Positively charged Pt-based cocatalysts: an orientation for achieving efficient photocatalytic water splitting

Abstract

Considerable effort has been made to develop efficient water-splitting photocatalysts, which are generally composed of light-absorbing semiconductors and so-called cocatalysts that play a key role in accelerating the reaction kinetics of water splitting. Platinum (Pt)-based catalysts have been widely regarded as excellent cocatalysts for photocatalytic water splitting, but the relationship between their valence state and reaction activity for photocatalytic water splitting had never been summarized. This review presents positively charged Pt nanoparticles, clusters and even single atoms that can boost the reaction activity dramatically, and summarises several strategies and techniques to tailor the positive valence states of Pt-based cocatalysts such as by using oxygen atoms. By tailoring the valence states of Pt-based cocatalysts, the capacities of these cocatalysts for water splitting can be significantly enhanced. Moreover, the operando analyses that can determine the real structures of this type of cocatalyst under working conditions are also discussed. These developments suggest that positively charged Pt-based cocatalysts could determine the direction and efficiency of photocatalytic water splitting, which may provide an ideal orientation to design state-of-the-art catalysts for achieving efficient photocatalytic water splitting.

---

Jinze Liu

Jinze Liu received his Bachelor's degree in Chemical Engineering and Technology from the East China University of Science and Technology (ECUST) in 2017. He is currently a Master’s candidate in Material Chemical Engineering under the supervision of Prof. Chunzhong Li and Prof. Yuhang Li at the ECUST. His current research is focused on the design and fabrication of novel photo/electro-catalysts for water splitting and CO2 reduction.

Yuhang Li

Yuhang Li received his PhD degree in Materials Science and Engineering from the East China University of Science and Technology (ECUST) in 2016. He joined the Key Laboratory for Ultrafine Materials of Ministry of Education as a postdoctoral fellow in 2016–2018 at the ECUST. In September 2019, he joined the Sargent group, University of Toronto, as a visiting professor. He is currently an associate professor of the School of Materials Science and Engineering at the ECUST. His research interests are focused on the design and fabrication of novel photo/electro-catalysts for water splitting and CO2 reduction.

Xiaodong Zhou

Xiaodong Zhou is a Professor at the East China University of Science and Technology. He received his Bachelor's degree in Polymer Materials Science and Engineering (1990) and PhD degree in Polymer Materials Science from Nanjing University of Science and Technology (1996). He was a post-doctoral researcher (1996–1998) at the East China University of Science and Technology and has been working there since 1998. His current research interests include interface design and control of multiphase multicomponent polymer materials, preparation and molding technology of thermoplastic polymer composites and high performance and practical technology of biological resource polymer materials.

Hao Jiang

Hao Jiang received his PhD degree in Materials Science and Engineering from the ECUST, China, in 2009. He joined Temasek Laboratories, Nanyang Technological University (NTU) in Singapore, as a research scientist in 2009–2011. Currently he is a full professor in the Key Laboratory for Ultrafine Materials of Ministry of Education at the ECUST, and was the winner of the National Science Fund for Excellent Young Scholars in 2015. His current research interests focus on the design and preparation of novel hierarchical nanomaterials and their application in electrochemical energy storage and conversion.

Hua Gui Yang

Hua Gui Yang completed his PhD degree at the National University of Singapore in 2005. He joined the University of Queensland in 2007 as a Postdoctoral Research Fellow. After finishing his postdoctoral training, he came back to China and took a professor position at the East China University of Science and Technology at the end of 2008. His main research interests are focused on the rational design and fabrication of functional materials for solar-energy conversion.

Chunzhong Li

Chunzhong Li received his BS (1989), MS (1992) and PhD (1997) degrees from the ECUST. He became a full professor at the School of Materials Science and Engineering in 1998, and now he is the dean of the School of Chemical Engineering at the ECUST. He was elected as a Fellow of the Royal Society of Chemistry in 2015, became the Cheung Kong Distinguished Professor in 2014, and the winner of the National Science Fund for Distinguished Young Scholars in 2009. His research interests include functionalization and fabrication of nanomaterials for new energy, environment and relevant applications.

