Enhancing the Performance of Atomically Dispersed Carbon-Based Catalysts Through Metallic/Nonmetallic Elements Co-Doping Towards Oxygen Reduction

Meng Pengfei; Zhang Xiaorong; Liao Shijun; Deng Yijie

doi:10.7536/PC220214

Progress in Chemistry ›› 2022, Vol. 34 ›› Issue (10) : 2190-2201 DOI: 10.7536/PC220214

research-article

Scan to read & Share

Enhancing the Performance of Atomically Dispersed Carbon-Based Catalysts Through Metallic/Nonmetallic Elements Co-Doping Towards Oxygen Reduction

Author information +

History +

PDF

Abstract

Metal-nitrogen-carbon (M-N_x-C), possessing prominent advantages of high reactivity, high selectivity, facile synthesis, have presented potential to replace the conventional platinum-based catalysts. However, when these catalysts are used in the oxygen reduction process of fuel cells, the low active site density and insufficient durability restrict application. It is found that modulating the electronic structure and spatial configuration of the active site by doping with various metal/nonmetal elements can significantly enhance the oxygen reduction activity and stability of M-N_x-C catalysts, which has become a popular research topic in the field. This article reviews the main research works in recent years at home and abroad on the doping of various metal/nonmetal elements to enhance the performance of M-N_x-C carbon-based catalysts, including the studies on doping of metal elements and doping of nonmetal elements. The article also summarizes and points out the problems and challenges faced by M-N_x-C carbon-based catalysts, and gives an outlook on their development prospects and future research directions.

Graphical abstract

Keywords

non-platinum catalyst / carbon catalyst / oxygen reduction / co-doping / single-atom catalyst

Cite this article

Download citation ▾

Meng Pengfei, Zhang Xiaorong, Liao Shijun, Deng Yijie. Enhancing the Performance of Atomically Dispersed Carbon-Based Catalysts Through Metallic/Nonmetallic Elements Co-Doping Towards Oxygen Reduction. Progress in Chemistry, 2022, 34(10): 2190-2201 DOI:10.7536/PC220214

登录浏览全文

4963

注册一个新账户忘记密码

1 Introduction

2 M-N_x-C catalysts

2.1 The synthetic methods of M-N_x-C

2.2 The catalytic mechanism of M-N_x-C

3 Multi-metal elements co-doping M-N_x-C

3.1 Fe with other metals co-doping

3.2 co with other metals Co-doping

4 Non-metals elements co-doping M-N_x-C catalysts

4.1 N,S co-doping

4.2 N,P co-doping

4.3 Other co-doping types

5 Conclusion and outlook

1 Introduction

Since the Industrial Revolution, due to the excessive dependence of human society on fossil fuels, the content of greenhouse gases such as carbon dioxide on the earth has increased sharply, and the resulting climate problem has posed a serious threat to the sustainable development of nature and human society. The Chinese government has made a solemn commitment that "China's carbon dioxide emissions will reach their peak by 2030 and strive to achieve carbon neutrality by 2060".

At present, all kinds of clean energy development and research are widely carried out in the world, which is more and more concerned by the capital market and supported by governments^[1]. Among them, the theoretical research and technical development of new energy conversion and storage devices, such as fuel cells and metal-air batteries, have attracted much attention. However, as the core reaction of electrochemical energy conversion and storage system, the oxygen reduction reaction (ORR) with slow kinetics and high overpotential limits the energy conversion efficiency of the whole system. At present, platinum-based catalysts are considered to be the most ideal ORR catalysts, but the high price of platinum and the scarcity of platinum in the earth have resulted in the high cost of proton exchange membrane fuel cells (PEMFCs) and metal-air batteries, which hinders the large-scale commercialization process^[2,3]. Therefore, the development of efficient, stable and inexpensive non-platinum ORR catalysts has become the key to break through the related technologies^[4⇓~6].

Among the reported non-platinum catalysts, transition metal-doped carbon-based catalysts with porphyrin-like structure (M-N_x-C) are currently the most important class of non-noble metal catalysts due to their cheap and readily available raw materials and oxygen reduction activity comparable to platinum (especially in the rotating disk electrode test)^{[7⇓⇓⇓⇓~12]}. However, due to the low intrinsic activity/active site density and easy corrosion of M-N_x-C catalyst, there is still a certain gap between the performance and commercial Pt/C in the single cell test. In recent years, researchers have carried out a lot of research on improving the activity and stability of M-N_x-C catalysts, and have made a lot of progress. Based on this, this paper reviews the recent progress in the field of M-N_x-C catalysts to improve the oxygen reduction activity and stability of M-N_x-C catalysts by controlling the electronic structure and spatial configuration of active sites, including metal doping, non-metal doping and other studies, which will be of great significance for the further development of this field.

2 M-N_x-C single-atom catalyst

2.1 M-N_x-C single-atom catalyst and preparation method thereof

Porous carbon materials containing transition metals and nitrogen elements are generally defined as transition metal-nitrogen-carbon catalysts (M-N_x-C,M=Fe, Co, Mn, Cu, etc.)^[13]. M-N_x-C single atom catalysts (SACs) are obtained when isolated transition metal atoms anchored on carbon support are directly used as active sites by minimizing the geometric size to increase the number of active sites^[14]. The synthesis of M-N_x-C catalyst usually involves four main steps: preparation of precursor — primary pyrolysis at high temperature — acid washing — secondary pyrolysis at high temperature^{[15⇓⇓~18]}. The main methods for the preparation of M-N_x-C single-atom catalysts in recent years will be described below.

Hard template method. Usually, the precursor is filled into a hard template with a rigid skeleton, and the precursor is subjected to chemical or electrochemical reaction for a controlled time, and then the template is removed by high temperature carbonization and acid/alkali treatment, and finally the porous nanomaterials with ordered shape are obtained^[19]. At present, silica nanospheres and metal oxides are commonly used as hard templates^[20]^[21]. Chen et al. Used SiO₂ nanospheres as hard templates^[22]. During the experiment, the surface of SiO₂ nanospheres was modified firstly to make it negatively charged, and the Fe³⁺ was adsorbed on the surface of nanospheres by electrostatic interaction. Subsequently, the histidine molecule is strongly bound to the Fe³⁺ through coordination, thereby coating the surface of the silica nanosphere. After a series of treatments such as high temperature pyrolysis, annealing, HF pickling to remove the SiO₂ hard template, the Fe-N-C carbon nanosphere catalyst with hollow morphology was formed (the synthesis schematic diagram is shown in Figure 1). Chen et al. Used cheap whey powder as carbon source, and silica spheres and zinc chloride powder as two hard template materials^[23]. In the preparation process, it is necessary to dissolve a certain amount of whey powder in a proper amount of acetone, and then add different proportions of ZnCl₂ and SiO₂. After the mixture was stirred homogeneously, it was then heated and dried under stirring to remove the solvent. The solid material ground into fine powder is put into an alumina boat to be pyrolyzed at a high temperature, and the obtained carbonized material is firstly refluxed in a potassium hydroxide solution to remove SiO₂. Washing with HCl solution and a large amount of deionized water to remove Zn residue. By adjusting the ratio of the two hard templates, the hierarchical pore structure of carbon materials was regulated and optimized.

