Electrocatalytic Reduction of CO2 by Group 6 M ... - ACS Publications

May 26, 2016 - Gaia Neri , Paul M. Donaldson , and Alexander J. Cowan ... Victor S. Batista , Kyle A. Grice , Mehmed Z. Ertem , and Alfredo M. Angeles...
0 downloads 0 Views 716KB Size
Download PDF
Article pubs.acs.org/IC

Electrocatalytic Reduction of CO2 by Group 6 M(CO)6 Species without “Non-Innocent” Ligands Kyle A. Grice* and Cesar Saucedo Department of Chemistry, DePaul University, 1110 West Belden Avenue, McGowan South Suite 100, Chicago, Illinois 60614, United States S Supporting Information *

ABSTRACT: To understand the electrocatalytic CO2 reduction of metal carbonyl complexes without “non-innocent” ligands, the electrochemical responses of group 6 M(CO)6 (M = Cr, Mo, or W) and group 7 M2(CO)10 (M = Mn or Re) complexes were examined under Ar and CO2 at a glassy carbon electrode. All of the complexes showed changes in their cyclic voltammograms under CO2. The group 6 hexacarbonyl species show a significant increase in current under CO2 during metal-based reduction, corresponding to catalytic reduction of CO2. Bulk electrolysis experiments with Mo(CO)6 showed that CO was the primary product. The group 7 dimers showed very little change during metal-based reduction, but return oxidation responses disappeared, indicative of a chemical reaction after exposure to CO2 without catalysis. Addition of H2O, a proton source, to the solutions under CO2 decreased the catalytic current of the group 6 carbonyls and had no effect on the responses of the group 7 carbonyls. The group 6 M(CO)6 species are notable in that that they are effective catalysts without the need for an added “non-innocent” ligand such as 2,2′-bipyridine.

■

INTRODUCTION The catalytic reduction of CO2 to fuels and commodity chemicals has garnered significant interest in recent years. One approach to CO2 reduction is the use of homogeneous electrocatalysts, which are of particular interest because they can be operated with renewable energy sources such as solar power and are amenable to study with a variety of analytical techniques.1−3 Significant effort has been spent on complexes with metals from groups 7−10 (e.g., Mn, Re, Fe, Ru, Co, and Ni), and many different electrocatalytic systems have been explored. However, reports of electrocatalysts based on homogeneous complexes of the earlier metals (groups 3−6) are much rarer, and this area represents a direction of substantial promise. Complexes of group 6 metals are particularly attractive because of their presence in nature, such as the molybdenum center in the active sites of enzymes capable of interconverting CO2 and formate.4,5 Recent reports have indicated that group 6 M(bpy)(CO)4 (M = Cr, Mo, or W; bpy = substituted or parent 2,2′-bipyridyl ligand) species, as well as Mo(L)(CO)4 and W(L)(CO)4 complexes coordinated by similar “non-innocent” ligands are competent electrocatalysts for the reduction of CO2 to CO (Chart 1).6−9 These complexes are isoelectronic and isostructural with the well-characterized and highly effective group 7 Re(bpy)(CO)3X and Mn(bpy)(CO)3X electrocatalysts (X = anionic ligand such as Cl− or neutral ligand with a counterion) that have been studied for several decades.10 The “non-innocent” natures of bipyridine or imine ligands are reflected in the ability of the ligands to participate directly in the catalytic cycle by accepting and donating electrons.11,12 For © 2016 American Chemical Society

example, upon two-electron reduction of Re(bpy)(CO)3Cl to form the catalytically active Re(bpy)(CO)3−, one electron ends up on the metal and the other resides in the π* orbital of the bipyridine ligand.13 There are many other examples of noninnocent ligands in CO2 reduction,14,15 and non-innocent ligands can play important roles in a variety of catalytic processes.16 In addition to the complexes shown in Chart 1, heterogeneous electrochemical reduction of CO2 by molybdenum metal and heterogeneous molybdenum oxides has also been reported,17−19 as have heterogenized monomeric Cr complexes for CO2 reduction.20 These reports of group 6 complexes led us to reexamine the electrochemistry of the chromium, molybdenum, and tungsten hexacarbonyls. We were also interested in the reactivity of the group 6 hexacarbonyl M(CO)6 complexes because of promising reports in the literature. The parent group 6 hexacarbonyls have been shown to be precursors to CO2 reduction reactions via more than one pathway. N. John Cooper and co-workers have shown that the doubly reduced M(CO)52− (M = Cr, Mo, or W) dianions stoichiometrically react with CO2 to yield M(CO)6 and carbonate.21 Darensbourg has shown that the anionic hydride species Cr(CO)5H− can insert CO2 to produce formate.22 Tyler and co-workers have also shown that W(CO) 5 L − anions produced photochemically from W2(CO)102− and a monodentate ligand can reduce CO2 to CO and formate.23 These stoichiometric approaches suggest Received: April 8, 2016 Published: May 26, 2016 6240

