Unexpected Solvent Effect in Electrocatalytic CO2 to CO

Article pubs.acs.org/IC

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

Unexpected Solvent Effect in Electrocatalytic CO2 to CO Conversion Revealed Using Asymmetric Metalloporphyrins Soumalya Sinha and Jeffrey J. Warren* Department of Chemistry Simon Fraser University 8888 University Drive Burnaby, British Columbia V5A 1S6, Canada

Downloaded via KAOHSIUNG MEDICAL UNIV on September 14, 2018 at 14:35:12 (UTC). See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles.

S Supporting Information *

ABSTRACT: Rapid and efficient electrochemical CO2 reduction is an ongoing challenge for the production of sustainable fuels and chemicals. In this work, electrochemical CO2 reduction is investigated using metalloporphyrin catalysts (metal = Mn, Fe, Co, Ni, Cu) that feature one hydroxyphenyl group, and three other phenyl groups, in the porphyrin heterocycle (5-(2-hydroxyphenyl)-10,15,20-triphenylporphyrin, TPOH). These complexes, which are minimal versions of related complexes bearing up to eight proton relays, were designed to allow more straightforward determination of the role of the 2-hydroxylphenyl functional group. The iron-substituted version of TPOH supports robust reduction of CO2 in acetonitrile solvent, where carbon monoxide is the only detected product. Addition of weak Brønsted acids (1 M water or 8 mM phenol) gives rise to almost 100-fold enhancement in turnover frequency. Surprisingly, the iron analogue is a poor catalyst when the solvent is changed to dimethylformamide. These results lead to the proposal of a model where the hydroxyphenyl group behaves as a local proton source, a hydrogen bond donor to CO2-bound intermediates, and a hydrogen bonding partner to Brønsted acids. The observations from this model suggest improvements for existing electrocatalytic CO2 reduction systems.

■

INTRODUCTION

The development of technology that can capture and/or convert CO2 to other products is of great interest. Utilizing reduced products derived from CO2 in downstream processes is a broadly promising strategy for addressing concerns about how to manage greenhouse gas emissions and for developing alternative sources of chemical fuels and feedstocks (e.g., hydrocarbons).1,2 For example, the development of an artificial carbon cycle could provide a bridge between existing infrastructure and future global energy demands. Research into the conversion of CO2 to CO has reintensified in the past decade, and catalysts that can carry out the CO2 reduction reaction with good kinetics parameters and high turnover numbers are still being developed. The principal factors that still must be improved in CO2 reduction catalyst development are selectivity, stability, and energetics, especially for production of a desirable product (e.g., CO, HCO2H, or C2H4).1,3 The backdrop for these challenges is that almost all reduction reactions of CO2 can be described as proton-coupled electron transfer (PCET) reactions because they depend on addition of protons (H+) and electrons (e−). Specifically, delivery of H+ and e− at the right time and in the correct stoichiometry is a crucial consideration for CO2 reduction electrocatalysis. Herein, we describe an investigation of metal-substituted 5-(2-hydroxyphenyl)10,15,20-triphenylporphyrin (MTPOH) electrocatalysts (Figure 1) for electrocatalytic CO2 reduction that addresses these issues and presents a new perspective on catalyst design. Special emphasis is given to ClFeTPOH, which is a very active homogeneous catalyst in organic solvent. © XXXX American Chemical Society

Figure 1. Asymmetric metalloporphyrins (MTPOH).

To address the requirement of e− and H+ input for CO2 reduction, a great deal of work on electrocatalysts has involved derivatives of metalloporphyrins4 or metal-diimine5,6 coordination complexes. These molecular catalysts include groups capable of multielectron redox chemistry (often at the ligand) and intra- and/or intermolecular proton sources (Brønsted acids). These complementary design features account for the requisite PCET steps. For metal-diimine complexes, catalytic overpotentials can be changed by modifying the functional groups on the ligand backbone: for example, in complexes with ClRe(CO)3 cores.5,6 While it is possible to tune the reduction potentials in those complexes, those shifts in E°′ do not necessarily yield higher catalyst activity at lower overpotentials. Recent designs of metalloporphyrin catalysts, specifically those using iron, focus on the role of phenolic proton relays to enhance the kinetics of electrocatalytic CO2 reduction. Synthetic incorporation of hydroxyphenyl groups results in catalysts with more appealing benchmark parameters by increasing the efficiency of Received: June 29, 2018

A

DOI: 10.1021/acs.inorgchem.8b01814 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry proton delivery.7 For example, iron 5,10,15,20-tetraarylporphyrin derivatives that have hydroxyl functional groups at the 2- and 6-positions of the phenyl group dramatically accelerate electrocatalytic CO2 reduction in DMF solvent with high Faradaic yield for CO.8 Such functional groups can participate in proton transfer reactions and stabilize reduced CO2-bound intermediates via hydrogen bonding.9 The above examples support an important paradigm for electrocatalytic CO2 reduction: proton delivery is a key component of catalyst activity, but lower overpotentials often come at the cost of lower turnover frequency. Understanding and overcoming these challenges to the development of highly active catalysts are key goals. Herein, we report the homogeneous electrochemical CO2 reduction in acetonitrile (MeCN) solvent using a series of metal-substituted porphyrins that have a 2-hydroxyphenyl proton relay at one meso position and three phenyl groups at the remaining meso positions (Figure 1). The Fe-containing analogue is the subject of the bulk of this work. Incorporation of only one proton relay allows for a more straightforward exploration of the effect of pendant groups as they relate to the proton-coupled reduction of CO2. This work investigates the importance of solvent choice in CO2 reduction reactions and supports the concept that a minimal number of ancillary groups is sufficient to enhance CO2 reduction electrocatalysis.

