7.1 Introduction

The ever‐increasing human population and galloping dependency on advanced electronic devices have led to a seminal growth in global energy consumption [1]. Consequently, advancement in energy conversion and storage technologies are essential. Conventional fossil fuels that cover transportation and industrial sectors supply almost 90% of global energy consumption. The energy conversion through combustion of fossil fuels and other chemical energy components has been a major worldwide concern of environmental pollution [2, 3]. As an estimation, there was 17 TW worldwide energy consumption in 2013, which is expected to be at least doubled by 2050 [4]. Therefore, green and renewable energy sources are essential to address the challenges of fossil fuel consumption [5–7]. The clean and sustainable energy sources, including wind, geothermal, hydropower, and solar, all have limitations related to either their cost or technological drawbacks and application difficulties. Owing to these concerns, intense research attempts are being performed on developing highly efficient and environment‐friendly energy conversion and storage technologies, such as fuel cells, metal–air batteries, and water splitting. These technologies are fundamentally operated through various electrocatalyses [8–10].

Recently, oxygen reduction reaction (ORR), oxygen evolution reaction (OER), hydrogen evolution reaction (HER), carbon dioxide reduction reaction (CO₂RR), hydrogen peroxide reduction reaction (HPRR), and methanol/ethanol oxidation reaction (MOR/EOR) have attracted intense research attention for clean and renewable electrochemical energy storage and conversion applications. ORR is related to the fuel cells, whereas OER is crucial for metal–air batteries. The electrochemical or photoelectrochemical water‐splitting‐based hydrogen fuel generation is dictated by HER. The efficiency of these energy conversion techniques depends on the mechanism of catalytic activities of these reactions. The basic mechanism of different redox reactions could be understood by the electrochemical potential, nature of electrolyte, and the interfacial properties between electrodes and electrolytes. Different electrochemical potentials are generally identified through polarization curves, as depicted in Figure 7.1.

For an electrochemical reaction, the standard potential reflects the thermodynamic equilibrium potential for the reaction to take place. Practically, reactions are very unlikely to initiate at the equilibrium potential. The electrochemical reaction proceeds only after overcoming the activation energy barrier. As described in Figure 7.1, the onset potential is the lowest (for the anodic reaction) or the highest (for the cathodic reaction) potential for a reaction to occur at given electrode conditions [11], e.g. the onset potential is where the onset of reduction occurs (Figure 7.1). The overpotential (η) is the driving force for a specific reaction under practical conditions and is defined as the potential difference between the applied potential and the equilibrium potential. Ideally, a catalyst should be able to perform electrochemical reactions just above the equilibrium potential, with zero overpotential. But practically, this does not happen because of the binding energies of intermediates. Therefore, a minimal overpotential tending toward the equilibrium potential resembles a good electrocatalytic behavior. Under charge transfer limitations, the electrochemical reaction rate is expressed by Butler–Volmer equation as

7.1

where j is the current density, j⁰ is the equilibrium (η = 0) current density, α is anodic or cathodic transfer coefficient, n is the transferred electron number, F is the Faraday constant, R is the gas constant, and T is the temperature. A larger exchange current density (j⁰) implies a faster reaction rate. The trajectory of Equation 7.1 is presented in Figure 7.2a. The value of j⁰ can be significantly increased by fabricating anode and/or cathode with nanostructured materials with large active surface area keeping the same geometric area constant because j⁰ is normalized to mere geometric area. In this context, three‐dimensional (3D) isotropic or anisotropic nanostructured materials would be of interest.

At an equilibrium region (η∼0), Butler–Volmer equation can be simplified as a linear relationship between current density and overpotential as

7.2

However, for regions away from equilibrium (η ≫ 0), only one between cathodic and anodic terms presumes as the reverse reaction decays and Butler–Volmer equation is often replaced by Tafel equation [12], which is a linear relationship between η and log j as

7.3

Using Equation (7.3), the value of j⁰ is evaluated from Tafel plot, as shown in Figure 7.2b. The slope thus obtained, which is also called Tafel slope, is proportional to the anodic or cathodic transfer coefficient (α). The value of α generally reveals the possible reaction pathway and the rate‐determining steps. At overpotentials further away from equilibrium (η ⋙ 0), the reaction rate is only limited to the diffusion rate of the electroactive species or mass transport. Thus, the current density or rate of a reaction is evaluated by the mass transfer and does not change by increasing the overpotential (for example, ORR in Figure 7.1).

However, the current density may be increased eventually with a further increase in the overpotential, indicating the onset of parasitic reactions [13]. In this respect, the rotating disk electrode (RDE) is commonly used to understand the mechanism of reaction controlled by mass transport or diffusion. Using a RDE, the number of electrons transferred during the reaction is determined by Levich equation as

7.4

When the reactions are controlled through a combination of kinetics of mass transport and diffusion, the Koutecky–Levich equation comes forward, which is useful in estimating kinetic current density and the transferred electron number. Koutecky–Levich equation can be expressed as

7.5

Furthermore, a ring electrode (such as Pt or Au) is generally embedded around the disk electrode in a RDE as a second working electrode. This introduces a rotating ring disk electrode (RRDE), which allows to detect the by‐product of an electrochemical reaction [14, 15].

Therefore, several physical parameters, including the overpotential and exchange current density, are crucial to postmortem the electrocatalytic activity of electrocatalysts. Despite numerous reports of 2D electrocatalysts, 3D nanostructured electrocatalysts are very important for further advancement in this field. The 3D porous carbon framework could provide an ideal backbone support with a large surface area for catalytically active sites, a multidimensional conductive network for efficient electron transport, a large space to accommodate the electrolyte/reactant diffusion, and an excellent mechanical stability.

Among metal‐based catalysts, noble metals (e.g. platinum, iridium, and palladium) and their alloys have been used in electrochemical reactions [16]. Owing to the high cost, limited availability, low selectivity, poor durability, susceptibility to gas poisoning, and adverse environmental effects of metal‐based catalysts, however, soaring efforts are being carried out to replace them with cost‐effective alternatives. As a contemporary alternative, earth‐abundant, transition metal (Fe, Co, Zn, Sn, Ti, W, Mo, and Ni)‐based oxide, carbide, nitride, oxynitride, and carbonitride catalysts are being investigated [16–22]. Unfortunately, the applicability of such non‐precious metal‐ and metal‐oxide‐based catalysts has often been hindered by their low intrinsic conductivities due to interfacial complexity of the hybrid structures as well as their thermodynamic instability. For improving the transport conductivity, these transition metal‐based catalysts are generally supported by carbon‐based nanomaterials, such as graphene, carbon nanotubes, carbon nitride, fullerene, mesoporous carbon, activated carbon, and biomass‐derived carbon [22]. Therefore, despite numerous recent reports, the large‐scale commercial application of such electrocatalysts for renewable electrochemical energy technologies are often impeded [23].

A new class of catalyst based on heteroatom‐doped carbon nanomaterial was introduced in 2009, which could replace platinum to efficiently catalyze ORR in fuel cells [24, 25]. Since then, carbon‐based catalysts have increasingly been utilized as efficient catalysts due to their large surface area, excellent electrical and thermal conductivities, appreciable mechanical properties, and light weight [26]. These new metal‐free catalysts have been demonstrated to be efficient for OER [10, 27] and HER [28, 29] as well. They are also utilized for I⁻/I³⁻ reduction in dye‐sensitized solar cells [26], CO₂ reduction for various hydrocarbon‐based fuel production [27], biosensing as well as environmental parameter monitoring [10], and even for the effective production of commodity chemicals [26, 28, 29]. Recently, heteroatom‐doped, heteroatom‐co‐doped, or tri‐doped carbon nanomaterials were shown to act as efficient metal‐free bifunctional electrocatalysts for the ORR and OER in rechargeable metal–air batteries [22], the ORR and HER in regenerative fuel cells [30], and simultaneously for ORR, OER, and HER [31–33]. The catalytic performance of such heteroatom‐doped, carbon‐based catalysts is often comparable or even superior to noble‐metal‐based catalysts. Such an interesting performance has been attributed to the doping‐induced charge redistribution around the heteroatom dopants, modulation in transport properties, and sometimes synergistic effects in doped carbon materials. Figure 7.3 summarizes the revolution of heteroatom‐doped, carbon‐based, metal‐free catalysts [22].