It is worth noting that in the preparation of M-N_x-C catalysts by hard template method, some strong acids/bases are often needed to remove the template, which not only leads to a long and complicated preparation process,Moreover, it is easy to cause the collapse of the nanostructure, and it is difficult to replicate the porous structure of the template well, while the soft template method can not only provide an orderly three-dimensional structure, but also avoid the tedious steps of removing the template^[24]. At present, the commonly used soft templates mainly include surfactants and polymers, such as surfactants P123, F127 and other block copolymers^[25,26]. In the synthesis process, firstly, aggregates or micelles with certain specific structural characteristics are formed between molecules through intermolecular or intramolecular interactions (hydrogen bonds, electrostatic, etc.),These aggregates or micelles are subsequently used as templates to allow inorganic substances to be deposited on their surface or inside by various synthetic methods, resulting in particles with different morphologies and porous structures^[27]. For example, Lee et al. Reported a simple soft template method to synthesize ordered mesoporous carbon materials with FeN_x/C active sites by direct incorporation of Fe-CNAR precursor (complex of Fe²⁺ with 1,10-phenanthroline) during block copolymer self-assembly^[28]. In the preparation process, they first dissolved F127 in a co-solvent of ethanol and HCl to obtain a polymer solution, and then mixed FeCl₂·4H₂O and 1,10-phenanthroline in a solvent of ethanol and HCl to obtain an iron complex solution. Finally, the resol resin, the tetraethyl orthosilicate and the iron complex solution are added to the polymer solution. After stirring for 1 H, the solution was poured into a dish. After being evaporated at 50 deg C and dried at 100 deg C, the obtained polymer nanocomposite is subjected to high-temperature treatment, acid washing and secondary pyrolysis to obtain the prepared carbon-based catalyst. It is found that the formation of a stable micellar structure is the key to the soft template method. However, there are relatively few compounds that can be used as surfactants, and it is difficult to effectively control the morphology and size of the product by the soft template method. Recently, the self-templated synthesis of M-N_x-C catalysts using MOFs has also received a lot of attention because of their regular three-dimensional structure, high specific surface area, and highly ordered channels^[29⇓~31]. For example, Xing et al. Used the ordered pore structure of ZIF-8 materials with CrCl₃·6H₂O as the chromium source to synthesize a single-atom chromium catalyst with a Cr-N₄ configuration after pyrolysis^[32]. The synthesis diagram is shown in Figure 2. In the preparation process, CrCl₃·6H₂O and ZIF-8 were dispersed in ethanol solution, chromium ions were adsorbed into the microporous structure of ZIF-8 during the ultrasonic process, and Cr was captured/fixed by the CN_x structure in the carbon support during the pyrolysis process to form a CrN₄ structure.

Although the preparation methods of single-atom catalysts have been widely studied, it is still a great challenge to achieve stable industrial production of single-atom catalysts in small batches (gram scale) or even large batches (kilogram scale). Recently, the vapor deposition method has provided a new idea to solve the large-scale industrial preparation of single-atom catalysts^[33]. M-N_x-C catalyst prepared by CVD method can be roughly divide into two types: 1) porous nanomaterials capture volatile metal compounds, I. E. volatile compounds containing metal atoms (Fe, Mn, Cu, etc.) migrate to that surface of porous nanomaterials through the gas phase under the action of gas flow^[34]^[35]^[36]; 2) Metal oxides (CaO, ZnO, etc.) doped with transition metals such as Fe are used as templates, and the precursors are deposited on the templates as nitrogen and carbon sources. During the vapor deposition process, controllable carbonization and in-situ heteroatom doping are beneficial to anchoring single metal atoms to achieve high metal loading and effective exposure of active sites M-N_x^[33,37]. Wu et al. Used the coordination effect of NH₃ and copper atoms on the surface of copper foam to directly convert the bulk metal into a single-atom catalyst^[36]. During the preparation, copper foam and ZIF-8 were placed in a porcelain boat. First, ZIF-8 was pyrolyzed in an argon atmosphere at high temperature to form pyrolytic C-ZIF-8 at defect sites. In ammonia atmosphere, ammonia molecules separate the copper atoms on the surface from the copper foam, forming volatile Cu(NH₃)_x molecules based on strong Lewis acid-base interaction. Then, the Cu(NH₃)_x was captured by the nitrogen-rich carbon support, and the Cu-SAs/N-C single-atom catalyst was produced after pyrolysis at high temperature. Wu et al. Prepared Fe-doped ZnO substrate by mixing Fe(NO₃)₃·9H₂O with Zn(Ac)₂·2H₂O via sol-gel method^[38]. Then, 2-methylimidazole and Fe-ZnO nanosheets were placed in two temperature zones, respectively. The evaporated gaseous 2-MeIm flowed with argon and was subsequently deposited on Fe-ZnO nanosheets, resulting in a porous carbon catalyst with highly dispersed single-atom Fe.

2.2 Catalysis Principle of M-N_x-C Single Atom Catalyst

Oxygen reduction reaction is a multi-electron, multi-proton and multi-step complex reaction. Under different electrolyte conditions, the reduction path and reduction products of ORR are also different. According to different reaction pathways, the oxygen reduction reaction can be divided into two pathways: four-electron reduction and two-electron reduction^[39]. The two-electron path often involves the generation of hydrogen peroxide, which is reduced to free radicals under the Fenton reaction, and these free radicals attack the proton exchange membrane, resulting in the degradation of the membrane and a sharp decline in battery performance. Therefore, rational design of the catalyst to catalyze oxygen reduction according to the four-electron path, avoiding the generation of hydrogen peroxide, can significantly improve the stability of the catalyst. The key to rational design of catalysts is to study the catalytic mechanism of active sites. It is now generally accepted that the catalytic sites of a M-N_x-C catalyst include a nonmetal part and a metal part, and the metal-containing part MN_x is considered to be the source of the high activity of the M-N_x-C catalyst compared with the nonmetal part (C-N_x). The typical metal active site is composed of non-noble metal atom M (mainly Fe, Co, Ni, Cu, Zn, etc.) and its coordinated nitrogen atom or other atoms^[40,41]. The coordination structure of the MN_x site may vary with different synthesis methods and precursors. Shui et al. Prepared Fe-N-C catalysts with the same density of active sites but different coordination numbers of N atoms (from 1 to 5) by controlling the pyrolysis temperature^[42]. FeN₄ has the best ORR activity and excellent PEMFC performance as evidenced by electrochemical characterization and theoretical calculations. In addition, the oxygen reduction activity of the catalyst is closely related to the type of metal element in the MN_xstructure. Nørskov et al., Zagal et al., Qiao et al. Respectively established the oxygen reduction activity volcano diagram of different MN₄structures through DFT calculations and a series of experiments, as shown in Fig. 3^[43]^[44]^[45]. It was found that the adsorption of O₂ and the dissociation of OOH^- play a crucial role in the process of oxygen reduction catalyzed by carbon-based catalysts. Peng et al. Prepared a series of nitrogen and transition metal Co-doped catalysts (transition metal = Mn, Fe, Co, Ni, Cu) using melamine, aniline and a series of transition metal chlorides, and proved through experiments that the activity of different metal center M — N — C catalysts is Fe > Co > Cu > Mn > Ni^[46]. Obviously, the moderate oxygen adsorption promoting O = O dissociation is the key to obtain high activity. On the other hand, researchers have been trying to reveal the dynamic transformation process of the active site and reaction intermediate in the ORR reaction.

