocatalytic CO2 reduction under visible-light irradiation by
ruthenium CNC pincer complexes
Yasuhiro Arikawa,*[a] Itoe Tabata,[a] Yukari Miura,[a] Hiroki Tajiri,[a] Yudai Seto,[a] Shinnosuke Horiuchi,[a]
Eri Sakuda,[a] and Keisuke Umakoshi[a]
[a]
Prof. Dr. Y. Arikawa, I. Tabata, Y. Miura, H. Tajiri, Y. Seto, Dr. S. Horiuchi, Prof. Dr. E. Sakuda, Prof. Dr. K. Umakoshi
Division of Chemistry and Materials Science
Graduate School of Engineering, Nagasaki University
Bunkyo-machi 1-14, Nagasaki 852-8521 (Japan)
E-mail: arikawa@nagasaki-u.ac.jp
Supporting information for this article is given via a link at the end of the document.
Abstract: Photocatalytic CO2 reduction using
a
ruthenium
photosensitizer, a sacrificial reagent (BI(OH)H) (1,3-dimethyl-2-(o-
hydroxyphenyl)-2,3-dihydro-1H-benzo[d]imidazole), and a ruthenium
catalyst were carried out. The catalysts contain a pincer ligand, 2,6-
bis(alkylimidazol-2-ylidene)pyridine (CNC) and a bipyridine (bpy).
Our photocatalytic reaction system resulted in HCOOH as a main
product (selectivity 70 ~ 80%), a small amount of CO, and H2.
Comparative experiments (a coordinated ligand (NCMe vs. CO) and
substituents (tBu vs. Me) of the CNC ligand in the catalyst) were
performed. The TONHCOOH of carbonyl-ligated catalysts are higher
than those of acetonitrile-ligated catalysts, and the carbonyl catalyst
with the smaller substituents (Me) reached TONHCOOH = 5634 (24h),
which is the best performance among the experiments.
(PF6)2
(PF6)2
L
N
Ru
N
N
N
N
N
N
N
N
R
H
N
N
N
Ru
N
N
N
R
OH
L = NCMe, R = tBu (2a)
L = NCMe, R = Me (2b)
L = CO, R = tBu (3a)
L = CO, R = Me (3b)
[Ru(dmbpy)3](PF6)2
BI(OH)H
Figure 1. Structures of catalysts, photosensitizer, and sacrificial reagent in this
study.
To realize a sustainable society, we must address the problem
of energy/resource depletion. One possible solution is the use
of carbon dioxide (CO2), which is abundant and ubiquitous in the
atmosphere. However, to do so CO2 has to be reduced to useful
substances, because it is the final carbon oxidation product.
Ideally, sunlight (especially visible light) could be used as the
energy for the reduction.
Against this background, CO2
reduction by visible light has been actively studied in recent
years.[1]
Both homogeneous and heterogeneous[2] systems for CO2
reduction have been explored. Although the latter is more
practical, at present the former system appears to be more
effective. The homogeneous system generally consists of a
photosensitizer, a sacrificial reagent, and a catalyst which can
be easily adjusted. Recently, the non-noble metal based
Figure 2.
X-ray crystal structures of the cation part of
[(tBuCNC)Ru(bpy)(CO)](PF6)2 (3a) (left) and [(tBuCNC)Ru(bpy)(HOMe)](PF6)2,
(right). One of the two independent molecules of 3a in the unit cell is shown.
Thermal ellipsoids are set at 50% probability level. The counter PF6 ions,
crystallization solvents, and hydrogen atoms except for methanol protons in
the methanol complex are omitted for clarity.
molecular catalyst have been extensively investigated[1c, 3]
.
It
has been proposed that the photosensitizer excited by visible
light is reductively quenched by the sacrificial reagent to be a
powerful reductant, and one electron is transferred from the
reductant to the catalyst at a time.[1] Finally, CO2 is reduced on
the catalyst. In this study, we investigated the CO2 reducing
ability of ruthenium complexes having a 2,6-bis(alkylimidazol-2-
ylidene)pyridine (CNC) ligand as the catalyst under visible light
(l ≥ 500 nm) using a photosensitizer [Ru(dmbpy)3](PF6)2 (dmbpy
= 4,4’-dimethylbipyridine) and a sacrificial reagent BI(OH)H (1,3-
dimethyl-2-(o-hydroxyphenyl)-2,3-dihydro-1H-benzo[d]imidazole)
which is very effective against this photosensitizer.[4] The
performance of the catalyst was evaluated by the differences of
substituents on the CNC ligand and ligands on the ruthenium
atom. In 2015, we have reported that the CNC ruthenium
complexes can readily incorporate CO2 from air.[5a] This is the
reason why we choose the CNC complexes as the catalyst.