■

where M = Mn, Fe, Co, Ni, Cu) was prepared and characterized. The electronic absorption spectra of the H2TPOH ligand16 and the metal complexes17−21 have extinction coefficients (ε) and λmax values identical with those for the corresponding 5,10,15,20-tetraphenylporphyrin (TPP) complexes (see the Supporting Information). On the basis of this observation, the addition of a single hydroxyl group to a phenyl group does not significantly affect the electronic properties of the aromatic system and metal center in these complexes. The electrochemical behavior of all of the above MTPOH complexes was first investigated in Ar-sparged MeCN solution and then in CO2-saturated solution. Cyclic voltammograms (CVs) for each complex under Ar and under CO2 are set out in the Supporting Information. In Ar-sparged MeCN, all complexes show mostly reversible electrochemical behavior (see the Supporting Information). However, most of these metalloporphyrins are modest to poor CO2 reduction catalysts in MeCN. ClFeTPOH, on the other hand, showed noteworthy activity and will be the subject of the remainder of this report. CVs for ClFeTPOH show two chemically reversible redox waves and a redox wave near the edge of the electrochemical window that is best described as quasi-reversible in MeCN or DMF solution (Table 1 and Figure 2). These features are Table 1. Potentials for ClFeTPOH in Deoxygenated 0.1 M n Bu4NPF6 DMF and MeCN Solutions

EXPERIMENTAL SECTION

All reagents were purchased from Sigma-Aldrich unless otherwise noted and used without further purification. Solvents were from J.T. Baker. Gases were obtained from Praxair Canada. Mass spectra were collected by using a Bruker microFlex MALDI-TOF or Agilent 6210 electrospray ionization mass spectrometer. UV−visible spectra were recorded in MeCN solvent using a Cary100Bio UV−visible spectrophotometer. GC experiments were carried out using an Agilent 6890 gas chromatograph equipped with a Restek ShinCarbon ST Micropacked Column and thermal conductivity detector. Additional details are provided in the Supporting Information. The ligand, 5-(2-hydroxyphenyl)-10,15,20-triphenylporphyrin (TPOH),10 was prepared using a modified literature procedure.11 Metalation with the appropriate metal chloride salts was done according to the literature in refluxing DMF.12 Metalloporphyrins were obtained as precipitates upon addition of water and a few drops of 1 M HCl solution. The solids were filtered, purified, and recrystallized from toluene. The complexes were characterized by UV−vis spectroscopy, MALDI-TOF, and ESI-MS mass spectrometry; these spectra are set out in the Supporting Information. Electrochemical measurements were performed on a Gamry Instruments Interface1000 potentiostat, using a conventional threeelectrode cell with a basal plane graphite working electrode13,14 (3 mm by 3 mm surface area), Pt-wire counter electrode, and nonaqueous Ag/0.01 M AgNO3 in MeCN reference electrode (CH Instruments). Working electrode surfaces were prepared by lightly abrading with 4000 SiC paper, washing thoroughly with deionized water followed by sonication (120 s) in isopropyl alcohol, and briefly drying with a heat gun. Complexes were dissolved to 1 mM concentrations with 0.1 M nBu4NPF6, and cyclic voltammograms (CVs) were collected at a 100 mV s−1 scan rate unless otherwise noted. Ferrocene was used as an external standard for all electrochemical experiments, and all potentials are reported with respect to the ferrocenium− ferrocene couple (Cp2Fe+/0).15 Conversion to the normal hydrogen electrode (NHE) in MeCN requires addition of 0.63 V to all potentials reported here.15

E1/2 (V vs Cp2Fe+/0) DMF MeCN

−0.75, −1.61, −2.23 −0.66, −1.46, −2.06

Figure 2. Cyclic voltammograms for ClFeTPOH in Ar-sparged DMF (left) and MeCN (right) solutions. Scans were recorded at a scan rate of 100 mV s−1, and both solutions contained 0.1 M nBu4NPF6.

consistent with the FeIII/II, FeII/I, and FeI/0 redox couples, respectively.2,6 The current at the potential for FeI/0 is more pronounced at scan rates greater than 100 mV s−1. The additional waves observed in the reverse sweep in MeCN were not present after replacing the chloride counterion with trifluoromethanesulfonate (OTf−; see the Supporting Information). Porphyrin-based CO2 reduction electrocatalysts have long been investigated in DMF solutions.22 In CO2-saturated DMF solution, CVs of ClFeTPOH showed a 2-fold current enhancement at the same potential as for the FeI/0 couple (Figure 3). In contrast, in CO2-saturated MeCN solution, CVs for ClFeTPOH showed a 20-fold increase in current density. Addition of weak Brønsted acids to electrochemical reaction solutions is known to enhance CO2 reduction.23 To test for such improvements in CO2 reduction by ClFeTPOH, CVs were collected in DMF and MeCN with water added to a final concentration of 1 M. The addition of H2O to the CO2-saturated DMF solution

■

RESULTS A series of new first-row transition-metal complexes of 5-(2-hydroxyphenyl)-10,15,20-triphenylporphyrin (MTPOH, B

DOI: 10.1021/acs.inorgchem.8b01814 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

