Catalyst–electrolyte interface chemistry for electrochemical CO₂ reduction†

Young Jin Sa‡ ^ab, Chan Woo Lee‡ ^c, Si Young Lee‡ ^ad, Jonggeol Na ^e, Ung Lee *^adf and Yun Jeong Hwang *^adg
aClean Energy Research Center, Korea Institute of Science and Technology (KIST), Seoul 02792, Republic of Korea. E-mail: ulee@kist.re.kr; yjhwang@kist.re.kr
^bDepartment of Chemistry, Kwangwoon University, Seoul 01897, Republic of Korea
^cDepartment of Chemistry, Kookmin University, Seoul 02707, Republic of Korea
^dDivision of Energy and Environmental Technology, KIST School, Korea University of Science and Technology (UST), Seoul 02792, Republic of Korea
^eDivision of Chemical Engineering and Materials Science, Ewha Womans University, Seoul 03760, Republic of Korea
^fGreen School, Korea University, Seoul 02841, Republic of Korea
^gDepartment of Chemical and Biomolecular Engineering and Yonsei-KIST Convergence Research Institute, Yonsei University, Seoul 03722, Republic of Korea

First published on 11th August 2020

---

---

Young Jin Sa

Prof. Young Jin Sa received his PhD degree in Chemistry from Ulsan National Institute of Science and Technology (UNIST) in 2018. He participated in energy-related research about the development and characterization of active electrocatalytic nanomaterials for hydrogen fuel cells and electrolyzers. He then joined the clean energy research center at Korea Institute of Science and Technology (KIST) as a postdoctoral fellow and conducted CO2 electroreduction research. He started his independent career from September 2019 as an assistant professor of Kwangwoon University, Republic of Korea. His research interest is synthesis of nanomaterials and understanding of selective electrocatalytic processes.

Chan Woo Lee

Prof. Chan Woo Lee received his PhD in material science and engineering from Seoul National University. During his PhD, his research focus was on the reaction mechanism of electrochemical CO2 conversion based on electrokinetic and in situ analyses under the supervision of Prof. Ki Tae Nam. After his PhD years, he worked at the clean energy research center of the Korea Institute of Science and Technology (KIST) as a postdoctoral fellow and now is an assistant professor at the Department of Applied Chemistry in Kookmin University. His research interest is in the energy and environmental science based on electrocatalysis and nanomaterials.

Si Young Lee

Si Young Lee received his Bachelor's degree in Chemistry from Kyung Hee University in 2015. He moved to the clean energy research center at Korea Institute of Science and Technology (KIST) in 2015 and is currently competing research as a PhD candidate with Professor Yun Jeong Hwang. His research is mainly focused on analyzing the chemical variables of electrocatalysts for renewable energy sources and CO2 reduction.

Jonggeol Na

Prof. Jonggeol Na is now an assistant professor at Ewha Womans University. He performed a postdoctoral research fellow at Carnegie Mellon University, USA and Korea Institute of Science and Technology (KIST), South Korea. He received his BSc and PhD in Chemical and Biological Engineering at Seoul National University where his research focused on the computational science approach to design and optimization of sustainable process systems. Currently, his research is focused on autonomous discovery of non-intuitive process system designs through artificial intelligence and multiscale simulations for accelerating the conceptual design of non-traditional electrical energy-based processes that improve sustainability in the chemical industry.

Ung Lee

Prof. Ung Lee received his PhD in chemical and biological engineering from Seoul National University. During his PhD, his research focused on the CO2 capture and utilization process design and optimization based on waste heat recovery under the supervision of Prof. Chonghun Han. After his PhD years, he worked at AVT.SVT in Aachen University in Germany as a postdoctoral fellow with Prof. Alexander Mitsos. Currently, he is a senior research scientist in Korea Institute of Science and Technology and also the adjunct professor in Korea University. His research interests are in the reactor and process design/optimization for CO2 utilization systems.

Yun Jeong Hwang

Dr Yun Jeong Hwang received her Bachelor's and Master's degree in Chemistry from Korea Advanced Institute of Science and Technology. She then studied photoelectrochemical water splitting during her PhD at the University of California, Berkeley. In June 2012, she joined Korea Institute of Science and Technology (KIST) where she started her research on the electrocatalyst for CO2 reduction and water oxidation as a sustainable carbon cycle technology. She is currently a principal researcher of KIST and a recipient of KIST Young Fellowship, and currently serving as an Associate Editor of Journal of Materials Chemistry A (Royal Society of Chemistry).

---

1. Introduction

Combustion of fossil fuels releases significant amounts of greenhouse gases into the atmosphere, resulting in the continuous accumulation of CO₂, and consequently, an imbalance in the carbon cycle. The global atmospheric CO₂ concentration was 410 ppm in 2018, accelerating at a rate of 2.87 ppm per year, which is ∼100 times higher than the rate at the end of the last ice age.¹ In response to climate change-related issues, including global warming,² ocean acidification,³ and ecosystem destruction⁴ caused by greenhouse gas emissions, 196 states signed the Paris Agreement on climate change in 2016. This agreement states that the global average temperatures should be maintained well below 1.5 °C, above pre-industrial levels.⁵ The primary sources of carbon emissions are automobiles, factories, and electricity generation plants.⁶ To adequately mitigate carbon emissions and promote a sustainable carbon cycle, carbon capture and utilization (CCU) technologies have to be incorporated with renewable wind, solar or hydropower electricity sources, which also extend the usage of renewable energy resources. Specifically, the electrochemical CO₂ reduction enables the storage of the intermittent renewable electricity in chemical forms under ambient conditions, producing various raw chemicals for use in production processes and transportation fuels.^7–9

In recent decades, studies on the electrochemical CO₂ reduction at the fundamental level have been intensified to understand the reaction pathways and catalytic properties. Furthermore, research efforts are focused on industry-scale applications and parameter optimization for improving the economic feasibility of the CO₂ electroreduction technology. Therefore, the research scope on the electrochemical CO₂ reduction encompasses electrokinetic analyses, in situ spectroscopic analyses, labelling experiments, mechanistic studies, density functional theory (DFT) calculations, catalyst material design, electrode structure modification, product evaluations, and electrolytic device engineering and optimization. The initial investigations into the electrocatalytic activities of polycrystalline single metal electrodes, combined with theoretical studies on key intermediates and the binding affinity, contributed to the understanding of the electrochemical CO₂ conversion process to target products.^10–12 However, polycrystalline single metal electrodes exhibit low activity and/or product selectivity. Furthermore, the scaling relations of the binding energies of various intermediates make it difficult to control the catalytic activity selectively.¹³ These limitations incentivized the modulation of active sites, including open-packed facets and grain boundaries, along with edge and corner sites,^10,14–19via nanostructuring, as well as the development of new active sites through alloying and interfacial structuring.^20,21 In addition, electrokinetic, in situ, and isotopic analyses provided more direct experimental evidence for the reaction mechanisms of CO₂ reduction reaction (CO₂RR).^22–24 In addition, theoretical studies have been conducted to support molecular insights into the catalytic process.²⁵ Recent studies have focused more on the system components, including electrolytic cells, electrolytes, and gas-diffusion-layer (GDL) electrodes, to ensure high catalytic performance in terms of the current density and faradaic efficiency (FE) for practical implementation.^26–29 As the research on CO₂ electroreduction progresses, increasingly complex components are developed to increase the efficiency. Since each component creates an interface, which influences the overall activity, the catalyst–electrolyte interface should be adequately understood. An in-depth understanding of the catalyst–electrolyte interface, regarding the reaction intermediates, ionic distribution, arrangement of catalyst surface atoms, and transport behaviour, would facilitate the identification of the main factors influencing the CO₂RR activity. The importance of such an interfacial science is further rationalized by the inherent limitations of the catalyst such as unstable surface states, as well as the extrinsic fine control of the electrocatalytic process.

Herein, we review the current understanding of the catalyst–electrolyte interface and its impact on the CO₂RR performance, including FE, overpotential, and current density, as well as strategies for controlling the interface. First, based on previous technoeconomic analysis reports, the economic feasibility of the CO₂ electroreduction technology is evaluated. The technoeconomic results help to determine the key operational parameters that should be prioritized for practical implementation, such as the FE, operation potential, overpotential, and current density (i.e. production rate) depending on the CO₂ electroreduction condition. Subsequently, we introduce the fundamentals and challenges associated with the electrochemical CO₂RR. To address the issues, the kinetic equations of the CO₂RR are provided as a starting point. The mathematical expressions help to understand the factors that regulate the catalytic performance and identify the important challenges related to the catalyst–electrolyte interface. In the following chapter, beginning with the impact of the catalyst material on the activity/selectivity, we discuss, in detail, the strategies to control the catalyst–electrolyte interface via ligand chemistry, electrolyte engineering, mass transport kinetics, three-phase boundary, and gas-phase conversion. The review articles published in recent years on the electrochemical reduction of CO₂ primarily focus on the catalyst (Table S1, ESI†), while about 10% of the currently emerging review articles focus on the catalyst–electrolyte interface. We believe that this review, based on technoeconomic perspectives and catalyst-surface modulation chemistry, can provide further insights into the catalyst–electrolyte interface.

2. Technoeconomic perspectives of electrochemical CO₂RR

The electrochemical reduction of CO₂ can produce hydrocarbons and oxygenates, which can serve as essential commodity fuels and chemicals.³⁰ However, achieving an economical CO₂RR is challenging because the target products (e.g. CO, olefin, paraffin, alcohols, etc.) are usually produced on a massive scale in the petrochemical industry. To secure the economic feasibility of CO₂RR processes, several studies have presented performance guidelines related to the current density, FE, conversion, and catalyst durability.

Early studies focused on the relationship between the minimum energy requirements and the economic feasibility of the CO₂ reduction process. These studies provided a quick screening method, which involved major cost contributors (e.g. electricity cost). Lu and Jiao estimated the CO₂RR costs with minimum overpotentials to identify economically feasible products.³¹ Among the major CO₂ reduction products, CO, formic acid, formaldehyde, and propanol were proposed to be economically feasible. They assumed that electricity cost ($0.07 kW per h) accounted for a significant proportion. When other costs, including the capital and operational costs, and raw material prices are considered further, more rigorous results would be obtained. Meanwhile, Palmer et al. analyzed a CO₂RR using sunlight,³² and classified formic acid, acetaldehyde, and allyl alcohols as economically feasible CO₂RR products, while methane, ethylene, methanol, ethanol, propanol, and formaldehyde could not meet the lower boundary requirement for economic feasibility. To simplify the analysis, they assumed that electrochemical cell operates under an equilibrium voltage, determined by the Nernst equation, which provides positive results, but does not reflect the actual catalytic performance. When considering the overpotential of the CO₂RR, the boundary requirement for economic feasibility will become more stringent.

The economic feasibility analysis based on minimum energy requirements can serve as a quick method for identifying feasible CO₂RR products; however, more rigorous models considering electrochemical cell performance parameters (e.g. current density, overpotential, FE, and durability) were also developed to identify the minimum cell performance targets and realistic economic potentials. Verma et al. presented a CO₂RR performance target through a gross margin-based technoeconomic analysis.³³ Using their model, the effects of the current density, operating cell potential, FE, as well as the catalyst durability, can be evaluated. The gross margin model predicted that CO and formic acid are the only economically viable products when the CO₂ capture costs are considered. For CO production, the maximum operating cell potential responsible for the economic viability of the CO₂RR process decreases linearly as the FE of CO production decreases. Thus, syngas production can be another attractive option. The minimum current density of 465 mA cm⁻², which guarantees a zero gross margin, was determined with the cell potential of 1.6 V. Interestingly, they also presented a target catalyst durability relative to the current density. The minimum current density dramatically increases when the catalyst durability decreases below 2000 h. Bushuyev et al. determined the electricity costs, energy conversion, and FE targets for CO, formic acid, ethylene, propanol, methanol, ethanol, and ethylene glycol using a simplified technoeconomic analysis.³⁰ They claimed that the levelized costs of these compounds are lower than current market prices with achievable FEs and current densities. The cost evaluations for their analysis are based on the following: electricity cost, 2 cents per kW per h; electrolyzer cost, $500 per kW; 60% energy conversion efficiency; and 90% FE. Luna et al. carried out a similar analysis for CO, ethanol, and ethylene production and described the production costs in terms of the electrolyzer energy conversion efficiency and electricity.³⁴ For their analysis, they employed a 90% FE, a current density of 500 mA cm⁻², and an electrolyzer cost of $300 per kW. The sensitivity analysis in their study showed that the prices of all products compete with current market prices when the electricity cost falls below 4 cents per kW per h, with an energy efficiency of at least 60%.

More recently, the technoeconomic analysis of the electrochemical CO₂ reduction process was carried out on a more comprehensive level to account for not only the cell performance but also auxiliary operations, such as separation, recycling, and transportation processes. Jouny et al. presented a comprehensive technoeconomic study that considered the CO₂ capture process, product-specific separation processes, and the electrolyte recyclability, along with the electrochemical CO₂ reduction system.³⁵ They presented a parametric sensitivity analysis that reflected the worst, base, and best-case scenarios. Consistent with other studies, CO and formic acid are proposed as the only profitable products under the base-case condition (i.e. electricity cost, $0.03 kW per h; FE, 90%; current density, 300 mA cm⁻²; energy conversion, 40%; CO₂ capture cost, $40 per ton). Conversely, n-propanol might also be a promising product under the optimistic case scenario (i.e. electricity cost, $0.02 kW per h; FE, 100%; current density, 500 mA cm⁻²; energy conversion, 70%; negligible CO₂ capture cost). Chae et al. developed a monolithic device for CO₂RR by integrating photovoltaic and electrochemical cells, and they carried out a comprehensive technoeconomic analysis based on the device performance.³⁶ They concluded that the levelized cost of CO is about 10 times higher than the current market price due to the high electrolyzer cost. Thus, they recommended a relatively high current density and CO₂ conversion to secure the technoeconomic feasibility of the photo-electrochemical CO production process. Spurgeon and Kumar analysed CO₂ reaction pathways leading to liquid products, using a net present value-based technoeconomic analysis.³⁷ The CO₂ reaction pathway included the hybrid reaction of electrochemical CO₂ reduction to CO, the Fischer–Tropsch reaction, direct and cascade electrolysis for ethanol production, and electrochemical formic acid production. They reported that the costs of current state-of-the-art CO₂ reduction technologies cannot compete with present fuel production costs; however, formic acid production can be competitive with its commercial bulk production cost and the optimistic scenario of $0.03 kW per h for electricity, 96% FE, 200 mA cm⁻² current density, 75% conversion, and 3.2 V cell potential. Herron et al. proposed a generalized framework for accessing the energy cost of a solar driven fuel system.³⁸ The framework considered CO₂ capture and transportation options, product-specific separation systems, and the recyclability of unreacted raw materials. The methanol production process from CO₂ was presented as a case study, and the market price of methanol was achievable at a single-pass conversion of 38% or higher and 40% selectivity. Rumayor et al. presented a comprehensive technoeconomic CO₂RR evaluation for formic acid production.³⁹ They considered catalyst durabilities of 2.5 and 4.45 years in their process model and assumed a cell potential and FE of 4.3 V and 42.3%, respectively. The result of this study indicated that even the formic acid production process is not yet profitable under the current conditions, considering the formic acid commercial process as a benchmark. Na et al. carried out general technoeconomic analyses of 16 possible CO₂RR products and demonstrated that weak economic feasibility is achieved if the current density is less than 2 A cm⁻², when a complete process, including product separation and recycling, is considered.⁴⁰ They, however, demonstrated that electrochemical coproduction (i.e. CO₂RR at the cathode and organic oxidation at the anode) has a recognizable economic potential and may compete in the current market with available technologies.

Meanwhile, recent progress demonstrated electrochemical CO₂RR performances. Specifically, Arquer et al. achieved a high current density of 1.3 A cm⁻² for the electrochemical CO₂RR experimentally by introducing a catalyst–ionomer bulk heterojunction. This study brightens the prospect that the CO₂RR technology can economically replace fossil-fuel-based chemical production technologies and that the catalytic activity improvement, achieved by adjusting the interface, plays an important role in ensuring economic feasibility.⁴¹

The economic feasibility of CO₂ reduction systems is a controversial issue. However, a comprehensive technoeconomic analysis can provide insight into the catalytic requirements and research priorities. For a CO₂RR to be economically feasible, substantial improvement of the key performance parameters is required (e.g. FE, overpotential, and current density). The performance parameters can be classified into linear and nonlinear parameters. The overpotential is a linear parameter because it is linearly correlated to the electricity consumption and the corresponding operating cost. In contrast, an increase in the FE and current density corresponds to an exponentially decrease in the production costs. For example, the electrolyzer cost can be as high as 80% of the total equipment cost when the current density is lower than 100 mA cm⁻², and it rapidly decreases as the current density increases due to the inverse proportional relationship between the equipment cost and the current density.³⁶ FE also exhibits similar characteristics, although the current density may exert a significant influence because the electrolyzer is more expensive than conventional separation systems when both the FE and current density are low.⁴⁰ In general, the production cost is lowered when the key performance parameters are improved in the order of the current density, FE, and overpotential (i.e. when the current density and FE are low). In contrast, the overpotential can be the most influential parameter under a high current density and FE condition. Thus, catalytic activity research should focus primarily on improving the current density at the early stage (e.g. low FE and current density) and lowering the overpotential when optimum current density and FE are achieved.

