oxide composite catalysts for bifunctional oxygen reduction and evolution in reversible alkaline fuel cells: A mini review

Journal of Power Sources xxx (2017) 1e14

Contents lists available at ScienceDirect

Journal of Power Sources journal homepage: www.elsevier.com/locate/jpowsour

Nanocarbon/oxide composite catalysts for bifunctional oxygen reduction and evolution in reversible alkaline fuel cells: A mini review Mengjie Chen a, Lei Wang b, Haipeng Yang b, **, Shuai Zhao c, Hui Xu c, Gang Wu a, * a

Department of Chemical and Biological Engineering, University at Buffalo, The State University of New York, Buffalo, NY, 14260, United States College of Materials Science and Engineering, Shenzhen University, Shenzhen, Guangdong, 518060, China c Giner Inc., Newton, MA, 02466, United States b

h i g h l i g h t s

g r a p h i c a l a b s t r a c t


The bifunctional ORR/OER catalysts are for reversible alkaline fuel cells.
The most promising nanocarbon/oxide composites catalysts were discussed.
A synergistic effect between oxides and nanocarbons was highlighted.
Perspective and current challenges of nanocarbon/oxide catalysts were outlined.

a r t i c l e i n f o

a b s t r a c t

Article history: Received 20 June 2017 Received in revised form 14 August 2017 Accepted 16 August 2017 Available online xxx

A reversible fuel cell (RFC), which integrates a fuel cell with an electrolyzer, is similar to a rechargeable battery. This technology lies on high-performance bifunctional catalysts for the oxygen reduction reaction (ORR) in the fuel cell mode and the oxygen evolution reaction (OER) in the electrolyzer mode. Current catalysts are platinum group metals (PGM) such as Pt and Ir, which are expensive and scarce. Therefore, it is highly desirable to develop PGM-free catalysts for large-scale application of RFCs. In this mini review, we discussed the most promising nanocarbon/oxide composite catalysts for ORR/OER bifunctional catalysis in alkaline media, which is mainly based on our recent progress. Starting with the effectiveness of selected oxides and nanocarbons in terms of their activity and stability, we outlined synthetic methods and the resulting structures and morphologies of catalysts to provide a correlation between synthesis, structure, and property. A special emphasis is put on understanding of the possible synergistic effect between oxide and nanocarbon for enhanced performance. Finally, a few nanocomposite catalysts are discussed as typical examples to elucidate the rules of designing highly active and durable bifunctional catalysts for RFC applications. © 2017 Elsevier B.V. All rights reserved.

Keywords: Reversible alkaline fuel cells Electrolyzers Oxygen reduction Oxygen evolution Bifunctional catalysts Nanocomposites

1. Introduction

* Corresponding author. ** Corresponding author. E-mail addresses: [email protected] (H. Yang), [email protected] (G. Wu).

Recently, substantial attention has been paid on the development of innovative energy storage and conversion technologies, which should be environmental friendly with highly efficiency. Among others, electrochemical energy systems, especially fuel cells

http://dx.doi.org/10.1016/j.jpowsour.2017.08.062 0378-7753/© 2017 Elsevier B.V. All rights reserved.

Please cite this article in press as: M. Chen, et al., Journal of Power Sources (2017), http://dx.doi.org/10.1016/j.jpowsour.2017.08.062

2

M. Chen et al. / Journal of Power Sources xxx (2017) 1e14

and water electrolyzers, have been studied worldwide over the last decades associated with utilization and generation of hydrogen and oxygen, which are ideally compatible with renewable energy sources and sustainability [1e3]. Recently, a new concept of integrating a fuel cell and an electrolyzer together was introduced to realize periodic energy storage and conversion with reduced cost and increased energy density. This kind of device is so-called reversible fuel cells (RFCs). It is similar to a rechargeable battery. RFCs could provide higher theoretical specific energy (3660 Wh/kg) and packaged specific energy (400e1000 Wh/kg), which are nearly 5 times higher than that of batteries [4]. A reversible fuel cell can be operated in two different modes. One is fuel cell mode and the other is electrolyzer mode. At fuel cell mode, RFCs convert the chemical energy (e.g., H2 and O2) into electricity and H2O with relatively high efficiency in an environmental friendly manner. Oppositely, when RFCs are operated at electrolyzer mode, they can store the electricity generated from solar or wind through water splitting into H2 and O2. The fuel cell mode is comparable to the discharging process of a battery for generating electricity, while the electrolyzer mode is the charging process by utilizing electricity for storing energy into chemicals. Therefore, in a RFC, the oxygen electrode should be active and stable for both the oxygen reduction reaction (ORR) and the oxygen evolution reaction (OER) as the cathode in fuel cell and as the anode in electrolyzer mode, respectively. A fuel cell cathode for the ORR needs to shift to the anode for the OER in the electrolyzer. Therefore, it is highly demanded to develop bifunctional catalysts that are active for the ORR and the OER, which represents one of grand challenges in electrocatalysis society [5]. Theoretical simulation predicted that there is no single active site capable of catalyzing both ORR and OER simultaneously [6]. For example, the state-of-the-art catalysts for the ORR are Pt-based catalysts. Nevertheless, Pt is not a good choice for the OER occurring at high potential ranges, during which Pt oxides would form and cover at the surface causing a decrease of activity. The most active and stable catalysts for the OER are Ir and Ru oxides due to their high activity and stability along with good electrical and ionic conductivities. Likewise, they still cannot be used as bifunctional catalysts due to poor activity for the ORR. Therefore, development of PtIr nanocomposite catalysts consisting of active ORR and OER components has become one of the effective solutions to overcome the technical barrier [7]. However, it is highly undesirable to use large amount of these precious metal-based catalysts because they are very scarce and expensive. Pursuing highly efficient PGM-free nanocomposite catalysts simultaneously active for the ORR and the OER therefore becomes highly demanded for the attractive RFC technologies [8e11]. In addition to insufficient catalyst activity, stability remains as a grand challenge due to the fact that most of PGM-free catalysts are not stable in such oxidative ORR/OER environments within a wide electrochemical potential window (0.6e1.9 V). It should be noted that a favorable electrolyte medium is of vital importance for the RFC technology. Compared to acidic media, RFCs operated in alkaline media have many benefits. At first, the alkaline medium provides a relatively less corrosive condition and makes the catalysts more stable [12,13]. In particular, oxides and carbon-based PGM-free catalysts are much more stable in alkaline medium [14,15]. Although these oxygen reactions (ORR/OER) are kinetically sluggish, they would become relatively favorable in alkaline media [2,16e19]. For the OER, the hydroxide is oxidized to generate oxygen and water in alkaline media. For the ORR, two distinct pathways can be observed during the process. The preferred one is that oxygen reacts with water to produce hydroxide directly through a four-electron pathway. The other route goes through two separate two-electron transfer steps and yields hydrogen peroxide that is a harmful intermediate and is undesirable owing to low efficiency.

At present, a variety of PGM-free materials have been studied as a bifunctional catalysts including various transition metal oxides and nanocarbons. Transition metal oxides are able to shift among cationic oxidation states, which provides active redox sites and tolerates the oxidative condition during ORR and OER [16,20]. Moreover, transition metal oxides have some extra abilities, with which precious metals could not compete. For example, transition metal oxides are able of being doped with secondary and third metals, which creates the opportunity to develop complex oxide catalysts, such as perovskite and spinel oxides, simultaneously active for the ORR and the OER. However, their insufficient electrical conductivity and low surface area largely limit them to be used for practical catalysts. In addition, relative to their high activity for the OER, these oxide catalysts often suffer from low ORR activity. Consequently, highly conductive and ORR-active materials are demanded to integrate with oxides to achieve the intrinsic high OER activity and improve ORR performance. Another promising materials are carbon nanomaterials, which often own high electrical conductivity and high active surface area for electrocatalysis [19,21e25]. In addition, they can be easily doped by heteroatoms and transition metals with enhanced catalytic activity [26]. Especially, the doped pyridinic nitrogen sites in carbonbased catalysts could serve as adsorption sites of O2 to promote ORR by 4-electron route in alkaline media. The graphitic nitrogen in carbon planes can change the electronic structure of carbon inducing catalytic activity [27e29]. In addition to being used as catalysts, carbon materials could serve as supports to improve the conductivity of oxide catalysts and promote the dispersion of oxide nanocrystals [30,31]. Commonly used carbon materials are carbon blacks, carbon nanotubes, and graphene [32]. However, due to the thermodynamic instability of carbon, carbon itself cannot be considered as a stable OER catalyst candidate even in alkaline media. To overcome their weaknesses and combine their advantages, the most effective solution is to develop nanocarbon/oxide composite catalysts for the challenging bifunctional oxygen catalysis. Currently, it is not fully understood the active site structure in such nanocomposite catalysts during ORR and OER. But it is a consensus that the porous morphology and large electrochemical surface area often lead to better catalytic activity. Importantly, optimal combination of oxides and nanocarbon can provide a great opportunity to achieve outstanding stability [33]. In this mini review, we discuss the rules of design and synthesis of nanocarbon/oxide composite catalysts with an emphasis on their possible synergistic effect. Recent progress is summarized to provide insightful structureproperty correlations. 2. Nanocarbon/oxide nanocomposite catalysts Although transition metal oxide has been considered as one kind of promising bifunctional catalyst, it is difficult to achieve their intrinsic electrocatalytic activity due to their low electrical conductivity and low surface area, primarily limiting kinetic reaction rates and power density in the reversible fuel cells [34e36]. One effective solution is to integrate transition metal oxides with conductive materials by promoting conductivity or changing their electronic structure [37,38]. Carbon material is a promising candidate, not only because it could provide high electrical conductivity, but also its high surface areas can be applied as effective catalyst supports [1,12,35,39]. Furthermore, the oxide catalysts are relatively more active towards OER due to the difference of optimal active sites between these two reactions. Oppositely, carbon materials with appropriate heteroatom doping often exhibit high ORR activity in alkaline media, but not good for the OER. Therefore, substantial research is focused on synthesis of nanocarbon/oxides

Please cite this article in press as: M. Chen, et al., Journal of Power Sources (2017), http://dx.doi.org/10.1016/j.jpowsour.2017.08.062

