Highly Efficient and Robust Photocatalytic Systems for CO2 Reduction Consisting of a Cu(I) Photosensitizer and Mn(I) Catalysts

Article abstract of DOI:10.1021/jacs.8b10619

The development of highly efficient, selective, and durable photocatalytic CO₂ reduction systems that only use earth-abundant elements is key for both solving global warming and tackling the shortage of energy and carbon resources. Here, we successfully developed CO₂ reduction photocatalysts using [Cu₂(P₂bph)₂]²⁺ (CuPS) (P₂bph = 4,7-diphenyl-2,9-di(diphenylphosphinotetramethylene)-1,10-phenanthroline) as a redox photosensitizer and fac-Mn(X₂bpy)(CO)₃Br (Mn(4X)) (X₂bpy = 4,4′-X₂-2,2′-bipyridine (X = -H and -OMe) or Mn(6mes) (6mes = 6,6′-(mesityl)₂-2,2′-bipyridne)) as the catalyst. The most efficient photocatalysis was achieved by Mn(4OMe): The total quantum yield of CO₂ reduction products was 57%, the turnover number based on the Mn catalyst was over 1300, and the selectivity of CO₂ reduction was 95%. Electronic and steric effects of the substituents (X) in the Mn complexes largely affected both the photocatalytic efficiency and the product selectivity. For example, the highest selectivity of CO formation was achieved by using Mn(6mes) (selectivity S_CO = 96.6%), whereas the photocatalytic system using Mn(4H) yielded HCOOH as the main product (S_HCOOH = 74.6%) with CO and H₂ as minor products (S_CO = 23.7%, S_H2 = 1.7%). In these photocatalytic reactions, CuPS played its role as an efficient and very durable redox photosensitizer, while remaining stable in the reaction solution even after a turnover number of 200 had been reached (the catalyst used had a turnover number of over 1000).

Full text of DOI:10.1021/jacs.8b10619

Article
pubs.acs.org/JACS
Cite This: J. Am. Chem. Soc. XXXX, XXX, XXX−XXX
Highly Efficient and Robust Photocatalytic Systems for CO2
Reduction Consisting of a Cu(I) Photosensitizer and Mn(I) Catalysts
Hiroyuki Takeda,† Hiroko Kamiyama,† Kouhei Okamoto,† Mina Irimajiri,† Toshihide Mizutani,†
Kazuhide Koike,‡ Akiko Sekine,† and Osamu Ishitani*,†
†Department of Chemistry, School of Science, Tokyo Institute of Technology, 2-12-1-NE-1 O-okayama, Meguro-ku, Tokyo
152-8550, Japan
‡National Institute of Advanced Industrial Science and Technology, Onogawa 16-1, Tsukuba 305-8569, Japan
S
* Supporting Information
ABSTRACT: The development of highly efficient, selective, and
durable photocatalytic CO2 reduction systems that only use earth-
abundant elements is key for both solving global warming and tackling
the shortage of energy and carbon resources. Here, we successfully
developed CO2 reduction photocatalysts using [Cu2(P2bph)2]2+ (CuPS)
(P2bph = 4,7-diphenyl-2,9-di(diphenylphosphinotetramethylene)-1,10-
phenanthroline) as a redox photosensitizer and fac-Mn(X2bpy)(CO)3Br
(Mn(4X)) (X2bpy = 4,4′-X2-2,2′-bipyridine (X = −H and −OMe) or
Mn(6mes) (6mes = 6,6′-(mesityl)2-2,2′-bipyridne)) as the catalyst. The
most efficient photocatalysis was achieved by Mn(4OMe): The total
quantum yield of CO2 reduction products was 57%, the turnover
number based on the Mn catalyst was over 1300, and the selectivity of CO2 reduction was 95%. Electronic and steric effects of
the substituents (X) in the Mn complexes largely affected both the photocatalytic efficiency and the product selectivity. For
example, the highest selectivity of CO formation was achieved by using Mn(6mes) (selectivity SCO = 96.6%), whereas the
photocatalytic system using Mn(4H) yielded HCOOH as the main product (SHCOOH = 74.6%) with CO and H2 as minor
products (SCO = 23.7%, SH = 1.7%). In these photocatalytic reactions, CuPS played its role as an efficient and very durable
2
redox photosensitizer, while remaining stable in the reaction solution even after a turnover number of 200 had been reached
(the catalyst used had a turnover number of over 1000).
photoanode (CoOx-adsorbed TaON) for water oxidation. This
photoelectrochemical cell photocatalytically reduces CO2 by
using water as the reductant and only visible light as the energy
source:2
INTRODUCTION
■
Recent concerns about global warming and the shortage of
energy and carbon resources are the primary reasons behind
the development of artificial photosynthetic systems that can
reduce CO2 to useful chemicals by using solar light as the
energy source. From this viewpoint, photocatalytic CO2
reduction systems that consist of metal complexes have
attracted attention as a key component of artificial photosyn-
thesis because of their high efficiency and the selectivity of the
products. Since relatively low-energy multielectron reduction
of CO2 is necessary, as shown in eqs 1−3, such photocatalytic
systems consist of two components: a redox photosensitizer,
which initiates the photochemical one-electron transfer from a
reductant to a catalyst, and the catalyst, which accepts
electrons from the redox photosensitizer, activates CO2, and
introduces the electrons into CO2. Recently, metal-complex
photocatalytic systems for CO2 reduction have been
successfully applied to semiconductor hybrid systems, which
can reduce CO2 using water as the reductant.1,2 A binuclear
complex consisting of a Ru(II) redox photosensitizer and a
Re(I) catalyst was adsorbed onto a p-type semiconductor
electrode, which works as a molecular photocathode for CO2
reduction, and was combined with an n-type semiconductor
CO2 + e− → CO•2−
(1)
CO2 + 2e− + 2H+ → HCO2H
(2)
CO2 + 2e− + 2H+ → CO + H2O
(3)
Although various combinations of metal-complex redox
photosensitizers and catalysts have been reported, many of
them used noble- and/or rare-metal complexes. Highly
efficient and durable photocatalysts have been constructed by
using Re(I) metal complexes,3 Ru(II) complexes, or both
metals,4 and quantum yields Φ > 50% and turnover number
(TON) > 1000 were reported. Fossil fuel-related CO2
emissions reached 32 2.7 Gt/year in 2010 and have been
increasing by a few percent every year.5 It is not feasible to
treat this large volume of CO2 by using rare-earth metals such
Received: October 2, 2018
© XXXX American Chemical Society
A
DOI: 10.1021/jacs.8b10619
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
as Re (only 0.7 ppb in the earth crust)6 or Ru (1 ppb in the
reported as an electrochemical catalyst for CO2 reduction by
Deronzier and co-workers,15,16 electrocatalysis by these Mn
complexes has been widely investigated by several groups.17−24
These studies clarified that the catalytic behavior of the Mn
complexes and their mechanisms is very different from the
Re(I) complexes, even though the structure and the ligands are
the same. In most of the systems that use Mn complexes, the
major product was CO. We first applied this Mn complex to a
photocatalytic system for CO2 reduction by combining it with
[RuII(N^N)3]2+ as a redox photosensitizer. This system
produced HCOOH as the major product with a quantum
yield of 5.3%, and both TON(PS)HCOOH and TON-
(CAT)HCOOH were 150.25 Although some other groups also
reported photocatalytic systems including Mn-complex cata-
lysts, Ru and Zn porphyrin complexes were used as the redox
photosensitizers, and the photocatalytic behavior was low
(ΦHCOOH < 4%, TON(PS) < 13, TON(CAT) < 130).26−28
Although, as described above, CO2 reduction photocatalytic
systems without noble elements have recently become the
center of attention, their photocatalytic properties are still
much lower than the reported photocatalytic systems that
consist of noble and/or rare metal complexes.
earth crust)6 as photosensitizers and/or catalysts.
Reports on molecular photocatalytic systems for CO2
reduction consisting of only earth-abundant elements have
recently been increasing. Co(II) and Ni(II) macrocyclic
complexes (Co: 25 ppm, Ni 84 ppm in the earth crust)6
have been investigated as catalysts for CO2 reduction in both
photocatalytic and electrocatalytic systems for a long time. Of
these photocatalytic systems, Yanagida and Fujita et al.7
reported the most efficient photocatalytic systems using Co(II)
complexes as the photosensitizer, however p-terphenyl only
uses UV light (ΦCO+HCOOH = 25% under irradiation at λex
=
313 nm). The durability of this system was also low;
TON(CAT)CO+HCOOH < 20 (TON based on the catalyst
used). Other photocatalytic systems that use Co and Ni
complexes are combined with rare-metal photosensitizers,
mostly [Ru(N^N)3]2+. In addition, in most of the reported
cases, high efficiency, durability, and selectivity of the
photocatalytic reduction of CO2 were not achieved.8
We previously reported a photocatalytic system with a
dinuclear Cu(I) complex as a redox photosensitizer (CuPS in
Scheme 1) and FeII(2,9-Me2phen)2(NCS)2 (phen = 1,10-
We herein report the most efficient and durable CO2
reduction photocatalytic systems that only contain earth-
abundant metal complexes, that is, a combination of CuPS as
the redox photosensitizer and fac-MnI(4,4′-X2-bpy)(CO)3Br or
fac-MnI{6,6′-(mesityl)2-bpy}(CO)3Br, Mn(4X) or Mn(6mes),
respectively (Scheme 1); ΦCO+HCOOH = 57%, TON-
(PS)CO+HCOOH = 263, TON(CAT)CO+HCOOH = 1314 in the
case of Mn(4OMe). By changing the substituents, the
distribution of the reduction products of CO2 could also be
controlled; CO:HCOOH = 93:6 for Mn(6mes) and 6:94 for
Mn(4H).