As we will discuss later, the early stage of the CO₂RR catalyst studies has been mostly performed in H-cell type electrolyzers, where CO₂ gas is bubbled through the electrolyte, resulting in a low partial current density for CO₂RR due to the limitation of the CO₂ solubility and diffusion. Considering the limited partial current density for the CO₂RR, FE improvement has been a high research priority, and the intrinsic/extrinsic catalytic activities are controlled by modifying the catalyst and catalyst–electrolyte interface to suppress the competitive hydrogen evolution reaction (HER) occurring in the aqueous electrolyte. Meanwhile, the CO₂RR current density could be increased by more than a hundred times, and a current density of over 1 A cm⁻² can be achieved by applying GDL-based electrolyzers. As the GDL-based electrolyzer creates new types of interfaces as a high flux of CO₂ gas is directly fed to the catalyst surface, the electrolyzer design and non-catalytic operating parameters affect the transport behavior of CO₂ molecules and protons at the catalyst–electrolyte interface as well as the intrinsic activity of the catalytic active sites, and thus are deemed crucial for the FE and overpotential, particularly when C₂₊ chemicals are targeted as products of the CO₂RR. Therefore, engineering directions suitable for these new interfaces should be developed to improve the FE and overpotential. To understand the importance of the catalyst interface on the molecular level, the fundamentals of the CO₂RR and the effect of the interface engineering are discussed in the following sections.

3. Fundamentals of electrochemical CO₂ reduction

3.1. A schematic of the conventional H-type cell system

Electrochemical CO₂RRs produce C_xH_yO_z chemicals by multiple electron and proton transfers to CO₂, as expressed by eqn (1) below:

xCO2 + nH+ + ne− → CxHyOz + mH2O	(1)

In the H-type cell (Fig. 1a), gaseous CO₂ is continuously supplied to the catholyte to dissolve and saturate CO₂ in the aqueous electrolyte by the chemical equilibrium between the gas and liquid phases (eqn (2)).⁴² The concentration of the dissolved CO₂ (c_CO2) is given by Henry's law:⁴²

CO2(g) ⇌ CO2(aq)	(2)

The aqueous CO_2(aq), dissolved in the bulk electrolyte, can be transported to the cathode surface by convection and diffusion, where it finally undergoes proton–electron transfer. Normally, CO_2(aq) is considered a carbon source, while protons have different sources included in the electrolyte, such as bicarbonate, water, hydronium ions, and carbonic acid.^43,44 Through isotopic experiments, it was established that CO₂–H₂O–HCO₃⁻ complex molecules act as a carbon source.⁴⁵

CO₂ molecules have complex acid–base equilibria in aqueous electrolytes. When gaseous CO_2(g) is dissolved, most of the CO₂ remains as solvated molecular CO_2(aq), while only a small portion (∼1/1000) of dissolved CO₂ is hydrated to carbonic acid (H₂CO₃).⁴² These CO₂ species participate in acid dissociation reactions to form HCO₃⁻ and H⁺, as shown in eqn (3). Conventionally, the solvated molecular CO₂ and carbonic acid are expressed with a single term “CO_2(aq)” in the equation.²⁵ Furthermore, the dissociation of HCO₃⁻ and the self-ionization of water are involved in the chemical equilibria (eqn (4) and (5)). CO₂-saturated 0.1 M KHCO₃ and 0.5 M KHCO₃ are popular electrolytes for CO₂RRs in conventional H-type cells, and the bulk pH values under the CO₂ partial pressure of 1 atm are 6.8 and 7.2, respectively; the bulk concentration of CO_2(aq) is approximately 33 mM (Fig. 1b).⁴²

CO2(aq) + H2O ⇌ HCO3− + H+ pKa1 = 6.37 (25 °C)	(3)

HCO3− ⇌ CO32− + H+ pKa2 = 10.25 (25 °C)	(4)

H2OH+ + OH− pKw = 14 (25 °C)	(5)

Alkali metal cations (M⁺) such as Li⁺, Na⁺, K⁺, and Cs⁺ are used as supporting electrolyte cations. Various ionic species in the electrolyte, including M⁺, H⁺, HCO³⁻, OH⁻, and CO₃²⁻, facilitate ionic conduction. The membrane inserted between the catholyte and anolyte aids the mobility of ions, while inhibiting the crossover of gaseous products. Nafion membranes selectively transport cations and additionally inhibit the crossover of anionic liquid products, including formate and acetate. However, anion-exchange membranes are preferred to reduce the polarization losses induced by the membrane, because the exiting anions are major charge carriers.⁴² At the anode, the oxygen evolution reaction (OER) proceeds normally with water oxidation and functions as a counter half-reaction for CO₂RR to complete the electrochemical circuit.

3.2. Challenges of CO₂RRs

Electrochemical CO₂RRs can be categorized by the type of products formed, as listed in Table 1, such as HCOOH,^8,46–49 CO,^50,51 CH₄,^52–54 CH₃OH,⁵⁵ C₂H₄,^56–61 C₂H₅OH,^62–64 CH₃COOH,^65,66 C₃H₇OH,^67–70 C₃H₄O₂,⁷¹ C₄H₄O₃,⁷¹ and graphitic carbon.⁷² As the products are formed via the activation of stable CO₂ molecules and a series of electron–proton transfer steps, the thermodynamic equilibrium potential for full reaction equations does not reflect the energetic advantages of the CO₂RR. The energetics of each reaction step determine the actual onset potentials. Previously, various reaction pathways and intermediates were proposed based on experimental observations and theoretical calculations. These are still debating points due to the highly complex and multiple reaction steps. A comprehensive summary can be found in recent review articles.^26,73–75 Here, we briefly address a specific reaction pathway for each product to explain the key challenges associated with the CO₂RR (Table 1).^{55,66,70–73,76–81}

---

One of the challenges is that intermediates participating in a specific reaction pathway have similar binding affinity to the surface. For example, for CO production, *COOH and *CO are involved via the carbon atom. *COOH needs to be stabilized to decrease the reaction overpotential, while a low *CO binding energy is required for easy desorption from the catalyst surface. However, it is difficult to control the binding affinities individually on single metal electrodes because of their linear scaling relation. Similarly, in CH₄ production, the binding affinities of the intermediates are correlated,^76,77,82,83 which prevents selective electrocatalytic reduction.

Furthermore, various reaction pathways have common intermediates that can branch out into different intermediates or products. *CO is the shared key intermediate not only in CO production but also in the production of most hydrocarbons, including CH₄, C₂H₄, CH₃OH, and C₂H₅OH. In another example, the CH₄ and CH₃OH pathways have *CHO as a common intermediate but proceed with different intermediates (*CHOH or *OCH₂) after *CHO, according to the binding affinity of the respective catalysts.^55,84 The C₂H₄, C₂H₅OH, and CH₃COOH pathways have *CO–CO or its protonated species as common intermediates, although the final product can vary substantially depending on the subsequent steps.^66,76 Due to these features, obtaining a single product with a high FE (e.g. control of product selectivity) is very challenging especially when the products are targeted to C₂₊ chemicals.

More importantly, the undesirable HER can competitively occur because proton donors are also required for CO₂RR. As the binding affinity of adsorbed H (*H) is correlated with those of C-binding intermediates according to theoretical studies,⁷⁴ selectively suppressing the HER over CO₂RR becomes another major issue to be addressed, especially when the catalyst–electrolyte surface has an overwhelming approach of proton donors including water molecules.

Besides the product selectivity and overpotential issues, to achieve high CO₂RR production rates (i.e. partial current density), the reactant concentrations at the catalyst surface should be sufficiently high. However, in the conventional H-type cell, CO₂ molecules are dissolved and supplied in the liquid electrolyte, and the CO₂ concentration is depleted at the catalyst surface due to the low CO₂ solubility and inadequate mass transport from the bulk electrolyte, in contrast to the scenario where the high CO₂RR rate is achieved. A simple modelling study reported that the local CO₂ concentration (at the catalyst surface) is close to ∼0 mM at a current density of 25 mA cm⁻², when 0.1 M KHCO₃ aqueous electrolyte was used.⁸⁵

3.3. Kinetic equations and key factors for CO₂RRs

The thermodynamics, kinetics, and reaction mechanisms of the CO₂RR can address the major challenges associated with the CO₂RR, such as the product selectivity, overpotential, and current density. In this section, several equations on the rates of CO production and HER are introduced (Table 2). The reaction mechanisms are referred from previous reports, where the acid reaction mechanism is assumed.^77,84,86 Such mathematical expressions indicate which factors affect the CO₂RR and provide systematic insights for improving the CO₂RR performance.

---

From the equations, assuming that the c1 step is rate-limiting in the CO production, it can be identified that lowering the free energy of reaction (ΔG⁰_c1) would lead to an increase in the forward rate constant and reaction rate of step c1. As ΔG⁰_c1 is a function of the *COOH binding energy (E_b(COOH)) with the same directionality,⁸⁷ strengthening the *COOH binding affinity is expected to improve the CO production. If we assume that the c3 step is rate-limiting, e.g. strong CO binding, lowering the CO binding energy will increase the CO desorption and CO production rates. Given that the intermediate binding energies depend on the catalyst materials as well as the binding sites,^10,87–90 developing catalysts that can stabilize *COOH and destabilize *CO with a broken scaling relation becomes one of the important strategies for enhancing the CO production rate at low overpotentials.

Note that the free energy of reaction and intermediate binding energies can also be influenced by components of the reaction environment, such as the electrolyte ions and coadsorbates present at the catalyst–electrolyte interface.^10,91–97 These chemical species not only change the electronic property of the catalysts but also impose adsorbate–intermediate interactions, which affect the intermediate binding energies and the free energy of reaction. Furthermore, one can observe that the rate equation of CO production is correlated with the CO₂ partial pressure, i.e. the reactant concentration (Table 2). This implies that a high CO₂ concentration corresponds to increased CO production rates. As described in the previous section, the CO₂ concentration at the catalyst surface can be depleted, limiting the reaction rate as dissolved CO₂ is fed to the catalyst surface through a long pathway. Utilizing a GDL electrode with short pathways allows for a much higher CO₂ concentration at the catalyst surface,⁸¹ inducing several orders of magnitude higher current density. Na et al. reported that many CO₂RR products (e.g. CO, formic acid, n-propanol, acetaldehyde, acetone, acetic acid, ethylene glycol, etc.) have relatively low production costs compared to current market prices when the current density exceeds 2 A cm⁻², even if the complete product separation and recycling costs are considered.⁴⁰ Accordingly, the engineering of the catalyst–electrolyte interfaces allows for the control of the catalytic activity beyond just combining and engineering the materials. Additionally, we can consider the contribution of the free energy of activation. The rate constant includes the free energy of activation at the reversible potential (ΔG^#_ci(U⁰_ci)), where the free energies of the initial and final states involved in the reaction pathway are the same.⁷⁷ This term depends on the electrocatalyst materials, as well as the electrolyte environment.^84,98–100

Furthermore, similar approaches for suppressing the HER to enhance the FE of the CO₂RR are available. Increasing the free energy of reaction of the h1 step by destabilizing *H for the catalysts with a low *H binding affinity can lower the rate constants (Table 2). In contrast, for catalysts with a high *H binding affinity, stabilizing *H can elevate the free energy of reaction and the forward rate constant of the h2 step. Therefore, modifying the *H binding affinity using catalysts, ions, and adsorbates can be an effective strategy for suppressing the HER. In addition, the high local pH formed at the electrode surface during the CO₂ electroreduction can be increased by the low electrolyte buffer strength and the slow diffusion of proton donors from the bulk electrolyte region, which can retard the HER rate (e.g. local pH effect),^44,101,102 thereby enhancing the CO₂RR product selectivity.

Using the parameters included in the kinetic equations, advanced catalysts can be designed, and the focus needs to be extended to the interfacial area of the catalyst and electrolyte to address the problematic issues. Engineering the catalyst–electrolyte interface can provide useful solutions for breaking the scaling relations imposed on the intermediate binding energies, and for controlling the coadsorbates, electrolyte ions and concentration of chemical species.

4. Design of interfaces between the catalyst and the surface modulator

4.1. The impact of catalysts on CO₂RRs

The relative binding energy between the catalyst surface atom and the reaction intermediates is the key factor influencing the CO₂RR activity and selectivity. The binding properties were revealed to be closely related to the type and structure of the catalysts. Here, we briefly introduce experimental results and theoretical insights on single metal electrodes, as an example, to understand the effect of the catalyst on the CO₂RR performance. First, the type of metal primarily determines the type of major products formed, as shown in the representative FE data for various single metal electrodes (Fig. 2a).¹¹ Single metals can be categorized based on the by major products: HCOO⁻, CO, and H₂. Cu metal is specially classified because it can produce a variety of hydrocarbons and alcohols, ranging from C₁ to C₃ chemicals. Such a different electrocatalytic trend can be explained by the selective binding affinity of metals to certain reaction intermediates (Fig. 2b). The horizontal and vertical black lines indicate energy values corresponding to ΔG_CO = 0 and ΔG_H = 0, respectively. The HCOO⁻ group metals, such as Cd, Pb, and Sn, possess poor binding energies for *H, *COOH, and *CO, which are intermediates in the H₂ and CO evolution pathways; however, they relatively favour the binding of *OCHO, which is a key intermediate in the production of HCOOH.^12,74,75,103

In contrast, H₂ group metals, such as Ni, Pt, and Rh, have strong binding affinities for *CO, while their *H binding affinities are appropriate for the HER, approaching the line of ΔG_H = 0 (Fig. 2b).^10,104 Accordingly, H₂ becomes the dominant product, while *CO is poisoned at the metal surface. Cu and CO group metals, including Au, Ag, and Zn, commonly have medium binding affinities for *H with ΔG_H > 0 (Fig. 2b). However, there is a difference in their *CO binding energies (ΔG_CO > 0 for CO group metals and ΔG_CO < 0 for Cu). Considering that CO can be adsorbed on the Cu surface, *CO can undergo further electron–proton transfers to produce hydrocarbons and alcohols.⁸² Conversely, *CO can be desorbed from CO group metals due to the positive free energy of adsorption.

However, it should be noted that the major product of CO₂RR depends on the applied potentials even on the same electrode. It has been reported that the selectivity of a certain product basically forms a parabolic curve as a function of the potential with optimum selectivity.⁴⁷ Furthermore, Pd nanoparticles (NPs) produce HCOO⁻ selectively at low overpotentials, while CO evolves as a dominant product at high overpotentials.¹⁰⁵ On Cu-based electrodes, the major product changes from HCOO⁻ or CO to C₂H₄ to CH₄ depending on the applied potential.¹⁰⁶ These phenomena are related to the change in the complex factors, including the reaction kinetics, preferential reaction pathways, binding sites, adsorbate coverages, number of active sites, defects of the catalyst surface, and local concentration of reactants, with the potential difference.²⁵

More quantitative information on the overpotential and product selectivity of CO₂RR can be obtained using the concept of limiting potential (U_L), which is an effective mean of the theoretical onset potential.⁸² In addition, comparing the limiting potentials of CO₂RR and HER can provide information on the ability of electrocatalysts to drive the CO₂RR selectively and suppress the HER.^10,98

The reported limiting potentials on (211) metal surfaces as a function of *CO binding energy, where the adsorbate–adsorbate interactions by CO adsorption are considered under 0.5 ML *CO coverage, are shown in Fig. 3.¹⁰ The black and red lines represent the dependence of U_L on the *CO binding energy for the elementary steps of the CO₂RR and HER. The H₂ group metals (Pt, Rh, Pd, and Ni) are positioned at the top of the volcano plot for the HER, showing U_L,HER values close to 0 V (vs. RHE). In contrast, U_L,CO2RR is much more negative than U_L,HER with *CO → *CHO as the potential-determining step (PDS), due to their high *CO binding energies. This can explain why the H₂ group metals predominantly produce H₂. For CO group metals such as Au and Ag, CO₂ → *COOH is the PDS due to the low *CO binding energy, and *CO can be desorbed from the surface, while the |U_L,CO2RR − U_L,HER| values are as small as 0.40 and 0.46 V. Consequently, Au and Ag metals are CO-selective with low HER activities. The Cu metal is near the centre of the volcano plot composed of CO₂ → *COOH and *CO → *CHO lines. The moderate *CO binding allows for the easiest *CO hydrogenation among single metals to produce various hydrocarbon products, including CH₄.

