Emerging nanostructured carbon-based non-precious metal electrocatalysts for selective electrochemical CO₂ reduction to CO

Xinyue Wang ^a, Qidong Zhao ^c, Bin Yang ^a, Zhongjian Li ^a, Zheng Bo ^d, Kwok Ho Lam ^e, Nadia Mohd Adli ^f, Lecheng Lei ^a, Zhenhai Wen *^b, Gang Wu *^f and Yang Hou *^a
aKey Laboratory of Biomass Chemical Engineering of Ministry of Education, College of Chemical and Biological Engineering, Zhejiang University, Hangzhou 310027, China. E-mail: yhou@zju.edu.cn
^bCAS Key Laboratory of Design and Assembly of Functional Nanostructures, Fujian Provincial Key Laboratory of Nanomaterials, Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences, Fuzhou, Chin. E-mail: wen@fjirsm.ac.cn
^cSchool of Petroleum and Chemical Engineering, Dalian University of Technology, Panjin Campus, No. 2 Dagong Road, New District of Liaodong Bay, Panjin 124221, Liaoning Province, China
^dState Key Laboratory of Clean Energy Utilization, College of Energy Engineering, Zhejiang University Hangzhou, China
^eDepartment of Electrical Engineering, The Hong Kong Polytechnic University, Hunghom, Hong Kong
^fDepartment of Chemical and Biological Engineering, University at Buffalo, The State University of New York, Buffalo, NY, 14260, USA. E-mail: gangwu@buffalo.edu

First published on 15th October 2019

---

---

Zhenhai Wen

Zhenhai Wen received his M.Sc. from the Beijing University of Technology in 2004 and Ph.D. degree from the Chinese Academy of Sciences, China in 2008. He then worked at the Max Planck Institute for Polymer Research in Germany as a Humboldt postdoctoral research scholar and then moved to the University of Wisconsin-Milwaukee as a postdoctoral researcher in 2010. He joined the Fujian Institute of Research on the Structure of Matter, Chinese Academy of Sciences in 2015. His research topics mainly include the design and synthesis of functional nanostructures and exploration of their applications in electrochemical energy conversion and storage systems.

Gang Wu

Gang Wu is an associate professor in the Department of Chemical and Biological Engineering at the University at Buffalo (USA). He completed his Ph.D. studies at the Harbin Institute of Technology (China) in 2004 followed by postdoctoral training at Tsinghua University (China; 2004–2006), the University of South Carolina (USA; 2006–2008), and the Los Alamos National Laboratory (LANL, USA; 2008–2010). He was then promoted to be a staff scientist at the LANL. Then he joined the University at Buffalo (USA) as a tenure-tack assistant professor in 2014 and was promoted to a tenured associate professor in 2018. His research focuses on functional materials and catalysts for electrochemical energy storage and conversion.

Yang Hou

Yang Hou received his Ph.D. degree from the School of Environmental Science & Technology at the Dalian University of Technology in 2011, and then did ∼6 years of postdoctoral research in the Department of Chemistry at the University of California, Riverside, Department of Mechanical Engineering, University of Wisconsin-Milwaukee, and Center for Advancing Electronics Dresden, Technische Universität Dresden. He has been a Professor at the College of Chemical and Biological Engineering at Zhejiang University since March 2017. His current research focuses on the design and synthesis of low-dimensional nanomaterials for energy and environmental applications.

---

1. Introduction

Conversion of carbon dioxide (CO₂) into value-added fuels or chemicals holds great promise for reducing CO₂ emissions and facilitating the storage of renewable energy.¹ Compared with traditional CO₂ conversion methods, electrochemical CO₂ reduction (ECR) technology driven by renewable energy-generated electricity has been widely recognized as a promising pathway owing to its multiple advantages, including (I) mild operating conditions, (II) tunable conversion efficiency and selectivity by applied voltages or current, and (III) potential integration into energy storage systems as a unit.^2–4

As a means of energy conversion and storage, the ECR is a promising strategy to convert CO₂ into desirable chemical fuels, including carbon monoxide (CO), methane, ethylene, ethane, etc. for energy storage. In particular, CO gas is widely utilized as an essential raw material for Fischer–Tropsch synthesis in industrial production and also as a reducing agent during metallurgical processes.⁵ Compared with the multi-electron reduction reactions forming methane or ethylene products, the formation of CO molecules involves a two-electron transfer process that significantly reduces the ECR reaction barrier and the conversion mechanism is relatively simple.⁶ As a result, ECR to CO has been widely studied in the past few years, and it is generally accepted that electrochemical CO₂-to-CO conversion involves the following four crucial steps:^7–10

CO2 + * = *CO2	(I)

*CO2 + H+ + e− = *COOH	(II)

*COOH + H+ + e− = *CO + H2O	(III)

*CO = CO + *	(IV)

Theoretical calculations demonstrated that electrocatalysts are the key components for CO₂-to-CO conversion because they not only function to better stabilize *COOH intermediates and to enhance the electron transport, but also play a paramount role in converting *COOH into *CO. Meanwhile, an optimal adsorption/desorption capability of electrocatalysts is of crucial for promoting CO production. Up to now, various precious metal catalysts, such as Au and Ag based materials with high activity and selectivity,^11–13 have been extensively studied as electrocatalysts towards CO production.^12,14–16 Unfortunately, the high price and scarcity of these precious metal electrocatalysts hinder their large-scale applications. In recent years, non-precious metal catalysts, such as alloys, transition metal compounds, and transition metal-based molecular catalysts,¹⁷ have been widely reported to replace precious metal materials to realize CO₂-to-CO conversion. For practical ECR applications, essential requirements of high activity and selectivity, low cost, and high stability should be considered. Despite non-precious metal ECR electrocatalysts addressing the cost issue, they can hardly meet all of the aforementioned requirements.^18,19 Recently, nanostructured carbon-based non-precious metal electrocatalysts (Nano-CNMs) have received extensive attention owing to their low cost, excellent stability, and controllability, allowing for their use as supports for active material and composition tuning, as well as morphology and structure fine-tuning to efficiently catalyze the ECR. The emergence of a large number of novel Nano-CNMs with excellent ECR performance has motivated us to review this emerging research area.

In this review, we summarize recent developments in Nano-CNM materials and mainly focus on the relationship between electrocatalytic performance and structure/composition. Based on the catalyst formulation and configuration, they can be divided into three categories: metal-free heteroatom-doped carbon, transition metal–heteroatom co-doped carbon, and carbon/metal compound hybrid materials. Synthesis methods and corresponding ECR performances of these Nano-CNM materials are summarized in Tables 1 and 2 for general comparisons. Future opportunities and remaining challenges of Nano-CNM materials for highly efficient CO production via ECR are highlighted.

