Preparation of MoS₂/TiO₂ based nanocomposites for photocatalysis and rechargeable batteries: progress, challenges, and perspective

Abstract

The rapidly increasing severity of the energy crisis and environmental degradation are stimulating the rapid development of photocatalysts and rechargeable lithium/sodium ion batteries. In particular, MoS₂/TiO₂ based nanocomposites show great potential and have been widely studied in the areas of both photocatalysis and rechargeable lithium/sodium ion batteries due to their superior combination properties. In addition to the low-cost, abundance, and high chemical stability of both MoS₂ and TiO₂, MoS₂/TiO₂ composites also show complementary advantages. These include the strong optical absorption of TiO₂vs. the high catalytic activity of MoS₂, which is promising for photocatalysis; and excellent safety and superior structural stability of TiO₂vs. the high theoretic specific capacity and unique layered structure of MoS₂, thus, these composites are exciting as anode materials. In this review, we first summarize the recent progress in MoS₂/TiO₂-based nanomaterials for applications in photocatalysis and rechargeable batteries. We highlight the synthesis, structure and mechanism of MoS₂/TiO₂-based nanomaterials. Then, advancements and strategies for improving the performance of these composites in photocatalytic degradation, hydrogen evolution, CO₂ reduction, LIBs and SIBs are critically discussed. Finally, perspectives on existing challenges and probable opportunities for future exploration of MoS₂/TiO₂-based composites towards photocatalysis and rechargeable batteries are presented. We believe the present review would provide enriched information for a deeper understanding of MoS₂/TiO₂ composites and open avenues for the rational design of MoS₂/TiO₂ based composites for energy and environment-related applications.

---

Biao Chen

Biao Chen is a Ph. D candidate under the supervision of Prof. Naiqin Zhao at the School of Materials Science and Engineering at Tianjin University. He received his B.S. degree from Tianjin University in 2013. His current research focuses on the design, fabrication and properties of nanostructures for energy and environmental applications, including lithium-ion batteries, sodium-ion batteries, lithium–sulfur batteries, and solar cells. He has published six papers as the first author in Nano Energy, Nanoscale, ACS Applied Materials & Interfaces, etc.

Naiqin Zhao

Naiqin Zhao is a professor in the School of Materials Science and Engineering at Tianjin University. She obtained her B.Sc. and Ph.D. at Tianjin University. She was a visiting scholar at the Illinois Institute of Technology and The Hong Kong Polytechnic University, and a visiting professor at Tohoku University and Vanderbilt University. Her research interests are phase transformation and properties of alloys and composites, synthesis and characteristics of nanophase materials and nanocomposites, and their applications in energy storage materials. She has published 187 papers in Advanced Materials, Advanced Energy Materials, ACS Nano, Nano Energy, etc.

---

1. Introduction

Currently, the severe energy crisis and environmental degradation are rampant due to the increasing depletion of fossil fuels. In order to address these problems, developing clean energy and new energy storage technologies that are environmentally friendly and inexpensive is urgently necessary. Among current candidates, photocatalysis associated with solar energy and rechargeable lithium/sodium ion batteries are emerging as promising methods to address this difficult situation.^1–10 Thus, considerable research efforts have been devoted to exploring appropriate materials for practical applications. Importantly, it is well known that TiO₂ is a star material in various energy-related devices, especially photocatalysis^11–16 and batteries,^17–24 because it has outstanding physical and chemical properties, including chemical stability, long durability, nontoxicity, abundance, and low cost. For photocatalysis, which was first achieved by Fujijshma and Honda in 1972, TiO₂ was employed for the water photolysis reaction under ultraviolet (UV) light irradiation.¹¹ Since this breakthrough, TiO₂ has been the most widely studied material to remove pollutants^25–27 and obtain hydrogen.^28–30 However, the efficient utilization of TiO₂ still suffers from rapid recombination of electron–hole pairs and low surface activation for redox reactions. In addition, its wide band gap energy (3.2 eV) allows only absorption of UV light, which is a small portion of natural light. To improve the photoactivity of TiO₂, loading of a co-catalyst, especially noble metal, is a typical method to accelerate the separation of electron–hole pairs and provide active sites for molecular reactions.^31–33 However, the scarcity and high cost of noble metals limit their widespread application. Moreover, TiO₂/noble metal systems still can only be used for absorption of UV light. Recently, nano-sized MoS₂ has attracted a great deal of attention as a promising substitute for noble metal co-catalysts because of its abundance, low cost, high activity, and excellent chemical stability.^34–40 Moreover, the band gap of MoS₂ changes from 1.3 to 1.9 eV as its thickness decreases to the size of a monolayer.^41–43 Therefore, the obtained MoS₂/TiO₂ heterojunction system shows high charge carrier separation and migration efficiencies, numerous active sites for reaction, and a broad light harvesting range that endows it with excellent photocatalytic activity. Therefore, MoS₂ has been very frequently investigated in recent years.

Rechargeable lithium ion batteries (LIBs) and sodium ion batteries (SIBs) are of great importance in energy storage.^44–47 Among potential candidates, TiO₂ has been extensively explored as a promising anode material due to its advantages of low volume variation and safe operating voltage during the charge–discharge process.^17–24 However, the low theoretical specific capacity of TiO₂ limits its application. In contrast, MoS₂ with a layered structure shows high theoretical specific capacity but low structural stability, leading to poor electrochemical performance.^48–54 Therefore, constructing composites composed of TiO₂ and MoS₂ is an attractive strategy to exploit their mutual advantages and address shortages to solve these problems. The study of MoS₂/TiO₂-based nanocomposites as anode materials for LIBs or SIBs has attracted significant attention in recent years.

Based on the abovementioned analyses, we believe that the MoS₂/TiO₂ based materials are attracting increasing interest in various environmental and energy-related applications as demonstrated in a number of scientific publications on MoS₂/TiO₂ based materials in the past four years (Fig. S1†). Recently, several review articles have focused on the synthesis and application of TiO₂-based^11–24 and MoS₂-based^{34–40,48–54} composites. For example, in 2014, Tian and co-authors²¹ studied one-dimensional TiO₂-based heterostructures for broad applications, including photocatalysis, dye-sensitized solar cells, sensors, LIBs, biomedicine, catalysis, and supercapacitors. Wang et al.⁴⁹ summarized the synthesis of MoS₂-based composites and their broad electrochemical energy storage applications, including LIBs, SIBs, and supercapacitors. The reported reviews discuss a wide range of MoS₂-based materials and TiO₂-based materials in various fields, including electrocatalysis, photocatalysis, electronic devices, sensors, bio-applications, LIBs, SIBs, solar cells, and supercapacitors; however, MoS₂/TiO₂ composites, which are indeed an important and exciting topic in these areas, have rarely been discussed. Therefore, there is an urgent need to focus on the recent development and progress in the synthesis of MoS₂/TiO₂-based composites for environment and energy-related applications, especially in photocatalysis and rechargeable batteries (LIBs and SIBs).

In this review, we aim to provide a clear understanding of the basic introduction, progress, and challenges of MoS₂/TiO₂-based nanomaterials for photocatalysis, LIBs, and SIBs. First, the synthetic methods for MoS₂/TiO₂-based nanomaterials are introduced and classified. Then, the applications of MoS₂/TiO₂-based nanomaterials in photocatalysis, LIBs, and SIBs, as well as their interesting associated structures, functions, and mechanisms, are elucidated. The major challenges and issues facing MoS₂/TiO₂-based nanomaterials are also discussed. Finally, some suggestions towards achieving high performance devices and perspectives on the future research direction of MoS₂/TiO₂-based composites are proposed.

2. Properties

Prior to our review of MoS₂/TiO₂ composites, first, in this section, the individual crystal structures and electrical properties of TiO₂ and MoS₂ would be briefly introduced since the properties and applications of composite materials normally depend on their individual components.

2.1 Crystal structures and electrical properties of TiO₂

There are four major polymorphs of TiO₂ according to research: rutile, anatase, brookite and TiO₂ bronze (TiO₂-B). As shown in Fig. 1 and Table 1, these structures are composed of distorted TiO₆ octahedra, where each Ti⁴⁺ cation is surrounded by six O²⁻ anions; namely, the polymorphs of TiO₂ rely on the arrangement of TiO₆ octahedra. Rutile, anatase, brookite and TiO₂-B contain TiO₆ octahedra joined by shared edges, corners, both corners and edges, and both sheets and edges, respectively.^55,56 The space groups of the four polymorphs are summarized in Table 1.¹⁸ Rutile has a tetragonal structure, with a = 0.459 nm and c = 0.296 nm; anatase also exhibits a tetragonal structure, with a = 0.536 nm and c = 0.953 nm; brookite presents an orthorhombic structure, with a = 0.915 nm, b = 0.544 nm, and c = 0.514 nm; and TiO₂-B possesses a monoclinic structure, with a = 1.217 nm, b = 0.374 nm, c = 0.651 nm, and β = 107.29°.^57–59 It is well known that the crystal structures/phases of TiO₂ can significantly affect the performance of LIBs, SIBs and photocatalysts. The structures and properties of voids in polymorphs directly influence the diffusion and storage of Li-ions and Na-ions. As can be seen in Fig. 1, TiO₂-B has a very open crystal structure that can facilitate diffusion and provide large storage sites for Li-ions and Na-ions; thus, this material is of great interest in LIBs and SIBs.^18,20 The differences in lattice structures also result in different electrical properties, which is important for photocatalysis performance. TiO₂ is a n-type semiconductor, with band gaps of 3.0, 3.2 and 2.96 eV for the rutile, anatase and brookite phases, respectively.⁶⁰ According to the molecular orbitals in TiO₂, the valence band of TiO₂ is formed of hybrid orbitals between the 3d orbitals of titanium and the 2p orbitals of oxygen. Meanwhile, the conduction band consists of the 3d orbitals of titanium.⁶¹ The large band gap means that electrons in TiO₂ can be excited by near-UV light from the valence band to the conduction band, leaving holes. The probability of combination of electron–hole pairs is reduced due to dissimilar parity. Among the four phases, brookite and TiO₂-B have been rarely reported in photocatalysis; however, anatase TiO₂ has been widely accepted to possess superior photocalytic activity due to its rapid hole trapping process.⁶² Moreover, in addition to its polymorphs, the most important aspect of TiO₂ is its exposed surface since many physical and chemical processes, particularly photocatalytic reactions, occur on the surface.²⁸

Fig. 1 Crystal structures of TiO₂ polymorphs: (a) rutile; (b) anatase; (c) brookite; and (d) TiO₂-B. Purple spheres represent Ti atoms, and blue octahedra represent TiO₆ blocks. Oxygen atoms at the corner of the octahedra are omitted for clarity. Reproduced with permission from ref. 55. Copyright 2015, Royal Society of Chemistry.

Table 1 Crystal structures and electrical property parameters of four common TiO₂ polymorphs

Polymorphs	Crystal structures				Electrical properties
	Space group	Crystal system	Unit cell (Å)	Ref.	Band gap (eV)	Conductivity	Bonding	Ref.
Rutile	P42/mnm	Tetragonal	a = 4.59, c = 2.96	21 and 55	3.0 indirect	n-Type semiconductor	Covalent bonding	60
Anatase	I41/amd	Tetragonal	a = 3.79, c = 9.51	21 and 55	3.2 indirect	n-Type semiconductor		60
Brookite	Pbca	Orthorhombic	a = 9.17, b = 5.46, c = 5.14	21 and 55	2.96 indirect	n-Type semiconductor		60
TiO2-B	C2/m	Monoclinic	a = 12.17, b = 3.74, c = 6.51, β = 107.29°	21 and 55	N/A

---

2.2 Crystal structures and electrical properties of MoS₂

As shown in Fig. 2 and Table 2, according to previous investigations,^63,64 there are three polymorphs of bulk MoS₂: 1T-MoS₂, 2H-MoS₂, and 3R-MoS₂, where the number represents the number of layers per repeat unit while the letter denotes the symmetry. 1T-MoS₂ has a tetragonal unit cell that contains 1 layer with a = 0.560 nm and c = 1.838 nm.⁶⁵ 2H-MoS₂ has a hexagonal unit cell that contains 2 layers with a = 0.315 nm, c = 1.230 nm, and γ = 120°,⁶⁶ and 3R-MoS₂ exhibits a rhombohedral unit cell that contains 3 layers with a = 0.317 nm, c = 1.838 nm, and γ = 120° (Fig. 2a).⁶⁷ Fundamentally, the three polymorphs vary in the Mo atom coordination and the stacking orders of each layer.^63,64 It is well known that a single layer of MoS₂ consists of three atomic layers, where one Mo atomic layer is surrounded by double S atomic layers. Generally, there are three possible positions of the S sheets and Mo sheets, which can be labelled using the designations A, B, and C and the designations a, b, and c, respectively;⁶⁸ thus, any single MoS₂ layer can be described as a combination of a lowercase letter between two uppercase letters. As shown in Fig. 2b, 1T-MoS₂ can be defined as AbC, as it displays octahedral coordination (O_h) of the Mo atom; namely, each S atom forms the apex of the octahedron, with the Mo atom located in its center. 2H-MoS₂ and 3R-MoS₂ can be described as AbA BaB AbA… and AbA BcB CaC AbA…, respectively, because both 2H-MoS₂ and 3R-MoS₂ represent trigonal prismatic coordination (D_3h) of the Mo atom due to its d⁴sp hybrid orbitals.

Fig. 2 (a) Schematics of the structural polytypes of MoS₂: 2H (hexagonal symmetry, two layers per repeat unit, trigonal prismatic coordination), 3R (rhombohedral symmetry, three layers per repeat unit, trigonal prismatic coordination), and 1T (tetragonal symmetry, one layer per repeat unit, octahedral coordination). (b) Schematic crystal structures of three polytypes: 1T, 2H, and 3R. The dashed lines show how the top views and the lateral views match with each other. In the 1T and 2H polytypes, atoms of the underlying layers overlap exactly with the upper layers. However, in the 3R polytype, some atoms of the underlying layers do not overlap with the above layers; thus, they are shown faintly in the top view. Reproduced with permission from: (a) ref. 63. Copyright 2015, Nature Publishing Group; (b) ref. 64. Copyright 2015, Royal Society of Chemistry.

Table 2 Crystal structures and electrical property parameters of three common MoS₂ polymorphs

Polymorphs	Crystal structures				Electrical properties
	Space group	Stacking order	Unit cell (Å)	Ref.	Band gap (eV)	Conductivity	Bonding	Ref.
1T	P3̄m1	AbC	a = 5.60, c = 18.38	65	N/A	Metallic	Strong Mo–S covalent bond (about 2 eV) in single layer and weak van der Waals forces (about 200 meV) between S–Mo–S layers	65
2H	P63/mmc	AbA BaB	a = 3.15, c = 12.30, γ = 120°	66	Bulk 1.29 (indirect)	Semiconductor (P-type)		69
					Monolayer 1.9 (direct)
3R	R3m	AbA BcB CaC	a = 3.17, c = 18.38 γ = 120°	67	1.29 (indirect)	Semiconductor (P-type)		64

---

Among the different types, 2H-MoS₂ is the most popular material in electronic, optical, and catalytic applications. Many scientists have thoroughly discussed its electronic properties. Fig. 3 shows that bulk MoS₂ is an indirect semiconductor with a band gap of 1.29 eV. However, MoS₂ crystals exhibit crossover from an indirect- to a direct-gap semiconductor in the monolayer limit.⁶⁹ Experiments reveal a progressive confinement-induced shift in the direct-gap with decreasing number of layers, from the bulk value of 1.29 eV to over 1.90 eV. Therefore, in the monolayer, 2H-MoS₂ is a direct-gap semiconductor with a band gap of 1.9 eV.⁶⁹ In the 3R and 1T phases, the properties of 3R-MoS₂ are similar to those of 2H-MoS₂;^64,70 the 1T-MoS₂ phase has also attracted the attention of scientists due to its interesting properties in different areas,^71–73 although the 1T-MoS₂ phase has not yet been found in nature. From band structure calculations, a negative temperature coefficient and Pauli paramagnetism were observed for the electronic conductivity of 1T phase. Thus, unlike 2H phase (Fermi level in the d band gap), the 1T phase (Fermi level within the d band) is metallic. These structural and electronic distortions indicate that 1T-MoS₂ is easily accessible for all electron transfer and exchange reactions.⁶⁵

Fig. 3 Lattice structures of MoS₂ in both the in- and out-plane directions and simplified band structure of bulk MoS₂, showing the lowest conduction band c1 and the highest split valence bands v1 and v2. A and B are the direct-gap transitions, and I is the indirect-gap transition. E′_g is the indirect-gap for the bulk, and E_g is the direct-gap for the monolayer. Reproduced with permission from ref. 69. Copyright 2010, American Physical Society.

3. Synthesis, structure and mechanism

Currently, increasing numbers of researchers are devoting effort to developing synthetic methods to obtain MoS₂/TiO₂-based composites with uniform distribution, strong interaction, and unique structures for various applications. It is generally accepted that the synthetic strategies of the composites can be roughly divided into two categories: ex situ synthetic strategies and in situ synthetic strategies. The synthetic strategies of MoS₂/TiO₂-based composites are illustrated in Fig. 4. We will carefully summarize recent developments as follows.

Fig. 4 Illustrations of ex situ and in situ synthetic strategies of MoS₂/TiO₂-based composites.

3.1 Ex situ synthetic strategy

In the ex situ synthetic strategy (Fig. 4a), MoS₂ and TiO₂ are prepared separately in advance; then, the MoS₂/TiO₂ composites are synthesized by drop-casting,⁷⁴ sonication assembly,^75–77 impregnation,^78–81 ball-milling,^82,83 sol–gel,⁸⁴ or hydrothermal/solvothermal assembly^85–87 methods. Table 3 summarizes the various methods and parameters.

Table 3 A summary of typical MoS₂/TiO₂-based composites synthesized by ex situ synthetic strategies

Composite	SEM/TEM image	Reaction conditions	Ref. (year)
Reproduced with permission from: ref. 75. Copyright 2015, Springer; ref. 76. Copyright 2017, De Gruyte; ref. 78. Copyright 2015, Royal Society of Chemistry; ref. 79. Copyright 2015, American Chemical Society; ref. 83. Copyright 2015, Royal Society of Chemistry; ref. 82. Copyright 2016, Wiley-VCH; ref. 84. Copyright 2017, Elsevier; ref. 85. Copyright 2014, Elsevier.
Sonication assembly
TiO2–MoS2(1T)		A mixture of a dispersion of MoS2 in ethanol and an aqueous dispersion of TiO2 was sonicated for 3 min	75 (2015)
MoS2/TiO2		Mesoporous TiO2 was added to a MoS2 suspension and sonicated at a frequency of 50/60 Hz for 60 min	76 (2017)
Impregnation
MoS2/TiO2 mesocrystals		TiO2 mesocrystals were impregnated with exfoliated MoS2 nanosheets in Milli-Q water	78 (2015)
MoS2/TiO2 nanotubes		The composites were obtained by impregnation of TiO2 nanotubes with a centrifuged solution of nanosized MoS2 particles in isopropyl alcohol	79 (2015)
Ball-milling
TiO2/MoS2		2.0 g TiO2 was mixed with MoS2, agate balls, and 4.0 mL of absolute ethanol; then, the mixtures were ball-milled at 300 rpm for 4 h	83 (2015)
Black-TiO2/MoS2/TiO2 nanosheets		3.0 g TiO2 was mixed with 0.18 g MoS2, agate balls, and 5.0 mL of absolute ethanol; then, the mixtures were ball-milled at 300 rpm for 5 h	82 (2016)
Sol–gel
MoS2/TiO2/hollow microfibers		Carboxymethyl cellulose, TiO2, and MoS2 were added to 100 mL alcohol; then, the mixture was stirred and sonicated for 2 h. The obtained paste was calcined in N2	84 (2017)
Hydrothermal/solvothermal assembly
3D MoS2/P25/graphene		Degussa P25 grade and a solution of MoS2 nanosheets were added to a GO dispersion. The suspension was hydrothermally treated at 180 °C for 12 h	85 (2014)

---

The ex situ synthetic strategy has many advantages, including low cost and scalable production. However, the methods require multiple complex and time-consuming steps to prepare raw MoS₂ and TiO₂ materials. Moreover, it is difficult to control the preparation of the composites; MoS₂ and TiO₂ are randomly dispersed, with weak interactions.

