A metal/semiconductor contact induced Mott–Schottky junction for enhancing the electrocatalytic activity of water-splitting catalysts

Abstract

Hydrogen, with clean and high gravimetric energy density features, has been considered as an ideal energy carrier. Electrochemical water splitting that can convert the intermittent electricity generated from wind and solar energy into storable hydrogen is a promising technique. The large-scale utilization of water electrolysis is constrained by the sluggish hydrogen evolution reaction (HER) and oxygen evolution reaction (OER). Therefore, cost-effective and highly efficient electrocatalysts are needed to reduce the intrinsic energy barriers and thus enhance the energy conversion efficiency. The Mott–Schottky junction at the metal/semiconductor interface can modulate the electron density of active sites and optimize the chemisorption of water-splitting intermediates via the Mott–Schottky effect, and thus is drawing growing attention. Herein, we summarize the recent advances in Mott–Schottky electrocatalysts for the OER, HER, and overall water-splitting. Specifically, the strategies for preparing Mott–Schottky catalysts and the relationships between the structure/composition and catalytic performance are profoundly discussed. Finally, the challenges and prospects for the future development and application of Mott–Schottky catalysts in water electrolysis are briefly discussed.

---

1. Introduction

Considering the overuse of fossil fuels and the associated environmental concerns, it is of vital importance to develop sustainable energy alternatives to the traditional fossil fuels.^1,2 Energy harvested from wind and sunlight is a clean and renewable energy source and offers an appealing approach to electricity generation. However, wind and solar energy is usually intermittent and dependent on daily and seasonal variation. Therefore, efficiently converting this renewable energy into chemical fuel which can be stored and transported is greatly required.^3,4 Hydrogen, with carbon-free emissions and high gravimetric energy density, has been commonly considered as a promising renewable energy carrier to store the abundant but intermittent sustainable energy.^5,6 In view of negligible environment pollution and accessible water resources, electrochemical water splitting technology that can convert the intermittent electricity generated from wind and solar energy into high-purity hydrogen in a sustainable way is attracting more and more attention.^7,8 Generally, electrochemical water splitting proceeds through two half reactions: oxygen evolution reaction (OER) and hydrogen evolution reaction (HER).⁹ Despite providing a sustainable way to produce hydrogen fuel, the large-scale utilization of water electrolysis is constrained due to the strongly uphill reactions of the HER and OER, resulting in large energy barriers and slowing down the water-splitting kinetics. In practice, excess input voltages (named overpotential, η) must be applied to initiate water electrolysis, typically operating at a high practical voltage of 1.8–2.0 V higher than the theoretical voltage of 1.23 V.^10,11 The η is mainly used to overcome the intrinsic energy barriers present in both OER and HER processes.¹² To reduce the kinetic energy barriers and speed up the mass production of hydrogen, active and robust electrocatalysts that can lower the η are essential. Currently, Pt-based materials are the most active electrocatalysts for the HER, while Ir/Ru-based compounds are the best-performing electrocatalysts for the OER, but their scarcity and high price severely limit their widespread commercialization.¹³ In this regard, the exploration of highly efficient earth-abundant electrocatalysts that can replace expensive noble electrocatalysts is key for the practical application of electrochemical water splitting technology.

Both the OER and HER take place on the surface and/or at the interface of the corresponding electrocatalysts.¹⁴ The electrocatalytic activity of water-splitting catalysts is thus generally determined by two factors: the number of accessible surface active sites and the intrinsic catalytic activity of each site.^15–18 Therefore, in order to improve the efficiency of water splitting, an elaborately designed electrocatalyst should expose abundant surface active sites and exhibit optimized physicochemical properties to enhance its catalytic activity toward the HER and/or OER.^19–23 Specifically, nanostructure engineering is effective for increasing the number of active sties while alloying or hybridizing may endow the electrocatalysts with high intrinsic activity.^24–29 Accordingly, various non-noble-metal-based electrocatalysts have been reported to replace precious-metal-based materials, for instance, transition-metal-based carbides,^30–32 phosphides,^33–37 nitrides^38–40 and chalcogenides^41–43 for the HER; and transition metal oxides,^44–47 hydr(oxy)oxides,^48–51 phosphides^52–54 and nitrides^55–57 for the OER. For example, the Yang group reported an interior-modified Pt–Co₂P as an alkaline HER catalyst.⁵⁸ The strong dopant–host interaction between Pt (a trace amount) and Co₂P was beneficial for water dissociation and hydrogen adsorption, thus boosting the alkaline HER activity of Pt–Co₂P. The Wang group demonstrated that Co³⁺ in the center of the octahedron (Co³⁺(Oh)) was the optimal configuration of cobalt cations for the OER.⁵⁹ Although some of the reported electrocatalysts show excellent activities closer to those of precious-metal-based catalysts in terms of η to reach a geometric area normalized current density of 10 mA cm⁻², the intrinsic activity of these earth-abundant HER/OER electrocatalysts is still lower than those of noble-metal-based materials.

As is well known, a Mott–Schottky junction (or a Schottky junction), formed at the metal/semiconductor interface, such as dispersing a transition metal on the surface of a semi-conductive substance (carbonaceous materials, transition metal based semiconductors and so on), can accelerate interfacial charge transfer, redistribute interfacial charges and modulate the chemisorption of reaction intermediates via the Mott–Schottky effect, thus enhancing the intrinsic activity of electrocatalysts (Scheme 1). Very recently, Mott–Schottky junctions have been explored to boost the catalytic activity of catalysts (Mott–Schottky catalysts) for hydrogenation, aerobic oxidation, artificial photosynthesis, water splitting and so on.^60–64 For example, the Li group reported a Cu/PI Mott–Schottky catalyst by coupling Cu nanoparticles to semi-conductive polyimide (PI) for electrochemical reduction of nitrogen to ammonia under ambient aqueous conditions.⁶⁵ The authors claimed that the Mott–Schottky effect could tune the electron densities of Cu nanoparticles and boost the pre-adsorption of N₂ molecules, thus enhancing the activity of Cu nanocatalysts toward the N₂ reduction reaction. Su and co-workers constructed a Mott–Schottky catalyst by encapsulating Co nanoparticles inside nitrogen-rich carbon shells (Co@NC).⁶¹ The Mott–Schottky effect at the Co/NC interface modifies the electron density of Cu nanoparticles effectively, promoting the catalytic activity of Cu nanoparticles for base-free oxidation of alcohols. Recently, Biswas et al. assembled semiconductive SnS₂ with metallic nanoporous gold (NPG) to prepare a SnS₂/NPG Mott–Schottky electrocatalyst for nitrogen fixation.⁶⁴ Experimental and theoretical investigations demonstrated that the d-band center of Sn atoms near the Mott–Schottky junction was lowered, leading to a facile adsorption of N₂ on the surface active sites of Sn. Therefore, the SnS₂/NPG Mott–Schottky electrocatalyst showed a high faradaic efficiency for the nitrogen reduction reaction.

Scheme 1 Schematic illustration of the Mott–Schottky junction in electrocatalysts.

For the water-splitting process, the unique physicochemical properties of Mott–Schottky junctions could not only provide abundant active sites, but also promote both the adsorption and desorption processes of the OER and/or HER, thereby forming highly efficient Mott–Schottky electrocatalysts for water splitting. In this review, we will focus on the rational design of Mott–Schottky electrocatalysts and recent progress of the Mott–Schottky effect in water electrolysis. We specially review the strategies for preparing Mott–Schottky catalysts, and the relationships between the structure/composition and catalytic performance. Based on the discussions, several possible solutions and future directions are provided finally. We hope that this review could provide helpful guidance for developing cost-effective and highly efficient Mott–Schottky water-splitting electrocatalysts.

2. Metal/semiconductor contact induced Mott–Schottky junction

2.1 Construction of the Mott–Schottky junction

A Mott–Schottky junction is the interface region formed at the contact interface between a metal and semiconductor.⁶³ For example in metal/n-type semiconductor, after stacking or stitching a metal with a semiconductor, the difference in work functions (ϕ) drives free electron transfer at the interface between the metal and semiconductor, followed by Mott–Schottky junction formation at the metal/semiconductor interface (Fig. 1).⁶⁶ The ϕ is the minimum energy needed to extract an electron from a material at 0 K.⁶⁷ As shown on the left of Fig. 1, when the n-type semiconductor work function (ϕ_s) < the metal work function (ϕ_m), the electrons will transfer from the semiconductor to the metal until the Fermi level of the semiconductor (E_F,s) and metal (E_F,m) is the same, followed by band bending and the formation of a Schottky barrier (ϕ_SB). The ϕ_SB allows the charges to irreversibly transport across the heterointerface, thus resulting in charge accumulation or depletion along the contact interface.⁶⁸ In this case, the metal/semiconductor contact can induce a charged interface, with an enriched electron density on the metal side, while the semiconductor side is positively charged because of the electrostatic induction. As compared with the corresponding bulk, the electrons are depleted near the semiconductor surface, which is called the depletion layer. As shown on the right of Fig. 1, when ϕ_s > ϕ_m, the electrons will transfer from the metal to the semiconductor leading to the accumulation of the electrons near the semiconductor surface and thus forming an accumulation layer. Generally, when E_F,m is below E_F,s, the free charge will flow to the contacted metal side and thus causing the E_F,s to decrease, and vice versa. It can be seen from the above that the Mott–Schottky junction can rectify the surface charge density at metal/semiconductor interfaces, which is described as the Mott–Schottky effect in state physics. Therefore, it can be expected that the adsorption of reaction intermediates and electron transport process will be affected by the charged contact interfaces.

Fig. 1 Energy band diagrams of typical n-type semiconductor and metal contacts. ϕ, work function; χ_s, electron affinity of the semiconductor; E_F, Fermi level; E_vac, vacuum energy; E_c, conduction band; E_v, valence band. Reproduced with permission.⁶⁶ Copyright 2012, American Chemical Society.