Scheme 1. Structure of the CuI Complex Redox
Photosensitiser, CuPS, and the MnI Complex Catalysts,
Mn(4X) and Mn(6mes)
RESULTS AND DISCUSSION
■
Photocatalytic Reduction of CO2. A N,N-dimethylace-
toamide (DMA)−triethanolamine (TEOA) (4:1, v/v) solution
containing the Mn complex (Mn(4H), Mn(4OMe) or
Mn(6mes)) as the catalyst (0.05 mM), CuPS as a redox
photosensitizer (0.25 mM), and 1,3-dimethyl-2-phenyl-2,3-
dihydro-1H-benzo[d]imidazole (BIH; 10 mM) as a reductant
was irradiated at λex = 436 nm. In the case using Mn(4H) as
the catalyst, the main product was HCOOH, and the TON
based on the catalyst used (TON(CAT)) was 157 for 2 h
irradiation (eq 4, Figure S2a).
phenantroline) as a catalyst (ΦCO = 6.7%, TON(PS) = 55,
TON(CAT) = 273)9 (Cu: 60 ppm, Fe: 56300 ppm in the
earth crust).8 Although homoleptic-type complexes, that is,
+
CuI(2,9-Me2phen)2 and its derivatives, have an advantage of
light absorption at longer wavelength up to 600 nm and were
used as a photosensitizer in photocatalytic H2 production
systems,10 their weak oxidation power in the excited state
caused lower efficiency of the photocatalytic reaction. Beller
and his co-workers recently reported another mixed system
consisting of a mononuclear Cu(I) photosensitizer complex
and Fe(II) cyclopentadienone carbonyl complexes as catalysts,
where ΦCO = 13.3% (TON(PS) = 97, TON(CAT) = 487).11
Robert and co-workers reported a combination of an organic
photosensitizer and a Fe quaterpyridine catalyst for CO2
reduction to give CO (ΦCO = 1.1%, TON(PS) = 341,
TON(CAT) = 1365).12 They also reported CH4 formation
using a Fe porphyrin in a combined system with an Ir(III)
I
complex(0.05mM) + CuPS(0.25mM)/hv (436nm)
CO2 ⎯M⎯⎯n⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯⎯→ CO and/or HCOOH
BIH(10mM)in DMA−TEOA(4:1v/v)
(4)
A certain amount of CO was also produced (TON(CAT)CO
= 50), whereas H2 evolution was minimal (TON(CAT)H
4). It should also be noted that almost all of the BIH was
consumed in the 2 h irradiation because BIH (40 μmol) is a
two-electron donor as shown in eq 5, and the production of all
=
2
photosensitizer (ΦCH = 0.18%, TON(PS) = 0.79, TON-
4
(CAT) = 79).13,14
The other notable earth-abundant metal complexes for the
catalysis of CO2 reduction are Mn(I) complexes, fac-
[MnI(N^N)(CO)3L]n+ (N^N = diimine ligand; L = Cl−,
Br−, NCS−, MeCN) (Mn: 950 ppm in the earth crust).6 Since
fac-MnI(bpy)(CO)3Br (bpy = 2,2′-bipyridine) was first
the reduction products requires two electrons (the total
amount of the reduction products was 41 μmol). Therefore,
the termination of the photocatalytic CO2 reduction was
attributed to the complete consumption of BIH.
B
DOI: 10.1021/jacs.8b10619
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 1. Quantum yield determination of photocatalytic CO2 reduction using Mn catalysts (0.05 mM) and CuPS (0.25 mM) in the presence of
BIH (10 mM) in a CO2-saturated DMA−TEOA (4:1, v/v) solution. (a) Mn(4H), (b) Mn(4OMe), and (c) Mn(6mes). The photocatalytic
reactions were performed at 25 °C under 436 nm monochromatic light (intensity: 2.0 × 10−8 einstein s−1).
Figure 2. Photocatalytic CO2 reduction using a higher concentration of BIH (0.1 M); [Mn(4OMe)] = 0.05 mM, [CuPS] = 0.25 mM; CO2-
saturated DMA−TEOA (4:1, v/v) solution, λex = 436 nm, 25 °C. Time courses of (a) product formation and (b) decrease of BIH and CuPS in the
solution during the photocatalytic reaction.
Table 1. Photocatalytic Properties of Systems Consisting of CuPS as the Redox Photosensitiser and Mn(I) Complexes As the
a
Catalyst
b
c
d
quantum yield
products /μmol (TON(CAT) , distribution/%)
Mn(I) complex
CO
HCOOH
CO
HCOOH
H2
Mn(4H)
Mn(4OMe)
Mn(6mes)
0.02
0.33
0.41
0.30
0.24
0.03
9.9 (50, 23.7)
32.7 (164, 66.3)
41.5 (208, 96.6)
31.3 (157, 74.6)
12.9 (65, 28.5)
0.9 (5, 0.2)
0.7 (4, 1.7)
0.2 (1, 5.2)
0.1 (0.5, 0.2)
a
The photocatalytic reaction performed using a 4 mL CH3CN−TEOA solution containing CuPS (0.25 mM), a Mn complex (0.05 mM), and BIH
b
c
(10 mM) under irradiation by 436 nm monochromatic light from a Hg lamp. Intensity of the irradiated light was 2.0 × 10−8 einstein s−1. The
amount of products and their distribution (Figure S2). Turnover number based on the catalyst used.
d
In the case of Mn(4OMe), the main product was CO with
TON(CAT)CO = 150 (Figure S2b), and less than half the
amount of HCOOH (TON(CAT)HCOOH = 65) along with a
ΦCO = 2%. In the case of using Mn(4OMe), ΦCO = 33% and
ΦHCOOH = 24%. Notably, the total quantum yield of the CO2
reduction products (Φtotal) reached 57% in the photocatalytic
system using Mn(4OMe). To the best of our knowledge, this
is the highest quantum yield for CO2 reduction using abundant
elements. The photocatalytic system using Mn(6mes) showed
the highest quantum yield of CO formation (ΦCO = 41% with
ΦHCOOH = 3%).
To check the durability of the photocatalytic systems
consisting of the Cu photosensitizer and the Mn catalyst, the
amount of BIH was increased to 0.1 M in the Mn(4OMe)
system (Figure 2a). After 24 h irradiation, TON(CAT)CO
reached 1004 with TON(CAT)HCOOH = 310 and TON-
small amount of H2 (TON(CAT)H = 11) were produced after
2
2 h irradiation. No further production of the reduction
products after 1 h irradiation supports the full consumption of
BIH. A very high selectivity of CO formation could be
achieved by using Mn(6mes) as the catalyst (Figure S2c),
where TON(CAT)CO = 208, TON(CAT)HCOOH = 5, and
TON(CAT)H = 3.
2
The formation quantum yields of the products were
determined during the initial stage of the photocatalytic
reactions, where the formation of the products increased in
proportion to the irradiation time (Figure 1). In the
photocatalytic system using Mn(4H), ΦHCOOH = 30% and
(CAT)H = 68 (TON(PS)s were 201, 62, and 14, respectively).
2
Figure 2b shows the changes in the concentration of BIH and
CuPS during the photocatalytic reaction. This clearly indicates
C
DOI: 10.1021/jacs.8b10619
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
that the consumption of BIH was also a reason for decreasing
the rate of the photocatalytic reaction.29 It also indicates that
CuPS is a very stable and efficient redox photosensitizer, as
decomposition was only 2% even after 12 h irradiation where
the TON(PS) of all the reduction products was over 200 and
most of the BIH was already consumed. As can be seen in the
video (see Video S1), this system generated vigorous bubbles
of CO from a few minutes after the photoirradiation, and the
formation of bubbles was still observed even after 5.5 h.
Interestingly, the selectivity of CO formation was 76% after 36
h irradiation, which is higher than that obtained in the
quantum yield determination (58%). The concentration of
BIH, which is a strong base, and/or the enhancement of the
electron injection rate into the catalyst might be the reasons for
these observations.
