1
Carbon‐Based, Metal‐Free Electrocatalysts for Renewable Energy Technologies

Li Tao Zhaohui Xiao Ruilun Wang and Shuangyin Wang

Hunan University, State Key Laboratory of Chem/Bio‐Sensing and Chemometrics, Provincial Hunan Key Laboratory for Graphene Materials and Devices, College of Chemistry and Chemical Engineering, Changsha, 410082, Hunan, PR China

1.1 Introduction

The traditional fossil resources cannot meet the rapidly increasing energy demand, a series of challenges, such as environment pollution, energy security, and health issues, are accompanied with energy over consumption. To solve these energy problems, developing renewable and sustainable energy storage/conversion technologies to optimize the traditional energy system is urgently needed.

Various renewable energy technologies, including fuel cells, Li‐ion batteries, metal–air batteries, all‐vanadium redox flow batteries (VRFBs), water splitting, etc., have been extensively developed. In most of these energy technologies, electrocatalysis is a core process. For example, fuel oxidation and oxygen reduction reactions (ORRs) are the core reactions in the anodes and cathodes, respectively, of proton exchange membrane fuel cells (PEMFCs); oxygen evolution reaction (OER) is a bottleneck reaction of metal–air batteries and water‐splitting devices. The electrocatalysts are also needed to promote the redox chemistry of vanadium ions on the graphite electrodes in all‐VRFBs. Overall, the design of advanced electrocatalysts is of critical importance to realize high‐performance renewable energy technologies.

Electrochemical water splitting can produce oxygen (OER) and hydrogen (hydrogen evolution reaction, HER), which converts electricity to chemical energy. In contrast, the hydrogen oxidation reaction (HOR) and ORR can assemble fuel cells to convert chemical energy to electrical energy. Similarly, the ORR and OER can act as the reversible half reaction of rechargeable metal–oxygen batteries. However, these renewable energy technologies were hampered by the sluggish kinetics of these electrochemical reactions and consequently demand a high overpotential to drive the reactions. For example, in PEMFCs, the poor performance of the cathode is the major cause of efficiency reduction and hinders the progress of fuel cell commercialization. Therefore, these electrode reactions generally need to be catalyzed by catalysts to achieve the desired performance for practical applications. However, considering the high cost and scarcity of the currently used noble metal catalysts, their large‐scale applications in relevant clean energy technologies have been greatly limited. Thus, it is very necessary to develop cheap alternative electrocatalysts with high activity, high durability, and low cost to facilitate the realization of clean energy devices (Figure 1.1).

Image described caption and surrounding text.

Figure 1.1 (a) Hydrogen and oxygen cycles for energy storage and conversion. HER and OER can realize energy storage in chemical form. For energy conversion, ORR and HOR (MOR) are two half‐cell reactions of fuel cell. (b) Scheme of the overpotentials associated with HER, OER, ORR , and HOR. (c) Scheme of the anion‐exchange membrane electrolyzer. (d) Scheme of a fuel cell in alkaline electrolytes.

Source: Reproduced with permission [1]. Copyright 2015, Royal Society of Chemistry.

1.2 Oxygen Reduction Reaction

Compared with the traditional fossil fuels, fuel cells are a more efficient and clean device, which has been considered the most promising energy conversion device. Using renewable energy sources, such as hydrogen (H2) or small organic molecule (CH3OH, CH3CH2OH, and CHOOH) as fuel, fuel cells can directly convert the chemical energy to electricity accompany with oxygen as a reactant at room temperature. Compared with the traditional foil engine, fuel cells would not be limited by the thermal efficiency and the energy conversion efficiency can convert up to 50–70% of available fuel to electricity [2].

In fuel cells, ORR is the process of O2 absorption from electrolyte and decomposition on the electrode. Actually, the oxygen molecule and H2O molecule are always absorbed on the active sites of the electrode. From the previous researches, the ORR mechanism and various ORR electrocatalysts have been deeply studied. The ORR process is a multi‐electron reduction reaction, in which there are many elementary reactions and intermediate products. As well known, the electrocatalysis of ORR occurs mainly in two ways. The first one is the one‐step, four‐electron reaction, in which O2 is efficiently reduced to H2O in an acid electrolyte or is reduced to OH− in an alkaline electrolyte. The other one is the two‐step, two‐electron reactions, in which O2 is reduced to H2O2 in an acid electrolyte or O2 is reduced to HO2− in an alkaline electrolyte. To develop more efficient ORR electrocatalyst, the reaction process and mechanism must be studied in depth. Many works about the reaction pathway and the intermediates in the ORR process have been conducted; the accepted four‐electron reaction elementary steps according to the associative mechanism for carbon materials electrocatalyst are shown as follows (the dissociative has the same mechanisms with the associative both in acid and in alkaline and is not shown here) [3]:

In acid solution:

1.1vol-2-c01-math-0001
1.2vol-2-c01-math-0002
1.3vol-2-c01-math-0003
1.4vol-2-c01-math-0004
1.5vol-2-c01-math-0005

In alkaline solution:

1.6vol-2-c01-math-0006
1.7vol-2-c01-math-0007
1.8vol-2-c01-math-0008
1.9vol-2-c01-math-0009
1.10vol-2-c01-math-0010

where * stands for the active sites of the electrocatalyst surface, and (l) and (g) refers to the liquid and gas phases, respectively. The O*, OH*, and OOH* are adsorbed intermediates. The electrolyte H2O plays the role of proton donor in the alkaline condition rather than H3O+. For the ORR process, both of the overpotentials of the ORR and the reaction free energy ΔG of elementary steps can be obtained by density functional theory (DFT) calculation. It has been reported that the rate‐determining step is the adsorption of O2 as OOH* or desorption of OH* as water on the active sites of the electrocatalyst surface [4]; thus, the property of surface active sites in the electrocatalyst may play a major role.

For the multistep and multiple adsorbed intermediates in the ORR process, a series of works have been conducted to develop highly efficient electrocatalysts. In all of these works, carbon‐based materials are considered the most promising materials to replace Pt‐based electrocatalysts. For the carbon‐based electrocatalysts, they include metal‐free carbon electrocatalyst and carbon/metal hybrid electrocatalyst. To enhance the electrocatalytic activity of carbon‐based electrocatalyst, various strategies such as heteroatom doping, creating defects, surface molecular functionalization, nonmetallic carbon materials recombination, and metal compounds loading have been developed, as shown in Figure 1.2.