More recently, E. T. Papish et al. have reported several CO2
reduction reactions using similar CNC complexes.[6] In this
report, their comparisons are included.
We prepared four types of the CNC complexes; tert-butyl and
methyl substituents on the CNC ligand and CH3CN and CO
ligands on ruthenium (Figure 1). Although
1
This article is protected by copyright. All rights reserved.

10.1002/chem.201905840
Chemistry - A European Journal
COMMUNICATION
significant blue shift in the lowest-energy absorption bands,
because complexes 3 have a carbonyl ligand which lower the
dp(Ru) energy level due to p-back donation. It is clear that the
[Ru-NCMe] (2a and 2b) absorb the irradiated light (≥ 500 nm),
but not the [Ru-CO] (3a and 3b). Cyclic voltammetry (CV) was
used to evaluate the efficacy of complexes 2 and 3 to reduce
CO2 (Table 1 and Figure S1). Their first reduction waves, which
are quasi-reversible, would be attributed to the bpy-based redox
(bpy•–/bpy). Although the effect of substituent groups (tBu vs.
Me) of the CNC ligand on the redox potentials was small, the
axial ligands (NCMe vs. CO) affected largely to their redox
potentials. The reduction potentials of [Ru-CO] (3a and 3b) are
more positive than those of [Ru-NCMe] (2a and 2b), which is
reasonable because of the strong electron-withdrawing
character of CO ligand. In any case, the electron transfer from
the photosensitizer [Ru(dmbpy)3]2+ (E1/2(PS–/PS) = –1.74 V vs.
Ag/AgNO3)[8] to their catalysts would be exergonic processes.
Photocatalytic CO2 reduction was performed in a glass tube
containing DMA/TEOA (dimethylacetamide/triethanolamine) (5.0
mL, 4:1 v/v), catalyst 2 or 3 (10 µM), [Ru(dmbpy)3](PF6)2 (50 µM),
1.6
1.4
1.2
1.0
0.8
0.6
0.4
0.2
0.0
2a
2b
3a
3b
300
350
400
450
500
550
600
650
Wavelength / nm
Figure 3. UV-vis absorption spectra of 2a, 2b, 3a, and 3b.
Table 1. The first reduction potentials of catalysts[a]
Complex
E1/2red (V)[b]
-1.39
Epc (V)
-1.42
-1.51
-1.24
-1.28
Epa (V)
-1.35
-1.43
-1.17
-1.19
2a
2b
3a
3b
and BI(OH)H (0.10 M).
The CO2-saturated solution was
-1.47
irradiated at l ≥ 500 nm using a 450 W high-pressure Hg lamp
combined with a K2CrO4 solution filter. Formate was analyzed
by capillary electrophoresis, while the gaseous products (CO
and H2) were analyzed by gas chromatography (GC-TCD).
Table 2 summarizes the results of the photocatalytic reactions
and total turnover numbers of the reduced products are shown
in Figure 4. The time courses of obserbed products (TON) using
3b are shown in Figure 5. Use of complexes 2 and 3 led to
formation of formic acid as a main product with H2 and a small
-1.21
-1.24
[a] Measured in MeCN (1.0 mM). All values of the first reduction peaks are
reported versus Ag/AgCl. Values are measured by CV in 0.1 M nBu4NPF6
using Pt counter and glassy carbon working electrode. [b] Quasi-reversible.
[(RCNC)Ru(bpy)Cl]PF6 (1) and [(RCNC)Ru(bpy)(NCMe)](PF6)2
(2; [Ru-NCMe]) were prepared referring to literature methods,[5]
[(RCNC)Ru(bpy)(CO)](PF6)2 (3; [Ru-CO]) were synthesized by
the substitution reactions of 1 in the presence of AgPF6 through
MeOH-ligated intermediates, which are confirmed as
[(tBuCNC)Ru(bpy)(HOMe)](PF6)2 by X-ray crystallographic
analysis (Figure 2).
