Photocatalytic Reduction of Carbon Dioxide to CO and HCO2H Using fac-Mn(CN)(bpy)(CO)3

Article abstract of DOI:10.1021/acs.inorgchem.6b00379

Studies are reported regarding the use of Mn(CN)(bpy)(CO)₃ (1) as a catalyst for CO₂ reduction employing [Ru(dmb)₃]²⁺ as a photosensitizer in mixtures of dry N,N-dimethylformamide-triethanolamine (N,N-DMF-TEOA) or acetonitrile-TEOA (MeCN-TEOA) with 1-benzyl-1,4-dihydronicotinamide as a sacrificial reductant. Irradiation with 470 nm light for up to 15 h yields both CO and HCO₂H with maximum turnover numbers (TONs) as high as 21 and 127, respectively, with product preference dependent on the solvent. Further data suggests that upon single electron reduction this catalyst avoids the formation of a Mn-Mn dimer and instead undergoes a disproportionation reaction, which requires 2 equiv of [Mn(CN)(bpy)(CO)₃]^?- to generate 1 equiv each of the active catalyst [Mn(bpy)(CO)₃]^- and the starting compound 1. Additional characterization by cyclic voltammetry (CV) and infrared spectroelectrochemistry (IR-SEC) indicates that the stability of the singly reduced [Mn(CN)(bpy)(CO)₃]^?- differs slightly in the N,N-DMF-TEOA solvent system compared to the MeCN-TEOA system. This contributes to the observed selectivities for HCO₂H vs CO production.

Full text of DOI:10.1021/acs.inorgchem.6b00379

Article
pubs.acs.org/IC
Photocatalytic Reduction of Carbon Dioxide to CO and HCO2H Using
fac-Mn(CN)(bpy)(CO)3
Po Ling Cheung,† Charles W. Machan,† Aramice Y. S. Malkhasian,‡ Jay Agarwal,§
and Clifford P. Kubiak*,†
†Department of Chemistry and Biochemistry, University of California, San Diego, California 92093, United States
‡Chemistry Department, King Abdulaziz University, Jeddah, 21589, Saudi Arabia
§Center for Computational Quantum Chemistry, University of Georgia, Athens, Georgia 30602, United States
S
* Supporting Information
ABSTRACT: Studies are reported regarding the use of Mn(CN)(bpy)(CO)3 (1) as a catalyst for CO2 reduction employing
[Ru(dmb)3]2+ as a photosensitizer in mixtures of dry N,N-dimethylformamide-triethanolamine (N,N-DMF-TEOA) or
acetonitrile-TEOA (MeCN-TEOA) with 1-benzyl-1,4-dihydronicotinamide as a sacrificial reductant. Irradiation with 470 nm
light for up to 15 h yields both CO and HCO2H with maximum turnover numbers (TONs) as high as 21 and 127, respectively,
with product preference dependent on the solvent. Further data suggests that upon single electron reduction this catalyst avoids
the formation of a Mn−Mn dimer and instead undergoes a disproportionation reaction, which requires 2 equiv of
[Mn(CN)(bpy)(CO)3]•− to generate 1 equiv each of the active catalyst [Mn(bpy)(CO)3]− and the starting compound 1.
Additional characterization by cyclic voltammetry (CV) and infrared spectroelectrochemistry (IR-SEC) indicates that the
stability of the singly reduced [Mn(CN)(bpy)(CO)3]•− differs slightly in the N,N-DMF-TEOA solvent system compared to the
MeCN-TEOA system. This contributes to the observed selectivities for HCO2H vs CO production.
focus of the present work, we note that there is a
complementary focus on the development of photosensitizers
that do not use heavy metals for similar reasons.12−14
INTRODUCTION
■
Global warming from anthropogenic greenhouse gas emissions
and the decreasing availability of fossil fuels continues to
highlight the potential for photocatalytic conversion of carbon
dioxide (CO2) to generate chemicals or fuel precursors, such as
synthesis gas (CO/H2). These products can be converted into
liquid fuels through the industrial Fischer−Tropsch reac-
tion.1−3 Therefore, the development of artificial photosynthetic
systems to harness renewable energy for this purpose is
consequently important. Molecular catalysts with Ru- and Re-
carbonyl cores and ligands based on 2,2′-bipyridine (bpy) have
been investigated extensively for their photocatalytic perform-
ance in this context.4−9 Complexes of this type have been used
to convert CO2 to carbon monoxide (CO) and/or formic acid
(HCO2H) using a photosensitizer to initiate photochemical
electron transfer along with a sacrificial reductant to donate
electrons and sustain catalysis. However, the cost of Ru and Re
precludes their use on a large scale, underscoring the need for
earth-abundant alternatives, e.g., Mn.10,11 Although it is not the
Prior reports have focused on the electrochemical15 and