3.2 In situ synthetic strategy

For the in situ synthetic strategy, the synthesis process involves ionic reactions; this results in more uniform dispersion and complex interactions compared with the ex situ synthetic strategy. Therefore, the in situ synthetic strategy is receiving an increasing amount of attention. The in situ synthetic strategy for preparing MoS₂/TiO₂ composites can be divided into four groups: one-step synthesis (Fig. 4b), multi-step synthesis starting with MoS₂ (Fig. 4c), multi-step synthesis starting with TiO₂ (Fig. 4d), and multi-step synthesis with indirect methods (Fig. 4e). We will introduce the four groups in detail as follows.

3.2.1 One-step synthesis. As shown in Fig. 4b, the one-step synthesis method is simple and rapid. During the process, the Ti, Mo, and S sources are firstly mixed homogeneously; then, the precursor is converted into TiO₂/MoS₂ composites through a hydrothermal or solvothermal process.^88–93Table 4 summarizes the various preparation parameters of the one-step synthesis method. The key step of this method is the preparation of the homogeneous precursor, which is composed of Ti, Mo, and S sources. The used Mo sources, such as (NH₄)₂MoS₄, (NH₄)₆Mo₇O₂₄·4H₂O, and (NH₄)₂Mo₃S₁₃·2H₂O, and the S sources, such as CH₄N₂S and C₂H₅NS, are all water-soluble salts, whereas Ti sources such as Ti(OBu)₄, TiOP, and TiCl₄ are always water-insoluble or hydrolytic. Therefore, organic solvents such as ethanol and diethylene glycol or alkalis such as NaOH are added to stabilize the Ti, Mo, and S sources. The obtained precursors are further treated at low temperature (180 °C to 240 °C) for 12 to 24 h to prepare MoS₂/TiO₂ composites. In order to improve the crystallization degree of MoS₂, the hydrothermal products must be annealed at a high temperature in Ar atmosphere.

Table 4 A summary of MoS₂/TiO₂ based composites synthesized by in situ synthetic strategies with one-step growth

Composites	SEM/TEM images	Methods	Ti sources	Mo source/S sources	Additives	Conditions	Ref. (year)
TiOP is titanium isopropoxide, Ti(OBu)4 is tetrabutyl titanate, NMP is N-methyl-1-pyrrolidone, IPA is isopropanol. Reproduced with permission from: ref. 88. Copyright 2013, Springer; ref. 89. Copyright 2015, Royal Society of Chemistry; ref. 90. Copyright 2016, Royal Society of Chemistry; ref. 91. Copyright 2016, Elsevier; ref. 93. Copyright 2016, Elsevier; ref. 92. Copyright 2017, Royal Society of Chemistry.
MoS2/TiO2 nanoparticles		Solvothermal	Ti(OBu)4	(NH4)2MoS4	Diethylene glycol	200 °C for 12 h	88 (2013)
Nano-TiO2-decorated MoS2 nanosheets		Hydrothermal	TiCl4	(NH4)6Mo7O24·4H2O/CH4N2S	Ethanol	220 °C for 24 h	89 (2015)
Carbon doped TiO2–MoS2		Solvothermal	TiOP	(NH4)2Mo3S13·2H2O	NMP-IPA	180 °C for 20 h	90 (2016)
TiO2/MoS2		Hydrothermal	TiCl4	Na2MoO4·2H2O/C2H5NS	NaOH	240 °C for 24 h	91 (2016)
MoS2/CdS/TiO2 nanoparticles		Solvothermal	Ti(OBu)4	Na2MoO4·2H2O/CH4N2S	CdCl2·2.5H2O ethanol	200 °C for 24 h	93 (2016)
MoS2–TiO2		Hydrothermal	(NH4)2TiF6	Na2MoO4·2H2O/C2H5NS	Citric acid	180 °C for 24 h	92 (2017)

---

However, as with the ex situ strategy, it is still difficult to obtain uniform distribution of MoS₂ and TiO₂; this is because the nucleation and growth of MoS₂ and TiO₂ always occur simultaneously, although the interaction between MoS₂ and TiO₂ is stronger and more complex. For example, Sabarinathan et al.⁹² described the synthesis of MoS₂–TiO₂ mixture heterostructures via a one-step hydrothermal method in which 0.04 M Na₂MoO₄·2H₂O, 0.08 M C₂H₅NS, 0.04 M citric acid, and different amounts of (NH₄)₂TiF₆ were dissolved in 50 mL of deionized water and further maintained at 180 °C for 24 h. As shown in Fig. 5, the TEM images present a random distribution of MoS₂. The HR-TEM images show that the MoS₂ and TiO₂ in all the samples contact each other through good interfaces, demonstrating strong interaction between the two components. However, the SEM images show that the TiO₂ agglomerates into increasingly large particles from S2 to S6 as the content of the Ti source increases, suggesting that TiO₂ was nonuniformly dispersed in the random MoS₂. Therefore, despite the urgent need, it is still difficult to prepare uniformly distributed MoS₂/TiO₂ composite materials through a one-step synthesis strategy.

Fig. 5 (a, d, and g) FESEM, (b, e, and h) TEM and (c, f, and i) HRTEM images of samples S2, S4, and S6. S2, S4, and S6 contain 0.0015 M, 0.0050 M, and 0.015 M (NH₄)₂TiF₆, respectively. Reproduced with permission from ref. 92. Copyright 2017, Royal Society of Chemistry.

3.2.2 Multi-step synthesis: MoS₂ first. In this strategy, as shown in Fig. 4c, the MoS₂ is prepared in advance as the substrate for in situ growth of TiO₂; the resulting composites are TiO₂-coated MoS₂ heterostructures. This strategy requires one additional step compared with the one-step synthesis strategy, namely advance preparation of MoS₂; thus, it appears to be more complicated. However, actually, the multi-step method involves obviously different mechanisms of nucleation and growth because the MoS₂ acts as a template and also provides active sites for TiO₂ nucleation in this strategy. The synthetic methods of MoS₂ have been summarized systematically in many profound reviews to date;^48–50 thus, we only highlight the methods of decorating TiO₂ onto the surface of the MoS₂ substrate here. A great variety of methods for preparing pure TiO₂ materials have been reported.^{13,15–18,21} Particularly, decoration of TiO₂ onto the surface of MoS₂ is always performed at low temperature and under oxygen-deficit conditions because the MoS₂ substrate has excellent chemical stability in inert conditions but is prone to oxidization in oxygen-rich and relatively high temperature conditions. In this section, the reported synthetic strategy includes hydrothermal/solvothermal^75,94–99 and atomic layer deposition¹⁰⁰ methods. Table 5 summarizes the detailed reaction parameters and conditions.

Table 5 A summary of MoS₂/TiO₂ based composites synthesized by in situ synthetic strategies where the MoS₂ substrate is prepared in advance and then decorated with TiO₂

Composites	SEM/TEM images	Methods	Ti sources	Additives	Conditions	Ref. (year)
Ti(OBu)4 is tetrabutyl titanate, DEG is diethylene glycol, DMF is N,N-dimethylformamide, GO is graphene oxide, and P123 is PEG-PPG-PEG triblock copolymer. Reproduced with permission from: ref. 97. Copyright 2012, American Chemical Society; ref. 75. Copyright 2015, Springer; ref. 96. Copyright 2016, Institute of Physics Publishing; ref. 98. Copyright 2016, Elsevier; ref. 100. Copyright 2017, American Chemical Society.
TiO2/MoS2/graphene		Hydrothermal	Ti(OBu)4	GO, ethanol	180 °C for 10 h	97 (2012)
TiO2–MoS2(2H)		Hydrothermal	Ti(OBu)4	Ethanol, DMF	200 °C for 20 h	75 (2015)
TiO2/MoS2 nanosheets		Hydrothermal	Ti(OBu)4	DEG, HF	180 °C for 12 h	96 (2016)
TiO2/g-C3N4/MoS2		Solvothermal	TiCl4	g-C3N4, ethanol, glycerol, (NaPO3)6	140 °C for 3 h	98 (2016)
MoS2-1 nm–TiO2		Atomic layer deposition	TiCl4	N/A	75 °C at the speed of ∼0.8 Å per cycle	100 (2017)

---

3.2.2.1 Hydrothermal and solvothermal methods. The hydrothermal and solvothermal methods are conducted at low temperatures (≤200 °C) in a closed system with aqueous solution or organic solvent above ambient temperature and pressure. These methods have been widely used to decorate TiO₂ onto the surface of MoS₂.^75,94–99 The morphology, size, and content of TiO₂ in the obtained composites can be easily adjusted by controlling the reaction conditions. However, the morphologies of the reported composites mostly depend on their corresponding MoS₂ substrates. The bulk MoS₂ templates result in TiO₂ agglomeration and undispersed MoS₂/TiO₂ composites due to the insufficient growth sites of MoS₂.^94,95 Conversely, few-layered and well-dispersed MoS₂ substrates possess large specific surface areas with abundant active sites for TiO₂ decoration, leading to nano-TiO₂-coated MoS₂ nanosheet composites with homogeneous distribution and strong interaction.^75,96 For example, Bai et al. prepared few-layered MoS₂ nanosheets through a chemical exfoliation method; the MoS₂ and Ti source suspensions were then obtained in additional steps. Finally, the TiO₂–MoS₂(2H) hybrid was synthesized by in situ growth of TiO₂ nanocrystals on MoS₂ nanosheets through a hydrothermal process.⁷⁵ However, prefabricated MoS₂ nanosheets must always be dispersed in a particular solution to avoid agglomeration; namely, additional operations such as washing, centrifugation and drying are required to purify the MoS₂ nanosheets, which complicates the process. To address these situations, many researchers have introduced stable substrates with large specific surface areas, such as graphene oxide (GO)⁹⁷ and g-C₃N₄,^98,99 to prepare MoS₂/GO and MoS₂/g-C₃N₄ composites in advance instead of pure MoS₂ nanosheets. The MoS₂/GO and MoS₂/g-C₃N₄ nanosheets are readily prepared by hydrothermal methods and collected by filtration and drying. The thus-obtained powders can be readily used in the next hydrothermal or solvothermal process. Moreover, the large specific surface areas of GO and g-C₃N₄ are beneficial to improve the dispersity of MoS₂ and TiO₂, along with other unique properties. For example, Xiang et al. firstly reported the synthesis of TiO₂ nanoparticles on a layered MoS₂/graphene hybrid photocatalyst by a two-step hydrothermal process.⁹⁷ The TEM image (Table 5) of the resulting TiO₂/MoS₂/graphene composite shows that the MoS₂/graphene nanosheets serve as a novel support that is uniformly decorated with TiO₂ nanoparticles. Moreover, the superior electron mobility of the introduced graphene can greatly improve the photocatalytic performance. Therefore, these multi-component composites with controllable synthesis, homogeneous distribution, and unique properties are very important.
3.2.2.2 Atomic layer deposition method. Atomic layer deposition (ALD) has been widely used to prepare nano-TiO₂ covering layers. Recently, Ren and co-workers¹⁰⁰ firstly described the preparation of ultrathin TiO₂-modified MoS₂ nanosheet arrays by an ALD process at 75 °C in which TiCl₄ and H₂O were used as the Ti and O sources, respectively. The thickness of the TiO₂ layer was effectively controlled by controlling the number of depositions. It can be clearly seen that the morphology of TiO₂-modified MoS₂ nanosheet arrays are closely related to those of the MoS₂ template. The method enables the preparation of uniform TiO₂ particles or continuous coating layers on the MoS₂ surface. However, this method is still limited by low yields due to the low deposition rate and limited working thickness.

3.2.3 Multi-step synthesis: TiO₂ first. As shown in Fig. 4d, in this strategy, TiO₂ with various morphologies and properties is synthesized as a template for in situ growth of MoS₂; the resulting composites are MoS₂-coated TiO₂ heterostructures. This strategy is similar to that described above. The conceptual difference between the two strategies is the sequence of preparation of TiO₂ and MoS₂ in the MoS₂/TiO₂ composites. However, the TiO₂ substrate has better chemical stability and a greater variety of morphologies compared with the MoS₂ substrate. Therefore, MoS₂/TiO₂ composites can be prepared under more varied conditions by this strategy, and their morphologies can be readily controlled by adjusting the TiO₂ substrate. As a result, this strategy is more widely used for preparing MoS₂/TiO₂-based composites. As many extensive reviews have summarized the synthetic methods of TiO₂ in detail,^{13,15–18,21} we will not summarize the synthesis of TiO₂ substrate here; instead, we will focus on the methods of decorating MoS₂ onto the surface of TiO₂. Although many reviews^48–50 have summarized the synthetic methods of MoS₂, the TiO₂ substrate/template creates new challenges and possibilities under limited conditions. This synthetic strategy has involved hydrothermal or solvothermal techniques,^101–146 chemical bath deposition,¹⁴⁷ and photo-deposition^148,149 methods to date. Table 6 summarizes the various preparation methods of MoS₂/TiO₂-based composites where the TiO₂ substrate was prepared in advance followed by decoration with MoS₂.

Table 6 A summary of typical MoS₂/TiO₂ based composites synthesized by in situ synthetic strategies where the TiO₂ substrate is prepared in advance and then decorated with MoS₂

Composites	SEM/TEM images	Methods	TiO2 surface states	Mo sources/S sources	Additives	Conditions	Ref.
DMF is dimethyl formamide, DMAC is dimethylacetamide, CTAB is cetyltrimethyl ammonium bromide, P123 is PEO-PPO-PEO, PEG = poly(ethylene glycol), and PPG = poly(propylene glycol). Ref. 102, 122, 127, and 132. Copyright 2014, Royal Society of Chemistry; ref. 126 and 144. Copyright 2015, Royal Society of Chemistry; ref. 123 and 131. Copyright 2016, Royal Society of Chemistry; ref. 113, 116, 136, and 141. Copyright 2017, Royal Society of Chemistry; ref. 147. Copyright 2017, Hindawi; ref. 143. Copyright 2017, Springer; ref. 148. Copyright 2004, American Chemical Society; ref. 138. Copyright 2015, American Chemical Society; ref. 101. Copyright 2013, Wiley-VCH; ref. 107, 121, and 134. Copyright 2016, Wiley-VCH; ref. 118 and 120. Copyright 2017, Wiley-VCH; ref. 103, 115, and 137. Copyright 2015, Elsevier; ref. 105, 109, 124, 128, 129, 130, and 139. Copyright 2016, Elsevier; ref. 111, 115, 117, 145, and 149. Copyright 2017, Elsevier.
3.2.3.1 Hydrothermal and solvothermal methods
(a) TiO 2 surface state: rough TiO 2
TiO2@MoS2 nanobelts		Hydrothermal	Rough	Na2MoO4·2H2O/C2H5NS	N/A	200 °C for 24 h	101 (2013)
TiO2@MoS2 nanobelt		Hydrothermal	Rough	Na2MoO4·2H2O/CH4N2S	N/A	220 °C for 24 h	102 (2014)
TiO2@MoS2 nanofiber		Hydrothermal	Rough	Na2MoO4·2H2O/C2H5NS	N/A	220 °C for 24 h	103 (2015)
MoS2/TiO2 nanofiber		Hydrothermal	Rough	Na2MoO4·2H2O/L-cysteine	N/A	180 °C for 24 h	105 (2016)
TiO2@MoS2 nanotube array		Hydrothermal	Rough	Na2MoO4·2H2O/C2H5NS	N/A	200 °C for 24 h	107 (2016)
TiO2/MoS2 hollow sphere		Hydrothermal	Rough	Na2MoO4·2H2O/C2H5NS	N/A	220 °C for 24 h	109 (2016)
MoS2–TiO2 nanorod		Hydrothermal	Rough	Na2MoO4·2H2O/CH4N2S	N/A	220 °C for 24 h	111 (2017)
MoS2@hydrogenated-TiO2 nanobelt		Hydrothermal	Rough	Na2MoO4·2H2O/L-cysteine	N/A	200 °C for 24 h	113 (2017)
MoS2@TiO2 nanotube array		Hydrothermal	Rough	Na2MoO4·2H2O/C2H5NS	N/A	220 °C for 24 h	115 (2017)
MoS2/TiO2 cylinder		Hydrothermal	Rough	Na2MoO4/CH4N2S	N/A	200 °C for 24 h	116 (2017)
TiO2@MoS2 nanofiber		Hydrothermal	Rough	Na2MoO4·2H2O/C2H5NS	N/A	220 °C for 24 h	117 (2017)
TiO2@distorted 1T-MoS2 nanofibers		Hydrothermal	Rough	Na2MoO4·2H2O/C2H5NS	N/A	220 °C for 24 h	118 (2017)
(b) TiO 2 surface state: intermediate layer
TiO2@carbon@MoS2 nanotubes		Hydrothermal	TiO2@C prepared in advance	Na2MoO4·2H2O/CH4N2S	Glucose	200 °C for 24 h	120 (2017)
(c) TiO 2 surface state: lattice match
TiO2@MoS2 nano-onions		Hydrothermal	Ti0.87O2 precursor prepared instead[Editor31]	Na2MoO4·2H2O/C2H5NS	N/A	200 °C for 24 h	121 (2016)
(d) Reaction parameter: acidic conditions
MoS2-TiO2 nanowires		Hydrothermal	Rough	MoO3/KSCN	HCl	240 °C for 24 h	122 (2014)
TiO2-B/MoS2 nanowire arrays		Hydrothermal	Rough	MoO3/C2H5NS	HCl	220 °C for 20 h	123 (2016)
MoS2/TiO2/Si nanowires		Hydrothermal	Rough	Na2MoO4·2H2O/L-cysteine	HCl	180 °C for 12 h	124 (2016)
(e) Reaction parameter: organic binder
MoS2@TiO2 nanotubes		Solvothermal	Rough	Na2MoO4·2H2O/CH4N2S	Glucose	200 °C for 24 h	127 (2014)
TiO2@MoS2 nanowires		Hydrothermal	Rough	Na2MoO4·2H2O/CH4N2S	Glucose	200 °C for 24 h	126 (2015)
Carbon-coated TiO2@MoS2 nanosheets		Hydrothermal	Rough	Na2MoO4·2H2O/CH4N2S	Glucose	200 °C for 24 h	128 (2016)
TiO2@MoS2/C nanosheet arrays		Hydrothermal	Rough	Na2MoO4·2H2O/CH4N2S	Glucose	200 °C for 22 h	129 (2016)
(f) Reaction parameter: surfactant
TiO2@MoS2 microspheres		Hydrothermal	Smooth	Na2MoO4·2H2O/CH4N2S	CTAB	180 °C for 36 h	130 (2016)
MoS2/TiO2/Ti nanosheet arrays		Solvothermal	Smooth	Na2MoO4·2H2O/CH4N2S	P123	230 °C for 24 h	131 (2016)
(g) Reaction parameter: organic solvent
MoS2/r-GO/P25		Solvothermal	Two-phase	Na2MoO4·2H2O/C2H5NS	GO, DMAC	150 °C for 10 h	132 (2014)
TiO2/MoS2 nanospheres		Solvothermal	Rough with mesopores	(NH4)2MoS4	DMF	180 °C for 24 h	134 (2016)
MoS2@TiO2 microspheres		Solvothermal	Rough with mesopores	Na2MoO4·2H2O/L-cysteine	Ethanol	220 °C for 24 h	136 (2017)
(h) TiO 2 nanosheets with exposed {001} facets
MoS2/TiO2{001} nanosheets		Hydrothermal	F-ion	Na2MoO4·2H2O/C2H5NS	N/A	210 °C for 24 h	137 (2015)
2D–2D MoS2/TiO2 nanosheets		Hydrothermal	F-ion	Na2MoO4·2H2O/CH4N2S	N/A	200 °C for 24 h	138 (2015)
MoS2/TiO2(001) nanosheets		Hydrothermal	F-ion	Na2MoO4·2H2O/CH4N2S	N/A	200 °C for 24 h	139 (2016)
MoS2/TiO2 nanosheets		Hydrothermal	F-ion	Na2MoO4·2H2O/CH4N2S	N/A	180 °C for 24 h	141 (2017)
MoS2/TiO2 nanosheets		Hydrothermal	F-ion	Na2MoO4·2H2O/C2H5NS	N/A	200 °C for 24 h	143 (2017)
(i) TiO 2 nanosheets with exposed {001} facets + organic binder
MoS2@TiO2 nanosheets		Hydrothermal	F-ion	Na2MoO4·2H2O/CH4N2S	Glucose	200 °C for 24 h	144 (2015)
G/TiO2@C/MoS2 nanosheets		Hydrothermal	F-ion with GO	Na2MoO4·2H2O/CH4N2S	Glucose	200 °C for 24 h	145 (2017)
3.2.3.2 Chemical bath deposition method
MoS2/TiO2 film		Chemical bath deposition	Rough	(NH4)6Mo7O24·4H2O/CH4N2S	N2H4·H2O	80 °C for 45 min	147 (2017)
3.2.3.3 Photoinduced deposition method
MoS2/P25	N/A	Photodeposition	Two-phase	(NH4)2MoS4	N2, N2H4·H2O	Irradiated with a 300 W Xe lamp	148 (2004)
MoS2/CdS-TiO2		Photodeposition	Rough	(NH4)2MoS4	N2	Irradiated with a 300 W Xe lamp	149 (2017)