Very recently, 3D carbon‐based, metal‐free catalysts for different electrochemical‐reaction‐based energy storage applications are on enormous attention [34–36]. Figure 7.4 presents the structural details of different carbon materials with their schematic and experimentally obtained images. The ideal 3D graphene‐CNT structure has not yet been experimentally realized, although extensive experimental efforts have been made to obtain a structure very close to the ideal 3D isotropic carbon structure, as shown in Figure 7.4 [40, 41]. 3D hierarchical materials are very promising due to the large specific surface areas for reaction, which are crucial for efficient energy storage in three dimensions. However, there are many critical issues to fabricate efficient 3D electrocatalysts. In this chapter, we discuss recent developments on heteroatom‐doped and undoped 3D carbon catalysts for various electrochemical reactions toward efficient energy conversion and storage, along with the elucidation of various factors influencing the overall performance of such 3D metal catalysts, and related existing challenges.

7.2 3D Carbon Catalysts for Oxygen Reduction Reaction (ORR)

ORR is the most important reaction for biological respiration and technologically crucial in energy‐converting systems, such as fuel cells and metal–air batteries [42–47]. In a fuel cell, conversion of direct and efficient electrochemical energy to electric power is easily attained with no greenhouse gas emission. Other than high‐temperature solid‐oxide fuel cells (SOFCs), polymer electrolyte membrane fuel cells (e.g. proton exchange membrane fuel cells, PEMFC) and direct methanol fuel cells (DMFCs) are well known owing to their lower operation temperature and longer life span, promising for next‐generation lightweight vehicles and portable electronics [48]. For both PEMFC and DMFC, the key component, a Pt‐containing catalyst, is anchored on a porous conductive material (e.g. carbon black) acting as an anode and cathode to catalyze hydrogen oxidation reaction (HOR) and alcohol oxidation reaction (AOR) or ORR, respectively. The sluggish kinetics for HOR or AOR at a low temperature (∼80–100 °C) and a high energy input for ORR often becomes crucial, resulting in a large polarization resistance.

Oxygen reduction is a multistep reaction related to either four‐electron (4e⁻) path to directly generate H₂O (in an acidic medium) and OH⁻ (in an alkaline medium) or two‐electron (2e⁻) process with H₂O₂ (in an acidic medium) and HO₂⁻ (in an alkaline medium) as the intermediate species, depending on the inherently electrocatalytic properties of catalysts. The detailed reaction mechanisms in different media with thermodynamic electrode potential at standard condition are presented below [49, 50].

ORR in acid media:

ORR in alkaline media:

A higher selectivity of 4e⁻ reduction indicates better electrocatalytic ORR efficiency of a given catalyst.

Among heteroatom‐doped 3D carbon (C; electronegativity, ε = 2.50) catalysts, nitrogen (N; ε = 3.07) doping in different carbon architectures, such as MWCNT/rGO hybrid, rGO‐based aerogels, hydrogels, xerogels, graphene foam, porous carbon (mesoporous and macroporous), carbon capsules, and polymer‐ and MOF‐derived carbon networks, has been reported [42, 51–86]. Carbon nitride (C₃N₄)‐based carbon composites are also popular in this context [83–87]. Sulfur (S; ε = 2.44)‐doped carbon aerogel and graphene foam have also been examined for ORR applications [88–90]. Co‐doping (among various heteroatoms, such as N,B (ε = 2.01); S,P (ε = 2.06); F (ε = 4.17); and O (ε = 3.50), of carbon structures are often performed to tune the bandgap and electrochemical catalytic activities. S/N [91–100], B/N [101–104], P/N [105–108], and F/N [109] as co‐dopants and N/P/O [110] as a tri‐dopant have recently been used for demonstrating the bifunctional and trifunctional carbon‐based electrocatalysts for energy applications.

The synthesis and fabrication of the above‐mentioned carbon electrodes generally involve the functionalization of carbon structures, followed by electrostatic self‐assembly, lyophilization, and pyrolysis. Pyrolysis is very important for improving the conducting properties of the 3D carbon network structure and for removing undesired chemical agents. Heteroatom doping is generally performed through mixing proper chemical with the carbon nanostructures before pyrolysis or during pyrolysis through feeding different gases at different temperatures. The obtained carbon structures are often processed through activation steps. The activation process is normally performed either by low‐temperature (∼100–200 °C) heating with KOH or H₂SO₄ solution (called chemical activation) or by pyrolysis at high temperature (∼700–1000 °C) with gases such as HN₃, N₂, H₂, and CO₂ (called gas activation). Activation process helps to open up nanopores, clean carbon surfaces by removing unnecessary chemical constituents, and increase active defect sites on the carbon lattice, leading to an improvement in electrocatalytic performance. With an increasing number of publications on the carbon‐based, metal‐free electrocatalysts, the main purpose has remained in increasing the specific surface area, improving the overall electrical conductivity of the obtained 3D structure and uplifting the density of intrinsic active sites. Sometimes, it is misunderstood that a simple agglomeration of 2D carbon nanoflakes can form a 3D electrode [39]. Such anisotropic 3D structure lacks in uniform conduction properties, and the pores are usually blocked due to random stacking of carbon layers. Therefore, a real isotropic 3D carbon hierarchy is still very rare. Many reports with 3D carbon electrodes with some extent of structural order can be found and are described below.

Nitrogen‐doped 3D carbon materials are very well known to have high ORR activity and conductivity together with good performance stability [26, 44]. For instance, N‐doped ordered mesoporous graphitic arrays have been demonstrated to exhibit high catalytic activity, high durability, and excellent methanol tolerance in alkaline solutions [57]. Figure 7.5 illustrates fabrication process for a highly conductive, ultralight (2.5 ± 0.5 mg cm⁻³), neat, and versatile nitrogen‐doped graphene nanoribbon (GNR) aerogel through hydrothermally unzipping MWCNTs. The obtained aerogel can act as a novel ORR catalyst with superb electrocatalytic activity and better stability than commercial Pt/C catalysts in an alkaline medium. The obtained N‐doped GNRs have shown remarkable ORR performance in acidic solutions [42]. N‐doped carbon nanofiber (N‐CNF)‐based aerogels have also demonstrated excellent ORR electrocatalysis [68]. N‐CNF aerogel was prepared by direct pyrolysis of a cheap, green, and mass‐producible biomass, i.e. bacterial cellulose, followed by NH₃ activation.

Figure 7.6 shows that the N‐CNF aerogel inherits the 3D nanofibrous network of bacterial cellulose and meanwhile possesses high density of N‐containing active sites (5.8 at.%) and large BET surface area (916 m² g⁻¹). Such N‐CNF aerogel exhibits superior ORR activity (onset potential of 0.85 V versus reversible hydrogen electrode (RHE)), high selectivity (electron transfer number of 3.97 at 0.8 V), and excellent electrochemical stability (97.5% retention after 10 000 potential cycles) in alkaline (0.1 M KOH) media (Figure 7.6). Such performance is better than that of NH₃‐treated carbon blacks, carbon nanotubes, and reduced graphene‐oxide‐based aerogels reported as metal‐free catalysts [68]. This N‐CNF aerogel has also demonstrated promising performance in 0.5 M H₂SO₄ media with 0.83 V onset potential.

3D cross‐linking hierarchically porous structure can be obtained through a modified classical carbon aerogel method using NH₃ as an activating agent [80]. Figure 7.7 presents the detailed synthesis process and its electrochemical properties. The structure was abbreviated as LHNHPC having specific surface area of 2600 m² g⁻¹, high content of N dopants (3.12 at.%), and is well compared with benchmark Pt/C catalyst. Comparatively, the LHNHPC exhibited similar electrocatalytic activity toward ORR, superior durability, and excellent methanol tolerance in basic media. Such excellent electrochemical properties of LHNHPC are mainly attributed to the synergistic contribution of its unique hierarchical pore structure, the rich N doping, and the large surface area [80].

