Research Article

Cite This: ACS Sustainable Chem. Eng. 2019, 7, 19631−19639 pubs.acs.org/journal/ascecg

α‑ Electrochemical CO2 Fixation to Methylbenzyl Bromide in Divided Cells with Nonsacrificial Anodes and Aqueous Anolytes Jury J. Medvedev, Xenia V. Medvedeva, Feng Li, Thomas A. Zienchuk, and Anna Klinkova* Department of , , 200 University Avenue West, Waterloo, Ontario N2L 3G1, Canada

*S Supporting Information

ABSTRACT: Electrocarboxylation of organic halides repre- sents a CO2 utilization strategy and a green alternative for the synthesis of many industrially relevant carboxylic acids. However, current electrocarboxylation methods rely on the utilization of sacrificial metal anodes, which are not sustainable, require high voltages, and complicate the understanding of the reaction mechanism. Here, we demonstrate the feasibility of performing electrocarboxylation reactions in divided cells with aqueous anolytes and nonsacrificial anodes, thereby eliminat- ing the reliance on sacrificial anodes and opening the door for coupling of this important reduction process with various electrooxidation reactions requiring aqueous . Specifically, we report a detailed study of electrocarboxylation of (1-bromoethyl)benzene at a silver cathode coupled with an oxygen evolution reaction at a platinum anode in a divided cell with organic and aqueous compartments separated by ion-exchange membranes of different types. We examine how operating parameters, including membrane type, applied potential, substrate concentration, , and temperature affect the overall process and the reaction product distribution. Based on the extensive experimental results, we propose a detailed mechanism for major electrochemical product formation accounting for both aprotic and protic environments. Systematic analysis and mechanistic insights presented in this study are expected to enable a rational catalyst, electrolyte, and system design tailored to fi ff electroorganic CO2 xation with di erent organic substrates to obtain industrially relevant carboxylic acids at practical potentials and currents.

KEYWORDS: CO2 utilization, electrocarboxylation, electroorganic synthesis, cell design, membrane, reaction mechanism

■ INTRODUCTION products in EC reactions also causes electrode passivation, resulting in large current fluctuations (potentiostatic electrol- As climate change continues to accelerate, novel technologies 15 that reduce or use excess greenhouse gas emission will play a ysis) or increase in the cell voltage (constant current electrolysis).16 In addition, cathode passivation hinders the Downloaded via UNIV OF WATERLOO on February 2, 2020 at 19:35:03 (UTC). vital role in mitigating the negative impact of human activity 15 fi on our environment. One approach is to develop catalytic completion of the reaction. Replacing sacri cial anodes with alternative anodic reactions at nonsacrificial anodes could offer technologies that convert CO2 into value-added chemicals See https://pubs.acs.org/sharingguidelines for options on how to legitimately share published articles. using renewable electricity.1 Besides direct CO electro- a potential solution to overcome these limitations. While 2 divided cells with nonsacrificial anodes have been reported for reduction to single- and multicarbon small molecules, CO2 17−20 can be electrochemically coupled to various precursors to EC of several organic substrates, for halides any mentions − 21,22 produce a plethora of carboxylic acids.2 5 are limited to unsuccessful attempts. Specifically, electrocarboxylation (EC) of organic halides Despite EC reactions of organic halides with Mg or Al (R−X) has attracted attention as a green alternative for the anodes giving high yields of carboxylation products, their exact synthesis of nonsteroidal anti-inflammatory drugs, such as formation mechanism remains unclear. Electrochemical − 6 reduction of R X in the presence of CO2 is a three-step ibuprofen, and important precursors in industrial processes, • • 7,8 3,4 process: (1) halide reduction to a radical R , (2) R reduction such as cyanoacetic acid, an, and many others. It has been − shown that EC of organohalides with different cathodes (e.g., to the corresponding anion R , and (3) nucleophilic addition − − 23 2 5 fi of R to CO2 with a formation of carboxylate anion. Ag, Cu, and Ni) in undivided cells with sacri cial Mg or Al 9−14 anodes generally results in the formation of corresponding Mechanistic interpretations in previous works operated on 2+ carboxylic acids with moderate to high yields in the form of the assumption that electrochemically produced Mg ions − magnesium (or aluminum) salts.9 14 Although the formation of these salts simplifies the work-up procedure, it negatively Received: August 13, 2019 affects the reaction atom economy due to the anode Revised: November 4, 2019 consumption in the course of reaction. Formation of insoluble Published: November 12, 2019

© 2019 American Chemical Society 19631 DOI: 10.1021/acssuschemeng.9b04647 ACS Sustainable Chem. Eng. 2019, 7, 19631−19639 ACS Sustainable Chemistry & Engineering Research Article participated in the reaction only by interacting with the anodes, we contemplated performing this reaction in a divided carboxylate anion to form a corresponding salt24 (Scheme 1, electrochemical cell with a well-established process at the anodic side. Direct CO electrolyzers typically rely on the pathway I). 2 − oxygen evolution reaction (OER) as an anodic process,35 37 Scheme 1. Possible Pathways of Organic Halide although other oxidation reactions are now considered as well Carboxylation in Undivided Electrochemical Cells with Mg to minimize the cell voltage for their commercial implementa- Sacrificial Anodes tion.35,38 The mechanism of OER and catalysts for this reaction are well-established due to their importance in water splitting for electrochemical hydrogen production.39,40 The difficulty of adopting OER for EC lies in poor solubility of organic substrates in aqueous electrolytes and the need of an aprotic environment for the organic substrate to avoid intermediate protonation.23,41 We hypothesized that this complication can be overcome by introducing an appropriate membrane that can render the aprotic and aqueous compart- ment segregation. Cation-exchange membranes (CEMs) have been previously used in EC reactions in fully organic electrolytes to provide counterions for the carboxylate anions generated at the cathode.3,17,20 However, released halide ions (Hal−) accumulate in the catholyte, causing unwanted side − However, R can also uptake Mg2+ ion to form a Grignard reactions (alkene formation due to the Hal−-catalyzed 25,26 3,42,43 reagent that selectively reacts with CO2, producing a elimination reaction). In addition, in the case of an 27 corresponding carboxylate with quantitative yields (Scheme aqueous anolyte and an aprotic catholyte, CEM may enable 1, pathway II). In fact, the formation of Grignard reagents from competing the hydrogen evolution reaction (HER) at the organic halides in electrochemical cells with Mg anodes has 25,26 cathode by allowing the passage of protons. An anion-exchange been observed in the absence of CO2. Moreover, the membrane (AEM), whose performance has never been studied majority of reported EC of R−X was performed at constant 2−5,9−14 in EC reactions, could address these issues; however, with this current conditions, resulting in inevitable large membrane, counterions for the carboxylate anions can only be potential fluctuations.16 Consequently, at high negative 2+ 0 ° − provided by the catholyte, resulting in its consumption. A potentials, the reduction of Mg to Mg [E = 2.38 V vs bipolar membrane (BPM), which consists of anion and cation- standard hydrogen electrode (SHE)] at the cathode could also exchange layers, combines the properties of both and facilitate the Grignard reagent formation28 (Scheme 1, pathway potentially may either solve or exacerbate the problems III), in addition to enabling competing CO2 electroreduc- 29 associated with CEM and AEM. tion. Other sacrificial anodes (e.g., Al and Zn) have similar reactivity: generated Zn2+ or Al3+ can be cathodically reduced Here, we report a detailed study of EC of organic halides in to their metal forms, which, in addition to causing cathode a divided cell with organic and aqueous compartments 16 separated by various ion-exchange membranes using (1- passivation, produce organometallic compounds by reacting − with organic halides.30,31 These organometallic compounds bromoethyl)benzene (R Br) as a model substrate molecule. can react with CO , producing corresponding carboxylic This new strategy for EC of organic halides does not require − 2 fi acids.32 34 sacri cial anodes, instead relaying on a stable Pt anode and Thus, the chemistry of R−X reduction in the presence of OER as an anodic process. We investigate how operating fi CO2 in undivided cells with sacri cial anodes is rather complex parameters (membrane type, applied potential, substrate and convoluted, with carboxylate anions potentially originating concentration, electrolyte, and temperature) affect the overall from the electrochemically assisted Grignard reagent, rather process and the reaction product distribution. Based on the than the direct EC process. extensive experimental results, we propose a detailed Aiming to assess the potential of purely electrochemical mechanism for major electrochemical products formation fi fi xation of organohalides to CO2 decoupled from sacri cial supported by density functional theory (DFT) calculations.

