Research Progress on the Application of Carbon-Based Materials in Electrocatalytic CO₂ Reduction Reaction

Abstract

The conversion of CO₂ into high-value-added chemicals and fuels using electricity generated from renewable energy sources is one of the most promising methods to reduce the dependence of human society on fossil fuels and to alleviate environmental problems. The performance of catalysts is one of the most important factors restricting the development of this technology, and in recent years, carbon materials have been the hot spot of research in the field of CO₂ electrocatalytic reduction catalysts. In this paper, the progress of the application of carbon materials in CO₂ electrocatalytic reduction reaction (ECR) is reviewed in detail. Three aspects of carbon materials directly as metal-free carbon material catalysts for CO₂ reduction, metal-centered coatings in metal catalysts, and support for metals, are comprehensively described, respectively, including the preparation strategy of catalysts, the mechanism of action and structural characteristics of catalysts, the distribution of products and the catalytic performance of catalysts. Finally, the problems and challenges faced by the field are summarized, and the outlook is presented in various areas, including catalyst preparation, performance enhancement, and deepening mechanism research.

1. Introduction

The widespread consumption of fossil fuels has led to a steady increase in atmospheric carbon dioxide concentrations, causing a range of serious ecological and environmental challenges [1,2,3]. An emerging and highly promising strategy to mitigate humanity’s reliance on fossil fuels and address the ecological challenges posed by excessive CO₂ emissions involves leveraging electricity where CO₂ is catalytically reduced into premium-grade chemical feedstocks and renewable fuels powered by sustainable electricity [4,5]. The reduction products of CO₂ are mainly C₁ products (e.g., CO, HCOOH, CH₄) and C₂₊ products (e.g., C₂H₄, CH₃COOH), and different reaction pathways control their formation. The proposed reaction mechanism consists of three key stages: CO₂ adsorption/activation, generation of intermediates via proton-coupled electron transfer (PCET), with the reaction trajectory splitting based on *COOH and *CO surface retention strength and final product expulsion dynamics (Figure 1). Initially, CO₂ molecules are absorbed and activated on the catalyst surface. Subsequent PCET reactions generate key intermediates such as *COOH and *OCHO, which further evolve into *CO (essential for downstream products) and HCOOH. The fate of *CO (the core intermediate of the C₁ and C₂₊ products) is largely determined by its binding energy to the catalyst. When *CO is weakly adsorbed, it readily desorbs from the surface to form gaseous CO. Conversely, strong adsorption favors further PCET-driven transformations: (1) hydrogenation to produce the *COH intermediate, which in turn produces the C₁ product (e.g., CH₄), or (2) coupling of the *CO and *COH species, which initiates the formation of the C-C bond and ultimately the C₂₊ product. This adsorption ability-dependent selectivity emphasizes the importance of catalyst design in directing the reaction pathway [6].

The catalysts used in the electrocatalytic reduction of CO₂ include metal-free catalysts and metal catalysts, and the metal catalysts include metal/metal compounds and metal/carbon composite catalysts. Nevertheless, the high thermodynamic stability of CO₂ molecules, the sluggish reaction kinetics of CO₂ electrochemical reduction reactions (ECR), and the complexity of reaction products present significant hurdles. Consequently, developing catalysts with superior activity, selectivity, and durability is crucial for advancing this technology [7,8]. Carbon-based materials have emerged as a focal point in ECR research. These materials exhibit three key merits: economic viability, global abundance, and outstanding charge transport characteristics, good chemical stability, i.e., chemical properties are not easily altered, and a well-developed porous architecture, making them increasingly attractive for these applications [9]. The catalysts currently applied to CO₂ electrocatalytic reduction mainly contain carbon-based nonmetallic and metallic catalysts. Carbon-based non-metallic catalysts are carbon materials directly used as catalysts for ECR through activation or doping with heteroatoms. Metallic catalysts in which carbon or heteroatom-doped carbon materials play a great role as metal-based or coatings in the ECR. At present, carbon materials in CO₂ reduction mainly contain two aspects, one is as a carbon-based metal-free catalyst to catalyze the reduction reaction of CO₂ directly. The alternative configuration employs metallic substrates or coatings, synergistically integrating the merits of both metallic and carbonaceous materials to enhance catalytic efficacy. Fu et al. catalysts prepared using nitrogen-doped carbon materials at high temperatures (800 °C) (ANBC800) showed the best performance: 89.3% FE_CO (−0.82 V vs. RHE) and a partial current density of 1.59 mA∙cm⁻² for CO [10].

Compared with pure metal, when carbon material is used as a base or coating for metal, if the base is rich in pore structure, it can also enhance CO₂ adsorption capacity and promote mass transfer. Architecturally, the N-doped carbon coating surrounding the metal nanoparticles provides essential protection against electrochemical corrosion while maintaining the reactivity of the vulnerable metal cores. For example, Sun used a low-coordination nickel single-atom catalyst (L-Ni-NC-C) whose unique Ni–N₁ C₂ coordination structure was prepared by nitrogen elimination. The catalyst showed excellent performance in ECR and efficiently converted CO₂ to CO [11]. Dutta prepared mesoporous copper foam catalysts by electrodeposition, which showed good selectivity for C₂ (C₂H₄ and C₂H₆) with a Faraday efficiency (FE_C2) of 55% [12]. Wang et al. prepared core (hierarchically structured Sn-Cu alloy/Sn) and shell (amorphous SnO_x) structured Sn_2.7Cu nanomaterials by electrodeposition, and used a reversible hydrogen electrode as a reference electrode to achieve a current density of up to 406.7 ± 14.4 mA∙cm⁻² and a Faraday efficiency of up to 98.0 ± 0.9% when applying a voltage of −0.70 V. When an applied voltage of −0.55 V, the current density and Faraday efficiency are 243.1 ± 19.2 mA∙cm⁻² and 99.0 ± 0.5%, respectively, after the reduction reaction has been carried out for 40 h [13]. Zhang synthesized a nano-homogeneous junction catalyst (h-Cu₂O) by wet chemistry. h-Cu₂O achieved a Faraday efficiency (FE) of 73.7% for ethylene in an H-type electrolytic cell at −1.4 (V vs. RHE), with a partial current density of 38.2 mA∙cm⁻² [14].

Unlike carbonaceous materials, metallic catalysts demonstrate enhanced electrocatalytic activity owing to their unique d-orbital electron configuration. However, metal nanoparticles are more favorable for the hydrogen evolution reaction (HER) in aqueous solution and are easily poisoned by adsorbed CO, and on the other hand, metals are easily oxidized when exposed to atmospheric conditions. For metal oxides as catalysts in ECR, some oxidized material will inevitably be reduced to the metal state during the electrocatalytic reaction, and therefore, if the characterization of the material is not in situ, this may lead to bias. The active substances used for CO₂ reduction are also controversial when it comes to metal oxides as electrocatalysts. The limited operational stability arises from concurrent metal leaching through corrosive dissolution and activity loss via particle coalescence, as well as a significant reduction in the stability of the catalyst, all of which hinders practical application.

