Optimization of electronic structure by defect engineering for electrocatalytic carbon dioxide reduction reaction

Abstract

The electrochemical CO₂ reduction reaction (eCO₂RR) serves as an effective method to mitigate greenhouse gas emissions and convert them into valuable chemicals. For the practical application of the eCO₂RR, it is crucial to utilize an electrocatalyst that exhibits both high activity and stability. Research has shown that defect engineering can precisely modulate the key intermediates involved in catalysis, thereby influencing the activity, selectivity, and stability of the reaction products. This paper reviews the advancements in copper (Cu)-based materials for eCO₂RR, specifically examining the effects of doping, vacancies, surface engineering, twin crystals, and heterojunctions. It describes the specific changes in the resulting defect structures and their impact on the eCO₂RR process, with particular emphasis on changes in electron behavior. Finally, the challenges and potential future directions are briefly discussed, such as the operando characterization methodologies, and machine learning (ML) for accelerating materials discovery.

---

Jinghan He

Jinghan He is currently a PhD student in Prof. Xiangdong Yao's group, State Key Laboratory of Inorganic Synthesis and Preparative Chemistry, College of Chemistry, Jilin University (JLU). Her research interests focus on the design and synthesis of defective Cu-based electrocatalysts for electrocatalytic carbon dioxide reduction.

Jianxin Qiang

Jianxin Qiang is currently a PhD student in the Department of Analytical Chemistry, College of Chemistry, Jilin University (JLU). Her research interests focus on the design and synthesis of carbon-based catalytic materials.

Yang-Fan Xu

Yang-Fan Xu received his bachelor's degree and PhD degree from Sun Yat-sen University in 2013 and 2018, respectively. He is now an associate professor in the School of Advanced Energy, Sun Yat-sen University. His current research interest focuses on the catalyst design and mechanism study of photothermal catalytic CO2 hydrogenation reactions.

Zhan Shi

Zhan Shi received his PhD degree in Inorganic Chemistry from Jilin University in 2002. He is currently a professor at the State Key Laboratory of Inorganic Synthesis & Preparative Chemistry of Jilin University. In recent years, his research interests have mainly focused on coordination chemistry, crystal engineering and nano chemistry.

Keke Huang

Keke Huang received his PhD degree in Inorganic Chemistry from Jilin University in 2007. He is currently a professor at the State Key Laboratory of Inorganic Synthesis & Preparative Chemistry of Jilin University. His main research field is inorganic solid chemistry. In recent years, his research interests have mainly focused on solid-state chemistry, composite materials, and related energy applications.

Xiangdong Yao

Prof. Xiangdong Yao received his PhD degree in Materials Engineering from the University of Queensland (Australia) in 2005. He joined Griffith University in 2009 as an Associate Professor then a full Professor from 2013, and was the group leader of advanced energy materials. In July 2022, he joined Sun Yat-sen University as a Full Professor and Dean of the School of Advanced Energy, and Prof. Yao's current research focuses on energy materials and technology, especially hydrogen energy and defect electrocatalysis.

---

1. Introduction

In recent decades, rapid social progress and extensive combustion of fossil fuels have led to a continuous elevation in atmospheric carbon dioxide concentrations, significantly contributing to the greenhouse effect and exacerbating global warming issues.^1–5 According to statistics from the Energy & Climate Intelligence Unit, over 130 countries and territories have set targets for “zero carbon” or “carbon neutrality”. Consequently, effective measures must be implemented to reduce CO₂ concentration in the atmosphere.^6,7 To this end, various strategies are being employed to mitigate the effect of CO₂, including CO₂ capture, simulating photosynthesis to convert CO₂ into valuable chemicals, and developing renewable resources to replace fossil fuels, all of which aim at achieving a carbon-neutral cycle.^8–10

The eCO₂RR has a broad application scenario to convert CO₂ into value-added products.¹¹ At present, industrial processes like power generation, chemical synthesis, etc., usually release a relatively high concentration of carbon dioxide, which thereby facilitates an easier, energy-effective and cost-effective carbon dioxide capture.^12–15 Compared with carbon dioxide storage, the conversion of carbon dioxide into value-added chemicals by electrocatalysis using renewable electricity can be considered an attractive process with both economic and environmental benefits.^16,17 For example, the carbon dioxide can be electrocatalytically converted into formic acid. Formic acid is, on one hand, an important raw chemical molecule used in the chemical industry, and on the other hand, also a good liquid hydrogen storage carrier, so as to facilitate the storage and transportation of hydrogen.^18–21 In addition, carbon dioxide electrocatalytic reduction can produce CO, and CO has a very wide range of industrial uses, such as when coupled with hydrogen through Fischer–Tropsch synthesis to fabricate various chemicals and fuels.^22,23 Carbon dioxide electrocatalytic reduction can also directly synthesize alkenes and alkanes that can replace the petrochemical process to produce gasoline or to synthesize monomers for polymer synthesis.

In current eCO₂RR research, Ag, Au, Zn, Bi and Cu have been extensively studied,^24–35 among which Cu-based materials are receiving more attention because of their advantages like moderate price, excellent conductivity, element abundance and ability for C₂₊ product synthesis. However, Cu-based catalysts also face some challenges in eCO₂RR, such as poor stability and low selectivity for a single product. Several strategies have been developed to settle the above issues, represented by the defect engineering strategy.

Defect structures in carbon materials have been the subject of extensive study and continuous development over the past decade. Initially, heteroatom doping such as sulfur (S) was reported to enhance the catalytic performance of graphene.³⁶ However, the origin of electrocatalytic activity in heteroatom-modified graphene remains a topic of debate. For example, some researchers argue that the introduced heteroatoms serve as the active sites in the doped graphene structure,³⁷ while others contend that the true activity centers are the defective structures caused by the incorporation of foreign elements, with heteroatoms primarily facilitating the redistribution of the original environment of graphene.³⁸ Recent work by Yao's team on defect-facilitated electrocatalysis has lent strong support to the latter mechanism, through experimental and theoretical design and study of a series of intrinsic defects (such as vacancy and topological defects) in defective graphene.^39,40 As a result, a defect-derived catalytic mechanism has been established. Subsequently, Yao et al. further introduced the concept of complex defects, significantly enhancing the electrocatalytic activity of intrinsic defective graphene and the stability of heteroatoms (such as atomic metal atoms) through the synergistic effect between heteroatoms/isolated metal species and defects. Intrinsic defective graphene has also been used as an ideal substrate for the fabrication of high-efficiency electrocatalysts. Therefore, it is necessary to carry out research on other electrocatalytic directions in defect engineering.

Herein, the impact on the electronic structure and adsorption behavior in eCO₂RR over Cu-based materials will be discussed, highlighting the structure–activity relationship. The representative configurations of the main defective Cu-based catalysts are depicted in Fig. 1. Importantly, the focus will be on unraveling the relationship between defect structure and catalytic performance, addressing the unique electrochemical properties of different defect structures. In summary, the ultimate goal is to aid researchers designing more efficient Cu-based catalysts that can sustainably produce high-value chemicals and reduce the environmental impact of CO₂ emissions; this direction not only promises a new avenue for renewable energy utilization, but also aims to deepen the understanding and application of defect engineering in the precise design of catalyst materials.

Fig. 1 Schematic diagram of the defect engineering for Cu-based catalysts.

2. Fundamentals of CO₂ electroreduction

To fully analyze the role played by defect structures in defective Cu-based materials for eCO₂RR, a deep understanding of the reaction route for generating different products and the key intermediates involved is essential. Given that metallic Cu and Cu-based materials facilitate the formation of a variety of products during CO₂ reduction, the design of these products involves multiple coordination chemical controls, including common C₁ and C₂ products.^41–43 The C₁ products involve fewer electrons, whereas C₂ products involve more electron transfers. The specific reaction formulas are outlined in Table 1.⁴⁴

Table 1 The standard potentials for reducing CO₂ in aqueous solution at standard conditions (1.0 atm and 25 °C)

Half electrochemical thermodynamic reactions	E° (V vs. SHE pH = 7)	E° (V vs. RHE pH = 7)
CO2 + e^− → CO2˙^−	−1.90	−1.48
2H^+ + 2e^− → H2(g)	−0.42	0.00
CO2 + 2H^+ + 2e^− → HCOOH(l)	−0.55	−0.13
CO2 + 2H^+ + 2e^− → CO(g) + H2O	−0.52	−0.10
CO2 + 4H^+ + 4e^− → HCHO(l) + H2O	−0.48	−0.06
CO2 + 4H^+ + 4e^− → CH3OH(l) + H2O	−0.39	+0.03
CO2 + 8H^+ + 8e^− → CH4(g) + 2H2O	−0.25	+0.17
CO2 + 12H^+ + 12e^− → C2H4(g) + 4H2O	−0.38	+0.08
CO2 + 12H^+ + 12e^− → C2H5OH(l) + 3H2O	−0.35	+0.09
CO2 + 14H^+ + 14e^− → C2H6(g) + 4H2O	−0.28	+0.14
CO2 + 18H^+ + 18e^− → C3H7OH(l) + 5H2O	−0.30	+0.10

---

It's important to note that eCO₂RR is a highly complex process involving multiple electron and proton transfer steps across various reaction pathways, occurring at the three-phase interfaces.^45,46 Here, we break down the reaction steps into three crucial stages.

2.1 CO₂ adsorption and activation

The adsorption and activation of CO₂ at the active site are critical for the subsequent reduction reaction, as they directly determine the key intermediates and reaction products. Optimizing these steps can also help inhibit competing reactions, primarily the HER, at the optimal energy level.^47,48 Generally, CO₂ exhibits two adsorption states on catalysts: one is the formation of linear molecules through physical adsorption, and the other is the formation of charged species though chemical adsorption^49–51 (Fig. 2a and b). When only electrons are involved in the activation process, CO₂^δ− intermediates typically form on catalysts with C atoms as bonding sites; if both protons and electrons are involved, the intermediate with C as the bonding atom is *COOH. This intermediate can undergo hydroxyl removal to form *CO, which can be further converted to various products, including C₁ and C₂ products.⁵² Conversely, if the O atom acts as the bonding atom instead of the C atom, a *OCOH intermediate is formed, which is subsequently converted to HCOOH.⁵³

Fig. 2 (a and b) The potential structures of adsorbed CO₂^δ˙⁻ on catalysts surface. Reproduced from ref. 51 with permission from Elsevier, copyright 2008. (c) Possible reaction pathways for the electrocatalytic reduction of CO₂ to CO. Reproduced from ref. 53 with permission from Elsevier, copyright 2017 (d) reaction schemes of major pathways considered for CO reduction toward C₁ and C₂₊ products. Reproduced from ref. 56 with permission from Springer Nature, copyright 2019. (e) Proposed mechanism for the reduction of CO to C₂ products at high potentials on Cu (100). Reproduced from ref. 54 with permission from the American Chemical Society, copyright 2018. (f) Schematic representations of the species involved in the pathways to different products. Reproduced from ref. 61 with permission from the WILEY, copyright 2013. (g) Lowest kinetic pathways for the eight-electron reduction of CO to C₂H₄. Reproduced from ref. 63 with permission from PNAS, copyright 2017.

2.2 *CO intermediate formation

The intermediates produced by the adsorption and activation processes are crucial for the subsequent reduction reaction. The most important intermediate is *CO, which is produced from the *COOH intermediate through a proton-coupled electron transfer (PCET) process that removes the H₂O molecule. When the *CO adsorption energy is low, the *CO species can be directly desorbed from the catalyst (e.g., Ag and Au) to form CO gas (this is a typical mechanism of electrochemical CO₂RR to form CO) (Fig. 2c).^53,54 Alternatively, according to the Sabatier principle,⁵⁴ a moderate *CO adsorption energy (e.g., on Cu) can lead to further coupling and formation of C–C bonds, which promotes the subsequent production of C₂ products. Additionally, the coverage of *CO is critical for the conversion of CO₂ to C₂, as high coverage of *CO occupies most of the catalytic active sites, reducing hydrogen adsorption and inhibiting HER, while increasing the chance of C–C coupling, thereby enhancing the conversion efficiency of CO₂ to C₂.⁵⁵

2.3 C–C coupling

One of the most distinctive capabilities of Cu-based electrocatalysts for CO₂ reduction is their ability to facilitate the formation of multi-carbon products through C–C coupling. However, there is significant debate regarding the mechanism of this C–C coupling, particularly whether it involves concurrent proton and electron transfers. Fig. 2d shows the formation paths of some key intermediates.⁵⁶ In addition, several intermediates have been proposed as critical nodes where the mechanistic pathways diverge towards either the ethylene (C₂H₄) pathway or the ethanol (C₂H₅OH) pathway. These intermediates include *COCO, *COCHO, *CHCOH, *CCH, and *CH₂CHO.

The first pathway is the direct dimerization of adsorbed *CO intermediates to produce *CO*CO intermediates by C–C coupling. Specifically, Kim et al. found by time-resolved attenuated total reflective surface-enhanced infrared absorption spectroscopy (ATR-SEIRAS) that C–C coupling occurs only by *CO dimerization to *OCCO, without the involvement of *CHO.⁵⁷ In the actual measurements, it was found that CO dimerization and CO adsorption occurred simultaneously, while the kinetics of proton coupling reduction to *CHO were relatively slow. In other mechanistic studies, it has been found that the dimerization of *CO–CO* is a chemical coupling reaction that does not involve the transfer of electrons and protons, in which case the Tafel slope implies that the rate-limiting step is to protonation of *OCCO to *OCCOH and the production of C₂. Chan et al. found that when the pH increases from 7 to 13, the overpotential of the C₂ product changes by about 0.36 V, which means that the C₂ activity increases by more than three orders of magnitude.⁵⁶ Specific analyses found that the generation of C₂ at low overpotential was limited by the rate at which the first proton–electron was transferred to the OCCO* intermediate, resulting in a traditional standard hydrogen electrode scales dependence, while at high overpotential, it was limited by CO coverage. In contrast, Hori et al. found that the partial current density of C₂H₄ and C₂H₅OH was exponentially related to the applied potential, regardless of pH.⁵⁸ Therefore, the relationship between the activity produced by C₂ and pH needs to be further discussed.

The second possible reaction route is the direct coupling of *CO intermediates to protons to generate *CHO intermediates or *COH intermediates.^54,59 Some studies have shown that *CO, in combination with protonated *CHO intermediates, is more stable for adsorption on high-potential Cu (100) surfaces, while Cu (111) is preferred over the full potential range. Specifically, Garza et al. proposed a roadmap for the production of C₂ compounds on the Cu surface based on experimental results and density functional theory (DFT) simulations.⁵⁴ As shown in Fig. 2e, *COCHO is a key intermediate that determines the selectivity between C₂H₄ and C₂H₅OH on the Cu(100) surface, and can be further converted to acetic acid and C₂H₄via the C₂H₄ pathway, or to acetaldehyde and C₂H₅OH via the C₂H₅OH pathway. Koper et al. used Fourier transform infrared spectroscopy to detect *OCCOH in a LiOH solution with low overpotential, an intermediate that was only observed during CO reduction on Cu (100) and not on Cu (111). Specifically, the bands at 1191 and 1584 cm⁻¹ are caused by C–O–H and C–O tensile vibrations and, in the simplest case, correspond to hydrodimers (*OCCOH) or, in general, to a combination of this adsorbate with other C₁ and C₂ adsorbents. This work shows that *OCCOH is the most stable intermediate formed by hydrogenation of CO on Cu (100).⁶⁰

Koper and Bell et al. proposed that the CH₂CHO* species is the selectivity-determining intermediate, which originates from the proton and electron transfer of the OC**COH or *OCHCHO* intermediates.^54,61 As shown in Fig. 2f, CH₂CHO* hydrogenolysis produces C₂H₄, while CH₂CHO* hydrogenation produces CH₃CHO*, which in turn produces C₂H₅OH.⁶¹ Koper and co-workers showed that the barrier for C₂H₄ formation is 0.2 eV lower than the barrier for the formation of C₂H₅OH via CH₃CHO*. In addition, Wang et al. pointed out that *CH₂CHO intermediates can form *OCHCH*O species through dimerization between *CHO intermediates, which are then converted to CH₂ and CH*O₂ through H to remove O through the PCET process, and finally produce C₂H₄.⁶²

Finally, *CHCOH intermediates and *CCH were proposed by Goddard III and colleagues.⁶³ They used the Eley–Rideal (ER) mechanism and the Langmuir–Hinshelwood (LH) mechanisms to explain. Specifically, as shown in Fig. 2g, the dihydroxylation of *COH–COH results in the formation of *C–COH, and when *C–COH is reduced to *CH–COH, *CH–COH is dehydroxylated to *CH–C by ER. *CH–C is reduced to *CH₂–C instead of *CH–CH because *CH₂–C formation energy is lower than *CH–CH formation energy. Finally, C₂H₄ is formed from *CH₂–CH. In addition, *CH*COH species are key intermediates for the production of C₂H₄ and C₂H₅OH, with *CH*COH losing its OH group to form *C*CH, which is then hydrogenated to form C₂H₄, while *CH*COH is directly hydrogenated to form *CHCHOH, which is then hydrogenated to produce C₂H₅OH.

3. Defect engineering for Cu-based catalysts

A variety of Cu-based materials are designed as eCO₂RR electrocatalysts, including metallic Cu, Cu-based oxides, hydroxides, sulfides, alloys, and other composites.^34,64–69 A common challenge with bulk Cu-based materials is the limited number of catalytically active sites exposed, which necessitates higher overpotentials and reaction energies to drive the eCO₂RR. This often results in competitive HER and poor single-product catalytic performance. To address these issues, defect engineering is employed to modify the electronic structure and surface properties of Cu-based materials. This approach has proved to be an effective strategy to enhance catalytic performance for eCO₂RR. Unlike traditional categorizations based on synthesis, composition, or morphology, this focus on defect engineering strategies relates directly to their impact on catalytic activity.

3.1 Doping

Heteroatom doping is recognized as a potent strategy to tune the physicochemical properties of electrocatalysts. Introducing heteroatoms can disrupt the crystal lattice's periodicity and adjust the electronic structure, leading to improved adsorption behavior of reaction intermediates and enhanced electrocatalytic performance. It is worth noting that the type of dopant and doping degree should be carefully modified because excessive doping may produce serious defects or lattice distortion inside or on the surface of the material, resulting in a lower charge transport efficiency. Therefore, it is important to adopt an appropriate doping for eCO₂RR when intending to promote charge transfer, change the local charge distribution of the catalyst, and strengthen the bonding between the electron-rich active centers.⁷⁰ This section delves into how heteroatom doping, including both metal and non-metal elements, regulates the electronic structure and its relationship with the adsorption behavior and electrocatalytic performance of key organic intermediates.

