Strong effect-correlated electrochemical CO₂ reduction

Yu-Feng Tang ^a, Lin-Bo Liu ^a, Mulin Yu ^a, Shuo Liu ^a, Peng-Fei Sui ^b, Wei Sun ^a, Xian-Zhu Fu ^c, Jing-Li Luo ^bc and Subiao Liu *^a
aSchool of Minerals Processing and Bioengineering, Central South University, Changsha, Hunan 410083, China. E-mail: subiao@csu.edu.cn
^bDepartment of Chemical and Materials Engineering, University of Alberta, Edmonton, Alberta T6G 1H9, Canada
^cCollege of Materials Science and Engineering, Shenzhen University, Shenzhen, China

First published on 20th August 2024

---

---

Yu-Feng Tang

Yu-Feng Tang graduated from Southwest Petroleum University in June, 2022, with a bachelor's degree in Materials Science and Engineering. In September 2022, he joined Prof. Subiao Liu's group at Central South University with a research passion in electrochemical CO2 reduction.

Lin-Bo Liu

Lin-Bo Liu received his master's degree in Engineering at the School of Mineral Processing and Bioengineering, Central South University, as part of Prof. Subiao Liu's group in 2023 and then he joined Prof. Subiao Liu's group to pursue his PhD focused on water splitting at room temperature and water electrolysis at elevated temperatures.

Mulin Yu

Mulin Yu received his master's degree of Science from Central South University in 2020. Currently, he is a PhD candidate in Prof. Subiao Liu's group at the School of Minerals Processing and Bioengineering, Central South University. His research mainly focuses on specific architectures of metallic nanomaterials toward electrochemical CO2 reduction.

Shuo Liu

Shuo Liu received his master's degree in Engineering from Zhengzhou University and joined Prof. Subiao Liu's group as a PhD candidate in September, 2022. His research interest mainly concentrates on designing advanced perovskite oxides to drive CO2 electrolysis at elevated temperatures in solid oxide electrolysis cells.

Peng-Fei Sui

Peng-Fei Sui received his BS and MS degrees at China University of Petroleum (East China) in 2015 and 2018, respectively. Currently, he is working as a PhD candidate in Prof. Jing-Li Luo's group at the Department of Chemical and Materials Engineering, University of Alberta. His research focuses on electrochemical conversion of CO2 to value-added fuels.

Wei Sun

Wei Sun is the Dean of the Graduate School at Central South University and a Distinguished Changjiang Scholar of the Ministry of Education. He currently serves as the Secretary-General of the Chinese Mineral Processing Council. He has received 14 provincial and ministerial awards, including the Second Prize of the National Science and Technology Progress Award and the Hunan Provincial Natural Science Award. He holds over 100 authorized patents, has published more than 300 papers, and has authored four monographs.

Xian-Zhu Fu

Xian-Zhu Fu is currently Professor in the College of Materials Science and Engineering, Shenzhen University. He received his PhD in Chemistry from Xiamen University in 2007. Later, he joined the Department of Materials and Chemical Engineering at the University of Alberta, Canada, as a post-doctoral research fellow and Lawrence Berkeley National Lab as a visiting scholar. From 2012 to 2017, he worked at the Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences. His research interests focus on electrochemistry/electrocatalysts for energy materials and devices, electronic materials, and processes.

Jing-Li Luo

Jing-Li Luo is Distinguished Professor at Shenzhen University, China, Emeritus Professor at the University of Alberta, and Fellow of the Canadian Academy of Engineering. She obtained her PhD in Materials Science and Engineering from McMaster University, Canada, in 1992. She served as the Canadian Research Chair in Alternative Fuel Cells from 2004 to 2015. Her research focuses on fuel cells, energy storage research, clean energy technology, and corrosion control.

Subiao Liu

Subiao Liu is Professor at the School of Minerals Processing and Bioengineering, Central South University. His research mission is to develop novel classes of nanostructured materials for electrochemical CO2 reduction to high-value fuels at room temperature, alkane dehydrogenation to alkene in proton-conducting solid oxide fuel cells (SOFC), CO2 electrolysis in solid oxide electrolysis cells (SOEC) at elevated temperatures, and other energy conversion and storage techniques (e.g., oxygen evolution/reduction reaction and hydrogen evolution reaction).

---

1. Introduction

Due to deforestation and the overexploitation of fossil fuel with the industrial revolution, CO₂ concentration in the atmosphere has risen from 277 ppm to a record-high level of over 400 ppm today.^1,2 This value is projected to reach 600 ppm by 2100, significantly exceeding the certified safety threshold of 350 ppm and causing severe environmental issues such as global warming, desertification, and species extinction.^3,4 Thus, to overcome the greenhouse effect, substantial efforts have been devoted to developing different technologies for the non-conversion utilization and direct conversion of CO₂. As a result, various approaches (e.g., biological,⁵ thermochemical,⁶ electrochemical,⁷ and photochemical⁸ conversion) have been developed with varying levels of success for the electrosynthesis of high-value fuels/chemicals (e.g., CO, CH₄, HCOOH, C₂H₄ and C₂H₅OH) by employing CO₂ as a carbon source.⁹ In contradistinction, electrochemical CO₂ reduction (ECR) to carbon-neutral products has ignited particular interest since it is a process controlled by electrochemical potential at ambient temperature and pressure in modular, compact, flexible and scalable setups while enabling the possibility to achieve CO₂-neutral electrosynthesis by integrating renewable energy sources (e.g., solar, wind and tidal).¹⁰ However, the centrosymmetric linear CO₂ molecule is fortified with two robust CO double bonds, creating a high energy barrier for its activation; the subsequent proton-coupled multi-electron transfer steps make the overall reaction kinetics sluggish; and the concomitant hydrogen evolution reaction (HER) competes with the target product selectivity.^11,12 Moreover, CO₂ shows an extremely low solubility of ∼33 mmol L⁻¹ in aqueous media under the ambient conditions of ∼25 °C and 1 atm, posing a challenge in achieving industrial-scale current density (j) due to an insufficient supply of dissolved CO₂ near the electrode.^13,14 Accordingly, in pursuit of the ultimate goal of commercial ECR for the electrosynthesis of high-value fuels/chemicals, over the few past decades, a plethora of studies have focused on developing active, selective and stable electrocatalysts by optimizing electrolytes, devices, electrode assembly methods, and external environment factors to achieve an optimal balance among these components.

Given the importance and predominance of electrocatalysts, various avant-garde design strategies, including control of the size and shape of solid-state electrocatalysts to achieve an optimal exposure of ECR-preferred active sites and/or facets,^15–17 the coupling of different metal elements to design alloy or multi-metal center atomic electrocatalysts to yield an optimal electronic structure and electron distribution,^18,19 the incorporation of defects (e.g., vacancies, dislocations, and grain boundaries) to construct extra high-energy active sites on solid-state electrocatalysts,^20,21 interface construction to adjust interfacial electron interaction in solid-state electrocatalysts,^22,23 and ligand modification on solid-state electrocatalysts and altering the ligands in molecular electrocatalysts to tune the local electron transfer between active metal centers and ligands for optimal local electrostatics,^24–26 have been employed to augment the intrinsic catalytic activity and selectivity of various electrocatalysts. In addition, the derivative effects induced by the specific nanostructures of electrocatalysts on their ECR performances should not be overlooked. For example, the confinement effect to protect electrocatalysts from space mobility and/or being oxidized, stabilize the key intermediates, and limit the mass transfer,²⁷ the strain effect to influence the lattice atomic electron orbital overlap capable of changing the electronic structures of electrocatalysts and tuning their binding energies to various reaction intermediates,²⁸ and the field-induced effect to adjust the reaction pathways caused by locally enhanced fields (e.g., electric, thermal, and magnetic fields) in regions of high curvature.^29,30 In the meantime, different electrolytes [e.g., aqueous solution,³¹ organic solvent,^32,33 and ionic liquid (ILs)^34–37] and their associated modulations of cations,³⁸ anions,³⁹ and organic additives⁴⁰ have also been proven to crucially impact the ECR performances, such as the solute effect with different buffering capacities to tailor the local pH, the non-aqueous solvent effect (e.g., organic solvents and ILs) to change the CO₂ trapping and polarization abilities of different solvents,^34,41,42 solvents with different charge-shielding abilities to modulate the dipole moments and local electrostatics of molecular electrocatalysts,^43,44 and cation and anion effects to influence the local electric field at electric double layer (EDL), the potential mechanism for CO₂ activation and the electrode surface reconstruction.^45–47 Likewise, external environment factors (e.g., binders for ink preparation,^48,49 reaction pressure and temperature,^50,51 gas feed purity,⁵² and flow rate⁵³) often significantly affect the ECR performances, such as the ionomer effects with different functional groups, affecting the distribution of electrocatalysts on the electrodes and the enrichment/rejection of intermediates,⁵⁴ pressure and temperature effects, changing the physicochemical properties of materials and the solubility and diffusion of the reactants/products,⁵⁵ the gas impurity effect, inducing the surface reconstruction of the electrocatalysts and the competitive occupancy of their active sites,⁵⁶ and the flow rate effect of the gas and/or electrolyte, tuning the mass transfer and the coverage of intermediates at the electrode–electrolyte interface.⁵³

However, although numerous attempts have been made to improve the ECR performances from various perspectives, there is still a lack of a deep and systematic understanding on the precisely introduced effects induced by different electrocatalysts, electrolytes, and external environment factors, together with the underlying intrinsic relationships among various as-generated effects and ECR performances. According to a thorough survey of previous studies, it can be found that the existing reviews predominantly focused on either various design strategies of electrocatalysts or a specific effect, thereby failing to holistically present all aspects of this subject capable of grasping the essential interconnectedness and discrepancies amongst diverse effects. Therefore, this review aims to bridge this lacuna by highlighting this topic with a more extensive purview. More importantly, was embarks on a comparative analysis encompassing different effects that are strongly correlated with ECR, including material intrinsic effects (i.e., size and shape effect, composition effect, defect effect, interface effect, and ligand effect), structure-induced effects (i.e., confinement effect, strain effect, and field-induced effect), electrolyte effects (i.e., solute effect, solvent effect, cation effect, and anion effect), and environment effects (i.e., ionomer effect, pressure and temperature effect, gas impurity effect, and flow rate effect) with an emphasis on elaborating how these effects shape the ECR catalytic activity and selectivity. In addition, we present the challenges and prospects on how to accurately introduce these effects and how ECR can be adroitly harnessed to reach closer to its thermodynamic and kinetic limits. It is believed that this review will enrich the fundamental understanding of the myriad of effects induced by various design strategies toward ECR and furnish guiding principles for their practical applications.

2. Material intrinsic effects

2.1. Size and shape effect

The size and shape of materials directly determine the number of active sites in solid-state electrocatalysts toward ECR.^57,58 Despite the general consensus that smaller solid-state electrocatalysts often exhibit significantly better catalytic activity due to the presence of more exposed active sites, not all are favorable for ECR, resulting in a nonlinear relationship between catalytic activity and selectivity with a variation in size. Typically, Au nanoparticles (NPs) with varying sizes (i.e., 4, 6, 8, and 10 nm) were synthesized for ECR to CO and the Au NP was modeled as a truncated octahedron to establish the relationship between the density of active sites and NP diameter (Fig. 1a).⁵⁹ The results showed that the highest CO selectivity by the 8 nm Au NPs was attributed to the optimal proportion of edge sites, which was thermodynamically validated to benefit the stabilization of*COOH rather than the corner sites for HER. Beyond Au NPs, the size effects in other metal nanostructures, such as Ag,⁶⁰ Pd,⁶¹ Cu,⁶² Bi,¹⁵ In,⁶³ Sn,⁶⁴ and Zn,⁶⁵ have also been extensively explored. For instance, it was found that the overall catalytic activity significantly increased as the size of Cu NPs decreased, particularly below 5 nm, together with a remarkable decline in CH₄ and C₂H₄ selectivity due to the inhibition of the subsequent protonation induced by the rich low-coordination number sites (CN < 8) on the Cu NPs below 5 nm.⁶² Consequently, Rong et al. studied the size-dependent selectivity in Cu-based electrocatalysts with smaller sizes of <1.5 nm⁶⁶ and found that when CO₂ was used as the feed gas, the reduction in size significantly exacerbated the competitive HER, whereas when it was replaced with CO, the main product gradually shifted from C₂H₄ to CH₄ with a decrease in size. Further, density functional theory (DFT) calculations indicated that the low-coordination Cu sites strongly adsorbed *H and *CO₂, but weakly adsorbed *CO, which well explained the increased H₂ selectivity with a decrease in the size of the Cu-based electrocatalysts. However, upon directly using CO as the feed gas, the C–C coupling was thermodynamically unfavorable at the low-coordination Cu sites, given that the sufficient *CO and *H coverage facilitated the protonation of *CO to form CH₄. In contrast, a reduction in size resulted in an increase in the number of coordination defects in the surface atoms on In nanocrystals with different sizes,⁶³ which enhanced their affinity to all the intermediates (i.e., *COOH, *OCHO, and *H). Among them, the 15 nm In nanocrystal exhibited a considerably high HCOOH selectivity of over 95% at a lower η regardless of its overall catalytic activity, while HER was preferred at smaller sizes, thus affecting the HCOOH formation (Fig. 1b). Apparently, a smaller size of solid-state electrocatalysts does not necessarily equate to better ECR performances. Moreover, due to the varying binding energies for various reaction intermediates (e.g., *H, *COOH, and *OCHO), the optimal size to achieve the best ECR performance also differs across solid-state electrocatalysts with distinct reactive centers.

Besides size tuning, controlling the shape of various metal nanostructures (e.g., Au,⁶⁸ Ag,⁷⁰ Pd,⁷¹ Cu,⁷² Sn,⁷³ and Zn⁷⁴), as another efficacious means of tuning the ratio of exposed active sites and facets, has attracted particular attention. Persuasively, comparative studies were conducted on Au octahedra (OD), rhombic dodecahedra (RD), and truncated ditetragonal prisms (TDP) using scanning electrochemical cell microscopy (SECCM), as shown in Fig. 1c.⁶⁷ It was found that Au RD with (110) facets exhibited a higher j than Au OD with (111) facets and Au TDP with high-index (310) facets due to its lower coordination number (CN = 7) of atoms. It is worth noting that the undercoordinated atoms are indicated by an abundance of unpaired electrons, making them favorable as active sites for ECR. In addition, the edge sites on Au NPs have been previously demonstrated to be more conducive to the formation of *COOH, while the corner sites are prone to adsorbing *H. Therefore, nanowires (NWs), with an abundance of edge sites, are considered exemplary nanostructures for ECR. For instance, an enhanced CO faradaic efficiency (FE_CO) of over ∼94% was observed on ultrathin Au NWs with a diameter of 2 nm compared with Au NPs,⁶⁸ which was attributed to the significantly increased edge-to-corner site ratio on Au NWs relative to Au NPs (Fig. 1d). Aiming to gain insight into the shape-activity relationship, DFT calculations were performed in a pioneer study on various Ag facets.⁷⁰ It was observed that the Gibbs free energies (ΔGs) required for forming *COOH and *CO on the Ag(100) and edge sites were substantially lower than that on Ag(111), implying that accurate shape control can expose energetically favorable facets/sites and induce a desirable ECR performance. Inspired by this, triangle Ag nanoplates and Ag nanocubes (NCs) were also synthesized to maximize the exposure of edge sites and Ag(100) facets, as shown in Fig. 1e,^69,70 both of which exhibited an enhanced j and near unity FE_CO compared to similarly sized Ag NPs, well establishing a representative paradigm for deciphering the shape-activity relationships of various Ag nanostructures for ECR.

In the case of Cu-based electrocatalysts, the variation in product distribution induced by the exposure of different facets associated with various specific shapes becomes more elusive. Early studies on single-crystal Cu electrodes indicated that Cu crystal facets played a critical role in ECR, where Cu(111) favored the formation of CH₄, whereas Cu(100) and other higher-index facets [i.e., Cu(911), Cu(711), and Cu(511)] favored the formation of the C₂₊ product,^75,76 which inspired the synthesis of Cu-based electrocatalysts with various shapes. Typically, Roberts et al. first evaluated the ECR performance on Cu NCs fully encapsulated by (100) facets and found that Cu NCs exhibited preferential selectivity for C₂H₄ compared to polycrystalline Cu foils.⁷⁷ Later, Buonsanti and colleagues further studied the size effect on Cu NCs, concluding that among 24, 44, and 63 nm Cu NCs, the NCs with a side length of 44 nm exhibited the highest C₂H₄ FE of 41%, which is likely due to their highest (100)/(110) ratio.⁷⁸ Similarly, the size-dependent selectivity was further explored on Cu ODs wrapped by eight (111) facets,⁷⁹ and because of their largest (110)/(111) ratio, the Cu ODs with the smallest size of 75 nm exhibited the highest CH₄ FE of 55%.

Thus, it can be concluded that reducing the size of NPs is not necessarily equal to an increase in selectivity for the target product,^80,81 but precisely controlling their shape can effectively offset the uncontrollability of product selectivity upon decreasing the size of solid-state electrocatalysts, thus achieving a desirable FE for the target product at a high j. However, it is worth noting that the variation trend of the surface active site distribution caused by size tuning differs for solid-state electrocatalysts with various shapes. Therefore, to design high-performance electrocatalysts with the maximum exposure of active sites for ECR, both size and shape effects should be synergistically considered, probably with the help of big data tools such as machine learning.

2.2. Composition effect

Composition normally impacts the intrinsic catalytic reactivity of an electrocatalyst via the electron redistribution induced by the difference in electronegativity of various elements, which further tunes the binding energies toward various key intermediates (e.g., *COOH, *OCHO, *CO, and *H), consequently influencing the ECR performance.^82,83 Different from CO production on Ag and Au, Pd demonstrates a more complex product distribution for ECR (e.g., CO, HCOOH, and H₂).⁷¹ Moreover, the strong Pd–C bond acts as a double-edged sword, i.e., effectively reducing the ΔG for the formation of *COOH, while making the desorption of *CO difficult, which results in the so-called “CO poisoning”. Accordingly, Gao et al. modulated the electronic structure of Pd by introducing Ag to form Pd_1−xAg_x alloys,⁸⁴ and confirmed that the alloy with an Ag/Pd ratio of 1:3 (i.e., Pd_0.75Ag_0.25) exhibited the maximum FE_CO of ∼95.3%. The DFT calculation results in Fig. 2a suggested that overlapping between C 2p and Pd 3d existed in the *COOH configuration at a low Ag/Pd ratio of 1:3, similar to pure Pd, implying that strong binding to *COOH was maintained. Conversely, the overlapping between C 2p and Pd 3d in the *CO configuration was diminished, indicating a weakened binding energy to *CO. A gradual increase in the Ag ratio further weakened the binding energy to *CO and reduced the binding energy to *COOH. In the case of HCOOH production on Pd, Sn was introduced to form Pd_xSn_y alloys, and a near 100% FE_HCOOH was achieved at the optimal Pd/Sn ratio of 1:1,⁸⁵ where the oxophilic Sn induced the formation of PdSnO₂ species on the Pd surface, contributing to an enhanced binding energy to *OCHO and a thermodynamically favorable reaction pathway (Fig. 2b). In fact, insufficient Sn failed to attract enough O atoms to oxidize Pd, while excessive Sn resulted in the formation of rich SnO₂ on the surface, which prevented the oxidation of Pd (Fig. 2c). Likewise, benefiting from the enhanced surface oxophilicity by introducing Sn on the surface of Cu, the improved formation of C₂H₅OH on a bimetallic Cu–Sn electrocatalyst was obtained,⁸⁶ together with suppressed C₂H₄ selectivity compared to bare Cu. Specifically, *OHC_β–C_αH₂ served as the key intermediate in the bifurcation of C₂H₄ and C₂H₅OH evolution, where the protonation of C_α resulted in the production of *OHC_β–C_αH₃, which was further reduced to C₂H₅OH, whereas the protonation of C_β led to the formation of *OH₂C_β–C_αH₂, which produced C₂H₄. Moreover, the enhanced surface oxygenophilicity induced by Sn simultaneously reduced the positive charge on C_α and the negative charge on C_β, both favoring the protonation of C_α rather than C_β. Moreover, the introduction of different elements inevitably alters the lattice parameters of an electrocatalyst, which changes its surface atomic arrangement and microstructure, consequently redirecting the reaction pathway and selectivity.⁸⁷ For instance, Gunji et al. investigated the composition effect by introducing secondary metal elements with different sizes in Pd to form PdM (M = Zn, Cu, Sn, Ag) alloys,⁸⁸ among which PdZn and PdCu exhibited higher selectivity for HCOOH versus CO on PdSn and PdAg. In the case of metals with atomic radii smaller than Pd (e.g., Zn and Cu), the compressive strain on the Pd surface allowed the formation of *OCHO, whereas the surface tensile strain on PdSn and PdAg induced by metals with a larger atomic size inhibited the formation of *OCHO due to steric hindrance, leading to the formation of CO rather than HCOOH.