---

---

2. Metal-free heteroatom-doped carbon materials

In the last few years, heteroatom-doped carbon materials, such as nitrogen (N), fluorine (F), sulfur (S), and boron (B) doped nanocarbons, have been widely developed as efficient electrocatalysts for the ECR, and great progress has been made.²³ Although these reported heteroatom-doped carbon materials are metal-free catalysts, their catalytic activities could compete well with those of precious metal containing materials. The reason for the outstanding ECR electrocatalytic activity can be attributed to the introduction of heteroatoms, resulting in conjugation of sp²–sp² linkages or the delocalization of π-orbital electrons.²⁴ Benefiting from the different electronegativity of heteroatoms (N, B, F, etc.) from that of carbon atoms, the introduced heteroatoms could disrupt the lattice arrangement of carbon and produce positive or negative groups, such as B–C and N–C bonds, which can interact with CO₂ or *COOH intermediates, thus reducing the energy barrier of the reaction and facilitating the ECR.²⁵ In light of these benefits, numerous heteroatom-doped nanocarbon electrocatalysts have been developed to produce CO via ECR.

2.1. Metal-free N-doped carbon materials

Among others, N-doped nanocarbon materials are considered ideal ECR electrocatalysts for CO production, owing to their good selectivity and stability.²⁶ Density functional theory (DFT) calculations showed that the N dopants could change the spin density of adjacent carbon atoms and induce a higher density of donor states close to the Fermi level, which could lead to an increased electron conductivity of N-doped nanocarbons.²⁷ The similar size of N and C atoms makes C atoms easily to be replaced by N atoms. According to the different chemical environments of N atoms, the N dopants are divided into four types, pyridinic-N, graphitic-N, pyrrolic-N, and oxidized-N, as shown in Fig. 1a.²⁰ Although most previous studies demonstrated that pyridinic-N contributed to the high ECR activities due to its positive effect on the rate-determining step,^21,28,29 there is still a debate on the role of pyridinic-N and pyrrolic-N in ECR catalysis. For instance, pyrrolic-N-rich carbon nanotubes (CN-H-CNTs) can be prepared by a hydrothermal steam etching strategy.³⁰ The X-ray photoelectron spectroscopy (XPS) results showed that pyridinic-N could be partially removed while the percentage of pyrrolic-N would accordingly increase due to the etching treatment; as a result, the CN-H-CNTs showed an increased CO faradaic efficiency (FE) from 60% to 88% for the ECR. The researchers also believed that the presence of pyrrolic-N greatly inhibited the competitive reaction of water splitting.

Oppositely, effective synthesis methods have been developed to regulate the type of N doping and identified pyridinic N as real active sites. As a typical example, N-doped 3D graphene foam (3D-NG) was prepared through the chemical vapor deposition (CVD) method and achieved a CO FE of 85% at a low overpotential of −0.47 V.²⁸ The high resolution N 1s XPS spectrum revealed that the content of pyridinic-N was up to 4.1 at% under pyrolysis at 800 °C and the activity was optimal. Therefore, the high ECR activity was attributed to the formation of active sites induced by pyridinic-N dopants. In addition, it should be noted that the synthesis method can be further extended to fabricate pyridinic-N doped carbon nanotubes (NCNTs), which manifested a maximum CO selectivity of 80%, slightly lower than that of 3D-NG, and an unprecedented overpotential of −0.18 V for the NCNTs still makes this method more energy efficient and practical for applications.²¹ The pyridinic-N in the NCNTs was recognized as the real catalytic sites to selectively convert CO₂ for CO production, as it is beneficial for the formation of intermediates and desorption of CO molecules (Fig. 1b).

Apart from the complex CVD method, other simple synthesis methods have also been employed to prepare N-doped Nano-CNM catalysts. For example, N-doped graphene nanoribbon networks (N-GRW) with tunable types and contents of N dopants were successfully synthesized by a high-temperature pyrolysis approach (Fig. 1c). The N-GRW electrocatalyst exhibited superior ECR catalytic activity, with a higher CO FE of 87.6% than that of the above-mentioned 3D-NG at a mild overpotential of −0.49 V. As shown in Fig. 1d, pyridinic-N was blocked by the phosphate anion, resulting in a decrease of the catalytic activity. Thus, the activity could be attributed to the activation of pyridinic N and the DFT results also supported this experimental observation (Fig. 1e).²² Along this line, the high-temperature pyrolysis method was further utilized to synthesize N-doped carbon nanotubes (NCNTs) with different contents of pyridinic-N by adjusting the employed precursors. When the pyridinic-N content increased from 0.3 to 1.1 at%, the CO FE of NCNTs was enhanced from 14% to 80%, demonstrating that the pyridinic-N in the NCNTs likely acted as the primary active sites for the ECR.²⁹ In addition to graphene and CNTs, other carbon-based precursors (e.g., polymers) were also employed to synthesize Nano-CNMs. For instance, metal-free N-doped carbon nanofibers (CNFs) were obtained by pyrolysis treatment of an electrospun polyacrylonitrile (PAN) solution,³¹ and the obtained CNFs could convert CO₂ into CO with a CO FE of 98% at a very low overpotential of −0.17 V, which was the lowest among those of all the above-mentioned ECR electrocatalysts. Raman spectroscopy suggested that N defects are integrated into the carbon lattice. Theoretical calculations and experimental characterization suggested that the high ECR performance was attributed to the interactions between the dopant N atoms and their adjacent C atoms, which redistributed the charge and spin density of prepared CNFs.

In addition to chemicals, plant biomass materials can also be used as raw precursors to synthesize the Nano-CNM catalysts. For instance, N-doped hierarchical honeycomb porous carbon (NDC) was synthesized by a facile one-step pyrolysis treatment of wheat flour. In spite of the NDC only displaying a CO selectivity of 83.7%, the durable NDC electrocatalyst can continuously convert CO₂ to CO for more than two days.²⁰ Besides, solid fossil fuels can be directly used as carbon sources through doping to prepare Nano-CNMs. As a typical example, bituminous coal as a precursor was treated with ammonia to prepare N-doped porous carbon (CNPC).³² In order to explore the effect of the N configurations, the calcination temperature was adjusted. XPS analysis showed that the relative content of pyridinic-N increased to the maximum at 1100 °C and the CO FE reached the maximum at the same time. Thus, pyridinic-N is considered to play an important role. In addition, commercial activated carbon BAX derived N-doped porous nanocarbons were successfully prepared through a simple carbonization treatment at high temperature under a nitrogen (N₂) flow, and the achieved N-doped porous nanocarbon leads to CO formation with 40% FE via ECR.³³ Notably, the linear relationship between the content of pyridinic-N and CO FE was fitted, which proved that pyridinic-N is crucial for the ECR.