3.1.1 Cationic doping. Sargent et al. highlighted the use of adsorption energy as a descriptor of eCO₂RR activity through a comprehensive screening of metal-doped Cu catalysts (X–Cu, where X includes Cr, Mo, W, Mn, Re, Fe, Ru, Co, Rh, Ir, Ni, Pt, and Pd) under acidic conditions.⁷¹ The results showed that Pd has a strong affinity for *CO, while the values of ΔG_OCCOH* (Gibbs free energy) and ΔG_OCCOH* − ΔG_CHO* were lowest, showing the best C₂₊ selectivity. Further analysis showed that the introduction of Pd into Cu could enhance the local coverage of *CO, and reduced the desorption of *CO, which is beneficial for the C–C coupling reaction. In addition, the high affinity of CO* competed with the H* active site, thus weakening the H binding energy and inhibiting the HER process, further improving the selectivity to C₂₊ products over Pd–Cu. The experimental results showed that Pd–Cu catalyst synthesized under acidic conditions had a single-pass carbon efficiency of up to 60% for the conversion of CO₂ to C₂₊ product at a current density of 500 mA cm⁻² (Fig. 3a and b).⁷¹ Li et al. synthesized a bimetallic Cu/Pd catalyst through a two-step method, achieving impressive eCO₂RR performance at high current densities with a Faraday efficiency of 66.2% and a partial current density of 463.2 mA cm⁻². DFT calculations results showed that the adsorption energy of Cu/Pd bimetallic catalyst for *CO intermediates was more negative than that of pure Cu, indicating that Cu/Pd bimetallic catalyst had stronger adsorption of *CO. This enhanced *CO adsorption capacity could improve the coverage of *CO on the catalyst surface, thereby promoting C–C coupling. In addition, DFT analysis also found that the Cu/Pd catalyst enhanced the adsorption of *H intermediates, which was consistent with the high C₂H₅OH/C₂H₄ faradaic efficiency observed in the experiments. Further analysis of Gibbs free energy changes showed that Cu/Pd bimetallic catalyst promoted the asymmetric coupling reaction of *CO–*CHO by enhancing the adsorption of *H and *CO intermediates, which facilitated the selective formation of C₂₊ products (Fig. 3c and d).⁷² Zhang et al. introduced a novel Pd–Cu₃N catalyst featuring charge-separated Pd^δ−–Cu^δ+ atomic pairs, stabilizing the Cu^δ+ site.⁷³In situ characterization and DFT analyses revealed that the negatively charged Pd^δ− site, along with the adjacent Cu^δ+ site, displayed an excellent *CO binding capacity, synergistically promoting the *CO dimerization process to produce C₂ products. The activation energy and enthalpy change of the CO dimerization reaction were further compared, and it was found that the Pd^δ−–Cu⁰ APs (atom pairs) (100) surface was the most favorable for CO dimerization (Fig. 3e and f).⁷³ The faradaic efficiency of C₂ product on Pd–Cu₃N increased from 5.6% to 78.2%, making a 14-fold enhancement. In addition, Xiao et al. developed the Pd–Cu₂O catalyst through in situ electrodeposition by incorporating Pd into Cu₂O, achieving a faradaic efficiency of 63.8% for C₂H₄, 1.5 times higher than that for Cu₂O at the same potential.⁷⁴ This improvement was attributed to the Pd site facilitating electron transfer from Cu₂O to Pd, stabilizing Cu⁺ in Cu₂O by reducing the local electron density, and stabilizing *CO intermediates, thereby accelerating C–C coupling kinetics and C₂H₄ generation. Zhou et al. created a catalyst with three distinct catalytic sites by introducing atomically dispersed Pd onto Cu. DFT calculation of reaction free energy difference between Cu (100) and Pd–Cu(100) was further studied. On pure Cu (100), hydrogenation of *CO to *COH was the rate-determine step, which limited the further conversion of CO. On Pd–Cu (100) surface, the introduction of Pd atoms promoted the hydrogenation of *CO, resulting in a more active reaction site for CO conversion⁷⁵ (Fig. 3g and h). The synergistic effect of the three catalytic sites significantly improved the selectivity and activity for C₂₊ products, with the faradaic efficiency of the C₂₊ products reaching 80.8% at a high current density of 0.88% Pd–Cu at 0.8 A cm⁻². In addition, at 1.4 A cm⁻², the FE of the C₂₊ product was above 70%.⁷⁵

Fig. 3 (a) Two-dimensional C₂₊ activity and C₂₊versus C₁ selectivity plot of CO₂RR to C₁ and C₂₊ products. (b) Free energy diagram of CO₂RR via the CHO pathway and the OCCOH pathway. Solid and dashed lines represent Pd–Cu and Cu, respectively. Reproduced from ref. 71 with permission from Springer Nature, copyright 2022. (c) The adsorption energy of *H and *CO intermediates on Cu and Cu/Pd model. (d) The Gibbs free energy of various C–C coupling processes on Cu and Cu/Pd model. Reproduced from ref. 72 with permission from Wiley, copyright 2023. (e) CO adsorption energies of different Cu catalysts. (f) Energy profile of the CO dimerization step on Cu(111), Cu(100), reduced Cu^δ+ sites (100), and reduced Pd^δ−–Cu^δ+ APs (100). Reproduced from ref. 73 with permission from the American Chemical Society, copyright 2023. (g) Calculated ΔG for the CO₂ reduction to *OCCOH on Cu (100) model. (h) Calculated ΔG for the CO₂ reduction to *OCCOH on Pd–Cu (100) model. Reproduced from ref. 75 with permission from Wiley, copyright 2024. (i) Distribution of CO binding energy for overall surface (from 0% to 30%) and Cu_0.80Zn_0.20 NPs. (j) Blue: the relationship between ΔE_*CO and the charge transfer to *CO. Orange: the relationship between the ΔE_*CO and the –ICOHP of the M–C bond (the inset figure shows the charge transfer between the active site and *CO). Reproduced from ref. 77 with permission from Wiley, copyright 2022. (k) Reaction free energy. (l) Desorption energy of C₂H₅OH and C₂H₄ products on different models and the difference between the two desorption energies. Reproduced from ref. 79 with permission from American Chemical Society, copyright 2023.

In addition to Pd, the incorporation of other metals such as Ce, Zn, Pb, Al, Au, and Ag into Cu-based catalysts has shown significant effects on the performance of eCO₂RR. Each of these dopants modifies the catalyst's properties in unique ways, influencing the reduction pathways and product selectivity. Shan et al. introduced Ce into Cu nanoparticles using a one-step method, enhancing the faradaic efficiency of C₂H₄ to reach 53% at 150 mA cm⁻², which was 2.8 times higher than that of Cu nanoparticles (NP). The improvement was largely attributed to the electronic effects of Ce atoms, which reduced particle size and increased the number of catalytic sites. Ce doping also stabilized CuO_x and enhanced the adsorption of *CO, promoting C–C coupling and increasing the selectivity for C₂H₄ production.⁷⁶ In addition, the study by Zhen et al. found that in CuZn bimetallic catalysts, the *CO adsorption free energy (ΔE_*CO) at the Zn-doped site was weaker than that at the pure Cu site, which altered the *CO adsorption to an optimal range for subsequent C₂₊ synthesis. Further analysis showed that the adsorption energy of *CO was linearly related to the charge transfer from the CuZn active site to *CO. In addition, the bonding properties between active sites and *CO intermediates were explored by crystal orbital Hamilton family (COHP) and composite COHP (ICOHP) analysis. The results showed that the CuZn synergistic effect promoted the transfer of electrons from the active site to *CO, thus regulating the interaction between the catalyst surface and *CO intermediates, and eventually promoting selectivity of C₂₊ products in eCO₂RR⁷⁷ (Fig. 3i and j). In order to solve the problem of inefficiency and instability caused by the change of the valence state of Cu in the process of electroreduction, stabilization by doping with other metals is an effective method; for example, Ma et al. designed a Pb-doped Cu₂O catalyst with controllable Cu⁰–Cu⁺ sites, which showed a faradaic efficiency of 83.9% at −1.1 V vs. RHE and a current density of 203.8 mA cm⁻². In situ X-ray absorption spectroscopy (XAS) and Raman spectroscopy showed that Pb doping in Cu₂O could stabilize the Cu⁰–Cu⁺ structure, enhanced *CO adsorption and promoted C–C coupling, thus improving C₂ product formation. The Gibbs free energy analysis found that the addition of Pb reduced the reaction energy barrier from *CO₂ to *COOH and promoted the C–C coupling process.⁷⁸ Zhang et al. could adjust the oxyphilic of Cu-based catalysts by introducing Al, enhancing the electroreduction of CO₂ to C₂₊ alcohols. Al–Cu/Cu₂O showed 84.5% C₂₊ faradaic efficiency, with significant stability, and the FE of C₂₊ alcohol reached 55.2%. The corresponding current density and formation rate of C₂₊ alcohol was 354.2 mA cm⁻² and 1066.8 μmol cm⁻² h⁻¹, respectively. The ratio of alcohol to C₂H₄ was more than 2, which was significantly higher than 0.91 of Cu/Cu₂O. DFT further revealed that doping Al in Cu would affect the cleavage of C–O and Cu–C bonds, thus facilitating the C₂H₅OH synthesis pathway over the C₂H₄ synthesis pathway. By comparing the free energy difference from *CH₂CHO with *CH₃CHO or *C₂H₄ in the two models, it was found that the addition of Al dopant changed the affinity of oxygen-containing intermediates at the Cu site and enhanced the ethanol formation path. In addition, calculation of desorption energy of the target product also showed that Al doping was beneficial for reducing the energy gap between the *CH₃CHO step and *C₂H₄, thus favoring the selective formation of C₂H₅OH⁷⁹ (Fig. 3k and l). Wei et al. showed a faradaic efficiency of nearly 65.3% for multi-carbon products by doping a small amount of Au nanoparticles on the surface of Cu nanowires, which was much higher than that of pure Cu NWs (39.7%). In addition, the *CO intermediates generated by the Au site overflowed to the Cu site, and at the same time, the interfacial electrons transferred from Cu to Au induced electron-deficient Cu, which was conducive to the adsorption of *CO, and the Cu_99.3–Au_0.7 nanowire had the highest d-band center, thereby enhancing the adsorption of *CO intermediates and further producing multi-carbon products.⁸⁰ Different from the previous increase in the adsorption of *CO, Wang et al. modified Ag to cubic Cu₂O, and the faradaic efficiency and energy efficiency of Cu₂O/Ag_2.3% to C₂H₅OH were 40.8% and 22.4%, respectively, and Cu₂O/Ag_2.3% showed asymmetric C–C coupling to stabilize the reaction intermediate, thereby increasing C₂H₅OH yield at high current density. In situ studies confirmed that the redispersion of silver in Cu significantly optimized the coordination number and oxide state of Cu. In this way, the CO binding strength changed, resulting in a hybrid adsorption structure that initiated asymmetric C–C coupling to stabilize the C₂H₅OH intermediate, thereby increasing C₂H₅OH yield.⁸¹

In summary, the strategic doping of metal heteroatoms into Cu-based catalysts significantly affects their electronic structure and the adsorption behavior of reaction intermediates. While all these dopants increase the *CO concentration on the catalyst surface, the specific structure of the parent Cu-based catalysts leads to varied reduction products. The doping of heteroatoms not only influences the CO₂ reduction pathway but also stabilizes the oxidation state of Cu and adjusts the oxygen affinity on the surface, enhancing the adsorption of reactive oxygen species and *CO intermediates.

3.1.2 Anionic doping. In addition to cationic doping, anionic doping plays a crucial role in modulating the performance and stability of catalyst atomic sites. On one hand, doping with elements such as O, N, S, and other elements can alter the electronic structure of active sites, thereby regulating the adsorption/desorption processes of intermediates by promoting electron transfer or changing charge distribution, ultimately tuning the catalytic performance.^82–85 On the other hand, employing halogen doping can polarize the surrounding atoms, significantly change the electron transfer and band center of the active site, and thus affect the adsorption behavior of reaction intermediates.⁸⁶

For example, Cheng et al. introduced Si into CuO by a selective dissolution method to generate Cu–O–Si units to stabilize the Cu⁺/Cu⁰ site. In situ Raman spectroscopy and DFT calculations confirmed that Cu⁺/Cu⁰ at the CuO–Si interface was the active site to promote the conversion of eCO₂RR to C₂ products, and the faradaic efficiency of C₂ products increased to 81.9% at −100 mA cm⁻². Si incorporation significantly improved the hybridization of O 2p and Cu 3d orbitals, thereby enhancing the Cu–O bond and stabilizing the Cu⁺/Cu⁰ site, which was essential to promote C–C coupling by decreasing the energy barrier of *OCCO formation and enhancing C₂ product selection.⁸⁷ As shown in Fig. 4a, the *CO adsorption energy related to C–C coupling on Cu₂O@Cu was lower than those on CuO and Cu, indicating that Cu₂O@Cu structure could enhance the C–C coupling process. The projected state density (PDOS) analysis (Fig. 4b) further studied the electronic structure between the active site and the adsorption intermediate. After the adsorption of *OCCO, the electron density of the Cu d orbital in the Cu₂O@Cu structure changed significantly, creating a notable density state migration to the Fermi level, indicating the electron transfer from the Cu⁺/Cu⁰ site to *OCCO to support the C–C coupling mechanism. Liang et al. synthesized S-doped Cu₂O electrocatalyst, and in situ surface reconstruction of S-doped Cu₂O was performed to generate active S-adsorbed metal Cu sites, suppress H₂ and CO pathways and promote HCOOH formation. S–Cu₂O catalyst at −1.2 V vs. RHE had the highest faradaic efficiency of 81.4%, and the HCOOH yield was 7.0 times higher than Cu₂O. Fig. 4c and d compare the Gibbs free energies of HCOOH and CO formation pathways for eCO₂RR adsorbed on Cu (111) and S-adsorbed Cu (111). For the CO formation path, the potential limiting step on both surfaces was the formation of *COOH, but the barrier of *COOH formation on S-adsorbed Cu (111) was significantly higher. On the other hand, the CO₂-to-HCOOH pathway on Cu (111) and S-adsorption Cu (111) had different potential limiting steps; the former was *OCHO to HCOOH, the latter was *CO₂ to *OCHO. In addition, the CO₂-to-HCOOH reaction barrier on S-adsorbing Cu (111) was lower than that on Cu (111). As shown in Fig. 4e and f, charge transfer occurred on both surfaces after adsorption of *COOH and *OCHO. Compared with Cu (111), S-adsorbed Cu (111) enabled more electron transfer from Cu (111) to *OCHO, while fewer electrons were transferred to *COOH. This suggests that S-adsorption could stabilize *OCHO intermediates and inhibit *COOH formation, thereby favoring HCOOH production over CO.⁸⁸

Fig. 4 (a) Activation energies for C–C coupling on CuO, Cu and Cu₂O@Cu. (b) The pDOS of Cu d orbitals of Cu₂O@Cu with/without *OCCO adsorption. Reproduced from ref. 87 with permission from American Chemical Society, copyright 2024. Free energy diagram for eCO₂RR on (c) Cu(111) and (d) S-adsorbed Cu(111). The PDOS of (e) Cu(111) and (f) S-adsorbed Cu(111). Reproduced from ref. 88 with permission from Wiley, copyright 2023. (g) Computed oxygen vacancy formation energy in B–Cu₂O (111) and Cu₂O (111) models. (h) PDOS plots of Cu 3d and O 2p for B–Cu₂O and Cu₂O. Charge density difference for (i) B–Cu₂O and (j) Cu₂O models after one electron injection. Reproduced from ref. 91 with permission from Royal Society of Chemistry, copyright 2022. (k) Schematic diagram of eCO₂RR to C₂H₄ and C₂H₅OH over fluorine-modified copper. Reproduced from ref. 62 with permission from Springer Nature, copyright 2020. (l) Gibbs free energy of C–C coupling on Cu, Cu₂O and Cu–Cu₂O. (m) A proposed reaction mechanism for the eCO₂RR to C₂H₄. Reproduced from ref. 94 with permission from Wiley, copyright 2022. (n) Schematic diagram of copper–copper iodide catalyst. Reproduced from ref. 95 with permission from Wiley, copyright 2021.

Li et al. proved through experiments that Cu doped with boron and phosphorus is more favorable to eCO₂RR. Compared with undoped Cu electrodes, linear adsorbed *CO intermediates on Cu–B and Cu–P electrodes were more easily converted into C₂₊ products. Specifically, P doping enhanced the formation of CH₄ and C₂₊ products, while B doping significantly improved the selectivity of C₂₊/C₁.⁸⁹ Kong et al. prepared a series of P-doped Cu catalysts to electro-reduce CO₂ to C₂₊ products by adjusting the surface electronic structure of Cu. It was found that when the molar ratio of P to Cu was 8.3%, the faradaic efficiency of C₂₊ product reached 64%. The C₂₊ current density of Cu_0.92P_0.08 was the highest at 300 mA cm⁻², reaching 176 mA cm⁻². The specific analysis showed that the introduction of P could stabilize the surface Cu^δ+ species and improve the activity of C₂₊ products by adjusting the adsorption intensity of *CO. The mechanistic studies showed that with the increase of the molar ratio of P to Cu, the center of the d band shifted upward, the bonding state of Cu decreased, while the antibonding state increased at the same time, the interaction between the adsorbate and the Cu surface was enhanced, the adsorption strength of *CO was enhanced, and the weak adsorption of *CO on Cu could promote the desorption of *CO and reduce the coverage of *CO and the possibility of C–C coupling.⁹⁰ In addition to the above-mentioned doped P stabilizing the valence state of Cu, this can also be achieved by doping with B; for example, Yao et al. successfully stabilized the active Cu⁺ species with high application potential by adding B to the Cu₂O lattice. The C₂H₄/CO ratio on B–Cu₂O catalyst at −1.2 V was 2.5, which was about 3.5 times higher than that on Cu₂O catalyst. Fig. 4g shows that the oxygen vacancy formation energy of B–Cu₂O was significantly higher than Cu₂O, indicating that B-doping enhanced the Cu–O bond and made it more difficult for lattice oxygen to escape from the surface. In Fig. 4h, it was found that the binding states of O 2p and Cu 3d in B–Cu₂O had a higher overlap degree, indicating that B doping enhanced the binding between Cu and O atoms. Fig. 4i and j shows that during the reduction process, the B–Cu₂O could effectively buffer the attack of excess electrons on the Cu–O bond, while the Cu₂O easily caused the charge to accumulate around the Cu atom and destroy the Cu–O bond. This shows that the electrophilic effect of B atom could protect the Cu atom from electron attack, inhibit the removal of lattice oxygen, and improve the stability of catalyst.⁹¹ Similarly, Zhou et al. used hexagonal boron nitride (h-BN) nanosheets to modify Cu₂O nanoparticles to stabilize Cu⁺. Compared with Cu₂O, the C₂H₄/CO ratio in the eCO₂RR of Cu₂O-BN increased by 1.62 times. Experimental and theoretical studies confirmed that there was a strong electronic interaction between the two components in Cu₂O–BN, which strengthened the Cu–O bond. Electrophilic h-BN received partial electron density from Cu₂O, protected Cu–O bonds from electron attack during eCO₂RR, and stabilized Cu⁺ species during long-term electrolysis. The well-retained Cu⁺ species enhanced the selectivity of the C₂ product and improved the stability of Cu₂O-BN.⁹²

In addition to the doping mentioned above, halogens are an effective means of improving eCO₂RR performance; for example, Ma et al. modified fluorine on the Cu (F–Cu) surface to have a current density of 1.6 A cm⁻² and a C₂ faradaic efficiency of 80% in the flow cell (Fig. 4k). The incorporation of fluorine enhanced water activation, CO adsorption, and hydrogenation of adsorbed *CO into higher active *CHO intermediates.⁶² Yan et al. introduced a F–Cu catalyst to improve the efficiency of C₂ product formation, and at high current densities, the F–Cu catalyst for C₂ products exhibited a faradaic efficiency of 70.4%, which was mainly attributed to the altered behavior of *CO intermediates. The results showed that at low potentials, CO₂ can be reduced to CO, and the adsorption of CO on F–Cu could be enhanced, while the adsorption potential of *CO on F–Cu was lower, which promoted the production of *CO and enabled it to be stable on the surface instead of escaping in gaseous form. At the same time, the addition of fluorine also optimized the hydrophobicity of the F–Cu catalyst to a certain extent, thereby slightly inhibiting the competitive HER.⁹³ In addition to F ions, other halogen ions also have a significant promotion effect on eCO₂RR, such as the AgI–CuO catalyst reported by Yang et al. At −1.00 V_RHE, the faradaic efficiency of C₂₊ product on AgI–CuO electrocatalyst was about 68.9%, and the partial current density of C₂₊ was about 18.2 mA cm⁻². DFT results showed that the C–C coupling process from CO to the O*CCO intermediate at the Cu–Cu₂O heterojunction was more favorable than that on individual Cu and Cu₂O. This was because the strong electronegative iodide adsorbed on Ag surface promoted the activation and conversion of CO₂ molecules to CO, and formed a high local CO coverage on the modified Ag surface, and further promoted the C–C coupling reaction via the local spillover of CO. In addition, the orbital distribution and bond orientation adjustment at the Cu⁰/Cu⁺ interface significantly reduced the energy barrier of C–C coupling, thus improving the selectivity of eCO₂RR for C₂₊ products⁹⁴ (Fig. 4l and m). Li et al. designed a defective Cu–CuI composite catalyst with an enriched Cu⁰/Cu⁺ interface by physically mixing Cu nanoparticles and CuI powders. A significant C₂ current density of 591 mA cm⁻² was achieved in the presence of Cu species and adsorbed I species, which improved *CO adsorption and promoted C–C coupling⁹⁵ (Fig. 4n). The Cu–CuI catalyst was investigated at a high current of  550 mA cm⁻² at − 0.8 V vs. RHE for 85 h, and its stability was higher than that of individual Cu and CuI. It was found that a reconstruction of CuI introduced many low coordinated sites and the C–C coupling was thereof promoted. In addition, DFT calculation showed that iodine atom improved the CO adsorption energy, making the adsorption energy more negative. At the same time, the PDOS study found that the presence of iodine reduced the d-band center of Cu atoms and promoted C–C coupling and affected the stability. At the same time, Li et al. introduced a halogen stabilization strategy for Cu-based electrocatalysts. They prepared a novel Cu–Cl catalyst with a faradaic efficiency of 53.8% for C₂. Residual Cl-stabilized Cu⁰–Cu⁺ active centers and catalyst-stabilized structures were identified as key to maintaining C₂ FE. This suggests that halide ions could facilitate the C–C coupling step and enhance the formation and stability of C₂ products by inducing mixed oxidation states of Cu.⁹⁶ Thus, the specific adsorption of halogen ions on the Cu surface can affect the electrochemical process in several ways: changing the potential distribution at the interface, altering the adsorption energy at sites near adsorbed ions, reducing overpotential, and inhibiting HER.⁹⁷

In summary, the doping of metal and non-metal elements can improve the electrocatalytic reaction activity for the following six reasons: (1) the geometric and electronic structure of the metal can be changed by the position isolation effect; (2) the adsorption and desorption behavior of the reaction products on the electrocatalyst can be adjusted by adjusting the d-band center of the metal; (3) increasing the number of active sites on the catalyst surface can improve the catalytic activity of the reaction; (4) catalyst activity can be increased by reducing deactivation in the reaction; (5) by changing the surface chemical properties of the catalyst, the selectivity of the catalyst for the reaction is improved; (6) adding other metals to the lattice can form lattice strain, which affects the bonding strength of organic intermediates on the electrocatalyst.