photosensitizer-driven16,17 reactions of fac-MnBr(bpy)(CO)3
(the fac label will be omitted from here on). During
electrocatalysis, MnBr(bpy)(CO)3 is first reduced by a single
electron, after which the Br− ligand dissociates and a metal−
metal bonded dimer, [Mn(bpy)(CO)3]2, is formed.18,19 Further
reduction of the dimer with two electrons yields 2 equiv of the
proposed active species, [Mn(bpy)(CO)3]−; this reduced
species can react with CO2 in the presence of added Brønsted
acids to produce CO in nonaqueous media.18,20,21 The
intermediate steps are proposed to involve a hydroxycarbonyl
species, [Mn(η1-COOH)(bpy)(CO)3], which forms from the
initial reaction of the [Mn(bpy)(CO)3]− anion with CO2/H+
Received: February 15, 2016
© XXXX American Chemical Society
A
DOI: 10.1021/acs.inorgchem.6b00379
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
a
Table 1. Summary of Catalytic Experiments for Photocatalysis
entry
Mn(CN)(bpy)(CO)3/mM
[Ru(dmb)3]2+/mM
Catalytic Experiments in N,N-DMF-TEOA
irradation time/h
HCO2H TON
CO TON
H2 TON
HCO2H Φ
CO Φ
1
2
3
4
5
6
0.10
0.50
1.00
0.10
0.50
1.00
0.10
0.50
1.00
0.10
0.50
1.00
15
15
15
6
36
21
3.9
0.64
0.45
0.37
0.20
0.25
0.0073
0.0096
0.031
0.027
0.011
0.039
0.033
0.0010
0.0047
0.0066
0.0018
0.0033
0.0053
3.2
9.5
2.3
16
11
2.8
6
0.92
0.83
6
5.2
Catalytic Experiments in MeCN-TEOA
7
0.10
0.50
1.00
0.10
0.50
1.00
0.10
0.50
1.00
0.10
0.50
1.00
15
15
15
6
4.5
7.9
5.9
3.2
4.7
3.4
3.1
0.56
0.24
0.17
0.27
0.14
0.14
0.0011
0.0027
0.0029
0.00
0.0020
0.0069
0.0080
0.0032
0.011
8
2.1
9
1.2
10
11
12
0.00
0.00
0.00
6
0.00
6
0.00
0.021
Variable Concentration Studies in N,N-DMF-TEOA
13
14
15
0.10
0.10
1.00
0.50
1.00
0.50
15
15
15
130
130
9.1
7.1
3.0
1.2
0.026
0.032
0.014
0.0018
0.0018
0.0070
1.6
5.8
0.24
Variable Concentration Studies in MeCN-TEOA
16
17
18
0.10
0.10
1.00
0.50
1.00
0.50
15
15
15
8.1
9.0
1.4
19
21
4.2
1.4
0.0021
0.0022
0.0032
0.0048
0.0053
0.0096
1.3
0.18
a
All runs were sparged with CO2 for 30 min prior irradiation with 470 nm monochromatic light for 15 h.
a
Table 2. Summary of Control Experiments for Photocatalysis
entry
Mn(CN)(bpy)(CO)3/mM
[Ru(dmb)3]2+/mM
sparging gas
HCO2H TON
CO TON
H2 TON
HCO2H Φ
CO Φ
Control Reactions in N,N-DMF
19
20
0.50
0
0
CO2
CO2
CO2
Ar
1.9
2.1
1.5
5.8
0.00
2.8
0.0022
0.0069
0.00
0.0025
0.0018
0.0082
0.50
0.50
0.50
0.50
0.50
0
5.7
b
21
0.50
0.50
0.50
0
0.00
2.6
0.26
c
22
23
24
25
Ar
3.8
1.5
7.8
0.0047
0.0040
0.0030
0.00
0.0018
0.0011
0.0030
0.0021
Ar
3.2
0.87
2.3
1.3
0.50
0.50
Ar
2.2
0.00
1.4
b
26
0.50
Ar
0.00
2.4
Control Reactions in MeCN
27
28
0.50
0
0
CO2
CO2
CO2
Ar
0.00
0.72
0.00
0.55
0.00
2.0
0.60
2.2
2.2
2.4
0.00
1.2
0.00
0.0024
0.00078
0.0025
0.0026
0.0021
0.50
0.50
0.50
0.50
0.00093
0.00
b
29
0.50
0.50
0.50
0.00
4.4
30
0.00066
0.00
b
31
Ar
1.4
a
b
c
All runs were irradiated with 470 nm monochromatic light for 15 h. TEOA is not used in the solvent system. Irradiation time = 0 h.
and requires an additional reduction to achieve the highest rate
of catalysis.20,22 Consistent with the electrochemical mecha-
nism detailed above, Takeda et al.16 found that, as a catalyst,
MnBr(bpy)(CO)3 forms the [Mn(bpy)(CO)3]2 dimer rapidly
following a photoinduced one-electron reduction by a photo-
sensitized [Ru(dmb)3]2+ complex. Time-resolved experiments
competent catalyst for the electrocatalytic reduction of CO2
with high Faradaic yield for CO (98
3%).26 With the
substitution of the Br− ligand for the pseudohalogen cyanide
(CN−), the mechanism of catalytic CO2 reduction was altered
from that described above. Specifically, CN− did not readily
dissociate like Br− following a one-electron reduction,
mitigating the formation of the Mn−Mn dimer species as an
intermediate to the active state, [Mn(bpy)(CO)3]−, which was
instead generated after a disproportionation reaction. As a
result, we were interested in investigating the photocatalytic
properties of 1 to see if the alternate reduction mechanism
would also impact the distribution of two-electron products
(H2, CO, and HCO2H).16
estimate the dimerization rate constant to be rapid, with 2kdim
1.3
=
0.1 × 109 M−1 s−1.19 The authors16 suggested a
monomeric, five-coordinate manganese radical species [Mn-
(bpy)(CO)3]• results from photoinduced homolytic dimer
cleavage; this radical is also proposed to be the active catalyst
for reduction of CO2, with HCO2H as the major product.
