Photocatalytic CO2 Reduction by Trigonal-Bipyramidal Cobalt(II

Article Cite This: Inorg. Chem. XXXX, XXX, XXX−XXX

pubs.acs.org/IC

Photocatalytic CO2 Reduction by Trigonal-Bipyramidal Cobalt(II) Polypyridyl Complexes: The Nature of Cobalt(I) and Cobalt(0) Complexes upon Their Reactions with CO2, CO, or Proton Tomoe Shimoda,†,‡ Takeshi Morishima,‡ Koichi Kodama,‡ Takuji Hirose,‡ Dmitry E. Polyansky,† Gerald F. Manbeck,† James T. Muckerman,† and Etsuko Fujita*,† †

Chemistry Division, Brookhaven National Laboratory, Upton, New York 11973-5000, United States Graduate School of Science and Engineering, Saitama University, Saitama, 338-8570, Japan

‡

S Supporting Information *

ABSTRACT: The cobalt complexes CoIIL1(PF6)2 (1; L1 = 2,6bis[2-(2,2′-bipyridin-6′-yl)ethyl]pyridine) and CoIIL2(PF6)2 (2; L2 = 2,6-bis[2-(4-methoxy-2,2′-bipyridin-6′-yl)ethyl]pyridine) were synthesized and used for photocatalytic CO2 reduction in acetonitrile. X-ray structures of complexes 1 and 2 reveal distorted trigonal-bipyramidal geometries with all nitrogen atoms of the ligand coordinated to the Co(II) center, in contrast to the common six-coordinate cobalt complexes with pentadentate polypyridine ligands, where a monodentate solvent completes the coordination sphere. Under electrochemical conditions, the catalytic current for CO2 reduction was observed near the Co(I/0) redox couple for both complexes 1 and 2 at E1/2 = −1.77 and −1.85 V versus Ag/AgNO3 (or −1.86 and −1.94 V vs Fc+/0), respectively. Under photochemical conditions with 2 as the catalyst, [Ru(bpy)3]2+ as a photosensitizer, tri-p-tolylamine (TTA) as a reversible quencher, and triethylamine (TEA) as a sacrificial electron donor, CO and H2 were produced under visible-light irradiation, despite the endergonic reduction of Co(I) to Co(0) by the photogenerated [Ru(bpy)3]+. However, bulk electrolysis in a wet CH3CN solution resulted in the generation of formate as the major product, indicating the facile production of Co(0) and [Co−H]n+ (n = 1 and 0) under electrochemical conditions. The one-electron-reduced complex 2 reacts with CO to produce [Co0L2(CO)] with νCO = 1894 cm−1 together with [CoIIL2]2+ through a disproportionation reaction in acetonitrile, based on the spectroscopic and electrochemical data. Electrochemistry and time-resolved UV−vis spectroscopy indicate a slow CO binding rate with the [CoIL2]+ species, consistent with density functional theory calculations with CoL1 complexes, which predict a large structural change from trigonal-bipyramidal to distorted tetragonal geometry. The reduction of CO2 is much slower than the photochemical formation of [Ru(bpy)3]+ because of the large structural changes, spin flipping in the cobalt catalytic intermediates, and an uphill reaction for the reduction to Co(0) by the photoproduced [Ru(bpy)3]+.

■

INTRODUCTION In recent years, energy, natural resources, and the environment have become indisputably important issues. Many scientists have expended great effort to address these issues through technologies that convert sunlight into electricity or chemical bonds (i.e., solar fuels).1−6 Among these, the development of carbon dioxide (CO2) reduction to energy-rich C1 chemicals such as carbon monoxide (CO), formic acid, methanol (CH3OH), and methane using molecular complexes, metal electrodes, and semiconductor materials has attracted significant attention.7−26 Quite a few cobalt molecular complexes have been studied for their binding of CO2,27−30 electrochemical CO2 reduction14,31−38 and photochemical CO2 reduction,39−43 especially in nonaqueous media because in aqueous systems proton reduction competes with CO2 reduction.44 We have prepared new cobalt complexes with pentadentate ligands 2,6-bis[2-(2,2′-bipyridin-6′-yl)ethyl]pyridine (L1) and 2,6-bis[2-(4-methoxy-2,2′-bipyridin-6′-yl)© XXXX American Chemical Society

ethyl]pyridine (L2) (Chart 1) for the investigation of CO2 reduction. L2 is expected to exhibit stronger σ donation by methoxy groups, making the Co center more electron-rich, leading to higher association constants with CO2 and other substrates. Here we report the synthesis of new Co(II) complexes, 1 and 2, their X-ray structures, the spin states of the Co(II) center, and experimental and theoretical studies of the binding of CO2, CO, or a proton to the reduced Co complexes. Interestingly, these Co(II) complexes have trigonalbipyramidal (TBP) geometry without an additional monodentate ligand such as a solvent molecule. For photocatalytic CO2 reduction, we employed the conditions reported by Shan and Schmehl,45 which use [Ru(bpy)3]2+ as a photosensitizer, tri-p-tolylamine (TTA) as a commercially available efficient reductive quencher, and triethylamine (TEA) as a sacrificial Received: February 20, 2018

A

DOI: 10.1021/acs.inorgchem.8b00433 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry Chart 1. Structures of the Cations of Various Cobalt Complexes Discussed in This Study

electron donor. The quenching rate constant of *[Ru(bpy)3]2+ by 0.04 M TTA was reported as 1.03 × 108 M−1 s−1 with a quenching efficiency of 0.77. The quantum yield of the oneelectron-reduced [Ru(bpy)3]+ of 0.62 is a dramatic improvement over the efficiency for the quenching of *[Ru(bpy)3]2+ by TEA alone.45 We envisioned that the high quantum yield for the formation of [Ru(bpy)3]+ under those conditions would enhance the photochemical CO2 reduction with complexes 1 and 2. While Co(II) is reduced to Co(I) more efficiently during photolysis in the presence of TTA than in its absence, the rates of CO production are nearly identical in both experiments. The thermal reaction of CO2 reduction and the release of CO seem to be much slower than the formation of [Ru(bpy)3]+ for several reasons, which we will discuss.

■

were prepared under high vacuum with vacuum-distilled CH3CN and reduced gradually by introducing small portions to the amalgam chamber. NMR samples of Co(I) species were prepared in CD3CN using a homemade glassware apparatus equipped with a NMR tube, a 0.1 mm optical cell, and Ha/Hg chamber separated by a glass frit and flame-sealed in the NMR tube after confirmation of the formation of the Co(I) species by UV−vis spectroscopy. Electrochemistry and Spectroelectrochemistry. Electrochemical measurements were carried out with a BASi Epsilon potentiostat or a CH Instruments 620E electrochemical analyzer. Cyclic voltammetry (CV) was performed using a glassy carbon disk (3 mm) working electrode, a platinum wire counter electrode, and a silver/silver nitrate (Ag/AgNO3) reference electrode in CH3CN containing 0.1 M tetrabutylammonium hexafluorophosphate (TBAPF6) as a supporting electrolyte. The scan rate was 100 mV s−1 unless otherwise noted. Solutions were purged with argon or CO2 (to saturation at 0.28 M) before cyclic voltammograms were taken. Before and after CV measurements, the potentials were calibrated with reference to the Fc+/0 couple (Fc+/0 = 90 mV vs Ag/AgNO3). Bulk electrolysis was performed in a two-compartment cell with a mercury pool working electrode and a silver wire pseudoreference separated from the platinum counter electrode by a fine glass frit. Experiments were performed for 10 min with 20−30 turnovers of products formed with no sign of catalyst deactivation except for the experiments in the absence of water, in which case rapid precipitation of the deactivated catalyst was observed. Hydrogen and CO were detected by gas chromatography (GC). Formate was quantified by 1H NMR using an internal standard. IR spectroelectrochemical measurements were carried out using a home-built electrochemical cell equipped with a gold working electrode, a platinum reference electrode, and a solid Ag/AgCl reference electrode (LF-1 from Innovative Instruments, Inc.) using a VeeMAX III Variable Angle Specular Reflectance Accessory (PIKE Technologies, Inc.) installed in a Bruker IFS 66/s spectrometer. Solutions containing 100 mM TBAPF6 electrolyte and ca. 1 mM cobalt complex that have been purged with argon, CO2, or CO were loaded into a gastight syringe (Hamilton Co.) and delivered into the cell through a gastight fluidic connection. Solutions were exchanged with at least three cell volumes between runs. A square-wave voltammetry scan was performed prior to each bulk electrolysis experiment to determine redox-wave potentials versus reference electrode. For spectroelectrochemical runs, the cell was biased at a constant potential and spectral data were collected with 2 cm−1 resolution using a mercury−cadmium telluride detector (Kolmar Technologies) at regular time intervals. The spectra were backgroundcorrected for the absorption of solvent and electrolyte. Photocatalysis. Each sample in a CH3CN solution with 300 μM [Ru(bpy)3](ClO4)2, 40 mM TTA, 0.55 M TEA, and a 180 μM cobalt

EXPERIMENTAL DETAILS

Materials. All chemicals and solvents used for synthesis were of reagent-grade quality and were used without further purification. Acetonitrile (CH3CN) used for photochemical studies and Na/Hg reduction was purified using a procedure similar to one published previously.28 CH3CN used for electrochemical studies was dried by passage through two alumina columns in a solvent-dispensing system designed by J. C. Meyer. [Ru(bpy)3]Cl2 was synthesized according to literature methods46,47 or obtained from Strem Chemicals and converted to the [Ru(bpy)3](ClO4)2 salt using AgClO4. (Warning! The perchlorate salts used in this study may be potentially explosive and hazardous.) L1 was prepared by a method similar to the one published previously.48 The detailed synthesis procedures of complexes 1 and 2 are shown in the Supporting Information (SI), together with those of L1 and L2. Certified gas mixtures [3, 10, and 30% CO2 balanced with N2] from MG Industries were used for electrochemistry measurements. The 1H NMR spectra were recorded with a Bruker Avance 500, 400, or 300 spectrometer at 298 K, and coupling constants are reported in hertz. The effective magnetic moment of complex 2 was determined by the Evans method in CD3CN with 5 vol % benzene.49−53 Mass spectrometry was performed on a LCQ ADVANTAGE MAX (Finnigan) mass spectrometer using CH3OH as the eluent. Elemental analyses were conducted by Robertson Microlit Laboratories (Ledgewood, NJ). Electronic absorption spectra were recorded using an Agilent 8454 UV−vis spectrophotometer. Fourier tranform infrared (FT-IR) spectra were recorded with a Jasco FT-IR 400 spectrometer or a Bruker IFS 66/S spectrometer. Sodium reduction was performed in a homemade airtight vessel equipped with a quartz spectrophotometric cell separated by a fine glass frit from a second compartment containing 0.5% sodium in mercury. Samples B


