Compared with the overlap between the reduction potential region and the hydrogen evolution reaction (HER) in the ECR process of noble metals, the overlap between the reduction potential region and the HER reaction of transition metals is less, so various transition metals are more suitable for the ECR process in kinetics^[23]. Fan et al. Formed a Ni single-atom catalyst (Ni-SAC) by redispersing a commercial centimeter-sized Ni foam at high temperature and adding melamine, as shown in fig. 1^[15]. It was found that the CO desorption on the active Ni unit site in the catalyst was weak, and the ^*COOH was easily produced. The as-synthesized Ni-SAC exhibits superior ECR catalytic performance, and the total current can be increased to 800 mA in an integrated gas-phase electrochemical zero-gap reactor, while maintaining the FE_CO of each unit cell above 90%, which has great potential for industrial application.

Li et al. Prepared a high-performance Ni-N-C single-atom catalyst by pyrolysis of ZIF-8 precursor^[28]. This unique two-step synthesis strategy can accurately control the carbon structure/morphology, metal content, and Ni-N bonding structure, and then upgrade the ECR performance. The performance test found that the synthesized catalyst exhibited excellent CO generation performance in an industrial flow cell, achieving a CO current density as high as 726 mA·cm^-2 with a FE greater than 90%, making it one of the best catalysts for the reduction of CO₂ to CO.

Fe is in the same group as Ni in the periodic table and has similar properties, so Fe can also be designed as a metal active center for ECR. Gu et al. Synthesized a catalyst (Fe³⁺-N-C) with dispersed single-atom iron sites by pyrolysis of Fe-doped ZIF-8 precursor^[29]. The performance test found that the catalyst produced CO at an overpotential as low as 80 mV, and the partial current density reached 94 mA·cm^-2 with a potential of 340 mV. Subsequent experiments confirmed that the active sites are discrete Fe³⁺, which coordinate with the pyrrole nitrogen (N) atoms of the N-doped carbon support to maintain the + 3 oxidation state during electrocatalysis, and that the CO adsorption rate of the Fe³⁺ site is faster and the CO absorption capacity is weaker than that of the conventional Fe²⁺ site, thus obtaining its superior activity.

Compared with the traditional NiN₄ site, the CoN₄ site has a lower energy barrier for the formation of ^*COOH, which makes the initial potential for the formation of CO lower and more conducive to the ECR process of electroreduction of CO₂ to CO^[30]. Yang et al. Synthesized three-dimensional netlike nanofibrous Co single-atom catalysts (CoSA/HCNFs) with a continuous porous structure by constructing free-standing, cross-linked, and high-yield carbon films^[17]. The performance test found that the CoSA/HCNFs had 92% FE as well as 211 mA·cm^-2 current density in the flow cell, respectively. The experiment showed that the continuous porous structure of three-dimensional network CoSA/HCNFs nanofibers can promote a large electrochemical active surface area, which is beneficial to the transport of reactants, thus producing abundant effective cobalt single atoms and improving the ECR catalytic activity of the catalyst.

Pan et al. Designed a Co single-atom catalyst (Co-N₅/HNPCSs) anchored on polymer-derived hollow N-doped porous carbon spheres through an N coordination strategy, as shown in Fig. 2^[31]. The performance test found that the Co-N₅/HNPCSs has excellent ECR performance, with FE_CO exceeding 90% over a wide potential range from − 0.57 to − 0.88 V, and exceeding 99% at − 0.73 and − 0.79 V. After electrolysis for 10 H, the CO current density and FE_CO remained almost unchanged, showing excellent catalytic activity and stability. Subsequent experiments found that the active site of Co-N₅/HNPCSs was CO-N-5, and this unique active site reduced the activation energy of the rate-limiting step, which was beneficial to the rapid formation of intermediate ^*COOH and the rapid desorption of Co, thus improving the ECR catalytic performance of Co-N₅/HNPCSs.