Table 1 summarizes the photocatalytic properties of the
systems studied. In this section, we conclude that the
combination of CuPS and the Mn catalysts is very efficient
and durable visible-light-driven photocatalytic systems for CO2
reduction. The distribution of CO and HCOOH can be
controlled by changing the substituents on the bpy ligand of
the Mn catalysts (CO:HCOOH was 93:6 and 6:94 for
Mn(6mes) and (Mn(4H), respectively).
The crystal structure of CuPS was successfully obtained by
X-ray analysis as shown in Figure 3. The crystallographic data
and the selected bond lengths and angles are summarized in
Tables S1 and S2. Two Cu-complex units in one molecule had
the same structure, and the distance between the two Cu+ ions
was 8.5 Å, which is too far for any interaction. The angle
between the planes containing NCuN and PCuP was 88.5°,
which is required for a long lifetime of the excited state of this
type of CuI complexes.30 The P−Cu−P bite angle was 113.4°,
which was smaller than that of the corresponding mononuclear
complex, CuI(2,9-Me2phen)(PPh2Pr)2(PF6) with one 2,9-
Me2phen and two monodentate phosphine ligands,9 indicating
the angles were restricted by tethering the quarter dentate
ligand with the phen structure in CuPS. It was shown that the
phenyl planes at the 4,7-positions of the phen moiety in the
ligands were twisted against the phen plane (51.7 and 68.9°)
because of steric repulsion between hydrogen atoms. This
result indicates that the π-conjugation between the phenyl
groups and phen moiety in the ligands is not strong in the
ground state of CuPS. The bond lengths around the Cu and P
atoms and the phen−C lengths were similar to those in other
CuI complexes, that is, Cu−P (2.3 Å), Cu−N (2.1 Å), P−C
(1.8 Å), and C[phen]−C (1.5 Å).
The photocatalytic reactions using CuPS and Mn catalysts
were conducted in DMA solutions. The photophysical and
electrochemical properties of CuPS were remeasured in DMA,
as these properties have only previously been measured in
MeCN,9 and, in general, these properties of many Cu(I)
complexes are highly dependent on the solvent.31 Figure 4a
shows the UV−vis absorption spectrum of CuPS measured in
DMA. The broad absorption band with λmax = 384 nm is
attributed to the singlet metal-to-ligand-charge-transfer
(1MLCT) transition from CuI to the phen moiety. The
molar extinction coefficient at the absorption maximum was
ε384 nm = 9800 M−1 cm−1.
Synthesis, Structure, and Properties of CuPS. In one of
the previous papers,9 we briefly reported the synthesis of CuPS
and its properties in MeCN. As CuPS works as a magnificent
redox photosensitizer, we herein report more details about its
structure and properties, especially in DMA and DMA−TEOA
solutions. We successfully improved its synthesis method by
optimizing the reaction condition (eq 6); the details are
described in the Experimental Section. The isolated yield was
12%.
Figure 4a also shows the emission spectrum from CuPS
excited at λex = 444 nm, due to the delayed fluorescence via the
1MLCT excited state which is thermally accessible from the
Figure 3. ORTEP drawing of the Cu complex: CuPS(PF6)2·5MeCN·MeOH. Displacement ellipsoids are shown at the 50% probability level. H
atoms, PF6 anions and solvent molecules were omitted for clarity.
D
DOI: 10.1021/jacs.8b10619
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 4. (a) UV−vis absorption and emission spectra (excitation at 444 nm) and (b) emission decay (excited at 401 nm and monitored at 650
nm) of CuPS in DMA at 25 °C. (c) UV−vis absorption spectra of Mn(4H) (red), Mn(4OMe) (blue), and Mn(6mes) (green).
3MLCT excited state.32 The emission maximum and quantum
yield were λmax = 634 nm and Φem = 2.7% in DMA,
respectively.
Emission decay was measured at λ = 650 nm by using the
single-photon counting method (Figure 4b, Supporting
Information); just after excitation at λex = 401 nm, a rapid
decay was observed within the time resolution of the apparatus
(1 ns), which is attributable to the dynamics relevant to the
structural distortion, had been reached,33 and then a slower
decay was observed with τem = 810 ns, which is attributed to
the delayed fluorescence.32
Figure 5a shows the cyclic voltammogram (CV) of CuPS;
one reversible redox couple was observed at E1/2 = −1.92 V vs
Ag/AgNO3. This shows that the one-electron-reduced species
(OERS) of CuPS is stable in the time scale of the CV
measurement. Figure 6a shows the changes in the UV−vis
absorption of an MeCN solution containing CuPS (0.25 mM)
and Et4NBF4 (0.1 M) as a supporting electrolyte during flow
electrolysis at various applied potentials under an Ar
atmosphere. The dependence of the absorption at λ = 720
nm and the current on the applied voltage are shown in Figure
6b, which indicate that the OERS of CuPS is stable and its
molar extinction coefficient is ε = 5500 M−1 cm−1 at λ = 720
Figure 5. CVs of CuPS (0.25 mM, a) and Mn(4X) (0.5 mM: b, X =
nm.
H; c, X = OMe: d) or Mn(6mes) (0.5 mM) in DMA solutions
Reaction Mechanism. In all the UV−vis absorption
containing 0.1 M Et4NBF4 as a supporting electrolyte under an Ar
1
spectra of the Mn complexes, a broad MLCT band was
atmosphere: WE, glassy carbon (ϕ 3 mm); CE, Pt wire; RE, Ag/
observed in the visible region (Figure 4c). The absorption
AgNO3 (0.01 M); scan rate, 0.2 V s−1.
1
maxima of the MLCT bands and the absorbance at λ = 436
nm (the wavelength of the irradiation light in the photo-
catalytic reactions) are summarized in Table 2. Since the
concentration of CuPS was five times higher than those of the
Mn complexes, the absorbance of CuPS was found to be 8.6−
10.6 times higher than those of the Mn complexes in the
reaction solutions.
The excited-state lifetimes of the Mn complexes are too
short to initiate the photocatalytic reactions.34 Therefore, the
photoinduced electron transfer from BIH to CuPS should be
the initial process in the photocatalytic CO2 reduction, which
was investigated by emission quenching and transient
absorption spectroscopy as follows. Emission from CuPS was
efficiently quenched by BIH, as shown in Figure 7a, from
which a Stern−Volmer analysis showed linear plots with an
intercept at 1 (Figure 7b). From the slope of this line (KSV
=
kqτem = 1.04 mM−1) and the emission lifetime τem = 810 ns, the
E
DOI: 10.1021/jacs.8b10619
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 6. UV−vis spectral changes obtained during the flow electrolysis of CuPS (0.25 mM) in MeCN containing Et4NBF4 (0.1 M) as a
supporting electrolyte under an Ar atmosphere (flow rate: 0.2 mL min−1, carbon felt as a working electrode, Pt wire as a counter-electrode, Ag/
AgNO3 (0.01 M) as a reference electrode). The electrolyzed solution was introduced into an optical cell (1.5 mm path length) for measuring the
absorption spectrum. Gray circles show the current, and the blue squares show the absorbance at λ = 720 nm.
of the excited state(s) was only observed in the early stage;
however, it rapidly decayed and a new broad absorption band
appeared at λmax ∼ 750 nm, which was attributed to the OERS
of CuPS (CuPS•−) (Figure 8b). These results clearly indicate
that the reductive quenching of the excited state(s) of CuPS
(*CuPS) proceeded by BIH to give CuPS•− (Figure 6).
Figure 8c shows the time course of the absorbance change at
λ = 720 nm. There were two kinds of increase in the transient
absorption: a fast increase within 100 ns from laser irradiation
and then a following slow rise up until 5 μs after the
irradiation. Since, under these reaction conditions, 91% of the
excited Cu complex was reductively quenched by BIH (10
mM) and the emission lifetime of CuPS was 71 ns, which
coincides with the fast absorption rise in the transient
absorption measurement, the first increase at λ = 720 nm
can be mainly attributed to the photochemical electron transfer
from BIH to *CuPS to produce CuPS•−. It is noteworthy that
this absorption change should include a decrease in *CuPS,
which has a weaker absorption at this wavelength (Figure 8a).
After 100 ns after laser irradiation, most of the *CuPS should
disappear. Since it has been reported that BIH works as a two-
electron donor as shown in eq 4,35 the slower increase in the
transient absorption at λ = 720 nm should be caused by the
reduction of the ground state of CuPS by the deprotonated
Table 2. Light-Absorbing Properties of the Cu and Mn
Complexes
a
b
c
complex
CuPS
Mn(4H)
Mn(4OMe)
Mn(6mes)
λmax /nm (ε /M−1 cm−1
)
A436
384
417
403
388
(9800)
(2640)
(3000)
(2390)
0.95
0.11
0.09
0.09
a
b
c
1MLCT absorption maxima. Molar extinction coefficient. Absorb-
ance at 436 nm under photocatalytic condition (CuPS, 0.25 mM; Mn
complexes, 0.05 mM; a 1 cm pass length).
quenching rate constant was determined as kq = 1.3 × 109 M−1
s−1.