Article

Inorganic Chemistry

Figure 1. X-ray structures of the cations of complexes 1 (left) and 2 (right). complex, unless otherwise noted, was prepared using a homemade glassware apparatus equipped with a 1 cm × 1 cm rectangular optical cell and a valve in a glovebox. The total volume of the apparatus was 12.5 mL, and 4 mL of solution was used for photolysis. The system was purged with CO2 for 20 min. Some experiments were carried out with 0.59 M TEA without TTA to investigate the effects of TTA. The sample was irradiated by a 150 W xenon lamp with a UV cutoff filter (410 nm ≤ λ) at room temperature under constant stirring. Gaseous products such as CO and H2 were analyzed by gas chromatography on an Agilent 6890N gas chromatograph with HP-MOLSIV (30 m × 0.32 mm × 12 μm) and GS-CARBON-PLOT (15 m × 0.32 mm × 1.5 μm) columns. CO was quantified via a methanizer using a calibrated flame ionization detector. The products in the solution were analyzed using a Dionex ICS-1600 ion chromatography instrument with a 4.5 mM sodium carbonate and 0.8 mM sodium bicarbonate eluent solution. The turnover number (TON) was calculated by the ratio of CO produced to the initial concentration of the cobalt complex catalyst. The quantum yields of Co(I) and CO formation were determined using a 405 nm laser diode for 60 s (60 mW) and 15 min (25 mW) irradiation, respectively. The amounts of Co(I) and CO were determined by optical spectroscopy and GC, respectively. Determination of the CO Binding Rate to the Photoproduced [CoIL2]+. Samples containing 2 mM p-terphenyl (TP) as a photosensitizer, 0.1 mM complex 2, and 150 mM TEA as a sacrificial electron donor were degassed using three freeze−pump−thaw cycles on a vacuum line and filled with different partial pressures of CO using a Baratron gauge for pressure measurements. The concentration of CO in solution was calculated using CO partial pressure and its solubility in CH3CN (0.0083 M under 1 atm of CO). Prepared samples were excited with 266 nm laser pulses (fourth harmonic of Nd/YAG SpectraPhysics Lab 170, 2 ns pulse width, 1 mJ pulse−1), and the absorbance at 580 nm was probed using the filtered (long pass filter above 500 nm) output of a 75 W xenon arc lamp in a 90° configuration detected by a photomultiplier tube (Hamamatsu, R928) and recorded with a digital oscilloscope (Tektronics DPO 4054B). One kinetic trace was recorded per each laser pulse, and the sample was stirred between laser pulses. The final kinetic traces were the average of eight acquisitions and were fit to a single-exponential function. Collection and Refinement of X-ray Data. Single crystals of 1 and 2 suitable for X-ray crystallographic analysis were grown by recrystallization from CH3CN/CH3OH mixed solvents, respectively. X-ray diffraction data were collected using a Bruker Smart APEX II diffractometer with graphite-monochromated Mo Kα radiation (0.71073 Å). The structures were solved by direct methods54 using SIR-2004 and refined using the SHELXL-2013 program. In the leastsquares refinement, the anisotropic temperature parameters were used for all of the non-hydrogen atoms. Hydrogen atoms were placed at

calculated positions and allowed to ride on the atom to which they were attached. The isotropic thermal parameters for the hydrogen atoms were determined from the atom to which they are attached. The data were corrected for absorption using the multiscan method (SADABS).55 A summary of the crystallographic data is given in Table S1. Density Functional Theory (DFT) and Wave-Function-Based Calculations. DFT calculations (RKS states for closed shells and UKS states for open shells) were carried out using the Gaussian 09 program package,56 the B3LYP functional,57−60 and the LANL2DZ basis set and effective core potentials for all atoms.61−64 All species considered were fully optimized in a SMD continuum model of the CH3CN solvent,65 and frequency analyses were carried out to exclude any transition-state geometries and to confirm a minimum in the energy. To provide insight into determination of the spin state of the cobalt(II) center, we performed higher-level wave-function-based ab initio electronic structure calculations on complex 1 (to take advantage of its C2 symmetry and to simplify the electronic structure) using the Gaussian09 program package,56 in particular its CASSCF facility.66−71 First a converged UHF quartet wave function at the geometry of the experimental crystal structure (omitting the two PF6− counterions) was obtained. Then CASSCF calculations with an active space of seven electrons in five orbitals were carried out separately for spin multiplicities 4 and 2, iteratively exchanging active-space orbitals that were doubly occupied in all configurations with d orbitals in order to obtain an active space consisting of the five d orbitals of the cobalt center. A final CASSCF(7,5) calculation was carried out with the optimized active space, followed by MP2 calculations on the CASSCF wave function to include the effect of dynamic correlation on the relative energy of the quartet and doublet states. To explore the electronic states of the singly reduced species of 4[CoII(L1)]2+, a similar procedure was carried out for states of triplet and singlet multiplicity. Additional B3LYP/LANL2DZ calculations were performed to explore the effects of geometry change and solvation on the electronic states of the one-electron-reduced species.

■

RESULTS AND DISCUSSION Synthesis, Characterization, and X-ray Structures. The synthesis procedures for ligands L1 and L2 and complexes 1 and 2 along with their characterization data are provided in Figures S1−S11. For the synthesis of complexes 1 and 2, CoCl2 and ligands L1 and L2, respectively, were dissolved in a CH3OH/CHCl3 (8/1, v/v) mixture and stirred for 30 min at room temperature. A saturated KPF6 aqueous solution was added, and the precipitate was collected by filtration. Crystals suitable for the X-ray diffraction study were obtained by C


Article

Inorganic Chemistry

NCS, NO3, and ClO4) complexes with TBP geometry are 4.12−4.55 μB.76 We measured the effective magnetic moment of complex 2 at room temperature by the Evans method as 4.25 μB (Figure S12), which is close to 3.88 μB for three unpaired electrons, indicating a high-spin Co(II) complex. The UV−vis spectrum of complex 2 in CH3CN (Figure 2) shows an intense band at 295 nm from a ligand π−π* transition

recrystallization from CH3CN/CH3OH to yield complex 1 as dark-brown plates and from CH3CN/diethyl ether to yield complex 2 as purple plates. ORTEP drawings of the cations of 1 and 2 are shown in Figure 1. Crystal data and a summary of selected bond distances and angles of the cations of 1 and 2 are provided in Tables S1 and S2, respectively. The polypyridine ligand L1 binds to one Co(II) ion in a pentadentate fashion to afford complex 1 in TBP geometry without coordination of any solvent molecules. In complex 1, the asymmetric unit contained half of an L1 ligand, and the nitrogen atoms of the central (N3) and terminal (N1 and N1*) pyridines occupy the equatorial positions while the axial positions are occupied by the other two nitrogen atoms (N2 and N2*). The equatorial Co−N bond distances (2.032−2.048 Å) are much shorter than those at the axial positions (2.145 Å). Within the trigonal plane involving N1, N3, N1*, and the cobalt center, the N1−Co− N1* angle is about 20° smaller than the N1−Co−N3 and N3− Co−N1* angles. However, the N2−Co−N2* angle is 175.14(7)°. Such a geometry is in contrast with the corresponding Ru(II) complex of L1 previously reported, where the additional coordination of CH3CN on the plane formed by the central and two terminal pyridine nitrogen atoms affords a six-coordinate, square-bipyramidal geometry.48 The RuIIL1 geometry deviates largely from octahedral because of the constraints caused by the ethylene linkers (with N−Ru−N angles of the ligand nitrogen atoms ranging from 80.33° to as large as 106.82° except those involving the two nitrogen atoms of a 2,2′-bipyridine ligand). Complex 2 has a TBP geometry similar to that of complex 1 and crystallized with two CH3CN molecules. The asymmetric unit in complex 2 contains one molecule of L2. The equatorial Co−N bond distances (2.011− 2.041 Å) are also much shorter than those at the axial positions (2.149−2.159 Å) in complex 2. Despite the different measurement temperatures, the Co−N1 and Co−N5 distances of complex 2 are shorter than the Co−N1 and Co−N1* distances of complex 1, respectively. This stronger coordination is attributable to the electron-donating ability of the methoxy groups attached to the terminal pyridines of ligand L2. While complex 2 does not have C2 symmetry because of the positions of the anion and solvated CH3CN, the cation of 2 has near-C2 symmetry. A TBP geometry of the Co(II) complexes with a pentadentate ligand is very rare, and only a few X-ray structures have been previously reported.72,73 As shown in Chart 1, those structures involve cobalt complexes with rigid five-coordinate polyamine ligands such as the encapsulated octaazamacrocyclic ligand [4,14,19-tris(methoxymethyl-1,4,6,9,12,14,19,21octaazabicyclo[7.7.7]tricosane], complex 3, 72 and 3,7diazabicyclo[3.3.1]nonane with four tertiary amine groups and one pyridyl group, complex 4.73 While the geometry of complex 3 is similar to those of complexes 1 and 2 with long Co−N axial bond distances, complex 4 has a rather distorted TBP geometry, probably because of the strained nonsymmetrical nature of the ligand. The X-ray crystal structures of Co(II) complexes of 2,6-bis[[methyl(pyrid-2-ylmethyl)amino]N-methyl]pyridine (5),74 tetradentate bis[2-(3,5-dimethylpyrazol-1-yl)ethyl][(pyrazol-1-y)methyl]amine (6),75 and tris(2pyridylmethyl)amine43 also reveal distorted TBP geometries. Complex 6 is a high-spin Co(II) complex with effective magnetic moments of 4.50 μB at 300 K and 3.68 μB at 2 K. Complex 3 has an effective magnetic moment of 4.18 μB at 290 K,72 while the magnetic moments of a series of [Co(NTB)X]+ (7; NTB = tris(benzimidazolyl-2-methyl)amine and X = Br,