Analogous Mn radical species have been previously detected as
a product of the photoassisted cleavage of Mn−Mn bonds.23−25
In a recent report we described the electrocatalytic behavior
of Mn(CN)(bpy)(CO)3 (1), which was found to be a
Herein, the use of 1 as part of a photocatalytic system for the
reduction of CO2 is reported. The results suggest that, similar
to the previously described electrochemical mechanism, 1
B
DOI: 10.1021/acs.inorgchem.6b00379
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Scheme 1. Proposed Mechanisms for the Formation of HCO2H and/or CO from the Photocatalytic Reaction with
16,17
Mn(CN)(bpy)(CO)3 (1, top) and MnBr(bpy)(CO)3
(bottom)
undergoes a disproportionation reaction involving two singly
reduced species, [Mn(CN)(bpy)(CO)3]•−, to produce 1 equiv
of the catalytically active species [Mn(bpy)(CO)3]− and 1
equiv of 1. Indeed, under a variety of experimental conditions 1
is capable of photocatalytically converting CO2 to both CO and
HCO2H, although at lower rates than previously reported for
MnBr(bpy)(CO)3. Interestingly, the preferred catalytic product
shifts from CO to HCO2H when changing the solvent from
MeCN to N,N-DMF. Supplementary electrochemical and
infrared spectroelectrochemical studies support the dispropor-
tionation mechanism, and indicate that the N,N-DMF-TEOA
mixture is noninnocent with respect to the production of
HCO2H. This is noteworthy because many studies have
focused on the electro- and photocatalytic performance of
molecular catalysts related to MnX(bpy)(CO)3,4−9,16,17 but few
studies investigate the role of solvent.8,27,28 Indeed, some
reports note that N,N-DMF can be hydrolyzed to produce
formate either during photocatalysis or during analysis if
significant amounts of water are present.27,28
(TEOA) (4:1 v/v) or N,N-dimethylformamide (N,N-DMF)
and TEOA (4:1 v/v) that contained 1, [Ru(dmb)3]2+ as a
photosensitizer, and 1-benzyl-1,4-dihydronicotinamide (BNAH,
0.1 M) as a sacrificial reductant. After irradiation at 470 nm
with an LED laser, the gaseous products (CO, H2) were
quantified by GC and liquid products (HCO2H) with 1H NMR
(see Experimental Methods). A key finding is that the main
product of the photochemical reaction was HCO2H in the N,N-
DMF-TEOA solvent system, while CO was the primary
product in MeCN-TEOA, vide infra.
In the following sections we highlight two conclusions that
are supported by these experimental results: (i) complex 1
follows a disproportionation mechanism under photocatalytic
conditions and (ii) varying the solvent system from N,N-DMF-
TEOA to MeCN-TEOA affects the product distribution
between HCO2H and CO. We discuss the origin of these
conclusions and introduce further experimental results in
support.
Disproportionation Mechanism. In Scheme 1, we
present a proposed mechanism for the photocatalytic reduction
of CO2 by 1 based on the photocatalytic results presented
above and additional experiments detailed below. In the
photosensitization step, the reduced [Ru(dmb)3]•+ ([Ru-
(dmb)3]2+/•+ = −1.85 V vs Fc/Fc+) transfers an electron to
the ligand of Mn complex 1, forming a [Mn(CN)(bpy)-
(CO)3]•− species, 2 [eq 1 in Scheme 2; [Mn(CN)(bpy)-
(CO)3]0/•− = −1.91 V vs Fc/Fc+] {[Ru(dmb)3]2+*/+ = 0.24 V;
RESULTS AND DISCUSSION
■
Photocatalytic Experiments. Thirty-one photocatalytic
experiments with different reaction conditions were conducted,
and the results are summarized in Tables 1 and 2. In a typical
photocatalytic experimentadapted from a previous re-
port17a quartz cell of known volume was charged with
solvent mixtures of acetonitrile (MeCN) and triethanolamine
C
DOI: 10.1021/acs.inorgchem.6b00379
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
(CO)3]− and 1. At high concentrations of 1 and [Ru(dmb)3]2+,
mass transport limitations would be expected to affect this
process, and limit the overall efficiency. This can be shown in a
simplified kinetic description of the overall reaction as shown in
Scheme 2 above. Inhibition by increasing concentration of 1 is
an expected consequence of the disproportionation reaction in
eq 2, if it is assumed to be the rate-limiting step for the
formation of 3 and the reaction of 3 with CO2 is assumed to be
fast relative to k−2.
Scheme 2. Simplified Description of the Elementary Steps in
Photochemical Reduction of CO2 by 1
This mechanism is further validated by the observation that
the TON increases when the ratio of Ru photosensitizer to Mn
complex (entries 1, 7, 13, 14, 16 and 17) is increased. The
observed increase in catalysis levels off after the ratio of
[Ru(dmb)3]2+: 1 reaches 5:1. At this ratio, relative to
experiments with equimolar amounts of photosensitizer and
catalyst (entries 1−12), there is a 3.6× enhancement in the
HCO2H-TON and 2.3× in the CO-TON in the N,N-DMF-
TEOA system. Likewise, the MeCN-TEOA system shows a
1.8× increase in the HCO2H-TON and 2.4× in the CO-TON.