Given the important application of carbon materials in CO₂ electrocatalytic conversion, this review systematically examines metal-free carbonaceous catalysts and hybrid metal-carbon catalytic systems, establishing fundamental principles for engineering advanced carbon-based catalytic materials. This paper firstly describes the non-metallic carbon materials used as catalysts, secondly describes the carbon materials used as metal coating, and finally describes the carbon materials used as metal supports. To ensure the comprehensiveness and representativeness of the selected studies, a comprehensive literature search, including multiple databases, was conducted for this review. The databases searched included Web of Science, Scopus, PubMed, and Google Scholar. Search terms were selected based on the thematic core of the review articles. Search terms included “carbon-based materials”, “electrocatalytic CO₂ reduction”, “catalyst preparation”, “catalytic properties”, and “reaction properties”.

2. Non-Metallic Carbon Materials as Catalysts for Catalyzing the CO₂ Reduction Reaction

2.1. Carbon Material Catalyst

The inherent structural features of carbon matrices—including engineered defects, tunable surface functionalities, and modifiable electronic configurations—enable precise tailoring of active centers for optimized catalytic behavior in carbonaceous catalysts. Metal-free carbonaceous catalysts have demonstrated remarkable efficacy in ECR. Xu et al. [15] reported a facile synthetic protocol for fabricating a series of monolayer graphene nanosheets (GNDs) densely functionalized with oxygen-containing moieties. These GNDs were synthesized via hydrothermal treatment at varying temperatures, utilizing graphene oxide and nitric acid as the primary precursors. Among the synthesized samples, the GNDs demonstrated optimal electrocatalytic performance, evidenced by the highest recorded current density and Faraday efficiency (FE) (Figure 2). Density functional theory (DFT) calculations were employed to systematically elucidate the impact of oxygen-containing functional groups, including carboxyl (-COOH), hydroxyl (-OH), and epoxide (-C-O-C) moieties, on the ECR activity of defective (d⁻) graphene structures. The DFT analysis considered both edge (e⁻) and central (c⁻) positions of surface-attached functional groups, with the active sites induced by these oxygen-containing groups distinctly marked by red circles, with the red dots representing oxygen atoms. The results indicated that the defective graphene structure significantly enhanced the ECR activity mediated by oxygen-containing functional groups. Comparative analysis further revealed that the carboxyl group (-COOH) on the graphene surface, in synergy with neighboring oxygen-containing functional groups (e.g., hydroxyl (-OH), carbonyl (-C=O), and epoxy (-C-O-C) groups) on the GNDs, substantially decreased the activation energy barrier for the conversion of CO₂ to *COOH, thereby facilitating the electrocatalytic process. Zhang et al. [16] pyrolyzed MOF-5 to obtain defect-containing carbon materials (D-C-X). Their experimental results revealed a direct proportionality between carbon defect density and ECR activity, with spectroscopic characterization confirming sp²-hybridized topological defects (pentagons/octagons)—rather than peripheral defects —as the dominant active centers.

2.2. N-Doped Carbon Material Catalysts

Heteroatom doping and morphology/structure design are two effective receptive approaches to strengthen the electrocatalytic response of metal-free carbon-based materials. Heteroatom doping with N, F, P, and S changes the electronic properties of carbon atoms. It creates defects and forms active catalytic sites. Different shapes, such as nanowires, nanosheets, and 3D porous structures, can enhance active sites, which facilitate mass transfer and improve the local environment. Consequently, ECR performance improves. Partially doped carbon materials enhance mass transfer. ECR is promoted by domain-limited effects that they also have, along with their rich pore structure and high surface area, which makes this possible.

Numerous research works have indicated that the incorporation of N atoms into the carbon skeleton generates defects. These defects elevate the charge density around the C atoms and enhance the performance of the ECR, and that pyrrole N, pyridine N, carbon adjacent to N species, and defects may be active sites. N-doped carbon materials are prepared directly from carbon materials (graphene, carbon nanotubes, activated carbon, etc.), and N dopants, or N-doped carbon materials are constructed by using coal, biomass, etc., as a carbon source and N dopant as an N source.

In N-doped carbon materials featuring a reduced cylindrical mesoporous pore structure, active pyridine and pyrrole N sites are evenly dispersed on the inner surface of the cylindrical channel [17]. The uniform dispersion of active pyridine and pyrrole N sites on the inner surface of the cylindrical channel in N-doped carbon materials promotes the adsorption and activation of CO₂. As a result, *CO intermediates are generated, which efficiently provide electrons for the dimerization reaction of CO intermediates. Moreover, the channel surfaces, characterized by regular arrangement and high electron densities, play a role in stabilizing these intermediates and *CO to a certain extent. Subsequently, nitrogen heteroatoms, in conjunction with the synergistic effect of the cylindrical channel configuration, contribute to the dimerization and proton-electron transfer of the key intermediate *CO, thereby facilitating the generation of ethanol (

Table 1. Catalytic performance of catalysts made of N-doped carbon materials.

Catalysts	Primary Product	Faraday Efficiency (FE) (%)	Stability (h)	Optimal Potential (V)	Current Density at the Optimal Potential (mA∙cm−2)	Electrolyzer	Ref.
c-NC	C2H5OH	77.00%	24	−0.56	\	H-type cell	[17]
NG-T	CO	95.00%	\	−0.72	9.07	flow cell	[18]
NGQDs	C2+	90.00%	\	−0.86	23.00	flow cell	[19]
N-doped 3D graphene foams	CO	85.00%	5	−0.58	−1.80	H-type cell	[20]
NPC	CO	95.00%	10	–0.67	−4.80	H-type cell	[21]
NCNTs	CO	80.00%	10	−0.26	−2.25	flow cell	[22]
PEI-NCNTs/GC	HCOO−	87.00%	20	−1.80	9.50	H-type cell	[23]
g-C3N4/MWCNTs	CO	60.00%	50	−0.95	2.50	H-type cell	[24]
NRMC	CO	80.00%	\	−0.60	−2.90	H-type cell	[25]
NC800	CO	97.80%	12	−0.86	6.70	H-type cell	[26]
ANBC800	CO	89.30%	80	−0.82	−1.59	flow cell	[10]
MSC-HA	C2H5OH	90.14%	30	−0.71	24.23	H-type cell	[27]
N-doped carbon foams	CO	95.00%	\	−0.50	\	H-type cell	[28]

). In N-doped graphene-like carbon materials (Figure 3b) [17], the actual active center is the C atom that is adjacent to the graphitic nitrogen species, not the nitrogen species (Figure 3a) [18]. The higher N-doping concentration, larger porosity, and ultrathin graphene-like planar structure in the catalyst promote charge transfer, which improves the catalytic performance (Table 1).