2.2 The effects of the Mott–Schottky junction for electrocatalysis

2.2.1 The effect on the interfacial electronic structure. As mentioned above, electrocatalytic reactions including both the OER and HER take place on the surface and/or at the interface of the corresponding electrocatalysts. Therefore, the catalytic ability of electrocatalysts is dependent on the surface electronic structure and atomic arrangements.⁶⁹ In order to improve their catalytic performance, it is of significance to modulate the surface charge distribution of the catalyst via rational structural design. In this regard, the built-in electric field at the Mott–Schottky junction can drive charge transfer along the interface and thus lead to charge redistribution on the surface structure, which is an effective method to accelerate electron transport from the surface of the catalyst to the reaction intermediate.^70,71 The promoted electron transfer and abundant active sites synergistically lead to better electrocatalytic performance. For example, the Sun group designed a typical Co@NC heterostructure as a model Mott–Schottky catalyst for lithium–sulfur batteries by a metal–organic framework-templated method (Fig. 2).⁷² Experimental and theoretical investigations revealed the charge density redistribution and a built-in electric field at the heterointerface of Co@NC, which promoted Li⁺/e⁻ transport and improved the kinetics of polysulfide reduction and Li₂S oxidation in Li–S electrochemistry. With similar considerations in mind, the Cabot group dispersed metallic CoFeP on n-type C₃N₄ tubes (t-CN) to construct a Mott–Schottky catalyst for lithium–sulfur batteries by an electrostatic self-assembly method.⁷³ The authors claimed that the optimized electronic structure and charge redistribution at the Mott–Schottky interface contributed to its superior rate performance. Undoubtedly, the electron redistribution at the metal/semiconductor interface via the Mott–Schottky effect can also facilitate the electrocatalytic reactions of the HER and/or OER. For example, Yang et al. encapsulated a Co/Co₂P Mott–Schottky junction in N,P-doped carbon nanotubes as a Mott–Schottky electrocatalyst for bifunctional oxygen electrocatalysis.⁷⁴ DFT calculations confirmed the facile electron transfer from metallic Co to semiconductive Co₂P, upshifting the d-band center of Co/Co₂P and thus enhancing its catalytic performance toward oxygen electrocatalysis.

Fig. 2 (a) Energy band diagrams of Co and NC before and after the Mott–Schottky type contact. (b and c) Differential charge density redistributions at the interface between Co and NC. Yellow represents charge accumulation, while blue represents charge depletion. Reproduced with permission.⁷² Copyright 2021, American Chemical Society.

2.2.2 The effect on chemical free energy. The electrocatalytic reaction including both the HER and OER can be concluded to proceed by the initial adsorption of the reactant, followed by the formation of reaction intermediates and the final desorption step.^12,75 Based on the Sabatier principle, the binding between the reaction species and the active sites should be neither too strong nor too weak.^76,77 Therefore, to achieve higher activity, the chemisorption and desorption free energy of the electrocatalysts should be optimized. Recently, the construction of a Mott–Schottky junction has been explored to tune the chemisorption and desorption behaviors of the electrocatalysts to enhance their inherent activities toward electrocatalytic reactions.^78–80 For example, Yuan et al. constructed Mott–Schottky Bi–BiVO₄ heterostructures for electrochemical urea synthesis by a NaBH₄ reduction method.⁸¹ DFT calculations revealed that the targeted adsorption of N₂ and CO₂ were facilitated through the formed space-charge region at the Bi–BiVO₄ interface (Fig. 3a and b). In addition, the space-charge region could also activate N₂ and CO₂ molecules and optimize the Gibbs free energy of the corresponding reaction intermediates (Fig. 3c). Benefitting from the Mott–Schottky effect, the Bi–BiVO₄ electrocatalyst exhibited an improved urea yield rate and faradaic efficiency. Sun et al. designed Mott–Schottky Mo–MoS₂ heterojunctions as an efficient electrocatalyst for the HER under all pH conditions by an in situ lithiation method.⁸² The authors claimed that the heterointerfaces of Mo–MoS₂ not only resulted in the electron redistribution of sulfur sites, but also facilitated the proton and H₂O adsorption, leading to an enhanced HER activity over the whole pH range.

Fig. 3 DFT-calculated adsorption energies for (a) N₂ and (b) CO₂ on pristine BiVO₄ and Bi–BiVO₄. (c) Energetic pathway of electrolytic urea production. Reproduced with permission.⁸¹ Copyright 2018, Wiley-VCH.

To sum up, the rational combination of a semiconductor and metal to build a delicate Mott–Schottky junction in electrocatalysts can significantly improve the catalytic reactivity of the electrocatalysts. The constructed Mott–Schottky junction could not only adjust the charge densities of atoms near the interface, but also optimize the reaction kinetics, thus showing enhanced activity toward specific catalytic reactions.^65,83 With these considerations, rational integration of a semiconductor and metal to construct a delicate Mott–Schottky catalyst is a proof-of concept strategy for designing efficient earth-abundant HER/OER electrocatalysts, because the charged surfaces in Mott–Schottky catalysts have important influence on the charge transfer process and absorption of ions involved in the HER and OER. Research on Mott–Schottky catalysts toward water electrolysis is a rapidly emerging research focus and recent research efforts have triggered excellent syntheses and electrochemical application of Mott–Schottky catalysts in the water splitting reaction (Table 1). In this review, we discuss recent significant advances in developing Mott–Schottky electrocatalysts in the electrochemical water splitting application, along with the underlying mechanisms. Based on the discussions, some challenges and future directions about Mott–Schottky electrocatalysts are provided.

Table 1 Mott–Schottky electrocatalysts for the HER and OER, respectively

Mott–Schottky catalyst	Application	Electrolyte	η 10 (mV)	Tafel slope (mV dec^−1)	Ref.
MoB/g-C3N4	HER	1 M KOH	133	46	63
MoC/NPC@CNTs	HER	0.5 M H2SO4	175	62	91
FeNi/DCF/NC	HER	1 M KOH	169	98	92
Co/TiO2	HER	1 M KOH	229	88	93
Co@CoMoO4	HER	1 M KOH	46	85	94
Co/a-WOx	HER	1 M KOH	36.3	53.9	95
Co/Co2P@NPCNTs	HER	1 M KOH	310	71.5	74
NiFe LDH/NiS	OER	1 M KOH	230	110.7	100
Co/Co3O4@N-CNTs	OER	1 M KOH	275	99.2	101
Co2P2O7@N,P–C	OER	1 M KOH	270	49.1	104
Co/CoSe@NC	OER	1 M KOH	335	77.6	106
Ni/Ni0.2Mo0.8N@N–C	OER	1 M KOH	260	46	108

---

3. Properties of Mott–Schottky catalysts for water electrolysis

3.1 Hydrogen evolution reaction

3.1.1 Brief introduction of the HER mechanism. The HER is a two-electron transfer reaction and may proceed through the Volmer–Heyrovsky or the Volmer–Tafel pathway, which is roughly discerned from the Tafel slope:^84,85

Volmer step (Tafel slope = 120 mV dec⁻¹):

H⁺ + * + e⁻ → H* (in acidic medium)

H₂O + * + e⁻ → H* + OH⁻ (in basic medium)

Heyrovsky step (Tafel slope = 40 mV dec⁻¹):

H* + H⁺ + e⁻ → H₂ + * (in acidic medium)

H* + H₂O + e⁻ → H₂ + OH⁻ + * (in basic medium)

Tafel step (Tafel slope = 30 mV dec⁻¹):

2H* → H₂ + 2*

where * denotes an active site and H* is the catalytic intermediate of the adsorbed hydrogen atom. The reaction rate of the HER is largely determined by the Gibbs free energy of hydrogen adsorption (ΔG_H). When ΔG_H is more positive, in other words, the electrocatalyst surface adsorbs hydrogen too weakly, and the Volmer step (the adsorption step) will be the rate determining step. However, if the adsorption is too strong, the Heyrovsky/Tafel step (the desorption step) will be the rate determining step. Therefore, the highly active electrocatalysts towards the HER should adsorb hydrogen neither too weakly nor too strongly. As shown in the Volcano plot (Fig. 4), noble-metal-based electrocatalysts with a near-zero ΔG_H may occupy the top of the volcano, that is to say, they will display the highest catalytic activity for the HER.⁸⁶ It is worth noting that most HER electrocatalysts exhibit much lower activity in basic solutions than that in acidic medium.^87,88 Strmcnik et al. attributed the poor catalytic performance of alkaline HER catalysts to the sluggish water dissociation process in alkaline medium.⁸⁹ To evaluate the performance of a HER electrocatalyst, the corresponding parameters should be accurately measured or calculated, for instance, overpotential, Tafel plot, turnover frequency (TOF), faradaic efficiency and stability. Significantly, the catalytic activity of a HER electrocatalyst is related to the electronic structure and adsorption energy of the surface active sites.

Fig. 4 Volcano plots of various metals towards the HER. Reproduced with permission.⁸⁶ Copyright 2016, Springer Nature.

As mentioned above, the optimal HER catalyst always has a ΔG_H close to zero. In this regard, constructing a Mott–Schottky junction may help to modify the surface electronic structures and optimize the adsorption energy of the electrocatalyst, facilitating the HER kinetics. In addition, by constructing a Mott–Schottky junction, the interfacial electron transfer can also be accelerated, further enhancing the HER activity of the electrocatalyst. The Mott–Schottky effect has been demonstrated as a very promising way to boost the activity of electrocatalysts for specific catalytic reactions and extended to rationally designing Mott–Schottky electrocatalysts for the HER.