As shown in eq 1 of Scheme 2, an increased amount of
photosensitized Ru (Ru*) relative to 1 in solution could
generate more equivalents of the singly reduced species
[Mn(CN)(bpy)(CO)3]•− 2 (Scheme 2). Therefore, more
equivalents of the active anion [Mn(bpy)(CO)3]− 3, would
be formed in eq 2, resulting in an increase in catalyst TONs [eq
3]. After the optimized ratio is reached, the concentration of
photosensitizer is no longer a limiting factor and the observed
TONs do not change at higher ratios of [Ru(dmb)3]2+: 1.
Lastly, photodecomposition (mole equiv of CO ≈ 2 relative
to [1]) in both the MeCN-TEOA and N,N-DMF-TEOA
solvent systems was observed when there was no [Ru-
(dmb)3]2+, CO2, or TEOA present (entries 19−31 in Table
2). These experiments also showed that [Ru(dmb)3]2+
contributed to H2, HCO2H, and CO production, although in
a much lower quantity than 1 (entries 20, 24, and 28).16,17,44
Solvent Effects. The observation that the photocatalytic
system is more selective for CO in MeCN-TEOA than in N,N-
DMF-TEOA (which produces more HCO2H) suggests the
possibility of solvent interactions during the photochemical
mechanism. In control reactions with the N,N-DMF-TEOA
solvent system (entries 19−20, 22−25), we note that there is
always a small amount of residual HCO2H which is
independent of photocatalysis. The lack of residual HCO2H
in the absence of TEOA (entries 21 and 26) suggests the
TEOA is vital for the production of HCO2H in N,N-DMF.
Furthermore, in the case of entry 22 (a control workup of entry
23 before any irradiation occurred), HCO2H was detected by
1H NMR immediately after mixing the solution in the dark. No
HCO2H was found in the MeCN-TEOA system in analogous
experiments. Residual HCO2H is only observed from the
interaction between N,N-DMF and TEOA; this observation is
consistent with some recent reports: Vos27 and Ishida28
pointed out that N,N-DMF hydrolyzes spontaneously to give
HCO2H in the presence of water or TEOA. A plot (Figure S1)
of HCO2H-TON vs the concentration of complex 1 shows high
linearity for both the 15 and 6 h irradiation times in the N,N-
DMF-TEOA solvent system, suggesting that the amount of
HCO2H formed from N,N-DMF and TEOA interaction is
insignificant under catalytic conditions. Solvent decomposition
alone does not explain the large difference in product
distribution between the two solvent systems.
BNAH0/•+ = 0.20 V vs Fc/Fc+}.26,29−32 The reduced
photosensitizer [Ru(dmb)3]•+ is capable of driving catalysis at
this potential, but at a diminished rate. Indeed, in comparison
to MnBr(bpy)(CO)3 much lower rates are observed.16 Two
equivalents of 2 can undergo disproportionation to generate
the active [Mn(bpy)(CO)3]− anion, 3, and the starting material
1 [Eqn (2) in Scheme 2]. It is presumed that HCO2H
formation in this mechanism results from the protonation of
species 3 (by TEOA, TEOA•+, or TEOAH+) to generate a Mn
hydride, an expected intermediate before CO2 insertion, during
catalysis.17,33−37 A preference for HCO2H is observed in N,N-
DMF-TEOA solution, indicating that the formation of an
intermediate hydride species may be favored relative to
MeCN.17 It is also possible for species 3 to form a
hydroxycarbonyl species (preferred in MeCN-TEOA) through
a direct attack of CO2 with subsequent protonation by TEOA
or TEOAH+ (generated by the deprotonation of BNAH•+, the
expected sacrificial coproduct from photosensitization).38,39
The resulting η1-COOH coordination mode generally yields
CO as a product.40−42 We note another formate formation
pathway. It is possible that species 2 may abstract a hydrogen
atom from TEOA or BNAH•+ to generate a hydride species
(with presumptive CN− loss) and that the photosensitizer is
observed to generate formate in control reactions by an
ostensibly analogous mechanism.43 This presumptive hydride
species generated by 2 could also then react with CO2 to give
HCO2H. Such reaction pathways are expected to be in
competition with the disproportionation mechanism and
eventual catalysis by 3.
The proposed mechanism for the photocatalytic reduction of
CO2 by MnBr(bpy)(CO)3 based on the previous literature16,17
is also presented for comparison (Scheme 1). In this reaction
scheme, the formation of a Mn−Mn bonded dimer is an
intermediate step to the formation of a neutral five-coordinate
radical species from photoassisted bond cleavage. It is likely at
this point in the reaction that the photosensitizer provides the
additional electron required to complete the reduction of CO2.
The redox potential for [Mn(bpy)(CO)3]•/− has not been
directly measured, but when using the structural analogue
MnBr(6,6′-dimesityl-2,2′-bipyridine)(CO)3, the first reduction
observed by cyclic voltammetry becomes a two electron process
at −1.55 V vs Fc/Fc+ through the prevention of dimerization,
suggesting this can be considered the approximate maximum
potential required to generate the anion.22
In both the N,N-DMF-TEOA and MeCN-TEOA solvent
systems, the turnover number (TON) for HCO2H, CO, and
H2 decreased as the concentration of the Mn complex increased
(entries 2, 8, 13, 15, 16, and 18). This observation is consistent
with the proposed mechanism of photocatalytic CO2 reduction
involving an intermediate disproportionation step, vide infra.26
Two equivalents of 1 are required to be reduced by one
electron each from [Ru(dmb)3]•+ before a disproportionation
reaction yields 1 equiv each of the active catalyst [Mn(bpy)-
To further understand the role of solvent in product
selectivity, cyclic voltammetry (CV) was performed both
D
DOI: 10.1021/acs.inorgchem.6b00379
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
under Ar and CO2 saturation. In Figure 1, scans were
performed with 1 (1 mM) and tetrabutylammonium
Figure 3. Cyclic voltammetry of Mn(CN)(bpy)(CO)3 1 under Ar
saturation (black), in the presence of TEOA (blue), and under CO2
saturation (red) in dry MeCN at 100 mV s−1. Conditions: 1 mM 1 in
0.1 TBAPF6/MeCN; glassy carbon working electrode, Pt wire counter
electrode, Ag/AgCl pseudoreference electrode; referenced to internal
Fc standard.
