The Mechanism of Room-Temperature Ionic-Liquid-Based Electrochemical CO2 Reduction: A Review

Electrochemical CO2 conversion technology is becoming indispensable in the development of a sustainable carbon-based economy. While various types of electrocatalytic systems have been designed, those based on room-temperature ionic liquids (RTILs) have attracted considerable attention because of their high efficiencies and selectivities. Furthermore, it should be possible to develop more advanced electrocatalytic systems for commercial use because target-specific characteristics can be fine-tuned using various combinations of RTIL ions. To achieve this goal, we require a systematic understanding of the role of the RTIL components in electrocatalytic systems, however, their role has not yet been clarified by experiment or theory. Thus, the purpose of this short review is to summarize recent experimental and theoretical mechanistic studies to provide insight into and to develop guidelines for the successful development of new CO2 conversion systems. The results discussed here can be summarized as follows. Complex physical and chemical interactions between the RTIL components and the reaction intermediates, in particular at the electrode surface, are critical for determining the activity and selectivity of the electrocatalytic system, although no single factor dominates. Therefore, more fundamental research is required to understand the physical, chemical, and thermodynamic characteristics of complex RTIL-based electrocatalytic systems.

Generally, an electrochemical CO 2 conversion system is composed of a cathode part where the CO 2 molecule is directly reduced to other chemicals (e.g., CO, CH 3 OH and C 2 H 4 etc.) on the catalyst surface (e.g., Ag, Au, Cu, and Zn etc.), and an anode part where counter oxidation reaction takes place (e.g., water oxidation in aqueous system). These two cathode and anode parts are present together in the same reaction medium or separated by a proton exchange membrane to take an electron transfer at the electrode surface. Braunschweig et al. also suggested the complex formation of [Emim] and CO 2 at the interface [52]. Deng et al. investigated the adsorption behavior of the cations and anions of 1-butyl-3methylimidazolium methylsulfate ([Bmim][MS]) in aqueous solution at the air-liquid surface, and the relationship between the surface structure and surface tension in terms of the RTIL concentration in solution [53]. Although this experiment was performed with an air-liquid interface instead of an electrode-solvent interface, the results provide important insights into the molecular structures and their correlation with the physical properties. Using the SFG spectra of aqueous [Bmim][MS] measured at different mole fractions (Figure 2), the authors found that the amphiphilic imidazolium cation is surface-active and behaves like a surfactant even at very low mole fractions. As the mole fraction of the RTIL increased to 0.1, the surface became saturated with [Bmim], and [MS] began to adsorb to the [Bmim] surface layer. Since the surface tension rapidly decreased at low concentrations, the authors suggested that the cations affect the surface tension to a greater extent than the anions.
To explore the potential-dependent structural changes in the interfacial double layer region during the electrochemical reduction of CO 2 , Rey et al. employed a non-resonant (NR) SFG method to study the [Emim][BF 4 ] system with 0.3 mol % water on an Ag electrode surface [54]. NR-SFG is similar to NR second-harmonic generation (SHG) measurement which is frequently used in the study of electrified interface system. The authors observed that the NR intensity minima appeared at a similar potential for both the Ar and CO 2 saturated cases, thereby indicating that the NR intensity changes are closely related to the intrinsic properties of the RTIL itself. As shown in Figure 3, the curvature of the potential-dependent NR intensity was different from the threshold potential for CO 2 reduction, −1.33V vs. Ag/AgCl in both forward and reverse scans. Therefore, they concluded that the intrinsic structural transition in the RTIL system controls the potential of CO 2 reduction, and this NR intensity behavior could be a descriptor for the design of new RTIL-based electrochemical systems.  [MS]) in aqueous solution at the air-liquid surface, and the relationship between the surface structure and surface tension in terms of the RTIL concentration in solution [53]. Although this experiment was performed with an air-liquid interface instead of an electrode-solvent interface, the results provide important insights into the molecular structures and their correlation with the physical properties. Using the SFG spectra of aqueous [Bmim][MS] measured at different mole fractions (Figure 2), the authors found that the amphiphilic imidazolium cation is surface-active and behaves like a surfactant even at very low mole fractions. As the mole fraction of the RTIL increased to 0.1, the surface became saturated with [Bmim], and [MS] began to adsorb to the [Bmim] surface layer. Since the surface tension rapidly decreased at low concentrations, the authors suggested that the cations affect the surface tension to a greater extent than the anions.