Figure 1. Cell view and ion transport in EC/OER reactions performed in a divided cell with the Ag cathode, R−X/TBAI/ACN catholyte, Pt anode, KHCO3/H2O anolyte, and CEM (a), BPM (b), or AEM (c).

19632 DOI: 10.1021/acssuschemeng.9b04647 ACS Sustainable Chem. Eng. 2019, 7, 19631−19639 ACS Sustainable Chemistry & Engineering Research Article ■ RESULTS AND DISCUSSION at ∼−1.2 V), while they overlapped in the Ar-saturated Membrane Effects. A divided electrochemical cell with electrolyte. The separation of these peaks in the presence of two compartments separated by an ion-exchange membrane CO2 can be explained by faster kinetics of the second 48 fi reduction step in the presence of CO2. Thus, we identi ed was used to segregate an organic catholyte for EC and an − fi ff three potential ranges (Figure 2): (a) E1 > 1.0 V, where aqueous anolyte for OER. Speci cally, to study the e ect of a • − − reactions of R are dominant; (b) 1.5 V < E2 < 1.0 V, which membrane type on the outcome of EC reaction, we tested (1) fi − fl is promising for chemical xation of CO2 by R ; and (c) E3 < a CEM based on a chemically stabilized per uorosulfonic acid/ − poly(tetrafluoroethylene) copolymer in its protonated form, 1.5 V, where direct CO2 reduction can occur. (2) an AEM with high proton blocking capability based on Consequently, we selected three potentials from these ranges for testing the membranes performance: −0.8, −1.4, polyaromatic structure with quaternary ammonium bromide, − and (3) a BPM consisting of an anion exchange layer and a and 1.7 V. Figure 3 shows the distribution of major cation-exchange layer (Figure 1).44 For all three types of membranes, we performed electrolysis using a CO2-saturated organic electrolyte (tetrabutylammonium iodide in acetonitrile, TBAI/ACN) containing R−Br in the cathodic compartment and aqueous sodium bicarbonate buffer in the anodic compartment. A nanostructured Ag electrode45 [Figure S1, Supporting Information (SI)] prepared by chemical vapor deposition was selected as a cathode material due to its higher activity in EC of R−Br compared to other metals,23 and a Pt mesh was used as a standard OER anode. The distributions of reaction products and current stability (Figure S2, SI) were assessed at various fixed potentials. First, to select meaningful potentials for the potentiostatic tests, we performed cyclic voltammetry (CV) of a pure electrolyte (Ar- saturated), the electrolyte with R−Br and CO , and the Figure 3. Initial assessment of the membranes at an arbitrary (50 2 mM) precursor concentration. Bar graph shows the distribution of electrolyte containing only one of these two components of EC − 46,47 major liquid products in the electrolysis of 50 mM R Br in CO2- (Figures 2 and S2d). In agreement with previous reports, saturated ACN + 0.1 M TBAI in a divided cell with AEM, CEM, or BPM at −0.8, −1.4, and −1.7 V. The product distribution is shown as a molar fraction of their sum, i.e., [EC products] + [R−R] + [R−OR] +[R−H] + [R−OH].

electrochemical products in the liquid phase obtained at these three potentials using different membranes and an arbitrary concentration of R−Br (50 mM). The reactions occurred in the diffusion-controlled regime at this R−Br concentration as confirmed by a linear relationship between the current and square root of scan rate (Figure S3, SI).49 The product distribution was determined by 1H NMR analysis of reaction mixtures using an internal standard (Table S1 and − Figure S4, SI). As a result of EC, carboxylate RCO2 or its Figure 2. Determination of potential ranges: cathodic sweeps of CV ester RCO2R [via second-order nucleophilic substitution recorded at 50 mV s−1 scan rate at Ag in Ar-saturated ACN + 0.25 M − − (SN2) reaction of RCO2 with R Br] was formed. Combined tetrabutylammonium bromide (TBABr); CO -saturated ACN + 0.25 ffi − 2 yields and Faradaic e ciencies (FEs) of RCO2 and RCO2R M TBABr; Ar-saturated ACN + 0.25 M TBABr + 25 mM (1- (EC products) represent EC selectivity in the electrolysis. bromoethyl)benzene; CO2-saturated ACN + 0.25 M TBABr + 25 9−14,50 − • Other products of electrochemical nature included mM (1-bromoethyl)benzene. (a) E1 > 1.0 V: reactions of R ; (b) − − − − −1.5 V < E < −1.0 V: reactions of R−; and (c) E < −1.5 V: direct dimer R R, alcohol R OH, ether R O R, and alkylarene 2 3 R−H. Other side products, such as alkenes and their CO2 reduction. oligomers,42,43,51 can be formed as a result of nonelectrochem- ical transformations of reactive intermediates produced during ∼− the onset potential of CO2 reduction was 1.5 V and the electrolysis and are not included in Figure 3 (for detailed onset potential of R−Br reduction was ∼− 0.5 V (the latter analysis of the formation of minor products, see the Detailed − ° • − corresponds to one electron transfer to R Br: E RBr/R = 0.49 Reaction Mechanism below and Table S1 and Figure S4, SI). V vs SCE in ACN,47 or ∼−0.34 V vs Ag/Ag+; for conversion, At −0.8 V with AEM, the only electrochemical product was − − − − see the SI, page S16), but when both CO2 and R Br were R R (86% yield). At a less negative potential ( 0.6 V), R R present in the electrolyte, a rapid chemical reaction between was obtained in 95% yield and with 96% FE (Table S2, SI) due − ° −• ≈− CO2 and electrochemically generated R (E R/R 0.97 V to the suppression of undesirable side processes, such as vs SCE; ∼−0.82 V vs Ag/Ag+)23 was evident from a elimination reactions catalyzed by Br− generated during the significantlyhighercurrentatpotentialsbelow∼−1V electrolysis.42,43 Thus, we demonstrated that a selective C−C compared to the CV lines of single components.48 Two coupling of organic halides can be performed at low potentials reduction peaks were observed in CV scans of R−Br in the in divided cells with AEM. In contrast, at −0.8 V with BPM → • ∼− • → − − − − CO2-saturated electrolyte (RBr R at 0.85 V; R R and CEM, besides R R, ester R OR and ethylbenzene R H