Figure 1. Cyclic voltammetry of Mn(CN)(bpy)(CO)3 1 under Ar
saturation. Conditions: 1 mM 1 in 0.1 TBAPF6/N,N-DMF; glassy
carbon working electrode, Pt wire counter electrode, Ag/AgCl
pseudoreference electrode; referenced to internal Fc standard.
DMF-TEOA and MeCN-TEOA solvent systems by IR
spectroscopy. IR-SEC allows the starting, intermediate, and
product species in solution to be characterized as a function of
potential and time.18,26 When the potential of the IR-SEC cell is
shifted stepwise from resting to that of the first reduction in
MeCN with added TEOA and 1 (∼−1.9 V vs Fc/Fc+), two
new bands appear at 1911 and 1810 cm−1 with the concomitant
disappearance of the original bands (Figure 4a), which are
hexafluorophosphate (TBAPF6; 0.1 M) as the supporting
electrolyte in N,N-DMF under argon (Ar) at scan rates from
100 to 2000 mV/s. The first one-electron, quasireversible redox
feature is observed at −1.95 V vs Fc/Fc+ (Epc) with peak-to-
peak separation (ΔEp) of 74 mV. The process is diffusion
limited as indicated by the linear relationship between the
square root of the scan rate and the peak current (Figure S2).
As shown in Figure 2, the addition of TEOA to the solution
Figure 4. Infrared spectra of Mn(CN)(bpy)(CO)3 (1) at controlled
potentials (a) in dry MeCN with TEOA under N2, and (b) in dry
MeCN with TEOA under CO2. Conditions: 4 mM 1 in 0.1 TBAPF6/
MeCN; glassy carbon working electrode, Pt counter electrode, Ag
pseudoreference electrode.
Figure 2. Cyclic voltammetry of Mn(CN)(bpy)(CO)3 1 under Ar
saturation (black), in the presence of TEOA (blue), and under CO2
saturation (red) in dry N,N-DMF at 100 mV s−1. Conditions: 1 mM 1
in 0.1 TBAPF6/N,N-DMF; glassy carbon working electrode, Pt wire
counter electrode, Ag/AgCl pseudoreference electrode; referenced to
internal Fc standard.
assigned to [Mn(bpy)(CO)3]−. When the MeCN-TEOA
solution of 1 is sparged briefly with CO2 and this experiment
is repeated, a band for CO2 is observed at 2340 cm−1. At the
potential of the first reduction, this band decreases in intensity,
consistent with catalytic consumption of CO2 (Figure S3).
Similar to the case when TEOA is present under N2 saturation
conditions, the only Mn carbonyl bands observed are at 1911
and 1810 cm−1, consistent with the formation of the active
catalyst species [Mn(bpy)(CO)3]−.
By comparison, conducting the same experiments in N,N-
DMF showed a similar redshift of the Mn carbonyl stretching
modes to the active catalyst at the potential of the first
reduction, albeit at a slower rate on the experimental time scale
(∼5 min, Figure 5a). The singly reduced species, [Mn(CN)-
(bpy)(CO)3]•−, is more stable under these conditions relative
to the results in MeCN. The IR bands visible at resting
potential (2026, 1936, and 1927 cm−1) almost completely shift
to lower frequencies (2003, 1909, and 1893 cm−1) when the
cell potential is changed from resting to that of the first
reduction (∼−1.9 V vs Fc/Fc+; ∼1 min). Previous
computations predicted that in MeCN [Mn(CN)(bpy)-
(CO)3]•− would have observable IR modes at 2010, 1909,
and 1895 cm−1; the high frequency mode was experimentally
under Ar showed a slight current attenuation (0.7×) of the
peak current of the first reduction wave (ca. −1.95 V vs Fc/
Fc+), which is attributed to dilution. When CO2 is present, the
peak current increased to 2.1× that observed under Ar. The
current enhancement is consistent with the catalytic reduction
of CO2 by a [Mn(bpy)(CO)3]− active species.26
Interestingly, when TEOA was added to dry MeCN, the
redox feature became less reversible and the peak current
showed a slight enhancement (2×) under Ar (Figure 3). Such
irreversibility is not observed in the absence of TEOA.26 The
presence of CO2 increased both the irreversibility of the wave
and the peak current, suggesting that the interaction of MeCN
with TEOA and the reduced species is relevant at this potential,
but not for N,N-DMF. This provides important clues for the
different product distributions between two solvent systems
observed during photocatalysis.