To explore the potential-dependent structural changes in the interfacial double layer region during the electrochemical reduction of CO2, Rey et al. employed a non-resonant (NR) SFG method to study the [Emim][BF4] system with 0.3 mol % water on an Ag electrode surface [54]. NR-SFG is similar to NR second-harmonic generation (SHG) measurement which is frequently used in the study of electrified interface system. The authors observed that the NR intensity minima appeared at a similar potential for both the Ar and CO2 saturated cases, thereby indicating that the NR intensity changes are closely related to the intrinsic properties of the RTIL itself. As shown in Figure 3, the curvature of the potential-dependent NR intensity was different from the threshold potential for CO2 reduction, −1.33V vs. Ag/AgCl in both forward and reverse scans. Therefore, they concluded that the intrinsic structural transition in the RTIL system controls the potential of CO2 reduction, and this NR intensity behavior could be a descriptor for the design of new RTIL-based electrochemical systems.   The circles denote the threshold potential for CO2 reduction and the NR intensity minimum. There was a close association between the onset of CO2 reduction and the nonresonant intensity minimum. (a) First forward (reduction) scan; (b) First reverse (oxidation) scan. Reprinted with permission from ref. [54]. Copyright 2015 American Chemical Society.
Osawa's group has reported various studies using surface-enhanced infrared absorption spectroscopy (SEIRAS) for probing the RTIL/electrode interface characters. The most representative cases are potential dependent in-situ SEIRAS measurements of the interface between the Au electrode and pure 1-butyl-3-methylimidazolium bis(trifluoromethanesulfonyl)amide ( [55,56]. In their measurements, the authors were able to observe the potential-dependent restructuring of the solvent components, such as the cations and anions of [Bmim][TFSA] or the water molecules at the Au electrode surface by direct integration of specific   The circles denote the threshold potential for CO2 reduction and the NR intensity minimum. There was a close association between the onset of CO2 reduction and the nonresonant intensity minimum. (a) First forward (reduction) scan; (b) First reverse (oxidation) scan. Reprinted with permission from ref. [54]. Copyright 2015 American Chemical Society.
Osawa's group has reported various studies using surface-enhanced infrared absorption spectroscopy (SEIRAS) for probing the RTIL/electrode interface characters. The most representative cases are potential dependent in-situ SEIRAS measurements of the interface between the Au electrode and pure 1-butyl-3-methylimidazolium bis(trifluoromethanesulfonyl)amide ( [55,56]. In their measurements, the authors were able to observe the potential-dependent restructuring of the solvent components, such as the cations and anions of [Bmim][TFSA] or the water molecules at the Au electrode surface by direct integration of specific . The circles denote the threshold potential for CO 2 reduction and the NR intensity minimum. There was a close association between the onset of CO 2 reduction and the nonresonant intensity minimum. (a) First forward (reduction) scan; (b) First reverse (oxidation) scan. Reprinted with permission from ref. [54]. Copyright 2015 American Chemical Society.
