2022 Volume 90 Issue 10 Pages 101004
The supporting electrolyte is an essential component of electrochemical reactions. Although there have been many reports on the influence of the type of electrolyte and its concentration on reaction efficiency in electrosynthesis, very few reports have systematically discussed the reasons for such effect. In several reaction systems, we have found that the coordination of anions from the supporting electrolyte to cationic organic species generated in electrochemical oxidation dramatically changes the reaction efficiency. In this comprehensive paper, we review these case studies, generalize the findings learned from them, and provide guidelines for strategic electrolyte design.
To achieve carbon neutrality in human activities, the chemical industry is facing an urgent demand to switch from conventional thermal processes to electrical processes. The synthetic method based on electrical energy is known as electrosynthesis. It is attractive because of its higher energy conversion efficiency compared to a classical chemical process relying on thermal energy. There is a high expectation that sustainable production of chemicals would be achieved by combining electrosynthesis technology with the electricity derived from renewable energy sources.1–3
Electrosynthesis is not only a green synthetic process but is also unique in that it can quickly generate active reaction intermediates.4 Electrochemical electron transfer driven by electrical energy at the electrode surface enables the oxidation/reduction of closed-shell organic compounds and generates highly reactive intermediates such as radical cations and radical anions under mild conditions. Although chemical redox agents can also potentially generate similar intermediates, electrochemical reactions are advantageous because the process uses electrons as reagents and does not produce any waste associated with the reaction. It is also noteworthy that the oxidation and reduction occur at a spatiotemporally separated reaction field in an electrochemical system. Thus the intrinsic reactivity of cationic/anionic intermediates is preserved without accompanying back-electron transfer.
To further advance electrosynthesis technology for the next generation, it is essential to properly control the reactivity of the active intermediates generated by electrochemical electron transfer. In the general chemical reaction, the stability and reactivity of intermediates are governed by the reaction medium, such as solvent. In the case of electrochemical reactions, the electrolyte is a medium for the reaction, composed of a solvent and a supporting electrolyte. The general requirements for the electrolyte are as follows:
(1) Solvents need to solubilize both organic substrates and supporting electrolytes. To dissolve supporting electrolytes, an organic solvent with a certain dielectric constant must be used. Typically, polar solvents such as acetonitrile are used, but solvents with relatively low dielectric constants such as dichloromethane and tetrahydrofuran can also be used.
(2) Quaternary alkylammonium salts are often used for the cationic part due to their high solubility in polar organic solvents. There are a variety of options for anionic moiety. Typically, conjugate bases of strong acids such as TsO−, TfO−, BF4−, ClO4−, TFSI−, FSI−, PF6−, and B(C6F5)4− are used [TsO− = p-toluenesulfonate, TfO− = trifluoromethane sulfonate, TFSI− = bis(trifluoromethane)sulfonimide, FSI− = bis(fluorosulfonyl)imide]. These anions are chemically stable, inert to redox reactions, have a wide potential window, and readily available. Halides such as Br−, Cl−, and I− are sometimes used as the anionic part, but these anions are susceptible to oxidation. They can function as electron mediators, as well as the supporting electrolyte.
In this comprehensive paper, the author summarizes the effect of electrolytes on electrochemical reactions based on the case studies of radical cation Diels-Alder reaction and redox chemistry of tellurophene. It is also briefly summarized that the strategic design of electrolytes to solubilize alkali metal fluoride into an organic solvent enables safe and economical electrochemical fluorination. Based on these studies, the author hopes to give insight into the rational design of electrolytes for the desired electrochemical reaction.
Before moving to the case studies, the author would like to summarize the idea of donor numbers (DNs) briefly. DN is a quantitative measure of the Lewis basicity of a molecule and was proposed by Gutmann. In addition to solvents,5 the Lewis basicity of anions has been experimentally investigated.6 DNs for selected solvent and anions (measured for ionic liquid of 1-ethyl-3-methylimidazolium, emim) are listed in Tables 1 and 2.
| Solvent | DNs for solvent/kcal mol−1 |
|---|---|
| 1,2-dichloroethane | 0 |
| dichloromethane | 1 |
| acetonitrile | 14.1 |
| water | 18 |
| tetrahydrofurane | 20 |
| N,N-dimethylformamide | 26.6 |
aData from Ref. 5.
| Ionic liquidb | DNs for anions/kcal mol−1 |
|---|---|
| [emim][PF6] | −6.2 |
| [emim][TFSI] | 7.2 |
| [emim][BF4] | 7.3 |
| [emim][ClO4] | 7.6 |
| [emim][TfO] | 20.5 |
aData from Ref. 6. b[emim] = 1-ethyl-3-methylimidazorium.