DOI: 10.1021/acs.inorgchem.6b00875 Inorg. Chem. 2016, 55, 6240−6246

Article

Inorganic Chemistry Chart 1. “Non-Innocent” Ligands Used on M(L)(CO)4 Species for CO2 Reduction

recorded with an ABB FTLA2000 spectrometer using a Buck Scientific demountable liquid cell equipped with KBr windows and a 0.1 mm spacer. 1H and 13C{1H} nuclear magnetic resonance (NMR) spectra were acquired with a Bruker Avance 300 MHz NMR spectrometer at 298 K. Gas chromatography was performed on an Agilent 7820 instrument equipped with a HP-Molesieve column and a gas sampling valve. A TCD detector was used with N2 as the carrier gas. Calibration curves for carbon monoxide and hydrogen were created by injecting known quantities of CO or H2 into the electrochemical cell and then sampling the headspace (see Figures S22 and S23 of the Supporting Information). Tetrabutylammonium oxalate was synthesized as described in literature reports, and spectral data matched literature values.34,35 Electrochemistry. Electrochemical experiments were performed with an eDaq ER466 potentiostat. For cyclic voltammograms, the experiments were performed using 2.5 mL of a solution in a 5 mL cell with a 1 mm diameter glassy carbon working electrode or a 1 mm diameter platinum working electrode for the Pt experiments, a Ag/ AgCl or Ag wire pseudoreference, and a Pt/Ti alloy rod or Pt wire counter electrode. The working electrode was cleaned with diamond polish prior to each use. Unless otherwise noted, cyclic voltammograms were recorded at a scan rate of 250 mV/s using the metal complexes of interest at 5 mM. Ferrocene (Fc) was added to cyclic voltammograms as an internal reference (the E1/2 of the Fc+/Fc couple was set to 0 V). Reduction potentials versus the saturated calomel electrode (SCE) can be determined by adding 380 mV (in acetonitrile) or 470 mV (in N,N-dimethylformamide) to the values referenced to the Fc+/Fc couple.36,37 Solutions were sparged with Ar or CO2 for several minutes prior to analysis. Controlled potential electrolysis experiments were performed in a 25 mL three-neck roundbottom flask using 25 mL of 0.1 M tetrabutylammonium hexafluorophosphate in acetonitrile with a carbon rod working electrode (1.9 cm2 surface area submerged in solution), a Pt wire counter electrode, and a Ag wire pseudoreference. The pseudoreference electrode was separated from the bulk of the stirring solution with a porous CoralPor frit (purchased from Bioanalytical Systems Inc.), and the counter was protected from the bulk of the solution and the working electrode with a medium-porosity glass frit. For experiments with Pt as the working electrode, a coiled Pt wire was used in place of the carbon rod. The glassware was oven-dried prior to use, and new septa were secured in place with electrical tape for each experiment. Solutions were analyzed by IR and NMR after bulk electrolysis. For 1H and 13C{1H} NMR spectroscopy analysis, 100 μL of CD3CN (with 1% TMS) was added as a lock solvent to 400 μL of the reaction solution in an NMR tube.

that the group 6 M(CO)6 complexes could potentially be adapted to an electrocatalytic system. The electrochemical behaviors of group 6 M(CO)6 species under reducing potentials have been studied in aprotic media, dating back to the work of Dessy and co-workers in the 1960s.24 Pickett and Pletcher’s report a decade later,25 Seurat’s work,26 Milo and Gautier’s report,27 and, more recently, Amatore and co-workers’ report28 also examined the electrochemistry of group 6 hexacarbonyls. In a study that stands out as an early mention of electrocatalytic CO2 reduction, Pickett and Pletcher reported an electrochemical study of Mo and Cr hexacarbonyl species under CO2 at a platinum metal electrode in acetonitrile.29 The authors stated that the primary products were oxalate, which seems to be at odds with the findings of Cooper, Darensbourg, and Tyler mentioned in the paragraph above, none of whom observe oxalate. The heterogeneous reduction of CO2 at the platinum electrode has been reported to form oxalate in acetonitrile,30,31 but even the observation of oxalate production at platinum has been disputed.32 Homogeneous electrocatalysts for the formation of oxalate from CO2 are exceedingly rare.33 It was unclear if the M(CO)6 species were acting as homogeneous electrocatalysts in the system or if oxalate was even being formed. To examine this, we explored the electrochemistry of M(CO)6 (M = Cr, Mo, or W) at a glassy carbon electrode. We also explored the electrochemistry of the M2(CO)10 (M = Mn or Re) species to determine if these group 7 compounds were capable of CO2 reduction without the use of the “non-innocent” ligands, and our findings are reported below.