It is worth noting that when exploring the composition effects in alloy electrocatalysts, attention should not only be given to the effect of different ratios of elements on ECR performances but also the actual arrangement and distribution of the relevant elements. For instance, Ma et al. investigated a series of CuPd alloy electrocatalysts with ordered, disordered, and phase-separated atomic arrangements for ECR and found that the ordered CuPd favored C₁ (i.e., CO and CH₄) product formation with a C₂ (i.e., C₂H₄ and C₂H₅OH) FE of <5%, whereas the phase-separated CuPd favored C₂ product generation with an FE of ∼64%.⁹⁰ This marked difference was believed to be caused by the proximity of Cu atoms in the phase-separated structure, which facilitated the dimerization of adjacent adsorbed *CO into *COCOH intermediates and their further transformation into C₂ products. By contrast, CuPd mainly existed in intermetallic form in the ordered structure, and the *CO adsorbed on Cu atoms tended to form *CHO intermediates, resulting in the generation of CH₄. Similarly, Chen et al. suggested that Cu in the AgCu alloy exhibited high mobility, which could migrate to the surface, detach from the alloy electrocatalyst, and recrystallize as a new phase.⁹¹ Although the initially homogeneously mixed AgCu alloy was within the thermodynamic solubility limit, both Ag and Cu showed a trend to phase separate into Cu- and Ag-rich domains during ECR, implying that the Ag–Cu heterointerfaces were the true active centers for the enhanced C₂H₅OH product selectivity generally observed on AgCu alloy electrocatalysts.

Another prevalent approach for composition regulation is doping. Doping the anion sites of metal compounds can introduce additional charge carriers, and thus modulate the electronic structure of electrocatalysts. Typically, S doping has been suggested to be an efficient strategy to promote the formation of HCOOH.^92–95 For instance, the maximum HCOOH FE of 97.06% was observed on S-doped Bi₂O₃ nanosheets⁹² due to the fact that the introduction of S with a much lower electronegativity distorted the electron distribution of adjacent Bi, leading to the formation of numerous under-coordinated active Bi sites. Additionally, the charge delocalization induced by the electronic distortion further enhanced the electron density on these under-coordinated active Bi sites, thus strengthening the binding energy to *OCHO. Although Cu-based electrocatalysts generally favor the formation of C₂₊ products, S-doped Cu₂O (S-Cu₂O) was reported to display an HCOOH FE of up to 81.4%,⁹³ which is 7-fold higher than that of pure Cu₂O. A further study revealed that the in situ reconstruction of the S-covered Cu sites completely altered the ECR pathway on Cu(111), where the spontaneous electron transfer from Cu to S created a strong charge accumulation zone, creating an additional dipole force that exclusively enhanced the adsorption of *OCHO, while inhibiting the formation of *COOH, both ultimately resulting in enhanced HCOOH selectivity on S-Cu₂O. The highly electronegative halogens (i.e., F⁻, Cl⁻, Br⁻, and I⁻) are another class of elements widely used for doping, which can stabilize the positive valence states of active metal centers through strong electronic interactions. Typically, Ko et al. achieved a high HCOOH FE of 90% on F-doped SnO₂ electrocatalysts, which stabilized at a j of 100 mA cm⁻² for over 7 days.¹⁹ Compared to bare SnO₂, the introduction of F not only enhanced the interaction between *OCHO and the electrocatalyst surface, but also induced rapid electron transfer from Sn to F through the strong F–Sn bonds, which allowed Sn to maintain a positive valence state for extended periods at higher applied potentials and j, thus favoring the formation of HCOOH.

Besides anion doping, doping the cation sites of metal compounds holds potential to modulate the electronic environment of the active metal centers, thus affecting the bonding strength. To address the rapid deactivation of metal compounds at industrial js, Chi et al. doped Zn in In₂S₃ (ZnIn₂S₄) for ECR,⁸² which maintained an HCOOH FE of 99.3% for over 60 h at a j of 300 mA cm⁻². A series of in situ characterizations indicated that S leaching was the primary reason for the decreased stability of In₂S₃, which could be prevented by doping Zn, as further confirmed by DFT calculations. Specifically, the tetrahedral and OD sites in ZnIn₂S₄ were fully occupied after doping Zn compared to the common tetrahedral vacancies in In₂S₃, implying that the introduction of Zn led to a strong local covalent characteristic in ZnIn₂S₄, making bond breakage between In/Zn and S kinetically difficult. Moreover, Hu et al. demonstrated that the occupied and unoccupied orbitals near the Fermi level of ZnO could be modulated by substituting Zn sites with heteroatoms (Mo and Cu) possessing distinct outer shell electrons.⁸⁹ The newly formed unoccupied orbitals in Cu-doped ZnO (i.e., Cu–ZnO) contributed to the enhanced hybridization and bonding of adsorbates, which strongly enhanced the binding energies to both *H and *COOH. However, overly strong bonding was detrimental, where strong *H adsorption promoted the competing HER and tight *COOH adsorption created a high formation energy for the subsequent evolution from *COOH to *CO, resulting in poor CO selectivity on Cu–ZnO. In contrast, besides the unoccupied orbitals in Mo-doped ZnO (Mo–ZnO), two distinct new occupied orbitals lowered the d-band center of Mo–ZnO (Fig. 2d), which synergistically modulated the adsorption of *COOH for enhanced CO selectivity and weakened the adsorption of *H, suppressing HER.

As the size of NPs decreases to the atomic scale, i.e., single-atom electrocatalysts (SAECs), many elements display significantly different ECR performances compared to their corresponding bulk forms as a result of their unique electronic structures and coordination environments.⁹⁶ Normally, transition metals (e.g., Fe,⁹⁷ Co,⁹⁸ and Ni⁹⁹) are remarkably more active for HER rather than ECR, but their single-atom counterparts have been demonstrated to efficiently promote ECR to CO. Moreover, N-doped carbon (N–C) materials as supports are often employed to anchor atomically isolated metal atoms, usually with a coordination form of –N₄ to prevent atomic aggregation. To date, most SAECs for ECR are in the form of M–N–C, where the strong interaction between the central metal atom and the N–C framework leads to a residual charge on the central metal atom, thereby enhancing the intrinsic ECR activity. To explore the composition effect induced by different metal centers in SAECs, Wang et al. prepared three M–N–C SAECs (M = Fe, Co, and Ni, Fig. 3a)¹⁰⁰ and found that the ECR activity followed the order of Co–N–C > Ni–N–C > Fe–N–C, which is different from the selectivity trend with the sequence of Ni–N–C > Co–N–C > Fe–N–C. DFT calculations indicated that the rate-determining steps toward ECR at the Ni–N₄ and Fe–N₄ sites are *COOH formation and *CO desorption, respectively, while Co–N₄ exhibited a compromise in energy demand for both *COOH formation and *CO desorption, thus delivering the optimal ECR activity (Fig. 3b). Nevertheless, the largest thermodynamic potential difference between ECR and HER on the Ni–N₄ site well explained the best selectivity on Ni–N–C (Fig. 3c). Recently, attention has shifted toward constructing dual metal atom sites (e.g., Cu–Fe,¹⁰¹ Ni–Zn,¹⁰² and Ni–Co¹⁰³) capable of tuning the electronic structure of metal centers and breaking the conventional linear-scaling relationship of the intermediates, while maintaining the merits of SAECs. As mentioned earlier, Ni–N–C normally shows a higher energy barrier for *COOH formation, while Fe–N–C possesses a larger *CO desorption energy. To reach a balance, a dual-atom electrocatalyst (DAEC) with Ni and Fe sites (Ni/Fe–N–C) was synthesized,¹⁰⁴ which showed a turnover frequency (TOF) of 7682 h⁻¹, exceeding that of the individual Ni–N–C (3690 h⁻¹) and Fe–N–C (813 h⁻¹). Close inspection of the reaction pathway in Fig. 3d suggested that Ni/Fe–N–C behaved highly similar to Fe–N–C, where CO passivated the Fe–Ni site and the CO-adsorbed Ni/Fe–N lowered the desorption energy of *CO. Moreover, the distance of 2.5 Å between the Ni and Zn atoms in the Ni–Zn DAECs is in the range of the specified cutoff distance between individual Ni and Zn atoms (Fig. 3e),¹⁰² indicating the presence of strong electronic interaction in the atomically dispersed bimetallic sites. Consequently, this led to a synchronized increase in electron density at both the Ni and Zn sites (Fig. 3f), thus enhancing the binding energy to *CO₂ and the subsequent ECR process with the highest TOF over 1200 h⁻¹.

In addition, the diversity of coordinating elements is another crucial variable in regulating the local electronic environment in SAECs. Once the coordinating N atom is replaced by another element, the original structure distorts into an asymmetric pattern, and the induced change in the electronegativity of the coordinating atom leads to an electron redistribution capable of modulating the intrinsic activity of an electrocatalyst.¹⁰⁶ Compared to N, S and P have a larger diameter and lower electronegativity, which act as electron donors to the coordinating metal atoms.^105,107 For instance, Li et al. prepared an Fe–N/P–C electrocatalyst by partially replacing the N atom with P atom for ECR and achieved the maximum TOF of 509 h⁻¹, which is almost 2.5-fold higher than that on Fe–N–C.¹⁰⁵ Due to the weak electronegativity of P, the oxidation state of the Fe atom center could be stabilized (Fig. 3g), which further prevented it from being aggregated into Fe NPs during ECR. Moreover, the DFT results suggested that the P and N co-coordinated single-atom Fe site could transfer more electrons to drive ECR for the formation of CO.

Notably, uniformly distributed alloys may undergo rapid elemental segregation to form phase-separated structures during ECR,⁹¹ while parts of doped electrocatalysts also undergo in situ dissolution to construct heterogeneous structures,^108,109 suggesting that only relying on ex situ characterization techniques and theoretical simulations probably results in the misinterpretation of the actual composition effect. Thus, future studies should emphasize the importance of detecting the dynamic evolution of electrocatalysts to clarify the true presence and roles of each component for ECR, and consequently reveal the real composition effect to enable the customization of high-performance electrocatalysts.

2.3. Defect effect

Crystals normally possess periodic structures, and the emergence of defects (e.g., point defects,¹¹⁰ line defects,¹¹¹ and planar defects¹¹²) signifies the absence or deviation from this periodicity. This inevitably causes lattice distortion and/or an intricate atomic rearrangement, creating abundant electrocatalytically active sites on the solid-state electrocatalysts. Among them, the point defects pertain to the loss, substitution, or interstitial occupancy of atomic lattice points, resulting in localized destruction within the crystal structure. Vacancies, as a principal form of point defects, have been widely reported to promote the catalytic activity of electrocatalysts for ECR.^21,113 For instance, enhanced catalytic activity was observed by leaching Fe from Au-Fe alloy to create an abundance of surface Au vacancies, which lowered the ΔG for the formation of *COOH.¹¹⁴ Similarly, the potent stabilizing effect of *OCHO on In and Bi vacancies was reported to promote ECR for the formation of HCOOH.^115,116 In addition to the cation vacancies in pure metals, the effects of anion vacancies on ECR have also been extensively studied. Normally, oxygen vacancies with excessive electrons can be generated upon the loss of oxygen from a pristine metal oxide, where the excessive electrons are either being trapped within the oxygen vacancy or acquired by neighboring metal ions rather than being delocalized on the surface. Therefore, the reducible oxygen vacancy serves as an electron donor to activate the electron-deficient C atom in CO₂. For instance, Hu et al. introduced oxygen vacancies in SnO₂ through thermal annealing under a reducing atmosphere.¹¹⁷ Compared to the pristine SnO₂, SnO₂ with oxygen vacancies exhibited a narrowed bandgap and enabled an increase in number of charge carriers (Fig. 4a). Moreover, the increased carrier density, coupled with the reduced bandgap guaranteed swifter charge transfer within the oxygen vacancy-confined SnO₂, which favored the adsorption and activation of CO₂. To further decipher the interrelationship between oxygen vacancies and ECR performance, Huang et al. synthesized two InO_x nanorods (NRs) with different oxygen vacancy concentrations by simply calcining InO_x NRs under different atmospheres (i.e., air and H₂).¹¹⁸ DFT calculations suggested that the d-band center shifted upward with an increase in the oxygen vacancy concentration (Fig. 4b), which reduced the bonding states and increased the antibonding states, thus bolstering the binding energy for *OHCO for the formation of HCOOH. Furthermore, the defect effect can be further enhanced by integrating vacancy engineering and the doping strategy. For instance, Wang et al. developed Cu-doped CeO₂ for selective ECR to CH₄¹¹⁹ and observed that the strong interaction between CeO₂ and Cu resulted in well-dispersed Cu species in the form of single atoms, which aggregated ∼3 oxygen vacancies in the proximate position, giving rise to active sites for CH₄ production. Additionally, DFT calculations implied that the activation of CO₂ was difficult when one or two oxygen vacancies were clustered around the Cu sites (Fig. 4c), whereas a Cu configuration conjoined with three oxygen vacancies could effectively activate CO₂.

Besides point defects, grain boundaries (GBs), as planar defects formed by the disruption of periodicity in the arrangement of adjacent grains in polycrystalline materials, are also considered highly active sites in solid-state electrocatalysts.^122,123 Kanan et al. demonstrated that ECR to CO was more active at the GBs rather than the grain surface on an Au electrode using SECCM.¹²⁰ Specifically, the j on the GB was 2 to 2.5 times higher than that on the adjacent grains (Fig. 4d), which provided solid evidence for the high activity of GBs for ECR. In a step forward to elucidating the quantitative correlation between GBs and ECR performance, Feng et al. fabricated Au NPs on carbon nanotubes via vapor deposition, followed by annealing treatment at various temperatures to obtain different GB densities.¹²⁴ They found that the Au NP size was gradually enlarged upon increasing the temperature, causing a diminished surface GB density and associated decrease in both j and FE_CO, while further examination confirmed the linear relationship between CO partial j and surface GB density. It is noteworthy that the heightened catalytic activity induced by GB comes at the expense of structural stability, which is not conducive to practical ECR application. In seeking a balance between catalytic activity and structural stability, twin GBs, as a particular type of boundary interface common in face-centered cubic crystals (e.g., Au, Ag, and Cu), are usually more stable than high-angle GBs, thereby serving as a compromise in addressing the dichotomy between catalytic activity and structural stability.^20,121,125 Particularly, a Cu electrocatalyst with dense twin GBs was prepared for ECR, which exhibited a significantly enhanced CH₄ FE of 86.1%.¹²¹ DFT calculations revealed that compared to the planer Cu(111), twin GBs lowered the activation energy barrier (G_act) for *CO protonation, while maintaining a similar or even higher C–C coupling energy barrier (Fig. 4e), thus resulting in enhanced CH₄ selectivity. The prominent role of twin GBs in guiding the product selectivity of ECR was also evident in the study by Li et al. on five-fold twinned Cu NWs.¹²⁶ Although the Cu(100) facet has been claimed to favor C₂H₄ formation in Section 2.1, a CH₄ FE as high as 55% was observed on 20 nm Cu NWs completely wrapped by Cu(100) facets. Further experimental exploration revealed that the twin GBs on bare Cu NWs rapidly degraded during ECR, accompanied by a fast shift in product selectivity from CH₄ to C₂H₄, whereas wrapping the Cu NWs with reduced graphene oxide stabilized the twin GBs, which achieved long-term high selectivity for CH₄. Obviously, the introduction of various defects (e.g., vacancies, dislocations, and GBs) creates a substantial quantity of highly active sites in an inherently high-energy state, which are difficult to remain stable during ECR, and thus developing feasible strategies to delay or even prevent the degradation of defective sites during ECR should be highlighted in future research.

2.4. Interface effect

The interface represents the boundary between two spatial regions occupied by different matter or by matter in different physical states, which has been considered a form of unique planar defects in solid-state electrocatalysts and delivers enhanced catalytic activity similar to GBs. However, unlike GBs, the interface is constructed by two different crystal structures, which are regarded as the loading phase and support phase. More importantly, the interface can well disperse and stabilize the loading phase and tune the electronic structure of electrocatalyst via a strong interfacial electronic interaction.^21,127 A study on the metal–metal interface in Ag–Cu nanodimers (NDs) indicated that the Ag–Cu interface induced spontaneous electron transfer from Cu to Ag, which not only maintained the oxidative state of the Cu sites, but also enhanced the electron density of Ag sites, collectively improving the CO₂ adsorption and conversion abilities on the Ag sites toward *CO formation.¹²⁸ Due to the higher binding energy for *CO on the Cu sites, the accumulated *CO tended to flow to the adjacent Cu sites, and then just in time to promote the subsequent C–C coupling for C₂H₄ production (Fig. 5a). Besides promoting C₂H₄ formation, Luan et al. constructed an Ag–Cu interface on the specific Cu(111) facet to facilitate asymmetric C–C coupling for C₂H₅OH production with an FE of up to 56.5%¹²⁹ and claimed that *CH₂ was efficiently accumulated on Cu(111), followed by its asymmetric *CH₂–*CO coupling with *CO formed on Ag. Moreover, Chen et al. created abundant exposed Sn–Bi interfacial sites for HCOOH production,¹³⁰ and the calculated partial density of state (PDOS) results clarified that the addition of Bi caused the electronic state of Sn to increase from the Fermi level, indicating that electrons were transferred from Sn to Bi (Fig. 5b). Notably, the p-band center of the Sn–Bi interface model was located between that of pure Sn and the Sn–Bi alloy, showing the optimal oxidation state. Unlike pure Sn with a higher electron density and Sn in the SnBi alloy with fewer electrons, Sn in the Sn–Bi interface possessed a medium electron density, which failed to increase the adsorption of *COOH for a favored ECR to HCOOH by generating a stronger p–d interaction through π back-bonding or σ bonding (Fig. 5c).

Besides the metal–metal interface, the construction of a metal–metal oxide interface on a solid-state electrocatalyst has also attracted particular attention, and has been demonstrated to improve the ECR performance.^21,133,134 Representatively, Bao et al. observed that CO₂ was first adsorbed on the Ag–CeO_x interfacial boundaries rather than on Au or CeO_x,¹³⁵ and adsorption loops were gradually formed, intuitively proving that the Au–CeO_x interfaces were more favorable for CO₂ adsorption and activation. In addition, the electron distribution at the Ag–ZnO interface was beneficial to enhance the ECR performance,¹³⁶ where a decrease in electron density around the Zn and Ag atoms was achieved, together with an increase in electron density around the O atoms. This resulted in an obvious upward shift in the d-band center for Ag in Ag–ZnO, which enhanced the binding energy to *COOH and delivered high CO selectivity. More interestingly, a potential-dependent ECR to specific products throughout the potential window on the Ag–SnO_x interface was established in Fig. 5d, and an FE_CO of 85% at −0.6 V vs. reversible hydrogen electrode (V_RHE) and FE_HCOOH of 83% at −1.0 V_RHE were obtained.¹³¹ The potential-dependent selectivity was attributed to the modulation of the energy demand for the formation of intermediates toward different reaction pathways on the Ag–SnO_x interface. Specifically, the activation energy barrier of 1.46 eV for the HCOOH pathway was higher than that of 1.17 eV for the CO pathway. Therefore, CO dominated the main products at a low η, but the electrocatalyst preferentially generated HCOOH when the activation energy required for both pathways could be overcome at a high η. This was because the ΔG for the formation of *OCHO (−0.41 eV) was more favorable than that of *COOH (0.34 eV). Moreover, the most recent studies indicate that the true active sites for forming C₂₊ products on Cu-based electrocatalysts are the Cu^δ+/Cu⁰ interfacial sites.^108,137,138 For example, Zhang et al. reported a high C₂H₅OH FE of 53.5% on Cu₂O NCs with abundant “embossed” structural GBs.¹³⁷ The in situ spectroscopy and DFT calculations collectively suggested that the dual-phase Cu(I) and Cu(0) interfacial sites stabilized by the GBs concurrently promoted the adsorption of *CO and *OH, which favored the hydrogenation of the key *CHCOH intermediate to form *CHCHOH for the C₂H₅OH pathway, while inhibiting its dehydration to produce *CCH for the C₂H₄ pathway, thus increasing the C₂H₅OH selectivity. Furthermore, using near-ambient pressure scanning tunneling microscopy, Jensen et al. reported that a single Cu or CuO film was inert to ECR, but the Cu–CuO interfacial sites were observed to react immediately with CO₂ at near ambient pressure,¹³⁸ implying that the Cu–CuO interfacial sites were the real active sites on Cu-based electrocatalysts during ECR.