3.1.2 Mott–Schottky catalysts for the HER. 2D carbonaceous materials, such as g-C₃N₄ and heteroatom doped carbon materials, could serve as the semiconductor side for the construction of Mott–Schottky catalysts. Zhuang and coworkers reported a MoB/g-C₃N₄ Mott–Schottky catalyst by simply blending tetragonal α-MoB with n-type g-C₃N₄ (Fig. 5a–e).⁶³ A temperature-dependent resistivity test and local density of states (DOS) analysis showed that the tetragonal α-MoB exhibited metallic characteristics. UV/vis analysis suggested that n-type g-C₃N₄ possessed a wide band gap of 2.7 eV, making it suitable for Mott–Schottky contact formation. As shown in Fig. 5b, the MoB/g-C₃N₄ Mott–Schottky catalyst exhibited better HER catalytic activity than MoB in 1 M KOH, indicating the role of the Mott–Schottky contact in boosting the HER activity of MoB/g-C₃N₄. X-ray photoelectron spectroscopy (XPS) and energy-loss near-edge structure (ELNES) measurements are conducted to unravel the electronic structures of MoB/g-C₃N₄ (Fig. 5c and d). These results confirmed that the local electron density of Mo atoms in MoB/g-C₃N₄ increased (donated by g-C₃N₄), indicating that the Mott–Schottky effect could create a fraction of Mo atoms in MoB/g-C₃N₄ in a reduced state. It should be noted that photoluminescence (PL) together with ultraviolet photoemission spectroscopy (UPS) analyses validated the existing Mott–Schottky effect. Benefiting from the increased local electron density of MoB and efficient charge transfer across the MoB/g-C₃N₄ Mott–Schottky junction, the kinetic reaction barriers of the HER process on the Mott–Schottky catalyst were dramatically reduced (Fig. 5e). Moreover, MoB/g-C₃N₄ demonstrated stability for > 48 h at 10 mA cm⁻² and exhibited negligible attenuation after 5000 cycles, indicating its excellent stability for the HER. In addition, Xia et al. developed a MoN/g-C₃N₄ Mott–Schottky catalyst by a self-assembly and annealing method, which showed an enhanced catalytic activity towards photocatalytic H₂ production.⁹⁰

Fig. 5 (a) Energy band diagrams of the Schottky junction between MoB and g-C₃N₄. (b) Polarization curves of MoB and MoB/g-C₃N₄ in 1 M KOH. (c) XPS spectra of Mo 3d_5/2 and (d) ELNES of Mo-M_2,3 edges for MoB in comparison with MoB/g-C₃N₄. (e) Free energy diagram of the HER on MoB and MoB/g-C₃N₄. Reproduced with permission.⁶³ Copyright 2018, Wiley-VCH. (f) The preparation procedure for the synthesis of MoC/NPC@CNTs. (g) LSV curves of MoC/NPC@CNTs with different pyrrole amounts in 0.5 M H₂SO₄. (h) Mott–Schottky plots for MoC/NPC800 and MoC/NPC@CNTs800. (i) XPS spectra of Mo 3d for MoC/NPC800 and MoC/NPC@CNTs800. Reproduced with permission.⁹¹ Copyright 2020, Royal Society of Chemistry.

Wang et al. prepared a MoC based Mott–Schottky catalyst by carbonization of polyoxometalates and polypyrrole nanocomposites (Fig. 5f).⁹¹ The MoC nanoparticles were uniformly embedded in N,P co-doped carbon and further covered on the carbon nanotubes (CNTs), constructing a MoC/NPC@CNT Mott–Schottky electrocatalyst. The optimal MoC/NPC@CNTs showed excellent HER performance with a small overpotential (η) of 175 mV to reach a current density of 10 mA cm⁻² in 0.5 M H₂SO₄ (Fig. 5g). Based on the small positive slope of Mott–Schottky plots in Fig. 5h and Hall results, the authors claimed that the Mott–Schottky effect existed between N,P co-doped carbon and metallic MoC. Notably, the introduced CNTs could further increase the electron density around E_f, due to its impact on facilitating the electron transfer from carbon to MoC. As shown in Fig. 5i, the XPS results indicated that the Mo³⁺/Mo²⁺ ratio in MoC/NPC (1.382) or MoC/NPC@CNTs (1.372) was far lower than that of pure MoC (10.9). The smaller Mo³⁺/Mo²⁺ ratio caused by the Mott–Schottky effect could modulate Mo–H bond strength for optimal ΔG_H, accelerating the Volmer reaction step and thus enhancing the HER activity. In addition, the optimized MoC/NPC@CNTs was stable for 12 h at 10 mA cm⁻² and the deviation after 1000 cyclic voltammetric (CV) scans was negligible. Similarly, Wang et al. prepared a Mott–Schottky catalyst based on metallic FeNi/defect-rich CoFe₂O₄ decorated on semiconducting nitrogen-doped carbon,⁹² which boosts the charge transfer ability of transition metal oxides and provides a promising way to enhance the HER activity of oxides in an alkaline environment.

Besides, metal oxides have been explored as the semiconductor side for the construction of Mott–Schottky catalysts for the HER as well. Yu and co-workers succeeded in dispersing TiO₂ nanoparticles on the surface of microscale Co dendrites by a one-pot hydrothermal method (Fig. 6a), constructing a Co/TiO₂ Mott–Schottky catalyst.⁹³ The high-resolution TEM and inverse fast Fourier transform (FFT) images presented the interfacial structure between Co and TiO₂ (Fig. 6b). DOS plots of the d-orbital and total orbital showed that Co/TiO₂ had a negative shift of the d-band center and higher electron density than Co (Fig. 6c and d). The planar-averaged electron density difference on Co/TiO₂ (Fig. 6e), together with XPS analysis indicated that the free electrons would flow from TiO₂ to Co, confirming the formation of a Mott–Schottky junction. DFT calculations showed that the Mott–Schottky junction in Co/TiO₂ could effectively optimize the ΔG_H and thus increase the HER catalytic activity (Fig. 6f). As a result, the as-prepared Co/TiO₂ Mott–Schottky catalyst displayed superior HER catalytic activity with an η of 229 mV to deliver a current density of 10 mA cm⁻² and a TOF of 0.052 s⁻¹ in 1 M KOH (Fig. 6g). Additionally, the Co/TiO₂ Mott–Schottky catalyst displayed excellent stability, as evidenced by the results of accelerated degradation tests (ADTs) and chronopotentiometric measurements. In another study, Xiang et al. successfully prepared a self-standing Co@CoMoO₄ catalyst by calcination of cobalt carbonate hydroxide@CoMoO₄.⁹⁴ The positive shift of the binding energies of Co 2p and Mo 3d indicated the possible construction of a Mott–Schottky junction between metallic Co and n-type CoMoO₄. Benefiting from the synergistic effect, the resulting Co@CoMoO₄ catalyst afforded a current density of 10 mA cm⁻² at an extremely low overpotential of 46 mV in 1 M KOH. Most recently, Chen and co-workers reported a Co/a-WO_x Mott–Schottky electrocatalyst with a hollow porous nanowire structure through a self-template etching method.⁹⁵ The porous tubular structure exposed abundant heterointerfaces, maximizing the Mott–Schottky effect. Experimental and theoretical investigations demonstrated facile electron transfer and redistribution across the heterointerface, which endowed the Co nanoparticles with the properties of accelerating water dissociation and a-WO_x with optimized adsorption energy for hydrogen intermediates. Benefitting from maximizing the Mott–Schottky effect and the unique nanostructure, Co/a-WO_x showed superior alkaline HER activity with a small overpotential of 36.3 mV to acquire 10 mA cm⁻² and a low Tafel slope (53.9 mV dec⁻¹) as well as stability for 200 h in 1 M KOH.

Fig. 6 (a) Illustration of the preparation and HER procedure of the Co/TiO₂ Mott–Schottky catalyst. (b) HRTEM image of Co/TiO₂. Inset shows the inverse FFT in region II. DOS plots of the d-orbital (c) and total orbital contribution to Co and Co/TiO₂ (d). (e) The difference of planar-averaged electron density for the Co/TiO₂ Mott–Schottky catalyst. (f) ΔG_H* of various samples. (g) LSV curves of various samples. Reproduced with permission.⁹³ Copyright 2019, American Chemical Society.

3.2 Oxygen evolution reaction

3.2.1 Brief introduction of the OER mechanism. Compared with the HER, the OER is a complicated 4-electron process along with the adsorption of various oxygen intermediates, leading to more sluggish kinetics. Thus, a large η is needed to initiate the OER, making it the bottleneck for the water splitting reaction. The four-electron reaction pathway in acidic medium has been proposed as follows:^96,97

* + H₂O → H⁺ + HO* + e⁻

HO* → H⁺ + HO* + e⁻

O* + H₂O → H⁺ + HOO* + e⁻

HOO* → H⁺ + O₂ + e⁻

Under basic conditions, the four-step process of the OER is described as follows:

* + OH⁻ → HO* + e⁻

HO* + OH⁻ → H₂O + O* + e⁻

O* + OH⁻ → HOO* + e⁻

OH⁻ + HOO* → * + H₂O + O₂ + e⁻

where * denotes an active site, while HO*, O* and HOO* represent the oxygen-containing absorption intermediates. Unlike the HER, the binding energies of the three important intermediates are strongly correlated. As proposed by previously reported articles,^97,98 a scaling relationship between the binding energies of adsorbed HOO* and HO* intermediates (ΔG_OOH = ΔG_OH + 3.2 ± 0.2 eV) is screened theoretically (Fig. 7a). Take the overall energy barrier of 4.92 eV into consideration, this scaling relationship indicates that the rate-determining step is probably the formation of O* (step 2) or HOO* (step 3) with a minimized Gibbs free energy of ≈1.6 eV. An optimal OER catalyst should bind the key oxygen adsorbate neither too strongly nor too weakly on the basis of the Sabatier principle.⁷⁶ Therefore, the difference (ΔG_O* − ΔG_HO*) is commonly regarded as an effective descriptor to evaluate the catalytic activity of catalysts and is translated to the OER volcano plot (Fig. 7b). As shown in the OER volcano plot, noble-metal-based electrocatalysts (RuO₂) with a moderate bonding energy of oxygen intermediates will show the highest catalytic activity toward the OER. Similar to the HER, the Mott–Schottky effect is also a very promising strategy for optimizing the binding energies of oxygen intermediates to facilitate the kinetics of the OER.