■

EXPERIMENTAL SECTION

General Considerations. Unless otherwise stated, all chemicals were used as provided by commercial suppliers. Acetonitrile was purified using an Innovative Technology PureSolv system under N2. Dimethylformamide (DMF) was purchased in anhydrous form and stored over activated molecular sieves prior to use. Tetrabutylammonium hexafluorophosphate (TBAH) was recrystallized from highperformance liquid chromatography-grade methanol and dried in a vacuum oven prior to use. Carbon dioxide (“bone dry”) and argon were purchased from American Gases Corp. Commercial ferrocene was purified by sublimation prior to use. Infrared spectra were 6241

DOI: 10.1021/acs.inorgchem.6b00875 Inorg. Chem. 2016, 55, 6240−6246

Article

Inorganic Chemistry

■

RESULTS AND DISCUSSION Cyclic Voltammetry of Group 6 and Group 7 Carbonyl Compounds under Ar and CO2. Cyclic voltammetry of Cr(CO)6, Mo(CO)6, and W(CO)6 at a glassy carbon electrode under Ar and CO2 was performed in both N,N-dimethylformamide (DMF) and acetonitrile (ACN). The electrochemistry under Ar in DMF (Table 1) and ACN (Table S1 in the

(Table 1). We performed scan rate studies under CO2 and found that the current was relatively constant at the scan rates studied compared to faster or slower scan rate regimes (Figures S11 and S12). Via comparison among the three species, the Mo(CO)6 complex displays the largest catalytic response in DMF as determined by icat/ip values. Interestingly, these icat/ip values are all considerably higher than what was reported by Pickett and Pletcher, a value of ∼3 for Mo(CO)6 using platinum as a working electrode. We examined the CV of Mo(CO)6 using a platinum working electrode in acetonitrile and also obtained an icat/ip value of ∼3 (Figure S13), which was much smaller than that at a glassy carbon working electrode in acetonitrile (icat/ip = 27.9). The lower current seen with a platinum electrode is likely due to poisoning of the electrode by CO released by the complex upon reduction. The overall electrochemical stability of the group 6 compounds under CO2 was assessed by performing 50 sequential cyclic sweeps on each catalyst in acetonitrile under CO2 at a glassy carbon electrode (Figures S8−S10). The peak location of the CO2 reduction did not significantly change, and a decrease in current height coincided with the appearance of gas bubbles at the electrode. Bubbles at the electrode result in a smaller surface area, which gives a smaller current. The current was restored when the solution was agitated to dislodge the bubbles for the Mo and W compounds, suggesting that these complexes do not degrade or rapidly deposit on the electrode. The activity of Cr(CO)6 did not appear to be restored, indicating that this compound was not stable over longer electrochemical experiments. After the 50 cycles, the glassy carbon working electrode for each catalyst was then rinsed with acetonitrile, placed in a fresh solution of TBAH and acetonitrile, and examined by cyclic voltammetry. No electrochemical responses were observed under CO2 for these electrodes, indicating that the active catalysts do not deposit on the electrode. It was noted that the return wave at the catalytic peak in our cyclic voltammograms under CO2 was not always reproducible, sometimes crossing over the reductive wave as in Figure 1, but the forward sweep, corresponding to a linear sweep voltammogram, was reproducible and consistent. We attribute the irreproducible return wave behavior to the sensitive catalytic intermediates and other species formed during reduction under CO2. Inhibition due to products or reactions of intermediates followed by subsequent reactivation at further potentials can also explain this behavior, a motif that has been observed in electrocatalysis, particularly oxidations.41−43 The electrochemistry of the group 7 Re2(CO)10 and Mn2(CO)10 dimers was also explored under Ar and CO2. These group 7 dimers can be electrochemically reduced to generate M(CO)5− anions that are isoelectronic with the group 6 M(CO)52− dianions, and the electrochemical behavior of the group 7 dimers under argon at a glassy carbon electrode matched literature reports.25,44−46 In stark contrast to the group 6 complexes, the Mn and Re species did not show catalytic behavior in the voltammograms under CO2. However, the return oxidation behaviors changed under CO2, indicative of reaction with CO2 after reduction, but without catalytic turnover (Figures S15 and S16). Cyclic Voltammetry under CO2 with Added Water. The addition of proton sources to group 7 CO2 reduction catalysts such as Mn(bpy)(CO)3X species has been shown to increase the rate of CO2 electrocatalysis, and in some cases, catalysis will not proceed without added proton sources.47−49