The important reports on ORR performance using different N‐doped 3D carbon materials are summarized in Table 7.1. Sulfur has very similar electronegativity as carbon, and some researchers have shown superior ORR performance in S‐doped 3D aerogel [88] and graphene foam [89] than their undoped carbon structures (Table 7.1). Such performance was attributed to the thiophene‐like form of S within carbon lattice [89]. The hydrophobicity introduced by sulfur as thiophenic configuration is believed to be crucial for withdrawing oxygen from the electrolyte and its physical adsorption on the surface during ORR [87, 89]. Furthermore, the location of these groups in smaller pores seemed to have some effects on the physical adsorption process. There is also an indication that other bulky sulfur surface groups in configurations with oxygen (sulfoxide, sulfones, and sulfonic acids) located in larger pores (mesopores) attract an electrolyte with dissolved oxygen to the pore system [90]. Therefore, more research is required to explore the S‐doping effect on ORR activities in 3D hierarchical carbon materials.

There have been various strategies to incorporate different elements in 3D carbon materials for multifunctional electrocatalysis. Co‐doped carbon electrodes have already demonstrated similar or superior catalytic performance in comparison with metal‐based catalysts. For example, hollow‐structured 3D carbon sphere nanocomposites (N,S‐hcs) doped with nitrogen and sulfur were synthesized through a soft template method and high‐temperature treatment [95]. The N,S‐hcs nanomaterials demonstrated excellent catalytic activity for ORR. Figure 7.8 shows the synthesis and catalytic performance of N,S‐hcs. Co‐doping of mesoporous carbon with N/S and the related large surface area were attributed to the enhanced catalytic activity of the resulting materials. Moreover, the N,S‐hcs electrocatalysts possessed four‐electron oxygen reduction selectivity, long‐term durability, and high resistance to methanol poisoning, all of which represented improvements over the conventional Pt/C electrocatalyst.

Except N/S co‐doping, N/P‐, B/N‐, N/F‐, and P/C₃N₄‐co‐doped 3D carbon materials have also been fabricated and examined with a purpose to achieve superior ORR catalysts than Pt/C. Table 7.2 enlists all such efforts with their ORR performance. Among various co‐doped 3D carbon catalysts, N/S and N/P co‐doping seem to be comparatively favorable for ORR catalysis. Furthermore, three elemental doping (tri‐doped) in 3D carbon structure has also been studied [32, 33], although very rare, as ORR catalysts. Figure 7.9 shows another such study, in which a scalable method is introduced to obtain N–P–O tri‐doped freestanding 3D graphene foam using one‐pot red phosphorus‐assisted technique. In particular, the solid carbon precursor was gradually exfoliated through the slowly released gases (e.g. pH₃, H₂, and CO₂) and metallic K during the reaction [110]. As a result, a large amount of nanopore formation and compactness in 3D carbon structure were achieved.

The as‐fabricated 3D hierarchical porous graphene (3D‐HPG) was almost isotropic in nature having pores ranging from 50 nm to 1 μm. The 3D‐HPG was also utilized as the supercapacitor electrode and a metal‐free catalyst for ORR, offering ultrahigh‐specific capacitance of 426 F g⁻¹ (424 F cm⁻³), as well as excellent ORR catalytic performance in alkaline electrolyte with 0.93 V onset potential and 92% current density retention even after 10 000 s of reaction, whereas Pt/C had only 81% retention [110]. Furthermore, the assembled all‐solid‐state cell exhibited both high gravimetric (25.3 Wh kg⁻¹) and volumetric (25.2 Wh l⁻¹) energy density, which are among the highest values of the state‐of‐the‐art carbon‐only supercapacitors. Density functional theory (DFT) calculations indicated that N–P–O tri‐doping could significantly enhance the charge delocalization with benefited electrochemical activity.

It should be mentioned that pure carbon nanocages without any apparent dopants or physically adsorbed polyelectrolyte have performed good ORR performance. DFT calculations indicated that the high ORR activities intrinsically associated with the pentagon and zigzag edge defects [111]. Utilizing this concept, ORR catalyst based on graphene quantum dots supported by GNRs was developed through a one‐step reduction reaction, with ORR performance comparable or even better than that of a Pt/C electrode [112]. Such electrocatalytic performance has been attributed to the presence of surface and edge defects at the interfaces and edges of quantum dots and GNRs, together with efficient charge transfer through the intimate interfaces between quantum dots and GNRs. Such quantum defect and interface‐based strategy can also be adapted for 3D carbon structures without and with heteroatom doping to develop defect‐induced ORR catalysis with detailed mechanistic studies. In this context, 3D carbon materials derived from biomass (e.g. shrimp‐shell, chitin, and keratin) [113–116] and food‐grade (D‐glucose) [117] component have also been demonstrated to exhibit efficient ORR activities in microbial fuel cells [118].

7.3 Carbon‐Based 3D Electrocatalysts for Oxygen Evolution Reaction (OER)

OER produces molecular oxygen through several proton/electron‐coupled steps [119, 120], and is pH sensitive. In acidic and neutral electrolytes, two water molecules (H₂O) are oxidized into four protons (H⁺) and oxygen molecule (O₂), whereas, for basic solution, hydroxyl groups (OH⁻) are oxidized into H₂O and O₂. The equilibrium half‐cell potentials () at 1 atm and 25 °C for OER are shown below [121]:

• OER in alkaline solution:
  
• OER in acidic solution:
  

To avoid the pH influence on the applied potential and to maintain the working voltage around 1.23 V, reversible hydrogen electrode (RHE) is generally used as a reference [121]. The evolution of O₂ molecule needs transfer of four electrons, and favorable kinetics for OER process takes place in multistep reactions with single‐electron transfer at each step. Clearly, accumulation of energy at each step triggers sluggishness in OER kinetics, resulting in a large overpotential. Therefore, highly active OER electrocatalyst is required to overcome the energy barrier [122, 123]. For an ideal OER catalyst, related overpotential must be low with high catalytic stability, low cost, and earth‐abundance for large‐scale commercialization.

Table 7.3 summarizes heteroatom‐doped 3D carbon materials and their OER performance. Apart from N‐doped porous carbon [27] and C₃N₄/CNT composite catalysts [124], N/O‐co‐doped hydrogel [125] demonstrated a lower onset potential (1.28 V) for OER in 0.1 M KOH electrolyte, although N‐doped porous carbon exhibited a better OER catalytic stability [27]. In an acid medium (0.5 M H₂SO₄), N/O hydrogel showed better OER performance as compared with commercial IrO₂ electrode [125]. S/N‐co‐doped foam and rGO/CNT composite electrodes have also performed efficient OER in an alkaline medium (0.1 M KOH) even with a very low doping amount (0.5–1.26 at.%) due to synergistic effect, modification in electronic configuration in the catalysts, and modulation in conducting properties [126, 127]. However, the catalytic stability in S/N‐doped electrodes is worse as compared with N‐ or N/O‐doped, carbon‐based catalysts. In this context, P/N‐ [128] or P/C₃N₄ [129]‐co‐doped 3D carbon catalysts have demonstrated excellent OER stability (>93% current density retention) for up to 30 h continuous operation in an alkaline medium (0.1–1 M KOH). In particular, hierarchically porous P‐ and N‐co‐doped carbon nanofibers were prepared by directly growing polymer nanowires on conductive carbon paper through an electrochemically induced polymerization process in the presence of aniline monomer and phosphonic acid, followed by carbonization, as shown in Figure 7.10 [128]. The resultant carbon catalyst exhibited robust stability (little activity attenuation after 12 h continuous operation) and high activity with a low overpotential (310 mV at 10 mA cm⁻²) toward electrocatalytic oxygen production, comparable with that of the precious iridium oxide (IrO₂) benchmark [128]. Experimental measurements revealed that dual doping of N and P led to an increased active surface area and abundant active sites in comparison with the single‐doped and pristine carbon counterparts. Density functional theory calculations indicated that N and P dopants could coactivate the adjacent C atoms, inducing synergistically enhanced activity toward OER, as shown in Figures 7.10f–g [128]. The catalytic performances could be theoretically examined by calculating the free energy diagrams based on experimentally evidenced 4e⁻ transferring processes at different electrode potentials (Figure 7.10h). This implies that co‐doping of N and P into the carbon framework can significantly reduce the overall free energy relative to the single‐doped and pristine carbons. When the applied potential (U) is set to 0 V, the first three steps, corresponding to the electron transfer for adsorbing HO*, O*, and OOH*, respectively, are endothermic reactions, whereas the fourth step relating to desorption of O₂ is exothermic. The third electron transfer step with the highest reaction energy barrier is the rate‐determining step. However, when the potential is increased to 0.907 V (0.505 V in overpotential, Figures 7.10h,i), all the elementary reaction pathways become downhill and exothermic, and thus, OER can occur spontaneously. The OER overpotential is remarkably decreased via co‐doping, signifying that OER activity can be improved overall by N,P co‐doped [128].