2.2. Metal-free other heteroatom (F, S, or B)-doped carbon materials

Similar to N atoms, the more electronegative F and S atoms and the less electronegative B atoms (relative to C atoms) could also modify the electronic structure of adjacent carbon atoms, achieving an asymmetric distribution of spin density.³⁴ These dopings resulted in similar electrocatalytic properties to N-doped Nano-CNMs.^34–37 For instance, a F-doped carbon (F–C) electrocatalyst with a porous structure was synthesized through a facile pyrolysis process (Fig. 2a).³⁸ Benefitting from the F dopants, the F–C catalyst exhibited a maximum CO FE of 90% at a low overpotential of −0.51 V for ECR (Fig. 2b). Compared to pristine carbon, F–C possessed abundant defects and a large surface area as confirmed by Raman spectroscopy and N₂ adsorption–desorption, respectively. With activation of F atoms, the neighboring carbon atoms tightly adsorbed the *COOH intermediate and favorably acted as an energy barrier for the hydrogen evolution reaction (HER) based on DFT calculations (Fig. 2c). Similarly, although the active sites of S atoms could stabilize the *COOH intermediates, the S-doped carbon (S–C) materials tended to produce methane with a small amount of CO formation (CO FE of ∼2%).³⁹ Likewise, B-doped carbon materials may be disinclined to produce CO. For example, B-doped graphene (BG) could mainly convert CO₂ to formic acid, and there was a small amount of CO formation.⁴⁰ In view of the unsatisfactory ECR performance of single heteroatom dopants, multiple heteroatom co-doped carbon materials were developed to increase the type and activity of heteroatom dopants. For instance, S, N co-doped carbon materials synthesized via a pyrolysis treatment of electrospun polymer nanofibers (Fig. 2d) were reported to exhibit a much higher ECR catalytic activity than the S-doped carbon materials.⁴¹ The CO selectivity of S, N co-doped hierarchically porous carbon nanofibers (NSHCFs) was significantly improved to 94% at 100 mA cm⁻², but it was a little bit lower than that of N-doped CNFs. It was believed that pyridinic-N was more active than thiophene S, and the N dopant was more likely to be capable of stabilizing the *COOH intermediates. On the other hand, previous reports demonstrated that the S dopant could also act as a coadjutant to increase the content of pyridinic-N and suppress the formation of graphitic-N.⁴² X-ray photoelectron spectroscopy (XPS) results showed that, after doping of S into N-doped carbon, the content of pyridinic-N is increased from 43% to 52% and that of graphitic-N decreased from 52% to 44%, thus resulting in the maximum CO FE increasing from 75% to 92%. Such an increase in CO FE could be attributed to the S dopants, which provide more edge positions to host high-density pyridinic-N.

3. Transition metal heteroatom co-doped carbon materials

Although doping with heteroatoms improves catalytic performance of pristine nanocarbon,⁴⁵ further introduction of transition metals into heteroatom-doped nanocarbons could greatly enhance the ECR catalytic activity. Because their partially filled d-orbitals close to the Fermi level could chemically optimize the electronic structures. This enhances the reaction kinetics, subsequently promoting the formation of intermediates.^46,47 In addition, low cost, high selectivity, and high energy efficiency make transition metal heteroatom co-doped carbon materials a promising candidate for ECR electrocatalysts.⁴⁸

3.1. Transition metal–N co-doped carbon materials

Transition metal–N co-doped carbon materials (M–N–C), such as Ni–N–C, Fe–N–C, Co–N–C, Mn–N–C, etc., have recently been reported with unique MN_x active sites.^49–51 In these M–N–C materials, N atoms are coordinated with a transition metal center forming catalytically active sites. The active sites are similar to those reported for the molecules structure of phthalocyanine, which has been proposed to have ECR catalytic activity. M–N–C materials are considered promising alternative electrocatalysts not only due to the outstanding catalytic performance, but also they can be prepared by using cheap precursors via a simple synthesis method.⁵² DFT calculations revealed that the M–N–C materials possess much lower absorption free energy for *COOH intermediates than transition metal nanoparticles (NPs) directly loaded on the surface of nanocarbons.⁵³ In addition, the interactions of transition metal with N atoms could increase the rate of charge transfer and greatly reduce the local work function of nanocarbons to prompt the ECR.⁵⁴ In light of the unique structural characteristics, different synthesis methods were widely developed to prepare the M–N–C materials (Fig. 3a). For example, Ni and N modified graphene (Ni–N–Gr) was synthesized by annealing the metal complex and graphene oxide (GO) composite at high temperatures. The Ni and N elements were observed to disperse uniformly on the graphene by energy-dispersive X-ray (EDX) analysis, while no Ni aggregates were observed in the X-ray diffraction (XRD) pattern. The Ni–N–Gr exhibited a CO FE of >90% at potentials from −0.7 to −0.9 V.⁵⁵ Such a high ECR selectivity could be attributed to the unique formation of a Ni–N–C structure, which could be well retained without obvious NP aggregation using an appropriate heat treatment process. Notably, this was the first report on a Ni–N–C catalyst with excellent ECR performance. To increase more isolated Ni–N_x–C sites in catalysts with excellent ECR performance, a nanocarbon catalyst with a high loading amount of Ni single atoms (5.44 wt%) was prepared by a pyrolysis treatment of zeolitic imidazolate frameworks (ZIF) containing Ni ions, denoted as C–Zn_xNi_y ZIF-8 (Fig. 3c). After a pyrolysis, all elements were uniformly distributed in the C–Zn_xNi_y ZIF-8 (Fig. 3b).⁵⁶ By regulating the Ni content, the C–Zn₁Ni₄ ZIF-8 achieved a CO FE of over 92% in a wide potential window between −0.53 and −1.03 V vs. RHE, much higher than the afore-mentioned Ni–N–Gr material's FE for CO production, which was only 90%. The high ECR catalytic activities might be due to the fact that the Ni–N₄–C structure was mostly retained to a great extent avoiding the conversion of the Ni atoms into Ni NPs. Along this line, other Ni–N–C materials were prepared. As shown in Fig. 3d, Ni species highlighted by red circles were atomically dispersed in graphene, and were not present as nanoclusters or particles. They displayed a maximum CO FE of 99% at a current density of 28.6 mA cm⁻².⁵⁷ Such an ECR activity of Ni–N–C could significantly suppress the competitive HER because the presence of the Ni–N–C structure efficiently inhibited the formation of HER intermediates. Theoretical calculations revealed that, after absorbing CO₂ molecules, the free energy of the *COOH intermediates formed by Ni–N₄ sites was lower than that of the NC structure. Ni–N₄ significantly reduced the rate determining step of the reaction barrier, which contributed to the high ECR activity for CO production (Fig. 3e). However, for industrial applications, conversion rates are more important than selectivity. Thus, a Ni–N–C gas diffusion electrode (GDE) and flow cell were constructed to realize practical applications. Notably, when the current density was up to 200 mA cm⁻², the CO yield was significantly better than that of a noble Ag catalyst.⁵⁸ Besides the Ni–N–C materials, other transition metal–N co-doped carbon based catalysts, such as Fe–N, Co–N, Mn–N, etc., with excellent ECR performances have also been reported.^59–62 For instance, isolated Fe³⁺–N–C was prepared via a pyrolysis treatment of iron containing ZIF-8, and the oxidation state of Fe³⁺ species was well maintained in this system to improve CO₂ adsorption. When operating in a flow-cell, the CO partial current density reached the industrial application level of nearly 100 mA cm⁻² at a low overpotential of 340 mV.⁵⁹ The effect of the valence state of the metal center on ECR activity was discussed in depth. The higher valence (+3) of the metal center was not only beneficial for CO₂ adsorption, but also weakened the CO binding at the metal center, which indicated that the reaction was not limited by the CO desorption step. As a result, a higher current density of Fe³⁺–N–C via ECR could be achieved as compared to that of Fe²⁺–N–C. Due to the variable valences of the Mn element, there are few studies on Mn-based nanocarbon materials as efficient electrocatalysts for ECR, but the electrocatalytic activity of Mn species could still be enhanced by optimizing their electronic structure. As a typical example, chlorine (Cl) and N dual-coordinated Mn sites dispersed on graphene (Cl,N-Mn/G) catalyst were prepared by a one-step pyrolysis method using Mn-ethylenediamine-Cl polymer. The structure of the Cl atom axial suspension on Mn–N₄ was confirmed by Mn K-edge X-ray absorption near-edge structure (XANES) and extended X-ray absorption fine structure (EXAFS). Compared with only N coordinated Mn dispersed graphene (N–Mn/G), the Cl, N–Mn/G showed a superior turnover frequency (TOF) of 38347 h⁻¹ and outstanding selectivity, which was attributed to the fact that the unique electronic structure of Mn species significantly reduced the formation barriers of *COOH intermediates.⁶¹

