Recent Advances in Transition-Metal-Mediated Electrocatalytic CO₂ Reduction: From Homogeneous to Heterogeneous Systems

Abstract

Global climate change and increasing demands for clean energy have brought intensive interest in the search for proper electrocatalysts in order to reduce carbon dioxide (CO₂) to higher value carbon products such as hydrocarbons. Recently, transition-metal-centered molecules or organic frameworks have been reported to show outstanding electrocatalytic activity in the liquid phase. Their d-orbital electrons are believed to be one of the key factors to capture and convert CO₂ molecules to value-added low-carbon fuels. In this review, recent advances in electrocatalytic CO₂ reduction have been summarized based on the targeted products, ranging from homogeneous reactions to heterogeneous ones. Their advantages and fallbacks have been pointed out and the existing challenges, especially with respect to the practical and industrial application are addressed.

1. Introduction

Carbon dioxide (CO₂) is a necessary substance for the growth of all living organisms on the earth and a vital raw material in many industrial processes [1]. Unfortunately, with the rising global population and the increasing usage of fossil fuels as a result of the dramatic development of industry, this has led to an unprecedented CO₂ generation, making global warming a pressing issue. On this topic, a vast majority of governments all over the world have expressed their concerns by introducing a carbon tax and increasing their investment in the issue of climate change. Therefore, discovering a new energy for reducing CO₂ emission and developing viable protocols for capturing and converting CO₂ into useful materials has become critical and urgent with respect to sustainable development.

In order to fulfill the high energy demands of modern society, the conversion and utilization of CO₂ [2] seem to be a more attractive and promising solution than merely capturing and geologically sequestrating CO₂ [3,4]. Since the 1970s, many plausible strategies have proved that CO₂ can be converted by chemical methods [5], by photocatalytic and electrocatalytic reduction [6], and by other methodologies [7]. However, after decades of investigation, there are still obstacles that hinder the practical application of transforming CO₂ into value-added materials and fuels. Among them, the most urgent barriers confronted are the high cost of capturing CO₂ and the high energy requirements for CO₂ conversion. These challenges make the catalysts (both photo- and electro-) a feasible and promising cutting-edge topic in terms of energy sustainability and environmental protection.

In recent years, electrocatalytic CO₂ reduction approaches have been investigated by many groups in the clean energy community [8]. The high thermodynamic stability of CO₂ induces challenges for its reduction; therefore more stable molecules, such as CO, formic acid, formaldehyde, methanol, methane etc., are generally produced through proton-coupled multi-electron steps (Equations (1)–(6),

Table 1. Selected standard potentials of CO₂ reduction in aqueous solutions ¹.

Equation Number	Half-Electrochemical Thermodynamic Reactions	E° (V)
Equation (1)	CO2 + 2H+ + 2e− = CO(g) + H2O	−0.53
Equation (2)	CO2 + 2H+ + 2e− = HCO2H	−0.61
Equation (3)	CO2 + 4H+ + 4e− = HCHO + H2O	−0.48
Equation (4)	CO2 + 6H+ + 6e− = CH3OH + H2O	−0.38
Equation (5)	CO2 + 8H+ + 8e− = CH4 + 2H2O	−0.24
Equation (6)	2CO2 + 12H+ + 12e− = C2H4 + 4H2O	−0.34
Equation (7)	CO2 + e− = CO2•−	−1.90

¹ pH = 7 in aqueous solution vs. standard hydrogen electrode (SHE), 25 °C, 1 atmosphere gas pressure, and 1 M for other solutes.