Likewise, modifying the metal oxide–metal oxide interface has been verified as an effective strategy to adjust and stabilize the oxidation states of active metal sites,¹³⁹ with strong emphasis on the intense electron transfer and/or electron redistribution, as exemplified on Sn-based oxides. Wu et al. studied the electron transfer at the SnO₂–Sn₃O₄ interface constructed by surface hydrothermal treatment for ECR,¹⁴⁰ where the intense electron redistribution induced by the electric field maintained a rich and stable supply of active Sn²⁺ species for the formation of HCOOH. In the meantime, Liu et al. proposed a dynamic CeO₂-mediated Sn⁰/Sn^δ+ redox cycle mechanism at SnO₂–CeO₂ interfaces for ECR (Fig. 5e).¹³² In more detail, a portion of Ce⁴⁺ was first reduced to Ce³⁺ under a negative potential to create rich oxygen vacancies during ECR and promote H₂O dissociation for the generation of active *OH and *H species. Specifically, *OH oxidized Sn⁰ to the highly active Sn^δ+, while *H assisted the conversion of *CO₂˙⁻ to the crucial *OCHO. Finally, the *OH species was reduced to OH⁻ and diffused into the bulk electrolyte and Sn^δ+ was reduced to Sn⁰ under a negative potential.

2.5. Ligand effect

In numerous studies on the synthesis of colloidal nanostructures, a common route for structural control is the introduction of surface ligands, which can stabilize the specific morphologies, while preventing the aggregation of nanostructures.^71,78 Generally, the presence of surface ligands on solid-state electrocatalysts is considered to have a detrimental effect on ECR, given that they may compete with CO₂ to occupy the surface active sites. However, a substantial number of recent studies indicated that surface ligands play dominant roles in shaping the local electrocatalytic environment in multiple ways.^141,142 It has been concluded that the surface ligands on solid-state electrocatalysts have an effect on ECR mainly based on three aspects, including tuning the surface electronic structure, stabilizing the key intermediates, and regulating the mass transfer. Remarkably, the electron-donating ability of surface ligands has been confirmed to crucially impact the electronic structure of the active centers. Chang et al. fabricated electron-rich N-heterocyclic carbenes (NHC) on Au NPs to enhance their intrinsic catalytic activity (Fig. 6a)¹⁴³ and observed that the surface NHC ligands enriched the electron density on the surface of Au NPs and promoted the first electron transfer to CO₂. A follow-up study extended the use of NHC on Pd foil and achieved increased catalytic activity and a remarkably enhanced selectivity for C₁ products (i.e., CO and HCOOH) by 32 fold compared to the bare foil.¹⁴⁴ Although the binding energies for the key intermediates are mainly influenced by the electronic structure, such as the position of the d-band center, the linear-scaling relationship between the binding energies to *CO and *COOH still poses a challenge for further improving the ECR performance. Kim and colleagues proposed a strategy to break this linear-scaling relationship, and initially suggested the use of p-band elements to induce partial covalent bond formation in metal electrocatalysts and further adjust their electronic structure based on theoretical calculations.¹⁴⁵ Among the various p-band elements, S and As exhibit outstanding effects in enhancing the ECR performances on Ag-based electrocatalysts. This concept was later confirmed by experiments.¹⁴⁶ In detail, Ag NPs were synthesized by a colloidal method in the presence of cysteamine molecules, which served as an anchoring agent and a surface regulator containing S. Upon introducing cysteamine, the disruption of symmetry caused changes in the spatial localization of unpaired electrons, especially on the superatomic surface of Ag, which imparted more covalent characteristics to the Ag–COOH bond, thus effectively stabilizing the key *COOH intermediate. However, the local unpaired electrons performed poorly in stabilizing the donor–acceptor bond between Ag and *CO, which favored the desorption of *CO from the electrocatalyst surface, thus breaking the linear-scaling relationship and enhancing the CO selectivity (Fig. 6b).

In addition, the functional groups on ligands often play a vital role in stabilizing the key intermediates.¹⁴⁹ In the aforementioned Ag-cysteamine system, Wang et al. used molecular dynamics simulation to explore the process of CO₂ molecules approaching the surface of the cysteamine-covered Ag NPs,¹⁴⁷ where a constraining force was applied to drive the adsorption of CO₂ and the ΔG for CO₂ activation (ΔG_act) was calculated as a function of the distance (Z) from the Ag surface. It was found that the ΔG_act reached the minimum value at Z = 3 Å on the clean Ag surface, suggesting that the CO₂ molecules only underwent weak physical adsorption. By contrast, the amine on cysteamine offered an additional H-bond to CO₂ on the cysteamine-covered Ag surface (Fig. 6c), which induced the minimum ΔG_act at a smaller Z of 2.12 Å, indicating the strong chemical adsorption of CO₂ on the cysteamine-covered Ag surface. Similar results have been further verified on Cu-based electrocatalysts.^150,151 Considering that ECR involves proton-coupled multi-electron transfer steps, the local protonation and product selectivity at the electrocatalyst (solid)–electrolyte (liquid) interface can be accurately controlled by introducing surface organic molecules with different affinities to the proton source (e.g., H₂O and HCO₃⁻).¹⁵² Typically, Buonsanti et al. used alkane chains to change the hydrophobicity of the Ag surface and concluded that the short-tail ligands failed to block H₂O molecules, predominately leading to H₂ formation, whereas the long-tail ligands induced a steric hindrance and a limited CO₂ transport.¹⁵³ Therefore, the optimal ECR performance with balanced permeability to H₂O and CO₂ was obtained on the surface of Ag with medium-length alkane chains. Similarly, Lin et al. modified the surface of Cu using alkanethiols with different alkyl chain lengths to study the effect of ligand-induced hydrophobicity on the ECR product selectivity.¹⁵⁴ The results showed that with an increase in the chain length (i.e., carbon atoms from 4 to 18), the HER was strongly inhibited, while the C₂H₅OH FE was first increased, and then decreased. Moreover, it has been demonstrated that gradually increasing the hydrophobicity resulted in an increase in the *CO/*H ratio, which was capable of enhancing the C₂H₅OH selectivity, whereas the subsequent decrease in C₂H₅OH selectivity was attributed to the excessive hydrophobicity, leading to the insufficient *H and limited proton transfer process for ECR to form C₂H₅OH. Building on this, Geng et al. employed Ag and azobenzene as a model electrocatalyst and a model surface ligand, respectively, proposing the possibility of directly using the functional groups in surface ligands as the proton source for ECR and a dynamic promotion mechanism (Fig. 6d).¹⁴⁸ Due to the short-range H-bonds (O–H–N), the proton in the electrophilic triazolium directly participated in the protonation of adsorbed CO₂ to form *COOH with a lower ΔG and better stabilization, where the deprotonated-triazolium returned to the triazolium by capturing a proton from H₂O, and again served as the proton source for ECR.

Nevertheless, the ligand effects in molecular electrocatalysts differ from that in solid-state electrocatalysts, which can be classified into two categories, i.e., inductive effect and through-space effect. The inductive effect usually refers to the electron redistribution benefiting from electronic structure tuning of a molecular electrocatalyst enabled by electron-withdrawing or electron-donating ligands.¹⁵⁵ To validate this, Azcarate et al. synthesized a series of FeTPP (TPP = meso-tetraphenyl porphyrin) derivatives substituted with varying degrees of perfluoro and o,o′-methoxy meso aryl groups.^156,157 According to the results, as the number of perfluoro groups increased, the electron-withdrawing inductive effect was gradually strengthened, which resulted in a decrease in the nucleophilicity of the Fe center (Fig. 7a), finally lowering the η and the maximum TOF. By contrast, the electron-donating substituents tended to increase the TOF but at the expense of an increase in η (Fig. 7b), suggesting that focusing only on the inductive effect is insufficient to reach a win–win situation for molecular electrocatalysts. Accordingly, creating specific interactions between certain ligands and central metals with a through-space effect can activate an alternative mechanistic pathway for ECR, which exhibits no conflicts between TOF and η as controlled by the linear-scaling relationship of the pure inductive effect.¹⁵⁸ Due to the specific spatial orientations of ligands, the ECR performances are strongly promoted by stabilizing the reaction intermediates through an intramolecular H-bond and/or increasing the local proton donor concentration. For example, after introducing eight –OH groups at the ortho position of the porphyrin phenyl in FeTPP, a significant enhancement in ECR activity was observed beyond the linear-scaling relationship.¹⁵⁹ It was believed that the –OH group acted both as an intramolecular proton relay to facilitate the CO₂ protonation and C–O bond cleavage steps and as an H-bond stabilizer for the Fe–CO₂ intermediate. Moreover, Han and colleagues reported that when N,N-di(2-picolyl)ethylenediamine (DPEN) units are attached to Fe porphyrins, the created polypyridine and/or amine sites can effectively capture H₂O molecules to form an H-bond network, facilitating the proton transfer process during ECR.¹⁶⁰ In addition, protonated and positively charged DPEN units were demonstrated to stabilize the negatively charged *CO₂ intermediates through electrostatic interaction and H-bonds. With one step forward, the effects of different proton donors (i.e., phenol, guanidyl, and sulfonic acid groups) on a dibenzofuran scaffold were studied in a series of Fe hangman porphyrins by Margarit and colleagues.¹⁶¹ Although high CO selectivity was observed for all the molecular electrocatalysts, their reaction rate considerably differed, following the order of phenol > guanidyl > sulfonic acid, which was strongly related to the H-bond between the terminal groups and *CO₂ (Fig. 7c). The theoretical calculations indicated that the easily deprotonatable sulfonic acid base (pK_a ≈3) cannot form an H-bond with the Fe–CO₂ intermediate and is not protonated by the PhOH ligand with a pK_a of ∼18. This resulted in strong coulombic repulsion between the negatively charged Fe–CO₂ intermediate and the deprotonated sulfonic acid base, leading to poor ECR activity for the sulfonic acid-substituted derivatives. Moreover, due to the more favorable H-bonding interaction of the guanidyl group with the porphyrin ligand than that with CO₂, lower ECR activity was observed for the guanidyl-modified derivatives as compared to that modified with phenol. Besides the intramolecular H-bond, enhanced stabilization of the negatively charged Fe–CO₂ intermediate could be also achieved, benefiting from the coulombic spatial interaction with the positively charged pendant groups in proximity to the active sites.¹⁶² Consequently, the reaction rate could be significantly increased by introducing 4 positively charged trimethylbenzene amine groups at the para position of the TPP phenyl ring,¹⁵⁶ whereas due to the unfavorable electrostatic effect, the opposite trend was observed when it was substituted with a negatively charged sulfonate group. Clearly, as shown in Fig. 7d, the “all-in-one” strategy by integrating different ligand effects into one molecular electrocatalyst is an effective approach to break the linear-scaling relationship between TOF and η.¹⁵⁸ Moreover, Guo et al. found that Fe porphyrins, when substituted with two positively charged trimethyl benzylamine groups, exhibited a better ECR performance compared to that replaced with only one group.¹⁶³ The theoretical insights indicated that the two positively charged groups separately but cooperatively worked for ECR instead of simply stacking-up, where one enhanced the CO₂ adsorption ability by stabilizing the negatively charged *CO₂, while the other interacted with the phenol molecule via electrostatic interaction.

3. Structure-induced effects

3.1. Confinement effect

Spatial confinement refers to the restriction of the active sites within a confined space, mainly benefiting ECR by limiting the space mobility of a solid-state electrocatalyst, stabilizing the key intermediates and changing their mass transfer. Extensive experimental evidence has demonstrated that metal nanostructures tend to aggregate into large NPs, leading to a decrease in the number of active sites and degradation in the stability of solid-state electrocatalysts for ECR.¹⁶⁴ Nevertheless, utilizing porous structures to confine the space displacement of nanostructures has been perceived as an effective approach to tackle this issue. It was mentioned in Section 2.1 that the product distribution on Cu clusters is strongly size-dependent based on theoretical simulations, indicating that Cu clusters with a smaller size are more favorable for CH₄ generation. However, the difficulty in synthesizing sub-nanometer Cu clusters and their severe agglomeration during ECR limit their further applications. In this case, Hu and colleagues prepared 1.0 nm Cu clusters with around 10 atoms being confined within a defect-rich microporous carbon structure,¹⁶⁵ which achieved a maximum CH₄ FE of 81.7%, which is consistent with previous DFT results. Recently, Zhu et al. confined SnCu_xO_2+x clusters in pure siliceous Mobil-5 zeolites (SnCu_xO_2+x@MFI) for an Li–CO₂ battery¹⁶⁶ and found that the accumulation of *CO intermediates in the channels of the zeolites was beneficial for the subsequent multi-step protonation process, which resulted in a high FE of 66.6% for CH₄ generation and superb long-term stability for over 100 cycles at a cutoff specific capacity of 1000 mA h g⁻¹. Similarly, Xie et al. synthesized ultrathin Sn quantum sheets confined within graphene to prevent their oxidation and agglomeration (Fig. 8a).¹⁶⁷ The thermogravimetric analysis displayed a rapid weight loss in air due to the decomposition of graphene from 390 °C to 570 °C (Fig. 8b), whereas a weight increase was observed above 570 °C, resulting from the oxidation of metallic Sn, demonstrating the protective role played by graphene to maintain the stability of the Sn quantum sheets, even in air up to 570 °C. In contrast, the reference 15 nm Sn NPs and their mixture with graphene both showed a weight increase above 200 °C, indicating the easy oxidation of metallic Sn without the graphene confinement.

In addition to stabilizing small-sized electrocatalysts during ECR, confined-space edge sites are thought to be able to stabilize the reaction intermediates through strong electronic interactions. It is commonly accepted that noble metals (e.g., Au and Ag) can effectively catalyze ECR to CO, but their high η limits their further application. Thus, to address this issue, Cheng et al., using DFT calculations, demonstrated that the confinement effect formed in the gap of ≤7 Å between adjacent Au or Ag NPs could significantly stabilize the key *COOH intermediate via a strong electronic interaction induced by the ends of the narrow gaps.¹⁷¹ This highlights the superiority of designing electrocatalysts with sub-nanometer gaps for ECR due to the intermediate stabilization effect caused by the confined space, but their accurate synthesis is challenging. Thus, to address this, novel SnO_x NPs with highly controlled sub-nanometer interfacial spacings of <10 Å were synthesized using a lithium electrochemical tuning method.¹⁶⁸ DFT calculations showed that the sub-nanometer-scale gaps significantly improved ECR over HER due to the strong stabilization of *COOH and *OCHO, particularly at a distance of 6–7 Å (Fig. 8c), whereas the rate-determining intermediate of *H for HER was independent of the gap distance. Moreover, Wang and colleagues proposed the construction of sub-nanometer gaps by in situ fragmenting colloidal Cu₂O superparticles for ECR.¹⁷² The experimental and computational results revealed that the sub-nanometer gaps contained coordination-unsaturated sites with a strong binding energy to *CO, which increased the local *CO concentration and reduced the formation energy for the *C₂ intermediates, thus delivering a high C₂H₄ FE of 53.2% compared to that of 21% on commercial Cu NPs.

It is well known that the main hurdle in obtaining multi-carbon products is the difficulty in effectively coupling C–C bonds on Cu-based electrocatalysts. In this case, the confinement effect holds potential to suppress the loss of carbon intermediates, thereby benefiting C–C coupling to C₂₊ products.^{169,170,173,174} To achieve this, Sargent et al. investigated the confinement effect of a cavity on the selectivity of multi-carbon products¹⁶⁹ and found that the cavity well confined the locally generated C₂ species (Fig. 8d), leading to an increase in the surface coverage and residence of the C₂ intermediates required for the generation of C₃. Ultimately, it resulted in a high C₃ production rate. Importantly, it was a volcano-type curve was observed for the C₃ product selectivity as the cavity opening angle changed, where that with a medium opening angle exhibited the highest C₃H₇OH FE of 21%. The further finite element simulations clarified that the production rates of C₂ and C₃ were both slow at smaller angles of <30° due to the excessive confinement limiting CO transportation into the cavity, while the cavity failed to retain the generated C₂ intermediates at larger angles of >180°. Only at intermediate angles between 45–90° the confinement effect ensured the retention of the C₂ intermediates without limiting the CO inflow, thus leading to a high C₃/C₂ selectivity and overall C₃ production rate (Fig. 8e). However, the single-layer Cu shell exposed large opened pores, leaving ample space and opportunities for the carbon intermediates to escape. Accordingly, Zeng et al. prepared a Cu electrocatalyst with a hollow multi-shell structure (HOMSs) and verified that owing to the increase in the diffusion barrier and the pathway length for outward moving carbon species, the catalytic activity and selectivity for C₂₊ products were significantly improved upon increasing the shell number.¹⁷⁰ However, although the HOMSs with one shell may locally confine the C₁ intermediates within their cavity, the desorbed C₁ species still had a high probability to escape as C₁ products. Thus, by increasing the shell number, the diffusion barrier was lifted and the pathway length was extended (Fig. 8f), which effectively inhibited the escape of the locally generated C₁ species and promoted the C–C coupling to C₂₊ products.

Besides regulating the mass transfer of intermediates, the confined space also functions to limit the transfer of other substances (e.g., H⁺, K⁺, OH⁻, and HCO₃⁻) in the electrolyte, thus influencing the EDL microenvironment and the final ECR performance. Recent studies have demonstrated that the specifically adsorbed *OH from the electrolyte and the surface *O residue from the lattice oxygen promote the C–C coupling via H-bonds.¹⁷⁵ However, due to the electrostatic repulsion, the adsorption of *OH/*O is quite weak, especially at negative potentials during ECR. Accordingly, Pan et al. synthesized CuO nanofibers with bicontinuous mesopores to induce a strong confinement effect for *OH/*O transfer toward ECR¹⁷⁶ and achieved a C₂ (i.e., C₂H₄, C₂H₅OH, and CH₃COOH⁻) FE of up to 74.7%. The in situ Raman spectra suggested that the abundant bicontinuous mesoporous channels could maintain the *OH/*O adsorption at negative potentials, and further DFT results implied that *OH/*O reduced the binding energy to *CO, which lowered the energy barriers for both *CO–CO dimerization and *CO hydrogenation, and consequently enabled *CO–CHO coupling. Additionally, the confinement effect can increase the local pH in the EDL by inhibiting H⁺ from approaching the active sites and preventing the dissipation of the locally formed OH⁻, thus holding great potential for acidic ECR. Typically, Li and colleagues encapsulated Ag NPs within mesoporous hollow carbon spheres with a pore size of 2.6 nm and achieved a CO FE of over 95% in a K₂SO₄ electrolyte with a pH of 1.1.¹⁷⁷ Further characterizations and theoretical simulations demonstrated that due to the local accumulation of OH⁻ and the depletion of H⁺, the confinement effect led to the rapid neutralization of the acidic electrolyte in the EDL and maintained an alkaline local microenvironment capable of effectively suppressing HER.