As can be learned from the case studies shown in the following sections, it is necessary to design electrolytes with appropriate coordinating properties to accomplish desired electrochemical reactions. In other words, it is of paramount importance to understand the coordination ability of the electrolyte of interest. For example, if one designs an electrolyte composed of solvents with small donor numbers such as dichloromethane, dichloroethane, nitromethane, and fluorinated alcohol in combination with weakly-coordinating electrolytes (WCAs) such as PF6− or B(C6F5)4−, the electrolyte should be regarded as a “weakly-coordinating electrolyte”. On the other hand, if one wants to design a “coordinating electrolyte”, acetonitrile is the solvent of choice, and TfO− or TsO− could be used for the anion. The author would like to emphasize that if the solvent is coordinating, i.e., of high donor number, the electrolyte should be considered “coordinating” even when WCAs are used as an anion.
Radical cations are chemical species produced by one-electron oxidation of closed-shell, electrically neutral organic compounds. Because radical cations are highly reactive and short-lived, their reactivity is still less understood than those of organic cations, anions, radicals, and carbenes, commonly used as intermediates in organic chemistry. However, with the recent active application of photoredox catalysis7 and electrosynthesis8 in synthetic organic chemistry, radical cations generated by single-electron transfer are now regarded as an essential intermediate in modern organic chemistry.
Among the vast library of reactions related to radical cation intermediates, the radical cation Diels-Alder reaction is one of the characteristic reactions (Fig. 1a).9,10 Radical cation Diels-Alder reaction is a [4+2] cycloaddition reaction the same as the general Diels-Alder reaction. Nevertheless, it is unique that both diene and dienophile are electron-rich, i.e., an electron-mismatched combination. The Diels-Alder reaction does not usually proceed with such a combination of substrates. Still, in the radical cationic Diels-Alder reaction, single-electron oxidation of one of the substrates (generally the dienophile) produces a radical cation, which is electron-deficient, and the cycloaddition reaction proceeds. Since the net reaction is redox neutral, the resulting Diels-Alder product in the radical cation state oxidizes starting material to close the catalytic cycle. In other words, the reaction is hole catalytic, as proposed by Bauld and co-workers in the 1980s.11
Radical cation Diels-Alder reaction of trans-anethole and isoprene. (a) Mechanism of the reaction. (b) Methodology-dependent reaction outcome of the radical cation Diels-Alder reaction. Reproduced with permission of Springer Nature from Ref. 12.
We found an interesting “reaction method dependence” of the radical cationic Diels-Alder reaction. Radical cation Diels-Alder reaction of trans-anethole with isoprene was chosen as a model reaction, and the reaction was carried out by three different redox systems: photoredox catalysis, chemical oxidation, and electrochemical oxidation (Fig. 1b).12 The photoredox catalysis and chemical oxidation gave high yields of the Diels-Alder product, whereas the equivalent electrochemical oxidation gave dramatically lower yields. We were intrigued by this phenomenon and finally came to the idea that the supporting electrolyte may influence the reaction. Electrochemical reactions significantly differ from other reaction systems. Specifically, in the electrochemical reactions, the supporting electrolyte (in this case, Bu4NTfO) is added at a concentration of 0.1 M (mol dm−3). We thus hypothesized that the significant excess of TfO− added relative to the substrate would likely affect the reactivity of the intermediates.