---

3.2.3.1 Hydrothermal and solvothermal methods. Hydrothermal and solvothermal methods are bottom-up strategies to prepare MoS₂ nanosheets using Mo ions and S ions. Meanwhile, the TiO₂ substrate/template shows excellent chemical stability in these conditions. Therefore, hydrothermal and solvothermal methods are the most commonly used to synthesize MoS₂ on the surface of TiO₂. Generally, as shown in Fig. 6, this method contains two steps: nucleation and growth. Of these, the nucleation step plays a crucial role. The added TiO₂ in the reaction solution acts as the substrate or template for MoS₂ nucleation, which is called heterogeneous nucleation.¹⁵⁰ It is found that the surface states (crystal lattice and surface) of the TiO₂ substrate/template have strong effects on the heterogeneous nucleation of MoS₂ on the TiO₂ substrate or template. After successful nucleation, the crystallization of MoS₂ hetero-epitaxy occurs on the TiO₂ substrate/template. The size and density of these MoS₂ nanosheets can be readily controlled by adjusting the reaction parameters, such as growth time, growth temperature, initial reactant concentration, pH value, and additives. Since 2013, when Zhou et al.¹⁰¹ first synthesized a few-layered MoS₂ nanosheet-coated TiO₂ nanobelt composite via a hydrothermal method at 200 °C for 24 h using Na₂MoO₄·2H₂O as the Mo source and C₂H₅NS as the S source, a large number of MoS₂/TiO₂-based composites have been reported to be fabricated using this method. There has been an upsurge of research interest in the controlled synthesis of MoS₂-coated TiO₂ composites with various morphologies.^101–142 As mentioned above, the hydrothermal or solvothermal controllable synthesis of MoS₂/TiO₂-based composites are determined by two aspects: TiO₂ surface states and reaction parameters. Various MoS₂/TiO₂-based composites have been synthesized by adjusting controllable factors, which will be extensively discussed below. The TiO₂ surface states directly influence nucleation and growth. Among these, a small lattice mismatch (<5%) between two compounds with similar crystal structures can promote epitaxial nucleation and growth.¹⁵⁰ In general, there are four different crystallographic forms of TiO₂ (Table 1) and three different crystallographic forms of MoS₂ (Table 2). Among these forms, anatase TiO₂ ⁷⁹ and 2H-MoS₂ ⁶⁴ are the most stable and common crystal phases under hydrothermal synthesis conditions. For anatase TiO₂, the natural and artificial crystal exposes the most stable (101) plane (rectangular cell: 5.48 Å × 3.78 Å).¹²¹ Meanwhile, anatase TiO₂ nanosheets with exposed (001) planes (hexagonal cell: 3.79 Å × 3.79 Å) have gained considerable attention and great research interest due to their novel properties.^151–155 2H-MoS₂, which is a typical layered structure material, shows a dominant (002) surface (hexagonal cell: 2.41 Å × 2.41 Å).¹²¹ The unit structures are shown in Fig. 7. The lattice of the 2H-MoS₂ (002) plane is elongated by 36.2% and 5.6% to match the anatase (101) plane and (001) plane, respectively. The large lattice mismatch between MoS₂ and TiO₂, especially common anatase TiO₂ with exposed (101) planes, impedes the epitaxial nucleation and growth of MoS₂ nanosheets on the TiO₂ substrate/template. Meanwhile, the TiO₂ substrate is always negatively charged, and the commonly used Mo sources are Mo-containing anions such as MoO₄²⁻ and MoS₄²⁻; thus, the charge incompatibility between negatively charged TiO₂ and Mo-containing anions causes them to repel each other. As a result, the formed MoS₂ nanosheets tend to assemble into microspheres instead of coating the TiO₂ surface (Fig. 8). To address this issue, many studies have explored the abovementioned two aspects: TiO₂ surface states and reaction parameters.

Fig. 6 Schematic of the nucleation and growth of MoS₂ nanosheets on a TiO₂ template. Reproduced with permission from ref. 103. Copyright 2015, Elsevier.

Fig. 7 Top views of projected structures of (a) 2H-MoS₂ (002) plane, (b) anatase TiO₂ (101) plane and (c) anatase TiO₂ (001) plane. Reproduced with permission from: (a, b) ref. 121. Copyright 2016, Wiley-VCH; (c) ref. 155. Copyright 2014, American Chemical Society.

Fig. 8 Growth of MoS₂ nanosheets on the smooth surface of untreated TiO₂ nanobelts. Reproduced with permission from ref. 101. Copyright 2013, Wiley-VCH.

3.2.3.1 (a) TiO ₂ surface state: rough TiO ₂ . To modify the surface state of TiO₂, the surface roughness of the substrate/template is a key driving force of the nucleation. Therefore, many studies have introduced rough TiO₂ as the substrate to decorate MoS₂ nanosheets, as shown in Table 6. For example, Zhou and co-workers¹⁰¹ added an acid treatment process prior to the calcination of H₂Ti₃O₇ nanobelts to obtain a rough TiO₂ surface, providing high energy nucleation sites for the nucleation and growth of MoS₂ nanosheets. As a result, few-layered MoS₂ nanosheet-coated TiO₂ nanobelt heterostructures were successfully prepared (see Table 6). Liu et al.¹⁰³ obtained porous TiO₂ nanofibers after removal of PVP by calcination of the electrospinning precursor; MoS₂ nanosheets were grown on the resulting porous TiO₂ nanofibers via a simple hydrothermal process (see Table 6). This modification strategy has been widely explored and studied;^101–119 it is key to obtain TiO₂ with sufficient surface roughness to overcome the obstacle of lattice mismatch between TiO₂ and MoS₂. The high surface roughness of TiO₂ can facilitate the growth of MoS₂ nanoparticles to serve as seeds for later growth of MoS₂ nanosheets.¹⁵⁰ However, the processes for obtaining high surface roughness are time-consuming and complex.

3.2.3.1 (b) TiO ₂ surface state: intermediate layer. In addition, introducing an intermediate layer between TiO₂ and MoS₂ is an effective method for the growth of MoS₂ nanosheets on the surface of Ti₂O. The intermediate layer should serve as a bonding interface to create strong adhesion between MoS₂ and TiO₂. Generally, carbon materials, especially amorphous carbon with abundant functional groups, can readily coat TiO₂ ^156–158 and support MoS₂ ^159–161 with strong chemical bonds. Therefore, amorphous carbon can be used as an ideal intermediate layer material in this process. Recently, as shown in Fig. 9a, Wang and co-workers¹²⁰ reported the synthesis of hierarchical tubular structures by sequential coating of a nitrogen-doped carbon (NC) layer and MoS₂ nanosheets on TiO₂ nanotubes (designated as TiO₂@NC@MoS₂). According to Fig. 9b and c, we can observe that the NC layer tightly coats the TiO₂ surface in TiO₂@NC. The prefabricated TiO₂@NC is then used to grow MoS₂ nanosheets. After the hydrothermal and calcination processes, as shown in Fig. 9d and e, the NC intermediate layer acts as a bonding interface to maintain good contact between the TiO₂ nanotubes and MoS₂ nanosheets in the obtained TiO₂@NC@MoS₂; this is a very stable architecture. The NC intermediate layer can also increase electrical conductivity, which is beneficial to rechargeable batteries. In contrast, the carbon intermediate layer may impede the delivery of holes between MoS₂ and TiO₂, which is detrimental to photocatalytic applications. Therefore, the introduced carbon intermediate layer requires careful design according to specific applications. This modification strategy is facile and effective. However, it requires multiple steps; thus, it is time-consuming and complex.

Fig. 9 (a) Schematic of the synthesis process of TiO₂@NC@MoS₂ tubular nanostructures. (I) TiO₂ coating, (II) PDA coating, (III) annealing in N₂ and acid etching, and (IV) growth of MoS₂ nanosheets. (b, d) FESEM and (c, e) TEM images of (b, c) TiO₂@NC nanotubes and (d, e) annealed TiO₂@NC@MoS₂ nanotubes. Reproduced with permission from ref. 120. Copyright 2017, Wiley-VCH.

3.2.3.1 (c) TiO ₂ surface state: lattice match. Although the above two modification strategies can effectively be used to grow MoS₂ nanosheets on the surface of TiO₂, core–shell architectures cannot be successfully prepared. Because the significant lattice mismatch between TiO₂ and MoS₂ remains, a dendritic/dumbbell-like morphology is more likely to form by the above two modification strategies.^101–120 In order to prepare core–shell nanostructures with high-quality interfaces, the as-exposed TiO₂ surface should be well designed and adjusted. Recently, Dai et al.¹²¹ used unilamellar Ti_0.87O₂ nanosheets as a starting material and subsequently epitaxially grew MoS₂ on TiO₂. The formation of these core–shell nano-onions is attributed to an amorphous layer-induced hetero-growth mechanism. As shown in Fig. 10, during the hydrothermal process, the Ti_0.87O₂ nanosheets recrystallized and transformed into distorted anatase nanocrystals.¹²¹ The abundant Ti vacancies may lead to the rearrangement of Ti and O atoms during the phase transformation process, thus forming an amorphous layer outside the anatase nucleus. This amorphous layer significantly facilitates the nucleation of MoS₂ on the anatase nanocrystals. Subsequently, MoS₂ grains continue to grow epitaxially on the amorphous surface of the anatase nanoparticles. The amorphous layer at the anatase TiO₂/MoS₂ interface gradually disappears with prolonged hydrothermal reaction time. Finally, TiO₂@MoS₂ nano-onions with a high-quality interface are obtained (see Table 6) despite the large lattice mismatch between TiO₂ and MoS₂.¹²¹ The obtained TiO₂@MoS₂ nano-onions show superior structural stability when they are evaluated as LIB anodes.

Fig. 10 Schematic of the proposed formation process for TiO₂@MoS₂ nano-onions. Reproduced with permission from ref. 121. Copyright 2016, Wiley-VCH.

3.2.3.1 (d) Reaction parameter: acidic conditions. Apart from the TiO₂ surface states, the reaction parameters are also important for the growth of MoS₂. Among these, the pH value is an important factor in hydrothermal processes. When slightly soluble MoO_x is used as the Mo source, the MoO_x converts to MoS₂ by itself and does not coat onto TiO₂. Therefore, some researchers have added inorganic acid to the solution to convert MoO_x to Mo ions upon hydrothermal treatment; the MoS₂ then nucleates and grows on the rough TiO₂ surface.^122,123 TiO₂ and MoS₂ show high structural stability under weakly acidic conditions. Moreover, the TiO₂ surface can provide more high-energy nucleation sites in acidic conditions upon hydrothermal treatment.^101,102 Therefore, weakly acidic conditions can further facilitate the nucleation and growth of MoS₂ on the surface of TiO₂, leading to MoS₂-coated TiO₂ hierarchical structures with uniform coverage.^124,125

3.2.3.1 (e) Reaction parameters: organic binder. In addition, the additive also plays a key role in the hydrothermal process. Glucose has been widely used as an organic binder in the preparation of uniform MoS₂-based composites due to its abundant hydroxyl groups.^162–164 Herein, some researchers used glucose to promote the growth of MoS₂ nanosheets on the surface of TiO₂.^126–129 For example, Li and co-workers¹²⁶ prepared TiO₂ nanowires with rough surfaces (Fig. 11a) as a substrate; then, the rough TiO₂ nanowires were added to two solutions for a hydrothermal reaction, where one solution contained Na₂MoO₄ and CN₂H₄S and the other contained excess glucose. The two solutions were transferred into stainless steel autoclaves and maintained at 200 °C for 24 h. After the hydrothermal reaction, as shown in Fig. 11b, although the TiO₂ possessed a rough surface, few MoS₂ nanosheets were coated onto the surface of TiO₂ without the assistance of glucose. Most of the MoS₂ sheets aggregated into thick flakes, probably due to insufficient roughness. Upon the addition of glucose, the MoS₂ coated uniformly on the surface of TiO₂ with smaller size and thickness, as shown in Fig. 11c. Therefore, glucose promoted the growth of MoS₂ nanosheets on the surface of TiO₂ with uniform coverage. In addition, MoS₂ nanosheets can also grow on rough anatase TiO₂ nanosheets with exposed (001) faces with the assistance of glucose.¹²⁸ We have explored the reaction mechanism of glucose-assistance in our previous work.¹²⁸ As shown in Fig. 12a, the glucose molecules can be absorbed on the surface of rough ultrathin TiO₂ (UT-TiO₂) and simultaneously tightly adsorb MoO₄²⁻via their abundant hydroxyl groups. Then, the glucose is carbonated on the surface of UT-TiO₂, and MoS₂ nuclei form on the carbon surface during the hydrothermal process in a relatively short reaction time of 8 h (Fig. 12b). MoS₂ grows homogeneously and finally forms few-layered 2D interconnected MoS₂ networks in 24 h (Fig. 12c). In summary, both a certain degree of roughness of TiO₂ and glucose are crucial for the formation of uniform TiO₂@MoS₂ nanocomposites. The rough surface of TiO₂ can provide accessible active sites for the adsorption of glucose molecules. The adsorbed glucose can adsorb large Mo ions tightly via its abundant hydroxyl groups. Therefore, the glucose acts as a binder to promote the growth of MoS₂ nanosheets on TiO₂ with uniform coverage.

Fig. 11 FESEM images of (a) rough TiO₂ and TiO₂–MoS₂ composites (b) without and (c) with the addition of glucose. Reproduced with permission from ref. 126. Copyright 2015, Royal Society of Chemistry.

Fig. 12 Schematic of the synthesis of UT-TiO₂/C@DR-MoS₂. (a) Homogeneous solution containing UT-TiO₂, C₆H₁₂O₆, Na₂MoO₄ and CN₂H₄S. (b) UT-TiO₂/C/MoS₂-8 h and (c) UT-TiO₂/C@DR-MoS₂ obtained with hydrothermal reaction times of 8 h and 24 h, respectively. Reproduced with permission from ref. 128. Copyright 2016, Elsevier.

3.2.3.1 (f) Reaction paraments: surfactant. The surfactant has an important impact on the interface state of the solution system; thus, it can effectively control the nucleation and growth. In general, the most commonly used Mo sources in hydrothermal and solvothermal processes are Na₂MoO₄·2H₂O, (NH₄)₆Mo₇O₂₄·4H₂O, and (NH₄)₂MoS₄ inorganic salts (see Table 6). They form Mo-containing anions when they are dissolved in water. Therefore, a cationic surfactant is helpful to increase the interaction between TiO₂ and negatively charged Mo-containing species. Among these, cetyltrimethyl ammonium bromide (CTAB) is a well-known cationic surfactant which has also been widely used in the preparation of MoS₂-based composites by hydrothermal methods.^165–167 Recently, Xu et al.¹³⁰ added 0.5 wt% CTAB surfactant to a reaction solution containing Na₂MoO₄·2H₂O, C₂H₅NS, and smooth TiO₂ microspheres. As shown in Fig. 13, the CTAB is adsorbed on the negatively charged TiO₂ surface due to electrostatic interactions. Due to the adsorption of CTA⁺ on the surface of TiO₂, the surface is positively charged and can further absorb MoO₄²⁻ anions without repelling them. During the hydrothermal process, the MoO₄²⁻ is reduced to MoS₂ nanosheets in situ by H₂S released by C₂H₅NS on the TiO₂ microspheres. In conclusion, cationic surfactants can effectively aid MoS₂ nucleation and growth on the TiO₂ surface to form TiO₂@MoS₂ composites. Moreover, Zhou and co-workers¹³¹ reported vertically aligned MoS₂ nanosheets on the surface of TiO₂ with the assistance of nonionic surfactant P123. Therefore, the additition of surfactant is beneficial for controlling the morphology of MoS₂/TiO₂ composites; this requires further exploration and research.

Fig. 13 Schematic of the synthesis process of core–shell TiO₂@MoS₂ composites. Reproduced with permission from ref. 130. Copyright 2016, Elsevier.

3.2.3.1 (g) Reaction parameters: organic solvent. The solvent is also important in the preparation of MoS₂/TiO₂ composites. Generally, the solvent dissolves the Mo and S sources and TiO₂ substrate to prepare a homogeneous precursor solution. Water is widely used as a solvent to dissolve soluble inorganic salts. Meanwhile, it is also necessary to add organic solvent to dissolve some organic Mo and S sources or inhibit hydrolysis under certain conditions.^132–136 For example, Zhao et al.¹³³ added ethanol and oleic acid to water to inhibit MoCl₅ hydrolysis, and the uniform solution ensures the successful preparation of TiO₂@MoS₂ composites (see Table 6). Liu and co-workers¹³⁵ used sulfur powder, which cannot be dissolved in water, as a S source. Thus, the authors chose ethanol and octylamine as solvents to dissolve the sulfur power. As a result, MoS₂/TiO₂ composites were successfully prepared.

3.2.3.1 (h) TiO ₂ nanosheets with exposed {001} facets. 2D anatase TiO₂ nanosheets with exposed {001} facets have attracted especially great interest in photocatalysis and rechargeable batteries due to their unique properties. Therefore, many researchers have devoted much effort to using 2D TiO₂ nanosheets as a substrate to grow 2D MoS₂ nanosheets.^137–143 Herein, we also summarized the preparation strategy and mechanism of 2D–2D MoS₂/TiO₂ nanosheets. Generally, TiO₂ nanosheets are prepared by employing a fluorine morphology controlling agent; thus, the obtained exposed TiO₂ {001} facets are covered by ≡Ti–F bonds. This unique surface state has an important impact on the nucleation and growth of the MoS₂ nanosheets. The ≡Ti–F bonds on the surface are unstable during hydrothermal treatment,¹⁶⁸ and they tend to be replaced by hydroxyl groups (eqn (1)).¹⁶⁹ Therefore, the obtained ≡Ti–OH bonds can absorb anionic Mo sources, such as MoO₄²⁻ and MoS₄²⁻,¹⁷⁰ which are then reduced in situ by the S source to form 2D–2D MoS₂/TiO₂ nanosheets. The TEM images in Table 6 show the 2D–2D structures of the obtained nanosheets.

≡Ti–F + H₂O → ≡Ti–OH + H⁺ + F˙⁻ (1)

However, the number of transformed hydroxyl groups on the surface of the TiO₂ nanosheets is limited by the specific surface area of TiO₂. Generally, the reported specific surface area of TiO₂ nanosheets is around 50 m² g⁻¹,^137–143 which is significantly lower than that of the carbon substrate. Therefore, only small amounts of MoS₂ nanosheets can be loaded on the TiO₂ nanosheet surface (see Table 6). The loading amounts of MoS₂ are 0.73 wt%, 2 wt%, 5 at%, 1 wt%, 0.5 wt%, 5 at% and 5 wt% in ref. 137, 138, 139, 140, 141, 142, and 143, respectively. The small amounts of MoS₂ nanosheets in MoS₂/TiO₂ nanosheets severely limit their application in rechargeable batteries because the MoS₂ is necessary to provide high capacity.