Figure 2. UV−vis absorption spectra of complex 2 in CH3CN using (a) 5.3 × 10−5 M and (b) 1.0 × 10−2 M complex 2 solutions.

and a weak shoulder peak at 325 nm likely from a metal-toligand charge-transfer (MLCT) transition. Weak d−d transitions with molar absorptivity of less than 50 M−1 cm−1 are observed in the visible region (Figure 2b), similar to a reported TBP complex.72 Time-dependent DFT calculations of complex 1 also indicate very weak d−d transitions in the visible region (see the SI). Electrochemistry. The cyclic voltammograms of complexes 1 and 2 in CH3CN display three reversible redox couples at E1/2 = −1.10, − 1.77, and −2.27 V for 1 and E1/2 = −1.27, − 1.85, and −2.36 V for 2 (vs Ag/AgNO3), assignable to Co(II/ I), Co(I/0), and Co(0/1−) (Figure 3a). Here we write the doubly and triply reduced species as Co(0) and Co(1−), respectively, but the actual electronic structures are more accurately described as [CoI(L•−)] and [Co0(L•−)]− according to the spectral changes observed for complex 2 and free ligand L2 upon Na/Hg reduction and DFT calculations of complex 1 (see below). The potentials of complex 2 are cathodically shifted compared to complex 1 because of the electrondonating methoxy groups (Figure 3a). Irreversible reduction of the free ligands occurred at potentials more negative than −2.5 V (Figure S15). However, when these ligands coordinate to the cobalt center, the reduction potentials seem to shift anodically, corresponding to the process [CoI(L)]/[CoI(L•−)]. In solutions containing CO2, the first reductions are unchanged, but increased current attributable to catalytic CO2 reduction is observed at the second reductions of complexes 1 and 2 (Figure 3b for 2). The catalytic rate constants (kcat) for 1 and 2 were calculated from icat. using eqs 1−3. Here icat. is the catalytic current, ip is the peak current in the absence of catalysis, n = 2 is the electrochemical stoichiometry for CO2 reduction to CO, F is the Faraday constant, A (0.152 π cm2) is the electrode surface area, R is the gas constant, T is the absolute temperature, v is the scan rate, and DCo is the catalyst diffusion coefficient (Figures S16 and S17). The CO 2 concentration at 25 °C is 0.28 M under 1 atm of CO2.77 icat. = nFA[Co] DCok 0[CO2 ] D

(1) DOI: 10.1021/acs.inorgchem.8b00433 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Figure 3. (a) Cyclic voltammograms of 0.5 mM complexes 1 and 2 in 0.1 M TBAPF6/CH3CN under argon. (b) Cyclic voltammograms of 0.5 mM complex 2 under argon and CO2/N2 (or CO2/argon) gas mixtures. Electrode = glassy carbon, scan rate = 100 mV s−1, potential = volts versus Ag/ AgNO3, and E1/2(Fc+/0) = 0.09 V.

icat. n = ip 0.4463 kcat. = k 0[CO2 ]

RTk 0[CO2 ] k = 0.72 0 [CO2 ] Fv v

the pKa of water in CH3CN might be balanced to accelerate CO2 reduction while precluding proton reduction catalysis. In the 2−10% range of water by volume, the catalytic current at the Co(I/0) couple is enhanced 2-fold compared to that when no water was added, but the current decreases slightly at the highest water content (Figure S24). Because the solubility of CO2 in CH3CN drops from 280 mM in a pure solvent to roughly 215 mM at 10% water,85 the small drop in the catalytic current might be due to the change of [CO2] without a major change in the selectivity for CO2 reduction. Bulk electrolysis experiments were used to quantify the products of catalysis using various water concentrations at an applied potential equivalent to that of the peak current by CV. In the absence of water, the catalytic current was not sustained and a pale-gray precipitate was observed. CO was produced with low Faradaic efficiency (FE) and in substoichiometric quantities (Table 1). In the presence of water, catalysis was

(2) (3)

The current increases linearly with the square root of the CO2 concentration (Figures S18−S21). The k0 values of 0.94 and 6.8 M−1 s−1 for CO2 reduction with complexes 1 and 2, respectively, were determined from eq 1 using the measured diffusion coefficients of 1.0 × 10−5 and 9.5 × 10−6 cm2 s−1, respectively. Using eq 3, the kcat. values of 0.26 and 1.9 s−1, respectively, were determined in a CO2-saturated solution. The values obtained from eqs 2 and 3 are smaller (0.1 and 0.9 s−1, respectively), but within the experimental error, assuming that the product may be reactive at more negative applied potentials because the limiting current was not observed. These rate constants are smaller than those calculated for the doubly reduced species of precious metal complexes [Ru(tpy)(bpy)(CH3CN)]2+ (kcat. = 5.5 s−1)78 and [Re(bpy)(CO)3Cl] (kcat. = 14 s−1)79 under 1 atm of CO2 in CH3CN using a similar method. Reduction of CO2 to formate or CO requires an oxide acceptor, such as a proton or another molecule of CO2 (to yield carbonate).9,11 These considerations motivate the examination of CO2 reduction catalysis in the presence of weak acids.80−84 The cyclic voltammogram of complex 2 under argon in the presence of acetic acid or trifluoroacetic acid shows a positive shift of the Co(II/I) couple, followed by a catalytic wave consistent with the binding of H+ to the singly reduced species and catalytic reduction of the acid to H2 (Figure S22). Given the high reactivity for proton reduction using strong acids, CO2 reduction was not investigated in this medium. On the other hand, the Co(II/I) couple is unchanged when a weaker acid such as phenol or water is added to the solution (Figures S23 and S24). Catalytic H2 production is observed at the Co(I/0) couple using phenol and slightly cathodic of the Co(I/0) couple using water under argon. At low concentrations of phenol, selective reduction of CO2 may be possible (Figure S25), but when a larger amount of phenol (>0.3 M) is added, the catalytic current does not show any dependency on the CO2 concentration, indicating that proton reduction is the dominant pathway. A comparison of the CV data under argon or CO2 using 2−10% water suggests that

Table 1. Summary of Bulk Electrolysis Dataa FE,b % entry

water content %

CO

H2

HCO2H

HCO2H/ CO/H2

1 2 3 4

0 2 5 10

14c 22 ± 4 9±3 4.2 ± 1.3

5.9 ± 1.0 7.2 ± 1.8 9.5 ± 1.1

64 ± 4 75 ± 12 64 ± 8

10.8:3.7:1 10.4:1.2:1 6.7:0.4:1

a

Reaction conditions: 0.5−1 mM [CoIIL2]2+, 0.1 M Bu4NPF6, mercury pool working electrode, silver wire pseudoreference, platinum mesh counter electrode, CO2-saturated CH3CN ([CO2] = 0.28 M), potential = −1.9 V vs Ag/AgNO3. bThe FEs for CO, H2, and formate were averaged over two runs and were corrected for the background current. cThis is an estimated value because under 1 turnover was obtained and at least 2 equiv of charge was consumed by reduction of the catalyst.

sustained and no precipitate was observed. The reaction yielded CO, H2, and formate with a constant FE for formate but decreasing CO and increasing H2 as water was increased to 10% by volume. When the cyclic voltammogram in dry CH3CN (Figure 3b) is compared to that in bulk electrolysis (Table 1, entry 1), the absence of catalysis in the latter suggests a stoichiometric CO2 reduction to form [Co0CO], which inhibits catalysis owing to E


Article

Inorganic Chemistry

Figure 4. Left panel: Photocatalytic CO production in dry CH3CN containing 180 μM 1 (purple diamonds) or 2 (black down triangles), 300 μM [Ru(bpy)3](ClO4)2, 40 mM TTA, and 0.55 M TEA under irradiation 410 nm ≤ λ. Right panel: Photocatalytic CO production in H2O/CH3CN for complex 2 with similar conditions: 2% water (black squares), 5% water (red circles), and 10% water (blue triangles).

Table 2. Photocatalytic Reductions in CH3CNa entry

catalyst

1 2 3 4 5 6b 7 8 9c

blank 1 2 2 2 2 2 2 2

water content, %

catalyst, μM

CO, μmol

2 5 10

0 180 60 120 180 180 180 180 180

0.04 2.5 0.95 2.2 3.7 5.2 4.3 6.3 8.5

H2, μmol

2.1 3.8 5.8 28 23 12

HCOO−, μmol 1.2 1.0 0.8

2.1 2.3

TONCO

TOFCO, h−1

TONH2

3.4 4.0 4.6 5.1 7.2 6.0 8.8 12

2.7 4.0 3.5 3.5

8.8 7.9 8.1

3.6 3.1 4.0

39 32 17

Reaction conditions: 0−180 μM cobalt complex catalyst, 300 μM [Ru(bpy)3](ClO4)2 as a photosensitizer, 40 mM TTA as a quencher, 0.55 M TEA as a sacrificial electron donor, irradiation with a xenon lamp (λ ≥ 410 nm) in CO2-saturated CH3CN ([CO2] = 0.28 M). All of the reactions were performed at 20 °C for 4 h. TONCO was calculated based on the cobalt complex. The amounts of CO and H2 and the TON and TOFCO values were averaged over two runs, with errors of less than 20% for CO and 25% for H2. In addition, TOFCO was calculated based on the TON for CO at an initial 1 h irradiation. bAfter a second run was conducted by the addition of 1.2 μmol of [Ru(bpy)3](ClO4)2 to the solution after the experiment of entry 5 was completed. cCloudy solution due to the limited solubility of TTA. a