3.2 Vacancy

Recently, research has demonstrated that vacancies are an effective means to enhance the eCO₂RR. However, the exact role played by vacancies in the eCO₂RR process remains somewhat elusive. The creation of surface and/or bulk vacancy sites can impact eCO₂RR performance by altering the interaction process, including surface adsorption and desorption behavior, as well as affecting the bulk material properties like electronic structure and conductivity. A deeper understanding of the action mechanism is crucial not only for grasping how vacancies promote the eCO₂RR by modulating electrocatalyst performance but also for guiding the design of materials with high activity and selectivity. This section provides a comprehensive overview of the effects of introducing vacancies on the eCO₂RR performance of various Cu-based materials.

Yang et al. investigated the effects of Cu₂O with varying coverage of surface-bonded hydroxyl group coverage (θ_OH) and different contents of surface oxygen vacancies (O_V) on eCO₂RR and HER. Through DFT and ab initio molecular dynamics simulations, they found that high θ_OH or low O_V conditions led to dominant CO₂ activation to COOH* in HER due to the stronger COOH* adsorption. However, the conversion from COOH* to CO* was the potential-limiting step. Interestingly, the reduction of Cu₂O under these conditions was carried out simultaneously, which inhibited the formation of carbonaceous products. When θ_OH decreased or O_V increased, H* became more stable than COOH*. When θ_OH content was moderate, the activity of eCO₂RR was relatively high (Fig. 5a and b). In addition, Cu₂O with moderate θ_OH was best suited for hydrocarbon generation because it is located near the top of the volcanic activity relationship. On the other hand, since the adsorption energies of COOH* and H* are almost equal, it is likely that high selectivity of eCO₂RR will be obtained. In general, Cu₂O with moderate θ_OH was the most promising catalyst for eCO₂RR in terms of activity and selectivity.⁹⁸ Similarly, Varandili reported on the synthesis of novel Cu/CeO_2−x nanocrystalline heterodimers (HDs) and their study of the catalytic behavior of eCO₂RR. Overcoming the synthesis challenges posed by the high lattice mismatch between Cu and CeO_2−x by using a colloidal seed growth method, the formation of the interface in Cu/CeO_2−x HDs resulted in faradaic efficiency of methane at −1.2 V vs. RHE up to 54%. These values were approximately 5 times higher than those obtained using a physical mixture of isolated Cu and CeO_2−x nanocrystals of similar size under the same operating conditions. DFT indicated that CO₂ molecules were stabilized through a bidentate adsorption over Ce and Cu sites, then a large amount of surface binding CO* was produced to cover the surface of Cu. Moreover, this high coverage of CO* on the Cu surface can hinder the exposure of HER active sites, thus inhibiting the occurrence of HER reaction. In addition, the heterodimers’ design broke the scaling relationship between CHO*/CO* intermediates over the conventional single metallic catalysts, resulting in a better eCO₂RR performance (Fig. 5c).⁹⁹ Zhu et al. designed Cu vacancies on the surface of the AuCu alloy catalyst, which significantly decreased the overpotential and improved the CO faradaic efficiency. The mechanism of improved performance of eCO₂RR on dealloyed Au₃Cu nanocubes (De-Au₃Cu) was further elucidated. The alloyed Au₃Cu achieved an astonishing CO faradaic efficiency of 90.2% at −0.38 V vs. RHE and an impressive CO mass current density of 418.60 A g⁻¹ at −0.8 V vs. RHE. The free energy of De-Au₃Cu (100) to absorb CO₂ and protonation to *COOH near the vacancy was 1.04 eV, which was significantly lower than that of Au₃Cu. Therefore, it could be inferred that vacancies exhibited a higher affinity for CO₂ adsorption compared with other metal locations. Both Au₃Cu and De-Au₃Cu required an energy barrier of about 0.2 eV to desorb CO, with De-Au₃Cu having a COOH energy barrier 0.24 eV lower than Au₃Cu. At the same time, they exhibited comparable energy barriers in *CO desorption and *H absorption, suggesting that their reactivity is similar. As a result, De-Au₃Cu exhibited the best performance in CO production of CO₂RR¹⁰⁰ (Fig. 5d).

Fig. 5 (a) Adsorption energies of COOH* and H* at different θ_OH. (b) U_L as a function of θ_OH for the three elementary reactions at site A. Reproduced from ref. 98 with permission from Wiley, copyright 2018. (c) Breaking of CO*/CHO* scaling relations due to additional through bidentate adsorption. Reproduced from ref. 99 with permission from American Chemical Society, copyright 2019. (d) Calculated free energy of Au, Au₃Cu, and dealloyed Au₃Cu for CO supposed reactive pathways for eCO₂RR. Reproduced from ref. 100 with permission from American Chemical Society, copyright 2018. (e) Gibbs free energy changes of eCO₂RR route on different samples. Reproduced from ref. 102 with permission from Wiley, copyright 2021. (f) CO* on the catalyst surfaces (the circle in the center of yellow copper atoms represents the oxygen vacancy). Reproduced from ref. 103 with permission from Wiley, copyright 2022. (g) Schematic for boosted C₂H₅OH generation over V_Se-Cu_2−xSe. Reproduced from ref. 105 with permission from Wiley, copyright 2023. (h) Corresponding energy diagrams of CO–CO coupling on Cu₃N_x-12.5%, Cu₃N_x-25%, Cu₃N_x-37.5%, and Cu₃N_x-50% at 0 V versus RHE, respectively. Reproduced from ref. 106 with permission from Wiley, copyright 2021. (i) Faradaic efficiencies of alcohols (C₂H₅OH and propanol) and C₂H₄ on different catalysts at the potential of −0.95 V versus RHE. Reproduced from ref. 107 with permission from Springer Nature, copyright 2018. (j) Mechanism of n-propanol formation on adjacent CuS_x-DSV, showing the dimerization of CO–CO followed by CO–OCCO coupling. Reproduced from ref. 110 with permission from Springer Nature, copyright 2021.

The presence of vacancies can increase the coverage of intermediates and improve eCO₂RR performance. For example, Gu et al. synthesized a defect-rich Cu structure (called Cu-DS) that adsorbed *CO species under CO-rich conditions. In the flow cell, the Cu-DS catalyst converted CO₂ to C₂ alcohol with a faradaic efficiency of 70% and a high current density of more than 100 mA cm⁻², whereas on a Cu control with a flat surface without a CO adsorbate (called Cu–C), C₂H₄ dominated the product distribution. Cu-DS increased the alcohol-to-C₂H₄ ratio by a factor of 54 compared with Cu–C. DFT calculations showed that surfaces rich in defect sites enhanced local CO production, thereby increasing surface *CO coverage and CO₂ selectivity for alcohols.¹⁰¹ Wang et al. found that bimetallic CuInSe₂ (V-CuInSe₂) with Se vacancies exhibited highly selective CO₂RR to CO electrocatalyst. Unlike monometallic selenides (CuSe₂ and In₂Se₃), a metallic feature was found on CuInSe₂ in which the valence and conduction bands overlap. Through the study of Gibbs free energy, it was found that the energy barrier of the RDS_eCO2RR on different samples follows in the order of V-CuInSe₂ (0.88 eV) < CuInSe₂ (0.97 eV) < In₂Se₃ (1.01 eV) < CuSe₂ (1.17 eV), indicating that a facilitated eCO₂RR process was achieved on the CuInSe₂ and V-CuInSe₂. (Fig. 5e) It is important to note that the adsorption of intermediates on the CuInSe₂ surface was moderated due to the interaction of the In and Cu orbitals, so eCO₂RR was more favorable than monometallic selenides. At the same time, HER was very disadvantageous in terms of energy, because lattice reconfiguration will occur after hydrogen adsorption on the CuInSe₂ surface. Specifically, after the introduction of surface Se vacancies, electron redistribution occurred in V-CuInSe₂, and more electrons were distributed around the surface In and Cu atoms. The Bader charge analysis also showed that the electron number of the surface In and Cu atoms increased, and the electron number of the surface Se atoms to reduce the selenium vacancies promoted the delocalization of electrons and further optimized the eCO₂RR pathway.¹⁰²

In addition, the presence of vacancies can also modulate the surface electronic structure. For example, Wang et al. used ionic liquid [Omim]Cl (1-octyl-3-methylimidazole chloride) as a bifunctional structure guide agent to prepare Cu₂O nanoparticles with rough surface and abundant oxygen vacancy. [Omim]⁺ played the role of surfactant to regulate the morphology of Cu₂O, and Cl⁻ promoted the formation of oxygen vacancy by cooperating with Cu⁺. The resulting Cu₂O nanoparticles were further dispersed on self-made graphite nanosheets to prepare a composite catalyst, which showed excellent catalytic performance in a flow cell, with a high faradaic efficiency of C₂ (78.5 ± 2%) and a commercial-level current density of 123.1 mA cm⁻² at −1.1 V vs. RHE. DFT calculation calculated the adsorption energy (ΔE_ads) of CO* on the surface structure of oxygen vacancy Cu₂O (111) (Cu₂O-O_V) and pure Cu₂O (111) surface. The ΔE_ads of CO* on Cu₂O-O_V (−1.56 eV) was much higher than that of Cu₂O (−1.31 eV), indicating that the adsorption of CO* on Cu₂O-O_V was stronger, and the Gibbs free energy of OC–COH* generated by Cu₂O-O_V was 1.36 eV (Fig. 5f). It was much lower than the Gibbs free energy of 3.38 eV for Cu₂O to generate OC–COH*. This showed that the surface of Cu₂O-O_V was conducive to the formation of C₂. Due to the abundant oxygen vacancies on Cu₂O-O_V, the CO* coverage rate on the Cu surface was increased, and the reaction of 2CO* + H⁺ → OC–COH* was accelerated, resulting in the highest selectivity of C₂.¹⁰³ Li et al. synthesized a CuInSe₂ catalyst with selenium vacancies, achieving a faradaic efficiency of 85.2% for the C₂ product. Operando infrared spectroscopy and DFT calculations indicated that the formation of surface Se vacancies brought Cu⁺ atoms closer together, activating Cu sites effectively and facilitating the generation of crucial intermediates (*CO and *CHO). This process lowered the C–C coupling barrier, leading to the production of C₂. The presence of metallic Cu shortened the distance between the second-nearest Cu⁰–Cu⁺ bridged by oxygen atoms, resulting in the preferential formation of C₂H₅OH from *OC₂H₄ rather than C₂H₄ through C–O bond cleavage.¹⁰⁴ In addition, vacancies can change the distance between metal bonds. Wang et al. prepared ultra-thin 2D Cu_2−xSe with abundant Se vacancies, where the spatial distance of Cu–Cu around Se vacancies changed from 4.16 Å to 2.51 Å due to lattice stress, thereby effectively shortening the distance. In addition, the medium space distance induced by selenium vacancy could significantly reduce the Gibbs free energy of the asymmetric *CO–*CHO coupling process, effectively change the local charge distribution, reduce the valence state of Cu atoms, and improve the electron donating ability of the double active site. V_Se-Cu_2−xSe samples could catalyze eCO₂RR to C₂H₅OH with high selectivity in the potential range of −0.4 to −1.6 V, and the faradaic efficiency reached 68.1% at −0.8 V. In situ Fourier transform infrared spectroscopy and DFT calculation results showed that the optimal spatial distance of the active site could reduce the ΔG (Gibbs free energy) from C–C to C₂H₅OH and increase the ΔG from C–C to formic acid (Fig. 5g).¹⁰⁵ Zheng et al. demonstrated that high-density nitrogen vacancies on cubic copper nitride (Cu₃N_x) could serve as efficient electrocatalytic centers for CO–CO coupling, forming the key OCCO* intermediate for C₂ product formation. Cu₃N_x with different nitrogen densities was prepared using an electrochemical lithium tuning strategy. By calculating the adsorption energy of CO–CO coupling on Cu₃N (100) surfaces, it was found that when the nitrogen vacancy (V_N) percentage was in the range from 0 to 25%, the Cu₃N crystal could not induce CO–CO coupling. When the V_N proportion was increased to 37.5%, CO–CO coupling was favorable. However, further increase in the V_N content to 50% would cause a negative effect on CO–CO coupling. Compared with pure Cu with a similar Cu–Cu distance (2.556 Å), nitrogen-deficient Cu₃N exhibited abundant negative charges at the nitrogen vacancy positions, enhancing the adsorption of CO* and lowering the energy barrier for CO–CO coupling, thus enabling high selectivity and activity towards C₂ products. The Cu₃N_x catalyst with abundant nitrogen vacancies exhibited the highest faradaic efficiency of 81.7 ± 2.3% for C₂ products at −1.15 V, corresponding to a partial current density of −307 ± 9 mA cm⁻² for C₂ production, outperforming the reversible hydrogen electrode (Fig. 5h).¹⁰⁶

The introduction of metal atom vacancy defects affects the electrocatalytic performance by adjusting the electronic structure of adjacent atoms, thereby adjusting the energy barrier of rate-limiting reaction intermediates. Zhuang et al. reported a class of core–shell vacancy engineering (CSVE) catalysts that utilized sulfur atoms in nanoparticle core and Cu vacancies in shells to achieve efficient eCO₂RR to propanol and C₂H₅OH (Fig. 5i). These catalysts would selectively shift from competing C₂H₄ reactions to liquid alcohols.¹⁰⁷ In addition, the alcohol-to-ethylene ratio increased from 0.18 on bare Cu nanoparticles to 1.2 on CSVE catalysts, representing a six-fold improvement on synthesis rate while the overall C₂₊ selectivity remained unchanged. A C₂ alcohol productivity of 126 ± 5 mA cm⁻² and a faradaic efficiency selectivity of 32 ± 1% were achieved. On the original Cu, both C₂H₄ and C₂H₅OH had a lower thermodynamic energy barrier, while the introduction of surface Cu vacancies slightly increased the energy barrier for C₂H₄ production. Interestingly, the presence of vacancies on the copper shell with a Cu₂S core increased the energy barrier of the C₂H₄ pathway (1.148 eV), while the C₂H₅OH pathway was largely unaffected (0.427 eV). These results suggested that subsurface S atoms and Cu vacancy defects together inhibited C₂H₄ production through this common pathway, thereby shifting the equilibrium in favor of C₂H₅OH. Compared with Cu₂S nanoparticles, pure Cu nanoparticles, bulk Cu₂S and bulk Cu without vacancies under the same conditions, the partial current density of C₂ alcohols was increased by 6, 19, 46 and 44 times. Li et al. proved that SNC@Cu₂S without Cu vacancy had higher faradaic efficiency for HCOOH, while the main product of SNC@Cu_xS and SNC@Cu_1.96S with Cu vacancy was CO.¹⁰⁸ On the one hand, the Cu vacancy in Cu_1.96S changed the electronic structure of the S site and greatly improved the H* formation barrier. The Cu vacancy weakened the interaction between the adsorbent and the metal, providing suitable binding energy for the *COOH intermediate. In addition, Cu vacancy also reduced the charge transfer resistance, improved the CO₂ adsorption capacity, and thus promoted the eCO₂RR activity of SNC@Cu_1.96S. According to DFT calculations, the Cu vacancy formed in Cu–S altered the electronic structure of the S site so that H* occupied a large Gibbs free energy, which in turn inhibited HCOOH formation. The reduction of Cu content in Cu_1.96S reduced the Fermi energy level (E_f) by 0.21 eV compared with pure Cu₂S, and the Cu d band center dropped from −2.31 eV to −2.72 eV. The dense state of Cu_1.96S was further away from E_f than Cu₂S, and the hybrid form of Cu_1.96S widened the bandwidth of Cu 3d and shifted the s and p electrons slightly downward. The Cu vacancy surface (Cu_1.96S) was composed of a small number of surface and subsurface Cu cations, and the p–d charge transfer between the remaining Cu atoms and the S anion increased, thereby redistributing the surface electronic structure, which was favorable for the adsorption configuration of a single Cu–C bond, and the product was converted from HCOOH to CO. In addition, Chen et al. designed Cu vacancy defects through fast electrochemical reconstruction of Cu oxide or hydroxide nanowires to produce C₂₊.¹⁰⁹In situ ATR-SEIRAS showed that the three Cu vacancy complexes enhanced the adsorption of the critical intermediate *CO, thus promoting the dimerization of *CO. Operando Raman and in situ ATR-SEIRAS measurements showed that *CO dimerized into *OCCO and then hydrogenated into *OCCOH, while DFT calculations showed that the adjacent atoms induced by triple-Cu vacancies had enrichment and different charge distributions. This helped to stabilize the *CO intermediate and reduced the C–C coupling barrier by 0.4 eV. The triple-Cu vacancy association enriched and redistributed the local negative charge, enhanced the adsorption of *CO intermediates, reduced the energy barrier of CO–CO coupling, and thus increased the C₂₊ yield. Based on these advantages, the C₂₊ selectivity was 9 times higher than that of catalysts with lower vacancy concentration.

Zheng et al. employed an electrochemical lithium tuning strategy to synthesize dual sulfur vacancies, where the density of sulfur vacancies could be regulated by the number of charge–discharge cycles (Fig. 5j). The dual sulfur-deficient Cu catalyst exhibited a faradaic efficiency of 15.4 ± 1% for propanol at −1.05 V and a high partial current density of 9.9 mA cm⁻² in the flow cell at −0.85 V. DFT calculations indicated that the dual sulfur vacancies formed on hexagonal Cu sulfide could serve as effective electrocatalytic centers for stable CO* and OCCO* dimerization, further enabling the coupling of *CO and *OCCO to form C₃, which cannot be achieved on single sulfur-deficient or sulfur-free Cu surfaces.¹¹⁰ Wang et al. proposed an effective method to improve the electrochemical reduction of CO₂ to HCOOH by controlling the surface polarization of Cu₂SnS₃. This resulted in HCOOH current density of 408.3 mA cm⁻² and a faradaic efficiency of 91.7% at −1.2 V vs. RHE. More specifically, the surface polarization of Cu₂SnS₃ was changed by controlling the concentration of S vacancies. In addition, the local charge redistribution caused by sulfur vacancies explained the altered CO₂ evolution pathway, while the low Gibbs free energy formed by the intermediate *OCHO guaranteed its high selectivity for HCOOH, which could be explained by the coordination structure changes in which Cu and Sn atoms participated in Cu₂SnS₃. The presence of sulfur vacancy on the surface caused the distortion of Cu₂SnS₃ surface lattice, which affected the surface symmetry and changed the surface polarization direction, thus reducing the surface polarization. Sulfur vacancy not only reduced the surface dipole moment of Cu₂SnS_3−x, but also promoted charge redistribution and local coordination changes, thereby alleviating the energy barrier of the rate-determining step, changing the CO₂ reduction pathway, and inhibiting HER. The hydrophobicity and enhanced stability of *OCHO intermediates made Cu₂SnS_3−x cathodes very effective.