3.2. Strain effect

The strain effect refers to the change in the electronic structure of solid-state electrocatalysts resulting from variations in the equilibrium distances between atoms in a lattice, which modulates the binding energies to the key intermediates and the final ECR performances.¹⁷⁸ Representatively, Zeng et al. conducted a comparative study on the strain effect of similarly sized Pd octahedrons and icosahedrons for ECR (Fig. 9a).¹⁷⁹ It was confirmed that tensile strain on the Pd icosahedron led to a decrease in the Pd d-band overlap and a shift in the d-band center toward the Fermi level (Fig. 9b), which enhanced the binding energy to *COOH and promoted ECR to CO. Besides tensile strain, Clark et al. claimed that incorporating a small amount of Ag in the surface of Cu induced compressive strain in Cu surface atoms, achieving higher selectivity for ECR over HER.¹⁸⁰ This was rationalized by the shift in the valence band density of compressively strained metals toward a higher binding energy, which resulted in a reduced interaction with the electron orbitals of the adsorbates in general, consequently causing the occupancy of the anti-bonding orbital and weak adsorption (Fig. 9c). Although it synchronously weakened the binding abilities to the intermediates for ECR and HER, the adsorbed *H at the hollow position on Cu(111) was more likely to destabilize than the adsorbed *CO at the top position due to the existence of compressive strain, thereby selectively suppressing HER, while promoting ECR.

In fact, exclusively studying the strain effect is challenging given that the introduction of strain often involves morphology control and/or surface alloying, which inevitably introduce other factors such as crystal facets and/or a second component that interfere the study of the pure strain effect.^179,183 In view of this, Kim et al. epitaxially grew gradually thinning Cu(001) films on an Si(001) single-crystal substrate to exclusively study the effect of pure strain on ECR.¹⁸¹ It was demonstrated that the increase in tensile strain resulting from a decrease in the film thickness ranging from 100 to 20 nm significantly improved the C₂₊/C₁ selectivity due to the upward shift in the Cu d-band center caused by the tensile strain and the concomitant increase in the binding energy for *CO and *H. Moreover, the enhanced surface coverage of *CO and *H may promote the protonation of the carbonaceous intermediates (e.g., *CO → *CHO), leading to a decrease in the desorption and further reduction of the C₁ intermediates, ultimately resulting in a decrease in the partial j for C₁ products. Interestingly, an increase in the partial j for C₂₊ products was not observed due to the inherent scaling relationship between *CO and *H, which contributed to the enhanced binding energy for *H on the Cu surface with the introduction of tensile strain, thus occupying the adsorption sites for ECR intermediates. Consequently, an increase in j for HER and a decrease in j for C₁ products were obtained, while the j for the C₂₊ products remained stable (Fig. 9d). Likewise, different Cu precursors [i.e., Cu(OH)₂, Cu₂(OH)₂CO₃, and CuO] were used to synthesize Cu electrocatalysts with a similar size but at a different degree of strain, and the results were contrasted with that reported by Kim et al.¹⁸² Although the C₂ product selectivity also remained unchanged with an increase in the tensile strain, the production of CH₄ significantly promoted, while that of H₂ was inhibited (Fig. 9e). DFT calculations revealed that the tensile strain indeed increased the binding energy for *CO by lifting the Cu d-band center upward, but its effect on the subsequent C–C coupling and HER was negligible, favoring the continued protonation of *CO to form CH₄.

3.3. Field-induced effect

The field-induced effect is considered as the tuning of the reactant concentration and reaction pathways induced by a locally enhanced field (e.g., electric, thermal, and magnetic fields) in regions with high curvatures of a solid-state electrocatalyst.¹⁸⁴ For instance, Liu et al. investigated the influence of a local electric field induced by an Au nanoneedle tip on ECR performance.¹⁸⁵ Finite element simulations suggested that the enhanced local electric field originated from the free electrons migrating to the charged metal electrode with the highest curvatures because of the weak electrostatic repulsion. This led to a 20-fold increase in the concentration of K⁺ ions adsorbed on the surface of the Au nanoneedle tip (Fig. 10a). The high K⁺ concentration favored the adsorption and activation of CO₂ and stabilized the key intermediates, thus improving the ECR performance (Fig. 10b). Apparently, the intensity of the local electric field mainly depends on the sharpness of nanostructured electrocatalysts. Based on this, Lee et al. synthesized three Bi nanostructures (i.e., dot, flake, and dendrite) to study the influence of the degree of sharpness on ECR performance, among which the sharpest Bi nanoflakes exhibited an enhanced j and FE_HCOOH.¹⁸⁶ The finite element simulations revealed that changing the thickness of the Bi nanoflakes from 500 to 10 nm resulted in a 5- and 3-fold increase in the electric field intensity at the corner and edge sites on the Bi nanoflakes, respectively (Fig. 10c). In addition, the electric field at the apex further increased 2.5 times as the corner angle decreased from 70° to 5° (Fig. 10d).

Moreover, close inspection of the nanoneedle array electrode showed that the generated electric field by two close needle tips will be intensified as the distance between adjacent nanoneedles gradually decreases and the enhancement degree is primarily determined by the distance between the nanoneedles, i.e., the proximity effect. To validate this, Yu et al. assembled a closely arranged CdS tip array structure¹⁸⁷ and observed that the concentration of K⁺ increased 28-fold as the gap width decreased from 40 to 10 nm, which further non-covalently interacted with CO₂ and quickly stabilized CO₂ at the tip of the CdS nanoneedle (Fig. 10e). Besides increasing the electron density, high-curvature metal nanostructures also promote the mutual collision of electrons, which leads to an increase in the local temperature. In particular, Yang et al. proposed a synergistic electro-thermal field strategy for ECR by covering polytetrafluoroethylene (PTFE) on Cu nanoneedles, where a locally intensified electro-thermal field was generated at the tip.²⁹ Indeed, the electron density at the tip increased by 2 times as the PTFE coverage ratio increased from 0 to 99% (Fig. 10f), which enhanced the electric and thermal fields by 2- and 3-folds at the tip, respectively. The intensified electric field reduced the ΔG required for the formation of *CO, while the enhanced thermal field accelerated the C–C coupling (Fig. 10g), both of which synergistically promoted the formation of the C₂ product.

However, despite the huge progress realized in structure-induced effects for ECR, the existing studies often struggle to establish exclusive one-to-one effect-performance relationships, which is challenging due to the presence of other inevitable interference effects. For instance, finite element simulations indicated that electrocatalysts with sharp edges or needle-like structures can induce strong a local electric or thermal field to enhance the ECR performance but cannot rule out the shape effect. Likewise, theoretical calculations could clarify the tensile and compressive strain effects on the electronic structures of electrocatalysts and their further influence on the binding energies to reaction intermediates. However, the existing approaches for introducing the strain effect usually involve the shape tuning of electrocatalysts and/or the incorporation of a second component with a different atomic radius. This inevitably introduces other effects (e.g., shape effect and/or composition effect), making the in-depth understanding of the strain effect inaccurate. Therefore, future research should focus on exclusively exploring a specific effect capable of establishing a pure one-to-one relationship between the structure-induced effect and the ECR performance without interferences.

4. Electrolyte effects

4.1. Solute effect

Generally, ECR takes place at the electrode–electrolyte-gas triple-phase interface of a solid-state electrocatalyst, and thus one can postulate that in addition to the electrocatalyst itself, another determinant of an efficient ECR is the electrolyte employed, together with the constituent solute concentration, which impacts the local EDL environment. Therefore, a profound understanding of the electrolyte effect is paramount for assembling effective systems for ECR. Serving as a reaction medium for ECR, the electrolyte not only acts as an ion carrier, but also tunes the formation of specific intermediates and/or products.³¹ Apparently, the selection of the electrolyte and its associated inherent properties emerge as pivotal factors in determining the ECR performance, as shown in Table 1. Considering that ECR involves multiple proton-coupled electron transfer steps, proton-rich aqueous media are preferred as the electrolyte. Upon dissolving CO₂ in an aqueous solution, a series of complicated reactions occur to reach a solubility equilibrium, as illustrated in Fig. 11a.¹⁸⁸ The dissolved CO₂ is progressively converted into HCO₃⁻ and CO₃²⁻ with an increase in pH, which inevitably decreases the CO₂ concentration in the solution. However, an excessively high concentration of H⁺ may favor HER over ECR in alkaline electrolyte with a pH below 4.3. Therefore, ECR is often conducted in a near-neutral solution with KHCO₃ as the solute to ensure an adequate CO₂ concentration and diminish the H⁺ concentration in static cells.¹⁸⁹ Beyond KHCO₃, a concentrated solution with a solute of KOH has been demonstrated to play a beneficial role in impeding HER and bolstering the C₂₊ product selectivity when using a gas diffusion electrode (GDE) in flow cells.¹⁹⁰ Although it is articulated that the dissolved CO₂ will be converted into the inert and ECR-detrimental HCO₃⁻ and CO₃²⁻ in highly alkaline media, the unique architecture of GDE shortens the CO₂ diffusion distance at the gas–liquid interface. This guarantees that CO₂ interacts with the electrocatalyst for its reduction to the target product before converting to HCO₃⁻ and/or CO₃²⁻, which inhibits HER and ensures a high CO₂ concentration due to elevated alkalinity.

---

However, it is worth noting that under aqueous conditions, the selectivity and activity of molecular electrocatalysts toward ECR tend to decrease, which is primarily due to two reasons, as follows: (i) hydrophobic molecular electrocatalysts normally exhibit poor solubility in polar aqueous electrolytes, which limits their accessibility to the electrolyte and (ii) the metal centers are prone to adsorb H⁺ and prefer the dominant pathway for the competitive HER.^155,204 Likewise, the ECR performances of molecular electrocatalysts in proton-free organic solvents are also poor due to the lack of an H⁺ source, and thus the addition of a Lewis acid or Brønsted acid is significant.^205,206 In general, Fe porphyrins produce CO in the presence of mono (e.g., Li⁺ and Na⁺) or divalent (e.g., Mg²⁺, Ca²⁺, and Ba²⁺) cations,²⁰⁵ coupled with HCOOH as a by-product. However, the introduction of weak/medium organic acids (e.g., CF₃CH₂OH and PhOH) has a pronounced effect on ECR,²⁰⁷ resulting in ultrahigh FEs for CO production without H₂ and HCOOH being detected. Thus, it is believed that in the presence of a Lewis acid or Brønsted acid, the molecular electrocatalyst-mediated ECR to CO undergoes a two-electron “push–pull” mechanism. Initially, the ligand-assisted active metal center interacts with CO₂ to form the key M–CO₂ intermediate, where the electron density is “pushed” from the nucleophilic metal center to the electrophilic CO₂ molecule. Following this, the M–CO₂ intermediate is stabilized by forming ion pairs (Lewis acid) or H-bonds (Brønsted acid), which helps to “pull” the electron density from *CO₂ and facilitates the C–O cleavage step.

4.2. Solvent effect

Various non-aqueous electrolytes (e.g., organic solvents and ILs) have also been developed with the aim to improve the ECR performance.^208,209 In the case of organic solvents, one prominent advantage over aqueous media is the higher CO₂ solubility, and most organic solvents present a diminished concentration of H⁺, thus effectively curbing the HER.^33,201 Moreover, alternative reaction pathways favoring specific products can be facilitated within organic solvents, which occur primarily in three ways, as follows: (i) the dimerization of two *CO₂ molecules to yield oxalic acid, (ii) the disproportionation of *CO₂ and CO₂ to produce CO and CO₃²⁻, and (iii) the protonation of CO₂˙⁻ radicals to form HCOOH in the presence of small amounts of H₂O.^202,210 Acetonitrile, usually possessing an 8-fold higher CO₂ solubility (∼270 mM) than aqueous solution, has been subjected to extensive studies in the field of ECR.^32,211,212 More interestingly, the HER-preferred Mo₂C exhibited high ECR activity and selectivity in acetonitrile.³² The electrochemical infrared reflection absorption spectroscopy (EC-IRRAS) analyses and DFT calculations consistently elaborated that acetonitrile decomposed to form the 3-aminonitrile anion, which interacted with CO₂ to form 3-aminonitrile carboxylate (Fig. 11b and c), followed by its subsequent reduction to CO. It is noteworthy that ECR within acetonitrile is quite sensitive to the presence of H₂O, where the reaction pathways and product selectivity undergo a significant shift, even with trace amounts of H₂O as low as 46 ppm. Simply, Aljabour et al. examined the impact of minute amounts of H₂O (1% w/v) in acetonitrile on ECR over Co₃O₄ nanofibers,²¹³ and surprisingly noticed that HCOOH was produced in the presence of H₂O, whereas the primary product resulting from the disproportionation of CO₂˙⁻ and CO₂ in the neat acetonitrile was CO.

In addition, ILs with outstanding ionic conductivities and strong CO₂ adsorption capabilities have been intensively explored as electrolytes for ECR.^196–200 Rosen et al. first reported the use of an IL, 1-ethyl-3-methylimidazolium tetrafluoroborate ([EMIM][BF₄]), as an electrolyte for ECR to CO at extremely low η on Ag electrodes.²¹⁴ Afterwards, various cation and anion-based ILs, such as [EMIM][Tf₂N],²¹⁵ [EMIM][Cl],²¹⁶ and [BMIM][PF₆],⁴¹ have been used as electrolytes for ECR. It is well recognized that due to the functional groups on triazolium and/or imidazolium rings, most ILs have potential to capture and activate CO₂ with a lower ΔG for ECR either by chemical adsorption or by forming complexes with CO₂.²¹⁴ Furthermore, the H atoms at the C4 and C5 positions of the imidazolium cation have been proven to be binding sites for CO₂˙⁻ (Fig. 11d).²⁰³ Concurrently, ILs can influence the electronic structure of electrocatalysts and the local microenvironment at the electrode–electrolyte interface. For instance, the [BMIM]⁺ in [BMIM][PF₆] formed a strong local electric field at the EDL,⁴¹ which induced a positive shift in the d-band center of Ni, while [PF₆]⁻ formed a hydrophobic layer to control the mass transfer of CO₂ and H₂O (Fig. 11e). More importantly, a variety of ILs allow the customization of the electrolyte to achieve the most desirable ECR performance. Accordingly, Zhou et al. used Ag, Au, Cu, and Pt as electrode materials to conduct comparative experiments for ECR in [EMIM]Cl with different H₂O concentrations (20, 40, 60, and 80 wt%),²¹⁷ and stated that Ag exhibited the highest CO selectivity in the IL containing 20 wt% H₂O due to the strong H-bonds formed between the H atoms in H₂O and Cl⁻.

4.3. Cation effect

Upon initiating ECR, the negatively charged cathode induces the accumulation of cations and depletion of anions, resulting in the formation of an EDL at the electrode–electrolyte interface including the inner Helmholtz plane (IHP) and the outer Helmholtz plane (OHP).^43,193,218 The alkali metal cations are directly involved in the EDL under negative potentials, prone to hydrate and accumulate in the OHP through electrostatic adsorption, consequently changing the property of the electrode–electrolyte interface, and further influencing the ECR performance.^189,195 Currently, there are three main theories regarding the impact of alkali metal cations on ECR, as presented in Fig. 12a, including: (i) modification of the local electric field, (ii) pH buffering in EDL, and (iii) stabilization of the key intermediates, but there is still no consensus.

An early study by Murata and Hori indicated that the product selectivity on Cu electrodes was highly dependent on the type of alkali metal cations in the electrolyte.²²³ It was observed that the C₂ selectivity increased with an increase in the Stokes radius of the hydrated alkali metal cation from Li⁺ to Cs⁺, which was rationalized by the adjustment in the OHP potential caused by the change in cation size (Fig. 12b). Specifically, the specific adsorption of smaller alkali metal cations (e.g., Li⁺) with strong hydration was inhibited on the electrode surface, whereas larger weakly hydrated alkali metal cations (e.g., Cs⁺) were more easily adsorbed on the electrode surface.²¹⁹ Consequently, this changed the OHP potential and impacted the H⁺ concentration in EDL, which finally tailored the product selectivity. However, considering that the equilibrium potential of alkali metal cation adsorption on metal electrodes is mostly more negative than −2 V versus the standard hydrogen electrode,²²⁴ the adsorption of specific alkali metal cations may not occur at the ECR operating potential. Markovic and colleagues demonstrated that the activity of redox reactions was mainly influenced by the non-covalent interactions between the hydrated cations and the adsorbed species rather than the covalent and electrostatic interactions.²²⁵ This suggests that the change in the OHP potential derived from the specific adsorption of alkali metal cations may not accurately explain the improved ECR performances in electrolytes with larger cations. Subsequently, Singh et al. proposed a new viewpoint on the influence of alkali metal cation size toward ECR, which was attributed to the hydrolysis of hydrated alkali metal cations near the electrode.²²⁰ As the size of the alkali metal cations increase, the pK_a of the hydrated alkali metal cation decreased, which introduced more H⁺ and caused a decrease in the local pH (Fig. 12c). This further alleviated the polarization and energy loss caused by the increase in local pH under the ECR operating conditions, and ultimately promoted the evolution of CO, C₂H₄, and C₂H₅OH rather than H₂ and CH₄. In view of the strong correlation of pH with the local CO₂/HCO₃⁻ ratio, Ayemoba et al. observed the pH changes at the interface between the Au electrode and electrolyte using in situ surface-enhanced infrared absorption spectroscopy, further confirming that the differences in ECR activity and selectivity under different alkali metal cation environments were due to the decrease in pK_a caused by the hydrolysis of the alkali metal cation.²²¹ Moreover, they clarified that Singh and colleagues overestimated the decrease in the pK_a due to alkali metal cation hydrolysis, especially for larger alkali metal cations, due to the overestimated charge density on the electrode surface. Notably, the impact of cations on local H⁺ concentration is particularly obvious in acidic ECR, and the H⁺ coverage at the electrode–electrolyte interface is sensitive to the K⁺ concentration. Due to the electrostatic shielding effect induced by the accumulated cations at OHP (Fig. 12d), the surface H⁺ coverage sharply decreased as the K⁺ concentration increased from 0 to 3 M, ultimately suppressing HER and favoring ECR.²²⁶ In addition, Gallent et al. provided another viewpoint on the changes in hydrocarbon selectivity caused by the alkali metal cations on Cu electrodes,²²⁷ claiming that metal cations acted as the catalytic promoters and selectively stabilized the reaction intermediates, where larger cations (i.e., Rb⁺ and Cs⁺) were more effective in stabilizing the intermediates than smaller cations (i.e., Li⁺ and Na⁺). Similarly, Guo et al. concluded that the alkali metal cations (e.g., Na⁺ and K⁺) stabilized the Fe–CO₂ adduct through electrostatic interaction, which enhanced the catalytic activity of Fe porphyrins toward ECR to varying degrees.²²⁸ Nevertheless, it is worth noting that due to its smaller hydration radius, K⁺ contributed to a faster migration rate caused by the electric field-induced effect, which increased the interfacial electron transfer ability, thus exhibiting a stronger cation effect and a better ECR performance compared to Na⁺.

Furthermore, metal cations with higher valence states (e.g., La³⁺ and Zr⁴⁺) were found to be more effective than monovalent alkali metal cations in improving the ECR performance on CuSnPb alloy electrodes due to the fact that the reduction of the negatively charged CO₂˙⁻ could be further promoted with an increase in surface charge.²²⁹ Additionally, alkylammonium cation additives with specific configurations were also demonstrated to impact the ECR performance.²²² In the presence of two spatially bulky alkylammonium cations (propyl N⁴⁺ and butyl N⁴⁺), the intermolecular interaction between the adsorbed surface *CO and the interfacial H₂O was screened to prevent the protonation of the *CO dimer to form C₂H₄ (Fig. 12e).