Photochemical CO2 Reduction. Shan and Schmehl45 reported efficient reductive quenching of the excited state of [Ru(bpy)3]2+ when TTA and TEA are used in CH3CN as the quencher and sacrificial electron donor, respectively. We anticipated that such conditions could improve the efficiency of a photocatalytic CO2 reduction due to the high quantum efficiency of producing [Ru(bpy)3]+ (over 60%). Using conditions similar to those of Schmehl’s group, dry CH3CN solutions containing 180 μM 1, 300 μM [Ru(bpy)3](ClO4)2, 40 mM TTA, and 0.55 M TEA were irradiated using a 150 W xenon lamp with a 410 nm long pass filter at 20 °C (Figure 4). Complex 1 catalyzed the reduction of CO2 to CO for 4 h with a TON of 3.4 and a maximum TOF of 2.7. When using complex 2, CO generation was improved with a TON of 5.1 and TOF of 3.5. H2 was produced along with CO with the product ratio H2/CO ∼ 1.5−2 despite the absence of water. In this experiment, the TEA oxidation products are potential proton sources. To examine the dependence of the CO production activity on the concentration of the cobalt complex, we conducted photolysis experiments using catalyst concentrations of 60, 120, and 180 μM. As the concentration of the cobalt complex increases, the amount of CO generation increases proportionally (Figure S26). The initial TOF values are almost the same, indicating a first-order reaction. After 5 h of irradiation, a fresh aliquot of 300 μM [Ru(bpy)3](ClO4)2 (1.2 μmol) was added and the TON increased further to 7.2. This experiment suggests that the decomposition of [Ru-

the strong CO binding. To investigate the mechanism, a concentrated sample was electrolyzed with 13CO2 for analysis by NMR spectroscopy. After electrolysis, the 1H NMR spectrum was identical with the free ligand and the 13C NMR showed a single enhanced resonance at 163 ppm, which can be assigned to carbonate.86 In the absence of protons, CO2 will function as the oxide acceptor to produce carbonate and a cobalt carbonyl complex. The putative Co−CO species was observed by spectroelectrochemical IR experiments (see below), indicating that, although some ligand dissociation occurred in the electrolysis experiment, a portion of the catalyst remained with a strongly associated CO ligand, consistent with the low TON for CO release. The addition of water shifts the major catalytic product to formate. A small amount of H2 is also produced, and its yield is increased at the expense of CO as the water content is increased. The production of formate indicates that the doubly reduced complex preferentially binds H+ over CO2 and shows that the insertion of CO2 into the metal hydride bond is favored over protonolysis of the hydride by water to yield H2. On the other hand, CV data with other proton sources such as phenol show large currents under argon, indicating that the cobalt hydride will react with acids that are stronger than water in CH3CN. The data do not clearly indicate the oxidation state of the cobalt hydride during CO2 insertion. Given the negative potentials, it is likely reduced at least to Co(I), which will increase the hydricity and selectivity for formate. F


Article

Inorganic Chemistry

Figure 5. UV−vis absorption spectra of the cobalt species for 5.3 × 10−5 M complex 2 upon Na/Hg reduction in CH3CN: (a) complex 2 (black) and the one-electron-reduced species (red); (b) the one-electron-reduced species (red) and the two-electron-reduced species (blue).

(bpy)3]2+ causes limited durability of our photochemical system and that the cobalt complex remains active as a catalyst for CO2 reduction to CO. To explore the effect of a proton source on CO generation, we carried out photolysis experiments with 2, 5, and 10% water. These experiments yielded 6.0, 8.8, and 12 turnovers of CO, respectively, and 39, 32, and 17 turnovers of H2, which are considerably higher than the H2 TON in the absence of water (Table 2 and Figure S27). While we observed some precipitation of TTA with 10% water, the decreasing combined turnovers with increasing water are consistent with the lower current observed by CV above. The increase in the selectivity for CO2 reduction over proton reduction suggests that the proton might facilitate the formation of a metallocarboxylic acid Co−COOH intermediate but not enough to completely suppress the competitive binding of a proton to Co(0). An increase in the CO TON might imply that proton-induced cleavage of this Co−COOH intermediate to cobalt carbonyl and water is accelerated by the higher water content of the reaction mixture. A small amount of formate was also generated in photolysis experiments (formate/CO ≤ 0.5); however, [Ru(bpy)3]2+ is known to decompose to active CO2 reduction catalysts during reductive photolysis, and a comparable amount of formate was produced in a control experiment without the cobalt catalyst under our conditions. The different selectivities between electrochemical and photochemical experiments are noteworthy. The dominance of H2 in the photochemical reaction shows that the mixture contains an acid that is stronger than water and is likely TEAH+ derived from the deprotonation of TEA oxidation products. TEAH+ may react with a Co(II) hydride, whereas CO2 insertion likely requires further reduction of the hydride. We also carried out photochemical CO2 reduction using 0.59 M TEA without 40 mM TTA in CH3CN solutions containing 2% water and found that the amounts of both CO and H2 formation are almost the same as those with TTA. However, in CH3CN solutions containing 10% water, the formation of both CO and H2 with only TEA decreased to less than half of the amounts compared to those in the system with TTA. Because the system with TTA contains a considerable amount of precipitate, a comparison of these results may not be meaningful. As an initial step toward investigating the mechanism of CO2 reduction, the quantum yield for the formation of Co(I) in the absence of CO2 was determined using CH3CN solutions containing 2% water, 300 μM [Ru(bpy)3]2+, 180 μM complex

2, 40 mM TTA, and 0.55 M TEA (or 0.59 M TEA without TTA) under 405 nm irradiation (60 mW power) for 60 s. The quantum yields for Co(I) formation with and without TTA are 1.1 × 10−2 and 7.5 × 10−3, respectively. Electron transfer to Co(II) appears to be inefficient considering that the reported quantum yield for the formation of [Ru(bpy)3]+ via TTA/TEA reductive quenching under similar conditions is 0.62. In the presence of CO2, the quantum yields for CO and H2 formation (ΦCO and ΦH2, respectively) based on two photons to one molecule of gas were determined using 15 min of 405 nm irradiation (25 mW) to be 2.8 × 10−3 and 3.1 × 10−3, respectively, with the above solution with TTA. UV−Vis Spectra of the Reduced Species. The cyclic voltammograms of complex 2 in CH3CN exhibited three reversible redox couples (Figure 3a). Among them, the Co(II/ I) and Co(I/0) couples at potentials E1/2 = −1.27 and −1.85 V vs Ag/AgNO3, respectively, indicate that complex 2 can be chemically reduced by a sodium amalgam (Na/Hg; E0 = −1.96 V vs NHE87). UV−vis spectra of the one- and two-electronreduced ligand L2 species prepared by Na/Hg reduction (Figure S28) are distinctly different from the UV−vis spectra of the one- and two-electron-reduced cobalt complex species (Figure 5). The CH3CN solution of complex 2 is pale pink with visible absorptions, as shown in Figure 2b, but almost colorless for a 5.3 × 10−5 M solution (Figure 5a, black line). Upon Na/ Hg reduction, the color changed to blue for the one-electronreduced species and to brown for the two-electron-reduced species. Both reduced species have MLCT bands with high molar absorptivities in the visible region. It is of interest to investigate the electronic structure of these species. The 1H NMR spectrum of the one-electron-reduced species in Figure S29 shows sharp aliphatic and aromatic proton signals of the ligand, suggesting that this is a low-spin Co(I) d8 species and not the Co(II) ligand radical anion [CoII(L2•−)], indicating spin flipping during the reduction. These chemical shifts are different from those of the free ligand, as shown in Figure S11. The pattern of the NMR signal indicates that the ligand has a symmetrical configuration around the cobalt center. The twoelectron-reduced species can be assigned as [CoI(L2•−)] because a characteristic ligand radical anion absorption at ∼350 nm appears, accompanied by a decrease in the broad π−π* absorption at 298 nm (Figure 5), as observed in the reduction of L2 (Figure S28). In order to investigate the reactivity of the reduced species with CO2, the solution was saturated with CO2 after the preparation by Na/Hg (Figure S30). The singly reduced [CoIL2]+ did not react with CO2 in 24 h of observation. In G


Article

Inorganic Chemistry

investigated under a 10% and 100% CO atmosphere (1 atm of CO in CH3CN: 0.0083 M). A reversible peak with E1/2 = −1.27 V vs Ag/AgNO3, assigned as Co(II/I) under argon changed to an irreversible two-electron wave, and the Co(I/0) couple at −1.85 V was absent under 100% CO (Figure 7). When the concentration of CO was reduced 10-fold, the CVs (Figure S31) showed the current at −1.3 V, intermediate between a one- or two-electron wave and a residual reversible couple at −1.85 V with a current less than that under argon. The ratio of the current at each peak is scan-rate-dependent (Figures 7, right panel, and S31), with the highest current at −1.3 V at slow scan rates. This response indicates that the bimolecular reaction of Co(I) with CO is slow at this concentration relative to the sweep rate, and the residual current at −1.85 V is the Co(I/0) couple without CO binding. The reaction of Co(I) with CO was probed further by laser flash-quenching experiments using CH3CN solutions containing TP, TEA, and complex 2 under various CO concentrations. Upon 266 nm laser excitation, the singlet excited state of TP is rapidly quenched by TEA to form TP•−, which has a reduction potential of −2.4 V (vs SCE in dimethylamine).40,88 TP•− rapidly reduces Co(II) to Co(I), and the disappearance of Co(I) could be monitored. Figure S32 shows that the rate of Co(I) decay is linear with the concentration of CO (1 atm of CO in CH3CN: 0.08 M) and kCO is 890 M−1 s−1, indicating a very slow reaction compared to the typical rates of 106−108 M−1 s−1 for the binding of CO to transition-metal complexes.12 IR spectroelectrochemistry was carried out to detect [CoIL2]+ and [Co0L2] species in CH3CN under argon, CO, and CO2. The results are shown in Figures S33 and S34. The CC stretching frequencies around 1600 cm−1 shift to lower energy upon the reduction of [CoIIL2]2+ under argon. Under a CO atmosphere and an applied potential of 100 mV more negative than that of the Co(II/I) couple, νCO was observed at 1894 cm−1. In order to confirm the identity of observed species, the [Co0L2] complex was prepared by Na/Hg reduction of the parent [CoIIL2]2+ in dry CH3CN, followed by saturation with CO gas. The IR spectrum of the resulting [Co0L2(CO)] species featured a single band at νCO = 1894 cm−1, indicating that the same species was produced during the electrolysis experiment. This finding indicates that the Co(I/0) reduction potential of [CoIL2(CO)]+ is less negative than that of the [CoIL2]+ complex, in agreement with the two-electron wave

contrast, when CO2 was added to the two-electron-reduced species [CoI(L2•−)], a notable color change from brown to almost colorless was observed. The spectrum of this final product was similar to the spectrum of 2; however, λmax is shifted from 298 nm in 2 to 278 and 286 nm with a shoulder at ∼298 nm in the product (Figure 6). In the absence of