CO2-saturated MeCN with 8 mM PhOH or 10 mM TFE. Addition of PhOH and TFE resulted in 40-fold and 35-fold catalytic current increases, respectively, in CO2-saturated MeCN solutions (see the Supporting Information). Scans in the absence of CO2 showed a ca. 2.5-fold current increase in the presence of 8 mM PhOH, consistent with some degree of proton reduction;29 CO2 is required to observe the larger catalytic waves. In contrast, addition of 10 mM TFE showed only negligible increases in current under Ar. Addition of PhOH or TFE beyond the above concentrations resulted in incremental shifts of the catalytic wave toward more negative potentials with a minor change in current. The stability of ClFeTPOH to prolonged CO2 reduction in MeCN was assessed using controlled-potential electrolysis (CPE) experiments. Gas chromatography (GC) and FT-IR spectroscopy were used to identify the gaseous products, and GC was used for quantification. CPE experiments were carried out with an applied potential of −2.3 V versus Cp2Fe+/0 using MeCN solutions with 1 M H2O and saturating CO2. The current density decreased by about 50% during the first 30 min of electrolysis and then remained stable for ∼2 h (see the Supporting Information). CO production was confirmed by GC against an authentic sample and by its characteristic peak at 2173 cm−1 in gas FT-IR spectra (see the Supporting Information). GC analyses carried out by transferring the reaction head space after 5 and 20 min showed only CO as the reduced product, yielding 96% Faradaic efficiency after 2 h of electrolysis. 1 H NMR analyses carried out during the course of and after CPE experiments showed no other soluble products (e.g., formic acid, methanol). On the basis of the total current passed, 0.5 mol of CO was produced during 2 h of electrolysis. Variable scan rate experiments were carried out to determine kinetics parameters for CO2 reduction by ClFeTPOH in MeCN. The turnover frequency (TOF) was calculated using “foot-ofthe-wave” analyses (FOWA) that allow straightforward comparisons to other CO2 reduction catalysts. Full details are given in the Supporting Information. The slope of the linear fit obtained from FOWA gives the pseudo-first-order rate constant for CO2 reduction by ClFeTPOH as 4.5 × 103 s−1 (logTOF = 3.7 s−1) at −2.3 V applied potential in CO2saturated MeCN + 1 M H2O solution. For comparison, the analogous experiments in CO2-saturated DMF + 1 M H2O give a value of 1.3 × 102 s−1 (log TOF = 2.11 s−1). Using an equation derived for the second-order turnover number (TON(2)),30,31 an intrinsic turnover number of about 4000 for ClFeTPOH is observed during 2 h of electrolysis in MeCN. Note that this calculation assumes zero consumption/ deactivation of catalyst.30,31 CVs recorded in CO2-saturated MeCN + 1 M H2O with scan rates greater than 1 V s−1 show a precatalytic wave at −1.7 V, near the onset of the catalytic wave, but at potentials more negative than that of the FeII/I couple (Figure 5). This precatalytic feature is similar to what is found for other porphyrin CO2 reduction catalysts.9 The current density at the potential of the precatalytic wave increases linearly with the square root of scan rate (see the Supporting Information). The addition of water at a concentration between 0 and 1 M did not dramatically affect the precatalytic wave but did result in higher current density for the catalytic wave. In contrast, when the solvent is DMF, the current from the precatalytic wave was very weak. As for MeCN, the addition of water to DMF solutions of ClFeTPOH did not dramatically affect the observed precatalytic current (Figure 5).

Figure 3. Comparison of CVs for ClFeTPOH in Ar- or CO2-saturated 0.1 M nBu4NPF6 DMF (left) and MeCN (right) solutions. CVs were recorded at a scan rate of 100 mV s−1.

Figure 4. Comparison of current enhancement for ClFeTPOH between CO2-saturated 0.1 M nBu4NPF6 DMF (left) and MeCN (right) solutions + 1 M H2O. Scans were recorded at a scan rate of 100 mV s−1. The maximum current density in CO2-saturated 0.1 M n Bu4NPF6 DMF + 1 M H2O solution is shown in the inset.

only marginally improves this activity without changing the overpotential (Figure 4, left). Control experiments without CO2 show only minor current increases (see the Supporting Information). In contrast to the experiments in DMF, addition of 1 M H2O to ClFeTPOH in CO2-saturated MeCN solution gave rise to currents that are ca. 60-fold larger than those with respect to ClFeTPOH alone (Figure 3, right). In these solutions, the addition of H2O in excess of 1 M caused systematic decreases in current density without any apparent precipitation of catalysts. Control experiments using ClFeTPOH in Ar-sparged MeCN show a ≥5-fold current increase with the addition of H2O at 1 M concentration (see the Supporting Information). This unexpectedly different CO2 reduction activity of ClFeTPOH in DMF and MeCN prompted a more careful consideration of the underlying factors that could contribute to such different responses. As a control case, analogous experiments using [FeTPP]+ also were conducted in MeCN. Note that ClFeTPP is insoluble in MeCN; therefore, the corresponding OTf− salt was prepared by addition of 1 equiv of AgOTf to ClFeTPP. The presence of the single 2-hydroxyphenyl group shifts the current onset potential by about +0.25 V and gives larger current densities (see the Supporting Information). The addition of stronger Brønsted acids to ClFeTPOH in MeCN also was explored. Water is a poor acid (pKa,MeCN ≥ 3124); therefore, phenol (PhOH, pKa,MeCN = 29.425) and trifluoroethanol (TFE26,27) also were tested. PhOH is a commonly added Brønsted acid in CO2 reduction electrocatalyst testing,7,28 though the speciation of dissolved CO2 is not as well defined under these conditions. These acids display background contributions at potentials more negative than that for ClFeTPOH (i.e., a bare electrode) (see the Supporting Information). Such behavior is well established for weak acids in MeCN.29 Electrocatalytic activity was monitored in C

DOI: 10.1021/acs.inorgchem.8b01814 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

This value is in good agreement with the value calculated (−1.50 V) using eq 1, where the pKa value is 23.4 and the standard potential (E°CO2/CO,MeCN) is −0.12. E = E° − 0.059pK a

(1)