In summary, vacancy defects can improve the performance of eCO₂RR in the following potential ways: they increase the adsorption coverage of intermediates, adjust the surface electronic structure, change the atomic spacing, suppress side reactions, and promote charge transfer.

3.3 Surface engineering

Cu and its compounds, when engineered at the surface level, can alter the electronic structure of the catalyst, and influence the binding strength of adsorbates. This “surface effect” is an outcome of how the electronic properties of an electrocatalyst are shaped by its size, shape, and structure. As the size of the electrocatalyst diminishes, its surface area expands, offering more active sites for the eCO₂RR. Changes in the electrocatalyst's morphology affect the proportion of edges and corners, which in turn influences the binding strength of intermediates. Lastly, alterations in the catalyst's composition directly impact the intermediates and the final product formation.

3.3.1 Size. The size effect has been extensively studied for the role it plays in modifying the electronic structure of electrocatalysts and in regulating the adsorption and desorption kinetics of key intermediates.^111,112 Huang et al. compared the adsorption of *CO and *O on icosahedral Cu NPs of different sizes (13, 55, 147, 309, 561) on the Cu(111) surface. As the size of Cu NPs increased, the binding strength of *CO and *O gradually weakened, with *CO and *O binding energies appearing to converge to the limit of crystalline Cu (111) for Cu NPs with only 309 atoms. For CO, HCOOH and H₂ (two-electron products), the U_L (limiting potential) was strongly dependent on the binding strength of the key intermediate such as CO, COOH and H as shown in Fig. 6a–c. Cu₃₀₉ is positioned on the weak binding side of the volcano plot, suggesting insufficient interaction between the reaction intermediates and the Cu surface to generate highly active electronic products. Cu₁₃ and Cu₅₅ are on the strong binding side of the volcano plot, indicating overly strong binding of reaction intermediates to the Cu surface, hindering further reduction or desorption. Cu₁₄₇ is located near the top of the volcano plot, implying Cu₁₄₇ close to optimal binding energy for key intermediates responsible for electronic product formation. The varying strengths of adsorption of reaction intermediates lead to higher U_L. With decreasing Cu NP size, the d-band center shifts linearly upward, enhancing the adsorption of reaction intermediates, controlling the hydrogenation activity of CO₂, and exhibiting significant side effects in the CO₂ electroreduction process.¹¹² In addition, Wallace et al. oxidized Cu meshes in air at 550 °C for varying durations to obtain Cu_xO NWs of different lengths.¹¹³ The average length of Cu_xO NWs increased with increasing annealing time, where the average lengths of Cu_xO-4, Cu_xO-6, Cu_xO-8, Cu_xO-10, and Cu_xO-12 were 1.5 μm, 2 μm, 2.3 μm, 2.6 μm, and 2.7 μm respectively. The Tafel slopes of all Cu NW electrodes indicated that the initial electron transfer to form CO₂˙⁻ intermediates was the rate-determining step (Fig. 6d and e). This suggests that length-dependent selectivity may be attributed to the presence of two distinct active sites favoring the formation of *COOH for CO binding and *OCHO for HCOOH oxidation intermediates. For shorter Cu NWs, the subsequent formation of *COOH and *CO intermediates leading to CO production was favored with a high CO/HCOOH ratio of 2.52. CO selectivity increased with increasing electrode surface roughness, while HCOOH showed the opposite trend. The 2.3 μm exhibited the highest CO selectivity due to its highest surface roughness. In contrast, the 2.7 μm with the lowest surface roughness showed the highest HCOOH selectivity. HCOOH was the dominant product on longer nanowires with a low CO/HCOOH ratio of 0.63. Selectivity was also influenced by the local pH at the electrode–electrolyte interface, where longer nanowires promoted the generation of *OCHO intermediates and more HCOOH production at higher local pH, while shorter nanowires favored the formation of *COOH intermediates and subsequent CO production at lower local pH. No C₂ products were observed on the Cu NWs, possibly due to the short residence time of *CO intermediates on these short-length nanowires inhibiting the dimerization of CO to C₂ products. In another example, Rong et al. employed an acetylene-directed site capture method to prepare size-gradient Cu catalysts ranging from single atoms (SAs) to sub-nanometer clusters (SCs, 0.5–1 nm) and to nanometer clusters (NCs, 1–1.5 nm) on a graphene-like carbon support (Fig. 6f). In eCO₂/CORR experiments, the product distribution strongly depended on the size of NCs. Typically, the catalyst with smaller NCs was less active and less selective for CO in favor of H₂ toward eCO₂RR. Increasing the size of Cu nanoclusters could enhance the catalytic activity and selectivity towards C₂₊ in eCORR. In comparison with eCORR, the three Cu/GDY catalysts exhibited higher activity and selectivity towards H₂, indicating that the binding of CO₂ to Cu metal was weaker than that of CO, which could not effectively suppress the HER. At −1.0 V vs. RHE, H₂ was the main product, with hydrogen evolution dominating. As the size of Cu NCs increased, the activity and selectivity towards CO were enhanced, while the generation of H₂ was significantly suppressed. The stability for eCO₂/CORR originated from the confinement environment from the hole enclosed by six triple bonds and the strong metal–support interaction from orbital overlaps and electron transfer between Cu and C atoms. Decreasing the size to SAs led to many low-coordinated single sites, resulting in stronger adsorption of H* and CO₂ compared with bulk Cu. However, the weak binding of CO* hindered subsequent deep reduction or coupling reactions. In the eCO₂RR, Cu/GDY only produced C₁ compounds and H₂.¹¹⁴

Fig. 6 (a–c) Volcano plots using CO binding energy as a descriptor for U_L of reduction CO₂ to CO, COOH binding energy for HCOOH and H binding energy for H₂. Reproduced from ref. 112 with permission from Elsevier, copyright 2019. Tafel plots of Cu NWs for (d) CO production and for (e) HCOO⁻ production. Reproduced from ref. 113 with permission from Elsevier, copyright 2021. (f) Schematic illustration for the synthesis of Cu/GDY. Reproduced from ref. 114 with permission from Wiley, copyright 2020. (g) Linear sweep voltammetry of the CO₂ electroreduction on Cu NP catalysts. Reproduced from ref. 115 with permission from American Chemical Society, copyright 2014. (h) ECSA-normalized partial current densities for CO₂RR at −1.25 V vs. RHE plotted versus the {100}/{111} and {110}/{111} facet ratios. Reproduced from ref. 117 with permission from Royal Society of Chemistry, copyright 2014.

Strasser et al. synthesized Cu NP catalysts with an average size of 2–15 nm and found that as the size of the Cu particles decreased, the catalytic activity and selectivity towards H₂ and CO significantly increased, especially for NPs smaller than 5 nm. Changes in the population of low-coordinated surface sites and their stronger chemisorption were linked to surging H₂ and CO selectivity, higher catalytic activity, and smaller hydrocarbon selectivity. In stark contrast to the Cu foil electrode, the Cu NPs exhibited significantly higher catalytic activity (larger negative current density) as the Cu NP size decreased (Fig. 6g). The authors proposed a site-counting model and found below 2 nm, particle size effects were often referred to as “catalytic finite size effects”, and small variations in size induced drastic changes in the NP's electronic structure, while quantum effects could become non-negligible. At intermediate particle sizes, (5–15 nm) the spherical particle model predicted low and constant populations of (100) and (111) facets, yet constant hydrocarbon selectivity was observed for Cu NPs compared with Cu bulk surfaces. For these larger NPs, weaker binding of CO and H was expected, favoring hydrocarbon formation.¹¹⁵ Buonsanti et al. monitored the structural changes and catalytic behavior of three different sizes of Cu NCs (16 nm, 41 nm, and 65 nm), revealing the unusual nanoclustering degradation mechanism of nanoparticle electrocatalysts. The CO₂ adsorption and the negative potentials were identified as the parameters accounting for the nanoclustering, and smaller Cu nanoparticles and nanoclusters with more low coordination sites favored HER at the expense of eCO₂RR.¹¹⁶ They also synthesized octahedral Cu nanocrystals (O_h-NCs) in the range of 75–310 nm and found that smaller Cu O_h-NCs greatly enhanced eCO₂RR.¹¹⁷ Angles and edges play an important role in nanostructured catalysts to the extent that unique mechanisms can occur at the boundaries between faces. By quantifying the ratio of (100)/(111) and (110)/(111), it was found that smaller Cu Oh-NCs contribute a higher fraction of corners and edges to the total surface area (Fig. 6h). Catalytic sites at the interfaces between two faces are crucial for promoting eCO₂RR and may stabilize key intermediates such as *CO and *COOH. For larger sizes, Cuenya et al. discovered through their research on the dynamic changes in Cu that the loss of Cl leads to a drastic structural transformation of the entire cubic volume. Under eCO₂RR conditions, the cube morphology of Cu on C underwent drastic changes, including roughness and loss of (100) faces, loss of Cu atoms at edge and corner locations, and reduction of CuO_x species. The selectivity of eCO₂RR versus HER decreased with decreasing cube size, which was attributed to the fact that the cube shape changed more dramatically under eCO₂RR conditions than smaller cubes.¹¹⁸

Supported Cu single-atom catalysts (Cu-SACs) have shown excellent performance in eCO₂RR, converting CO₂ into products such as CO, CH₃OH, CH₄, C₂H₄, C₂H₅OH, and CH₃COCH₃. The Cu active sites in Cu-SACs can undergo electron transfer with atoms on the support, so the coordination environment of Cu-SACs is richer, and a variety of Cu-SACs can be designed by adjusting different coordination environments. Wang et al. designed and synthesized Cu-substituted ceria nanorod materials (Cu–CeO₂), achieving high dispersion of Cu on ceria; at the same time, the specially exposed (110) surface of ceria nanorods was the crystal surface that was most likely to produce oxygen vacancies, and the structure of multiple oxygen vacancies was conducive to the catalytic reduction reaction of the Cu–CeO₂. Among them, the highest CH₄ selectivity (58%) was at −1.8 V, and the current density reached 70 mA cm⁻². DFT results predicted that the Cu–CeO₂ surface had a unique structure with multiple V_O aggregated around the Cu site, which could effectively activate CO₂ molecules and promote the efficient eCO₂RR. The single-atom dispersion of Cu reduced the connectivity between reaction sites and the probability of C–C coupling, so CH₄ was the superior product of the Cu–CeO₂ catalyst. The generation of the first oxygen vacancy was spontaneous, as a pair of Ce⁴⁺–O²⁻ on the (110) surface was replaced by a pair of Cu²⁺–V_O to maintain charge balance. The second and third V_O formed were respectively preferred to the nearest neighbors of the Cu site. Further increasing the number of V_O around the Cu site to 4 significantly destabilized this structure. Therefore, doping with Cu on CeO₂ is favorable for the formation of single-atom Cu sites with an unconventional high V_O value of 3. In addition, charge analysis showed that the Cu site was reduced from Cu²⁺ with two V_O to Cu¹⁺ with three V_O¹¹⁹ (Fig. 7a–e).

Fig. 7 (a–c) Structure models of CeO₂(110) doped with one single Cu site, with (a) 1 oxygen vacancy (V_O); (b) 2 V_O's; and (c) 3 V_O's. (d and e) Structure models of these V_O-bound, single-atomic Cu site on CeO₂ for CO₂ adsorption and activation. Reproduced from ref. 119 with permission from American Chemical Society, copyright 2018. (f and g) Free energy profiles (at −0.78 V versus RHE) for CO₂ activation on Cu, Cu@Cu₂O and APC of Cu₁⁰–Cu₁^x+ on Pd₁₀Te₃ nanowires. (g) Configurations of physisorbed CO₂ and chemisorbed CO₂ on Cu–APC. Reproduced from ref. 121 with permission from Springer Nature, copyright 2019. (h) and (i) The density of states (DOS) and partial DOS (PDOS) for the CoCu-DASC, Co-SAC, and Cu-SAC catalysts. Reproduced from ref. 122 with permission from Wiley, copyright 2022.

Zhao et al. designed single-atom Cu encapsulated on N-doped porous carbon (Cu-SA/NPC) catalysts for the reduction of CO₂ to multi-carbon products.¹²⁰ Acetone was identified as the main product with a FE of 36.7% and a productivity of 336.1 μg h⁻¹. DFT calculations showed that the coordination of Cu with four pyrrolic-N atoms was the main active site, which reduced the reaction free energy required for CO₂ activation and C–C coupling. Cu single atoms coordinated with pyrrolic N species (Cu-pyrrolic-N₄) had higher catalytic activity than non-coordinated pyrrolic N species and should be responsible for the eCO₂RR to acetone on Cu-SA/NPC. For the reduction of CO₂ to acetone, the formation of the reaction intermediates required the concerted action of Cu and the coordinated pyrrolic N species, which facilitated C–C coupling to form C₂ and C₃ species. Therefore, for acetone and other oxygenates, the selectivity-determining step should occur after *CO formation.

In addition, dual-atom catalysts (DACs) can maintain 100% atomic utilization and excellent selectivity of single-atom catalysts, and improve catalytic activity through synergistic effects between adjacent metal atoms. The introduction of a second metal site near the Cu metal increases the metal loading and promotes the activation of CO₂. Current research on DACs focuses on two subcategories: (1) homologous pair DACs, where the two metal atoms are the same; (2) heterologous pair DACs, where the two metal atoms are different. For example, Jiao et al. reported a Cu₁⁰–Cu₁^x+ atomic pair catalyst (Cu-APC) with a highly active atomic interface, which was anchored on a one-dimensional Pd₁₀Te₃ alloy nanowire. Extended X-ray absorption fine structure (EXAFS) fitting results showed that the Cu–Cu bond length was 2.49 Å, which was shorter than the bond length of Cu foil (2.56 Å). The Cu–Cu coordination number was 2.1, which was much smaller than the coordination number of Cu foil, indicating that Cu existed in the defects of Pd₁₀Te₃ in the form of Cu_x and atomically dispersed. Further characterization on the Pd₁₀Te₃ surface indicated that the doped Cu atoms bonded with the O atoms to form a stable configuration (i.e., two Cu atoms formed Cu₁⁰–Cu₁^x+ atomic pair structure, while the other two underlying Cu atoms accommodated the defects around the O atoms for stabilizing Cu-APC). The H₂O molecules adsorbed on Cu₁^x+ stabilized the chemisorbed CO₂ molecules on adjacent Cu₁⁰, thus promoting CO₂ activation (Fig. 7f and g).¹²¹ Unlike single-atom and homogeneous diatomic catalysts, heterogeneous diatomic catalysts can adjust the electronic structure and coordination environment of the metal active center by introducing another metal single atom. Yi et al. synthesized a diatomic center catalyst (CoCu-DASC) composed of Co–Cu hetero diatomic pairs. X-ray absorption near edge structure spectroscopy (XANES) showed that the valence state of Co atoms in Co-SAC and CoCu-DASC was different, indicating that the introduction of Cu atoms changed the electronic structure of Co atoms. Similarly, the valence state of Cu in CoCu-DASC was different from that of Cu-SAC. In the H cell, the CO faradaic efficiency of CoCu-DASC was 99.2% at −0.6 V. In the flow cell, the CO FE of CoCu-DASC was always higher than 90% in the range of 100 mA cm⁻² to 500 mA cm⁻². DFT showed that the Co–Cu diatomic sites reduced the activation energy of the *COOH intermediate and promoted the desorption of *CO. According to the total density of states (DOS) calculation, the d-band centers of Co or Cu atoms of CoCu DASC (Co), CoCu DACC (Cu), Co SAC and Cu SAC were −1.45, −3.51, −0.93 and −3.38 eV, respectively (Fig. 7h and i). The more negative the d-band center value, the weaker the binding of the intermediate to the active site. However, the binding of the intermediate to the active site should not be too strong or too weak, indicating that a moderate d-band center value is better. The d-band center of CoCu DASC (Co) was more moderate.¹²²

When Cu clusters are loaded on the substrate, strong metal–substrate interactions are formed, thereby regulating the catalytic activity and selectivity of CO₂. Hu reported a simple dual-domain strategy to synthesize Cu clusters with a particle size of about 1 nm and about 10 atoms on a carbon matrix, and found that Cu clusters catalyzed the high selectivity of eCO₂RR to generate CH₄, and the maximum faradaic efficiency was as high as 81.7%. DFT results showed that the adsorption intensity of Cu clusters for *CO and *H intermediates was higher than that of Cu nanoparticles, thereby inhibiting the formation of CO and H₂. The strong interaction between Cu clusters and carbon defects could further regulate the electronic structure of Cu clusters, and improve the stability of Cu clusters and the selectivity of CH₄. Corresponding structural models of Cu (111) for Cu NP, Cu cluster with 13 atoms (Cu₁₃), and Cu₁₃ (Cu₁₃/DG) loaded on defective graphene (DG) were investigated. According to the gas adsorption results, the adsorption energies of *CO and *H on Cu₁₃ clusters were calculated to be −0.66 and −0.44 eV, respectively, which were much lower than those on Cu (111) (−0.29 and −0.23 eV), indicating that the adsorption strength of Cu₁₃ clusters for the above two intermediates was much higher than that of Cu NP. The elevated central position of the Cu cluster resulted in a stronger adsorption intensity compared with the larger NPs. In the case of Cu₁₃ cluster/DG, Bader charge analysis showed that electrons were transferred from the Cu₁₃ cluster to the defective graphene on the Cu₁₃ cluster/DG, further confirming the existence of a strong metal–support interaction (SMSI) effect between the Cu cluster and the defective carbon. In addition, the high heat emission of the adsorption energy of Cu₁₃ clusters on the graphene layer with two carbon vacancies was −7.78 eV, indicating that the SMSI effect could significantly improve the stability of Cu clusters. The SMSI effect also affected ΔG_CO*, and the value of Cu₁₃ cluster/DG was −0.55 eV, which was slightly higher than the value of Cu₁₃ cluster (−0.66 eV), suggesting that tuning the electronic structure of Cu₁₃ by SMSI effect could slightly reduce the adsorption strength of *CO.¹²³

In this section, it is evident that particle size influences catalytic behavior through various factors. Specifically: (1) reducing the size of the catalyst enhances the surface-to-volume ratio, which in turn improves the utilization rate of metal atoms. (2) As the particle size decreases, the increase in low-coordination sites leads to perturbations in the electronic structure, which generally elevates reactivity. (3) Changes in particle size affect the binding strength of different reaction intermediates, thereby influencing the selectivity of the final product.

3.3.2 Morphology. Currently, various morphologies of Cu-based catalysts for the eCO₂RR have been developed and investigated, including nanoparticles, nanocubes, core–shell structures, nanowires, nanoarrays, and nanofoams.^124–129 These diverse structures influence the catalytic activity and selectivity due to their unique surface properties and interaction with reactants.