3.2. Transition metal–other heteroatom co-doped carbon materials

Despite certain research studies, the application of M–N–C in the ECR is still in its infancy; to improve the overall catalytic activity, different types of Nano-CNM catalysts comprising transition metal and heteroatom (N, P, S, etc.) co-doped nanocarbons were also constructed. Theoretical studies have shown that, due to the strong electronegativity of N adjacent atoms, transition metal atoms have higher reaction free energy for adsorption of intermediate products, which increases the potential barrier in the reaction process. Thus, by introducing suitable foreign atoms, the electron donating properties of transition metal atoms can be adjusted to reduce the potential barriers and further improve the intrinsic catalytic activity of M–N–C materials.⁶³ For instance, N, P co-doped 3D mesoporous carbon frameworks (3D N,P Co-MPC) were prepared by a self-growth-templating method, achieving a CO FE of 63% for the ECR.⁶⁴ Besides, N and S co-doped Fe-containing highly porous carbon (Fe–NS–C) was obtained through a pyrolysis treatment of 2-pyrrolecarboxaldehyde and 2-thiophenecarboxaldehyde organic gels. The achieved Fe–NS–C carbon gel exhibited a CO FE of nearly 85%, which was comparable to that of the control Fe–N–C sample.⁶⁵ It could be found that, besides the N dopants, the introduction of additional S atoms into the Nano-CNMs could not only facilitate the formation of *COOH intermediates, but also increase the porosity and specific surface area of carbons, thus improving ECR activity and selectivity.

4. Carbon-based hybrid materials

Compared to heteroatom-doped carbon materials and transition metal–heteroatom co-doped carbon materials, carbon-based hybrid materials, as popular Nano-CNMs, could allow for combining the advantages of each material, such as large surface area, high electrical conductivity, and dense active sites.⁶⁹ Owing to these unique advantages, carbon-based hybrid materials have been recently utilized as a class of effective ECR electrocatalysts for CO production. In the hybrid system, several different kinds of electrocatalysts containing transition metals, transition metal compounds, and metal-free nanocarbon materials were introduced into nanocarbon based supports, acting as mediums for charge transport and active centers to enhance ECR performance.

4.1. Transition metal/carbon-based hybrid materials

Transition metal/carbon-based hybrid materials have been considered efficient Nano-CNM catalysts for the ECR because they combine the advantages of transition metals as active sites to reduce the energy barriers with the advantages of nanocarbon materials as supports to provide a high electron transfer rate and large surface area for the deposition of active materials.⁷⁰ For instance, transition metal (Cu, Ni, Fe, Co, etc.) NPs were uniformly deposited on the surface of carbon black (M-CB) by an incipient-wetness impregnation of transition metal ions into CB (Fig. 4a and b).⁶⁶ All of these transition metal NPs loaded on M-CB showed certain ECR catalytic activity; however the overall ECR performances of transition metal NPs-CB for CO production were still not as competitive as the HER performances (CO FE < 10%). This result was mainly because transition metal NPs suffer from difficult control of the loading amounts, thus leading to the excessive NPs serving as the active sites for the HER. Following this route, the loading methods of transition metal NPs were further improved for controlled synthesis of transition metal/carbon-based hybrid materials. For instance, a unique carbon-based nanocomposite consisting of Cu NPs and reduced GO (rGO) supported on a Cu substrate (Cu-rGO) was prepared by using a facile electrochemical reduction method via cyclic voltammetry. Compared with Cu particles, the NP size after compositing with rGO was more uniform (Fig. 4c and d).⁶⁷ The CO FE of the Cu-rGO nanocomposite reached 40% as compared with that of the above-mentioned M-CB. The enhanced ECR catalytic activity could be attributed to the synergistic coupling effects between the Cu NPs and rGO, which makes CO₂ molecules easier to be absorbed. Considering that the doped carbon materials could facilitate the ECR reaction better than nanocarbon materials, transition metal NPs/doped carbon materials were further designed. For instance, in situ encapsulation of Ni NPs into N-doped carbon nanosheet and nanotube hybrid substrates (Ni/N-CHSs) was achieved by a pyrolysis treatment of a metal ion–benzene complex and dicyandiamide (DCDA) as the nitrogen precursor (Fig. 4e).⁶⁸ TEM and EDX revealed that the Ni NPs were completely encapsulated by carbon layers instead of sticking to the surface. The Ni/N-CHSs showed a CO FE of 93.1% for the ECR, which was much higher than that of the controlled N-doped carbon sample. DFT calculations demonstrated that the Ni NP assisted N-doped carbon could efficiently tune the electron distribution, thus resulting in the reduction of the reaction barrier for *COOH formation (Fig. 4f and g). Moreover, the positive role of the carbon layer coating the NPs has been gradually observed. For example, the carbon layer coated Ni NPs supported on N-doped carbon (Ni-NC@C) was synthesized by pyrolysis of nickel metal organic frameworks (MOFs) on carbon in argon, and a CO FE of 94% at an overpotential of −0.59 V was achieved. Structural characterization showed that most of the crystalline Ni NPs were well wrapped by carbon shells, which resulted in a suppressed HER performance compared with that of bare Ni NPs.⁷¹