Table 1. Electrochemical Behavior of Group 6 Species under CO2 in DMF at 250 mV/s species

ip vs Fc under Ar (V)

ip vs Fc under CO2 (V)

icat/ip

estimated TOF (s−1)

Cr(CO)6 Mo(CO)6 W(CO)6

−2.745 −2.603 −2.475

−3.175 −3.248 −3.080

7.0 22 18

24 238 159

Supporting Information) is consistent with the previous reports at platinum or mercury electrodes,26 displaying an irreversible reduction at negative potentials and a small oxidation peak at more positive potentials at the return sweep. The reduction wave corresponds to overall reduction by two electrons, loss of CO, and formation of pentacarbonyl dianion.27 Scan rate dependence studies of Mo(CO)6 indicated that this reduction corresponds to a freely diffusing species, as there is a linear relationship between peak current and the square root of the scan rate (see Figure S1). When these species were exposed to an atmosphere of CO2, the current significantly increased at the reduction wave in the cyclic voltammograms [see Figure 1 for Mo(CO)6 in DMF; the other group 6 species displayed very similar responses in both DMF and ACN (Figures S3−S7)].

Figure 1. Cyclic voltammetry of Mo(CO)6 in DMF under Ar and under CO2 at 250 mV/s with a glassy carbon working electrode.

Cyclic voltammetry can be used to analyze the catalytic activity of electrocatalysts. The icat/ip ratio is often used to characterize catalyst turnover frequencies (TOFs) but can only be used when there is a reversible response for ip in a specific electrochemical regime.38−40 Therefore, we are hesitant to draw conclusions about the specific activity (i.e., specific turnover frequency values for comparison to those of other catalysts) from the icat/ip values for these species but report estimated TOFs for comparison between the hexacarbonyl complexes 6242

DOI: 10.1021/acs.inorgchem.6b00875 Inorg. Chem. 2016, 55, 6240−6246

Article

Inorganic Chemistry

Controlled Potential Electrolysis. To examine the products of CO2 reduction, controlled potential electrolysis (“bulk electrolysis”) experiments were performed on Mo(CO)6, the species with the largest icat/ip ratio in DMF. The experiments were performed in ACN because it offered a better FTIR spectral window compared to that in DMF. In the IR cell used, DMF/TBAH was found to completely obscure the regions of 1000−1800 and 2730−3100 cm−1 whereas ACN/ TBAH obscured only the regions of 1000−1070, 1330−1540, 2230−2310, and 2900−3050 cm−1. Controlled potential electrolyses were performed with stirring 1 mM solutions of Mo(CO)6 under CO2 or Ar in ACN/TBAH. At an applied potential of −2.8 V versus Fc (−2.4 V vs SCE) using a carbon rod with a submerged surface area of 1.9 ± 0.1 cm2, rapid bubble formation was observed at the electrode throughout the course of the experiment. The headspaces of the bulk electrolysis experiments were examined by gas chromatography after 30 min, and carbon monoxide was observed with 83−95% faradaic efficiency based on 2 mol of e−/mol of CO (Figure 3).

Therefore, a proton source was added to the M(CO)6 (M = Cr, Mo, or W) and M2(CO)10 (Mn or Re) species to determine the effect of added protons on CO2 reduction. Because of the negative potentials at which catalysis occurs with the group 6 compounds, water was used as the proton source. Stronger acids can be directly reduced at glassy carbon at the potentials used,50 which would complicate interpretation of the results. The electrochemistry of Mo(CO)6 and the other group 6 carbonyls under argon with added water showed very little change from the “dry” solutions. It should be noted that an anhydrous solution will absorb water during the course of sparging with gases,51 and therefore, even “dry” solutions likely had some concentration of water. Therefore, the direct reduction of water to form H2 is not catalyzed by the group 6 complexes on the cyclic voltammetry time scales that were used. However, under CO2 in the presence of increasing amounts of water (45, 90, and 135 μL in 2.5 mL of solution, corresponding to 0.98, 1.93, and 2.84 M H2O, respectively), the catalytic current decreased as compared to the electrochemical response under CO2 in the absence of added water, indicating suppression of catalytic activity (Figure 2). A scan rate study

Figure 3. Controlled potential electrolysis experiment showing current vs time for Mo(CO)6 in ACN under CO2 using a carbon rod electrode. This particular run showed a higher average current density (15.1 mA/cm2) with an 83% faradaic efficiency for CO.