Nano-sized N-doped graphene quantum dots (NGQDs) have been demonstrated as catalysts for the reduction of carbon dioxide to polycarbonate-containing hydrocarbons and oxygenated compounds, and N-doping defects on the edge sites and unique nanostructures endow NGQDs with high activity and selectivity (Figure 3c) [19]. Three-dimensional (3D) graphene foam having defective nitrogen was used as a catalyst to bring down the amount of carbon dioxide, and the pyridine N defects were the sites with the highest CO₂ reduction activity (Figure 3d) [20], which reduced the free energy of the generated intermediate *COOH and favored the CO generation, and catalyst’s 3D multilevel pore structure enabled the electrolyte to penetrate smoothly, which increased the interface area created by the graphene foam and electrolyte, resulting in excellent catalytic performance. Coal-based N-doped porous carbon (NPC) catalysts encapsulate N atoms in micropores, and due to the spatial domain-limiting effect, they can efficiently convert CO₂ into CO products (Figure 3e) [21]. Polyacrylonitrile-based N-doped carbon nanofibers (CNFs) have an active center of positivity in the form of a positively charged carbon atom during the conversion of CO₂ to CO (Figure 4a) [29], and their excellent catalytic performance is also associated with their nanofiber structure and the strong binding capacity of the key intermediates. The pyrrole N defect in the N-doped carbon nanotube arrays (NCNT) obtained based on the liquid-phase CVD method is the most active site, and due to the high conductivity, better catalytic site (pyridine N defect), low freestanding energy of CO₂ activation, and significant hydrogen precipitation barriers of the NCNT (Figure 4b) [22], it exhibits promising catalytic properties in the electrocatalytic transformation of CO₂ to CO.

In NCNTs modified by polyethyleneimine (PEI-NCNTs/GC) composites, PEI stabilizes the intermediate CO₂ by hydrogen bonding, and thus can reduce CO₂ to yield formate at a minimal onset potential (Figure 4c) [23]. The g-C₃N₄/MWCNTs composite catalyst facilitates the reduction of CO₂ to CO. The active site is the covalent linkage of carbon and nitrogen established between g-C₃N₄ and MWCNTs [24]. In addition, its catalytic performance is associated with its significantly larger BET surface area, as well as its favorable electrical conductivity. Removal of some of the nitrogen species in the N-doped carbon materials by high-temperature heating to obtain carbon materials (NRMC), which resulted in the creation of edge defect sites in the carbon materials, decreased the energy barrier of the ECR, inhibited the hydrogen evolution reaction (HER), and helped modulate the selectivity of N-doped mesoporous carbon catalysts in generating CO [25].

Qin et al. [26] prepared carbon black, which was then utilized as a carbon source. In addition, formamide served as a nitrogen source. This was done by the hydrothermal reaction-calcination method. N-doped carbon material (NC800), through a systematic study, it was evident that nitrogen doping was an essential factor in improving the selectivity of the catalyst. A positive correlation was identified between the N/(N+O) ratio, Faraday efficiency, and local current density of the catalyst. We discovered that the CO selectivity and the CO local current density both decreased rapidly, owing to the limited nitrogen presence, but the excess oxygen content inhibits their electrocatalytic activity; the presence of oxygen functional groups on carbon materials inhibits their electrochemical properties and reduces the activity of ECR (Figure 4d) [26].

2.3. N Co-Doped Catalysts with Non-Metallic Elements Such as S, P, and F

Synergistically co-doping nitrogen and nonmetallic elements such as sulfur, phosphorus, and fluorine can greatly increase the activity and selectivity of catalysts. This increase is due to the electronic coupling effects among these different elements. The implementation of this doping method did not merely optimize the electronic framework of the catalysts and elevate the concentration of active sites; it also strengthened the ability of reactant molecules to be adsorbed and activated. Among them, N doping modulates the electron cloud distribution of carbon-based materials, the introduction of S and P promotes charge redistribution and the formation of new active centers. At the same time, the highly electronegative F element enhances the redox performance of the catalysts (

Table 2. Catalytic properties of N, S, F, P co-doped carbon materials catalysts.

Catalysts	Primary Product	Faraday Efficiency (FE) (%)	Stability (h)	Optimal Potential (V)	Current Density at the Optimal Potential (mA∙cm−2)	Electrolyzer	Ref.
NSHPC	CO	87.8%	8	0.45	5.49	H-type cell	[30]
NSHCF]	CO	94.0%	36	−0.70	−103	H-type cell	[32]
NSCNW	CO	93.4%	20	−0.49	−5.93	H-type cell	[33]
NF-C	CO	90.0%	10	−0.49	1.9	H-type cell	[31]
NPC	CO	70.4%	10	−0.80	0.36	H-type cell	[28]
NPC	CO	86.0%	12	−0.45	0.81	H-type cell	[34]
NPCA	CO	91.4%	35	−1.80	−22.86	H-type cell	[35]
NPCA	CO	99.1%	24	−2.40	−143.6	flow cell.	[36]
NPF-CNTs	CO	\	\	1.16	0.10	H-type cell	[37]

). Takada et al. found that N, P-co-doped carbon-based catalysts with N provided more favorable CO₂ adsorption sites, while P atoms near the N atoms stabilized the *COOH key intermediate, which is a key intermediate for the generation of CO, thus facilitating the conversion of ECR to CO [28]. According to Li et al., the joint doping of pyridinic N and carbon-bonded S atoms effectively lowered the Gibbs free energy of *COOH, which is an essential intermediate during CO generation. As a result, the conversion of CO₂ to CO was facilitated [30]. Pan et al. found that pyridine N was the active site in the N and F co-doped porous carbon layer (NF-C), and the presence of F changed the electronic properties of N, such as charge density and density of states, to further activate the pyridine N active site. At the same time, the weak CO adsorption leads to the easy desorption of CO from the catalyst surface, so that high CO selectivity can be obtained [31].

Guo et al. [30] prepared N and S co-doped multistage pore structure carbon materials (NSHPC) by pyrolysis of glucosamine hydrochloride and trimeric thiocyanate precursors based on a SiO₂ hard template; the amount of active sites can be increased by introducing N and S species (pyridine N, graphite N, and thiophene). Co-doping synergistically activates π-electrons in sp² carbon, while the multistage pore structure greatly increases the exposure of active sites for reactive species, facilitating electrolyte transport. Consequently, N, S co-doping effectively improves catalytic activity and selectivity (Figure 5a) [30]. He et al. [32] prepared N, S co-doped multistage porous structure carbon nanofibers (NSHCF) by electrostatic spinning and calcination, and based on comparative experiments and DFT calculations, they inferred that the main active site contains pyridine N, and the presence of sulfur in the C-S configuration exerts a positive influence on boosting the catalytic performance, and N and S co-doping effectively reduces the Gibbs free energy of the key intermediate *COOH. Additionally, the ECR has significantly more accessible active sites due to the hierarchical porous structure of NSHCF (Figure 5b) [32].

Kim et al. [33] prepared N, S co-doped carbon nanonetwork catalysts, in which the presence of abundant thiophene S in the catalysts altered the electronic properties of the pyridinium-nitrogen sites in the N-doped carbon, resulting in enhanced CO₂ adsorption and activation, and the 3D nano-web junctions having a high specific surface area, which makes the active sites highly exposed and facilitates the rapid transport of electrons and related intermediates, leading to an efficient ECR (Figure 5c) [33].