Laser-pulse irradiation (λ = 355 nm, 10 ns fwhm) of a
DMA−TEOA (4:1, v/v) solution containing only CuPS (0.05
mM) induced a new transient absorption band with absorption
maxima at 500 and 630 nm (Figure 8a). The absorption
decayed with τ = 840 ns without undergoing any change in the
shape. Since this lifetime was almost identical to the emission
lifetime from CuPS (τem = 810 ns as described above), this
transient absorption is therefore attributed to arise mostly from
3
the MLCT excited state of CuPS, which might have a small
contribution from its 1MLCT excited state in thermal
equilibrium. In the presence of BIH (10 mM), the absorption
36
product of BIH•+, that is, BI• (Ep = − 2.06 V vs Fc+/Fc ).
ox
Figure 7. Emission spectra (excitation at 444 nm) of CuPS in DMA under an Ar atmosphere in the presence of various amounts of BIH (a) and
the Stern−Volmer plot using the emission intensity (I) observed at 630 nm (b). I0 is the emission intensity without any BIH.
F
DOI: 10.1021/jacs.8b10619
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 8. Transient absorption spectra measured after irradiation with a 355 nm laser pulse of a DMA−TEOA (4:1, v/v) solution containing CuPS
(0.05 mM) (a) in the absence or (b) in the presence of BIH (10 mM) under an Ar atmosphere. (c) Transient absorption change (red solid curve)
monitored at λ = 720 nm and the excitation laser pulse intensity profile (green solid curve).
Scheme 2 summarizes the mechanism of the photochemical
reduction of CuPS by BIH, including the back-electron-
shown in Figure 9. From this figure, we could determine the
formation yield of CuPS•− as 16.2 0.6 μM in the initial stage,
a
Scheme 2. Initial Processes of the Photocatalytic Reaction.
a
I
abs: Absorbed photons. k0: Intrinsic decay rate constant of *CuPS.
kq: Quenching rate constant of *CuPS by BIH. kesc: Escape rate
constant from the solvated radical ion pair to the individually solvated
free radical ions. krec: Back-electron-transfer rate constant from
CuPS•− to BIH•+ in the solvated ion cage. krec′: Back-electron-transfer
rate constant between the solvated CuPS•− and BIH•+: kdp:
Deprotonation rate constant of BIH•+ by the base species, mostly
TEOA: ket: Electron-transfer rate constant from BI• to the ground-
state CuPS.
Figure 9. Concentration changes of *CuPS (blue curve) and CuPS•−
(red curve) as calculated from the transient absorption change of a
DMA−TEOA (4:1, v/v) solution containing CuPS (0.05 mM) in the
presence of BIH (10 mM) under an Ar atmosphere.
transfer processes from CuPS•− to the one-electron-oxidation
species of BIH (BIH•+) both in the state of the solvated radical
ion pair [CuPS•−···BIH•+]sol and the individually solvated free-
radical ions.
mostly attributed to the photochemical electron transfer from
For quantifying these CuPS•−-generation processes, the rise
in absorption at λ = 720 nm (ΔA(t)) was analyzed by using
the model shown in Scheme 2. Since, in the initial stage up to
250 ns after laser irradiation, both *CuPS and CuPS•− should
coexist in the reaction solution, the absorbance attributed to
BIH to *CuPS, and 7.5
0.5 μM in the slower process
attributed to the electron transfer from BI• to the ground state
of CuPS. Since the absorbed photon number was determined
as 25.8 × 10−6 einstein dm−3 (Supporting Information), the
total quantum yield of formation of CuPS•− was ΦOER = 1.20
0.06, which can be divided into Φ1 = 0.82 for the initial
photochemical electron-transfer process and Φ2 = 0.38 for the
slower process since the difference in the time scales between
these two processes was very large. Our assumption of the
absorbed photon number is the lowest possible value because
we assumed that the formation quantum yield of the lowest
3MLCT excited state of the Cu photosensitizer, that is, *CuPS,
*CuPS was calculated by using the molar extinction coefficient
of *CuPS (2700 M−1 cm−1 at 720 nm; see Supporting
Information) and the time profile of *CuPS, which was
obtained from the emission decay kinetics (Figure S4). By
using this data and the molar extinction coefficient of CuPS•−
(5500 M−1 cm−1 at 720 nm: Figure 6), ΔA(t) can be
converted into the concentrations of *CuPS and CuPS•− as
G
DOI: 10.1021/jacs.8b10619
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Table 3. Redox Potentials of the Mn Complexes and Their Reduced Species As Observed by Flow Electrolysis
a
redox potential (Ep) /V vs Ag/AgNO3
λmax/nm (ε/M−1 cm−1
)
b
complex
cathodic
anodic
Mn−Mn dimer
(11,500)
Mn−
Mn(4H)
Mn(4OMe)
Mn(6mes)
−1.59, −1.77
−1.73, −1.93
−1.69
−1.37, −0.49
−1.55, −0.69
−1.43
805
840
nd
569
566
572
(13,600)
(8,800)
(10,600)
(7,500)
c
c
nd
a
b
c
Peak potential. Two-electron reduced 5-coordinate 18-electron species ([Mn(N^N)(CO)3]−). Not detected.
which was used for determining the absorbed photon, is unity.
It has been reported that the rate constant of intersystem
crossing (kISC) is much larger than that of intrinsic direct
deactivation from S1 to S0 (kS) not only in Cu(I) bis-diimine
complexes33 but also in mononuclear Cu(I) diimine bi-
sphosphine complexes (kISC ∼ 1011 s−1, kS ∼ 107 s−1),37 which
means that the quantum yields of the intersystem crossing
were almost unity. It has been also reported that another Cu(I)
dimer complex Cu2Cl2(N^P)2 (N^P = 2-pyridyldiphenylphos-
phine), of which Cu(I)s interact each other through the Cl
ligands, the intersystem crossing rate increases due to
enhanced spin orbit coupling.38 Although such intramolecular
interaction might be small in CuPS because the two Cu(I)
centers largely separated each other (8.5 Å), it should be
reasonable to assume that the quantum yield of the intersystem
crossing to produce *CuPS is almost unity. It is also
noteworthy that Φ2 should depend on the light intensity
because the back-electron-transfer process, which lowers Φ2,
complies with second-order kinetics (Scheme 2; both the
concentration of BIH•+ and CuPS•− depend on light
intensity), whereas Φ1 is independent of light intensity.
Since, here, Φ2 was determined by using a very strong laser
light for the excitation, the Φ2 value in the photocatalytic
reactions should be higher than that in the laser flash
photolysis experiment.
From the results of this investigation, it should be interesting
to compare the data obtained here with that of other redox
photosensitizers, typically [RuII(dmb)3]2+ (dmb = 4,4′-
dimethyl-bpy), which has been frequently used in the
photocatalytic reactions for CO2 reduction; ΦOER = 0.59
(higher light intensity, i.e., 1.1 × 10−8 einstein s−1) and 0.95
(lower light intensity, i.e., 1.1 × 10−9 einstein s−1) with [BIH]
= 0.1 M (100% quenching of the excited Ru complex),35 which
was determined by irradiation using a high-pressure mercury
lamp with a much lower light intensity than the laser light used
in this study. Therefore, the formation efficiency of OERS is
higher in the case of CuPS when compared to [RuII(dmb)3]2+.
This is another advantage of CuPS.
The next electron-transfer process should be from the
produced CuPS•− to the Mn complexes to produce the OERS
of the Mn complexes because in all the cases it is an exergonic
reaction; their CVs are illustrated in Figure 3b−d, and their
reduction potentials are summarized in Table 3. In order to
identify and quantify the intermediates formed during the
photocatalytic reactions, the UV−vis absorption spectral
changes during the flow electrolysis of the Mn complexes
were first studied in MeCN containing Et4NBF4 (0.1 M) as a
supporting electrolyte under an Ar atmosphere (Figure 9). In
the case of Mn(4H) (Figure 9a), at applied potentials between
−1.40 and −1.53 V vs Ag/AgNO3, Mn−Mn dimer formation
was observed, which is characterized by a rise in the absorption
bands at ca. 650 and 800 nm.15,39 It was reported that the
Mn−Mn dimer is produced via the coupling of the 17-electron
species, that is, •Mn(bpy)(CO)3 (eq 7). Electrolysis at more
negative potentials changed the spectrum to one which had a
strong absorption band at around λmax = 550 nm
corresponding to the 5-coordinated 18-electron species, that
is, [Mn(bpy)(CO)3]− (eq 8)15,39 with isosbestic points at 400,
500, and 700 nm.