Figure 6. UV spectra of [CoIIL2]2+ (black), [CoIL2]+ CO (green), [CoIL2]+ + CO + one electron (blue), and [Co0L2] + CO2 (red) in CH3CN.

continued reduction, Co(II) carbonyl is a plausible intermediate. To investigate this possibility, the reduced complex was exposed to CO after Na/Hg reduction. The Co(I) species reacted with CO, as shown by the spectral shift in Figure 6; however, the resulting spectrum does not match either that of complex 2 or that obtained by the reaction of [CoI(L2•−)] with CO2. Instead, the spectrum seems to be that of a mixture of the two species. The reduction of this mixture by Na/Hg produced a spectrum identical with that obtained by the reaction of [CoI(L2•−)] with CO2. This finding was rather puzzling, because the reaction of [CoI(L2•−)] with CO2 should yield a product describable as Co(II): either as the metallocarboxylate or the carbonyl if C−O bond cleavage occurs (carbonate is the other product of this reaction). Therefore, we further investigated the reaction of the reduced complex 2 with CO using electrochemistry, laser flash photolysis, and IR spectroelectrochemistry. Reaction of the Reduced Complex 2 with CO. The cyclic voltammograms of complex 2 in CH3CN were

Figure 7. Left: Cyclic voltammograms of [CoL2]2+ under argon (red) and 100% CO (black) with a scan rate of 100 mV s−1. The small peak around −1600 mV is due to an Fe(CO)5 contaminant from the CO cylinder. Right: Current changes associated with scan rate changes under 10% CO. See the selected cyclic voltammograms in Figure S31. H


Article

Inorganic Chemistry

Figure 8. DFT-calculated structures of 2[Co0(κ5-L1)(CO2)], left, and 2[Co0(κ3-L1)(CO)], right. The CO2 adduct might be more correctly described as 2[CoII(κ5-L1)(CO22−)].

[CoL1] complex’s HOMO indicates ligand-based electron density; see Figure S41) and UV−vis spectrum and its CO and CO2 adducts here. The CO 2 adduct is distorted octahedral, and CO 2 coordinates on the opposite side of the terminal N2 with a Co−C distance of 2.032 Å. The distances between cobalt and the coordinating nitrogen atoms are between 2.014 and 2.404 Å (Figure 8 and Table S6). The adduct can be described as a 19electron system, 2[CoII(κ5-L1)(CO22−)]. The CO adduct cannot have a [Co0(κ5-L1)(CO)] coordination geometry with a 21-electron system; therefore, it changes to a distorted tetragonal [Co0(κ3-L1)(CO)] geometry with a dangling bipyridine moiety, as shown in Figure 8. Proposed Mechanism of Photocatalytic Reactions. The extensive characterization and mechanistic experiments discussed above can be compiled to illustrate likely mechanisms for CO2 reduction catalysis. The Co(II) precatalyst is reduced inefficiently as shown above, consistent with the requisite spin flip from the high-spin quartet Co(II) to the low-spin singlet Co(I). CV data and CO2 binding experiments showed that the doubly reduced “Co(0)” oxidation state is needed to initiate catalysis; however, the [Ru(bpy)3]2+/+ couple, relevant in photochemical electron transfer, is about 200 mV more positive than the Co(I/0) potential of complex 2 (Figure S39). The moderately endergonic reduction of Co(I) by [Ru(bpy)3]+ can be driven by fast subsequent steps in the reaction; however, the radical products of the one-electron oxidation of triethylamine. TEA•+ (Et3N•+), and Et2NC•HCH3 can also reduce the complex in the photochemical milieu. The standard potentials of these radicals are not known. The following reactions might occur to initiate catalysis.

observed electrochemically and the observation of an approximately 1:1 mixture of species when the Na/Hgprepared Co(I) was exposed to CO. In that experiment, disproportionation of the [CoIL2(CO)]+ species yields [CoIIL2]2+ and [Co0L2(CO)]. No evidence for the reaction between [CoIL2]+ and CO2 was found during bulk electrolysis in 12CO2-saturated solutions at a potential of about 100 mV more negative than the Co(II/I) couple, which is consistent with CV measurements (see above). However, at potentials more negative than Co(I/0), the 1894 cm−1 peak of [Co0L2(CO)] was observed. Normalizing the intensities to the absorption of the CC stretching shows that about 30% of the catalyst exists as Co(0) carbonyl during continued electrolysis. In addition, vibrational frequencies for HCO3−/CO32− were evident at 1685, 1646, and 1304 cm−1.86 This observation indicates that at this potential (−2.05 V vs Fc+/0) complex 2 has entered the catalytic cycle, during which CO2 functions as the oxide acceptor to yield [CoIIL2(CO)]2+, which subsequently releases CO or converts to [Co0L2(CO)] by reduction. No clear spectroscopic signatures of other forms of the catalyst, e.g., [Co0L2(CO2)], were observed, suggesting that the reactions of these intermediates are not rate-limiting. The C−O vibrational frequencies of [CoL2(CO)] are found at 1849 cm−1 under 13CO2 (Figure S40). The result clearly shows that the generated CO originates from the reduction of CO2. Exposure of [Co0L2(CO)] to air resulted in a loss of absorbance of the CO stretch, but the UV−vis spectrum remained unchanged. Because the CO adduct is unlikely to be a 21-electron species, dissociation of one or more pyridyl ligands is possible. The nature of these species is investigated below by DFT calculations. DFT Calculations on Complex 1 and Its CO2, CO, and Hydride Complexes. DFT calculations on complex 1 and its reduced species were carried out at the B3LYP/LANL2DZ level of theory and the SMD continuum solvation model for CH3CN as the solvent. Because complexes 1 and 2 are very similar in terms of their electronic and geometric structures, we carried out calculations only for complex 1, but we believe our results also to be applicable to complex 2. The detailed computational procedure is shown in the SI, and calculated structures of 4[CoII(κ5-L1)]2+, 3[CoI(κ5-L1)]+, 1[CoI(κ5-L1)]+, 2 [CoI(κ5-L1•−)], 2[Co0(κ5-L1)(CO2)], 2[Co0(κ3-L1)(CO)], 1 [CoIII(κ5-L1)(H)]2+, and 2[CoII(κ5-L1)(H)]+ are summarized in Table S4. We discuss 2[Co0(κ5-L1)], which is actually 2 [CoI(κ5-L1•−)] based on the DFT calculations (i.e., the

[Ru(bpy)3 ]2 + + hν → [*Ru(bpy)3 ]2 +

(4)

[*Ru(bpy)3 ]2 + + TTA → [Ru(bpy)3 ]+ + TTA•+

(5)

TTA•+ + Et3N → TTA + Et3N•+

(6)

Et3N•+ + Et3N → Et3NH+ + Et 2NC•HCH3

(7)

2+

[Ru(bpy)3 ]+ + 4[CoIIL2]

+

→ [Ru(bpy)3 ]2 + + 1[CoIL2] I

(8) DOI: 10.1021/acs.inorgchem.8b00433 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry +

Et 2NC•HCH3 + 1[CoIL2]

→ Et 2N+CHCH3 + 2[CoI(L2•−)]

2+

→ 2[CoII(κ 3‐L)(CO)]

(9)

[CoI(L2•−)] + CO2 → 2[CoIIL2(CO2 2 −)]

2

CONCLUSIONS We have presented in this work a new type of cobalt complex with TBP geometry that can be a CO2 reduction catalyst for both electrochemical and photochemical approaches when water is added as a proton source. Efficient reductive quenching of the excited state of [Ru(bpy)3]2+ by the reversible TTA electron donor previously investigated by Shan and Schmehl45 has been applied to photocatalytic CO2 reduction. Unfortunately, the benefits of the TTA/TEA reductive quenching system were not realized to their fullest potential because the reduction of CO2 is much slower than the photochemical formation of [Ru(bpy)3]+. Nevertheless, we were able to observe several stoichiometric reactions and elucidate many of the mechanistic details of catalysis by a variety of spectroscopic techniques supported by DFT calculations. Knowledge of the reaction bottlenecks is particularly useful in advancing the next generation of catalysts, and several were determined. (1) A spin flip occurs during the reduction from Co(II) to Co(I), and the reaction is inefficient. (2) The reduction to “Co(0)” (more strictly [Co(I)L•−]) by [Ru(bpy)3]+ is endergonic and may require the potent reductant Et2NC•CH3 that is formed by a secondary reaction of Et3N•+. (3) A small structural change from a TBP to a distorted octahedral geometry occurs during the addition of CO2 to “Co(0)”. (4) A significant geometrical change involving the dissociation of two Co−N bonds is required during CO2 reduction to form the distorted tetrahedral [CoII(κ3-L)(CO)]2+ complex. (5) The CO formed may recoordinate to the reduced complex and act as an inhibitor.