Infrared spectroelectrochemistry (IR-SEC) was used to
monitor the microscale electrolysis of complex 1 in the N,N-
E
DOI: 10.1021/acs.inorgchem.6b00379
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
Figure 5. Infrared spectra of Mn(CN)(bpy)(CO)3 (1) at controlled potentials (a) in dry N,N-DMF under N2, (b) in dry N,N-DMF with TEOA
under N2, and (c) in dry N,N-DMF with TEOA under CO2. Conditions: 4 mM 1 in 0.1 TBAPF6/N,N-DMF; glassy carbon working electrode, Pt
counter electrode, Ag pseudoreference electrode.
observed at 2007 cm−1.26 The presence of TEOA and CO2
300 Bio UV-Vis spectrophotometer with a quartz cuvette (1 cm path
decreases the stability of [Mn(CN)(bpy)(CO)3]•−, as
length).
evidenced by a more rapid shift to [Mn(bpy)(CO)3]− on this
Photochemical Reactions. A solution of acetonitrile (MeCN)
and triethanolamine (TEOA) (∼20 mL; 4:1 v/v) or N,N-dimethyl-
time scale (1909 and 1811 cm−1; ∼5 min). In comparison to
formamide (N,N-DMF) and TEOA (∼20 mL; 4:1 v/v), which
MeCN-TEOA, however, a more significant amount (as
contained Mn(CN)(bpy)(CO)3, [Ru(dmb)3]2+ as a photosensitizer,
estimated by IR intensity) of the singly reduced species is
and 1-benzyl-1,4-dihydronicotinamide (BNAH, 0.1 M) as a sacrificial
reductant was prepared in a 36 mL quartz cell (NSG Precision Cell,
Inc.; path length = 2 cm) sealed with a rubber septum. In the
still observed under these conditions.17 This difference in the
relative stability of [Mn(CN)(bpy)(CO)3]•− in the MeCN-
and N,N-DMF-based solvent systems could explain the
observed product selectivities. The higher stability of species
2 observed in N,N-DMF would allow more time for competing
hydrogen atom abstraction pathways involving TEOA or
BNAH•+ to generate HCO2H via a Mn hydride species.
concentration study, solutions containing 0.1, 0.5, and 1.0 mM of both
Mn(CN)(bpy)(CO)3 and [Ru(dmb)3]2+ were prepared and meas-
ured. The solution was sparged with dry Ar or CO2 gases for 30 min
prior to irradiation. The solution was irradiated with a 470 nm LED
(ThorLabs, Inc.; bandwidth fwhm = 25 nm) with temperature
maintained at 25.0
experiment. The light intensity was measured using a NOVA II power
0.1 °C, and constant stirring throughout the
CONCLUSION
meter.
■
Product Analysis from Photocatalysis.17 The gaseous products,
H2 and CO, were analyzed by GC-TCD (Hewlett-Packard 7890A
Series gas chromatograph) with two molsieve columns (30 m × 0.53
mm × 25 μm film). The 1 mL injection was split between two
columns, which use N2 and He as carrier gases to measure the quantity
of H2 and CO, respectively. Turnover numbers (TONs) for the
product were calculated as the moles of product divided by the moles
of catalyst.
Using [Ru(dmb)3]2+ as a photosensitizer in dry N,N-DMF-
TEOA or MeCN-TEOA solvent mixtures with Mn(CN)(bpy)-
(CO)3 as a catalyst results in photocatalytic CO2 reduction
through an intermediate disproportionation mechanism. The
formation of HCO2H and CO from CO2 decreases with
increasing concentration of the Mn complex from 0.1 to 1.0
mM, while the formation of products increases with a higher
ratio of [Ru(dmb)3]2+:1. Control reactions have also
demonstrated that the N,N-DMF solvent is not innocent in
the formation of HCO2H in the presence of TEOA. Even in the
dark, N,N-DMF and TEOA generate a measurable amount of
HCO2H. Supplemental CV and IR-SEC studies indicate that
although the reduction mechanism in MeCN and N,N-DMF is
similar, there is an observable difference in stability of the
[Mn(CN)(bpy)(CO)3]•−. It is clear from these studies that the
process of photosensitization by [Ru(dmb)3]2+ with BNAH as a
sacrificial reductant and the role of TEOA are mechanistically
complex and require further investigation to rationally optimize
the overall reaction.
The liquid products were analyzed by a previously reported 1H
NMR (400 MHz Mercury NMR Instrument) method:10 a known
amount of ferrocene (5−8 mg) was added to the irradiated solution in
a 5.0 mL volumetric flask to serve as an internal standard, and the
solution was sonicated for 10 min. A 1.0 mL syringe was used to
transfer a 0.8 mL aliquot of the resulting solution to a 2.0 mL
volumetric flask containing 0.1 mmol of Verkade’s base (2,8,9-
triisopropyl-2,5,8,9-tetraaza-1-phosphabicyclo[3,3,3]undecane).
CD3CN was added to the 2.0 mL mark and the resulting solution was
sonicated for another 10 min. The solution was divided into three
NMR samples, and each one was run for 128 scans on a Mercury 400
MHz spectrometer at 298 K. The whole workup process was carried
out in the dark.
Cyclic Voltammetry (CV). CV studies were performed using a
BASi Epsilon potentiostat. A 3 mm diameter glassy carbon working
electrode (BASi), a platinum (Pt) wire couter electrode, and a silver/
silver chloride (Ag/AgCl) wire separated from the bulk solution by a
CoralPor tip as a pseudoreference electrode were used in a single
compartment cell for all experiments. Experiments were run with and
without ferrocene as an internal reference. Electrolyte solutions were
composed of dry MeCN or N,N-DMF, containing 1 mM of catalyst,
0.1 M TBAPF6 as supporting electrolyte, and with or without 1.88 M
TEOA (solvent: TEOA = 4:1 v/v) unless otherwise noted. The
electrolyte was purged with Ar or CO2 for before cyclic voltammo-
grams were recorded and stirred between successive experiments.