Liu et al. Used a synchrotron radiation-based characterization technique to in situ detect the dynamic change of the iron-active structure OH-Fe-N₂ with unsaturated coordination during the reaction^[47]. As shown in Fig. 4, under open circuit voltage, the pre-adsorption of hydroxyl groups occurs in the active center, forming a HO-Fe-N₄ coordination structure. When the oxygen reduction potential of 0.90 V is applied, the longer two-coordinate Fe-N path is broken due to the pulling effect of OH, and the coordinatively unsaturated HO-Fe-N₂ catalytic active structure is formed. At the same time, the dynamic adsorption and desorption of the key reaction intermediate (* OOH) occurs on the active site. The active structure of HO-Fe-N₂ remains stable during the catalytic process, which ensures the efficient oxygen reduction reaction.

It is worth noting that some progress has been made in the study of the reaction mechanism of oxygen at the active site and the dynamic transformation process of the reaction intermediate by using noble metal (Pt) as a model catalyst.Most of the oxygen reduction reaction intermediates are short lived, very low in content, and have a complex electrochemical environment, so it is still a challenge to accurately clarify the catalytic mechanism of ORR so far^[48,49]. In particular, for M-N_x-C catalysts, the complexity of the active site structure and the dynamic process that the active site undergoes during the reaction increase the difficulty of resolving the catalytic mechanism. Therefore, there is an urgent need to develop advanced in situ characterization and fine structure characterization techniques to study the reduction process of oxygen at the active site.

3 Multi-metal co-doped M-N_x-C single-atom catalyst

In recent years, the research results show that the ORR activity of metal-nitrogen-carbon catalysts with single transition metal type is comparable to that of commercial platinum carbon in alkaline medium, but there is still a certain gap in the performance of acidic medium^[50]. The possible reason is that the metal loading of these catalysts is low, and the isolated metal atoms are easy to aggregate and dissolve out during the catalytic process, resulting in insufficient activity and stability of the catalysts^[51,52]. Recently, it has been reported that the introduction of a second component metal element into a single transition metal M-N_x-C can improve the activity and stability of the M-N_x-C by changing the geometry and chemical environment of the single metal active site^[53]. Among many transition metal (such as Fe, Co, Cu, Ni, Mn, etc.) doped carbon-based catalysts, Fe and Co-centered M-N_x-C catalysts have been widely studied because of their optimal oxygen reduction activity^[44,54,55]. It is summarized as follows.

3.1 Co-doping of iron with other metals

Sun et al. Synthesized a single-atom electrocatalyst with FeNi-N₆ active sites^[56]. The percentages of Fe and Ni in the catalyst detected by ICP-OES were 1.45 wt% and 1.47 wt%, respectively, and the total metal content was close to 3 wt%. The introduction of nickel metal sites provides a large number of catalytic active sites, and at the same time, the H₂O₂ yield of the catalyst in the ORR process is only 1% to 4%. Under acidic conditions, the FeNi-N₆ exhibits excellent durability, and the half-wave potential decay is only 12 mV after 5000 cycles. Chen et al. Prepared the electrocatalyst Fe, Mn-N/C with two metal M-N_x active sites by one-step pyrolysis of a metal-organic framework (ZIF-8)^[57]. The LSV test results showed that the catalyst containing Fe-N_x and Mn-N_x sites had more excellent ORR activity than the catalyst with only Fe-N_x active sites. The synergetic interaction between Fe and Mn ions effectively lowers the energy barrier of the ORR process, and first-principles calculations show that Fe/Mn-N/C requires less energy in the protonation process from O^* to OH^* in the ORR process compared with Fe-N/C. The catalyst exhibited superior ORR performance with a half-wave potential of 0.904 V and a current density of 33.33 mA·cm², which is 4.9 times higher than that of 20%Pt/C(6.76 mA·cm²).

Chen et al. Simply prepared the precursor by mixing iron-copper metal salts, PVP, and nano-sized CaCO₃ particles, and after a series of treatments, an efficient catalyst with the coexistence of FeN₄ and CuN₄ single atoms was obtained^[58]. The electrochemical test showed that the activity of Fe/Cu SA with bimetallic site was significantly better than that of Fe SA or Cu SA with single site, indicating the promotion effect of Cu site on the ORR activity of Fe site. From the electron local function (ELF) images in the literature, it can be seen that the electron density of iron atoms is obviously increased in the Fe/Cu SA structure. Bader charge calculations show that the charge transfer number of Fe decreases from − 0.96 to − 0.88 e after the introduction of Cu. This strongly demonstrates the key role of the neighboring sites in tuning the electronic structure of the iron atom, i.e., preventing the charge density of the iron atom from falling, and thus enhancing the electron density around the iron atom. Higher electron density promotes the adsorption and activation of O₂ molecules.

Shui et al. Used density functional theory to study the effect of Zn on the electrocatalytic performance of Fe-N-C catalyst, and the calculation results showed that the Fe-Zn-N_xstructure is thermodynamically stable, and the second metal zinc atom weakens the binding strength between Fe-O atoms and reduces the ORR activation energy barrier by modifying the electronic structure of Fe-N-C, thus improving the catalytic activity of Fe-N C^[59].

In addition, Shui et al. Also used platinum atoms to modify the Fe-N-C catalyst at the atomic scale, and they grafted a single atom of Pt to the active center of Fe-N₄ through a bridging oxygen molecule to obtain a new active site Pt₁-O₂-Fe-N₄^[60]. The catalyst exhibited slightly enhanced ORR activity in acidic medium and significantly enhanced stability in the full cell. Accelerated voltage cycling tests were performed on Pt₁@Fe-N-C and Fe-N-C between 0.3 and 1.1 V. As shown in Fig. 5, the E_1/2 of Pt₁@Fe-N-C decayed by only 12 mV after 10 000 cycles, which was much smaller than the 36 mV attenuation of Fe-N-C. These results suggest that grafting Pt₁O₂ onto Fe₁N₄ molecules is an effective strategy to improve the durability of Fe-N-C electrocatalysts. This enhanced stability benefits from the protection of Fe sites from hydrogen peroxide destruction by the Pt₁-O₂-structure. The synergistic effect between the grafted atom Pt and the original active center Fe was further demonstrated. This study also provides a new idea for improving and expanding the performance of carbon-based metal catalysts through single-atom and single-atom grafting technology.