3.2.3.1 (i) TiO ₂ nanosheets with exposed {001} facets + organic binder. In order to increase the amount of MoS₂ in 2D–2D MoS₂/TiO₂ nanosheets, in our previous studies, we firstly introduced the organic binder glucose to promote the nucleation and growth of MoS₂ on the surface of TiO₂ nanosheets.^144,145 As shown in Fig. 14, the flourinated-TiO₂ (F-TiO₂) nanosheets can capture glucose molecules with electrostatic interactions. The adsorbed glucose with abundant hydroxyl groups can then absorb a large amount of MoO₄²⁻.¹⁷⁰ During the hydrothermal process, the ≡Ti–F bonds are removed, the glucose is carbonized in situ on the TiO₂ surface, and the MoS₂ nanosheets are formed in situ on the amorphous carbon. Finally, 3D few-layered MoS₂-coated TiO₂ nanosheet (3D FL-MoS₂@TiO₂) core–shell nanocomposites were prepared with a relatively large ratio of 47 wt% for MoS₂. This stable core–shell TiO₂@MoS₂ composite with 47 wt% MoS₂ shows excellent electrochemical performance as a LIB anode. We have also explored the relationship between glucose addition and the prepared structure.¹⁴⁵ The result shows that an appropriate amount of glucose is essential for obtaining ideal composites in which the surface of TiO₂ is entirely covered with the grown MoS₂ shell.

Fig. 14 Schematic of the fabrication of 3D FL-MoS₂@TiO₂ using F-TiO₂ as template and C₆H₁₂O₆ as binder. (a) F-TiO₂. (b) C₆H₁₂O₆-coated F-TiO₂. (c) C₆H₁₂O₆-coated F-TiO₂ adsorbed by MoO₄²⁻. (d) Hydrothermal 3D FL-MoS₂@TiO₂ before thermal treatment. (e) 3D FL-MoS₂@TiO₂. Reproduced with permission from ref. 144. Copyright 2015, Royal Society of Chemistry.

Moreover, we expanded this strategy to prepare core-double shell carbon-supported MoS₂ nanosheet-coated P25 (P25@C@MoS₂) composite using fluorinated P25 and glucose binder.¹⁴⁶

In summary, hydrothermal and solvothermal methods with TiO₂ as a substrate/template are effective and excellent methods for preparing core–shell TiO₂@MoS₂ composites. The synthetic process and structure can be carefully controlled by adjusting the TiO₂ surface states and reaction parameters for different applications. Moreover, the shapes and dimensionalities of the resulting MoS₂/TiO₂-based composites are determined by those of the TiO₂ substrate/template. Various morphologies, such as nanowires, nanofibers, nanobelts, nanotubes, nano-onions, cylinders, flowers and micro-spheres, have been synthesized to date (see Table 6).

3.2.3.2 Chemical bath deposition method. Compared with hydrothermal and solvothermal methods, chemical bath deposition is a common low-temperature aqueous technique for depositing thin MoS₂/TiO₂ films with large areas.¹⁷¹ Recently, Phung and co-workers¹⁴⁷ used chemical bath deposition to prepare MoS₂/TiO₂ heterostructure thin films. Firstly, TiO₂ thin films were prepared on glass substrates using a dip-coating process. Secondly, the prepared TiO₂ thin films were immersed into a solution containing (NH₄)₆Mo₇O₂₄·4H₂O, CH₄N₂S, and N₂H₄·H₂O. The bath was maintained at 80 °C for different deposition times. Finally, MoS₂ was well deposited on the surface of the TiO₂ films to fabricate MoS₂/TiO₂ heterostructure films.
3.2.3.3 Photoinduced deposition method. Photodeposition is also a simple and economical technique for depositing inorganic materials.¹⁷² In 2004, Ho et al.¹⁴⁸ firstly used photoinduced deposition to prepare nanosized MoS₂-decorated TiO₂ composites. As shown in Fig. 15, a 300 W high-pressure mercury lamp was used as a UV light source, and hydrazine was added to react with the positively charged holes. The electrons photogenerated in situ converted the MoS₄²⁻ anions to MoS₂ directly on the surface of the TiO₂ particles under UV light irradiation. Notably, it is worth pointing out that MoS₂ exhibits quantum-sized nanoclusters due to the mild reaction environment. The quantum-sized MoS₂ nanoclusters can significantly improve the visible light activity of photocatalysis. Moreover, this method can be extended to prepare WS₂/TiO₂ composites using WS₄²⁻ anions.¹⁴⁸ The reaction mechanism of this method is based on photocatalysis.

Fig. 15 Proposed model of in situ photoinduced fabrication of MS₂ (M = W and Mo)-sensitized TiO₂. Reproduced with permission from ref. 148. Copyright 2004, American Chemical Society.

3.2.4 Multi-step synthesis: indirect method. In the above in situ synthesis methods starting with TiO₂, the MoS₂ directly nucleates and then grows on the TiO₂ surface. The synthesis process is mostly controlled by the nucleation. However, the large lattice mismatch between MoS₂ and TiO₂ impedes the nucleation; this is considered to be the most major issue of this technique. In order to address this issue, some researchers have proposed using indirect methods to avoid direct nucleation of MoS₂ on the TiO₂ substrate. The methods can be subdivided into two groups: preparing the MoO_x/TiO₂ composite in advance followed by sulfuration,^173–176 and preparing the MoS_x/TiO₂ composite in advance followed by reduction.^177,178 MoO_x and MoS_x probably have low lattice mismatch with MoS₂; therefore, MoO_x/TiO₂ and MoS_x/TiO₂ can be prepared more readily. More interestingly, the indirect method shows a unique growth mechanism compared with direct methods. Table 7 summarizes the synthesis parameters.

Table 7 A summary of MoS₂/TiO₂-based composites synthesized by in situ synthetic strategies where MoO_x/TiO₂ or MoS_x/TiO₂ precursors are prepared in advance and then transformed into MoS₂/TiO₂ composites

Composite	SEM/TEM image	Methods	TiO2 surface states	Mo source/S source^b	Additive^b	Conditions^b	Ref. (year)
a Represents the first step. b Represents the second step.
3.2.4.1 MoO x /TiO 2 precursors first
MoS2–TiO2 nanofibers		Two-step: impregnation^a, thermal sulfuration^b	MoOx/TiO2 precursor prepared in advance	N/A/H2S	N/A	500 °C for 2 h	173 (2015)
MoS2/TiO2 nanotube arrays		Two-step: sputtering^a, thermal sulfuration^b	MoO3/TiO2 precursor prepared in advance	N/A/H2S	N/A	500 °C for 30 min	174 (2016)
TiO2/MoS2(E) nanowires		Two-step: impregnation^a, thermal sulfuration^b	MoO2/TiO2 precursor prepared in advance	N/A/S	N/A	400 °C for 1 h under vacuum	175 (2016)
Flower-like MoS2/TiO2		Two-step: hydrothermal^a, hydrothermal sulfuration^b	MoO2/TiO2 precursor prepared in advance	N/A/CH4N2S	N/A	200 °C for 48 h	176 (2017)
3.2.4.2 MoS x /TiO 2 precursors first
Nano-MoS2/TiO2		Two-step: chemical bath deposition^a, thermal reduction^b	MoS3/TiO2 precursor prepared in advance	N/A	H2	450 °C for 30 min	177 (2010)
MoS2/P25		Two-step: deposition^a, thermal reduction^b	MoSx/TiO2 precursor prepared in advance	N/A	H2	350 °C	178 (2017)

---

3.2.4.1 MoO_x/TiO₂ precursors first. In the direct methods described above, the MoS₂/TiO₂ composites are prepared by reaction between Mo ions and S ions on TiO₂ substrates. However, for the indirect method, as shown in Fig. 4e, the MoS₂/TiO₂ composites are formed by the transformation of solid MoO_x/TiO₂ precursors. The solid transformation mechanism in the indirect method is obviously different from the growth mechanism with ionic reactions in the direct method. For example, He et al.¹⁷⁵ reported a unique heterostructure with MoS₂ standing on TiO₂ along its edges (named TiO₂/MoS₂(E)). As shown in Fig. 16, the prefabricated TiO₂/MoO₂ composite was sulfurized in a chemical deposition system. During the sulfurizing process, the sulfurization began on the surface of MoO₂ and then diffused to the deep interior. The sulfur diffusion occurred more rapidly parallel to the basal plane because of the weak interaction between the MoS₂ layers.¹⁷⁹ Therefore, the growth direction is the same as the sulfur diffusion direction, from the surface of MoO₂ to the interface between MoO₂ and TiO₂. The interface between MoO₂ and TiO₂ is the last area to be sulfurized; thus, the lattice match rule is avoided. Finally, connection of the MoS₂ nanoplates with TiO₂via edges was achieved for the first time. Moreover, compared to the direct methods described above, the effects of surface energy on the growth process of MoS₂ are suppressed. Although the indirect method requires multiple steps, the unique formation mechanism is interesting and powerful and can offer a new synthesis strategy for fine structures and interface designs.

Fig. 16 Schematic of the sulfurization process. The inset TEM images correspond to TiO₂/MoO₂ and TiO₂/MoS₂(E)3%, respectively. Reproduced with permission from ref. 175. Copyright 2016, Wiley-VCH.

3.2.4.2 MoS_x/TiO₂ precursors first. Similar to the above method, the MoS₂/TiO₂ composites are fabricated by the transformation of solid MoS_x/TiO₂ precursors. For example, Girel et al.¹⁷⁸ prepared MoS_x/TiO₂ solids in advance by aqueous deposition; then, the obtained precursors were treated under H₂ flow at 350 °C to reduce the precursors to MoS₂/TiO₂ composites.

4. Applications and enhancement strategies

4.1 Photocatalysis

MoS₂/TiO₂ composites have been extensively studied as photocatalysts for photocatalytic degradation, hydrogen evolution, and CO₂ reduction. In general, TiO₂ and MoS₂ serve as catalyst and cocatalyst, respectively. As shown in Fig. 17, under light irradiation, TiO₂ is excited to generate electron–hole pairs with photon energy higher than its band gap energy (E_g). Unfortunately, the excited electrons (e⁻) and holes (h⁺) are unstable and can easily recombine. The loaded cocatalyst MoS₂ can effectively accept photogenerated charge carriers to reduce recombination. Moreover, the MoS₂ cocatalyst can provide large catalytically active sites for photocatalytic progress. The separated holes can be used as oxidants for degradation of organic pollutants, while the separated electrons can be used as reductants for hydrogen evolution and CO₂ reduction (Fig. 17).

Fig. 17 Schematic of organics degradation, photocatalytic hydrogen evolution, and CO₂ photoreduction over a TiO₂ photocatalyst loaded with MoS₂ cocatalyst. VB is the valance band, and CB is the conduction band.

4.1.1 Photocatalytic degradation. Degradation of pollutants in water using solar energy is an attractive technology for environmental protection. MoS₂/TiO₂ composites are widely used for the photocatalytic degradation of organic pollutants, such as RhB,^{77,84,104,105,107,124,132,134,137} MB,^{81,86,87,92,117,139,143,147,148} MO,^{82,85,94,95,98,106,114} direct green 6⁷⁶ 4-ATP,⁹⁰ Rs,⁹³ and phenol.¹⁰⁹ According to previous reports,^86,92,148 the generated electron–hole pairs (eqn (2)) are formed under light irradiation. Then, the photogenerated holes react with water molecules on the surface of the composites, forming hydroxyl radicals (˙OH) (eqn (3)). The holes and obtained ˙OH can oxidize the organic molecules directly on the surface of the composites. In addition, the generated electrons can react with O₂ molecules in air to form superoxide radical anions (˙O₂⁻) (eqn (4)); then, ˙O₂⁻ can transform into ˙OH (eqn (5)–(7)) via complex and selective steps that are difficult to achieve. Therefore, the photogenerated holes and resulting ˙OH are the main oxidants for the degradation of organic pollutants. Table 8 summarizes the reported MoS₂/TiO₂-based photocatalysts for pollutant degradation.

MoS₂/TiO₂ + hv → MoS₂ (h⁺)/TiO₂ (e⁻) (2)

h⁺ + H₂O → {˙OH} + H⁺ (3)

e⁻ + O₂ → {˙O₂⁻} (4)

˙O₂⁻ + H₂O → {˙HO₂} + OH⁻ (5)

˙HO₂ + H₂O → H₂O₂ + {˙OH} (6)

H₂O₂ → 2˙OH (7)

Table 8 A summary of MoS₂/TiO₂-based photocatalysts for pollutant degradation

Materials/(MoS2 wt%)	Pollutant content/(mg L^−1)	Light source	Activity	Photoactivity enhancement	Ref. (year)
RHB is rhodamine B, MB is methylene blue, MO is methyl orange, 4-ATP is 4-aminothiophenol, Rs is resorcinol, TC is tetracycline, DP is degradation percentage, K is rate constant, Kapp is average rate constant.
Pollutant: RhB
MoS2/TiO2{001} nanosheets/0.73%	10	300 W Xe lamp λ > 400 nm	DP of over 98.3% in 100 min	1.35 × >TiO2	137 (2015)
TiO2–MoS2 nanobelts/40%	10	500 W Xe lamp visible light	K 0.0239 min^−1	4.78 × >TiO2	104 (2015)
MoS2/r-GO/P25	10	150 W Xe lamp simulated sunlight	DP of over 80% in 80 min	4 × >P25	132 (2015)
MoS2/P25/0.1%	10	Simulated sunlight	K 0.221 min^−1	1.5 × >P25	77 (2016)
MoS2/TiO2 nanofibers	10	50 W mercury lamp λ = 313 nm 10 mW cm^−2	DP of 98.2% in 30 min	7.01 × >TiO2	105 (2016)
TiO2@MoS2 nanotube arrays	10	230 W mercury lamp λ = 365 nm	DP of 85.3% in 120 min	1.14 × >TiO2	107 (2016)
MoS2/TiO2/Si nanowires	5 × 10^−6 M	Visible fiber lamp λ 420 to 700 nm 0.5 W cm^−2	DP of 82% in 180 min	2.9 × >TiO2/Si	124 (2016)
TiO2/MoS2 nanospheres	5	500 W Xe lamp visible light	DP of over 97.8% in 180 min	1.9 × >TiO2	134 (2016)
MoS2/TiO2/hollow microfibers	20	500 W halogen lamp λ > 390 nm	DP of 38% in 20 min	N/A	84 (2017)
Pollutant: MB
MoS2/P25	8	300 W Xe lamp λ 400 to 660 nm	DP of ∼40% in 4 h	N/A	148 (2004)
MoS2–TiO2/7.1%	5 ppm	UV lamp	K 0.034 min^−1	2.13 × >P25	87 (2011)
Z-Scheme g-C3N4/TiO2/MoS2/0.5%	10	500 W Xe lamp 280 to 800 nm	DP of 99.54% in 45 min	N/A	81 (2016)
TiO2-RGO/MoS2	10	Natural sunlight	DP of ∼100% in 100 min	N/A	86 (2016)
MoS2/TiO2(001) nanosheets/5%	4 × 10^−5 M	250 W mercury lamp UV-vis light	K 0.04797 min^−1	3.4 × >TiO2	139 (2016)
MoS2–TiO2	5	400 W Xe lamp visible light	DP of 99.33% in 12 min	1.75 × >P25	92 (2017)
Ag3PO4/TiO2@MoS2 nanofibers/3.5%	2.5	300 W Xe lamp visible light	DP of ∼95% in 16 min	2.5 × >Ag3PO4	117 (2017)
MoS2/TiO2 nanosheets/4.36%	1 × 10^−5 M	300 W Xe lamp λ > 420 nm	DP of over 83.26% in 30 min	1.4 × >TiO2	143 (2017)
MoS2/TiO2 film	10 ppm	Visible light	DP of 60% in 150 min	N/A	147 (2017)
Pollutant: MO
3D MoS2/P25/graphene/0.5%	20	300 W Hg lamp UV light	DP of 100% in 15 min	∼1.4 × >P25	85 (2014)
TiO2/MoS2@zeolite	20	500 W Xe lamp visible light	K app 2.304 h^−1	2.65 × >TiO2@zeolite	94 (2015)
TiO2/MoS2	20	500 W Xe lamp λ > 464 nm	DP of 97% in 60 min	3.46 × >P25	95 (2015)
Black-TiO2/MoS2/TiO2 nanosheets	10	300 W Xe lamp λ > 420 nm	K 0.01788 min^−1	17.5 × >TiO2	82 (2016)
TiO2/g-C3N4/MoS2	20	500 W Xe lamp simulated sunlight	DP of 90% in 60 min	3.11 × >TiO2	98 (2016)
MoS2/TiO2 nanobelts/5%	20	300 W mercury lamp λ = 356 nm/350 W Xe lamp λ > 420 nm	DP of ∼100% in 15 min/∼100% in 120 min	1.08/50 × >TiO2	106 (2016)
Nitrogen doped-TiO2−x@MoS2	10	300 W Xe lamp λ > 420 nm	DP of 91.8% in 120 min	21.3 × >TiO2	114 (2017)
Pollutant: direct green 6
MoS2/TiO2/0.3%	60	Sunlight 0.75 W m^−2	DP of >95% in 3 days	2.2 × >TiO2	76 (2017)
Pollutant: 4-ATP
Carbon-doped TiO2-MoS2/0.7%	10	84 W light source λ > 420 nm	DP of 40% in 60 min	2.22 × >TiO2	90 (2016)
Pollutant: Rs
MoS2/CdS/TiO2 nanoparticles	50	500 W Xe lamp UV-vis light	DP of 21.19% in 60 min third cycle	1.25 × >CdS/TiO2	93 (2016)
Pollutant: phenol
TiO2/MoS2 hollow spheres	10	300 W Xe lamp λ > 420 nm	DP of 78% in 150 min	N/A	109 (2016)

---

4.1.2 Photocatalytic hydrogen evolution. The photocatalytic hydrogen reduction using MoS₂/TiO₂-based photocatalysts is a sustainable and economical technique for addressing the energy crisis. The photogenerated electrons on the surface of the MoS₂ cocatalyst can reduce H₂O molecules to H₂ directly. Since Kanda and co-workers¹⁸⁰ first prepared MoS₂/TiO₂ composites through in situ photoinduced deposition as photocatalysts for hydrogen evolution, many researchers have devoted effort to exploring low-cost and high-activity MoS₂/TiO₂-based photocalysts.^{75,78,82,83,88,97,99,101,103,106,108,114,118,119,122,138,140,149,174,175,180} The synthetic methods have been summarized in section 3. Similar to photocatalytic degradation, the photocatalytic hydrogen evolution activity of MoS₂/TiO₂ composites depends on their light harvesting, charge carrier separation and migration, and catalytically active sites. A variety of strategies have been reported to increase photocatalytic hydrogen evolution performance, especially in the last three years. Examples include (i) MoS₂ engineering for promoting photocatalytic activity; (ii) interface engineering and graphene cocatalyst loading for improving charge carrier separation and migration efficiencies; (iii) visible light activity catalysis loading, photosensitizer loading and element doping for broadening the light absorption range. The strategies for improving the utilization efficiency of solar energy will be discussed in detail below. The recent MoS₂/TiO₂-based photocatalysts for photocatalytic hydrogen evolution are summarized in Table 9. Under visible light, the best photocatalytic hydrogen production achieved to date among reported MoS₂/TiO₂-based composites is 16 700 μmol g⁻¹ h⁻¹ in the presence of 5 vol% TEOA as a sacrificial agent.¹²²

Table 9 A summary of MoS₂/TiO₂ based photocatalysts for photocatalytic hydrogen evolution