The UV-visible (UV-vis) absorption spectra of 2 and 3 are
shown in Figure 3. According to a literature,[7] the lowest-energy
transitions would contain dp(Ru) → p*(bpy) MLCT without dp(Ru)
→ p*(RCNC) MLCT character. Compared to 2, 3 display a
amount of CO (entry 1-4). Best performance is TONHCOOH
=
5634 (24h) for 3b (entry 4). Even in the absence of the catalyst,
a small amount of HCOOH and CO were detected because
decomposition of the photosensitizer [Ru(dmbpy)3]2+ would
generate other active catalysts (entry 5).[1e] However, presence
of the catalyst (2 or 3) increased the total amount of the reduced
products (HCOOH and CO) more than about 10 folds. Control
experiments in the absence of the photosensitizer (entry 6), the
Table 2. The results of photocatalytic CO2 reduction for 24 h[a], [b]
Product (
Sacrificial
µ
mol)
CO
TON (selectivity %)
Entry
Catalyst
Sensitizer
HCOOH
H2
HCOOH
CO
H2
reagent
BI(OH)H
BI(OH)H
1
2a
2b
3a
3b
–
[Ru(dmbpy)3]2+
[Ru(dmbpy)3]2+
[Ru(dmbpy)3]2+
[Ru(dmbpy)3]2+
[Ru(dmbpy)3]2+
–
165
230
190
282
21
0
6.5
11
6.5
15
3
28
72
25
95
8.9
4.4
1
3296 (83)
129 (3)
224 (4)
129 (3)
300 (4)
-
556 (14)
1438 (23)
505 (11)
1897 (24)
-
2
4593 (73)
3
BI(OH)H
BI(OH)H
BI(OH)H
BI(OH)H
–
3792 (86)
4
5634 (72)
5
-
6
3b
3b
3b
3b
0.43
~ 0
0
0
0
0
0
8.6 (9)
~ 0
87 (91)
20 (100)
307 (100)
0
7
[Ru(dmbpy)3]2+
[Ru(dmbpy)3]2+
[Ru(dmbpy)3]2+
0
8[c]
9[d]
BI(OH)H
BI(OH)H
0
15
0
0
0
0
0
[a] In a typical run, a CO2-saturated DMA/TEOA (4/1) solution containing the catalyst (10 µM), sensitizer (50 µM), and sacrificial reagent (0.1 M) was irradiated for
24h using a 450 W high-pressure Hg lamp combined with a K2CrO4 solution filter under CO2 atmosphere. [b] Products (µmol) values were averaged over three
reactions. [c] Under Ar atmosphere. [d] Under dark. TON = mol number of the product/mol number of the catalyst.
2
This article is protected by copyright. All rights reserved.

10.1002/chem.201905840
Chemistry - A European Journal
COMMUNICATION
The
similar
CNC
ruthenium
complex,
[(CNCOMe)RuCl(NCMe)2]OTf, where
a
4-methoxy group is
introduced to the pyridyl moiety in CNC ligand, has been
reported.[6a,b] The use of this complex for the photocatalytic CO2
reduction in CH3CN with photosensitizer [Ir(ppy)3] resulted in
selective production of CO (TONCO = 227 (selectivity 97%)), but
in DMF with photosensitizer [Ru(bpy)3]2+ production of formate
was dominant (TONHCOOH = 143 (selectivity 90%)). More
recently, alternation of imidazole to benzimidazole moiety and
coordination of a bpy ligand led to formation of CO (TONCO = 55)
without photosensitizer.[6c] Interestingly, the low loading of the
catalyst (1.0 nM) afforded TONCO = 33,000. In contrast, our
system shows high performance in the photocatalytic CO2
reduction, leading to production of formate.
9000
HCOOH
8000
CO
7000
H2
6000
5000
4000
3000
2000
1000
0
2a
2b
3a
3b
Comparison of TON (24 h) of HCOOH based on differences
in the axial ligands ([Ru-NCMe] vs. [Ru-CO]) showed that
TONHCOOH (24 h) of [Ru-CO] (3a and 3b) are higher than those
of [Ru-NCMe] (2a and 2b), respectively. This may be due to the
presence of the absorption band (l ≥ 500 nm) in [Ru-NCMe],
which reduce the performance of the catalyst. Alternatively,
differences in the substituents (R = tBu vs. Me) resulted in higher
TON for the catalyst having Me substituents (2b and 3b), but
with lower selectivity of HCOOH because of increment of H2
Figure 4. Total turnover number (TON) graphs of the reduced products
(HCOOH, CO, and H2) after 24 h.
7000
HCOOH
t
production. The steric hindrance of the Bu substituents should
6000
CO
account for the results.[9] The bulkier substituents (tBu) should
prevent access of CO2 to the metal center, whereas the smaller
substituents (Me) should allow access of protons as well as CO2
to the metal center, leading to lower selectivity of HCOOH.