Fig. 7 (a) The scaling relationship between the binding energies of adsorbed HOO* and HO* intermediates on various metal oxides. (b) OER volcano plot for metal oxides against G_O* − G_HO*. Reproduced with permission.⁹⁷ Copyright 2011, Wiley-VCH.

3.2.2 Mott–Schottky catalysts for the OER. Metal (oxy)hydroxides, oxides and LDHs have been actively investigated as the semiconductor side for the synthesis of Mott–Schottky catalysts for the OER.^99–103 Wen et al. prepared a NiFe LDH/NiS Mott–Schottky catalyst by electrodepositing NiFe LDH onto NiS nanosheets (Fig. 8a–f).¹⁰⁰ The total densities of states (TDOS) proved that NiS was metallic, while NiFe LDH exhibited semiconductor characteristics with a band gap of 1.51 eV (Fig. 8c). The NiFe LDH/NiS contact induced an active interface region (Mott–Schottky junction), where the electrons will transfer from metallic NiS to semiconductive NiFe LDH as revealed by DFT calculations (Fig. 8d and e) and XPS. Furthermore, the d-band center of Ni (Fe) atoms near the Mott–Schottky junction shifted to a high energy level, resulting in the optimal binding energy of oxygen-containing intermediates. Benefitting from the interfacial charge redistribution of the Mott–Schottky junction and porous hierarchical structure, the resultant NiFe LDH/NiS Mott–Schottky catalyst exhibited impressive OER performance with a low η₁₀ and η₁₀₀₀ of 230 and 325 mV (Fig. 8f), respectively. Additionally, NiFe LDH/NiS possessed an excellent stability with negligible potential decay for over 100 h of a continuous test at 400 mA cm⁻² in 1 M KOH. Similarly, the Li group reported a NiFe(OH)_x/Ni₃S₂ Mott–Schottky junction by in situ assembly of a NiFe(OH)_x electrocatalyst onto the as-prepared Ni₃S₂ layer supported on a Ni mesh,⁹⁹ and revealed the charge redistribution at the interface between NiFe(OH)_x and Ni₃S₂. Thus, the resultant NiFe(OH)_x/Ni₃S₂/Ni electrode showed extraordinary OER performance in 1 M KOH.

Fig. 8 (a) Illustration of the preparation of NiFe LDH/NiS. (b) HRTEM image and corresponding FFT patterns of NiFe LDH/NiS. (c) The calculated total density of states (TDOS) for various samples. (d) Charge density difference and (e) corresponding 2D data plot for the NiFe LDH/NiS model. (f) LSV curves of different electrodes. Reproduced with permission.¹⁰⁰ Copyright 2021, Wiley-VCH. (g) HRTEM image of Co/Co₃O₄@N-CNTs. (h) Energy band diagrams of Co₃O₄ and Co. (i) Illustration of the interfacial charge transfer between Co₃O₄ and Co. Reproduced with permission.¹⁰¹ Copyright 2022, Elsevier B. V.

Co/Co₃O₄ Mott–Schottky heteronanoparticles embedded in N-doped carbon nanotubes (Co/Co₃O₄@N-CNTs) were successfully synthesized by a hydrogel-bridged pyrolysis strategy (Fig. 8g–i).¹⁰¹ The high resolution transmission electron microscopy (HRTEM) image of Co/Co₃O₄ heteronanoparticles exhibited the interface between Co₃O₄ and Co (Fig. 8g), verifying the construction of Mott–Schottky interfaces. According to energy band diagrams (Fig. 8h), the work function of metallic Co (5.0 eV) was lower than that of semiconductive Co₃O₄ (6.5 eV). Thus, a built-in electric field was generated in the Co/Co₃O₄ heteronanoparticles, which would drive the electron transfer from the metallic Co side to the semiconductive Co₃O₄ side until reaching equilibrium Fermi levels (Fig. 6i). DFT calculation results together with XPS analysis further demonstrated the above charge redistribution across the interface between Co and Co₃O₄. Such charge redistribution can energetically help the adsorption of oxygen intermediates and reduce the overpotentials for the OER and ORR. Regarding the OER, the as-prepared Co/Co₃O₄@N-CNTs showed a lower overpotential and smaller Tafel plot than other contrast samples in 1 M KOH. In addition, the N-doped carbon nanotube (N-CNT) shell could expose more active components, enhance electron transfer efficiency and protect the anchored active sites, further boosting the activity and stability of Co/Co₃O₄@N-CNTs. With similar considerations in mind, Co/CoO_x Mott–Schottky junctions were embedded in BS-NCNTs to prepare a bifunctional Co/CoO_x@BS-NCNT electrocatalyst by a template-free self-assembly method,¹⁰³ which exhibited enhanced activities towards the OER and ORR.

Other transition metal based semiconductors, such as transition metal nitrides, phosphides, chalcogenides and pyrophosphate, were also explored as a matrix to construct a Mott–Schottky junction as well.^104–108 Liang and co-workers prepared Co₂P₂O₇@N,P–C nanocages by a ligand exchange reaction and subsequent pyrolysis process (Fig. 9a–e).¹⁰⁴ The HRTEM image showed the contact interface between Co₂P₂O₇ and N, P co-doped carbon (N,P–C) (Fig. 9a). A temperature-dependent resistivity test demonstrated the metallic characteristics of N,P–C, as also confirmed by the XPS valence band spectrum. Thus, a Mott–Schottky junction is constructed between semiconductive Co₂P₂O₇ and metallic N,P–C based on the theory of semiconductor physics, in which a built-in electric field was formed. On the basis of UPS results, the work function of metallic N,P–C (5.2 eV) was higher than that of semiconductive Co₂P₂O₇ (3.8 eV) (Fig. 9b). Therefore, the valence electrons will transfer from Co₂P₂O₇ NPs to the N,P–C layer until the Fermi levels reach equilibrium driven by the built-in electric field (Fig. 9c), which was also proved by XPS and FT-IR analysis. In this case, the valence electron density of Co can be remarkably reduced, thereby optimizing its occupancy of the e_g orbital and then contributing to the adsorption energies of oxygen intermediates (Fig. 9d). On account of the enhanced intrinsic catalytic activity caused by the interfacial charge polarization of the Mott–Schottky junction, Co₂P₂O₇@N,P–C exhibited an extraordinary OER performance with a very high turnover frequency value and low overpotential in basic medium. In addition, the resultant Co₂P₂O₇@N,P–C electrocatalyst showed outstanding catalytic stability towards the OER with no obvious overpotential increase after a 100 h chronopotentiometry (CP) test, owing to the protection of the robust N,P–C layer.

Fig. 9 (a) HRTEM image of the as-prepared Co₂P₂O₇@N,P–C nanocage catalyst. (b) Work function values of various samples (inset: corresponding UPS spectra). (c) Energy band diagrams of Co₂P₂O₇ and N,P–C before and after the construction of the Mott–Schottky heterojunction. (d) Free energy diagram of Co₂P₂O₇@N,P–C for the OER. (e) The OER LSV curves of different samples. Reproduced with permission.¹⁰⁴ Copyright 2019, Elsevier B. V. (f) HRTEM image with FFT patterns (inset) of Co/CoSe@NC. (g) Schematic energy band diagrams of Co and CoSe before and after forming a Schottky contact. (h) Free energy diagrams of CoSe, Co and Co/CoSe for the OER. Reproduced with permission.¹⁰⁶ Copyright 2022, Elsevier B. V.

Li and co-workers constructed a Co/CoSe Mott–Schottky junction encapsulated in N-doped carbon by an in situ partial reduction strategy.¹⁰⁶ The in situ partial reduction of CoSe to Co ensured a close contact interface between metallic Co and semiconductive CoSe, which was directly visualized by the HRTEM image and corresponding FFT patterns (Fig. 9f). According to Mott–Schottky and Tauc plots, the energy band diagrams of the Co/CoSe Mott–Schottky junction were proposed (Fig. 9g). The lower Fermi levels of semiconductive CoSe than that of metallic Co resulted in electron transfer from Co to CoSe, which was verified by XPS analysis and DFT calculations. Consequently, a Mott–Schottky heterojunction is constructed and interfacial charge polarization is created, thus leading to a hole abundant region on the metallic side. DFT calculations revealed that the hole-rich Co center in the Co/CoSe Mott–Schottky junction possessed strong oxidation capacity, which is conducive to the OER (Fig. 9h). Because of the charge redistribution of the Mott–Schottky junction and the mesoporous carbon encapsulation, the as-prepared Co/CoSe@NC showed superior OER and ORR performance in alkaline media. Focusing on this topic, the Huang group designed a Co₉S₈/Co Mott–Schottky junction supported on defective reduced graphene oxide (Co₉S₈/Co-rGO) as a bifunctional oxygen catalyst (OER and ORR).¹⁰⁵ The constructed Co₉S₈/Co Mott–Schottky junction is conducive to the adsorption of oxygen intermediates towards catalyzing OER and ORR processes. Benefiting from the charge density redistribution across the Mott–Schottky junction and optimized oxygen/ion diffusion pathways, the as-formed Mott–Schottky Co₉S₈/Co–rGO catalyst exhibits enhanced catalytic activity towards the OER and ORR with a facile reaction mechanism.