) [6]. In contrast, single electron reduction of CO₂ needs relatively high potential energy due to the large amount of transformation energy between the linear molecule and the bent radical anion (Equation (7), Table 1). In addition, the kinetics of CO₂ electroreduction involves a complicated reaction mechanism, typically generating a mixture of the aforementioned products. Even in the presence of electrocatalysts, the reaction rates are still very low. In fact, it is the insufficient catalytic activity, selectivity, and stability of currently used electrocatalysts that restrain the practical usage and technological commercialization of the electrocatalytic CO₂ conversion systems. Comparing with the gas phase transformation process, which simplifies product separation and is not bothered by CO₂ solubility issues, liquid phase electrocatalytic reduction of CO₂ displays advantages of higher productivity and less-chaotic outputs [9].

Transition-metal elements possessing partially filled d orbital electrons are believed to have the capability to facilitate bonding with CO₂ in the formation of active intermediates. In 1975, Aresta and Nobile first discovered a crystal structure of CO₂ bound to a transition metal complex [10]. In this finding, an η²-bidentate binding mode with significant bending in the CO₂ structure, involving the carbon atom and one oxygen atom, was reported, which proved the advantages of transition-metal elements as electrocatalysts to mediate the reduction of CO₂. Thus, these elements and their related compounds are most commonly explored as electrocatalysts.

Although many review articles have been published relating to CO₂ reduction reactions [8,11,12,13], they mainly focus on electrocatalyst effects. In this review, selected recent advances in developing and applying the electrocatalysts with transition metal centers is presented and discussed. Meanwhile, the demonstration of electrocatalytic CO₂ reduction is begun with homogeneous reactions discussing the influence of different ligands, followed by heterogeneous reactions with product categories.

2. Useful Notations

Basically, there are four fundamental parameters to compare and evaluate the performance of electrocatalysts for CO₂ reduction. The definitions of them are shown as follows:

(1) Catalytic selectivity (CS).

CS refers to the parts occupied by the intended products from all the reaction products, and this parameter describes the effectiveness of the reaction. From the definition we can tell that a higher CS represents a better focusing on the formation of a given product.

(2) Turnover number (TON) and Turnover frequency (TOF).

TON is the total number of the substrate molecules that a catalyst converts into product molecules, and TOF is the TON in unit time. The parameters are used as a yard stick to measure the efficiency of the catalysts.

(3) Faradaic efficiency (FE).

FE is defined as the percentage of electrons employed to generate a desired product. The FE can be simply calculated by the quotient of moles of electrons consumed in reaction product formation and the total moles of electrons transferred from anode to cathode. This parameter is not only related to the product selectivity, but also refers to the energetic efficiency.

$${\mathit{\epsilon }}_{Faradaic}=\frac{\mathit{\alpha }\mathit{n}\mathit{F}}{\mathit{Q}}$$

where, α is the number of electrons transferred (e.g., α = 8 for reduction of CO₂ to CH₄), n is the number of moles of a desired product, F is Faraday’s constant (96,485 C/mol), while Q is the total charge passed.

(4) The Tafel slope.

The Tafel plot is constructed by the logarithm of the current density changing with overpotential, and the Tafel slope is the gradient of the plot. It is noteworthy that a lower Tafel slope value indicates a better performance of electrocatalysts. As for CO₂ electroreduction, the Tafel slope value can imply the reaction mechanism.

3. Homogeneous CO₂ Reduction

In the homogeneous electroreduction of CO₂, the soluble electrocatalysts act as the electron transfer agents to facilitate and accelerate the reaction [14,15]. Ideally, the reaction can be driven near the thermodynamic potential, E° (products/substrates). However, in most cases, the difference between the applied electrode potential, V_applied, and E° (products/substrates) cannot be eliminated. As such, a significant overvoltage is required in the process of direct electroreduction of CO₂. In this situation, in order to increase the conversion efficiency, catalysts that possess formal potential, E° (Cat^n+/0), well matched to E° (products/substrates) with proper rate constant, k_cat, need to be developed. Additionally, at the electrode, the heterogeneous rate constant, k_h, must be high for V_applied near E° (Cat^n+/0) to further boost the conversion efficiency. A general layout for a homogenous electrocatalytic reduction of CO₂ is shown in Scheme 1.