Multifunctional electrocatalysts have attracted focused research attention due to inherent cost reduction using same catalyst for various redox reactions. In this context, bifunctional 3D carbon catalysts with ORR and OER activities have been summarized in Table 7.4. Chemically and electrochemically synthesized N‐, P‐, C₃N₄‐, B‐doped and N‐C₃N₄, P/C₃N₄‐, P/S‐, B/N‐, N/S‐co‐doped carbon 3D electrodes have demonstrated electrocatalytic activities for both ORR and OER (Table 7.4) [47, 58, 84, 129–138]. Along with recent attempts for conversion of metal–organic frameworks, metal‐based coordination compounds have also been researched for producing 3D carbon networks. In this context, a 3D interconnected hollow N‐doped graphene shells, consisting of three to four layers with small, uniform mesopores of c. 4 nm, were obtained from Ni(II) molecular coordination compound, as shown in Figure 7.11 [130]. The resultant 3D mesoporous graphene material exhibited superior bifunctional electrocatalytic activity and durability in ORR and OER, comparable with that of noble metal catalysts such as Pt/C and Ir/C [130]. Such high bifunctional activity of the 3D mesoporous graphene material was attributable to its very large surface area, numerous defect sites associated with its highly curved structure (Table 7.4). In addition, the presence of N and Ni species in the carbon framework might have also contributed to the high catalytic activity, showing promises for use in regenerative fuel cells as well as rechargeable metal–air batteries [130].

N‐doped mesoporous carbon nanosheet and MWCNT hybrid composite can be synthesized through hydrothermal method by the pyrolysis of glucose, urea, and CNTs for developing efficient bifunctional catalysts, as shown in Figure 7.12 [133]. The resulting hybrids exhibited remarkable OER activities with a low onset potential (1.50 V versus RHE) and an overpotential of only 320 mV at 10 mA cm⁻² in an alkaline medium (0.1 M KOH). The N‐MCN/CNT hybrids exhibited comparable catalytic performance, but better durability, compared with those of the Pt/C (20 wt%) catalyst for ORR. The demonstrated ultrahigh catalytic performance of the hybrids could originate from the large specific surface area (594.1 m² g⁻¹), high N content (10.7 at.%), and mesoporous structure for a full exposure of active sites, an improved capability for mass/electron transport, an easy adsorption/release of oxygen gas bubbles, and a high structural stability [133].

As can be seen above, C₃N₄‐based electrode performs well for both ORR and OER activities mostly in alkaline media (KOH). Heteroatom doping of C₃N₄ can further enhance the catalytic activities. Particularly, P‐doped C₃N₄ nanoflowers supported by a carbon fiber paper showed an ORR onset potential of −0.94 V and OER onset potential of 1.53 V [129]. Furthermore, P/S‐co‐doped C₃N₄ sponge sandwiched with C nanocrystals (P,S‐CNS) exhibited a high surface area (1474 m² g⁻¹) and superior ORR and OER bifunctional catalytic activities than those of Pt/C and RuO₂, respectively, in terms of the limiting current density and onset potential (Figures 7.13 and 7.14) [47]. The resultant electrode exhibited an excellent suitability and durability as the oxygen cathode for primary and rechargeable Zn–air batteries. The resulting primary Zn–air battery showed a high open‐circuit voltage of 1.51 V, a high discharge peak power density of 198 mW cm⁻², a specific capacity of 830 mAh g⁻¹, and a better durability for 210 h after mechanical recharging. A small charge‐discharge voltage polarization of 0.80 V at 25 mA cm⁻², good reversibility, and stability have also been observed for rechargeable Zn–air batteries in three‐electrode system. DFT analyses indicate that the excellent electrocatalytic performance of P,S‐CNS was attributable to the co‐doping‐induced efficient mass/charge transport and high conductivity of the C₃N₄/C composite, as indicated by energy diagrams in Figure 7.14 [47].

The recent development of flexible and wearable electronics requires flexible power sources. However, flexible electrode fabrication for ORR and OER applications is still a big challenge. In this regard, the successful fabrication of a flexible, large‐area 3D porous N‐doped carbon microtube (NCMT) sponge via a simple and low‐cost process by pyrolyzing cotton is promising [134]. Due to its unique structure with a micron‐scale hollow core and well‐graphitized and interconnected porous walls, the NCMT sponge exhibited incomparable electrocatalytic activity for the ORR and OER with a small potential difference of 0.63 V between the OER current density at 10 mA cm⁻² and the ORR current density at 3 mA cm⁻² [134].

7.4 Carbon‐Based 3D Electrocatalysts for Hydrogen Evolutions Reaction (HER)

Despite its excellence as an emission‐free fuel, hydrogen is primarily produced from steam reforming of natural gas – a process that produces carbon dioxide emissions. In contrast, electrochemical reduction of water can produce H₂ via HER with no harmful emission. However, Pt is required as the most active electrocatalyst for HER, and the high cost of Pt has hindered the large‐scale production of H₂ from HER. Therefore, it is important to develop low‐cost and highly efficient electrocatalysts for HER, and carbon‐based catalysts have been demonstrated to be a promising alternative. As a result, much effort has been devoted to fabricate 2D carbon‐based HER catalysts [139–142], although 3D carbon architectures for HER activities are emphasized in recent development [143–150]. Nevertheless, certain heteroatom‐doped carbon materials have recently been shown to exhibit high promising catalytic activity and stability toward HER in acidic, basic, and neutral conditions.

Generally speaking, there are different mechanisms, Volmer/Heyrovsky/Tafel, for HER as shown below [141]:

The most sensitive factor, which influences HER rate, is the electrochemical balance between adsorption and desorption known as the Sabatier principle, generally presented by a “volcano plot” [140–142]. In cases where the substrate interaction is too inferior, the Volmer reaction is inhibited. If it is well strong, however, the Tafel/ Heyrovsky reaction is hindered. Therefore, a well‐maintained sorption on the catalyst surface is important and the interfacial property between the catalyst and electrolyte plays a crucial role in regulating the HER process. Therefore, the relative adsorption free energy of the H* intermediate acts as an indicator of the catalytic activity and is normally used to evaluate HER performance for a catalyst. Conway and Tilak [140] proposed a model for inspecting the surface coverage associated with underpotential deposition of hydrogen and formulated the difference in the Tafel slope for various mechanisms.

Table 7.5 summarizes 3D carbon‐based, heteroatom‐doped catalysts for HER [144–150]. N‐doped, plasma‐etched 3D graphene has demonstrated promising HER performance with −0.06 V (versus RHE) onset potential in acid (0.5 M H₂SO₄) electrolyte medium [146]. C₃N₄ networks with and without heteroatom doping have also been tested for HER in both acid and alkaline media [147, 148, 150]. In 0.5 M H₂SO₄ electrolyte, graphitic C₃N₄ nanoribbon network showed HER onset potential of −0.11 V (versus RHE) (Figures 7.15a–k), whereas P‐doped graphene and C₃N₄ composite catalyst demonstrated an onset potential of about −0.076 V (versus RHE) [147, 150]. Of particular interest, porous C₃N₄ nanolayers with nitrogen‐doped graphene sheets demonstrated the lowest onset potential of −0.008 V (versus 0 mV for Pt/C, versus RHE at 0.5 mA cm⁻²) in 0.5 M H₂SO₄ electrolyte [148]. The preparation and electrochemical results for this flexible, 3D porous carbon catalyst are given in Figures 7.15l–p [148], which show a high exchange current density of 0.43 mA cm⁻², and remarkable durability (negligible activity loss >5000 cycles). Such excellent HER performance could stem from synergistic effects originating from (i) highly exposed active sites generated by introduction of in‐plane pores into C₃N₄ and exfoliation of C₃N₄ into nanosheets, (ii) hierarchical porous structure of the hybrid film, and (iii) 3D conductive graphene network [148]. Other 3D carbon HER catalysts include poly(3,4‐dinitrothiophene)/SWCNT composite [145] and N/P‐co‐doped carbon nanofiber network [149], as listed in Table 7.5 along with their analogs.