Wang et al. [31] reported a N, F co-doped porous carbon layer (NF-C) with high density and high activity of pyridine N sites for CO₂ reduction, which was synthesized by using NH₄F as a soft template, dicyandiamide was heated to generate layered g-C₃N₄, and glucose was immobilized as a carbon source in between the g-C₃N₄ layers to generate carbon nanosheets. In the catalyst, pyridine N is the active site, and the presence of F changes the electronic properties, such as charge density and density of states of N, further activating the pyridine N active site to become a highly active and selective reduction site. NF-C’s extensive surface area means that the N and F species are highly exposed, allowing CO₂ to come into contact with the active site. The porous carbon structure provides channels for CO₂ and CO transport and reduces the mass transfer resistance. The abundance of N atoms in NF-C, primarily in the form of pyridine N, allows for the formation of numerous active sites (Figure 5d) [31].

Due to the different sizes and electronegativity of the elements nitrogen and phosphorus, co-doping of N and P triggers the repositioning of adjacent carbon atoms and alters their charges, thereby enhancing the catalytic activity of carbon materials. Takada et al. [28] described a one-step synthesis of N, P-co-doped carbon materials (NPCs) with biobased glycine and phytanic acid, whereby the ratio of the syngas generated by CO₂ reduction can be controlled by adjusting the contents of N and P. A single-step synthesis was employed to prepare N- and P-doped carbons. Through experiments and DFT calculations, it was revealed that the negatively charged C atoms neighboring N and P atoms exhibited the highest activity for CO₂ conversion, and the introduction of N atoms produced more favorable CO₂ adsorption sites, while the P atoms helped to reduce the Gibbs free energy of adsorption of the key intermediate *COOH on the negatively charged C atoms (Figure 5e) [28]. Xie et al. [34] used anhydrous glucose, urea, and phytic acid as raw materials, synthesized N and P co-doped carbon materials, and demonstrated that phosphorus-nitrogen co-doping activated the neighboring carbon atoms in the graphene structure to become the active centers of catalysis, synergistically promotes the uptake of CO₂, the passing of the primary electrons to CO₂ and the stabilization of the intermediate *COOH. This enhances the activity and selectivity.

Carbon aerogels are very important non-homogeneous catalyst materials based on their low density, relatively large specific surface area, and high porosity of three-dimensional interconnections. Wang et al. [35] prepared N and P co-doped carbon aerogel (NPCA) catalysts using phytic acid and urea as precursors, with pyridine N as the main active center, N activating CO₂ and combining with protons to generate intermediate *COOH. When P is present, the aerogel forms a porous structure resembling a honeycomb, enhancing the extent of carbon imperfections and facilitating faster proton migration (Figure 5f) [35]. Sun et al. [36] prepared N and P co-doped carbon aerogel (NPCA) catalysts using g-C₃N₄ and phytic acid as precursors. Pyridine N is the active center of the ECR and promotes the performance of the ECR, and P can impede the adsorption of *H. Together, N and P promote the improvement of the reaction performance (Figure 5g) [36].

Although scholars at home and abroad have conducted extensive research on the application of carbon-based, metal-free catalysts in ECR, such catalysts generally have the problems of low catalytic activity, the reduction product is CO, the study of the active centers in the catalysts is often performed with the help of theoretical calculations, the doping of heteroatoms requires the use of costly reagents, the position of the dopant cannot be accurately controlled by any doping technique, and the current lack of high-precision techniques for synchronous detection and control of the position, size, structure and/or chemical properties of dopant catalytic active centers (CACs) at the atomic level is a key issue. The problems of simultaneous detection and control of dopant position/size/structure and/or chemical properties of CACs at the atomic level with high precision are the drawbacks that limit the large-scale application of carbon-based metal-free catalysts.

3. Carbon as a Metal Coating

It has been found that the presence of a carbon coating can form rich interfaces and produce point-in-action with metal centers, which not only increases the quantity of available active sites, accelerates mass and electronic transmission, optimizes intermediate adsorption and protects metal centers from corrosion by the electrolyte, but also maintains a higher specific capacity by reducing the aggregation and agglomeration of metal particles throughout the electrocatalytic process [38], which can, in turn, optimize catalyst performance (

Table 3. Catalytic CO₂ reduction properties of carbon materials as coatings.

Catalysts	Primary Product	Faraday Efficiency (FE) (%)	Stability (h)	Optimal Potential (V)	Current Density at the Optimal Potential (mA∙cm−2)	Electrolyzer	Ref.
Cu-NxC	C2H4, CH3CH2OH	80.00	\	−1.1	\	\	[39]
Ni@NC	CO	98.00	100	−0.87	220.0	H-type cell	[40]
CuxO/CN-10	C2H4	42.20	10	−1.20	25.0	H-type cell	[41]
In2O3@C	HCOOH	97.00	10	−1.27	144.2	flow cell	[42]
In2O3@NC	HCOOH	97.10	60	−1.70	190.0	flow cell	[43]
CuxO@C	CH3CH2OH	46.00	50	−1.00	−166.0	flow cell	[38]
FeNPs-C	CO	90.00	\	\	\	flow cell	[44]
Ni@N-C/rGO	CO	88.00	10	−0.97	20.0	H-type cell	[45]
Ni@NCNT-C	CO	100.00	40	−0.27	230.0	H-type cell	[46]
Ni-NC-ATPA@C	CO	93.70	24	−1.10	22.7	H-type cell	[47]
Ni@NC@NCNT	CO	94.10	43	−1.10	48.0	H-type cell	[48]
NiSA/NP	CO	98.00	10	−2.30	310.0	flow cell	[49]
Ni-N-C/CNT	CO	97.00	22.5	−1.22	−60.0	H-type cell	[50]
Ni@NCNT/Gr	CO	90.00	24	−0.71	−37.4	H-type cell	[51]
Ni@NCNT-700	CH3CH2OH	38.50	22	−0.50	128.0	flow cell	[52]
NiNCNT	CO	99.30	40	−0.35	−300	flow cell	[53]
Ni@NCNT/Gr	CO	>90	20	−0.50	−9.20	H-type cell	[51]

). Carbon material-coated metal catalysts can be prepared by high-temperature pyrolysis of metal organic framework (MOF) materials, high-temperature pyrolysis of metals with N and C sources, and chemical vapor deposition (CVD).

3.1. Pyrolysis of MOF Materials

Pyrolysis of MOF materials is one of the methods to prepare metal catalysts coated with carbon materials, M-BTC (H₃BTC homophthalic acid), M-MI (2-methylimidazole) (M=Ni, Co) pyrolysis, to obtain carbon layer, N-doped carbon coated transition metal nanoparticles (M@C/M@NC) catalysts, and by comparative experiments and DFT calculations, it was confirmed that the pyrrole N in Ni@NC served as an active site for adsorption contained adsorption-active regions for CO₂ interaction, and the metal NPs induced electronic modifications in nitrogen species, facilitating more efficient engagement with reaction intermediates (COOH/CO) and significantly improving ECR activity (Figure 6a) [40].