Figure 2. Cyclic voltammograms of Mo(CO)6 in DMF solvent and CO2 with increasing amounts of water added at 250 mV/s, showing diminishing icat values with increasing amounts of water.

The currents were observed to fluctuate somewhat over time, but average current densities were found to be 7.9 mA/cm2, indicating significant and sustained catalysis. Hydrogen was found in trace amounts (≤1% faradaic efficiency). FTIR spectra of the solution after bulk electrolysis were recorded, and peaks consistent with carbonate or bicarbonate were observed at 1680 and 1646 cm−1.34 No evidence of formate was observed by NMR or FTIR in these experiments. The IR peak for the C−O stretches in Mo(CO)6 was also present at 98% of the initial signal strength even after bulk electrolysis under CO2 for 2 h, which produces 8.1 turnovers of CO, supporting the homogeneous electrocatalytic nature of Mo(CO)6 and its stability under the reaction conditions. To help verify the identities of the specific soluble products that were formed during the CO2 reduction bulk electrolysis, 13 CO2 was used during selected experiments. A 13C{1H} NMR spectrum of the solution after bulk electrolysis for 2 h showed only a signal assigned as carbonate (165 ppm), and no evidence of oxalate or formate was observed on the basis of 13C{1H}

was performed for Mo(CO)6 under CO2 in the presence of water, and we found that the catalytic peak was smaller in the presence of water at all scan rates (Figures S11 and S12). This diminished current with water could have at least two different causes. Water could be inhibiting catalysis by interacting with the metal center to block the site where CO2 binds. Alternatively, a different reaction could be occurring between CO2 and the reduced complex in the presence of water, thus shifting from one catalytic cycle to another. The behavior of the group 7 complexes was significantly different from that of the group 6 compounds. In the presence of added water under CO2, the voltammograms of Mn2(CO)10 and Re2(CO)10 complexes did not show significant differences from their voltammograms under CO2 in dry solvent (Figures S15 and S16). This was consistent with a recent report in which a Mn(L)(CO)3X species expels the L ligand under CO2 reduction catalysis and forms Mn(CO)5−, which does not show any catalytic activity.52 6243

DOI: 10.1021/acs.inorgchem.6b00875 Inorg. Chem. 2016, 55, 6240−6246

Article

Inorganic Chemistry

Scheme 1. Potential Catalycic Cycles for CO2 Reduction by Group 6 M(CO)6 Species in the Presence and Absence of Proton Donors

during catalysis and therefore is a homogeneous catalyst. In the absence of added proton sources, catalysis is more rapid, resulting in carbon monoxide and carbonate (Scheme 1). No oxalate was detected by 13C{1H} NMR spectroscopy during the 13 CO2 bulk electrolysis experiments, indicating that it was not formed via electrocatalysis by these species. The platinum electrode also primarily yielded CO formation, albeit at a rate slower than that of the carbon working electrode. Therefore, we conclude that the oxalate observed by Pickett and Pletcher was possibly due a misinterpretation of products obtained from carbonate, or that perhaps their experiment was conducted at a potential at which different reactions were operating. The potential used during controlled potential electrolysis was not specified in their report. In the presence of water as a proton donor and under reducing conditions, the group 6 hexacarbonyls could form M(CO)5H− or a similar hydride compound. In fact, there is literature precedent that proton sources such as methanol react rapidly with M(CO)52− to form M(CO)5H−.53,54 The hydride complexes can then insert CO2 to produce formate or react with another proton donor to form hydrogen. In our studies, we observe significant amounts of hydrogen under CO2 with water but do not observe a catalytic response in the presence of H2O and Ar. This is reconciled by the fact that CO2 and H2O produce an environment more acidic than that of H2O alone due to the formation of small amounts of carbonic acid, something that must be considered when dealing with CO2 reduction, even when in “aprotic” solvents.55,56 These two cycles (the proton-free cycle and the proton donor cycle, shown in Scheme 1) are likely competing, and further experimental, computational, and spectroscopic work will be needed to elucidate each of the steps in both cycles. It is also possible that transient monoreduced 19-electron species such as M(CO)5(L)− are involved, and this will need to be examined via further mechanistic studies. The fact that no catalysis was observed for the group 7 carbonyls on the cyclic voltammetry time scale shows that these manganese or rhenium species require a non-innocent ligand to help shuttle the reducing equivalents for CO2 reduction. However, they do react with CO2, which indicates that a