19633 DOI: 10.1021/acssuschemeng.9b04647 ACS Sustainable Chem. Eng. 2019, 7, 19631−19639 ACS Sustainable Chemistry & Engineering Research Article were formed, with [R−R]/[R−OR]/[R−H] ratios of 8:1:2 (CEM) and 7:1:2 (BPM). At −1.4 V, the formation of EC products was observed with all three membranes. However, AEM showed the best selectivity with [EC products]/[R−R]/[R−OH] ratio of 1.6:1:0.2. In the case of CEM and BPM, [EC products]/ [R−R] ratio dropped to 0.7:1 and 0.3:1, respectively, in addition to the formation of R−OH, R−H, and R−OR. At −1.7 V, EC selectivity improved with all membranes and [EC products]/[R−R] ratio increased to 2.8:1, 2.8:1, and 2:1 for AEM, CEM, and BMP, respectively. The formation of R− H was observed only with CEM and BPM membranes, due to proton transfer from the anodic compartment, which was absent in the case of AEM. While full conversion of R−Br was achieved with AEM at −1.7 V, the use of CEM or BPM led to R−Br conversions below 100%. Visual examination of the cathodes after CEM and BPM electrolysis at −1.7 V (Figure S5d,e, SI) revealed the formation of solid products on their surface (primarily KBr and fi traces of K2CO3), that resulted in signi cant passivation of the cathodes and a subsequent reaction rate decay. In contrast, electrolysis with AEM led to no (at −1.4 V) or minor (at −1.7 V) electrode passivation (Figure S5b,c, SI): due to the absence of K+ ions in the cathodic compartment (see Figure 1), Figure 4. Assessment of electrolysis stability and kinetics. (a) ff − possible insoluble by-products were limited to TBA2CO3, but Performance of di erent membranes at 1.7 V: calculated (see the only at potentials above the onset for CO2 reduction leading to SI, page S7) and measured reaction rate decays during the electrolysis 2− of 50 mM R−Br in ACN + 0.1 M TBAI in a divided cell at −1.7 V the formation of CO and CO3 . We note that attempts to with AEM, CEM, and BPM. Experimentally observed trends (solid minimize the passivation when using CEM or BPM by using − − − 0.1 M sulfuric acid as anolyte led to the decrease in EC lines) at 1.4 V (b) and 0.8 V (c) during electrolysis of 50 mM R Br in ACN + 0.1 M TBAI with AEM, and calculated rate decays selectivity and FE due to competing HER at the cathode considering 1e− (dotted lines) and 2e− (dashed lines) reduction of (Table S2, SI). R−Br (see the SI, page S7). Trends in the reaction performance with different membranes can be illustrated by the dependence of passed charge as a function of time, Q(t)(Figure 4). We note that the Another noteworthy difference in the performance of the electrochemical performance of the nanoporous Ag electrodes membranes is related to the oxidation process at the anode. in EC was the same after at least 10 sequential 5 h reactions; Since TBAI was used in the cathodic compartment, OER therefore, we ascribe observed changes in current to the observed at the anode in the beginning of electrolysis was reaction kinetics or electrode passivation with insoluble salts ultimately replaced by iodine formation when AEM and BPM (Figure S5). At −1.7 V in the case of AEM, the Q(t) plot was − were used. close to the ideal scenario of 2e reduction of R−Br, The ion transport and corresponding side reactions considering the depletion of R−Br in the course of reaction responsible for the observed product distribution variation (for calculation details, see the SI, page S7). When electrode with different membranes are summarized in Figure 1. In the passivation is negligible, the deviation from the calculated beginning of electrolysis, only two processes are possible at the − trend at later stages of the electrolysis can also indicate a faster cathode with all three membranes: R X reduction and CO2 − − − − 2− than calculated depletion of R Br, suggesting its participation reduction, which generate anions RCO2 ,R ,X , and CO3 ̅. in nonelectrochemical transformations (e.g., SN2 reaction with Stabilization of these anions in the case of CEM is performed − + + RCO2 ; see the Detailed Reaction Mechanism below). In by H /K transferred from the anodic side (Figure 1a). − − CEM and BPM experiments at 1.7 V (Figure 4a), Q(t) Carboxylate anions RCO2 combine with protons producing curves significantly deviated from the calculated trend, in the target acid, while the inorganic anions form insoluble agreement with the observed cathode passivation. As a result of potassium salts KX (X = Br, I) and K2CO3, which passivate the this drastic decrease in current before the completion of the cathode. In the case of AEM (Figure 1c), stabilization of reaction, only 60−80% conversion of R−Br was achieved with anions can only be performed with TBA+ cations from the CEM and BPM. At less negative potentials (− 1.4 and −0.8 electrolyte, while I− along with Br− from the starting organic V), cathode passivation with BPM was noticeably lower and bromide passes through the membrane to the anodic full conversion of R−Br could be achieved. Poor performance compartment. The initially negligible concentration of these of CEM with respect to R−Br conversion was observed at all anions in the anolyte gradually increases in the course of tested potentials (Figure S6, SI). Importantly, electrolysis with electrolysis. With enough I− in the anolyte, iodine formation AEM at less negative potentials closely followed the calculated dominates over OER due to a lower oxidation potential of − − trends of 2e and 1e reduction of R−Br at −1.4 and −0.8 V, iodide anions compared to that of water (0.54 and 1.23 V vs respectively (Figure 4b). This information is important for SHE, respectively).52 To maintain OER at the anode, ff − − further elucidation of the reaction mechanisms in di erent electrolytes with less reactive anions such as Br or ClO4 potential ranges (see the Detailed Reaction Mechanism can be used. In the case of BPM (Figure 1b), all of the below). processes described above take place. Despite BPM combining