Low-temperature pyrolysis of Cu-BTC yielded a composite catalyst (CuO_x@C) of CuO_x covered by carbon materials, which could effectively prevent the destabilization of Cu⁺ species during CO₂ electroreduction, thus promoting C-C coupled reactions. Furthermore, the carbon coating could modulate the hydrogenation of the critical intermediate *HOCCH, thereby promoting ethanol synthesis (Figure 6b) [54]. MIL-68- In₂O₃@NC material was obtained by pyrolysis of NHNH₂ precursor, and C in the coating successfully modulated the electronic structure of In₂O₃ to optimize the adsorption of carbon dioxide intermediates, and the nitrogen had a strong adsorption capacity for carbon dioxide, which improved its catalytic performance for the conversion of CO₂ to CO (Figure 6c) [43]. Porous N-doped carbon layers (FeNPs-C) materials deposited on the surface of Fe nanoparticles were produced by calcination using ZIF-8 and iron salts as precursors. The surface of the Fe nanoparticles is the principal active site for the ECR. At the same time, the porous carbon layers play a vital role in enriching the CO₂ and altering the local chemical microenvironment. The nitride-doped carbon layer creates a hydrophobic atmosphere that enhances CO₂ adsorption and prevents water molecules from reaching the catalyst surface. Overall, the porous carbon layer promotes the ECR and prevents the HER (Figure 6d) [44]. Pyrolytic nickel diamine-dicarboxylic acid MOF was used for CO₂ electrocatalytic reduction by assembling carbon-encased Ni nanospheres onto N-doped carbon on carbon-based. And experiments combined with DFT calculations indicated that Ni nanoparticles stabilized the adsorption of intermediary *COOH on Gr/Ni(111) without affecting the *CO desorption capacity, and the pyrrole-nitrogen and graphitic-nitrogen sites in N-containing carbon nanostructures based on N-doped carbon with pyrrole nitrogen and graphite nitrogen sites enhanced the catalytic activity of carbon-based. The catalyst can effectively inhibit the HER [47].

3.2. Metals with N- and C-Sources

Wang et al. [39] constructed an N-rich graphitic carbon protective coating (N_xC) on the surface of Cu nanoparticles through the reaction of Cu nanoparticles (NPs) with TCNQ (7,7,8,8-tetracyano-p-benzoquinone) to obtain Cu-N_xC core-shell structured catalytic materials. The charge distribution characteristics of the Cu substrate was shown to be unaffected by the N_xC capping layer. The N species at the Cu/N_xC interface were found to be crucial in enriching and activating CO₂, which in turn enabled the bond reorganization process on the Cu surface. This resulted in a greater surface coverage of CO intermediates, thereby promoting the formation of C2 products from C-C couplings (Figure 7a) [39]. Fu et al. [41] developed an efficient heterostructured catalyst (Cu_xO/CN-10) through the pyrolysis of urea and copper acetate. Carbon nitride (CN) was deposited on the surface of Cu_xO, and the heterostructure enhanced the electron transfer efficiency between the electrode and the electrolyte. CN increased the charge density of the Cu_xO active site, thereby conferring the catalyst with a substantial adsorption capacity of CO₂ and intermediates, thus enriching CO₂. The catalyst was enriched with CO₂ at the interface of Cu and CN, which resulted in a substantial enhancement of the catalyst’s selectivity towards hydrocarbons [10].

3.3. Chemical Vapor Deposition (CVD) Method

Composites of In₂O₃ covered with carbon materials (In₂O₃@C) were prepared by Miao by in situ CVD and the appropriate thickness of the carbon coating effectively enhanced the p-orbital electronic configuration of In₂O₃, which was conducive to the enhanced stability of the intermediate *OCHO, promoted the electron transfer, and protected the In₂O₃ from the corrosion of the electrolyte, and significantly improved the performance of the ECR (Figure 7b) [42].

Miao et al. [45] used Ni-4,4′-bipyridine and GO as precursors and N-rich graphitic carbon-coated Ni nanoparticles loaded on graphene oxide (Ni@N-rGO), and the outcomes of experiments and theoretical calculations confirmed that the pyrrole N of the coating layer was the optimal active site, which was conducive to the adsorption and activation of CO₂, whereas the transfer of electrons from the Ni3d orbitals to the N π orbitals of the doped carbon skeleton, which enhanced the electron density of the skeleton, the desorption of the key intermediate *CO was improved, and thus its catalytic CO₂ to CO conversion performance was enhanced (Figure 7c) [45].

3.4. Carbon Nanotubes (CNTs) as a Coating

Carbon nanotubes (CNTs), when acting as a coating, have a vast surface area and numerous active sites, providing ample opportunity for catalytic reactions. They can also be doped with various heteroatoms to modify their properties and functions. Metal encapsulation in carbon nanotubes changes the electronic structure of the catalyst, protects the metal center, and contributes to limiting the spatial domains of the carbon nanotubes.

Zhou et al. [46] synthesized nickel-encapsulated, N-doped carbon nanotube catalysts (Ni@NCNT-C) via thermally annealing ZIF-8 and nickel nitrate (Ni(NO₃)₂) precursors. They then investigated the effect of annealing temperature and Ni content on the structure of the catalyst and its CO₂ catalytic activity. The effects of annealing temperature, Ni content on the catalyst morphology and catalytic activity of ECR were investigated, and the catalyst showed excellent catalytic performance when the annealing temperature was 800 °C and n(Ni)/n(ZIF-8) was 0.1, and the experiments showed that the pyridine N was the active site for the reaction (Figure 8a) [46]. Song et al. anchored atomically dispersed Ni-N-C sites and Ni nanoclusters on bamboo-like carbon nanotubes and encapsulated Ni nanoparticles in the nanotubes to obtain NiNCNT catalysts, which exhibited highly efficient ECR activity at an industrial grade current density, and were evaluated by a flow cell with a Faraday efficiency of 99.3% at a potential of −0.35 V. The Faraday efficiency of CO was 99.3% at a current density of 300 mA∙cm⁻² with a stability test time of 40 h. The excellent electrocatalytic reducibility can be attributed to the introduction of Ni nanoclusters, which enhanced the electron transfer and the local electron density of the Ni 3d orbitals, favoring the formation of the key intermediate *COOH intermediate [53]. Ni NPs were loaded on N-doped carbon nanotubes/graphene (NCNT/Gr) composites to produce Ni@NCNT/Gr. A combination of experimental and theoretical calculations demonstrated that Ni NPs are the active sites for the reaction, small Ni nanoparticle sizes provide richer adsorption sites with high surface/volume ratios for the key intermediates, *COOH and *CO, and pyrrole N has lone pair of electrons, and suitable pyrrole N doping increases the charge density and exposed active sites. In addition, the stable hybridized NCNT/Gr substrate promotes matter/electron transport, maximizes active site distribution, and prevents Ni nanoparticles from reacting with the electrolyte. The performance evaluation was carried out at H-cell, and the Faraday efficiency of the catalyst was >90% at CO in the potential range of −0.71 to −0.91 (V vs. RHE), with a current density of −37.4 mA∙cm⁻², and the activity was not significantly altered in the 24 h long-term stability test experiments [51]. The internal particles of the catalyst, which are encapsulated in carbon nanotubes, are prone to drifting and aggregating during the reaction process. This hurts the catalyst’s long-term electrochemical performance. Zhang et al. [48] encapsulated Ni nanoparticles coated with a N-doped carbon layer in N-doped carbon nanotubes (NCNTs) by the solvent heating-evaporation-calcination method (Ni@NC@NCNT), and N-doped carbon layers (NCs) and NCNTs have a double protection effect on the metal Ni, have a dual protective effect, protecting the metal core from corrosion in the electrolyte and enhancing the mechanical strength of the catalyst, and the domain-limited space well maintains the local alkalinity and inhibits the hydrogen precipitation reaction. The large specific surface area and plentiful pyridine N and Ni^δ+ sites mean it has a superior CO₂ adsorption capacity and strength (Figure 8b) [48].