M. Chen et al. / Journal of Power Sources xxx (2017) 1e14

nanocomposite, composed of OER active oxides and ORR active nanocarbons as shown in Fig. 1 [16,17]. Commonly used carbon materials include amorphous carbon black, 1D carbon nanotubes, 2D graphene, and 3D graphite. Most often studied transition metal oxides for the ORR and OER in alkaline media include Fe, Co, Ni and Mn-based oxides, which were comprehensively summarized in our recent review [1]. In addition to well-defined perovskite and spinel oxides, binary and ternary mixed oxides have also been explored in terms of their good catalytic activity for the ORR and the OER. Their structural diversity, as well as their ability to be mixed, doped, and combined with other materials, make transition metal oxides a highly attractive component in nanocomposites for bifuncational catalysts. Their catalytic activity and stability are greatly dependent on the structural properties associated with synthesis methods. Therefore, development of oxide catalysts requires carefully optimizing such properties, which could be predicted by theoretical understanding [40]. On the other hand, ORR/OER activity of carbon can be significantly enhanced by doping of heteroatoms (e.g. N, P, S) and transition metals (e.g. Fe, Co, Ni) in aqueous electrolyte [19,41,42]. It is believed that the microstructure and electronic transport of carbon materials can be modified by doping heteroatoms [8,11]. In these cases, carbon materials not only play a role as the support, but also as an ORR catalyst [24]. For example, the N atoms could substitute C atoms and exhibit sp2 hybridization in C-N bonds. The C atoms bonded with N dopants should be positively charged due to the larger electronegativity of N atom than that of C atom. Thus, carbon atoms could become more active for O2 adsorption [8,11,19]. As for the role played by transition metals, some experts hold the opinion that the transition metals could serve as an electrochemically active part of the catalytic site such as FeN4 and CoN4, while others just believe that the metals are involved in tuning the nitrogen doping, surface areas, and structures of carbon [9,43]. However, recent experimental and modeling simulation results tends to support that the atomic FeN4 moieties embedded into carbon plane are likely the active sites with favorable binding energy for O2 adsorption and bond dissociation during the ORR [44e47]. Considering its intrinsic electrical conductivity and promoted ORR activity, functionalized nanocarbons by nitrogen or transition metals can enhance the overall performance of oxides/nanocarbon composite catalysts. 2.1. Transition metal oxides In principle, transition metal oxide catalysts are more stable, especially in harsh oxidative environments [12,48]. Among these carbon-free catalysts, due to the catalytic activity for both ORR and OER in essence, perovskites have been one of very popular oxide

3

catalysts [49e52]. This type of metal oxides can be described by a general formula of ABO3, where A often is rare-earth or alkaline earth metals and B commonly represents transition metals. In an ideal unit cell, the A-site cations tend to be larger and more electropositive than B-site cations. The A-site cations coordinate with twelve-fold oxygen, whereas B-site cations are with six-fold oxygen coordination [52]. This unique formulation allows perovskite to freely adjust the physicochemical properties by changing metal cations. By substituting metal cations and producing oxygen deficiency, desired electronic structure could be achieved progressively. Consequently, both ORR and OER activity of this modified perovskite can be improved owing to the update of some parameters such as transition metal oxygen covalent bond and electron density of the d-band of transition metal [34,53,54]. The mechanisms of ORR and OER on perovskite oxide catalysts have been studied for over 40 years. It was found that the current density at certain fixed potential is inversely proportional to the enthalpy of formation of metal hydroxide and the rate-determining step has a big correlation with hydroxide intermediates [55]. Although the detailed reaction mechanisms have not been conclusive, it is believed that surface hydroxide species can be replaced by an adsorbed O 2 species to facilitate oxygen reduction [56]. The reverse process is believed to be OER mechanism. Moreover, it has been demonstrated the rate-determining step on oxygen reduction and oxygen evolution varies depending on the number of eg electrons in transition metals [53]. For OER, if this number is less than one, the reaction where OOH group becomes O2 2 by a deprotonation process would be the rate-determining step (RDS). If not, the RDS would be the process where OH group reacts with O2 which bonded on the transition metal to form an OOH group. As for ORR, there are similar cases. In the last decades, some measurable parameters such as eg orbital filling and covalence of transition metal-oxygen bond have been studied for explaining the intrinsic activity of OER/ORR. These parameters are based on molecular orbital theory and energy band theory. It was reported that the activity of OER is directly related to eg filling of transition metal in perovskite oxides which can be affected by metal oxidation states and spin states [10]. Because of the strong overlap between eg orbital of transition metal and oxygen-related adsorbate, the electron transfer between surface cation and adsorbed intermediates can be more directly promoted. Suntivich and co-workers proposed that eg filling of surface transition metal cations can greatly influence the binding of OER intermediates to the oxide surface and thus the OER activity. In most cases, the closer the value is to the number 1, the better the OER activity of the catalysts [10]. Moreover, the covalence of transition metal-oxygen bond plays an important role during ORR and OER process by changing the spatial overlapping between transition metal 3d orbital and oxygen 2p orbital. The charge-transfer interval between the metal and oxygen can be

Fig. 1. Scheme of designing principles of bifunctional ORR/OER catalyst by integrating oxide nanocrystals and nanocarbon (a) 2D graphene and (b) 1D carbon tubes. Reprint with permission from Refs. [16,17], Copyright Wiley-VCH.

Please cite this article in press as: M. Chen, et al., Journal of Power Sources (2017), http://dx.doi.org/10.1016/j.jpowsour.2017.08.062

4

M. Chen et al. / Journal of Power Sources xxx (2017) 1e14

reduced by enhancing the covalence of metal-oxygen bond in order to improve the charge transfer between transition metal and oxygen ions [10]. There are some strategies to adjust these parameters such as substitution of metal cations and creation of deficiency/ vacancy of oxygen. By adjusting metal cations on A-site, the oxidation state and spin state of metal cations on B-site can be changed because the positive charge contribution on A-site has been altered. Moreover, the oxygen vacancy can act as an acceptor or donor of oxygen, simultaneously improving the charge transfer between adsorbent and adsorbate, thus leading to semiconducting behavior over a thin surface layer on catalysts. Recently, Grimaud and co-workers conducted a systematic study on double perovskites [57]. The results showed that the presence of oxygen vacancies stabilizes the dxz, dyz and d2z molecular orbitals in the t2g and eg-parentage orbitals, respectively. This change results in the formation of oxygen-deficient octahedral symmetry which is socalled square pyramidal symmetry (C4v). It is believed that the type of lanthanide could affect the oxygen vacancy content and oxidation state of cobalt, which both make a difference on OER/ORR activity [57]. Spinel oxides, with a general formula of AB2O4, have been focused on over decades due to their promising ability to catalyze ORR and OER simultaneously [58]. The typical spinel oxide is cubic structure consisting of A-type cations, B-type cations, and oxygen anions that occupy octahedral sites, tetrahedral sites, and FCC lattice sites, respectively [59]. The cubic unit cell of spinel oxides is composed of 8 tetragonal sub-units. Although spinel oxides have complicated structure due to their multiple structural degrees of freedom, excellent versatile abilities can be obtained by presenting deficiencies, exchanging the types of metal cations and switching the positions of the two present cations to form a reverse spinel structure. This may improve the transfer of electrons during the ORR and OER. In addition to highly ordered perovskite and spinel oxides, other mixed oxide containing Co, Ni, and Mn such as Co3O4, CoNiOx, and CoxMn1xO also demonstrated great promise for ORR and OER catalysts [17,18]. Unlike perovskite and spinel oxides, simple mixed oxides can present single oxidation state, which are easier to be prepared. Relative to single-metal oxides that usually show insufficient electrical conductivity and low reactive surface areas, mixed oxides provide greater opportunity to enhance kinetics during these electrochemical reactions. Because the dissimilarity in lattice strain between Ni and Co creates the difference in their redox potentials and structural ordering. It has been reported that the possible mechanism of fourelectron pathway for ORR on these oxides in alkaline electrolyte could be described by one four-step reaction [60e62]:

A þ H2 O þ e 4AH þ OH

(1a)

2AH þ O2 4ðAHÞ2 /O2;ads

(1b)

ðAHÞ2 /O2;ads þ e /AH/Oads þ OH þ A

(1c)

AH/Oads þ e 4A þ OH

(1d)

Where A represents the catalytic center. Firstly, the catalyst proceeds one protonation process (eq. (1a)) followed by the O2 adsorption reaction (eq. (1b)) where single oxygen molecule adsorbs onto two AH sites. Eq. (1c) is the rate-determining step and yields hydroxide species, while the final step (eq. (1d)) is the reduction of Oads species. During these steps, the transition metals act as oxygen donor or acceptor by changing their valence. Notably, the amount of available active sites and adsorption affinity of O2 are

related to the catalysis of the oxygen reduction [63,64]. This mechanism for ORR is only tentative because it correlates with the 4-electron pathway. And the 2-electron transfer pathway where includes the formation of hydrogen peroxide intermediates could be possible in terms of different oxide catalysts. One possible mechanism of OER on spinel oxides in alkaline media was proposed by Singh [65] which is similar to the mechanism on perovskites proposed by Bockris [66]:

A þ OH 4AOH þ e

(2a)

AOH þ OH /AO þ H2 O þ e

(2b)

2AO42A þ O2

(2c)

First two steps correlate to electron-transfer process and the third reaction are related to oxygen generation. This mechanism is also not fixed due to the probability of other mechanisms containing the formation of hydrogen peroxides [66]. 2.2. Effective synthesis strategies of oxide catalysts The synthesis methods of catalyst are very important and will directly affect the catalytic performance of the catalyst. Here, we introduce several common synthesis methods of oxide catalysts and then compare their advantages and disadvantages (Table 1). Solid state route is one traditional method which can be easy to operate. Firstly, the metal precursors should be mixed together. And then they should be grounded to form a uniform powder. Next, the powder is heat-treated in certain condition and then purified by annealing at fixed temperature. Notably, the temperature and duration should be based on the selected metal ions and the desired products. Another common method is the sol-gel method. First, the metal precursors are mixed and dissolved in a specific solvent, heated while stirring, and refluxed to form a gel. Thereafter, by centrifugation or filtration, a solid is obtained. The final catalyst can be obtained by calcination in air. This method promotes uniform catalyst particles, but pH value should be under control accurately. Because pH affects the hydroxylation of the metal [67,68]. By controlling the temperature and atmosphere, oxides with different performance can be obtained [69]. In addition to the two methods above, co-precipitation is also a typical method. By mixing two kinds of metal salts in deionized water, a uniform solution can be formed. Subsequently, in order to get the precipitates, one organic base should be added into the solution while stirring. After centrifugation and drying process, the final desired nano-sized catalysts can be obtained by heat treatment at certain temperature and atmosphere [70]. The duration and temperature for heat treatment should be dependent on the types of metal salts [71]. The size of oxide particles synthesized through this method can reach as small as 10 nm. However, due to the difference between hydrolysis rate of metal ions, sometimes this kind of perovskite or spinel oxides show low purity and low mass activity [68]. Hydrothermal synthesis is another common method which prepares catalyst powder by hydrothermal reaction. The hydrothermal process essentially involves a dissolution/precipitation mechanism driven by the difference in solubility between the soluble reactant phase and the insoluble product. Hydrothermal method provides a special physical and chemical environment that cannot be obtained under atmospheric pressure for the reaction and crystallization of various precursors. Compared with other preparation methods, this method has the advantages of complete crystal development, small particle size and less agglomeration [36,72e74]. In general, most methods mentioned above require