To test this hypothesis, we intentionally added supporting electrolytes to Ru-catalyzed photoredox radical cation Diels-Alder reaction and investigated their effect on reactivity.12 When 0.1 M Bu4NX [X− = TsO−, TfO−, ClO4−, BF4−, B(C6F5)4−] salt was added to the photoredox system shown in Fig. 2, the yield of the target product was lower in all the systems than the yield when adding no salt. The decrease in yield was more pronounced for the more coordinating anion. These results indicate that the coexistence of donor anions decreases the efficiency of the radical cation Diels-Alder reaction. Furthermore, the addition of 1,1,1,3,3,3-hexafluoroisopropanol (HFIP), which has anion binding ability,13,14 to the reaction solution increased the reaction yield. HFIP is a solvent frequently used in one-electron oxidation reactions to stabilize radical cations, but the implication for such stabilization has rarely been discussed in detail. It is suggested that the interaction of anions with HFIP in this system may reduce the interaction between radical cations and anions and, as a consequence, kinetically improve the reactivity of radical cations.
Radical cation Diels-Alder reaction of trans-anethole and isoprene by Ru-based photocatalyst with or without Bu4NX salt [0.1 M, X = TsO−, TfO−, ClO4−, BF4−, PF6−, B(C6F5)4−] and 1,1,1,3,3,3-hexafluoroisopropanol (0.5 M, HFIP) as additives, respectively. n.d.: not detected. Reproduced with permission of Springer Nature from Ref. 12.
Immediately after our group, Yoon and co-workers reported a very similar experimental system.15 They investigated the effect of the counter ion of a photoredox catalyst on the radical cation Diels-Alder reaction and found that the yield was lower for coordinating anions and that the addition of an anion receptor increased the reaction efficiency. They found that the interaction between the counter anions and the Ru complexes at both ground and excited states significantly affects the reaction efficiency, as well as the interaction between anions and radical cation intermediate.
Structural optimization and electrostatic potential mapping using DFT calculations were performed to gain insights into how the anion coordination affects the trans-anethole radical cation.7 The optimized structure of trans-anethole bounded by the anion was obtained for anions with relatively high DNs, indicating that the cation and anion interact in the inner-sphere and the electrophilicity of trans-anethole radical cation is reduced (Fig. 3). On the other hand, no bonding interactions were observed for weaker anions such as BF4− and PF6−, and the anion remained in the outer sphere. This suggests that the binding interaction of the donor anion with the radical cation is one of the reasons for the decrease in reaction efficiency.
Computationally optimized geometric structures and electrostatic potential mappings of ion pairs of trans-anethole radical cation with various counter anions. Structural optimizations were performed with B3LYP/6-31+G(d,p) level basis set with density functional theory. Red and blue indicate negative and positive charge densities, respectively, with a common scale for all ion pairs so that the surface can be compared visually (Iso value = 0.004). Reproduced with permission of Springer Nature from Ref. 12.
We attempted to design an electrolyte for an efficient electrochemical radical cation Diels-Alder reaction based on these results.12 Scheme 1 shows the reaction using Bu4NTfO as the supporting electrolyte, suggesting that the strong interaction of TfO− with the radical cation reduced the reaction efficiency. Therefore, the yield was dramatically improved when the solvent was changed to HFIP, expecting the binding of TfO− via hydrogen bonding interaction. When using TFE, which is expected to have the same anion binding property as HFIP, the reaction also proceeded efficiently. Furthermore, the reaction proceeded even when dichloromethane was used as the solvent and HFIP was added as an additive.
Electrochemical radical cation Diels-Alder reaction of trans-anethole and isoprene with various solvents. HFIP: 1,1,1,3,3,3-hexafluoroisopropanol, TFE: 2,2,2-trifluoroethanol, CF: carbon felt. Reproduced with permission of Springer Nature from Ref. 12.
These results indicate that anion coordination management is vital in the radical cationic Diels-Alder reaction. This idea for the design of the electrolyte is successfully applied to other types of electrochemical reactions via radical cation intermediates.16–18
The former section demonstrates the importance of electrolyte design in conducting the electrochemical radical cation Diels-Alder reaction. Electrochemical systems essentially require ca. 0.1 M of electrolyte, thus a relatively high concentration of anions must be present in the reaction media. Therefore, careful electrolyte design is essential for the successful electrochemical radical cation Diels-Alder reaction. Interestingly, Chiba and co-workers have developed an electrolyte with high performance in hole catalytic cycloaddition reactions.19,20 This electrolyte consists of LiX (X = ClO4−, TSI−, TFSI−) salts dissolved in nitroalkane (nitromethane, nitroethane, nitropropane) at a high concentration (>1 M), and has achieved extremely high efficiency in hole catalytic reactions such as the electrochemical Diels-Alder reaction and [2+2] cyclization.21 Intriguingly, the reaction proceeds without loss of electrophilicity of the radical cation in this electrolyte, despite a high concentration of the supporting salt. In this context, we investigated the physicochemical properties of this LiX/nitroalkane electrolyte.