3.2 Cobalt co-doped with other metal

Compared with Fe-N-C catalyst, Co-N-C catalyst has lower Fenton reaction activity. Recently, it has been found that the catalytic activity of the metal center can be further enhanced by constructing Fe and Co bimetallic sites. Xing et al. Used Fe-ZnCo-ZIF-8 precursors to prepare carbon-based catalysts in which both Fe and Co are atomically dispersed and have Fe-Co diatomic active sites^[61]. A spontaneous electron-withdrawing OH ligand is introduced into the active site of the catalyst, which lowers the d-band center of the Fe site and forms a triangular Fe-Co-OH coordination with the Fe-Co diatomic active site. At the Fe-Co-OH active site, the binding degree of the ORR intermediate is weakened, the O — O bond cleavage is accelerated, and the ORR activity is enhanced. Zhang et al. Also reported a general route for large-scale production of Fe-Co-NC bimetallic atom catalysts using formamide as a precursor^[62].

Sun et al. Prepared a Zn/Co-N-C catalyst with Zn/Co bimetallic sites by pyrolyzing the precursor, and the catalyst showed excellent catalytic activity under both acidic and alkaline conditions, as shown in Fig. 6, and the half-wave potential was as high as 0.796 V vs RHE in V vs RHE^[63]. When the catalyst was used in H₂-O₂PEMFC, its peak power density reached 705 mW·cm^-2.

Hu et al. Obtained cobalt-nickel-supported nitrogen-doped carbon nanotubes (CoNi-NCNTs) by pyrolysis of cobalt-nickel metal salts and melamine resin on carbon nanotubes^[64]. The rotating ring-disk electrode technique showed that the oxygen reduction reaction on the Co/Ni-NCNT catalyst proceeded in a nearly ideal four-electron transfer path, and the hydrogen peroxide yield was less than 5%, which was far ahead of the Co-NCNT catalyst. They proposed that the introduction of Ni can coordinate with the Co-N-C structure to form a Co/Ni-N-C structure, which will greatly improve the number of electron transfer and the stability of active centers.

Amal et al. Prepared a copper-cobalt binary catalyst, and found that the existence of the copper monoatomic site near the cobalt monoatomic site changed the electronic structure of the cobalt monoatomic active site, which was conducive to the occurrence of oxygen reduction reaction, thus greatly improving the apparent reactivity^[65]. In addition, the cobalt-copper binary catalyst has lower hydrogen peroxide selectivity than the cobalt single-atom catalyst, which effectively improves the reaction stability and activity of the catalyst, and makes it have better activity and stability than the commercial 20% Pt/C. The theoretical and experimental results show that the electron-rich cobalt atom is the active site for oxygen reduction, while the copper atom mainly acts as the inhibitor of hydrogen peroxide.

4 Nonmetal element doped M-N_x-C single-atom catalyst

In addition to metal doping, the electronic structure of planar M-N₄ can also be effectively controlled by doping non-metallic elements (S, P, B, etc.) To break the symmetry of their structure and optimize the adsorption strength of intermediates to improve the ORR activity.

4.1 N and S co-doping

In nitrogen and sulfur co-doped carbon-based catalysts, sulfur with larger atomic radius and lower electronegativity (χ = 2.58) than nitrogen can induce electron spin redistribution and polarization configuration transformation, and enhance the ORR activity of carbon-based catalysts.

Chen et al confirmed that S doping could induce the transformation of spin polarization configuration of Fe through Fe Mössbauer spectroscopy and electron paramagnetic resonance spectroscopy, and the lower spin state after S doping promoted the desorption process of OH^*, improved the selectivity of the catalyst, and promoted the whole ORR process to be a four-electron transfer path^[66]. Our group prepared a Fe, N, S co-doped UIO-66—NH₂ derived carbon catalyst (Fe/N/S-PC) by double solvent diffusion pyrolysis method^[67]. Sulfur doping increases the number of nitrogen active sites, and the formed thiophene structure changes the electronic structure of the carbon skeleton. In addition, compared with the catalyst without sulfur doping, the layered porous structure of the derived carbon material has higher specific surface area and mesoporosity.As shown in fig. 7, Fe/N/S-PC exhibits superior activity to commercial Pt/C at both acidic and alkaline conditions, and has a greater current density. The excellent ORR performance of Fe/N/S-PC is attributed to its high mesoporous porosity, highly active active sites, and large specific surface area.

The doping of sulfur atoms into different positions around the active site of M-N₄ will make the electronic structure of the catalyst and the adsorption free energy of the ORR step different, which will lead to different degrees of activity improvement of carbon-based catalysts. Sun et al., based on the formamide liquid-phase synthesis system, precisely realized the precise doping of S elements at different positions on the periphery of the Fe-N₄ by designing molecular-scale reactions (as shown in Fig. 8A)^[68]. The DFT simulation results show that the d-band gap of Fe in Fe-NS₂-C is much narrower than that of Fe atoms in other Fe-NS-C, which is attributed to the improvement of the local electronic conductivity of the Fe-N₄ part by the adjacent S doping. In addition, combined with electron microscopy and synchrotron radiation techniques, they found that the atomic structure of the active site of the catalyst was Fe(N₃)(N-C-S) configuration, and the Fe-NS-C material doped with Fe-N₄ near sulfur showed excellent catalytic activity and stability for oxygen reduction. As shown in Fig. 8 B, the onset potential of Fe-NS-C is about 1.09 V, which exceeds the onset potentials of Pt/C and Fe-N-C by 90 and 80 mV, respectively.

In addition, there are two ways to affect the electronic structure of metal active centers by non-metallic element doping: (1) using the long-range force of heteroatoms on the carbon substrate to affect the electronic structure^[67,69]; (2) The electronic structure of the metal atom is directly tuned by changing the coordination atom species of the metal active center^[70⇓~72]. Therefore, in addition to the strategy of incorporating sulfur atoms into the carbon skeleton, some recent studies have attempted to use sulfur atoms to participate in the coordination of metal sites and replace nitrogen atoms in order to achieve the effect of regulating the electronic structure of metal sites.

Zhu et al. Synthesized N and S co-doped Fe-N-C materials by one-step pyrolysis of laver precursor coated with FeCl₃^[72]. After further analysis of the coordination environment of Fe atom by synchrotron radiation, it is found that the ratio of Fe-N to Fe-S local coordination bonds is close to 3 ∶ 1, suggesting that the active site of the catalyst is a Fe-N₃S₁. The electrochemical test results show that the sulfur-doped catalyst exhibits better ORR activity and four-electron selectivity than the catalyst without S element. The elevated ORR activity is attributed to the optimization of the charge and spin distribution of the Fe-N_xstructure by sulfur doping.

In addition to single metal site catalysts, heteroatom doping is also often applied in carbon-based catalysts with bimetallic sites. Zhao et al. Synthesized sulfur-doped modified Zn,Co-N_x-C-S_y bimetallic catalyst by a facile simultaneous coordination pyrolysis method using chitosan as carbon source, sodium diethyldithiocarbamate as sulfur source, cobalt acetate as cobalt source and zinc chloride as zinc source^[51]. The experimental results demonstrate that this active site has two advantages in ORR catalysis: (1) the Zn-Co bimetallic site enhances binding to O₂, promotes O — O bond activation, and lowers the dissociation barrier of the O — O bond in the ^*OOH+e^-→^*O+OH^- step; (2) S-doping can regulate the charge around Zn and Co active centers and enhance the interaction with oxygen-containing substances by reducing the free energy of the ^*O₂+e^-+H₂O→^*OOH+OH^-step. Electrochemical tests showed that the as-prepared Zn,Co-N_x-C-S_y exhibited excellent electrocatalytic performance, with a half-wave potential 67 mV higher than that of commercial Pt/C (0.893 V vs 0.826 V), and chronoamperometry showed that the sample had excellent stability (∼ 4.4% current loss after 20 000 s test).