Please cite this article in press as: M. Chen, et al., Journal of Power Sources (2017), http://dx.doi.org/10.1016/j.jpowsour.2017.08.062

M. Chen et al. / Journal of Power Sources xxx (2017) 1e14

5

Table 1 Comparison of synthesis methods for oxide Catalysts. Synthesis method

advantages

disadvantages

solid state sol-gel co-precipitation hydrothermal

simple process uniform particle size nano-sized particles low temperature and well-dispersed particles

require high temperature and long duration need to control pH value carefully impurity and low mass activity high pressure

extremely high temperature and long annealing duration in order to get crystalline structure oxides, which brings about low mass activity towards both ORR and OER due to less active sites or lower surface area [35]. 2.3. Selection of nanocarbon components As discussed above, the carbon materials could help the oxides achieve their intrinsic activity by changing the electronic structure and improving electrical conductivity. Moreover, well-dispersed and uniform oxides particles can be obtained by using carbon supports with high surface areas. However, the carbon materials can be oxidized to CO2 thermodynamically above 0.207 V (vs NHE) at room temperature [75], implying that the corrosion reaction will reach significant level at high potential (>1.3 V) during the OER. Thus, the catalytic activity could be undermined owing to less accessibility of active sites resulting from the carbon corrosion. Moreover, the corrosion of carbon support could not only weaken the interaction between the oxides and carbon support, leading to the separation of carbon and oxides, but also increase the hydrophilicity of the surface which impedes the gas transport as the pores are filled with water [76]. Thus, this is a great challenge to select highly stable nanocarbon support. However, in less harsh alkaline environment, some highly graphitized carbon nanomaterials still hold the promise to be applied for ORR/OER bifunctional catalysis after proper modification of structure and morphology [42,77e94]. As we all know that the high graphitization degree improves the stability of carbon materials due to high ordered structures and low deficiencies. In addition, we found that the graphitized carbon layers in a close structure, i.e. carbon nanotubes (CNTs) are the most stable when compare to other carbon structures. As shown in Fig. 2, CNTs showed better stability than carbon black and reduced graphene oxides (rGO) during the potential cycling (0e1.9 V vs. RHE) in alkaline media because of its higher graphitization degree and tube structure. Note, however, that not all the carbon nanotube materials were durable during the ORR/OER potential cycling tests [3]. Thus, seeking for highly robust and active carbon support becomes research directions of many groups by modifying the doping and adjusting the structure. Furthermore, we studied the effect of tube size on stability by using various commercially available multi-walled nanotubes (MWNTs) with different tube size ranging from 8 to 50 nm. As shown in Fig. 3, interestingly, the tubes with smaller sizes demonstrated better stability, while larger tubes significantly loss both ORR and OER activity during the potential cycling (0e1.9 V in 0.1 M NaOH). Thus, this will provide a general guidance to engineer the structures of carbon tubes for maximum activity and stability during the ORR and OER. It should be noted that these solutions just lower the rate of carbon corrosion and can't solve this problem fundamentally. 2.4. The possible synergy between oxides and nanocarbons Until now, many groups have reported the studies of bifunctional oxide/nanocarbon composite catalysts, providing the concept of synergistic effect between the oxides and carbon

materials that results in better activity and durability than single component [16,33,95]. Meanwhile, there are different theories proposed to explain this phenomenon including electronic effect (ligand effect) and interfacial chemical bonding. The first theory is so-called electronic effect which is widely studied in heterogeneous catalysis [96e98]. The electronic effect is described as the change of electron density of certain metal ions that further affects the performance of catalysts. Most recently, Fabbri et al. and coworkers reported that electronic effect plays a crucial role in enhancing ORR and OER catalytic activity of oxide/nanocarbon composite catalysts compared to the individual component in alkaline media [99]. They prepared Ba0.5Sr0.5Co0.8Fe0.2O3-d (BSCF) catalyst via a modified sol-gel method and further deposited onto acetylene black carbon (BSCF/AB) by an ultrasonication method. The activity of BSCF and BSCF/AB composite catalysts towards ORR/ OER is shown in Fig. 4a [99]. The BSCF/AB composite catalyst showed better ORR performance than individual BSCF oxide catalyst. The overpotential of BSCF/AB composite was 130 mV lower than single BSCF. As for OER, the BSCF/AB composite catalyst exhibited a potential of 1.55 V (vs. RHE) at 10 A/gmetal, which was around 100 mV lower than single BSCF. To demonstrate the electronic effect between oxide and carbon material, XANES spectra were recorded for Co, Sr, Fe K-edges and Ba L1 edge for the single BSCF oxide and the BSCF/AB composite. Although adsorption edges for Sr, Ba and Fe were identical, the Co K-edge showed a clear difference between BSCF/AB composite catalyst and single BSCF (Fig. 4b). The composite catalyst showed an adsorption edge similar to that of CoO and the adsorption edge of single BSCF was much closer to Co3O4, indicating that the electron density of cobalt ions was increased by surface functionalized acetylene black. The reduction of cobalt oxidation state modified the lattice structure for BSCF, which was also proven by XRD data, resulting in the enhancement of the intrinsic BSCF conductivity [100], which may be responsible for the observed increase of the catalytic activity in the composite catalyst. Moreover, it has been reported that the adsorption energy of oxygen-related species could be altered by changing the oxidation state of metal cations, which could affect the electrocatalysis of oxygen [101e103]. This work provided a novel understanding of synergistic effect between oxides and carbon materials by introducing the concept of electronic effect. There are some other literatures providing the concept of electronic effect including copper-centered MOF/GO composite [104], Pt75Ni25/C hybrids [105] and manganese oxide/CNT catalysts [106]. All research mentioned above provided a general idea that the presence of carbon leads to better performance of catalysts because of electronic effect. Another theory provided for illustrating the synergistic effect is interfacial chemical bonding. This interfacial structure is usually observed in chemically synthesized composite catalysts in the form of chemical bonds [107e109]. In 2011, Liang et al., reported that the high ORR/OER performance of Co3O4 nanocrystals/N-doped reduced graphene oxide hybrid catalysts resulted from the synergetic chemical bonding between Co3O4 nanocrystal and N-doped graphene oxide [110]. They proved that the formation of Co-N-C and Co-O-C bonds resulted in the high ORR/OER performance of this hybrid catalysts. In 2014, Liu et al., and co-workers reported a

Please cite this article in press as: M. Chen, et al., Journal of Power Sources (2017), http://dx.doi.org/10.1016/j.jpowsour.2017.08.062

6

M. Chen et al. / Journal of Power Sources xxx (2017) 1e14

Fig. 2. ORR and OER polarization curves of different carbon materials in O2 saturated and in absence of O2 in 0.1 M NaOH electrolyte, before (black) and after (red) 1000 cycles (0e1.9 V at a scan rate of 500 mV s1). SEM images for different carbon materials are also shown. Reprint with permission from Ref. [3]. Copyright Wiley-VCH. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

modified carbon black-LaMnO3 hybrids with high activity for the ORR [108]. They attributed the significantly promoted activity to the formation of covalent bonds (CeOeM) between the carbon and the oxide nanoparticles, which could effectively enhance the ORR kinetics. This interfacial chemical bond was confirmed by showing the negative shift of peak related to C-O bond and by exhibiting the additional peak recorded at 533.2 eV in the O 1s spectra (Fig. 4c and d). It was reported that the bond was formed during the sintering process and it was found that the higher contents of covalent CeOeM bonds resulted in better electrochemical performance. The same researchers successfully synthesized a carbon-supported LixCo3-xO4 nanocrystals with average particle size of 4 nm. This catalyst exhibited good performance towards ORR in alkaline media due to the formation of C-O-Co bond and oxygen defects [107]. Another example provided by Liang et al., also demonstrated the effect of strongly coupling bond [109]. They applied a simple twostep method to synthesize CoO/NCNTs catalyst, which showed high ORR performance that outperformed commercial Pt/C catalyst in alkaline electrolyte, mainly via a 4-electron pathway. The strongly coupled bond in inorganic/nanocarbon hybrid probably contributed to the high activity and stability under a highly corrosive condition of 10 M NaOH at 80  C. These researches showed the effect of interfacial chemical bond, providing a promising design of catalyst with high performance.

catalysts. Thus, the interaction between the particles is very weak, resulting in limited activity and durability due to the interfacial resistance. In order to maximize the synergistic effect between oxides and carbon materials, efforts have been paid on searching for effective chemical methods that bring a stronger interaction between oxides and nanocarbons. This is a difficult process that requires extremely accurate control because redox reaction always occurs between oxides and carbon materials during high temperature treatment, leading to the destruction of oxides structure. However, some useful methods have been designed successfully, including chemical vapor deposition [91,93,111,112], sol-gel route [113,114], impregnation method [115], and hydrothermal process [62,116,117]. Recent bifunctional catalysts towards ORR and OER based on oxides or oxide/nanocarbon composites are summarized in Table 2. 2.6. Several typical examples of nanocarbon/oxide catalysts During the development of high-performance nanocarbon/oxide catalysts, many excellent works have been reported from different groups cross the world in last five years [42,77e94]. In this part, to provide a general understanding of designing effective nanocarbon/oxide catalysts, the correlation between synthesis, strucure/morphology, and property was discussed by introduing several catalysts based on our recent effort in the field.