For instance, Cu nanowire arrays have been synthesized through the electrochemical reduction of CuO nanowire arrays on Cu foil. By modifying the length of the Cu nanowires, researchers can adjust the selectivity of the electrocatalytic reduction of carbon dioxide to hydrocarbons, effectively suppressing the production of CH₄ and H₂. These nanowires have a long and dense structure, which limits the diffusion of generated OH⁻ ions on the electrode surface, causing an increase in local pH at the electrode/electrolyte interface, thereby affecting the reaction pathways.¹²⁴ Zhang et al. prepared micron-long Cu nanowires with diameters of 25 nm and 50 nm for the electrochemical reduction of CO/CO₂ to C₂ hydrocarbons. They found that the 50 nm nanowires displayed higher activity and selectivity towards C₂ hydrocarbons (C₂H₄/C₂H₆) compared with the 25 nm nanowires, achieving a faradaic efficiency of 60% at −1.1 V. At −0.9 V, these nanowires were less efficient for reducing CO₂, producing CO, HCOOH, and C₂H₄ with a faradaic efficiency of 35%, with C₂H₄ being the main gas product (22% faradaic efficiency) at −1.5 V.¹²⁵ Another example involves Cu nanowires with abundant surface steps, which exhibited very high faradaic efficiency for C₂H₄ (77.40 ± 3.16%) and maintained this performance for over 200 hours. DFT calculations indicated that the adsorption energy barrier for the *CO intermediate at two adjacent active sites on the Cu step surface was significantly lower than on Cu (100), demonstrating excellent thermodynamic stability of the C₂ pathway and a low C–C coupling energy barrier at these step sites. These stepped surfaces also promoted the formation of C₂ products by inhibiting the C₁ pathway and H₂ production.¹³⁰ In contrast to the above nanowires, Kim et al. synthesized branched CuO nanoparticles and cubic CuO nanoparticles, comparing their eCO₂RR performance. They observed that the main copper species were reduced to Cu during the reaction. The branched form of the electrocatalyst exhibited more than 70% C₂H₄ faradaic efficiency and 30% H₂ faradaic efficiency in a neutral aqueous solution, without any by-products, while also showing high durability (up to 12 hours). Compared with the cubic form, the branched Cu oxide structure formed highly active domains with interfaces and connections during activation, which resulted in a large surface area and high local pH, enhancing both the selectivity and activity for C₂H₄ production.¹³⁰

In addition, Buonsanti et al. studied the eCO₂RR selectivity over Cu catalysts using Ag-decorated Cu nano-octahedra (Cu_oh) catalyst with (111) facets and Ag-decorated Cu nano-cube (Cu_cub) catalyst with (100) facets. The selectivity and activity for C₂H₅OH synthesis and C₂H₄ synthesis were promoted while the selectivity and activity of CH₄ and H₂ were inhibited after decoration of Ag when compared with the bare Cu catalyst, as demonstrated from the partial current densities normalized by the electrochemically active surface area (ECSA). J_ECSA (CH₄) decreased on both Cu_oh–Ag and Cu_cub–Ag compared with the bare Cu NCs at all potentials, indicating that the production rate of methane was actually suppressed (Fig. 8a). In contrast, J_ECSA (C₂H₄) and J_ECSA (C₂H₅OH) were significantly increased on both catalysts after Ag decoration (Fig. 8b and c), but J_ECSA (C₂H₅OH) was significantly higher on Cu_oh–Ag than that on Cu_cub–Ag. DFT calculation showed that under high *CO coverage, the (111) facets of Cu_oh–Ag tended to drive the coupling of *CH_x–*CO and subsequent reduction to C₂H₅OH, while the (100) facets of Cu_cub–Ag tended to promote ethylene synthesis. Thus, compared with Cu_cub–Ag catalyst, Cu_oh–Ag catalyst had more active sites for ethanol and thereof a better ethanol selectivity.¹³¹ Cao et al. could effectively adjust the chemical composition of Cu-based nanoreactors by adjusting the geometry of MOF precursors. The optimized pore structure and high content of Cu–N₄/Cu₂O active site exhibited excellent catalytic performance in eCO₂RR. DFT calculation found that CO₂ molecules were first reduced to CO molecules under the catalysis of the Cu–N₄ site. The CO molecule was then transferred to the Cu₂O/Cu site for further deep reduction. By analyzing the change of the free energy of the intermediate, when the coverage of *CO on the surface of Cu₂O (111) increased from 2/8 to 5/8, the free energy of *CO hydrogenation to *CHO decreased from 0.89 eV to 0.47 eV. The results indicated that the higher concentration of *CO produced at Cu–N₄ site could facilitate the formation of CH₄ on the surface of Cu₂O (111) (Fig. 8e). By using DFT and molecular dynamics (MD) simulation methods, Fang et al. studied Cu-based catalysts with different coordination numbers to design catalysts with high selectivity to ethylene. Fig. 8e shows that reducing the coordination number of Cu decreased the adsorption energy of *CO on the Cu surface, and decreased the activation energy of C–C coupling, both of which were beneficial for the formation of ethylene (Fig. 8f). This provides theoretical evidence for designing an efficient CO₂-to-ethylene catalyst via lowering the coordination number of Cu. Furthermore, MD simulation was used to explain the Cu and Cu₂O reconstruction toward various metallic Cu state during the eCO₂RR (Fig. 8g). By constructing several Cu_xO catalyst models, the study found that the crystal face of Cu₂O was similar to the corresponding metallic Cu face, and after removing oxygen from the Cu oxide, the resulting Cu surface could retain some of the properties of Cu₂O (Fig. 8h). Using the Wulff construction, different polyhedral structures of Cu_xO were designed and the surface and bulk phase reconstruction process after removing surface oxygen and bulk oxygen was studied by MD simulation (Fig. 8i). The results showed that the proportion of low-coordination Cu sites in Cu_xO increased, and the low-coordination numbers ratio of Cu atoms in the reduced Cu layer increased (Fig. 8i). The experiment verified that Cu₂O catalyst with low-coordination Cu site could generate ethylene at a current density of 800 mA cm⁻², faradaic efficiency more than 70%, and maintain stability at a cell voltage of 3.5 V for 230 hours.¹³³ Gao et al. synthesized Cu₂O NPs with different morphologies and exposed crystal facets, which significantly influenced the selectivity and activity of C₂H₄ production.¹³⁴ The cubic Cu₂O (c-Cu₂O) NPs with (100) facets, octahedral Cu₂O (o-Cu₂O) NPs with (111) facets, and truncated octahedral Cu₂O (t-Cu₂O) NPs with (111) and (100) facets showed varying degrees of selectivity for C₂H₄, with t-Cu₂O exhibiting the highest selectivity C₂H₄ faradaic efficiency (59%), followed by o-Cu₂O (45%) and c-Cu₂O (38%) (Fig. 8j–l). Tafel slope analysis revealed that t-Cu₂O NPs had the lowest Tafel slope (75 mV dec⁻¹), indicating the lowest activation energy for eCO₂RR compared with o-Cu₂O (82 mV dec⁻¹) and c-Cu₂O (97 mV dec⁻¹). The adsorption of CO on the Cu₂O (100) facets and at the interface between (100) and (111) facets was found to be stronger than on the Cu₂O (111) facets. This promoted C–C coupling to produce C₂₊ products. The adsorption of C₂H₄ at the interface between Cu₂O (100) and (111) facets was weaker than on the Cu₂O (100) facet, suggesting that c-Cu₂O with (100) facet facilitated C–C coupling to generate C₂₊ products. Furthermore, the superior catalytic performance of t-Cu₂O compared with c-Cu₂O and o-Cu₂O could be attributed to the lower Fermi level on the (111) facet of Cu₂O, which could promote charge transfer between Cu₂O (111) and (100) facets and further enhance the multi-electron participation kinetics in C₂H₄ generation within the Cu₂O NPs surrounded by (111) and (100) facets.

Fig. 8 ECSA-normalized partial current densities (J_ECSA) for (a) CH₄, (b) C₂H₄, and (c) C₂H₅OH of the Cu_oh–Ag and Cu_cub–Ag catalysts. Reproduced from ref. 131 with permission from American Chemical Society, copyright 2021. (d) The free energy of RDS with different *CO coverage on Cu₂O(111). Reproduced from ref. 132 with permission from Elsevier, copyright 2022. (e) *CO adsorption changes with the coordination number (CN) of the Cu sites. (f) Energy profile of C–C coupling on Cu sites with various CNs. (g) Schematic illustration of Cu and Cu₂O reconstruction. (h) Comparison of the spatial configuration of Cu atoms in Cu and Cu₂O. The dotted boxes on the Cu atoms highlight the similar Cu atom configurations and crystal face relationships between Cu and Cu₂O. (i) Cu_xO reconstruction toward metallic Cu via oxygen removal. (j) The correlation of the low CN ratio of Cu and Cu₂O. Reproduced from ref. 133 with permission from Wiley, copyright 2024. Formation of C₂H₄ on the (k) (100) facets of c-Cu₂O NPs, (l) (111) facets of o-Cu₂O NPs, and (m) (100) and (111) facets of t-Cu₂O NPs Reproduced from ref. 134 with permission from Wiley, copyright 2020.

3.4 Twin crystal

Grain boundaries (GBs) are a crucial structural feature in crystalline materials, particularly in polycrystalline materials, playing a significant role in influencing the physical and chemical properties of these materials. In the context of electrocatalytic eCO₂RR, GBs impact the process in several ways. (1) Enhancing electrocatalytic activity: by strategically designing and manipulating the structure and density of GBs, it is possible to significantly improve the reaction efficiency and selectivity. (2) Regulating reaction pathways: the interaction between lattice strain and the forces between the catalyst and adsorbed species can disrupt the linear proportional relationships, thereby adjusting the reaction pathways and facilitating the efficient generation of specific products. (3) Long-term stability: adjusting the grain boundary density can help maintain stability over extended periods.

Wang et al. analyzed the reason for high C₂H₄ faradaic efficiency produced by La₂CuO₄ and found that XAS results showed that twin boundaries (TBs) in La₂CuO₄ generated strain, resulting in an increase in the length of Cu–O and Cu–La bonds, thereby enhancing FE_C2H4.¹³⁵ DFT indicated that the high performance of La₂CuO₄ may come from the (113) twin boundary and the (111) twin boundary. Fig. 9a and b show the different structural configurations and real spatial 3D orbital contour plots of (113) and (111) surface of La₂CuO₄. Furthermore, the POS of (113) and (111) surfaces showed the different subtle electronic structure (Fig. 9c and d). Notably, the gap states were constructed by the joint contribution of La, Cu, and O sites, which significantly promoted the electron transfer by the annihilation of the barriers. Although the electronic contribution on the (111) surface also showed the main contribution of Cu and La for both valence band maximum (VBM) and conduction band minimum (CBM), respectively, the electron transfer barrier still existed, which could not be alleviated by the gap states. Such a barrier determined that the electron transfer efficiency on the (111) surface was consequently much lowered, especially for the C–C coupling. The PDOS of the key adsorbates of the C2 route on (113) and (111) surface is further illustrated (Fig. 9e and f). On the (113) surface, from the adsorption of the initial reactant to the final product C₂H₄, the dominant peak of the PDOS is linear correlation, which guarantees the efficient electron transfer in the reaction coordinates. Meanwhile, Cu 3d orbitals demonstrated a highly stable valence state. On the contrary, on (111) surfaces the linear correlation has weakened, which was attributed to the C–C coupling with a larger electron transfer barrier induced by the overbinding of CO. In addition, the Cu 3d band center has a slight downshifting, indicating that the local electroactivity on the surface was affected by the adsorption of intermediates. By comparing the PDOS of both adsorbed CO, the (113) surface showed a more evident reduction ability due to the larger downshifting of CO orbital. Meanwhile, the (111) surface exhibited a stronger tendency of electron transfer with a proton, which leads to the high production of H₂ (Fig. 9g). Tang et al. precisely engineered Sn atoms into the TBs of Cu through pulse electrodeposition and boundary segregation.¹³⁶ As Sn atoms were incorporated into the one-dimensional TB lattice, the faradaic efficiency of HCOOH significantly improved to 95%, with a 431% increase in local current density. The conversion frequency of the Sn atoms on TBs reached 4.6 × 10⁴ s⁻¹, two orders of magnitude higher than surface-doped Sn and pure Cu TB active centers. TBs are commonly regarded as defect sites exposing low-coordination atoms. When Sn atoms were positioned at the GBs, due to the lower U_L value on the HCOOH pathway compared with the CO pathway, the Cu–Sn sites played a dominant role in HCOOH generation. In samples with low Sn content, there was a weak correlation between HCOOH content and Sn concentration. However, in annealed samples enriched with Sn, the HCOOH content increased with rising annealing temperature. The Sn atoms tended to locate closer to Cu atoms rather than other Sn atoms, indicating the discrete single-atom dispersion of Sn. The binding free energy of the crucial intermediate H* was −0.06 eV on Cu–Cu sites, while it increased to 0.51 eV on Cu–Sn sites, demonstrating the inhibitory effect of Sn atoms on the HER. Park et al. analyzed the eCO₂RR selectivity over a series of CuO catalysts that were treated with plasma under different atmospheres (e.g., N₂, O₂). Through plasma treatment, more defective catalytic active sites could be obtained. As shown in Fig. 9h, the atomic charges obtained by Bader charge analysis showed that the partial positive charges increased on the O₂/N₂–Cu surface compared with those on bare Cu surface. The results revealed that the codoping of O and N radicals induced the partial oxidation of Cu to be Cu^δ+. In addition, CO is likely to be adsorbed at a face-center crystal site on the bare Cu surface, while on O₂/N₂–Cu, CO was likely adsorbed at the top site between O and N site at subsurface, accompanied by a reduction of Cu and a stronger attraction of CO molecules. In Fig. 9i, the PDOS of hybridized C 2p peaks also revealed that the splitting between bonding and antibonding was larger on the O₂/N₂–Cu surface compared with that on the Cu (111) surface. The strong binding affinity toward CO on the O₂/N₂–Cu surface indicated that a deeper reduction reaction for C₂₊ products could be facilitated, as demonstrated by the high binding energy of CO as well as the low thermodynamic barriers for hydrogenation and C–C coupling (Fig. 9i).¹³⁷

Fig. 9 (a) Structural configurations and real spatial 3D orbital contour plots of (a) (113) and (b) (111) surface of La₂CuO₄. PDOS of (c) (113) and (d) (111) surface of La₂CuO₄. PDOS of key adsorbates of the C₂ route on (e) (113) and (f) (111) surface of La₂CuO₄. (g) PDOS of CO* and H* adsorbates on (113) and (111) surfaces of La₂CuO₄. Reproduced from ref. 135 with permission from American Chemical Society, copyright 2021. (h) Atomic charges on Cu for the bare Cu (111) and ON–Cu surfaces, where V represents the octahedron site in the Cu (111) surface. (i) PDOS of CO-adsorbed surfaces. (j) Free-energy diagram of key hydrogenation intermediates to form *OHCCO at U = 0 V. Reproduced from ref. 137 with permission from American Chemical Society, copyright 2023. (k) Schematic diagram of the mechanism over ED-Cu catalyst. Reproduced from ref. 139 with permission from Elsevier, copyright 2023. (l) Quantum mechanical simulations of the rate-limiting steps of C1 and C2 pathways on tw-Cu(111). (m) Activation free energies G_act for C1 and C2 pathways on tw-Cu(111) (orange bars) and planar Cu(111) (gray bars). (n) Energetics of the pathways on planar Cu(111) (left) and tw-Cu(111) (right). Reproduced from ref. 142 with permission from American Chemical Society, copyright 2023.

Wu et al. used a CO atmosphere to reduce Cu₂O to a stable Cu₂O–Cu nanocube hybrid catalyst (Cu₂O(CO)), which had nanograin boundaries with different crystal faces, and C₂₊ faradaic efficiency reached 77.4%.¹³⁸ That is, stable Cu⁺ species produced abundant Cu⁰/Cu⁺ interfacial sites during the eCO₂RR process, thereby increasing the yield of C₂₊ by increasing the surface coverage of *CO in the eCO₂RR process and decreasing the C–C coupling energy barrier. In addition, the abundant nanograin boundary structure of the Cu₂O(CO) catalyst prevented the reconfiguration of the surface structure and improved the catalytic stability. Bian et al. synthesized Cu nanoribbons with rich grain boundaries (GBs) to produce C₂H₄ from eCO₂RR. Specifically, CO₂ was first activated at the GBs of the catalyst to generate adsorbed *CO₂, and then adsorbed *H attacked adsorbed *CO₂ from H₂O dissociation to generate *COOH intermediates, further accelerating *CO production. The active site on GBs was conducive to the adsorption of *CO intermediates, increased the local coverage of *CO, and greatly enhanced the dimerization kinetics of CO to produce ethylene and other C₂₊ products (C₂H₅OH, CH₃COOH, and C₃H₇OH) (Fig. 9k). The Cu nanoribbons were formed by the superposition of multiple crystal faces.¹³⁹ The current density of Cu nanoribbons reached 700 mA cm⁻², the FE of C₂H₄ and C₂₊ reached 67.2% and 82.1%, and the in situ Raman show that abundant GBs enhanced the activation of CO₂ and significantly promoted the formation and adsorption of *CO intermediates, which accelerated C–C coupling to form *OCCO and *OCCOH intermediates and improved the production of C₂H₄ and other C₂₊ products. He et al. prepared a series of core–shell metal (Cu/In) oxides with abundant GB between amorphous In₂O₃ and cubic Cu₂O by template-assisted co-precipitation, and tested the synthesis of syngas by eCO₂RR. The phases of Cu₂O and In₂O₃ were independent in the bimetal oxide and did not form any alloy oxide phase, so Cu₂O and In₂O₃ could maintain their crystal structure and chemical properties in the bimetal oxide. During the eCO₂RR process, Cu₂O and In₂O₃ were completely reduced to the metals Cu and In. Derived Cu/In had a maximum CO faradaic efficiency (80%) at −0.77 V vs. RHE and great stability of 10 h in H cells.¹⁴⁰ Kong et al. synthesized small Cu NPs with enriched micro-grain boundaries (RGBs-Cu) through spatial confinement and in situ electroreduction.¹⁴¹In situ spectroscopy and theoretical calculations showed that the adsorption of *CO intermediates was significantly enhanced by the small size of Cu grain boundaries due to the presence of a large number of low-coordination and disordered atoms. In addition, these GBs generated in situ at high current conditions exhibited excellent stability during the eCO₂RR process, creating a stable CO-rich microenvironment. The high local CO concentration around the catalyst surface could reduce the energy barrier of C–C coupling and significantly increase the faradaic efficiency of multi-carbon products in neutral and basic electrolytes. Specifically, the developed RGBs-Cu electrocatalyst achieved a faradaic efficiency of 77.3% in a multi-carbon product and maintained stability for more than 134 hours at a constant current density of −500 mA cm⁻².

Unlike the twin (tw) structure composed of the crystal faces above, the high (111) oriented Cu foils with double GBs synthesized by Cai et al. obtained, under −1.2 ± 0.02 V (vs. RHE) conditions, methane up to 86.1 ± 5.3%.¹⁴² DFT calculated the activation barriers of C₁ and C₂₊ pathways with key intermediates on tw-Cu (111) (Fig. 9l). In Fig. 9m, it was found that in the rate-determining C₁ pathway (the reduction of *CO to *CHO (*COH) by proton-coupled electron transfer), tw-Cu (111) had a lower activation (reaction) free energy than on planar Cu (111), which enhanced CH₄ production on tw-Cu (111). For the rate-limiting step in the C₂ pathway, two key intermediates represented by *COH–CHO and *COH–*COH showed that for *COH–*CHO formation, the tw-Cu (111) had higher barrier than planar Cu (111), while for *COH–*COH formation, tw-Cu (111) had lower barrier than planar Cu (111). Therefore, it was confirmed that compared with planar Cu (111), tw-Cu (111) reduced barriers for rate-limiting steps in the C₁ pathway, while increasing barriers for C–C coupling to form the most favored product *COH–*COH on tw-Cu (111). At −1.2 V vs. RHE, rate-determining steps on tw-Cu (111) for C₁ synthesis were lower than that of in C₂ synthesis, which could enhance CH₄ production and effectively limit C₂H₄ formation. In contrast, the barriers for C₁ synthesis and C₂ synthesis were similar at the same applied potential over planar Cu (111), resulting in a similar preference for CH₄ and C₂H₄ synthesis (Fig. 9n). Chen et al. controlled the growth of Cu with rich GBs by using polyvinylpyrrolidone as an additive, achieving a high faradaic efficiency of 70% for C₂H₄ and C₂H₅OH.¹⁴³ Through ATR-SEIRAS, they confirmed that the presence of GBs enhanced the adsorption of the key intermediate (*CO) on the surface, thereby promoting further reduction. A lower wavenumber of the CO_ad band indicated stronger adsorption of CO on the GBs–Cu surface, and the GBs could enhance and stabilize the adsorption of CO at more negative potentials. The binding energy of CO at the protruding GB site (GB₁) and the concave GB site (GB₂) was stronger than that at the terrace sites (t₁–t₇), and the GBs also altered the coordination of the surrounding atoms. Consequently, the binding energy of CO at the t₁, t₂, and t₇ sites was slightly higher than that at the other terrace sites. The formation energies of the *COCHO intermediate at the GB₁ and GB₂ sites were 0.46 eV and 0.58 eV, respectively, which were lower by 0.24 eV and 0.12 eV than those at the t₄ terrace site (0.70 eV), indicating that the presence of GBs promoted C–C coupling. Zhang et al. used a pulsed electrochemical deposition strategy to control the GB density in Cu catalysts.¹⁴⁴ The GB-Cu catalyst exhibited robust performance for the electrochemical conversion of CO₂ to C₂H₄ over a wide potential range, with a selectivity of 46.2%. GBs regulated the adsorption behavior of *CO on Cu surface, which is a key intermediate species to produce multi-carbon species. The main difference between GB-Cu and Cu (111) planes was the presence of uncoordinated Cu atoms at the GB site. In situ Raman demonstrated that as the density of GB increased, the C≡O stretch increased and the Cu–CO stretch decreased, indicating an increased possibility of C–C coupling rather than CO desorption and the formation of C₁ products. They found the maximum faradaic efficiency of C₂ product formation and the energy efficiency of the full-cell of 20.2% were achieved at a specific current density of 303.6 mA cm⁻². GB-Cu was implemented in a 25 cm² MEA electrolyzer and exhibited more than 62% selectivity over 70 h, with a current retention rate of 88.4% at an applied potential of −3.80 V.