19634 DOI: 10.1021/acssuschemeng.9b04647 ACS Sustainable Chem. Eng. 2019, 7, 19631−19639 ACS Sustainable Chemistry & Engineering Research Article the properties of both CEM and AEM, its efficiency and Besides R−R and EC products, the formation of R−OH was selectivity were the lowest (Figure 3). High yields of also observed at potentials more negative than −1.5 V with protonation product R−H were detected at all potentials, maximum yield at −2 V (16%). The yield of the R−OH was along with low yields of EC products and low R−Br similar at both concentrations. Since R−OH formation 55 conversion (Table S2, SI). This low performance of BPM requires the presence of H2O in the reaction medium, this could be associated with enhanced water dissociation in the observation suggests that at potentials below −1.6 V water membrane layers53 and membrane degradation due to starts to pass through AEM; therefore, applicable potentials for delamination of the layers in the organic−aqueous system.54 EC in ACN/AEM/water system lay above this potential value. Since AEM showed the best overall performance, further Effect of the Starting Precursor Concentration. To investigation of the effect of reaction parameters on the course further elucidate the influence of the initial R−Br concen- of electrolysis was performed using this membrane. We note tration on the product distribution, a series of electrolyses of that AEM also showed good stability for at least 50 h of R−Br with different initial concentrations (from 2.5 to 100 continuous or sequential electrolysis without any significant mM) were performed at −1.4 V. Figure 5b shows that the visual changes or decrease in reaction selectivity. lower the starting concentration of R−Br, the lower the yield Effect of the Applied Potential. A series of electrolyses of R−R, with an almost linear trend. The formation of side of R−Br at different cathode potentials were performed to products (appearing as additional signals between 3.5 and 6.5 assess the potential influence on the product distribution in ppm in the NMR spectra of the reaction mixture; see Figure more detail (Figure 5a). The dimer R−R was the major S4, SI) was also noticeably suppressed at lower starting R−Br concentrations. Consequently, the yield of EC products increased nearly exponentially with the decreasing concen- tration of the precursor. These trends indicated that at low concentrations of the R−Br radical coupling and non- electrochemical side reactions were suppressed. The nature ff − of the e ect is closely related to the ratio between CO2 and R Br (or its reduction intermediates) in the reaction mixture: if − CO2 is readily available once R is formed, the following nucleophilic addition (Ad ) proceeds selectively with the N − formation of the corresponding carboxylic anion (RCO2 ). In addition, with the decrease in concentration, a maximum yield of EC products was observed at less negative potentials, due to a higher contribution of direct CO2 reduction to CO (Figure S7a, SI).36,37 Overall, the maximum yields of EC products were achieved with AEM at −1.4 V in dilute solutions (up to quantitative at 2.5 mM). Although in dilute R−Br solutions R−R formation could be completely suppressed, other side reactions were still observed due to the increased ratio of water traces (from the anolyte) and reactive anion R− generated in EC in the reaction mixture, in addition to a higher contribution of competing CO2 reduction (see Detailed Reaction Mechanism section). Effect of the Charge Passed. Considering the effects described above and gradual depletion of the starting material in the course of the reaction, we examined how the reaction selectivity was changing in time with the charge passed. The contribution of direct CO2 electroreduction to CO and its Figure 5. Trends in the major product distribution based on applied − potentials and starting R−Br concentrations. (a) Yields of the major impact on the selectivity of EC with R Br depletion were also investigated. We note that only minimal contribution of direct products in the electrolysis of 50 mM (empty markers) and 25 mM − (solid markers) R−Br (AEM) as a function of applied potential. (b) CO2 reduction is expected at 1.4 V, as it is above the Yields of the major products at −1.4 V (AEM) as a function of the apparent onset potential for CO2 reduction, according to CV initial concentration of R−Br. Error bars correspond to standard sweeps (Figure 2). In addition, the membrane separating deviations. The lines are given for eye guidance. aprotic organic and aqueous electrolytes could allow traces of water to cross over, resulting in subsequent HER at the + product (yields up to 86%) at E > −1.4 V vs Ag/Ag . The yield cathode. The distribution of gas products (CO and H2)asa of R−R significantly decreased at more negative potentials function of reaction time was evaluated using in-line gas approaching 0 at −2 V (only trace amounts were detected by chromatography (GC) analysis (Figure 6a) in a series of − NMR). EC products dominated at −1.8 V (35%) and −1.6 V electrolyses of 50 mM R Br in the presence of CO2 at (45%) at 50 and 25 mM concentrations of R−Br, respectively. different potentials (−0.8, −1.4, −1.7, −2.2 V; Figure S8, SI). − A higher yield of EC products along with a lower yield of R−R With respect to gas products, at 1.4 V, FEs of CO and H2 at a lower starting R−Br concentration (while CO were <1% until the depletion of R−Br, indicating the 2 − concentration was the same) at almost all potentials points dominance of the R Br reduction over the direct CO2 − − at the importance of optimizing the [R Br]/[CO2] ratio to reduction and HER. With the depletion of R Br, FEs of gas maximize the EC yield (see Effect of the Starting Precursor products started to grow and reached maxima (FECO = 78% − Concentration and Detailed Reaction Mechanism sections). and FEH2 = 22%) after the full consumption of R Br (note

19635 DOI: 10.1021/acssuschemeng.9b04647 ACS Sustainable Chem. Eng. 2019, 7, 19631−19639 ACS Sustainable Chemistry & Engineering Research Article

Figure 6. Evaluation of consistency of product distribution in the course of electrolysis. Product distribution during electrolysis of 50 mM R−Br with AEM: (a) FEs of gas and liquid products (GC data) and charge passed in time at −1.4 V showing the effect of R−Br depletion on the process (the corresponding current density plot is shown in Figure S2b). (b, c) FEs of EC products (b) and R−R (c) as a function of charge passed for electrolysis at different potentials. Error bars correspond to standard deviations. The lines in (b) and (c) are given for eye guidance.