4.2. Transition metal compound/carbon-based hybrid materials

In addition to the transition metal/carbon-based catalysts, transition metal compound/carbon-based materials with high valences of transition metal ions could also show great promise by providing chemical functionality that stabilizes the incipient negative charge on CO₂ molecules or by mediating electron transfer directly.⁷⁷ Among various transition metal compounds, transition metal oxides are the most common and widespread form in nature. However, on account of the low number of active sites and poor CO selectivity, transition metal oxides are not ideal ECR electrocatalysts for CO production.⁷⁸ Up to now, only a few carbon-based transition metal oxide hybrids have been reported as efficient Nano-CNMs for the ECR. As a typical example, NiO impregnated into multi-walled carbon nanotubes (MWCNTs) to form a NiO/MWCNT composite was achieved by a pyrolysis treatment of MWCNTs impregnated with Ni salt (Fig. 5a).⁷² The CO FE of the NiO/MWCNT composite was only about 5%, which was much lower than that of other heterogeneous carbon-based materials. Following this step, a hybrid ECR catalyst formed by loading Co₃O₄ on the surface of NG (NG-Co₃O₄) was prepared by a hydrothermal treatment of Co ions adsorbed NG (Fig. 5b and c).⁷³ Although the CO FE of 10–20% for NG-Co₃O₄ was far higher than that for NiO/MWCNT, and even comparable to that for the above-mentioned transition metal NPs-CB, the CO selectivity of the NG-Co₃O₄ catalyst still remained unsatisfactory for practical applications (Fig. 5d). Apart from transition metal oxides, nanocarbon-based transition metal sulfides have also been reported as efficient ECR electrocatalysts due to their prominent catalytic features and high conductivities. A representative experiment was to load molybdenum sulphide (MoS_x) onto the surface of polyethylenimine (PEI)-modified rGO to form an rGO-PEI-MoS_x hybrid (Fig. 5e).⁷⁴ Compared to NG-Co₃O₄ and NiO/MWCNT materials, the rGO-PEI-MoS_x hybrid showed outstanding ECR activity and CO selectivity with a high CO FE of 85.1%, mainly attributed to the unique structure of MoS_x, in which each Mo atom was coordinated to six S ligands. The unique structure was similar to that of the formate dehydrogenase (FDH) enzyme which can catalyze the reduction of CO₂. Besides, the introduced PEI layer could further enhance the ECR activity of Mo-S sites and stabilize the *COOH intermediate.

4.3. Metal-free/carbon-based hybrid materials

Although both transition metal/carbon and transition metal compound/carbon catalysts exhibited high ECR catalytic performances, the high cost, sensitivity to poisoning, and difficulty of recovery largely limited their practical application in CO₂–CO conversion, which motivated researchers to pay more attention to metal-free/carbon-based hybrid ECR materials. Among metal-free/carbon-based hybrid catalysts, g-C₃N₄ is an important metal-free material which has received more and more attention in the fields of photocatalysis and electrocatalysis. Despite the low electrical conductivity of g-C₃N₄ its electrochemical properties can be revealed by complexing with other nanocarbons, giving rise to unique chemical and electronic coupling effects between them. DFT calculations revealed that g-C₃N₄ as a molecular scaffold could appropriately modify the electronic structure of the loading materials, lowering the free energy of intermediates and consequently facilitating their stabilization. Also, the adsorption capacity of CO₂ molecules can be improved because of the stronger electronegativity of N atoms compared with C atoms. As for the ECR, DCDA precursors was recently reported to form g-C₃N₄ and MWCNT composite through a polycondensation process (Fig. 6a).⁷⁵ The sp³ C–N peak was presented in the XPS C 1s spectra of g-C₃N₄/MWCNT, which indicated that the g-C₃N₄ was covalently attached onto MWCNTs. The obtained g-C₃N₄/MWCNT hybrid achieved high ECR catalytic activity with a CO FE of 60% (Fig. 6b) and over 50 h durability, in which g-C₃N₄ provided more active sites and the covalent C–N bonding in the MWCNTs enhanced the ECR activity. Similar to this work, a N-doped carbon (CN)-coated MWCNT (CN-MWCNT) nanocomposite was prepared via a pyrolysis treatment of the g-C₃N₄ and MWCNT mixture (Fig. 6c and d), and excellent ECR catalytic properties were obtained with approximately 98% FE for CO production.⁷⁶.

5. Summary and outlook

In summary, this review provides an overview of the recent Nano-CNM ECR catalyst development for electrocatalytic conversion of CO₂ into CO. Emerging Nano-CNMs include metal-free heteroatom-doped carbon, transition metal–heteroatom co- doped carbon and carbon-based hybrid materials; great progress has been made for Nano-CNMs in improving ECR selectivity, activity, and stability. For metal-free heteroatom (e.g., N, F, S, and B) doped carbon materials, the advantages of superior durability and low cost enable them to be competitive candidates to replace noble metal catalysts. Typically, the introduction of N atoms shows certain capability to regulate the distribution of charge density and increase the electron transfer rate of nanocarbons, thus boosting the overall ECR performances. For further enhancing ECR performance, the types and contents of heteroatom dopants need to be finely tuned by using different synthesis methods, such as stream etching, CVD, and pyrolysis. For transition metal–heteroatom co-doped carbon materials, further introduction of transition metal atoms could address the problems of low ECR catalytic activity of metal-free heteroatom-doped carbon materials. The partially filled d orbitals of transition metals can overcome the inherent activation barriers while enhancing reaction kinetics to improve the ECR activity. Furthermore, carbon-based hybrid materials possess a good flexibility because they can combine the advantages of different catalyst components with their various assembly possibilities.

Among discussed carbon-based catalysts in this review, we believe that the transition metal and heteroatom co-doped carbon presents the state of the art catalysts for CO₂ reduction to CO due to their encouraging activity, selectivity, and stability. However, researches on selective electrocatalysis for CO production via ECR is still in the developing stage, and more fundamental and insightful studies as well as industry design to achieve sufficiently high current density are needed.

During the research on ECR electrocatalysts, the standardization of data collection was a problem worthy of attention. Especially for Nano-CNMs, several factors could affect performance evaluation of studied electrocatalysts, such as selection of a test benchmark, the influence of the hydrodynamics of the electrochemical cell, and the effect of impurities in the electrolyte.⁸⁹ Among these factors, the impurities are easy to ignore but they potentially impact the quality of data collection. Due to its outstanding stability, Pt was chosen as the counter electrode in most three-electrode cell configurations. However, Pt was inevitably dissolved in the electrolyte and then deposited on the cathode during electrochemical cycling, thus resulting in the inaccurate evaluation of intrinsic activity of electrocatalysts. Therefore, several suitable substitutes have been proposed to avoid the effects of the Pt impurity, for example, glassy carbon was used as the counter electrode.⁹⁰ However, since Pt was used as the counter electrode in the initial ECR study, this measurement method has even been deployed for performance evaluation with precious metal-free materials, and only a few researchers realized this problem and tried to replace the Pt electrode. Therefore, if Pt is used as the counter electrode, control experiments are necessary to exclude the possible contamination of Pt to the studied catalysts.

In spite of certain breakthroughs that have been made in ECR selectivity, activity, and stability, how to balance the above three indicators is still a question. The types and contents of heteroatom dopants, the structures of active sites, and the morphologies of catalysts have significant effects on the ECR performances of Nano-CNMs. Consequently, advanced synthetic methods need to be developed to realize controllable synthesis of Nano-CNMs.

Unlike the explicit HER mechanism, CO production via ECR involving competitive reactions makes the study of the ECR mechanism more difficult in modeling and theoretical calculations. Although a few assumptions about catalysts and specific reaction processes have been proposed, a comprehensive and in-depth understanding of the ECR catalytic mechanism is still lacking. Further theoretical calculation models and methods accompanied by experimental explorations could help understand the reaction process, thus significantly improving the ECR performance.