Hu et al. [49] prepared Ni single-atom synergistic nanoparticle catalysts (NiSA/NP) by direct solid-phase pyrolysis, where Ni nanoparticles existed inside NCNTs and Ni single atoms were anchored on CNTs. This structure had a higher content of active sites, and the Ni nanoparticles supplied electrons to the Ni-N-C sites via the carbon nanotubes, which reinforced the adsorption of the *COOH intermediate at the center of the Ni monoatom, resulting in the efficient conversion of CO₂ at high current densities (Figure 8c) [49].

Chu et al. [50] used 2-methylimidazole as the N source and carbon nanotubes as the base. The catalysts were synthesized using a high-energy ball milling-calcination method to anchor N-doped carbon layer-coated Ni nanorods on CNTs (Ni-N-C/CNTs), which reduced CO₂ to CO with high efficiency. DFT and COHP analyses showed that Ni could significantly enhance the binding strength of N-C to the *COOH intermediate. However, the N-C capping layer weakened the binding strength of the intermediate *CO to the catalyst, which facilitated the improvement of CO₂ conversion efficiency (Figure 8d) [50]. The Ni@NCNT/Gr composite structure catalyst designed by Das et al. has a high CO Faraday efficiency. Graphene (Gr) with a high specific surface area in the catalyst acts as a substrate, which facilitates the adsorption of CO₂, and NCNTs act as a conductor of electricity while preventing the stacking of the Gr layer to a certain extent, which provides an active reaction for the Ni/NCNT interfacial Sites. The optimized pyrrole N doping, the small size of Ni nanoparticles, and the stable structure of the catalyst acted synergistically to increase the charge density, reduce the reaction, and expose a large number of active sites. Additionally, the stable NCNT/Gr substrate facilitates mass and electron transfer, optimizes active site distribution, and prevents Ni nanoparticles from reacting with the electrolyte (Figure 8e) [51]. Zhou et al. [52] synthesized N-doped carbon nanotubes-coated nickel nanoparticles catalysts (Ni@NCNT-700) by CVD, and the electron-donating effect of Ni nanoparticles resulted in enhanced electronegativity of N and tighter binding of CO to the catalyst, in addition to the domain-limiting effect of NCNTs; these could reduce the Gibbs free energy of C-C coupling while hindering the desorption of HER and intermediates *CO, and thus enable highly selective conversion of CO₂ to ethanol.

4. Carbon Materials as Catalyst-Based

Composite catalysts are fabricated through the immobilization of diverse metallic components (including elemental metals or their oxides) onto carbonaceous substrates (

Table 4. Catalytic CO₂ reduction properties of carbon materials as a basis for catalysts.

	Primary Product	Faraday Efficiency (FE) (%)	Stability (h)	Optimal Potential (V)	Current Density at the Optimal Potential (mA∙cm−2)	Electrolyzer	Ref.
Cu-10/NC	C2H4	37.0	7	−1.20	\	H-type cell	[55]
Bi2O3/p-rGO	HCOOH	94.3	39	−1.09	−16.8	H-type cell	[56]
SnO2/NC	HCOOH	87.6	20	−1.13	33.6	H-type cell	[57]
Fe-N-C(SACs)	CO	73.0	9	−0.60	\	H-type cell	[58]
Ni-NC-T	CO	98.0	20	−1.17	58.0	H-type cell	[59]
Cu/BN-C	HCOOH	70.0	12	−1.00	20.8	H-type cell	[60]
Fe-N2O2/NC	CO	95.5	12	−1.20	22.19	H-type cell	[61]
CoCu-N-C	CO	76.5	\	\	\	\	[62]
Bi2O3/PC	HCOOH	91.5	16	−1.10	150.0	H-type cell	[63]
SnO2@N-GPC	HCOO−	96.3	10	−1.20	104.7	flow cell	[64]
In/N-dG	HCOOH	100.0	14	−1.17	700.0	flow cell	[65]
Ni-N-C(P)-8	CO	99.0	10	−0.90	−20.0	H-type cell	[66]
H-Fe-NC SACs	CO	94.6	10	−0.98	−23.5	H-type cell	[67]
CuxO@NC	C2H4	48.4	19	−0.98	−8.32	H-type cell	[68]
Ni6@Ni-N3	CO	99.7	10	−1.15	500	flow cell	[69]
ZnOQDs/P-NC	CO	95.3	24	2.2	−21.6	H-type cell	[70]
NiFe-N bridge	CO	83	20	−0.5	−9.2	H-type cell	[71]

). The carbon matrix exhibits dual functional advantages in composite catalyst systems: it enhances catalytic efficiency by generating active sites, modulating surface electronic properties (including Fermi level alignment and acid-base characteristics), and facilitating charge/mass transport dynamics, which synergistically improve intermediate adsorption capacity and reduce reaction energy barriers. Simultaneously, the structural framework effectively suppresses metal nanoparticle coalescence during electrochemical operation, while robust metal-support interactions ensure prolonged catalytic durability through sustained activity maintenance and physical confinement effects. N-doped carbon materials have demonstrated substantial potential when serving as bases in ECR and have gained extensive utilization. Carbon-based materials were obtained by pyrolyzing metal salts with C and N sources from assembled precursors, MOF materials, biomass, and coal carbon. Interactions between metals and carbon substrates or the selection of proper carbon-based materials are employed to gain atomically dispersed or characteristic coordination structures.

4.1. Self-Assembled Precursor Pyrolytic Carbon-Based

Paul et al. [72] engineered a series of M-N4 coordination catalysts (M = Mn³⁺, Fe³⁺, Co²⁺, Ni²⁺, Cu²⁺, Zn²⁺, Sn⁴⁺) by immobilizing metallic porphyrins on carbon black through pyrolysis-acid leaching protocols. Electrochemical evaluation revealed distinct product selectivity correlated with metal center characteristics: Fe/Ni-based catalysts exhibited superior CO generation with >80% Faradaic efficiency, whereas Sn-modified variants predominantly yielded formate. Notably, Cu-incorporated catalysts demonstrated C-C coupling capability, producing CH₄ and C₂H₄ at combined efficiencies exceeding 30%. A hypothesis was put forward that CO may be an essential intermediate during the formation of multi-carbon hydrocarbons, by the experimental results and theoretical calculations of Cu-N-C catalysts.