4.4. Anion effect

Similarly, anions also crucially influence the ECR performance. As previously discussed, KHCO₃ solution can regulate the local CO₂ and H⁺ concentrations through a series of chemical reactions and directly serve as a proton donor to participate in ECR (Fig. 13a, left). In fact, Bell et al. studied the ECR performances on a Cu electrode in aqueous solutions with different KHCO₃ concentrations²³⁰ and found that the partial j of CO, HCOO⁻, C₂H₄, and CH₃CH₂OH remained almost unchanged with an increase in the KHCO₃ concentration, whereas that of H₂ and CH₄ significantly increased. The difference was because the rate-determining steps for the production of CO, HCOO⁻, C₂H₄, and CH₃CH₂OH did not involve the addition of H atoms, whereas the generation of H₂ and CH₄ was the opposite. Clearly, the pK_a of HCO₃⁻ is four orders of magnitude lower than that of H₂O, indicating that HCO₃⁻ can serve as an H⁺ donor for both ECR and HER. Upon increasing the KHCO₃ concentration, the enhanced H⁺ supply promoted the production of H₂ and CH₄. Also, Bell et al. observed that the presence of ClO₄⁻, SO₄²⁻, PO₄³⁻, and BO₃³⁻ showed no significant effect on the formation of CO, HCOOH, C₂H₄, and CH₃CH₂OH on the Cu electrode, but the generation of H₂ and CH₄ was quite sensitive to the concentration of these anions, which could be also attributed to the different abilities of these buffering anions to act as proton donors. Thus, the use of anions with low buffering capacities is believed to suppress the formation of H₂ and CH₄.

Another widely studied category of anions is the halide ions (i.e., F⁻, Cl⁻, Br⁻, and I⁻), which can form covalent or coordination bonds with metal atoms or functional groups on the electrode surface.^191,194 This strong chemical interaction allows anions to accumulate in the IHP through specific adsorption, introducing extra van der Waals interaction with the intermediates. Cho et al. prepared functionalized Au electrodes by electroplating in an aqueous solution containing CN⁻ or Cl⁻ and experimentally verified that the Cl-functionalized Au indeed showed significantly enhanced CO selectivity compared to pristine Au (Fig. 13b).²³¹ DFT calculations indicated that the van der Waals stabilization effect on *COOH was enhanced with an increase in the coverage of adsorbed Cl⁻ (1/9 and 2/9 monolayer) on Au, but a further enhanced Cl⁻ coverage of 3/9 monolayer resulted in a lower binding strength for *COOH (Fig. 13c), which is probably due to the competitive adsorption of Cl⁻ on the surface of Au with CO₂ and the reaction intermediates. Analogously, Hsieh et al. prepared a coral-nanostructured Ag electrocatalyst in the presence of Cl⁻ using a redox method, and the obtained Ag surface retained a certain amount of Cl⁻, which was confirmed to enhance the intrinsic catalytic activity for ECR.²³² The further reduction of the coral-nanostructured Ag electrocatalyst under an H₂ atmosphere could remove the adsorbed Cl⁻ on its surface and retain its porous structure, but resulted in a decrease in j for CO by 54%, persuasively demonstrating the promoting effect of Cl⁻ on CO formation. Clearly, halide ions are beneficial for the adsorption of CO (Fig. 13d), and the desirable generation and stabilization of *CO are conducive to further C–C coupling for the formation of multi-carbon products on Cu-based electrocatalysts.²² Based on this, Gao et al. added KX (X = Cl⁻, Br⁻, and I⁻) to 0.1 M KHCO₃ solution to study the underlying effects of different halide ions on ECR over a plasma pre-oxidized Cu electrocatalyst.²³³ Indeed, the presence of Cl⁻, Br⁻, and I⁻ improved the ECR performance with a reduction in η, following the order of Cl⁻ < Br⁻ < I⁻. Furthermore, the adsorbed halides on the Cu surface could impart some charges to the C atom in the CO₂ molecule and form covalent X–C bonds, thus facilitating the initial electron transfer to produce the bent CO₂^•− from the linear CO₂ (Fig. 13a, middle). Thus, it can be concluded that the ECR performance strongly depends on the specific adsorption ability of the halide ion (I⁻ > Br⁻ > Cl⁻), which also well explains the activity trend on the oxidized Cu foil (I⁻ > Br⁻ > Cl⁻ > no halides). Besides enhancing the CO₂ adsorption ability, halide ions can also induce the surface reconstruction of the electrocatalyst (Fig. 13a, right). For example, Tan et al. proposed a route to improve the CO tolerance of a Pd electrocatalyst by tailoring the microenvironment of the electrocatalyst-electrolyte interface in a KCl-saturated solution (Fig. 13e).¹⁹² The results showed that Cl⁻ triggered the dynamic surface reconstruction of Pd NPs to form O–Pd–Cl species with weak CO adsorption ability in the EDL during ECR (Fig. 13f), which endowed Pd NPs with a high CO tolerance.

5. Environment effects

5.1. Ionomer effect

Generally, ionomers are polymers composed of repeating organic units and connected to the main chain by covalent bonds, which can dissociate ions in H₂O and present unique physical properties (e.g., viscosity, hydrophilicity/hydrophobicity, and conductivity). Therefore, ionomers often serve as binders to anchor solid-state electrocatalyst and manipulating the ionomer structures and their associated hydrophilic/hydrophobic groups can not only change the mass transfer abilities of CO₂ and H₂O to the electrode surface, but also stabilize the chemically adsorbed intermediates on electrocatalysts on the electrode surface. Moreover, they can interact with the H-bonds, and/or induce an electrostatic effect due to their specific functional groups. Nafion as a representative ionomer, which is composed of hydrophobic perfluorocarbon backbones and hydrophilic sulfonic acid side chains, has been widely used for ECR. Specifically, the hydrophobic perfluorocarbon backbone in Nafion endows it with high mechanical strength and chemical stability, allowing it to remain stable under extreme chemical and physical conditions (e.g., strong acids and bases). By contrast, the hydrophilic sulfonic acid side chains are the key functional components in Nafion, which facilitate H⁺ transfer through ion exchange. Thus, the dual characteristics of Nafion play a crucial role in electrode preparation, where the hydrophobic PTFE chains help in the uniform deposition and fixation of electrocatalysts on the electrode, while the hydrophilic sulfonic acid ends facilitate the effective contact between the electrolyte and the electrocatalysts, thus enhancing the electrocatalytic efficiency. Ding et al. declared that the ECR performance was significantly influenced by varying the Nafion/solvent formulation (i.e., different ratios of n-PrOH and H₂O) in terms of catalytic activity, selectivity, and stability.²³⁴ Representatively, the H₂ FEs were reduced with a gradual increase in H₂O content in the electrocatalyst ink due to the fact that Nafion was dissolved in the low dielectric constant solvent of n-PrOH, which led to the formation of loose ionomer aggregates with linked sulfonic acid-terminal groups. In contrast, the high dielectric constant solvent of H₂O yielded tightly packed aggregates with strong interactions, resulting in a high connectivity of the inner network, reduced swelling, enhanced ionic conductivity and decreased H₂O uptake upon forming a Nafion film (Fig. 14a). Therefore, a high H₂O content between 50–75 vol% in the electrocatalyst ink is recommended for the optimized quality of Nafion coating.

However, considering the high cost of Nafion, substantial efforts have been devoted to exploring more affordable alternatives. For instance, Pham and colleagues investigated the effects of three ionomers with differing hydrophilicities [i.e., polyacrylic acid (PAA), Nafion, and fluorinated ethylene propylene (FEP)] on the ECR performances over Cu NPs (Fig. 14b),⁴⁸ among which the electrode coated with FEP exhibited the highest C₂₊ partial j of over 600 mA cm⁻² with 77% C₂₊ FE, while that with PAA performed the worst because of their different hydrophobic and lipophilic properties. It is well known that H⁺ and dissolved CO₂ in the electrolyte competitively participate in HER and ECR, respectively, but the electrode surface with Cu-PAA could inevitably enrich more H⁺ than dissolved CO₂ due to the strong hydrophilicity of PAA, delivering poor ECR selectivity. By contrast, the electrocatalyst surface with the highly hydrophobic FEP created localized channels, where only CO₂ could pass through the electrode and increase the local concentrations of CO₂ and *CO, which promoted the subsequent C–C coupling and improved the C₂ product selectivity. In addition to mass transfer control, the chemistry interplay between the groups on the ionomers and the reaction intermediates has been proven to impact the ECR performance. Chang and co-workers examined the ECR performances over Cu electrocatalysts using ionomer binders containing different types of functional groups (i.e., –COOH and –CF₂).²³⁸ It was found that PAA with –COOH groups favored the generation of HCOOH in the entire potential range, whereas polyvinylidene difluoride (PVDF) with –CF₂ groups favored the formation of CH₄. DFT calculations suggested that due to the H-bonds between the *OCHO and the –COOH groups, *OCHO bonded more strongly on COOH–Cu(111) than on Cu(111) and CF₂–Cu(111). Nevertheless, the H-bonds between *OCHO and the –CF₂ groups were not observed on CF₂–Cu(111) probably due to the hydrophobic nature of the –CF₂ group. Consequently, this difference in the binding affinity of the reaction intermediates endowed distinct thermodynamically favorable ECR pathways on PAA-Cu and PVDF-Cu with different product distributions.

5.2. Pressure and temperature effect

In aqueous ECR systems, CO₂ molecules are dissolved in the electrolyte, and then diffused to the electrode surface, followed by their adsorption and further reduction to various target products. Therefore, increasing the saturated CO₂ concentration adjacent to the cathode is an efficient way to enhance the ECR performance. According to Henry's law, an increase in CO₂ partial pressure can improve the concentration of dissolved CO₂ in the electrolyte and its mass transfer rate. Given this, Scialdone et al. evaluated the ECR performances on an Sn electrode in an undivided stainless steel reactor at various pressures. The results revealed that at the same j, the HCOOH selectivity increased as the pressure increased from 1 to 5 bar, implying that enhancing CO₂ partial pressure indeed amplified the CO₂ solubility and diffusion rate, which further improved the reactant concentration adjacent to the electrode surface. However, upon further increasing the pressure from 5 to 10 bar, the HCOOH selectivity and partial j suffered from a bottleneck, suggesting that the rate-determining step was gradually shifted from the saturated CO₂ diffusion to the proton-coupled electron transfer. However, this situation was interrupted, resulting in a continuous increase in HCOOH selectivity as the applied potential was further increased, demonstrating that increasing the CO₂ partial pressure could promote ECR by breaking the limitation of the local CO₂ concentration in EDL. Moreover, many electrocatalysts (e.g., Fe,²³⁹ Pt,²⁴⁰ Ni,²⁴¹ and C²⁴²) primarily being selected for HER under a standard atmospheric pressure of 1 atm have been demonstrated to exhibit enhanced selectivity for CO or HCOOH when pressurized. The significant increase in the local CO₂ concentration in the EDL under high pressure conditions results in more active sites on the electrode surface being occupied by CO₂ molecules rather than H⁺, thus favoring ECR, while inhibiting HER. Also, the pressure effect was proven to be capable of tuning the product selectivity, especially on Cu-based electrocatalysts. Upon applying a high CO₂ pressure of 58 atm in a KOH electrolyte with borate, a high acetate FE of 87% was achieved on a Cu(OH)₂-derived Cu/CuO_x electrocatalyst (Fig. 14c).²³⁵ The in situ Raman spectroscopy analysis indicated that under high-pressure conditions, this Cu-based electrocatalyst facilitated the direct formation of an oxygen-bound bidentate intermediate (*OCO*) on Cu(I) (Fig. 14d), and its subsequent coupling with a second C₁ species/intermediate to form CH₃COO⁻ rather than the conventional *CO dimerization. Besides tuning the selectivity, increasing the pressure also inhibits the nucleation and growth of bubbles, which reduces the obstruction of the ion conduction pathway, thus lowering the ohmic loss caused by the bubble interference.²⁴³ Furthermore, under high-pressure conditions, smaller bubbles are more easily detached from the electrode surface, leaving more of the electrode surface exposed to minimize the coverage of the active regions on the electrode.

As another critical environmental factor, the effect of temperature is complex given that it impacts many aspects of ECR (e.g., the solubility of CO₂ and its diffusion ability, the hydrolysis equilibrium of H₂O and H₂CO₃, the thermodynamic equilibrium potentials, and the electrolyte conductivity). Indeed, increasing temperature presents a promising avenue to improve the reaction kinetics and the electrolyte conductivity but is accompanied by a decrease in CO₂ solubility. Vos et al. studied the pure temperature effect unrelated to CO₂ concentration by adjusting the CO₂ partial pressure at different temperatures as the rectification. The CO₂ concentration at 70 °C under 1 atm was selected as the standard concentration, and the required CO₂ partial pressure to achieve the same CO₂ concentration was additionally determined for all other temperatures. Under the same CO₂ concentration at various temperatures, the j_CO increased as the temperature increased (Fig. 14e).²³⁶ Thus, it is reasonable to posit that when the CO₂ concentration in the electrolyte is high enough to offset the decrease caused by an increase in temperature, the temperature has a net positive effect on ECR. Building on this, Li and colleagues immobilized Au NPs on a porous carbon substrate to construct a superhydrophobic electrode for ECR,⁵⁰ which possessed a solid–liquid–gas triple-phase interface structure similar to the GDE, thus facilitating a rapid CO₂ supply through the gas phase to maintain a high CO₂ concentration even at relatively high temperatures. It was found that within the triple-phase interface structure, the partial j and the TOF for CO increased by 240% as the reaction temperature increased from 8 °C to 60 °C due to the approximately 26.2 times higher local CO₂ concentration than that within the double-phase electrode at 60 °C. Clearly, this strategy well addresses the issue of low CO₂ solubility at high temperatures, and ultimately achieve rapid ECR reaction kinetics induced by the high temperatures without sacrificing the local CO₂ concentration.

Although the primary products on most ECR electrocatalysts show negligible changes with temperature, studies on Cu electrocatalysts have shown that the product distribution varies in different temperature ranges during ECR.^237,244 Typically, Vos et al. evaluated the ECR product distribution on a Cu electrode by gradually increasing the reaction temperature in two different ranges.⁵⁵ The in situ Raman analysis revealed that due to the increase in *CO coverage and local pH, the selectivity for CH₄ and HCOOH decreased with an increase in temperature in the first range from 18 °C to 48 °C, whereas that for C₂H₄ product increased and that for H₂ remained stable. Nevertheless, HER dominated the overall catalytic activity in the second range from 48 °C to 70 °C because of the surface reconstruction of Cu induced by high temperature. Moreover, Jo et al. designed an ECR system capable of cooling the local electrode environment without impacting the bulk electrolyte temperature.²³⁷ Given that the bulk electrolyte temperature would not be influenced by the localized cooling on the electrode surface, the temperature effect could be independently studied based on the changes in carbonate equilibrium with temperature (Fig. 14f). Apparently, the decrease in local temperature on the Cu electrode surface improved the CH₄ selectivity due to the notable restriction in the diffusion of *CO (Fig. 14g), i.e., the key intermediate for CH₄ evolution, together with the prolonged residence of *CO on the electrode surface.

5.3. Gas impurity effect

From an application perspective, the most viable CO₂ sources for ECR should take from industrial emissions, thus inevitably containing varying levels of gas impurities (e.g., N₂, SO_x, NO_x, CO, and NH₃). Thus, to realize the direct use of industrial CO₂ emissions, it is crucial to understand the impact of gas impurities on the ECR performance. Generally, gas impurities can induce alterations in the local environment at the electrode–electrolyte interface, mainly involving (i) the occupation of active sites on the electrocatalyst surface to compete with CO₂ adsorption and transformation of the key intermediates, (ii) chemical interaction with the electrocatalytically active surface to induce surface reconstruction, and (iii) participation in the ECR process to alter the reaction pathway, as clearly shown in Fig. 15a. Typically, Wang et al. evaluated the ECR performance on Cu electrocatalysts with different CO₂ concentrations (i.e., 25%, 50%, 75%, and 100%) by adjusting the ratio of CO₂ to inert N₂ in the feed gas,²⁴⁵ and observed an increase in CH₄ FE and a decrease in C₂H₄ FE upon introducing inert N₂ gas impurity. Although the inert N₂ gas impurity did not directly participate in ECR on the Cu surface, it diluted the CO₂ concentration in the feed gas and reduced the coverage of *CO₂, and subsequently *CO, which disfavored C–C coupling to form C₂H₄ and resulted in enhanced CH₄ selectivity. Upon further increasing the N₂ gas impurity, a decrease in CH₄ FE was observed after reaching the peak due to the mass transport limitation of CO₂, where the generated *CO was insufficient to protonate for the formation of CH₄. Besides the inert gas impurity, the effects of some redox-active gas impurities (e.g., NO_x and SO_x) on ECR have also been investigated. Luc et al. studied the impact of SO₂ in the feed gas on ECR over Ag, Sn, and Cu in the near-neutral electrolyte.⁵² The results showed that due to the thermodynamic favorability, SO₂ preferentially underwent reduction on the electrode surface, thereby degrading the overall FE of ECR. Further inspection showed that Ag and Sn were fully recovered upon stopping the SO₂ feed, and still maintained their selectivity for ECR to CO and HCOOH, respectively. However, owing to the new generation of more stable Cu₂S on the electrocatalyst surface, an irreversible transition was detected on the Cu electrocatalysts for HCOOH production, accompanied by the suppression of multi-carbon product formation (Fig. 15b). Similarly, Ko et al. demonstrated that the presence of minor nitric oxide compounds (e.g., N₂O, NO, and NO₂) in the feed gas could induce a significant decrease in the total FE for ECR due to their more positive reduction potentials than CO₂.⁵⁶ Despite lowering the FE for ECR, minor NO_x in CO₂ feed gas was proven to not alter the properties of the electrocatalyst during ECR, which could deliver a similar ECR performance to the initial stage once pure CO₂ was resumed. Besides undergoing competitive adsorption and/or reduction with CO₂, introducing N-containing gas impurities also assists the co-reduction to form a variety of value-added products. Notably, Jouny et al. reported the formation of C–N bonds in the presence of NH₃, and the FE for acetamide production was close to 40% at a j of 300 mA cm⁻².²⁴⁶

5.4. Flow rate effect

The influence of flow rate refers to the effects of the electrolyte and inlet gas flow rate on the ECR performances. It is understandable that the flow rate of the electrolyte and inlet gas changes the mass transfer of the reactants, key intermediates, and products on the electrode surface. To gain insights into this, Koper et al. developed a rotating ring-disc electrode (RRDE) setup to quantitatively study the role of the electrolyte flow rate in the competition between ECR and HER on an Au disk electrode.²⁴⁷ It was identified that with an increase in the electrolyte flow rate, the selectivity of CO over H₂ on the Au electrode was significantly enhanced (Fig. 15c). Moreover, the partial j of H₂ was observed to significantly decrease with an increase in the electrolyte flow rate, but that of CO remained almost constant, suggesting that the enhanced CO selectivity was actually derived from the suppressed HER rather than the increase in ECR rate. Given that an increase in HER rate was observed on Au RRDE as the pH increased, it was concluded that an increase in electrolyte rotation speed could facilitate the release of OH⁻ from the electrode surface and reduce the local micro-environmental alkalinity, thereby inhibiting HER and improving the CO selectivity. However, Kim et al. obtained a different conclusion based on their study on the effect of the electrolyte flow rate on ECR over Cu using an RRDE.²⁵⁰ The results indicated that as the rotation speed increased, the j for ECR decreased, while that for HER increased, together with a shift in the product selectivity from CH₄ to CO. It was inferred that the mass transfer of dissolved CO on the electrode surface reduced the surface coverage of *CO at a high electrolyte flow rate owing to the equilibrium between *CO and dissolved CO at the electrode–electrolyte interface, thus limiting the further reduction of *CO to hydrocarbons and resulting in changes in the product selectivity.