To accurately analyze the origin of active sites and clarify the reaction mechanisms, advanced characterization can reveal the changes in catalysts during the reaction process such as operando techniques using X-ray powder diffraction, X-ray photoelectron spectroscopy, Raman spectroscopy, X-ray absorption spectroscopy, etc.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was financially supported by the National Natural Science Foundation of China (21922811, 51702284, and 21878270), the Zhejiang Provincial Natural Science Foundation of China (LR19B060002), and the Startup Foundation for Hundred-Talent Program of Zhejiang University (Y. Hou). G. Wu acknowledges the support from the National Science Foundation (CBET-1804326).

Notes and references

1. H. Rao, L. C. Schmidt, J. Bonin and M. Robert, Nature, 2017, 548, 74–77 CrossRef CAS PubMed .
2. M. Asadi, K. Kim, C. Liu, A. V. Addepalli, P. Abbasi, P. Yasaei, P. Phillips, A. Behranginia, J. M. Cerrato, R. Haasch, P. Zapol, B. Kumar, R. F. Klie, J. Abiade, L. A. Curtiss and A. Salehi-Khojin, Science, 2016, 353, 467–470 CrossRef CAS PubMed .
3. A. K. Singh, J. H. Montoya, J. M. Gregoire and K. A. Persson, Nat. Commun., 2019, 10, 443 CrossRef PubMed .
4. J. Gallagher, Nat. Energy, 2016, 1, 15024 CrossRef .
5. K. Xu, B. Sun, J. Lin, W. Wen, Y. Pei, S. Yan, M. Qiao, X. Zhang and B. Zong, Nat. Commun., 2014, 5, 5783 CrossRef CAS PubMed .
6. W. Ju, A. S. Varela, P. Strasser, A. Bagger, J. Rossmeisl, G.-P. Hao, V. Bon, S. Kaskel, A. S. Varela, I. Sinev, C. B. Roldan and C. B. Roldan, Nat. Commun., 2017, 8, 944 CrossRef PubMed .
7. I. Azcarate, C. Costentin, M. Robert and J.-M. Savéant, J. Am. Chem. Soc., 2016, 138, 16639–16644 CrossRef CAS PubMed .
8. S. Back, J. Lim, N. Y. Kim, Y. H. Kim and Y. Jung, Chem. Sci., 2017, 8, 1090–1096 RSC .
9. M. Bersani, K. Gupta, A. K. Mishra, R. Lanza, S. F. R. Taylor, H.-U. Islam, N. Hollingsworth, C. Hardacre, N. H. de Leeuw and J. A. Darr, ACS Catal., 2016, 6, 5823–5833 CrossRef CAS .
10. Z. Lei, Z. Zhi-Jian and G. Jinlong, Angew. Chem., Int. Ed., 2017, 56, 11326–11353 CrossRef PubMed .
11. S. Zhao, Z. Tang, S. Guo, M. Han, C. Zhu, Y. Zhou, L. Bai, J. Gao, H. Huang, Y. Li, Y. Liu and Z. Kang, ACS Catal., 2018, 8, 188–197 CrossRef CAS .
12. Y. Fang and J. C. Flake, J. Am. Chem. Soc., 2017, 139, 3399–3405 CrossRef CAS PubMed .
13. M. Dunwell, Q. Lu, J. M. Heyes, J. Rosen, J. G. Chen, Y. Yan, F. Jiao and B. Xu, J. Am. Chem. Soc., 2017, 139, 3774–3783 CrossRef CAS PubMed .
14. S. Back, M. S. Yeom and Y. Jung, ACS Catal., 2015, 5, 5089–5096 CrossRef CAS .
15. M. Cho, J.-W. Seo, J. T. Song, J.-Y. Lee and J. Oh, ACS Omega, 2017, 2, 3441–3446 CrossRef CAS PubMed .
16. E. L. Clark, C. Hahn, T. F. Jaramillo and A. T. Bell, J. Am. Chem. Soc., 2017, 139, 15848–15857 CrossRef CAS PubMed .
17. Y. Wu, J. Jiang, Z. Weng, M. Wang, D. L. J. Broere, Y. Zhong, G. W. Brudvig, Z. Feng and H. Wang, ACS Cent. Sci., 2017, 3, 847–852 CrossRef CAS PubMed .
18. N. Elgrishi, M. B. Chambers, X. Wang and M. Fontecave, Chem. Soc. Rev., 2017, 46, 761–796 RSC .
19. Y. He, X. Zhuang, C. Lei, L. Lei, Y. Hou, Y. Mai and X. Feng, Nano Today, 2019, 24, 103–119 CrossRef CAS .
20. F. Li, M. Xue, G. P. Knowles, L. Chen, D. R. MacFarlane and J. Zhang, Electrochim. Acta, 2017, 245, 561–568 CrossRef CAS .
21. W. Jingjie, Y. Ram Manohar, L. Mingjie, P. S. Pranav, T. Chandra Sekhar, M. Lulu, Z. Xiaolong, Z. Xiao-Dong, I. Y. Boris, L. Jun and M. A. Pulickel, ACS Nano, 2015, 9, 5364–5371 CrossRef PubMed .
22. S. Liu, H. Yang, X. Huang, L. Liu, W. Cai, J. Gao, X. Li, T. Zhang, Y. Huang and B. Liu, Adv. Funct. Mater., 2018, 28, 1800499 CrossRef .
23. T. Ma, Q. Fan, H. Tao, Z. Han, M. Jia, Y. Gao, W. Ma and Z. Sun, Nanotechnology, 2017, 28, 472001 CrossRef PubMed .
24. X. Duan, J. Xu, Z. Wei, J. Ma, S. Guo, S. Wang, H. Liu and S. Dou, Adv. Mater., 2017, 29, 1701784 CrossRef PubMed .
25. T. Ma, Q. Fan, X. Li, J. Qiu, T. Wu and Z. Sun, J. CO2 Util., 2019, 30, 168–182 CrossRef CAS .
26. A. Vasileff, Y. Zheng and S. Z. Qiao, Adv. Energy Mater., 2017, 7, 1700759 CrossRef .
27. L. Qiao, W. Zheng, H. Xu, L. Zhang and Q. Jiang, J. Chem. Phys., 2007, 126, 164702 CrossRef CAS PubMed .
28. J. Wu, M. Liu, P. P. Sharma, R. M. Yadav, L. Ma, Y. Yang, X. Zou, X. D. Zhou, R. Vajtai, B. I. Yakobson, J. Lou and P. M. Ajayan, Nano Lett., 2016, 16, 466–470 CrossRef CAS PubMed .