To visualize the influence of the types and concentrations of the supporting electrolyte on the reaction efficiency, the radical cation Diels-Alder reaction was performed using nitromethane solvents dissolving LiFSI, LiTFSI, and LiClO4 (Fig. 4).22 To highlight the difference in efficiency of hole catalysis depending on the concentration of Li salts in the electrolytes, the electrolysis was intentionally stopped after the passage of 0.03 F/mol of charge (equals to 0.03 electron oxidation per trans-anethole). In other words, the reaction was quenched before the complete consumption of the starting materials. We then calculated turnover frequency (TOF), defined as the amount of cyclized product formed per electrochemically generated radical cation species for each experiment. A plot of TOF versus the concentrations of Li salt in the electrolytes is shown in Fig. 4. The progress of the reaction was confirmed in all electrolytes, and the TOF increased at the higher concentration of supporting electrolytes. This suggests a unique reaction field is formed by dissolving these support electrolytes at high concentrations.
Electrochemical radical cation Diels-Alder reaction of trans-anethole and isoprene in various concentrations of LiX/CH3NO2 electrolyte. Reproduced with permission of John Wiley & Sons, Inc. from Ref. 22.
Electrolyte with a high concentration (>1 M) of Li salt has been actively studied as a super-concentrated electrolyte, especially for lithium-ion battery applications in recent years.23 Lithium salts in electrolytes can be classified into solvated solvent-separated ion pairs (SSIP), contacted ion pairs (CIP), in which the ion pairs associate one-to-one, and multiple aggregates (AGG) (Fig. 5a). SSIP is also called free species, whereas CIP and AGG can be classified as bound species. When lithium salt is dissolved at a high concentration, the activity of the solvent decreases; in this situation, free species decreases, and a portion for bound species is increasingly observed.
Investigation of the solution chemistry of ClO4− salt in CH3NO2. (a) Classifications of in-solution state of ion pairs. (b) Raman spectra for LiClO4 or Bu4NClO4 in CH3NO2 at concentrations of 0.1 M, 1.0 M, and 2.0 M. Reproduced with permission of John Wiley & Sons, Inc. from Ref. 22.
We analyzed the LiX/nitroalkane electrolyte using Raman spectroscopy.22 Figure 5b shows representative data for LiClO4/CH3NO2 electrolyte in the concentration range of 0.1–2.0 M. A broad absorption band at 942 cm−1 appeared with increasing concentration of LiClO4 (Fig. 5b). This suggests that ClO4− interacts with Li+ in the LiClO4/CH3NO2 electrolyte and forms bound species.24 As a control experiment, Raman spectroscopic measurements of Bu4NClO4/CH3NO2 were also performed, and no formation of bound species was observed. In LiFSI/CH3NO2 and LiTFSI/CH3NO2, the appearance of bound species was also confirmed by Raman spectroscopy. The formation of bound species is expected to reduce the donor nature of the anion because it becomes hydrodynamically larger. This may be a key factor for efficient hole catalysis. We concluded that this leads to a kinetic stabilization of the radical cation and a highly efficient reaction.
Aromatic compounds play a central role in organic electronic materials because of their redox responsiveness involving π-electrons. Five-membered ring aromatic compounds containing group 16 elements, known as chalcogenophenes, are electron-rich heteroaromatics and exhibit stable oxidation behavior. While thiophenes have been the main focus of material science, organic devices based on tellurophenes, five-membered ring aromatic compounds containing tellurium, have recently attracted much attention.25 Seferos and co-workers reported seminal work on the synthesis and redox properties of π-extended tellurophene and polymers.26,27 They demonstrated that 2,5-diphenyltellurophene (PT) undergoes oxidation by reacting Br2, ICl, and XeF2 to give novel tellurophene compounds possessing Te(IV) center, PT-X2 (Fig. 6a).28
Chemical and electrochemical redox chemistry of π-extended tellurophenes. Reproduced with permission of Springer Nature from Ref. 29.