Unlike the reduction products of other active center metals, which are usually limited to CO, Cu can electroreduce CO₂ to common CO, and can also electroreduce CO₂ to hydrocarbons or oxygenates^[32]. Therefore, it is of great significance to design Cu as the active center metal of single-atom catalysts for ECR. Chen et al. Designed a monoatomic site Cu catalyst (Cu/p-Al₂O₃SAC) supported by ultrathin porous Al₂O₃ enriched in Lewis acid sites^[33]. The performance test found that Cu/p-Al₂O₃SAC effectively promoted the methanation of CO₂, and the corresponding current density of CH₄ formation was 153.0 mA·cm^-2 at − 1.2 V (vs RHE). Theoretical calculations show that Lewis acid sites in metal oxides, such as Al₂O₃, can adjust the electronic structure of the Cu atom by optimizing the intermediate absorption, thereby promoting CO₂ methanation.

The study of single-atom catalysts is helpful for researchers to deepen the study of ECR process mechanism and find more active sites suitable for ECR process^[34]. Compared with transition metals, main group metals (In, Sn, etc.) have high HER potential and relatively low ECR potential, so they have great potential In CO₂RR process^[33,35,36]. With the development of research, some single-atom catalysts with main group elements as independent active sites have been developed and applied to the ECR process.

The electroreduction of CO₂ to formic acid is considered to be cost-effective. In can electroreduce CO₂ to formic acid, so it is necessary to design In single-atom catalyst. Zhang et al. Prepared a catalyst with atomically dispersed indium (In) sites (In-N-C) by a pyrolysis strategy based on the In/Zn zeolite imidazole framework^[37]. Interestingly, due to the strong electronic interaction between atomically dispersed In sites and neighboring N atoms on the carbon skeleton and the low energy barrier for the formation of ^*OCHO intermediate on isolated In sites, the efficiency of CO₂RR to formic acid is effectively improved. The performance test found that the FE of CO₂ to formate reached 80%. The production rate of formic acid catalyzed by In-N-C is higher than that of other single-atom catalysts or nanocatalysts between -0.6 V and 1.1 V. In addition, the turnover frequency of In-N-C is up to 26 771 h^-1 at 0.99 V, which is an order of magnitude larger than that of other single-atom catalysts.

Sn-based catalysts are characterized by low toxicity and low cost, so the design and synthesis of single-atom catalysts with Sn as the active site is of great significance for the ECR process^[38⇓~40]. Ni et al. Constructed a single-atom Sn catalyst (denoted as FNC-SnOF) with a fluorine atom axially coordinated Sn-C₂O₂structure through an indirect trapping strategy, and this unique non-N coordination design changed the mechanism of CO₂RR on the Sn single-atom catalyst, so that the reduction product changed from HCOO^- to CO^[41]. The FE of FNC-SnOF for the reduction of V vs RHE to CO is higher than 90. 0% in a wide electrochemical window (-0. 2 V to-0. 6 V vs RHE), and the peak value can reach 95. 2%. The anode energy conversion efficiency and current density reached 70.7% and 186 mA·cm^-2, respectively.

The spherical 3s orbital of Mg has a non-directional delocalization character and thus a weak interaction with CO, and Mg can be designed as a single-atom catalyst for the ECR process. Wang et al. Synthesized a Mg single-atom catalyst for the CO₂RR process by embedding atomic-level dispersed Mg into graphitized C₃N₄(Mg-C₃N₄) through a simple heat treatment process^[42]. The performance test showed that the synthesized Mg SAC had good ECR performance, exhibiting a high turnover frequency of about 18 000 times per hour in H cell, a large current density of 300 mA·cm^-2 in flow cell, and a high FE_CO(≥90%) in KHCO₃ electrolyte.

To sum up, the active sites of single-atom catalysts have been widely studied, ranging from noble metals to transition metals, and then to main group metals, and have shown excellent ECR performance. However, industrial catalysis has put forward higher requirements for the activity, selectivity and stability of ECR single-atom catalyst at high current density.Therefore, it is necessary to regulate the ECR single-atom catalyst to enhance its catalytic activity, product selectivity and long-term stability, so as to adapt to more complex industrial environments, meet more diverse product needs and long-term operation conditions.