Materials/(MoS2 wt%)	Sacrificial agents	Light sources	Photoactivity (μmol g^−1 h^−1)	Ref. (year)
TEOA is triethanolamine, FI is fluorescein, ZnTCPP is Zn(II)-5,10,15,20-tetrakis (4-carboxyphenyl) porphyrin, and EY is Eosin Y.
MoS2/TiO2	5 vol% formic acid	300 W Xe lamp λ > 300 nm	73	180 (2011)
TiO2/MoS2/graphene/0.5%	25 vol% ethanol/H2O	350 W Xe lamp λ = 365 nm 20 mW cm^−2	2066.3	97 (2012)
TiO2@MoS2 nanobelts/50%	0.35 M Na2S and 0.25 M Na2SO3 solution	300 W Xe lamp λ 280 to 700 nm 600 mW cm^−2	1600	101 (2013)
MoS2/TiO2 nanoparticles/1.0%	20 vol% methanol/H2O	300 W Xe lamp simulated sunlight 100 mW cm^−2	120	88 (2013)
EY-MoS2-TiO2 nanowires	5 vol% TEOA/H2O	300 W Xe lamp λ > 420 nm	16 700	122 (2014)
TiO2–MoS2(2H)	25 vol% methanol/H2O	300 W Xe lamp λ < 400 nm 2.7 mW cm^−2	∼175	75 (2015)
TiO2–MoS2(1T)	25 vol% methanol/H2O	300 W Xe lamp λ < 400 nm 2.7 mW cm^−2	∼2000
MoS2/TiO2 mesocrystals/10%	10 vol% lactic acid/H2O	UV-LED source 100 mW cm^−2	550	78 (2015)
TiO2/MoS2/4.0%	15 vol% methanol/H2O	300 W Xe lamp 250 to 380 nm	150.7 μmol h^−1	83 (2015)
TiO2@MoS2 nanofibers/60%	0.35 M Na2S and 0.25 M Na2SO3 solution	300 W Xe lamp 168 mW cm^−2 visible light/UV-vis	490/16 800	103 (2015)
ZnTCPP–MoS2/P25 nanoparticles/1.0%	0.2 M TEOA	300 W Xe lamp λ > 420 nm	10.2 μmol h^−1	119 (2015)
Black-TiO2/MoS2/TiO2 nanosheets	20 vol% methanol/H2O	300 W Xe lamp λ > 420 nm	560	82 (2016)
2D–2D MoS2/TiO2 nanosheets/0.5%	10 vol% methanol/H2O	300 W Xe lamp	2145	138 (2016)
MoS2/boron doped-TiO2 nanosheets/1.0%	33 vol% methanol/H2O	300 W Xe lamp λ > 400 nm 150 mW cm^−2	500	140 (2016)
g-C3N4/MoS2/TiO2/0.9%	33 vol% methanol/H2O	300 W Xe lamp λ > 400 nm	125 μmol h^−1	99 (2016)
MoS2/TiO2 nanotube arrays	50 vol% methanol/H2O	AM 1.5 solar 100 mW cm^−2	∼10 μL h^−1	174 (2016)
TiO2/MoS2(E) nanowires/0.5%	20 vol% ethanol/H2O	300 W Xe lamp UV light	4300	175 (2016)
MoS2/TiO2 nanobelts/5%	20 vol% methanol/H2O	300 W Xe lamp simulated sunlight 100 mW cm^−2	75	106 (2016)
FI-flower-like MoS2@TiO2/14.6%	7.5 vol% TEOA/acetone-H2O	Visible light	10 046	108 (2016)
Nitrogen doped-TiO2−x@MoS2	20 vol% methanol/H2O	AM 1.5 solar power system	1882	114 (2017)
EY-TiO2@distorted 1T-MoS2 nanofibers	15 vol% TEOA/H2O	300 W Xe lamp λ 420 to 780 nm 168 mW cm^−2	∼10 000	118 (2017)
MoS2/CdS-TiO2/1%	10 vol% lactic acid	300 W Xe lamp λ > 420 nm	280 μmol h^−1	149 (2017)

---

4.1.3 Photocatalytic CO₂ reduction. In addition to photocatalytic hydrogen evolution, the photoreduction of CO₂ into organic fuel sources based on MoS₂/TiO₂ photocatalysis shows great potential applications in both reducing greenhouse effects and addressing the energy crisis. Although MoS₂/TiO₂-based composites have been widely used in photocatalytic degradation and hydrogen evolution, only a few reports have focused on the photocatalytic reduction of CO₂. This is because CO₂ reduction is a multi-electron transfer and multi-step process. Recently, Tu et al.¹⁴¹ firstly reported the use of MoS₂/TiO₂ nanosheets as photocatalysts for CO₂ reduction to CH₃OH. The reaction pathway of CH₃OH generation in aqueous solution is CO₂ → HCOOH → HCHO → CH₃OH.¹⁸¹ Photocatalytic measurements confirmed that the MoS₂ cocatalyst can effectively promote charge carrier separation and provide catalytically active sites. The active sites of Mo-terminated edges on MoS₂ benefit the stabilization of the intermediate products (CH_xO_y) (Fig. 18). MoS₂/TiO₂ nanosheets with 0.5 wt% MoS₂ photocatalyst achieved the highest CO₂ reduction activity of 10.6 μmol g⁻¹ h⁻¹ under UV-vis light; this is 1.9 times higher than that of pure TiO₂ nanosheets (Fig. 18). Moreover, its activity is also higher than those of TiO₂ nanosheet photocatalysts loaded with 0.5 wt% Pt and Au, demonstrating that MoS₂ exhibits better and more selective photocatalytic reduction of CO₂ to CH₃OH than Pt and Au noble cocatalysts. However, the reported CO₂ reduction activity is still low; this requires further improvement and optimization.

Fig. 18 (a) Schematic of the charge separation and transfer in a MoS₂/TiO₂ nanosheet composite for the photoreduction of CO₂ to CH₃OH. (b) Comparison of the photocatalytic activity of pure TiO₂ nanosheets and MoS₂/TiO₂ hybrid nanosheets with different MoS₂ amounts for CO₂ photoreduction. Reproduced with permission from ref. 141. Copyright 2017, Royal Society of Chemistry.

4.1.4 Strategies for photocatalytic activity enhancement.
4.1.4.1 MoS₂ engineering. MoS₂ cocatalysts offer catalytically active sites for direct photocatalytic progress. Therefore, it is important to optimize MoS₂ with more catalytically active sites. MoS₂ has two main phases (2H phase and 1T phase) in practical research. Generally, MoS₂ exhibits the stable 2H phase in MoS₂/TiO₂-based composites. For 2H-MoS₂, the active sites are located along the edges, while the basal surface of MoS₂ is catalytically inert.¹⁸² Moreover, 2H-MoS₂ is a p-type semiconductor and thus has low charge carrier transport mobility. Therefore, these disadvantages of 2H-MoS₂ strongly limit its photocatalytic activity as a co-catalyst. In order to address these issues, engineering of MoS₂ has been widely researched. MoS₂ engineering in MoS₂/TiO₂-based composites mainly involves three methods. First, some researchers prepared small-sized 2H-MoS₂ in the in-plane direction to increase the exposed edge sites and reduce the charge carrier transport pathway.^{77,79,80,87,104,132,148,177,180} As a result, the small-sized MoS₂/TiO₂ composites show significantly improved photocatalytic activity. The dispersion of small-sized MoS₂ and the interaction between MoS₂ and TiO₂ should be well considered when using this modified method. Second, the surface defects of 2H-MoS₂ can provide additional active edge sites; however, they can also decrease the charge transfer kinetics. In order to disrupt the inverse relationship between MoS₂ surface defects and charge transfer rate, some researchers chose to dope MoS₂ together with defect engineering. The results demonstrated that oxygen-doped defective MoS₂ shows increased active edge sites and charge transfer rates compared with pure MoS₂.^183–185 Therefore, the photocatalytic activity of MoS₂ and its composites were significantly improved. However, there are no related reports of combining defect engineering with element doping in MoS₂/TiO₂-based composites; this topic is worthy of exploration. Third, some researchers have used 1T-MoS₂ instead of 2H-MoS₂ in MoS₂/TiO₂-based composites, as 1T-MoS₂ can not only offer active sites on its edge sites and basal surface, but also possesses excellent charge carrier transfer ability.^74,75,118 Moreover, unlike the case of 2H-MoS₂, the photogenerated electrons can participate directly in the reduction reaction on 1T-MoS₂ near the TiO₂, with no need to migrate to the far 2H-MoS₂ active edge sites (Fig. 19). This further decreases the chance of charge carrier recombination due to the greatly shortened transfer length of the electrons. As a result, the TiO₂–MoS₂(1T) sample exhibited 2 and ∼8 times higher RhB degradation and hydrogen evolution than the TiO₂–MoS₂(2H) sample, respectively, in the work of Bai et al. (Fig. 19).⁷⁵ Notably, 1T-MoS₂ is unstable under ambient conditions; it can readily change into 2H-MoS₂ during hydrothermal processes and low-temperature calcination.⁷⁸ Therefore, researchers always prepare 1T-MoS₂ in advance and then mix it with TiO₂ through ex situ strategies under mild conditions.^74,75 Thus, the interface contact between MoS₂ and TiO₂ is weak, which is harmful to charge carrier transfer. Recently, Wang and co-workers¹¹⁸ prepared TiO₂@distorted 1T-MoS₂ nanofibers through transformation of TiO₂@2H-MoS₂ that was fabricated by an in situ hydrothermal method; 1T-MoS₂ contacted TiO₂via a good interface. The stability of 1T-MoS₂ should be carefully considered when preparing and using 1T-MoS₂/TiO₂-based composites.

Fig. 19 Schematic of the charge-transfer behavior and H₂ evolution active sites for (a) 1T-MoS₂ and (b) 2H-MoS₂. (c) Photocatalytic degradation curves and (d) photocatalytic average rates of hydrogen production of RhB with MoS₂ NSs, TiO₂ NCs, TiO₂–MoS₂(1T), and TiO₂–MoS₂(2H) as catalysts under UV light irradiation. Reproduced with permission from ref. 75. Copyright 2015, Springer.

4.1.4.2 Interface engineering. The generated charge carriers originate by transfer from TiO₂ to the MoS₂ cocatalyst through the interface between MoS₂ and TiO₂; therefore, this interface plays a key role in charge carrier separation and migration. Generally, MoS₂/TiO₂-based composites prepared from ex situ strategies have poor-quality interfaces for charge carrier transfer compared with those prepared by in situ strategies.⁷⁵ For the in situ strategies, solution chemical methods are commonly used to prepare MoS₂/TiO₂-based composites,^88–149 where MoS₂ nanoplates contact the TiO₂ surface along the basal planes. The contact area and type directly determine the interface properties. In order to promote the interface between the TiO₂ and MoS₂ planes, interface engineering mainly involves two aspects. First, the TiO₂ surface is an important factor. Generally, readily obtained and stable 0D TiO₂ and 1D TiO₂ are used as substrates to prepare MoS₂/TiO₂-based composites. However, these species always show weak contact with 2D MoS₂. Recently, 2D TiO₂ nanosheets have attracted considerable interest in photocatalysis due to their unique properties.¹⁵² Moreover, the contact specific surface areas of 0D TiO₂/2D MoS₂, 1D TiO₂/2D MoS₂, and 2D TiO₂/2D MoS₂ gradually increase if the TiO₂ materials have the same specific surface areas. Therefore, TiO₂ nanosheets can provide larger contact specific surface areas for charge carrier transfer from TiO₂ to MoS₂. Recently, some researchers prepared MoS₂/TiO₂ nanosheets via in situ strategies.^137–143 For example, Yuan et al.¹³⁸ prepared 2D–2D MoS₂/TiO₂ nanosheets via a simple hydrothermal method (see Table 6). The 2D MoS₂ makes tight contact with the large surface area of 2D TiO₂. As a result, 2D–2D MoS₂/TiO₂ nanosheets with 0.5 wt% MoS₂ photocatalyst exhibited high activity of 2145 μmol g⁻¹ h⁻¹, which is 2.2 times higher than that of a 2D MoS₂/0D P25 composite with 0.5 wt% MoS₂; this demonstrates that the 2D–2D MoS₂/TiO₂ nanosheets with large contact area show excellent photocatalytic activity. However, the common configurations originating from solution chemical methods also exhibit low numbers of exposed active edge sites.¹⁸⁶ Moreover, the photogenerated charge carriers in TiO₂ are transferred to the active edge sites of 2H-MoS₂ through the MoS₂ layers. This hop transfer leads to much higher electron resistance.¹⁰³ In order to address these problems, MoS₂ is also an important factor affecting the contact interface properties. Second, some researchers have attempted to disrupt the traditional contact of MoS₂ nanoplates with the TiO₂ surface along the basal planes (TiO₂/MoS₂(B)). They fabricated MoS₂ standing on the TiO₂ surface, resulting in larger exposure of active edge sites and a regulated transport pathway.^103,175 Liu and co-workers¹⁰³ firstly prepared vertically standing MoS₂ nanosheets on porous TiO₂ nanofibers. The MoS₂ nanosheets could vertically grow from inside to outside the TiO₂ nanofibers due to the high number of pores in TiO₂. Therefore, the obtained TiO₂@MoS₂ composite enabled high exposure of the active edge sites of MoS₂, leading to high photocatalytic hydrogen evolution rates of 16 800 and 490 μmol g⁻¹ h⁻¹ under UV-vis and visible light, respectively.

Moreover, the interface contact between standing MoS₂ and TiO₂ requires further exploration and research. Recently, He et al.¹⁷⁵ developed a new strategy to control the contact of MoS₂ nanoplates with the TiO₂ surface along the MoS₂ edges (TiO₂/MoS₂(E)) (Fig. 20a), which is obviously different from the contact type in traditional TiO₂/MoS₂(B) composites (Fig. 20b). The designed photocatalyst showed excellent electrical contact and an optimized electron transfer pathway along the basal planes (Fig. 20c). As a result, TiO₂/MoS₂(E) with 3 wt% MoS₂ exhibited dramatically enhanced photocatalytic activity of 4300 μmol g⁻¹ h⁻¹ under UV light, which is over 8 times that of TiO₂/MoS₂(E) with the same MoS₂ content (Fig. 20d).

Fig. 20 Schematic of the preparation of TiO₂/MoS₂ composites. (a) Wet chemical method for preparing TiO₂/MoO₂(B) composite, where MoS₂ nanoplates contact the TiO₂ surface along (B) the basal planes; (b) CVD strategy to prepare TiO₂/MoS₂(E) composite, where MoS₂ nanoplates contact the TiO₂ surface along the edges (E); (c) schematic of the photoexcited electron transport pathway on TiO₂/MoS₂(E); (d) comparison of H₂ evolution rates on TiO₂, TiO₂/MoS₂(E)3%, and TiO₂/MoS₂(B)3%. Reproduced with permission from ref. 175. Copyright 2016, Wiley-VCH.

4.1.4.3 Graphene cocatalyst loading. The commonly used 2H-MoS₂ and anatase TiO₂ are semiconductors, which have poor electrical conductivity. Therefore, the photocatalytic activity of MoS₂/TiO₂ composites is limited by their intrinsically poor electrical conductivity. Graphene, or reduced graphene oxide (RGO), has high specific surface area and superior electrical conductivity; thus, it is widely used as a cocatalyst to prevent agglomeration and enhance the electron transfer of catalysts.⁹⁷ Therefore, some researchers^85,86,97 have introduced graphene or RGO cocatalysts in MoS₂/TiO₂ composites to construct ternary hybrids. For example, Xiang et al.⁹⁷ synthesized a TiO₂/MoS₂/RGO (T/MG) composite photocatalyst via a two-step hydrothermal method, where TiO₂ nanoparticles were grown on the surface of layered MoS₂/RGO (MG). Under irradiation, the photogenerated electrons on TiO₂ can not only transfer to the MoS₂ cocatalyst but also transfer to the G cocatalyst (Fig. 21a), which can effectively enhance the separation and migration efficiencies of the charge carriers. As a result, the ternary T95M5.0G sample exhibited a high photocatalytic activity of 2066.3 μmol g⁻¹ h⁻¹, which is 4 times that of a binary T/100M0G sample (Fig. 21b).

Fig. 21 (a) Schematic of the charge transfer in TiO₂/MG composites. (b) Photocatalytic H₂ evolution of TiO₂/MG composites. Photocatalytic H₂ production experiments were performed in 25% (v/v) ethanol/water solutions under UV irradiation using the photocatalyst TiO₂/MG composites with different MoS₂ and graphene contents in the MG hybrid as cocatalysts. Reproduced with permission from ref. 97. Copyright 2012, American Chemical Society.

4.1.4.4 Visible light active catalyst loading. TiO₂ catalysts with high E_g can only absorb UV light. However, natural sunlight comprises little UV light and 43% visible light. Therefore, it is important to broaden the light harvesting of MoS₂/TiO₂ photocatalysts to include visible light. One approach is to combine MoS₂/TiO₂ photocatalysts with a visible light activity catalyst. Recently, some researchers have introduced narrow-gap catalysts, such as C₃N₄,^81,98,99 CdS,^93,149 and Ag₃PO₄,¹¹⁷ into MoS₂/TiO₂ composites to form ternary hybrids. For example, Yang et al.⁹⁹ prepared g-C₃N₄/MoS₂/TiO₂ composites via a simple two-step method. The optimized sample 10C₃N₄/90MoS₂/TiO₂ with 0.1 wt% C₃N₄, 0.9 wt% MoS₂ and 99 wt% TiO₂ exhibited a high photocatalytic activity of 125 μmol h⁻¹ under visible light, which is 4.5 times that of binary 100MoS₂/TiO₂ with 1 wt% MoS₂ and 99 wt% TiO₂; this demonstrates that loading a visible light active catalyst in MoS₂/TiO₂ can effectively improve its visible light harvesting and photocatalytic activity.
4.1.4.5 Visible light-active photosensitizer loading. In addition to loading of visible light-active catalysts, introducing visible light-active photosensitizers in MoS₂/TiO₂ composites is an effective strategy to increase their visible light harvesting. Recently, some researchers have used visible light-active ZnTCPP,¹¹⁹ FI,¹⁰⁸ and EY^118,122 photosensitizers in MoS₂/TiO₂ composites (see Table 9). For example, MoS₂/TiO₂ materials were sensitized with ZnTCPP to realize H₂ evolution in a 0.2 M TEOA solution under visible light (λ > 420 nm).¹¹⁹ As shown in Fig. 22a, only the sensitized ZnTCPP–MoS₂–TiO₂ exhibited strong absorption at 415, 560, and 600 nm, which is similar to the adsorption of ZnTCPP. This result indicated that the ZnTCPP was successfully anchored on the surface of the MoS₂–TiO₂ composite and significantly improved its visible light harvesting. The ZnTCPP photosensitizer was excited under visible light irradiation; then, the photogenerated electrons were injected into TiO₂ CB, and the electrons catalyzed the reduction of water to H₂ at the surface of the MoS₂ cocatalyst (Fig. 22b). In addition, the oxidized ZnTCPP photosensitizer exhibited more positive oxidation potential than TEOA, ensuring the regeneration of the ZnTCPP dye. As a result, the ZnTCPP–MoS₂–TiO₂ sample obtained a high photocatalytic activity of 10.2 μmol h⁻¹ under visible light, which is significantly larger than that of the MoS₂–TiO₂ sample.

Fig. 22 (a) Absorption spectra of TiO₂, MoS₂/TiO₂ and ZnTCPP-MoS₂/TiO₂. (b) Proposed mechanism for visible-light-induced hydrogen production over ZnTCPP–MoS₂/TiO₂, S: ZnTCPP, D: TEOA; E_S+/S* = E_S+/S − E_0–0. Reproduced with permission from ref. 119. Copyright 2016, Elsevier.

4.1.4.6 Element doping. In order to broaden the absorption of MoS₂/TiO₂ photocatalysts from the UV region to the visible light region, impurity element doping in TiO₂ is another effective strategy.¹⁴ The element doping should meet at least two requirements: (1) introducing sub-bandgap states or narrowing the band gap of TiO₂; and (2) ensuring that the photoexcited carriers can transfer to reactive sites within their lifetimes.¹⁴ Three kinds of doping have been reported to date. The first is doping with anionic species such as B¹⁴⁰ and C.⁹⁰ The doping of anionic elements in TiO₂ can reduce the VB edge potential;¹⁴ thus, the band gap between the VB and CB edge potentials is reduced. For example, Yang et al.¹⁴⁰ used boron-doped TiO₂ (B-TiO₂) nanosheets as a substrate to prepare a MoS₂/B-TiO₂ composite. As shown in Fig. 23, the CB and VB edge potentials of pure TiO₂ are located at 0.46 and 2.70 eV, respectively. After B doping of TiO₂, the VB edge potential of B-TiO₂ decreased to 2.40 eV; therefore, the B-TiO₂ substrate exhibited a narrower band gap and broadened its light harvesting to the visible light region. The second type of doping is cationic metals such as Ti³⁺.^82,113 Ti³⁺-doped TiO₂ possesses increased CB edge potential; therefore, the band gap between the VB and CB edge potentials is reduced. Ti³⁺ doping is always performed in a reduction atmosphere at high temperature. In order to ensure the TiO₂ photoreduction activity, the CB minimum of cationic metal-doped TiO₂ should be higher than the H₂/H₂O level.¹⁴ Therefore, careful control of the degree of cationic metal doping is necessary. The last type of doping is co-doping of anionic and cationic species, such as N and Ti³⁺.¹¹⁴ N and Ti³⁺ co-doped TiO₂ can reduce the VB edge potential and increase the CB edge potential simultaneously; thus, the band gap is significantly reduced. As a result, the photocatalytic activity under visible light is significantly enhanced.