Furthermore, engineering the material factors of single metal electrocatalysts, such as the particle size, exposed facet, grain boundary, vacancy, edge, and corner sites, has been reported to significantly influence the overpotential and product selectivity because the intermediate binding energies are highly sensitive to the atomic arrangement of active sites.^14,107 For example, in comparison with Au(111), the overpotentials of CO production on the (211) facet, edge, corner, and grain boundary sites decrease by 0.40, 0.42, 0.50, and 0.48 V, respectively, varying with the intermediate binding energy.^10,88,89,108 Moreover, the value of U_L,CO2RR − U_L,HER decreases, indicating the relative suppression of the HER. Such under-coordinated sites with high surface energies are known to contribute to the modification of the binding energy. Similarly, anisotropic nanomaterials and nanostructured catalysts with rough and curved surfaces have been applied to maximize the surface catalytically active sites.^109,110

The chemical identity of electrocatalysts and their surface geometric arrangements have helped in broadening the spectrum of materials from monometallic to bimetallic catalysts of various combinations of transition metals. Attempts to develop catalysts can be found in other review articles involving doping, alloying, mixing patterns, and supported structures, etc.^20,21,111 Bimetallic materials have effective tunability in controlling heteroatomic binding sites, surface strains, d-band centres, and interface creation, leading to the change in the intermediate binding strength. The movement of the key intermediate across the separate active sites is also proposed with bimetallic or multicomponent catalysts in the concept of tandem catalysts. Furthermore, bimetallic candidates with defined structural information can be found using DFT calculations combined with machine learning.¹¹² It is highly anticipated that a variety of high-performance multimetallic catalysts with proper binding affinities will be developed shortly.

Besides catalyst development, the surface molecular approach has emerged in recent years, because it is a straightforward and simple method for tuning the activity and reaction pathways at the catalyst interfaces. The following sections focus on design strategies for modifying catalyst interfaces using surface modulators. The interface conditions have attracted significant interest as the chemical properties of a heterogeneous catalyst surface can vary extensively and distinctively based on the interaction with diverse surface molecules. In addition, as the catalyst size approaches the nanoscale region, additional surfaces are exposed, which allow for more effective interfacial engineering. The interaction between the catalyst and the organic molecules (or polymers) on the surface can exert four possible catalytic synergistic effects on the current density, FE, and overpotential (Fig. 4): (i) the electronic modification of catalytic materials, (ii) stabilization of reaction intermediates, (iii) regulation of proton delivery, and (iv) structural transformation. In the following subsections, we review the recent studies where activity and selectivity enhancements were achieved using surface modulators, as summarized in Table 3. The promotional effect is divided into four categories, and for each category, the catalytic materials, surface modulators, responsible functional groups of the modulators, and proposed chemistry are described.

---

4.2. Electronic modification

The adsorption strength between the catalysts and reaction intermediates determines the overall catalytic activity, where optimum binding strength is required to achieve the best activity. In addition, the relative binding energy with different intermediate species governs the thermodynamically favoured reaction pathways, which suggests that designed CO₂ reduction catalysts should prefer the adsorption of carbon-containing species over hydrogen adsorption for high selectivity. The adsorption energetics of CO₂ reduction electrocatalysts are mainly regulated by their electronic structures (i.e. the d-band position). Therefore, the modulation of electronic structures is one of the most effective methods for designing catalytic materials. In the case of CO production by CO₂RR, balancing the *H, *CO, and *COOH adsorption abilities of the catalytic materials is crucial for obtaining high activity and selectivity.⁸² Ag and Au have the highest activities for CO generation, attributed to their optimum electronic structures for *CO and *COOH binding energetics. High binding energy of *COOH, the intermediate species involved in the rate determining step, is preferable to decrease the overpotential, while high binding energy of *CO slow down the CO desorption from the catalyst surface. However, the improvement of their activities is challenged by the linear scaling relation of the *CO and *COOH binding energies, where the correlation line is slightly offset from the theoretical apex. Kim et al. suggested a strategy to break this intrinsic correlation of d-band metals and to boost the activity of Ag-based catalysts.¹¹³ It was first proposed computationally that p-band elements can induce a partial covalency to metal catalysts and modulate the electronic structures. Among various p-band elements, S and As were expected to be highly effective in enhancing the CO₂ reduction performance of Ag-based catalysts. Later, this concept was realized experimentally.⁹⁶ Ag NPs were colloidally synthesized in the presence of cysteamine molecules, which acted as both an anchoring agent and S-containing surface modulator. A maximum CO FE of 84.4% was achieved with 5 nm Ag NPs at −0.75 V (vs. RHE), while an Ag foil showed a maximum efficiency of 70.5% at a 400 mV higher overpotential. The improvement was attributed to the unpaired electron localization due to cysteamine, which broke the scaling relation and improved *CO and *COOH binding energies toward the vertex (Fig. 5a and b).⁹⁶

Since cysteamine molecules contain both thiol (S) and amine (N) functional groups, the respective role of each group should be investigated. Hwang and co-workers prepared thiol- and amine-capped Ag NPs.⁹⁷ Oleylamine-capped Ag NPs exhibited a maximum CO FE of 94.2% at −0.75 V (vs. RHE) and maintained a high selectivity of >80% over a wide potential range. In contrast, an FE of 65.5% was obtained with dodecanethiol-capped Ag NPs. DFT calculations showed that the amino group destabilizes only *H binding, thereby maintaining the *COOH energetics, while the thiol group improves both *H and *COOH binding strengths toward the volcano apex, increasing the effectiveness of the Ag–amine interaction for CO₂ reduction.⁹⁷

The promotional effect of surface amine molecules significantly depends on their molecular structure, surface coverage, and anchoring sites. Wallace and co-workers investigated the structural effect of surface amines to improve the CO₂RR activity of Au NPs.¹¹⁴ The authors found that linear amines generally promoted CO₂RR, while branched amines resulted in a decrease in activity. Among the linear amines, long-chain amines were more effective than short-chain ones, which is related to the optimal coverage achieved with long-chain amine molecules. The activity improvement by amines was only noticeable with small Au NPs, implying the existence of synergistic effects between the amine group and the undercoordinated Au sites.¹¹⁴

The electron donating and withdrawing abilities of surface molecules play a significant role in tuning the electronic structure of the catalytic centre. This strategy is analogous to the method by which the activity of molecular catalysts is improved.¹¹⁵ Chang's group successfully enhanced the activity of Au NPs using electron-rich N-heterocyclic carbine (NHC).¹¹⁶ The NHC molecules were spontaneously adsorbed on the Au NPs, replacing the parent surfactant due to the strong interaction with the Au NPs (Fig. 5c). NHC-functionalized Au NPs showed a 7.6 times higher CO production rate with a high FE of 83% compared to those of oleylamine-capped Au NPs (FE = 53%). Tafel analysis suggested that the surface NHC molecules strongly donated σ-electrons to Au NPs, thereby facilitating the first electron transfer step from Au to CO₂, which is thought to be the general rate-determining step (RDS) of CO₂RR (Fig. 5d).¹¹⁶ The same group later extended the NHC functionalization approach to Pd foils to enhance the production of C1 chemicals (i.e. formate and CO) by 32-fold.¹¹⁷ The activity and selectivity changes were considered sensitive to the structure of the surface NHC derivatives, suggesting the potential of fine-tuning the performance through surface modulation.

Regarding the electrocatalyst applications, the performance of powder-type catalytic materials is typically tested by dispersing them in a solvent (i.e. an alcohol/H₂O mixture) and depositing them on a working electrode substrate. To prevent the mechanical detachment of the catalyst during the reaction (specifically during gas-evolving reactions), a polymer binder was used. Nafion, a specialized polymer for proton conduction, was predominantly used in the preparation of the catalyst electrode. Despite the significant Nafion amounts in the catalyst layer (i.e. 1 mg of the catalyst and 10 μL of a 5 wt% Nafion solution result in a catalyst to Nafion mass ratio of 1:∼0.5), its role in electrocatalysis has barely been investigated. Flake et al. investigated the role of Nafion and polyvinylidene fluoride (PVDF) on the surface of an Au₂₅ cluster and 5 nm Au NPs.¹¹⁸ The main structural difference between both polymers was the presence of sulfonate groups in Nafion. The authors found that Nafion-containing catalysts exhibited considerably higher CO FE than PVDF-containing ones and the bare Au foil. X-ray photoelectron spectroscopy (XPS) measurements revealed that the Nafion-containing Au showed significant shifts in the Au 4f binding energy. This indicated that the sulfonate group altered the electronic state of Au and optimized the adsorption energy of the key reaction intermediates.¹¹⁸ Furthermore, Chen and co-workers systematically investigated the effect of polymer binder types.¹¹⁹ Five types of polymers were selected, including Nafion, polyfluoroethylenes, poly(vinyl alcohol), and poly(acrylic acid). The electrocatalytic performance of the Au NPs supported on carbon black indicated that F-containing polymers (including Nafion) promoted the selective CO₂RR to CO. DFT calculations suggested a strong HER-suppression effect of C–F functional groups, which led to a high CO selectivity. These results opened the possibility of utilizing polymers with these groups, other than Nafion, as alternatives for CO₂ reduction, and highlighted out the importance of polymer binders as catalytic activity modulators.¹¹⁹

4.3. Stabilization of reaction intermediates and CO₂

The adsorption mode and geometry are other important factors influencing the activity and selectivity of catalytic processes. The surface structure has direct relevance to the adsorption properties; therefore, surface molecules affect the binding energies of intermediates. Catalytic materials have a limited number of possible surface configurations due to their crystal structures and surface energies. Combining core catalysts with surface modulators that interact with the reactant (CO₂) and reaction intermediates can create intriguing synergies to lower the activation energy, thereby stabilizing the reactants and intermediates. Meyer et al. investigated the synergy between N-doped carbon nanotube (NCNT) catalysts and polyethyleneimine (PEI).¹²⁰ PEI-coated NCNT exhibited a maximum formate FE of 87% and a three-fold higher intrinsic activity compared to those of NCNT. PEI–NCNT and NCNT have a common RDS, i.e. the first electron transfer to CO₂, as indicated by their similar Tafel slopes. The authors assumed that the promotional effect of the PEI layer was attributed to the stabilization of the CO₂˙⁻ intermediate radical by the hydrogen bonding network of PEI and the increase in the local CO₂ concentration.¹²⁰ Other than the Meyer group's report, almost all organic modifier studies utilized Ag, Au, and Cu as platform materials. Since synergistic catalytic effects between carbon nanomaterials and surface modulators have rarely been reported, intensive investigations should be conducted in the future.

Hydrogen bonding at the catalyst surface enormously benefits the CO₂ electrocatalysis. Regarding the above-mentioned Ag–cysteamine system (Section 4.2), the Goddard group assumed that cysteamine could have a stabilizing as well as an electronic effect on CO₂ and conducted detailed quantum mechanics calculations.¹²¹ It was revealed that cysteamine not only exerted an effect on the electronic structure, but also stabilized the chemisorption of CO₂ molecules with the support of the opposite NH₂ group using hydrogen bonding networks (Fig. 6a). Experimentally, they showed that the modification of Ag NPs with C₁₁ analogue of cysteamine (C₂) only moderately decreased the overpotential.¹²¹

Wang et al. successfully doubled the hydrocarbon production by coating Cu(OH)₂ nanowires with amino acids (Fig. 6b).¹²² The promotional effect was general for various types of amino acids and several Cu-based electrodes, including Cu nanowires, Cu foil, and annealed Cu. The authors attempted to reveal the origin of the activity enhancement by investigating the role of each functional group in amino acid (i.e. amine and carboxyl groups). Other groups, such as carboxyl, nitro, and quinone, lowered the production rate of hydrocarbons, indicating that the activity improvement was attributed to the amine group. DFT calculations explained that the NH₃⁺ group (protonated amine under experimental conditions) could form a hydrogen-bond with *COOH and *CHO intermediates, where the latter is an important intermediate for hydrocarbon production (Fig. 6c).¹²² Similar results were obtained by Zhang et al., where methyl carbamate containing amine groups on Cu foil could enhance CH₄ formation up to 81.6% at −2.13 V (vs. RHE).¹²³ They suggested that the protonated amine (–NH₃⁺) could stabilize *CHO intermediates by hydrogen bonding and partial charge transfer from Cu to C. The *CHO intermediates are important for CH₄ production rather than CO dimerization.

The reports on the promotional effects of surface modulators occasionally contradict one another. Andreoli et al. prepared Cu foams modified with poly(acrylamide), poly(acrylic acid), and poly(allylamine).¹²⁴ Only the poly(acrylamide)-modified electrode exhibited a remarkable two-fold activity improvement for C₂H₄ production at −0.96 V (vs. RHE), while the poly(acrylic acid)-modified one showed only a slight increase. The poly(allylamine)-based Cu foams showed a HER selectivity of almost 100%, which is rather contradictory to previous work that proposed the promotional role of the amine group.¹²² DFT calculations revealed that the acrylamide oligomers on Cu surfaces are beneficial because they (i) increase the local CO coverage and (ii) stabilize the CO–CO dimers, which are important intermediates for C₂₊ chemical production. It is assumed that amine groups may elevate the CO₂RR rate cooperatively with other groups on Cu surfaces.¹²⁴

Liu et al. demonstrated the possibility of utilizing porphyrin molecules as surface promoters.¹²⁵ Typically, surface modulators partially block catalytic sites; however, owing to the open structure of such a tetradentate ligand, reactants could access almost the entire surface of the functionalized porphyrin–Au hybrid. The functionalized Au NPs exhibited a 300 mV lower overpotential, higher CO FE, and much better long-term stability (72 h) compared to the oleylamine-coated Au NPs. The enhancement by the porphyrin molecule was explained by the fast electron transfer and stabilization of the *COOH intermediate.¹²⁵

4.4. Regulation of proton delivery

The CO₂RR selectivity for each product is affected by the relative reaction kinetics of the competitive and/or shared intermediates. In particular, the chemical species involved in the RDS are important in determining the activity and selectivity. As CO₂RR proceeds via multiple reaction steps involving electrons and/or protons, the products have different pH-dependences according to the involvement of the proton in the RDS. The CO₂RR also competes with HER; therefore, the pH of the electrolyte can change the product selectivity of the CO₂RR and H₂ production. This change is not limited by the pH of the bulk electrolyte and is sensitively affected by the local pH gradient near the catalyst–electrolyte interface. For example, CH₄ formation is proposed to involve a proton-transfer RDS, which forms *CHO, implying that a high pH is not favourable.^126–128 The proton reduction (i.e. the HER), in most cases, also depends on the electrolyte pH as the major chemical species of the proton donor vary (i.e. proton, water, bicarbonate, or other protonated chemical species) and, thus, affect the reaction pathway. Typically, a large overpotential is required in an alkaline electrolyte,¹²⁹ and at a low pH, the reactions with the proton-transfer RDS would dominate. In contrast, the production rates of C₂H₄ and C₂H₅OH from CO₂RR on the Cu electrode are proposed to be independent of pH since the rate-determining C–C bond formation step is considered not to involve proton transfers.¹³⁰ Therefore, a high local pH is generally pursued for the selective generation of C₂ products and well suppress the H₂ and CH₄ formations. However, a bulk alkaline electrolyte is not suitable for a typical H-type cell, where dissolved CO₂ molecules are fed through the aqueous electrolyte, due to the rapid equilibration of CO₂ with OH⁻ to generate bicarbonate/carbonate, leading to the absence of usable CO_2(aq).

Therefore, the pH is one of the crucial factors in controlling the selectivity of targeted products from the CO₂RR. In the H-type cell, neutral electrolytes are widely used for the CO₂RR.⁴² Even under neutral conditions, the advantage of a high pH in the CO₂RR can be exploited by roughening the electrode surface or introducing a porous structure, so that a local pH gradient can be created near the catalyst surface due to the insufficient mass transfer under the CO₂RR conditions. This aspect will be discussed in Section 5. Another approach involves the direct feeding of gaseous CO₂ to the catalyst surface through the reactor design, enabling the use of the alkaline electrolyte on one side of the catalyst layer (also discussed in Section 5).

The addition of surface organic molecules that have different affinities to proton and water (water is also a proton source) can elaborately control the local proton transfer at the catalyst–electrolyte interface and the reaction selectivity. Earlier works carried out by Bocarsly et al. showed the possibility of tuning the reaction pathways of CO₂ electroreduction, using proton-mediating molecules. The authors demonstrated the generation of a six-electron product (CH₃OH) on hydrogenated Pd and Pt electrodes in a pyridine-containing aqueous solution.^131,132 These results are impressive because these metals cannot independently produce CH₃OH in aqueous solutions, which highlights the proton-mediating role of pyridine.