Scheme 2. Proposed Mechanism and Side Reactions of Electroreduction of Organic Halides and CO in Different Potential a 2 Ranges

a − − − − * −• (a) E1 > 1.0 V, (b) 1.5 V < E2 < 1.0 V, and (c) E3 < 1.5 V. ( )CO2 intermediate can react with CO2 followed by reductive 2− disproportionation to CO and CO3 . significantly lower currents compared to the average for the EC increased at −1.4 V (consistent with the decrease in FEs of R− duration; Figures 6a and S2b). At a less negative potential R), in agreement with the concentration dependence trend, (−0.8 V; Figure S8a, SI), the formation of gas products was but at −1.7 V, it slightly decreased (Figure 6b). This different suppressed and R−Br reduction was the major process. At behavior undefined suggests a variation in the EC reaction more negative potentials (−1.7 and −2.2 V; Figure S8b,c, SI), mechanism at these two potentials. Indeed, since at −1.7 V − − R Br reduction was still the dominant process over CO and CO2 reduction was more pronounced than at 1.4 V (Figure H2 formation initially, although CO formation was observed 2), the EC reaction could take place through CO2 reduction to −• − earlier and at higher partial current density values. a radical anion CO2 and its subsequent reaction with R Br. −• To understand the main changes in the distribution of major In this case, a competing process was CO2 disproportiona- 56 2− liquid products in the course of electrolysis, the reaction tion to CO and CO3 ̅, which became more prominent with mixtures were analyzed by 1H NMR after passing every 0.3 F the decreasing R−Br concentration, explaining the reduction in − mol 1 until the completion of the reaction (Figures 6b,c and EC product yields with electrolysis progression. S9). At −0.8 V, the formation of EC products was not Effect of the Temperature. Since the concentration of − − observed and FE of R R was constant until complete R Br CO2 increases with the decreasing temperature, the latter −1 ff − • − consumption after passing 1.0 F mol . At more negative a ects the ratio of R Br (or R /R )andCO2 and potentials (−1.4 and −1.7 V), FEs of R−R were initially lower consequently the product distribution. Indeed, the electrolysis and gradually dropped with the electrolysis time (Figure 6c). of 50 mM R−Br at −1.2 V at a reduced temperature favored This behavior was associated with the decrease in R−Br EC products, with their yields increasing from 15% at room concentration in the course of electrolysis, in agreement with temperature to 30% at 0 °C(Figure S7b, SI), while the yield of the suppression of the dimerization reaction observed in dilute R−R dropped from 35 to 27%. At the same time, the R−Br solutions (Figure 5b). With respect to EC product combined yield of EC products and R−R was higher at a distribution in time at more negative potentials, it gradually reduced temperature, indicating that electroorganic side

19636 DOI: 10.1021/acssuschemeng.9b04647 ACS Sustainable Chem. Eng. 2019, 7, 19631−19639 ACS Sustainable Chemistry & Engineering Research Article fi − reactions were suppressed (con rmed by NMR). A similar formation of R R at high negative potentials represents an SN2 ff − temperature e ect was also observed for more negative reaction of active anions with the starting R Br. The [CO2]/ potentials (Figure S7c, SI). [RBr] ratio plays an important role in this process: if the Detailed Reaction Mechanism. A general mechanism for reaction is too fast, the concentration of CO at the cathode 2 − electroreduction of organic halides and CO in divided cells surface is not sufficient for it to react with every generated R 2 − with nonsacrificial anodes that considers all possible pathways therefore, some of R reacts with R−Br or participates in other consistent with our experimental data and earlier re- undesirable side reactions. 2−5,9−14,50 − ports is summarized in Scheme 2. The pathways In dilute R Br solutions, CO2 is present in excess, leading to are separated into blocks according to the following potential the suppression of side reactions and a selective formation of − • − ranges: E1 > 1.0 V, corresponding to R formation and its the carboxylate anion. The formed RCO2 then reacts with − − + + + subsequent transformations (Scheme 2a); 1.5 V < E2 < 1.0 cations, i.e., H or K for BPM, CEM, or TBA for AEM. Since − V, corresponding to R formation and its reactions without TBA salt is more nucleophilic than the acid, the rate of its SN2 − − competing CO2 reduction (Scheme 2b); E3 < 1.5 V, showing reaction with R Br is higher and therefore higher yields of the the influence of CO reduction on the course of EC of organic ester are observed for AEM.57,62 2 − fi halides (Scheme 2c). At more negative potentials (E3 < 1.5 V), a signi cant The electrochemical R−Br activation starts with the cleavage formation of CO starts only near a full conversion of R−Br. • of the C−Hal bond with the formation of R , followed by the With the increase in [CO2]/[RBr] ratio associated with the − − reduction of the radical to a corresponding carbanion R , depletion of R Br, CO2 reduction to CO occurs through the 57 • −• 63 which then reacts with CO2. The generation of R either formation of CO2 intermediate. The absence of CO in the occurs via the initial formation of anion radical intermediate beginning of the reaction is likely due to a higher affinity of the •− −23,58 − RBr followed by the elimination of Br or via a Ag cathode to R Br than CO2, or due to the reaction of 58 −• − • 61 fi −• concerted mechanism. Besides the direct electrochemical CO2 with R Br (or R ). Speci cally, CO2 reacting with − • reduction of the starting bromide, the halogen exchange with excess R Br can form CO2R intermediate (similarly to 59 • electrolyte (TBAI) can also occur. Indeed, the analysis of the CO2H intermediate formation in CO2 electroreduction in reaction mixture before a full conversion of the starting aqueous media).63 This intermediate can further be reduced −1 ̅ − material (after passing 0.3−1.2 F mol ) showed the presence with the formation of either CO and RO anion or CO2R . The of both R−Br and R−I in reaction mixtures (Figure S9, SI) latter is an active carbanion that can react with R−Br to form with bromide being dominant. Since the standard reduction the ester (Scheme 2c). potential of (1-iodoethyl)benzene is more positive than that DFT calculations of energy barriers for the initial steps of 47 − for (1-bromoethyl)benzene (ΔE ≈ 0.15 V), the presence of R Br and CO2 reduction on the Ag surface show their I− in the electrolyte can lower the energy requirement for R• dependence on the applied potential (Figures 7 and S13). generation and thereby decrease the cell potential. At potentials less negative than E(R•/R−), only radical transformations are possible (E1, Scheme 2a). Indeed, the majority of R• underwent the radical coupling reaction, giving the dimer R−R as a main reaction product (up to 95% at −0.6 V). Based on DFT calculations, the dimerization likely occurs in solution, as the energy barrier for the radical coupling in solution is lower than on the electrode surface (Figures S10− S12,SI).Besidesdimerization,R• can be involved in disproportionation, H-atom abstraction, nucleophilic SN1-like reactions, rearrangements, and addition to unsaturated systems (e.g., styrene) originating from the elimination reaction in the Figure 7. Calculated energy barriers for the initial steps of reduction ff − − starting bromide mediated by di erent nucleophiles (e.g., Br ) of starting R Br and CO2. generated during electrolysis.42,43,51,59 The presence of R−H and R−OR in the ratio 2:1 was observed when using BPM and CEM. Typically, the formation of both products is attributed Specifically, at low potentials, the difference is large, with − + 50 fi − to the R transformations in the presence of water or H . signi cantly higher barriers for CO2 reduction, leading to R Alternatively, the formation of R−H can also be explained by Br reduction being the primary departure point for the the abstraction of hydrogen atom from the solvent (or any chemical transformations. At −1.5 V vs SHE (approximately other H donors) by R•,56,60 or via disproportionation.60,61 Our −1.59 V vs Ag/Ag+; Figure S14, SI), the trend switches, and at + experimental results show that the presence of H in the more negative potentials, CO2 reduction is more favorable catholyte is critical for the formation of R−H and R−OR, since than the reduction of R−Br (Scheme 2c). This result further they were not observed in the AEM experiments at −0.8 V. A supports the mechanism differentiation described in Scheme 2. constant ratio between these two products indicated that they originated from the initial disproportionation reaction, ■ CONCLUSIONS catalyzed by H+ (Scheme 2c). Carbocations R+ generated In conclusion, we demonstrated the principal feasibility of during disproportionation react with the traces of water to performing EC reactions in divided cells with aqueous analytes form R−OH, which subsequently undergoes an alkylation and nonsacrificial anodes. We laid out the critical role of the reaction to produce R−OR. membrane separator type and showed that AEM has a superior − − At potentials 1.5 V < E2 < 1.0 V, the reduction of radicals performance with the highest EC selectivity and the lowest R• to R− becomes a dominant process (Scheme 2b), which cathode passivation, compared to those of CEM and BPM. leads to a dramatic decrease in the R−R yield (Figure 5a). The This approach opens up the possibility of tuning the overall