The value of ECO2/CO,DMF is more challenging to address. Using a value for E° derived in the same way as the value for MeCN, we take E°CO2/CO,DMF = −0.73 V35 vs Cp2Fe+/0. The reported pKa for H2CO3 in DMF is 7.37, which was obtained using quantum chemical calculations.8,36 Those same calculations lead to a pKa value of 17.03 in MeCN, lower than the recently measured value of 23.4.37 Taking pKa(H2CO3) = 23.4 in MeCN and converting it to a pKa value in DMSO gives pKa(H2CO3) = 12.2 ± 1.0. This extrapolated pKa in DMSO is similar to that for acetic acid in DMSO (pKa = 12.6)38 and in DMF (pKa = 13.5),39 just as the pKa in MeCN is near that of CH3COOH.33 Using pKa = 12.2 in eq 1 gives ECO2/CO,DMF = −1.4 ± 0.1 V vs Cp2Fe+/0. This value is still an estimate and should be experimentally refined, but we think that it is more physically reasonable because it is anchored to the experimental pKa value in MeCN. The potentials for CO2/CO conversion described above in DMF and MeCN, and comparative overpotentials (η) are given in Table 2.

Figure 5. CVs for ClFeTPOH in CO2-saturated 0.1 M nBu4NPF6 MeCN solution in the presence of 0.2 M H2O, 0.5 M H2O, and 1 M H2O (left). The CV for the same catalyst in CO2-saturated 0.1 M n Bu4NPF6 DMF + 1 M H2O solution has been compared with that of in MeCN solution under identical conditions (right). All CVs were recorded at a scan rate of 1 V s−1. Only reductive waves in the forward directions are shown for clarity.

■

DISCUSSION A series of first-row transition-metal complexes of 5-(2-hydroxyphenyl)-10,15,20-triphenylporphyrin were prepared. Of those complexes, the iron analogue (ClFeTPOH) was the most active for CO2 reduction. In contrast to literature reports for related porphyrins, ClFeTPOH has much less activity in DMF than in MeCN solvent. Overall, the activity of this catalyst in MeCN approached that of related molecules with a greater number of hydroxyphenyl groups on the porphyrin aromatic groups. The value of log TOF calculated from the FOWA for ClFeTPOH in MeCN + 1 M H2O solution (log TOFmax = 3.7) is similar to that for chloroiron(III) 5,10,15,20-tetrakis(2′,6′-dihydroxyphenyl)porphyrin (log TOFmax ≤ 4.28) in DMF + 2 M H2O or chloroiron(III) 5,15-bis(2′,6′-dihydroxyphenyl)-10,20bis(pentafluorophenyl)porphyrin (log TOFmax = 4.0) in DMF + 3 M PhOH.4 For comparison, our log TOFmax value is smaller than that of an iron porphyrin with one strongly H-bond donating o-amide group (log TOFmax = 6.7 in DMF with 100 mM PhOH)28 or of an iron porphyrin with electrostatically stabilizing trimethylamine groups (log TOFmax = 6 in DMF with 0.1 M water and 3 M PhOH).32 Catalytic Tafel plots29 (Figure 6) were constructed to compare ClFeTPOH to other porphyrin catalysts. Full details

Table 2. Reduction Potentials for CO2 and the Overpotential for ClFeTPOH in MeCN and DMFa solvent

E°CO2/CO (V)

ECO2/CO,1MH2O (V)

η for ClFeTPOH (V) (ECO2/CO,1MH2O − Eelectrolysis)

MeCN DMF

−0.1235 −0.7335

−1.5434 −1.4

0.2,b 0.36c 0.6,b 0.28c

a

Reduction potentials are as described in the text. Additional details are given in the Supporting Information. bTaken at the current onset in MeCN (−1.73 V) and DMF (−1.98 V) vs Cp2Fe+/0. cTaken at the potential of Ecat/2 vs Cp2Fe+/0.40 For both cases, the solutions were saturated with CO2 in the presence of 1 M H2O.

The correlation between log TOF vs η shown in the catalytic Tafel plots (Figure 6) indicates that the single 2-hydroxyphenyl group in TPOH is sufficient to induce significant catalytic activity in wet MeCN solution. All else being equal, ClFeTPOH reaches maximum turnover frequencies at lower potentials in MeCN than in DMF. The addition of a single 2-hydroxyphenyl group in ClFeTPOH does not substantially affect reduction potentials or optical spectra of ClFeTPOH in DMF with respect to those for the parent ClFeTPP; therefore, it is not likely that the electronic features of the iron center are altered. ClFeTPOH is more active with the addition of other Brønsted acids (PhOH or TFE) without any significant change in overpotential. The calculated log TOFmax values for ClFeTPOH are very similar for added PhOH, TFE, and H2O, despite the slightly different pKa values for each compound. We were surprised that the addition of only 8 mM PhOH induces such active catalysis in comparison to catalysts in DMF, where greater PhOH concentrations are needed to reach the same current densities. The reactivity of ClFeTPOH for CO2 reduction in DMF is distinct from the reactivity in MeCN under the same conditions. In fact, the behavior of ClFeTPOH in CO2-saturated DMF + 1 M H2O resembles that reported for ClFeTPP.41 In DMF, larger concentrations of Brønsted acids (PhOH36 or TFE22) are needed to reach the currents that we observe with ClFeTPOH in wet MeCN. The dramatic change in log

Figure 6. Comparative catalytic Tafel plots for ClFeTPOH with (red) and without (black) 1 M H2O in CO2-saturated 0.1 M nBu4NPF6 (left) MeCN solution (using 1.54 V vs Cp2Fe+/0) and (right) DMF solution (1.4 V vs Cp2Fe+/0).

of the construction of these plots are given in the Supporting Information. One important factor in the construction of these plots is the reduction potential for the transformation being catalyzed. The potential (ECO2/CO,MeCN) in CO2-saturated MeCN + 1 M H2O solution is −1.54 V vs Cp2Fe+/0, where the pKa value for H2CO3 in MeCN + 1 M H2O is 23.4.33,34 D