To systematically study the proton delivery capability of surface modulators, Flake et al. prepared three molecules; 2-mercaptopropionic acid (MPA), 4-pyridylethylmercaptan (PEM), and cysteamine, with different pK_a values of each functional group, and attached them onto an Au foil.¹³³ The thiol group at one end formed a strong Au–S covalent bond, and the group at the other end induced functionality at the Au–modulator interface. The Au–MPA electrode showed almost 100% HER selectivity, which indicated that the facile proton donating ability of the carboxyl group (pK_a = 3.7) promoted the HER. For Au–PEM, a three-fold increase in formate production was observed (Fig. 7a). This result was attributed to the moderate pK_a value (5.2), resulting in an appropriate amount of adsorbed H and inhibiting the first electron transfer to CO₂ to form *COOH, the reaction intermediate for CO (Fig. 7b). The adequate *H coverage supported the formation of the *OCHO intermediate and subsequently the formate. The cysteamine-modified Au showed similar FEs for the products compared to the bare Au foil, while the current densities were doubled.¹³³

Recently, Toma et al. presented new insights into the catalytic performance of Cu–polymer composites.¹³⁴ The authors prepared oxide-derived (OD) Cu foils and 10 types of polymers. The polymer-coated OD-Cu was tested for CO₂ reduction at −0.7 V (vs. RHE), where hydrocarbon products are barely produced. Protic polymers enhance the HER activity as the protic groups facilitate proton delivery to the electrode. Interestingly, there was a distinct correlation between their hydrophobicity (contact angle) and the formic acid/CO product ratio (Fig. 7c). Using reactive molecular dynamics simulations, the water density surrounding the modified OD-Cu surface was calculated. The hydrophilicity and hydrophobicity influenced the nanoscale water structure in the vicinity of the electrode, which, in turn, impacted the hydrogen binding energy and product selectivity (Fig. 7d).¹³⁴

4.5. Change in catalyst structures

Another key aspect involves studying the dynamic changes in the catalyst and/or surface organic molecules to understand the promotional roles of surface modulators. A strong interaction between the catalyst and surface molecules can induce structural modifications that may lead to activity and selectivity changes. For instance, Chernyshova et al. investigated Cu foils coated with poly(4-vinyl pyridine).¹³⁵ The modified Cu electrode exhibited a maximum FE of 40% toward formate production in a 0.1 M KHCO₃ solution at −1.3 V (vs. Ag/AgCl). XPS and Raman analyses revealed the formation of a pyridyl–Cu^II complex (Cu^II–N bond) during the electrolysis and showed that this structure was stable for 30 h of operation. This phenomenon was attributed to the in situ dissolution of Cu atoms and their subsequent complexation by the pyridyl group of the polymer.¹³⁵ Xu et al. examined Au–thiol interface systems.¹³⁶ The formation of strong covalent Au–S bonds can withdraw the surface Au atoms, thus creating many defect sites that are favourable for CO₂RR (Fig. 8a). The defect sites result in a two-fold enhancement in both the CO FE and current density.¹³⁶ The restructuring of the Au surface was also predicted using DFT calculations when NHC was supported on Au.¹¹⁶ The Au atoms could be pulled from the lattice by strong interactions with two carbene molecules.

Some organic molecules increased the CO₂ reduction product selectivity by blocking specific active sites. The Agapie group demonstrated the promoted C₂₊ chemical production on polycrystalline Cu using N-substituted pyridinium and phenanthrolinium additives in the electrolyte (10 mM).^137,138 The authors achieved a total FE of ∼70–80% for C₂₊ products at −1.1 V (vs. RHE) in the presence of N-tolylpyridinium chloride¹³⁷ and N,N′-ethylene-phenanthrolinium dibromide.¹³⁸ Such a high selectivity was ascribed to the efficient suppression of CH₄ and the H₂ formation. The N-arylpyridinium additives induced the highest C₂₊ selectivity. The authors assumed that the improvement by the pyridinium additives was attributed to the selective poisoning of the Cu active sites for CH₄ and H₂ production and the increased local pH near the electrode.¹³⁷ For phenanthrolinium bromide, the morphology of the Cu surface was transformed to cube-like nanostructures with (100) Cu facets, which are known to be efficient for C–C coupling.¹³⁸ The site-blocking strategy could specifically be beneficial if the active site for the HER can be selectively inhibited. Jaramillo et al. showed that aqueous 1-ethyl-3-methylimidazolium solutions (0.1 M) can suppress the HER activities of Cu, Ag, and Fe metals by up to 75%.¹³⁹ However, it should be cautioned that the inhibition effect was not observed at high pH values and the molecule was susceptible to both reductive and oxidative potentials which could induce its breakdown.¹³⁹

Some combinations of electrocatalysts and surface organic molecules could remarkably promote structural changes that induce significant activity improvement. Hwang et al. prepared cysteamine-capped Cu₂O nanocubes (NCs) via one-pot polyol reductions.⁵⁷ Interestingly, chronoamperometric measurement revealed that within 6 h, the C₂H₄ FE increased from 27% to 57.3% (Fig. 8b). High-resolution transmission electron microscopy (HR-TEM) analysis indicated that the initial ∼20 nm-sized Cu₂O NCs were fragmented into fine Cu₂O nanocrystals with a domain size of 2–4 nm, implying a close packing with numerous domain boundaries (Fig. 8c).⁵⁷ In preceding studies, Cu-based CO₂ reduction electrocatalysts generally suffer from severe aggregation and deformation of their initial morphologies, resulting from an in situ dissolution–redeposition process under applied negative potentials.^140–142 In this regard, the electrochemical fragmentation phenomena were attributed to the presence of surface cysteamine molecules.

4.6. Comparison of activity improvement by surface modulators

To summarize this section, the molecular functionalization of the catalyst surface has been recently proven to be a simple and straightforward approach for the extrinsic modification of the CO₂RR activity and selectivity. This method changes the catalyst–electrolyte interface, which can lead to optimization of the catalyst structure and the binding mode, consequently stabilizing the reaction intermediates, and modification of the transport behaviour of protons and water. These surface additives provide an opportunity to fine-tune the catalyst–electrolyte interface, which cannot be fully achieved only with catalyst design and electrolyte engineering. The electrocatalytic activities of reported catalysts before and after the addition of a surface promoter are summarized to show the extent of improvement. Fig. 9 displays the maximum CO₂-to-CO FEs and CO partial current densities of Ag- and Au-based catalysts. Bare Ag foil exhibits mediocre maximum CO FEs ranging from 70 to 80%, only at highly negative potentials (<−1.1 V vs. RHE). The activity increase is remarkable when Ag-based NPs are surface-functionalized, as evidenced by the 300–600 mV decrease in the overpotential at the maximum CO FE, as well as the increase in the maximum CO FEs to over 90% (Fig. 9a). For Au-based catalysts, the significant increase in the maximum FE from ∼30% to ∼100% is noticeable, while the decrease in the overpotential at the maximum CO FE is also significant (100–400 mV) (Fig. 9b). Similar trends in the activity improvement are observed in terms of the CO partial current density, where the overpotential is reduced by up to 600 mV to obtain the same current density (Fig. 9c and d). Notably, the surface functionalization strategy combined with nanostructure leads to a similar degree of activity enhancement that has been accomplished by the catalyst design. For instance, the generation of abundant grain boundaries in polycrystalline Au electrodes led to the overpotential decrease of ∼200 mV and an increase in the CO FE by 20%, which can be achieved by tuning the interfaces with the same catalysts (Fig. 9b and d).⁵⁰ Furthermore, we note that particle-type Au catalysts exhibited significant improvement than the Au foil with surface modulators (Fig. 9d). This implies that the interface tuning effect can be maximized with high-surface-area catalysts, and the surface structure can affect the interaction between the surface atoms and promoter molecules.

The CO₂RR performance was also summarized for Cu-based catalysts (Fig. 10). Among many chemicals made from Cu-based catalysts, the activity for only C₂H₄ is shown as it is a special multicarbon product that is the most attractive and intensively studied. The surface promoters at Cu interfaces can induce an increase in the C₂H₄ selectivity by ∼10–50%, while bare Cu foils exhibit low C₂H₄ FEs of 5–20%. Although quantitative discussions on the other products are not presented here due to the lack of numerical data, Table 3 shows that the surface modulators facilitate varying selectivity between CO and formate at low overpotentials, C₂H₄ and C₂H₅OH at intermediate overpotentials, and CH₄ at high overpotentials. We envisioned that the design of appropriate molecules can break the trend of potential dependence in Cu-based catalysts for efficient CO₂RRs to produce hydrocarbons.

4.7. Future guidelines for surface modulation research

Many organic modulators possess multiple functional groups in a single molecule, which play different roles in electrocatalysis at the interfaces. In some cases, the effect of a certain functional group may dominate that of another group. If the functional groups have similar catalytic effects at the interfaces, we could measure their average properties or an unexpected new result from their cooperative interaction. Therefore, it is necessary to identify the role of each group to clearly understand the chemistry through systematic investigation.

We note that the surface modulator covers the catalyst or sometimes changes the surface arrangement of the catalyst. These phenomena can impair the activity comparison of modified catalysts with unmodified ones. Therefore, researchers are encouraged to measure the electrochemically active surface area (ECSA) and discuss the ECSA-normalized CO₂ reduction activity. Among various reported methodologies, the double layer capacitance measurement is the simplest and, thus, the most popular method for estimating the ECSA.^143,144 However, extreme care must be taken in both the measurement and interpretation because organic modulators are expected to exhibit double layer capacitance. The underpotential deposition method is preferred to this method because probe atoms (mostly Pb) are selectively deposited on the metal catalyst.¹⁴⁵

Another consideration is that the surface density of the modulator can critically affect the electrocatalytic performance as shown in Wallace et al.¹¹⁴ and Wang et al.¹²² These studies commonly suggest that low or excessive amounts of surface modulators are rather detrimental to the activity. Therefore, it is recommended to specify the used amount and to quantify the surface coverage and surface contents of the modulators to acquire consistent results in future studies.

Even the molecules that bind strongly with catalysts can electrochemically detach at CO₂RR potentials. Flake and Xu et al. revealed, using attenuated total reflectance infrared (ATR-IR) spectroscopy before and after electrochemical tests, that a significant number of surface-bound thiols were lost after the operation.¹³⁶ This result necessitates the estimation of the binding stability of modulators on catalyst surfaces under negative potentials in future studies, as well as the confirmation whether or not the enhancement is attributed to the surface molecules.

Until now, most surface modulator studies have been carried out in aqueous H-type electrolysis cells for fundamental investigation. The catalyst–electrolyte interface is an important factor that must be considered also in other types of CO₂ electrolysis cells utilizing gas-diffusion electrodes (GDEs), which have recently emerged to obtain commercially relevant current densities. The surface functionalization also provides a great opportunity to modify the catalyst–electrolyte–CO₂ three-phase interface, as demonstrated by Sargent group for the improved production of C₂ chemicals in both microfluidic and membrane electrode assembly cells, which are discussed in detail in Section 5.^146,147 We envision that the surface modulator strategies for activity enhancement will be an interesting research area, regardless of the reactor type.

Finally, it was found that most of the studies focused on the effect of N-containing compounds, with a few cases of S- and F-groups. There is still an extensive research scope for research to improve the catalyst performance and study new electrochemical, catalytic phenomena, as well as the underlying chemistry. Future studies are encouraged to further investigate other organic promoters with a variety of functional groups.

5. Understanding catalyst–electrolyte interfaces

The electrocatalytic activity of CO₂ reduction catalysts is typically measured using an electrochemical system, where CO₂ and the proton source are mediated by an aqueous electrolyte. When an electrode is immersed in the electrolyte, an electric double layer is constructed at the electrode–electrolyte interface, where covalently bonded species and reaction intermediates are present in the inner Helmholtz plane (IHP), and hydrated ions are situated in the outer Helmholtz plane (OHP), held by electrostatic forces (Fig. 11a). In addition, there are several dynamic equilibria involving CO₂ and H₂O, which can potentially vary the local ion distribution and pH at the interface as the CO₂RR proceeds (Section 3.1). Therefore, the electrolyte is the other important component that can control the CO₂RR activity at the catalyst interface.

Concerning the ion types and applied electrode potential, the ionic distribution at the electric double layer is determined, which strongly affects many factors for the CO₂RR, as schematically illustrated in Fig. 11b. At the potentials where the CO₂RR occurs, the electrode is negatively polarized. This situation results in the attraction of the hydrated cations to the electrode owing to Coulomb interactions and, subsequently, the increase in the population of cations at the OHP. The cations at the OHP elevate the local electrode potential, which can change the charge transfer kinetics. The cations also generate a local electric field, which improves the stability of the covalently adsorbed intermediates at the IHP. Conversely, they may rather directly interact with the polar intermediates by Coulomb interactions. In addition, the hydrated cations act as pH-buffering agents, which determine the local pH, pH-dependent reaction mechanism, and proton transport. Meanwhile, specifically adsorbed ions at the IHP can have multiple roles: they can alter the electronic structure of the surface catalyst atoms, block the active sites, and interact with the intermediates via van der Waals forces. These catalytic effects are greatly influenced by the type and concentration of the electrolyte. Despite the importance of the electrolyte-dependent electric double layer structure on the electrocatalytic process, its contribution has been less understood compared to the impact of heterogeneous catalysts. This is mainly because the characterization of electrolytes or their interaction with the catalyst surface during CO₂RR is not well-established yet. The role of electrolytes is typically proposed based on experimental CO₂RR activity trends and simulation results. Mass diffusion and the chemical balance of the individual chemical species in the electrolyte must be considered to understand the role of the electrolyte.

5.1. Effect of the electrolyte type on the interface

Bicarbonate (or carbonate) is a popular electrolyte for CO₂RR as it provides near-neutral pH, and the dissolved CO₂ concentration is relatively high.^148–152 The role of the bicarbonate has been a subject of controversy; however, recent studies have attempted to characterize the catalyst surface using in situ spectroscopy for clarification. Xu et al. attempted to identify the role of the bicarbonate anion in the CO₂RR on CO-selective metal electrodes using in situ attenuated total reflectance-surface enhanced infrared absorption spectroscopy (ATR-SEIRAS).⁴⁵ With ¹³C isotope labelling and electrokinetic experiments, it was revealed that surface bicarbonate anions rapidly equilibrated with aqueous CO₂ rather than the dissolved CO₂ used for the reduction reaction. Shao and co-workers arrived at a similar conclusion using a Cu electrode. In addition, the authors proposed that even the dissolved CO₂ required pre-equilibration with bicarbonate, which is more easily attracted to the surface cations than CO₂.¹⁵¹

Bicarbonate is a general electrolyte for CO₂RR; however, its corresponding alkali metal cations appear to have a significant effect on the CO₂RR activity and selectivity, owing to the relatively high population of the cations at the OHP. Numerous previous papers have shown that large alkali cations enhance the FE toward C₂ products on Cu and C₁ products on Ag and Sn.^153–155 In an earlier work, Hori et al. explained that large cations are specifically adsorbed more easily than small ones due to the relatively small hydration numbers.¹⁵³ Many specifically adsorbed cations could elevate the potential at the OHP and decrease the local proton concentration, thereby suppressing the HER. The Bell group explained cation-size-dependent CO₂RR activity trends with the interfacial pH (Fig. 12a and b).¹⁵⁶ It was claimed that the cation size significantly affects the hydrolysis constant of the hydrated cation. Consequently, the pK_a value of Li⁺ was calculated to be three times higher than that of Cs⁺. The hydrated Cs⁺ acted like a buffer, maintaining the low pH near the electrode and increasing the local CO₂ concentration compared to Li⁺ by 28 times (Fig. 12c).¹⁵¹ This cation-dependent pH at the electrode–electrolyte interface was experimentally demonstrated by Cuesta et al. using in situ ATR-SEIRAS. The authors utilized the ratio between the interfacial CO₂ and HCO₃⁻ concentrations, which is related to the pH.¹⁵⁷

The Bell group provided new insights regarding the cationic promotion effect, particularly at relatively low overpotentials, where the local pH is marginally built.¹⁵⁸ They found that the HER and CH₄ partial currents remained steady, while formate, C₂H₄, and C₂H₅OH formations proceeded using large alkali cations. DFT calculations revealed that the high density of the large cations at the OHP increased the interfacial dipole field compared to the case with small ones, which improved the stability of key reaction intermediates with high dipole moments (e.g. *CO₂, *CO, and *OCCO). The cation-size independent production of H₂ and CH₄ was attributed to the zero dipole moment of *H and *CHO, which are the reaction intermediates of these products (Fig. 12d).⁹⁵ It was suggested that the interfacial field effect could be maximized using high-valence cations with small radii.^27,158 However, transition metal ions (Fe²⁺, Zn²⁺, etc.) significantly deactivate the metal electrode by electrodeposition, even in trace amounts.^159–161