Such precedents encouraged us to investigate the electrochemical redox behavior of tellurophene. We found that the selectivity of the oxidation at the tellurium center or the π-system is governed by the coordination of solvent and counter anions (Figs. 6b–6f).29 We first confirmed the electrochemical oxidation of PT in the presence of halide anions to give hypervalent compound PT-X2 (Fig. 6b). Interestingly, WCAs such as ClO4−, BF4−, and PF6− and the neutral donor CH3CN also coordinate with tellurium in the electrochemical oxidation process to give the hypervalent compound PT-L2, evidenced by the spectroelectrochemistry measurement, bulk electrolysis, and computation based on density functional theory (Fig. 6c). This result suggests the strong Lewis acidity of the tellurium center. On the other hand, B(C6F5)4−, the even less coordinating anion, did not coordinate to tellurium and gave radical cationic species (Fig. 6e). In other words, two patterns of oxidation reactions proceed in the electrochemical oxidation of PT: two-electron oxidation induced by ligand (anion/solvent) coordination and one-electron oxidation in which positive charges and radicals delocalize in the π-electron system.
Based on these results, we hypothesized that if the radical cation state could be thermodynamically stabilized by extending the π-system, the selectivity of one- and two-electron oxidation could be altered.29 To verify this, we investigated the electrochemical redox behavior of 2,5-bis(2-(9,9′-dimethylfluorenyl))tellurophene (FT), which has a more extended π-electron system compared to PT. When BF4− and B(C6F5)4− were used for the oxidation of FT, hypervalent compound FT-L2 and the radical cation FT•+ were selectively given, respectively, based on the spectroelectrochemistry analyses. On the other hand, as expected, the coexistence of the radical cation FT•+ and the two-electron oxidized form FT-L2 resulted when PF6−, which exhibits donor properties intermediate between BF4− and B(C6F5)4−, was used (Fig. 6f).
These results suggest that the coordination of the electrolyte-derived anion determines the selectivity of the oxidation pathway and provide important insight into the electrooxidation process of tellurophene.
Fluorine-containing molecules are widely used in our daily life, such as pharmaceuticals30 and agrochemicals31 owing to their various effects originating from the unique properties of fluorine atoms. To produce fluorine-containing organic materials sustainably, electrochemical fluorination has been developed.32,33 Electrochemical fluorination uses stable fluoride ions as the fluorine source, and the cation species generated by the electrochemical oxidation react with fluoride anion to give fluorinated species. Although it has an advantage in that it does not use stoichiometric fluorinating agents, most of the electrochemical fluorination research that has been developed so far uses HF-based ionic liquids, which raises safety and economic concerns. For practical electrochemical fluorination, it is essential to develop electrochemical fluorination reactions using inexpensive and safe fluorine-source, such as alkali metal fluorides (MFs).34,35
Dissolving MF in organic solvent at specific concentrations can be achieved by controlling the coordination environment of MF by the solvent. Generally, MF is dissolved in an organic solvent using polyether such as crown ether, where ether linkage surrounds alkali metal cation by donating a lone pair. While these systems are promising to generate “naked” fluoride anion, such F− are highly basic and thus amenable to inducing side reactions other than nucleophilic attack. Therefore, a strategy to solubilize fluoride ions by hydrogen bonding donors such as alcohol has been actively developed in the field of fluorine chemistry.36–38
In this context, we were prompted to design an electrolyte that contains MF in high concentrations with a moderate anodic potential window.39 After examining various solvents, we found that the use of fluorinated alcohols was effective in solubilizing MF (Fig. 7a). CsF dissolved in the fluorinated alcohols (HFIP and TFE) is remarkably high concentrations, e.g., CsF/HFIP and CsF/TFE gave electrolyte concentrations up to 2 M and 3 M, respectively (Fig. 7b). KF was also applicable, and 1 M of KF was soluble in both HFIP and TFE (Fig. 7b). Viscosity and conductivity measurements were performed for all the electrolytes, giving insight into the dominating factor of the ionic conductivity depending on the concentration. Linear sweep voltammetry analyses revealed that the anodic potential window of all these electrolytes was slightly narrower than the original fluorinated alcohol, presumably due to the inevitable contamination by the adventitious water. Nevertheless, the electrolyte showed a moderate anodic potential window (up to around 2.0 V vs. SCE), allowing electrochemical oxidation of various organic compounds.