4.2 N and P co-doping

Compared with the nitrogen atom, the phosphorus atom has a larger atomic radius and low electronegativity. When P replaces part of the N atom to coordinate with the metal, the asymmetric N/P coupling caused by the difference in physical and chemical properties between P and N will optimize the charge distribution and electronegativity of the active center, and also improve the performance of the catalyst.

Chen et al. Developed a nitrogen and phosphorus two-coordinated single-atom iron-carbon based catalyst^[73]. As shown in Fig. 9, the PPy hydrogel was formed by the polymerization reaction of pyrrole, and phytic acid and FeCl₃ were introduced as phosphorus and iron sources. The hydrogels were then pyrolyzed in nitrogen and etched in sulfuric acid to obtain N-P bi-coordinated iron active site composite carbon nanosheets. By means of various characterizations and simulation calculations, it is proved that the Fe atom in the catalyst is coordinated by three nitrogen atoms and one phosphorus atom, forming a square planar configuration. The experimental and theoretical calculations show that the N and P doubly coordinated iron sites are beneficial to the adsorption/desorption process of oxygen intermediates, which can accelerate the kinetics of oxygen reduction reaction and achieve high catalytic activity.

Zhang et al. Studied the principle that distant phosphorus atoms improve the catalytic activity of edge-type FeN₄ through the delocalization of remote electrons^[74]. DFT calculations are performed to study the local electron distribution of the edge-type FeN₄P₂, and it is found that the defect accommodation of the edge-type FeN₄P₂structure is more favorable for the four-electron ORR pathway compared with the in-plane FeN₄P₂. The actual PEMFC performance test also showed that the catalyst had good activity and stability, and the cell exhibited a high open-circuit voltage of 0.98 V and a maximum power density of 0.625 W/cm² at H₂/O₂ and 80 ° C, as shown in Fig. 10.

Zhang et al. Developed a novel cross-linked polyphosphazene sphere, and obtained mesoporous carbon nanospheres with N/P coordinated with Co to form Co-N₂P₂ sites after carbonization^[75]. Different from the traditional method of introducing heteroatoms by gridding before pyrolysis, the heteroatoms such as N and P provided by the cross-linked copolymer can directly coordinate with the metal to obtain uniform active sites, so that the Co content in the catalyst is as high as 10.12%. As shown in Fig. 11, the DFT calculation results show that compared with the Co-N₄ site, the O₂ molecule has a lower adsorption free energy on Co-N₂P₂, indicating that Co-N₂P₂ can bind O₂ more effectively to catalyze ORR.

Li et al. Used a phosphorus-doped metal-organic framework (MOF) as a porous support to prepare a single-atom catalyst with Co₁-PNC and Ni₁-PNC planar configurations by multistep synthesis, which showed excellent ORR catalytic activity in alkaline solution, benefiting from the moderate reduction of oxygen intermediate binding interactions by the N/P-coordinated Co/Ni sites^[76]. Electrochemical tests in 0.1 mol/L KOH solution saturated with O₂ showed that Co₁-PNC/Ni₁-PNC exhibited the optimal ORR onset potential and half-wave potential (E_onset=1.00 V,E_1/2=0.88 V), which were superior to other sample performances in the experiment. In addition, the Tafel slope value of Co₁-PNC/Ni₁-PNC (63 mV/dec) is lower than that of Pt/C (87 mV/dec) and Co₁-NC/Ni₁-NC(93 mV/dec), and the lower Tafel slope implies excellent ORR kinetics.

4.3 Co-doping of other non-metallic element

In addition to the N/S and N/P co-doping methods mentioned above, in recent years, N/Cl, N/O and N/B co-doping methods have also been explored for application in M-N_x-C catalysts.

Li et al. Prepared sulfur and chlorine co-doped FeCl₁N₄/CNS catalyst by thermal migration method^[77]. In the active site, the Cl and N atoms are located in the first coordination shell of the metal site, while the S atom is far away from the metal active site. The obtained catalyst exhibited very excellent ORR activity. Electrochemical measurements showed that the half-wave potential of Pt/C was 0. 921 V in 0. 1 mol/L potassium hydroxide solution, which was 79 mV higher than that of commercial Pt/C (0. 842 V). This work demonstrates experimentally and by DFT that highly improved ORR can be successfully achieved by simultaneously tuning the Fe active site for short-range interactions with chlorine and long-range interactions with sulfur.

Unlike the electron-rich P and S atoms, the B atom lacks electrons, but B doping can also change the electroneutrality of the sp² hybridized carbon, resulting in positively charged sites favorable for oxygen adsorption^[78]. Xu et al. prepared a carbon nanosheet catalyst codoped with N, B and cobalt single atoms through a soft template self-assembly pyrolysis method, and the preparation schematic diagram is shown in Figure 12A^[79]. The excellent ORR activity was exhibited under alkaline conditions with a half-wave potential (E_1/2) of 0.83 V due to the O₂ adsorption on the B-atom doped Co-N-C site, which lowers the reaction energy barrier of oxygen molecules on the active site (Fig. 12 B).

Inspired by the structure and function of Mn coenzyme factor in biological enzymes, Chen et al. Synthesized a series of metal coordination structures with Mn-N₁O₃, Mn-N₂O₂, Mn-N₃O₁^[70]. Density functional theory calculations revealed that the catalyst containing the Mn-N₃O₁structure exhibited the fastest ORR kinetics. The nitrogen-oxygen double coordination structure adjusts the d electron orbit of the Mn atom in the catalyst, and the synthesized electrocatalyst has higher ORR catalytic performance. Chen et al. Also found that the formation of Ca coordination by N and O atoms in the carbon matrix adjusts the p-orbital electronic structure of Ca atom, which makes the number of States (PDOS) of the highest peak Ca atom move close to the Fermi level, thus promoting the adsorption and electron transfer of ORR intermediates^[80].

5 Conclusion and prospect

In this paper, the main research work on improving the performance of carbon-based catalysts for atomic-level dispersed M-N_x-C by doping various metal/nonmetal elements at home and abroad in recent years is systematically reviewed. Obviously, the oxygen reduction performance of atomically dispersed M-N_x-C catalysts has made great progress and shown attractive prospects through the unremitting efforts of many researchers, but there are still many challenges to promote their practical application in fuel cell membrane electrode assembly.