DOI: 10.1021/acs.inorgchem.8b01814 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Table 3. Selected Hydrogen Bond Acidity (α2H) and Basicity (β2H) Parameters43,44

TOFmax that we observe by changing the solvent from DMF to MeCN suggests that known catalysts could be much more active than originally reported. This idea is explored in more detail in the following paragraphs. The observation of a precatalytic wave in CVs of ClFeTPOH in CO2-saturated MeCN (Figure 5) allows additional characterization of CO2 reduction in MeCN. This wave is attributed to a single-electron-transfer process, coupled with a single H+ transfer into a [FeI−CO2] intermediate at the rate-determining step.9 Figure 7 shows the catalytic cycle that

H2O PhOH MeCN DMSO TFE DMF

α2H

β2H

0.35 0.60 0.09 0 0.57 0

0.38 0.22 0.44 0.78 0.18 0.66

solvent could sequester Brønsted acid additives or H2CO3. Figure 7 also omits any homo- or heteroconjugation effects, which are known to be important in electrocatalysis in organic solvent.45 H bonding of ionic species also is a consideration, since they can have very strong H bonding properties.46 As evidence of the importance of H bonding to the porphyrin 2-hydroxyphenyl group, we carried out 1H NMR experiments in DMF-d7 and MeCN-d3 using ZnTPOH (see the Supporting Information). The 1H resonance for the hydroxyphenyl −OH group is at 9.64 ppm in DMF-d7, a characteristic downfield shift for protons in a strong H bond.47 In MeCN-d3, the hydroxyphenyl −OH resonance appears at 6.65 ppm. The identity of the OH protons was confirmed by the addition of a small amount of D2O to each respective tube. Finally, as an independent confirmation of this hypothesis, we collected analogous CVs and 1H data in DMSO solvent. ClFeTPOH is a poor catalyst in DMSO and shows a downfield chemical shift for the 2-hydroxyphenyl −OH group (see the Supporting Information). The H-bonding properties also will affect the ability of 2-hydroxylphenyl groups to H-bond with Brønsted acid additives. For example, in cases where phenol is added, form 2B can be present at higher concentrations in MeCN than in DMF. A strongly H bond accepting solvent will sequester a greater population of a donating Brønsted acid (in analogy to form 2A). The simple model in Figure 7 should be taken as limiting cases that are consistent with our experiments in MeCN and in DMF. Also note that H bonding between S: and the Brønsted acid is not shown in Figure 7, which is likely an important factor. The actual equilibria are likely to be more complex. Importantly, considering this model, or a related model, of H bonding may be useful as a guiding principle in designing conditions for CO2-reduction catalysts.

Figure 7. Catalytic cycle for CO2 reduction by porphyrins with pendant H-bonding groups.23 The rightmost scheme shows limiting cases of H bonding to ClFeTPOH. The porphyrin ring is abbreviated by a thick black line. S: and H-A denote solvent and added Brønsted acid, respectively.

has been extensively outlined for porphyrin-catalyzed CO2 reduction.9,23,41 The process involved with the precatalytic wave is the conversion of the Fe(I) porphyrin (compound 5) to compound 3. While the individual forward rate constants cannot be determined, the equilibrium binding constants (KCO2) for CO2 can be estimated from CV data.28,42 For ClFeTPOH, KCO2 (conversion of 1 to 2, Figure 2) is about 5 M−1 in both DMF and in MeCN (see Supporting Information). The similarity of the binding constants suggests that solvent does not strongly affect the initial CO2 binding step to the reduced porphyrin. To rationalize the difference in behavior of ClFeTPOH in MeCN and in DMF, we considered the H-bonding properties of those solvents. One approach is to compare the ability of each solvent to donate and to accept H bonds. Likewise, we can compare the ability of phenol (as a model for 2-hydroxyphenyl substituents) to patriciate in H bonding. Here, we use Abraham’s hydrogen bond acidity (α2H)43 and basicity (β2H)44 as metrics (Table 3). DMF is a strong H-bond acceptor but does not donate H bonds. In contrast, MeCN is a moderate H-bond acceptor and a weak H-bond donor. The Brønsted acid additives (H2O, PhOH, TFE) are all H-bond donors and acceptors. In particular, phenol is a very good H-bond donor. The partial model proposed above for H bonding can be added in the CO2 reduction cycle in Figure 7. It is important to emphasize that H-bonding equilibria could be involved in all steps. For simplicity, we focus on H bonding of compound 2 on the basis of our data. The model in Figure 7 suggests that, in DMF solvent, a higher proportion of the total catalyst concentration is in form 2A. In MeCN, this population can be lower since MeCN is a weaker H-bond acceptor. Likewise, the

■

CONCLUSIONS Metalloporphyrins with a single 2-hydroxyphenyl proton relay group were synthesized, characterized, and investigated for their electrochemical CO2 reduction properties. The iron analogue, ClFeTPOH, is a poor CO2 reduction catalyst in DMF, which is a commonly used solvent to explore CO2 reduction by porphyrins. In contrast, ClFeTPOH is a good catalyst in CO2saturated MeCN + 1 M H2O (or +8 mM PhOH). The maximum turnover frequency was 4.5 × 103 s−1 with relatively stable current levels observed during 2 h of electrolysis. The Faradaic efficiency was 96%, and CO is the only reduced product detected. The activity in MeCN, with only one hydroxyphenyl group, is noteworthy and allowed us to uncover an unappreciated solvent effect for CO2 reduction chemistry. We suspect that known iron porphyrin catalysts could be more active than has been reported or require lower concentrations of Brønsted acids if solvent and/or hydrogen-bonding effects are optimized. This also may hold true for other catalysts48,49 that have been investigated in DMF. E