NMR spectroscopy. Tetrabutylammonium oxalate was independently synthesized, and the 13C{1H} NMR spectrum of the synthesized oxalate was used to confirm that oxalate was not present in any of the bulk electrolyses (Figures S20 and S21).35 When water (0.98 M) was included in the electrochemical cell, the bulk electrolysis current was decreased compared to that of the anhydrous reactions, consistent with the cyclic voltammetry experiments. When the headspace was analyzed by gas chromatography, carbon monoxide was observed at low faradaic efficiencies (35%), and hydrogen was also observed (47% faradaic efficiency). Formate was detected by 1H NMR in 7% faradaic yield. The average current density was also lower (3.5 mA/cm2) than the current density for our “dry” bulk electrolyses (average of 7.9 mA/cm2 from several runs). This is consistent with water suppressing the catalytic behavior in the cyclic voltammograms. We also sought to determine if the platinum electrode used by Pickett and Pletcher would give a very different product distribution. Therefore, we used a coiled platinum wire with a submerged surface area of ∼1.4 cm2. Bulk electrolysis under our standard conditions of Mo(CO)6 in acetonitrile for 30 min resulted in a 95% faradaic efficiency for CO and a 7% faradaic efficiency for H2. We were unable to detect any oxalate by IR. The current density of the Pt experiment (3.7 mA/cm2) was lower than that of our controlled potential electrolysis experiments using a carbon rod, consistent with the lower icat/ip values that were observed using a platinum electrode in cyclic voltammetry when compared to the same conditions using a glassy carbon working electrode. Preliminary Mechanistic Considerations and Relevance to Known Group 6 and 7 CO2 Reduction Systems. The reduction of the chromium, molybdenum, and tungsten hexacarbonyl species is irreversible, even at fast scan rates. Therefore, a rapid and irreversible chemical step is occurring after reduction, and literature data are consistent with generation of a M(CO)52− species. This dianionic species can then interact with CO2 and result in catalytic reduction of CO2, consistent with the known stoichiometric reactivity.21 Electrochemical studies with Mo(CO)6 show that it is freely diffusing, does not plate on the electrode, and does not decompose 6244

DOI: 10.1021/acs.inorgchem.6b00875 Inorg. Chem. 2016, 55, 6240−6246

Article

Inorganic Chemistry [Re(CO)5(CO2)]− or [Mn(CO)5(CO2)]− species or a similar compound is likely formed and could potentially be isolated and characterized. This also indicates that a key role of the noninnocent bpy ligand in catalysis by Re(bpy-R)(CO)3X and Mn(bpy-R)(CO)3X species is the facilitation of the steps required for C−O bond scission and catalyst turnover prior to reaction with CO2. What stands out is the fact that the group 6 hexacarbonyls are effective catalysts without the need for noninnocent ligands that help in the catalytic cycle. This difference between the group 6 and group 7 carbonyls could be justified on the basis of nuclear charge arguments. A group 6 M(CO)52− species will have a negative charge that experiences a nuclear charge less effective than that of the comparable group 7 species and therefore will be a more effective reducing agent. However, this is a very simple description, and a deeper understanding of the roles of the metals and ligands in the group 6 systems is needed to compare with the calculations that have been performed on the group 7 complexes.57−59 Further studies to elucidate the mechanism are ongoing in our laboratory.

Minority Participation (ILSAMP) program for support (National Science Foundation Grant 1411219).