---

1. Introduction

Photocatalytic water splitting to produce hydrogen (H₂) as a solar fuel is considered a promising way to address environmental issues.¹ This reaction is an uphill reaction, for which the ΔG can be as high as +237 kJ mol⁻¹.² Thus, to promote the production of H₂, photocatalysts are usually involved to lower the reaction barriers.^3–6 Since Fujishima and Honda demonstrated the capacity of TiO₂ coupled with a platinum (Pt) electrode for photoelectrocatalytic water splitting in the early 1970s, extensive research has been conducted to fabricate high-efficiency photocatalytic water-splitting systems, which can be illustrated as a photoelectrocatalytic-like cell without applying bias.^7–10

The major steps involved in the photocatalytic water-splitting reaction are as follows: (I) semiconductor photocatalysts convert light energy to electron–hole carriers, (II) carriers separate and migrate to the sites of cocatalysts on photocatalysts, and (III) water reduction and/or oxidation occurs on cocatalysts. Photocatalytic water splitting can occur when these three steps are completed, and numerous strategies have been proposed to promote these processes.¹¹ For example, semiconductors with narrow bandgaps absorb longer wavelengths of solar light, employing nanosized materials accelerates photogenerated carrier migration, and loading suitable cocatalysts reduces and/or oxidizes water molecules. Notably, it is necessary to study the half reactions of H₂ and O₂ evolution using sacrificial agents to identify possible photocatalysts for overall water-splitting systems.³ Moreover, investigations of the half reactions could provide insight into detailed mechanisms during water splitting. In water-splitting systems, semiconductor photocatalysts usually require the help of appropriate cocatalysts to increase the rate of the water-splitting reaction because cocatalysts that trap electrons from the bulk of the photocatalyst are key to complete proton reduction (splitting water into H₂).^12–15 Therefore, metal nanoparticles or clusters are widely used as cocatalysts to facilitate electron transfer and catalytic processes at the interfaces between photocatalysts and solutions.

Cocatalysts supply the active sites of water-splitting reactions because most bare photocatalysts have negligible H₂ evolution activity, even with the help of sacrificial agents.⁵ Researchers have proposed some reasons for this negligible activity, such as the fast recombination of photogenerated carriers before they migrate to the surface for water splitting and the insufficient rate of catalytic processes compared to the recombination step. Among all cocatalysts developed to date, platinum (Pt)-based nanoparticles and clusters have been considered the most universal cocatalysts over many different kinds of semiconductors, including metal oxides, metal sulfides, metal nitrides, etc.,^16–18 and all have been demonstrated to remarkably improve the H₂ evolution activity. Pt-based cocatalysts can act as electron sinks and provide adsorption sites for protons, thus facilitating the proton reduction reaction. The superior trapping ability of Pt may be because its work function (that is, the lowest Fermi level) is the highest among those of many metals and is also greater than those of most semiconductors, forming a Schottky barrier which facilitates carrier separation at the interface.¹⁹ In addition, researchers have established a volcano-like curve between the activity of H₂ generation and the adsorption energy of protons and have demonstrated the excellent activation energy for proton reduction by Pt.²⁰ Thus, Pt has been regarded as the most powerful cocatalyst for photocatalytic water splitting because of its electronic and catalytic properties.

Since an increase in the activity of Pt nanoparticles with a mean size less than 5 nm was observed experimentally, various efforts have been devoted to studying the descriptors of the catalytic activity,^16,21–25 such as the size and support interaction of Pt-based cocatalysts. Moreover, consideration of the undesirable back reaction of water formation by H₂ and O₂ is required because this reaction is spontaneous when using metallic Pt as a cocatalyst.^26,27 To promote H₂ and O₂ evolution by water splitting, the addition of carbonate salts to the reaction solution was suggested in early studies, and the fabrication of core–shell structures was suggested in recent literature.^28,29 Many researchers have discovered that Pt-based cocatalysts with positive valence states can enhance total performance in photocatalytic water splitting. And Pt-based cocatalysts with positive valence states can not only promote the reaction activity of photocatalysts but also solve the problem of the back-reaction of water formation by H₂ and O₂. Some literature studies also have reported the effect mechanisms of positively charged Pt-based cocatalysts on the performance of photocatalytic water splitting by the combination of theoretical simulations and operando characterization. Although a number of high-quality review articles describing Pt-based cocatalysts for water splitting have been published,^{2–5,13–15} there are few reviews devoted entirely to the effect of valence states on the activity of these cocatalysts. Here, we will focus on the role of the valence states of Pt-based cocatalysts in photocatalytic systems (Fig. 1) and highlight the use of operando characterization to determine the real structures of the cocatalysts during water splitting. Additionally, some interesting reaction mechanisms associated with positively charged Pt-based cocatalysts are also described and discussed.