The past decades have witnessed the appearance of many effective homogeneous CO₂ reduction electrocatalysts [16]. In the following subsections, recently published seminal works employing these catalysts are summarized and classified into two major categories depending on the ligand type: (1) metal catalysts with macrocyclic ligands [17]; and (2) metal catalysts with polydentade ligands [18]. It should be noted that this section focuses mainly on introducing recent literature about transition metal centered homogeneous catalysts for CO₂ electroreduction and thus may omit old literature and other types of CO₂ activation reactions.

3.1. Metal Complexes with Macrocyclic Ligands

In 2012, Saveant and co-workers reported a superior molecular system that can efficiently reduce CO₂ to produce CO by Fe complexes [19]. Introducing phenolic groups in all ortho and ortho’ positions of the phenyl groups for the modification of tetraphenylporphyrin ligand facilitated the catalysis efficiency of the reaction by the electrogenerated iron(0) complex (Scheme 2). At a low overpotential (0.47 V), the catalyst manifested a CO FE above 90% through 50 million turnovers over 4 h electrolysis with no degradation observed. Due to the phenolic hydroxyl substituents, the enhanced activity appeared to be due to the high local concentration of protons. Meanwhile, inspired by the Ni-Fe containing metalloenzyme, a cofacial iron(0) tetraphenyl porphyrin dimer, o-Fe₂DTPP (Scheme 2), was also reported to be highly effective in the electroreduction of CO₂ [20]. The electrogenerated Fe⁰(por) species exhibited excellent CO selectivity from the aspects of a high FE of 95% and a TOF of 4300 s⁻¹ at a moderate overpotential of 0.66 V.

Another pentadendate N₅ ligand (2,13-dimethyl-3,6,9,12,18-penta-azabicyclo[12.3.1]octadeca-1 (18),2,12,14,16-pentaene) was used to synthesize two effective electrocatalysts [21]. The synthesized Co^II complex selectively catalyzed CO₂ to generate CO with a high efficiency, while the Fe^III centered electrocatalyst exhibited a high selectivity for the HCOOH product at low overpotential (Scheme 3). The phenomenon that changing the metal from Co to Fe allows the switching of catalysis product over two electrons reduction of CO₂ is due to different protonation and reduction pathways. Within the Co-centered complex mediated process, a C-O bond cleavage could be expected which furnishes an OH⁻ and a CO molecule as the final product (Scheme 3a). However, Fe complexes prefer to form a η¹-OCOH coordinated intermediates through the catalytic pathway and eventually release a formate molecule (Scheme 3b). Besides, both catalysts showed sustainable catalytic activity over a long period of time and as well as turnover numbers.

3.2. Metal Complexes with Polydentade Ligands

Selectively generating CO product through a Ru-based catalyst was achieved by the use of a bulky bipyridine ligand, 6,6′-dimesityl-2,2′-bipyridine (mesbpy) (Scheme 4) [22]. The Ru complex exhibited a TOF of 1300 s⁻¹ and 95% FE for CO in the presence of Brønsted acids. Based on the mechanistic electrochemical and spectro-electrochemical studies, it was found that the cooperative redox response of the bipyridine ligand and Ru metal center at negative potentials, as well as the inhibition of Ru-Ru bond formation through steric interactions played critical roles in realizing such a remarkable performance.

In addition, the manganese tricarbonyl bromide complexes incorporating bidentate ligands, were proved to be very effective and robust as a new group of catalysts for CO₂ reduction [23,24]. In 2014, Agarwal et al. reported the electrocatalytic reduction of CO₂ using two Mn complexes with N-heterocyclic carbene (NHC) ligands (Scheme 5) [25]. In a wet CH₃CN (5% water) solution, the complexes were able to selectively transform CO₂ to CO at the potentials of −1.35 V vs. standard calomel electrode (SCE) and −1.46 V vs. SCE, respectively. As compared to [MnBr(bpy)(CO)₃], both species exhibited enhancement of catalytic current densities at the voltages. TOF values for converting CO₂ to CO with the assistance of two Mn complexes were 0.08 s⁻¹ and 0.07 s⁻¹, respectively. It is noteworthy that throughout the electrolysis, H₂ product was not observed, suggesting the excellent selectivity of the developed electrocatalysts, and an average Faradaic efficiency of 34.6% was obtained at the 4 h mark.