  1. Heteroatom doping: Both the experiment and the theoretical calculations have demonstrated that the doped heteroatoms can induce asymmetric spin density and effectively change the electronic arrangement of carbon materials, which would reduce the adsorption barrier and produce local changes for ORR. Interestingly, the dual‐doping (B–N, N–S, etc.) and triple‐doping (N–P–S, B–N–P) show much better catalytic performance than mono‐doping carbon materials due to the synergistic effect.
  2. Defecting: For the special electron structure (higher charge densities) of the edge sites, the defective carbon atoms show much better ORR active sites than the basal plane. Many works have been conducted to create defects on carbon materials; the defective carbon materials reveal much higher ORR activity.
  3. Surface molecule functionalization: Apart from the intramolecular charge transfer induced by heteroatom doping, the carbon materials have been modified by electron‐acceptor molecules or electron‐donor molecules in covalent or non‐covalent ways. By the charge transfer between the functionalized molecule and the carbon materials, the ORR performance of the carbon materials could be significantly increased.
  4. Metal‐free carbon materials combination: Due to the different electron structures of different carbon materials, synthesis of highly porous metal‐free carbon materials combination, the electron structure of these materials can be significantly optimized and numerous active sites could be generated, which exhibited high ORR performance.
  5. Metal compounds loading: For the catalytic performance of the metal compounds itself and the stronger interaction between the carbon materials, many works about metal compounds have been loaded on the carbon materials except high ORR performance.
Image described caption and surrounding text.

Figure 1.2 (a) Calculated charge density distribution for the NCNTs and the schematic representations of possible adsorption modes of an oxygen molecule at the CCNTs (top) and NCNTs (bottom) [5]. (b) Illustration of charge transfer process and oxygen reduction reaction on PDDA‐CNT [6]. (c) The air‐saturated droplet was deposited on the basal plane of the HOPG electrode [7]. (d) HAADF image of DG, hexagons, pentagons, heptagons, and octagons was labeled in orange, green, blue, and red, respectively [8].

Source: Copyright 2016, Wiley‐VCH.

Source: Copyright 2014, Wiley‐VCH.

Source: Copyright 2011 American Chemical Society.

Source: Copyright 2009 American Association for the Advancement of Science.

These methods could cause the local charge transfer of the materials, which could lead electron modulation to provide desirable electronic structures for catalysis. By revealing the mechanism of charge transfer effect to enhance the ORR catalytic performance, the presenting critical issues, challenges, and perspectives of different functionalized carbon‐based materials should be introduced. In this chapter, the understanding of the carbon‐based catalytic materials will be more distinctive and knowledge will help researchers to control and even tailor the characteristics of metal‐free catalysts for energy‐related electrocatalysis. In the following, we first briefly introduce the doped carbon materials as there are numerous reviews about it. Second, the defecting carbon materials, metal‐free carbon materials combination, and metal compounds loaded on carbon materials were comprehensively and critically reviewed. Finally, the deficiencies and challenges about charge transfer effects enhanced in carbon‐based materials were briefly pointed out.

1.2.1 Heteroatom‐Doped Carbon Materials

As Dai and coworkers successfully developed vertically aligned nitrogen‐doped carbon nanotube (VA‐NCNT) by pyrolysis of iron(II) phthalocyanine as a highly efficient electrocatalyst for ORR in 2009 [5], heteroatom doping as an efficient way to tailor the property of carbon materials has been widely investigated. In Dai's work, it is reported that VA‐NCNTs are used as a metal‐free electrocatalyst with a much better electrocatalytic activity than platinum for oxygen reduction in alkaline media. The nitrogen‐induced charge delocalization could also change the chemisorption mode of O2 onto the nitrogen‐doped carbon nanotube (NCNT) electrodes. The parallel adsorption could effectively weaken the OO bonding to facilitate ORR. Therefore, nitrogen doping in carbon can efficiently create the metal‐free active sites for electrochemical reduction of O2. The incorporation of electron‐accepting nitrogen atoms in the conjugated nanotube carbon plane appears to impart a relatively high positive charge density on adjacent carbon atom, which provides a one‐step, four‐electron pathway for the ORR on VA‐NCNTs with a high performance. The ORR active sites have been characterized on highly oriented pyrolytic graphite (HOPG) model catalysts with well‐defined N doping. The ORR active site is created by pyridinic‐N which creates Lewis basic sites. The specific activities per pyridinic‐N in the HOPG model catalysts are comparable with those of the N‐doped graphene powder catalysts. Thus, the ORR active sites in N‐doped carbon materials are carbon atoms with Lewis basicity next to pyridinic‐N [9]. To determine the active site conclusively, the researchers develop four types of model catalysts with well‐defined π conjugation based on HOPG. The active sites and adsorption properties of the nitrogen‐doped carbon surfaces are examined. According to the local scanning tunneling microscopy/spectroscopy measurements combined with DFT calculations, carbon atoms adjacent to pyridinic‐N possess a localized density of states in the occupied region near the Fermi level [10]. This suggests that the carbon atoms can act as Lewis bases owing to the possibility of electron pair donation. Acidic CO2 molecule is adsorbed only on the ORR‐active pyri‐HOPG catalyst, which proves that the Lewis basic site is created by pyridinic‐N on the HOPG surface. It is generally known that oxygen molecules can be adsorbed on Lewis base sites [11]. Because O2 adsorption is the initial step of the ORR, the Lewis base site created by pyridinic‐N is thus suggested to be the active site for ORR.

Recently, Yang et al. [12] reported boron‐doped carbon nanotubes (BCNTs) as ORR electrocatalysts with improved activities relative to undoped carbon nanotubes (CNTs). On the basis of experimental analyses and theoretical calculations, they concluded that the B atoms in the BCNT lattice are positively charged and act as the active sites for ORR. In contrast to all‐CNTs, CNTs containing both B and N atoms (BCN nanotubes), either in an aligned or in a nonaligned form, have tunable bandgaps by means of their chemical composition. Unlike CNTs, the bandgap of BCN nanotubes is independent of the diameter and chirality. This unique structure–property relationship makes BCN nanotubes attractive candidates for potential uses in many areas where CNTs have been exploited. In particular, the superb thermal stability and chemically tunable bandgap of BCN nanotubes provide tremendous opportunities to tune nanotube electronic properties for their use as an efficient metal‐free ORR electrode, even at elevated temperatures. Dai et al. have, for the first time, prepared vertically aligned BCN nanotubes (VA‐BCNs) by pyrolysis of melamine diborate, a single‐compound source of carbon, boron, and nitrogen, for BCN nanotube growth. Due to a synergetic effect arising from co‐doping of CNTs with boron and nitrogen, the resultant VA‐BCN nanotube electrode has higher electrocatalytic activity for ORR in alkaline medium than its counterparts doped with boron or nitrogen alone (i.e. VA‐BCNT or VA‐NCNT). The observed superior ORR performance with good tolerance to methanol and carbon monoxide and excellent durability for the VA‐BCN nanotube electrode compared with a commercial Pt/C electrode opens up avenues for the development of novel, efficient, metal‐free ORR catalysts by co‐doping. Thereafter, the authors have successfully developed a facile low‐cost approach to mass production of BCN graphene with tunable N/B‐doping levels simply by thermal annealing graphene oxide (GO) in the presence of boric acid under ammonia atmosphere. The resultant metal‐free BCN graphene samples exhibited ORR electrocatalytic activities even better than the commercial Pt/C electrocatalyst. In good agreement with the experimental observations, the first‐principle calculations revealed a doping‐level‐dependent energy bandgap, spin density, and charge density. BCN graphene with a modest N‐ and B‐doping level exhibited the best ORR electrocatalytic activity, fuel selectivity, and long‐term durability, along with an excellent thermal stability and porosity.