Fig. 1 (a) Valence state effects of Pt-based cocatalysts on photocatalytic water splitting. (b) Simulations of H₂ oxidation with different valence states of Pt-based cocatalysts. (c and d) Ball and stick model of a PtO cocatalyst. Reproduced with permission.³⁰ Copyright 2013, Springer Nature.

2. Valence state effects of Pt-based cocatalysts

2.1. Tailoring with oxygen atoms to suppress the HOR

In 2013, Li et al. synthesized a PtO/TiO₂ photocatalyst that exhibits enhanced performance and reported that the valence state of Pt can influence photocatalytic water splitting by controlling unidirectional H₂ oxidation (Fig. 2). According to the STEM characterization in Fig. 2b, the mean size of a PtO-cluster is found to be around 1 nm, and further from the Fourier-transformed spectrum of the Pt L₃-edge extended X-ray absorption fine structure, the Pt species of PtO/TiO₂ is indicated to be the PtO phase (Fig. 2c). In addition to the core–shell structures used for inhibiting the back reaction, Pt in a high valence state can suppress H₂ back-oxidation. In this study, the abilities of both Pt and PtO as cocatalysts to suppress the H₂ oxidation reaction were studied in an aqueous methanol solution. In the light reaction, the photocatalytic H₂ production rate of the Pt cocatalyst is 1/4 that of PtO. Interestingly, the H₂ pressure on the Pt cocatalyst decreases faster than that on PtO in the dark reaction (Fig. 2d). The photocatalytic water splitting benefited from the suppressing effect of the PtO clusters on the H₂ oxidation reaction. In this photocatalyst system, there are two roles for the PtO cocatalyst: (i) providing efficient H₂ evolution active sites and (ii) suppressing reverse H₂ oxidation. The X-ray photoelectron spectral analysis results indicate that the Pt species in the cocatalysts are in the form of metallic Pt and PtO.³⁰

Fig. 2 (a) H₂ oxidation can be suppressed and catalyzed by PtO and metallic Pt cocatalysts. (b) Scanning transmission electron microscopy image of a PtO cocatalyst loaded on TiO₂. The inset shows the size distribution of PtO. Scale bar, 5 nm. (c) The PtO cocatalyst contains only Pt–O bonds, as detected using X-ray absorption fine structure spectra. (d) Photocatalytic H₂ generation and H₂ oxidation rates using methanol as a sacrificial agent with alternating light on and off in 2 h intervals. Reproduced with permission.³⁰ Copyright 2013, Springer Nature.

Moreover, by using PtO cocatalysts and by utilizing advanced electron microscopy and simulation calculations, Li et al. demonstrated in 2015 that TiO₂ crystal facets do not play a critical role in photocatalytic reactions (Fig. 3). The possible electron transfer between the cocatalyst and host catalyst is illustrated in Fig. 3a. They have systematically investigated the true photoreactivity order of the {001} and {101} facets of TiO₂ by means of loading PtO clusters as a cocatalyst (Fig. 3c and d). The valence states of the Pt-based cocatalyst are the main reason why the photoactivity of the PtO-loaded photocatalyst is higher than that of the Pt-loaded photocatalyst,³¹ which could be inferred from Fig. 3b. Furthermore, to obtain structural information about the PtO catalyst during water splitting, the structural changes of the PtO cocatalyst were detected by operando X-ray absorption fine structure spectroscopy during water splitting (Fig. 4a–d), in which a Pt–O bond length of 2.13 Å and a coordination number of 2.5 were quantitatively determined under working conditions (Fig. 4f). The PtO cocatalyst can be observed by STEM with an average size of 1.5 nm, and the PtO cocatalyst is uniformly coated on the surface (Fig. 4e). The results show that the structure of PtO during the reaction is quite different from that of PtO with fewer O atoms and longer bond lengths, and even these longer bonds are shorter than the Pt–Pt bond in metallic Pt, consistent with the higher valence state of the PtO cocatalyst than that of metallic Pt. These results indicate the important role of valence states of the PtO cocatalyst in the unidirectional inhibition of the H₂ oxidation reaction.³²