2.5. Synthesis strategies for oxides/nanocarbon nanocomposites The performance of oxide/nanocarbon composites is not only related to the innate properties of the oxide or carbon material, but also to the synthesis methods. There are many ways to combine transition metal oxides with carbon materials. In general, all of them can be divided into two categories: physical and chemical methods. The physical method simply mixes two or more different catalysts together, for example, by means of grinding or ball milling, to achieve uniform dispersion and good conductivity of the

2.6.1. Carbon black supported oxygen-deficient perovskite catalysts Compared to other carbon blacks such as Ketjenblack and Blackpearl, Vulcan XC-72 (CABOT, Billerica, MA, USA) are very stable in oxidative environments, which has been extensively used for support for Pt catalysts in polymer electrolyte membrane fuel cells. We developed a novel oxygen-deficient BaTiO3-x perovskite oxide catalyst by using sol-gel method followed by high-temperature treatment in vacuum atmosphere [118]. To further improve the overall catalytic performance, 10 wt% Vulcan XC-72 was used as a

Please cite this article in press as: M. Chen, et al., Journal of Power Sources (2017), http://dx.doi.org/10.1016/j.jpowsour.2017.08.062

M. Chen et al. / Journal of Power Sources xxx (2017) 1e14

7

Fig. 3. The ORR/OER polarization curves of carbon nanotubes as function of diameter sizes from 8 to 50 nm, before (black) and after (red) 1000 cycles (0e1.9 V at a scan rate of 500 mV s1 in O2 saturated 0.1 M NaOH). Reprint with permission from Ref. [3]. Copyright Wiley-VCH. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

support to further improve the catalytic activity for the ORR and OER in alkaline media. The prepared oxide particles with an average size of 110 nm were uniformly mixed with Vulcan carbon particles (50 nm). The phase structures of the BaTiO3 and its oxygen content were controlled by heat temperature and atmosphere environment. Single phase t-BaTiO3 was generated by using air for calcination. In addition to t-BaTiO3, the sample calcined in argon also contained TiC and BaCO3. Both of these two samples showed low ORR and OER activity, which suggested that t-BaTiO3, TiC, and BaCO3 were inactive for oxygen electrocatalysis. Surprisingly, heat treatment of BaTiO3 at 1300  C under vacuum conditions leads to the formation of oxygen-deficient hexagonal BaTiO2.76 crystal structure as determined by neutron analysis (Fig. 5). With a uniform size distribution around 110 nm and BET surface area of 24.7 m2 g1, the oxygen-deficient hexagonal BaTiO3-x -rich catalyst presented excellent ORR and OER catalytic activity simultaneously. In particular, the measured OER activity was higher than state-ofthe-art IrO2 catalyst at relatively low potential (<1.6 V), reflected by negative onset potentials and higher current densities. The exceptionally improved catalytic activities probably resulted from the oxygen vacancies in h-BaTiO3 crystal structures, facilitating reactants adsorption and charge transfer. This work demonstrated an effective strategy to further engineering perovskite oxides by creating oxygen vacancy to deform crystalline structures and surface electronic structures of oxides with enhanced catalytic activity

for bifunctional catalysts. A more detailed understanding of the mechanism for oxygen-deficiency on perovskite and other oxides by using theoretical computation and simulation can generate rational rules for advanced catalyst design and synthesis [125].

2.6.2. CNT supported single metal oxide composites Recently, we have successfully developed a cobalt oxide catalyst supported on oxidized carbon nanotubes (Co3O4/oCNTs), showing decent ORR/OER bifunctional activity, as well as advanced durability compared to other hybrid structures reported in literatures [126]. The TEM images of NH3-treated Co3O4/oCNT (Fig. 6a) indicated the Co3O4 particle size was between 20 and 50 nm in diameter with blunt edges. High resolution TEM image shown in Fig. 6b clearly confirmed the hybrid structure of a single Co3O4 particle supported on CNT from the NH3-treated sample. Part of the Co3O4 particle was imbedded into the CNT even after heat treatment and a strong sonication, suggesting a strong interaction between CNT and Co3O4 particles. The ORR polarization curve of Co3O4/oCNTs was compared with commercial 50 wt% Pt/C (TEC10E50E, Tanaka) (Fig. 6c), with the same total catalyst loading, in O2-saturated 1.0 M KOH electrolyte. Both the ORR onset and half-wave potentials of Co3O4/oCNTs were only about 0.1 V negative of that of commercial Pt/C. In addition, the OER onset potential, 1.43 V, was only 0.05 V lower than that of commercial iridium black (Fig. 6d), equaling the best-known Co3O4-based hybrid OER catalysts [127]. The excellent

Please cite this article in press as: M. Chen, et al., Journal of Power Sources (2017), http://dx.doi.org/10.1016/j.jpowsour.2017.08.062

8

M. Chen et al. / Journal of Power Sources xxx (2017) 1e14

Fig. 4. (a) ORR and OER test of BSCF and BSCF/AB composite catalyst in oxygen saturated 0.1 M KOH electrolyte. (b) XANES spectra at the Co K edge for the BSCF and the BSCF/AB composite catalysts. (c) Scheme of ORR mechanism for the modified carbon black-LaMnO3 hybrid. (d) O 1s spectra of carbon black-LaMnO3 catalyst. Reprint with permission from Ref. [99], Copyright Wiley-VCH.

Table 2 Comparison of bifunctional catalysts towards ORR and OER in alkaline media. Catalyst

Synthesis method

ORR activity

OER potential @ 10 mA cm2

Reference

h-BaTiO3-x

sol-gel

onset potential 0.87 V vs RHE

[118]

NiCo2O4 MnO2-LaNiO3 La0.5Sr0.5CoO2.91/AC Mn oxide thin film CoO-NiO-NiCo/NCNT Ni-NiO/N-rGO Co-CoO/N-rGO La0.8Sr0.2MnO3 Mn-Co Oxide/NCNTs Co3O4/N-rmGO CoO/NCNTs NCNT/CoxMn1xO

electrospinning co-precipitation micro-emulsion electrodeposition hydrothermal hydrothermal hydrothermal co-precipitation thermal scission hydrothermal solvothermal hydrothermal

onset potential 0.93 V vs RHE e half-wave potential 0.77 V vs RHE 0.73 V @ 3 mA/cm2 vs RHE half-wave potential 0.83 V vs RHE e half-wave potential 0.78 V vs RHE e onset potential ~0.92 V vs RHE 0.83 vs RHE onset potential ~20 mV negative to that of Pt/C onset potential ~0.96 V vs RHE

onset potential 1.3 V vs RHE 1.62 V vs RHE 0.675 V vs MOE 1.83 V vs RHE 1.77 V vs RHE 1.5 V vs RHE 1.47 V vs RHE 1.62 V vs RHE onset potential 0.60 V vs Ag/AgCl onset potential 1.495 V vs RHE 1.54 vs RHE ~20 mV more negative than Ir/C @ 50 mA cm2 Overpotential of 0.34 V@10 mA cm2

bifunctional activities demonstrated by the Co3O4/oCNTs were among the highest when compared to other literatures [62,128]. From 0.5 V to 0.75 V, linear KouteckyLevich plots of Co3O4/oCNTs were presented in Fig. 6e by plotting J1 vs. u1/2 [129]. The average number of electrons transferred in ORR was around 4, indicating an efficient four-electron transfer mechanism. The stability of Co3O4/

[119] [120] [121] [5] [17] [16] [16] [122] [123] [62] [124] [18]

oCNTs was investigated over 2000 cycles by potential cycling from 0 to 1.9 V, at scan rate of 500 mV/s. The intermediate polarization curves were collected at 1600 rpm, 20 mV/s (Fig. 6f), suggesting that there was no significant change for both activities even after 2000 cycles in this harsh oxidizing environment, which has been rarely presented in previous reports.

Please cite this article in press as: M. Chen, et al., Journal of Power Sources (2017), http://dx.doi.org/10.1016/j.jpowsour.2017.08.062

M. Chen et al. / Journal of Power Sources xxx (2017) 1e14

9

Fig. 5. (a) Polarization plots of ORR and OER for Ir-550 and BTO-1300VAC in 0.1 M NaOH and (b) structure of oxygen-deficicney BaTiO3-x. Reprint with permission from Ref. [118], Copyrith Elsevier Ltd.

Fig. 6. (a) TEM image of Co3O4/oCNT. (b) High-resolution TEM image of Co3O4/oCNT. (c) ORR polarization curves of Co3O4/oCNT, reference catalyst of Pt/C (rotating rate: 1600 rpm, scan rate: 20 mV/s) in 1.0 M KOH electrolyte. (d) OER polarization curves of Co3O4/oCNT, reference catalyst of Ir black (rotating rate: 1600 rpm, scan rate: 20 mV/s) in 1.0 M KOH electrolyte. (e) KouteckyLevich plots of Co3O4/oCNT from 0.5 V to 0.75V. (f) Durability test of Co3O4/oCNT for 2000 cycles. Reprint with permission from Ref. [126], Copyright Elsevier Ltd.

2.6.3. CNT supported binary metal oxide composites Using binary metal oxides, furthermore, we developed one unique core-shell NiCo@CoO-NiO nanocomposite bifunctional electrocatalyst supported on nitrogen-doped multiwall carbon nanotubes (NCNT/NiCo@CoO-NiO) via hydrothermal synthesis (Fig. 7) [17]. The core-shell structures were determined by using XRD and XPS. XRD, as a bulk technique, indicated that the dominate phase in the sample was metallic CoNi alloy structure. Meanwhile, XPS, a relatively surface layer characterization, further suggested

the CoO and NiO were rich on the surface layer. Therefore, the prepared CoNi oxides were core-shell structures. Such unique configuration with an alloy core can address the poor electrical conductivity of traditional oxide catalysts. The oxide shells provided excellent activity and durability towards the ORR and OER in alkaline electrolyte. As seen from SEM and TEM images (Fig. 7a and b), NiCo@CoO-NiO composite nanoparticles were successfully deposited on the nitrogen-doped carbon nanotubes. For ORR tests (Fig. 7c and e), the NCNT/NiCo@CoO-NiO catalyst exhibited an E1/2

Please cite this article in press as: M. Chen, et al., Journal of Power Sources (2017), http://dx.doi.org/10.1016/j.jpowsour.2017.08.062