The ability of N‐doped carbon as a catalyst for ORR was also developed through the additional doping of B and/or P. It is suggested that additional doping of B and/or P enhances asymmetric atomic spin density. The additional doping of B and/or P modifies the carbon characteristics and improves the ORR activity. The B doping increases the portion of pyridinic‐N sites among various N‐doping types and magnifies the degree of the sp2 carbon structures. However, the P doping enhances the charge delocalization of the carbon atoms and constructs morphology with many open edge sites that are split and wrinkled. The ORR activities of the catalysts were increased due to the additional B or P doping. Therefore, it can be concluded that the charge delocalization of the carbon atoms or the number of open edge sites are the main factor in determining the ORR activity of N‐doped carbon.

1.2.2 Surface Molecule Functionalization

Previous work attributed the improved electrocatalytic activity of the N‐doped carbon to the electron‐withdrawing ability of the nitrogen atoms to create a net positive charge on the adjacent carbon atoms. In addition to the charge‐induced favorable O2 adsorption, the positively charged carbon atoms can readily attract electrons from the anode to enhance the ORR. Uncovering this new ORR mechanism in NCNT electrodes is significant as the same principle could be applied to the development of various other metal‐free efficient ORR catalysts for fuel cell and many other applications. In this regard, Wang et al. have discovered that functionalization of CNTs, simply through physical adsorption, with polyelectrolyte chains containing positively charged nitrogen moieties could also create a net positive charge on carbon atoms of nitrogen‐free CNTs via intermolecular charge transfer. It was found that the polyelectrolyte‐adsorbed all‐CNTs possess remarkable electrocatalytic activities for ORR. The ease with all‐CNTs can be converted into efficient metal‐free ORR electrocatalysts simply by adsorption‐induced intermolecular charge transfer suggests considerable room for cost‐effective preparation of various metal‐free catalysts for oxygen reduction, and even new catalytic materials for applications beyond fuel cells. Thereafter, composites of poly(3,4‐ethylenedioxythiophene):poly(styrenesulfonate) (PEDOT:PSS) and reduced graphene oxide (RGO) have been prepared and used as electrocatalysts for ORR. Hybridization of PEDOT:PSS with RGO and the subsequent treatment with H2SO4 can significantly increase the electrical conductivity of PEDOT because of the alternation of PEDOT resonating ground states through strong π–π interaction and segregation and partial removal of PSS chains from PEDOT:PSS [13]. On the other hand, the positively charged PEDOT backbones have electron‐withdrawing ability to create net positive charges on RGO sheets via intermolecular charge transfer [6, 14]. Accordingly, the resultant PEDOT:PSS/RGO composite films are expected to exhibit synergistically enhanced electrocatalytic activity for ORR.

On the other hand, a new concept of molecular doping of graphene used as an ORR electrocatalyst was presented. As a strong electron‐withdrawing group, nitrobenzene was doped onto graphene via covalent CC bonding between nitrobenzene and the graphene surface to give nitrobenzene‐doped graphene (denoted as NB‐graphene), and it has a strong electron‐accepting ability to withdraw an electron from carbon atoms in the graphene sheet [15]. The ORR catalytic activity of the molecule‐doped graphene was enhanced through the charge transfer process between the linked molecule, nitrobenzene, and graphene. Raman characterizations confirmed the charge transfer process. The electrochemical characterizations proved that the molecular‐doping‐induced charge transfer strategy could enhance the catalytic activity for ORR like N‐doped graphene. It is believed that graphene doped by other molecules with strong electron‐withdrawing ability may also work well as metal‐free electrocatalysts for ORR. In addition, the first‐principle calculations were performed to predict the structure and catalytic activities of graphene nanoribbon (GNR) adsorbed with tetracyanoethylene (TCNE) [16]. The adsorption of TCNE results in electron transfer from graphene to TCNE and redistribution of charge on the GNRs. The existence of Stone–Wales defect enhances the adsorption of TCNE and the active sites are mostly located near the edges. These predictions show that TCNE‐adsorbed graphene could be a good metal‐free catalyst to replace Pt‐based catalysts for ORR.

1.2.3 Defective Carbon

Carbon‐related materials have become one of the most popular ORR catalysts. As we know, after the heteroatom or molecule doping, more defects would form as to be the active sites for ORR. The defects induced by heteroatom or molecule doping to improve the ORR activities have been extensively reported. Recently, the intrinsic defects of the carbon have been explored. When studying the heteroatom‐doped carbon materials, the improved ORR activity was attributed to the charge redistribution after inducing the heteroatom dopants, which can change the O2 chemisorption mode to effectively weaken the O−O bonding and further facilitate the ORR process. Inspiring by these conclusions, Wang et al. observed that the distribution of electronic density of a graphene model excitedly found that the electron distribution on a graphene sheet is inhomogeneous. It can be obviously seen that the edge carbon is highly charged compared with the basal plane carbon (Figure 1.3A). This observation encourages finding out whether the edge carbon is more active than the basal plane carbon. To prove the guess that the edge carbon is more active than the basal plane carbon, our group designed a micro‐electrochemical testing system (Figure 1.3B) to study the ORR on HOPG surface as the working electrode. An air‐saturated droplet with a diameter of approximately 15 μm was deposited at a specified location on the HOPG surface using a microinjection system. The standard three electrodes make the whole system collected. Then, the tip was moved to any specified locations, the edge‐rich areas and the basal plane surface. On different locations, the electrochemical signals could be collected. It can be obviously seen that the edge carbon is more active than the basal plane carbon and the edge carbon can be considered as the intrinsic defects of the carbon for ORR. The edge carbon not only has a larger current density but also has a lower overpotential than the basal plane carbon. This research provided the direct evidence to prove that the edge carbon is more active than the basal plane carbon. This conclusion could provide a general principle to guide researchers to design efficient, carbon‐based, metal‐free electrocatalysts for ORR. The carbon materials with more defects have better activities. This interesting finding can guide researchers to prepare defect‐rich carbon materials for ORR.

(A) Digital capture and screenshot of Rectangular armchair and zigzag edges with atomic values. (B) Schematic diagram of (a) micro apparatus for the ORR electrochemical experiment. (b, c) Optical photograph of the HOPG surface as the working electrode with the air-saturated droplet (encircled )deposited on the edge and basal plane of the HOPG surface. (d) LSV curves of the ORR tested for a droplet located either on the edge (dashed) or on the basal (solid) plane of the HOPG surface.