Along with the development of 3D carbon‐based bifunctional electrocatalysts for ORR–OER [10], bifunctional 3D carbon catalysts for ORR–HER or OER–HER have recently been under intensive investigation [151–154], as seen in Table 7.6. It is interesting to notice that activated carbon doped with N through pyrolysis in NH₃ environment demonstrated both HER (onset potential −0.34 V versus RHE) and ORR (onset potential 0.882 V versus RHE) activities and stability (above 90%) in either acid or alkaline medium [151]. Compared with the single N‐doped carbon ORR–HER bifunctional catalysts, however, N/P‐co‐doped 3D carbon network, obtained through pyrolysis of GO, melamine and phytic acid supermolecular aggregate at 1000 °C, performed better HER activities with −0.14 V (versus RHE) onset potential in 0.5 M H₂SO₄ electrolyte, whereas it displayed 0.92 V onset potential versus RHE in 0.1 M KOH electrolyte [30, 153].

Among 3D bifunctional catalysts for HER and OER, carbon fiber/CNT/C₃N₄ composite and N/P/O‐tri‐doped porous carbon materials are particularly promising [152, 154]. Figure 7.16 shows the synthetic scheme and electrochemical evaluation for homologous carbon‐based, metal‐free electrolyzer composed of C₃N₄–CNT–CF and S–C₃N₄–CNT–CF as OER and HER bifunctional catalysts in both basic and acidic media [152]. The well‐designed structure with the hierarchical 3D conductive CNT–CF network improved the charge transport ability and facilitated the electrolyte penetration of the electrodes. The high dispersion and longtime stability of layered C₃N₄ ensured the excellent OER electrocatalytic activities (1.52 V onset potential versus RHE in 1 M KOH), and subsequent sulfur doping improved the HER performance of the C₃N₄‐based electrode with −0.15 V onset potential versus RHE in 0.5 M H₂SO₄ electrolyte [152]. This 3D metal‐free bifunctional carbon catalyst exhibited high catalytic ability, low onset potential, and longtime duration in various pH environments, thus promoting self‐supported metal‐free systems for efficient, low‐cost, and environment‐friendly water splitting [152]. The N/P/O‐tri‐doped, carbon‐based catalyst performed HER and OER activities both in acid (0.5 M H₂SO₄) and alkaline (1 M KOH) media, as seen in Table 7.6 [154]. Compared with bifunctional electrocatalysts, the development of tri/multifunctional carbon catalysts is still in infancy. However, continued research in this area would be of value.

7.5 Carbon‐Based 3D Electrocatalysts for Carbon Dioxide Reduction Reaction (CO₂RR)

Electrochemical conversion of CO₂ into energy‐rich fuels and chemicals has recently attracted intensive research interest. In this reduction process, CO₂ is recycled and closed loop of fuel combustion and waste CO₂ reduction mitigate greenhouse gas in the atmosphere. At the same time, intermittent electricity generation can be stored in an energy‐dense, portable form in chemical bonds. However, the structural stability of CO₂ molecules makes its conversion kinetically challenging, which requires a large overpotential. Therefore, the related electrochemical reduction efficiency strongly depends on the activity and selectivity of the cathodic electrocatalyst. Numerous heterogeneous catalysts have been examined with an emphasis to increase catalytic activity and product selectivity, although the CO₂ reduction mechanism remains not fully understood. Figure 7.17 indicates the increasing trend in electrochemical and photoelectrochemical CO₂ reduction efforts [155, 156].

The thermodynamic redox potentials for different possible CO₂ reduction half reactions leading to different products are given in Table 7.7. These half reaction potentials only reflect the minimum thermodynamic potential to enable the reaction and depend on the electrolyte media as well. However, the reaction kinetics, including the activation energy, reaction rate, and mechanistic pathway cannot be forecast by the thermodynamic potentials alone.

Generally speaking, CO₂ reduction competes with the HER, which also occurs at a comparable thermodynamic potential. Therefore, ideal electrocatalysts for CO₂ reduction should minimize the activation barrier for CO₂RR relative to HER, driving CO₂ reduction selectively at low overpotential with high reaction rates (i.e. currents) [155–158]. The mechanism and relative energy requirement of photocatalytic CO₂RR on a semiconductor photocatalyst surface can be well understood, as described in Figure 7.18.

In comparison with bulk metal (e.g. Ag) catalysts, carbon nanofibers comprising N‐doped graphitic layers showed more than an order of magnitude higher CO₂ reduction current density under identical conditions (Figures 7.19a–g) [31]. It has been demonstrated that CO₂RR occurs primarily at reduced carbon sites rather than the electronegative nitrogen atoms in these heteroatom‐doped graphitic catalysts. Another key advantage for the N‐doped carbon nanofibers is its extended stability without significant decay in the catalytic activity even in acidic media, as evidenced by the long‐term performance and unchanged density of pyridinic nitrogen atoms. In another study, N‐doped CNTs exhibited a very low overpotential (0.18 V) for CO₂RR with a maximum faradaic efficiency for CO of 80% at −1.05 V versus the cell potential [160] and was stable even after 10 h continued operation. The observed outstanding CO₂RR activity was attributed to the high electrical conductivity of N‐doped CNTs, in which the pyridinic‐N atoms increased local active sites and lowered the free energy for CO₂ activation due to a lone‐pair of electrons at pyridinic defects [160]. Alternatively, CO₂RR can also be performed with boron‐doped graphene [161] or N‐doped CNTs decorated with polyethylenimine macromolecules, which acted as a cocatalyst and reduced CO₂ to formate at high current density and selectivity at low overpotentials [162]. N‐doped nanodiamond has even been tried as CO₂RR catalyst to produce acetate, preferentially over formate, with an onset potential of −0.36 V versus RHE and up to ∼78% faradaic efficiency [163]. These results indicate the applicability of nonmetallic catalysts for the formation of higher (C₂⁺) carbon products by CO₂RR.

Furthermore, S‐doped (CPS) and S,N‐co‐doped polymer‐derived 3D carbon (CPSN) catalysts were also examined for CO₂RR in aqueous solutions, in which CPSN showed better catalytic performance compared with CPS for the formation of CO (11.3%) and CH₄ (0.18%) at a potential of −0.99 V (versus RHE) (Figures 7.19h–o) [159]. It was revealed that the better catalytic activity of CPSN is due to increase in active sites (nitrogen‐based) and the basic surface. It has been proposed that the pyridinic group (negative charge) facilitated electron transfer for conversion of CO₂ to COOH intermediate, which underwent further reduction to form CO [159]. As far as the published work on CO₂RR is concerned, it would be unfair to comment that not much has yet been achieved in this area. There remains, however, much work to do.

7.6 Carbon‐Based 3D Electrocatalysts for H₂O₂ Reduction (HPRR)

Hydrogen peroxide (H₂O₂) is becoming significant due to its potential applications in various sectors, including analytical chemistry, biomedical sensing, and energy conversion. Despite few reports, proper understanding and development of efficient carbon‐based catalysts for HPRR are still at nascent stage [164–169]. So far, the catalysts developed for HPRR are mainly focused on nitrogen‐doped and/or N/S‐co‐doped graphene materials [164, 168, 169]. Figure 7.20 shows the morphology and electrochemical performance for a metal‐free “graphene alloy” developed from reduced graphene oxide (NS‐rGO) co‐doped with N/S via a simple one‐step thermal annealing procedure using a mixture of 5‐amino‐2‐mercapto‐1,3,4‐thiadiazole (AMT) and GO [169]. The obtained metal‐free NS‐rGO composite showed better electrocatalytic activity toward the HPRR compared with rGO. The enhanced performance was attributed to the synergistic effect of N and S co‐doping and high electrical conductivity [169]. The DFT investigation on N graphene for HPRR has revealed that adsorption of H₂O₂ on the N graphene and pristine graphene surface was classified as physisorption and occurred through a weak interaction without the formation of a chemical bond [168]. When H⁺ was introduced into the system, a series of electron transformation processes for the H₂O₂ reduction reaction occurred spontaneously, including the breakage of the OO bond, the formation of an OC chemical bond between oxygen and graphene, and the creation of water molecules. The calculations of the relative energy for each of the reaction steps and the value of the onset potential for H₂O₂ reduction reaction suggested that the reactivity of the pristine and N‐doped graphene has the following order: pyridinic‐N graphene > pyrrolic‐N graphene > graphitic‐N graphene > pristine graphene [168]. Figure 7.20f shows the relative energy levels and configurations of the reaction intermediates for HPRR onto N graphene. The overall HPRR can be expressed by the following equation [168]: H₂O₂ + 2H⁺ + 2e⁻ → 2H₂O (1.763 V versus SCE).