The electronic structure of the metal active site can be changed by changing the coordination structure of the metal active site, and then the ECR activity of the catalyst can be improved. Jia et al. Constructed a highly defective Ni-pyridine nitrogen single-atom catalyst (NiNV SAC) via a nitrogen vacancy (NV) -induced coordination reconstruction strategy^[43]. Compared with other Ni-N-C single-atom catalysts, the performance of Ni-NV SAC showed that the CO selectivity and production efficiency were significantly improved, with a high FE_CO of 96% at a low overpotential of 590 mV and a large CO current density of 33 mA·cm^-2 at 890 mV. The excellent performance of Ni-NV SAC originates from the fact that the local N environment of the defect reduces the constraint on the central Ni atom, which provides enough space for the adsorption and activation of CO₂ molecules, resulting in a lower energy barrier for CO₂ reduction.

In addition to changing the coordination number of the metal active site, the ECR catalytic activity of single-atom catalysts can also be manipulated by changing the atom coordinated to the metal active site. Li et al. Designed a two-step reaction to synthesize an axial chlorine (Cl) atom-coordinated FeNC single-atom catalyst (FeN₄Cl/NC) with excellent electroreduction CO₂ performance by pyrolysis of Fe-loaded two-dimensional zeolitic imidazole framework nanosheets and low temperature in hydrochloric acid solution^[44]. The performance test found that FeN₄Cl/NC has a FE_CO of 90.5%, a low overpotential of 490 mV, and a high turnover frequency of 1566 h^-1, which is one of the best iron-based CO₂RR catalysts reported in recent years. Theoretical calculations show that the introduction of an axial chlorine atom can adjust the electronic structure of the Fe atom in the catalytically active FeN₄ site, thereby promoting the desorption of ^*CO and inhibiting the adsorption of ^*H, thus improving the activity and selectivity of CO₂RR.

The introduction of other non-metallic heteroatoms or heteroatom groups such as P, S, — CN, etc., into the single-atom catalyst can also change the local environment and enhance the ECR performance of the catalyst. Sun et al. Designed a P-induced single-Fe-site single-atom Fe catalyst (Fe-SAC/NPC), which exhibited a FE_CO of about 97% at a low overpotential of 320 mV and a Tafel slope of only 59 mV·dec^-1, comparable to that of gold catalyst^[45]. Subsequent experiments found that P mainly introduced a single P atom into Fe-SAC/NPC in the form of P — C bonds, and the presence of a single P atom increased the electron density of the Fe center and significantly promoted the formation of ^*COOH, resulting in excellent CO₂RR performance at low overpotential. Subsequently, Li et al. Prepared a P-tuned Fe-N-C catalyst (Fe-N/P-C) by pyrolyzing a mixture of activated carbon black with Fe³⁺, urea, and triphenylphosphine in argon atmosphere^[46]. It was found that FE-N/P-C exhibited excellent electroreduction of CO₂ to CO, with a high Fe of 98% at a low overpotential of 0.34 V and a mass-normalized turnover frequency of up to 508.8 h^-1. Theoretical calculations show that the tuning of P in the iron single-atom catalyst reduces the oxidation state of the iron center and lowers the free energy barrier for the formation of the ^*CO intermediate, thus maintaining the electrocatalytic activity and stability of the iron single-atom catalyst.

Wang et al. Synthesized a Ni single-atom catalyst (Ni@C₃N₄-CN) incorporating a cyano group by a pyrolysis aerogel method, as shown in Fig. 3^[47]. The performance test found that Ni@C₃N₄-CN has excellent ECR performance with a TOF of about 22 000 h^-1,FE_CO≥90%. Even at low CO₂ concentrations, the FE_CO remained greater than 90%. In addition, the current density of the flow battery assembled with Ni@C₃N₄-CN reaches 300 mA·cm^-2,FE_CO≥90%, which has a good prospect for industrial application. Theoretical calculations show that the excellent performance of Ni@C₃N₄-CN is derived from the introduction of -CN, which weakens the d-π conjugation on the metal site and improves the electron density of CO₂ activation, which is beneficial to the formation of ^*COOH.