Huang et al. synthesized unique star-shaped dodecahedral Cu nanoparticles (SD-Cu NPs) as electrocatalysts, characterized by TBs and multiple stacking faults, resulting in low overpotential for CH₄ and high production efficiency of C₂H₄ at 52.43% ± 2.72%.¹⁴⁵ The results indicated that surface layer faults and twin defects increased the CO binding energy, enhancing the eCO₂RR performance of SD-Cu NPs. It was found that the formation energy of *CHO was less than that of *COH, and DFT calculations predicted that CH₄ on TBs was formed through *CHO. A comparison of the energy required for *OC–CHO formation and subsequent C₂H₄ generation on (111) and TBs revealed a significant advantage on TBs. The observed increase in C₂H₄ production in SD-Cu NPs could be attributed to the enhanced CO binding energy and increased surface CO coverage. Stacking faults in individual TBs resulted in a stronger CO binding energy compared with a single TB on the Cu (111) surface. DFT calculations demonstrated that TBs with higher concentrations of CO could catalyze the reduction of CO₂ to C₂H₄ rather than CH₄.

3.5 Heterostructure

The fabrication of heterostructures, which involves combining two or more phases, is a crucial strategy for optimizing the electronic structure of catalysts. The core feature of heterostructures lies in their ability to create mismatches in energy levels at interfaces. The uneven charge distribution at the heterostructure interface, especially at semiconductor–metal junctions, triggers the Mott–Schottky effect. This effect leads to the formation of electron transport channels and the partial reconstruction of the electronic structure. Atoms on both sides of the heterostructure interface undergo the most significant alterations in their electronic structure, often becoming highly active sites. The influence on the electronic structure diminishes with increasing distance from the interface. Thus, a fundamental strategy to enhance catalytic activity is to strengthen the interface between the two components of the heterostructure.

By designing Cu/Ag interfaces, Zhang et al. showed that the *CO–*CHO coupling induced by interface effects can promote the generation of C₂H₄ at high current density.¹⁴⁶ All the prepared Cu/Ag catalysts showed excellent eCO₂RR performance against C₂H₄, and Cu₅₂Ag₄₈ reached up to 69.2% C₂H₄ faradaic efficiency at 450 mA cm⁻². DFT calculation showed that Cu/Ag interface reduced the energy barrier of *CO hydrogenation and promoted the asymmetric coupling of *CO–*CHO, which was the key to obtain high C₂H₄ faradaic efficiency. In addition, in situ Raman spectroscopy could detect both *CHO intermediates and asymmetrically coupled intermediates O*CCHO. DFT calculations confirmed that CO₂ could be easily converted to adsorbed *CO, whether Cu (111) or Cu (111)/Ag. The hydrogenation rate of *CO at the Cu (111)/Ag interface was faster, and the main substance should be *CHO. The accelerated hydrogenation of *CO at the Cu/Ag interface was not due to sufficient *H coverage, and the reduction of H adsorption and coverage reduced the selectivity of C₂H₅OH, resulting in an increase in the selectivity of C₂H₄ in C₂₊ products. Hydrogenation of *CO to *CHO and coupling with *CO was the most thermodynamically advantageous C–C coupling pathway. Zhang et al. adopted an active hydrogen (*H) intermediate strategy based on CuAl₂O₄/CuO catalyst derived from layered precursors for C₂H₅OH production. The CuAl₂O₄/CuO catalyst had a FE of 2.5 times higher for C₂H₅OH than CuO and showed a continuous durability of 150 hours in flow cell.¹⁴⁷ In addition, in situ attenuated total reflection Fourier-transform infrared spectroscopy of D₂O dissociation and desorption showed that CuAl₂O₄ mediated local *H coverage on the surface, thus promoting the subsequent hydrogenation reaction of the intermediate, which was conducive to the production of C₂H₅OH. DFT calculation clearly showed that the in situ CuAl₂O₄ stable Cu⁺ site could increase the binding energy of *CO and *COH, facilitate the hydrogenation of *HCCOH to *HCCHOH, reduce the energy barrier of *OC–*COH coupling to form C–C, and promote the formation of C₂.

Zhang et al. using the interface effect, found the optimized Cu₉Zn₁/Cu_0.8Zn_0.2Al₂O₄ catalyst had a faradaic efficiency of 88.5% for C₂₊ products with a current density of 400 mA cm⁻² at −1.15 V vs. RHE.¹⁴⁸In situ Raman spectroscopy and attenuate total reflectance-infrared absorption spectroscopy showed that CuZnAl₂O₄ enhanced both the adsorption activation of CO₂ and the dissociation of H₂O at the alloy/oxide interface. Then, *CO₂ was transferred to CuZn alloy through the interface, and the surface *H_ad attacked *CO₂ to form *COOH intermediates, accelerating the formation of *CO. CuZnAl₂O₄ oxides or positively charged Cu or Zn at the interface strengthened the adsorption of *CO. Subsequently, the high *CO surface coverage enhanced the C–C coupling step on the CuZn alloy surface, promoted the formation of C₂₊ intermediates, and further protonated to form *OCCOH. After that, the selectivity of C₂₊ products was determined by the subsequent *OCCOH hydride oxidation step. Here the adsorption and desorption of key intermediates were regulated.

Zhang et al. developed N₂ cold plasma to make a stable activation of Cu⁰/Cu⁺-on-Ag interface. The C₂ faradaic efficiency of Ag@Cu–CuN_x was 72% at −1.0 V vs. RHE, and the partial current density was −14.9 mA cm⁻². Low-temperature N₂ plasma could promote the formation of CuN_x nanoclusters, obtain Cu⁺/Cu⁰-on-Ag interface, and expose a large amount of Cu⁰. Due to the high stability of copper nitride, this structure not only ensured the stability of Cu⁺, but also ensured the electrocatalytic activity of CO₂. The s and p central antibonding states of the atom Cu⁰ in Ag@Cu–CuN_x had more electrons than those of Ag@Cu and Ag@CuO_x. The electron density difference at Cu⁰/Cu⁺ interface caused by electron structure reconstruction was conducive to *COOH dissociation to CO formation. The enhanced activity and stability of eCO₂RR into C₂ composites could be attributed to the Cu⁰/CuN_x synergistic effect. The presence of Cu⁺ on the Ag surface not only stabilized Cu⁺, but also improved the adsorption of *CO, which was conducive to C–C coupling to produce C₂ products.¹⁴⁹ Xiang et al. designed a heterostructure Cu@Cu_0.4W_0.6 catalyst for eCO₂RR to C₂ products under the guidance of theoretical calculations.¹⁵⁰ X-ray diffraction (XRD) and transmission electron microscopy (TEM) confirmed the heterostructure of the Cu@Cu@Cu_0.4W_0.6 catalyst, which was spherical in shape and 15.6 ± 0.5 nm in size. X-ray photoelectron spectroscopy showed that the electron transfer from Cu to tungsten (W) in the heterostructure Cu@Cu_0.4W_0.6 catalyst was more than that of pure Cu_0.4W_0.6. The pure Cu@Cu_0.4W_0.6 catalyst was dominated by H₂ evolution in eCO₂RR, and the selectivity and activity of the Cu@Cu_0.4W_0.6 catalyst for C₂ products were significantly improved, which was closely related to the mass ratio of Cu@Cu_0.4W_0.6 interface. Finally, the faradaic efficiency of the C₂ product in the alkaline electrolyte of the Cu@Cu_0.4W_0.6 catalyst was 60.9% at −1.0 V vs. RHE and the partial current density was 121.8 mA cm⁻². DFT showed that the heterostructure Cu@Cu_0.4W_0.6 inhibited H₂ evolution, which was conducive to the generation of CO and asymmetric CO–CHO coupling, which was related to the charge redistribution at the Cu@Cu_0.4W_0.6 interface. The high O affinity of W favored the cleavage of C–O bonds of *C₂H₃O intermediates and promoted the formation of C₂H₄. Wang et al. used an electrodeposition calcination process to prepare Cu₃Sn/Cu₆Sn₅ intermetallic heterojunctions. In the 0.1 M NaHCO₃ electrolyte, eCO₂RR produced 82% faradaic efficiency of HCOOH at −1.0 V vs. RHE, with a current density of 18.9 mA cm⁻² for up to 42 h. A current density of 148 mA cm⁻² and 87% faradaic efficiency for HCOOH production were achieved by using a gas diffusion electrode and a 1 M KOH electrolyte. The combination of experimental and theoretical calculations showed that the high catalytic activity between Cu₆Sn₅ and Cu₃Sn intermetallic compounds was mainly due to the interface between Cu₆Sn₅ and Cu₃Sn, in which the adsorption of HCOO* intermediates was stronger than that of COOH*, and the free energy of hydrogen adsorption shifted upward, resulting in inhibition of hydrogen evolution and selective HCOOH formation.⁴⁴

4. Summary and outlook

This paper investigates the impact of electronic structure optimization on eCO₂RR behavior and elucidates the underlying mechanisms. We have reviewed various defect strategies employed to adjust the electronic structure, including heteroatom doping, vacancy engineering, morphological engineering, twin crystal formation, and heterostructure engineering. These approaches modify the electronic structure of electrocatalysts, enhance electronic interactions, and increase the density of active sites. Such effective regulation aids in controlling the adsorption and desorption processes of key intermediates and products on the electrocatalyst, leading to improved activity, selectivity, and stability.

Despite the acknowledged efficacy of electronic structure modulation as a strategy for enhancing eCO₂RR, several challenges still exist. For instance, the dynamics of electrons and reaction intermediates during eCO₂RR add complexity, complicating the elucidation of structure–performance relationships. In this context, in situ or operando characterization techniques are crucial for understanding these dynamic interactions. Although DFT is commonly used for adsorption energy calculations, direct observations of adsorption behavior have not been well correlated. At present, the role played by in situ/operando methods in eCO₂RR research is to monitor the behavior of intermediates and products, and detect the microenvironment near the electrode surface. However, in most cases, it is difficult to capture dynamic information due to the influence of bubble formation on the sample on detection, and electrocatalysts often undergo chemical state transformation, which requires the combination of multiple in situ characterization techniques for a deep and comprehensive understanding of the catalytic mechanism. We believe that future research into defective catalyst materials using advanced operando technology is critical to solving scientific challenges.

Furthermore, research on Cu-based catalysts is still not exhaustive. Although various methods have been employed to enhance catalyst performance, transitioning from excellent laboratory results to practical, large-scale applications continues to face significant hurdles. Thus, developing more efficient and scalable catalysts remains critically important.

To achieve real-world application of eCO₂RR, the stability of the defect Cu-based catalyst is important and needs to be considered from many aspects. First, as for the catalyst material itself, the defect needs to be within a reasonable range to promote the eCO₂RR. While the current research on Cu-based materials is focused on the low current density, it is necessary to test their stability at large current density over 1 A cm⁻² in the future. Second, as for the purity of carbon dioxide, due to the accompanying impurities like NO_x, SO_x, the active site may be poisoned and hence worsen the stability. Therefore, developing poisoning-resistant catalysts or improving the purity of CO₂ is necessary. Third, as for the device, in industrial-scale electrolysis the temperature and pressure undulation is much more significant than in lab-scale research. Along with the local pH change and the problem of salt accumulation in long-term operation, the management of heat, pressure, and electrolyte should be taken into account.

In the field of electrocatalysis for the eCO₂RR, the development of efficient and effective datasets is crucial. Accurate prediction of the activity, selectivity, and stability of different electrocatalysts is essential in designing superior materials for eCO₂RR electrocatalysis. Machine learning can aid in this process by analyzing large datasets and identifying patterns and correlations between various properties of the electrocatalysts.^151,152 Through high-throughput calculation of some key descriptors such as adsorption energies, d-band center, and coordination number by well-constructed machine learning models, the catalytic activity, optimal composition, active sites, and eCO₂RR pathway over various possible materials can be predicted and understood.¹⁵³ Accurate datasets with precise measurements of key properties can aid in the development of reliable machine learning models and enable the discovery of novel electrocatalysts with improved eCO₂RR performance.

A bridging relationship across interdisciplinary research should be established to guide the targeted preparation and regulation of catalysts, while clarifying the reaction pathway of eCO₂RR. With existing technology it is still difficult to explain the changes in catalyst structure relative to products at different potentials, as the relationship between structure and products has not been well understood. Therefore, it is necessary to link interdisciplinary fields such as materials science, chemical engineering, electrochemistry, and computational science to better design catalysts for eCO₂RR, while clarifying the structure–activity relationship between the structure and the product.

Bridging these gaps will deepen our understanding of eCO₂RR catalyst performance through defect engineering, leading to enhanced material efficiency and stability at larger scales and higher currents. Beyond catalyst design, the influence of the catalyst reaction system must also be considered. Current systems such as flow cells and MEAs are constrained by membrane limitations and spatial constraints, which impede progress to some extent. Therefore, the rational design of both the catalyst and the system in which it is used is essential.

Data availability

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

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the Ministry of Science and Technology (MOST) of China through the Key Project of Research & Development (2021YFF0500502), Guangdong-Shenzhen Joint Research Fund (2023A1515111016).