7.7 Conclusions and Perspectives

As can be seen from the above discussions, various heteroatom‐doped or co‐doped or even tri‐doped carbon‐based, 3D catalysts with excellent electrocatalytic properties have been developed to replace noble or non‐precious metal‐based electrocatalysts for various redox reactions important to energy conversion and storage, such as ORR, OER, HER, CO₂RR, and HPRR. Heteroatom‐doping‐induced enhancement in electrochemical efficiency of these catalysts can be attributed to doping‐induced synergistic effects, improvement in conducting and transport properties, modulation in electronic band structure, charge transfer, and also improvement in interfacial interaction between the catalyst surface and electrolyte/reactant molecules. Most of the reported carbon‐based 3D catalysts are simple agglomeration/composite of 2D carbon allotropes, including graphene oxide, graphene flakes, CNT/buckypapers, carbon fiber, nanoribbons, and onion‐like carbons, without structural continuity and mechanical stability. In this respect, isotropic 3D carbon structures with hierarchically porous architectures are attractive for electrocatalytic applications, although they are difficult to make. The 3D porous carbon framework could not only provide an ideal backbone support with a large surface area for catalytically active sites and a multidimensional conductive network for efficient electron transport but also provide a large space to accommodate the electrolyte/reactant diffusion and an excellent mechanical stability. Templated chemical vapor deposition or metal‐oxide framework (MOF)‐derived 3D carbon materials could perform attractive electrochemical activities; however, their abundant structural interfaces increase the overall resistance within the fabricated 3D carbon catalyst electrodes and the related mechanical reliability is often poor. Therefore, much more research effort is required to design and develop cost‐effective 3D carbon‐based electrocatalysts with well‐defined structures and properties toward a commercial success for clean and renewable energy technologies, which has long been hindered by the production cost, scarcity of precious and even non‐precious metal catalysts, and the intrinsically associated environmental concerns. With all the high‐performance carbon‐based, metal‐free electrocatalysts already reviewed in this chapter, and many more to be developed, we are confident about their widespread applications for energy conversion and storage in near future with commercial success.

Acknowledgments

We thank our colleagues for their contributions to the work cited. We are also grateful for financial support from NSF (CMMI‐1400274, DOD‐AFOSR‐MURI (FA9550‐12‐1‐0037), NASA (NNX16AD48A, NNC16CA42C), Key Program of National Natural Science Foundation of China (51732002, 51433005), and The National Key Research and Development Program of China (2017YFA0206500).