Figure 10b shows the relationships of both the current and
the absorbance at λ = 569 and 805 nm with the applied
potential. These results clearly indicate that the Mn−Mn dimer
accepted two electrons to quantitatively yield [Mn(bpy)-
(CO)3]−. The molar extinction coefficients of the Mn−Mn
dimer is 11,500 M−1 cm−1 at λ = 805 nm and that of
[Mn(bpy)(CO)3]− is 13,600 M−1 cm−1 at λ = 569 nm. Similar
spectral changes were observed in the case of Mn(4OMe),
except that the applied potentials were more negative in the
case of Mn(4OMe) (Figure 10c and d). The molar extinction
coefficients are summarized in Table 3. In the case of
Mn(6mes), the spectral changes and the potential−current
relationship were very different from those of the other Mn
complexes, as shown in Figure 10e; no formation of the Mn−
Mn dimer was observed. This is reasonable because the
sterically bulky mesityl groups should prevent the dimerization
of the 17-electron species.18 Actually, the CV of Mn(6mes)
shows a two-electron reduction wave at Ep = −1.63 V (Figure
5d). This was also supported by Figure 10f where only one
two-electron reduction curve was observed at E = −1.60 V to
−2.20 V.
Figures 11 and 12 show the UV−vis absorption spectral
changes of a DMA−TEOA (4:1, v/v) solution containing the
Mn complexes (0.05 mM), CuPS (0.25 mM) and BIH (10
mM) during irradiation under an Ar atmosphere and under a
CO2 atmosphere, respectively.
Under an Ar atmosphere, the Mn−Mn dimers rapidly
formed in the initial stage (λmax = 810 nm) in the case of
Mn(4H). Further irradiation caused a decrease in the dimer
accumulation, and the simultaneous formation of the 5-
coordinate 18-electron species ([Mn(bpy)(CO)3]−, λmax = 565
nm) was observed (Figure 10a). The formation of the Mn−
Mn dimer was similarly observed in the case using Mn(4OMe)
(Figure 11b). Further irradiation induced a decrease in the
absorption of the dimer and the formation of the
corresponding 5-coordinate 18-electron species. In the case
of Mn(6mes), on the other hand, the formation of the dimer
was not observed while the corresponding 5-coordinate 18-
electron species accumulated after irradiation for 30 s (λmax
=
H
DOI: 10.1021/jacs.8b10619
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 10. UV−vis spectral changes during the flow electrolysis of Mn(4H) (a), Mn(4OMe) (c), and Mn(6mes) (e) under an Ar atmosphere in
MeCN containing Et4NBF4 (0.1 M) as a supporting electrolyte, carbon felt as a working electrode, Ag/AgNO3 (0.01 M) as a reference electrode,
and coiled Pt wire as a counter-electrode at a flow rate of 0.2−0.3 mL min−1. The electrolyzed solution was introduced into an optical cell (1.5 mm
path length) to measure the UV−vis absorption spectrum. On the right-hand side, the relationship of current and absorbance with applied voltage
is illustrated (gray indicates current): (b) Mn(4H), absorption at 805 nm (green) and 565 nm (blue); (d) Mn(4OMe), absorption at 840 nm
(green) and 566 nm (blue); (f) Mn(6mes), absorption at 572 nm (blue). The absorption maxima of the reduced species are summarized in Table
3.
573 nm), followed by the accumulation of CuPS•−. Reduction
of Mn(6mes) by CuPS•− should proceed because it is a
thermodynamically favorable process (Table 2) and it
produces the OERS of Mn(6mes). Release of the Br− ligand
from the OERS of Mn(6mes) gave the 17-electron species,
and then its further reduction yielded the 5-coordinate 18-
electron species. This reduction process might proceed by
either another CuPS•− or BI• (eq 9).
Under a CO2 atmosphere, the Mn−Mn dimer steadily
accumulated as the photocatalytic CO2 reduction proceeded in
the cases using Mn(4H) and Mn(4OMe) (Figure 12a and b).
In contrast to these Mn complexes, almost no accumulation of
the reduced Mn complex was observed in the case of
Mn(6mes) (Figure 12c). These results provide important
information on the intermediates of the photocatalytic
reduction of CO2 (Scheme 3). In the case of Mn(6mes), the
reactive intermediate should be the 5-coordinate 18-electron
species,18 which reacts with CO2 to give CO selectively. In the
case of Mn(4OMe), on the other hand, the Mn−Mn dimer
probably reduces CO2 giving CO.16 Although a similar
I
DOI: 10.1021/jacs.8b10619
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 11. UV−vis absorption spectral changes of a DMA−TEOA (4:1, v/v) solution containing the Mn complexes (0.05 mM): Mn(4H) (a);
Mn(4OMe) (b); Mn(6mes) (c) and CuPS (0.25 mM), and BIH (10 mM) during irradiation under an Ar atmosphere; λext = 436 nm, 25 °C. (d)
Without a Mn complex.
reaction of the Mn dimer made from Mn(4H) might also
proceed, its reaction rate with CO2 was probably slower25 than
that of the dimer made from Mn(4OMe) because the
oxidation potential of the dimer with X = H was more
negative than that when X = OMe (the anodic scan in Table
3). In both cases, using either Mn(4H) or Mn(4OMe), the 5-
coordinate 18-electron species probably contributed to the
CO2 reduction because its absorption was not observed in the
CO2 atmosphere even though that of the Mn−Mn dimer was
observed as the photocatalytic reaction proceeded. We cannot
identify the product distribution of this reaction at the present
stage.
Mn(6mes) and the Mn−Mn dimer and/or the 5-coordinate
18-electron species in the cases using Mn(4H) or Mn(4OMe).
EXPERIMENTAL SECTION
■
Materials. DMA was dried over 4 Å molecular sieves and distilled
at reduced pressure. TEOA was distilled at reduced pressure.
Tetraethylammonium tetrafluoroborate (Et4NBF4) was recrystallized
twice from MeCN-ethyl acetate and dried in vacuo at 100 °C
overnight just before use. A sacrificial reductant, BIH,32 was prepared
by the method previously defined. Mn(4H)25 and Mn(6mes)18 were
synthesized according to the literature method. Synthetic procedures
of Mn(4OMe) are described in the Supporting Information. All other
reagents were of reagent-grade quality and were used without further
purification.
CONCLUSION
Synthesis. PPh2PrBr. Under a N2 atmosphere, n-BuLi (4.1 mL, 6.6
mmol, 1.60 M in hexane) was added dropwise to a mixture of
tetrahydrofuran (THF; 8.3 mL) containing HPPh2 (1.3 mL, 6.5
mmol) at 0 °C. This mixture was stirred for 50 min at room
temperature. The obtained red solution was transferred to a closed
flask containing 1,3-dibromopropane (2.7 mL, 26.9 mmol) at room
temperature through a cannula under a N2 atmosphere. The solution
was stirred for 20 min at room temperature. The organic solvents
were then removed with a rotary evaporator, and the residue was
extracted with CH2Cl2 (6 mL) after the addition of H2O (6 mL). The
organic layer was dried by using Na2SO4, and then the solvent was
removed using a rotary evaporator. The obtained oil was purified by
column chromatography on silica gel (40−50 μm, N60 (Kanto)) with
CH2Cl2−n-hexane (1:4, v/v) to give 396.6 mg of a colorless oil
containing PPh2PrBr as a mixture with 25 mol % of Ph2P-
■
We successfully constructed a new, efficient, selective, and
durable photocatalytic system for CO2 reduction, which only
consists of earth-abundant metal complexes, that is, CuPS and
Mn(4OMe) (the maximum values: ΦCO+HCOOH = 57%,
TON(CAT)CO+HCOOH = 1314, TON(PS)CO+HCOOH = 263,
selectivity (CO + HCOOH):H2 = 95:5). The Cu(I) dinuclear
complex CuPS can work as an outstanding redox photo-
sensitizer, which has a long excited-state lifetime and a much
stronger reduction power of its OERS (E1/2 = −1.92 V vs Ag/
AgNO3) compared to most other photosensitizers (e.g., E1/2
=
−1.65 V in the case of [RuII(bpy)3]2+). The product selectivity
between CO and HCOOH could be controlled from
CO:HCOOH = 93:6 (Mn(4H)) to 6:94 (Mn(6mes)) by
changing the substituents of the bpy ligand. The mechanistic
studies suggested that the intermediate(s) for CO2 reduction is
(are) the 5-coordinate 18-electron species in the case using
1
(CH2)3PPh2, as judged by the H NMR spectrum. This compound
was used without further purification for the following procedures as
1
described below because of the instability of PPh2PrBr. H NMR
(chloroform-d): δ (ppm) 7.44−7.32 (m, 10H, PPh2−), 3.48 (t, J = 7.2
J
DOI: 10.1021/jacs.8b10619
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Figure 12. UV−vis absorption spectral changes during the photocatalytic reactions using the Mn complexes (0.05 mM): Mn(4H) (a); Mn(4OMe)
(b); Mn(6mes) (c), CuPS (0.25 mM), and BIH (10 mM) in a CO2-saturated DMA−TEOA (4:1, v/v) solution; λext = 436 nm, 25 °C. (d)
Without a Mn complex.