Figure 1.3 (A) Charge distribution on the two representative graphene sheets of C72H24. (B) (a) Micro apparatus for the ORR electrochemical experiment. (b, c) Optical photograph of the HOPG surface as the working electrode with the air‐saturated droplet deposited on the edge and basal plane of the HOPG surface. (d) LSV curves of the ORR tested for a droplet located either on the edge or on the basal plane of the HOPG surface [7].

Source: Copyright 2014, Wiley‐VCH.

With this exciting finding, many people try to prepare metal‐free carbon materials with more intrinsic defects and some in‐depth researches can help us to clearly understand the ORR mechanism. Dai's group reported that graphene quantum dots (GQDs) supported by GNRs have ultra‐high electrocatalytic performance for ORR [17]. It revealed that numerous surface/edge defects on the GQD/GNR surfaces and at their interface act as the active sites. Electrocatalytic performance of the GQD–GNR hybrid was very excellent and the half‐wave potential of it is as better as that of the Pt/C catalyst, but much better than that of the GQD and graphene alone. The results showed that the defect‐rich carbon materials have a very good ORR activity. So after the understanding of the positive effect of the intrinsic defects on ORR activity, researchers must develop good methods to prepare the metal‐free carbon materials with more edge defects. Recently, Hu's group successfully prepared carbon nanocages (CNCs) with abundant holes, edges, and positive topological disclinations but without any dopants to address the influence of intrinsic carbon defects on ORR activity [18]. The CNCs display much typical defective locations, such as the corner, the broken fringe, and the hole. They clearly indicate the significant contribution of intrinsic carbon defects to ORR. According to the theoretical results, the pentagon and zigzag edge defects are the main ORR active sites, thus the most contributors for the high ORR activity of the CNCs. The underlying reason of these defects that have high ORR activity is attributed to their different electronic structures. The defects of zigzag edge have a portion of the active unpaired π electrons, which are located at each edge of the carbon atom. The unpaired π electrons can effectively facilitate electron transfer to O2 and the OOH* is easily formed with a free energy change of −0.222 eV. In addition, Zhang's group also used Mg(OH)2 as a template precursor to prepare a novel heteroatom‐doped and edge‐rich graphene material for both ORR and OER [19]. Their results also revealed the critical importance of edge and topological defects in the activity origin of metal‐free nanocarbon materials for oxygen electrocatalysis. All the nitrogen‐doping‐induced sites near the edge exhibited a much lower overpotential, further indicating the great contributions of the edge defects. Such kinds of findings suggest the researchers to reconsider the origin of ORR activity for the doped carbon‐based electrocatalysts because of the unavoidable coexistence of a large number of defects.

Recently, Wang and coworkers developed a novel Ar plasma method to treat the surface of the graphene to prepare dopant‐free and edge‐rich graphene as highly efficient, metal‐free electrocatalysts for ORR, as shown in Figure 1.4. Plasma technology can be used to modify and clean the surface of materials. In this work, the authors used plasma to etch the graphene and then dig many holes in graphene to expose a number of edge defects on the basal plane. The SEM images of pristine graphene (G) and plasma‐treated graphene (P‐G) were shown in Figure 1.4B, both of them showed similar surface morphology, indicating that Ar plasma treatment did not cause obvious structural damage on graphene. The graphene can reserve its ideal electronic conductivity. However, we can clearly see from the TEM images that there are many holes with a diameter of around 15 nm on the graphene treated by Ar plasma. On the one hand, this novel structure can reserve the good electrical conductivity of graphene; on the other hand, it can also provide abundant active sites, which are beneficial for the electrocatalysis of ORR. After the plasma treatment, the increasing of defect level have led to exposing of extra edge after the plasma treatment. The electrochemical data showed that the ORR activity enhanced a lot after the plasma treatment. The ORR activity of P‐G was even better than some of the doped carbon materials, which can guide people to rethink about the origin of ORR activity for doped carbon materials.

Image described caption and surrounding text.

Figure 1.4 (A) Illustration of the preparation of the edge‐rich and dopant‐free graphene by the Ar plasma etching. (B) SEM and TEM images of pristine graphene (a, c) and Ar‐ plasma‐treated graphene (b, d). (C) CV curves of G and P‐G at a scan rate of 50 mV s−1 in N2‐saturated and O2‐saturated 0.1 M KOH. (D) RDE voltammograms of G and P‐G in an O2‐saturated 0.1 M KOH with a scan rate of 10 mV s−1 [20].

Source: Copyright 2016, Royal Society of Chemistry.

1.3 Electrochemical Water Splitting (HER and OER)

Electrochemical water splitting (2H2O − O2 + 2H2) is considered as a promising approach to generate hydrogen fuel [21]. In order to prompt hydrogen generated at the cathode with appreciable rate at relatively low applied voltages, the sluggish kinetics of OER and HER usually requires catalyst materials such as ruthenium (RuO2) or Pt to lower the energy barriers [22]. However, the noble‐metal‐based catalysts are costly and their supply is not sustainable, which has severely restricted the large‐scale implementation of this technology. Recently, significant research efforts have been devoted to the development of OER and HER catalysts based on transition metal chalcogenides, such as Ni, Fe, and Co [23].

1.3.1 Hydrogen Evolution Reaction

As we know, Pt‐based materials are the best electrocatalysts for HER due to their ideal interaction with H or OH species. However, the high price and the limited reserves of Pt definitely prohibit its large‐scale commercialization [24]. Therefore, it is desired to develop cost‐effective but highly efficient alternatives to replace the noble metals. In recent years, much effort has been devoted to study transition metal sulfides (Mo, Co, Fe, Ni, Cu) [25], phosphides [26], nitrides [27], and carbides [28] as HER catalysts. Although some of them showed high HER activity, the essential HER activity of these catalysts are based on the interaction of metal‐H bond. The common oxidation and corrosion susceptibility greatly limits their utilization in acidic proton‐exchange, membrane‐based electrolysis for continuous hydrogen production [29]. Therefore, exploring metal‐free alternatives is extremely necessary and important.

Heteroatoms (e.g. N, S, P, B, or others) as a dopant into carbon can modulate carbon's physical and chemical properties to obtain more reactive sites [28b]. More importantly, this process can produce carbon‐based materials with improved ability to adsorb the atomic/molecular species undergoing catalytic reactions and without substantially compromised conductive properties [30]. These types of heteroatom‐doped structures may provide opportunities for further development of low‐cost, metal‐free catalysts with high activities and long lifetimes. It has been known that the difference in electronegativity and size between the heteroatoms (N,P, B, and S) and carbon can polarize adjacent carbon atoms to facilitate ORR [12, 31, 32], which may be applicable to the hydrogen evolution process [33].