References

1. 1. Zhu, Y., Zhou, W., Yu, J. et al. (2016). Chem. Mater. 28: 1691–1697.
2. 2. Nakayama, M., Fujii, Y., Fujimoto, K. et al. (2016). RSC Adv. 6: 23377.
3. 3. Yeo, B.S. and Bell, A.T. (2011). J. Am. Chem. Soc. 133: 5587–5593.
4. 4. Moniz, S.J.A., Shevlin, S.A., Martin, D.J. et al. (2015). Energy Environ. Sci. 8: 731.
5. 5. Solomon, S., Plattner, G.K.G., Knutti, R., and Friedlingstein, P. (2009). Proc. Natl. Acad. Sci. U.S.A. 106: 1704–1709.
6. 6. Primio, R.D., Horsfield, B., and Vega, M.G. (2000). Nature 406: 173–176.
7. 7. Maeda, K. and Domen, K. (2010). J. Phys. Chem. Lett. 1: 2655–2661.
8. 8. Cheng, F.Y. and Chen, J. (2012). Chem. Soc. Rev. 41: 2172–2192.
9. 9. Li, Y.G. and Dai, H.J. (2014). Chem. Soc. Rev. 43: 5257–5275.
10. 10. Zhang, J., Zhao, Z., Xia, Z., and Dai, L. (2015). Nature Nanotech. 10: 444–452.
11. 11. Zhang, J., Li, H., Guo, P. et al. (2016). J. Mater. Chem. A 4: 8497–8511.
12. 12. Shinagawa, T., Garcia‐Esparza, A.T., and Takanabe, K. (2015). Sci. Rep. 5: 13801.
13. 13. Lavacchi, A., Miller, H., and Vizza, F. (2013). Nanotechnology in Electrocatalysis for Energy, 25–61. New York: Springer.
14. 14. Li, Y., Zhou, W., Wang, H. et al. (2012). Nat. Nanotechnol. 7: 394–400.
15. 15. Wang, D.W. and Su, D. (2014). Energy Environ. Sci. 7: 576–591.
16. 16. Shao, M., Chang, Q., Dodelet, J.P., and Chenitz, R. (2016). Chem. Rev. 116: 3594–3657.
17. 17. Ohmori, T., Takahashi, H., Mametsuka, H., and Suzuki, E. (2000). Phys. Chem. Chem. Phys. 2: 3519–3522.
18. 18. Kanan, M.W. and Nocera, D.G. (2008). Science 321: 1072–1075.
19. 19. Goux, A., Pauporte, T., and Lincot, D. (2006). Electrochim. Acta 51: 3168–3172.
20. 20. Seaboldand, J.A. and Choi, K.S. (2011). Chem. Mater. 23: 1105–1112.
21. 21. Li, X., Walsh, F.C., and Pletcher, D. (2011). Phys. Chem. Chem. Phys. 13: 1162–1167.
22. 22. Dai, L. (2017). Cur. Opin. Electrochem. doi: https://doi.org/10.1016/j.coelec.2017.06.004.
23. 23. Wu, J., Xue, Y., Yan, X. et al. (2012). Nano Res. 5: 521–530.
24. 24. Gong, K., Du, F., Xia, Z. et al. (2009). Science 323: 760–764.
25. 25. Shui, J., Wang, M., Du, F., and Dai, L. (2015). Sci. Adv. 1: e1400129.
26. 26. Liu, X. and Dai, L. (2016). Nat. Rev. Mater. 1: 16064.
27. 27. Zhao, Y., Nakamura, R., Kamiya, K. et al. (2013). Nat. Commun. 4: 2390.
28. 28. Zheng, Y., Jiao, Y., Li, L.H. et al. (2014). ACS Nano 8: 5290–5296.
29. 29. Liu, J., Liu, Y., Liu, N. et al. (2015). Science 347: 970–974.
30. 30. Zhang, J., Qu, L., Shi, G. et al. (2016). Angew. Chem. Int. Ed. 55: 2230–2234.
31. 31. Kumar, B., Asadi, M., Pisasale, D. et al. (2013). Nat. Commun. 4: 2189.
32. 32. Zhang, J. and Dai, L. (2016). Angew. Chem. Int. Ed. 55: 13296–13300.
33. 33. Hu, C. and Dai, L. (2017). Adv. Mater. 29: 1604942.
34. 34. Shehzad, K., Xu, Y., Gao, C., and Duan, X. (2016). Chem. Soc. Rev. 45: 5541–5588.
35. 35. Wu, Z.Y., Liang, H.W., Chen, L.F. et al. (2016). Acc. Chem. Res. 49: 96–105.
36. 36. Zhang, J., Zhao, F., Zhang, Z. et al. (2013). Nanoscale 5: 3112–3126.
37. 37. Pawlak, R., Kawai, S., Fremy, S. et al. (2012). J. Phys. Condens. Matter 24: 084005.
38. 38. Odom, T.W., Huang, J.L., Kim, P., and Lieber, C.M. (1998). Nature 391: 62–64.
39. 39. Yu, T.Q., Jauregui, L.A., Wu, W. et al. (2011). Nat. Mater. 10: 443–449.
40. 40. Du, F., Yu, D., Dai, L. et al. (2011). Chem. Mater. 23: 4810–4816.
41. 41. Xue, Y., Ding, Y., Niu, J. et al. (2015). Sci. Adv. 1: 1400198.
42. 42. Chen, L., Du, R., Zhu, J. et al. (2015). Small 11 (12): 1423–1429.
43. 43. Zheng, Y., Jiao, Y., Jaroniec, M. et al. (2012). Small 8 (23): 3550–3566.
44. 44. Dai, L., Xue, Y., Qu, L. et al. (2015). Chem. Rev. 115: 4823–4892.
45. 45. Zheng, Y., Liu, J., Liang, J. et al. (2012). Energy Environ. Sci. 5: 6717–6731.
46. 46. Tang, J., Liu, J., Torad, N.L. et al. (2014). Nano Today 9: 305–323.
47. 47. Shinde, S.S., Lee, C.H., Sami, A. et al. (2017). ACS Nano 11 (1): 347–357.
48. 48. Andersson, M., Beale, S.B., Espinoza, M. et al. (2016). Appl. Energy 180: 757–778.
49. 49. Yeager, E. (1986). J. Mol. Catal. 38: 5–25.
50. 50. Bard, A.J. and Faulkner, L.R. (1980). Electrochemical Methods: Fundamentals and Applications. New York: Wiley.
51. 51. Lee, J.S., Jo, K., Lee, T. et al. (2013). J. Mater. Chem. A 1: 9603–9607.
52. 52. Yang, Y. and Chang, H. (2017). J. Mat. Res. doi: 10.1557/jmr.2017.372.
53. 53. Li, H., Kang, W., Wang, L. et al. (2013). Carbon 54: 249–257.
54. 54. Zhou, X., Bai, Z., Wu, M. et al. (2015). J. Mater. Chem. A 3: 3343–3350.
55. 55. Men, B., Sun, Y., Li, M. et al. (2016). ACS Appl. Mater. Interfaces 8: 1415–1423.
56. 56. Lin, Z., Waller, G.H., Liu, Y. et al. (2013). Nano Energy 2: 241–248.
57. 57. Liu, R.L., Wu, D.Q., Feng, X.L., and Mullen, K. (2010). Angew. Chem. Int. Ed. 49: 2565–2569.
58. 58. Jin, H., Zhang, H., Zhong, H., and Zhang, J. (2011). Energy Environ. Sci. 4: 3389–3394.
59. 59. Wohlgemuth, S.A., Fellinger, T.P., Jaker, P., and Antonietti, M. (2013). J. Mater. Chem. A 1: 4002–4009.
60. 60. Ito, Y., Qiu, H.J., Fujita, T. et al. (2014). Adv. Mater. 26: 4145–4150.
61. 61. Tao, G., Zhang, L., Chen, L. et al. (2015). Carbon 86: 108–117.
62. 62. Guo, Q., Zhao, D., Liu, S. et al. (2014). Electrochim. Acta 138: 318–324.
63. 63. Liu, J., Cunning, B.V., Daio, T. et al. (2016). Electrochim. Acta 220: 554–561.
64. 64. Jin, H., Li, J., Chen, F. et al. (2016). Electrochim. Acta 222: 438–445.
65. 65. Zhou, X., Tang, S., Yin, Y. et al. (2016). Appl. Energy 175: 459–467.
66. 66. Tang, S., Zhou, X., Xu, N. et al. (2016). Appl. Energy 175: 405–413.
67. 67. Hu, W., Yoshida, N., Hirota, Y. et al. (2017). Electrochem. Commun. 75: 9–12.
68. 68. Liang, H.W., Wu, Z.Y., Chen, L.F. et al. (2015). Nano Energy 11: 366–376.
69. 69. Song, L.T., Wu, Z.Y., Liang, H.W. et al. (2016). Nano Energy 19: 117–127.
70. 70. Hu, C., Xiao, Y., Zhao, Y. et al. (2013). Nanoscale 5: 2726–2733.
71. 71. Yu, D., Wei, L., Jiang, W. et al. (2013). Nanoscale 5: 3457–3464.
72. 72. Ding, W., Li, L., Xiong, K. et al. (2015). J. Am. Chem. Soc. 137: 5414–5420.
73. 73. Liang, H.W., Zhuang, X., Bruller, S. et al. (2014). Nat. Commun. 5: 4973.
74. 74. Xuan, C., Wu, Z., Lei, W. et al. (2017). Chem. Cat. Chem. 9: 809–815.
75. 75. Wu, M., Wang, K., Yi, M. et al. (2017). ACS Catal. 7: 6082–6088.
76. 76. Eisenberg, D., Stroek, W., Geels, N.J. et al. (2016). Phys. Chem. Chem. Phys. 18: 20778–20783.
77. 77. Li, L., Dai, P., Gu, X. et al. (2017). J. Mater. Chem. A 5: 789–795.
78. 78. Fu, Y., Tian, C., Liu, F. et al. (2016). Nano Res. 9 (11): 3364–3376.
79. 79. Wen, Z., Ci, S., Hou, Y., and Chen, J. (2014). Angew. Chem. 126: 6614–6618.
80. 80. Wang, Y., Liu, H., Wang, K. et al. (2017). Appl. Catal., B 210: 57–66.
81. 81. Yu, D., Zhou, L., Tang, J. et al. (2017). Ind. Eng. Chem. Res. 56: 8880–8887.
82. 82. Yang, H.B., Miao, J., Hung, S.F. et al. (2016). Sci. Adv. 2: e1501122.
83. 83. Zheng, Y., Jiao, Y., Chen, J. et al. (2011). J. Am. Chem. Soc. 133 (50): 20116–20119.
84. 84. Wu, X., Li, S., Wang, B. et al. (2017). Microporous Mesoporous Mater. 240: 216–226.
85. 85. Liang, J., Zheng, Y., Chen, J. et al. (2012). Angew. Chem. 124: 3958–3962.
86. 86. Fu, X., Hu, X., Yan, Z. et al. (2016). Chem. Commun. 52: 1725–1728.
87. 87. Tian, J., Ning, R., Liu, Q. et al. (2014). ACS Appl. Mater. Interfaces 6: 1011–1017.