19637 DOI: 10.1021/acssuschemeng.9b04647 ACS Sustainable Chem. Eng. 2019, 7, 19631−19639 ACS Sustainable Chemistry & Engineering Research Article fi cell potential by combining electroorganic CO2 xation with (9) Yang, H.-P.; Zhang, H.-W.; Wu, Y.; Fan, L.-D.; Chai, X.-Y.; various anodic reactions and performing the reaction Zhang, Q.-L.; Liu, J.-H.; He, C.-X. A Core−Shell-Structured Silver continuously using flow electrolyzers. Importantly, we Nanowire/Nitrogen-Doped Carbon Catalyst for Enhanced and elucidated the interplay of different reaction pathways in this Multifunctional Electrofixation of CO2. ChemSusChem 2018, 11, − system and proposed a complete reaction mechanism 3905 3910. accounting for all observed products. By optimizing the (10) Wu, L.-X.; Sun, Q.-L.; Yang, M.-P.; Zhao, Y.-G.; Guan, Y.-B.; Wang, H.; Lu, J.-X. Highly Efficient Electrocatalytic Carboxylation of reaction parameters, we were able to achieve quantitative yields − − 1-Phenylethyl Chloride at Cu Foam Cathode. Catalysts 2018, 8, 273 of EC at high [CO2]/[R Br] ratios. With further system fl 282. optimization, potentially including a transition to a ow (11) Wang, H.-M.; Sui, G.-J.; Wu, D.; Feng, Q.; Wang, H.; Lu, J.-X. reactor, this approach holds promise for continuously carrying Selective electrocarboxylation of bromostyrene at silver cathode in fi out electroorganic CO2 xation to obtain industrially relevant DMF. Tetrahedron 2016, 72, 968−972. carboxylic acids at practical potentials and currents. (12) Senboku, H.; Nagakura, K.; Fukuhara, T.; Hara, S. Three- component coupling reaction of benzylic halides, carbon dioxide, and ■ ASSOCIATED CONTENT N,N-dimethylformamide by using paired electrolysis: sacrificial *S Supporting Information anode-free efficient electrochemical carboxylation of benzylic halides. − The Supporting Information is available free of charge at Tetrahedron 2015, 71, 3850 3856. (13) Yang, H.; Wu, L.; Wang, H.; Lu, J. Cathode made of compacted https://pubs.acs.org/doi/10.1021/acssuschemeng.9b04647. silver nanoparticles for electrocatalytic carboxylation of 1-phenethyl − Additional experimental and electrochemical data, SEM bromide with CO2. Chin. J. Catal. 2016, 37, 994 998. images of Ag cathode, details of qualitative and (14) Yang, H.-P.; Yue, Y.-N.; Sun, Q.-L.; Feng, Q.; Wang, H.; Lu, J.- quantitative product analyses (NMR and GC), addi- X. Entrapment of a chiral cobalt complex within silver: a novel tional DFT results, and details of calculations (PDF) heterogeneous catalyst for asymmetric carboxylation of benzyl − bromides with CO2. Chem. Commun. 2015, 51, 12216 12219. (15) Scialdone, O.; Galia, A.; Belfiore, C.; Filardo, G.; Silvestri, G. ■ AUTHOR INFORMATION Influence of the Experimental System and Optimization of the Corresponding Author Selectivity for the Electrocarboxylation of Chloroacetonitrile to *E-mail: [email protected]. Cyanoacetic Acid. Ind. Eng. Chem. Res. 2004, 43, 5006−5014. (16) Datta, A. K.; Marron, P. A.; et al. Process development for ORCID electrocarboxylation of 2-acetyl-6-methoxynaphthalene. J. Appl. Anna Klinkova: 0000-0003-0337-0579 Electrochem. 1998, 28, 569−577. Notes (17) van Tilborg, W. J. M.; Smith, C. J. The electrosynthesis of The authors declare no competing financial interest. carboxylic acids from carbon dioxide and butadiene. Recl. Trav. Chim. Pays-Bas 1981, 100, 437−438. ■ ACKNOWLEDGMENTS (18) Grinberg, V. A.; Koch, T. A.; Mazin, V. M.; Mysov, E. I.; Sterlin, S. R. Electrochemical reduction of CO2 in the presence of 1,3- The authors thank the University of Waterloo, National butadiene using a hydrogen anode in a nonaqueous medium. Russ. Science and Engineering Research Council of Canada, and Chem. Bull. 1999, 48, 294−299. Canada Foundation for Innovation for financially supporting (19) Matthessen, R.; Fransaer, J.; Binnemans, K.; De Vos, D. E. this work and the Linda Nazar group for their help with DFT Paired Electrosynthesis of Diacid and Diol Precursors Using Dienes − calculations. and CO2 as the Carbon Source. ChemElectroChem 2015, 2,73 76. (20) Wagenknecht, J. H. Electrochemical Carboxylation of p- ■ REFERENCES Isobutilacetophenone. US4,582,577, April 15, 1986. (21) Wawzonek, S.; Shradel, J. M. Electrochemical Reduction of (1) Seh, Z. W.; Kibsgaard, J.; Dickens, C. F.; Chorkendorff, I.; Benzal Chloride in the Presence of Carbon Dioxide. J. Electrochem. Nørskov, J. K.; Jaramillo, T. F. Combining theory and experiment in Soc. 1979, 126, 401−403. electrocatalysis: Insights into materials design. Science 2017, 355, (22) Baizer, M. M.; Chruma, J. L. Electrolytic Reductive Coupling. No. eaad4998. XXI. Reduction of Organic Halides in the Presence of Electrophiles. J. (2) Yan, M.; Kawamata, Y.; Baran, P. S. Synthetic Organic − Electrochemical Methods Since 2000: On the Verge of a Renaissance. Org. Chem. 1972, 37, 1951 1960. Chem. Rev. 2017, 117, 13230−13319. (23) Huang, Y.-F.; Wu, D.-Y.; Wang, A.; Ren, B.; Rondinini, S.; (3) Matthessen, R.; Fransaer, J.; Binnemans, K.; De Vos, D. E. Tian, Z.-Q.; Amatore, C. Bridging the Gap between Electrochemical and Organometallic Activation: Benzyl Chloride Reduction at Silver Electrocarboxylation: towards sustainable and efficient synthesis of − valuable carboxylic acids. Beilstein J. Org. Chem. 2014, 10, 2484−2500. Cathodes. J. Am. Chem. Soc. 2010, 132, 17199 17210. (4) Cao, Y.; He, X.; Wang, N.; Li, H.-R.; He, L.-N. Photochemical (24) Silvestri, G. In Carbon Dioxide as a Source of Carbon, and Electrochemical Carbon Dioxide Utilization with Organic Biochemical and Chemical Uses;Aresta, M., Forti, G., Eds.; Kluwer: − Compounds. Chin. J. Chem. 2018, 36, 644−659. Dordrecht, The Netherlands, 1987; pp 360 361. (5) Rossi, L. Electrochemical Methodologies for the Carboxylation (25) Luder, D.; Ein-Eli, Y. Electrochemical Grignard Reagent − Reactions in Organic Synthesis. An Alternative Re-use of Carbon Synthesis for Ionic-Liquid-Based Magnesium Air Batteries. Chem- Dioxide. Curr. Green Chem. 2015, 2,77−89. ElectroChem 2014, 1, 1319−1326. (6) Mena, S.; Sanchez, J.; Guirado, G. Electrocarboxylation of 1- (26) Luder, D.; Kraytsberg, A.; Ein-Eli, Y. Catalyst-Free Electro- chloro-(4- isobutylphenyl)ethane with a silver cathode in ionic chemical Grignard Reagent Synthesis with Room-Temperature Ionic liquids: an environmentally benign and efficient way to synthesize Liquids. ChemElectroChem 2014, 1, 362−365. Ibuprofen. RSC Adv. 2019, 9, 15115−15123. (27) Dowson, G. R. M.; Dimitriou, I.; Owen, R. E.; Reed, D. G.; (7) Batanero, B.; Barba, F.; et al. Paired Electrosynthesis of Allen, R. W. K.; Styring, P. Kinetic and economic analysis of reactive Cyanoacetic Acid. J. Org. Chem. 2004, 69, 2423−2426. capture of dilute carbon dioxide with Grignard reagents. Faraday (8) Isse, A. A.; Gennaro, A. Electrochemical Synthesis of Discuss. 2015, 183,47−65. Cyanoacetic Acid from Chloroacetonitrile and Carbon Dioxide. J. (28) Seyferth, D. The Grignard Reagents. Organometallics 2009, 28, Electrochem. Soc. 2002, 149, D113−D117. 1598−1605.