Electrolytes composed of alkali metal fluoride and fluorinated alcohol. (a) Hydrogen bonding interaction of metal fluoride (MF, M+ = K+, Cs+) and fluorinated alcohol. HFIP: 1,1,1,3,3,3-hexafluoroisopropanol, TFE: 2,2,2-trifluoroethanol. (b) Photograph of MF/fluorinated alcohol electrolyte. (c) Packing diagram for CsF/(HFIP)3 and CsF/(TFE)2 with anisotropic displacement ellipsoids shown at the 50 % probability level. The labeling of atoms is as follows: black, carbon; red, oxygen; green, fluorine; blue, nitrogen; purple, cesium; white, hydrogen. (d) Electrochemical fluorination of triphenyl methane with 0.3 M CsF dissolved in CH3CN/HFIP (8/2). Reprinted with permission from Ref. 39. Copyright 2021 American Chemical Society.
Single crystals obtained from CsF/HFIP and CsF/TFE electrolytes showed that both CsF solvated with fluorinated alcohols, giving a solid-state structure of CsF surrounded by fluorinated alcohol; Cs+ keeps a strong electrostatic interaction with F−, but multiple fluorinated alcohols hydrogen bond to F− (Fig. 7c).39 It is concluded that such flexible and multiple hydrogen-bonding interactions enabled the high solubility of MF.
We then applied this novel electrolyte for electrochemical fluorination of triphenylmethane.39 Unexpectedly, 0.3 M CsF/HFIP electrolyte did not give any fluorinated compound, presumably due to the defluorination of the resulting fluorotriphenylmethane via the hydrogen bonding interaction of HFIP. Instead, the desired reaction smoothly proceeded by using 0.3 M MF (M = Cs+, K+) dissolved in a mixed solvent of CH3CN/HFIP (8/2 in vol.) to give a fluorinated compound in quantitative yield (Fig. 7d). Although the scope of the reaction is still limited, this work provides some direction for designing electrolyte that enables oxidative fluorination using alkali metal fluoride salt as a safe and economical fluorine source. This electrolyte was also successfully applied to the electrochemical fluorination based on flow bipolar electrochemistry.40
In recent years, electrosynthesis has become one of the most topical fields in organic synthesis, and fascinating transformations are reported on a daily basis. Despite the growing interest, a rational design of the electrolyte for the desired reaction is still unclear, which is one of a barrier for beginners to starting electrosynthesis. In this paper, the author showcased how the electrolyte affects the electrochemical oxidation reactions and how to design electrolytes in a rational way. The author hopes this paper will be of some help for the researchers to start electrosynthesis.
The author would like to express his deepest gratitude to Prof. Shinsuke Inagi (Tokyo Institute of Technology), Prof. Kazuhiro Chiba (Tokyo University of Agriculture and Technology), and Prof. Masato Atobe (Yokohama National University) for their exceptional guidance and continuous encouragement. The author would also like to express his sincere gratitude to all the faculty members, students, and staff who have been involved in my research. This work is supported by a Kakenhi Grant-in-Aid (JP22H02118) from the Japan Society for the Promotion of Science (JSPS).
Naoki Shida: Funding acquisition (Lead), Resources (Lead), Supervision (Lead), Writing – original draft (Lead), Writing – review & editing (Lead)
The authors declare no conflict of interest in the manuscript.
Japan Society for the Promotion of Science: JP22H02118
This comprehensive paper is based on the author's achievements for the Award entitled “Analysis of coordination behavior of electrolytes and its application to strategic electrolyte design in organic electrolytic reactions”.
N. Shida: ECSJ Active Member
Naoki Shida (Assistant Professor, Graduate School of Science and Engineering, Yokohama National University)
Naoki Shida received his Ph.D. degree from Tokyo Institute of Technology. He then worked as a postdoctoral fellow with a fellowship from JSPS. In 2018, he started his academic career as a specially appointed assistant professor at Tokyo Institute of Technology in the group of Prof. Shinsuke Inagi. In 2020, he joined the group of Prof. Mahito Atobe at Yokohama National University as an assistant professor.