[CoIIL(CO2 2 −)] + CO2 + CO + CO32 −

(11)

[CoIIL(CO2 2 −)] + CO2

■

2

2+

→ 2[CoII(κ 3‐L)(CO)] 2+

2

[CoII(κ 3‐L)(CO)] 2

II

+ CO32 −

(12)

2+

→ [Co (κ ‐L)(NCCH3)]

+ CO

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.8b00433. Detailed synthetic procedures, NMR spectra, summary of the crystal data, bond distances and angles, cyclic voltammograms, UV−vis and FT-IR spectra of various cobalt species, additional photochemical data and tables of DFT, CASSCF, and CASSCF+MP2 results, Cartesian coordinates of the C2-symmetrized crystal structure of 1, and DFT-calculated structures of key reduced species and their adducts (PDF)

(13)

[CoI(L•−)] + CO → 2[Co0(κ 3‐L)(CO)]

2

(14) +

[CoIIL(CO2 2 −)] + H+ → 2[CoII(κ 5‐L)(COOH)]

2

(15) 2

+

[CoII(κ 5‐L)(COOH)] + CO2 2+

→ 2[CoII(κ 3‐L)(CO)]

+ HCO3−

ASSOCIATED CONTENT

S Supporting Information *

+ CH3CN

3

(17)

■

2

2+

+ H 2O

Under electrochemical or photochemical conditions, the availability of protons to function as oxide acceptors must be considered. In this case, the protonation of Co(II) metallocarboxylate (eq 15) is followed by C−O bond cleavage induced by CO2 (eq 16) or H+ (eq 17). Both reactions yield the same short-lived Co(II) carbonyl species. Spectroelectrochemical data revealed the formation of bicarbonate during electrolysis as evidence for eq 16. The IR experiments also show the occurrence of eq 14 during catalytic conditions. While the Co(II) complex releases CO in exchange for CH3CN, Co(0) coordinates to CO, representing a mechanism of product inhibition because CO2 addition is precluded by this strong equilibrium.

(10)

As mentioned in the theoretical characterization above, the structures of 2[CoI(L1•−)] and 2[CoII(κ5-L1)(CO22−)] are distorted TBP and distorted octahedral, respectively, and these results should agree for complex 2. Therefore, the coordination of CO2 can proceed with minimal geometric constraints. On the other hand, 2[CoII(κ3-L1)(CO)]2+ (not observed experimentally) and 2[Co0(κ3-L1)(CO)] (observed experimentally) are distorted tetragonal, and generating 2[CoII(κ3-L1)(CO)]2+ from 2[CoII(κ5-L1)(CO22−)] requires cleavage of a C−O bond and significant structural rearrangement. In the Na/Hg preparation of 2[CoI(L2•−)] and its reaction with CO2, there are no protons available, and CO2 must function as the oxide acceptor via eq 11 or 12. Interestingly, the reaction of CO2(g) with the two-electron-reduced 2[CoII(κ5L1)(CO22−)] complex to form the spin-forbidden 4[Co(κ5L1)]2+ along with CO(g) and CO32−(solv) (eq 11) is predicted by DFT calculations to be only ca. 2.3 kcal mol−1 endergonic. Equation 12 is more plausible because the resultant UV−vis spectrum is different from the initial 4[CoIIL2]2+ spectrum. The net DFT free-energy change for eq 12 is −138.4 kcal mol−1 with L1, i.e., extremely exoergic, but also requiring a lot of rearrangement of the complex (κ5 to κ3) and reactants, which, consistent with experimental data, would probably be quite slow. The Co(II) carbonyl complex was not observed by IR, consistent with the rapid exchange for CH3CN, which is calculated to be exergonic by 10.6 kcal mol−1 (eq 13). Further evidence for exchange via eq 13 is found in the similarity between the UV spectra after stoichiometric reactions of 2 [CoI(L2•−)] with CO2 or CO [ΔG° = −5.6 kcal mol−1 with L1 (eq 14) and confirmed by IR data]. These spectra are identical and are characterized by three π−π* transitions consistent with a dangling L2 ligand in contrast to the single broad π−π* absorbance of 4[CoII(κ5-L2)]2+. The spinforbidden conversion of 2 [Co II (κ 3 -L2)(NCCH 3 )] 2+ to 4 [CoII(κ5-L2)]2+ and CH3CN was not observed in the stoichiometric reaction of Co0L2 with CO2, in agreement with the calculated ΔG° of 33.8 kcal mol−1.

→ 2[CoII(κ 5‐L)]

+

[CoII(κ 5‐L)(COOH)] + H+

2

(16) J


Article

Inorganic Chemistry Accession Codes

(11) Morris, A. J.; Meyer, G. J.; Fujita, E. Molecular Approaches to the Photocatalytic Reduction of Carbon Dioxide for Solar Fuels. Acc. Chem. Res. 2009, 42 (12), 1983−1994. (12) Schneider, J.; Jia, H. F.; Muckerman, J. T.; Fujita, E. Thermodynamics and kinetics of CO2, CO, and H+ binding to the metal centre of CO2 reduction catalysts. Chem. Soc. Rev. 2012, 41 (6), 2036−2051. (13) Yamazaki, Y.; Takeda, H.; Ishitani, O. Photocatalytic reduction of CO2 using metal complexes. J. Photochem. Photobiol., C 2015, 25, 106−137. (14) Manbeck, G. F.; Fujita, E. A review of iron and cobalt porphyrins, phthalocyanines and related complexes for electrochemical and photochemical reduction of carbon dioxide. J. Porphyrins Phthalocyanines 2015, 19 (1−3), 45−64. (15) Berardi, S.; Drouet, S.; Francas, L.; Gimbert-Surinach, C.; Guttentag, M.; Richmond, C.; Stoll, T.; Llobet, A. Molecular artificial photosynthesis. Chem. Soc. Rev. 2014, 43 (22), 7501−7519. (16) Qiao, J.; Liu, Y.; Hong, F.; Zhang, J. A review of catalysts for the electroreduction of carbon dioxide to produce low-carbon fuels. Chem. Soc. Rev. 2014, 43 (2), 631−675. (17) Das, S.; Wan Daud, W. M. A. A review on advances in photocatalysts towards CO2 conversion. RSC Adv. 2014, 4 (40), 20856−20893. (18) Tu, W.; Zhou, Y.; Zou, Z. Photocatalytic Conversion of CO2 into Renewable Hydrocarbon Fuels: State-of-the-Art Accomplishment, Challenges, and Prospects. Adv. Mater. 2014, 26 (27), 4607−4626. (19) Habisreutinger, S. N.; Schmidt-Mende, L.; Stolarczyk, J. K. Photocatalytic Reduction of CO2 on TiO2 and Other Semiconductors. Angew. Chem., Int. Ed. 2013, 52 (29), 7372−7408. (20) Costentin, C.; Robert, M.; Saveant, J.-M. Catalysis of the electrochemical reduction of carbon dioxide. Chem. Soc. Rev. 2013, 42 (6), 2423−2436. (21) Windle, C. D.; Perutz, R. N. Advances in molecular photocatalytic and electrocatalytic CO2 reduction. Coord. Chem. Rev. 2012, 256 (21−22), 2562−2570. (22) Kumar, B.; Llorente, M.; Froehlich, J.; Dang, T.; Sathrum, A.; Kubiak, C. P. Photochemical and Photoelectrochemical Reduction of CO2. Annu. Rev. Phys. Chem. 2012, 63, 541−569. (23) Izumi, Y. Recent advances in the photocatalytic conversion of carbon dioxide to fuels with water and/or hydrogen using solar energy and beyond. Coord. Chem. Rev. 2013, 257 (1), 171−186. (24) Finn, C.; Schnittger, S.; Yellowlees, L. J.; Love, J. B. Molecular approaches to the electrochemical reduction of carbon dioxide. Chem. Commun. 2012, 48 (10), 1392−1399. (25) Gattrell, M.; Gupta, N.; Co, A. A review of the aqueous electrochemical reduction of CO2 to hydrocarbons at copper. J. Electroanal. Chem. 2006, 594 (1), 1−19. (26) Tanaka, K.; Ooyama, D. Multi-electron reduction of CO2 via Ru-CO2, -C(O)OH, -CO, -CHO, and -CH2OH species. Coord. Chem. Rev. 2002, 226 (1−2), 211−218. (27) Schmidt, M. H.; Miskelly, G. M.; Lewis, N. S. Effects of redox potential, steric configuration, solvent, and alkali metal cations on the binding of carbon dioxide to cobalt(I) and nickel(I) macrocycles. J. Am. Chem. Soc. 1990, 112 (9), 3420−3426. (28) Fujita, E.; Creutz, C.; Sutin, N.; Szalda, D. J. Carbon Dioxide Activation by Cobalt(I) Macrocycles: Factors Affecting CO2 and CO Binding. J. Am. Chem. Soc. 1991, 113 (1), 343−353. (29) Fujita, E.; Creutz, C.; Sutin, N.; Brunschwig, B. S. Carbon Dioxide Activation by Cobalt Macrocycles: Evidence of HydrogenBonding Between Bound CO2 and the Macrocycle in Solution. Inorg. Chem. 1993, 32 (12), 2657−2662. (30) Fujita, E.; Furenlid, L. R.; Renner, M. W. Direct XANES evidence for charge transfer in Co-CO2 complexes. J. Am. Chem. Soc. 1997, 119 (19), 4549−4550. (31) Chapovetsky, A.; Do, T. H.; Haiges, R.; Takase, M. K.; Marinescu, S. C. Proton-Assisted Reduction of CO2 by Cobalt Aminopyridine Macrocycles. J. Am. Chem. Soc. 2016, 138 (18), 5765− 5768.

CCDC 1825268−1825269 contain the supplementary crystallographic data for this paper. These data can be obtained free of charge via www.ccdc.cam.ac.uk/data_request/cif, or by emailing [email protected], or by contacting The Cambridge Crystallographic Data Centre, 12 Union Road, Cambridge CB2 1EZ, UK; fax: +44 1223 336033.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. ORCID

Koichi Kodama: 0000-0002-8567-9360 Gerald F. Manbeck: 0000-0002-6632-3895 Etsuko Fujita: 0000-0002-0407-6307 Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We thank Drs. David Shaffer and M. Zahid Ertem for useful discussions. The work carried out at Brookhaven National Laboratory (BNL) was supported by the U.S. Department of Energy, Office of Science, Division of Chemical Sciences, Geosciences, & Biosciences, Office of Basic Energy Sciences, under Contract DE-SC0012704. T.S. thanks the Japan Public− Private Partnership Student Study Abroad Program “Tobitate! Young Ambassador Program” for partial financial support during her stay at BNL.