■

(1) Savéant, J.-M. Chem. Rev. 2008, 108, 2348. (2) Benson, E. E.; Kubiak, C. P.; Sathrum, A. J.; Smieja, J. M. Chem. Soc. Rev. 2009, 38, 89. (3) Costentin, C.; Robert, M.; Saveant, J.-M. Chem. Soc. Rev. 2013, 42, 2423. (4) Reda, T.; Plugge, C. M.; Abram, N. J.; Hirst, J. Proc. Natl. Acad. Sci. U. S. A. 2008, 105, 10654. (5) Schuchmann, K.; Müller, V. Science 2013, 342, 1382. (6) Clark, M. L.; Grice, K. A.; Moore, C. E.; Rheingold, A. L.; Kubiak, C. P. Chemical Science 2014, 5, 1894. (7) Tory, J.; Setterfield-Price, B.; Dryfe, R. A. W.; Hartl, F. ChemElectroChem 2015, 2, 213. (8) Sieh, D.; Lacy, D. C.; Peters, J. C.; Kubiak, C. P. Chem. - Eur. J. 2015, 21, 8497. (9) Franco, F.; Cometto, C.; Sordello, F.; Minero, C.; Nencini, L.; Fiedler, J.; Gobetto, R.; Nervi, C. ChemElectroChem 2015, 2, 1372. (10) Grice, K. A.; Kubiak, C. P. In Advances in Inorganic Chemistry; Michele, A., Rudivan, E., Eds.; Academic Press: New York, 2014; Vol. 66, p 163. (11) Kaim, W. Inorg. Chem. 2011, 50, 9752. (12) Kaim, W. Eur. J. Inorg. Chem. 2012, 2012, 343. (13) Benson, E. E.; Sampson, M. D.; Grice, K. A.; Smieja, J. M.; Froehlich, J. D.; Friebel, D.; Keith, J. A.; Carter, E. A.; Nilsson, A.; Kubiak, C. P. Angew. Chem., Int. Ed. 2013, 52, 4841. (14) Thammavongsy, Z.; Seda, T.; Zakharov, L. N.; Kaminsky, W.; Gilbertson, J. D. Inorg. Chem. 2012, 51, 9168. (15) Lacy, D. C.; McCrory, C. C. L.; Peters, J. C. Inorg. Chem. 2014, 53, 4980. (16) Praneeth, V. K. K.; Ringenberg, M. R.; Ward, T. R. Angew. Chem., Int. Ed. 2012, 51, 10228. (17) Oh, Y.; Vrubel, H.; Guidoux, S.; Hu, X. Chem. Commun. 2014, 50, 3878. (18) Summers, D. P.; Leach, S.; Frese, K. W., Jr J. Electroanal. Chem. Interfacial Electrochem. 1986, 205, 219. (19) Oh, Y.; Hu, X. Chem. Commun. 2015, 51, 13698. (20) Ramos Sende, J. A.; Arana, C. R.; Hernandez, L.; Potts, K. T.; Keshevarz-K, M.; Abruna, H. D. Inorg. Chem. 1995, 34, 3339. (21) Lee, G. R.; Maher, J. M.; Cooper, N. J. J. Am. Chem. Soc. 1987, 109, 2956. (22) Darensbourg, D. J. Inorg. Chem. 2010, 49, 10765. (23) Silavwe, N. D.; Goldman, A. S.; Ritter, R.; Tyler, D. R. Inorg. Chem. 1989, 28, 1231. (24) Dessy, R. E.; Stary, F. E.; King, R. B.; Waldrop, M. J. Am. Chem. Soc. 1966, 88, 471. (25) Pickett, C. J.; Pletcher, D. J. Chem. Soc., Dalton Trans. 1975, 879. (26) Seurat, A.; Lemoine, P.; Gross, M. Electrochim. Acta 1978, 23, 1219. (27) Melo, A.; Gautier, J. L. Electrochim. Acta 1990, 35, 1879. (28) Amatore, C.; Krusic, P. J.; Pedersen, S. U.; Verpeaux, J.-N. Organometallics 1995, 14, 640. (29) Pickett, C. J.; Pletcher, D. J. Chem. Soc., Dalton Trans. 1976, 749. (30) Desilvestro, J.; Pons, S. J. Electroanal. Chem. Interfacial Electrochem. 1989, 267, 207. (31) Tomita, Y.; Teruya, S.; Koga, O.; Hori, Y. J. Electrochem. Soc. 2000, 147, 4164. (32) Christensen, P. A.; Hamnett, A.; Muir, A. V. G.; Freeman, N. A. J. Electroanal. Chem. Interfacial Electrochem. 1990, 288, 197. (33) Angamuthu, R.; Byers, P.; Lutz, M.; Spek, A. L.; Bouwman, E. Science 2010, 327, 313. (34) Cheng, S. C.; Blaine, C. A.; Hill, M. G.; Mann, K. R. Inorg. Chem. 1996, 35, 7704. (35) Beattie, J. W.; White, D. S.; Bheemaraju, A.; Martin, P. D.; Groysman, S. Dalton Trans. 2014, 43, 7979. (36) Pavlishchuk, V. V.; Addison, A. W. Inorg. Chim. Acta 2000, 298, 97.

■

CONCLUSIONS The commercially available and air-stable group 6 hexacarbonyl complexes are capable electrocatalysts for the reduction of CO2 to form CO, whereas the group 7 carbonyls do not catalyze CO2 reduction, even in the presence of added proton sources. Notably, the current response in the presence of added proton sources decreases for the group 6 carbonyl electrocatalysts. Different products were observed when water was added as compared to the reduction in the absence of added water. CO and carbonate were formed under anhydrous conditions, whereas hydrogen, CO, and formate were formed in the presence of added water. No oxalate was observed, in contrast to a previous report. Future work will focus on using experimental studies and computational approaches to understand the reasons behind the differences between the group 6 and 7 catalysts by elucidating the full mechanism of CO2 reduction by the group 6 M(CO)6 species. With that knowledge, we can turn to the discovery of ligands and conditions that promote the reduction of CO2 at less negative potentials on “early metal” (groups 3−6) complexes.