Osawa's group has reported various studies using surface-enhanced infrared absorption spectroscopy (SEIRAS) for probing the RTIL/electrode interface characters. The most representative cases are potential dependent in-situ SEIRAS measurements of the interface between the Au electrode and pure 1-butyl-3-methylimidazolium bis(trifluoromethanesulfonyl)amide ( [55,56]. In their measurements, the authors were able to observe the potential-dependent restructuring of the solvent components, such as the cations and anions of [Bmim][TFSA] or the water molecules at the Au electrode surface by direct integration of specific SEIRAS peaks during the cyclic voltammetry measurements. They observed the intrinsic behavior of the [Bmim][TFSA] components at the Au electrode surface in terms of the applied potential and found that the first interfacial layer is dominated by the cations at a negatively charged surface and anions at a positively charged surface. Thus, the mixing and exchanging of ions in the surface layers that occur during the potential scans are accompanied by a certain degree of hysteresis arising from the strong association between ionic components. More interesting behavior has been observed in the [Bmim][TFSA]/water mixture system with a water concentration (c w ) of 700 ppm. As shown in Figure 4, the band intensity profiles of v(OH) from water and v(CF 3 ) from the [TFSA] plotted against the potential cycle are very similar, which strongly suggests the association of water molecules with [TFSA] at the electrode surface. In addition, the authors suggested that the large intensity changes (hysteresis) in humid conditions compared to the dry state indicate the accelerated cation-anion exchange rate induced by water. [TFSA] components at the Au electrode surface in terms of the applied potential and found that the first interfacial layer is dominated by the cations at a negatively charged surface and anions at a positively charged surface. Thus, the mixing and exchanging of ions in the surface layers that occur during the potential scans are accompanied by a certain degree of hysteresis arising from the strong association between ionic components. More interesting behavior has been observed in the [Bmim][TFSA]/water mixture system with a water concentration (cw) of 700 ppm. As shown in Figure 4, the band intensity profiles of v(OH) from water and v(CF3) from the [TFSA] plotted against the potential cycle are very similar, which strongly suggests the association of water molecules with [TFSA] at the electrode surface. In addition, the authors suggested that the large intensity changes (hysteresis) in humid conditions compared to the dry state indicate the accelerated cation-anion exchange rate induced by water. Another useful technique to identify the characteristics of solid-liquid interfaces, especially the quantitative resistance factors in solid-liquid interface system, is electrochemical impedance spectroscopy (EIS) [57,58]. Yang [59]. From the Nyquist plots (Figure 5a,b) and the corresponding fitted equivalent circuit parameters, the authors suggested that the RTIL components are strongly absorbed on the electrode surface, forming a film layer; consequently, RTIL solutions show lower solution and charge transfer resistances. Thus, they concluded that the activation energy and overpotential of CO2 reduction are reduced by the interaction of CO2 and the RTIL layer formed at the electrode surface. Another useful technique to identify the characteristics of solid-liquid interfaces, especially the quantitative resistance factors in solid-liquid interface system, is electrochemical impedance spectroscopy (EIS) [57,58]. Yang [59]. From the Nyquist plots (Figure 5a,b) and the corresponding fitted equivalent circuit parameters, the authors suggested that the RTIL components are strongly absorbed on the electrode surface, forming a film layer; consequently, RTIL solutions show lower solution and charge transfer resistances. Thus, they concluded that the activation energy and overpotential of CO 2 reduction are reduced by the interaction of CO 2 and the RTIL layer formed at the electrode surface. As summarized above, researchers have carried out mechanistic investigations by experimentally observing the structural changes of the double layer formed at the electrode-liquid interface, and they tried to relate the experimental observations with the electrochemical CO2 reduction performance. They commonly observed strong adsorption of cations of RTILs onto the catalyst surface during electrochemical reaction even in low concentration of RTILs, and the detailed structural behaviors of RTIL components showed strong dependency on applied electric potential. It is obvious that the performance of electrochemical reaction is altered by RTIL components in the vicinity of the electrode surface. However, these spectroscopic observations alone may not be enough to elucidate the detailed mechanism due to the complicatedly entangled structural and electrostatic properties of double layer region, which are still barely known.
Beyond such experimental efforts, theoretical studies have also been extensively carried out to elucidate the underlying mechanism of electrochemical CO2 reduction in RTIL-based solvents [60][61][62][63][64][65][66][67][68]. In particular, Norskov's group has focused on the effect of the electric field generated by cations in the vicinity of the electrode surface. Urushihara et al. examined first-principles-based thermodynamic stability (Pourbaix diagram) of [Emim] at the Ag(111) water interface as a function of electrochemical potential and [Emim] concentration. In their simulations, the solvation effect was described using the combination of explicit water molecules, which were modeled adsorbed on the electrode surface, and implicit water, modeled as a dielectric medium [66].