■

REFERENCES

(1) Olah, G. A. Beyond oil and gas: The methanol economy. Angew. Chem., Int. Ed. 2005, 44 (18), 2636−2639. (2) Lewis, N. S.; Nocera, D. G. Powering the Planet: Chemical Challenges in Solar Energy Utilization. Proc. Natl. Acad. Sci. U. S. A. 2006, 103, 15729−15735. (3) Balzani, V.; Credi, A.; Venturi, M. Photochemical conversion of solar energy. ChemSusChem 2008, 1 (1−2), 26−58. (4) Appel, A. M.; Bercaw, J. E.; Bocarsly, A. B.; Dobbek, H.; DuBois, D. L.; Dupuis, M.; Ferry, J. G.; Fujita, E.; Hille, R.; Kenis, P. J. A.; Kerfeld, C. A.; Morris, R. H.; Peden, C. H. F.; Portis, A. R.; Ragsdale, S. W.; Rauchfuss, T. B.; Reek, J. N. H.; Seefeldt, L. C.; Thauer, R. K.; Waldrop, G. L. Frontiers, Opportunities, and Challenges in Biochemical and Chemical Catalysis of CO2 Fixation. Chem. Rev. 2013, 113 (8), 6621−6658. (5) Lewis, N. S. Introduction: Solar Energy Conversion. Chem. Rev. 2015, 115 (23), 12631−12632. (6) Assen, N. v. d.; Müller, L. J.; Steingrube, A.; Voll, P.; Bardow, A. Selecting CO2 Sources for CO2 Utilization by Environmental-MeritOrder Curves. Environ. Sci. Technol. 2016, 50, 1093−1101. (7) White, J. L.; Baruch, M. F.; Pander, J. E., III; Hu, Y.; Fortmeyer, I. C.; Park, J. E.; Zhang, T.; Liao, K.; Gu, J.; Yan, Y.; Shaw, T. W.; Abelev, E.; Bocarsly, A. B. Light-Driven Heterogeneous Reduction of Carbon Dioxide: Photocatalysts and Photoelectrodes. Chem. Rev. 2015, 115 (23), 12888−12935. (8) Wang, W.-H.; Himeda, Y.; Muckerman, J. T.; Manbeck, G. F.; Fujita, E. CO2 Hydrogenation to Formate and Methanol as an Alternative to Photo- and Electrochemical CO2 Reduction. Chem. Rev. 2015, 115 (23), 12936−12973. (9) Rakowski DuBois, M.; DuBois, D. L. Development of Molecular Electrocatalysts for CO2 Reduction and H2 Production/Oxidation. Acc. Chem. Res. 2009, 42 (12), 1974−82. (10) Benson, E. E.; Kubiak, C. P.; Sathrum, A. J.; Smieja, J. M. Electrocatalytic and homogeneous approaches to conversion of CO2 to liquid fuels. Chem. Soc. Rev. 2009, 38 (1), 89−99. K


Article

Inorganic Chemistry (32) Kramer, W. W.; McCrory, C. C. L. Polymer coordination promotes selective CO2 reduction by cobalt phthalocyanine. Chem. Sci. 2016, 7 (4), 2506−2515. (33) Elgrishi, N.; Chambers, M. B.; Artero, V.; Fontecave, M. Terpyridine complexes of first row transition metals and electrochemical reduction of CO2 to CO. Phys. Chem. Chem. Phys. 2014, 16 (27), 13635−13644. (34) Elgrishi, N.; Chambers, M. B.; Fontecave, M. Turning it off! Disfavouring hydrogen evolution to enhance selectivity for CO production during homogeneous CO2 reduction by cobalt-terpyridine complexes. Chem. Sci. 2015, 6 (4), 2522−2531. (35) Chan, S. L. F.; Lam, T. L.; Yang, C.; Yan, S. C.; Cheng, N. M. A robust and efficient cobalt molecular catalyst for CO2 reduction. Chem. Commun. 2015, 51 (37), 7799−7801. (36) Lin, S.; Diercks, C. S.; Zhang, Y. B.; Kornienko, N.; Nichols, E. M.; Zhao, Y. B.; Paris, A. R.; Kim, D.; Yang, P.; Yaghi, O. M.; Chang, C. J. Covalent organic frameworks comprising cobalt porphyrins for catalytic CO2 reduction in water. Science 2015, 349 (6253), 1208− 1213. (37) Isaacs, M.; Canales, J. C.; Riquelme, A.; Lucero, M.; Aguirre, M. J.; Costamagna, J. Contribution of the ligand to the electroreduction of CO2 catalyzed by a cobalt(II) macrocyclic complex. J. Coord. Chem. 2003, 56 (14), 1193−1201. (38) Isaacs, M.; Armijo, F.; Ramirez, G.; Trollund, E.; Biaggio, S. R.; Costamagna, J.; Aguirre, M. J. Electrochemical reduction of CO2 mediated by poly-M-aminophthalocyanines (M = Co, Ni, Fe): polyCo-tetraaminophthalocyanine, a selective catalyst. J. Mol. Catal. A: Chem. 2005, 229 (1−2), 249−257. (39) 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 CO2. J. Am. Chem. Soc. 1993, 115 (2), 601−609. (40) Ogata, T.; Yanagida, S.; Brunschwig, B. S.; Fujita, E. Mechanistic and Kinetic Studies of Cobalt Macrocycles in a Photochemical CO2 Reduction System: Evidence of Co-CO2 Adducts as intermediates. J. Am. Chem. Soc. 1995, 117 (25), 6708−6716. (41) Ogata, T.; Yamamoto, Y.; Wada, Y.; Murakoshi, K.; Kusaba, M.; Nakashima, N.; Ishida, A.; Takamuku, S.; Yanagida, S. PhenazinePhotosensensitized Reduction of CO2 Mediated by a Cobalt-Cyclam Conmplex Through Electron and Hydrogen-Transfer. J. Phys. Chem. 1995, 99 (31), 11916−11922. (42) Ouyang, T.; Hou, C.; Wang, J.-W.; Liu, W.-J.; Zhong, D.-C.; Ke, Z.-F.; Lu, T.-B. A Highly Selective and Robust Co(II)-Based Homogeneous Catalyst for Reduction of CO2 to CO in CH3CN/ H2O Solution Driven by Visible Light. Inorg. Chem. 2017, 56, 7307− 7311. (43) Chan, S. L. F.; Lam, T. L.; Yang, C.; Lai, J.; Cao, B.; Zhou, Z. Y.; Zhu, Q. H. Cobalt(II) tris(2-pyridylmethyl)amine complexes [Co(TPA)X]+ bearing coordinating anion (X = Cl−, Br−, I− and NCS−): synthesis and application for carbon dioxide reduction. Polyhedron 2017, 125, 156−163. (44) Tinnemans, A. H. A.; Koster, T. P. M.; Thewissen, D.; Mackor, A. Tetraaza-Macrocyclic Cobalt(II) and Nickel(II) Complexes as Electron-Transfer Agents in the Photoelectrochemical and Electrochemical Reduction of Carbon Dioxide. Recl. Trav. Chim. Pays-Bas 1984, 103, 288−295. (45) Shan, B.; Schmehl, R. Photochemical Generation of Strong One-Electron Reductants via Light-Induced Electron Transfer with Reversible Donors Followed by Cross Reaction with Sacrificial Donors. J. Phys. Chem. A 2014, 118, 10400−10406. (46) Braddock, J. N.; Meyer, T. J. Kinetics of the Oxidation of Fe(H2O)62+ by Polypyridine Complexes of Ruthenium(III). Negative Enthalpies of Activation. J. Am. Chem. Soc. 1973, 95, 3158−3162. (47) Lin, C.-T.; Bottcher, W.; Chou, M.; Creutz, C.; Sutin, N. Mechanism of the Quenching of the Emission of Substituted Polypyridineruthenium(II) Complexes by Iron(III), Chromium(III), and Europium(III) Ions. J. Am. Chem. Soc. 1976, 98, 6536−6544.