19638 DOI: 10.1021/acssuschemeng.9b04647 ACS Sustainable Chem. Eng. 2019, 7, 19631−19639 ACS Sustainable Chemistry & Engineering Research Article

(29) Luo, P.-P.; Zhang, Y.-T.; Chen, B.-L.; Yu, S.-X.; Zhou, H.-W.; (49) An, C.; Kuang, Y.; Fu, C.; Zeng, F.; Wang, W.; Zhou, H. Study Qu, K.-G.; Kong, Y.-X.; Huang, X.-Q.; Zhang, X.-X.; Lu, J.-X. on Ag−Pd bimetallic nanoparticles for electrocatalytic reduction of Electrocarboxylation of Dichlorobenzenes on a Silver Electrode in benzyl chloride. Electrochem. Commun. 2011, 13, 1413−1416. DMF. Catalysts 2017, 7, 274−284. (50) Isee, A. A.; Gennaro, A. Electrochemical reduction of benzyl (30) Yakhvarov, D. G.; Ganushevich, Y. S.; Dobrynin, A. B.; bromide in the presence of carbon dioxide. Indian J. Chem. 2003, 42A, − Krivolapov, D. B.; Litvinov, I. A.; Sinyashin, O. G. Electrochemical 751 757. (51) Klymenko, O. V.; Buriez, O.; Labbe, E.; Zhan, D.-P.; Rondinini, Reduction of ZnBr2 in the Presence of Organic Halides. Russ. J. Electrochem. 2009, 45, 139−144. S.; Tian, Z.-Q.; Svir, I.; Amatore, C. Uncovering the Missing Link (31) Bala, K.; Singh, B. Direct Electrochemical Synthesis of Unique between Molecular and Electrocatalysis: Mechanism of the Reduction of Benzyl Chloride at Silver Cathodes. Organoaluminium Halides and Their Coordination Complexes at − Sacrificial Aluminium Anode. Asian J. Chem. 2018, 30, 1070−1074. ChemElectroChem 2014, 1, 227 240. (32) Sakakura, T.; Choi, J.-C.; Yasuda, H. Transformation of Carbon (52) Zhao, Y.; Gu, S.; Gong, K.; Zheng, J.; Wang, J.; Yan, Y. Iodine Redox-Mediated Electrolysis for Energy-Efficient Chlorine Regener- Dioxide. Chem. Rev. 2007, 107, 2365−2387. ation from Gaseous HCl. J. Electrochem. Soc. 2017, 164, E138−E143. (33) Kobayashi, K.; Kondo, Y. Transition-Metal-Free Carboxylation (53) Pourcelly, G. Electrodialysis with Bipolar Membranes: 2009 of Organozinc Reagents Using CO2 in DMF Solvent. Org. Lett. , Principles, Optimization, and Applications. Russ. J. Electrochem. 11, 2035−2037. − ’ 2002, 38, 919 926. (34) Yur ev, V. P.; Kuchin, A. V.; Tolstikov, G. A. Reaction of (54) Ramdin, M.; Morrison, A.; de Groen, M.; van Haperen, R.; de aluminum trialkyls with carbon dioxide. Russ. Chem. Bull. 1974, 23, Kler, R.; van den Broeke, L. J. P.; Trusler, J. P. M.; de Jong, W.; Vlugh, 817−819. T. J. H. High Pressure Electrochemical Reduction of CO2 to Formic (35) Verma, S.; Lu, S.; Kenis, P. J. A. Co-electrolysis of CO2 and Acid/ Formate: A Comparison between Bipolar Membranes and glycerol as a pathway to carbon chemicals with improved Cation Exchange Membranes. Ind. Eng. Chem. Res. 2019, 58, 1834− technoeconomics due to low electricity consumption. Nat. Energy 1847. 2019, 4, 466−474. (55) Chougala, B. M.; S amundeeswari, S.; Holiyachi, M.; Shastri, L. (36) Shi, J.; Shen, F.-X.; Shi, F.; Song, N.; Jia, Y.-J.; Hu, Y.-Q.; Li, Q.- A. Mild, Efficient and Catalyst-Free Hydroxylation of Alkyl Halides in Y.; Liu, J.-X.; Chen, T.-Y.; Dai, Y.-N. Electrochemical reduction of Water: Significant Enhancement of Water Nucleophilicity in Dipolar − CO2 into CO in tetrabutylammonium perchlorate/propylene Solvents. ChemistrySelect 2017, 2, 1290 1296. carbonate: Water effects and mechanism. Electrochim. Acta 2017, (56) Singh, M. R.; Goodpaster, J. D.; Weber, A. Z.; Head-Gordon, 240, 114−121. M.; Bell, A. T. Mechanistic insights into electrochemical reduction of (37) Shen, F.-x.; Shi, J.; Shi, F.; Chen, T.-y.; Li, Y.-f.; Li, Q.-y.; Dai, CO2 over Ag using density functional theory and transport models. − Y.