Nanocarbons (e.g. graphene and CNT) are often selected to support HER nanocatalysts for promoting the catalytic performances (activity and stability) of the latter. Generally, nanocarbons could provide either favorable structural supports (e.g. large surface area) or cooperative electrical effects (e.g. enhanced electron transfer and transport) or both in those cases. However, these nanocarbon materials are HER‐inert intrinsically. Recently, heteroatom‐doped nanocarbons were directly used to catalyze HER. The charms of nanocarbons as “non‐conventional” HER catalysts might mainly originate from the low cost and excellent physicochemical properties (e.g. high chemical stability) of carbon materials. Intentional introduction of some heteroatoms is necessary to make the catalytically inactive pristine nanocarbon materials highly active. This might lead to the creation of defect sites that can modulate the physical and chemical properties of nanocarbons and, more importantly, the addition of reactive sites that mediate the conversion of atomic/molecular species to the desired products. This is particularly well demonstrated by heteroatom‐doped, graphene‐based HER catalysts. Sathe et al. reported a B‐substituted graphene with an enhanced activity for HER. This material was synthesized by doping defective graphene (DG) with B atoms using borane tetrahydrofuran (BH3‐THF) as the borylating agent. In electrochemical water reduction, the material exhibited a lower overpotential than its undoped counterpart. To explore the effects of various dopants (N, B, O, S, P, and F) in graphene toward HER activity, Qiao's group conducted DFT calculations to study the electronic properties of differently doped graphene models (Figure 1.5). The DFT calculation provided us some important information: (i) N and O acted as electron acceptors for the adjacent C, whereas F, S, B, and P served as electron donors. (ii) N‐ and P‐co‐doped graphene had the most favorable H* adsorption–desorption property among several doped graphene models, indicating the best HER catalytic activity. (iii) The different H* adsorption behavior on graphene correlated with graphene's valence orbital energy. Based on theoretical predictions, the authors prepared N,P‐co‐doped graphene that exhibited a much lower HER overpotential than that of other pure and single‐doped graphene samples and a performance comparable with some of the traditional metal‐containing catalysts. In another study, Qiao's group found that coupling graphitic carbon nitride (g‐C3N4) with nitrogen‐doped graphene could lead to a highly active composite HER catalyst. The excellent catalytic performance of the resulting hybrid material originated from the synergistic effect of g‐C3N4 and N‐doped graphene. g‐C3N4 provided highly active hydrogen adsorption sites, and N‐doped graphene facilitated the electron transfer process for proton reduction. In order to reveal the origin of high catalytic activity, Deng et al. conducted DFT calculations on heteroatom‐doped CNTs. Their results showed that the introduction of metal and nitrogen dopants synergistically optimized the electronic structure of CNTs, and the adsorption free energy of H atoms on CNTs. Furthermore, they suggested that the predominant route of HER in this catalytic system was based on the Volmer–Heyrovsky mechanism.

(a) Bar graph with chemical structural diagram inset for NBO population analysis of F; gN; O; pN; S; B P nonmetallic heteroatoms in graphene matrix with Site 1 to 3 marked. (b)Free energy (ΔGH*) diagram for HER at the equilibrium potential (URHE=0 V) for N- and/or P-doped graphene models. (c) Line graph depicting relationship between ΔGH* and Ediff for six models.

Figure 1.5 (a) NBO population analysis of six different nonmetallic heteroatoms in graphene matrix. pN and gN represent pyridinic and graphitic type of N, respectively. Inset shows the proposed doping sites for different elements, sites 1 and 2 are the edge and center in‐plane sites, respectively, and site 3 is an out‐of‐plane center site in graphene. (b) The calculated free energy (ΔGH*) diagram for HER at the equilibrium potential (URHE = 0 V) for N‐ and/or P‐doped graphene models. (c) Relationship between ΔGH* and Ediff for various models [33a].

Source: Copyright 2014 American Chemical Society.

Intermolecular doping of graphene with heteroatoms, such as nitrogen (N), boron (B), phosphorus (P), fluorine (F), or sulfur (S), is an effective way to tailor its electronic structure and (electro)‐chemical properties [34]. In particular, the co‐doping of two elements that have different electronegativities to that of carbon, for example, B/N [35], S/N [36], and P/N [33a] couples, could lead to unique electron‐donor properties of carbon by the so‐called synergistic coupling effect between two heteroatoms. Both experimental tests and theoretical analysis confirmed that such effect could largely boost the electrocatalytic activities of graphene materials toward electrocatalytic ORR and HER [33a, 35, 36]. Taking N,P‐co‐doped graphene (N,P‐G) as an example [33a], DFT calculations predicted that the N and P heteroatoms could co‐activate the adjacent C atom in the graphene matrix by affecting its valence orbital energy levels and leading to a reduced ΔGH*. The experiments indeed showed that the synthesized N,P‐G catalyst exhibited a much lower HER overpotential than that for all investigated pure and single‐doped graphene samples.

Graphitic carbon nitride (g‐C3N4) is an inert electrocatalyst; however, by chemical coupling of g‐C3N4 with nitrogen‐doped graphene, the hybrid (C3N4@NG) exhibited a favorable HER activity, comparable with some of the metallic catalyst [37]. On the theoretical perspective, ΔGH* values for pure g‐C3N4 and graphene‐N surfaces are opposite in signs, indicating that hydrogen adsorption is either too strong (g‐C3N4) or too weak (graphene‐N). Chemical coupling of g‐C3N4 and graphene‐N into a uniform hybrid results in a mediated adsorption–desorption behavior (ΔGH* tends to zero), which facilitates the overall HER kinetics. Such an atomic‐level HER mechanism on the surface of g‐C3N4@NG clearly indicates the origin of its strikingly high electrocatalytic activity.

Recently, Dai et al. have developed a metal‐free HER catalyst based on 3D porous graphitic carbons co‐doped with nitrogen and phosphorus prepared by self‐assembling melamine and phytic acid into melamine‐phytic acid supermolecular aggregate (MPSA) in the presence of GO (MPSA/GO), followed by pyrolysis. The obtained catalysts exhibit high catalytic activities toward both HER with a low overpotential and ORR via a four‐electron path way. A high peak power density was achieved when using the pyrolyzed MPSA/GO as the air electrode. Although the low‐cost and efficient metal‐free, carbon‐based ORR and HER bifunctional electrocatalysts could advance regenerative fuel cells and water‐splitting systems to the marketplace, the newly developed simple and scalable methodology should open avenues for low‐cost, large‐scale production of high‐performance, carbon‐based bifunctional catalysts for renewable energy technologies and beyond.