The electrocatalytic process of CO₂RR is largely affected by the cation type due to the high population of cations under negative reaction potentials. In addition to the contributions of cations, anions contribute significantly to control the electrocatalytic trends by modifying the interface. First, the buffering capability of an anion can affect the product distribution.¹³⁰ For Cu-based electrodes, non-buffering anions, such as SO₄²⁻ and ClO₄⁻, increase C₂H₄ and C₂H₅OH selectivity with a high C₂/C₁ ratio, while phosphate dominantly increases the H₂ selectivity at a low overpotential and CH₄ at a high overpotential. Bicarbonate produced both C₂ and C₁ with a mediocre C₂/C₁ ratio. In a non-buffering electrolyte, the locally elevated OH⁻ concentration at the interface can even suppress the formation of H₂ and CH₄, whose RDSs involve proton transfers. In contrast, the buffering anion can mitigate the pH changes by acting as a proton donor, where its donating power depends on the pK_a of the anion, and thus increase the H₂ and CH₄ selectivity. The production rates of the other products (CO, formate, C₂H₄, and C₂H₅OH), independent of the proton transfer, are constant across the entire pH range.¹³⁰ The second effect of anions involves a more direct modification of the interface, including catalyst structure change and interaction with the reaction intermediates, enabled by specifically adsorbed anions that have strong affinity to the catalyst surface in the IHP. Halides have been most actively investigated as promoting anions for the CO₂RR, particularly for C₂ production on Cu-based electrodes.¹⁶² Ogura and co-workers achieved very high C₂H₄ FEs using a CuBr-modified Cu mesh in the presence of 3 M KBr.^163,164 The high efficiency was attributed to the stabilization of methylene intermediate radicals by CuBr.^163,164 Recent works provided activity trends on Cu-based electrodes based on halide-containing electrolytes, where the CO₂ reduction current density increased in the order of Cl⁻ < Br⁻ < I⁻ (Fig. 13a and b).^165–168 However, there were inconsistencies in the product distributions. Strasser et al. achieved improved CH₄ production in the presence of halides.¹⁶⁵ Cuenya et al. reported that C₂H₄, CH₄, and formate productions were promoted,¹⁶⁶ and Yeo et al. were able to increase the C₂H₄ and C₂H₅OH selectivity using halide anions.¹⁶⁷ It is assumed that these research groups have their independent protocols for preparing the Cu foil electrodes, leading to different local surface structures. This, in turn, could affect the adsorption of halide anions, which is facet dependent.¹⁶⁹

However, it is consistently suggested that electrochemical cycling or electroreduction conditions induces surface reconstruction, which increases the roughness (effective surface area) and/or in situ exposure of the Cu(100) surface, thus favouring the C–C coupling process (Fig. 13c and d).^{165–168,170} Another explanation of the boosting effect of halide is the change of electronic structure of the Cu surface. It was suggested that halide can suppress the complete reduction of Cu and stabilize Cu⁺ (or Cu₂O) species under the highly reductive conditions, which has been demonstrated to be effective for C–C bond formation (Fig. 13e).^166,171 Interestingly, the presence of iodide in the electrolyte enhanced the long-term stability as well as the activity; however, more detailed research is required.¹⁶⁸

Halides can also affect the CO₂RR activity of other metals, such as Ag and Zn, selectively producing CO.^172–177 For Zn, however, different trends were observed for OD-Zn catalysts and reported by Hwang et al., where the CO production activity increased in the order of I⁻ < Br⁻ < Cl⁻. The presence of halide ions during the preparation of OD-Zn catalysts can modulate the surface roughness as well.¹⁷³ Zhao et al. showed that the presence of Cl⁻ led to the formation of surface Zn–Cl bonds, which could suppress the HER more effectively and promote reductive CO₂ adsorption from electron-rich Cl species.¹⁷⁷ The adsorption of a halide anion on the catalyst surface has been proposed to modulate the binding energies of the intermediates. The interaction between the anion and the catalyst is also proposed to modulate the electronic states of the catalyst and affect the activity as well. The proposed origin of the halide effect is not general for the CO₂RR but varies depending on the electrode materials. It is assumed that the electronic effect appears dominant for Ag- and Zn-based catalysts because no dramatic structural deformation occurs. For Cu-based catalysts, however, the promotional effect of halides is attributed to the combination of structural and electronic effects. These effects presumably synergize because in situ roughening increases the density of specifically adsorbed halides, and the in situ generation of Cu(100) can promote halide adsorption, owing to relatively strong adsorption energy on the facet.¹⁶⁹ Determining the effect of each factor will be a challenging task due to the great difficulty associated with decoupling these effects. One aspect that is worth considering to explain the anion-dependent activity is the interfacial electric field created by specifically adsorbed anions, revealed to be critical for the cation.¹⁶⁹

Oh et al. investigated the effect of CN⁻ and Cl⁻ anions on Au surfaces.¹⁷⁸ Such specifically adsorbed anionic species can directly interact with the electrocatalyst and/or adsorbed reaction intermediates. CN- and Cl-adsorbed Au surfaces exhibited a higher current density of CO FE than bare Au. DFT calculations suggested that the adsorbed CN⁻ and Cl⁻ could stabilize the *COOH intermediate mainly via van der Waals interactions. Although such anions have been considered as catalyst poisons, they can be utilized as promotors in the CO₂RR, providing interface tuning of the adsorbed anions. However, most of the specifically adsorbed anions were unstable and easily detached under the respective operating conditions.¹⁷⁸

The electrolyte effect on non-metallic electrocatalysts has rarely been documented. Doped carbon nanomaterials have attracted immense attention as potential catalysts for selective CO or formate production.¹⁷⁹ Due to their substantially different catalyst surface properties (and presumably electric double layer structure), diverse electrocatalytic trends are expected with this electrolyte type, necessitating the study on carbon-based electrodes. Einaga's group investigated the influence of the electrolyte type on the CO₂ reduction activity of boron-doped diamond (BDD) electrodes.^180–182 In hydroxide-HCl-based solutions, BDD produced formate rather than CO, regardless of the cation type.¹⁸⁰ The highest FE of 71% was obtained with Rb⁺-based electrolytes. Interestingly, the formate selectivity was higher at low concentrations of large cations and high concentrations of small ones. Subsequently, the same group systematically examined the influence of the electrolyte composition using various types of alkali metal-based halide solutions.¹⁸¹ Regardless of the halide type, the formate selectivity was relatively high when large cations were present in the electrolyte. However, two different trends were noticeable compared with the cases in previous studies using metal electrodes: (i) comparable or higher FE for formate was obtained in Rb⁺-based (not Cs⁺-based) electrolytes, particularly in RbBr solutions (FE = 95%); and (ii) the current density was larger with small cations. In addition, the authors found that halides (Cl⁻, Br⁻, and I⁻), as well as sulfate anions, could efficiently produce formate through their specific adsorptions. Regarding the anions, improved selectivity was obtained with small anions. Recent work by the same group showed a higher formate selectivity for BDD in KCl solution than in KClO₄ solution.¹⁸² They assumed that the adsorption of CO₂˙⁻ radical intermediates occurs preferentially in the KCl solution.

5.2. Regulation of mass diffusion near the catalyst interface

In CO₂RR, the mass diffusion of the reactants (i.e. CO₂, H⁺, and H₂O) is another important parameter that controls the catalytic activity beyond the intrinsic activity of the catalyst. The mass diffusion of CO₂ limits the catalytic activity at a current density of several tens of mA cm⁻² due to the low aqueous CO₂ solubility of 33 mM under ambient conditions (i.e. room temperature, 1 atm, and neutral pH).^42,183 The temperature and pressure dependences of the gas solubility, described by Henry's law, have widely been applied by researchers to improve the low solubility.^184–186 However, because the reaction proceeds at a low temperature and high pressure, it is difficult to achieve a perfect solution due to the limited ion mobility at low temperatures and the instability of the reactor operation at high pressures. Therefore, the mass diffusion control of reactants (i.e. CO₂, H⁺, and H₂O) near the electrode surface is important. In addition, the mass diffusion of the proton sources affects the catalytic activity, both in liquid electrolytes and in GDE type cells. In this section, we describe two categories of results obtained by the mass diffusion control of CO₂, H⁺, and OH⁻: (i) the local pH effect and (ii) the application of a three-phase interface by modifying the catalyst structure.

5.3. Design of CO₂ electrolyzers using GDEs

The research on improving the CO₂ mass transfer by changing the catalyst structure has been successful to some extent; however, exceeding the CO₂RR current density of ∼30 mA cm⁻² is still difficult even with a high overpotential in the aqueous H-cell.²⁰⁴ There is a technical mismatch with the oxidation current density in CO₂ electrolyzers because the current density for the OER increases significantly as the overpotential increases, while the CO₂RR current density rather decreases due to the extensive HER activity. To achieve a high CO₂ reduction current density at a practical level, it is desirable to feed CO₂ gas directly to the catalyst layer of the GDE instead of dissolving CO₂ in the aqueous electrolyte. These CO₂ gas-phase electrolyzers can achieve current densities of hundreds of mA cm⁻² for CO₂ conversion.^61,205–208 In this section, (i) the characteristics of CO₂ electrolyzers using GDEs and (ii) the importance of interface control in this system are discussed.

6. Conclusions and perspectives

Modulating the catalyst–electrolyte interface has emerged as a promising strategy for increasing the efficiency of the electrochemical CO₂RR. This strategy can provide additional direction to overcome the limitations of the catalyst design, such as scaling relations and Sabatier volcano-like relations, that have been observed in simple metal catalyst surfaces. Indeed, the interface tuning significantly improved the FE and partial current density at relatively low overpotentials, which, thus far, has been accomplished by the design of catalyst structures and compositions (Fig. 9 and 10). The modification of the catalyst–electrolyte interfaces includes adding foreign molecules (or polymers), changing the electrolyte compositions and concentrations, and designing novel reactors and electrode structures, as well as the combination of these methods. These modification strategies of the catalyst–electrolyte interfaces significantly affect both the intrinsic and extrinsic catalytic activities.

The surface organic modulators adsorbed on the catalyst surface impact the interfacial chemistry by interacting with catalysts, reactants, and intermediates, which primarily affect the intrinsic activity of the active sites. Meanwhile, the organic additives on the catalyst surface also influence the extrinsic catalytic activities by modifying the mass transport of the protons or CO₂ molecules through the catalyst layer via chemical interactions. In the conventional and fundamental H-cell systems, as well as the CO₂ gas-fed electrolyzers, it was established that the surface treatment can effectively improve the current density and FE activity toward more practical levels. The review of recent studies provides insights into the possible functions of the surface additives, divided into four categories with respect to the interfacial chemistry (Fig. 4). Although the effectiveness of the molecular tuning strategy has been well demonstrated, the promoted electrocatalysis is yet to be fully understood. Therefore, more fundamental studies are required to exploit the effect of surface modulators further. However, there are great challenges associated with identifying the role of the surface modulators, because activity improvement originates from a complicated interplay of multiple factors (e.g. cysteamine can both change the electronic structure of the catalyst and provide a hydrogen bonding network), which are generally difficult to decouple. The elaborate design of molecules and advanced (operando) surface-sensitive spectroscopies will address the respective roles. Fortunately, there are many organic molecules available from databases. Using computational studies, researchers could conduct screening tests to identify the best organic modulator or elucidate the core factors by experimentally changing the atoms and functional groups of the organic modulators, which provides new research directions for optimizing the catalytic performance.

Interestingly, nature already utilizes this concept in many enzymes to effectively catalyse biological reactions. These enzymes are composed of an active metal centre (or cluster) and surrounding peptides. The organic components around the active centres support enzymatic reactions by binding reactants and facilitating the transfer of protons and electrons. For example, cytochrome c oxidase, a biological oxygen reduction catalyst in the respiratory chain, comprises a haem-Cu binuclear centre and an adjacent tyrosine moiety.²³³ The binuclear centre cooperatively adsorbs O₂ molecules while tyrosine plays a key role as a proton and electron transfer agent.²³⁴ The natural oxygen-evolving complex in photosystem II is composed of a Mn₄CaO₅ cluster and its surrounding amino acids.²³⁵ The amino acids facilitate the delivery of protons and electrons and the formation of hydrogen bonding networks with H₂O molecules.²³⁶ We envision that lessons from nature could aid the design of interfaces between heterogeneous catalysts and surface organic modulators by providing a more fundamental understanding of the synergistic processes.

The electrolyte compositions and concentrations are also highly crucial for determining the CO₂RR activity by tuning the electric double layer. The electric double layer structure affects not only the reaction environments (pH and CO₂ concentration) but also the adsorption strength and charge transfer kinetics. Despite extensive efforts to reveal the cation effect using computational methodologies, the function of anions (particularly for specifically adsorbed anions) remains elusive. In addition, the distribution of charged species is purported to be sensitive to the electronic structure of catalysts and applied potentials. A material-dependent study will provide profound insights into the electrolyte and double layer effects. However, due to the difficulty in identifying the double layer structure, experimental and computational breakthroughs are strongly pursued to characterize and describe the structure of the electric double layer, which will, in turn, afford further guidelines.

Since many dynamic equilibria, ion movements, and consumptions of H⁺ and CO₂ are involved at the interfaces during the CO₂RR, the in situ mass transport of reactants (CO₂, H⁺, H₂O) and ions is critical for improving the activity and selectivity. The regulation of the mass transport through meso-macroscopic structuring has particularly shown effectiveness in improving the selectivity and overpotential by generating local pH and boosting the local CO₂ mass transport. However, these strategies led to only a marginal increase in the current density of a few tens of mA cm⁻², which does not meet the requirement for practical applications. The fundamental limitation of the CO₂ supply through the slow CO₂ diffusion and low CO₂ concentration in the aqueous media has been successfully circumvented in the novel reactor design, where a distinct catalyst (solid)–electrolyte (liquid)–CO₂ (gas) three-phase interface is created compared to that of a conventional H-cell configuration. The application of a flow cell and MEA CO₂RR reactor could considerably increase the current density to commercially relevant levels (>1 A cm⁻²) by interface modifications.⁴¹

The control of the three-phase interface is the most important criterion for achieving the maximum activity in flow cell/MEA systems. This is because the proton (or H₂O in alkaline solution) and H₂O (as a steam in MEA) transports become significant as gaseous CO₂ is provided at a sufficient rate. Therefore, the performance of those devices is critically determined by extrinsic factors, such as the electrode thickness, hydrophobicity, and binder contents. In particular, controlling the hydrophobicity of the electrode in the CO₂ gas-fed electrolyzer is crucial which affects the gas–electrolyte–catalyst boundaries. More importantly, it is closely relevant to the flooding of the device, which causes the destruction of the three-phase boundary and a sudden decrease in performance. The balanced CO₂ flow and water control are other key factors involved in the three-phase boundary generation and destruction.

Since high current densities had been achieved with the CO₂ gas-fed electrolyzers, intensive technoeconomic analyses were conducted. The technoeconomic feasibility of CO₂RR is still a controversial issue because different results can be derived, depending on the assumptions, such as the electricity costs, electrocatalytic performance, and target chemicals. In general, low-level economic analyses considering electrolyzers and simplified models show more promising results than detailed process models, which include separation and recycling processes. Importantly, the technoeconomic analyses suggest the necessity for substantial improvement of the electrocatalytic performance (FE, current density, and overpotential) and the relative importance of each performance criterion: the current density has the greatest impact on improving the economic feasibility at the early stage. The FE and overpotential, which determine the separation and electricity costs, respectively, are the secondary major factors, which become more influential as the current density increases, suggesting the future research direction for practical application of CO₂RR technologies.

A catalyst–electrolyte interface development map in Fig. 18 displays the CO₂RR current density as a function of the overpotential.^{41,57,59,61,96,147,202,205,221,222,230,232} Regardless of the electrochemical reactor system, the modification of the catalyst–electrode interface has successfully increased the CO₂RR current density at relatively low overpotentials. The most remarkable improvement is achieved when the CO₂RR current density is increased by ∼100-fold, attributed to the adjustment of the three-phase boundary for any type of electrocatalyst. The technoeconomic analyses highlight that the FE has a high impact on the economic feasibility at a high current density. This correlation suggests the next goal in CO₂RR research: a huge improvement in the target product selectivity in CO₂ electrolyzers to reduce the separation costs. For these purposes, the catalyst–electrolyte interface requires systematic studies, which have been rarely conducted due to the complicated involvement of various interfacial factors. The surface modulators in the H-cells and the recent study on the effect of molecular additives in CO₂ electrolyzer suggest important perspectives and guidelines for enhancing the CO₂RR performance.¹⁴⁷

As shown in Fig. 18, despite the significant differences in the method of CO₂ supply and the electrode structure in the H-cells and CO₂ gas-fed electrolyzers, the high FEs of the reported catalysts demonstrated in H-cells have been well translated to the GDE reactors by interface tuning. It is highly probable that the interfacial chemistry knowledge gained from the H-cell studies, to some extent, can be also applied in the CO₂ gas-fed electrolyzers. In addition, the recent activity and selectivity enhancements in high-current reactors, achieved by a molecular functionalization strategy,¹⁴⁷ also suggest the presence of common interfacial chemistries in the H-cell and CO₂ gas-fed electrolyzers. Therefore, although the research on high-current reactors has practical importance, the importance of the H-cell study should not be undervalued as it can still provide a well-established foundation for the interface research. Furthermore, the difference in the interfacial processes between H-cells and GDE reactors needs to be clarified.