References

1. O. S. Bushuyev, P. De Luna, C. T. Dinh, L. Tao, G. Saur, J. van de Lagemaat, S. O. Kelley and E. H. Sargent, What Should We Make with CO₂ and How Can We Make It?, Joule, 2018, 2, 825–832.
2. L. Li, S. Wang, C. Wan, C. Xu, M. Zuo, Y. Sun and Y. Xie, New paradigm of in situ characterization for next-generation CO₂ electroreduction towards multi-carbon products over Cu-based catalysts, Next Energy, 2023, 1, 100027.
3. R. G. Grim, J. R. Ferrell III, Z. Huang, L. Tao and M. G. Resch, The feasibility of direct CO₂ conversion technologies on impacting mid-century climate goals, Joule, 2023, 7, 1684–1699.
4. X. Yang, C. P. Nielsen, S. Song and M. B. McElroy, Breaking the hard-to-abate bottleneck in China's path to carbon neutrality with clean hydrogen, Nat. Energy, 2022, 7, 955–965.
5. J. Zhong, X. Zhang, L. Guo, D. Wang, C. Miao and X. Zhang, Ongoing CO₂ monitoring verify CO₂ emissions and sinks in China during 2018–2021, Sci. Bull., 2023, 68, 2467–2476.
6. M. Jouny, W. Luc and F. Jiao, Correction to “General Techno-Economic Analysis of CO₂ Electrolysis Systems”, Ind. Eng. Chem. Res., 2020, 59, 8121–8123.
7. L. Li, X. Li, Y. Sun and Y. Xie, Rational design of electrocatalytic carbon dioxide reduction for a zero-carbon network, Chem. Soc. Rev., 2022, 51, 1234–1252.
8. C. W. Jones, A.-H. A. Park and P. Wright, The Complex Interplay of Separations, Reactions and Storage in Pursuit of Carbon Neutrality: A Central Role for Nanomaterials, Acc. Chem. Res., 2023, 56, 3545–3546.
9. S. Overa, B. H. Ko, Y. Zhao and F. Jiao, Electrochemical Approaches for CO₂ Conversion to Chemicals: A Journey toward Practical Applications, Acc. Chem. Res., 2022, 55, 638–648.
10. T. O’ Carroll, X. Yang, K. J. Gordon, L. Fei and G. Wu, Ethylene Electrosynthesis via Selective CO₂ Reduction: Fundamental Considerations, Strategies, and Challenges, Adv. Energy Mater., 2024, 14, 2401558.
11. Y. Kwon, B. Wu, N. Zhang, D. Hand, T. Mou, X. Han and Q. Chang, Recent advances in integrated capture and electrochemical conversion of CO₂, MRS Commun., 2024, 14, 728–740.
12. T. Lei, D. Wang, X. Yu, S. Ma, W. Zhao, C. Cui, J. Meng, S. Tao and D. Guan, Global iron and steel plant CO₂ emissions and carbon-neutrality pathways, Nature, 2023, 622, 514–520.
13. D. Grant, T. Hansen, A. Jorgenson and W. Longhofer, A worldwide analysis of stranded fossil fuel assets’ impact on power plants’ CO₂ emissions, Nat. Commun., 2024, 15, 7517.
14. H. Na, Y. Yuan, T. Du, T. Zhang, X. Zhao, J. Sun, Z. Qiu and L. Zhang, Multi-process production occurs in the iron and steel industry, supporting ‘dual carbon’ target: An in-depth study of CO₂ emissions from different processes, J. Environ. Sci., 2024, 140, 46–58.
15. S. M. Jarvis and S. Samsatli, Technologies and infrastructures underpinning future CO₂ value chains: A comprehensive review and comparative analysis, Renewable Sustainable Energy Rev., 2018, 85, 46–68.
16. S. Mondal and S. C. Peter, A Perspective on Electrochemical Point Source Utilization of CO₂ and Other Flue Gas Components to Value Added Chemicals, Adv. Mater., 2024, 36, 2407124.
17. M. Jiang, H. Wang, M. Zhu, X. Luo, Y. He, M. Wang, C. Wu, L. Zhang, X. Li, X. Liao, Z. Jiang and Z. Jin, Review on strategies for improving the added value and expanding the scope of CO₂ electroreduction products, Chem. Soc. Rev., 2024, 53, 5149–5189.
18. H. Wen, Y. Liu, S. Liu, Z. Peng, X. Wu, H. Yuan, J. Jiang and B. Li, Heterogeneous Catalysis in Production and Utilization of Formic Acid for Renewable Energy, Small, 2024, 20, 2305405.
19. K. Peramaiah, M. Yi, I. Dutta, S. Chatterjee, H. Zhang, Z. Lai and K.-W. Huang, Catalyst Design and Engineering for CO₂-to-Formic Acid Electrosynthesis for a Low-Carbon Economy, Adv. Mater., 2024, 2404980.
20. D. S. Tran, N.-N. Vu, H.-E. Nemamcha, C. Boisvert, U. Legrand, A. G. Fink, F. Navarro-Pardo, C.-T. Dinh and P. Nguyen-Tri, Design of electrocatalysts and electrodes for CO₂ electroreduction to formic acid and formate, Coord. Chem. Rev., 2025, 524, 216322.
21. F. Valentini, V. Kozell, C. Petrucci, A. Marrocchi, Y. Gu, D. Gelman and L. Vaccaro, Formic acid, a biomass-derived source of energy and hydrogen for biomass upgrading, Energy Environ. Sci., 2019, 12, 2646–2664.
22. B. Chang, Z. Min, N. Liu, N. Wang, M. Fan, J. Fan and J. Wang, Electrocatalytic CO₂ reduction to syngas, Green Energy Environ., 2024, 9, 1085–1100.
23. K. Cheng, J. Kang, Q. Zhang and Y. Wang, Reaction coupling as a promising methodology for selective conversion of syngas into hydrocarbons beyond Fischer-Tropsch synthesis, Sci. China: Chem., 2017, 60, 1382–1385.
24. X. Deng, D. Alfonso, T.-D. Nguyen-Phan and D. R. Kauffman, Resolving the Size-Dependent Transition between CO₂ Reduction Reaction and H₂ Evolution Reaction Selectivity in Sub-5 nm Silver Nanoparticle Electrocatalysts, ACS Catal., 2022, 12, 5921–5929.
25. S. Liu, X.-Z. Wang, H. Tao, T. Li, Q. Liu, Z. Xu, X.-Z. Fu and J.-L. Luo, Ultrathin 5-fold twinned sub-25 nm silver nanowires enable highly selective electroreduction of CO₂ to CO, Nano Energy, 2018, 45, 456–462.
26. S. Liu, H. Tao, L. Zeng, Q. Liu, Z. Xu, Q. Liu and J.-L. Luo, Shape-Dependent Electrocatalytic Reduction of CO₂ to CO on Triangular Silver Nanoplates, J. Am. Chem. Soc., 2017, 139, 2160–2163.
27. D.-R. Yang, L. Liu, Q. Zhang, Y. Shi, Y. Zhou, C. Liu, F.-B. Wang and X.-H. Xia, Importance of Au nanostructures in CO₂ electrochemical reduction reaction, Sci. Bull., 2020, 65, 796–802.
28. S. Li, A. V. Nagarajan, X. Du, Y. Li, Z. Liu, D. R. Kauffman, G. Mpourmpakis and R. Jin, Dissecting Critical Factors for Electrochemical CO₂ Reduction on Atomically Precise Au Nanoclusters, Angew. Chem., Int. Ed., 2022, 61, e202211771.
29. H. S. Jeon, I. Sinev, F. Scholten, N. J. Divins, I. Zegkinoglou, L. Pielsticker and B. R. Cuenya, Operando Evolution of the Structure and Oxidation State of Size-Controlled Zn Nanoparticles during CO₂ Electroreduction, J. Am. Chem. Soc., 2018, 140, 9383–9386.
30. M. P. L. Kang, M. J. Kolb, F. Calle-Vallejo and B. S. Yeo, The Role of Undercoordinated Sites on Zinc Electrodes for CO₂ Reduction to CO, Adv. Funct. Mater., 2022, 32, 2111597.
31. J. Li, J. Li, X. Liu, J. Chen, P. Tian, S. Dai, M. Zhu and Y.-F. Han, Probing the role of surface hydroxyls for Bi, Sn and In catalysts during CO₂ Reduction, Appl. Catal., B, 2021, 298, 120581.
32. S. Ruan, B. Zhang, J. Zou, W. Zhong, X. He, J. Lu, Q. Zhang, Y. Wang and S. Xie, Bismuth nanosheets with rich grain boundaries for efficient electroreduction of CO₂ to formate under high pressures, Chin. J. Catal., 2022, 43, 3161–3169.
33. 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, Size Effects of Highly Dispersed Bismuth Nanoparticles on Electrocatalytic Reduction of Carbon Dioxide to Formic Acid, J. Am. Chem. Soc., 2023, 145, 14133–14142.
34. D. Ma, C. Zhi, Y. Zhang, J. Chen, Y. Zhang and J.-W. Shi, A Review on the Influence of Crystal Facets on the Product Selectivity of CO₂RR over Cu Metal Catalysts, ACS Nano, 2024, 18, 21714–21746.
35. A. R. Woldu, Z. Huang, P. Zhao, L. Hu and D. Astruc, Electrochemical CO₂ reduction (CO₂RR) to multi-carbon products over copper-based catalysts, Coord. Chem. Rev., 2022, 454, 214340.
36. I.-Y. Jeon, S. Zhang, L. Zhang, H.-J. Choi, J.-M. Seo, Z. Xia, L. Dai and J.-B. Baek, Edge-Selectively Sulfurized Graphene Nanoplatelets as Efficient Metal-Free Electrocatalysts for Oxygen Reduction Reaction: The Electron Spin Effect, Adv. Mater., 2013, 25, 6138–6145.
37. R. Liu, D. Wu, X. Feng and K. Müllen, Nitrogen-Doped Ordered Mesoporous Graphitic Arrays with High Electrocatalytic Activity for Oxygen Reduction, Angew. Chem., Int. Ed., 2010, 49, 2565–2569.
38. S. Ni, Z. Li and J. Yang, Oxygen molecule dissociation on carbon nanostructures with different types of nitrogen doping, Nanoscale, 2012, 4, 1184–1189.
39. X. Yan, Y. Jia and X. Yao, Defects on carbons for electrocatalytic oxygen reduction, Chem. Soc. Rev., 2018, 47, 7628–7658.
40. Q. Wu, X. Yan, Y. Jia and X. Yao, Defective carbon-based materials: controllable synthesis and electrochemical applications, EnergyChem, 2021, 3, 100059.
41. H. Yang, S. Li and Q. Xu, Efficient strategies for promoting the electrochemical reduction of CO₂ to C₂₊ products over Cu-based catalysts, Chin. J. Catal., 2023, 48, 32–65.
42. S. Liu, B. Zhang, L. Zhang and J. Sun, Rational design strategies of Cu-based electrocatalysts for CO₂ electroreduction to C₂ products, J. Energy Chem., 2022, 71, 63–82.
43. Y. Zang, P. Wei, H. Li, D. Gao and G. Wang, Catalyst Design for Electrolytic CO₂ Reduction Toward Low-Carbon Fuels and Chemicals, Electrochem. Energy Rev., 2022, 5, 29.
44. J. Wang, J. Zou, X. Hu, S. Ning, X. Wang, X. Kang and S. Chen, Heterostructured intermetallic CuSn catalysts: high performance towards the electrochemical reduction of CO₂ to formate, J. Mater. Chem. A, 2019, 7, 27514–27521.
45. Y. Y. Birdja, E. Pérez-Gallent, M. C. Figueiredo, A. J. Göttle, F. Calle-Vallejo and M. T. M. Koper, Advances and challenges in understanding the electrocatalytic conversion of carbon dioxide to fuels, Nat. Energy, 2019, 4, 732–745.
46. S. Nitopi, E. Bertheussen, S. B. Scott, X. Liu, A. K. Engstfeld, S. Horch, B. Seger, I. E. L. Stephens, K. Chan, C. Hahn, J. K. Nørskov, T. F. Jaramillo and I. Chorkendorff, Progress and Perspectives of Electrochemical CO₂ Reduction on Copper in Aqueous Electrolyte, Chem. Rev., 2019, 119, 7610–7672.
47. M. B. Ross, P. De Luna, Y. Li, C.-T. Dinh, D. Kim, P. Yang and E. H. Sargent, Designing materials for electrochemical carbon dioxide recycling, Nat. Catal., 2019, 2, 648–658.
48. G. Zhang, Z.-J. Zhao, D. Cheng, H. Li, J. Yu, Q. Wang, H. Gao, J. Guo, H. Wang, G. A. Ozin, T. Wang and J. Gong, Efficient CO₂ electroreduction on facet-selective copper films with high conversion rate, Nat. Commun., 2021, 12, 5745.
49. L. Wang, W. Chen, D. Zhang, Y. Du, R. Amal, S. Qiao, J. Wu and Z. Yin, Surface strategies for catalytic CO₂ reduction: from two-dimensional materials to nanoclusters to single atoms, Chem. Soc. Rev., 2019, 48, 5310–5349.
50. X. Chang, T. Wang and J. Gong, CO₂ photo-reduction: insights into CO₂ activation and reaction on surfaces of photocatalysts, Energy Environ. Sci., 2016, 9, 2177–2196.
51. M. Gattrell, N. Gupta and A. Co, A review of the aqueous electrochemical reduction of CO₂ to hydrocarbons at copper, J. Electroanal. Chem., 2006, 594, 1–19.
52. M. Favaro, H. Xiao, T. Cheng, W. A. Goddard, J. Yano and E. J. Crumlin, Subsurface oxide plays a critical role in CO₂ activation by Cu(111) surfaces to form chemisorbed CO₂, the first step in reduction of CO₂, Proc. Natl. Acad. Sci. U. S. A., 2017, 114, 6706–6711.
53. Z. Sun, T. Ma, H. Tao, Q. Fan and B. Han, Fundamentals and Challenges of Electrochemical CO₂ Reduction Using Two-Dimensional Materials, Chem, 2017, 3, 560–587.
54. A. J. Garza, A. T. Bell and M. Head-Gordon, Mechanism of CO₂ Reduction at Copper Surfaces: Pathways to C₂ Products, ACS Catal., 2018, 8, 1490–1499.
55. M. Zheng, P. Wang, X. Zhi, K. Yang, Y. Jiao, J. Duan, Y. Zheng and S.-Z. Qiao, Electrocatalytic CO₂-to-C₂₊ with Ampere-Level Current on Heteroatom-Engineered Copper via Tuning *CO Intermediate Coverage, J. Am. Chem. Soc., 2022, 144, 14936–14944.
56. X. Liu, P. Schlexer, J. Xiao, Y. Ji, L. Wang, R. B. Sandberg, M. Tang, K. S. Brown, H. Peng, S. Ringe, C. Hahn, T. F. Jaramillo, J. K. Nørskov and K. Chan, pH effects on the electrochemical reduction of CO(₂) towards C₂ products on stepped copper, Nat. Commun., 2019, 10, 32.
57. Y. Kim, S. Park, S.-J. Shin, W. Choi, B. K. Min, H. Kim, W. Kim and Y. J. Hwang, Time-resolved observation of C–C coupling intermediates on Cu electrodes for selective electrochemical CO₂ reduction, Energy Environ. Sci., 2020, 13, 4301–4311.
58. Y. Hori, R. Takahashi, Y. Yoshinami and A. Murata, Electrochemical Reduction of CO at a Copper Electrode, J. Phys. Chem. B, 1997, 101, 7075–7081.
59. J. D. Goodpaster, A. T. Bell and M. Head-Gordon, Identification of Possible Pathways for C–C Bond Formation during Electrochemical Reduction of CO₂: New Theoretical Insights from an Improved Electrochemical Model, J. Phys. Chem. Lett., 2016, 7, 1471–1477.
60. E. Pérez-Gallent, M. C. Figueiredo, F. Calle-Vallejo and M. T. M. Koper, Spectroscopic Observation of a Hydrogenated CO Dimer Intermediate During CO Reduction on Cu(100) Electrodes, Angew. Chem., Int. Ed., 2017, 56, 3621–3624.
61. F. Calle-Vallejo and M. T. M. Koper, Theoretical Considerations on the Electroreduction of CO to C₂ Species on Cu(100) Electrodes, Angew. Chem., Int. Ed., 2013, 52, 7282–7285.
62. W. Ma, S. Xie, T. Liu, Q. Fan, J. Ye, F. Sun, Z. Jiang, Q. Zhang, J. Cheng and Y. Wang, Electrocatalytic reduction of CO₂ to ethylene and ethanol through hydrogen-assisted C–C coupling over fluorine-modified copper, Nat. Catal., 2020, 3, 478–487.
63. T. Cheng, H. Xiao and W. A. Goddard, Full atomistic reaction mechanism with kinetics for CO reduction on Cu(100) from ab initio molecular dynamics free-energy calculations at 298 K, Proc. Natl. Acad. Sci. U. S. A., 2017, 114, 1795–1800.
64. Y. Jia, D. Li, C. Liu, T. Song, L. Duan, F. Li, F. Li, Y. Ji, L. Sun and K. Fan, Regulating Cu Oxidation State for Electrocatalytic CO₂ Conversion into CO with Near-Unity Selectivity via Oxygen Spillover, Small, 2024, 20, 2402537.
65. M. Sun, A. Staykov and M. Yamauchi, Understanding the Roles of Hydroxide in CO₂ Electroreduction on a Cu Electrode for Achieving Variable Selectivity, ACS Catal., 2022, 12, 14856–14863.
66. Y. Wen, N. Fang, W. Liu, T. Yang, Y. Xu and X. Huang, Cu₇S₄ nanosheets enriched with Cu–S bond for highly active and selective CO₂ electroreduction to formate, J. Mater. Chem. A, 2023, 11, 10823–10827.
67. M. Wang, L. Wan, J. Cheng and J. Luo, Scalable preparation of a CuO nanosheet array via corrosion engineering for selective C–C coupling in CO₂ electroreduction, J. Mater. Chem. A, 2022, 10, 14070–14077.
68. S. Y. Lee, H. Jung, N.-K. Kim, H.-S. Oh, B. K. Min and Y. J. Hwang, Mixed Copper States in Anodized Cu Electrocatalyst for Stable and Selective Ethylene Production from CO₂ Reduction, J. Am. Chem. Soc., 2018, 140, 8681–8689.
69. Y. Li, C.-Z. Huo, H.-J. Wang, Z.-X. Ye, P.-P. Luo, X.-X. Cao and T.-B. Lu, Coupling CO₂ reduction with CH₃OH oxidation for efficient electrosynthesis of formate on hierarchical bifunctional CuSn alloy, Nano Energy, 2022, 98, 107277.
70. Y. Hu, X. Yan, L. Zhou, P. Chen, Q. Pang and Y. Chen, Improved Energy Transfer in Mn-Doped Cs₃Cu₂I₅ Microcrystals Induced by Localized Lattice Distortion, J. Phys. Chem. Lett., 2022, 13, 10786–10792.
71. Y. Xie, P. Ou, X. Wang, Z. Xu, Y. C. Li, Z. Wang, J. E. Huang, J. Wicks, C. McCallum, N. Wang, Y. Wang, T. Chen, B. T. W. Lo, D. Sinton, J. C. Yu, Y. Wang and E. H. Sargent, High carbon utilization in CO₂ reduction to multi-carbon products in acidic media, Nat. Catal., 2022, 5, 564–570.
72. X. Li, M. Qin, X. Wu, X. Lv, J. Wang, Y. Wang and H. B. Wu, Enhanced CO Affinity on Cu Facilitates CO₂ Electroreduction toward Multi-Carbon Products, Small, 2023, 19, 2302530.
73. Z. Zhang, S. Chen, J. Zhu, C. Ye, Y. Mao, B. Wang, G. Zhou, L. Mai, Z. Wang, X. Liu and D. Wang, Charge-Separated Pd^δ−–Cu^δ+ Atom Pairs Promote CO₂ Reduction to C₂, Nano Lett., 2023, 23, 2312–2320.
74. D. Xiao, X. Bao, M. Zhang, Z. Li, Z. Wang, Y. Gao, Z. Zheng, P. Wang, H. Cheng, Y. Liu, Y. Dai and B. Huang, Stabilizing Cu₂O for enhancing selectivity of CO₂ electroreduction to C₂H₄ with the modification of Pd nanoparticles, Chem. Eng. J., 2023, 452, 139358.
75. D. Zhou, C. Chen, Y. Zhang, M. Wang, S. Han, X. Dong, T. Yao, S. Jia, M. He, H. Wu and B. Han, Cooperation of Different Active Sites to Promote CO₂ Electroreduction to Multi-carbon Products at Ampere-Level, Angew. Chem., Int. Ed., 2024, 63, e202400439.
76. J. Shan, Y. Shi, H. Li, Z. Chen, C. Sun, Y. Shuai and Z. Wang, Effective CO₂ electroreduction toward C₂H₄ boosted by Ce-doped Cu nanoparticles, Chem. Eng. J., 2022, 433, 133769.
77. S. Zhen, G. Zhang, D. Cheng, H. Gao, L. Li, X. Lin, Z. Ding, Z.-J. Zhao and J. Gong, Nature of the Active Sites of Copper Zinc Catalysts for Carbon Dioxide Electroreduction, Angew. Chem., Int. Ed., 2022, 61, e202201913.
78. X. Ma, X. Song, L. Zhang, L. Wu, J. Feng, S. Jia, X. Tan, L. Xu, X. Sun and B. Han, Stabilizing Cu⁰–Cu⁺ sites by Pb-doping for highly efficient CO₂ electroreduction to C₂ products, Green Chem., 2023, 25, 7635–7641.
79. L. Zhang, J. Feng, L. Wu, X. Ma, X. Song, S. Jia, X. Tan, X. Jin, Q. Zhu, X. Kang, J. Ma, Q. Qian, L. Zheng, X. Sun and B. Han, Oxophilicity-Controlled CO₂ Electroreduction to C₂₊ Alcohols over Lewis Acid Metal-Doped Cu^δ+ Catalysts, J. Am. Chem. Soc., 2023, 145, 21945–21954.
80. Z. Wei, S. Yue, S. Gao, M. Cao and R. Cao, Synergetic effects of gold-doped copper nanowires with low Au content for enhanced electrocatalytic CO₂ reduction to multicarbon products, Nano Res., 2023, 16, 7777–7783.
81. P. Wang, H. Yang, C. Tang, Y. Wu, Y. Zheng, T. Cheng, K. Davey, X. Huang and S.-Z. Qiao, Boosting electrocatalytic CO₂–to–ethanol production via asymmetric C–C coupling, Nat. Commun., 2022, 13, 3754.