Different from RRDE, the impact of the electrolyte flow rate is even more pronounced in GDE-based flow cells. Yuan et al. designed a large-scale flow cell to assess the scaling behavior of IL-based electrolytes for ECR.²⁴⁸ A volcano-shaped relationship between FE_CO and the electrolyte flow rate on an Ag electrode was established. In detail, a high flow rate only allowed a short residence time of the electrolyte in the flow cell, which exacerbated the turbulent electrolyte flow and affected the ion transport, thereby reducing the catalytic activity and the CO selectivity. Conversely, a low electrolyte flow rate led to the sluggish removal of gaseous products from the flow cell and aggregation of bubbles at the cathode, which also decreased the ECR performance (Fig. 15d). Similarly, Tan et al. demonstrated that in a GDE-based flow cell, changes in the feed gas flow rate led to remarkable shifts in the product distribution on Cu at 200 mA cm⁻², where the FE for the C₂₊ products exponentially increased to 60.3% as the feed gas flow rate decreased from 40 to 5 sccm, but the FE for the C₁ products followed an inverse trend due to the different CO₂ concentrations in the different gas flow rates (Fig. 15e).²⁴⁹ Clearly, a faster gas flow rate enabling a higher local CO₂ concentration may lead to a greater population of unreacted *CO₂ species, which is not conducive to CO dimerization, while an appropriate gas flow rate can provide an optimal local *CO₂ concentration and guarantee sufficient *CO generation without exacerbating the competition of active surface sites, thus maximizing the *CO coverage on the Cu electrocatalyst surface and facilitating the C-C coupling to form C₂₊ products.

Notably, it is found that the key research points on ECR have gradually shifted from solely focusing on the synthesis of high-performance electrocatalysts for static cell testing to electrolyzer construction, electrode assembling, and environmental condition control for the flow cell production of target chemicals/fuels. However, the priorities to further scale-up industrial ECR applications should consider four points, as follows: (i) directly using waste gases as feedstocks from various fields (e.g., power plant, industrial combustion, iron and steel industry); (ii) running large stacks consisting of thousands of cells for large-scale production toward the target products; (iii) operating at high temperatures and pressures for desirable reaction kinetics and production rate; and (iv) performing accurate techno-economic analysis. Apparently, the widely used computational calculations, which primarily guide the electrocatalyst design and the EDL microenvironment regulation at the atomic level, cannot effectively guide industrial-scale ECR systems (e.g., system integration and process modeling), highlighting the necessity of adopting more reliable and feasible industrial simulation methods. In this case, finite element simulation under multi-physics field coupling allows the synchronous study of different effects (e.g., fluid, temperature, pressure, and electrolyte) on the ECR system performance, which is believed to be economic-effective and time-saving in accessing industrial-scale ECR systems.

6. Conclusions

The clean-energy-powered ECR to high-value chemicals/fuels holds great potential to realize a carbon-neutral footprint. Thus far, huge progress across multiple fronts, including electrocatalyst design (e.g., size and shape tuning, composition adjustment, defect incorporation, interface engineering, and ligand modification), electrolyte optimization (e.g., solute and solvent selection, and cation and anion modulation), and control of external environment parameters (e.g., ionomer selection, pressure and temperature adjustment, gas impurity exploration, and flow rate tailoring), has been made in pursuit of advancing industrial-scale ECR applications. However, there is still a lack of a comprehensive and lucid understanding on the underlying specific effects of all these strategies and their corresponding mechanisms for enhanced ECR performances. Moreover, these gaps in the knowledge base hinder further the optimization of ECR systems and their large-scale industrialization, thus inspiring us to present this review with a focus on the strong correlations between different effects induced by various design strategies and electrochemical ECR to high-value fuels/chemicals, together with the underlying issues and challenges faced by this technology in terms of further clarifying various effects in ECR, which are suggested as follows.

When examining the interconnection between a specific effect and ECR performance, interference often occurs by other influencing variables. Previous studies have unveiled conflicting findings regarding the interconnection of even the same effect with ECR performance across different timeframes, reaction systems, and/or operation circumstances, despite being substantiated by experiments and/or theoretical calculations. Thus, to gain deep insights into this, future research objectives should encompass two key aspects. Firstly, the associated experiments should be intellectually designed to independently investigate a pure effect from other variables, with the aim of determining the relationship in isolation. Secondly, constructing models that involve all potential influencing factors, and considering them to achieve the optimal ECR operation.

The conventional trial and error approaches are generally inefficient and laborious, while the existing descriptors, such as the classic Sabatier principle and d-band center theory, have certain limitations. Therefore, it is pivotal to develop more universal and measurable descriptors to guide the electrocatalyst design and accurately predict the effect-activity/selectivity relationship. In this case, theoretical calculations (e.g., DFT, machine learning, and finite element simulation) can provide beneficial guidance for the design of materials and understanding of their mechanisms. Specifically, machine learning can facilitate the high-throughput screening of materials, composition optimization, identification of the active sites, and prediction of the activity and selectivity, while finite element simulation imitates the collision of NPs and distribution of different parameters (e.g., current, potential, reaction rate, and CO₂ concentration adjacent to electrocatalyst/electrolyte) to understand the entire ECR process and the underlying mechanism. By coupling experimental validation, a rigorous closed loop between experiment and theory is formed, which promotes the discovery of a universal and precise electrocatalyst descriptor, thus paving an effective avenue to unlock the inherent effect-performance relationship.

Concurrently, the dynamic structural evolution (e.g., surface reconstruction, atomic rearrangement, in situ deposition and/or reduction, phase transition, coarsening, and agglomeration) of electrocatalysts during ECR usually changes their physicochemical and structural properties. Clearly, simply relying on ex situ characterizations usually underestimates the dynamic changes in the physiochemical and structural evolutions of electrocatalysts, and the real-time transitions of the reactants, key intermediates, and products during ECR, which probably deviate from the factual conclusions and distort the fundamental understandings of the true reaction pathways, the underlying mechanisms and the effect-performance relationships. This inspires us to pay more attention to the electrocatalyst before, during, and after ECR. The integration of in situ/operando techniques (e.g., in situ Fourier-transform infrared spectroscopy, in situ Raman spectroscopy, in situ X-ray photoelectron spectroscopy, in situ X-ray diffraction, in situ transmission electron microscopy, scanning electrochemical microscopy, and in situ synchrotron radiation) is an exciting direction, which provides numerous opportunities for in situ detection and real-time observation. This includes confirmation of the local coordination environments and near-surface oxidation states, identification of the reaction intermediates and electron transfer behaviors, and evaluation of the local electrocatalytic activity and selectivity as well as observation of the evolving structures and phases, all reflecting the most authentic relationships between different effects and ECR performances from various perspectives.

Data availability

No primary research results, software or code has been included and no new data were generated or analyzed as part of this review.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work is supported by the National Natural Science Foundation of China (NSFC, Grant No. 22109182), the Pilot Group Program of the Research Fund for International Senior Scientists (Grant No. 22350710789), the Natural Science Foundation of Hunan Province, China (2022JJ30684) and the Start-up Funding of Central South University (No. 206030104). This work is supported in part by the High-Performance Computing Center of Central South University.