10

M. Chen et al. / Journal of Power Sources xxx (2017) 1e14

of 0.83 V which was only negative by 40 mV compared to commercial Pt/C catalyst. The electron transfer number calculated from 0.3 to 0.8 V was 3.85, which indicated that this catalyst was able to catalyze ORR via a highly desirable 4-electron pathway. As for OER studies (Fig. 7d and f), the NCNT/NiCo@CoO-NiO exhibited a remarkable low overpotential of 0.27 V at 10 mA cm2, surpassing the state of the art IrO2 catalyst (0.39 V). Furthermore, the Tafel slope of NCNT/NiCo@CoO-NiO was much lower than any other catalysts mentioned in this literature. Notably, the significant catalytic activity of NCNT/NiCo@CoO-NiO catalyst was attributed to NiCo@CoO-NiO rather than NCNT, due to the poor activity exhibited by NCNT itself. In addition, The OER stability measurement was tested at a constant potential of 1.5 V for 20000 s. The current density increases from 8 to 12 mA cm2 at the starting 5000 s and then remained nearly constant at next 15000 s. The stability test was also conducted on IrO2 at a constant potential. The current density rapidly decreased to 4 mA cm2 during the starting 2000 s, indicating lower stability and sluggish reaction kinetics of IrO2 compared with NCNT/NiCo@CoO-NiO. In the ORR stability measurement, 97% of current density was remained for NCNT/ NiCo@CoO-NiO during the ORR operation of 40000 s. On the contrast, the relative current density of Pt/C decreased to 87%. This result suggested NCNT/NiCo@CoO-NiO was intrinsically more stable than Pt/C under ORR conditions. This work provided a new pathway to design bifunctional oxides/nanocarbon composite catalysts for ORR/OER in fuel cell. 2.6.4. In-situ formation of graphene tube/metal composites In addition to depositing prepared metal oxides onto commercially available carbon tubes, metals and tubes can be in-situ formed via a one-step carbonization process of inexpensive

dicyandiamide (DCDA) (Fig. 8a) [3]. Compared to traditional carbon nanotubes with diameter of 10e50 nm, the carbon tubes derived from DCDA are doped with nitrogen heteroatom and have larger sizes of diameter up to 500 nm. Besides the large diameter size, their thickness of wall is usually very thin and less than 10 layers of carbon planes (Fig. 8b) [3]. Therefore, such large-size and thinlayered carbon tubes are itemed as graphene tubes (GTs). Importantly, the size of tubes can be largely controlled by varying the types of transition metals (Fe, Co, and Ni) during the carbonization [130]. As shown in Fig. 8c, GTs derived from individual Fe yield diameter around 100 nm, while tubes derived from binary CoNi and ternary FeCoNi generate much larger diameters around 200 and 500 nm, respectively. It should be noted that metallic alloy/oxides particles formed during the tube growth and uniformly deposit onto tubes. Therefore, the nitrogen-doped graphene tubes supported metals/oxides are considered as a new class of oxide/ nanocarbon composite catalyst for bifunctional application. Their activity and stability were systematically studied using identical potential cycling conditions from 0 to 1.9 V. Unfortunately, the most active Fe-based tubes with the highest ORR and OER activity were not stable, showing significant degradation after 1000 cycles. However, the CoNi and FeCoNi-derived tubes demonstrated sufficiently high ORR and OER activity, also maintaining good stability (Fig. 8d). Further characterization indicated the tubes derived from CoNi and FeCoNi exhibited much high graphitization degree, thicker wall, and relatively high graphitic nitrogen doping, which were crucial for enhancing stability of nanocarbon. It should be noted that, after thoroughly acidic treatment to remove metal alloys/oxides, the stability of the FeCoNi-derived carbon tubes declined dramatically, indicating that the highly OER-active metal oxides were able to protect carbon from oxidizing in corrosive

Fig. 7. (a) SEM image of NCNT/NiCo@CoO-NiO. (b) TEM images of NCNT/NiCo@CoO-NiO. (c) ORR polarization curves of NCNT/NiCo@CoO-NiO, reference catalysts, and Pt/C (rotating rate: 1600 rpm, scan rate: 10 mV/s, the loading is 0.21 mg cm2) in 1.0 M KOH electrolyte. d) OER polarization curves of NCNT/NiCo@CoO-NiO, reference catalysts, and IrO2 (rotating rate: 1600 rpm, scan rate: 10 mV/s, the loading is 0.21 mg cm2) in 1.0 M KOH electrolyte. (e) The Tafel plots of NCNT/NiCo@CoO-NiO, reference catalysts, and Pt/C derived from ORR polarization. f) The Tafel plots of NCNT/NiCo@CoO-NiO, reference catalysts, and IrO2 derived from OER polarization curves. Reprint with permission from Ref. [17], Copyright WilleyVCH.

Please cite this article in press as: M. Chen, et al., Journal of Power Sources (2017), http://dx.doi.org/10.1016/j.jpowsour.2017.08.062

M. Chen et al. / Journal of Power Sources xxx (2017) 1e14

11

Fig. 8. Large size graphene tube supported metal catalysts for bifunctional ORR and OER catalysis. (a) synthesis procedure, (b) scheme of graphene tubes/metal composites, (c) tunable size of graphene tubes from 100 to 500 nm by using various metals, and (d) corresponding stability determined by potential cycling (0e1.9 V) in 0.1 M NaOH. Reprint with permission from Ref. [3], Copyright Elsevier Ltd.

environments. The unique metal/graphene tube nanocomposite as ORR/OER bifunctional catalysts will open up a new direction to explore the promotional role of carbon nanomaterials in oxygen electrocatalysis. 3. Summary and perspective Development of high-performance bifunctional catalysts simultaneously active and stable towards the ORR and the OER is highly demanded for reversible alkaline fuel cells, which is considered an innovative technology for effectively utilizing renewable energy sources. State-of-the-art catalysts for ORR and OER applied in fuel cells and electrolyzers are Pt- and Ir-based catalysts, respectively, which are not ideal options for large-scale deployment of such clean energy technologies. Exploring highly efficient PGM-free catalysts active for the ORR and the OER therefore becomes important to replace these precious metals. Due to the differences of active sties and reaction mechanisms, there is no any single site suitable for the ORR and OER simultaneously.

Nanocomposites consisting of ORR and OER active components would the most effective catalyst systems. Among studied catalyst formulations, both transition metal oxides and nanocarbons have attracted significant attention due to their intrinsic activity towards OER and ORR along with sufficient stability in alkaline media. However, because of their respective downsides including low surface-area and electrical conductivity for oxides and poor OER activity/stability for nanocarbons, each component alone cannot become high-performance bifunctional ORR/OER catalysts. To overcome their problems and combine their advantages, design and synthesis of nanocarbon/oxide nanocomposites therefore are highly desirable for the challenging bifunctional ORR/OER catalysis. The key is to achieve their optimal interfacial properties with a favorable synergy, which could be due to their electronic effect and chemical bonding. Engineering the interface between oxides and nanocarbon supports would strengthen their interactions and facilitate the charge transfer during the ORR and OER. In addition to chemical properties, catalyst morphology including microstructures, surface areas, and porosity also play important role in

Please cite this article in press as: M. Chen, et al., Journal of Power Sources (2017), http://dx.doi.org/10.1016/j.jpowsour.2017.08.062

12

M. Chen et al. / Journal of Power Sources xxx (2017) 1e14

boosting catalyst activity. Novel synthesis methods are required to prepare oxide/nanocarbon composites with intimate contact at nanoscale and molecular levels. A wide variety of oxides can be selected for nanocomposites including single oxides and complex oxides (i.e., perovskite and spinel). Compared to other carbon structures, carbon tubes have been discovered as the effective nanocarbon due to significantly enhanced OER stability in alkaline media. Despite promising performance was reported by others and us, highly stable nanocarbon/ oxide catalysts for long-term practical applications still remain a primary challenge. Overall, although the carbon corrosion cannot be completely avoided during the OER, the synergistic effect between oxides and nanocarbons can significantly enhance stability. Highly active and stable oxides can mitigate the corrosion rates of carbon in nanocomposites. This represents an effective strategy to further develop highly stable and active nanocomposites for reversible alkaline fuel cells. In this mini-review, we didn't discuss the electrode design and fabrication at membrane electrode assembly (MEA) level, but it is extremely important to transfer intrinsic activity/stability of nanocomposite catalysts into high performance of electrodes at device level. This will be one of our future focuses. Acknowledgements The authors acknowledge financial support for the start-up funding from the University at Buffalo along with National Science Foundation (CBET-1604392) and U.S. Department of Energy, Fuel Cell Technologies Office (FCTO) Incubator Program (DEEE000696). H. Yang thanks the financial support from the Guangdong Natural Science Foundation (2015A030313545) and the Shenzhen Science and Technology Research Grant (JCYJ20150324140036855). References [1] H. Osgood, S.V. Devaguptapu, H. Xu, J. Cho, G. Wu, Transition metal (Fe, Co, Ni, and Mn) oxides for oxygen reduction and evolution bifunctional catalysts in alkaline media, Nano Today 11 (2016) 601e625. [2] S. Gupta, W. Kellogg, H. Xu, X. Liu, J. Cho, G. Wu, Bifunctional perovskite oxide catalysts for oxygen reduction and evolution in alkaline media, ChemistryeAn Asian J. 11 (2016) 10e21. [3] S. Gupta, L. Qiao, S. Zhao, H. Xu, Y. Lin, S.V. Devaguptapu, X. Wang, M.T. Swihart, G. Wu, Highly Active, Stable Graphene, Tubes decorated with FeCoNi alloy nanoparticles via a template-free graphitization for bifunctional oxygen reduction and evolution, Adv. Energy Mater. 6 (2016) 1601198. [4] F. Mitlitsky, B. Myers, A.H. Weisberg, Regenerative fuel cell systems, Energy & Fuels 12 (1998) 56e71. [5] Y. Gorlin, T.F. Jaramillo, A bifunctional nonprecious metal catalyst for oxygen reduction and water oxidation, J. Am. Chem. Soc. 132 (2010) 13612e13614. [6] M. Busch, N.B. Halck, U.I. Kramm, S. Siahrostami, P. Krtil, J. Rossmeisl, Beyond the top of the volcano? e A unified approach to electrocatalytic oxygen reduction and oxygen evolution, Nano Energy 29 (2016) 126e135. [7] E. Antolini, Iridium as catalyst and cocatalyst for oxygen evolution/reduction in acidic polymer electrolyte membrane electrolyzers and fuel cells, ACS Catal. 4 (2014) 1426e1440. [8] K. Gong, F. Du, Z. Xia, M. Durstock, L. Dai, Nitrogen-doped carbon nanotube arrays with high electrocatalytic activity for oxygen reduction, Science 323 (2009) 760e764.
vre, E. Proietti, F. Jaouen, J.-P. Dodelet, Iron-based catalysts with [9] M. Lefe improved oxygen reduction activity in polymer electrolyte fuel cells, Science 324 (2009) 71e74. [10] J. Suntivich, K.J. May, H.A. Gasteiger, J.B. Goodenough, Y. Shao-Horn, A perovskite oxide optimized for oxygen evolution catalysis from molecular orbital principles, Science 334 (2011) 1383e1385. [11] G. Wu, K.L. More, C.M. Johnston, P. Zelenay, High-performance electrocatalysts for oxygen reduction derived from polyaniline, iron, and cobalt, Science 332 (2011) 443e447. [12] G. Wu, H.T. Chung, M. Nelson, K. Artyushkova, K.L. More, C.M. Johnston, P. Zelenay, Graphene-riched Co9S8-NC non-precious metal catalyst for oxygen reduction in alkaline media, ECS Trans. 41 (2011) 1709e1717. [13] J.S. Spendelow, A. Wieckowski, Electrocatalysis of oxygen reduction and small alcohol oxidation in alkaline media, Phys. Chem. Chem. Phys. 9 (2007) 2654e2675.