29. P. P. Sharma, J. Wu, R. M. Yadav, M. Liu, C. J. Wright, C. S. Tiwary, B. I. Yakobson, J. Lou, P. M. Ajayan and X. D. Zhou, Angew. Chem., Int. Ed., 2015, 54, 13701–13705 CrossRef CAS PubMed .
30. X. Q. Cui, Z. Y. Pan, L. J. Zhang, H. S. Peng and G. F. Zheng, Adv. Energy Mater., 2017, 7, 1701456 CrossRef .
31. B. Kumar, M. Asadi, D. Pisasale, S. Sinha-Ray, B. A. Rosen, R. Haasch, J. Abiade, A. L. Yarin and A. Salehi-Khojin, Nat. Commun., 2013, 4, 2819 CrossRef .
32. C. Li, Y. Wang, N. Xiao, H. Li, Y. Ji, Z. Guo, C. Liu and J. Qiu, Carbon, 2019, 151, 46–52 CrossRef CAS .
33. W. Li, B. Herkt, M. Seredych and T. J. Bandosz, Appl. Catal., B, 2017, 207, 195–206 CrossRef CAS .
34. J. Duan, S. Chen, M. Jaroniec and S. Z. Qiao, ACS Catal., 2015, 5, 5207–5234 CrossRef CAS .
35. W. Cermignani, T. E. Paulson, C. Onneby and C. G. Pantano, Carbon, 1995, 33, 367–374 CrossRef CAS .
36. T. J. Li, M. H. Yeh, W. H. Chiang, Y. S. Li, G. L. Chen, Y. A. Leu, T. C. Tien, S. C. Lo, L. Y. Lin, J. J. Lin and K. C. Ho, Sens. Actuators, B, 2017, 248, 288–297 CrossRef CAS .
37. D. Yuan, Z. Wei, P. Han, C. Yang, L. Huang, Z. Gu, Y. Ding, J. Ma and G. Zheng, J. Mater. Chem. A, 2019, 7, 16979–16983 RSC .
38. J. Xie, X. Zhao, M. Wu, Q. Li, Y. Wang and J. Yao, Angew. Chem., Int. Ed., 2018, 57, 9640–9644 CrossRef CAS PubMed .
39. W. Li, M. Seredych, E. Rodriguez-Castellon and T. J. Bandosz, ChemSusChem, 2016, 9, 606–616 CrossRef CAS PubMed .
40. N. Sreekanth, M. A. Nazrulla, T. V. Vineesh, K. Sailaja and K. L. Phani, Chem. Commun., 2015, 51, 16061–16064 RSC .
41. H. Yang, Y. Wu, Q. Lin, L. Fan, X. Chai, Q. Zhang, J. Liu, C. He and Z. Lin, Angew. Chem., Int. Ed., 2018, 57, 15476–15480 CrossRef CAS PubMed .
42. F. Pan, B. Li, W. Deng, Z. Du, Y. Gang, G. Wang and Y. Li, Appl. Catal., B, 2019, 252, 240–249 CrossRef CAS .
43. R. Wang, X. Sun, S. Ould-Chikh, D. Osadchii, F. Bai, F. Kapteijn and J. Gascon, ACS Appl. Mater. Interfaces, 2018, 10, 14751–14758 CrossRef CAS PubMed .
44. J. Xu, Y. Kan, R. Huang, B. Zhang, B. Wang, K.-H. Wu, Y. Lin, X. Sun, Q. Li, G. Centi and D. Su, Carbon, 2016, 9, 1085–1089 CAS .
45. Q. Cheng, K. Mao, L. Ma, L. Yang, L. Zou, Z. Zou, Z. Hu and H. Yang, ACS Energy Lett., 2018, 3, 1205–1211 CrossRef CAS .
46. D. M. Feng, Y. P. Zhu, P. Chen and T. Y. Ma, Catalysts, 2017, 7, 373 CrossRef .
47. J. Hao and W. Shi, Chin. J. Catal., 2018, 39, 1157–1166 CrossRef CAS .
48. M. Jia, Q. Fan, S. Liu, J. Qiu and Z. Sun, Curr. Opin. Green Sustain. Chem., 2019, 16, 1–6 CrossRef .
49. T. Wang, Q. Zhao, Y. Fu, C. Lei, B. Yang, Z. Li, L. Lei, G. Wu and Y. Hou, Small Methods, 2019, 1900210 CrossRef .
50. C. Lei, H. Chen, J. Cao, J. Yang, M. Qiu, Y. Xia, C. Yuan, B. Yang, Z. Li, X. Zhang, L. Lei, J. Abbott, Y. Zhong, X. Xia, G. Wu, Q. He and Y. Hou, Adv. Energy Mater., 2018, 8, 1801912 CrossRef .
51. C. Lu, J. Yang, S. Wei, S. Bi, Y. Xia, M. Chen, Y. Hou, M. Qiu, C. Yuan, Y. Su, F. Zhang, H. Liang and X. Zhuang, Adv. Funct. Mater., 2019, 29, 1806884 CrossRef .
52. A. S. Varela, W. Ju, A. Bagger, P. Franco, J. Rossmeisl and P. Strasser, ACS Catal., 2019, 9, 7270–7284 CrossRef CAS .
53. K. Wu, X. Chen, S. Liu, Y. Pan, W.-C. Cheong, W. Zhu, X. Cao, R. Shen, W. Chen, J. Luo, W. Yan, L. Zheng, Z. Chen, D. Wang, Q. Peng, C. Chen and Y. Li, Nano Res., 2018, 11, 6260–6269 CrossRef CAS .
54. H. B. Yang, S.-F. Hung, S. Liu, K. Yuan, S. Miao, L. Zhang, X. Huang, H.-Y. Wang, W. Cai, R. Chen, J. Gao, X. Yang, W. Chen, Y. Huang, H. M. Chen, C. M. Li, T. Zhang and B. Liu, Nat. Energy, 2018, 3, 140–147 CrossRef CAS .
55. P. Su, K. Iwase, S. Nakanishi, K. Hashimoto and K. Kamiya, Small, 2016, 12, 6083–6089 CrossRef CAS PubMed .
56. C. Yan, H. Li, Y. Ye, H. Wu, F. Cai, R. Si, J. Xiao, S. Miao, S. Xie, F. Yang, Y. Li, G. Wang and X. Bao, Energy Environ. Sci., 2018, 1204–121048 RSC .
57. X. Li, W. Bi, M. Chen, Y. Sun, H. Ju, W. Yan, J. Zhu, X. Wu, W. Chu, C. Wu and Y. Xie, J. Am. Chem. Soc., 2017, 139, 14889–14892 CrossRef CAS PubMed .
58. T. Möller, W. Ju, A. Bagger, X. Wang, F. Luo, T. Ngo Thanh, A. S. Varela, J. Rossmeisl and P. Strasser, Energy Environ. Sci., 2019, 12, 640–647 RSC .
59. J. Gu, C.-S. Hsu, L. Bai, H. M. Chen and X. Hu, J. Am. Chem. Soc., 2019, 364, 1091–1094 CAS .
60. Y. Pan, R. Lin, Y. Chen, S. Liu, W. Zhu, X. Cao, W. Chen, K. Wu, W.-C. M. Cheong, Y. Wang, L. Zheng, J. Luo, Y. Lin, Y. Liu, C. Liu, J. Li, Q. Lu, X. Chen, D. Wang and Y. Li, J. Am. Chem. Soc., 2018, 140, 4218–4221 CrossRef CAS PubMed .