As shown in Figure 6, they found that the coverage of [Emim] on the Ag(111) surface increased as a more negative bias potential was applied and as the concentration of RTIL in aqueous solution increases. Based on the theoretical Pourbaix diagram, they found that the coverage of [Emim] may be around one-ninth in the experimental condition of Rosen et al.'s work [33]. Thus, they suggested that the strong electric field on the electrode surface could have a positive effect by stabilizing the highly polarizable reaction intermediates, such as CO2 − or −COOH. As summarized above, researchers have carried out mechanistic investigations by experimentally observing the structural changes of the double layer formed at the electrode-liquid interface, and they tried to relate the experimental observations with the electrochemical CO 2 reduction performance. They commonly observed strong adsorption of cations of RTILs onto the catalyst surface during electrochemical reaction even in low concentration of RTILs, and the detailed structural behaviors of RTIL components showed strong dependency on applied electric potential. It is obvious that the performance of electrochemical reaction is altered by RTIL components in the vicinity of the electrode surface. However, these spectroscopic observations alone may not be enough to elucidate the detailed mechanism due to the complicatedly entangled structural and electrostatic properties of double layer region, which are still barely known.
Beyond such experimental efforts, theoretical studies have also been extensively carried out to elucidate the underlying mechanism of electrochemical CO 2 reduction in RTIL-based solvents [60][61][62][63][64][65][66][67][68]. In particular, Norskov's group has focused on the effect of the electric field generated by cations in the vicinity of the electrode surface. Urushihara et al. examined first-principles-based thermodynamic stability (Pourbaix diagram) of [Emim] at the Ag(111) water interface as a function of electrochemical potential and [Emim] concentration. In their simulations, the solvation effect was described using the combination of explicit water molecules, which were modeled adsorbed on the electrode surface, and implicit water, modeled as a dielectric medium [66].
As shown in Figure 6, they found that the coverage of [Emim] on the Ag(111) surface increased as a more negative bias potential was applied and as the concentration of RTIL in aqueous solution increases. Based on the theoretical Pourbaix diagram, they found that the coverage of [Emim] may be around one-ninth in the experimental condition of Rosen et al.'s work [33]. Thus, they suggested that the strong electric field on the electrode surface could have a positive effect by stabilizing the highly polarizable reaction intermediates, such as CO 2 − or −COOH. Moreover, Chen et al. evaluated the theoretical reaction energetics of CO2-to-CO on Ag(111) and Pt(111) electrodes in the presence of the electric field induced by the adsorbed cations [60]. The authors confirmed that the changes in the adsorption energies of the reaction intermediates that arise from the presence of explicit solvent and cations (K + or [Emim]) and can be effectively described by applying an external homogeneous electric field to the simulation cell. Therefore, they developed a micro-kinetic model that includes an external electric field to describe the reaction thermodynamics in the RTIL solvent system. Based on the reaction thermodynamics including the effect of the external electric field, the authors found that the CO2 adsorption step and the first reduction step forming *COOH (* denotes the species adsorbed on the catalyst surface) become easier, which explains the experimental observations of Rosen et al. (Figure 7) [33]. Despite some minor discrepancies, there is consensus in previous experimental and theoretical studies regarding the character of the interfacial region and its effect on the reaction energetics: the cations at the electrode surface are the main driving force increasing the activity of RTIL-based electrocatalytic systems. This is also consistent with the recent experimental studies reporting non-RTIL-based cationic field-induced activity enhancements in CO2 electrochemical reduction [65,69].  [60]. The authors confirmed that the changes in the adsorption energies of the reaction intermediates that arise from the presence of explicit solvent and cations (K + or [Emim]) and can be effectively described by applying an external homogeneous electric field to the simulation cell. Therefore, they developed a micro-kinetic model that includes an external electric field to describe the reaction thermodynamics in the RTIL solvent system. Based on the reaction thermodynamics including the effect of the external electric field, the authors found that the CO 2 adsorption step and the first reduction step forming *COOH (* denotes the species adsorbed on the catalyst surface) become easier, which explains the experimental observations of Rosen et al. (Figure 7) [33]. Moreover, Chen et al. evaluated the theoretical reaction energetics of CO2-to-CO on Ag(111) and Pt(111) electrodes in the presence of the electric field induced by the adsorbed cations [60]. The authors confirmed that the changes in the adsorption energies of the reaction intermediates that arise from the presence of explicit solvent and cations (K + or [Emim]) and can be effectively described by applying an external homogeneous electric field to the simulation cell. Therefore, they developed a micro-kinetic model that includes an external electric field to describe the reaction thermodynamics in the RTIL solvent system. Based on the reaction thermodynamics including the effect of the external electric field, the authors found that the CO2 adsorption step and the first reduction step forming *COOH (* denotes the species adsorbed on the catalyst surface) become easier, which explains the experimental observations of Rosen et al. (Figure 7) [33]. Despite some minor discrepancies, there is consensus in previous experimental and theoretical studies regarding the character of the interfacial region and its effect on the reaction energetics: the cations at the electrode surface are the main driving force increasing the activity of RTIL-based electrocatalytic systems. This is also consistent with the recent experimental studies reporting non-RTIL-based cationic field-induced activity enhancements in CO2 electrochemical reduction [65,69]. . Free-energy diagram for CO 2 (g) to CO(g) without (a) and with (b) the effects of electric field on Ag(111) at the equilibrium potential of the overall reaction (−0.11 V vs. SHE). The same pathway on Pt(111) is shown to contrast CO 2 and CO adsorption energies for the two metals in the absence of an applied electric field. Reprinted with permission from ref. [60]. Copyright 2016 American Chemical Society.
Despite some minor discrepancies, there is consensus in previous experimental and theoretical studies regarding the character of the interfacial region and its effect on the reaction energetics: the cations at the electrode surface are the main driving force increasing the activity of RTIL-based electrocatalytic systems. This is also consistent with the recent experimental studies reporting non-RTIL-based cationic field-induced activity enhancements in CO 2 electrochemical reduction [65,69]. However, a more complicated mechanism for determining the overall activity in RTIL-based systems is required. For example, no direct correlation is found among the chemical details of RTIL components and activity enhancement, indicating that the complex interactions between the RTIL components and the reaction intermediate species must be addressed. In other words, it is necessary to understand the complex physico-chemical nature of RTIL-based electrochemical systems, and thereby we can develop a general design principle for more enhanced RTIL-based electrochemical systems. The simulation method used in the previous theoretical studies is the first-principle density functional theory (DFT) which is known to predict thermodynamic properties of system with a relatively reliable accuracy level, in principle. However, considering the large size of the molecule and huge dynamical degrees of freedom of the RTIL-based solvent system, the computational burden should be concerned when the DFT method is applied. We conceive that the development/application of multiscale simulation methods (e.g., combined quantum mechanical/molecular mechanical (QM/MM) simulation) that can accurately describe the reaction energetics as well as the dynamical and structural characteristics of a complex catalytic system (with affordable computation burden) would be of importance in computational chemistry research field.

The Roles of RTILs in Chemical Interactions
Because experiments are based on ensemble measurements, the collective behavior of the reaction medium is often discussed, as summarized above. However, rather contradictorily, researchers have also tried to find mechanistic clues concerning the activity enhancement in the local intermolecular interactions between reaction intermediates and RTIL components (usually through the acidic hydrogen at the C2 position in imidazolium-based cations [70,71]). Thus, many experimental and theoretical studies have been performed to verify the effect of direct chemical interactions between the reaction intermediates and the RTIL components and to identify the correlation between reactivity and the chemical composition of the RTIL. Using DFT calculations, Wang 4 ]-CO 2 , and they evaluated the energies of every possible reaction intermediate, resulting in a mechanism for the imidazolium-based catalytic cycle (Figure 8). They argued that there exists a stable interaction between CO 2 and the RTIL components at the molecular level. However, they found that there was a large discrepancy between the estimated onset potential obtained via DFT calculations, and this was due to the absence of electrode surface effects.