Fig. 23 Schematic of the energy band structure of the MoS₂/B-TiO₂ heterostructure and the proposed charge transfer mechanism under visible light irradiation. Reproduced with permission from ref. 140. Copyright 2016, Royal Society of Chemistry.

4.2 Rechargeable batteries

4.2.1 Lithium-ion batteries. Lithium-ion batteries (LIBs) are regarded as green and promising power sources due to their long cycle life and high-energy densities. MoS₂/TiO₂ composites are promising anode materials for LIBs owing to their combined advantages of high specific capacity and strong structural stability. In these composites, TiO₂ stores Li⁺via a typical intercalation mechanism, while MoS₂ shows different storage mechanisms by controlling the cut-off voltage range. Table 10 summarizes the LIB performance of reported MoS₂/TiO₂-based anode materials.

Table 10 A summary of the LIB performance of MoS₂/TiO₂-based anode materials

Materials	MoS2 and TiO2 weight contents	Voltage range	Capacities (mA h g^−1) at different current densities (A g^−1) (after 100 cycles)	Capacities (mA h g^−1) at different current densities (A g^−1)	Ref. (year)
4.2.1.1 TiO 2 as an anode material for LIBs
MoS2–TiO2 nanofibers	6%, 94%	1 to 3 V	124 at 6 (1000)	188, 177, 120 at 1, 2, 40	173 (2015)
4.2.1.2 TiO 2 as a supplemental material in LIBs
TiO2@MoS2 nanobelts	50.1%, 49.9%	0.01 to 3 V	710 at 0.1	717, 417 at 0.1, 1	102 (2014)
MoS2@TiO2 nanotubes	66.7%, 33.3%	0.005 to 3 V	472 at 0.1	713, 461 at 0.1, 1	127 (2014)
Nano-TiO2-decorated MoS2 nanosheets	71.8%, 28.2%	0.01 to 3 V	604 at 0.1	650, 472 at 0.05, 1	89 (2015)
Firework TiO2@MoS2 microspheres	N/A	0.01 to 3 V	714 at 0.1 (200 cycles)	820, 450 at 0.1, 1	125 (2015)
TiO2@MoS2 nanowires	75%, 25%	0.01 to 3 V	544 at 0.1	724, 414 at 0.1, 1	125 (2015)
MoS2@TiO2 nanosheets	47%, 53%	0.005 to 3 V	714 at 0.1	684, 551, 467 at 0.1, 1, 2	144 (2015)
TiO2@MoS2 microspheres	79.6%, 20.4%	0.01 to 3 V	467 at 0.1 (80 cycles)	722, 464, 430 at 0.1, 1, 2	130 (2016)
TiO2@MoS2 nano-onions	73.7%, 26.3%	0 to 3 V	632 at 0.1	710, 600 at 0.1, 1	121 (2016)
Carbon-coated TiO2@MoS2 nanosheets	36.5%, 48.3%	0.005 to 3 V	805 at 0.1	786, 586, 508 at 0.1, 1, 2	128 (2016)
MoS2/TiO2/Ti nanosheet arrays	N/A	0.01 to 3 V	1189 at 1 (600 cycles)	990, 700, 565 at 0.1, 1, 5	131 (2016)
TiO2-B/MoS2 nanowire arrays	N/A	0.01 to 3 V	350 at 0.02	437, 158 at 0.02, 2	123 (2016)
TiO2@MoS2/C nanosheet arrays	N/A	0.05 to 3 V	643 at 0.1	1046, 162 at 0.1, 1.2	129 (2016)
Flower-like MoS2/TiO2	85.7%, 14.3%	0.01 to 3 V	802 at 0.1 (50 cycles)	660, 760 at 1, 0.1	176 (2017)
P25@carbon@MoS2	46.3%, 43.5%	0.005 to 3 V	716 at 0.1	775, 554 at 0.1, 1	146 (2017)
TiO2@carbon@MoS2 nanotubes	59.6%, 17.0%	0.01 to 3 V	770 at 0.2 (100 cycles), 590 at 1 (200 cycles)	925, 670, 612 at 0.1, 1, 2	120 (2017)
MoS2@TiO2 microspheres	70.3%, 29.7%	0.01 to 3 V	734 at 0.1 (200 cycles)	725, 472 at 1, 2	136 (2017)
Graphene@MoS2@TiO2 microspheres	66.9%, 29.7%	0.01 to 3 V	980 at 0.1 (200 cycles)	783, 602 at 1, 2
G/UT-TiO2@C/MoS2 nanosheets	58.5%, 35.7%	0.005 to 3 V	818 at 0.1 (100 cycles), 648 at 1 (400 cycles)	793, 661, 582 at 0.1, 1, 2	145 (2017)
TiO2-MoS2 arrays	94.6%, 5.4%	0.02 to 3 V	362 at 0.8 (300 cycles)	570, 380 at 0.1, 1.6	112 (2017)

---

4.2.1.1 TiO₂ as an anode material. TiO₂ materials have received much interest in LIBs owing to their safe working voltage of around 1.5 V. Therefore, TiO₂ has been extensively studied as a safe LIB anode material. However, the low theoretical specific capacity and poor rate capability of TiO₂ severely limits its applications. In order to address these issues, some researchers have employed MoS₂ to construct MoS₂/TiO₂ composites. For example, Zhuang et al.¹⁷³ used MoS₂–TiO₂ nanofibers with 94 wt% TiO₂ as a LIB anode material. Then, electrochemical tests with MoS₂–TiO₂ half cells were performed over a narrow voltage range of 1.0 to 3.0 V. From the cyclic voltammogram (CV) curves (Fig. 24a), it can be seen that pure TiO₂ shows two types of peaks. The pair of peaks at 1.7 V in the discharge curve and at 2.1 V in the charge curve were assigned to anatase TiO₂ (eqn (8)), while the cathodic peak at 1.1 to 1.5 V and the anodic peak at 1.5 to 1.8 V were attributed to TiO₂-B (eqn (9)). After loading 6 wt% of MoS₂ on the TiO₂ surface, the peaks of anatase TiO₂ can be clearly observed, while the cathodic peak of TiO₂-B overlaps with that of MoS₂. The peaks of MoS₂ were associated with the interaction reaction (eqn (10)).

TiO₂ + xLi⁺ + xe⁻ → Li_xTiO₂ (0 < x < 1) (8)

TiO₂-B + xLi⁺ + xe⁻ → Li_xTiO₂-B (0 < x < 1) (9)

MoS₂ + xLi⁺ + xe⁻ → Li_xMoS₂ (10)

Fig. 24 (a) CV curves of the electrodes in the potential window of 1 to 3 V vs. Li⁺/Li at a rate of 0.1 mV s⁻¹. (b) Cycling performance and columbic efficiencies of TiO₂ and MoS₂–TiO₂ electrodes at a rate of 6C for 1000 cycles. (c) Rate capabilities of the MoS₂–TiO₂ and TiO₂ samples. Reproduced with permission from ref. 173. Copyright 2015, Wiley-VCH.

The cycling performance of MoS₂, TiO₂, and the MoS₂–TiO₂ composite was evaluated at the current density of 6C between 1 and 3 V (Fig. 24b). MoS₂–TiO₂ showed better cycling stability and larger capacity than pure MoS₂ and TiO₂, suggesting that MoS₂ can accelerate surface electrochemical reactions to increase the capacity. Moreover, rate capability tests demonstrated that MoS₂–TiO₂ shows improved capacity retention and rate capability compared with pure TiO₂ (Fig. 24c), further indicating that MoS₂ can also promote Li⁺ diffusion, especially at high rates. In summary, MoS₂ can effectively improve the Li⁺ diffusion and storage capacity of MoS₂–TiO₂ composite through careful design at a high cut-off voltage. However, MoS₂/TiO₂ composites with TiO₂ serving as an anode material have not been widely explored.

4.2.1.2 TiO₂ as a supplemental material. In addition, TiO₂ has been widely used as a supplemental material in LIB anode materials because its excellent structural stability when associated with lithium can provide good protection of the alloy and conversion-type active materials from agglomeration and pulverization.¹⁹ In most reported MoS₂/TiO₂-based LIB anode materials, TiO₂ mainly acts as a mechanical support against structure and volume changes of MoS₂. Generally, there are two main configurations in MoS₂/TiO₂ composites: the first one is TiO₂ coated with MoS₂; the second one is MoS₂ loaded on the surface of TiO₂.¹⁹ The reported MoS₂/TiO₂-based composites for LIB applications belong to the second configuration.^{102,112,120,121,125–131,136,144–146,176} When they are used as LIB anode materials, MoS₂ contributes most of the capacity and TiO₂ protects the overall structure. For example, Mao et al.¹⁰² firstly used a TiO₂ nanobelt@few-layered MoS₂ (TiO₂@MoS₂) composite in LIBs in 2014. The electrochemical tests were carried out in the potential window of 0.01 to 3.0 V. As shown in Fig. 25a, TiO₂@MoS₂ exhibited the typical peaks of anatase TiO₂ (eqn (8)). MoS₂ can undergo a full conversion reaction with lithium after deep discharge at 0.01 V. Therefore, the cathodic peak at 0.5 V was attributed to the conversion reaction (eqn (11)). After the first discharge, MoS₂ disappeared and S and Mo were generated. The subsequent electrochemical peaks of Mo and S were obviously different from those of the initial MoS₂. The peak at 2.0 V in the discharge curves and the peak at 2.3 V corresponded to S (eqn (12)). The discharge/charge curves also demonstrate the reaction mechanism (Fig. 25b). The cycling tests indicated that TiO₂@MoS₂ exhibits significantly better cycling stability than MoS₂ and obviously larger reversible capacity than TiO₂ (Fig. 25c), indicating that TiO₂@MoS₂ has better structural stability than MoS₂ and more lithium storage sites than TiO₂. In other words, the TiO₂@MoS₂ composites simultaneously possess high structural stability and numerous lithium storage sites. Moreover, the TiO₂@MoS₂ composite showed good rate capability, with a reversible capacity of 417 mA h g⁻¹ at 1000 mA g⁻¹ (Fig. 25d). The SEM images of the TiO₂@MoS₂ composites before cycling and after 100 cycles directly demonstrate their structural stability (Fig. 25e and f). This electrochemical mechanism and improved LIB performance can also be found in subsequent research. Currently, increasing attention has been drawn to exploring MoS₂/TiO₂-based composites for application in LIBs.

Li_xMoS₂ + (4 − x)Li⁺ + (4 − x)e⁻ → Mo + 2Li₂S (11)

S + 2Li⁺ + 2e⁻ ↔ Li₂S (12)

Fig. 25 (a) CV curves of TiO₂@MoS₂ measured in the voltage range of 0.01 to 3.0 V with a scan rate of 0.1 mV s⁻¹, (b) discharge/charge curves for the first three cycles, (c) cycling performance of TiO₂@MoS₂, TiO₂ and MoS₂ tested in the range of 0.01 to 3.0 V vs. Li⁺/Li at a current density of 100 mA g⁻¹, (d) rate performance at different current densities (mA g⁻¹), SEM images of TiO₂@MoS₂ (e) before cycling and (f) after 100 cycles. Reproduced with permission from ref. 102. Copyright 2014, Royal Society of Chemistry.

4.2.2 Sodium-ion batteries. LIBs have been widely used in consumer electronics, transportation, and renewable energy storage devices. However, the resources of lithium are limited and cannot meet the increasing demand. Therefore, researchers are eagerly searching for more abundant materials. Recently, sodium has been considered as a promising alternative due to its earth abundance, low cost and similar electrochemical properties.⁶ MoS₂/TiO₂ composites have also been used as anode materials in sodium-ion batteries (SIBs).^100,123 The SIB performance of MoS₂/TiO₂-based anode materials is summarized in Table 11. For example, Liao and co-workers¹²³ prepared TiO₂-B/MoS₂ nanowire arrays through a simple hydrothermal method as a SIB anode material. Fig. 26a and b display the CV and discharge–charge curves of the TiO₂-B/MoS₂ electrode in a wide voltage range between 0.01 V and 3.0 V. The sodium ion storage mechanism of TiO₂-B/MoS₂ composites is similar to the lithium ion storage mechanism. The electrochemical reactions of the TiO₂-B/MoS₂ electrode are composed of the reactions of TiO₂ and MoS₂.^100,123 They involve Na⁺ intercalation into the TiO₂ nanosheets (eqn (13)), the insertion of Na⁺ in MoS₂ to form Na_xMoS₂ (eqn (14)), and the further conversion of Na_xMoS₂ into Mo and Na₂S (eqn (15)) in the first cycle. After the first cycle, the Na⁺ storage reaction in TiO₂ remains unchanged; however, the Na⁺ storage reaction of MoS₂ no longer occurs. The S transformed from MoS₂ can store sodium ions (eqn (16)). The cycling performance of TiO₂-B and TiO₂-B/MoS₂ nanowire array electrodes is exhibited in Fig. 26c. Both electrodes show excellent cycling stability due to the robust TiO₂ backbone. However, TiO₂-B/MoS₂ electrode demonstrates a much higher capacity than the TiO₂-B electrode even after 100 cycles, suggesting the MoS₂ material can provide numerous sodium ions storage sites. The rate capabilities of TiO₂-B and TiO₂-B/MoS₂ in Fig. 26d show the similar phenomenon in all tested current densities. Therefore, MoS₂/TiO₂-based composites exhibit great potential application in SIBs, which requires more attention and deeper research.

TiO₂ + xNa⁺ + xe⁻ ↔ Na_xTiO₂ (13)

MoS₂ + xNa⁺ + xe⁻ → Na_xMoS₂ (14)

Li_xMoS₂ + (4 − x)Na⁺ + (4 − x)e⁻ → Mo + 2Na₂S (15)

S + 2Na⁺ + 2e⁻ ↔ Na₂S (16)

Fig. 26 (a) CV curves and (b) discharge–charge profiles of TiO₂-B/MoS₂ nanowire array electrodes. (c) Cycling performance and (d) rate capability of TiO₂-B and TiO₂-B/MoS₂ nanowire array electrodes. Reproduced with permission from ref. 123. Copyright 2015, Royal Society of Chemistry.

Table 11 A summary of the SIBs performance of MoS₂/TiO₂-based anode materials

Material	TiO2 weight content	Voltage range	Capacities (mA h g^−1) at current different densities (A g^−1) (after 100 cycles)	Capacities (mA h g^−1) at different current densities (A g^−1)	Ref. (year)
TiO2-B/MoS2 nanowire arrays	N/A	0.01 to 3 V	191 at 0.02	214, 77, 48 at 0.02, 1, 4	123 (2016)
MoS2-1 nm–TiO2	N/A	0.01 to 2.5 V	74.1% of its second cycle capacity at 0.2 (150 cycles); 182 at 0.5 (200 cycles)	369, 754 at 0.8, 0.05	100 (2017)

---

4.2.3 Strategies for enhancement of battery performance.
4.2.3.1 Interface engineering. It is well known that the interfaces of composite materials play a key role in their applications. For MoS₂/TiO₂ composites, the interface between MoS₂ and TiO₂ determines the structural stability and interface electron transfer directly when used in rechargeable batteries. Generally, the reported MoS₂/TiO₂-based composites are synthesized by in situ strategies so the MoS₂ can make strong contact with TiO₂ through chemical bonds. However, MoS₂/TiO₂ composites are more likely to display dendritic/dumbbell-like morphologies^{102,120,126,127,129–131,176} due to the significant lattice mismatch between TiO₂ and MoS₂. Therefore, considerable amounts of MoS₂ cannot directly contact the TiO₂ backbone, which reduces the structural stability and electrical conductivity of the overall composites. There are two main strategies to improve the interface. First, some researchers constructed a nanoscopic/atomic intimate interface between TiO₂ and MoS₂ through a rationally designed hydrothermal method.¹²¹ As shown in Table 6, the core–shell TiO₂@MoS₂ nano-onion heterostructures show conformal coverage with a high-quality interface. Moreover, the high-quality interface between TiO₂ and MoS₂ also results in exciting synergistic effects of both components. As a result, when the TiO₂@MoS₂ nano-onions were evaluated as a LIBs anode, they displayed superior lithium ion storage performance, especially at high current densities (see Table 10). Second, we designed a “surface-to-surface” interaction between 2D TiO₂ and 2D MoS₂ with the assistance of glucose to improve the interface in our previous studies.^128,144 As shown in Fig. 27a and b, most of the 2D MoS₂ is tightly covered on the surface of TiO₂ in the 3D FL-MoS₂@TiO₂ composite. Therefore, there is a large interface between 2D MoS₂ and 2D TiO₂. Both the strong interface and large contact area between MoS₂ and TiO₂ enable high structural stability and rapid interface electron transfer. Also, there are synergistic effects between TiO₂ and MoS₂. As a result, the 3D FL-MoS₂@TiO₂ electrode shows excellent cycling stability and rate capability which are higher than those of MoS₂/TiO₂ electrodes without strong interfaces (Fig. 27c and d). Therefore, “surface-to-surface” contact between MoS₂ and TiO₂ is also an effective strategy to improve the interface.

Fig. 27 TEM images of (a, b) 3D FL-MoS₂@TiO₂. (c) Cycling performance of F-TiO₂, MoS₂, the MoS₂/TiO₂ composite and 3D FL-MoS₂@TiO₂. (d) Rate-capability performance of electrodes of MoS₂, MoS₂/TiO₂ composite and 3D FL-MoS₂@TiO₂. Reproduced with permission from ref. 131. Copyright 2015, Royal Society of Chemistry.

4.2.3.2 Defect engineering. In addition to the structural stability of the composites, Li⁺/Na⁺ diffusion also plays an important role in the electrochemical performance of MoS₂/TiO₂ composites. Generally, MoS₂ contributes the most storage capacity in reported MoS₂/TiO₂ composites. Therefore, the Li⁺/Na⁺ diffusion in MoS₂ is very important. In order to improve the diffusion of ions in MoS₂, we prepared 2D sandwich-like carbon-coated ultrathin TiO₂@defect-rich MoS₂ UT-TiO₂/C@DR-MoS₂ nanosheets (see Table 6) via a simple method with the assistance of glucose binder.¹²⁸ The 2D defect-rich MoS₂ is tightly anchored onto the 2D carbon-coated UT-TiO₂ nanosheets in the MoS₂ UT-TiO₂/C@DR-MoS₂ composite, indicating a strong interface and large contact area; this leads to high structural stability and fast interface electron transfer. Moreover, the defect-rich MoS₂ can significantly improve the Li-ion diffusion kinetics. Therefore, the UT-TiO₂/C@DR-MoS₂ electrode exhibits the best LIB performance when compared with previous work on MoS₂/TiO₂-based LIB anodes (Fig. 28a and b and Table 10). The superior LIB performance is attributed to the new synergistic effect between the strong interface (interface engineering) and the rich defects in MoS₂ (defect engineering). As shown in Fig. 28c, in the first discharge process, the rich defects in MoS₂ can provide more channels for Li-ion diffusion and shorten the Li-ions diffusion pathway, thus leading to higher Li-ion diffusion rates. As a result, the Li-ions can be transported to the interface and deeper locations to provide more storage sites. Therefore, we believe the rich defects in MoS₂ can enhance Li-ion diffusion in MoS₂. Importantly, it should be noted that the MoS₂ nanosheets break down into Mo and S nanoparticles after the first discharge. It is also necessary to study the relationship between the structures of defect-rich MoS₂ and the Li-ion diffusion kinetics. It is believed that the pristine structure would directly affect the subsequent structures. We systematically studied the Li-ion diffusion kinetics of the UT-TiO₂/C@DR-MoS₂ electrode after the first cycle by electrochemical impedance spectroscopy. The results confirm the excellent Li-ion diffusion kinetics in UT-TiO₂/C@DR-MoS₂. Therefore, defect engineering can effectively improve the diffusion of ions in MoS₂, which may also result in novel synergistic effects with other modification strategies.