First, the monoatomic active center faces the problem of how to improve the ORR selectivity and reduce the yield of hydrogen peroxide^[30]. The catalytic process involves the breaking and formation of chemical bonds^[81]. Because of the spatial configuration of the active site of M-N_x-C catalyst, oxygen is more likely to be adsorbed on the site in the superoxide state, which is more likely to produce hydrogen peroxide in the subsequent reaction. If the number of metal atoms in the active site is increased from a single atom to more than two atoms to form a M₂-N-C or even a M_x-N-C structure, the presence of more than two atoms in the active site breaks the O — O bond, allowing oxygen to be adsorbed on the site in the peroxy state, which can improve the four-electron reaction selectivity^[30,82]. Therefore, in the future, M-N_x-C catalysts may be developed in the direction of diatomic, triatomic or cluster M-N_x-C catalysts. It may be convenient to synthesize M_x-N-C catalysts by using the characteristics of polynuclear metal coordination complexes or clusters containing multiple metal atoms.

Second, currently synthesized single-atom catalysts have low metal loading and low active site density, which makes it difficult for the mass specific activity of the catalyst to reach the level of industrial application. In the membrane electrode assembly, the catalyst loading of M-N_x-C is usually 1~4 mg·cm^-2, which is much higher than that of commercial Pt/C catalyst (usually 0.2~0.5 mg·cm^-2), resulting in a thicker catalyst layer (up to 100 μm), which significantly increases the mass transfer resistance and the difficulty of water and thermal management^[20,83]. At present, some people use surfactants, a variety of ligand combinations and other means to adjust the metal loading, adjust the pore structure of the catalyst to expose more active sites, but there is still a distance from the goal of commercial application^[84]^[11].

Thirdly, it is difficult to accurately characterize the structure of atomic-level active centers, and the theory of explaining monatomic catalytic centers is still not perfect. Although there are transmission electron microscopy, spherical aberration electron microscopy, synchrotron radiation and other instruments for the characterization of catalysts, it is difficult to distinguish the subtle differences in the coordination environment. Especially in heterogeneous catalytic reactions, the dynamic structural transformation of single-atom catalysts is very common. With this in mind, it is important to employ more time-resolved techniques, such as in situ XAS, XPS, IR, and Raman spectroscopy, to follow the reaction intermediate state and detect the transformation of the active site under the reaction conditions^[85]. Therefore, in order to better explore the catalytic mechanism of atomic-scale structure, it is very important to develop and study high-precision instruments.

Fourthly, the synthetic methods of carbon-based single-atom catalysts still need to be further developed. The catalytic performance of carbon-based monatomic catalysts is closely related to the configuration of the active site, and the number of coordinating atoms, the type of coordinating atoms, and the type of metal atoms are all important influencing factors. At present, although some groups have realized the precise control of coordination elements and types in M-N_x-C, and the synthesis of atom-pair M-N_x-C has been reported sporadically, there is still a lack of relatively universal methods, and there is still a long way to go for the rational design of single-atom M-N_x-C catalysts^[70,73]^[86,87]. In the future, artificial intelligence and deep learning technology may be applied to the rational design of catalysts, enabling scientists to design catalyst structures more pertinently.

References

Publishing order | Descend order by publishing year | Descend order by cited within

[1]	Liu J G, Mooney H, Hull V, Davis S J, Gaskell J, Hertel T, Lubchenco J, Seto K C, Gleick P, Kremen C, Li S X. Science, 2015, 347(6225): 1258832.

[2]	Nørskov J K, Bligaard T, Rossmeisl J, Christensen C H. Nat. Chem., 2009, 1(1): 37.

[3]	Xia W, Mahmood A, Liang Z B, Zou R Q, Guo S J. Angew. Chem. Int. Ed., 2016, 55(8): 2650.

[4]	Debe M K. Nature, 2012, 486(7401): 43.

[5]	Fu J, Cano Z P, Park M G, Yu A P, Fowler M, Chen Z W. Adv. Mater., 2017, 29(7): 1604685.

[6]	Wang Z L, Xu D, Xu J J, Zhang X B. Chem. Soc. Rev., 2014, 43(22): 7746.

[7]	Ren Q, Wang H, Lu X F, Tong Y X, Li G R. Adv. Sci., 2018, 5(3): 1700515.

[8]	Xu J H, Xia C L, Li M, Xiao R. ChemElectroChem, 2019, 6(22): 5735.

[9]	He Q, Chen X H, Jia F Q, Ding W, Zhou Y Y, Wang J, Song X Y, Jiang J X, Liao Q, Li J, Wei Z D. ChemistrySelect, 2019, 4(27): 8135.

[10]	Qin X L, Huang Y, Wang K, Xu T T, Wang Y L, Dong W H. Electrochimica Acta, 2019, 322: 134745.

[11]	Jiao L, Wan G, Zhang R, Zhou H, Yu S H, Jiang H L. Angew. Chem. Int. Ed., 2018, 57(28): 8525.

[12]	Deng Y J, Luo J M, Chi B, Tang H B, Li J, Qiao X C, Shen Y J, Yang Y J, Jia C M, Rao P, Liao S J, Tian X L. Adv. Energy Mater., 2021, 11(37): 2101222.

[13]	Tang T, Ding L, Jiang Z, Hu J S, Wan L J. Sci. China Chem., 2020, 63(11): 1517.

[14]	Liu L C, Corma A. Chem. Rev., 2018, 118(10): 4981.

[15]	Li S P, Zhang G, Tu X M, Li J H. ChemElectroChem, 2018, 5(4): 701.

[16]	Lin Y X, Yang L, Zhang Y K, Jiang H L, Xiao Z J, Wu C Q, Zhang G B, Jiang J, Song L. Adv. Energy Mater., 2018, 8(18): 1870087.

[17]	Wang F L, Yang X D, Dong B X, Yu X, Xue H G, Feng L G. Electrochem. Commun., 2018, 92: 33.

[18]	Ban J J, Wen X H, Xu H J, Wang Z, Liu X H, Cao G Q, Shao G S, Hu J H. Adv. Funct. Mater., 2021, 31(19): 2010472.

[19]	Malgras V, Tang J, Wang J, Kim J, Torad N L, Dutta S, Ariga K, Hossain M S A, Yamauchi Y, Wu K C W. J. Nanosci. Nanotechnol., 2019, 19(7): 3673.

[20]	Wan X, Liu X F, Li Y C, Yu R H, Zheng L R, Yan W S, Wang H, Xu M, Shui J L. Nat. Catal., 2019, 2(3): 259.

[21]	Jin X X, Jiang Y, Hu Q, Zhang S H, Jiang Q K, Chen L, Xu L, Xie Y, Huang J H. RSC Adv., 2017, 7(89): 56375.

[22]	Chen Y F, Li Z J, Zhu Y B, Sun D M, Liu X E, Xu L, Tang Y W. Adv. Mater., 2019, 31(8): 1806312.

[23]	Yan S Y, Yu Z X, Liu C, Yuan Z W, Wang C J, Chen J S, Wei L, Chen Y. Chem. Asian J., 2020, 15(12): 1881.

[24]	Petkovich N D, Stein A. Chem. Soc. Rev., 2013, 42(9): 3721.

[25]	Kang H Q, Sui L N, Wang Y T, Zhang S, Dong H Z, Yu L Y, Dong L F. J. Electrochem. Soc., 2017, 164(12): A2328.

[26]	Yu L, Yu X Y, Lou X W D. Adv. Mater., 2018, 30(38): 1800939.