Scheme 3. Mechanism of the CO2 Reduction Catalyzed by the Mn(I) Complexes
1
Hz, 2H, BrCH2−), 2.21−2.17 (m, 2H, PPh2CH2−), 2.02−1.93 (m,
2H, PPh2CH2CH2−).
and dried in vacuo. Yield: 49.2 mg (12.0%). The H NMR spectrum
data was consistent with the reported one.9
CuPS. A cyclohexane solution (1.0 mL) containing lithium
diisopropylamide (1.5 M) was added dropwise to a THF solution
(15 mL) containing bathocuproine (bcp) (143.9 mg, 0.40 mmol) at
−45 °C under a dry Ar atmosphere and was then stirred for 15 min.
The solution was added via cannula into an Ar-deoxygenised THF
solution (15 mL) containing the product and PPh2PrBr (396.6 mg,
ca. 1 mmol), as described above, and CuI (76.0 mg, 0.40 mmol) at 0
°C, and then the solution was stirred for 16 h at room temperature.
The reaction solution was quenched with water containing an excess
amount of KPF6. The products were extracted with CH2Cl2 (70 mL)
four times. Column chromatography on alumina (aluminum oxide 90
standard (Merck)) with CH2Cl2/Et2O (1:1, v/v) provided a solution
containing CuPS as a PF6− salt. After evaporation, further purification
was achieved by recrystallization from an MeCN, Et2O and n-hexane
solution. The yellow precipitates were filtered off, washed with Et2O
General Procedures. 1H NMR spectra were measured in
chloroform-d using a JEOL ECA400II (400 MHz) system. UV−vis
absorption spectra were recorded on a JASCO V-565 or V-670
spectrometer. FT−IR spectra were measured by a JASCO FT/IR-610
or FT/IR-6600 spectrometer at 1 cm−1 resolution in an MeCN
solution. Emission spectra were recorded on a JASCO FP-6500
spectrometer. Emission quantum yields were collected using a
Hamamatsu Photonics C9920-02G absolute photoluminescence
quantum-yield measurement system consisting of a calibrated
integrating sphere and a multichannel spectrometer. Sample solutions
were degassed by freeze−pump−thaw cycles before the emission
measurements. CVs were measured using Et4NBF4 (0.1 M) as a
supporting electrolyte under an Ar atmosphere using a BAS
CHI620EX or CHI760Es electrochemical analyzer with a glassy-
carbon working electrode (diameter, 3 mm), an Ag/AgNO3 (0.01 M)
K
DOI: 10.1021/jacs.8b10619
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
as a reference electrode, and a Pt counter-electrode at a scan rate of
200 mV s−1. Procedures of single crystal X-ray analysis of CuPS and
emission decay lifetime measurement are described in the Supporting
Information.
mm o.d., 20 mm length) and consisted of carbon felt. The counter-
electrode was a Pt wire coiled around the porous glass tube. The Ag/
AgNO3 (0.01 M) reference electrode was positioned near the
counter-electrode. An MeCN solution containing a complex (0.5 mM
for Mn complexes: 0.25 mM for CuPS), and Et4NBF4 (0.1 M) as an
electrolyte, was purged with Ar for 20 min and was then passed into
the flow-through cell at a rate of 0.2 mL/min using a Fisher Scientific
Variable-Flow Peristaltic Pump with a Cole-Parmer Masterflex C-Flex
tubing. Immediately after electrolysis using a BAS CHI760Es
electrochemical analyzer, the solution was transferred to a quartz
cell (pass length 1.5 mm) for spectral measurements. The spectral
changes were recorded on a BAS SEC2000-UV−vis spectrometer
with a SEC2000-DH light source. The number of electrons reduced
per Mn complexes or Cu unit in CuPS (n) was determined using
Photocatalytic Reaction. A 4 mL DMA−TEOA (4:1, v/v)
solution containing CuPS (0.25 mM), the Mn complexes (0.05 mM),
and BIH (10 mM or 0.1 M) in a necked quartz cubic cell (1 cm pass
length; 11 mL volume) was bubbled with CO2 for over 30 min, and
the cell was sealed using a septum. The sample solution was irradiated
by an Ushio Optical Module high-pressure Hg lamp (BA-H500) with
a 436 nm band-pass filter (fwhm: 10 nm), purchased from Asahi
Spectra Co., and a CuSO4·5H2O aqueous solution (20 g L−1, 5 cm
pass length) filter. A neutral density (ND) glass filter was used to
adjust the light intensity. The solution was constantly mixed by a
magnetic stirrer bar during the photoirradiation. The temperature of
the reaction solution was maintained at 25 0.1 °C using an IWAKI
CTS-134A cooling thermo pump during irradiation. The UV−vis
absorption spectral changes of the reaction solutions during
photoirradiation were recorded on a Photal MCPD-6800 photo-
diode-array spectrometer.
For the determination of the quantum yields of the photocatalytic
reactions, the sample solution was irradiated using a SHIMADZU
QYM-01 photoreaction quantum yield evaluation system. A 4 mL
solution in an 11 mL quartz cubic cell (light path length: 1 cm) was
irradiated with 436 nm monochromatic light using a 300 W xenon
lamp (MAX-303 ASAHI SPECTRA) with a band-pass filter. The
solution was constantly mixed by a magnetic stirrer bar during the
photoirradiation. The temperature of the solutions was controlled at
25 °C using an IWAKI CTS-134A constant temperature system
during irradiation. The light intensity was controlled at 2.0 × 10−8
einstein s−1. The amount of formic acid in the reaction solution was
analyzed by a capillary electrophoresis system (Otsuka Electronics
Co. CAPI-3300I or 7100L) with a buffer solution (pH 6.0) consisting
of quinolinic acid, hexadecyltrimethylammonium hydroxide, and 2-
amino-2-hydroxymethyl-1,3-propanediol as the electrolyte. The
amounts of CO and H2 were analyzed using GC-TCD (GL science
GC323) with an active carbon column. The TONs for the products,
TON(CAT), were calculated as the amount of product divided by the
amount of the Mn catalyst used, whereas TON(PS) was calculated as
the amount of the product divided by the amount of the CuPS used.
The amounts of CuPS in the reaction solution before and after the
photocatalytic reaction were determined by the high-performance
liquid chromatography (HPLC) analysis40 with a system consisting of
a JASCO 880-PU pump, a pair of Shodex PROTEIN KW402.5
columns with a KW-LG guard column, a Rheodyne 7125 injector, and
a JASCO MD-2010 plus UV−vis photodiode-array detector. The
column temperature was kept at 40 °C in a JASCO 860-CO column
oven. An MeCN−MeOH (1:1, v/v) mixed solution containing
CH3COONH4 (0.5 M) was used as the mobile phase. For
quantitative analysis of BIH, an HPLC system consisting of a
JASCO 880-PU pump, a Develosil ODS-UG-5 column, a JASCO 880-
51 degasser and a JASCO MD-2010 plus UV−vis photodiode-array
detector was used.32 The column temperature was maintained at 30
°C using a JASCO-860-CO column oven. A mixed solution consisting
of MeCN−buffer (pH 7, NaOH−KH2PO4 (0.05 M), 3:2, v/v) was
used as the mobile phase.
n = (i − i0)/C0Fv
(10)
where i is the current (A) applied in the constant potential electrolysis
of the complexes, i0 (A) is the background current, C0 (M) is the
initial concentration of the Mn complexes or the Cu unit in CuPS
(0.5 mM), F is the Faraday constant (96,500 A/mol), and v is the
flow rate of the solution (0.26 mL/min for CuPS, 0.23 mL/min for
Mn(4H), and Mn(4OMe), 0.27 mL/min for Mn(6mes)). For
example, the quantitative one-electron reduction (n = 1) of each
complex unit requires 0.16 mA (= i − i0) according to this equation
when v = 0.2 mL/min is employed.
Time-Resolved UV−vis Transient Absorption Spectroscopy.
Third-harmonic-generation light at 355 nm produced by a Spectra-
Physics Quanta-Ray LAB-150-10 pulsed Nd:YAG laser was used for
excitation (10 ns fwhm). An Ushio 300-W Xe arc lamp was operated
in a pulse-enhanced mode (500 μs duration) using an XC-300 power
supply and a YXP-300 light pulsar (Eagle Shoji) as the monitoring
light source. The monitoring light beam was passed through the
quartz cuvette (10 × 10 × 40 mm) that contained a sample and was
directed into an R926 photomultiplier tube (Hamamatsu Photonics)
on a Jobin−Yvon HR-320 monochromator. Time profiles of the
monitoring light intensity were stored using a LeCroy WaveRunner
640zi oscilloscope (4 GHz bandwidth). Transient spectra were
obtained with an Andor Technology iStar H320T-18F-03 (690
channels; minimum gate width: 5 ns) ICCD detector head mounted
on the HR-320 monochromator. A DMA−TEOA (4:1, v/v) solution
that contained CuPS was bubbled with Ar prior to the laser flash
photolysis.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/jacs.8b10619.