Fig. 3 (a) Possible photogenerated electron migration when loading different valence states of Pt-based cocatalysts on TiO₂. (b) Photocatalytic H₂ generation rates of TiO₂ photocatalysts loaded with different valence states of Pt-based cocatalysts. Scanning transmission electron microscopy images of PtO cocatalysts on sheet-like (c) and octahedron-like (d) photocatalysts. Reproduced with permission.³¹ Copyright 2015, Elsevier.

Fig. 4 (a–d) Spectra of PtO as a cocatalyst detected by operando X-ray absorption fine structure spectroscopy. The numbers 1–5 indicate the initial time, addition of solvent, light on, light off and recycled sample, respectively. (e) Scanning transmission electron microscopy image of a PtO cocatalyst (in white circles) loaded on TiO₂. (f) Parameters such as Pt–O bond length and coordination number under different conditions. Reproduced with permission.³² Copyright 2017, The Royal Society of Chemistry.

Ren et al. reported an enhanced photocatalytic H₂ evolution performance of Pt/PtO heterojunction nanodots on TiO₂. In this report, a NaOH modification method is utilized to anchor the Pt/PtO nanodots onto porous TiO₂ with interparticle mesopores, resulting in enhanced photochemical H₂ production.

The formation of Pt²⁺ species is related to the influence of NaOH on the reduction rate of H₂PtCl₆. Metallic Pt can effectively separate the photogenerated carriers, and PtO can serve as a reaction site for H₂ evolution while inhibiting the H₂ oxidation reaction.³³ Manwar et al. introduced Pt/CeO₂ for donor-assisted photocatalytic H₂ evolution reactions, where the photocatalyst is prepared via the wet impregnation method. X-ray photoelectron spectra of the Pt component have been reported, indicating that the Pt species are in the form of Pt, Pt²⁺ and Pt⁴⁺. The existence of Pt and PtO together can promote H₂ evolution and suppress H₂ oxidation.³⁴

2.2. Tailoring with oxygen atoms to boost the HER and OER

2.2.1. Pt oxide cocatalysts on a TiO₂ photocatalyst. In 2010, Huang et al. investigated how redox-treated Pt/TiO₂ photocatalysts influence H₂ generation via the impregnation method. Moreover, oxidation-treated Pt cocatalysts, compared with reduction-treated samples, can increase the H₂ generation rate. X-ray photoelectron spectral analysis indicates that PtO exists after oxidation treatment. Furthermore, the Pt species are in the form of Pt⁰ in the photocatalytic system after reduction treatment. The H₂ production rate of the photocatalysts in an aqueous methanol solution is 3400 mmol h⁻¹ g⁻¹ when 0.2 wt% Pt is loaded. The energy band gap of this photocatalyst is higher than that of the reduction-treated catalyst, indicating that H₂ generation from water splitting can be easily realized on oxidation-treated PtO cocatalysts.³⁵ To further study the active sites on photocatalysts during the photocatalytic H₂ evolution reaction, Xing and coworkers investigated the active sites on the atomic level with a combination of experimental and theoretical methods (Fig. 5). The Fourier transformed spectra of the Pt L₃-edge EXAFS of the platinized photocatalysts are exhibited in Fig. 5c, which suggest the existence of Pt–O binding in the sample Pt/TiO₂-A-CN. The photocatalytic activities of Pt/TiO₂-A/B-CN can be much higher than those of Pt/TiO₂-A/B (Fig. 5d). The computational results can confirm the experimental observations that the oxidized Pt species loaded on TiO₂ can play key roles in increasing the H₂ production rate.³⁶ Parayil et al. prepared Pt–TiO₂ with high photocatalytic H₂ production from water by different methods. The Pt–TiO₂ material prepared by the hydrothermal method shows the highest surface area. To further clarify the individual roles of PtO₂ and Pt as cocatalysts in photocatalytic H₂ evolution, X-ray photoelectron spectral analysis was utilized. The results show that the Pt species are in the form of both metallic Pt and PtO₂ in the sample. The main reason that this material promotes photocatalytic H₂ evolution is that the existence of Pt^δ+ accelerates the efficient transmission of electrons and reduces the recombination of carriers.³⁷ Jin et al. designed a Pt/TiO₂ light harvesting reactor via an alkali modification process that enhances photocatalytic H₂ production. These researchers synthesized Pt/TiO₂ photocatalysts with a low content of Pt loaded onto yolk–shell 3-dimensional, porous TiO₂. The Pt species are in the form of both Pt⁰ and oxidized Pt²⁺. The photocatalyst demonstrates a superior H₂ generation rate of approximately 21 mmol h⁻¹ g⁻¹ with robust stability under light irradiation, which is 87 times that of the control sample that contains no Pt²⁺ species.³⁸