In contrast, many experiments have been carried out to compare the CO 2 reduction activities of various combination of cations and anions of RTILs to decipher the underlying mechanism and to screen and optimize the various combinations. Tanner (Figure 9). The chain length of the imidazolium cation had a less systematic effect, and the pyrrolidinium cation, which has no acidic protons, showed higher activity than the imidazolium cations [Emim] [73,74]). Thus, the authors suggested that the RTIL components play a complex role in the CO 2 reduction mechanism, presumably altering the character of the double-layer at the electrode surface, which contrasts with the prevailing simplified view focusing on the local chemical interaction between CO 2 and RTIL components or CO 2 solubility.   Zhao et al. reported a more extensive screening study of 13 ionic liquids or salts, including imidazolium, pyrrolidinium, ammonium, phosphonium, borate, pyridinium, and sulfonium, and 7 ammonium salts with an Ag electrode [75]. They performed electrochemical measurements in the presence of a low concentration of the above compounds (2.0 mM) dissolved in an aprotic (acetonitrile)   Zhao et al. reported a more extensive screening study of 13 ionic liquids or salts, including imidazolium, pyrrolidinium, ammonium, phosphonium, borate, pyridinium, and sulfonium, and 7 ammonium salts with an Ag electrode [75]. They performed electrochemical measurements in the presence of a low concentration of the above compounds (2.0 mM) dissolved in an aprotic (acetonitrile) Figure 9. CVs of the reduction of carbon dioxide on silver with a scan rate of Zhao et al. reported a more extensive screening study of 13 ionic liquids or salts, including imidazolium, pyrrolidinium, ammonium, phosphonium, borate, pyridinium, and sulfonium, and 7 ammonium salts with an Ag electrode [75]. They performed electrochemical measurements in the presence of a low concentration of the above compounds (2.0 mM) dissolved in an aprotic (acetonitrile) solution of 0.1 M tetrabutylammonium hexafluorophosphate ([Bu 4 N][PF 6 ]) to minimize background effects from pure solvents. They observed that most compounds enhanced the catalytic CO 2 reduction kinetics, but imidazolium and pyrrolidinium cations were the most active. However, pyrrolidinium-based RTILs showed better performance than those containing imidazolium in terms of long-term stability, which is ascribed to the different role of cations; pyrrolidinium is believed to change the character of the interfacial double layer, while imidazolium is believed to act as a co-catalyst.
Lau et al. performed another interesting screening experiment, and they concluded that the activity enhancement of imidazolium-based RTILs originates from the C4 and C5 hydrogen atoms rather than that at the C2 position (Figure 10a) [76]. They systematically synthesized various functionalized imidazolium-based RTILs, as shown in Figure 10b, and observed that the methylation of the C2 hydrogen did not affect the catalytic activity, while the methylation of C4 and C5 hydrogen significantly decreased the activity (Figure 10b,c). Thus, they suggested that the coordination of the reaction intermediates and the C4 and C5 hydrogen results in the stabilization of the intermediates (Figure 10d). They also screened various imidazolium-based RTILs functionalized at the C2 hydrogen and found that RTIL 4a in Figure 10b showed the best activity and stability. Interestingly, they showed a dramatic loss of activity after methylation of the C4 and C5 positions of RTIL 4a (8a in Figure 10b), which is consistent with their conclusion; that is, the formation of interactions via the C4 and C5 hydrogen atoms stabilizes the intermediates, enhancing the activity.  