[27]	Lee J, Kim J, Hyeon T. Adv. Mater., 2006, 18(16): 2073.

[28]	Mun Y, Kim M J, Park S A, Lee E, Ye Y, Lee S, Kim Y T, Kim S, Kim O H, Cho Y H, Sung Y E, Lee J. Appl. Catal. B Environ., 2018, 222: 191.

[29]	Liu J L, Zhu D D, Guo C X, Vasileff A, Qiao S Z. Adv. Energy Mater., 2017, 7(23): 1700518.

[30]	Yin P Q, Yao T, Wu Y E, Zheng L R, Lin Y, Liu W, Ju H X, Zhu J F, Hong X, Deng Z X, Zhou G, Wei S Q, Li Y D. Angew. Chem. Int. Ed., 2016, 55(36): 10800.

[31]	Zhang H G, Hwang S, Wang M Y, Feng Z X, Karakalos S, Luo L L, Qiao Z, Xie X H, Wang C M, Su D, Shao Y Y, Wu G. J. Am. Chem. Soc., 2017, 139(40): 14143.

[32]	Luo E G, Zhang H, Wang X, Gao L Q, Gong L Y, Zhao T, Jin Z, Ge J J, Jiang Z, Liu C P, Xing W. Angew. Chem. Int. Ed., 2019, 58(36): 12469.

[33]	Liu S W, Wang M Y, Yang X X, Shi Q R, Qiao Z, Lucero M, Ma Q, More K L, Cullen D A, Feng Z X, Wu G. Angew. Chem. Int. Ed., 2020, 59(48): 21698.

[34]	Deng Y J, Chi B, Li J, Wang G H, Zheng L, Shi X D, Cui Z M, Du L, Liao S J, Zang K T, Luo J, Hu Y F, Sun X L. Adv. Energy Mater., 2019, 9(13): 1802856.

[35]	Zhou Q Y, Cai J J, Zhang Z, Gao R, Chen B, Wen G B, Zhao L, Deng Y P, Dou H Z, Gong X F, Zhang Y L, Hu Y F, Yu A P, Sui X L, Wang Z B, Chen Z W. Small Methods, 2021, 5(6): 2100024.

[36]	Qu Y T, Li Z J, Chen W X, Lin Y, Yuan T W, Yang Z K, Zhao C M, Wang J, Zhao C, Wang X, Zhou F Y, Zhuang Z B, Wu Y E, Li Y D. Nat. Catal., 2018, 1(10): 781.

[37]	Gao Y, Duan X G, Li B, Jia Q Q, Li Y, Fan X B, Zhang F B, Zhang G L, Wang S B, Peng W C. J. Mater. Chem. A, 2021, 9(26): 14793.

[38]	Liu S W, Wang M Y, Yang X X, Shi Q R, Qiao Z, Lucero M, Ma Q, More K L, Cullen D A, Feng Z X, Wu G. Angew. Chem. Int. Ed., 2020, 59(48): 21698.

[39]	Tang C, Chen L, Li H J, Li L Q, Jiao Y, Zheng Y, Xu H L, Davey K, Qiao S Z. J. Am. Chem. Soc., 2021, 143(20): 7819.

[40]	Xu H X, Cheng D J, Cao D P, Zeng X C. Nat. Catal., 2018, 1(5): 339.

[41]	Nie Y, Li L, Wei Z D. Chem. Soc. Rev., 2015, 44(8): 2168.

[42]	Li Y C, Liu X F, Zheng L R, Shang J X, Wan X, Hu R M, Guo X, Hong S, Shui J L. J. Mater. Chem. A, 2019, 7(45): 26147.

[43]	Nørskov J K, Rossmeisl J, Logadottir A, Lindqvist L, Kitchin J R, Bligaard T, JÓnsson H. J. Phys. Chem. B, 2004, 108(46): 17886.

[44]	Zagal J H, Koper M T M. Angew. Chem. Int. Ed., 2016, 55(47): 14510.

[45]	Jiao Y, Zheng Y, Jaroniec M, Qiao S Z. J. Am. Chem. Soc., 2014, 136(11): 4394.

[46]	Peng H L, Liu F F, Liu X J, Liao S J, You C H, Tian X L, Nan H X, Luo F, Song H Y, Fu Z Y, Huang P Y. ACS Catal., 2014, 4(10): 3797.

[47]	Zhou W L, Su H, Li Y L, Liu M H, Zhang H, Zhang X X, Sun X, Xu Y Z, Liu Q H, Wei S Q. ACS Energy Lett., 2021, 6(9): 3359.

[48]	Dong J C, Zhang X G, Briega-Martos V, Jin X, Yang J, Chen S, Yang Z L, Wu D Y, Feliu J M, Williams C T, Tian Z Q, Li J F. Nat. Energy, 2019, 4(1): 60.

[49]	GÓmez-Marín A M, Rizo R, Feliu J M. Catal. Sci. Technol., 2014, 4(6): 1685.

[50]	Chen Y J, Ji S F, Wang Y G, Dong J C, Chen W X, Li Z, Shen rongan, Zheng L R, Zhuang Z B, Wang D S, Li Y D. Angew. Chem. Int. Ed., 2017, 56(24): 6937.

[51]	Liu D X, Wang B, Li H G, Huang S F, Liu M M, Wang J, Wang Q J, Zhang J J, Zhao Y F. Nano Energy, 2019, 58: 277.

[52]	Choi C H, Baldizzone C, Grote J P, Schuppert A K, Jaouen F, Mayrhofer K J J. Angew. Chem. Int. Ed., 2015, 54(43): 12753.

[53]	Calle-Vallejo F, Martínez J I, Rossmeisl J. Phys. Chem. Chem. Phys., 2011, 13(34): 15639.

[54]	Zagal J H. Coord. Chem. Rev., 1992, 119: 89.

[55]	van Veen J A R, van Baar J F, Kroese C J, Coolegem J G F, de Wit N, Colijn H A. Berichte Der Bunsengesellschaft Für Physikalische Chemie, 1981, 85(9): 693.

[56]	Zhou Y D, Yang W, Utetiwabo W, Lian Y M, Yin X, Zhou L, Yu P W, Chen R J, Sun S R. J. Phys. Chem. Lett., 2020, 11(4): 1404.

[57]	Gong S P, Wang C L, Jiang P, Hu L, Lei H, Chen Q W. J. Mater. Chem. A, 2018, 6(27): 13254.

[58]	Du C, Gao Y J, Chen H Q, Li P, Zhu S Y, Wang J G, He Q G, Chen W. J. Mater. Chem. A, 2020, 8(33): 16994.

[59]	Li Y C, Hu R M, Chen Z B, Wan X, Shang J X, Wang F H, Shui J L. Nano Res., 2021, 14(3): 611.

[60]	Zeng X J, Shui J L, Liu X F, Liu Q T, Li Y C, Shang J X, Zheng L R, Yu R H. Adv. Energy Mater., 2018, 8(1): 1701345.

[61]	Xiao M L, Chen Y T, Zhu J B, Zhang H, Zhao X, Gao L Q, Wang X, Zhao J, Ge J J, Jiang Z, Chen S L, Liu C P, Xing W. J. Am. Chem. Soc., 2019, 141(44): 17763.