Li et al. developed a hard-template strategy to successfully synthesize Mott–Schottky Ni/Ni_0.2Mo_0.8N nanosheets combined with N-doped carbon (Ni/Ni_0.2Mo_0.8N@N–C).¹⁰⁸ The authors claim that the Mott–Schottky Ni/Ni_0.2Mo_0.8N nanosheets provide a highway for charge transfer and reaction-relevant species, which will contribute to the generation of O_ads intermediates and thus promote the subsequent OER step. Moreover, the carbon frameworks not only optimize the electron transportation, but also prevent the agglomeration of Ni/Ni_0.2Mo_0.8N nanosheets. Besides, the N dopants modulate the electronic structures of carbon that favor the OER process. As a result, the as-obtained Ni/Ni_0.2Mo_0.8N@N–C exhibits excellent OER catalytic activity and stability, with a low η₁₀ of 260 mV and a negligible attenuation after 2000 CV cycles in 1 M KOH. The excellent structural stability of Ni/Ni_0.2Mo_0.8N@N–C was corroborated by SEM and XPS characterization. Similarly, Chen and co-workers constructed Co/Co_xM_y (M = P, N) nanosheets as a Mott–Schottky catalyst for the OER through a salt-templated strategy.¹⁰⁷ Benefiting from the Mott–Schottky components and hierarchically porous structure, the Co/Co_xM_y Mott–Schottky catalyst shows improved OER activity in alkaline medium.

3.3 Overall water splitting

3.3.1 Brief introduction of overall water splitting. Up to now, various earth-abundant electrocatalysts have been reported for the OER in alkaline media and the HER in acidic electrolytes. However, in order to work together for water electrolysis, catalysts capable of catalyzing both the OER and HER must work in the same electrolyte. In addition, water electrolysis is always performed in either a strong alkali or an acid so as to simplify the water splitting system design and reduce the overpotentials.^109,110 However, an earth-abundant electrocatalyst with high activity in alkali aqueous solution will probably be unstable or inactive in an acidic electrolyte. Therefore, it is of great significance to exploit highly efficient and stable bifunctional electrocatalysts for overall water splitting. Integrating the virtues of the OER and HER catalysts is an effective way to design highly efficient HER–OER heterostructures,^111–114 which can possess optimized adsorption energy for both oxygen and hydrogen intermediates, thus enhancing the catalytic activity of overall water splitting. This stimulates the investigation of Mott–Schottky junction electrocatalysts capable of catalyzing both the OER and HER (Table 2). Here, we summarize the recent advances in Mott–Schottky junction based bifunctional water splitting electrocatalysts with an emphasis on their rational design, structural characterization, and catalytic activity.

Table 2 Mott–Schottky electrocatalysts for overall water splitting

Mott–Schottky catalyst	Electrolyte	Current density	Overall voltage	Ref.
Co/CoP-5/Ni foam	1 M KOH	10 mA cm^−2	1.45 V	115
NC/CuCo/CuCoOx	1 M KOH	10 mA cm^−2	1.53 V	116
RuO2/Co3O4–RuCo@NC	0.5 M H2SO4	10 mA cm^−2	1.66 V	117
Ni–Mo2C/NC@NF	1 M KOH	10 mA cm^−2	1.59 V	118
Co–Mo2C-CNx	1 M KOH	10 mA cm^−2	1.68 V	119
WN-Ni@N,P-CNT	1 M KOH	10 mA cm^−2	1.57 V	121
Ni/CeO2@N–CNFS	1 M KOH	10 mA cm^−2	1.56 V	122

---

3.3.2 Mott–Schottky catalysts for overall water splitting. The integration of a transition-metal-based metal and semiconductor to construct a Mott–Schottky electrocatalyst is a proof-of-concept strategy to simultaneously catalyze both the OER and HER by means of the Mott–Schottky effect. Considering the highly catalytic activity of Co and CoP catalysts for the HER in both an alkali and an acid, the Li group developed Mott–Schottky Co/CoP nanoparticles as catalysts toward both the OER and HER in a wide pH range by a vacuum-diffusion method (Fig. 10a–e).¹¹⁵ By combining energy-dispersive X-ray spectroscopy line scan with HRTEM results (Fig. 10b and c), the formation of Co/CoP-5 nanoparticles was efficiently identified. The work function of semiconductive CoP was lower than that of Co, and thus electrons would transfer from semiconductive CoP to metallic Co due to the Mott–Schottky effect (Fig. 10e), which was directly verified by XPS analysis (Fig. 10d). The authors claimed that the electron redistribution at the interface would lower the work function of the Co “facet” and pull down the valence band of the CoP “facet”, resulting in the improved HER and OER activity of the Co/CoP Mott–Schottky electrocatalyst in a wide pH range, respectively. In addition, Co/CoP-5 showed a minimal potential decay after a 12 h stability test (a 10 mV increase of overpotential at 20 mA cm⁻² in 1 M KOH).

Fig. 10 (a) Synthetic process of carbon coated Co/CoP-5. (b) Compositional line profile of Janus Co/CoP-5. (c) HRTEM image of Janus Co/CoP-5. (d) Co 2p XPS spectra of various samples. (e) Energy band diagrams of Co and CoP before and after forming a Mott–Schottky contact. Reproduced with permission.¹¹⁵ Copyright 2017, Wiley-VCH. (f) TEM image of NC/CuCo/CuCoO_x nanowires. (g) Continuous charge transfer of the Mott–Schottky catalysts. (h) HER polarization curves of different samples. (i) OER polarization curves of various samples. Reproduced with permission.¹¹⁶ Copyright 2017, Wiley-VCH.

Afterwards, Hou and co-workers reported a NC/CuCo/CuCoO_x Mott–Schottky electrocatalyst by implanting nitrogen-doped carbon (NC) on a metal (CuCo)–semiconductor (CuCoO_x) nanowire array (Fig. 10f–i).¹¹⁶ In NC/CuCo/CuCoO_x, CuCo nanocrystals were dispersed on CuCoO_x nanowires, which were encapsulated by ultrathin NC shells with a thickness of ≈5 nm. The ultrathin NC shell could serve as a highway for continuous charge transfer between CuCo and CuCoO_x (Fig. 10g). NC/CuCo/CuCoO_x presented a small overpotential of 112 and 198 mV at a current density of 10 mA cm⁻² for the HER and OER in 1 M KOH (Fig. 10h and i), respectively. The as-assembled NC/CuCo/CuCoO_x‖NC/CuCo/CuCoO_x cell needed a voltage of just 1.53 V to reach a current density of 10 mA m⁻² for overall water splitting and the high activity was still maintained after 100 h of electrolysis. The authors proposed that the 3D hierarchical architectures, Mott–Schottky effect and nitrogen-doped carbon were responsible for the enhanced activities of NC/CuCo/CuCoO for the HER and OER.

To further explore the role of the Mott–Schottky effect in enhancing the activities of overall water splitting, various heteroatom-doped carbon confined Mott–Schottky electrocatalysts have been reported recently.^117–121 The Shao group developed a Ru-modified cobalt-based Mott–Schottky catalyst embedded in an N-doped carbon support (RuO₂/Co₃O₄–RuCo@NC) for overall water splitting in acidic media (Fig. 11a–g).¹¹⁷ The enlarged HRTEM image and corresponding FFT patterns revealed the existence of abundant interfaces (Fig. 11a–e). The Mott–Schottky interface would result in the redistribution of electrons at the interface between metallic RuCo and semiconductive RuO₂/Co₃O₄ (Fig. 11f). Such an electron redistribution at the interface positively shifted the flat-band potential of RuO₂/Co₃O₄–RuCo@NC to 2.33 V (Fig. 11g), indicating its more thermodynamically favorable OER process. Benefitting from the unique Mott–Schottky interfaces and favorable electron redistribution, the as-fabricated RuO₂/Co₃O₄–RuCo@NC Mott–Schottky catalysts displayed excellent activities for both the OER and HER with a low overpotential and long-term stability in 0.5 M H₂SO₄. The RuO₂/Co₃O₄–RuCo@NC electrolyzer required a cell voltage of 1.66 V to achieve 10 mA cm⁻² for overall water splitting and exhibited long-term stability with ∼5% attenuation after an 8 h stability test at 10 mA cm⁻² in 0.5 M H₂SO₄. Very recently, Xu et al. reported the large-area synthesis of Ni–Mo₂C/NC@NF through an assembly carburization strategy.¹¹⁸ The as-prepared Ni–Mo₂C/NC@NF Mott–Schottky catalyst could be used as a bifunctional electrode for overall water splitting with outstanding activity and stability in 1 M KOH solution, requiring a small HER and OER overpotential of 40 mV and 337 mV to reach a 10 mA cm⁻² current density (Fig. 11h and i). The Ni–Mo₂C/NC@NF cell needed small voltages of 1.59 and 1.80 V to gain 10 and 50 mA cm⁻² for overall water electrolysis in 1 M KOH, respectively. In addition, Ni–Mo₂C/NC@NF showed high stability with negligible current degradation for 150 h of continuous testing at 10 mA cm⁻². DFT calculations validated that the construction of Mott–Schottky interfaces enables beneficial electronic structures for catalyzing both the HER and OER (Fig. 11j and k).

Fig. 11 (a) HRTEM image of RuO₂/Co₃O₄–RuCo@NC. Enlarged HRTEM images in section I (b) and section II (d) and corresponding FFT patterns (c and e). (f) Energy band diagram of Mott–Schottky RuO₂/Co₃O₄–RuCo@NC. (g) Flat-band position of different samples. Reproduced with permission.¹¹⁷ Copyright 2019, American Chemical Society. (h) Polarization curves of various samples for the OER in 1 M KOH solution. (i) Polarization curves of different samples for the HER in 1 M KOH solution. The free energy diagrams of NC, Mo₂C/NC and Ni–Mo₂C/NC for the OER (j) and HER (k). Reproduced with permission.¹¹⁸ Copyright 2021, Elsevier B. V.