Notes and references

1. X. Zhang, Z. Zhang, H. Li, R. Gao, M. Xiao, J. Zhu, M. Feng and Z. Chen, Adv. Energy Mater., 2022, 12, 2201461 CrossRef CAS .
2. L. Wang, W. Chen, D. Zhang, Y. Du, R. Amal, S. Qiao, J. Wu and Z. Yin, Chem. Soc. Rev., 2019, 48, 5310–5349 RSC .
3. D. I. Armstrong McKay, A. Staal, J. F. Abrams, R. Winkelmann, B. Sakschewski, S. Loriani, I. Fetzer, S. E. Cornell, J. Rockström and T. M. Lenton, Science, 2022, 377, eabn7950 CrossRef PubMed .
4. H. Yang, M. Driess and P. W. Menezes, Adv. Energy Mater., 2021, 11, 2102074 CrossRef CAS .
5. R. M. Rodrigues, X. Guan, J. A. Iñiguez, D. A. Estabrook, J. O. Chapman, S. Huang, E. M. Sletten and C. Liu, Nat. Catal., 2019, 2, 407–414 CrossRef CAS .
6. X. Hu, X. Liu, X. Hu, C. Zhao, Q. Guan and W. Li, Adv. Funct. Mater., 2023, 33, 2214215 CrossRef CAS .
7. D. A. Salvatore, C. M. Gabardo, A. Reyes, C. P. O’Brien, S. Holdcroft, P. Pintauro, B. Bahar, M. Hickner, C. Bae and D. Sinton, Nat. Energy, 2021, 6, 339–348 CrossRef CAS .
8. X. Li, J. Yu, M. Jaroniec and X. Chen, Chem. Rev., 2019, 119, 3962–4179 CrossRef CAS PubMed .
9. B. Franta, Nat. Climate Change, 2018, 8, 1024–1025 CrossRef .
10. P. De Luna, C. Hahn, D. Higgins, S. A. Jaffer, T. F. Jaramillo and E. H. Sargent, Science, 2019, 364, eaav3506 CrossRef CAS PubMed .
11. W. Lai, Z. Ma, J. Zhang, Y. Yuan, Y. Qiao and H. Huang, Adv. Funct. Mater., 2022, 32, 2111193 CrossRef CAS .
12. P. Gao, S. Li, X. Bu, S. Dang, Z. Liu, H. Wang, L. Zhong, M. Qiu, C. Yang, J. Cai, W. Wei and Y. Sun, Nat. Chem., 2017, 9, 1019–1024 CrossRef CAS PubMed .
13. S. Jin, Z. Hao, K. Zhang, Z. Yan and J. Chen, Angew. Chem., Int. Ed., 2021, 60, 20627–20648 CrossRef CAS PubMed .
14. J. Chen and L. Wang, Adv. Mater., 2022, 34, e2103900 CrossRef PubMed .
15. G. Jia, Y. Wang, M. Sun, H. Zhang, L. Li, Y. Shi, L. Zhang, X. Cui, T. W. B. Lo, B. Huang and J. C. Yu, J. Am. Chem. Soc., 2023, 145, 14133–14142 CrossRef CAS PubMed .
16. W. Zhu, S. Kattel, F. Jiao and J. G. Chen, Adv. Energy Mater., 2019, 9, 1802840 CrossRef .
17. P. F. Sui, M. R. Gao, S. Liu, C. Xu, M. N. Zhu and J. L. Luo, Adv. Funct. Mater., 2022, 32, 2203794 CrossRef CAS .
18. H. Li, H. Huang, Y. Chen, F. Lai, H. Fu, L. Zhang, N. Zhang, S. Bai and T. Liu, Adv. Mater., 2023, 35, e2209242 CrossRef PubMed .
19. Y. J. Ko, J. Y. Kim, W. H. Lee, M. G. Kim, T. Y. Seong, J. Park, Y. Jeong, B. K. Min, W. S. Lee, D. K. Lee and H. S. Oh, Nat. Commun., 2022, 13, 2205 CrossRef CAS PubMed .
20. Q. Wu, C. Liu, X. Su, Q. Yang, X. Wu, H. Zou, B. Long, X. Fan, Y. Liao, L. Duan, Z. Quan and S. Luo, ACS Nano, 2023, 17, 402–410 CrossRef CAS PubMed .
21. Y. Wang, P. Han, X. Lv, L. Zhang and G. Zheng, Joule, 2018, 2, 2551–2582 CrossRef CAS .
22. P. P. Yang, X. L. Zhang, P. Liu, D. J. Kelly, Z. Z. Niu, Y. Kong, L. Shi, Y. R. Zheng, M. H. Fan, H. J. Wang and M. R. Gao, J. Am. Chem. Soc., 2023, 15, 8714–8725 CrossRef PubMed .
23. A. Xu, S.-F. Hung, A. Cao, Z. Wang, N. Karmodak, J. E. Huang, Y. Yan, A. Sedighian Rasouli, A. Ozden, F.-Y. Wu, Z.-Y. Lin, H.-J. Tsai, T.-J. Lee, F. Li, M. Luo, Y. Wang, X. Wang, J. Abed, Z. Wang, D.-H. Nam, Y. C. Li, A. H. Ip, D. Sinton, C. Dong and E. H. Sargent, Nat. Catal., 2022, 5, 1081–1088 CrossRef CAS .
24. X. Bai, L. Shi, Q. Li, C. Ling, Y. Ouyang, S. Wang and J. Wang, Energy Environ. Mater., 2021, 5, 892–898 CrossRef .
25. Y. J. Sa, C. W. Lee, S. Y. Lee, J. Na, U. Lee and Y. J. Hwang, Chem. Soc. Rev., 2020, 49, 6632–6665 RSC .
26. R. Cao, ChemSusChem, 2022, 15, e202201788 CrossRef CAS PubMed .
27. B. Ren, Z. Zhang, G. Wen, X. Zhang, M. Xu, Y. Weng, Y. Nie, H. Dou, Y. Jiang, Y. P. Deng, G. Sun, D. Luo, L. Shui, X. Wang, M. Feng, A. Yu and Z. Chen, Adv. Mater., 2022, 34, e2204637 CrossRef PubMed .
28. Z. Wei, J. Ding, X. Duan, G.-L. Chen, F.-Y. Wu, L. Zhang, X. Yang, Q. Zhang, Q. He, Z. Chen, J. Huang, S.-F. Hung, X. Yang and Y. Zhai, ACS Catal., 2023, 13, 4711–4718 CrossRef CAS .
29. B. Yang, K. Liu, H. Li, C. Liu, J. Fu, H. Li, J. E. Huang, P. Ou, T. Alkayyali, C. Cai, Y. Duan, H. Liu, P. An, N. Zhang, W. Li, X. Qiu, C. Jia, J. Hu, L. Chai, Z. Lin, Y. Gao, M. Miyauchi, E. Cortes, S. A. Maier and M. Liu, J. Am. Chem. Soc., 2022, 144, 3039–3049 CrossRef CAS PubMed .
30. Q. Chen, K. Liu, Y. Zhou, X. Wang, K. Wu, H. Li, E. Pensa, J. Fu, M. Miyauchi, E. Cortés and M. Liu, Nano Lett., 2022, 22, 6276–6284 CrossRef CAS PubMed .
31. A. Xu, N. Govindarajan, G. Kastlunger, S. Vijay and K. Chan, Acc. Chem. Res., 2022, 55, 495–503 CrossRef CAS PubMed .
32. T. Mairegger, H. Li, C. Griesser, D. Winkler, J. Filser, N. G. Hormann, K. Reuter and J. Kunze-Liebhauser, ACS Catal., 2023, 13, 5780–5786 CrossRef CAS PubMed .
33. W. Ni, Y. Guan, H. Chen, Y. Zhang, S. Wang and S. Zhang, Angew. Chem., Int. Ed., 2023, 135, e202303233 CrossRef .
34. Y. Sha, J. Zhang, X. Cheng, M. Xu, Z. Su, Y. Wang, J. Hu, B. Han and L. Zheng, Angew. Chem., Int. Ed., 2022, 61, e202200039 CrossRef CAS PubMed .
35. A. Fortunati, F. Risplendi, M. Re Fiorentin, G. Cicero, E. Parisi, M. Castellino, E. Simone, B. Iliev, T. J. S. Schubert, N. Russo and S. Hernandez, Commun. Chem., 2023, 6, 84 CrossRef CAS PubMed .
36. S. Dongare, O. K. Coskun, E. Cagli, K. Y. C. Lee, G. Rao, R. D. Britt, L. A. Berben and B. Gurkan, ACS Catal., 2023, 13, 7812–7821 CrossRef CAS PubMed .
37. D. V. Vasilyev, S. Shyshkanov, E. Shirzadi, S. A. Katsyuba, M. K. Nazeeruddin and P. J. Dyson, ACS Appl. Mater. Interfaces, 2020, 3, 4690–4698 CAS .
38. M. C. O. Monteiro, F. Dattila, B. Hagedoorn, R. García-Muelas, N. López and M. T. M. Koper, Nat. Catal., 2021, 4, 654–662 CrossRef CAS .
39. P. Sebastián-Pascual, A. S. Petersen, A. Bagger, J. Rossmeisl and M. Escudero-Escribano, ACS Catal., 2021, 11, 1128–1135 CrossRef .
40. S. Banerjee, X. Han and V. S. Thoi, ACS Catal., 2019, 9, 5631–5637 CrossRef CAS .
41. W. Ren, X. Tan, X. Chen, G. Zhang, K. Zhao, W. Yang, C. Jia, Y. Zhao, S. C. Smith and C. Zhao, ACS Catal., 2020, 10, 13171–13178 CrossRef CAS .
42. Z. Yao and R. Lin, Small, 2024, 20, e2306686 CrossRef PubMed .
43. S. J. Shin, H. Choi, S. Ringe, D. H. Won, H. S. Oh, D. H. Kim, T. Lee, D. H. Nam, H. Kim and C. H. Choi, Nat. Commun., 2022, 13, 5482 CrossRef CAS PubMed .
44. Z. Ma, Z. Yang, W. Lai, Q. Wang, Y. Qiao, H. Tao, C. Lian, M. Liu, C. Ma, A. Pan and H. Huang, Nat. Commun., 2022, 13, 7596 CrossRef CAS PubMed .
45. J. Wang, H. Yang, Q. Liu, Q. Liu, X. Li, X. Lv, T. Cheng and H. B. Wu, ACS Energy Lett., 2021, 6, 437–444 CrossRef CAS .
46. S. Sarkar, A. Maitra, S. Banerjee, V. S. Thoi and J. M. Dawlaty, J. Phys. Chem. B, 2020, 124, 1311–1321 CrossRef CAS PubMed .
47. Y. Kim, G. T. Yun, M. Kim, A. Jamal, I. Gereige, J. W. Ager, W. B. Jung and H. T. Jung, Angew. Chem., Int. Ed., 2024, 136, e202316264 CrossRef .
48. T. H. M. Pham, J. Zhang, M. Li, T. H. Shen, Y. Ko, V. Tileli, W. Luo and A. Züttel, Adv. Energy Mater., 2022, 12, 2103663 CrossRef CAS .
49. M. Chang, W. Ren, W. Ni, S. Lee and X. Hu, ChemSusChem, 2023, 16, e202201687 CrossRef CAS PubMed .
50. K. Li, S. Zou, J. Zhang, Y. Huang, L. He and X. Feng, ACS Catal., 2023, 13, 9346–9351 CrossRef CAS .
51. C. M. Gabardo, A. Seifitokaldani, J. P. Edwards, C.-T. Dinh, T. Burdyny, M. G. Kibria, C. P. O’Brien, E. H. Sargent and D. Sinton, Energy Environ. Sci., 2018, 11, 2531–2539 RSC .
52. W. Luc, B. H. Ko, S. Kattel, S. Li, D. Su, J. G. Chen and F. Jiao, J. Am. Chem. Soc., 2019, 141, 9902–9909 CrossRef CAS PubMed .
53. A. Perazio, C. E. Creissen, J. G. Rivera de la Cruz, M. W. Schreiber and M. Fontecave, ACS Energy Lett., 2023, 8, 2979–2985 CrossRef CAS .
54. 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 .
55. R. E. Vos, K. E. Kolmeijer, T. S. Jacobs, W. van der Stam, B. M. Weckhuysen and M. T. M. Koper, ACS Catal., 2023, 13, 8080–8091 CrossRef CAS PubMed .
56. B. H. Ko, B. Hasa, H. Shin, E. Jeng, S. Overa, W. Chen and F. Jiao, Nat. Commun., 2020, 11, 5856 CrossRef CAS PubMed .
57. H. Zhou, K. Liu, H. Li, M. Cao, J. Fu, X. Gao, J. Hu, W. Li, H. Pan, J. Zhan, Q. Li, X. Qiu and M. Liu, J. Colloid Interface Sci., 2019, 550, 17–47 CrossRef CAS PubMed .
58. H. C. Mi, C. Yi, M. R. Gao, M. Yu, S. Liu and J. L. Luo, ACS Appl. Mater. Interfaces, 2022, 14, 43257–43264 CrossRef CAS PubMed .
59. W. Zhu, R. Michalsky, O. Metin, H. Lv, S. Guo, C. J. Wright, X. Sun, A. A. Peterson and S. Sun, J. Am. Chem. Soc., 2013, 135, 16833–16836 CrossRef CAS PubMed .
60. X. Deng, D. Alfonso, T.-D. Nguyen-Phan and D. R. Kauffman, ACS Catal., 2022, 12, 5921–5929 CrossRef CAS .
61. D. Gao, H. Zhou, J. Wang, S. Miao, F. Yang, G. Wang, J. Wang and X. Bao, J. Am. Chem. Soc., 2015, 137, 4288–4291 CrossRef CAS PubMed .
62. R. Reske, H. Mistry, F. Behafarid, B. Roldan Cuenya and P. Strasser, J. Am. Chem. Soc., 2014, 136, 6978–6986 CrossRef CAS PubMed .
63. Y. Huang, X. Mao, G. Yuan, D. Zhang, B. Pan, J. Deng, Y. Shi, N. Han, C. Li, L. Zhang, L. Wang, L. He, Y. Li and Y. Li, Angew. Chem., Int. Ed., 2021, 60, 15844–15848 CrossRef CAS PubMed .
64. J. Gu, F. Heroguel, J. Luterbacher and X. Hu, Angew. Chem., Int. Ed., 2018, 57, 2943–2947 CrossRef CAS PubMed .
65. H. S. Jeon, I. Sinev, F. Scholten, N. J. Divins, I. Zegkinoglou, L. Pielsticker and B. R. Cuenya, J. Am. Chem. Soc., 2018, 140, 9383–9386 CrossRef CAS PubMed .
66. W. Rong, H. Zou, W. Zang, S. Xi, S. Wei, B. Long, J. Hu, Y. Ji and L. Duan, Angew. Chem., Int. Ed., 2021, 60, 466–472 CrossRef CAS PubMed .
67. S. Jeong, M. H. Choi, G. S. Jagdale, Y. Zhong, N. P. Siepser, Y. Wang, X. Zhan, L. A. Baker and X. Ye, J. Am. Chem. Soc., 2022, 144, 12673–12680 CrossRef CAS PubMed .
68. W. Zhu, Y. J. Zhang, H. Zhang, H. Lv, Q. Li, R. Michalsky, A. A. Peterson and S. Sun, J. Am. Chem. Soc., 2014, 136, 16132–16135 CrossRef CAS PubMed .
69. S. Liu, C. Sun, J. Xiao and J.-L. Luo, ACS Catal., 2020, 10, 3158–3163 CrossRef CAS .
70. S. Liu, H. Tao, L. Zeng, Q. Liu, Z. Xu, Q. Liu and J. L. Luo, J. Am. Chem. Soc., 2017, 139, 2160–2163 CrossRef CAS PubMed .
71. A. Klinkova, P. De Luna, C.-T. Dinh, O. Voznyy, E. M. Larin, E. Kumacheva and E. H. Sargent, ACS Catal., 2016, 6, 8115–8120 CrossRef CAS .
72. Z. Wang, G. Yang, Z. Zhang, M. Jin and Y. Yin, ACS Nano, 2016, 10, 4559–4564 CrossRef CAS PubMed .
73. P. F. Sui, M. R. Gao, Y. C. Wang, S. Liu and J. L. Luo, Adv. Energy Mater., 2024, 2401202 CrossRef .
74. J. Xiao, M. R. Gao, S. Liu and J. L. Luo, ACS Appl. Mater. Interfaces, 2020, 12, 31431–31438 CrossRef CAS PubMed .
75. Y. Huang, A. D. Handoko, P. Hirunsit and B. S. Yeo, ACS Catal., 2017, 7, 1749–1756 CrossRef CAS .
76. C. Hahn, T. Hatsukade, Y. G. Kim, A. Vailionis, J. H. Baricuatro, D. C. Higgins, S. A. Nitopi, M. P. Soriaga and T. F. Jaramillo, Proc. Natl. Acad. Sci. U. S. A., 2017, 114, 5918–5923 CrossRef CAS PubMed .
77. F. S. Roberts, K. P. Kuhl and A. Nilsson, Angew. Chem., Int. Ed., 2015, 127, 5179–5182 CrossRef PubMed .
78. A. Loiudice, P. Lobaccaro, E. A. Kamali, T. Thao, B. H. Huang, J. W. Ager and R. Buonsanti, Angew. Chem., Int. Ed., 2016, 55, 5789–5792 CrossRef CAS PubMed .
79. P. Iyengar, J. Huang, G. L. De Gregorio, C. Gadiyar and R. Buonsanti, Chem. Commun., 2019, 55, 8796–8799 RSC .
80. T. Zhang, Y.-F. Tang, M. Yu, S. Liu, L.-B. Liu, X.-Z. Fu, J.-L. Luo and S. Liu, Chem. Catal., 2024, 4, 100906 CrossRef CAS .
81. M. Yu, P. F. Sui, X. Z. Fu, J. L. Luo and S. Liu, Adv. Energy Mater., 2022, 13, 2203191 CrossRef .
82. L. P. Chi, Z. Z. Niu, X. L. Zhang, P. P. Yang, J. Liao, F. Y. Gao, Z. Z. Wu, K. B. Tang and M. R. Gao, Nat. Commun., 2021, 12, 5835 CrossRef CAS PubMed .
83. X. Cui, B. Gao, X. Zheng, G. Zhang, J. Wang, J. Liu, Y. Chen and Q. Xu, ChemCatChem, 2022, 14, e202201076 CrossRef CAS .
84. J. Zeng, W. Zhang, Y. Yang, D. Li, X. Yu and Q. Gao, ACS Appl. Mater. Interfaces, 2019, 11, 33074–33081 CrossRef CAS PubMed .
85. X. Bai, W. Chen, C. Zhao, S. Li, Y. Song, R. Ge, W. Wei and Y. Sun, Angew. Chem., Int. Ed., 2017, 56, 12219–12223 CrossRef CAS PubMed .
86. M. Li, N. Song, W. Luo, J. Chen, W. Jiang and J. Yang, Adv. Sci., 2023, 10, e2204579 CrossRef PubMed .
87. N. Wu, L. Xiao and L. Zhuang, J. Mater. Chem. A, 2019, 7, 20567–20573 RSC .
88. T. Gunji, H. Ochiai, T. Ohira, Y. Liu, Y. Nakajima and F. Matsumoto, Chem. Mater., 2020, 32, 6855–6863 CrossRef CAS .
89. J. Wang, G. Wang, J. Zhang, Y. Wang, H. Wu, X. Zheng, J. Ding, X. Han, Y. Deng and W. Hu, Angew. Chem., Int. Ed., 2021, 133, 7602–7606 CrossRef PubMed .
90. S. Ma, M. Sadakiyo, M. Heima, R. Luo, R. T. Haasch, J. I. Gold, M. Yamauchi and P. J. Kenis, J. Am. Chem. Soc., 2017, 139, 47–50 CrossRef CAS PubMed .
91. P. C. Chen, C. Chen, Y. Yang, A. L. Maulana, J. Jin, J. Feijoo and P. Yang, J. Am. Chem. Soc., 2023, 145, 10116–10125 CrossRef CAS PubMed .
92. S.-Q. Liu, M.-R. Gao, R.-F. Feng, L. Gong, H. Zeng and J.-L. Luo, ACS Catal., 2021, 11, 7604–7612 CrossRef CAS .
93. S. Liang, J. Xiao, T. Zhang, Y. Zheng, Q. Wang and B. Liu, Angew. Chem., Int. Ed., 2023, 135, e202310740 CrossRef .
94. X. Ma, Y. Zhang, T. Fan, D. Wei, Z. Huang, Z. Zhang, Z. Zhang, Y. Dong, Q. Hong, Z. Chen and X. Yi, Adv. Funct. Mater., 2023, 33, 2213145 CrossRef CAS .
95. Y. Wang, H. Xu, Y. Liu, J. Jang, X. Qiu, E. P. Delmo, Q. Zhao, P. Gao and M. Shao, Angew. Chem., Int. Ed., 2024, 136, e202313858 CrossRef .
96. Y. Wang, Y. Liu, W. Liu, J. Wu, Q. Li, Q. Feng, Z. Chen, X. Xiong, D. Wang and Y. Lei, Energy Environ. Sci., 2020, 13, 4609–4624 RSC .
97. X. Li, S. Xi, L. Sun, S. Dou, Z. Huang, T. Su and X. Wang, Adv. Sci., 2020, 7, 2001545 CrossRef CAS PubMed .
98. Y. Pan, R. Lin, Y. Chen, S. Liu, W. Zhu, X. Cao, W. Chen, K. Wu, W. C. Cheong, Y. Wang, L. Zheng, J. Luo, Y. Lin, Y. Liu, C. Liu, J. Li, Q. Lu, X. Chen, D. Wang, Q. Peng, C. Chen and Y. Li, J. Am. Chem. Soc., 2018, 140, 4218–4221 CrossRef CAS PubMed .
99. X. Sheng, W. Ge, H. Jiang and C. Li, Adv. Mater., 2022, 34, e2201295 CrossRef PubMed .
100. C. Wang, X. Hu, X. Hu, X. Liu, Q. Guan, R. Hao, Y. Liu and W. Li, Appl. Catal., B, 2021, 296, 120331 CrossRef CAS .
101. X. Wei, S. Wei, S. Cao, Y. Hu, S. Zhou, S. Liu, Z. Wang and X. Lu, Appl. Surf. Sci., 2021, 564, 150423 CrossRef CAS .
102. Y. Li, B. Wei, M. Zhu, J. Chen, Q. Jiang, B. Yang, Y. Hou, L. Lei, Z. Li, R. Zhang and Y. Lu, Adv. Mater., 2021, 33, e2102212 CrossRef PubMed .
103. X. Zou, C. Ma, A. Li, Z. Gao, Z. Shadike, K. Jiang, J. Zhang, Z. Huang and L. Zhu, ACS Appl. Energy Mater., 2021, 4, 9572–9581 CrossRef CAS .
104. W. Ren, X. Tan, W. Yang, C. Jia, S. Xu, K. Wang, S. C. Smith and C. Zhao, Angew. Chem., Int. Ed., 2019, 58, 6972–6976 CrossRef CAS PubMed .
105. K. Li, S. Zhang, X. Zhang, S. Liu, H. Jiang, T. Jiang, C. Shen, Y. Yu and W. Chen, Nano Lett., 2022, 22, 1557–1565 CrossRef CAS PubMed .
106. M. Qu, Z. Chen, Z. Sun, D. Zhou, W. Xu, H. Tang, H. Gu, T. Liang, P. Hu, G. Li, Y. Wang, Z. Chen, T. Wang and B. Jia, Nano Res., 2022, 16, 2170–2176 CrossRef .
107. S. Li, M. Ceccato, X. Lu, S. Frank, N. Lock, A. Roldan, X.-M. Hu, T. Skrydstrup and K. Daasbjerg, J. Mater. Chem. A, 2021, 9, 1583–1592 RSC .
108. S. Hong, H. G. Abbas, K. Jang, K. K. Patra, B. Kim, B. U. Choi, H. Song, K. S. Lee, P. P. Choi, S. Ringe and J. Oh, Adv. Mater., 2023, 35, e2208996 CrossRef PubMed .
109. P.-F. Sui, M.-R. Gao, R. Feng, S. Liu and J.-L. Luo, Chem. Catal., 2023, 3, 100842 CrossRef CAS .
110. J. Jiang, B. Huang, R. Daiyan, B. Subhash, C. Tsounis, Z. Ma, C. Han, Y. Zhao, L. H. Effendi, L. C. Gallington, J. N. Hart, J. A. Scott and N. M. Bedford, Nano Energy, 2022, 101, 107593 CrossRef CAS .
111. B. Qin, Q. Zhang, Y. H. Li, G. Yang and F. Peng, ACS Appl. Mater. Interfaces, 2020, 12, 30466–30473 CrossRef CAS PubMed .
112. C. Choi, T. Cheng, M. Flores Espinosa, H. Fei, X. Duan, W. A. Goddard, 3rd and Y. Huang, Adv. Mater., 2019, 31, e1805405 CrossRef PubMed .
113. S. Zhao, A. Liu, Y. Li, Y. Wen, X. Gao and Q. Chen, Nanomaterials, 2022, 12 CAS .
114. K. Sun, T. Cheng, L. Wu, Y. Hu, J. Zhou, A. Maclennan, Z. Jiang, Y. Gao, W. A. Goddard 3rd and Z. Wang, J. Am. Chem. Soc., 2017, 139, 15608–15611 CrossRef CAS PubMed .
115. M. Jiang, M. Zhu, H. Wang, X. Song, J. Liang, D. Lin, C. Li, J. Cui, F. Li, X. L. Zhang, Z. Tie and Z. Jin, Nano Lett., 2023, 23, 291–297 CrossRef CAS PubMed .
116. Q. Gong, P. Ding, M. Xu, X. Zhu, M. Wang, J. Deng, Q. Ma, N. Han, Y. Zhu, J. Lu, Z. Feng, Y. Li, W. Zhou and Y. Li, Nat. Commun., 2019, 10, 2807 CrossRef PubMed .
117. G. Liu, Z. Li, J. Shi, K. Sun, Y. Ji, Z. Wang, Y. Qiu, Y. Liu, Z. Wang and P. Hu, Appl. Catal., B, 2020, 260, 118134 CrossRef CAS .
118. J. Zhang, R. Yin, Q. Shao, T. Zhu and X. Huang, Angew. Chem., Int. Ed., 2019, 58, 5609–5613 CrossRef CAS PubMed .
119. Y. Wang, Z. Chen, P. Han, Y. Du, Z. Gu, X. Xu and G. Zheng, ACS Catal., 2018, 8, 7113–7119 CrossRef CAS .
120. R. G. Mariano, K. McKelvey, H. S. White and M. W. Kanan, Science, 2017, 358, 1187–1192 CrossRef CAS PubMed .
121. J. Cai, Q. Zhao, W. Y. Hsu, C. Choi, Y. Liu, J. M. P. Martirez, C. Chen, J. Huang, E. A. Carter and Y. Huang, J. Am. Chem. Soc., 2023, 145, 9136–9143 CrossRef CAS PubMed .
122. D. G. Park, J. W. Choi, H. Chun, H. S. Jang, H. Lee, W. H. Choi, B. C. Moon, K.-H. Kim, M. G. Kim, K. M. Choi, B. Han and J. K. Kang, ACS Catal., 2023, 13, 9222–9233 CrossRef CAS .