[62]	Li J J, Guan Q Q, Wu H, Liu W, Lin Y, Sun Z H, Ye X X, Zheng X S, Pan H B, Zhu J F, Chen S, Zhang W H, Wei S Q, Lu J L. J. Am. Chem. Soc., 2019, 141(37): 14515.

[63]	Lu Z Y, Wang B, Hu Y F, Liu W, Zhao Y F, Yang R O, Li Z P, Luo J, Chi B, Jiang Z, Li M S, Mu S C, Liao S J, Zhang J J, Sun X L. Angew. Chem. Int. Ed., 2019, 58(9): 2622.

[64]	Qiao M F, Wang Y, Wågberg T, Mamat X, Hu X, Zou G A, Hu G Z. J. Energy Chem., 2020, 47: 146.

[65]	Zhang Q R, Kumar P, Zhu X F, Daiyan R, Bedford N M, Wu K H, Han Z J, Zhang T R, Amal R, Lu X Y. Adv. Energy Mater., 2021, 11(17): 2100303.

[66]	Chen Z Y, Niu H, Ding J, Liu H, Chen P H, Lu Y H, Lu Y R, Zuo W B, Han L, Guo Y Z, Hung S F, Zhai Y M. Angewandte Chemie Int. Ed., 2021, 60(48): 25404.

[67]	Zheng L, Dong Y Y, Chi B, Cui Z M, Deng Y J, Shi X D, Du L, Liao S J. Small, 2019, 15(4): 1803520.

[68]	Jia Y, Xiong X Y, Wang D N, Duan X X, Sun K, Li Y J, Zheng L R, Lin W F, Dong M D, Zhang G X, Liu W, Sun X M. Nano Micro Lett., 2020, 12(1): 1.

[69]	Chen Y J, Ji S F, Zhao S, Chen W X, Dong J C, Cheong W C, Shen rongan, Wen X D, Zheng L R, Rykov A I, Cai S C, Tang H L, Zhuang Z B, Chen C, Peng Q, Wang D S, Li Y D. Nat. Commun., 2018, 9: 5422.

[70]	Yang Y, Mao K T, Gao S Q, Huang H, Xia G L, Lin Z Y, Jiang P, Wang C L, Wang H, Chen Q W. Adv. Mater., 2018, 30(28): 1801732.

[71]	Wang J, Li H G, Liu S H, Hu Y F, Zhang J, Xia M R, Hou Y L, Tse J, Zhang J J, Zhao Y F. Angew. Chem. Int. Ed., 2021, 60(1): 181.

[72]	Zhang J T, Zhang M, Zeng Y, Chen J S, Qiu L X, Zhou H, Sun C J, Yu Y, Zhu C Z, Zhu Z H. Small, 2019, 15(24): 1900307.

[73]	Yuan K, Lützenkirchen-Hecht D, Li L B, Shuai L, Li Y Z, Cao R, Qiu M, Zhuang X D, Leung M K H, Chen Y W, Scherf U. J. Am. Chem. Soc., 2020, 142(5): 2404.

[74]	Yin H B, Yuan P F, Lu B A, Xia H C, Guo K, Yang G G, Qu G, Xue D P, Hu Y F, Cheng J Q, Mu S C, Zhang J N. ACS Catal., 2021, 11(20): 12754.

[75]	Wei X, Zheng D, Zhao M, Chen H Z, Fan X, Gao B, Gu L, Guo Y, Qin J B, Wei J, Zhao Y L, Zhang G C. Angew. Chem. Int. Ed., 2020, 59(34): 14639.

[76]	Hu B T, Huang A J, Zhang X J, Chen Z, Tu R Y, Zhu W, Zhuang Z B, Chen C, Peng Q, Li Y D. Nano Res., 2021, 14(10): 3482.

[77]	Han Y H, Wang Y G, Xu R R, Chen W X, Zheng L R, Han A J, Zhu Y Q, Zhang J, Zhang H B, Luo J, Chen C, Peng Q, Wang D S, Li Y D. Energy Environ. Sci., 2018, 11(9): 2348.

[78]	Xue W D, Zhou Q X, Cui X, Jia S R, Zhang J W, Lin Z Q. Nano Energy, 2021, 86: 106073.

[79]	Guo Y Y, Yuan P F, Zhang J N, Hu Y F, Amiinu I S, Wang X, Zhou J G, Xia H C, Song Z B, Xu Q, Mu S C. ACS Nano, 2018, 12(2): 1894.

[80]	Lin Z Y, Huang H, Cheng L, Hu W, Xu P P, Yang Y, Li J M, Gao F Y, Yang K, Liu S, Jiang P, Yan W S, Chen S, Wang C L, Tong H G, Huang M X, Zheng W, Wang H, Chen Q W. Adv. Mater., 2021, 33(51): 2107103.

[81]	Cuesta A. ChemPhysChem, 2011, 12(13): 2375.

[82]	O’Keeffe M. Chem. Soc. Rev., 2009, 38(5): 1215.

[83]	Wu R, Song Y J, Huang X, Chen S G, Ibraheem S, Deng J H, Li J, Qi X Q, Wei Z D. J. Power Sources, 2018, 401: 287.

[84]	Yang L, Cheng D J, Xu H X, Zeng X F, Wan X, Shui J L, Xiang Z H, Cao D P. PNAS, 2018, 115(26): 6626.

[85]	Jiao L, Jiang H L. Chem, 2019, 5(4): 786.

[86]	Ye W, Chen S M, Lin Y, Yang L, Chen S J, Zheng X S, Qi Z M, Wang C M, Long R, Chen M, Zhu J F, Gao P, Song L, Jiang J, Xiong Y J. Chem, 2019, 5(11): 2865.

[87]	Zhu X F, Zhang D T, Chen C J, Zhang Q R, Liu R S, Xia Z H, Dai L M, Amal R, Lu X Y. Nano Energy, 2020, 71: 104597.

[88]	Zhang H B, Zhou W, Chen T, Guan B Y, Li Z, Lou X W D. Energy Environ. Sci., 2018, 11(8): 1980.

[89]	Zhang D Y, Chen W X, Li Z, Chen Y J, Zheng L R, Gong Y, Li Q H, Shen rongan, Han Y H, Cheong W C, Gu L, Li Y D. Chem. Commun., 2018, 54(34): 4274.

[90]	Zhu Z J, Yin H J, Wang Y, Chuang C H, Xing L, Dong M Y, Lu Y R, Casillas-Garcia G, Zheng Y L, Chen S, Dou Y H, Liu P R, Cheng Q L, Zhao H J. Adv. Mater., 2020, 32(42): 2004670.

Funding

National Key Research and Development Program of China(2017YFB0102900)

National Key Research and Development Program of China(2016YFB0101201)

National Natural Science Foundation of China(51971094)

National Natural Science Foundation of China(21476088)

National Natural Science Foundation of China(21776104)

Guangdong Provincial Department of Science and Technology(2015A030312007)

PDF (9633KB)

1686

Accesses

Citation

Detail

Sections

Recommended