In another important work [38], Cui et al. found that HER‐inert pristine CNTs could be activated by acidic oxidation to become an active HER electrocatalyst, and the catalytic performance of this catalyst could be further enhanced significantly by cathodic pretreatment. The authors also claimed that the HER activity of CNTs was correlated with the amount of surface acidic groups (‐COOH), which could act as proton relays. The nanocarbon catalyst reported here provides an illuminating example for creating low‐cost, metal‐free HER catalysts.

1.3.2 Oxygen Evolution Reaction

OER is a more challenging reaction for the design of advanced electrocatalysts. Luckily, many researchers have devoted much effort to develop metal‐free, carbon‐based electrocatalysts for OER. Dai and coworkers [39] have reported that 3D N,P‐co‐doped mesoporous nanocarbon foams can act as efficient ORR‐OER bifunctional catalysts. Their calculations revealed that the N,P co‐doping and the graphene edge effect are crucial for the bifunctional electrocatalytic activities. Wang and coworkers [40] have reported that edge‐selective P doping of graphene (G‐P) was synthesized for OER electrocatalysts [40]. The doped P plays an important role in catalyzing the OER.

In particular, being low‐cost, metal‐free catalysts for ORR and OER, nitrogen‐doped carbon nanomaterials have gained significant popularity recently. In the applications of nanocarbon materials, both as supports for metal nanoparticles or as metal‐free OER catalysts themselves, the coexistence of oxygen‐containing groups on the carbon materials is almost inevitable. This is because the oxidation of the nanocarbon substrates is usually required to introduce oxygen‐containing groups and defect sites for subsequent functionalization or doping with heteroatoms. However, the role of oxygen‐containing groups on the catalytic activity remains unclear. The unexpectedly high OER catalytic activity can be achieved using surface‐oxidized and electrochemically activated multiwalled carbon nanotubes (MWCNTs). Unlike previously reported carbon‐based catalysts, the oxidized MWCNTs consisted of only C and O on the surface, which enables us to study the role of O in catalyzing OER. This study reveals that the significant role of oxygen‐containing groups, especially ketonic (CO) groups, as the catalytic active sites for OER. Hence, by rational design, carbon nanomaterials bearing oxygen‐containing groups can be utilized as a new category of efficient OER catalysts.

Dai et al. reported a template‐free method for the scalable fabrication of three‐dimensional N‐ and P‐co‐doped mesoporous nanocarbon (NPMC) foams, simply by pyrolysis of polyaniline (PANi) aerogels synthesized in the presence of phytic acid. The resultant NPMCs show bifunctional catalytic activities toward ORR and OER (Figure 1.6). As a result, Zn–air batteries fabricated with NPMCs show good performance and long‐term stability. Typically, our primary battery based on the NPMC metal‐free air electrode operating in ambient air with an aqueous KOH electrolyte exhibited an open‐circuit potential of ∼1.48 V, an energy density of ∼835 Wh kgZn−1, and a peak power density of ∼55 mW cm−2, as well as excellent durability (over 240 h of operation after two mechanical recharges). A three‐electrode rechargeable battery using two NPMC metal‐free air electrodes to separate ORR and OER also exhibited good stability (600 cycles for 100 h of operation). First‐principles simulations revealed that the N and P co‐doping and the highly porous network of the carbon foam are crucial to generating bifunctional activity toward both ORR and OER. We anticipate that our nanomaterial will also be useful for other electrocatalytic applications.

Image described caption and surrounding text.

Figure 1.6 (a) Schematic illustration of the preparation process for the NPMC foams. An aniline (i)–phytic acid (ii) complex (iii) is formed (for clarity, only one of the complexed anilines is shown for an individual phytic acid), followed by oxidative polymerization into a three‐dimensional PANi hydrogel cross‐linked with phytic acids. As each phytic acid molecule can complex with up to six aniline monomers, phytic acid can be used as the cross‐linker and protonic dopant to directly form the three‐dimensional PANi hydrogel network; for clarity, only a piece of the two‐dimensional network building block is shown in the enlarged view under the three‐dimensional PANi hydrogel. The PANi hydrogel is freeze‐dried into an aerogel and pyrolyzed in Ar to produce the NPMC (for clarity, only a piece of the two‐dimensional NPMC network building block is shown in the enlarged view under the three‐dimensional NPMC). (b, c) SEM images of PANi aerogel (b) and NPMC‐1000 (c). Inset in (c): digital photo images of PANi aerogel before (left) and after (right) pyrolysis at 1000 °C. (d, e) High‐resolution TEM image (d) and TEM image (e, left), with corresponding element‐mapping images of NPMC‐1000 (e). The TEM image shows a piece of interconnected network‐like scaffold. The element‐mapping images for C, N, and P show a uniform distribution of the elements [39].

Source: Copyright 2015, Nature Publishing Group.

Wang and coworkers [40] have reported oxygen‐functionalized graphene on the surface of carbon fibers with Ar plasma treatment. Compared with pristine CC, the plasma‐etched carbon cloth (P‐CC) has a higher specific surface area and an increased number of active sites for OER. P‐CC also exhibits good intrinsic electron conductivity and excellent mass transport. DFT calculations predicted that P‐CC had an impressive electrocatalytic performance that was comparable with that of Ru‐based catalysts for OER, as a result of both defects and oxygen doping. The edge carbon, O content, and defects induced by Ar plasma play an important role in OER. Yao and coworkers [8] reported the assembly of a 2D graphene material possessing carbon defects (DG) obtained via a facile nitrogen removal procedure from a N‐doped precursor. The theoretical calculations support this hypothesis and the experimental work confirms the triple functions of the DG material. The simple approach we utilized to synthesize a triple‐functional material possesses overwhelming advantages over the current complicated fabrication process of bifunctional catalysts (multistep co‐doping and hybridization). To summarize, the defects derived by the removal of heteroatoms from graphene have been demonstrated, both experimentally and theoretically, to be effective for all three basic electrochemical reactions, e.g. ORR, OER, and HER. The activities of the DG for all the three reactions are much better than the N‐doping graphene.

1.4 Carbon‐Based Electrocatalysts for All‐Vanadium Redox Flow Battery

Redox flow batteries (RFBs) are a kind of new energy device. Owing to low cost, high performance, and long life, RFBs attract a lot of attention. As electrode, graphite felts are common materials. However, the low chemical activity limits the performance to improve. RFB mainly includes VRFBs, Li‐ion RFBs, and lead acid RFBs. Among these RFBs, due to active species existing in the electrolyte, the species do not undergo reaction with electrode in VRFBs. So VRFBs have the feature of long life. Although RFBs is a kind of promising energy storage device, it has faced many challenges in development.