88. 88. Seredych, M., László, K., Castellón, E.R., and Bandosz, T.J. (2016). J. Energ. Chem. 25: 236–245.
89. 89. Chen, L., Cui, X., Wang, Y. et al. (2014). Dalton Trans. 43: 3420–3423.
90. 90. Seredych, M., Laszlo, K., and Bandosz, T.J. (2015). Chem. Cat. Chem. 7: 2924–2931.
91. 91. Su, Y., Zhang, Y., Zhuang, X. et al. (2013). Carbon 62: 296–301.
92. 92. Jiang, S., Sun, Y., Dai, H. et al. (2015). Electrochim. Acta 174: 826–836.
93. 93. Wu, H., Shi, L., Lei, J. et al. (2016). J. Power Sources 323: 90–96.
94. 94. Gao, S., Liu, H., Geng, K., and Wei, X. (2015). Nano Energy 12: 785–793.
95. 95. Wu, Z., Liu, R., Wang, J. et al. (2016). Nanoscale 8: 19086–19092.
96. 96. Wohlgemuth, S.A., White, R.J., Willinger, M.G. et al. (2012). Green Chem. 14: 1515–1523.
97. 97. Wu, M., Qiao, J., Li, K. et al. (2016). Green Chem. 18: 2699–2709.
98. 98. Huang, Z., Zhou, H., Yang, W. et al. (2017). Chem. Cat. Chem. 9 (6): 987–996.
99. 99. Sahoo, M. and Ramaprabhu, S. (2017). Energy 119: 1075–1083.
100. 100. Fu, S., Zhu, C., Song, J. et al. (2017). Nano Res. 10 (6): 1888–1895.
101. 101. Wang, Z., Cao, X., Ping, J. et al. (2015). Nanoscale 7: 9394–9398.
102. 102. You, B., Kang, F., Yin, P., and Zhang, Q. (2016). Carbon 103: 9–15.
103. 103. Xu, C., Su, Y., Liu, D., and He, X. (2015). Phys. Chem. Chem. Phys. 17: 25440–25448.
104. 104. Patil, I.M., Lokanathan, M., and Kakade, B. (2016). J. Mater. Chem. A 4: 4506–4515.
105. 105. Jiang, H., Zhu, Y., Feng, Q. et al. (2014). Chem. Eur. J. 20: 3106–3112.
106. 106. Wang, J., Wu, Z.X., Han, L.L. et al. (2016). Chin. Chem. Lett. 27: 597–601.
107. 107. Jiang, H., Wang, Y., Hao, J. et al. (2017). Carbon 122: 64–73.
108. 108. Qiu, Y., Xin, L., Jia, F. et al. (2016). Langmuir 32: 12569–12578.
109. 109. Fu, S., Zhu, C., Song, J. et al. (2017). Chem. Eur. J. 23: 10460–10464.
110. 110. Zhao, Y., Huang, S., Xia, M. et al. (2016). Nano Energy 28: 346–355.
111. 111. Jiang, Y.F. (2015). ACS Catal. 5: 6707–6712.
112. 112. Jin, H. (2015). J. Am. Chem. Soc. 137: 7588–7591.
113. 113. Antolini, E. (2016). Renewable Sustainable Energy Rev. 58: 34–51.
114. 114. Liu, R., Zhang, H., Liu, S. et al. (2016). Phys. Chem. Chem. Phys. 18: 4095–4101.
115. 115. Wang, B., Li, S., Wu, X. et al. (2016). J. Mater. Chem. A 4: 11789–11799.
116. 116. Zhang, J., Zhou, H., Liu, X. et al. (2016). J. Mater. Chem. A 4: 15870–15879.
117. 117. Ba, H., Liu, Y., Phuoc, L.T. et al. (2016). ACS Catal. 6: 1408–1419.
118. 118. Yang, Y., Liu, T., Liao, Q. et al. (2016). J. Mater. Chem. A 4: 15913–15919.
119. 119. Gong, M. and Dai, H. (2015). Nano Res. 8 (1): 23–29.
120. 120. Busch, M., Halck, N.B., Kramm, U.I. et al. (2016). Nano Energy 29: 126–135.
121. 121. Tahir, M., Pan, L., Idrees, F. et al. (2017). Nano Energy 37: 136–157.
122. 122. Gong, M., Li, Y., Wang, H. et al. (2013). J. Am. Chem. Soc. 135: 8452–8455.
123. 123. Xia, Z. (2016). Nat. Energy 1: 16155.
124. 124. Ma, T.Y., Dai, S., Jaroniec, M., and Qiao, S.Z. (2014). Angew. Chem. Int. Ed. 53: 7281–7285.
125. 125. Chen, S., Duan, J., Jaroniec, M., and Qiao, S.Z. (2014). Adv. Mater. 26: 2925–2930.
126. 126. Yu, X., Zhang, M., Chen, J. et al. (2016). Adv. Energy Mater. 6: 1501492.
127. 127. Zhao, J., Liu, Y., Quan, X. et al. (2016). Electrochim. Acta 204: 169–175.
128. 128. Zhu, Y.P., Jing, Y., Vasileff, A. et al. (2017). Adv. Energy Mater. 7: 1602928.
129. 129. Ma, T.Y., Ran, J., Dai, S. et al. (2015). Angew. Chem. Int. Ed. 54: 4646–4650.
130. 130. Lee, K.J., Sa, Y.J., Jeong, H.Y. et al. (2015). Chem. Commun. 51: 6773–6776.
131. 131. Liu, Z., Zhao, Z., Wang, Y. et al. (2017). Adv. Mater. 29: 1606207.
132. 132. Park, H.W., Lee, D.U., Liu, Y. et al. (2013). J. Electrochem. Soc. 160 (11): A2244–A2250.
133. 133. Li, X., Fang, Y., Zhao, S. et al. (2016). J. Mater. Chem. A 4: 13133–13141.
134. 134. Li, J.C., Hou, P.X., Zhao, S.Y. et al. (2016). Energy Environ. Sci. 9: 3079–3084.
135. 135. Han, L.N., Wei, X., Zhang, B. et al. (2016). RSC Adv. 6: 56765–56771.
136. 136. Patra, S., Choudhary, R., Roy, E. et al. (2016). Nano Energy 30: 118–129.
137. 137. Huang, X., Wang, Q., Jiang, D., and Huang, Y. (2017). Catal. Commun. 100: 89–92.
138. 138. Pei, Z., Li, H., Huang, Y. et al. (2017). Energy Environ. Sci. 10: 742–749.
139. 139. Zhou, W., Jia, J., Lu, J. et al. (2016). Nano Energy 28: 29–43.
140. 140. Conway, B. and Tilak, B. (2002). Electrochim. Acta 47: 3571–3594.
141. 141. Mahmood, N., Yao, Y., Zhang, J.W. et al. (2017). Adv. Sci. 1700464.
142. 142. Sudhagar, P., Roy, N., Vedarajan, R. et al. (2016). Hydrogen and CO₂ reduction reactions: mechanisms and catalysts. In: Photoelectrochemical Solar Fuel Production: From Basic Principles to Advanced Devices (ed. S. Giménez and J. Bisquert), 105–160. Cham: Springer International Publishing.
143. 143. Jun, Y.S., Park, J., Lee, S.U. et al. (2013). Angew. Chem. Int. Ed. 52: 11083–11087.
144. 144. Talapaneni, S.N., Kim, J., Je, S.H. et al. (2017). J. Mater. Chem. A 5: 12080–12085.
145. 145. Xie, K., Wu, H., Meng, Y. et al. (2015). J. Mater. Chem. A 3: 78–82.
146. 146. Tiana, Y., Ye, Y., Wang, X. et al. (2017). Appl. Catal., A 529: 127–133.
147. 147. Zhao, Y., Zhao, F., Wang, X. et al. (2014). Angew. Chem. 126: 14154–14159.
148. 148. Duan, J., Chen, S., Jaroniec, M., and Qiao, S.Z. (2015). ACS Nano 9 (1): 931–940.
149. 149. Yan, D., Dou, S., Tao, L. et al. (2016). J. Mater. Chem. A 4: 13726–13730.
150. 150. Shinde, S.S., Sami, A., and Lee, J.H. (2015). Chem. Cat. Chem. 7: 3873–3880.
151. 151. Yan, X., Jia, Y., Odedairo, T. et al. (2016). Chem. Commun. 52: 8156–8159.
152. 152. Peng, Z., Yang, S., Jia, D. et al. (2016). J. Mater. Chem. A 4: 12878–12883.
153. 153. Zhang, J. and Dai, L. (2015). ACS Catal. 5: 7244–7253.
154. 154. Lai, J., Li, S., Wu, F. et al. (2016). Energy Environ. Sci. 9: 1210–1214.
155. 155. Kumar, B., Brian, J.P., Atla, V. et al. (2016). Catal. Today 270: 19–30.
156. 156. Lingampalli, S.R., Ayyub, M.M., and Rao, C.N.R. (2017). ACS Omega 2: 2740–2748.
157. 157. Wu, J., Yadav, R.M., Liu, M. et al. (2015). ACS Nano 9: 5364–5371.
158. 158. Xie, J., Huang, Y., and Yu, H. (2015). Front. Environ. Sci. Eng. 9: 861–866.
159. 159. Li, W., Seredych, M., Castellun, E.R., and Bandosz, T.J. (2016). Chem. Sus. Chem. 9: 606–616.
160. 160. Sharma, P.P., Wu, J., Yadav, R.M. et al. (2015). Angew. Chem. Int. Ed. Eng. 54: 13701–13705.
161. 161. Sreekanth, N., Nazrulla, M.A., Vineesh, T.V. et al. (2015). Chem. Commun. 51: 16061–16064.
162. 162. Zhang, S., Kang, P., Ubnoske, S. et al. (2014). J. Am. Chem. Soc. 136: 7845–7848.
163. 163. Liu, Y., Chen, S., Quan, X., and Yu, H. (2015). J. Am. Chem. Soc. 137: 11631–11636.
164. 164. Amirfakhri, S.J., Binny, D., Meunier, J.L., and Berk, D. (2014). J. Power Sources 257: 356–363.
165. 165. Adić, R.R., Marković, N.M., and Vešović, V.B. (1984). J. Electroanal. Chem. Interfacial Electrochem. 165: 105–120.
166. 166. Wilshire, J. and Sawyer, D.T. (1979). Acc. Chem. Res. 12: 105–110.
167. 167. Memming, R. (1969). J. Electrochem. Soc. 116: 785–790.
168. 168. Wu, P., Du, P., Zhang, H., and Cai, C. (2013). Phys. Chem. Chem. Phys. 15: 6920–6928.
169. 169. Zhang, T., Li, C., Gu, Y. et al. (2017). Talanta 165: 143–151.