DOI: 10.1021/acs.inorgchem.8b01814 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

(6) Sinha, S.; Berdichevsky, E. K.; Warren, J. J. Electrocatalytic CO2 Reduction Using Rhenium(I) Complexes with Modified 2-(2′Pyridyl)Imidazole Ligands. Inorg. Chim. Acta 2017, 460, 63−68. (7) Bhugun, I.; Lexa, D.; Savéant, J.-M. Catalysis of the Electrochemical Reduction of Carbon Dioxide by Iron(0) Porphyrins: Synergystic Effect of Weak Brönsted Acids. J. Am. Chem. Soc. 1996, 118 (7), 1769−1776. (8) 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 (6103), 90−94. (9) Costentin, C.; Passard, G.; Robert, M.; Savéant, 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. (10) D’Souza, F.; Chitta, R.; Gadde, S.; Zandler, M. E.; McCarty, A. L.; Sandanayaka, A. S. D.; Araki, Y.; Ito, O. Effect of Axial Ligation or π-π-Type Interactions on Photochemical Charge Stabilization in “Two-Point” Bound Supramolecular Porphyrin-Fullerene Conjugates. Chem. - Eur. J. 2005, 11 (15), 4416−4428. (11) Forneli, A.; Planells, M.; Sarmentero, M. A.; Martinez-Ferrero, E.; O’Regan, B. C.; Ballester, P.; Palomares, E. The Role of Para-Alkyl Substituents on Meso-Phenyl Porphyrin Sensitised TiO2 Solar Cells: Control of the eTiO2/Electrolyte+ Recombination Reaction. J. Mater. Chem. 2008, 18 (14), 1652. (12) Fleischer, E. B.; Palmer, J. M.; Srivastava, T. S.; Chatterjee, A. Thermodyamic and Kinetic Properties of an Iron-Porphyrin System. J. Am. Chem. Soc. 1971, 93 (13), 3162−3167. (13) Sinha, S.; Aaron, M. S.; Blagojevic, J.; Warren, J. J. Electrocatalytic Dioxygen Reduction by Carbon Electrodes Noncovalently Modified with Iron Porphyrin Complexes: Enhancements from a Single Proton Relay. Chem. - Eur. J. 2015, 21 (50), 18072− 18075. (14) Hanson, S. S.; Warren, J. J. Syntheses, Characterization, and Electrochemical Behavior of Alkylated 2-(2′-Quinolylbenzimidazole) Complexes of Rhenium (I). Can. J. Chem. 2018, 96 (2), 119−123. (15) Pavlishchuk, V. V.; Addison, A. W. Conversion Constants for Redox Potentials Measured versus Different Reference Electrodes in Acetonitrile Solutions at 25°C. Inorg. Chim. Acta 2000, 298 (1), 97− 102. (16) Mojarrad, A. G.; Zakavi, S. Photocatalytic Activity of the Molecular Complexes of Meso -Tetraarylporphyrins with Lewis Acids for the Oxidation of Olefins: Significant Effects of Lewis Acids and Meso Substituents: Photocatalytic Activity of the Molecular Complexes of Meso-Tetraarylporphyrins with Lewis Acids for the Oxidation of Olefins: Significant Effects of L. Eur. J. Inorg. Chem. 2017, 2017 (21), 2854−2862. (17) Chizhova, N. V.; Zvezdina, S. V.; Kataleva, Y. S.; Mamardashvili, N. Z. Preparation and Spectral Properties of βBromo-Substituted Mn(III) Tetraphenylporphyrinates. Russ. J. Gen. Chem. 2015, 85 (5), 1132−1135. (18) Chizhova, N. V.; Kumeev, R. S.; Mamardashvili, N. Z. Synthesis and Spectral Properties of Cobalt(II) and Cobalt(III) Tetraarylporphyrinates. Russ. J. Inorg. Chem. 2013, 58 (6), 740−743. (19) Dou, Z.; Xu, L.; Zhi, Y.; Zhang, Y.; Xia, H.; Mu, Y.; Liu, X. Metalloporphyrin-Based Hypercrosslinked Polymers Catalyze HeteroDiels-Alder Reactions of Unactivated Aldehydes with Simple Dienes: A Fascinating Strategy for the Construction of Heterogeneous Catalysts. Chem. - Eur. J. 2016, 22 (29), 9919−9922. (20) Tatar, A.; Dolenský, B.; Dvořaḱ ová, H.; Král, V. Selective Formation of Either Tröger’s Base or Spiro Tröger’s Base Derivatives from [2-Aminoporphyrinato(2-)]Nickel by Choice of Reaction Conditions. Tetrahedron Lett. 2012, 53 (45), 6015−6017. (21) Ye, L.; Ou, Z.; Fang, Y.; Xue, S.; Song, Y.; Wang, L.; Wang, M.; Kadish, K. M. Electrochemistry of Nonplanar Copper(II) Tetrabutano- and Tetrabenzotetraarylporphyrins in Nonaqueous Media. RSC Adv. 2015, 5 (94), 77088−77096. (22) Bhugun, I.; Lexa, D.; Saveant, J.-M. Ultraefficient Selective Homogeneous Catalysis of the Electrochemical Reduction of Carbon

The choice of an appropriate solution system is an important experimental factor as work continues to improve electrocatalytic CO2 reduction. The concepts demonstrated here have consequences for the development of new catalysts and efforts toward their incorporation into devices.50 Solvent effects will be critical considerations, especially if catalysts are to be investigated for application in prototype flow devices where reactions of CO2 occur at gas−catalyst interfaces (in the absence of solvent).51 In the context of this work, our asymmetric porphyrins maintain proton delivery and H-bonding properties while opening synthetic flexibility at other porphyrin positions. This presents new opportunities for porphyrin catalyst design. Proton delivery and appropriate stabilization of reduced intermediates along the catalytic cycle remain important for the control for proton-coupled redox catalysis, and the H-bonding model presented here should, in principle, be applicable to any electrocatalytic proton-coupled transformation.