■
S
Crystallographic Data for CuPS (CCDC reference:
1863041) embedding RES and structure factors (CIF)
1
Experimental details, synthesis and H NMR spectrum
of Mn(4OMe), the single-crystal X-ray crystallography
of CuPS, time courses of the photocatalytic CO2
reduction, details of the photochemical formation of
CuPS•−, evaluation of εT(λ), and the absorbed photon
numbers in the transient absorption measurement
(PDF)
Video S1: Movie of the beginning and ending of the
photocatalytic CO2 reduction using the system consist-
ing of CuPS, Mn(4OMe) and BIH under 450 nm light
irradiation for 5.5 h is available (AVI)
For taking the movie of the photocatalytic CO2 reduction, the
system consisting of CuPS, Mn(4OMe) and BIH was used. The CO2-
saturated DMA−TEOA (4:1, v/v) solution (4 mL) containing CuPS
(0.25 mM), Mn(4OMe) (0.05 mM), and BIH (0.1 M) was loaded in
the quartz cell closed at the top of the neck and was irradiated with
constant stirring for 5.5 h. The photoirradiation was carried out at λ =
450 nm using a Cell System IRS-450F LED lamp with an output
current controlled constantly at 2000 mA by a Cell System Iris-S
software. The temperature of the cell was maintained at 25 °C by a
water bath connected to an EYELA NCB-1210 temperature
controlling circulator during the photoirradiation.
AUTHOR INFORMATION
Corresponding Author
*ishitani@chem.titech.ac.jp
■
Flow Electrolysis. Flow electrolysis40,41,42 was used to obtain the
UV−vis absorption spectra of the OER species of the complexes. The
working electrode was contained in a porous glass tube (1 mm i.d., 2
ORCID
Akiko Sekine: 0000-0001-9220-944X
L
DOI: 10.1021/jacs.8b10619
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
Efficient Electrocatalyst for CO2 Reduction. Angew. Chem., Int. Ed.
2011, 50, 9903−9906.
(16) Bourrez, M.; Orio, M.; Molton, F.; Vezin, H.; Duboc, C.;
Deronzier, A.; Chardon-Noblat, S. Pulsed-EPR Evidence of a
Manganese(II) Hydroxycarbonyl Intermediate in the Electrocatalytic
Reduction of Carbon Dioxide by a Manganese Bipyridyl Derivative.
Angew. Chem., Int. Ed. 2014, 53, 240−243.
(17) Smieja, J. M.; Sampson, M. D.; Grice, K. A.; Benson, E. E.;
Froehlich, J. D.; Kubiak, C. P. Manganese as a Substitute for Rhenium
in CO2 Reduction Catalysts: the Importance of Acids. Inorg. Chem.
2013, 52, 2484−2491.
Osamu Ishitani: 0000-0001-9557-7854
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
■
This work was supported by JST CREST Grant Number
JPMJCR13L1. H.T. acknowledges the support from JSPS
KAKENHI grant number JP16K17876.
(18) Sampson, M. D.; Nguyen, A. D.; Grice, K. A.; Moore, C. E.;
Rheingold, A. L.; Kubiak, C. P. Manganese Catalysts with Bulky
Bipyridine Ligands for the Electrocatalytic Reduction of Carbon
Dioxide: Eliminating Dimerization and Altering Catalysis. J. Am.
Chem. Soc. 2014, 136, 5460−5471.
(19) Machan, C. W.; Stanton, C. J.; Vandezande, J. E.; Majetich, G.
F.; Schaefer, H. F.; Kubiak, C. P.; Agarwal, J. Electrocatalytic
Reduction of Carbon Dioxide by Mn(CN)(2,2′-bipyridine)(CO)3:
CN Coordination Alters Mechanism. Inorg. Chem. 2015, 54, 8849−
8856.
(20) Zeng, Q.; Tory, J.; Hartl, F. Electrocatalytic Reduction of
Carbon Dioxide with a Manganese(I) Tricarbonyl Complex
Containing a Nonaromatic α-Diimine Ligand. Organometallics 2014,
33, 5002−5008.
(21) Agarwal, J.; Stanton, C. J., III; Shaw, T. W.; Vandezande, J. E.;
Majetich, G. F.; Bocarsly, A. B.; Schaefer, H.F., III Exploring the Effect
of Axial Ligand Substitution (X = Br, NCS, CN) on the
Photodecomposition and Electrochemical Activity of [MnX(N−
C)(CO)3] Complexes. Dalton Trans. 2015, 44, 2122−2131.
(22) Ngo, K. T.; McKinnon, M.; Mahanti, B.; Narayanan, R.; Grills,
D. C.; Ertem, M. Z.; Rochford, J. Turning on the Protonation-First
Pathway for Electrocatalytic CO2 Reduction by Manganese Bipyridyl
Tricarbonyl Complexes. J. Am. Chem. Soc. 2017, 139, 2604−2618.
(23) (a) Franco, F.; Cometto, C.; Vallana, F. F.; Sordello, F.; Priola,
E.; Minero, C.; Nervi, C.; Gobetto, R. A Local Proton Source in a
[Mn(bpy-R)(CO)3Br]-type Redox Catalyst Enables CO2 Reduction
Even in the Absence of Brønsted Acids. Chem. Commun. 2014, 50,
14670−14673. (b) Franco, F.; Cometto, C.; Nencini, L.; Barolo, C.;
Sordello, F.; Minero, C.; Fiedler, J.; Robert, M.; Gobetto, R.; Nervi, C.
Local Proton Source in Electrocatalytic CO2 Reduction with
[Mn(bpy-R)(CO)3Br] Complexes. Chem. - Eur. J. 2017, 23, 4782−
4793.
REFERENCES
■
(1) Sato, S.; Arai, T.; Morikawa, T. Toward Solar-Driven
Photocatalytic CO2 Reduction Using Water as an Electron Donor.
Inorg. Chem. 2015, 54, 5105−5113.
(2) Sahara, G.; Kumagai, H.; Maeda, K.; Kaeffer, N.; Artero, V.;
Higashi, M.; Abe, R.; Ishitani, O. Photoelectrochemical Reduction of
CO2 Coupled to Water Oxidation Using a Photocathode with a
Ru(II)−Re(I) Complex Photocatalyst and a CoOx /TaON Photo-
anode. J. Am. Chem. Soc. 2016, 138, 14152−14158.
(3) Morimoto, T.; Nishiura, C.; Tanaka, M.; Rohacova, J.;
Nakagawa, Y.; Funada, Y.; Koike, K.; Yamamoto, Y.; Shishido, S.;
Kojima, T.; Saeki, T.; Ozeki, T.; Ishitani, O. Ring-shaped Re(I)
Multinuclear Complexes with Unique Photofunctional Properties. J.
Am. Chem. Soc. 2013, 135, 13266−13269.
(4) Tamaki, Y.; Ishitani, O. Supramolecular Photocatalysts for the
Reduction of CO2. ACS Catal. 2017, 7, 3394−3409.
(5) IPCC, 2014: Summary for Policymakers. Climate Change 2014:
Mitigation of Climate Change. Contribution of Working Group III to the
Fifth Assessment Report of the Intergovernmental Panel on Climate
Change; Edenhofer, O., Pichs-Madruga, R., Sokona, Y., Farahani, E.,
Kadner, S., Seyboth, K., Adler, A., Baum, I., Brunner, S., Eickemeier,
P., Kriemann, B., Savolainen, J., Schlomer, S., von Stechow, C.,
Zwickel, T., Minx, J. C., Eds.; Cambridge University Press:
Cambridge, United Kingdom, New York, 2014.
(6) Rumble, J. R. CRC Handbook of Chemistry and Physics, 98th ed.;
Taylor & Francis Group: Boca Raton, FL, 2017.
̈
(7) Matsuoka, S.; Yamamoto, K.; Ogata, T.; Kusaba, M.; Nakashima,
N.; Fujita, E.; Yanagida, S. Efficient and Selective Electron Mediation
of Cobalt Complexes with Cyclam and Related Macrocycles in the p-
Terphenyl-Catalyzed Photoreduction of Carbon Dioxide. J. Am.
Chem. Soc. 1993, 115, 601−609.
(8) Takeda, H.; Cometto, C.; Ishitani, O.; Robert, M. Electrons,
Photons, Protons and Earth-Abundant Metal Complexes for
Molecular Catalysis of CO2 Reduction. ACS Catal. 2017, 7, 70−88.
(9) Takeda, H.; Ohashi, K.; Sekine, A.; Ishitani, O. Photocatalytic
CO2 Reduction Using Cu(I) Photosensitizers with a Fe(II) Catalyst.
J. Am. Chem. Soc. 2016, 138, 4354−4357.
(24) Rawat, K. S.; Mahata, A.; Choudhuri, I.; Pathak, B. N-
Heterocylic Carbene-Based Mn Electrocatalyst for Two-Electron CO2
Reduction over Proton Reduction. J. Phys. Chem. C 2016, 120, 8821−
8831.