(48) Hamelin, O.; Menage, S.; Charnay, F.; Chavarot, M.; Pierre, J.L.; Pecaut, J.; Fontecave, M. New polydentate ligand and catalytic properties of the corresponding ruthenium complex during sulfoxidation and alkene epoxidation. Inorg. Chem. 2008, 47 (14), 6413− 6420. (49) Evans, D. F. 400. The determination of the paramagnetic susceptibility of substances in solution by nuclear magnetic resonance. J. Chem. Soc. 1959, 2003−2005. (50) Schubert, E. M. Utilizing the Evans method with a superconducting NMR spectrometer in the undergraduate laboratory. J. Chem. Educ. 1992, 69, 62−62. (51) Grant, D. H. Paramagnetic Susceptibility by NMR: The ″Solvent Correction″ Reexamined. J. Chem. Educ. 1995, 72, 39−40. (52) Bain, G. A.; Berry, J. F. Diamagnetic Corrections and Pascal’s Constants. J. Chem. Educ. 2008, 85, 532−536. (53) Piguet, C. Paramagnetic Susceptibility by NMR: The ″Solvent Correction″ Removed for Large Paramagnetic Molecules. J. Chem. Educ. 1997, 74 (7), 815−816. (54) Sheldrick, G. M. SHELXL, 5th ed.; Siemens Analytical Instruments Inc.: Madison, WI, 1994. (55) Sheldrick, G. M. SADABS; Bruker AXS Inc., 2007. (56) Frisch, M. J.; Trucks, G. W.; Schlegel, H. B.; Scuseria, G. E.; Robb, M. A.; Cheeseman, J. R.; Scalmani, G.; Barone, V.; Mennucci, B.; Petersson, G. A.; Nakatsuji, H.; Caricato, M.; Li, X.; Hratchian, H. P.; Izmaylov, A. F.; Bloino, J.; Zheng, G.; Sonnenberg, J. L.; Hada, M.; Ehara, M.; Toyota, K.; Fukuda, R.; Hasegawa, J.; Ishida, M.; Nakajima, T.; Honda, Y.; Kitao, O.; Nakai, H.; Vreven, T.; Montgomery, J. A., Jr.; Peralta, J. E.; Ogliaro, F.; Bearpark, M.; Heyd, J. J.; Brothers, E.; Kudin, K. N.; Staroverov, V. N.; Kobayashi, R.; Normand, J.; Raghavachari, K.; Rendell, A.; Burant, J. C.; Iyengar, S. S.; Tomasi, J.; Cossi, M.; Rega, N.; Millam, J. M.; Klene, M.; Knox, J. E.; Cross, J. B.; Bakken, V.; Adamo, C.; Jaramillo, J.; Gomperts, R.; Stratmann, R. E.; Yazyev, O.; Austin, A. J.; Cammi, R.; Pomelli, C.; Ochterski, J. W.; Martin, R. L.; Morokuma, K.; Zakrzewski, V. G.; Voth, G. A.; Salvador, P.; Dannenberg, J. J.; Dapprich, S.; Daniels, A. D.; Farkas, Ö .; Foresman, J. B.; Ortiz, J. V.; Cioslowski, J.; Fox, D. J. Gaussian 09, revision D.01; Gaussian, Inc.: Wallingford, CT, 2013. (57) Becke, A. D. Density-Functional Exchange-Energy Approximation with Correct Asymptotic Behavior. Phys. Rev. A: At., Mol., Opt. Phys. 1988, 38 (6), 3098−3100. (58) Lee, C. T.; Yang, W. T.; Parr, R. G. Development of the ColleSalvetti Correction-Energy Formula into a Funcitional of the ElectronDensity. Phys. Rev. B: Condens. Matter Mater. Phys. 1988, 37 (2), 785− 789. (59) Vosko, S. H.; Wilk, L.; Nusair, M. Accurate spin-dependent electron liquid correlation energies for local spin density calculations: a critical analysis. Can. J. Phys. 1980, 58 (8), 1200−1211. (60) Becke, A. D. Density-functional thermochemistry. III. The role of exact exchange. J. Chem. Phys. 1993, 98 (7), 5648−5652. (61) Dunning, T. H. J.; Hay, P. J. In Modern Theoretical Chemistry; Schaefer, H. F., III, Ed.; Plenum: New York, 1976; Vol. 3, pp 1−28. (62) Hay, P. J.; Wadt, W. R. Ab Initio Effective Core Potentials for Molecular Calculations - Potnetials for the Transition-Metal Atoms Sc to Hg. J. Chem. Phys. 1985, 82 (1), 270−283. (63) Wadt, W. R.; Hay, P. J. Ab initio effective core potentials for molecular calculations. Potentials for main group elements Na to Bi. J. Chem. Phys. 1985, 82, 284−298. (64) Hay, P. J.; Wadt, W. R. Ab initio effective core potentials for molecular calculations. Potentials for K to Au including the outermost core orbitals. J. Chem. Phys. 1985, 82, 299−310. (65) Marenich, A. V.; Cramer, C. J.; Truhlar, D. G. Universal Solvation Model Based on Solute Electron Density and on a Continuum Model of the Solvent Defined by the Bulk Dielectric Constant and Atomic Surface Tensions. J. Phys. Chem. B 2009, 113 (18), 6378−6396. (66) Hegarty, D.; Robb, M. A. Application of unitary group-methods to configuration-interaction calculations. Mol. Phys. 1979, 38, 1795− 1812. L


Article

Inorganic Chemistry (67) Eade, R. H. A.; Robb, M. A. Direct minimization in MC SCF theory - the Quasi-Newton method. Chem. Phys. Lett. 1981, 83, 362− 368. (68) Schlegel, H. B.; Robb, M. A. MC SCF gradient optimization of the H2CO –> H2 + CO transition structure. Chem. Phys. Lett. 1982, 93, 43−46. (69) Bernardi, F.; Bottoni, A.; McDouall, J. J. W.; Robb, M. A.; Schlegel, H. B. MCSCF gradient calculation of transition structures in organic reactions. Faraday Symp. Chem. Soc. 1984, 19, 137−147. (70) Frisch, M. J.; Ragazos, I. N.; Robb, M. A.; Bernhard Schlegel, H. An Evaluation of 3 Direct MC-SCF Procedures. Chem. Phys. Lett. 1992, 189, 524−528. (71) Yamamoto, N.; Vreven, T.; Robb, M. A.; Frisch, M. J.; Bernhard Schlegel, H. A Direct Derivative MC-SCF Procedure. Chem. Phys. Lett. 1996, 250, 373−378. (72) Suh, M. P.; Lee, J.; Han, M. Y.; Yoon, T. S. Template synthesis, properties, and crystal structure of a trigonal bipyramidal cobalt(II) octaazamacrobicyclic complex. Inorg. Chem. 1997, 36 (24), 5651− 5654. (73) Comba, P.; Haaf, C.; Wadepohl, H. Novel Bispidine Ligands and Their First-Row Transition Metal Complexes: Trigonal Bipyramidal and Trigonal Prismatic Geometries. Inorg. Chem. 2009, 48 (14), 6604−6614. (74) Song, X.-W.; Meng, Y.; Zhang, C.-L.; Ma, C.-B.; Chen, C.-N. A cobalt complex of a pentadentate aminopyridine ligand as an efficient catalyst for photocatalytic hydrogen generation. Inorg. Chem. Commun. 2017, 76, 52−54. (75) Lim, J. W.; Mikuriya, M.; Sakiyama, H. Trigonal-bipyramidal geometry in a cobalt(II) complex with an unsymmetrical tripodal ligand. Bull. Chem. Soc. Jpn. 2001, 74 (11), 2131−2132. (76) Thompson, L. K.; Ramaswamy, B. S.; Seymour, E. A. Cobalt(II) and zinc(II) complexes of the ’tripod’ ligand tris(2benzimidazylmethyl)amine. Some five-coordinate derivatives and some with mixed stereochemistries. Can. J. Chem. 1977, 55, 878−888. (77) Fujita, E.; Szalda, D. J.; Creutz, C.; Sutin, N. Carbon Dioxide Activation: Thermodynamics of CO2 Binding and the Involvement of Two Cobalt Centers in the Reduction of CO2 by a Cobalt(I) Macrocycle. J. Am. Chem. Soc. 1988, 110 (14), 4870−4871. (78) Chen, Z.; Chen, C.; Weinberg, D. R.; Kang, P.; Concepcion, J. J.; Harrison, D. P.; Brookhart, M. S.; Meyer, T. J. Electrocatalytic reduction of CO2 to CO by polypyridyl ruthenium complexes. Chem. Commun. 2011, 47, 12607−12609. (79) Smieja, J. M.; Kubiak, C. P. Re(bipy-tBu)(CO)3Cl-improved Catalytic Activity for Reduction of Carbon Dioxide: IR-Spectroelectrochemical and Mechanistic Studies. Inorg. Chem. 2010, 49 (20), 9283−9289. (80) Sampson, M. D.; Kubiak, C. P. Manganese Electrocatalysts with Bulky Bipyridine Ligands: Utilizing Lewis Acids To Promote Carbon Dioxide Reduction at Low Overpotentials. J. Am. Chem. Soc. 2016, 138 (4), 1386−1393. (81) Machan, C. W.; Sampson, M. D.; Kubiak, C. P. A Molecular Ruthenium Electrocatalyst for the Reduction of Carbon Dioxide to CO and Formate. J. Am. Chem. Soc. 2015, 137 (26), 8564−8571. (82) Costentin, C.; Drouet, S.; Robert, M.; Saveant, J. M. A Local Proton Source Enhances CO2 Electroreduction to CO by a Molecular Fe Catalyst. Science 2012, 338, 90−94. (83) Costentin, C.; Drouet, S.; Passard, G.; Robert, M.; Saveant, J. M. Proton-Coupled Electron Transfer Cleavage of Heavy-Atom Bonds in Electrocatalytic Processes. Cleavage of a C-O Bond in the Catalyzed Electrochemical Reduction of CO2. J. Am. Chem. Soc. 2013, 135 (24), 9023−9031. (84) Costentin, C.; Passard, G.; Robert, M.; Saveant, J. M. Pendant Acid-Base Groups in Molecular Catalysts: H-Bond Promoters or Proton Relays? Mechanisms of the Conversion of CO2 to CO by Electrogenerated Iron(0)Porphyrins Bearing Prepositioned Phenol Functionalities. J. Am. Chem. Soc. 2014, 136 (33), 11821−11829. (85) Tomita, Y.; Teruya, S.; Koga, O.; Hori, Y. Electrochemical Reduction of Carbon Dioxide at a Platinum Electrode in AcetonitrileWater Mixtures. J. Electrochem. Soc. 2000, 147, 4164−4167.

(86) Cheng, S. C.; Blaine, C. A.; Hill, M. G.; Mann, K. R. Electrochemical and IR Spectroelectrochemical Studies of the Electrocatalytic Reduction of Carbon Dioxide by [Ir2(dimen)4]2+ (dimen) 1,8-Diisocyanomenthane. Inorg. Chem. 1996, 35, 7704−7708. (87) Mussini, T.; Longhi, P.; Rondinini, S. Standard Potentials of Amalgam Electrodes in Aqueous Solutions, Temperature Coefficients and Activity Coefficients of Metals in Mercury. Pure Appl. Chem. 1985, 57, 169−179. (88) Meerholz, K.; Heinze, J. Multiple Reversible Electrochemical Reduction of Aromatic Hydrocarbons in Liquid Alkylamines. J. Am. Chem. Soc. 1989, 111, 2325−2326.

M


Photocatalytic CO2 Reduction by Trigonal-Bipyramidal Cobalt(II

Recommend Documents