4. Heterogeneous Electrocatalysis of CO₂ Reduction

Although homogeneous catalysts presented excellent activity and selectivity for electrocatalytic CO₂ reduction, the non-recyclable character and high cost still impeded their application in practical industrial utilization. However, in the heterogeneous electroreduction of CO₂, the selected electrode itself also performs as the electrocatalyst, which artfully overcomes the afore-mentioned difficulties.

A simple alternative for the transition from homogeneous to heterogeneous system in the arena of electrocatalytic CO₂ reduction is to functionalize or modify the electrode with catalytic effective organic metal complexes [26]. Recently, Reisner’s group reported a novel approach by immobilizing Mn(bipyridine) catalysts on a carbon nanotube electrode (Scheme 6) [27]. Anchored by a pyrene unit, the once robust homogeneous CO₂ reduction catalyst dramatically transformed to a quasi-heterogeneous system. In such a hybrid system, the CO₂ was effectively reduced with a catalytic onset overpotential of 0.36 V and with more than 1000 turnovers at 0.55 V. Selectively, CO as the main product was obtained at high catalyst loading, whereas formate was the dominant product when lower catalytic loading was adopted.

In addition, taking the advantage of a facile synthesis protocol without any further purification, with outstanding efficiency and great potential for large-scale applications, the heterogeneous electrocatalysts yield more promising materials than the homogeneous ones to be applied in real industrial processes. The carbon dioxide molecule is adsorbed onto the conductive materials and subsequent reduction occurs on the surface of the electrode. During the past decades, various transition-metal-centered heterogeneous electrocatalysts have been widely explored for mediating CO₂ reduction reactions [28]. Cu, Au, Sn, Hg, Ag, Zn, Pd, etc. have proved to be robust and effective as both metal electrodes and electrocatalysts. Meanwhile, several transition metal oxides and chalcogenides such as TiO₂, FeO_x, Cu₂O, ZnS, MoS₂, and VS₄, have been reported as versatile electrocatalysts for CO₂ reductions as well [29]. Typically, heterogeneous electrocatalysis reduction of CO₂ involves three major steps: (i) adsorption of CO₂ molecules onto the surface of the electrocatalysts; (ii) electron transfer and/or proton migration; (iii) configuration rearrangement and desorption of products. However, not all the aforementioned electrodes are able to bind the key intermediates generating from the first step tightly. Different electrocatalysts have their own protocols to provide diversity of low-carbon products. Therefore, the following sections summarize recent works categorized by different main products.

4.1. Selective Production of Carbon Monoxide

CO is one of the important molecules applied in industrial and domestic usage. The syngas consisting of CO and H₂ can be transferred to liquid hydrocarbons through the famous Fischer-Tropsch synthesis. For the formation of CO, the reaction initiates from the reductive adsorption of CO₂ on the catalysts’ surface to construct a COOH* intermediate. After further reduction by another electron coupled to proton transfer, the COOH* intermediate is desorbed from the electrode, providing CO and H₂O as the final products. From the reaction pathway, in order to selectively generate CO product, the catalysts should fulfill the following requirements: a strong binding with COOH* and a weak coordinating with the CO molecule. Recently, nanostructured catalysts with well-controlled size, composition and surface characteristics based on different active sites have been proven robust and effective for reducing CO₂ to CO.