In summary, we achieved TONHCOOH = 5634 (selectivity 72%)
(24h) by photocatalytic CO2 reduction using visible light. Our
ruthenium catalysts contain a CNC and a bpy ligands. The
carbonyl complex with smaller substituents (Me groups) (3b)
resulted in the best performance within the four complexes.
5000
4000
3000
2000
1000
0
H2
0
4
8
12
16
20
24
time / h
Acknowledgements
Figure 5. Turnover numbers (TON) for formation of HCOOH, CO, and H2 as a
function of irradiation time during photocatalytic reaction of 3b. Each data
point is the average of three runs.
This work was supported by the JSPS, KAKENHI grant number
JP17K05813, and by the priority research project of Nagasaki
University. We are grateful to M. Sadamitsu at Nagasaki
University for his technical assistance and Prof. Y. Sunada at
The University of Tokyo for elemental analyses.
sacrificial reagent (entry 7), or CO2 (entry 8) led to no formation
of HCOOH. Catalyst 3b can function both as a photosensitizer
and a catalyst, but the performance was very low (TONCO = 8.6)
(entry 6). Even under Ar, H2 was detected (TONH2 = 307) in our
photocatalytic reaction conditions (entry 8).
Keywords: photocatalytic CO2 reduction • pincer ligand • ligand
effect • carbonyl complexes • ruthenium
Isotopic labeling experiment under 13CO2 atmosphere was
performed to determine the source of the carbon atoms in the
produced formic acid. Before irradiation, in the 13C{1H} NMR
spectrum of a DMF-d7/TEOA (4:1 v/v) solution consisting of
catalyst 3b (10 µM), [Ru(dmbpy)3](PF6)2 (50 µM), and BI(OH)H
(0.10 M), a signal assignable to 13CO2 is observed at 125.2 ppm
(Figure S3). Irradiation resulted in both decrease of the 13CO2
signal and appearance of a signal at 167.2 ppm, which is
assignable to an equilibrium mixture of H13COOH and H13COO–
[1]
a) Y. Kuramochi, O. Ishitani, H. Ishida, Coord. Chem. Rev. 2018, 373,
333-356; b) Y. Tamaki, O. Ishitani, ACS Catalysis 2017, 7, 3394-3409;
c) H. Rao, L. C. Schmidt, J. Bonin, M. Robert, Nature 2017, 548, 74-77;
d) N. Elgrishi, M. B. Chambers, X. Wang, M. Fontecave, Chem. Soc.
Rev. 2017, 46, 761-796; e) Y. Yamazaki, H. Takeda, O. Ishitani, J.
Photochem. Photobiol. C: Photochem. Rev. 2015, 25, 106-137; f) K.
Kobayashi, K. Tanaka, Phys. Chem. Chem.l Phys. 2014, 16, 2240-
2250; g) C. D. Windle, R. N. Perutz, Coord. Chem. Rev. 2012, 256,
2562-2570; h) M. Schulz, M. Karnahl, M. Schwalbe, J. G. Vos, Coord.
Chem. Rev. 2012, 256, 1682-1705; g) H. Takeda, O. Ishitani, Coord.
Chem. Rev. 2010, 254, 346-354; h) A. J. Morris, G. J. Meyer, E. Fujita,
Acc. Che. Res. 2009, 42, 1983-1994.
1
[4]. The H NMR spectra confirmed the formation of the labeled
1
formic acid (8.56 ppm, a doublet signal, JCH = 185 Hz) along
with trace amount of signal for the non-labeled formic acid
(Figure S4). These results clearly show that the formic acid
originated from the reduction of CO2.
[2]
a) A. Nakada, R. Kuriki, K. Sekizawa, S. Nishioka, J. J. M. Vequizo, T.
Uchiyama, N. Kawakami, D. Lu, A. Yamakata, Y. Uchimoto, O. Ishitani,
K. Maeda, ACS Catalysis 2018, 8, 9744-9754; b) J. L. White, M. F.
Baruch, J. E. Pander, Y. Hu, I. C. Fortmeyer, J. E. Park, T. Zhang, K.
3
This article is protected by copyright. All rights reserved.

10.1002/chem.201905840
Chemistry - A European Journal
COMMUNICATION
Liao, J. Gu, Y. Yan, T. W. Shaw, E. Abelev, A. B. Bocarsly, Chem. Rev.