Lastly, in the CO₂ gas-fed electrolyzer research, Cu-based electrodes received significant research attention compared to CO₂-to-CO conversion catalysts because many valuable chemicals can be produced on Cu surfaces, although the product selectivity and overpotentials require further improvement. Several studies have demonstrated that CO₂-to-CO conversion catalysts, such as Ag or single-atom catalysts can produce CO with a high FE over 95% selectivity. However, the faradaic efficiency for single C₂₊ chemicals (i.e. C₂H₄ or C₂H₅OH) is below 80%, and high overpotentials are required for C–C coupling reactions on Cu-based electrodes. The generation of multiple products originates from the involvement of various reaction intermediates and mechanisms, such as C–C coupling, tautomerization, and isomerization. These surface processes can be tuned by the interfacial species to produce minor but economically feasible chemicals, such as acetaldehyde, acetone, acetic acid, and ethylene glycol, which have been identified as valuable targets over C₂H₄ or C₂H₅OH according to the technoeconomic analysis. This also provides the possibility of increasing the C₃₊ multicarbon selectivity by the interface modification, for instance, toward n-propanol, the most profitable product expected. Therefore, the study of the interface in commercially relevant reactors is encouraged to the increase FE or decrease overpotential for practical CO₂RR applications.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors appreciate the support from the Korea Institute of Science and Technology (KIST) institutional program and YU-KIST convergence program, and partially from the National Research Foundation (NRF) funded by the Ministry of Science and ICT, Korean government (No. 2019R1A2C2005521, NRF-2020R1C1C1006766, NRF-2020R1C1C1010963) and “Next Generation Carbon Upcycling Project” (Project No. 2020M1A2A6079141).