[14] G. Liu, X. Li, P. Ganesan, B.N. Popov, Studies of oxygen reduction reaction active sites and stability of nitrogen-modified carbon composite catalysts for PEM fuel cells, Electrochimica Acta 55 (2010) 2853e2858. [15] X. Li, G. Liu, B.N. Popov, Activity and stability of non-precious metal catalysts for oxygen reduction in acid and alkaline electrolytes, J. Power Sources 195 (2010) 6373e6378. [16] X. Liu, W. Liu, M. Ko, M. Park, M.G. Kim, P. Oh, S. Chae, S. Park, A. Casimir, G. Wu, Metal (Ni, Co)-Metal oxides/graphene nanocomposites as multifunctional electrocatalysts, Adv. Funct. Mater. 25 (2015) 5799e5808. [17] X. Liu, M. Park, M.G. Kim, S. Gupta, G. Wu, J. Cho, Integrating NiCo alloys with their oxides as efficient bifunctional cathode catalysts for rechargeable zinceair batteries, Angew. Chem. Int. Ed. 54 (2015) 9654e9658. [18] X. Liu, M. Park, M.G. Kim, S. Gupta, X. Wang, G. Wu, J. Cho, High-performance non-spinel cobaltemanganese mixed oxide-based bifunctional electrocatalysts for rechargeable zinceair batteries, Nano Energy 20 (2016) 315e325. [19] G. Wu, A. Santandreu, W. Kellogg, S. Gupta, O. Ogoke, H. Zhang, H.-L. Wang, L. Dai, Carbon nanocomposite catalysts for oxygen reduction and evolution reactions: from nitrogen doping to transition-metal addition, Nano Energy 29 (2016) 83e110. [20] G. Wu, M.A. Nelson, N.H. Mack, S. Ma, P. Sekhar, F.H. Garzon, P. Zelenay, Titanium dioxide-supported non-precious metal oxygen reduction electrocatalyst, Chem. Commun. 46 (2010) 7489e7491. [21] H. Sheng, M. Wei, A. D'Aloia, G. Wu, Heteroatom polymer-derived 3D highsurface-area and mesoporous graphene sheet-like carbon for supercapacitors, ACS Appl. Mater. Interfaces 8 (2016) 30212e30224. [22] Q. Li, R. Cao, J. Cho, G. Wu, Nanocarbon electrocatalysts for oxygen-reduction in alkaline media for advanced energy conversion and storage, Adv. Energy Mater. 4 (2014) 1301415. [23] Q. Li, R. Cao, J. Cho, G. Wu, Nanostructured carbon-based cathode catalysts for nonaqueous lithium-oxygen batteries, Phys. Chem. Chem. Phys. 16 (2014) 13568e13582. [24] H. Zhang, H. Osgood, X. Xie, Y. Shao, G. Wu, Engineering nanostructures of PGM-free oxygen-reduction catalysts using metal-organic frameworks, Nano Energy 31 (2017) 331e350. [25] E. Antolini, Graphene as a new carbon support for low-temperature fuel cell catalysts, Appl. Catal. B Environ. 123 (2012) 52e68. [26] J. Zhang, L. Dai, Heteroatom-doped graphitic carbon catalysts for efficient electrocatalysis of oxygen reduction reaction, ACS Catal. 5 (2015) 7244e7253. [27] Q. Li, P. Xu, W. Gao, S. Ma, G. Zhang, R. Cao, J. Cho, H.-L. Wang, G. Wu, Graphene/graphene tube nanocomposites templated from cage-containing metal-organic frameworks for oxygen reduction in Li-O2 batteries, Adv. Mater. 26 (2014) 1378e1386. [28] G. Wu, L. Li, J.-H. Li, B.-Q. Xu, Polyaniline-carbon composite films as supports of Pt and PtRu particles for methanol electrooxidation, Carbon 43 (2005) 2579e2587. [29] G. Wu, K.L. More, P. Xu, H.-L. Wang, M. Ferrandon, A.J. Kropf, D.J. Myers, S. Ma, P. Zelenay, Carbon-nanotube-supported graphene-rich non-precious metal oxygen reduction catalyst with enhanced performance durability, Chem. Commun. 49 (2013) 3291e3293. [30] Q. Li, H. Pan, D. Higgins, R. Cao, G. Zhang, H. Lv, K. Wu, J. Cho, G. Wu, Metalorganic framework derived bamboo-like nitrogen-doped graphene tubes as an active matrix for hybrid oxygen-reduction electrocatalysts, Small 11 (2015) 1443e1452. [31] G. Wu, Y.-S. Chen, B.-Q. Xu, Remarkable support effect of SWNTs in Pt catalyst for methanol electrooxidation, Electrochem. Commun. 7 (2005) 1237e1243. [32] W. Gao, D. Havas, S. Gupta, Q. Pan, N. He, H. Zhang, H.-L. Wang, G. Wu, Is reduced graphene oxide favorable for nonprecious metal oxygen-reduction catalysts? Carbon 102 (2016) 346e356. [33] X. Ge, Y. Liu, F.T. Goh, T.A. Hor, Y. Zong, P. Xiao, Z. Zhang, S.H. Lim, B. Li, X. Wang, Dual-phase spinel MnCo2O4 and spinel MnCo2O4/nanocarbon hybrids for electrocatalytic oxygen reduction and evolution, ACS Appl. Mater. interfaces 6 (2014) 12684e12691. [34] W.T. Hong, M. Risch, K.A. Stoerzinger, A. Grimaud, J. Suntivich, Y. Shao-Horn, Toward the rational design of non-precious transition metal oxides for oxygen electrocatalysis, Energy & Environ. Sci. 8 (2015) 1404e1427. [35] Y. Zhu, W. Zhou, Z. Shao, Perovskite/carbon composites: applications in oxygen electrocatalysis, Small 13 (2017) 1603793. [36] S. Guo, S. Zhang, L. Wu, S. Sun, Co/CoO nanoparticles assembled on graphene for electrochemical reduction of oxygen, Angew. Chem. 124 (2012) 11940e11943. [37] L. Wang, X. Zhao, Y. Lu, M. Xu, D. Zhang, R.S. Ruoff, K.J. Stevenson, J.B. Goodenough, CoMn2O4 spinel nanoparticles grown on graphene as bifunctional catalyst for lithium-air batteries, J. Electrochem. Soc. 158 (2011) A1379eA1382. [38] V. Neburchilov, H. Wang, J.J. Martin, W. Qu, A review on air cathodes for zinceair fuel cells, J. Power Sources 195 (2010) 1271e1291. pez-Sua rez, D. Lozano-Castello , A. Bueno[39] Q. He, Q. Li, S. Khene, X. Ren, F.E. Lo pez, G. Wu, High-loading cobalt oxide coupled with nitrogen-doped graLo phene for oxygen reduction in anion-exchange-membrane alkaline fuel cells, J. Phys. Chem. C 117 (2013) 8697e8707. [40] N. Tyminska, G. Wu, M. Dupuis, Water oxidation on oxygen-deficient barium titanate: a first principles study, J. Phys. Chem. C 121 (2017) 8378e8389.

Please cite this article in press as: M. Chen, et al., Journal of Power Sources (2017), http://dx.doi.org/10.1016/j.jpowsour.2017.08.062

M. Chen et al. / Journal of Power Sources xxx (2017) 1e14 [41] K. Mamtani, D. Jain, D. Dogu, V. Gustin, S. Gunduz, A.C. Co, U.S. Ozkan, Insights into oxygen reduction reaction (ORR) and oxygen evolution reaction (OER) active sites for nitrogen-doped carbon nanostructures (CNx) in acidic media, Appl. Catal. B Environ. (2017), http://dx.doi.org/10.1016/ j.apcatb.2017.1007.1086. [42] H.B. Yang, J. Miao, S.-F. Hung, J. Chen, H.B. Tao, X. Wang, L. Zhang, R. Chen, J. Gao, H.M. Chen, Identification of catalytic sites for oxygen reduction and oxygen evolution in N-doped graphene materials: development of highly efficient metal-free bifunctional electrocatalyst, Sci. Adv. 2 (2016) e1501122. [43] G. Wu, P. Zelenay, Nanostructured nonprecious metal catalysts for oxygen reduction reaction, Accounts Chem. Res. 46 (2013) 1878e1889. [44] Q. Jia, N. Ramaswamy, H. Hafiz, U. Tylus, K. Strickland, G. Wu, B. Barbiellini, A. Bansil, E.F. Holby, P. Zelenay, Experimental observation of redox-induced FeeN switching behavior as a determinant role for oxygen reduction activity, ACS Nano 9 (2015) 12496e12505. [45] S. Kattel, G. Wang, A density functional theory study of oxygen reduction reaction on MeeN4 (Me¼ Fe, Co, or Ni) clusters between graphitic pores, J. Mater. Chem. A 1 (2013) 10790e10797. [46] Z. Qiao, H. Zhang, S. Karakalos, S. Hwang, J. Xue, M. Chen, D. Su, G. Wu, 3D polymer hydrogel for high-performance atomic iron-rich catalysts for oxygen reduction in acidic media, Appl. Catal. B Environ. 219 (2017) 629e639. [47] S. Gupta, S. Zhao, O. Ogoke, Y. Lin, H. Xu, G. Wu, Engineering favorable morphology and structure of Fe-N-C oxygen-reduction catalysts via tuning nitrogen/carbon precursors, ChemSusChem 10 (2017) 774e785. [48] X. Wang, Y. Ke, H. Pan, K. Ma, Q. Xiao, D. Yin, G. Wu, M.T. Swihart, Cudeficient plasmonic Cu2exS nanoplate electrocatalysts for oxygen reduction, ACS Catal. 5 (2015) 2534e2540. [49] G. Wu, N. Li, D.-R. Zhou, K. Mitsuo, B.-Q. Xu, Anodically electrodeposited Coþ Ni mixed oxide electrode: preparation and electrocatalytic activity for oxygen evolution in alkaline media, J. Solid State Chem. 177 (2004) 3682e3692. [50] M. Yagi, E. Tomita, T. Kuwabara, Remarkably high activity of electrodeposited IrO2 film for electrocatalytic water oxidation, J. Electroanal. Chem. 579 (2005) 83e88. [51] K.J. May, C.E. Carlton, K.A. Stoerzinger, M. Risch, J. Suntivich, Y.-L. Lee, A. Grimaud, Y. Shao-Horn, Influence of oxygen evolution during water oxidation on the surface of perovskite oxide catalysts, J. Phys. Chem. Lett. 3 (2012) 3264e3270. [52] M. Pena, J. Fierro, Chemical structures and performance of perovskite oxides, Chem. Rev. 101 (2001) 1981e2018. [53] J. Suntivich, H.A. Gasteiger, N. Yabuuchi, H. Nakanishi, J.B. Goodenough, Y. Shao-Horn, Design principles for oxygen-reduction activity on perovskite oxide catalysts for fuel cells and metaleair batteries, Nat. Chem. 3 (2011) 546e550. [54] J. Duan, S. Chen, M. Jaroniec, S.Z. Qiao, Heteroatom-doped graphene-based materials for energy-relevant electrocatalytic processes, ACS Catal. 5 (2015) 5207e5234. [55] J.M. Bockris, T. Otagawa, V. Young, Solid state surface studies of the electrocatalysis of oxygen evolution on perovskites, J. Electroanal. Chem. Interfacial Electrochem. 150 (1983) 633e643. [56] J.B. Goodenough, R. Manoharan, M. Paranthaman, Surface protonation and electrochemical activity of oxides in aqueous solution, J. Am. Chem. Soc. 112 (1990). [57] A. Grimaud, K.J. May, C.E. Carlton, Y.-L. Lee, M. Risch, W.T. Hong, J. Zhou, Y. Shao-Horn, Double perovskites as a family of highly active catalysts for oxygen evolution in alkaline solution, Nat. Commun. 4 (2013) 2439. [58] A. Serov, N.I. Andersen, A.J. Roy, I. Matanovic, K. Artyushkova, P. Atanassov, CuCo2O4 ORR/OER bi-functional catalyst: influence of synthetic approach on performance, J. Electrochem. Soc. 162 (2015) F449eF454. [59] W. Bragg, The structure of magnetite and the spinels, Nature 95 (1915) 561. [60] M.S. El-Deab, T. Ohsaka, Manganese oxide nanoparticles electrodeposited on platinum are superior to platinum for oxygen reduction, Angew. Chem. Int. Ed. 45 (2006) 5963e5966. k, Carbon-supported manganese [61] I. Roche, E. Chaînet, M. Chatenet, J. Vondra oxide nanoparticles as electrocatalysts for the oxygen reduction reaction (ORR) in alkaline medium: physical characterizations and ORR mechanism, J. Phys. Chem. C 111 (2007) 1434e1443. [62] F. Cheng, J. Shen, B. Peng, Y. Pan, Z. Tao, J. Chen, Rapid room-temperature synthesis of nanocrystalline spinels as oxygen reduction and evolution electrocatalysts, Nat. Chem. 3 (2011) 79e84. [63] P. Strasser, S. Koh, T. Anniyev, J. Greeley, K. More, C. Yu, Z. Liu, S. Kaya, D. Nordlund, H. Ogasawara, Lattice-strain control of the activity in dealloyed coreeshell fuel cell catalysts, Nat. Chem. 2 (2010) 454e460. [64] J. Greeley, I. Stephens, A. Bondarenko, T.P. Johansson, H.A. Hansen, T. Jaramillo, J. Rossmeisl, I. Chorkendorff, J.K. Nørskov, Alloys of platinum and early transition metals as oxygen reduction electrocatalysts, Nat. Chem. 1 (2009) 552e556. [65] R. Singh, J. Pandey, N. Singh, B. Lal, P. Chartier, J.-F. Koenig, Sol-gel derived spinel M x Co3 xO4 (M¼ Ni, Cu; 0 x 1) films and oxygen evolution, Electrochimica Acta 45 (2000) 1911e1919. [66] J.O. Bockris, T. Otagawa, Mechanism of oxygen evolution on perovskites, J. Phys. Chem. 87 (1983) 2960e2971. [67] S. Malkhandi, B. Yang, A. Manohar, A. Manivannan, G.S. Prakash, S. Narayanan, Electrocatalytic properties of nanocrystalline calcium-doped lanthanum cobalt oxide for bifunctional oxygen electrodes, J. Phys. Chem. Lett. 3 (2012) 967e972.