61. B. Zhang, J. Zhang, J. Shi, D. Tan, L. Liu, F. Zhang, C. Lu, Z. Su, X. Tan, X. Cheng, B. Han, L. Zheng and J. Zhang, Nat. Commun., 2019, 10, 2980 CrossRef PubMed .
62. F. Pan, H. Zhang, K. Liu, D. Cullen, K. More, M. Wang, Z. Feng, G. Wang, G. Wu and Y. Li, ACS Catal., 2018, 8, 3116–3122 CrossRef CAS .
63. Y. Hou, M. Qiu, M. G. Kim, P. Liu, G. Nam, T. Zhang, X. Zhuang, B. Yang, J. Cho, M. Chen, C. Yuan, L. Lei and X. Feng, Nat. Commun., 2019, 10, 1392 CrossRef PubMed .
64. F. Pan, A. Liang, Y. Duan, Q. Liu, J. Zhang and Y. Li, J. Mater. Chem. A, 2017, 5, 13104–13111 RSC .
65. B. Dembinska, W. Kicinski, A. Januszewska, A. Dobrzeniecka and P. J. Kulesza, J. Electrochem. Soc., 2017, 164, H484–H490 CrossRef CAS .
66. S. Perez-Rodriguez, E. Pastor and M. J. Lazaro, J. CO2 Util., 2017, 18, 41–52 CrossRef CAS .
67. M. N. Hossain, J. Wen and A. Chen, Sci. Rep., 2017, 7, 1–10 CrossRef PubMed .
68. C. Z. Yuan, H. B. Li, Y. F. Jiang, K. Liang, S. J. Zhao, X. X. Fang, L. B. Ma, T. Zhao, C. Lin and A. W. Xu, J. Mater. Chem. A, 2019, 7, 6894–6900 RSC .
69. Y. X. Duan, F. L. Meng, K. H. Liu, S. S. Yi, S. J. Li, J. M. Yan and Q. Jiang, Adv. Mater., 2018, 30, 1706194 CrossRef PubMed .
70. C. Lei, Y. Wang, Y. Hou, P. Liu, J. Yang, T. Zhang, X. Zhuang, M. Chen, B. Yang, L. Lei, C. Yuan, M. Qiu and X. Feng, Energy Environ. Sci., 2019, 12, 149–156 RSC .
71. M. Jia, C. Choi, T. S. Wu, C. Ma, P. Kang, H. Tao, Q. Fan, S. Hong, S. Liu, Y. L. Soo, Y. Jung, J. Qiu and Z. Sun, Chem. Sci., 2018, 9, 8775–8780 RSC .
72. S. M. Bashir, S. S. Hossain, S. ur Rahman, S. Ahmed and M. M. Hossain, Electrocatalysis, 2015, 6, 544–553 CrossRef CAS .
73. P. Sekar, L. Calvillo, C. Tubaro, M. Baron, A. Pokle, F. Carraro, A. Martucci and S. Agnoli, ACS Catal., 2017, 7695–7703 CrossRef CAS .
74. F. Li, S.-F. Zhao, L. Chen, A. Khan, D. R. MacFarlane and J. Zhang, Energy Environ. Sci., 2016, 9, 216–223 RSC .
75. X. Lu, T. H. Tan, Y. H. Ng and R. Amal, Chem.–Eur. J., 2016, 22, 11991–11996 CrossRef CAS PubMed .
76. H. M. Jhong, C. E. Tornow, B. Smid, A. A. Gewirth, S. M. Lyth and P. J. Kenis, ChemSusChem, 2017, 10, 1094–1099 CrossRef CAS PubMed .
77. S. Guo, S. Zhao, X. Wu, H. Li, Y. Zhou, C. Zhu, N. Yang, X. Jiang, J. Gao, L. Bai, Y. Liu, Y. Lifshitz, S.-T. Lee and Z. Kang, Nat. Commun., 2017, 8, 1828 CrossRef PubMed .
78. J. Liu, C. Guo, A. Vasileff and S. Qiao, Small Methods, 2017, 1, 1600006 CrossRef .
79. K. Jiang, S. Siahrostami, T. Zheng, Y. Hu, S. Hwang, E. Stavitski, Y. Peng, J. J. Dynes, M. Gangishetty, D. Su, K. Attenkofer and H. Wang, Energy Environ. Sci., 2018, 11, 893–903 RSC .
80. T. Zheng, K. Jiang, N. Ta, Y. Hu, J. Zeng, J. Liu and H. Wang, Joule, 2019, 3, 265–278 CrossRef CAS ; Pan, B. Li, W. Deng, Z. Du, Y. Gang, G. Wang and Y. Li, Appl. Catal., B, 2019, 252, 240–249 CrossRef .
81. Y. Cheng, S. Zhao, B. Johannessen, J.-P. Veder, M. Saunders, M. R. Rowles, M. Cheng, C. Liu, M. F. Chisholm, R. De Marco, H.-M. Cheng, S.-Z. Yang and S. P. Jiang, Adv. Mater., 2018, 30, 1706287 CrossRef PubMed .
82. Z. Zhang, C. Ma, Y. Tu, R. Si, J. Wei, S. Zhang, Z. Wang, J.-F. Li, Y. Wang and D. Deng, Nano Res., 2019 DOI:10.1007/s12274-019-2316-2319 .
83. X. Qin, S. Zhu, F. Xiao, L. Zhang and M. Shao, ACS Energy Lett., 2019, 4, 1778–1783 CrossRef CAS .
84. T. N. Huan, N. Ranjbar, G. Rousse, M. Sougrati, A. Zitolo, V. Mougel, F. Jaouen and M. Fontecave, ACS Catal., 2017, 7, 1520–1525 CrossRef CAS .
85. C. Zhang, S. Yang, J. Wu, M. Liu, S. Yazdi, M. Ren, J. Sha, J. Zhong, K. Nie, A. S. Jalilov, Z. Li, H. Li, B. I. Yakobson, Q. Wu, E. Ringe, H. Xu, P. M. Ajayan and J. M. Tour, Adv. Energy Mater., 2018, 8, 1703487 CrossRef .
86. F. Yang, P. Song, X. Liu, B. Mei, W. Xing, Z. Jiang, L. Gu and W. Xu, Electrocatalysis, 2018, 57, 12303–12307 CAS .
87. Z. Chen, K. Mou, S. Yao and L. Liu, Angew. Chem., Int. Ed., 2018, 11, 2944–2952 CAS .
88. W. Zheng, C. Guo, J. Yang, F. He, B. Yang, Z. Li, L. Lei, J. Xiao, G. Wu and Y. Hou, Carbon, 2019, 150, 52–59 CrossRef CAS .
89. E. L. Clark, J. Resasco, A. Landers, J. Lin, L.-T. Chung, A. Walton, C. Hahn, T. F. Jaramillo and A. T. Bell, ACS Catal., 2018, 8, 6560–6570 CrossRef CAS .
90. J. G. Chen, C. W. Jones, S. Linic and V. R. Stamenkovic, ACS Catal., 2017, 7, 6392–6393 CrossRef CAS .

This journal is © The Royal Society of Chemistry 2019