References

1. Global carbon dioxide growth in 2018 reached 4th highest on record|National Oceanic and Atmospheric Administration, https://www.noaa.gov/news/global-carbon-dioxide-growth-in-2018-reached-4th-highest-on-record, accessed 18 December 2019.
2. M. R. Allen, D. J. Frame, C. Huntingford, C. D. Jones, J. A. Lowe, M. Meinshausen and N. Meinshausen, Nature, 2009, 458, 1163–1166 CrossRef CAS PubMed .
3. K. Caldeira and M. E. Wickett, Nature, 2003, 425, 365 CrossRef CAS PubMed .
4. H. Baumann, S. C. Talmage and C. J. Gobler, Nat. Clim. Change, 2011, 2, 38–41 CrossRef .
5. M. Hulme, Nat. Clim. Change, 2016, 6, 222–224 CrossRef .
6. Sources of Greenhouse Gas Emissions|Greenhouse Gas (GHG) Emissions|US EPA, https://www.epa.gov/ghgemissions/sources-greenhouse-gas-emissions, accessed 18 December 2019.
7. D. T. Whipple and P. J. A. Kenis, J. Phys. Chem. Lett., 2010, 1, 3451–3458 CAS .
8. C. Xia, P. Zhu, Q. Jiang, Y. Pan, W. Liang, E. Stavitsk, H. N. Alshareef and H. Wang, Nat. Energy, 2019, 4, 776–785 CrossRef CAS .
9. S. Hernández, M. A. Farkhondehfal, F. Sastre, M. Makkee, G. Saracco and N. Russo, Green Chem., 2017, 19, 2326–2346 RSC .
10. C. Shi, H. A. Hansen, A. C. Lausche and J. K. Nørskov, Phys. Chem. Chem. Phys., 2014, 16, 4720–4727 RSC .
11. Y. Hori, H. Wakebe, T. Tsukamoto and O. Koga, Electrochim. Acta, 1994, 39, 1833–1839 CrossRef CAS .
12. A. Bagger, W. Ju, A. S. Varela, P. Strasser and J. Rossmeisl, ChemPhysChem, 2017, 18, 3266–3273 CrossRef CAS PubMed .
13. Y. Li and Q. Sun, Adv. Energy Mater., 2016, 6, 1600463 CrossRef .
14. L. Zhang, Z.-J. Zhao and J. Gong, Angew. Chem., Int. Ed., 2017, 56, 11326–11353 CrossRef CAS PubMed .
15. Q. Lu, J. Rosen and F. Jiao, ChemCatChem, 2015, 7, 38–47 CrossRef CAS .
16. K. D. Yang, C. W. Lee, J. H. Jang, T. R. Ha and K. T. Nam, Nanotechnology, 2017, 28, 352001 CrossRef PubMed .
17. A. Verdaguer-Casadevall, C. W. Li, T. P. Johansson, S. B. Scott, J. T. McKeown, M. Kumar, I. E. L. Stephens, M. W. Kanan and I. Chorkendorff, J. Am. Chem. Soc., 2015, 137, 9808–9811 CrossRef CAS PubMed .
18. H.-E. Lee, K. D. Yang, S. M. Yoon, H.-Y. Ahn, Y. Y. Lee, H. Chang, D. H. Jeong, Y.-S. Lee, M. Y. Kim and K. T. Nam, ACS Nano, 2015, 9, 8384–8393 CrossRef CAS PubMed .
19. B. Kumar, V. Atla, J. P. Brian, S. Kumari, T. Q. Nguyen, M. Sunkara and J. M. Spurgeon, Angew. Chem., Int. Ed., 2017, 56, 3645–3649 CrossRef CAS PubMed .
20. J. He, N. J. J. Johnson, A. Huang and C. P. Berlinguette, ChemSusChem, 2018, 11, 48–57 CrossRef CAS PubMed .
21. C. W. Lee, K. D. Yang, D.-H. Nam, J. H. Jang, N. H. Cho, S. W. Im and K. T. Nam, Adv. Mater., 2018, 30, 1704717 CrossRef PubMed .
22. Y. Lum and J. W. Ager, Angew. Chem., Int. Ed., 2018, 57, 551–554 CrossRef CAS PubMed .
23. C. W. Lee, N. H. Cho, S. W. Im, M. S. Jee, Y. J. Hwang, B. K. Min and K. T. Nam, J. Mater. Chem. A, 2018, 6, 14043–14057 RSC .
24. R. Kas, O. Ayemoba, N. J. Firet, J. Middelkoop, W. A. Smith and A. Cuesta, ChemPhysChem, 2019, 20, 2904–2925 CrossRef CAS PubMed .
25. M. R. Singh, J. D. Goodpaster, A. Z. Weber, M. Head-Gordon and A. T. Bell, Proc. Natl. Acad. Sci. U. S. A., 2017, 114, E8812–E8821 CrossRef CAS PubMed .
26. M. G. Kibria, J. P. Edwards, C. M. Gabardo, C.-T. Dinh, A. Seifitokaldani, D. Sinton and E. H. Sargent, Adv. Mater., 2019, 31, 1807166 CrossRef PubMed .
27. S. Ringe, E. L. Clark, J. Resasco, A. Walton, B. Seger, A. T. Bell and K. Chan, Energy Environ. Sci., 2019, 12, 3001–3014 RSC .
28. H. K. Ju, G. Kaur, A. P. Kulkarni and S. Giddey, J. CO₂ Util., 2019, 32, 178–186 CrossRef CAS .
29. J.-B. Vennekoetter, R. Sengpiel and M. Wessling, Chem. Eng. J., 2019, 364, 89–101 CrossRef CAS .
30. O. S. Bushuyev, P. D. Luna, C. T. Dinh, L. Tao, G. Saur, J. van de Lagemaat, S. O. Kelley and E. H. Sargent, Joule, 2018, 2, 825–832 CrossRef CAS .
31. Q. Lu and F. Jiao, Nano Energy, 2016, 29, 439–456 CrossRef CAS .
32. C. Palmer, F. Saadi and E. W. McFarland, ACS Sustainable Chem. Eng., 2018, 6, 7003–7009 CrossRef CAS .
33. S. Verma, B. Kim, H.-R. M. Jhong, S. Ma and P. J. A. Kenis, ChemSusChem, 2016, 9, 1972–1979 CrossRef CAS PubMed .
34. P. D. Luna, C. Hahn, D. Higgins, S. A. Jaffer, T. F. Jaramillo and E. H. Sargent, Science, 2019, 364, eaav3506 CrossRef PubMed .
35. M. Jouny, W. Luc and F. Jiao, Ind. Eng. Chem. Res., 2018, 57, 2165–2177 CAS .
36. S. Y. Chae, S. Y. Lee, S. G. Han, H. Kim, J. Ko, S. Park, O.-S. Joo, D. Kim, Y. Kang, U. Lee, Y. J. Hwang and B. K. Min, Sustainable Energy Fuels, 2020, 4, 199–212 RSC .
37. J. M. Spurgeon and B. Kumar, Energy Environ. Sci., 2018, 11, 1536–1551 RSC .
38. J. A. Herron, J. Kim, A. A. Upadhye, G. W. Huber and C. T. Maravelias, Energy Environ. Sci., 2015, 8, 126–157 CAS .
39. M. Rumayor, A. Dominguez-Ramos and P. P. A. Irabien, J. CO₂ Util., 2019, 34, 490–499 CrossRef CAS .
40. J. Na, B. Seo, J. Kim, C. W. Lee, H. Lee, Y. J. Hwang, B. K. Min, D. K. Lee, H.-S. Oh and U. Lee, Nat. Commun., 2019, 10, 5193 CrossRef PubMed .
41. F. P. García de Arquer, G.-T. Dinh, A. Ozden, J. Wicks, C. McCallum, A. R. Kirmani, D.-H. Nam, C. Garbardo, A. Seifitokaldani, X. Wang, Y. C. Li, F. Li, J. Edwards, L. J. Richter, S. J. Thorpe, D. Sinton and E. H. Sargent, Science, 2020, 367, 661–666 CrossRef PubMed .
42. M. R. Singh, E. L. Clark and A. T. Bell, Phys. Chem. Chem. Phys., 2015, 17, 18924–18936 RSC .
43. A. Wuttig, M. Yaguchi, K. Motobayashi, M. Osawa and Y. Surendranath, Proc. Natl. Acad. Sci. U. S. A., 2016, 113, E4585–E4593 CrossRef CAS PubMed .
44. B. A. Zhang, T. Ozel, J. S. Elias, C. Costentin and D. G. Nocera, ACS Cent. Sci., 2019, 5, 1097–1105 CrossRef CAS PubMed .
45. M. Dunwell, Q. Lu, J. M. Heyes, J. Rosen, J. G. Chen, Y. Yan, F. Jiao and B. Xu, J. Am. Chem. Soc., 2017, 139, 3774–3783 CrossRef CAS PubMed .
46. C. W. Lee, N. H. Cho, K. T. Nam, Y. J. Hwang and B. K. Min, Nat. Commun., 2019, 10, 3919 CrossRef PubMed .
47. C. W. Lee, J. S. Hong, K. D. Yang, K. Jin, J. H. Lee, H.-Y. Ahn, H. Seo, N.-E. Sung and K. T. Nam, ACS Catal., 2018, 8, 931–937 CrossRef CAS .
48. X. Zheng, P. D. Luna, F. P. García de Arquer, B. Zhang, N. Becknell, M. B. Ross, Y. Li, M. N. Banis, Y. Li, M. Liu, O. Voznyy, C. T. Dinh, T. Zhuang, P. Stadler, Y. Cui, X. Du, P. Yang and E. H. Sargent, Joule, 2017, 1, 794–805 CrossRef CAS .
49. S. Gao, Y. Lin, X. Jiao, Y. Sun, Q. Luo, W. Zhang, D. Li, J. Yang and Y. Xie, Nature, 2016, 529, 68–71 CrossRef CAS PubMed .
50. Y. Chen, C. W. Li and M. W. Kanan, J. Am. Chem. Soc., 2012, 134, 19969–19972 CrossRef CAS PubMed .
51. M. Liu, Y. Pang, B. Zhang, P. D. Luna, O. Voznyy, J. Xu, X. Zheng, C. T. Dinh, F. Fan, C. Cao, F. P. García de Arquer, T. S. Safaei, A. Mepham, A. Klinkova, E. Kumacheva, T. Filleter, D. Sinton, S. O. Kelley and E. H. Sargent, Nature, 2016, 537, 382–386 CrossRef CAS PubMed .
52. K. P. Kuhl, E. R. Cave, D. N. Abram and T. F. Jaramillo, Energy Environ. Sci., 2012, 5, 7050–7059 RSC .
53. X. Liu, H. Yang, J. He, H. Liu, L. Song, L. Li and J. Luo, Small, 2018, 14, 1704049 CrossRef PubMed .
54. Y. Wang, Z. Chen, P. Han, Y. Du, Z. Gu, X. Xu and G. Zheng, ACS Catal., 2018, 8, 7113–7119 CrossRef CAS .
55. S. Mou, T. Wu, J. Xie, Y. Zhang, L. Ji, H. Huang, T. Wang, Y. Luo, X. Xiong, B. Tang and X. Sun, Adv. Mater., 2019, 31, 1903499 CrossRef PubMed .
56. C. W. Lee, S.-J. Shin, H. Jung, D. L. T. Nguyen, S. Y. Lee, W. H. Lee, D. H. Won, M. G. Kim, H.-S. Oh, T. Jang, H. Kim, B. K. Min and Y. J. Hwang, ACS Energy Lett., 2019, 4, 2241–2248 CrossRef CAS .
57. H. Jung, S. Y. Lee, C. W. Lee, M. K. Cho, D. H. Won, C. Kim, H.-S. Oh, B. K. Min and Y. J. Hwang, J. Am. Chem. Soc., 2019, 141, 4624–4633 CrossRef CAS PubMed .
58. Z.-Q. Liang, T.-T. Zhuang, A. Seifitokaldani, J. Li, C.-W. Huang, C.-S. Tan, Y. Li, P. D. Luna, C. T. Dinh, Y. Hu, Q. Xiao, P.-L. Hsieh, Y. Wang, F. Li, R. Quintero-Bermudez, Y. Zhou, P. Chen, Y. Pang, S.-C. Lo, L.-J. Chen, H. Tan, Z. Xu, S. Zhao, D. Sinton and E. H. Sargent, Nat. Commun., 2018, 9, 3828 CrossRef PubMed .
59. C.-T. Dinh, T. Burdyny, M. G. Kibria, A. Seifitokaldani, C. M. Gabardo, F. P. García de Arquer, A. Kiani, J. P. Edwards, P. D. Luna, O. S. Bushuyev, C. Zou, R. Quintero-Bermudez, Y. Pang, D. Sinton and E. H. Sargent, Science, 2018, 360, 783–787 CrossRef CAS .
60. D. Ren, Y. Deng, A. D. Handoko, C. S. Chen, S. Malkhandi and B. S. Yeo, ACS Catal., 2015, 5, 2814–2821 CrossRef CAS .
61. T. T. H. Hoang, S. Verma, S. Ma, T. T. Fister, J. Timoshenko, A. I. Frenkel, P. J. A. Kenis and A. A. Gewirth, J. Am. Chem. Soc., 2018, 140, 5791–5797 CrossRef CAS PubMed .
62. Y. C. Li, Z. Wang, T. Yuan, D.-H. Nam, M. Luo, J. Wicks, B. Chen, J. Li, F. Li, F. P. García de Arquer, Y. Wang, C.-T. Dinh, O. Voznyy, D. Sinton and E. H. Sargent, J. Am. Chem. Soc., 2019, 141, 8584–8591 CrossRef CAS PubMed .
63. D. Ren, B. S.-H. Ang and B. S. Yeo, ACS Catal., 2016, 6, 8239–8247 CrossRef CAS .
64. Y. Liu, Y. Zhang, K. Cheng, X. Quan, X. Fan, Y. Su, S. Chen, H. Zhao, Y. Zhang, H. Yu and M. R. Hoffmann, Angew. Chem., Int. Ed., 2017, 56, 15607–15611 CrossRef CAS PubMed .
65. Y. Liu, S. Chen, X. Quan and H. Yu, J. Am. Chem. Soc., 2015, 137, 11631–11636 CrossRef CAS PubMed .
66. W. Luc, X. Fu, J. Shi, J.-J. Lv, M. Jouny, B. H. Ko, Y. Xu, Q. Tu, X. Hu, J. Wu, Q. Yue, Y. Liu, F. Jiao and Y. Kang, Nat. Catal., 2019, 2, 423–430 CrossRef CAS .
67. T.-T. Zhuang, Y. Pang, Z.-Q. Liang, Z. Wang, Y. Li, C.-S. Tan, J. Li, C. T. Dinh, P. D. Luna, P.-L. Hsieh, T. Burdyny, H.-H. Li, M. Liu, Y. Wang, F. Li, A. Proppe, A. Johnston, D.-H. Nam, Z.-Y. Wu, Y.-R. Zheng, A. H. Ip, H. Tan, L.-J. Chen, S.-H. Yu, S. O. Kelley, D. Sinton and E. H. Sargent, Nat. Catal., 2018, 1, 946–951 CrossRef CAS .
68. D. Ren, N. T. Wong, A. D. Handoko, Y. Huang and B. S. Yeo, J. Phys. Chem. Lett., 2016, 7, 20–24 CrossRef CAS PubMed .
69. Y. Pang, J. Li, Z. Wang, C.-S. Tan, P.-L. Hsieh, T.-T. Zhuang, Z.-Q. Liang, C. Zou, X. Wang, P. D. Luna, J. P. Edwards, Y. Xu, F. Li, C.-T. Dinh, M. Zhong, Y. Lou, D. Wu, L.-J. Chen, E. H. Sargent and D. Sinton, Nat. Catal., 2019, 2, 251–258 CrossRef CAS .
70. J. Li, F. Che, Y. Pang, C. Zou, J. Y. Howe, T. Burdyny, J. P. Edwards, Y. Wang, F. Li, Z. Wang, P. D. Luna, C.-T. Dinh, T.-T. Zhuang, M. I. Saidaminov, S. Cheng, T. Wu, Y. Z. Finfrock, L. Ma, S.-H. Hsieh, Y.-S. Liu, G. A. Botton, W.-F. Pong, X. Du, J. Guo, T.-K. Sham, E. H. Sargent and D. Sinton, Nat. Commun., 2018, 9, 4614 CrossRef PubMed .
71. K. U. D. Calvinho, A. B. Laursen, K. M. K. Yap, T. A. Goetjen, S. Hwang, N. Murali, B. Mejia-Sosa, A. Lubarski, K. M. Teeluck, E. S. Hall, E. Garfunkel, M. Greenblatt and G. C. Dismukes, Energy Environ. Sci., 2018, 11, 2550–2559 RSC .
72. D. Esrafilzadeh, A. Zavabeti, R. Jalili, P. Atkin, J. Choi, B. J. Carey, R. Brkljača, A. P. O’Mullane, M. D. Dickey, D. L. Officer, D. R. MacFarlane, T. Daeneke and K. Kalantar-Zadeh, Nat. Commun., 2019, 10, 865 CrossRef PubMed .
73. J. S. Yoo, R. Christensen, T. Vegge, J. K. Nørskov and F. Studt, ChemSusChem, 2016, 9, 358–363 CrossRef CAS PubMed .
74. S. Nitopi, E. Bertheussen, S. B. Scott, X. Liu, A. K. Engstfeld, S. Horch, B. Seger, I. E. L. Stephens, K. Chan, C. Hahn, J. K. Nørskov, T. F. Jaramillo and I. Chorkendorff, Chem. Rev., 2019, 119, 7610–7672 CrossRef CAS PubMed .
75. C. W. Lee, N. H. Cho, K. D. Yang and K. T. Nam, ChemElectroChem, 2017, 4, 2130–2136 CrossRef CAS .
76. T. Cheng, H. Xiao and W. A. Goddard, Proc. Natl. Acad. Sci. U. S. A., 2017, 114, 1795–1800 CrossRef CAS PubMed .
77. H. A. Hansen, J. B. Varley, A. A. Peterson and J. K. Nørskov, J. Phys. Chem. Lett., 2013, 4, 388–392 CrossRef CAS PubMed .
78. C. S. Chen, J. H. Wan and B. S. Yeo, J. Phys. Chem. C, 2015, 119(48), 26875–26882 CrossRef CAS .
79. A. D. Handoko, K. W. Chan and B. S. Yeo, ACS Energy Lett., 2017, 2, 2103–2109 CrossRef CAS .
80. S. Hanselman, M. T. M. Koper and F. Calle-Vallejo, ACS Energy Lett., 2018, 3, 1062–1067 CrossRef CAS .
81. A. J. Garza, A. T. Bell and M. Head-Gordon, ACS Catal., 2018, 8, 1490–1499 CrossRef CAS .
82. A. A. Peterson and J. K. Nørskov, J. Phys. Chem. Lett., 2012, 3, 251–258 CrossRef CAS .
83. M. T. M. Koper, J. Electroanal. Chem., 2011, 660, 254–260 CrossRef CAS .
84. X. Liu, J. Xiao, H. Peng, X. Hong, K. Chan and J. K. Nørskov, Nat. Commun., 2017, 8, 15438 CrossRef CAS PubMed .
85. T. Burdyny and W. A. Smith, Energy Environ. Sci., 2019, 12, 1442–1453 RSC .
86. A. R. Zeradjanin, J.-P. Grote, G. Polymeros and K. J. J. Mayrhofer, Electroanalysis, 2016, 28, 2256–2269 CrossRef CAS .
87. K. Chan, C. Tsai, H. A. Hansen and J. K. Nørskov, ChemCatChem, 2014, 6, 1899–1905 CrossRef CAS .
88. S. Back, M. S. Yeom and Y. Jung, ACS Catal., 2015, 5, 5089–5096 CrossRef CAS .
89. C. Dong, J. Fu, H. Liu, T. Ling, J. Yang, S. Z. Qiao and X.-W. Du, J. Mater. Chem. A, 2017, 5, 7184–7190 RSC .
90. H. Shin, Y. Ha and H. Kim, J. Phys. Chem. Lett., 2016, 7, 4124–4129 CrossRef CAS PubMed .
91. S. A. Akhade, I. T. McCrum and M. J. Janik, J. Electrochem. Soc., 2016, 163, F477–F484 CrossRef CAS .
92. D. Bohra, I. Ledezma-Yanez, G. Li, W. de Jong, E. A. Pidko and W. A. Smith, Angew. Chem., Int. Ed., 2019, 58, 1345–1349 CrossRef CAS PubMed .
93. E. R. Cave, C. Shi, K. P. Kuhl, T. Hatsukade, D. N. Abram, C. Hahn, K. Chan and T. F. Jaramillo, ACS Catal., 2018, 8, 3035–3040 CrossRef CAS .
94. L. D. Chen, M. Urushihara, K. Chan and J. K. Nørskov, ACS Catal., 2016, 6, 7133–7139 CrossRef CAS .
95. J. Resasco, L. D. Chen, E. Clark, C. Tsai, C. Hahn, T. F. Jaramillo, K. Chan and A. T. Bell, J. Am. Chem. Soc., 2017, 139, 11277–11287 CrossRef CAS PubMed .
96. C. Kim, H. S. Jeon, T. Eom, M. S. Jee, H. Kim, C. M. Friend, B. K. Min and Y. J. Hwang, J. Am. Chem. Soc., 2015, 137, 13844–13850 CrossRef CAS PubMed .
97. C. Kim, T. Eom, M. S. Jee, H. Jung, H. Kim, B. K. Min and Y. J. Hwang, ACS Catal., 2017, 7, 779–785 CrossRef CAS .
98. A. T. Landers, M. Fields, D. A. Torelli, J. Xiao, T. R. Hellstern, S. A. Francis, C. Tsai, J. Kibsgaard, N. S. Lewis, K. Chan, C. Hahn and T. F. Jaramillo, ACS Energy Lett., 2018, 3, 1450–1457 CrossRef CAS .
99. H. Xiao, W. A. Goddard, T. Cheng and Y. Liu, Proc. Natl. Acad. Sci. U. S. A., 2017, 114, 6685–6688 CrossRef CAS PubMed .
100. X. Liu, P. Schlexer, J. Xiao, Y. Ji, L. Wang, R. B. Sandberg, M. Tang, K. S. Brown, H. Peng, S. Ringe, C. Hahn, T. F. Jaramillo, J. K. Nørskov and K. Chan, Nat. Commun., 2019, 10, 32 CrossRef CAS PubMed .
101. A. S. Hall, Y. Yoon, A. Wuttig and Y. Surendranath, J. Am. Chem. Soc., 2015, 137, 14834–14837 CrossRef CAS PubMed .
102. Y. Yoon, A. S. Hall and Y. Surendranath, Angew. Chem., Int. Ed., 2016, 55, 15282–15286 CrossRef CAS PubMed .
103. J. T. Feaster, C. Shi, E. R. Cave, T. Hatsukade, D. N. Abram, K. P. Kuhl, C. Hahn, J. K. Nørskov and T. F. Jaramillo, ACS Catal., 2017, 7, 4822–4827 CrossRef CAS .
104. Y.-J. Zhang, V. Sethuraman, R. Michalsky and A. A. Peterson, ACS Catal., 2014, 4, 3742–3748 CrossRef CAS .
105. D. Gao, H. Zhou, F. Cai, D. Wang, Y. Hu, B. Jiang, W.-B. Cai, X. Chen, R. Si, F. Yang, S. Miao, J. Wang, G. Wang and X. Bao, Nano Res., 2017, 10, 2181–2191 CrossRef CAS .
106. D. Ren, J. Fong and B. S. Yeo, Nat. Commun., 2018, 9, 925 CrossRef PubMed .
107. H. Liu, Y. Zhu, J. Ma, Z. Zhang and W. Hu, Adv. Funct. Mater., 2020, 30, 1910534 CrossRef CAS .
108. S. Back, J.-H. Kim, Y.-T. Kim and Y. Jung, ACS Appl. Mater. Interfaces, 2016, 8, 23022–23027 CrossRef CAS PubMed .
109. Z. Sun, T. Ma, H. Tao, Q. Fan and B. Han, Chem, 2017, 3, 560–587 CAS .
110. D. Sun, X. Xu, Y. Qin, S. P. Jiang and Z. Shao, ChemSusChem, 2020, 13, 39–58 CrossRef CAS PubMed .
111. A. Vasileff, C. Xu, Y. Jiao, Y. Zheng and S.-Z. Qiao, Chem, 2018, 4, 1809–1831 CAS .
112. M. Zhong, K. Tran, Y. Min, C. Wang, Z. Wang, C.-T. Dinh, P. D. Luna, Z. Yu, A. S. Rasouli, P. Brodersen, S. Sun, O. Voznyy, C.-S. Tan, M. Askerka, F. Che, M. Liu, A. Seifitokaldani, Y. Pang, S.-C. Lo, A. Ip, Z. Ulissi and E. H. Sargent, Nature, 2020, 581, 178–183 CrossRef CAS PubMed .
113. H.-K. Lim, H. Shin, W. A. Goddard, Y. J. Hwang, B. K. Min and H. Kim, J. Am. Chem. Soc., 2014, 136, 11355–11361 CrossRef CAS PubMed .