With these considerations, Zhang and co-workers fabricated a dual Mott–Schottky electrocatalyst (Co–Mo₂C-CN_x) by the carbonization of Co–Zn MOF to wrap Co–Mo₂C nanoparticles in porous nitrogen-doped carbon.¹¹⁹ The dual Mott–Schottky heterostructure of Co–Mo₂C-CN_x enabled the whole system present a metallic state, which was conducive to electron transfer. Moreover, the dual Mott–Schottky heterostructure will offer rich charge carriers for intermediate adsorption, thus accelerating the water adsorption/dissociation. The Yang group reported that an N,P co-doped CNT confined WN-Ni Mott–Schottky heterostructure (WN–Ni@N,P-CNT) could lead to enhanced HER and OER catalytic activity.¹²¹ In the Mott–Schottky heterogeneous electrocatalyst, the Mott–Schottky junction existing at the WN–Ni interface was beneficial to electron transfer and water dissociation, thereby promoting the overall water splitting. Subsequently, WN-Ni@N,P-CNT showed improved catalytic activity towards the ORR, HER and OER. Regarding the HER and OER, the optimal WN-Ni@N,P-CNT Mott–Schottky electrocatalyst presented a small overpotential of 70 and 268 mV at a current density of 10 mA cm⁻² for the HER and OER in 1 M KOH, respectively. Most recently, Li and co-workers reported a Ni/CeO₂@N–CNFS Mott–Schottky electrocatalyst by dispersing a Ni/CeO₂ junction on N-doped carbon fibers by an electrospinning reaction and subsequent carbonization process.¹²² Experiments and DFT calculations revealed that the Mott–Schottky junction of Ni/CeO₂ triggered charge redistribution across the constructed heterointerface, resulting in fast charge transfer and optimized adsorption energy for the corresponding reaction intermediates.

4. Conclusions and outlook

The rational combination of a semiconductor and metal to construct a delicate Mott–Schottky electrocatalyst is a proof-of-concept strategy to design high-performance HER and/or OER electrocatalysts. In this review, we present the recent advances in the development of Mott–Schottky water-splitting electrocatalysts with the emphasis on the preparation strategies and catalytic performance, aiming to understand the relationships between the structure/composition and catalytic performance. We introduced theoretical basics about Mott–Schottky junctions and the fundamental understanding of the reaction mechanisms for the HER and OER. We also shed light on the working principle of Mott–Schottky electrocatalysts, which can be attributed to abundant active sites, higher electron transfer efficiency, and the modified local electronic structure at the interface between the metal and semiconductor. Thankfully, the Mott–Schottky effect at the metal–semiconductor interface is capable of optimizing the chemisorption of water-splitting intermediates, which finally enhance the catalytic activity of Mott–Schottky electrocatalysts for the HER and/or OER. Although these pioneering studies have demonstrated the enhancement in water-splitting performance by the delicate construction of Mott–Schottky electrocatalysts, the exploration of Mott–Schottky electrocatalysts is still at the preliminary stage and some key challenges but promising aspects should be paid more attention for future investigations.

(1) Although various Mott–Schottky electrocatalysts have been designed for water splitting, Mott–Schottky electrocatalysts with higher activity and stability are needed to meet the demand for practical applications. Exploring novel and efficient Mott–Schottky junctions with an optimal work function difference between the semiconductor and metal to optimize the Schottky barrier and thus provide more possibilities to regulate the charge densities is particularly promising.^83,123 In this regard, DFT calculation together with experimental study is a very effective means to design and develop advanced Mott–Schottky electrocatalysts and may be one of the future research directions. It can be expected that significant breakthroughs in constructing Mott–Schottky electrocatalysts with an optimal work function difference will promote the development Mott–Schottky electrocatalysts for practical applications.

(2) To identify the fine structures of the heterointerface between a metal and semiconductor, some advanced characterization technologies are needed. The true information about the relatively complex heterostructure during the catalytic reaction is hard to accurately characterize only by ex situ characterization. In order to clarify the evolution of the heterostructure and the information of the local electronic structure under catalytic reaction conditions, some advanced in situ characterization techniques such as in situ TEM,¹²⁴operando XPS¹²⁵ and in situ XAS¹²⁶ are in urgent need to identify the fine interface structure of Mott–Schottky electrocatalysts.

(3) The catalytic mechanism of Mott–Schottky electrocatalysts for water splitting remain difficult to accurately demonstrate. Therefore, in-depth understanding of the mechanism of the complex HER and/or OER with the aid of DFT calculations should be another research direction in the future. The decisive descriptors of the Mott–Schottky effect and catalytic activity from DFT calculations are urgently needed.¹²⁷ In addition, theoretical calculations should be further developed to obtain more precise results.^128,129 In brief, DFT calculations combining electrochemical measurements and in situ characterization are an effective approach to study the reaction mechanisms. Definitely, a deep understanding of the mechanism will help to enhance the catalytic activity and guide the design of Mott–Schottky electrocatalysts.

(4) Although most Mott–Schottky electrocatalysts show improved water-splitting performance, there is still a big gap between their performance and the industrial requirement. The exploration of Mott–Schottky electrocatalysts with low cost, extraordinary activity and robust stability is highly desired. The construction of Mott–Schottky catalysts with high surface areas and abundant heterointerfaces to expose more active sites is an effective way to further enhance their catalytic activity. More importantly, the excellent stability of electrocatalysts which can meet the industrial requirement (delivering a large current for a few years) may play a pivotal role in practical water-splitting applications. In this regard, more studies should be conducted to further enhance the stability of Mott–Schottky electrocatalysts, such as multi-element doping, multi-component interfaces and so on.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors acknowledge the support from the start-up fund of Linyi University (40619025), Natural Science Foundation of Shandong Province (ZR2020QB032) and National Natural Science Foundation of China (NSFC) (22108113).