Recently, VRFB has been widely reported in some studies. Owing to low cost, carbon nanomaterials have been investigated as electrode materials or catalysts for electrochemical activity in the VRFB system to enhance battery performance. The class of 1D carbon nanomaterials can be further subdivided into the class of carbon nanofibers (CNFs) and CNTs, in which the distinct difference lies in the configuration of the underlying graphene planes created by the alignment of carbon atoms. CNFs have a unique morphology in that the graphene layer is tilted against fiber axis, resulting in exposed edge planes toward exterior surface providing active sites for ionic adsorption and chemical bonding directly with the reactants. CNTs, on the other hand, comprised concentric cylinders of a graphene sheet consisting of relatively inert basal planes, which has unique properties, such as good electric conductivity and chemical resistance to acids and bases. In these regards, simultaneous usage of both CNTs and CNFs could provide the best electrocatalytic performance. The CNF/CNT composite catalyst on the carbon felt (CF) surface for high‐performance VRFB was reported. This composite catalyst facilitates the electron and mass transfer kinetics resulting in increased battery efficiency at a high rate by lowering the overpotential for vanadium redox reaction [1]. Similarly, MWCNTs and functional MWCNTs were used as an electrode reaction catalyst and their performances were investigated for VO2+/VO2+ redox couple for VRFB [41].

To achieve these improvements, a unique hierarchical structure was prepared by growing a uniform and dense layer of 3D graphene on CF via a simple microwave plasma‐enhanced chemical vapor deposition process. It was observed that the carbon fibers in CF were successfully wrapped by vertically grown graphene nanowalls, which not only increase the electrode‐specific area but also exposing a high density of sharp graphene edges with good catalytic activities to the vanadium ions. It was also found that the graphene‐coated CF electrode exhibited excellent stability in the battery operation [42].

Recently, it has been widely accepted that surface‐active oxygen functional groups are effective in catalyzing redox reactions of active species and improving the wettability of CF. In order to populate the surface of CFs with these functional groups, many researchers have used various surface modification techniques, including electrochemical oxidization, heat treatment, plasma treatment, and irradiation treatment. Most of these approaches, however, suffer from long processing time, processing difficulty, and high production costs. Thus, it is necessary to develop a new, simple, and cost‐effective surface modification process, allowing for the formation of abundant and robust oxygen functional groups [43].

In some studies, a new treatment with mixed acid was reported under ultrasonication for highly effective and fast hydroxylation of carbon fibers. It is found that the AOH groups can be successfully introduced onto the surface of carbon fibers after the treatment with the mixed acid. The characterizations of electrochemical activities of the hydroxylated carbon fibers toward the redox reactions of V(II)/V(III) and V(IV)/V(V) indicate that the highly hydroxylated carbon fibers show high catalytic activity [44].

Atmospheric pressure plasma jets (APPJs) technology can be a rapid, cost‐effective, and large‐area‐compatible process for the surface modification of graphite felts to improve the energy efficiency of VRFBs. VRFB performance is examined and the implications of the APPJ‐based process are discussed as well [8]. The oxygen‐containing functional groups can enhance the performance of VRFB. However, what is the function of different groups is not solved. In a study, the GO was loaded on the surface electrode and then reduced under different negative potentials in the phosphate buffer solution (PBS) for a certain time to obtain the electrochemically reduced graphene oxide (ERGO) with different C:O ratios. The electrochemical results show the preferable degree of reduction for the electrode catalyst and the role played by oxygen functional groups on the surface of the ERGO during the redox reactions for the VRFB. This simple and environmental friendly approach also provides a new and an effective way to modify the carbon‐based materials used in VRFB, which could lead to some good properties unattainable by unmodified carbon‐based materials [45].

On the other hand, nitrogen‐doped carbon materials including nitrogen‐doped mesoporous carbon (N‐MPC) and nitrogen‐doped graphene foam (N‐GF) have been discovered as electrode materials for VRFBs. Compared with N‐MPC and N‐GF, NCNTs maybe more interesting due to their unique physical and chemical properties such as excellent electronic conductivity and high thermal/chemical stability. Accordingly, NCNTs have been recognized as advanced electrode materials in the energy storage field with improved activity and stability. In addition, the growth methods of NCNTs have been well established, which ensures large‐scale production of NCNTs as electrode materials for energy devices.

In some studies, the successful growth of NCNTs on graphene foam (GF) by a chemical vapor deposition (CVD) method is reported. The ideal porous GF substrate provides an excellent porous framework for the growth of CNTs (undoped and doped with nitrogen). The CNTs (or NCNTs) on the surface of GF could significantly increase the electrochemical surface area of the carbon materials due to their relatively small size, resulting in higher battery performance in VRFBs. The enriched porous structures of CNTs or NCNTs on GF could potentially facilitate the diffusion of electrolyte in the VRFB system. In addition, the N doping could further improve the electrode performance because of the modified electronic and surface properties of CNTs on GF [46].

Similarly, co‐doping is also an important idea. The N,O‐co‐doped CF is successfully prepared by plasma treatment as electrodes in all‐VRFBs. The N,O‐co‐doped CF was obtained by treating the CF with mixed N2 and O2 plasma. Through the plasma modification, N and O atoms could be successfully doped into the lattice of carbon species of CF. The N,O‐co‐doped CF has greatly improved the electrochemical performance of the battery due to the modified electronic properties, the enhanced affinity with electrolyte, and thus the improved electrocatalytic activity by the heteroatoms [47].

NCNT/graphite felts as electrode materials for VRFBs show excellent performance. However, except for nitrogen‐doped carbon, graphitic carbon nitride (C3N4) with a graphite‐like structure has drawn plenty of scientific interest due to its excellent chemical and thermal stability, high in‐plane nitrogen content, and appealing electronic structure. C3N4 has been widely investigated for a range of applications such as photocatalysis, CO2 reduction, electrocatalysis, and bio‐imaging applications. To the best of our knowledge, there are not any reports on the use of this material for VRFBs. Therefore, it is of essential interest to observe the electrocatalytic behavior of graphitic C3N4 for V(IV)/V(V) and V(II)/V(III) redox reactions and thus for VRFBs [48].