In 2013, Sun and co-workers studied the performance of monodispersed Au nanoparticles (NPs) of various sizes [30]. The edge sites of the NPs are considered to accelerate CO formation, and the 8 nm Au NPs with an optimal ratio of edge sites thus offered the maximum FE up to 90% at −0.67 V vs. reversible hydrogen electrode (RHE) (Figure 1a–e). Meanwhile, similar work was also reported by Strasser’s group with the sizes of Au NPs ranging from 1 nm to 8 nm [31]. As mentioned, the FE of CO declined with decreasing NP size. Both works suggested that 8 nm is the optimal size of Au NPs in CO₂ reduction. In the follow up studies, Sun and co-workers designed Au nanowires with high aspect-ratios and conducted electroreduction of CO₂ to CO at an onset potential of −0.2 V vs. RHE (Figure 1f,g) [32]. Under the catalysis of 500 nm length Au nanowires, the reduction FE reached 94% at −0.35 V, with a mass activity 1.84 A g⁻¹ (Figure 1h). Furthermore, Nam and co-workers proposed concave Au rhombic dodecahedrons as the electrodes with a high density of atomic steps on the surface, obtaining enhanced efficiency over that of Au film [33]. After long-term electrocatalytic testing, the morphologies of such concave rhombic dodecahedrons were still maintained. In addition, Kanan et al. constructed Au NPs with a high density of grain boundaries, which was proven to be a powerful approach to improve the catalytic activity of metal NPs [34].

Similarly, relevant studies demonstrated that the optimal size of Ag NPs for CO₂ reduction was 5 nm. Hwang and co-workers discovered the optimal choice by comparing three different sizes of easy-synthesized carbon-based Ag NPs (Figure 2a–d) [35]. Comparing to polycrystalline Ag foil, the 5 nm Ag/C catalysts provide four times enhanced CO FE at −0.75 V vs. RHE with the smallest overpotential about 0.63 V (Figure 2e). In Salehi-Khojin’s study, the 5 nm Ag NPs boosted the conversion of CO₂ to CO at a 10 times higher conversion rate than the bulk silver electrode in an ionic liquid 1-ethyl-3-methylimidazolium tetrafluoroborate (EMIM-BF₄) solution [36]. Moreover, after two-step de-alloying in an aqueous HCl solution, a two-dimensional nanoporous Ag electrocatalyst could be prepared from an Ag-Al precursor (Figure 2f) [37]. The readily made catalyst delivered a nearly 150 times larger surface area and exhibited at least 20 times higher activity in comparison with that of the polycrystalline silver in the electroreduction of CO₂ (Figure 2g). Such a remarkable performance was associated with the amazing stability of COOH* intermediates on the highly curved surface, thus overcoming the kinetic barrier to produce CO at a relatively small overpotential.

Apart from Au and Ag NPs, small Pd particles also presented good selectivity for CO formation in the CO₂ reduction process, with the FE approaching a maximum value of 91.2% at −0.89 V vs. RHE [38]. In addition, the catalytic activities of the Cu NPs increased with the decreasing sizes of the particles, however, not as profound as with the noble metal NPs [39]. Intriguingly, effectiveness can be promoted by changing the morphology. The Yin group prepared high-energy Cu surfaces for the electroreduction of CO₂ by an etching process [40]. The rhombic dodecahedra structure with abundant high-energy (110) facets originated from Cu nanocubes with exposed (100) facets (Figure 3a,b). It is noteworthy that the selectivity of specific binding agents is the key factor in determining the etched morphologies of Cu NPs. The formation of Cu rhombic dodecahedra was achieved in N,N-dimethylformamide solvent. Meanwhile, due to its strong capping of (100) facets of Cu, hexapod-shaped NPs with cubic arms formed under the promotion of the etching process on the corners and edges by oleylamine. The current density of the ready-made Cu electrode was nearly three times higher than that of the Cu nanocubes, at −1.4 V vs. RHE. Noticeably, the selectivity toward CH₄, C₂H₄, C₂H₆, and C₃H₈ was also higher on Cu(110) facets than on the original Cu(100) facets. These results indicated the enhanced catalytic activity and C₁, C₂ selectivity of the Cu rhombic dodecahedra catalyst (Figure 3c).