■

ASSOCIATED CONTENT

* Supporting Information S

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.6b00875. Additional electrochemical data, including various cyclic voltammetry and bulk electrolysis plots (PDF)

■

REFERENCES

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS K.A.G. acknowledges the support of DePaul University and the College of Science and Health (CSH) for startup funds as well as a Competitive Research Grant through DePaul University’s University Research Council (URC) for additional support. C.S. acknowledges the Illinois Louis Stokes Alliance for 6245

DOI: 10.1021/acs.inorgchem.6b00875 Inorg. Chem. 2016, 55, 6240−6246

Article

Inorganic Chemistry (37) Aranzaes, J. R.; Daniel, M.-C.; Astruc, D. Can. J. Chem. 2006, 84, 288. (38) Costentin, C.; Drouet, S.; Robert, M.; Savéant, J.-M. J. Am. Chem. Soc. 2012, 134, 11235. (39) Costentin, C.; Savéant, J.-M. ChemElectroChem 2014, 1, 1226. (40) Rountree, E. S.; McCarthy, B. D.; Eisenhart, T. T.; Dempsey, J. L. Inorg. Chem. 2014, 53, 9983. (41) Finkelstein, D. A.; Letcher, C. D.; Jones, D. J.; Sandberg, L. M.; Watts, D. J.; Abruña, H. D. J. Phys. Chem. C 2013, 117, 1571. (42) Wu, Q.-H.; Li, N.-H.; Sun, S.-G. J. Phys. Chem. B 2006, 110, 11383. (43) Meher, S. K.; Rao, G. R. ACS Catal. 2012, 2, 2795. (44) Lacombe, D.; Anderson, J.; Kadish, K. Inorg. Chem. 1986, 25, 2074. (45) Dessy, R. E.; Pohl, R. L.; King, R. B. J. Am. Chem. Soc. 1966, 88, 5121. (46) Diot, M.; Bousquet, J.; Lemoine, P.; Gross, M. J. Organomet. Chem. 1976, 112, 79. (47) Smieja, J. M.; Sampson, M. D.; Grice, K. A.; Benson, E. E.; Froehlich, J. D.; Kubiak, C. P. Inorg. Chem. 2013, 52, 2484. (48) Bourrez, M.; Molton, F.; Chardon-Noblat, S.; Deronzier, A. Angew. Chem., Int. Ed. 2011, 50, 9903. (49) Smieja, J. M.; Benson, E. E.; Kumar, B.; Grice, K. A.; Seu, C. S.; Miller, A. J. M.; Mayer, J. M.; Kubiak, C. P. Proc. Natl. Acad. Sci. U. S. A. 2012, 109, 15646. (50) McCarthy, B. D.; Martin, D. J.; Rountree, E. S.; Ullman, A. C.; Dempsey, J. L. Inorg. Chem. 2014, 53, 8350. (51) Hui, Y.; Webster, R. D. Anal. Chem. 2011, 83, 976. (52) Zeng, Q.; Tory, J.; Hartl, F. Organometallics 2014, 33, 5002. (53) Darensbourg, M. Y.; Deaton, J. C. Inorg. Chem. 1981, 20, 1644. (54) Darensbourg, M. Y.; Slater, S. J. Am. Chem. Soc. 1981, 103, 5914. (55) Matsubara, Y.; Grills, D. C.; Kuwahara, Y. ACS Catal. 2015, 5, 6440. (56) Pegis, M. L.; Roberts, J. A. S.; Wasylenko, D. J.; Mader, E. A.; Appel, A. M.; Mayer, J. M. Inorg. Chem. 2015, 54, 11883. (57) Keith, J. A.; Grice, K. A.; Kubiak, C. P.; Carter, E. A. J. Am. Chem. Soc. 2013, 135, 15823. (58) Riplinger, C.; Sampson, M. D.; Ritzmann, A. M.; Kubiak, C. P.; Carter, E. A. J. Am. Chem. Soc. 2014, 136, 16285. (59) Riplinger, C.; Carter, E. A. ACS Catal. 2015, 5, 900.

6246

DOI: 10.1021/acs.inorgchem.6b00875 Inorg. Chem. 2016, 55, 6240−6246