■

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.inorgchem.8b01814. Additional experimental details, additional electrochemical data and calculations, optical spectra, GC traces, and NMR data (PDF)

■

AUTHOR INFORMATION

Corresponding Author

*E-mail for J.J.W.: [email protected]. ORCID

Jeffrey J. Warren: 0000-0002-1747-3029 Funding

Simon Fraser University, the National Sciences and Engineering Research Council (RGPIN05559 to J.J.W.), and the Canadian Institute for Advanced Research Bio-Inspired Solar Energy and Azrieli Global Scholars Programs supported this work. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We acknowledge S. Grunert for helping us with gas chromatography and E. K. Berdichevsky for assisting with MALDI data acquisition.

■

REFERENCES

(1) Bushuyev, O. S.; De Luna, P.; Dinh, C. T.; Tao, L.; Saur, G.; van de Lagemaat, J.; Kelley, S. O.; Sargent, E. H. What Should We Make with CO 2 and How Can We Make It? Joule 2018, 2 (5), 825−832. (2) Keith, D. W.; Holmes, G.; St. Angelo, D.; Heidel, K. A Process for Capturing CO2 from the Atmosphere. Joule 2018, 2, 1573. (3) Jouny, M.; Luc, W.; Jiao, F. General Techno-Economic Analysis of CO 2 Electrolysis Systems. Ind. Eng. Chem. Res. 2018, 57 (6), 2165−2177. (4) Costentin, C.; Passard, G.; Robert, M.; Savéant, J.-M. Ultraefficient Homogeneous Catalyst for the CO2-to-CO Electrochemical Conversion. Proc. Natl. Acad. Sci. U. S. A. 2014, 111 (42), 14990−14994. (5) Smieja, J. M.; Kubiak, C. P. Re(Bipy-TBu)(CO) 3 Cl−improved Catalytic Activity for Reduction of Carbon Dioxide: IR-Spectroelectrochemical and Mechanistic Studies. Inorg. Chem. 2010, 49 (20), 9283−9289. F

DOI: 10.1021/acs.inorgchem.8b01814 Inorg. Chem. XXXX, XXX, XXX−XXX

Article

Inorganic Chemistry

Generalized Method for Proton-Coupled Electron-Transfer Reactions. ACS Energy Lett. 2017, 2 (8), 1886−1891. (38) Kolthoff, I. M.; Chantooni, M. K.; Bhowmik, S. Dissociation Constants of Uncharged and Monovalent Cation Acids in Dimethyl Sulfoxide. J. Am. Chem. Soc. 1968, 90 (1), 23−28. (39) Kolthoff, I. M.; Chantooni, M. K.; Smagowski, H. Acid-Base Strength in N,N-Dimethylformamide. Anal. Chem. 1970, 42 (13), 1622−1628. (40) Appel, A. M.; Helm, M. L. Determining the Overpotential for a Molecular Electrocatalyst. ACS Catal. 2014, 4 (2), 630−633. (41) Costentin, C.; Drouet, S.; Passard, G.; Robert, M.; Savéant, 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. (42) Gagne, R. R.; Allison, J. L.; Ingle, D. M. Unusual Structural and Reactivity Types for Copper(I). Equilibrium Constants for the Binding of Monodentate Ligands to Several Four-Coordinate Copper(I) Complexes. Inorg. Chem. 1979, 18 (10), 2767−2774. (43) Abraham, M. H.; Grellier, P. L.; Prior, D. V.; Morris, J. J.; Taylor, P. J. Hydrogen Bonding. Part 10. A Scale of Solute HydrogenBond Basicity Using Log K Values for Complexation in Tetrachloromethane. J. Chem. Soc., Perkin Trans. 2 1990, 521. (44) Abraham, M. H.; Grellier, P. L.; Prior, D. V.; Duce, P. P.; Morris, J. J.; Taylor, P. J. Hydrogen Bonding. Part 7. A Scale of Solute Hydrogen-Bond Acidity Based on Log K Values for Complexation in Tetrachloromethane. J. Chem. Soc., Perkin Trans. 2 1989, 699. (45) Rountree, E. S.; Dempsey, J. L. Reactivity of Proton Sources with a Nickel Hydride Complex in Acetonitrile: Implications for the Study of Fuel-Forming Catalysts. Inorg. Chem. 2016, 55 (10), 5079− 5087. (46) Abraham, M. H.; Zhao, Y. H. Determination of Solvation Descriptors for Ionic Species: Hydrogen Bond Acidity and Basicity. J. Org. Chem. 2004, 69, 4677−4685. (47) Bernet, B.; Vasella, A. 1H-NMR Analysis of Intra- and Intermolecular H-Bonds of Alcohols in DMSO: Chemical Shift of Hydroxy Groups and Aspects of Conformational Analysis of Selected Monosaccharides, Inositols, and Ginkgolides. Helv. Chim. Acta 2000, 83 (5), 995−1021. (48) 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. (49) Chapovetsky, A.; Welborn, M.; Luna, J. M.; Haiges, R.; Miller, T. F.; Marinescu, S. C. Pendant Hydrogen-Bond Donors in Cobalt Catalysts Independently Enhance CO2 Reduction. ACS Cent. Sci. 2018, 4 (3), 397−404. (50) Tatin, A.; Comminges, C.; Kokoh, B.; Costentin, C.; Robert, M.; Savéant, J.-M. Efficient Electrolyzer for CO2 Splitting in Neutral Water Using Earth-Abundant Materials. Proc. Natl. Acad. Sci. U. S. A. 2016, 113 (20), 5526−5529. (51) Weekes, D. M.; Salvatore, D. A.; Reyes, A.; Huang, A.; Berlinguette, C. P. Electrolytic CO2 Reduction in a Flow Cell. Acc. Chem. Res. 2018, 51 (4), 910−918.