According to recent studies, multi-component catalysts exhibited alternative properties to single-component catalysts. Takanabe and co-workers explored the CO₂ electroreduction performance of a Cu-In electrode prepared by electrochemical reduction of the oxide-derived Cu in a two-electrode system (Figure 3d,e). The as-prepared electrode selectivity reduced CO₂ to CO product, while suppressing the formation of H₂ [41]. After extensive trials, the CO was produced as almost the sole product of CO₂ reduction, reaching an FE up to 90% at −0.5 V vs. RHE (Figure 3f). The Tafel slope over 0.12 V dec⁻¹ indicated that the rate-limiting step in the mechanism was the initial single-electron transfer. Density functional theory (DFT) calculation showed the replacement of one Cu by In atom on the four-fold site which disfavored H adsorption by 0.12 eV, while the CO adsorption energy was constant. Also, the In atom enhanced the stability of COOH* by roughly 0.1 eV, thereby reducing the trend to release formate as final product.

As a costless and promising alternative to replace noble-metal catalysts in hydrogen evolution reaction (HER), MoS₂ also exhibited superior CO₂ reduction performance (Figure 4a) [42]. Comparing with the noble metals, such a cost-effective substitute selectively reduced CO₂ to CO with a high current density and low overpotential (0.054 V) in an ionic liquid (Figure 4b–d). Due to the metallic character and a high d-electron density, the Mo-terminated edges are the main characteristic related to its catalytic performance (Figure 4e–g). A thorough review on this material family was conducted by Jaramillo et al., so this review does not go into any details [43].

4.2. Selective Production of Formate

As one of the most commonly used raw materials in modern industrial processes, the selective production of formate through CO₂ electroreduction is one of the facilitated ways to satisfy the supply requirement. Recent attention has been paid to further increase the selectivity and decrease the overpotential through the electrolytic process.

In 2016, ultra-thin Co related nanosheets of 4-atoms-thick were prepared through a ligand-confined growth strategy by Xie et al. (Figure 5a,b) [44]. Compared with bulky samples, the surface Co atoms of the atomically thin layers produced formate during CO₂ electroreduction with higher intrinsic activity and selectivity at lower overpotential, 0.24 V (Figure 5c). The partially oxidized nanostructured cobalt electrocatalyst further attained stable current densities of about 100 A/m² over 40 h, with almost 90% FE for formate at −0.85 V vs. SCE. Gas chromatography quantified the remaining 10% of the charge causing H₂ evolution (Figure 5d). Based on mechanistic studies, it was speculated that the declined onset potential from 0.73 V to 0.68 V upon partial oxidation was related to the facilitated rate-determining chemical reaction by the cobalt oxides. Another fast-heating strategy was applied to generate 1.72 nm thick Co₃O₄ layers, which were ideal models of transition-metal-oxide-based atomic layers for CO₂ reduction (Figure 5e) [45]. Abundant active sites and high electrical conductivity of the as-prepared nanosheets exhibited promotion for the electroreduction of CO₂. At −0.88 V vs. SCE, the ultrathin Co₃O₄ nanosheet electrocatalyst had a current density of 6.8 A/m² with a formate FE of over 60% in 20 h (Figure 5f,g).