Diatomic pair is considered to be an effective means to enhance the ECR performance, because when two metal atoms form a diatomic pair in a single-atom catalyst, the two metals can cooperate to optimize the electronic structure and enhance the performance of the catalyst^[48]. Pan et al. Reported a two-atom pair catalyst (Cu-APC) of C {{\mathrm{u}}_1}^0-C {{\mathrm{u}}_1}^{x+} highly active atomic interface stabilized by Te surface defects on Pd₁₀Te₃ alloy nanowires, as shown in Fig. 4^[49]. The performance test shows that Cu-APC has good stability, and the current density hardly decreases within 180 min at-0. 78 V. Theoretical calculations show that the interaction between the C {{\mathrm{u}}_1}^0 -C {{\mathrm{u}}_1}^{x+} diatom pair in Cu-APC enhances the selectivity of the catalyst for CO₂RR, while inhibiting the HER reaction that competes with the ECR process, and therefore enhances the catalytic performance.

Yi et al. Synthesized a diatomic site catalyst consisting of a Co-Cu hetero-diatomic pair (CoCu-DASC)^[50]. Subsequent studies found that a diatomic pair was formed between Co and Cu of CoCu-DASC, and the cooperative catalysis of Co-Cu bimetallic sites reduced the activation energy and promoted the formation of the intermediate product ^*COOH, thus improving the catalytic activity. Compared with CO-SAC and Cu-SAC, CoCu-DASC showed excellent ECR performance, reaching a CO selectivity of 99.1% and a CO partial current density of 483 mA·cm^-2 in the electroreduction of CO₂, even exceeding the current density requirement at the industrial level, and showed excellent stability in the long-term durability test.

To sum up, the ECR activity of single-atom catalysts can be enhanced by coordination environment regulation, local environment regulation and diatomic pair regulation, and the current density of ECR single-atom catalysts can be improved, even reaching or exceeding the industrial current density. These regulatory strategies essentially change the electronic structure of the active site of the ECR single-atom catalyst, reduce the activation energy of the rate-limiting step, and facilitate the formation of intermediates, while enhancing the adsorption of intermediates and desorption of reduced products, thus enhancing the ECR performance of single-atom catalysts.

The metal monoatomic active site and the catalyst support are generally connected by covalent or ionic bonds, with metal-support interaction. Therefore, by optimizing the interaction between the metal single-atom active site and the catalyst support, the electronic structure of the active site of the ECR single-atom catalyst can be changed, thereby improving the current density of the ECR single-atom catalyst and enhancing the ECR performance of the single-atom catalyst^[51]. Zheng et al. Reported a Cu single-atom catalyst (Pd₁Cu SAC) supported on Pd^[52]. The performance test found that Pd₁Cu SAC has excellent ECR performance, showing an extremely large current density (1 A·cm^-2) as well as an extremely large FE (96%) in the ECR process of converting CO₂ to formate. At the same time, the Pd₁Cu SAC has excellent stability, and the pure formic acid solution can be continuously produced in the solid electrolyte reactor within 180 H at a current density of 100 mA·cm^-2. Theoretical calculations show that the unique metal-support interaction in Pd₁Cu SAC changes the electronic structure of the active site of the ECR process, optimizes the first protonation step of the ECR process, and converts the COOH^* path to the HCOO^* path, which is the reason why Pd₁Cu SAC exhibits a large current density as well as excellent ECR performance during ECR.

A large number of studies have shown that the production of renewable fuels by electrochemical CO₂ reduction is desirable, but the products of most CO₂RR processes are still limited to CO and formic acid (HCOOH), so enhancing the product selectivity of single-atom catalysts in ECR process and producing more diverse CO₂ reduction products are also the focus of research. Zhang et al. Designed a CuO/Ni single-atom tandem catalyst (CuO/Ni SA), as shown in Fig. 5^[53].