Infrared Spectroelectrochemistry. The experimental method,
cell design, and setup of the IR-SEC cell have been reported
previously.18 All measurements were made with a Pine Instrument
Company model AFCBP1 bipotentiostat. The IR-SEC cell is
composed of GC working electrode, Ag pseudoreference electrode,
EXPERIMENTAL METHODS
■
General. All reagents were obtained from commercial suppliers and
used as received unless otherwise noted. Acetonitrile (MeCN) was
obtained from a solvent system under argon (Ar); N,N-dimethylfor-
mamide (N,N-DMF) and methanol were dried over molecular sieves
(3 Å) for 3 days under N2 prior to use. Triethanolamine (TEOA) was
stored under N2. Tetrabutylammonium hexafluorophosphate
(TBAPF6) was recrystallized in MeOH twice and dried at 90 °C
before use. Mn(CN)(bpy)(CO)3 was synthesized according to a
literature procedure.26
1H NMR spectra were recorded on a Varian Mercury Plus 400 MHz
spectrometer and NMR chemical shifts were referenced to the proton
signal of ferrocene. Fourier transform infrared (FTIR) spectra were
recorded on a ThermoNicolet 6700 spectrophotometer running
OMNIC software. Absorption spectra were recorded on a CARY
F
DOI: 10.1021/acs.inorgchem.6b00379
Inorg. Chem. XXXX, XXX, XXX−XXX

Inorganic Chemistry
Article
and Pt counter electrode. All potentials were referenced to the
pseudoreference Ag/Ag+, ∼+200 mV higher than the Fc/Fc+ couple.
The solution containing 4 mM Mn complex and 0.1 M TBAPF6 in dry
MeCN or N,N-DMF with or without TEOA (solvent: TEOA = 4:1 v/
v) was sparged with N2 or CO2 and then injected into the IR-SEC cell
to form a thin liquid layer for bulk electrolysis. The potential was
changed stepwise and the solution was monitored with Fourier
transform reflectance IR off the electrode surface over time.
Preparation of Ru(dmb)3(PF6)2. The photosensitizer, Ru-
(dmb)3(PF6)2 (dmb = 4,4′-dimethyl-2,2′-bpyridine) was synthesized
from RuCl3•xH2O (91.7 mg, 0.169 mmol) and dmb (46.5 mg, 0.255
mmol) according to literature45 and further purified by recrystalliza-
tion with acetone-diethyl ether. All characterization was consistent
with previous reports; yield: 55.2 mg (34.6%).
(17) Fei, H. H.; Sampson, M. D.; Lee, Y.; Kubiak, C. P.; Cohen, S. M.
Inorg. Chem. 2015, 54, 6821.
(18) Machan, C. W.; Sampson, M. D.; Chabolla, S. A.; Dang, T.;
Kubiak, C. P. Organometallics 2014, 33, 4550.
(19) Grills, D. C.; Farrington, J. A.; Layne, B. H.; Lymar, S. V.; Mello,
B. A.; Preses, J. M.; Wishart, J. F. J. Am. Chem. Soc. 2014, 136, 5563.
(20) Riplinger, C.; Sampson, M. D.; Ritzmann, A. M.; Kubiak, C. P.;
Carter, E. A. J. Am. Chem. Soc. 2014, 136, 16285.
(21) Riplinger, C.; Carter, E. A. ACS Catal. 2015, 5, 900.
(22) Sampson, M. D.; Nguyen, A. D.; Grice, K. A.; Moore, C. E.;
Rheingold, A. L.; Kubiak, C. P. J. Am. Chem. Soc. 2014, 136, 5460.
(23) Meyer, T. J.; Caspar, J. V. Chem. Rev. 1985, 85, 187.
(24) Allen, D. M.; Cox, A.; Kemp, T. J.; Sultana, Q.; Pitts, R. B. J.
Chem. Soc., Dalton Trans. 1976, 1189.
(25) Van der Graaf, T.; Hofstra, R. M. J.; Schilder, P. G. M.; Rijkhoff,
M.; Stufkens, D. J.; Vanderlinden, J. G. M. Organometallics 1991, 10,
3668.
(26) Machan, C. W.; Stanton, C. J.; Vandezande, J. E.; Majetich, G.
F.; Schaefer, H. F.; Kubiak, C. P.; Agarwal, J. Inorg. Chem. 2015, 54,
8849.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.inorg-
chem.6b00379.
■
S
(27) Paul, A.; Connolly, D.; Schulz, M.; Pryce, M. T.; Vos, J. G. Inorg.
Chem. 2012, 51, 1977.
(28) Kuramochi, Y.; Kamiya, M.; Ishida, H. Inorg. Chem. 2014, 53,
3326.
NMR, FTIR, and UV−vis spectra data (PDF)
AUTHOR INFORMATION
Corresponding Author
*E-mail: ckubiak@ucsd.edu (C.P.K.).
Notes
(29) Kitamura, N.; Rajagopal, S.; Tazuke, S. J. Phys. Chem. 1987, 91,
3767.
(30) Connelly, N. G.; Geiger, W. E. Chem. Rev. 1996, 96, 877.
(31) Ross, H. B.; Boldaji, M.; Rillema, D. P.; Blanton, C. B.; White, R.
P. Inorg. Chem. 1989, 28, 1013.
■
(32) Fukuzumi, S.; Hironaka, K.; Nishizawa, N.; Tanaka, T. Bull.