Dioxide by an Iron(0) Porphyrin Associated with a Weak Broensted Acid Cocatalyst. J. Am. Chem. Soc. 1994, 116 (11), 5015−5016. (23) Costentin, C.; Robert, M.; Savéant, J.-M. Current Issues in Molecular Catalysis Illustrated by Iron Porphyrins as Catalysts of the CO2-to-CO Electrochemical Conversion. Acc. Chem. Res. 2015, 48 (12), 2996−3006. (24) Felton, G. A. N.; Vannucci, A. K.; Okumura, N.; Lockett, L. T.; Evans, D. H.; Glass, R. S.; Lichtenberger, D. L. Hydrogen Generation from Weak Acids: Electrochemical and Computational Studies in the [(η5-C5H5)Fe(CO)2]2 System. Organometallics 2008, 27 (18), 4671− 4679. (25) Kütt, A.; Movchun, V.; Rodima, T.; Dansauer, T.; Rusanov, E. B.; Leito, I.; Kaljurand, I.; Koppel, J.; Pihl, V.; Koppel, I.; et al. Pentakis(Trifluoromethyl)Phenyl, a Sterically Crowded and ElectronWithdrawing Group: Synthesis and Acidity of Pentakis(Trifluoromethyl)Benzene, -Toluene, -Phenol, and -Aniline. J. Org. Chem. 2008, 73 (7), 2607−2620. (26) Huffman, L. M.; Casitas, A.; Font, M.; Canta, M.; Costas, M.; Ribas, X.; Stahl, S. S. Observation and Mechanistic Study of Facile CO Bond Formation between a Well-Defined Aryl-Copper(III) Complex and Oxygen Nucleophiles. Chem. - Eur. J. 2011, 17 (38), 10643−10650. (27) Kütt, A.; Leito, I.; Kaljurand, I.; Sooväli, L.; Vlasov, V. M.; Yagupolskii, L. M.; Koppel, I. A. A Comprehensive Self-Consistent Spectrophotometric Acidity Scale of Neutral Brønsted Acids in Acetonitrile. J. Org. Chem. 2006, 71 (7), 2829−2838. (28) Nichols, E. M.; Derrick, J. S.; Nistanaki, S. K.; Smith, P. T.; Chang, C. J. Positional Effects of Second-Sphere Amide Pendants on Electrochemical CO2 Reduction Catalyzed by Iron Porphyrins. Chem. Sci. 2018, 9 (11), 2952−2960. (29) McCarthy, B. D.; Martin, D. J.; Rountree, E. S.; Ullman, A. C.; Dempsey, J. L. Electrochemical Reduction of Brønsted Acids by Glassy Carbon in AcetonitrileImplications for Electrocatalytic Hydrogen Evolution. Inorg. Chem. 2014, 53 (16), 8350−8361. (30) Costentin, C.; Drouet, S.; Robert, M.; Savéant, J.-M. Turnover Numbers, Turnover Frequencies, and Overpotential in Molecular Catalysis of Electrochemical Reactions. Cyclic Voltammetry and Preparative-Scale Electrolysis. J. Am. Chem. Soc. 2012, 134, 11235− 11242. (31) Costentin, C.; Drouet, S.; Robert, M.; Savéant, J.-M. Correction to Turnover Numbers, Turnover Frequencies, and Overpotential in Molecular Catalysis of Electrochemical Reactions. Cyclic Voltammetry and Preparative-Scale Electrolysis. J. Am. Chem. Soc. 2012, 134 (48), 19949−19950. (32) Azcarate, I.; Costentin, C.; Robert, M.; Savéant, J.-M. ThroughSpace Charge Interaction Substituent Effects in Molecular Catalysis Leading to the Design of the Most Efficient Catalyst of CO2-to-CO Electrochemical Conversion. J. Am. Chem. Soc. 2016, 138 (51), 16639−16644. (33) Matsubara, Y.; Grills, D. C.; Kuwahara, Y. Thermodynamic Aspects of Electrocatalytic CO2 Reduction in Acetonitrile and with an Ionic Liquid as Solvent or Electrolyte. ACS Catal. 2015, 5 (11), 6440−6452. (34) Matsubara, Y. Standard Electrode Potentials for the Reduction of CO2 to CO in Acetonitrile−Water Mixtures Determined Using a Generalized Method for Proton-Coupled Electron-Transfer Reactions. ACS Energy Lett. 2017, 2 (8), 1886−1891. (35) Pegis, M. L.; Roberts, J. A. S.; Wasylenko, D. J.; Mader, E. A.; Appel, A. M.; Mayer, J. M. Standard Reduction Potentials for Oxygen and Carbon Dioxide Couples in Acetonitrile and N, N -Dimethylformamide. Inorg. Chem. 2015, 54 (24), 11883−11888. (36) Azcarate, I.; Costentin, C.; Robert, M.; Saveant, J.-M. ̀ Dissection of Electronic Substituent Effects in Multielectron,Ä iMultistep Molecular Catalysis. Electrochemical CO2-to-CO Conversion Catalyzed by Iron Porphyrins. J. Phys. Chem. C 2016, 120, 28951− 28960. (37) Matsubara, Y. Standard Electrode Potentials for the Reduction of CO2 to CO in Acetonitrile−Water Mixtures Determined Using a G

DOI: 10.1021/acs.inorgchem.8b01814 Inorg. Chem. XXXX, XXX, XXX−XXX