References

  1. 1 Hong, W.T., Risch, M., Stoerzinger, K.A. et al. (2015). Energy Environ. Sci. 8: 1404–1427.
  2. 2 Wai, R.‐J., Lin, C.‐Y., Duan, R.‐Y., and Chang, Y.‐R. (2007). IEEE Trans. Ind. Electron. 54: 354–364.
  3. 3 (a) Zhang, L., Niu, J., Li, M., and Xia, Z. (2014). J. Phys. Chem. C 118: 3545–3553.(b) Yu, L., Pan, X., Cao, X. et al. (2011). J. Catal. 282: 183–190.
  4. 4 Rossmeisl, J., Logadottir, A., and Nørskov, J.K. (2005). Chem. Phys. 319: 178–184.
  5. 5 Gong, K., Du, F., Xia, Z. et al. (2009). Science 323: 760–764.
  6. 6 Wang, S., Yu, D., and Dai, L. (2011). J. Am. Chem. Soc. 133: 5182–5185.
  7. 7 Shen, A., Zou, Y., Wang, Q. et al. (2014). Angew. Chem. Int. Ed. 126: 10980–10984.
  8. 8 Jia, Y., Zhang, L., Du, A. et al. (2016). Adv. Mater. 28: 9532–9538.
  9. 9 Guo, D., Shibuya, R., Akiba, C. et al. (2016). Science 351: 361–365.
  10. 10 Kondo, T., Casolo, S., Suzuki, T. et al. (2012). Phys. Rev. B 86: 035436.
  11. 11 Metiu, H., Chrétien, S., Hu, Z. et al. (2012). J. Phys. Chem. C 116: 10439–10450.
  12. 12 Yang, L., Jiang, S., Zhao, Y. et al. (2011). Angew. Chem. Int. Ed. 123: 7270–7273.
  13. 13 Zhang, M., Yuan, W., Yao, B. et al. (2014). ACS Appl. Mater. Interfaces 6: 3587–3593.
  14. 14 Wang, S., Yu, D., Dai, L. et al. (2011). ACS Nano 5: 6202–6209.
  15. 15 Dou, S., Shen, A., Tao, L., and Wang, S. (2014). Chem. Commun. 50: 10672–10675.
  16. 16 Zhao, Z., Zhang, L., and Xia, Z. (2016). J. Phys. Chem. C 120: 2166–2175.
  17. 17 Jin, H., Huang, H., He, Y. et al. (2015). J. Am. Chem. Soc. 137: 7588–7591.
  18. 18 Jiang, Y., Yang, L., Sun, T. et al. (2015). ACS Catal. 5: 6707–6712.
  19. 19 Tang, C., Wang, H.‐F., Chen, X. et al. (2016). Adv. Mater. 28: 6845–6851.
  20. 20 Tao, L., Wang, Q., Dou, S. et al. (2016). Chem. Commun. 52: 2764–2767.
  21. 21 (a) Ouyang, C., Wang, X., and Wang, S. (2015). Chem. Commun. 51: 14160–14163.(b) Dou, S., Tao, L., Huo, J. et al. (2016). Energy Environ. Sci. 9: 1320–1326.
  22. 22 Liang, Y., Li, Y., Wang, H. et al. (2011). Nat. Mater. 10: 780–786.
  23. 23 (a) Gao, M.‐R., Xu, Y.‐F., Jiang, J. et al. (2012). J. Am. Chem. Soc. 134: 2930–2933.(b) Xu, L., Jiang, Q., Xiao, Z. et al. (2016). Angew. Chem. Int. Ed. 128: 5363–5367.(c) Dou, S., Li, X., Tao, L. et al. (2016). Chem. Commun. 52: 9727–9730.(d) Ouyang, C., Wang, X., Wang, C. et al. (2015). Electrochim. Acta 174: 297–301.(e) Wu, J., Shao, Y., Wang, B. et al. (2016). Plasma Process. Polym. .(f) Hu, K., Tao, L., Liu, D. et al. (2016). ACS Appl. Mater. Interfaces .
  24. 24 Stephens, I.E. and Chorkendorff, I. (2011). Angew. Chem. Int. Ed. 50: 1476–1477.
  25. 25 Voiry, D., Fullon, R., Yang, J. et al. (2016). Nat. Mater. 15: 1003–1009.
  26. 26 Shi, Y. and Zhang, B. (2016). Chem. Soc. Rev. 45: 1529–1541.
  27. 27 Xie, J., Li, S., Zhang, X. et al. (2014). Chem. Sci. 5: 4615–4620.
  28. 28 (a) Li, J.‐S., Wang, Y., Liu, C.‐H. et al. (2016). Nat. Commun. 7: 11204.(b) Zou, X. and Zhang, Y. (2015). Chem. Soc. Rev. 44: 5148–5180.
  29. 29 Popczun, E.J., McKone, J.R., Read, C.G. et al. (2013). J. Am. Chem. Soc. 135: 9267–9270.
  30. 30 Sathe, B.R., Zou, X., and Asefa, T. (2014). Cat. Sci. Technol. 4: 2023–2030.
  31. 31 Jin, J., Pan, F., Jiang, L. et al. (2014). ACS Nano 8: 3313–3321.
  32. 32 Yang, Z., Yao, Z., Li, G. et al. (2012). ACS Nano 6: 205–211.
  33. 33 (a) Zheng, Y., Jiao, Y., Li, L.H. et al. (2014). ACS Nano 8: 5290–5296.(b) Zhang, J., Qu, L., Shi, G. et al. (2016). Angew. Chem. Int. Ed. 128: 2270–2274.
  34. 34 (a) Jiao, Y., Zheng, Y., Jaroniec, M., and Qiao, S.Z. (2014). J. Am. Chem. Soc. 136: 4394–4403.(b) Paraknowitsch, J.P. and Thomas, A. (2013). Energy Environ. Sci. 6: 2839–2855.
  35. 35 Zheng, Y., Jiao, Y., Ge, L. et al. (2013). Angew. Chem. Int. Ed. 52: 3110–3116.
  36. 36 Liang, J., Jiao, Y., Jaroniec, M., and Qiao, S.Z. (2012). Angew. Chem. Int. Ed. 51: 11496–11500.
  37. 37 Zheng, Y., Jiao, Y., Zhu, Y. et al. (2014). Nat. Commun. 5: 3783.
  38. 38 Cui, W., Liu, Q., Cheng, N. et al. (2014). Chem. Commun. 50: 9340–9342.
  39. 39 Zhang, J.T., Zhao, Z.H., Xia, Z.H., and Dai, L.M. (2015). Nat. Nanotechnol. 10: 444.
  40. 40 Xiao, Z., Huang, X., Xu, L. et al. (2016). Chem. Commun. 52: 13008–13011.
  41. 41 Li, W., Liu, J., and Yan, C. (2011). Carbon 49: 3463–3470.
  42. 42 Li, W., Zhang, Z., Tang, Y. et al. (2016). Adv. Sci. 3: 1500276.
  43. 43 Kim, K.J., Lee, S.‐W., Yim, T. et al. (2014). Sci. Rep. 4: 6906.
  44. 44 Yue, L., Li, W., Sun, F. et al. (2010). Carbon 48: 3079–3090.
  45. 45 Li, W., Liu, J., and Yan, C. (2013). Carbon 55: 313–320.
  46. 46 Wang, S., Zhao, X., Cochell, T., and Manthiram, A. (2012). J. Phys. Chem. Lett. 3: 2164–2167.
  47. 47 Huang, Y., Deng, Q., Wu, X., and Wang, S. (2017). Int. J. Hydrogen Energy 42: 7177–7185.
  48. 48 Huang, Y., Huo, J., Dou, S. et al. (2016). RSC Adv. 6: 66368–66372.
..................Content has been hidden....................

You can't read the all page of ebook, please click here login for view all page.
Reset