Palmore and co-workers investigated the reduction of CO₂ on Cu foams with a hierarchical porosity [46]. Both the distribution of products and their FE presented a significant difference from those provided by smooth electropolished Cu electrodes. The highest FE of formate was up to 29% through increasing the thickness of the copper nanofoams which could suppress the HER process. The main reasons for obtaining such results were the high surface roughness, hierarchical porosity, and confinement of the reactive species. In addition to non-precious metals, Pd has the ability to convert CO₂ to formate with no overpotential. Unfortunately, within such an activity process, the low reaction rates and deactivation pathway have limited its application. Kanan et al. discovered that Pd NPs dispersed on a carbon support exhibited a high mass activity at an overpotential of 0.20 V [47]. In an aqueous solution containing CO₂⁻ and HCO₃⁻, the FE of formate could reach up to 100%. It was assumed that an electrochemically generated Pd hydride surface could be the inducement for the reduction of CO₂. The electrode could be contaminated by the formed CO products, and brief exposure to air would remove CO to restore activity. Meanwhile, Jaramillo and co-workers presented a facile tuning AuPd alloy thin film, as being effective in CO₂ electroreduction [48]. The alloys were found to generate 2e⁻ products (i.e., H₂, CO, and formate) during the conversion process. Different from the pure Au or Pd metals, the alloy thin films were more active and selective for formate production. These results indicated that the synergistic effect of Au and Pd in AuPd alloys fructified new performances rather than simple addition of the individual components.

4.3. Selective Production of Methanol

Methanol, as a liquid fuel, is one of the most important products generated directly from electrocatalysis reduction of the CO₂ process. Based on the mechanism, the protonation of a key intermediate CH₃O leads to the final formation of the methanol molecule. During this process, more protons could combine with carbon atoms, providing methane as a competitive product. Recently, nanostructured CuAu alloys with no organic additives were synthesized by electrochemical deposition with a nanoporous Cu film (NCF) template (Figure 6a) [49]. The alloy material exhibited superior catalytic activity and selectivity for the transformation of CO₂ into alcohols. The FE of methanol was 15.9% on the Cu_63.9Au_36.1/NCF electrode, which is almost 18 times higher than that of pure Cu (Figure 6b). In addition, the FE for formate in such an apparatus decreased with the increase of atomic percentage of Au in CuAu alloys. Meanwhile, Jung and co-workers further investigated the details of electroreduction of CO₂ to methanol, attempting to discover new effective catalysts through DFT calculations [50]. A few criteria, such as CO binding energy, OH binding energy, and H binding energy, were combined and collectively used as activity and selectivity determining descriptors [51]. Accordingly, the W/Au alloy was found to be a promising counterpart with increased efficiency and promoted selectivity toward methanol production in comparison to conventional Cu catalyst (Figure 6c,d).

4.4. Selective Production of Methane

As the most significant component of natural gas, methane has become an alternative fuel in both industrial and domestic facilities. Therefore, methanizing CO₂ through electrochemical methods is a promising process to convert greenhouse gases to valuable fuels. However, as the reduction of CO₂ to CH₄ involves an eight-electron process, the formation of a wide range of products, such as ethylene, hydrogen, carbon monoxide, and formate, can hardly be avoided. To date, the majority of studies on the CO₂ electroreduction to CH₄ adopted Cu-based materials as the catalyst. Alivisatos and co-workers proposed that the well-dispersed Cu NPs supported on glassy carbon exhibited up to 4-fold greater methanation current densities as compared to relatively pure copper foil electrodes [52]. The FE for CH₄ on such Cu NPs reached 80% at −1.25 V, which is the highest value till now. Also, the Tafel plot showed the reaction pathway was CO₂→CO₂⁻→CO₂-CO₂⁻→CO→CH₄, and the rate-limiting step was the formation of CO₂-CO₂⁻. Meanwhile, guided by the dramatic composition effect, various Cu catalysts were also investigated. Sun and Yan synthesized bimetallic CuPt NPs with diverse Cu/Pt ratios (Figure 7a) [53]. The highest FE for CH₄ was 21% at −1.8 V obtained on the sample of Cu₃Pt₁ NPs (Figure 7b). DFT calculations suggested that the protonation of adsorbed CO* to form adsorbed HCO* was the key step in the reaction process. Jin et al. prepared new Au₃Cu alloy NPs via a phase-stabilized synthesis, the size of which could be facilely tuned by controlling the amount of Au precursors (Figure 7c) [54]. Compared to Au NPs, the Au₃Cu surface could absorb more CO molecules, which promoted the formation of CH₄, thus leading to the distinct selectivity of CO₂ reduction products (Figure 7d). In addition, Mul and co-workers investigated the influence of KHCO₃ concentration and CO₂ pressure on the production selectivity of Cu electrodes [55]. The results revealed that the selectivity of CH₄ increased from 1% to 20% with the increase of electrolyte concentrations. Importantly, the low CO₂ pressures could be better in selectively producing CH₄ through the CO₂ reduction process.