(25) Takeda, H.; Koizumi, H.; Okamoto, K.; Ishitani, O.
Photocatalytic CO2 Reduction Using a Mn Complex as a Catalyst.
Chem. Commun. 2014, 50, 1491−1493.
(10) Lazorski, M. S.; Castellano, F. N. Advances in the Light
Conversion Properties of Cu(I)-Based Photosensitizers. Polyhedron
2014, 82, 57−70.
(26) Cheung, P. L.; Machan, C. W.; Malkhasian, A. Y. S.; Agarwal, J.;
Kubiak, C. P. Photocatalytic Reduction of Carbon Dioxide to CO and
HCO2H Using fac-Mn(CN)(bpy)(CO)3. Inorg. Chem. 2016, 55,
3192−3198.
́
(11) Rosas-Hernandez, A.; Steinlechner, C.; Junge, H.; Beller, M.
Earth-abundant Photocatalytic Systems for the Visible-Light-Driven
Reduction of CO2 to CO. Green Chem. 2017, 19, 2356−2360.
́
(27) Stanbury, M.; Compain, J. D.; Trejo, M.; Smith, P.; Goure, E.;
́
̀
(12) Guo, Z.; Cheng, S.; Cometto, C.; Anxolabehere-Mallart, E.; Ng,
S. M.; Ko, C. C.; Liu, G.; Chen, L.; Robert, M.; Lau, T. C. Highly
Efficient and Selective Photocatalytic CO2 Reduction by Iron and
Cobalt Quaterpyridine Complexes. J. Am. Chem. Soc. 2016, 138,
9413−9416.
Chardon-Noblat, S. Electrochimica Acta Mn-Carbonyl Molecular
Catalysts Containing a Redox-Active Phenanthroline-5,6-Dione for
Selective Electro- and Photoreduction of CO2 to CO or HCOOH.
Electrochim. Acta 2017, 240, 288−299.
(28) Zhang, J. X.; Hu, C. Y.; Wang, W.; Wang, H.; Bian, Z. Y. Visible
Light Driven Reduction of CO2 Catalyzed by an Abundant
Manganese Catalyst with Zinc Porphyrin Photosensitizer. Appl.
Catal., A 2016, 522, 145−151.
(29) There should be other reactions of BIH during this
photocatalytic reaction condition, that is, high concentration of BIH
(0.1 M) because, in the case using 10 times lower concentration of
BIH, almost all of the consumed BIH was converted to two-electron
oxidation product, that is, BI+ (eq 1).
(13) Rao, H.; Schmidt, L. C.; Bonin, J.; Robert, M. Visible-Light-
Driven Methane Formation from CO2 with a Molecular Iron Catalyst.
Nature 2017, 548, 74−77.
(14) Bonin, J.; Robert, M.; Routier, M. Selective and Efficient
Photocatalytic CO2 Reduction to CO Using Visible Light and an
Iron-Based Homogeneous Catalyst. J. Am. Chem. Soc. 2014, 136,
16768−16771.
(15) Bourrez, M.; Molton, F.; Chardon-Noblat, S.; Deronzier, A.
[Mn(bipyridyl)(CO)3Br]: An Abundant Metal Carbonyl Complex as
M
DOI: 10.1021/jacs.8b10619
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX

Journal of the American Chemical Society
Article
(30) Ruthkosky, M.; Kelly, C. A.; Castellano, F. N.; Meyer, G. J.
Electron and Energy Transfer from CuI MLCT Excited States. Coord.
Chem. Rev. 1998, 171, 309−322.
(31) McMillin, D. R.; Kirchhoff, J. R.; Goodwin, K. V. Exciplex
Quenching of Photo-Excited Copper Complexes. Coord. Chem. Rev.
1985, 64, 83−92.
(32) Czerwieniec, R.; Leitl, M. J.; Homeier, H. H. H.; Yersin, H.
Cu(I) Complexes − Thermally Activated Delayed Fluorescence.
Photophysical Approach and Material Design. Coord. Chem. Rev.
2016, 325, 2−28.
(33) (a) Siddique, Z. A.; Yamamoto, Y.; Ohno, T.; Nozaki, K.
Structure-Dependent Photophysical Properties of Singlet and Triplet
Metal-to-Ligand Charge Transfer States in Copper(I) Bis(Diimine)-
Compounds. Inorg. Chem. 2003, 42, 6366−6378. (b) Iwamura, M.;
Watanabe, H.; Ishii, K.; Takeuchi, S.; Tahara, T. Coherent Nuclear
Dynamics in Ultrafast Photoinduced Structural Change of Bis-
(Diimine)Copper(I) Complex. J. Am. Chem. Soc. 2011, 133, 7728−
7736. (c) Garakyaraghi, S.; Danilov, E. O.; McCusker, C. E.;
Castellano, F. N. Transient Absorption Dynamics of Sterically
Congested Cu(I) MLCT Excited States. J. Phys. Chem. A 2015,
119, 3181−3193. (d) McCusker, C. E.; Castellano, F. N. Design of a
Long-Lifetime, Earth-Abundant, Aqueous Compatible Cu(I) Photo-
sensitizer Using Cooperative Steric Effects. Inorg. Chem. 2013, 52,
8114−8120.
̌
(34) Vlcek, A., Jr.; Farrell, I. R.; Liard, D. J.; Matousek, P.; Towrie,
M.; Parker, A. W.; Grills, D. C.; George, M. W. Early Photochemical
Dynamics of Organometallic Compounds Studied by Ultrafast Time-
Resolved Spectroscopic Techniques. Dalton Trans. 2002, 701−712.
(35) Tamaki, Y.; Koike, K.; Morimoto, T.; Ishitani, O. Substantial
Improvement in the Efficiency and Durability of a Photocatalyst for
Carbon Dioxide Reduction Using a Benzoimidazole Derivative as an
Electron Donor. J. Catal. 2013, 304, 22−28.
(36) Zhu, X. Q.; Zhang, M. T.; Yu, A.; Wang, C. H.; Cheng, J. P.
Hydride, Hydrogen Atom, Proton, and Electron Transfer Driving
Forces of Various Five-Membered Heterocyclic Organic Hydrides and
their Reaction Intermediates in Acetonitrile. J. Am. Chem. Soc. 2008,
130, 2501−2516.
(37) (a) Czerwieniec, R.; Kowalski, K.; Yersin, H. Highly Efficient
Thermally Activated Fluorescence of a New Rigid Cu(I) Complex
[Cu(dmp)(phanephos)]+. Dalton Trans. 2013, 42, 9826−9830.
̈
(b) Bergmann, L.; Hedley, G. J.; Baumann, T.; Brase, S.; Samuel, I.
D. W. Direct Observation of Intersystem Crossing in a Thermally
Activated Delayed Fluorescence Copper Complex in the Solid State.
Sci. Adv. 2016, 2, e1500889.
(38) Hofbeck, T.; Monkowius, U.; Yersin, H. Highly Efficient
Luminescence of Cu(I) Compounds: Thermally Activated Delayed
Fluorescence Combined with Short-Lived Phosphorescence. J. Am.
Chem. Soc. 2015, 137, 399−404.
(39) Hartl, F.; Rossenaar, B. D.; Stor, G. J.; Stufkens, D. J. Role of an
Electron-Transfer Chain Reaction in the Unusual Photochemical
Formation of Five-Coordinated Anions [Mn(CO)3(α-diimine)]−
from fac-[Mn(X)(CO)3(α-diimine)] (X = halide) at Low Temper-
atures. Recl. des Trav. Chim. des Pays-Bas 1995, 114, 565−570.
(40) Takeda, H.; Yamamoto, Y.; Nishiura, C.; Ishitani, O. Analysis
and Isolation of Cationic Rhenium(I) and Ruthenium(II) Multi-
nuclear Complexes Using Size-Exclusion Chromatography. Anal. Sci.
2006, 22, 545−549.
(41) Ishitani, O.; George, M. W.; Ibusuki, T.; Johnson, F. P. A.;
Koike, K.; Nozaki, K.; Pac, C.; Turner, J. J.; Westwell, J. R.
Photophysical Behavior of a New CO2 Reduction Catalyst, Re-
+
(CO)2(bpy){P(OEt)3}2 . Inorg. Chem. 1994, 33, 4712−4717.
(42) Ohno, T.; Nozaki, K.; Haga, M. Photoexcited States of
Biruthenium(II) Compounds Bridged by 2,2′-Bis(2-Pyridyl)-
Bibenzimidazole or 1,2-Bis(2-(2-Pyridyl)Benzimidazolyl)Ethane.
Inorg. Chem. 1992, 31, 548−555.
N
DOI: 10.1021/jacs.8b10619
J. Am. Chem. Soc. XXXX, XXX, XXX−XXX