123. S. Liu, J. Xiao, X. F. Lu, J. Wang, X. Wang and X. W. D. Lou, Angew. Chem., Int. Ed., 2019, 58, 8499–8503 CrossRef CAS PubMed .
124. X. Feng, K. Jiang, S. Fan and M. W. Kanan, J. Am. Chem. Soc., 2015, 137, 4606–4609 CrossRef CAS PubMed .
125. J. Wang, C. Cheng, B. Huang, J. Cao, L. Li, Q. Shao, L. Zhang and X. Huang, Nano Lett., 2021, 21, 980–987 CrossRef CAS PubMed .
126. Y. Li, F. Cui, M. B. Ross, D. Kim, Y. Sun and P. Yang, Nano Lett., 2017, 17, 1312–1317 CrossRef CAS PubMed .
127. J. Li, S. U. Abbas, H. Wang, Z. Zhang and W. Hu, Nanomicro Lett., 2021, 13, 216 CAS .
128. J. Huang, M. Mensi, E. Oveisi, V. Mantella and R. Buonsanti, J. Am. Chem. Soc., 2019, 141, 2490–2499 CrossRef CAS PubMed .
129. P. Luan, X. Dong, L. Liu, J. Xiao, P. Zhang, J. Zhang, H. Chi, Q. Wang, C. Ding, R. Li and C. Li, ACS Catal., 2024, 14, 8776–8785 CrossRef CAS .
130. B. Ren, G. Wen, R. Gao, D. Luo, Z. Zhang, W. Qiu, Q. Ma, X. Wang, Y. Cui, L. Ricardez-Sandoval, A. Yu and Z. Chen, Nat. Commun., 2022, 13, 2486 CrossRef CAS PubMed .
131. Z. Cai, Y. Wu, Z. Wu, L. Yin, Z. Weng, Y. Zhong, W. Xu, X. Sun and H. Wang, ACS Energy Lett., 2018, 3, 2816–2822 CrossRef CAS .
132. H. Liu, B. Li, Z. Liu, Z. Liang, H. Chuai, H. Wang, S. N. Lou, Y. Su, S. Zhang and X. Ma, ACS Catal., 2023, 13, 5033–5042 CrossRef CAS .
133. Y.-F. Tang, T. Zhang, H.-C. Mi, M. Yu, P.-F. Sui, X.-Z. Fu, J.-L. Luo and S. Liu, Small Sci., 2023, 4, 2300169 CrossRef .
134. Y. F. Tang, S. Liu, M. Yu, P. F. Sui, X. Z. Fu, J. L. Luo and S. Liu, ACS Appl. Mater. Interfaces, 2024, 16(31), 41669–41676 CrossRef CAS PubMed .
135. D. Gao, Y. Zhang, Z. Zhou, F. Cai, X. Zhao, W. Huang, Y. Li, J. Zhu, P. Liu, F. Yang, G. Wang and X. Bao, J. Am. Chem. Soc., 2017, 139, 5652–5655 CrossRef CAS PubMed .
136. Z. Zhang, G. Wen, D. Luo, B. Ren, Y. Zhu, R. Gao, H. Dou, G. Sun, M. Feng, Z. Bai, A. Yu and Z. Chen, J. Am. Chem. Soc., 2021, 143, 6855–6864 CrossRef CAS PubMed .
137. T. Zhang, S. Xu, D. L. Chen, T. Luo, J. Zhou, L. Kong, J. Feng, J. Q. Lu, X. Weng, A. J. Wang, Z. Li, Y. Su and F. Yang, Angew. Chem., Int. Ed., 2024, e202407748 CAS .
138. S. Jensen, R. Cheula, M. Hedevang, M. Andersen and J. V. Lauritsen, Angew. Chem., Int. Ed., 2024, e202405554 CAS .
139. P.-F. Sui, S. Liu, C. Xu, J. Xiao, N. Duan, R. Feng and J.-L. Luo, Chem. Eng. J., 2022, 427, 131654 CrossRef CAS .
140. J. Wu, Y. Xie, S. Du, Z. Ren, P. Yu, X. Wang, G. Wang and H. Fu, Sci. China Mater., 2020, 63, 2314–2324 CrossRef CAS .
141. Q. Zhu, C. J. Murphy and L. R. Baker, J. Am. Chem. Soc., 2022, 144, 2829–2840 CrossRef CAS PubMed .
142. D. Yang, J. Wang, Q. Wang, Z. Yuan, Y. Dai, C. Zhou, X. Wan, Q. Zhang and Y. Yang, ACS Nano, 2022, 16, 15681–15704 CrossRef CAS PubMed .
143. 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 .
144. Z. Cao, J. S. Derrick, J. Xu, R. Gao, M. Gong, E. M. Nichols, P. T. Smith, X. Liu, X. Wen, C. Coperet and C. J. Chang, Angew. Chem., Int. Ed., 2018, 57, 4981–4985 CrossRef CAS PubMed .
145. H. K. Lim, H. Shin, W. A. Goddard, 3rd, Y. J. Hwang, B. K. Min and H. Kim, J. Am. Chem. Soc., 2014, 136, 11355–11361 CrossRef CAS PubMed .
146. 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 .
147. Z. Wang, L. Wu, K. Sun, T. Chen, Z. Jiang, T. Cheng and W. A. Goddard 3rd, J. Phys. Chem. Lett., 2018, 9, 3057–3061 CrossRef CAS PubMed .
148. X. Kong, J. Zhao, Z. Xu, Z. Wang, Y. Wu, Y. Shi, H. Li, C. Ma, J. Zeng and Z. Geng, J. Am. Chem. Soc., 2023, 145, 14903–14911 CrossRef CAS PubMed .
149. 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. H. Luo, M. J. Cheng, X. Wen and W. Liu, Angew. Chem., Int. Ed., 2018, 57, 12675–12679 CrossRef CAS PubMed .
150. Y. Qiu, H. Zhong, W. Xu, T. Zhang, X. Li and H. Zhang, J. Mater. Chem. A, 2019, 7, 5453–5462 RSC .
151. 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 .
152. H. Shang, S. K. Wallentine, D. M. Hofmann, Q. Zhu, C. J. Murphy and L. R. Baker, Chem. Sci., 2020, 11, 12298–12306 RSC .
153. J. R. Pankhurst, Y. T. Guntern, M. Mensi and R. Buonsanti, Chem. Sci., 2019, 10, 10356–10365 RSC .
154. Y. Lin, T. Wang, L. Zhang, G. Zhang, L. Li, Q. Chang, Z. Pang, H. Gao, K. Huang, P. Zhang, Z. J. Zhao, C. Pei and J. Gong, Nat. Commun., 2023, 14, 3575 CrossRef CAS PubMed .
155. H. A. Younus, N. Ahmad, W. Ni, X. Wang, M. Al-Abri, Y. Zhang, F. Verpoort and S. Zhang, Coord. Chem. Rev., 2023, 493, 215318 CrossRef CAS .
156. I. Azcarate, C. Costentin, M. Robert and J. M. Saveant, J. Am. Chem. Soc., 2016, 138, 16639–16644 CrossRef CAS PubMed .
157. I. Azcarate, C. Costentin, M. Robert and J.-M. Savéant, J. Phys. Chem. C, 2016, 120, 28951–28960 CrossRef CAS .
158. W. Nie, D. E. Tarnopol and C. C. L. McCrory, J. Am. Chem. Soc., 2021, 143, 3764–3778 CrossRef CAS PubMed .
159. C. Costentin, M. Robert and J.-M. Savéant, Science, 2012, 338, 90–94 CrossRef CAS PubMed .
160. J. Han, N. Wang, X. Li, H. Lei, Y. Wang, H. Guo, X. Jin, Q. Zhang, X. Peng, X.-P. Zhang, W. Zhang, U.-P. Apfel and R. Cao, eScience, 2022, 2, 623–631 CrossRef .
161. C. G. Margarit, C. Schnedermann, N. G. Asimow and D. G. Nocera, Organometallics, 2018, 38, 1219–1223 CrossRef .
162. H. Guo, Z. Liang, K. Guo, H. Lei, Y. Wang, W. Zhang and R. Cao, Chinese J. Catal., 2022, 43, 3089–3094 CrossRef CAS .
163. K. Guo, X. Li, H. Lei, H. Guo, X. Jin, X. P. Zhang, W. Zhang, U. P. Apfel and R. Cao, Angew. Chem., Int. Ed., 2022, 61, e202209602 CrossRef CAS PubMed .
164. C. Kim, F. Dionigi, V. Beermann, X. Wang, T. Moller and P. Strasser, Adv. Mater., 2019, 31, e1805617 CrossRef PubMed .
165. Q. Hu, Z. Han, X. Wang, G. Li, Z. Wang, X. Huang, H. Yang, X. Ren, Q. Zhang, J. Liu and C. He, Angew. Chem., Int. Ed., 2020, 59, 19054–19059 CrossRef CAS PubMed .
166. Y. Zhu, P. Li, X. Yang, M. Wang, Y. Zhang, P. Gao, Q. Huang, Y. Wei, X. Yang, D. Wang, Y. Shen and M. Wang, Adv. Energy Mater., 2023, 13, 2204143 CrossRef CAS .
167. 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 .
168. M. K. Kim, H. Lee, J. H. Won, W. Sim, S. J. Kang, H. Choi, M. Sharma, H. S. Oh, S. Ringe, Y. Kwon and H. M. Jeong, Adv. Funct. Mater., 2021, 32, 2107349 CrossRef .
169. T.-T. Zhuang, Y. Pang, Z.-Q. Liang, Z. Wang, Y. Li, C.-S. Tan, J. Li, C. T. Dinh, P. De 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 .
170. C. Liu, M. Zhang, J. Li, W. Xue, T. Zheng, C. Xia and J. Zeng, Angew. Chem., Int. Ed., 2022, 61, e202113498 CrossRef CAS PubMed .
171. K. Chang, X. Jian, H. M. Jeong, Y. Kwon, Q. Lu and M. J. Cheng, J. Phys. Chem. Lett., 2020, 11, 1896–1902 CrossRef CAS PubMed .
172. Z. Wang, Y. Li, X. Zhao, S. Chen, Q. Nian, X. Luo, J. Fan, D. Ruan, B. Q. Xiong and X. Ren, J. Am. Chem. Soc., 2023, 145, 6339–6348 CrossRef CAS PubMed .
173. B. B. Zhang, Y. H. Wang, S. M. Xu, K. Chen, Y. G. Yang and Q. H. Kong, RSC Adv., 2020, 10, 19192–19198 RSC .
174. Y. Pan, H. Li, J. Xiong, Y. Yu, H. Du, S. Li, Z. Wu, S. Li, J. Lai and L. Wang, Appl. Catal., B, 2022, 306, 121111 CrossRef CAS .
175. C. Li, Z. Guo, Z. Liu, T. Zhang, H. Shi, J. Cui, M. Zhu, L. Zhang, H. Li, H. Li and C. Li, ACS Catal., 2023, 13, 16114–16125 CrossRef CAS .
176. F. Pan, X. Duan, L. Fang, H. Li, Z. Xu, Y. Wang, T. Wang, T. Li, Z. Duan and K. J. Chen, Adv. Energy Mater., 2023, 14, 2303118 CrossRef .
177. X. Li, P. Zhang, L. Zhang, G. Zhang, H. Gao, Z. Pang, J. Yu, C. Pei, T. Wang and J. Gong, Chem. Sci., 2023, 14, 5602–5607 RSC .
178. R. P. Jansonius, L. M. Reid, C. N. Virca and C. P. Berlinguette, ACS Energy Lett., 2019, 4, 980–986 CrossRef CAS .
179. H. Huang, H. Jia, Z. Liu, P. Gao, J. Zhao, Z. Luo, J. Yang and J. Zeng, Angew. Chem., Int. Ed., 2017, 56, 3594–3598 CrossRef CAS PubMed .
180. E. L. Clark, C. Hahn, T. F. Jaramillo and A. T. Bell, J. Am. Chem. Soc., 2017, 139, 15848–15857 CrossRef CAS PubMed .
181. T. Kim, R. E. Kumar, J. A. Brock, E. E. Fullerton and D. P. Fenning, ACS Catal., 2021, 11, 6662–6671 CrossRef CAS .
182. Q. Lei, L. Huang, J. Yin, B. Davaasuren, Y. Yuan, X. Dong, Z. P. Wu, X. Wang, K. X. Yao, X. Lu and Y. Han, Nat. Commun., 2022, 13, 4857 CrossRef CAS PubMed .
183. J. Hao, Z. Zhuang, J. Hao, K. Cao, Y. Hu, W. Wu, S. Lu, C. Wang, N. Zhang, D. Wang, M. Du and H. Zhu, ACS Nano, 2022, 16, 3251–3263 CrossRef CAS PubMed .
184. J. Li, J. Pan, W. Yin, Y. Cai, H. Huang, Y. He, G. Gong, Y. Yuan, C. Fan, Q. Zhang and L. Wang, Chin. Chem. Lett., 2023, 34, 108049 CrossRef CAS .
185. M. Liu, Y. Pang, B. Zhang, P. De Luna, O. Voznyy, J. Xu, X. Zheng, C. T. Dinh, F. Fan, C. Cao, F. P. 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 .
186. S. Kim, W. J. Dong, S. Gim, W. Sohn, J. Y. Park, C. J. Yoo, H. W. Jang and J.-L. Lee, Nano Energy, 2017, 39, 44–52 CrossRef CAS .
187. F. Y. Gao, S. J. Hu, X. L. Zhang, Y. R. Zheng, H. J. Wang, Z. Z. Niu, P. P. Yang, R. C. Bao, T. Ma, Z. Dang, Y. Guan, X. S. Zheng, X. Zheng, J. F. Zhu, M. R. Gao and S. H. Yu, Angew. Chem., Int. Ed., 2020, 59, 8706–8712 CrossRef CAS PubMed .
188. J. Zosel, W. Oelßner, M. Decker, G. Gerlach and U. Guth, Meas. Sci. Technol., 2011, 22, 072001 CrossRef .
189. B. Deng, M. Huang, X. Zhao, S. Mou and F. Dong, ACS Catal., 2021, 12, 331–362 CrossRef .
190. C.-T. Dinh, T. Burdyny, M. G. Kibria, A. Seifitokaldani, C. M. Gabardo, F. P. García de Arquer, A. Kiani, J. P. Edwards, P. De Luna and O. S. Bushuyev, Science, 2018, 360, 783–787 CrossRef CAS PubMed .
191. J. Wang, Y. Qin, S. Jin, Y. Yang, J. Zhu, X. Li, X. Lv, J. Fu, Z. Hong, Y. Su and H. B. Wu, J. Am. Chem. Soc., 2023, 145, 26213–26221 CrossRef CAS PubMed .
192. X. Tan, C. Yu, X. Song, L. Ni, H. Xu, Y. Xie, Z. Wang, S. Cui, Y. Ren, W. Li, Y. Zhang and J. Qiu, Nano Energy, 2022, 104, 107957 CrossRef CAS .
193. J. Yang, J. Jiao, S. Liu, Y. Yin, Y. Cheng, Y. Wang, M. Zhou, W. Zhao, X. Tong, L. Jing, P. Zhang, X. Sun, Q. Zhu, X. Kang and B. Han, Angew. Chem., Int. Ed., 2024, e202410145 Search PubMed .
194. T. Yuan, T. Wang, G. Zhang, W. Deng, D. Cheng, H. Gao, J. Zhao, J. Yu, P. Zhang and J. Gong, Chem. Sci., 2022, 13, 8117–8123 RSC .
195. H. Liu, T. Yan, S. Tan, L. Sun, Z. Zhang, S. Hu, S.-H. Li, X. Kang, Y. Lei, L. Jiang, T. Hou, L. Liu, Q. Yu and B. Liu, J. Am. Chem. Soc., 2024, 146, 5333–5342 CrossRef CAS PubMed .
196. W. Guo, S. Liu, X. Tan, R. Wu, X. Yan, C. Chen, Q. Zhu, L. Zheng, J. Ma, J. Zhang, Y. Huang, X. Sun and B. Han, Angew. Chem., Int. Ed., 2021, 60, 21979–21987 CrossRef CAS PubMed .
197. L. Lu, X. Sun, J. Ma, D. Yang, H. Wu, B. Zhang, J. Zhang and B. Han, Angew. Chem., Int. Ed., 2018, 57, 14149–14153 CrossRef CAS PubMed .
198. Z. Min, B. Chang, C. Shao, X. Su, N. Wang, Z. Li, H. Wang, Y. Zhao, M. Fan and J. Wang, Appl. Catal., B, 2023, 326, 122185 CrossRef CAS .
199. C. Jiang, S. Zeng, X. Ma, J. Feng, G. Li, L. Bai, F. Li, X. Ji and X. Zhang, AIChE J., 2022, 69, e17859 CrossRef .
200. D. Yang, Q. Zhu, C. Chen, H. Liu, Z. Liu, Z. Zhao, X. Zhang, S. Liu and B. Han, Nat. Commun., 2019, 10, 677 CrossRef CAS PubMed .
201. F.-X. Shen, J. Shi, F. Shi, T.-Y. Chen, Y.-F. Li, Q.-Y. Li, Y.-N. Dai, B. Yang and T. Qu, Ionics, 2018, 25, 1921–1927 CrossRef .
202. C. Jiang, A. W. Nichols, J. F. Walzer and C. W. Machan, Inorg. Chem., 2020, 59, 1883–1892 CrossRef CAS PubMed .
203. G. P. Lau, M. Schreier, D. Vasilyev, R. Scopelliti, M. Gratzel and P. J. Dyson, J. Am. Chem. Soc., 2016, 138, 7820–7823 CrossRef CAS PubMed .
204. W. Chen, X. Jin, L. Zhang, L. Wang and J. Shi, Adv. Sci., 2024, 11, e2304424 CrossRef PubMed .
205. D. L. Bhugun and J.-M. Savéant, J. Am. Chem. Soc., 1994, 116, 5015–5016 CrossRef .
206. D. L. Hammouche, M. Momenteau and J.-M. Savéant, J. Am. Chem. Soc., 1991, 113, 8455–8466 CrossRef .
207. C. Costentin, S. Drouet, G. Passard, M. Robert and J. M. Saveant, J. Am. Chem. Soc., 2013, 135, 9023–9031 CrossRef CAS PubMed .
208. M. Konig, J. Vaes, E. Klemm and D. Pant, iScience, 2019, 19, 135–160 CrossRef PubMed .
209. S. Dongare, M. Zeeshan, A. S. Aydogdu, R. Dikki, S. F. Kurtoglu-Oztulum, O. K. Coskun, M. Munoz, A. Banerjee, M. Gautam, R. D. Ross, J. S. Stanley, R. S. Brower, B. Muchharla, R. L. Sacci, J. M. Velazquez, B. Kumar, J. Y. Yang, C. Hahn, S. Keskin, C. G. Morales-Guio, A. Uzun, J. M. Spurgeon and B. Gurkan, Chem. Soc. Rev., 2024 10.1039/d4cs00390j .
210. M. Moura de Salles Pupo and R. Kortlever, ChemPhysChem, 2019, 20, 2926–2935 CrossRef CAS PubMed .
211. S. Kaneco, K. Iiba, H. Katsumata, T. Suzuki and K. Ohta, J. Solid State Electrochem., 2006, 11, 490–495 CrossRef .
212. Y. Tomita, S. Teruya, O. Koga and Y. Hori, J. Electrochem. Soc., 2000, 147, 4164 CrossRef CAS .
213. A. Aljabour, H. Coskun, D. H. Apaydin, F. Ozel, A. W. Hassel, P. Stadler, N. S. Sariciftci and M. Kus, Appl. Catal., B, 2018, 229, 163–170 CrossRef CAS .
214. 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 .
215. L. Sun, G. K. Ramesha, P. V. Kamat and J. F. Brennecke, Langmuir, 2014, 30, 6302–6308 CrossRef CAS PubMed .
216. J. T. Feaster, A. L. Jongerius, X. Liu, M. Urushihara, S. A. Nitopi, C. Hahn, K. Chan, J. K. Norskov and T. F. Jaramillo, Langmuir, 2017, 33, 9464–9471 CrossRef CAS PubMed .
217. F. Zhou, S. Liu, B. Yang, P. Wang, A. S. Alshammari and Y. Deng, Electrochem. Commun., 2014, 46, 103–106 CrossRef CAS .
218. J. Gu, S. Liu, W. Ni, W. Ren, S. Haussener and X. Hu, Nat. Catal., 2022, 5, 268–276 CrossRef CAS .
219. 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 .
220. M. R. Singh, Y. Kwon, Y. Lum, J. W. Ager 3rd and A. T. Bell, J. Am. Chem. Soc., 2016, 138, 13006–13012 CrossRef CAS PubMed .
221. O. Ayemoba and A. Cuesta, ACS Appl. Mater. Interfaces, 2017, 9, 27377–27382 CrossRef CAS PubMed .
222. J. Li, X. Li, C. M. Gunathunge and M. M. Waegele, Proc. Natl. Acad. Sci. U. S. A., 2019, 116, 9220–9229 CrossRef CAS PubMed .
223. A. Murata and Y. Hori, Bull. Chem. Soc. Jpn., 1991, 64, 123–127 CrossRef CAS .
224. J. N. Mills, I. T. McCrum and M. J. Janik, Phys. Chem. Chem. Phys., 2014, 16, 13699–13707 RSC .
225. D. Strmcnik, K. Kodama, D. V. D. van der Vliet, J. Greeley, V. R. Stamenkovic and N. Marković, Nat. Chem., 2009, 1, 466–472 CrossRef CAS PubMed .
226. Y. Qiao, W. Lai, K. Huang, T. Yu, Q. Wang, L. Gao, Z. Yang, Z. Ma, T. Sun, M. Liu, C. Lian and H. Huang, ACS Catal., 2022, 12, 2357–2364 CrossRef CAS .
227. E. Perez-Gallent, G. Marcandalli, M. C. Figueiredo, F. Calle-Vallejo and M. T. M. Koper, J. Am. Chem. Soc., 2017, 139, 16412–16419 CrossRef CAS PubMed .
228. K. Guo, H. Lei, X. Li, Z. Zhang, Y. Wang, H. Guo, W. Zhang and R. Cao, Chinese J. Catal., 2021, 42, 1439–1444 CrossRef CAS .
229. A. Schizodimou and G. Kyriacou, Electrochim. Acta, 2012, 78, 171–176 CrossRef CAS .
230. J. Resasco, Y. Lum, E. Clark, J. Z. Zeledon and A. T. Bell, ChemElectroChem, 2018, 5, 1064–1072 CrossRef CAS .
231. M. Cho, J. T. Song, S. Back, Y. Jung and J. Oh, ACS Catal., 2018, 8, 1178–1185 CrossRef CAS .
232. Y.-C. Hsieh, S. D. Senanayake, Y. Zhang, W. Xu and D. E. Polyansky, ACS Catal., 2015, 5, 5349–5356 CrossRef CAS .
233. D. Gao, F. Scholten and B. Roldan Cuenya, ACS Catal., 2017, 7, 5112–5120 CrossRef CAS .
234. P. Ding, H. An, P. Zellner, T. Guan, J. Gao, P. Müller-Buschbaum, B. M. Weckhuysen, W. van der Stam and I. D. Sharp, ACS Catal., 2023, 13, 5336–5347 CrossRef CAS PubMed .
235. J. Li, Y. Kuang, X. Zhang, W.-H. Hung, C.-Y. Chiang, G. Zhu, G. Chen, F. Wang, P. Liang and H. Dai, Nat. Catal., 2023, 6, 1151–1163 CrossRef CAS .
236. R. E. Vos and M. T. M. Koper, ChemElectroChem, 2022, 9 Search PubMed .
237. S. W. Jo, J. Y. Kim, M. W. Lee, Y. Kim and H. S. Ahn, ACS Catal., 2023, 13, 5122–5126 CrossRef CAS .
238. Q. Chang, J. H. Lee, Y. Liu, Z. Xie, S. Hwang, N. S. Marinkovic, A. A. Park, S. Kattel and J. G. Chen, JACS Au, 2022, 2, 214–222 CrossRef CAS PubMed .
239. K. Hara, A. Kudo, T. Sakata and M. Watanabe, J. Electrochem. Soc., 1995, 142, L57 CrossRef CAS .
240. K. Hara and T. Sakata, J. Electrochem. Soc., 1997, 144, 539 CrossRef CAS .
241. Y. Hori and A. Murata, Electrochim. Acta, 1990, 35, 1777–1780 CrossRef CAS .
242. K. Hara, A. Kudo and T. Sakata, J. Electroanal. Chem., 1997, 421, 1–4 CrossRef CAS .
243. M. G. Kibria, J. P. Edwards, C. M. Gabardo, C. T. Dinh, A. Seifitokaldani, D. Sinton and E. H. Sargent, Adv. Mater., 2019, 31, e1807166 CrossRef PubMed .
244. S. T. Ahn, I. Abu-Baker and G. T. R. Palmore, Catal. Today, 2017, 288, 24–29 CrossRef CAS .
245. X. Wang, A. Xu, F. Li, S. F. Hung, D. H. Nam, C. M. Gabardo, Z. Wang, Y. Xu, A. Ozden, A. S. Rasouli, A. H. Ip, D. Sinton and E. H. Sargent, J. Am. Chem. Soc., 2020, 142, 3525–3531 CrossRef CAS PubMed .
246. M. Jouny, J. J. Lv, T. Cheng, B. H. Ko, J. J. Zhu, W. A. Goddard, 3rd and F. Jiao, Nat. Chem., 2019, 11, 846–851 CrossRef CAS PubMed .
247. A. Goyal, G. Marcandalli, V. A. Mints and M. T. M. Koper, J. Am. Chem. Soc., 2020, 142, 4154–4161 CrossRef CAS PubMed .
248. L. Yuan, L. Zhang, J. Feng, C. Jiang, J. Feng, C. Li, S. Zeng and X. Zhang, Chem. Eng. J., 2022, 450, 138378 CrossRef CAS .
249. Y. C. Tan, K. B. Lee, H. Song and J. Oh, Joule, 2020, 4, 1104–1120 CrossRef CAS .
250. C. F. C. Lim, D. A. Harrington and A. T. Marshall, Electrochim. Acta, 2017, 238, 56–63 CrossRef CAS .

This journal is © The Royal Society of Chemistry 2024