2015, 115, 12888-12935; c) R. Asahi, T. Morikawa, H. Irie, T. Ohwaki,
Chem. Rev. 2014, 114, 9824-9852.
[3]
a) H. Takeda, C. Cometto, O. Ishitani, M. Robert, ACS Catal., 2017, 7,
70-88; b) A. Rosas-Hernández, C. Steinlechner, H. Junge, M. Beller,
Green Chem., 2017, 19, 2356-2360; c) D. Hong, Y. Tsukakoshi, H.
Kotani, T. Ishizuka, T. Kojima, J. Am. Chem. Soc., 2017, 139, 6538-
6541; d) T. Ouyang, H.-J. Wang, H.-H. Huang, J.-W. Wang, S. Guo,
W.-J. Liu, D.-C. Zhong, T.-B. Lu, Angew. Chem. Int. Ed., 2018, 57,
16480-16485; e) Z. Guo, G. Chen, C. Cometto, B. Ma, H. Zhao, T.
Groizard, L. Chen, H. Fan, W.-L. Man, S.-M. Yiu, K.-C. Lau, T.-C. Lau,
M. Robert, Nature Catal., 2019, 2, 830-830; f) H. Shirley, X. Su, H.
Sanjanwala, K. Talukdar, J. W. Jurss, J. H. Delcamp, J. Am. Chem.
Soc., 2019, 141, 6617-6622.
[4]
Y. Tamaki, K. Koike, O. Ishitani, Chem. Sci. 2015, 6, 7213-7221;
BI(OH)H is more effective than BIH (1,3-dimethyl-2-phenyl-2,3-dihydro-
1H-benzo[d]imidazole) for the formation of formic acid, because
BI(OH)H donates two electrons and two protons in a step-by-step
process.
The reductive quenching rate constant of the excited
photosensitizer [Ru(dmbpy)3]2+ by BI(OH)H is 1.17 × 109 M-1s-1.
a) Y. Arikawa, T. Nakamura, S. Ogushi, K. Eguchi, K. Umakoshi, Dalton
Trans. 2015, 44, 5303-5305; b) T. Nakamura, S. Ogushi, Y. Arikawa, K.
Umakoshi, J. Organomet. Chem. 2016, 803, 67-72.
[5]
[6]
a) C. M. Boudreaux, N. P. Liyanage, H. Shirley, S. Siek, D. L. Gerlach,
F. Qu, J. H. Delcamp, E. T. Papish, Chem. Commun. 2017, 53, 11217-
11220; b) R. R. Rodrigues, C. M. Boudreaux, E. T. Papish, J. H.
Delcamp, ACS Appl. Energy Mater. 2019, 2, 37-46; c) S. Das, R. R.
Rodrigues, R. W. Lamb, F. Qu, E. Reinheimer, C. M. Boudreaux, C. E.
Webster, J. H. Delcamp, E. T. Papish, Inorg. Chem. 2019, 58, 8012-
8020.
[7]
[8]
[9]
L.-H. Chung, K.-S. Cho, J. England, S.-C. Chan, K. Wieghardt, C.-Y.
Wong, Inorg. Chem. 2013, 52, 9885-9896.
Y. Tamaki, K. Watanabe, K. Koike, H. Inoue, T. Morimoto, O. Ishitani,
Faraday Discuss. 2012, 155, 115-127.
see ref. 6c; Similar steric effect in the photocatalytic CO2 reduction has
been observed using the analogous CNC complex with Me or Ph
substituents. The performance of the complex with smaller substituents
(Me groups) is better than that of the other complex (Ph groups).
4
This article is protected by copyright. All rights reserved.

10.1002/chem.201905840
Chemistry - A European Journal
COMMUNICATION
Entry for the Table of Contents
(PF6
)
2
N
Ru
N
(PF6
)
2
N
N
L
N
N
N
N
N
R
N
N
N
N
H
Ru
N
N
R
OH
CO2
HCOOH
Photocatalytic CO2 reduction using a ruthenium photosensitizer, a sacrificial reagent, and a ruthenium catalyst were carried out. The
catalysts contain a pincer ligand, 2,6-bis(alkylimidazol-2-ylidene)pyridine (CNC) and a bipyridine (bpy). Our photocatalytic reaction
system resulted in HCOOH as a main product (selectivity 70 ~ 80%). The carbonyl catalyst with the smaller substituents (Me)
reached TONHCOOH = 5634 (24h).
5
This article is protected by copyright. All rights reserved.