Zhao et al. [55] fabricated copper-nitrogen codoped carbon composites (Cu/NC) through pyrolysis of Cu(NO₃)₂ and polyvinylpyrrolidone precursors. The resulting catalyst exhibited uniform dispersion of Cu nanoparticles on the carbon matrix, with particle dimensions expanding proportionally to copper loading. Excessive Cu content induced surface overcoverage, impairing CO₂ adsorption capacity. In optimal Cu/NC configurations, elevated Cu⁺ concentrations were observed to undergo electrochemical restructuring into undercoordinated Cu species, which stabilized *CO intermediates. Concurrently, synergistic interactions between pyrrolic N moieties and Cu-N coordination sites promoted C-C bond formation, as illustrated in the mechanistic diagram (Figure 9a) [55].

Wang et al. [59] synthesized a N-dual-doped carbon catalyst (Ni-NC-T) by pyrolyzing nickel nitrate and poly(ionic liquid) as dual precursors. This architecture achieved simultaneous anchoring of nickel single atoms (SAs) and nanoparticles (NPs) within the nitrogen-enriched carbon matrix. Crucially, multistage charge redistribution (Ni nanoparticles → carbon matrix → Ni single atoms) generated electron-enriched Ni-SA sites, enhancing CO₂ activation efficiency and *COOH intermediate adsorption affinity. This electronic modulation reduced the energy barrier of the rate-determining step while improving intermediate stabilization, collectively elevating catalytic performance as depicted in the mechanistic model (Figure 9b) [59]. Owing to its outstanding electrical conductivity and plentiful defect sites, reduced graphene oxide (rGO) is extensively employed as a base for immobilized metal-based catalysts in catalytic reactions. Niu et al. [56] prepared reduced graphene oxide-based (p-rGO) containing a mesoporous structure by using poly(ethylene oxide) (PEO) as a soft stencil and loaded Bi₂O₃ on p-rGO to obtain Bi₂O₃/p-rGO. p-rGO’s defective sites play an important role in providing abundant Bi³⁺ anchor sites to promote the uniform distribution of Bi₂O₃ NPs. As p-rGO has a rich mesopore and strong electron-donating ability, it promotes CO₂ diffusion. Also, the modified Bi₂O₃ nanoparticles exhibit enriched surface electron density, which effectively stabilizes critical reaction intermediates through electronic interactions, consequently enhancing their catalytic activation efficiency as illustrated in the mechanistic framework (Figure 9c) [56].

Tang et al. [58] fabricated a nitrogen-doped carbon-supported iron single-atom catalyst (Fe-N-C (SACs)) through a streamlined one-step calcination approach. The catalytic system derives its activity from Fe-N_x coordination centers, which synergize with the carbon matrix’s structural advantages: hierarchical porosity enhances CO₂ adsorption kinetics while facilitating CO product release, oxygen-containing functional groups improve surface hydrophilicity to promote electrolyte permeation, and interconnected pore networks optimize ionic diffusion pathways. This multifunctional architecture concurrently stabilizes reactive intermediates and accelerates interfacial charge transfer, ultimately achieving exceptional electrocatalytic performance as visualized in the structural mechanism diagram (Figure 9d) [58].

Wang et al. [60] developed a Cu/BN-C catalytic system by anchoring copper clusters on B, N-codoped porous carbon. The active centers originate from interfacial Cu-N bonds and pyridine-adjacent carbon atoms within the hybrid matrix. Nitrogen atoms act as anchoring sites for Cu cluster stabilization, while boron dopants modulate the surface electronic states of copper species. Optimized cluster dimensions combined with B/N coordination synergistically fine-tune the copper valence configuration, as evidenced by spectroscopic characterization in the electronic structure analysis (Figure 9e) [60].

Ma et al. [71] developed three distinct configurations of Ni and Fe atom pairs on nitrogen-doped carbon supports, namely NiFe-isolate, NiFe-N bridge, and NiFe bonding. The findings revealed that in the NiFe-N bridge catalyst, analysis and computational modeling demonstrated that Ni and Fe atoms, when positioned at an appropriate distance from each other, not only enhanced the adsorption of *COOH intermediates but also facilitated the desorption of *CO. This configuration exhibited better synergy, leading to higher CO₂ reduction activity and improved stability. The electrochemical performance of the H-cell was assessed at a potential of −0.5 V (vs. RHE), achieving 83% Faraday efficiency for CO, a current density of −9.2 mA·cm⁻², and showing no change in activity during a 20 h long-term stability test. Wu et al. [69] introduced coordination-unsaturated Ni atomic sites and Ni nanoclusters onto NCNTs to prepare Ni₆@Ni-N₃ catalysts, which realized the efficient conversion of CO₂ to CO at high current densities. DFT calculations indicated that the existence of Ni nanoclusters functioned to regulate the electronic configuration of Ni-N₃. This disrupted the orbital symmetry, substantially boosting the structural stability of the unsaturated Ni-N_x active site. Moreover, it promoted the adsorption of crucial *COOH intermediates and lowered the energy of the rate-determining step, thereby speeding up the ECR (Figure 9f) [69].

Min et al. [73] obtained catalysts with CuN₂O₂ active centers for the selective electrocatalytic transformation of CO₂ to CH₄ through the strategy of partial carbonization of Na₂[Cu(EDTA)] [55] (Figure 10a) [73]. Ren et al. [74] prepared CoCu-N-C catalyst, which was used in the CO₂ ECR and showed excellent CO selectivity, and the addition of Cu increased the Co dispersion, thus providing more Co-N_x active sites for the reaction. Moreover, the addition of Cu altered the electronic configuration of Co and boosted the charge-transfer efficiency (Figure 10b) [74]. Vulcan material is a commonly used conductive carbon-based material, and sometimes researchers use carbon black, graphene, and carbon nanofibers to prepare the microporous layer in gas diffusion electrodes. In this study, the Vulcan material was not used to prepare the microporous layer in gas diffusion electrodes (GDEs). Rather, it was used as a carbon substrate for catalysts for loading active components.

Wang et al. [67] utilized hollow porous carbon nanotubes as substrates to create highly loaded single-atom iron catalysts (H-Fe-NC SACs). The catalysts’ hollow and porous structure effectively improved the active sites’ accessibility and promoted mass transfer, accelerating the catalytic reaction kinetics and enabling efficient conversion of CO₂ to CO. Its Faraday efficiency reached 94.6% (Figure 10c) [67]. Guo et al. prepared ZnO_QDs/P-NC by loading ZnO quantum dots onto porous nitrogen-doped carbon substrates, which exhibited excellent CO₂ electrocatalytic performance, with a Faraday efficiency of 95.3% and a current density of 21.6 mA·cm⁻² for CO at a potential of −2.2 V vs. Ag/Ag+. The dispersed ZnO QDs provided highly active sites, the porous structure of the substrates was favorable for charge transport, and the large surface area of ZnO QDs/P/NC facilitates the acquisition of active sites [70].