82. Y. Yang, L. Chen, Z. Guo, S. Liu, P.-d. Wu, Z. Fang, K. Zhang and H. Li, Remote p–d orbital hybridization via first/second-layer coordination of Fe single atoms with heteroatoms for enhanced electrochemical CO₂-to-CO reduction, J. Mater. Chem. A, 2024, 12, 8991–9001.
83. G. Li, Y. Qin, Y. Wu, L. Pei, Q. Hu, H. Yang, Q. Zhang, J. Liu and C. He, Nitrogen and sulfur dual-doped high-surface-area hollow carbon nanospheres for efficient CO₂ reduction, Chin. J. Catal., 2020, 41, 830–838.
84. M. Wang, L. Kong, X. Lu and C.-M. L. Wu, Coordination Environment Engineering to Regulate the Adsorption Strength of Intermediates in Single Atom Catalysts for High-performance CO₂ Reaction Reduction, Small, 2024, 20, 2310339.
85. W.-F. Xiong, W.-Z. Cai, J. Wang, D.-H. Si, S.-Y. Gao, H.-F. Li and R. Cao, Br, O-Modified Cu(111) Interface Promotes CO₂ Reduction to Multicarbon Products, Small Methods, 2024, 2301807.
86. R. Wang, J. Liu, Q. Huang, L.-Z. Dong, S.-L. Li and Y.-Q. Lan, Partial Coordination-Perturbed Bi-Copper Sites for Selective Electroreduction of CO₂ to Hydrocarbons, Angew. Chem., Int. Ed., 2021, 60, 19829–19835.
87. L. Cheng, R. Wang, W. Si, Y. Deng, J. Li and Y. Peng, Enhancing CO₂ Electroreduction Performance through Si-Doped CuO: Stabilization of Cu⁺/Cu⁰ Sites and Improved C₂ Product Selectivity, ACS Catal., 2024, 14, 10324–10333.
88. S. Liang, J. Xiao, T. Zhang, Y. Zheng, Q. Wang and B. Liu, Sulfur Changes the Electrochemical CO₂ Reduction Pathway over Cu Electrocatalysts, Angew. Chem., Int. Ed., 2023, 62, e202310740.
89. H. Li, X. Qin, T. Jiang, X.-Y. Ma, K. Jiang and W.-B. Cai, Changing the Product Selectivity for Electrocatalysis of CO₂ Reduction Reaction on Plated Cu Electrodes, ChemCatChem, 2019, 11, 6139–6146.
90. X. Kong, C. Wang, H. Zheng, Z. Geng, J. Bao and J. Zeng, Enhance the activity of multi-carbon products for Cu via P doping towards CO₂ reduction, Sci. China: Chem., 2021, 64, 1096–1102.
91. Y. Yao, Y. Zhou, X. Liu, Y. Li, D. Wang, X. Chi, X. Wang, R. Zhao, H. Zhang, Y. Sun, Z.-Y. Yang, Y. Wei and Y.-M. Yan, Restraining lattice oxygen of Cu₂O by enhanced Cu-O hybridization for selective and stable production of ethylene with CO₂ electroreduction, J. Mater. Chem. A, 2022, 10, 20914–20923.
92. Y. Zhou, Y. Yao, R. Zhao, X. Wang, Z. Fu, D. Wang, H. Wang, L. Zhao, W. Ni, Z. Yang and Y.-M. Yan, Stabilization of Cu⁺ via Strong Electronic Interaction for Selective and Stable CO₂ Electroreduction, Angew. Chem., Int. Ed., 2022, 61, e202205832.
93. X. Yan, C. Chen, Y. Wu, Y. Chen, J. Zhang, R. Feng, J. Zhang and B. Han, Boosting CO₂ electroreduction to C₂₊ products on fluorine-doped copper, Green Chem., 2022, 24, 1989–1994.
94. R. Yang, J. Duan, P. Dong, Q. Wen, M. Wu, Y. Liu, Y. Liu, H. Li and T. Zhai, In Situ Halogen-Ion Leaching Regulates Multiple Sites on Tandem Catalysts for Efficient CO₂ Electroreduction to C₂₊ Products, Angew. Chem., Int. Ed., 2022, 61, e202116706.
95. H. Li, T. Liu, P. Wei, L. Lin, D. Gao, G. Wang and X. Bao, High-Rate CO₂ Electroreduction to C₂₊ Products over a Copper-Copper Iodide Catalyst, Angew. Chem., Int. Ed., 2021, 60, 14329–14333.
96. M. Li, Y. Ma, J. Chen, R. Lawrence, W. Luo, M. Sacchi, W. Jiang and J. Yang, Residual Chlorine Induced Cationic Active Species on a Porous Copper Electrocatalyst for Highly Stable Electrochemical CO₂ Reduction to C₂₊, Angew. Chem., Int. Ed., 2021, 60, 11487–11493.
97. J. J. Masana, B. Peng, Z. Shuai, M. Qiu and Y. Yu, Influence of halide ions on the electrochemical reduction of carbon dioxide over a copper surface, J. Mater. Chem. A, 2022, 10, 1086–1104.
98. P. Yang, Z.-J. Zhao, X. Chang, R. Mu, S. Zha, G. Zhang and J. Gong, The Functionality of Surface Hydroxy Groups on the Selectivity and Activity of Carbon Dioxide Reduction over Cuprous Oxide in Aqueous Solutions, Angew. Chem., Int. Ed., 2018, 57, 7724–7728.
99. S. B. Varandili, J. Huang, E. Oveisi, G. L. De Gregorio, M. Mensi, M. Strach, J. Vavra, C. Gadiyar, A. Bhowmik and R. Buonsanti, Synthesis of Cu/CeO_2-x Nanocrystalline Heterodimers with Interfacial Active Sites To Promote CO₂ Electroreduction, ACS Catal., 2019, 9, 5035–5046.
100. W. Zhu, L. Zhang, P. Yang, C. Hu, H. Dong, Z.-J. Zhao, R. Mu and J. Gong, Formation of Enriched Vacancies for Enhanced CO₂ Electrocatalytic Reduction over AuCu Alloys, ACS Energy Lett., 2018, 3, 2144–2149.
101. Z. Gu, H. Shen, Z. Chen, Y. Yang, C. Yang, Y. Ji, Y. Wang, C. Zhu, J. Liu, J. Li, T.-K. Sham, X. Xu and G. Zheng, Efficient Electrocatalytic CO₂ Reduction to C₂₊ Alcohols at Defect-Site-Rich Cu Surface, Joule, 2021, 5, 429–440.
102. J. Wang, X. Zheng, G. Wang, Y. Cao, W. Ding, J. Zhang, H. Wu, J. Ding, H. Hu, X. Han, T. Ma, Y. Deng and W. Hu, Defective Bimetallic Selenides for Selective CO₂ Electroreduction to CO, Adv. Mater., 2022, 34, 2106354.
103. W. Wang, Z. Ma, X. Fei, X. Wang, Z. Yang, Y. Wang, J. Zhang, H. Ning, N. Tsubaki and M. Wu, Joint tuning the morphology and oxygen vacancy of Cu₂O by ionic liquid enables high-efficient CO₂ reduction to C₂ products, Chem. Eng. J., 2022, 436, 135029.
104. S. Li, J. Yu, S. Zhang, W. Qiu, X. Tang, Z. Lin, R. Cai, Y. Fang, S. Yang and X. Cai, Operando Reconstruction of Porous Carbon Supported Copper Selenide Promotes the C₂ Production from CO₂RR, Adv. Funct. Mater., 2024, 34, 2311989.
105. H. Wang, X. Bi, Y. Yan, Y. Zhao, Z. Yang, H. Ning and M. Wu, Efficient Electrocatalytic Reduction of CO₂ to Ethanol Enhanced by Spacing Effect of Cu-Cu in Cu_21-xSe Nanosheets, Adv. Funct. Mater., 2023, 33, 2214946.
106. C. Peng, G. Luo, Z. Xu, S. Yan, J. Zhang, M. Chen, L. Qian, W. Wei, Q. Han and G. Zheng, Lithiation-Enabled High-Density Nitrogen Vacancies Electrocatalyze CO₂ to C₂ Products, Adv. Mater., 2021, 33, 2103150.
107. T.-T. Zhuang, Z.-Q. Liang, A. Seifitokaldani, Y. Li, P. De Luna, T. Burdyny, F. Che, F. Meng, Y. Min, R. Quintero-Bermudez, C. T. Dinh, Y. Pang, M. Zhong, B. Zhang, J. Li, P.-N. Chen, X.-L. Zheng, H. Liang, W.-N. Ge, B.-J. Ye, D. Sinton, S.-H. Yu and E. H. Sargent, Steering post-C-C coupling selectivity enables high efficiency electroreduction of carbon dioxide to multi-carbon alcohols, Nat. Catal., 2018, 1, 421–428.
108. S. Li, H. Duan, J. Yu, C. Qiu, R. Yu, Y. Chen, Y. Fang, X. Cai and S. Yang, Cu Vacancy Induced Product Switching from Formate to CO for CO₂ Reduction on Copper Sulfide, ACS Catal., 2022, 12, 9074–9082.
109. R. Chen, X. Zu, J. Zhu, Y. Zhao, Y. Li, Z. Hu, S. Wang, M. Fan, S. Zhu, H. Zhang, B. Ye, Y. Sun and Y. Xie, Dynamically Reconstructed Triple-Copper-Vacancy Associates Confined in Cu Nanowires Enabling High-Rate and Selective CO₂ Electroreduction to C₂₊ Products, Adv. Mater., 2024, 36, 2314209.
110. C. Peng, G. Luo, J. Zhang, M. Chen, Z. Wang, T.-K. Sham, L. Zhang, Y. Li and G. Zheng, Double sulfur vacancies by lithium tuning enhance CO₂ electroreduction to n-propanol, Nat. Commun., 2021, 12, 1580.
111. K. Rossi and R. Buonsanti, Shaping Copper Nanocatalysts to Steer Selectivity in the Electrochemical CO₂ Reduction Reaction, Acc. Chem. Res., 2022, 55, 629–637.
112. S. Liu and S. Huang, Size effects and active sites of Cu nanoparticle catalysts for CO₂ electroreduction, Appl. Surf. Sci., 2019, 475, 20–27.
113. A. Vijayakumar, Y. Zhao, K. Wang, Y. Chao, Z. Chen, C. Wang and G. G. Wallace, The length dependent selectivity on aligned Cu nanowires for C₁ products from CO₂ Electroreduction, Electrochim. Acta, 2021, 394, 139099.
114. W. Rong, H. Zou, W. Zang, S. Xi, S. Wei, B. Long, J. Hu, Y. Ji and L. Duan, Size-Dependent Activity and Selectivity of Atomic-Level Copper Nanoclusters during CO/CO₂ Electroreduction, Angew. Chem., Int. Ed., 2021, 60, 466–472.
115. R. Reske, H. Mistry, F. Behafarid, B. Roldan Cuenya and P. Strasser, Particle Size Effects in the Catalytic Electroreduction of CO₂ on Cu Nanoparticles, J. Am. Chem. Soc., 2014, 136, 6978–6986.
116. J. Huang, N. Hörmann, E. Oveisi, A. Loiudice, G. L. De Gregorio, O. Andreussi, N. Marzari and R. Buonsanti, Potential-induced nanoclustering of metallic catalysts during electrochemical CO₂ reduction, Nat. Commun., 2018, 9, 3117.
117. P. Iyengar, J. Huang, G. L. De Gregorio, C. Gadiyar and R. Buonsanti, Size dependent selectivity of Cu nano-octahedra catalysts for the electrochemical reduction of CO₂ to CH₄, Chem. Commun., 2019, 55, 8796–8799.
118. P. Grosse, D. Gao, F. Scholten, I. Sinev, H. Mistry and B. Roldan, Cuenya, Dynamic Changes in the Structure, Chemical State and Catalytic Selectivity of Cu Nanocubes during CO₂ Electroreduction: Size and Support Effects, Angew. Chem., Int. Ed., 2018, 57, 6192–6197.
119. Y. Wang, Z. Chen, P. Han, Y. Du, Z. Gu, X. Xu and G. Zheng, Single-Atomic Cu with Multiple Oxygen Vacancies on Ceria for Electrocatalytic CO₂ Reduction to CH₄, ACS Catal., 2018, 8, 7113–7119.
120. K. Zhao, X. Nie, H. Wang, S. Chen, X. Quan, H. Yu, W. Choi, G. Zhang, B. Kim and J. G. Chen, Selective electroreduction of CO₂ to acetone by single copper atoms anchored on N-doped porous carbon, Nat. Commun., 2020, 11, 2455.
121. J. Jiao, R. Lin, S. Liu, W.-C. Cheong, C. Zhang, Z. Chen, Y. Pan, J. Tang, K. Wu, S.-F. Hung, H. M. Chen, L. Zheng, Q. Lu, X. Yang, B. Xu, H. Xiao, J. Li, D. Wang, Q. Peng, C. Chen and Y. Li, Copper atom-pair catalyst anchored on alloy nanowires for selective and efficient electrochemical reduction of CO₂, Nat. Chem., 2019, 11, 222–228.
122. J.-d. Yi, X. Gao, H. Zhou, W. Chen and Y. Wu, Design of Co-Cu Diatomic Site Catalysts for High-efficiency Synergistic CO₂ Electroreduction at Industrial-level Current Density, Angew. Chem., Int. Ed., 2022, 61, e202212329.
123. Q. Hu, Z. Han, X. Wang, G. Li, Z. Wang, X. Huang, H. Yang, X. Ren, Q. Zhang, J. Liu and C. He, Facile Synthesis of Sub-Nanometric Copper Clusters by Double Confinement Enables Selective Reduction of Carbon Dioxide to Methane, Angew. Chem., Int. Ed., 2020, 59, 19054–19059.
124. M. Ma, K. Djanashvili and W. A. Smith, Controllable Hydrocarbon Formation from the Electrochemical Reduction of CO₂ over Cu Nanowire Arrays, Angew. Chem., Int. Ed., 2016, 55, 6680–6684.
125. H. Zhang, Y. Zhang, Y. Li, S. Ahn, G. T. R. Palmore, J. Fu, A. A. Peterson and S. Sun, Cu nanowire-catalyzed electrochemical reduction of CO or CO₂, Nanoscale, 2019, 11, 12075–12079.
126. N.-T. Suen, Z.-R. Kong, C.-S. Hsu, H.-C. Chen, C.-W. Tung, Y.-R. Lu, C.-L. Dong, C.-C. Shen, J.-C. Chung and H. M. Chen, Morphology Manipulation of Copper Nanocrystals and Product Selectivity in the Electrocatalytic Reduction of Carbon Dioxide, ACS Catal., 2019, 9, 5217–5222.
127. Z. Chang, S. Huo, W. Zhang, J. Fang and H. Wang, The Tunable and Highly Selective Reduction Products on Ag@Cu Bimetallic Catalysts Toward CO₂ Electrochemical Reduction Reaction, J. Phys. Chem. C, 2017, 121, 11368–11379.
128. J.-J. Lv, M. Jouny, W. Luc, W. Zhu, J.-J. Zhu and F. Jiao, A Highly Porous Copper Electrocatalyst for Carbon Dioxide Reduction, Adv. Mater., 2018, 30, 1803111.
129. Q. Li, W. Zhu, J. Fu, H. Zhang, G. Wu and S. Sun, Controlled assembly of Cu nanoparticles on pyridinic-N rich graphene for electrochemical reduction of CO₂ to ethylene, Nano Energy, 2016, 24, 1–9.
130. C. Choi, S. Kwon, T. Cheng, M. Xu, P. Tieu, C. Lee, J. Cai, H. M. Lee, X. Pan, X. Duan, W. A. Goddard and Y. Huang, Highly active and stable stepped Cu surface for enhanced electrochemical CO₂ reduction to C₂H₄, Nat. Catal., 2020, 3, 804–812.
131. P. Iyengar, M. J. Kolb, J. R. Pankhurst, F. Calle-Vallejo and R. Buonsanti, Elucidating the Facet-Dependent Selectivity for CO₂ Electroreduction to Ethanol of Cu–Ag Tandem Catalysts, ACS Catal., 2021, 11, 4456–4463.
132. W. Xiong, D. Si, J. Yi, Y. Huang, H. Li and R. Cao, Morphology and composition dependence of multicomponent Cu-based nanoreactor for tandem electrocatalysis CO₂ reduction, Appl. Catal., B, 2022, 314, 121498.
133. W. Fang, R. Lu, F.-M. Li, C. He, D. Wu, K. Yue, Y. Mao, W. Guo, B. You, F. Song, T. Yao, Z. Wang and B. Y. Xia, Low-coordination Nanocrystalline Copper-based Catalysts through Theory-guided Electrochemical Restructuring for Selective CO₂ Reduction to Ethylene, Angew. Chem., Int. Ed., 2024, 63, e202319936.
134. Y. Gao, Q. Wu, X. Liang, Z. Wang, Z. Zheng, P. Wang, Y. Liu, Y. Dai, M.-H. Whangbo and B. Huang, Cu₂O Nanoparticles with Both (100) and (111) Facets for Enhancing the Selectivity and Activity of CO₂ Electroreduction to Ethylene, Adv. Sci., 2020, 7, 1902820.
135. J. Wang, C. Cheng, B. Huang, J. Cao, L. Li, Q. Shao, L. Zhang and X. Huang, Grain-Boundary-Engineered La₂CuO₄ Perovskite Nanobamboos for Efficient CO₂ Reduction Reaction, Nano Lett., 2021, 21, 980–987.
136. C. Tang, Z. Chen, Y. Wang, T. Xiao, X. Li, C. Zheng, X. Xu and Z. Sun, Atomic Editing Copper Twin Boundary for Precision CO₂ Reduction, ACS Catal., 2022, 12, 11838–11844.
137. 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, Increasing CO Binding Energy and Defects by Preserving Cu Oxidation State via O₂-Plasma-Assisted N Doping on CuO Enables High C₂₊ Selectivity and Long-Term Stability in Electrochemical CO₂ Reduction, ACS Catal., 2023, 13, 9222–9233.
138. Q. Wu, R. Du, P. Wang, G. I. N. Waterhouse, J. Li, Y. Qiu, K. Yan, Y. Zhao, W.-W. Zhao, H.-J. Tsai, M.-C. Chen, S.-F. Hung, X. Wang and G. Chen, Nanograin-Boundary-Abundant Cu₂O-Cu Nanocubes with High C₂₊ Selectivity and Good Stability during Electrochemical CO₂ Reduction at a Current Density of 500 mA/cm², ACS Nano, 2023, 17, 12884–12894.
139. L. Bian, Z.-Y. Zhang, H. Tian, N.-N. Tian, Z. Ma and Z.-L. Wang, Grain boundary-abundant copper nanoribbons on balanced gas-liquid diffusion electrodes for efficient CO₂ electroreduction to C₂H₄, Chin. J. Catal., 2023, 54, 199–211.
140. R. He, X. Luo, L. Li, Y. Zhang, L. Peng, N. Xu and J. Qiao, Grain boundary and interface interaction of metal (copper/indium) oxides to boost efficient electrocatalytic carbon dioxide reduction into syngas, J. Colloid Interface Sci., 2024, 658, 1016–1024.
141. Y. Kong, H. Yang, X. Jia, D. Wan, Y. Zhang, Q. Hu and C. He, Constructing Favorable Microenvironment on Copper Grain Boundaries for CO₂ Electro-conversion to Multicarbon Products, Nano Lett., 2024, 24, 9345–9352.
142. 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, Highly Selective Electrochemical Reduction of CO₂ into Methane on Nanotwinned Cu, J. Am. Chem. Soc., 2023, 145, 9136–9143.
143. Z. Chen, T. Wang, B. Liu, D. Cheng, C. Hu, G. Zhang, W. Zhu, H. Wang, Z.-J. Zhao and J. Gong, Grain-Boundary-Rich Copper for Efficient Solar-Driven Electrochemical CO₂ Reduction to Ethylene and Ethanol, J. Am. Chem. Soc., 2020, 142, 6878–6883.
144. Y. Zhang, K. Qi, P. Lyu, E. Petit, H. Wu, W. Wang, J. Ma, Y. Wang, C. Salameh and D. Voiry, Grain-Boundary Engineering Boosted Undercoordinated Active Sites for Scalable Conversion of CO₂ to Ethylene, ACS Nano, 2024, 18, 17483–17491.
145. C. Choi, T. Cheng, M. Flores Espinosa, H. Fei, X. Duan, W. A. Goddard Iii and Y. Huang, A Highly Active Star Decahedron Cu Nanocatalyst for Hydrocarbon Production at Low Overpotentials, Adv. Mater., 2019, 31, 1805405.
146. Y. Zhang, F. Chen, X. Hao, Y. Liu, W. Wu, X. Zhang, Z. Zang, H. Dong, W. Wang, F. Lu, Z. Lu, H. Liu, H. Liu, F. Luo and Y. Cheng, Enhanced interfacial effect-induced asymmetric coupling boost electroreduction of CO₂ to ethylene, Appl. Catal., B, 2024, 344, 123666.
147. T. Zhang, B. Yuan, W. Wang, J. He and X. Xiang, Tailoring *H Intermediate Coverage on the CuAl₂O₄/CuO Catalyst for Enhanced Electrocatalytic CO₂ Reduction to Ethanol, Angew. Chem., Int. Ed., 2023, 62, e202302096.
148. Z.-Y. Zhang, H. Tian, L. Bian, S.-Z. Liu, Y. Liu and Z.-L. Wang, Cu-Zn-based alloy/oxide interfaces for enhanced electroreduction of CO₂ to C₂₊ products, J. Energy Chem., 2023, 83, 90–97.
149. Q. Zhang, J. Wang, F. Guo, G. He, X. Yang, W. Li, J. Xu and Z. Yin, Nitrogen cold plasma treatment stabilizes Cu⁰/Cu⁺ electrocatalysts to enhance CO₂ to C₂ conversion, J. Energy Chem., 2023, 84, 321–328.
150. D. Xiang, K. Li, M. Li, R. Long, Y. Xiong, D. Yakhvarov and X. Kang, Theory-guided synthesis of heterostructured Cu@Cu_0.4W_0.6 catalyst towards superior electrochemical reduction of CO₂ to C₂ products, Mater. Today Phys., 2023, 33, 101045.
151. M. Zhong, K. Tran, Y. Min, C. Wang, Z. Wang, C.-T. Dinh, P. De Luna, Z. Yu, A. S. Rasouli, P. Brodersen, S. Sun, O. Voznyy, C.-S. Tan, M. Askerka, F. Che, M. Liu, A. Seifitokaldani, Y. Pang, S.-C. Lo, A. Ip, Z. Ulissi and E. H. Sargent, Accelerated discovery of CO₂ electrocatalysts using active machine learning, Nature, 2020, 581, 178–183.
152. Y. Guo, X. He, Y. Su, Y. Dai, M. Xie, S. Yang, J. Chen, K. Wang, D. Zhou and C. Wang, Machine-Learning-Guided Discovery and Optimization of Additives in Preparing Cu Catalysts for CO₂ Reduction, J. Am. Chem. Soc., 2021, 143, 5755–5762.
153. D. H. Mok, H. Li, G. Zhang, C. Lee, K. Jiang and S. Back, Data-driven discovery of electrocatalysts for CO₂ reduction using active motifs-based machine learning, Nat. Commun., 2023, 14, 7303.

---