Bu4N][PF6]) to minimize background effects from pure solvents. They observed that most compounds enhanced the catalytic CO2 reduction kinetics, but imidazolium and pyrrolidinium cations were the most active. However, pyrrolidiniumbased RTILs showed better performance than those containing imidazolium in terms of long-term stability, which is ascribed to the different role of cations; pyrrolidinium is believed to change the character of the interfacial double layer, while imidazolium is believed to act as a co-catalyst. Lau et al. performed another interesting screening experiment, and they concluded that the activity enhancement of imidazolium-based RTILs originates from the C4 and C5 hydrogen atoms rather than that at the C2 position (Figure 10a) [76]. They systematically synthesized various functionalized imidazolium-based RTILs, as shown in Figure 10b, and observed that the methylation of the C2 hydrogen did not affect the catalytic activity, while the methylation of C4 and C5 hydrogen significantly decreased the activity (Figure 10b,c). Thus, they suggested that the coordination of the reaction intermediates and the C4 and C5 hydrogen results in the stabilization of the intermediates (Figure 10d). They also screened various imidazolium-based RTILs functionalized at the C2 hydrogen and found that RTIL 4a in Figure 10b showed the best activity and stability. Interestingly, they showed a dramatic loss of activity after methylation of the C4 and C5 positions of RTIL 4a (8a in Figure 10b), which is consistent with their conclusion; that is, the formation of interactions via the C4 and C5 hydrogen atoms stabilizes the intermediates, enhancing the activity. From the above experimental screening studies, it can be found that the performance of RTIL based CO 2 reduction system is determined by not only the individual role of each component but also the complex reaction environment around the active site. Beyond the local interactions between the reaction intermediate species and the RTIL components, many studies have focused on the thermodynamic properties of the RTILs themselves to provide an overall understanding of RTIL-containing reaction systems [62,63,68]. Marjolin et al. reported that it is possible to estimate the reactivity of electrochemical CO 2 reduction with aromatic N-heterocycles (ANH) by evaluating the Pourbaix diagram of ANHs in terms of the pH and electrode potential [62]. The authors showed that the triple point of a DFT-based molecular Pourbaix diagram successfully reproduced the experimental CO 2 reduction conditions ( Figure 11). Thus, they proposed that the theoretical Pourbaix diagram triple point could be a unifying descriptor to estimate which molecules would be good proton or hydride shuttling agents. reactivity of electrochemical CO2 reduction with aromatic N-heterocycles (ANH) by evaluating the Pourbaix diagram of ANHs in terms of the pH and electrode potential [62]. The authors showed that the triple point of a DFT-based molecular Pourbaix diagram successfully reproduced the experimental CO2 reduction conditions ( Figure 11). Thus, they proposed that the theoretical Pourbaix diagram triple point could be a unifying descriptor to estimate which molecules would be good proton or hydride shuttling agents. Figure 11. Molecular Pourbaix diagrams for (a) pyridinium, (b) imidazolium, (c) 1-ethyl-3-methyl imidazolium and (d) phenanthroline species. The diagrams depict electrochemical conditions (variable pH and E), where nonprotonated (white region), monoprotonated (cyan region), 2H + + 2e − reduced (purple region), and 3H + + 2e − reduced (yellow region) are the thermodynamically most stable states in water solvent. Overlaid in the diagrams are standard reduction potentials (SRPs) for 2H + + 2e − → H2 (H2, red dashed line), CO2 + 2H + + 2e − → CO + H2O (CO, blue dotted line), CO2 + 2H + + 2e − → HCO2H (HCO2H) and CO2 + H + + 2e − → HCO2 − (HCO2 − , gold dotted lines). At applied potentials, more negative than these lines, product species would be thermodynamically stable compared with reactant protons, electrons, and CO2. Reprinted with permission from ref. [62]. Copyright 2015 American Chemical Society.