114. Y. Zhao, C. Wang, Y. Liu, D. R. MacFarlane and G. G. Wallace, Adv. Energy Mater., 2018, 8, 1801400 CrossRef .
115. J. Schneider, H. Jia, J. T. Muckerman and E. Fujita, Chem. Soc. Rev., 2012, 41, 2036–2051 RSC .
116. Z. Cao, D. Kim, D. Hong, Y. Yu, J. Xu, S. Lin, X. Wen, E. M. Nichols, K. Jeong, J. A. Reimer, P. Yang and C. J. Chang, J. Am. Chem. Soc., 2016, 138, 8120–8125 CrossRef CAS PubMed .
117. Z. Cao, J. S. Derrick, J. Xu, R. Gao, M. Gong, E. M. Nichols, P. T. Smith, X. Liu, X. Wen, C. Copéret and C. J. Chang, Angew. Chem., Int. Ed., 2018, 57, 4981–4985 CrossRef CAS PubMed .
118. E. Andrews, S. Katla, C. Kumar, M. Patterson, P. Sprunger and J. Flake, J. Electrochem. Soc., 2015, 162, F1373–F1378 CrossRef CAS .
119. J. H. Lee, S. Kattel, Z. Xie, B. M. Tackett, J. Wang, C.-J. Liu and J. G. Chen, Adv. Funct. Mater., 2018, 28, 1804762 CrossRef .
120. S. Zhang, P. Kang, S. Ubnoske, M. K. Brennaman, N. Song, R. L. House, J. T. Glass and T. J. Meyer, J. Am. Chem. Soc., 2014, 136, 7845–7848 CrossRef CAS PubMed .
121. Z. Wang, L. Wu, K. Sun, T. Chen, Z. Jiang, T. Cheng and W. A. Goddard, J. Phys. Chem. Lett., 2018, 9, 3057–3061 CrossRef CAS PubMed .
122. M. S. Xie, B. Y. Xia, Y. Li, Y. Yan, Y. Yang, Q. Sun, S. H. Chan, A. Fisher and X. Wang, Energy Environ. Sci., 2016, 9, 1687–1695 RSC .
123. Y. Qiu, H. Zhong, W. Xu, T. Zhang, X. Li and H. Zhang, J. Mater. Chem. A, 2019, 7, 5453–5462 RSC .
124. S. Ahn, K. Klyukin, R. J. Wakeham, J. A. Rudd, A. R. Lewis, S. Alexander, F. Carla, V. Alexandrov and E. Andreoli, ACS Catal., 2018, 8, 4132–4142 CrossRef CAS .
125. Z. Cao, S. B. Zacate, X. Sun, J. Liu, E. M. Hale, W. P. Carson, S. B. Tyndall, J. Xu, X. Liu, X. Liu, C. Song, J. Luo, M.-J. Cheng, X. Wen and W. Liu, Angew. Chem., Int. Ed., 2018, 57, 12675–12679 CrossRef CAS PubMed .
126. K. J. P. Schouten, E. P. Gallent and M. T. M. Koper, J. Electroanal. Chem., 2014, 716, 53–57 CrossRef CAS .
127. H. Xiao, T. Cheng, W. A. Goddard and R. Sundararaman, J. Am. Chem. Soc., 2016, 138, 483–486 CrossRef CAS PubMed .
128. J. D. Goodpaster, A. T. Bell and M. Head-Gordon, J. Phys. Chem. Lett., 2016, 7, 1471–1477 CrossRef CAS PubMed .
129. J. Durst, A. Siebel, C. Simon, F. Hasché, J. Herranz and H. A. Gasteiger, Energy Environ. Sci., 2014, 7, 2255–2260 RSC .
130. J. Resasco, Y. Lum, E. Clark, J. Z. Zeledon and A. Bell, ChemElectroChem, 2018, 5, 1064–1072 CrossRef CAS .
131. G. Seshadri, C. Lin and A. B. Bocarsly, J. Electroanal. Chem., 1994, 372, 145–150 CrossRef CAS .
132. E. B. Cole, P. S. Lakkaraju, D. M. Rampulla, A. J. Morris, E. Abelev and A. B. Bocarsly, J. Am. Chem. Soc., 2010, 132, 11539–11551 CrossRef .
133. Y. Fang and J. C. Flake, J. Am. Chem. Soc., 2017, 139, 3399–3405 CrossRef CAS PubMed .
134. A. K. Buckley, M. Lee, T. Cheng, R. V. Kazantsev, D. M. Larson, W. A. Goddard, F. D. Toste and F. M. Toma, J. Am. Chem. Soc., 2019, 141, 7355–7364 CrossRef CAS PubMed .
135. S. Ponnurangam, C. M. Yun and I. V. Chernyshova, ChemElectroChem, 2016, 3, 74–82 CrossRef CAS .
136. Y. Fang, X. Cheng, J. C. Flake and Y. Xu, Catal. Sci. Technol., 2019, 9, 2689–2701 RSC .
137. Z. Han, R. Kortlever, H.-Y. Chen, J. C. Peters and T. Agapie, ACS Cent. Sci., 2017, 3, 853–859 CrossRef CAS PubMed .
138. A. Thevenon, A. Rosas-Hernández, J. C. Peters and T. Agapie, Angew. Chem., Int. Ed., 2019, 58, 16952–16958 CrossRef CAS PubMed .
139. J. T. Feaster, A. L. Jongerius, X. Liu, M. Urushihara, S. A. Nitopi, C. Hahn, K. Chan, J. K. Nørskov and T. F. Jaramillo, Langmuir, 2017, 33, 9464–9471 CrossRef CAS PubMed .
140. J. Huang, N. Hörmann, E. Oveisi, A. Loiudice, G. L. D. Gregorio, O. Andreussi, N. Marzari and R. Buonsantti, Nat. Commun., 2018, 9, 3117 CrossRef PubMed .
141. P. Grosse, D. Gao, F. Scholten, I. Sinev, H. Mistry and B. R. Cuenya, Angew. Chem., Int. Ed., 2018, 57, 6192–6197 CrossRef CAS PubMed .
142. D. Kim, C. S. Kley, Y. Li and P. Yang, Proc. Natl. Acad. Sci. U. S. A., 2017, 114, 10560–10565 CrossRef CAS .
143. C. C. L. McCrory, S. Jung, I. M. Ferrer, S. M. Chatman, J. C. Peters and T. F. Jaramillo, J. Am. Chem. Soc., 2015, 137, 4347–4357 CrossRef CAS PubMed .
144. E. L. Clark, J. Resasco, A. Landers, J. Lin, L.-T. Chung, A. Walton, C. Hahn, T. F. Jaramillo and A. T. Bell, ACS Catal., 2018, 8, 6560–6570 CrossRef CAS .
145. Y. Liu, S. Bliznakov and N. Dimitrov, J. Phys. Chem. C, 2009, 113, 12362–12372 CrossRef CAS .
146. F. Li, Y. C. Li, Z. Wang, J. Li, D.-H. Nam, Y. Lum, M. Luo, X. Wang, A. Ozden, S.-F. Hung, B. Chen, Y. Wang, J. Wicks, Y. Xu, Y. Li, C. M. Gabardo, C.-T. Dinh, Y. Wang, T.-T. Zhuang, D. Sinton and E. H. Sargent, Nat. Catal., 2020, 3, 75–82 CrossRef CAS .
147. F. Li, A. Thevenon, A. Rosas-Hernández, Z. Wang, Y. Li, C. M. Gabardo, A. Ozden, C. T. Dinh, J. Li, Y. Wang, J. P. Edwards, Y. Xu, C. McCallum, L. Tao, Z.-Q. Liang, M. Luo, X. Wang, H. Li, C. P. O’Brien, C.-S. Tan, D.-H. Nam, R. Quintero-Bermudez, T.-T. Zhuang, Y. C. Li, Z. Han, R. D. Britt, D. Sinton, T. Agapie, J. C. Peters and E. H. Sargent, Nature, 2020, 577, 509–513 CrossRef CAS PubMed .
148. R. M. Arán-Ais, D. Gao and B. R. Cuenya, Acc. Chem. Res., 2018, 51, 2906–2917 CrossRef PubMed .
149. D. Gao, R. M. Arán-Ais, H. S. Jeon and B. R. Cuenya, Nat. Catal., 2019, 2, 198–210 CrossRef CAS .
150. Y. Hori, A. Murata and R. Takahashi, J. Chem. Soc., Faraday Trans. 1, 1989, 85, 2309–2326 RSC .
151. S. Zhu, B. Jiang, W.-B. Cai and M. Shao, J. Am. Chem. Soc., 2017, 139, 15664–15667 CrossRef CAS PubMed .
152. A. Wuttig, Y. Yoon, J. Ryu and Y. Surendranath, J. Am. Chem. Soc., 2017, 139, 17109–17113 CrossRef CAS PubMed .
153. A. Murata and Y. Hori, Bull. Chem. Soc. Jpn., 1991, 64, 123–127 CrossRef CAS .
154. M. R. Thorson, K. I. Siil and P. J. A. Kenis, J. Electrochem. Soc., 2013, 160, F69–F74 CrossRef CAS .
155. D. Gao, I. T. McCrum, S. Deo, Y.-W. Choi, F. Scholten, W. Wan, J. G. Chen, M. J. Janik and B. R. Cuenya, ACS Catal., 2018, 8, 10012–10020 CrossRef CAS .
156. M. R. Singh, Y. Kwon, Y. Lum, J. W. Ager and A. T. Bell, J. Am. Chem. Soc., 2016, 138, 13006–13012 CrossRef CAS PubMed .
157. O. Ayemoba and A. Cuesta, ACS Appl. Mater. Interfaces, 2017, 9, 27377–27382 CrossRef CAS PubMed .
158. A. Schizodimou and G. Kyriacou, Electrochim. Acta, 2012, 78, 171–176 CrossRef CAS .
159. Y. Hori, H. Konishi, T. Futamura, A. Murata, O. Koga, H. Sakurai and K. Oguma, Electrochim. Acta, 2005, 50, 5354–5369 CrossRef CAS .
160. A. Wuttig and Y. Surendranath, ACS Catal., 2015, 5, 4479–4484 CrossRef CAS .
161. D. H. Won, H. Shin, M. W. Chung, H. Jung, K. H. Chae, H.-S. Oh, Y. J. Hwang and B. K. Min, Appl. Catal., B, 2019, 258, 117961 CrossRef CAS .
162. Y. Kwon, Y. Lum, E. L. Clark, J. W. Ager and A. T. Bell, ChemElectroChem, 2016, 3, 1012–1019 CrossRef CAS .
163. H. Yano, T. Tanaka and N. K. Ogura, J. Electroanal. Chem., 2004, 565, 287–293 CrossRef CAS .
164. K. Ogura, J. R. Ferrell, A. V. Cugini, E. S. Smotkin and M. D. Salazar-Villalpando, Electrochim. Acta, 2010, 56, 381–386 CrossRef CAS .
165. A. S. Varela, W. Ju, T. Reier and P. Strasser, ACS Catal., 2016, 6, 2136–2144 CrossRef CAS .
166. D. Gao, F. Scholten and B. R. Cuenya, ACS Catal., 2017, 7, 5112–5120 CrossRef CAS .
167. Y. Huang, C. W. Ong and B. S. Yeo, ChemSusChem, 2018, 11, 3299–3306 CrossRef CAS PubMed .
168. D. Gao, I. Sinev, F. Scholten, R. M. Arán-Ais, N. J. Divins, K. Kvashnina, J. Timoshenko and B. R. Cuenya, Angew. Chem., Int. Ed., 2019, 58, 17047–17053 CrossRef CAS PubMed .
169. I. T. McCrum, S. A. Akhade and M. J. Janik, Electrochim. Acta, 2015, 173, 302–309 CrossRef CAS .
170. F. S. Roberts, K. P. Kuhl and A. Nilsson, Angew. Chem., Int. Ed., 2015, 54, 5179–5182 CrossRef CAS PubMed .
171. S. Lee, D. Kim and J. Lee, Angew. Chem., Int. Ed., 2015, 54, 14701–14705 CrossRef CAS PubMed .
172. Y.-C. Hsieh, S. D. Senanayake, Y. Zhang, W. Xu and D. E. Polyansky, ACS Catal., 2015, 5, 5349–5356 CrossRef CAS .
173. D. L. T. Nguyen, M. S. Jee, D. H. Won, H.-S. Oh, B. K. Min and Y. J. Hwang, Catal. Commun., 2018, 114, 109–113 CrossRef CAS .
174. Y.-C. Hsieh, L. E. Betancourt, S. D. Senanayake, E. Hu, Y. Zhang, W. Xu and D. E. Polyansky, ACS Appl. Energy Mater., 2019, 2, 102–109 CrossRef CAS .
175. Y. Zhang, L. Liu, L. Shi, T. Yang, D. Niu, S. Hu and X. Zhang, Electrochim. Acta, 2019, 313, 561–569 CrossRef CAS .
176. H. Q. Fu, L. Zhang, L. R. Zheng, P. F. Liu, H. Zhao and H. G. Yang, J. Mater. Chem. A, 2019, 7, 12420–12425 RSC .
177. M. Zhao, H. Tang, Q. Yang, Y. Gu, H. Zhu, S. Yan and Z. Zou, ACS Appl. Mater. Interfaces, 2020, 12, 4565–4571 CrossRef CAS PubMed .
178. M. Cho, J. T. Song, S. Back, Y. Jung and J. Oh, ACS Catal., 2018, 8, 1178–1185 CrossRef CAS .
179. X. Duan, J. Xu, Z. Wei, J. Ma, S. Guo, S. Wang, H. Liu and S. Dou, Adv. Mater., 2017, 29, 1701784 CrossRef PubMed .
180. N. Ikemiya, K. Natsui, K. Nakata and Y. Einaga, RSC Adv., 2017, 7, 22510–22514 RSC .
181. M. Tomisaki, K. Natsui, N. Ikemiya, K. Nakata and Y. Einaga, ChemistrySelect, 2018, 3, 10209–20213 CrossRef CAS .
182. M. Tomisaki, S. Kasahara, K. Natsui, N. Ikemiya and Y. Einaga, J. Am. Chem. Soc., 2019, 141, 7414–7420 CrossRef CAS PubMed .
183. P. Lobaccaro, M. R. Singh, E. L. Clark, Y. Kwon, A. T. Bell and J. W. Ager, Phys. Chem. Chem. Phys., 2016, 18, 26777–26785 RSC .
184. M. Azuma, K. Hashimoto, M. Hiramoto, M. Watanabe and T. Sakata, J. Electroanal. Chem., 1989, 260, 441–445 CrossRef CAS .
185. A. Kudo, S. Nakagawa, A. Tsuneto and T. Sakata, J. Electrochem. Soc., 1993, 140, 1541–1545 CrossRef CAS .
186. K. Hara, A. Tsuneto, A. Kudo and T. Sakata, J. Electrochem. Soc., 1994, 141, 2097–2103 CrossRef CAS .
187. N. Gupta, M. Gattrell and B. MacDougall, J. Appl. Electrochem., 2006, 36, 161–172 CrossRef CAS .
188. J. Ryu, A. Wuttig and Y. Surendranath, Angew. Chem., Int. Ed., 2018, 57, 9300–9304 CrossRef CAS PubMed .
189. D. Raciti, M. Mao and C. Wang, Nanotechnology, 2018, 29, 044001 CrossRef PubMed .
190. K. Yang, R. Kas and W. A. Smith, J. Am. Chem. Soc., 2019, 141, 15891–15900 CrossRef CAS PubMed .
191. M. Dunwell, X. Yang, B. P. Setzler, J. Anibal, Y. Yan and B. Xu, ACS Catal., 2018, 8, 3999–4008 CrossRef CAS .
192. Y. Pocker and D. W. Bjorkquist, J. Am. Chem. Soc., 1977, 99, 6537–6543 CrossRef CAS .
193. W. Luo, J. Zhang, M. Li and A. Züttel, ACS Catal., 2019, 9, 3783–3791 CrossRef CAS .
194. D. L. T. Nguyen, C. W. Lee, J. Na, M.-C. Kim, N. D. K. Tu, S. Y. Lee, Y. J. Sa, D. H. Won, H.-S. Oh, H. Kim, B. K. Kim, S. S. Han, U. Lee and Y. J. Hwang, ACS Catal., 2020, 10, 3222–3231 CrossRef CAS .
195. A. S. Varela, M. Kroschel, T. Reier and P. Strasser, Catal. Today, 2016, 260, 8–13 CrossRef CAS .
196. K. D. Yang, W. R. Ko, J. H. Lee, S. J. Kim, H. Lee, M. H. Lee and K. T. Nam, Angew. Chem., Int. Ed., 2017, 56, 796–800 CrossRef CAS PubMed .
197. H. Song, M. Im, J. T. Song, J.-A. Lim, B.-S. Kim, Y. Kwon, S. Ryu and J. Oh, Appl. Catal., B, 2018, 232, 391–396 CrossRef CAS .
198. M. Ma, K. Djanashvili and W. A. Smith, Angew. Chem., Int. Ed., 2016, 55, 6680–6684 CrossRef CAS PubMed .
199. A. Dutta, M. Rahaman, N. C. Luedi, M. Mohos and P. Broekmann, ACS Catal., 2016, 6, 3804–3814 CrossRef CAS .
200. H. Yano, F. Shirai, M. Nakayama and K. Ogura, J. Electroanal. Chem., 2002, 533, 113–118 CrossRef CAS .
201. F. Lei, W. Liu, Y. Sun, J. Xu, K. Liu, L. Liang, T. Yao, B. Pan, S. Wei and Y. Xie, Nat. Commun., 2016, 7, 12697 CrossRef CAS PubMed .
202. J. Li, G. Chen, Y. Zhu, Z. Liang, A. Pei, C.-L. Wu, H. Wang, H. R. Lee, K. Liu, S. Chu and Y. Cui, Nat. Catal., 2018, 1, 592–600 CrossRef CAS .
203. D. Wakerley, S. Lamaison, F. Ozanam, N. Menguy, D. Mercier, P. Marcus, M. Fontecave and V. Mougel, Nat. Mater., 2019, 18, 1222–1227 CrossRef CAS PubMed .
204. B. A. Rosen, A. Salehi-Khojin, M. R. Thorson, W. Zhu, D. T. Whipple, P. J. A. Kenis and R. I. Masel, Science, 2011, 334, 643–644 CrossRef CAS PubMed .
205. Y. C. Li, D. Zhou, Z. Yan, R. H. Gonçalves, D. A. Salvatore, C. P. Berlinguette and T. E. Mallouk, ACS Energy Lett., 2016, 1, 1149–1153 CrossRef CAS .
206. N. Gutiérrez-Guerra, L. Moreno-López, J. C. Serrano-Ruiz, J. L. Valverde and A. de Lucas-Consuegra, Appl. Catal., B, 2016, 188, 272–282 CrossRef .
207. S. Ma, R. Luo, J. I. Gold, A. Z. Yu, B. Kim and P. J. A. Kenis, J. Mater. Chem. A, 2016, 4, 8573–8578 RSC .
208. H. Yang, J. J. Kaczur, S. D. Sajjad and R. I. Masel, J. CO2 Util., 2017, 20, 208–217 CrossRef CAS .
209. D. M. Weekes, D. A. Salvatore, A. Reyes, A. Huang and C. P. Berlinguette, Acc. Chem. Res., 2018, 51, 910–918 CrossRef CAS PubMed .
210. G. Wang, J. Pan, S. P. Jiang and H. Yang, J. CO₂ Util., 2018, 23, 152–158 CrossRef CAS .
211. S. Z. Oener, S. Ardo and S. W. Boettcher, ACS Energy Lett., 2017, 2, 2625–2634 CrossRef CAS .
212. Y. Hori, H. Ito, K. Okano, K. Nagasu and S. Sato, Electrochim. Acta, 2003, 48, 2651–2657 CrossRef CAS .
213. B. Smitha, S. Sridhar and A. A. Khan, J. Membr. Sci., 2005, 259, 10–26 CrossRef CAS .
214. M. B. McDonald, S. Ardo, N. S. Lewis and M. S. Freund, ChemSusChem, 2014, 7, 3021–3027 CrossRef CAS PubMed .
215. R. S. Reiter, W. White and S. Ardo, J. Electrochem. Soc., 2016, 163, H3132–H3134 CrossRef CAS .
216. R. S. Jayashree, S. K. Yoon, F. R. Brushett, P. O. Lopez-Montesinos, D. Natarajan, L. J. Markoski and P. J. A. Kenis, J. Power Sources, 2010, 195, 3569–3578 CrossRef CAS .
217. D. T. Whipple, E. C. Finke and P. J. A. Kenis, Electrochem. Solid-State Lett., 2010, 13, D109–D111 CrossRef .
218. M. G. Kibria, C.-T. Dinh, A. Seifitokaldani, P. D. Luna, T. Burdyny, R. Quintero-Bermudez, M. B. Ross, O. S. Bushuyev, F. P. García de Arquer, P. Yang, D. Sinton and E. H. Sargent, Adv. Mater., 2018, 30, 1804867 CrossRef PubMed .
219. S. Ma, M. Sadakiyo, R. Luo, M. Heima, M. Yamauchi and P. J. A. Kenis, J. Power Sources, 2016, 301, 219–228 CrossRef CAS .
220. H.-R. Jhong, F. R. Brushett and P. J. A. Kenis, Adv. Energy Mater., 2013, 3, 589–599 CrossRef CAS .
221. S. Verma, X. Lu, S. Ma, R. I. Masel and P. J. A. Kenis, Phys. Chem. Chem. Phys., 2016, 18, 7075–7084 RSC .
222. T. T. H. Hoang, S. Ma, J. I. Gold, P. J. A. Kenis and A. A. Gewirth, ACS Catal., 2017, 7, 3313–3321 CrossRef CAS .
223. E. Gauthier, Q. Duan, T. Hellstern and J. Benziger, Fuel Cells, 2012, 12, 835–847 CrossRef CAS .
224. S. Yu, X. Li, J. Li, S. Liu, W. Lu, Z. Shao and B. Yi, Energy Convers. Manage., 2013, 76, 301–306 CrossRef CAS .
225. K. Wagner, P. Tiwari, G. F. Swiegers and G. G. Wallace, Energy Environ. Sci., 2018, 11, 172–184 RSC .
226. S. Verma, Y. Hamasaki, C. Kim, W. Huang, S. Lu, H.-R. M. Jhong, A. A. Gewirth, T. Fujigaya, N. Nakashima and P. J. A. Kenis, ACS Energy Lett., 2018, 3, 193–198 CrossRef CAS .
227. J. Park, H. Oh, T. Ha, Y. I. Lee and K. Min, Appl. Energy, 2015, 155, 866–880 CrossRef CAS .
228. C. M. Gabardo, C. P. O'Brien, J. P. Edwards, C. McCallum, Y. Xu, C.-T. Dinh, J. Li, E. H. Sargent and D. Sinton, Joule, 2019, 3, 2777–2791 CrossRef CAS .
229. Z. Yin, H. Peng, X. Wei, H. Zhou, J. Gong, M. Huai, L. Xiao, G. Wang, J. Lu and L. Zhuang, Energy Environ. Sci., 2019, 12, 2455–2462 RSC .
230. S. Ren, D. Joulié, D. Salvatore, K. Torbensen, M. Wang, M. Robert and C. P. Berlinguette, Science, 2019, 365, 367–369 CrossRef CAS PubMed .
231. M. Wang, K. Torbensen, D. Salvatore, S. Ren, D. Joulié, F. Dumoulin, D. Mendoza, B. Lassalle-Kaiser, U. Işci, C. P. Berlinguette and M. Robert, Nat. Commun., 2019, 10, 3602 CrossRef PubMed .
232. C. Delacourt, P. L. Ridgway, J. B. Kerr and J. Newman, J. Electrochem. Soc., 2008, 155, B42–B49 CrossRef CAS .
233. S. Iwata, C. Ostermeier, B. Ludwig and H. Michel, Nature, 1995, 376, 660–669 CrossRef CAS PubMed .
234. R. A. Capaldi, Annu. Rev. Biochem., 1990, 59, 569–596 CrossRef CAS PubMed .
235. Y. Umena, K. Kawakami, J.-R. Shen and N. Kamiya, Nature, 2011, 473, 55–60 CrossRef CAS PubMed .
236. H. Ishikita and E.-W. Knapp, Biophys. J., 2006, 90, 3886–3896 CrossRef CAS PubMed .

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: 10.1039/d0cs00030b

‡ These authors contributed to equally to this work.

This journal is © The Royal Society of Chemistry 2020