4.5. Selective Production of C₂ Products

In the arena of electrochemical CO₂ reduction, Cu catalysts not only showed effective catalytic performance in generating C₁ products, the robust catalytic activity also presented potential for formation of C₂ products. According to recent studies, the selectivity of C₂H₄ can be effectively improved by surface treatments of Cu materials. For instance, a roughened copper electrode with copper NPs coated surfaces was found to be very effective in electrochemical CO₂ conversion towards ethylene production [56]. In addition, Baltrusaitis and co-workers investigated the effectiveness of cuprous oxide films of various thickness and [100], [110], and [111] orientation in the electrocatalytic reduction of CO₂ [57]. The experimental results demonstrated that the parent Cu₂O film thickness rather than the different orientation was the main factor that affected the selectivity of the process, which provided an FE up to 35% for C₂H₄ at −1.1 V. With thickness of Cu₂O film increasing, the selectivity for ethane soared, while that for ethylene declined. Meanwhile, a novel 3D chrysanthemum-like structured Cu nanoflower catalyst was prepared by Yu and co-workers, which expressed electrocatalytic CO₂ reduction selectively to C₂H₄ product with an FE value of about 10% at −0.9 V in aqueous solution [58].

Recently, Lewis and co-workers prepared Ni-Ga bimetallic films employing a drop-casting strategy with three different phases for reducing CO₂ to the highly reduced C₂ products including ethylene and ethane and C₁-product methane [59]. In aqueous bicarbonate electrolytes at neutral pH, the onset potential was found to be −0.48 V vs. RHE, which was 0.25 V more positive than that of polycrystalline copper. The Ni₅Ga₃ phase provided the highest yield of 0.25 mol h⁻¹ for acetate production. An isotope labeling experiment was also conducted with ¹³CO₂ to confirm that the origin of the generated C₂ products was from the reduction of CO₂.

5. Conclusions and Outlook

In recent years, a number of novel transition-metal-centered electrocatalysts have been developed for the electrocatalytic reduction of CO₂. Instead of focusing on optimization of the reaction conditions, most of the latest studies intended to purposefully design the structures and constitutions of the catalysts, as well as investigate the actual reaction mechanisms. This review summarizes the recent progress of CO₂ electroreduction that is dependent on its catalytic phase states. Although numerous works have shown powerful strategies in converting CO₂ to value-added C₁ or C₂ products, there are still several challenges and constraints remaining, such as moderate catalytic activity, insufficient product selectivity, and inadequate catalyst stability. It seems that the maturity of CO₂ electroreduction technology is still far away from fulfilling the requirements for industrial and commercial applications. To overcome these deficiencies, enormous efforts should be made on the following aspects: (1) Innovation of electrocatalysts to enhance activity and stability; (2) Further fundamental understanding through experimental and theoretical modeling; (3) Optimization of electrolytes, reaction apparatus, and system designs for practical applications.

In summary, in order to overcome excessive and environmentally harmful CO₂ emissions, and to remit energy shortage issues, it is of critical importance to develop electroreduction of CO₂ to generate value-added low-carbon fuels. With continued and extensive efforts focusing on overcoming the confronted difficulties on insufficient catalyst activity, product selectivity, and catalyst stability, we believe that e CO₂ electroreduction technology will become practical in the near future.