The in situ generation and rapid consumption of CO were achieved by the independent catalysis of CO₂-CO and CO-CO-C₂₊ in close proximity by Ni and Cu catalytic sites to reach high selectivity for the C₂₊ product. The performance test found that CuO/Ni SA achieved an extremely high partial current density of 1220.8 mA·cm^-2 in the C₂₊ product, while still maintaining excellent C₂₊ product (81.4% FE) and selectivity to ethylene (54.1% FE) and ethanol (28.8% FE).

Wei et al. Synthesized a single-atom catalyst with Fe-Cu diatomic center (Cu/Fe-NC) by the method of appending Cu, using Fe single-atom catalyst as a probe^[54]. The low limiting potential of Cu/Fe-NC for CH₃OH and CH₄ was found to be − 0.51 V, while the limiting potential of CO was − 1.18 V. Subsequent experiments showed that due to the construction of Fe-Cu diatomic centers, Fe-Cu synergy was generated, with Fe as the active site and Cu as the active promoter to improve the Fe activity and enhance the adsorption capacity, allowing CO₂RR to generate CH₃OH and CH₄ through a multi-electron pathway, thus promoting the generation of multi-electron products.

Yang et al. Proposed a simple strategy for large-scale synthesis of isolated Cu-decorated via carbon nanofibers to prepare catalysts with abundant Cu single-atom sites (Cu-SAs/TCNFs)^[55]. The performance test found that the Cu-SAs/TCNFs exhibited excellent selectivity and stability for the electroreduction of CO₂ to methanol (CH₃OH), and could produce high-purity methanol in the liquid phase with a FE of 44%, while having a partial current density of 93 mA·cm^-2 and an aqueous solution stability of more than 50 H. The free-standing and via structure of CuSAs/TCNF greatly reduces the intercalated metal atoms, resulting in an abundance of highly efficient Cu single atoms that can actually participate in the relatively high binding energy of the CO₂RR,Cu single atom to the ^*CO intermediate. Therefore, ^*CO can be further reduced to products such as methanol instead of being easily released from the catalyst surface as a CO product.

Cai et al. Firstly synthesized carbon dot-based SACs (Cu-CDs) with unique CuN₂O₂ sites by a low-temperature calcination process^[56]. It was found that Cu-CDs could convert CO₂ into methane with excellent performance, the FE efficiency reached 78%, and the selectivity of methane product was extremely high. The subsequent experiments showed that the support of the catalyst was partially carbonized by the low temperature calcination process, which led to the appropriate increase of the CH₄ energy barrier of Cu active sites and the change of electronic structure, reduced the overall endothermic energy of the key intermediate, and improved the selectivity of the catalyst for 8e^-ECR process.

To sum up, through some unique designs of single-atom catalysts, the selectivity of single-atom catalysts for ECR reduction products can be significantly enhanced, and more diverse ECR reduction products can be produced. The essence of these control methods is to change the ECR process on the active site of the single-atom catalyst, so that the undesorbed CO can continue to couple without releasing on the surface of the catalyst, so that the ^*CO can be further reduced to other multi-electron products.

Although the single-atom catalyst has shown good laboratory test performance in the ECR process, there is still a long way to go to realize its industrialization in the ECR process. This is because the current density in industrial production needs to reach the industrial current density (greater than 440 mA·cm^-2) or even higher, and at such a large current density.The ECR catalytic performance of single-atom catalysts may be significantly reduced for two reasons. The first is that many single-atom catalysts are constrained by their own conditions such as metal active sites and support properties, and can not achieve a larger current density. The second is that at high current density, various side reactions will occur on the surface of the catalyst, which is not conducive to the ECR process, thus leading to the decline of the ECR performance of single-atom catalysts^[61]. Unfortunately, many of the reported ECR single-atom catalysts cannot achieve such high current densities. In order to improve the current density of ECR monatomic catalyst and realize the industrial application of ECR monatomic catalyst, these practical problems need to be considered in the laboratory design and preparation of catalyst.