References

1. M. S. Faber and S. Jin, Energy Environ. Sci., 2014, 7, 3519–3542.
2. S. Chu and A. Majumdar, Nature, 2012, 488, 294–303.
3. M. G. Walter, E. L. Warren, J. R. McKone, S. W. Boettcher, Q. Mi, E. A. Santori and N. S. Lewis, Chem. Rev., 2010, 110, 6446–6473.
4. J. Zhang, Q. Zhang and X. Feng, Adv. Mater., 2019, 31, e1808167.
5. J. Wang, W. Cui, Q. Liu, Z. Xing, A. M. Asiri and X. Sun, Adv. Mater., 2016, 28, 215–230.
6. A. J. Bard and M. A. Fox, Acc. Chem. Res., 1995, 28, 141–145.
7. T. E. Mallouk, Nat. Chem., 2013, 5, 362–363.
8. P. Zhang, X. F. Lu, J. Nai, S.-Q. Zang and X. W. Lou, Adv. Sci., 2019, 6, 1900576.
9. Z. W. Seh, J. Kibsgaard, C. F. Dickens, I. Chorkendorff, J. K. Norskov and T. F. Jaramillo, Science, 2017, 355, 146.
10. K. Zeng and D. Zhang, Prog. Energy Combust. Sci., 2010, 36, 307–326.
11. B. M. Hunter, H. B. Gray and A. M. Muller, Chem. Rev., 2016, 116, 14120–14136.
12. Y. Wang, B. Kong, D. Zhao, H. Wang and C. Selomulya, Nano Today, 2017, 15, 26–55.
13. P. Zhai, Y. Zhang, Y. Wu, J. Gao, B. Zhang, S. Cao, Y. Zhang, Z. Li, L. Sun and J. Hou, Nat. Commun., 2020, 11, 5462.
14. C. Xie, Z. Niu, D. Kim, M. Li and P. Yang, Chem. Rev., 2019, 120, 1184–1249.
15. T. Sun, L. Xu, D. Wang and Y. Li, Nano Res., 2019, 12, 2067–2080.
16. J. D. Benck, T. R. Hellstern, J. Kibsgaard, P. Chakthranont and T. F. Jaramillo, ACS Catal., 2014, 4, 3957–3971.
17. J. Su, R. Ge, Y. Dong, F. Hao and L. Chen, J. Mater. Chem. A, 2018, 6, 14025–14042.
18. C.-X. Zhao, B.-Q. Li, M. Zhao, J.-N. Liu, L.-D. Zhao, X. Chen and Q. Zhang, Energy Environ. Sci., 2020, 13, 1711–1716.
19. J. Shan, Y. Zheng, B. Shi, K. Davey and S.-Z. Qiao, ACS Energy Lett., 2019, 4, 2719–2730.
20. L. Huang, R. Yao, X. Wang, S. Sun, X. Zhu, X. Liu, M. G. Kim, J. Lian, F. Liu, Y. Li, H. Zong, S. Han and X. Ding, Energy Environ. Sci., 2022, 15, 2425–2434.
21. Q. Liang, Q. Li, L. Xie, H. Zeng, S. Zhou, Y. Huang, M. Yan, X. Zhang, T. Liu, J. Zeng, K. Liang, O. Terasaki, D. Zhao, L. Jiang and B. Kong, ACS Nano, 2022, 16(5), 7993–8004.
22. B. Wang, C. Tang, H. F. Wang, X. Chen, R. Cao and Q. Zhang, Adv. Mater., 2018, 31, 1805658.
23. J. Chang, G. Wang, Z. Yang, B. Li, Q. Wang, R. Kuliiev, N. Orlovskaya, M. Gu, Y. Du and G. Wang, Adv. Mater., 2021, 33, 2101425.
24. C. G. Morales-Guio, L.-A. Stern and X. Hu, Chem. Soc. Rev., 2014, 43, 6555–6569.
25. J. Deng, H. Li, S. Wang, D. Ding, M. Chen, C. Liu, Z. Tian, K. S. Novoselov, C. Ma, D. Deng and X. Bao, Nat. Commun., 2017, 8, 14430.
26. H. Zhang, Y. Liu, T. Chen, J. Zhang, J. Zhang and X. W. Lou, Adv. Mater., 2019, 31, 1904548.
27. Y. Ito, T. Ohto, D. Hojo, M. Wakisaka, Y. Nagata, L. Chen, K. Hu, M. Izumi, J.-i. Fujita and T. Adschiri, ACS Catal., 2018, 8, 3579–3586.
28. W. Zhang, J. Chang, G. Wang, Z. Li, M. Wang, Y. Zhu, B. Li, H. Zhou, G. Wang, M. Gu, Z. Feng and Y. Yang, Energy Environ. Sci., 2022, 15, 1573–1584.
29. X. Liu, S. Xi, H. Kim, A. Kumar, J. Lee, J. Wang, N. Q. Tran, T. Yang, X. Shao, M. Liang, M. G. Kim and H. Lee, Nat. Commun., 2021, 12, 5676.
30. C. Yang, R. Zhao, H. Xiang, J. Wu, W. Zhong, W. Li, Q. Zhang, N. Yang and X. Li, Adv. Energy Mater., 2020, 10, 2002260.
31. D. Jin, L. R. Johnson, A. S. Raman, X. Ming, Y. Gao, F. Du, Y. Wei, G. Chen, A. Vojvodic, Y. Gogotsi and X. Meng, J. Phys. Chem. C, 2020, 124, 10584–10592.
32. Z. Zhao, Z. Zhu, F. Wang, S. Li, X. Bao, L. Zhang, S. Lin and Y. Yang, Chem. Eng. J., 2021, 415, 128885.
33. Y. Shi and B. Zhang, Chem. Soc. Rev., 2016, 45, 1529–1541.
34. R. Bernasconi, M. I. Khalil, C. Iaquinta, C. Lenardi, L. Nobili and L. Magagnin, ACS Appl. Energy Mater., 2020, 3, 6525–6535.
35. L. Xie, R. Zhang, L. Cui, D. Liu, S. Hao, Y. Ma, G. Du, A. M. Asiri and X. Sun, Angew. Chem., Int. Ed. Engl., 2017, 56, 1064–1068.
36. K. Liang, S. Pakhira, Z. Yang, A. Nijamudheen, L. Ju, M. Wang, C. I. Aguirre-Velez, G. E. Sterbinsky, Y. Du and Z. Feng, ACS Catal., 2018, 9, 651–659.
37. H. Kim, Y. Lee, D. Song, Y. Kwon, E.-J. Kim and E. Cho, Sustainable Energy Fuels, 2020, 4, 5247–5253.
38. H. Jin, Q. Gu, B. Chen, C. Tang, Y. Zheng, H. Zhang, M. Jaroniec and S.-Z. Qiao, Chem, 2020, 6, 2382–2394.
39. Z. Liu, X. Zhang, H. Song, Y. Yang, Y. Zheng, B. Gao, J. Fu, P. K. Chu and K. Huo, ChemCatChem, 2020, 12, 2962–2966.
40. J. Zhang, L. Zhang, L. Du, H. L. Xin, J. B. Goodenough and Z. Cui, Angew. Chem., Int. Ed. Engl., 2020, 59, 17488–17493.
41. T. Liu, P. Diao, Z. Lin and H. Wang, Nano Energy, 2020, 74, 104787.
42. V. Yadav, J. S. Lowe, A. J. Shumski, E. Z. Liu, J. Greeley and C. W. Li, ACS Catal., 2020, 10, 13305–13313.
43. S. Pakhira and S. N. Upadhyay, Sustainable Energy Fuels, 2022, 6, 1733–1752.
44. Z. Jin and A. J. Bard, Proc. Natl. Acad. Sci. U. S. A., 2020, 117, 12651.
45. I. Yamada, M. Kinoshita, S. Oda, H. Tsukasaki, S. Kawaguchi, K. Oka, S. Mori, H. Ikeno and S. Yagi, Chem. Mater., 2020, 32, 3893–3903.
46. Z. Liu, G. Wang, X. Zhu, Y. Wang, Y. Zou, S. Zang and S. Wang, Angew. Chem., Int. Ed. Engl., 2020, 59, 4736–4742.
47. Z. Xiao, Y.-C. Huang, C.-L. Dong, C. Xie, Z. Liu, S. Du, W. Chen, D. Yan, L. Tao and Z. Shu, J. Am. Chem. Soc., 2020, 142, 12087–12095.
48. D. Y. Chung, P. P. Lopes, P. Farinazzo Bergamo Dias Martins, H. He, T. Kawaguchi, P. Zapol, H. You, D. Tripkovic, D. Strmcnik, Y. Zhu, S. Seifert, S. Lee, V. R. Stamenkovic and N. M. Markovic, Nat. Energy, 2020, 5, 222–230.
49. S. Bera, W.-J. Lee, E.-K. Koh, C.-M. Kim, S. Ghosh, Y. Yang and S.-H. Kwon, J. Phys. Chem. C, 2020, 124, 16879–16887.
50. S. Hao, L. Chen, C. Yu, B. Yang, Z. Li, Y. Hou, L. Lei and X. Zhang, ACS Energy Lett., 2019, 4, 952–959.
51. P. Zhai, M. Xia, Y. Wu, G. Zhang, J. Gao, B. Zhang, S. Cao, Y. Zhang, Z. Li, Z. Fan, C. Wang, X. Zhang, J. T. Miller, L. Sun and J. Hou, Nat. Commun., 2021, 12, 4587.
52. Y. V. Kaneti, Y. Guo, N. L. W. Septiani, M. Iqbal, X. Jiang, T. Takei, B. Yuliarto, Z. A. Alothman, D. Golberg and Y. Yamauchi, Chem. Eng. J., 2021, 405, 126580.
53. C. Wang, W. Chen, D. Yuan, S. Qian, D. Cai, J. Jiang and S. Zhang, Nano Energy, 2020, 69, 104453.
54. J. Wu, D. Wang, S. Wan, H. Liu, C. Wang and X. Wang, Small, 2020, 16, 1900550.
55. H. Liu, J. Lei, S. Yang, F. Qin, L. Cui, Y. Kong, X. Zheng, T. Duan, W. Zhu and R. He, Appl. Catal., B, 2021, 286, 119894.
56. Y. Yuan, S. Adimi, X. Guo, T. Thomas, Y. Zhu, H. Guo, G. S. Priyanga, P. Yoo, J. Wang, J. Chen, P. Liao, J. P. Attfield and M. Yang, Angew. Chem., Int. Ed. Engl., 2020, 59, 18036–18041.
57. R.-Q. Yao, H. Shi, W.-B. Wan, Z. Wen, X.-Y. Lang and Q. Jiang, Adv. Mater., 2020, 32, 1907214.
58. Z. Li, W. Niu, Z. Yang, A. Kara, Q. Wang, M. Wang, M. Gu, Z. Feng, Y. Du and Y. Yang, Energy Environ. Sci., 2020, 13, 3110–3118.
59. Z. Liu, G. Wang, X. Zhu, Y. Wang, Y. Zou, S. Zang and S. Wang, Angew. Chem., Int. Ed. Engl., 2020, 59, 4736–4742.
60. Y. Wang, J. Yao, H. Li, D. Su and M. Antonietti, J. Am. Chem. Soc., 2011, 133, 2362–2365.
61. H. Su, K.-X. Zhang, B. Zhang, H.-H. Wang, Q.-Y. Yu, X.-H. Li, M. Antonietti and J.-S. Chen, J. Am. Chem. Soc., 2017, 139, 811–818.
62. X. H. Li and M. Antonietti, Chem. Soc. Rev., 2013, 42, 6593–6604.
63. Z. Zhuang, Y. Li, Z. Li, F. Lv, Z. Lang, K. Zhao, L. Zhou, L. Moskaleva, S. Guo and L. Mai, Angew. Chem., Int. Ed. Engl., 2018, 57, 496–500.
64. A. Biswas, S. Nandi, N. Kamboj, J. Pan, A. Bhowmik and R. S. Dey, ACS Nano, 2021, 15, 20364–20376.
65. Y. X. Lin, S. N. Zhang, Z. H. Xue, J. J. Zhang, H. Su, T. J. Zhao, G. Y. Zhai, X. H. Li, M. Antonietti and J. S. Chen, Nat. Commun., 2019, 10, 4380.