13

[68] J. Sunarso, A.A. Torriero, W. Zhou, P.C. Howlett, M. Forsyth, Oxygen reduction reaction activity of La-based perovskite oxides in alkaline medium: a thinfilm rotating ring-disk electrode study, J. Phys. Chem. C 116 (2012) 5827e5834. [69] F. Meyer, A. Dierstein, C. Beck, W. H€ artl, R. Hempelmann, S. Mathur, M. Veith, Size-controlled synthesis of nanoscaled aluminium spinels using heterobimetallic alkoxide precursors via water/oil microemulsions, Nanostructured Mater. 12 (1999) 71e74. [70] M. Yuasa, M. Nishida, T. Kida, N. Yamazoe, K. Shimanoe, Bi-functional oxygen electrodes using LaMnO3/LaNiO3 for rechargeable metal-air batteries, J. Electrochem. Soc. 158 (2011) A605eA610. [71] J. Suntivich, H.A. Gasteiger, N. Yabuuchi, Y. Shao-Horn, Electrocatalytic measurement methodology of oxide catalysts using a thin-film rotating disk electrode, J. Electrochem. Soc. 157 (2010) B1263eB1268. [72] G. Demazeau, Solvothermal processes: a route to the stabilization of new materials, J. Mater. Chem. 9 (1999) 15e18. [73] J. Ovenstone, K. Chan, C. Ponton, Hydrothermal processing and characterisation of doped lanthanum chromite for use in SOFCs, J. Mater. Sci. 37 (2002) 3315e3322. [74] M. Prabu, P. Ramakrishnan, S. Shanmugam, CoMn2O4 nanoparticles anchored on nitrogen-doped graphene nanosheets as bifunctional electrocatalyst for rechargeable zinceair battery, Electrochem. Commun. 41 (2014) 59e63. [75] J.P. Meyers, R.M. Darling, Model of carbon corrosion in PEM fuel cells, J. Electrochem. Soc. 153 (2006) A1432eA1442. [76] T. Kim, T. Xie, W. Jung, F. Gadala-Maria, P. Ganesan, B.N. Popov, Development of catalytically active and highly stable catalyst supports for polymer electrolyte membrane fuel cells, J. Power Sources 273 (2015) 761e774. [77] G.L. Tian, M.Q. Zhao, D. Yu, X.Y. Kong, J.Q. Huang, Q. Zhang, F. Wei, Nitrogendoped graphene/carbon nanotube hybrids: in situ formation on bifunctional catalysts and their superior electrocatalytic activity for oxygen evolution/ reduction reaction, Small 10 (2014) 2251e2259. [78] J. Masa, W. Xia, I. Sinev, A. Zhao, Z. Sun, S. Grützke, P. Weide, M. Muhler, W. Schuhmann, MnxOy/nc and CoxOy/NC nanoparticles embedded in a nitrogen-doped carbon matrix for high-performance bifunctional oxygen electrodes, Angew. Chem. Int. Ed. 53 (2014) 8508e8512. [79] Y. Su, Y. Zhu, H. Jiang, J. Shen, X. Yang, W. Zou, J. Chen, C. Li, Cobalt nanoparticles embedded in N-doped carbon as an efficient bifunctional electrocatalyst for oxygen reduction and evolution reactions, Nanoscale 6 (2014) 15080e15089. [80] J. Zhang, Z. Zhao, Z. Xia, L. Dai, A metal-free bifunctional electrocatalyst for oxygen reduction and oxygen evolution reactions, Nat. Nanotechnol. 10 (2015) 444e452. [81] K. Qu, Y. Zheng, S. Dai, S.Z. Qiao, Graphene oxide-polydopamine derived N, Scodoped carbon nanosheets as superior bifunctional electrocatalysts for oxygen reduction and evolution, Nano Energy 19 (2016) 373e381. [82] R. Li, Z. Wei, X. Gou, Nitrogen and phosphorus dual-doped graphene/carbon nanosheets as bifunctional electrocatalysts for oxygen reduction and evolution, ACS Catal. 5 (2015) 4133e4142. [83] Y. Hou, Z. Wen, S. Cui, S. Ci, S. Mao, J. Chen, An advanced nitrogen-doped graphene/cobalt-embedded porous carbon polyhedron hybrid for efficient catalysis of oxygen reduction and water splitting, Adv. Funct. Mater. 25 (2015) 872e882. [84] Z. Wang, S. Xiao, Z. Zhu, X. Long, X. Zheng, X. Lu, S. Yang, Cobalt-embedded nitrogen doped carbon nanotubes: a bifunctional catalyst for oxygen electrode reactions in a wide pH range, ACS Appl. Mater. Interfaces 7 (2015) 4048e4055. [85] M. Shen, C. Ruan, Y. Chen, C. Jiang, K. Ai, L. Lu, Covalent entrapment of cobalteiron sulfides in N-doped mesoporous carbon: extraordinary bifunctional electrocatalysts for oxygen reduction and evolution reactions, ACS Appl. Mater. Interfaces 7 (2015) 1207e1218. [86] Y. Liu, H. Jiang, Y. Zhu, X. Yang, C. Li, Transition metals (Fe, Co, and Ni) encapsulated in nitrogen-doped carbon nanotubes as bi-functional catalysts for oxygen electrode reactions, J. Mater. Chem. A 4 (2016) 1694e1701. [87] Q. Liu, J. Jin, J. Zhang, NiCo2S4@graphene as a bifunctional electrocatalyst for oxygen reduction and evolution reactions, ACS Appl. Mater. interfaces 5 (2013) 5002e5008. [88] Y. Gao, H. Zhao, D. Chen, C. Chen, F. Ciucci, In situ synthesis of mesoporous manganese oxide/sulfur-doped graphitized carbon as a bifunctional catalyst for oxygen evolution/reduction reactions, Carbon 94 (2015) 1028e1036. [89] B. Chen, R. Li, G. Ma, X. Gou, Y. Zhu, Y. Xia, Cobalt sulfide/N, S codoped porous carbon coreeshell nanocomposites as superior bifunctional electrocatalysts for oxygen reduction and evolution reactions, Nanoscale 7 (2015) 20674e20684. [90] T.S. Olson, S. Pylypenko, J.E. Fulghum, P. Atanassov, Bifunctional oxygen reduction reaction mechanism on non-platinum catalysts derived from pyrolyzed porphyrins, J. Electrochem. Soc. 157 (2010) B54eB63. [91] H.W. Park, D.U. Lee, M.G. Park, R. Ahmed, M.H. Seo, L.F. Nazar, Z. Chen, Perovskiteenitrogen-doped carbon nanotube composite as bifunctional catalysts for rechargeable lithiumeair batteries, ChemSusChem 8 (2015) 1058e1065. [92] B.Y. Xia, Y. Yan, N. Li, H.B. Wu, X.W.D. Lou, X. Wang, A metaleorganic framework-derived bifunctional oxygen electrocatalyst, Nat. Energy 1 (2016) 15006. [93] Z. Chen, A. Yu, D. Higgins, H. Li, H. Wang, Z. Chen, Highly active and durable coreecorona structured bifunctional catalyst for rechargeable metaleair

Please cite this article in press as: M. Chen, et al., Journal of Power Sources (2017), http://dx.doi.org/10.1016/j.jpowsour.2017.08.062