Although single-atom catalysts can catalyze a variety of ECR processes with different electron numbers, there are still many challenges to achieve excellent ECR process selectivity at high current densities. Therefore, it is of great significance to study the selectivity of single-atom catalysts for ECR process to improve the performance of ECR and obtain more diverse reduction products. The selectivity of single-atom catalysts for ECR process is mainly reflected in two aspects, one is the selectivity of reduction products, CO₂RR the reaction processes with different electron numbers reduce CO₂ to different products, however, most single-atom catalysts are limited to catalytic 2e^- process at high current density, the reduction products are mostly CO and HCOOH, and the reduction products of other multi-electron processes are rarely reported. On the other hand, people also need to pay attention to the selectivity of single-atom catalysts for ECR and HER reactions at high current density, which will greatly affect the ECR performance of single-atom catalysts at high current density. Therefore, it is of great significance to find and synthesize single-atom catalysts with lower ECR energy barrier and multi-electron ECR process at high current density for the large-scale application of single-atom ECR catalysts.

In-depth study of the deactivation mechanism of single-atom catalysts in the ECR process at high current density can provide ideas for the design of ECR single-atom catalysts and help researchers develop higher performance and more stable ECR single-atom catalysts. It is found that when the size of active metal particles is reduced to the atomic level, the surface energy of active metal particles will increase sharply, and the single atom will have an obvious aggregation trend. At the same time, under high current density, the polarization phenomenon of the electrode is more serious, and the ECR process will be more rapid, which will affect the adsorption and mass transfer process on the monatomic catalyst, and seriously affect the stability of the monatomic catalyst^[62⇓~64]. This is a great challenge for the industrial application of ECR single-atom catalysts. The stability of ECR single-atom catalyst is mainly reflected in two aspects, one is the stability during storage, which can reduce the cost of catalyst transportation. On the other hand, it is the long-period application stability of ECR single-atom catalyst at high current density, which is essential for the industrialization of ECR single-atom catalyst. Although the stability of single-atom ECR catalysts has been studied, the stability of single-atom ECR catalysts has not been fully studied due to the complexity of the reaction mechanism of ECR process, which greatly limits the large-scale application of single-atom ICR catalysts. Therefore, it is of great significance and challenge to study the stability of ECR monatomic catalysts and design ECR monatomic catalysts with high stability.

The use of reactors for the ECR process has become the most promising direction to achieve large-scale industrialization of the electroreduction of CO₂. The traditional ECR reactor is H-type cell or H-type reactor, which has the characteristics of simple operation and easy cleaning, and some successful configuration work has been carried out^[65]. However, this type of reactor has a low current density, usually only up to 45 mA·cm^-2, and it is difficult to achieve a commercial current density (about 200 mA·cm^-2)^[66,67]. At present, the common ECR reactor is carbon dioxide electrochemical flow cell, including carbon dioxide liquid phase electrolyzer, carbon dioxide gas phase electrolyzer and carbon dioxide solid phase electrolyzer^[68]. One of the most commonly used is the carbon dioxide liquid phase electrolyzer. The feature of this reactor is that the flowing catholyte can be designed. Although it has been reported that carbon dioxide liquid phase electrolyzer can realize ECR process at high current density, it requires system pressurization and high alkaline environment, which is dangerous^[69]. Although carbon dioxide solid phase electrolyzer can also be used in ECR process at high current density, it inevitably has some disadvantages, such as high temperature, carbon deposition, metal particle oxidation and single electroreduction product of CO₂, which need more in-depth study before large-scale application^[70,71]. The reported carbon dioxide gas phase ECR reactor has the advantages of higher product selectivity and stability under high current conditions, and can selectively produce alcohols or other C₂₊ products under low current density conditions, so it is a very potential ECR reactor, but it needs excellent catalysts to support^[72,73]. To sum up, most of the ECR reactors currently in use are facing the problem of poor ECR performance under high current conditions, which requires continuous improvement and the search for reactor catalysts with higher performance. Although single-atom catalysts have excellent ECR performance, how to realize the adaptability of ECR single-atom catalysts in ECR reactor under high current conditions is also a big challenge. Therefore, it is of great significance to design ECR single-atom catalysts with higher performance to meet the needs of ECR reactors under high current conditions, which is of great significance to realize the large-scale industrialization of CO₂ electroreduction.