Fig. 28 (a) Cycling performance and (b) rate capability performance of pure MoS₂, UT-TiO₂ and UT-TiO₂/C@DR-MoS₂ nanocomposites. (c) Illustration of the lithiation process of UT-TiO₂/C@DR-MoS₂ in the first cycle. Reproduced with permission from ref. 128. Copyright 2016, Elsevier.

4.2.3.3 Graphene loading. As mentioned above, interface engineering (Fig. 29a) and extra defect engineering (Fig. 29b) can effectively improve the structural stability, interface electron transfer, and lithium/sodium ion diffusion of MoS₂/TiO₂ composites. However, both MoS₂ and TiO₂ have poor electrical conductivity, which severely hinders the cycling capacity, especially at high rates. Therefore, some researchers have introduced superior conductive graphene in MoS₂/TiO₂ composites to further improve their electrical conductivities. For example, Pang et al.¹³⁶ prepared a graphene@MoS₂@TiO₂ hybrid where MoS₂@TiO₂ microspheres are encapsulated by graphene. When graphene@MoS₂@TiO₂ and MoS₂@TiO₂ were evaluated as LIB anodes, as shown in Table 10, graphene@MoS₂@TiO₂ showed significantly improved capacities in all tested current densities compared with MoS₂@TiO₂, demonstrating that the incorporation of graphene in MoS₂/TiO₂ composites can effectively improve their LIB performance due to enhanced electrical conductivity. The authors confirmed that the graphene@MoS₂@TiO₂ electrode possesses significantly better electrical conductivity than the MoS₂@TiO₂ electrode.¹³⁶ It has been demonstrated that graphene loading in MoS₂/TiO₂ composites can further improve their overall electrical conductivity and thus enhance their electrochemical performance. Therefore, in order to further improve the electrochemical performance of MoS₂/TiO₂ composites, we introduced graphene in TiO₂/MoS₂ composites together with interface engineering and defect engineering in our previous work (Fig. 29c).¹⁴⁵ A 2D G/TiO₂@C/MoS₂ composite was prepared by a simple hydrothermal method using 2D GO/TiO₂ nanosheets, Na₂MoO₄·2H₂O, CH₄N₂S, and glucose as substrate, Mo source, S source, and binder, respectively. The optimized G/TiO₂@C/MoS₂ composite was obtained by controlling the amounts of added glucose and CH₄N₂S. In the designed G/TiO₂@C/MoS₂ composite, defect-rich MoS₂ nanosheets are tightly anchored on the surface of G/UT-TiO₂via strong chemical bonds, resulting in high structural stability, large Li-ion diffusion kinetics, and fast electrical transport channels (Fig. 29d). In addition, the graphene loading creates abundant MoS₂/G and TiO₂/G interfaces, leading to extra Li-ion storage capacity via additional pseudocapacitive storage mechanisms.^187–191 As a result, the optimized anode exhibited excellent cycling stability and especially high rate long-life cycling performance (Fig. 29e) which is better than those of reported MoS₂/TiO₂ based anodes (Table 10). In summary, synergistic structural, electronic and ionic modulations of MoS₂/TiO₂-based composites are an effective method to achieve high-performance LIB anode materials and can be applied to the rational design of other anode materials.

Fig. 29 Schematic of anodes with Li-ion and electron transport paths depicting the reference 2D–2D MoS₂–TiO₂-based anode materials with (a) defect-free basal surface of MoS₂ shell, (b) many-defect MoS₂ shell and (c) many-defect MoS₂ shell and incorporation of graphene. (d) Illustration of the lithiation process of the optimized G/TiO₂@C/MoS₂ in the first cycle. (e) Cycle performance of the optimized G/TiO₂@C/MoS₂ at 0.1 and 1 A g⁻¹. Reproduced with permission from ref. 145. Copyright 2017, Elsevier.

5. Summary and perspectives

Both TiO₂ and MoS₂ are interesting materials in the fields of photocatalysis and rechargeable batteries because of their low cost, unique structures, and excellent physicochemical properties. However, TiO₂ and MoS₂ alone have several inherent drawbacks which severely restrict their practical applications. The properties of TiO₂ and MoS₂ are highly complementary. Therefore, in recent years, increasing numbers of researchers have devoted much effort to constructing MoS₂/TiO₂ composites for enhanced photocatalytic activity and rechargeable battery performance. In this review, we have summarized the recent advances in the synthesis methods and correlated mechanisms of MoS₂/TiO₂-based composites, as well as their applications in photocatalytic degradation, photocatalytic hydrogen evolution, photocatalytic CO₂ reduction, LIBs and SIBs, and various strategies for enhancement of their photocatalytic activities and performance in battery applications. The achieved progress demonstrates that coupling TiO₂ with MoS₂ in a suitable manner can effectively improve the photocatalytic activity and battery performance of pure TiO₂ and MoS₂. Although considerable developments have been realized in these areas, some fundamental and essential issues require more effort and further research.

Firstly, although both TiO₂ and MoS₂ show facile preparation, abundance, and low cost, the reported synthesis methods of MoS₂/TiO₂-based composites are complex, costly, and unsatisfactory. Among these, the ex situ synthesis strategy shows advantages of low cost and scalable production. However, the TiO₂ and MoS₂ in the obtained MoS₂/TiO₂-based composites exhibit heterogeneous dispersion with weak interface interaction. For the in situ strategy, the most frequently utilized methods are hydrothermal and solvothermal methods, leading to strong interface contact between MoS₂ and TiO₂ in the fabricated MoS₂/TiO₂ composites. However, the production yield ranging from milligrams to grams is quite low. Moreover, due to significant lattice mismatch between TiO₂ and MoS₂, the MoS₂/TiO₂-based composites prepared via solution methods show dendritic/dumbbell-like morphologies. Therefore, considerable amounts of MoS₂ cannot directly contact TiO₂, resulting in low-quality interfaces. The interface lies at the heart of the overall properties of these composites. Therefore, more attention must be paid to the controlled preparation of MoS₂/TiO₂-based composites with high-quality interfaces. In summary, it is highly desirable to develop a facile and inexpensive method for preparing MoS₂/TiO₂-based composites with high-quality interfaces. Understanding and controlling the synthesis process is believed to be the key step to achieve further breakthroughs in photocatalysis and rechargeable battery research.

Secondly, for photocatalysis applications, the photocatalytic activities of MoS₂/TiO₂-based photocatalysts suffer from poor visible-light harvesting, weak contact, deficient catalytic active sites, and low electrical conductivity. Therefore, much effort has been devoted to enhance their photocatalytic activities, such as visible-light harvesting expansion, interface engineering, MoS₂ engineering, and graphene loading. However, the reported studies of these enhancement strategies are limited; therefore, the photocatalytic activity obtained to date is still not very satisfactory. To further improve the photocatalytic activity of MoS₂/TiO₂-based photocatalysts, it is necessary to systematically understand and adjust the strategies with smart design and synthesis. In addition, researchers have paid more attention to the design and synthesis of MoS₂/TiO₂-based composites at the nanostructure level and less attention to fundamental understanding of the transport of charge carriers across the interface and the catalytic processes associated with the active sites at the molecular and atomic levels. Therefore, it is necessary to deeply study and understand the mechanisms of generation, transport, and reaction processes of these complex charge carriers based on density functional theory (DFT) calculations, X-ray absorption near edge structure (XANES) studies, atomic level high-angle annular dark field-scanning TEM (HADDF-STEM), and in situ time-resolved spectroscopy. In particular, additional sacrificial reagents such as methanol, ethanol, lactic acid, formic acid, TEOA, and Na₂S–Na₂S₂O₃ pairs have been added to the aqueous solution as electron donors for photocatalytic hydrogen evolution. However, these additional sacrificial reagents can damage the reaction stability, affect the cycle life and reduce the solar energy conversation efficiency, which is detrimental to practical applications. Therefore, there is an urgent need to develop overall water splitting systems that use MoS₂/TiO₂-based composites as photocatalysts. In addition, photocatalytic CO₂ reduction has gained increasing interest. However, the reported photolytic activity based on MoS₂/TiO₂ photocatalysts is low, and the CO₂ reduction mechanism is still unclear. It is necessary to improve the efficiency of the CO₂ reduction process and carefully investigate the photocatalytic CO₂ reduction mechanism.

Thirdly, in the case of rechargeable batteries, the electrochemical performance is determined by the storage sites, electron transport, Li/Na ion diffusion, and structural stability of the electrode. These properties are controlled by adjusting the structure, composition, and morphology of the electrode. The ideal MoS₂/TiO₂-based electrode should have the intriguing traits of a high-quality interface, large amounts of few-layer, defect-rich MoS₂, and loading of graphene or other materials with high electrical conductivity. Although significant advances have been achieved to enhance the performance of MoS₂/TiO₂-based LIBs, including high initial coulombic efficiency, large storage capacity, and high-rate long-cycle life, studies of NIBs have been rarely reported to date, probably due to the difficulty of Na-ion transport. Therefore, it is still necessary to investigate the structure, composition, and microscopic morphology of electrodes during battery operation via advanced in situ characterization techniques.

In summary, MoS₂/TiO₂-based composites show promising potential in the photocatalysis and rechargeable battery fields. However, this research is still in an early stage, and many challenges remain. We believe this review will aid understanding of the advances and challenges in the development of promising photocatalysts and rechargeable batteries based on MoS₂/TiO₂-based composites.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The authors acknowledge financial support by the National Natural Science Foundation of China (Grant No. 51472177, 11474216, and 51272173), the State Key Program of National Natural Science of China (Grant No. 51531004) and the China-EU Science and Technology Cooperation Project (Grant No. SQ2013ZOA100006). The authors also thank Tianshuai Wang and Lichao Guo for their help with the graphical abstract.