66. Z. Zhang and J. T. Yates Jr, Chem. Rev., 2012, 112, 5520–5551.
67. Y. Zhang, Y. Lin, T. Duan and L. Song, Mater. Today, 2021, 48, 115–134.
68. J. Liu, X. Yang, F. Si, B. Zhao, X. Xi, L. Wang, J. Zhang, X.-Z. Fu and J.-L. Luo, Nano Energy, 2022, 103, 107753.
69. Y. Yang, M. Luo, W. Zhang, Y. Sun, X. Chen and S. Guo, Chem, 2018, 4, 2054–2083.
70. G.-L. Chai, Z. Hou, D.-J. Shu, T. Ikeda and K. Terakura, J. Am. Chem. Soc., 2014, 136, 13629–13640.
71. Z. Sun, Y. Wang, L. Zhang, H. Wu, Y. Jin, Y. Li, Y. Shi, T. Zhu, H. Mao and J. Liu, Adv. Funct. Mater., 2020, 30, 1910482.
72. Y. Li, W. Wang, B. Zhang, L. Fu, M. Wan, G. Li, Z. Cai, S. Tu, X. Duan, Z. W. Seh, J. Jiang and Y. Sun, Nano Lett., 2021, 21, 6656–6663.
73. C. Zhang, R. Du, J. J. Biendicho, M. Yi, K. Xiao, D. Yang, T. Zhang, X. Wang, J. Arbiol, J. Llorca, Y. Zhou, J. R. Morante and A. Cabot, Adv. Energy Mater., 2021, 11, 2100432.
74. H. Yang, B. Wang, S. Kou, G. Lu and Z. Liu, Chem. Eng. J., 2021, 425, 131589.
75. H. Li, C. Chen, D. Yan, Y. Wang, R. Chen, Y. Zou and S. Wang, J. Mater. Chem. A, 2019, 7, 23432–23450.
76. A. Vojvodic and J. K. Nørskov, Science, 2011, 334, 1355–1356.
77. Y. Wu, Z. Jiang, X. Lu, Y. Liang and H. Wang, Nature, 2019, 575, 639–642.
78. M. Chen, S. Shu, J. Li, X. Lv, F. Dong and G. Jiang, J. Hazard. Mater., 2020, 389, 121876.
79. S. Chen, C. Wang, S. Liu, M. Huang, J. Lu, P. Xu, H. Tong, L. Hu and Q. Chen, J. Phys. Chem. Lett., 2021, 12, 4849–4856.
80. N. Wang, S. Ning, X. Yu, D. Chen, Z. Li, J. Xu, H. Meng, D. Zhao, L. Li and Q. Liu, Appl. Catal., B, 2022, 302, 120838.
81. M. Yuan, J. Chen, Y. Bai, Z. Liu, J. Zhang, T. Zhao, Q. Wang, S. Li, H. He and G. Zhang, Angew. Chem., Int. Ed. Engl., 2021, 60, 10910–10918.
82. Z. Sun, L. Lin, M. Yuan, H. Yao, Y. Deng, B. Huang, H. Li, G. Sun and J. Zhu, Nano Energy, 2022, 101, 107563.
83. Y. X. Liu, H. H. Wang, T. J. Zhao, B. Zhang, H. Su, Z. H. Xue, X. H. Li and J. S. Chen, J. Am. Chem. Soc., 2019, 141, 38–41.
84. Y. Jiao, Y. Zheng, M. Jaroniec and S. Z. Qiao, Chem. Soc. Rev., 2015, 44, 2060–2086.
85. W. Liu, H. Zhang, C. Li, X. Wang, J. Liu and X. Zhang, J. Energy Chem., 2020, 47, 333–345.
86. J. Greeley, T. F. Jaramillo, J. Bonde, I. Chorkendorff and J. K. Nørskov, Nat. Mater., 2006, 5, 909–913.
87. W. Yu, H. Huang, Y. Qin, D. Zhang, Y. Zhang, K. Liu, Y. Zhang, J. Lai and L. Wang, Adv. Energy Mater., 2022, 12, 2200110.
88. Z. Chen, Y. Xu, D. Ding, G. Song, X. Gan, H. Li, W. Wei, J. Chen, Z. Li, Z. Gong, X. Dong, C. Zhu, N. Yang, J. Ma, R. Gao, D. Luo, S. Cong, L. Wang, Z. Zhao and Y. Cui, Nat. Commun., 2022, 13, 763.
89. D. Strmcnik, M. Uchimura, C. Wang, R. Subbaraman, N. Danilovic, D. van der Vliet, A. P. Paulikas, V. R. Stamenkovic and N. M. Markovic, Nat. Chem., 2013, 5, 300–306.
90. K. Xia, Z. Chen, J. Yi, H. Xu, Y. Yu, X. She, Z. Mo, H. Chen, Y. Xu and H. Li, Ind. Eng. Chem. Res., 2018, 57, 8863–8870.
91. X. Ji, K. Wang, Y. Zhang, H. Sun, Y. Zhang, T. Ma, Z. Ma, P. Hu and Y. Qiu, Sustainable Energy Fuels, 2020, 4, 407–416.
92. Y. Wang, B. Liu, Y. Liu, C. Song, W. Wang, W. Li, Q. Feng and Y. Lei, Chem. Commun., 2020, 56, 14019–14022.
93. H. Z. Yu, Y. Wang, J. Ying, S. M. Wu, Y. Lu, J. Hu, J. S. Hu, L. Shen, Y. X. Xiao, W. Geng, G. G. Chang, C. Janiak, W. H. Li and X. Y. Yang, ACS Appl. Mater. Interfaces, 2019, 11, 27641–27647.
94. R. Xiang, Y. Duan, L. Peng, Y. Wang, C. Tong, L. Zhang and Z. Wei, Appl. Catal., B, 2019, 246, 41–49.
95. J. Chen, J. Zheng, W. He, H. Liang, Y. Li, H. Cui and C. Wang, Nano Res., 2022 DOI:10.1007/s12274-022-5072-1.
96. Z. P. Wu, X. F. Lu, S. Q. Zang and X. W. Lou, Adv. Funct. Mater., 2020, 30, 1910274.
97. I. C. Man, H. Y. Su, F. Calle-Vallejo, H. A. Hansen, J. I. Martínez, N. G. Inoglu, J. Kitchin, T. F. Jaramillo, J. K. Nørskov and J. Rossmeisl, ChemCatChem, 2011, 3, 1159–1165.
98. A. Vojvodic and J. K. Nørskov, Natl. Sci. Rev., 2015, 2, 140–143.
99. X. Wang, X. Zong, B. Liu, G. Long, A. Wang, Z. Xu, R. Song, W. Ma, H. Wang and C. Li, Small, 2022, 18, e2105544.
100. Q. Wen, K. Yang, D. Huang, G. Cheng, X. Ai, Y. Liu, J. Fang, H. Li, L. Yu and T. Zhai, Adv. Energy Mater., 2021, 11, 2102353.
101. B. Zhang, T. Lu, Y. Ren, L. Huang, H. Pang, Q. Zhao, S. Tian, J. Yang, L. Xu, Y. Tang and X. Tian, Chem. Eng. J., 2022, 448, 137709.
102. H. Meng, Z. Ren, S. Du, J. Wu, X. Yang, Y. Xue and H. Fu, Nanoscale, 2018, 10, 10971–10978.
103. P. Zhang, Z. Cai, S. You, F. Wang, Y. Dai, C. Zhang, Y. Zhang, N. Ren and J. Zou, J. Colloid Interface Sci., 2019, 557, 580–590.
104. D. Liang, C. Lian, Q. Xu, M. Liu, H. Liu, H. Jiang and C. Li, Appl. Catal., B, 2020, 268, 118417.
105. X. Wang, G. Zhan, Y. Wang, Y. Zhang, J. Zhou, R. Xu, H. Gai, H. Wang, H. Jiang and M. Huang, J. Energy Chem., 2022, 68, 113–123.
106. K. Li, R. Cheng, Q. Xue, P. Meng, T. Zhao, M. Jiang, M. Guo, H. Li and C. Fu, Chem. Eng. J., 2022, 450, 137991.
107. J. Chen, C. Fan, X. Hu, C. Wang, Z. Huang, G. Fu, J. M. Lee and Y. Tang, Small, 2019, 15, e1901518.
108. T. Li, Y. Hu, X. Pan, J. Yin, Y. Li, Y. Wang, Y. Zhang, H. Sun and Y. Tang, Chem. Eng. J., 2020, 392, 124845.
109. Y. Yan, B. Y. Xia, B. Zhao and X. Wang, J. Mater. Chem. A, 2016, 4, 17587–17603.
110. E. A. Hernández-Pagán, N. M. Vargas-Barbosa, T. Wang, Y. Zhao, E. S. Smotkin and T. E. Mallouk, Energy Environ. Sci., 2012, 5, 7582–7589.
111. B. H. Suryanto, Y. Wang, R. K. Hocking, W. Adamson and C. Zhao, Nat. Commun., 2019, 10, 1–10.
112. L. Dai, Z. N. Chen, L. Li, P. Yin, Z. Liu and H. Zhang, Adv. Mater., 2020, 32, 1906915.
113. W. Luo, Y. Wang, L. Luo, S. Gong, M. Wei, Y. Li, X. Gan, Y. Zhao, Z. Zhu and Z. Li, ACS Catal., 2022, 12, 1167–1179.
114. S. Kumaravel, K. Karthick, S. S. Sankar, A. Karmakar, R. Madhu, K. Bera and S. Kundu, Sustainable Energy Fuels, 2021, 5, 6215–6268.
115. Z.-H. Xue, H. Su, Q.-Y. Yu, B. Zhang, H.-H. Wang, X.-H. Li and J.-S. Chen, Adv. Energy Mater., 2017, 7, 1602355.
116. J. Hou, Y. Sun, Y. Wu, S. Cao and L. Sun, Adv. Funct. Mater., 2018, 28, 1704447.
117. Z. Fan, J. Jiang, L. Ai, Z. Shao and S. Liu, ACS Appl. Mater. Interfaces, 2019, 11, 47894–47903.
118. Z. Xu, S. Jin, M. H. Seo and X. Wang, Appl. Catal., B, 2021, 292, 120168.
119. P. Zhang, Y. Liu, T. Liang, E. H. Ang, X. Zhang, F. Ma and Z. Dai, Appl. Catal., B, 2021, 284, 119738.
120. J.-Y. Wang, T. Ouyang, N. Li, T. Ma and Z.-Q. Liu, Sci. Bull., 2018, 63, 1130–1140.
121. Q. Zhang, F. Luo, X. Long, X. Yu, K. Qu and Z. Yang, Appl. Catal., B, 2021, 298, 120511.
122. T. Li, J. Yin, D. Sun, M. Zhang, H. Pang, L. Xu, Y. Zhang, J. Yang, Y. Tang and J. Xue, Small, 2022, 18, e2106592.
123. K. He, T. Tadesse Tsega, X. Liu, J. Zai, X. H. Li, X. Liu, W. Li, N. Ali and X. Qian, Angew. Chem., Int. Ed. Engl., 2019, 58, 11903–11909.
124. Z. Fan, L. Zhang, D. Baumann, L. Mei, Y. Yao, X. Duan, Y. Shi, J. Huang, Y. Huang and X. Duan, Adv. Mater., 2019, 31, 1900608.
125. A. Bergmann, T. E. Jones, E. Martinez Moreno, D. Teschner, P. Chernev, M. Gliech, T. Reier, H. Dau and P. Strasser, Nat. Catal., 2018, 1, 711–719.
126. M. Wang, L. Árnadóttir, Z. J. Xu and Z. Feng, Nano-Micro Lett., 2019, 11, 1–18.
127. J. Suntivich, K. J. May, H. A. Gasteiger, J. B. Goodenough and Y. Shao-Horn, Science, 2011, 334, 1383–1385.
128. M. G. Mavros, T. Tsuchimochi, T. Kowalczyk, A. McIsaac, L.-P. Wang and T. V. Voorhis, Inorg. Chem., 2014, 53, 6386–6397.
129. J. L. Fajín, M. N. D. Cordeiro, F. Illas and J. R. Gomes, J. Catal., 2010, 276, 92–100.

---

Footnote

† YeXing Tong and Wei Liu contributed equally to this work.

---