Conclusions
So far, many experimental studies have reported the improved performance of RTIL-based electrocatalytic systems for CO2 conversion. It is believed that the unique physicochemical properties of the RTIL components enhance both the activity and selectivity. To develop a more advanced RTIL-based electrocatalytic system, we must find the underlying mechanisms that can be used to guide component design. Experimental studies have mainly focused on the observation of the structural characteristics (mainly physical) of the RTIL components at the electrode-solvent interface. The cationic species are primarily located at the electrode surface during electrolysis, interacting with the reaction intermediates or altering the character of the electric double layer (summarized in Figure 12). Along with these experimental efforts, theoretical works have mainly focused on quantitative thermodynamic analyses of the physicochemical interactions between the RTIL components and the reaction intermediates, primarily at the electrode-solvent interface or in the bulk region. In addition Figure 11. Molecular Pourbaix diagrams for (a) pyridinium, (b) imidazolium, (c) 1-ethyl-3-methyl imidazolium and (d) phenanthroline species. The diagrams depict electrochemical conditions (variable pH and E), where nonprotonated (white region), monoprotonated (cyan region), 2H + + 2e − reduced (purple region), and 3H + + 2e − reduced (yellow region) are the thermodynamically most stable states in water solvent. Overlaid in the diagrams are standard reduction potentials (SRPs) for 2H + + 2e − → H 2 (H 2 , red dashed line), CO 2 + 2H + + 2e − → CO + H 2 O (CO, blue dotted line), CO 2 + 2H + + 2e − → HCO 2 H (HCO 2 H) and CO 2 + H + + 2e − → HCO 2 − (HCO 2 − , gold dotted lines). At applied potentials, more negative than these lines, product species would be thermodynamically stable compared with reactant protons, electrons, and CO 2 . Reprinted with permission from ref. [62]. Copyright 2015 American Chemical Society.

Conclusions
So far, many experimental studies have reported the improved performance of RTIL-based electrocatalytic systems for CO 2 conversion. It is believed that the unique physicochemical properties of the RTIL components enhance both the activity and selectivity. To develop a more advanced RTIL-based electrocatalytic system, we must find the underlying mechanisms that can be used to guide component design. Experimental studies have mainly focused on the observation of the structural characteristics (mainly physical) of the RTIL components at the electrode-solvent interface. The cationic species are primarily located at the electrode surface during electrolysis, interacting with the reaction intermediates or altering the character of the electric double layer (summarized in Figure 12). Along with these experimental efforts, theoretical works have mainly focused on quantitative thermodynamic analyses of the physicochemical interactions between the RTIL components and the reaction intermediates, primarily at the electrode-solvent interface or in the bulk region. In addition to deciphering the underlying mechanisms, several screening studies have been conducted on various combinations of RTIL components to screen and optimize the combinations. Despite various in-depth systematic research efforts, unfortunately, we have found that there is still lack of general consensus about the actual role of RTIL components in electrochemical systems among various experimental reports, among theoretical investigations, and even between experiments and simulations. This limits researchers in deriving the key major factors determining the catalytic activity of the electrocatalytic system, the so-called descriptors. We speculate that this is due to the complicated entangled structural and electrostatic properties of RTIL at the double layer region. Therefore, we conclude that future studies will need to put research efforts on identifying not only the role of each ionic component of an RTIL separately, but also the cooperative and collective role of media in a physical and chemical manner at the electrode surface or in the region above the surface. We anticipate arriving at the RTIL design principle based on such fundamental efforts to characterize the physical, chemical, and thermodynamic properties of complex RTIL-based electrocatalytic systems in near future.

Conflicts of Interest:
The authors declare no conflict of interest. The founding sponsors had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, and in the decision to publish the results.

Abbreviations
The following abbreviations are used in this manuscript. Despite various in-depth systematic research efforts, unfortunately, we have found that there is still lack of general consensus about the actual role of RTIL components in electrochemical systems among various experimental reports, among theoretical investigations, and even between experiments and simulations. This limits researchers in deriving the key major factors determining the catalytic activity of the electrocatalytic system, the so-called descriptors. We speculate that this is due to the complicated entangled structural and electrostatic properties of RTIL at the double layer region. Therefore, we conclude that future studies will need to put research efforts on identifying not only the role of each ionic component of an RTIL separately, but also the cooperative and collective role of media in a physical and chemical manner at the electrode surface or in the region above the surface. We anticipate arriving at the RTIL design principle based on such fundamental efforts to characterize the physical, chemical, and thermodynamic properties of complex RTIL-based electrocatalytic systems in near future.

Conflicts of Interest:
The authors declare no conflict of interest. The founding sponsors had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, and in the decision to publish the results.