Chem. Soc. Jpn. 1983, 56, 2220.
The authors declare no competing financial interest.
(33) Sullivan, B. P.; Meyer, T. J. Organometallics 1986, 5, 1500.
(34) Agarwal, J.; Johnson, R. P.; Li, G. H. J. Phys. Chem. A 2011, 115,
2877.
(35) Chan, S. F.; Chou, M.; Creutz, C.; Matsubara, T.; Sutin, N. J.
Am. Chem. Soc. 1981, 103, 369.
(36) Koike, K.; Hori, H.; Ishizuka, M.; Westwell, J. R.; Takeuchi, K.;
Ibusuki, T.; Enjouji, K.; Konno, H.; Sakamoto, K.; Ishitani, O.
Organometallics 1997, 16, 5724.
(37) Franco, F.; Cometto, C.; Ferrero Vallana, F.; Sordello, F.; Priola,
E.; Minero, C.; Nervi, C.; Gobetto, R. Chem. Commun. 2014, 50,
14670.
(38) Tamaki, Y.; Watanabe, K.; Koike, K.; Inoue, H.; Morimoto, T.;
Ishitani, O. Faraday Discuss. 2012, 155, 115.
(39) Gholamkhass, B.; Mametsuka, H.; Koike, K.; Tanabe, T.; Furue,
M.; Ishitani, O. Inorg. Chem. 2005, 44, 2326.
(40) Keith, J. A.; Grice, K. A.; Kubiak, C. P.; Carter, E. A. J. Am.
Chem. Soc. 2013, 135, 15823.
(41) Agarwal, J.; Shaw, T. W.; Schaefer, H. F.; Bocarsly, A. B. Inorg.
Chem. 2015, 54, 5285.
(42) Keene, F. R., Sullivan, B. P. In Electrochemical and Electrocatalytic
Reactions of Carbon Dioxide; Sullivan, B. P., Krist, K., Guard, H. E., Ed.;
Elsevier: Amsterdam, 1993; p 118.
(43) Gerasimov, O. V.; Parmon, V. N. Russ. Chem. Rev. 1992, 61, 154.
(44) Priyadarshani, N.; Liang, Y. N.; Suriboot, J.; Bazzi, H. S.;
Bergbreiter, D. E. ACS Macro Lett. 2013, 2, 571.
(45) Hawecker, J.; Lehn, J. M.; Ziessel, R. J. Chem. Soc., Chem.
Commun. 1985, 56.
ACKNOWLEDGMENTS
■
P.L.C., C.W.M., and C.P.K. acknowledge support for this work
from the AFOSR through a Basic Research Initiative (BRI)
Grant (No. FA9550-12-1-0414). J.A. acknowledges support
from the NSF (CHE-1361178).
REFERENCES
■
(1) Lewis, N. S.; Nocera, D. G. Proc. Natl. Acad. Sci. U. S. A. 2006,
103, 15729.
(2) Balzani, V.; Credi, A.; Venturi, M. ChemSusChem 2008, 1, 26.
(3) Benson, E. E.; Kubiak, C. P.; Sathrum, A. J.; Smieja, J. M. Chem.
Soc. Rev. 2009, 38, 89.
(4) Ishida, H.; Terada, T.; Tanaka, K.; Tanaka, T. Inorg. Chem. 1990,
29, 905.
(5) Lehn, J. M.; Ziessel, R. J. Organomet. Chem. 1990, 382, 157.
(6) Kumar, B.; Smieja, J. M.; Kubiak, C. P. J. Phys. Chem. C 2010,
114, 14220.
(7) Tamaki, Y.; Morimoto, T.; Koike, K.; Ishitani, O. Proc. Natl. Acad.
Sci. U. S. A. 2012, 109, 15673.
(8) Morimoto, T.; Nakajima, T.; Sawa, S.; Nakanishi, R.; Imori, D.;
Ishitani, O. J. Am. Chem. Soc. 2013, 135, 16825.
(9) Shakeri, J.; Farrokhpour, H.; Hadadzadeh, H.; Joshaghani, M.
RSC Adv. 2015, 5, 41125.
(10) Smieja, J. M.; Sampson, M. D.; Grice, K. A.; Benson, E. E.;
Froehlich, J. D.; Kubiak, C. P. Inorg. Chem. 2013, 52, 2484.
(11) In CRC Handbook of Chemistry and Physics, 92nd ed.; Haynes,
W. M., Ed.; CRC Press: Boca Raton: FL, 2011; p 2011.
(12) Juris, A.; Balzani, V.; Barigelletti, F.; Campagna, S.; Belser, P.;
von Zelewsky, A. Coord. Chem. Rev. 1988, 84, 85.
(13) Matsuoka, S.; Yamamoto, K.; Ogata, T.; Kusaba, M.;
Nakashima, N.; Fujita, E.; Yanagida, S. J. Am. Chem. Soc. 1993, 115,
601.
(14) Loudet, A.; Burgess, K. Chem. Rev. 2007, 107, 4891.
(15) Bourrez, M.; Molton, F.; Chardon-Noblat, S.; Deronzier, A.
Angew. Chem., Int. Ed. 2011, 50, 9903.
(16) Takeda, H.; Koizumi, H.; Okamoto, K.; Ishitani, O. Chem.
Commun. 2014, 50, 1491.
G
DOI: 10.1021/acs.inorgchem.6b00379
Inorg. Chem. XXXX, XXX, XXX−XXX