14

M. Chen et al. / Journal of Power Sources xxx (2017) 1e14

battery application, Nano Lett. 12 (2012) 1946e1952. [94] C. Sun, F. Li, C. Ma, Y. Wang, Y. Ren, W. Yang, Z. Ma, J. Li, Y. Chen, Y. Kim, GrapheneeCo3O4 nanocomposite as an efficient bifunctional catalyst for lithiumeair batteries, J. Mater. Chem. A 2 (2014) 7188e7196. [95] G. Zhou, D.-W. Wang, L.-C. Yin, N. Li, F. Li, H.-M. Cheng, Oxygen bridges between NiO nanosheets and graphene for improvement of lithium storage, ACS Nano 6 (2012) 3214e3223. [96] D.A. Slanac, A. Lie, J.A. Paulson, K.J. Stevenson, K.P. Johnston, Bifunctional catalysts for alkaline oxygen reduction reaction via promotion of ligand and ensemble effects at Ag/MnO x nanodomains, J. Phys. Chem. C 116 (2012) 11032e11039. [97] X. Ma, H. Meng, M. Cai, P.K. Shen, Bimetallic carbide nanocomposite enhanced Pt catalyst with high activity and stability for the oxygen reduction reaction, J. Am. Chem. Soc. 134 (2012) 1954e1957. [98] T. Bligaard, J.K. Nørskov, Ligand effects in heterogeneous catalysis and electrochemistry, Electrochimica Acta 52 (2007) 5512e5516. [99] E. Fabbri, M. Nachtegaal, X. Cheng, T.J. Schmidt, Superior bifunctional electrocatalytic activity of Ba0. 5Sr0. 5Co0. 8Fe0.2O3-d/Carbon composite electrodes: insight into the local electronic structure, Adv. Energy Mater. 5 (2015). [100] A.S. Harvey, F.J. Litterst, Z. Yang, J.L. Rupp, A. Infortuna, L.J. Gauckler, Oxidation states of Co and Fe in Ba1 xSrx Co1 yFeyO3 d (x, y¼ 0.2e0.8) and oxygen desorption in the temperature range 300e1273 K, Phys. Chem. Chem. Phys. 11 (2009) 3090e3098. €tz, T. Schmidt, Developments and [101] E. Fabbri, A. Habereder, K. Waltar, R. Ko perspectives of oxide-based catalysts for the oxygen evolution reaction, Catal. Sci. Technol. 4 (2014) 3800e3821. [102] F. Calle-Vallejo, N.G. Inoglu, H.-Y. Su, J.I. Martínez, I.C. Man, M.T. Koper, J.R. Kitchin, J. Rossmeisl, Number of outer electrons as descriptor for adsorption processes on transition metals and their oxides, Chem. Sci. 4 (2013) 1245e1249. [103] A. Rabis, P. Rodriguez, T.J. Schmidt, Electrocatalysis for polymer electrolyte fuel cells: recent achievements and future challenges, ACS Catal. 2 (2012) 864e890. [104] M. Jahan, Z. Liu, K.P. Loh, A graphene oxide and copper-centered metal organic framework composite as a tri-functional catalyst for HER, OER, and ORR, Adv. Funct. Mater. 23 (2013) 5363e5372. [105] E. Antolini, J. Salgado, E. Gonzalez, Carbon supported Pt75M25 (M¼ Co, Ni) alloys as anode and cathode electrocatalysts for direct methanol fuel cells, J. Electroanal. Chem. 580 (2005) 145e154. [106] Z. Yang, X. Zhou, H. Nie, Z. Yao, S. Huang, Facile construction of manganese oxide doped carbon nanotube catalysts with high activity for oxygen reduction reaction and investigations into the origin of their activity enhancement, ACS Appl. Mater. interfaces 3 (2011) 2601e2606. [107] H. Liu, J. Liu, W. Song, F. Wang, Y. Song, Li x Co3-xO4 solid solution nanocrystals supported on carbon black as a superior electrocatalyst for oxygen reduction reaction, Mater. Lett. 139 (2015) 447e450. [108] J. Liu, X. Jin, W. Song, F. Wang, N. Wang, Y. Song, Facile preparation of modified carbon black-LaMnO3 hybrids and the effect of covalent coupling on the catalytic activity for oxygen reduction reaction, Chin. J. Catal. 35 (2014) 1173e1188. [109] Y. Liang, H. Wang, P. Diao, W. Chang, G. Hong, Y. Li, M. Gong, L. Xie, J. Zhou, J. Wang, Oxygen reduction electrocatalyst based on strongly coupled cobalt oxide nanocrystals and carbon nanotubes, J. Am. Chem. Soc. 134 (2012) 15849e15857. [110] Y. Liang, Y. Li, H. Wang, J. Zhou, J. Wang, T. Regier, H. Dai, Co3O4 nanocrystals on graphene as a synergistic catalyst for oxygen reduction reaction, Nat. Mater. 10 (2011) 780e786. [111] A. Weidenkaff, S.G. Ebbinghaus, T. Lippert, Ln1-xAx CoO3 (Ln¼ Er, La; A¼ Ca, Sr)/Carbon nanotube composite materials applied for rechargeable Zn/Air batteries, Chem. Mater. 14 (2002) 1797e1805. [112] D.U. Lee, H.W. Park, M.G. Park, V. Ismayilov, Z. Chen, Synergistic bifunctional

[113]

[114]

[115]

[116]

[117]

[118]

[119]

[120]

[121]

[122]

[123]

[124]

[125]

[126]

[127]

[128]

[129]

[130]

catalyst design based on perovskite oxide nanoparticles and intertwined carbon nanotubes for rechargeable zinceair battery applications, ACS Appl. Mater. interfaces 7 (2014) 902e910. J. Hu, L. Wang, L. Shi, H. Huang, Preparation of La1 xCaxMnO3 perovskiteegraphene composites as oxygen reduction reaction electrocatalyst in alkaline medium, J. Power Sources 269 (2014) 144e151. Z. Jian, P. Liu, F. Li, P. He, X. Guo, M. Chen, H. Zhou, Coreeshell-structured CNT@ RuO2 composite as a high-performance cathode catalyst for rechargeable LieO2 batteries, Angew. Chem. Int. Ed. 53 (2014) 442e446. N.I. Andersen, A. Serov, P. Atanassov, Metal oxides/CNT nano-composite catalysts for oxygen reduction/oxygen evolution in alkaline media, Appl. Catal. B Environ. 163 (2015) 623e627. €tz, O. Conrad, T.J. Schmidt, R. Mohamed, X. Cheng, E. Fabbri, P. Levecque, R. Ko Electrocatalysis of perovskites: the influence of carbon on the oxygen evolution activity, J. Electrochem. Soc. 162 (2015) F579eF586. J. Hu, Q. Liu, Z. Shi, L. Zhang, H. Huang, LaNiO3-nanorod/graphene composite as an efficient bi-functional catalyst for zinceair batteries, RSC Adv. 6 (2016) 86386e86394. C.-F. Chen, G. King, R.M. Dickerson, P.A. Papin, S. Gupta, W.R. Kellogg, G. Wu, Oxygen-deficient BaTiO3x perovskite as an efficient bifunctional oxygen electrocatalyst, Nano Energy 13 (2015) 423e432. M. Prabu, K. Ketpang, S. Shanmugam, Hierarchical nanostructured NiCo2O4 as an efficient bifunctional non-precious metal catalyst for rechargeable zinceair batteries, Nanoscale 6 (2014) 3173e3181. P.H. Benhangi, A. Alfantazi, E. Gyenge, Manganese dioxide-based bifunctional oxygen reduction/evolution electrocatalysts: effect of perovskite doping and potassium ion insertion, Electrochimica Acta 123 (2014) 42e50. Y. Zhao, L. Xu, L. Mai, C. Han, Q. An, X. Xu, X. Liu, Q. Zhang, Hierarchical mesoporous perovskite La0.5Sr0.5CoO2.91 nanowires with ultrahigh capacity for Li-air batteries, Proc. Natl. Acad. Sci. 109 (2012) 19569e19574. C. Jin, X. Cao, L. Zhang, C. Zhang, R. Yang, Preparation and electrochemical properties of urchin-like La0.8 Sr0.2 MnO3 perovskite oxide as a bifunctional catalyst for oxygen reduction and oxygen evolution reaction, J. Power Sources 241 (2013) 225e230. A. Zhao, J. Masa, W. Xia, A. Maljusch, M.-G. Willinger, G. Clavel, K. Xie, €gl, W. Schuhmann, M. Muhler, Spinel MneCo oxide in N-doped R. Schlo carbon nanotubes as a bifunctional electrocatalyst synthesized by oxidative cutting, J. Am. Chem. Soc. 136 (2014) 7551e7554. Y. Li, M. Gong, Y. Liang, J. Feng, J.-E. Kim, H. Wang, G. Hong, B. Zhang, H. Dai, Advanced zinc-air batteries based on high-performance hybrid electrocatalysts, Nat. Commun. 4 (2013) 1805. J. Kim, X. Yin, K.-C. Tsao, S. Fang, H. Yang, Ca2Mn2O5 as oxygen-deficient perovskite electrocatalyst for oxygen evolution reaction, J. Am. Chem. Soc. 136 (2014) 14646e14649. S. Zhao, B. Rasimick, W. Mustain, H. Xu, Highly durable and active Co3O4 nanocrystals supported on carbon nanotubes as bifunctional electrocatalysts in alkaline media, Appl. Catal. B Environ. 203 (2017) 138e145. S.K. Singh, V.M. Dhavale, S. Kurungot, Low surface energy plane exposed Co3O4 nanocubes supported on nitrogen-doped graphene as an electrocatalyst for efficient water oxidation, ACS Appl. Mater. interfaces 7 (2014) 442e451. G. Zhang, B.Y. Xia, X. Wang, Strongly coupled NiCo2O4-rGO hybrid nanosheets as a methanol-tolerant electrocatalyst for the oxygen reduction reaction, Adv. Mater. 26 (2014) 2408e2412. N. Anastasijevi c, V. Vesovi c, R. Ad zi c, Determination of the kinetic parameters of the oxygen reduction reaction using the rotating ring-disk electrode: Part I. Theory, J. Electroanal. Chem. interfacial Electrochem. 229 (1987) 305e316. X. Wang, Q. Li, H. Pan, Y. Lin, Y. Ke, H. Sheng, M.T. Swihart, G. Wu, Sizecontrolled large-diameter and few-walled carbon nanotube catalysts for oxygen reduction, Nanoscale 7 (2015) 20290e20298.

Please cite this article in press as: M. Chen, et al., Journal of Power Sources (2017), http://dx.doi.org/10.1016/j.jpowsour.2017.08.062