4.2. MOF Material Pyrolyzed Carbon-Based

SnO₂ quantum dots (QDs) loaded on ZIF-8-derived hollow N-doped carbon materials can obtain SnO₂/NC composite catalysts, which have abundant grain boundaries, and DFT calculations and experiments show that the catalyst can catalyze the conversion of CO₂ due to the abundant grain boundaries; the catalysts have abundant mesoporous structure to promote mass transfer, which is favorable for the dissolved CO₂ in the electrolyte to interact with the active sites within the electrolyte; the robust interfacial interaction promotes the electron transfer from NC based to SnO₂ QDs, thereby facilitating CO₂ adsorption, activation, and intermediate *CO₂- generation, and thus the catalysts have high selectivity and excellent stability for formate generation (Figure 11a) [57]. Jia et al. [61] fabricated N-N-doped carbon materials via ZIF-8 pyrolysis and incorporated Fe into the matrix. The resultant atomically dispersed FeN₂O₂ active site catalyst (Fe N₂O₂/NC) delivered outstanding catalytic performance in both ECR and ORR reactions. Theoretical studies revealed that the proportion of ligand center N and O significantly impacts reaction performance, with FeN₂O₂ facilitating the formation of *COOH intermediate (Figure 11b) [61].

Using MIL-101-NH₂ as a precursor, SnO₂ crystals were anchored on porous N-doped carbon nanoflowers to synthesize a composite catalyst (SnO₂@N-GPC) for efficient CO₂ electrocatalytic reduction into formate. Experimental results and DFT calculations showed that electron transfer from SnO₂ to N-GPC at the interface occurred. The electronic interactions between SnO₂ and N-GPC enhanced the catalyst’s phase stability, strengthened CO₂ adsorption on the catalyst, and accelerated the formation of crucial intermediates, thereby lowering the energy barrier for formate generation (Figure 11c) [64].

The carbon-based porous structure of the catalyst plays a crucial role in influencing the conversion of ECR to CO. Experiments and classical molecular dynamics simulations have demonstrated that Ni-N-C(P)-8 catalysts with a hierarchical porous structure exhibit excellent selectivity and activity in the electrochemical reduction of carbon dioxide. This is attributed to the abundant mesoporous pores, which not only facilitate ionic motion and gas diffusion but also reduce the mass transfer resistance of ECR to the active site Ni during the process (Figure 11d) [66].

4.3. Petroleum Asphalt-Based Carbon Material Based

Ning et al. fabricated composite catalysts (Bi₂O₃/PC) by loading Bi₂O₃ nanosheets on porous carbon (PC) via a hydrothermal method, using petroleum asphalt as the carbon source, enabling the preparation of hundreds of grams of catalysts. The high specific surface area of PC promotes Bi₂O₃ nanosheet dispersion. The carbon-based defective sites supply electrons to Bi through O-bridges, enhancing electron accumulation on Bi and lowering the energy barrier for *OCHO intermediate formation. The staggered 3D Bi₂O₃ network boosts CO₂ adsorption and eases electron transport. In a flow cell at a high current density of 1 A, the Faraday efficiency for formic acid reaches 91.5% (Figure 12a) [63].

4.4. Bio-Based Carbon-Based

Biomass material is a sustainable and naturally occurring carbon source offering benefits like ready accessibility, eco-friendliness, cost-effectiveness, and a unique inherent pore structure [75].

In nanoparticles (NPs) were anchored on chitosan-derived N-doped defective graphene to obtain a catalyst (In/N-dG), and the defective sites and multilayer structure of the catalyst not only enhanced the chemisorption of CO₂, but also led to the formation of an electron-rich catalytic environment around the In sites, which could promote the generation of HCOOH (Figure 12b) [65]. Guo et al. [68] utilized rice husk as a carbon material precursor and fabricated a series of Cu_xO@NC composite catalysts with varying biomass carbon content via a simple solvothermal method. These catalysts feature a large specific surface area and a synergistic effect between CuxO and NC, which can provide more active sites and enhance the charge transfer rate (Figure 12c) [68].

5. Conclusions and Outlook

Carbon materials are directly used as catalysts, metal based, and metal clads in CO₂ electrocatalytic reduction, which play an important role in the optimization and enhancement of catalyst activity, selectivity, and stability, for example, in Guo et al.’s study [63], CuxO was loaded with N-doped carbon material, and the Faraday efficiency of C₂H₄ could reach up to 48.43% (−0.98 V vs. RHE), whereas the Faraday efficiency of C₂H₄ under the same conditions with Cu_xO as the catalyst was only 28.78%. Wang et al. [76] used a silver-doped oxidatively derivatized copper nanosheet (CuAg_x% NSs) catalyst, which was prepared via a combined strategy of chemical etching + electrochemical reduction + ion exchange, and successfully synthesized a single-atom Ag-doped Cu nanosheet. The catalysts were significantly enhanced in ECR to generate multicarbon (C₂₊) products by atomically modulating the doping ratio of silver in Cu nanosheets.

But their research and development are still in their infancy, and the preparation of catalysts often involves complex steps such as ice bath, centrifugation, vacuum drying, rotary evaporation, pyrolysis, etc., which is a cumbersome preparation process, and may bring economic costs. The preparation process is cumbersome, which may bring economic and environmental burdens. There are limited types of common doping reagents, which can restrict the industrial production and application of catalysts in the electrocatalytic reduction of CO₂. There still exist high overpotentials and poor long-term stability of catalysts, i.e., with the extension of the reaction time, the active center on the carbon support is deactivated and the catalyst gradually loses its activity, especially at high current densities, which are more prominent and lead to low energy conversion efficiency, so the development of catalysts with better performance through a simple process is an urgent problem in this field that needs to be solved and is yet to be studied in depth. More crucially, in terms of metal catalysts, the roles of metal and heteroatom species in ECR are still disputed, and the true active sites of ECR, as well as the functions of dopants, remain uncertain, factors that impede the development of high-performance catalysts.

Among various metal catalysts, Cu shows a unique selectivity for ECR. The catalytic products of Cu-based catalysts include both C₁ and C₂₊ products. For instance, CuAg Janus nanostructures converted CO₂ to C₂H₄ with a Faraday efficiency of 50% (−1.2 V vs. RHE) [77]. Cu₃Pd alloy converted CO₂ to CH₄ with a Faraday efficiency of 43.23% (−1.8 V vs. RHE) and a current density of −269.68 mA·cm⁻² [78]. CuO_x@C catalysts achieved a Faradaic Efficiency (FE) of 46% for ethanol [38]. Cu₉₇Sn₃ catalysts in an alkaline flow cell delivered CO Faraday efficiencies up to 98% with only 30 mV overpotential, and attained 100 mA·cm⁻² current densities at 340 mV overpotential [79].

DFT plays an important role in the prediction of the reaction mechanism and active sites from a thermodynamic point of view in the CO₂ electrocatalytic reduction reaction in conjunction with experiments and in-situ technologies, mainly used to calculate the adsorption energy of key intermediates on the catalyst surface and the partitioned wavelet density of states (PDOS), etc., which can help us to understand the reaction mechanism and design the efficient catalyst precisely [80].