-n.;Yang,B.;Qu,T.FabricationofAg2Selectrodefor Proc. Natl. Acad. Sci. U.S.A. 2017, 114, E8812 E8821. − (57) Scialdone, O.; Galia, A.; Errante, G.; Isee, A. A.; Gennaro, A.; CO2reduction in organic media. Ionics 2019, 25, 1921 1927. (38) Li, T. F.; Cao, Y.; He, J. F.; Berlinguette, C. P. Electrolytic CO Filardo, G. Electrocarboxylation of benzyl chlorides at silver cathode 2 − Reduction in Tandem with Oxidative Organic Chemistry. ACS Cent. at the preparative scale level. Electrochim. Acta 2008, 53, 2514 2528. ́ Sci. 2017, 3, 778−783. (58) Andrieux, C. P.; Saveant, J.-M.; Tallec, A.; Tardivel, R.; Tardy, (39) Jiao, Y.; Zheng, Y.; Jaroniec, M.; Qiao, S. Z. Design of C. Concerted and Stepwise Dissociative Electron Transfers. electrocatalysts for oxygen- and hydrogen-involving energy conversion Oxidability of the Leaving Group and Strength of the Breaking − Bond as Mechanism and Reactivity Governing Factors Illustrated by reactions. Chem. Soc. Rev. 2015, 44, 2060 2086. α (40) Suen, N.-T.; Hung, S.-F.; Quan, Q.; Zhang, N.; Xu, Y.-J.; Chen, the Electrochemical Reduction of -Substituted Acetophenones. J. Am. Chem. Soc. 1997, 119, 2420−2429. H.-M. Electrocatalysis for the oxygen evolution reaction: recent (59) Han, H.; Kumar, R.; Tsarevsky, N. V. Responsive and 2017 − development and future perspectives. Chem. Soc. Rev. , 46, 337 Degradable Highly Branched Polymers with Hypervalent Iodine(III) 365. Groups at the Branching Points. Macromol. Rapid Commun. 2019, 40, (41) Deunf, E.; Labbe, E.; Verpeaux, J. N.; Buriez, O.; Amatore, C. No. 1900073. Direct electrochemical reduction of organic halide droplets dispersed (60) Franz, J. A.; Suleman, N. K.; Alnajjar, M. S. Absolute Rate − in water. RSC Adv. 2012, 2, 5398 5402. Constants for Hydrogen Atom Abstraction by Benzyl Radical from (42) Wolf, N. L.; Peters, D. G.; Mubarak, M. S. Electrochemical Thiophenol, Tributylstannane, Tributylstannane-d, and Dicyclohex- Reduction of 1-Halooctanes at Platinized Platinum Electrodes in ylphosphine and for the Cyclization of the 2-Allylbenzyl Radical. J. Dimethylformamide Containing Tetramethylammonium Tetrafluor- Org. Chem. 1986, 51,19−25. oborate. J. Electrochem. Soc. 2006, 153,E1−E4. (61) Yimin, D.; Lanli, N.; Hui, L.; Jiaqi, Z.; Linping, Y.; Qiuju, F. (43) Simonet, J.; Peters, D. G. Electrochemical Conversion of Cu-Ni Alloy Catalyzed Electrochemical Carboxylation of Benzyl Primary Alkyl Halides to Alkenes at Platinum Cathodes. J. Bromide with Carbon Dioxide in Ionic Liquid 1-Butyl-3-methyl- Electrochem. Soc. 2004, 151,D7−D12. imidazolium tetrafluoroborate. Int. J. Electrochem. Sci. 2018, 13, (44) Garcia-Herrero, I.; Alvarez-Guerra, M.; Irabien, A. Electrosyn- 1084−1095. thesis of dimethyl carbonate from methanol and CO2 using potassium (62) Chu, D.-B.; Li, X.-H.; Liu, X.-Y.; Yao, W.-L. Electrolytic fixation methoxide and the ionic liquid [bmim][Br] in a filter-press cell: a of CO2 by electrocarboxylation of RX on nanocrystalline TiO2-Pt study of the influence of cell configuration. J. Chem. Technol. cathode. Chin. J. Chem. 2004, 22, 1231−1234. Biotechnol. 2016, 91, 507−513. (63) Zhang, X.-G.; Jin, X.; Wu, D.-Y.; Tian, Z.-Q. Selective (45) Yuan, Z.; Dryden, N. H.; Vittal, J. J.; Puddephatt, R. Chemical Electrocatalytic Mechanism of CO2 Reduction Reaction to CO on Vapor Deposition of Silver. Chem. Mater. 1995, 7, 1696−1702. Silver Electrodes: A Unique Reaction Intermediate. J. Phys. Chem. C − (46) König, M.; Vaes, J.; Klemm, E.; Pant, D. Solvents and 2018, 122, 25447 25455. Supporting Electrolytes in the Electrocatalytic Reduction of CO2. iScience 2019, 19, 135−160. (47) Isse, A. A.; Lin, C. Y.; Coote, M. L.; Gennaro, A. Estimation of Standard Reduction Potentials of Halogen Atoms and Alkyl Halides. J. Phys. Chem. B 2011, 115, 678−684. (48) Niu, D.-F.; Xiao, L.-P.; Zhang, A.-J.; Zhang, G.-R.; Tan, Q.-Y.; Lu, J.-X. Electrocatalytic carboxylation of aliphatic halides at silver cathode in acetonitrile. Tetrahedron 2008, 64, 10517−10520.

19639 DOI: 10.1021/acssuschemeng.9b04647 ACS Sustainable Chem. Eng. 2019, 7, 19631−19639