Notes and references

1. D. Ravelli, D. Dondi, M. Fagnoni and A. Albini, Chem. Soc. Rev., 2009, 38, 1999–2011.
2. D. Bahnemann, Sol. Energy, 2004, 77, 445–459.
3. P. V. Kamat, J. Phys. Chem. C, 2007, 111, 2834–2860.
4. B. Dunn, H. Kamath and J.-M. Tarascon, Science, 2011, 334, 928–935.
5. M. Armand and J. M. Tarascon, Nature, 2008, 451, 652–657.
6. M. D. Slater, D. Kim, E. Lee and C. S. Johnson, Adv. Funct. Mater., 2013, 23, 947–958.
7. H. Pan, Y.-S. Hu and L. Chen, Energy Environ. Sci., 2013, 6, 2338–2360.
8. D.-D. Zhu, J. L. Liu and S. Z. Qiao, Adv. Mater., 2016, 28, 3423–3452.
9. Y. Tang, Z. Jiang, G. Xing, A. Li, P. D. Kanhere, Y. Zhang, T. C. Sum, S. Li, X. Chen, Z. Dong and Z. Chen, Adv. Funct. Mater., 2013, 23, 2932–2940.
10. J. Mao, M. Ge, J. Huang, Y. Lai, C. Lin, K. Zhang, K. Meng and Y. Tang, J. Mater. Chem. A, 2017, 5, 11873–11881.
11. A. Fujishima and K. Honda, Nature, 1972, 238, 37–38.
12. A. Fujishima, X. Zhang and D. A. Tryk, Surf. Sci. Rep., 2008, 63, 515–582.
13. X. Chen and S. S. Mao, Chem. Rev., 2007, 107, 2891–2959.
14. R. Asahi, T. Morikawa, T. Ohwaki, K. Aoki and Y. Taga, Science, 2001, 293, 269–271.
15. K. Nakata and A. Fujishima, J. Photochem. Photobiol., C, 2012, 13, 169–189.
16. Z. Jiang, Y. Tang, Q. Tay, Y. Zhang, O. I. Malyi, D. Wang, J. Deng, Y. Lai, H. Zhou, X. Chen, Z. Dong and Z. Chen, Adv. Energy Mater., 2013, 3, 1368–1380.
17. G.-N. Zhu, Y.-G. Wang and Y.-Y. Xia, Energy Environ. Sci., 2012, 5, 6652–6667.
18. Z. Yang, D. Choi, S. Kerisit, K. M. Rosso, D. Wang, J. Zhang, G. Graff and J. Liu, J. Power Sources, 2009, 192, 588–598.
19. T. Song and U. Paik, J. Mater. Chem. A, 2016, 4, 14–31.
20. Y. Mei, Y. Huang and X. Hu, J. Mater. Chem. A, 2016, 4, 12001–12013.
21. J. Tian, Z. Zhao, A. Kumar, R. I. Boughton and H. Liu, Chem. Soc. Rev., 2014, 43, 6920–6937.
22. Y. Tang, Y. Zhang, J. Deng, J. Wei, L. H. Tam, B. K. Chandran, Z. Dong, Z. Chen and X. Chen, Adv. Mater., 2014, 26, 6111–61118.
23. Y. Tang, Y. Zhang, J. Deng, D. Qi, W. R. Leow, J. Wei, S. Yin, Z. Dong, R. Yazami, Z. Chen and X. Chen, Angew. Chem., Int. Ed., 2014, 53, 13488–13492.
24. Y. Tang, Y. Zhang, W. Li, B. Ma and X. Chen, Chem. Soc. Rev., 2015, 44, 5926–5940.
25. N. Wu, J. Wang, D. N. Tafen, H. Wang, J.-G. Zheng, J. P. Lewis, X. Liu, S. S. Leonard and A. Manivannan, J. Am. Chem. Soc., 2010, 132, 6679–6685.
26. L. Ren, Y. Li, J. Hou, J. Bai, M. Mao, M. Zeng, X. Zhao and N. Li, Appl. Catal., B, 2016, 181, 625–634.
27. H. Dong, G. Zeng, L. Tang, C. Fan, C. Zhang, X. He and Y. He, Water Res., 2015, 79, 128–146.
28. H. Xu, S. Ouyang, L. Liu, P. Reunchan, N. Umezawa and J. Ye, J. Mater. Chem. A, 2014, 2, 12642–12661.
29. H. Xu, P. Reunchan, S. Ouyang, H. Tong, N. Umezawa, T. Kako and J. Ye, Chem. Mater., 2013, 25, 405–411.
30. L. Pan, S. Wang, J. Xie, L. Wang, X. Zhang and J.-J. Zou, Nano Energy, 2016, 28, 296–303.
31. J. Long, H. Chang, Q. Gu, J. Xu, L. Fan, S. Wang, Y. Zhou, W. Wei, L. Huang, X. Wang, P. Liu and W. Huang, Energy Environ. Sci., 2014, 7, 973–977.
32. T. Tachikawa, M. Minohara, Y. Hikita, C. Bell and H. Y. Hwang, Adv. Mater., 2015, 27, 7458–7461.
33. J. Jin, C. Wang, X.-N. Ren, S.-Z. Huang, M. Wu, L.-H. Chen, T. Hasan, B.-J. Wang, Y. Li and B.-L. Su, Nano Energy, 2017, 38, 118–126.
34. B. Han and Y. H. Hu, Energy Sci. Eng., 2016, 4, 285–304.
35. Y.-J. Yuan, H.-W. Lu, Z.-T. Yu and Z.-G. Zou, ChemSusChem, 2015, 8, 4113–4127.
36. X. Zou and Y. Zhang, Chem. Soc. Rev., 2015, 44, 5148–5180.
37. Y. Li, Y.-L. Li, B. Sa and R. Ahuja, Catal. Sci. Technol., 2017, 7, 545–559.
38. A. B. Laursen, S. Kegnaes, S. Dahl and I. Chorkendorff, Energy Environ. Sci., 2012, 5, 5577–5591.
39. S. Chen, J. Duan, Y. Tang, B. Jin and S. Z. Qiao, Nano Energy, 2015, 11, 11–18.
40. J. Ran, G. Gao, F. T. Li, T. Y. Ma, A. Du and S. Z. Qiao, Nat. Commun., 2017, 8, 13907.
41. H. S. Lee, S.-W. Min, Y.-G. Chang, M. K. Park, T. Nam, H. Kim, J. H. Kim, S. Ryu and S. Im, Nano Lett., 2012, 12, 3695–3700.
42. K. F. Mak, C. Lee, J. Hone, J. Shan and T. F. Heinz, Phys. Rev. Lett., 2010, 105, 136805.
43. E. S. Kadantsev and P. Hawrylak, Solid State Commun., 2012, 152, 909–913.
44. Y. Tang, Y. Zhang, X. Rui, D. Qi, Y. Luo, W. R. Leow, S. Chen, J. Guo, J. Wei, W. Li, J. Deng, Y. Lai, B. Ma and X. Chen, Adv. Mater., 2016, 28, 1567–1576.
45. Y. Zhang, O. I. Malyi, Y. Tang, J. Wei, Z. Zhu, H. Xia, W. Li, J. Guo, X. Zhou, Z. Chen, C. Persson and X. Chen, Angew. Chem., Int. Ed., 2017, 56, 14847–14852.
46. Y. Tang, J. Deng, W. Li, O. I. Malyi, Y. Zhang, X. Zhou, S. Pan, J. Wei, Y. Cai, Z. Chen and X. Chen, Adv. Mater., 2017, 29, 1701828.
47. Y. Zhang, X. Rui, Y. Tang, Y. Liu, J. Wei, S. Chen, W. R. Leow, W. Li, Y. Liu, J. Deng, B. Ma, Q. Yan and X. Chen, Adv. Energy Mater., 2016, 6, 1502409.
48. T. Stephenson, Z. Li, B. Olsen and D. Mitlin, Energy Environ. Sci., 2014, 7, 209–231.
49. T. Wang, S. Chen, H. Pang, H. Xue and Y. Yu, Adv. Sci., 2016, 4, 1600289.
50. X. Zhang, Z. Lai, C. Tan and H. Zhang, Angew. Chem., Int. Ed., 2016, 55, 8816–8838.
51. Y. Xiao, S. H. Lee and Y.-K. Sun, Adv. Energy Mater., 2017, 7, 1601329.
52. H. Li, Y. Shi, M.-H. Chiu and L.-J. Li, Nano Energy, 2015, 18, 293–305.
53. X. Xu, W. Liu, Y. Kim and J. Cho, Nano Today, 2014, 9, 604–630.
54. S. Wu, Y. Du and S. Sun, Chem. Eng. J., 2017, 307, 189–207.
55. Y. Zhang, Z. Jiang, J. Huang, L. Y. Lim, W. Li, J. Deng, D. Gong, Y. Tang, Y. Lai and Z. Chen, RSC Adv., 2015, 5, 79479–79510.
56. M. Ge, C. Cao, J. Huang, S. Li, Z. Chen, K.-Q. Zhang, S. S. Al-Deyab and Y. Lai, J. Mater. Chem. A, 2016, 4, 6772–6801.
57. D. T. Cromer and K. Herrington, J. Am. Chem. Soc., 1955, 77, 4708–4709.
58. A. Di Paola, M. Bellardita and L. Palmisano, Catalysts, 2013, 3, 36.
59. A. R. Armstrong, G. Armstrong, J. Canales and P. G. Bruce, Angew. Chem., Int. Ed., 2004, 43, 2286–2288.
60. W. Wunderlich, T. Oekermann, L. Miao, N. T. Hue, S. Tanemura and M. Tanemura, J. Ceram. Process. Res., 2004, 5(4), 343–354.
61. A. T. Paxton and L. Thiên-Nga, Phys. Rev. B: Condens. Matter, 1998, 57, 1579–1584.
62. Q. Shen, K. Katayama, T. Sawada, M. Yamaguchi, Y. Kumagai and T. Toyoda, Chem. Phys. Lett., 2006, 419, 464–468.
63. Q. H. Wang, K. Kalantar-Zadeh, A. Kis, J. N. Coleman and M. S. Strano, Nat. Nanotechnol., 2012, 7, 699–712.
64. I. Song, C. Park and H. C. Choi, RSC Adv., 2015, 5, 7495–7514.
65. F. Wypych and R. Schollhorn, J. Chem. Soc., Chem. Commun., 1992, 1386–1388.
66. R. G. Dickinson and L. Pauling, J. Am. Chem. Soc., 1923, 45, 1466–1471.
67. F. Jellinek, G. Brauer and H. Muller, Nature, 1960, 185, 376–377.
68. F. E. Wickman and D. K. Smith, Am. Mineral., 1967, 55, 1843–1856.
69. K. F. Mak, C. Lee, J. Hone, J. Shan and T. F. Heinz, Phys. Rev. Lett., 2010, 105, 136805.
70. A. Wilson and A. D. Yoffe, Adv. Phys., 1969, 18, 193–335.
71. G. Eda, H. Yamaguchi, D. Voiry, T. Fujita, M. Chen and M. Chhowalla, Nano Lett., 2011, 11, 5111–5116.
72. D. Voiry, M. Salehi, R. Silva, T. Fujita, M. Chen, T. Asefa, V. B. Shenoy, G. Eda and M. Chhowalla, Nano Lett., 2013, 13, 6222–6227.
73. R. Kappera, D. Voiry, S. E. Yalcin, B. Branch, G. Gupta, A. D. Mohite and M. Chhowalla, Nat. Mater., 2014, 13, 1128–1134.
74. Y. Pi, Z. Li, D. Xu, J. Liu, Y. Li, F. Zhang, G. Zhang, W. Peng and X. Fan, ACS Sustainable Chem. Eng., 2017, 5, 5175–5182.
75. S. Bai, L. Wang, X. Chen, J. Du and Y. Xiong, Nano Res., 2015, 8, 175–183.
76. L. Karimi, Pol. J. Chem. Technol., 2017, 19, 56–60.
77. D. Wang, Y. Xu, F. Sun, Q. Zhang, P. Wang and X. Wang, Appl. Surf. Sci., 2016, 377, 221–227.
78. P. Zhang, T. Tachikawa, M. Fujitsuka and T. Majima, Chem. Commun., 2015, 51, 7187–7190.
79. S. Cravanzola, L. Muscuso, F. Cesano, G. Agostini, A. Damin, D. Scarano and A. Zecchina, Langmuir, 2015, 31, 5469–5478.
80. B. Pourabbas and B. Jamshidi, Chem. Eng. J., 2008, 138, 55–62.
81. W.-K. Jo, T. Adinaveen, J. J. Vijaya and N. C. Sagaya Selvam, RSC Adv., 2016, 6, 10487–10497.
82. X. Liu, Z. Xing, H. Zhang, W. Wang, Y. Zhang, Z. Li, X. Wu, X. Yu and W. Zhou, ChemSusChem, 2016, 9, 1118–1124.
83. Y. Zhu, Q. Ling, Y. Liu, H. Wang and Y. Zhu, Phys. Chem. Chem. Phys., 2015, 17, 933–940.
84. Y. Yu, J. Wan, Z. Yang and Z. Hu, J. Colloid Interface Sci., 2017, 502, 100–111.
85. W. Han, C. Zang, Z. Huang, H. Zhang, L. Ren, X. Qi and J. Zhong, Int. J. Hydrogen Energy, 2014, 39, 19502–19512.
86. D. B. Nimbalkar, H.-H. Lo, P. V. R. K. Ramacharyulu and S.-C. Ke, RSC Adv., 2016, 6, 31661–31667.
87. I. Tacchini, E. Terrado, A. Anso and M. T. Marti, Micro Nano Lett. IET, 2011, 6, 932–936.
88. Q. Liu, Z. Pu, A. Asiri, A. Qusti, A. Al-Youbi and X. Sun, J. Nanopart. Res., 2013, 15, 2057.
89. X. Zhu, C. Yang, F. Xiao, J. Wang and X. Su, New J. Chem., 2015, 39, 683–688.
90. M. M. Ali and K. N. Y. Sandhya, New J. Chem., 2016, 40, 8123–8130.
91. L. Wang, C. Wang, W. Liu, Q. Chen and M. He, Tetrahedron Lett., 2016, 57, 1771–1774.
92. M. Sabarinathan, S. Harish, J. Archana, M. Navaneethan, H. Ikeda and Y. Hayakawa, RSC Adv., 2017, 7, 24754–24763.
93. M. Sun, Y. Wang, Y. Fang, S. Sun and Z. Yu, J. Alloys Compd., 2016, 684, 335–341.
94. W. Zhang, X. Xiao, L. Zheng and C. Wan, Appl. Surf. Sci., 2015, 358, 468–478.
95. W. Zhang, X. Xiao, L. Zheng and C. Wan, Can. J. Chem. Eng., 2015, 93, 1594–1602.
96. R. Xiaohui, Q. Xiang, S. Yongzhen, X. Si, X. Guanghua, Z. Zhen, H. Zongyu and Z. Jianxin, J. Phys. D: Appl. Phys., 2016, 49, 315304.
97. Q. Xiang, J. Yu and M. Jaroniec, J. Am. Chem. Soc., 2012, 134, 6575–6578.
98. W. Zhang, X. Xiao, Y. Li, X. Zeng, L. Zheng and C. Wan, Appl. Surf. Sci., 2016, 389, 496–506.
99. X. Yang, H. Huang, M. Kubota, Z. He, N. Kobayashi, X. Zhou, B. Jin and J. Luo, Mater. Res. Bull., 2016, 76, 79–84.
100. W. Ren, W. Zhou, H. Zhang and C. Cheng, ACS Appl. Mater. Interfaces, 2017, 9, 487–495.
101. W. Zhou, Z. Yin, Y. Du, X. Huang, Z. Zeng, Z. Fan, H. Liu, J. Wang and H. Zhang, Small, 2013, 9, 140–147.
102. M. Mao, L. Mei, D. Guo, L. Wu, D. Zhang, Q. Li and T. Wang, Nanoscale, 2014, 6, 12350–12353.
103. C. Liu, L. Wang, Y. Tang, S. Luo, Y. Liu, S. Zhang, Y. Zeng and Y. Xu, Appl. Catal., B, 2015, 164, 1–9.
104. H. Liu, T. Lv, C. Zhu, X. Su and Z. Zhu, J. Mol. Catal. A: Chem., 2015, 396, 136–142.
105. X. Zhang, C. Shao, X. Li, F. Miao, K. Wang, N. Lu and Y. Liu, J. Alloys Compd., 2016, 686, 137–144.
106. H. Li, Y. Wang, G. Chen, Y. Sang, H. Jiang, J. He, X. Li and H. Liu, Nanoscale, 2016, 8, 6101–6109.
107. L. Zheng, S. Han, H. Liu, P. Yu and X. Fang, Small, 2016, 12, 1527–1536.
108. B. Ma, P.-Y. Guan, Q.-Y. Li, M. Zhang and S.-Q. Zang, ACS Appl. Mater. Interfaces, 2016, 8, 26794–26800.
109. J. Wang, B. Wei, L. Xu, H. Gao and W. Sun, J. Chem. Mater. Lett., 2016, 179, 42–46.
110. X. Li, X. Li, J. Cheng, D. Yuan, W. Ni, Q. Guan, L. Gao and B. Wang, Nano Energy, 2016, 21, 228–237.
111. X. Liu, X. Huo, P. Liu, Y. Tang, J. Xu, X. Liu and Y. Zhou, Electrochim. Acta, 2017, 242, 327–336.
112. Y. Zhang, S. Zhao, X. Zeng, J. Xiao, P. Dong, J. Zhao, S. Sun, L. Huang and X. Li, J. Mater. Sci.: Mater. Electron., 2017, 28, 9519–9527.
113. C. Liu, J. Chen, H. Che, K. Huang, P. A. Charpentier, W. Z. Xu, W. Shi and H. Dong, RSC Adv., 2017, 7, 8429–8442.
114. X. Liu, Z. Xing, Y. Zhang, Z. Li, X. Wu, S. Tan, X. Yu, Q. Zhu and W. Zhou, Appl. Catal., B, 2017, 201, 119–127.
115. L. Yang, X. Zheng, M. Liu, S. Luo, Y. Luo and G. Li, J. Hazard. Mater., 2017, 329, 230–240.
116. R. Tang, R. Yin, S. Zhou, T. Ge, Z. Yuan, L. Zhang and L. Yin, J. Mater. Chem. A, 2017, 5, 4962–4971.
117. N. Shao, J. Wang, D. Wang and P. Corvini, Appl. Catal., B, 2017, 203, 964–978.
118. L. Wang, X. Liu, J. Luo, X. Duan, J. Crittenden, C. Liu, S. Zhang, Y. Pei, Y. Zeng and X. Duan, Angew. Chem., Int. Ed., 2017, 56, 7610–7614.
119. Y. Yuan, H. Lu, Z. Ji, J. Zhong, M. Ding, D. Chen, Y. Li, W. Tu, D. Cao, Z. Yu and Z. Zou, Chem. Eng. J., 2015, 275, 8–16.
120. S. Wang, B. Y. Guan, L. Yu and X. W. Lou, Adv. Mater., 2017, 29, 1702724.
121. R. Dai, A. Zhang, Z. Pan, A. M. Al-Enizi, A. A. Elzatahry, L. Hu and G. Zheng, Small, 2016, 12, 2792–2799.
122. M. Shen, Z. Yan, L. Yang, P. Du, J. Zhang and B. Xiang, Chem. Commun., 2014, 50, 15447–15449.
123. J.-Y. Liao, B. D. Luna and A. Manthiram, J. Mater. Chem. A, 2016, 4, 801–806.
124. A. Hamdi, L. Boussekey, P. Roussel, A. Addad, H. Ezzaouia, R. Boukherroub and Y. Coffinier, Mater. Des., 2016, 109, 634–643.
125. B. Guo, K. Yu, H. Fu, Q. Hua, R. Qi, H. Li, H. Song, S. Guo and Z. Zhu, J. Mater. Chem. A, 2015, 3, 6392–6401.
126. X. Li, W. Li, M. Li, P. Cui, D. Chen, T. Gengenbach, L. Chu, H. Liu and G. Song, J. Mater. Chem. A, 2015, 3, 2762–2769.
127. X. Xu, Z. Fan, S. Ding, D. Yu and Y. Du, Nanoscale, 2014, 6, 5245–5250.
128. B. Chen, E. Liu, F. He, C. Shi, C. He, J. Li and N. Zhao, Nano Energy, 2016, 26, 541–549.
129. G. Li, L. Yu, H. Hu, Q. Zhu, Y. Wang and Y. Yu, Electrochim. Acta, 2016, 212, 59–67.
130. W. Xu, T. Wang, Y. Yu and S. Wang, J. Alloys Compd., 2016, 689, 460–467.
131. Y. Zhou, Y. Liu, W. Zhao, F. Xie, R. Xu, B. Li, X. Zhou and H. Shen, J. Mater. Chem. A, 2016, 4, 5932–5941.
132. W. Gao, M. Wang, C. Ran and L. Li, Chem. Commun., 2015, 51, 1709–1712.
133. P. X. Zhao, Y. Tang, J. Mao, Y. X. Chen, H. Song, J. W. Wang, Y. Song, Y. Q. Liang and X. M. Zhang, J. Alloys Compd., 2016, 674, 252–258.
134. C. Wang, H. Lin, Z. Liu, J. Wu, Z. Xu and C. Zhang, Part. Part. Syst. Charact., 2016, 33, 221–227.
135. J. Liu, Y. Li, J. Ke, Z. Wang and H. Xiao, Catalysts, 2017, 7, 30.
136. Q. Pang, Y. Zhao, X. Bian, Y. Ju, X. Wang, Y. Wei, B. Liu, F. Du, C. Wang and G. Chen, J. Mater. Chem. A, 2017, 5, 3667–3674.
137. L. Cao, R. Wang, D. Wang, X. Li and H. Jia, Mater. Lett., 2015, 160, 286–290.
138. Y.-J. Yuan, Z.-J. Ye, H.-W. Lu, B. Hu, Y.-H. Li, D.-Q. Chen, J.-S. Zhong, Z.-T. Yu and Z.-G. Zou, ACS Catal., 2016, 6, 532–541.
139. J. Zhang, L. Huang, Z. Lu, Z. Jin, X. Wang, G. Xu, E. Zhang, H. Wang, Z. Kong, J. Xi and Z. Ji, J. Alloys Compd., 2016, 688, 840–848.
140. X. Yang, H. Huang, B. Jin, J. Luo and X. Zhou, RSC Adv., 2016, 6, 107075–107080.
141. W. Tu, Y. Li, L. Kuai, Y. Zhou, Q. Xu, H. Li, X. Wang, M. Xiao and Z. Zou, Nanoscale, 2017, 9, 9065–9070.
142. J. Zhang, L. Zhang, W. Yu, F. Jiang, E. Zhang, H. Wang, Z. Kong, J. Xi and Z. Ji, J. Am. Ceram. Soc., 2017, 100, 5274–5285.
143. C. Li, S. Zhang, Y. Zhou and J. Li, J. Mater. Sci.: Mater. Electron., 2017, 28, 9003–9010.
144. B. Chen, N. Zhao, L. Guo, F. He, C. Shi, C. He, J. Li and E. Liu, Nanoscale, 2015, 7, 12895–12905.
145. B. Chen, E. Liu, T. Cao, F. He, C. Shi, C. He, L. Ma, Q. Li, J. Li and N. Zhao, Nano Energy, 2017, 33, 247–256.
146. B. Chen, N. Zhao, C. Wei, J. Zhou, F. He, C. Shi, C. He and E. Liu, Appl. Surf. Sci., 2017, 401, 232–240.
147. H. N. T. Phung, V. N. K. Tran, L. T. Nguyen, L. K. T. Phan, P. A. Duong and H. V. T. Le, J. Nanomater., 2017, 2017, 3197540.
148. W. Ho, J. C. Yu, J. Lin, J. Yu and P. Li, Langmuir, 2004, 20, 5865–5869.
149. N. Qin, J. Xiong, R. Liang, Y. Liu, S. Zhang, Y. Li, Z. Li and L. Wu, Appl. Catal., B, 2017, 202, 374–380.
150. W. Shi, S. Song and H. Zhang, Chem. Soc. Rev., 2013, 42, 5714–5743.
151. H. G. Yang, C. H. Sun, S. Z. Qiao, J. Zou, G. Liu, S. C. Smith, H. M. Cheng and G. Q. Lu, Nature, 2008, 453, 638–641.
152. W.-J. Ong, L.-L. Tan, S.-P. Chai, S.-T. Yong and A. R. Mohamed, Nanoscale, 2014, 6, 1946–2008.
153. B. Chen, J. Sha, W. Li, F. He, E. Liu, C. Shi, C. He, J. Li and N. Zhao, ACS Appl. Mater. Interfaces, 2016, 8, 2495–2504.
154. E. Liu, J. Wang, C. Shi, N. Zhao, C. He, J. Li and J.-Z. Jiang, ACS Appl. Mater. Interfaces, 2014, 6, 18147–18151.
155. H. Liu, X. Wang, C. Pan and K. M. Liew, J. Phys. Chem. C, 2012, 116, 8044–8053.
156. T. Xia, W. Zhang, Z. Wang, Y. Zhang, X. Song, J. Murowchick, V. Battaglia, G. Liu and X. Chen, Nano Energy, 2014, 6, 109–118.
157. Y. Liu, A. A. Elzatahry, W. Luo, K. Lan, P. Zhang, J. Fan, Y. Wei, C. Wang, Y. Deng, G. Zheng, F. Zhang, Y. Tang, L. Mai and D. Zhao, Nano Energy, 2016, 25, 80–90.
158. Y. Li, C. Chen, M. Wang, W. Li, Y. Wang, L. Jiao and H. Yuan, J. Power Sources, 2017, 361, 326–333.
159. X.-Y. Yu, H. Hu, Y. Wang, H. Chen and X. W. Lou, Angew. Chem., Int. Ed., 2015, 54, 7395–7398.
160. Y. Li, Y. Liang, F. C. Robles Hernandez, H. Deog Yoo, Q. An and Y. Yao, Nano Energy, 2015, 15, 453–461.
161. L. Yang, W. Zhou, J. Lu, D. Hou, Y. Ke, G. Li, Z. Tang, X. Kang and S. Chen, Nano Energy, 2016, 22, 490–498.
162. S. Ding, J. S. Chen and X. W. Lou, Chem. – Eur. J., 2011, 17, 13142–13145.
163. Y. Li, L. Wang, T. Cai, S. Zhang, Y. Liu, Y. Song, X. Dong and L. Hu, Chem. Eng. J., 2017, 321, 366–374.
164. K. Chang and W. Chen, J. Mater. Chem., 2011, 21, 17175–17184.
165. Z. Wang, T. Chen, W. Chen, K. Chang, L. Ma, G. Huang, D. Chen and J. Y. Lee, J. Mater. Chem. A, 2013, 1, 2202–2210.
166. G. Huang, T. Chen, W. Chen, Z. Wang, K. Chang, L. Ma, F. Huang, D. Chen and J. Y. Lee, Small, 2013, 9, 3693–3703.
167. L. Yang, X. Wang, Y. Liu, Z. Yu, R. Li and J. Qiu, Catal. Sci. Technol., 2017, 7, 693–702.
168. X. H. Yang, Z. Li, C. Sun, H. G. Yang and C. Li, Chem. Mater., 2011, 23, 3486–3494.
169. H. G. Yang and H. C. Zeng, J. Phys. Chem. B, 2003, 107, 12244–12255.
170. K. Chang and W. Chen, Chem. Commun., 2011, 47, 4252–4254.
171. T. Ben Nasr, N. Kamoun, M. Kanzari and R. Bennaceur, Thin Solid Films, 2006, 500, 4–8.
172. X.-Q. Bao, D. Y. Petrovykh, P. Alpuim, D. G. Stroppa, N. Guldris, H. Fonseca, M. Costa, J. Gaspar, C. Jin and L. Liu, Nano Energy, 2015, 16, 130–142.
173. W. Zhuang, L. Li, J. Zhu, R. An, L. Lu, X. Lu, X. Wu and H. Ying, ChemElectroChem, 2015, 2, 374–381.
174. X. Zhou, M. Licklederer and P. Schmuki, Electrochem. Commun., 2016, 73, 33–37.
175. H. He, J. Lin, W. Fu, X. Wang, H. Wang, Q. Zeng, Q. Gu, Y. Li, C. Yan, B. K. Tay, C. Xue, X. Hu, S. T. Pantelides, W. Zhou and Z. Liu, Adv. Energy Mater., 2016, 6, 1600464.
176. X. Zhu, X. Liang, X. Fan and X. Su, RSC Adv., 2017, 7, 38119–38124.
177. K. H. Hu, X. G. Hu, Y. F. Xu and J. D. Sun, J. Mater. Sci., 2010, 45, 2640–2648.
178. E. Girel, E. Puzenat, C. Geantet and P. Afanasiev, Catal. Today, 2014, 292, 154–163.
179. D. Kong, H. Wang, J. J. Cha, M. Pasta, K. J. Koski, J. Yao and Y. Cui, Nano Lett., 2013, 13, 1341–1347.
180. S. Kanda, T. Akita, M. Fujishima and H. Tada, J. Colloid Interface Sci., 2011, 354, 607–610.
181. W. Tu, Y. Zhou and Z. Zou, Adv. Mater., 2014, 26, 4607–4626.
182. T. F. Jaramillo, K. P. Jørgensen, J. Bonde, J. H. Nielsen, S. Horch and I. Chorkendorff, Science, 2007, 317, 100–102.
183. Y. Sun, S. Gao, F. Lei and Y. Xie, Chem. Soc. Rev., 2015, 44, 623–636.
184. H. Lin, Y. Li, H. Li and X. Wang, Nano Res., 2017, 10, 1377–1392.
185. S. Zhang, H. Yang, H. Gao, R. Cao, J. Huang and X. Xu, ACS Appl. Mater. Interfaces, 2017, 9, 23635–23646.
186. H. Wang, D. Kong, P. Johanes, J. J. Cha, G. Zheng, K. Yan, N. Liu and Y. Cui, Nano Lett., 2013, 13, 3426–3433.
187. Y. Lv, B. Chen, N. Zhao, C. Shi, C. He, J. Li and E. Liu, Surf. Sci., 2016, 651, 10–15.
188. E. Liu, J. Wang, C. Shi, N. Zhao, C. He, J. Li and J.-Z. Jiang, ACS Appl. Mater. Interfaces, 2014, 6, 18147–18151.
189. B. Chen, Y. Meng, E. Liu, F. He, C. Shi, C. He, L. Ma and J. N. Q. Zhao, Nano Energy, 2017, 41, 154–163.
190. H. Liu, D. Su, R. Zhou, B. Sun, G. Wang and S. Z. Qiao, Adv. Energy Mater., 2012, 2, 970–975.
191. F. Xie, L. Zhang, D. Su, M. Jaroniec and S. Z. Qiao, Adv. Mater., 2017, 29, 1700989.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: 10.1039/c7nr07366f

---