Fig. 5 (a) Schematic preparation process of a Pt^δ+ cocatalyst. (b) Relationship between the catalytic rate and ΔG of H adsorption with different Pt-based cocatalysts. (c) The Pt^δ+ cocatalyst has only Pt–O bonds, as determined from the X-ray absorption fine structure spectra. (d) Photocatalytic H₂ generation rates and turnover frequencies of different valence states of Pt-based cocatalysts. Reproduced with permission.³⁶ Copyright 2013, The Royal Society of Chemistry.

In 2017, Sui et al. reported a Pt/TiO₂ photocatalyst fabricated by a facile hydrothermal method. In this photocatalyst, atomic Pt is loaded on TiO₂. The photocatalyst composition is 72% single atoms, 16% clusters of 0.4–1 nm, and 12% particles of 1–2 nm, as demonstrated by the size distribution frequency. The X-ray photoelectron spectral analysis results suggest that more deficient-state Pt^δ+ is present, while the Pt species in commonly used photocatalysts are mainly in the form of metallic Pt. The H₂ generation rate of the photocatalyst can reach 84.5 μmol h⁻¹, which is higher than that of the control sample due to the valence state effects of Pt-based cocatalysts.³⁹ Recently, Talas et al. modified PtO_x–TiO₂ photocatalysts by introducing tin. In this article, the valence state of Pt in the initial and recycled samples following high-temperature H₂ reduction and annealing was studied. Pt exists in the form of Pt⁰ and Pt²⁺ in the initial sample after calcination, while mainly Pt⁰ can be detected after the photocatalytic tests, indicating that the reduction of Pt^δ+ occurs during water splitting. The H₂ production can be hindered by reducing the Pt nanoparticles at the start of the H₂ evolution reaction in the case of the calcined samples, which has a negative effect on activating the photocatalyst by high-temperature hydrogenation. PtO_x not only prevents carrier recombination but also acts as the reaction site for the photocatalytic process.⁴⁰

2.2.2. Pt oxide cocatalysts on WO₃ photocatalysts. In addition to catalyzing H₂ generation, Pt oxide cocatalysts can also act as oxygen (O₂) generation sites on WO₃ photocatalysts, especially in Z-scheme photocatalytic systems.^41–48 In 2017, Qi et al. improved the photocatalytic performance using a PtO_x cocatalyst loaded on WO₃ as an O₂ generation photocatalyst under visible light with ruthenium-modified Ta₃N₅ as a H₂ generation photocatalyst and IO₃⁻/I⁻ as a redox mediator.⁴⁶ Moreover, Miseki et al. reported PtO_x as a cocatalyst for WO₃ photocatalysts treated with a Cs⁺ salt solution followed by H⁺ exchange treatment, achieving 3 times higher activity than that of untreated WO₃ as an O₂ generation photocatalyst. The highest apparent quantum yield of the Cs⁺ and H⁺ comodified PtO_x/WO₃ photocatalysts can be achieved at 420 nm with 20% IO₃⁻ as the redox mediator.⁴⁵ Further, Ma et al. used this PtO_x cocatalyst loaded on Cs⁺- and H⁺-treated WO₃ as a photocatalyst for O₂ generation and successfully carried out visible light-induced Z-scheme water splitting in combination with an oxysulfide H₂ generation photocatalyst.⁴² In addition, Maeda et al. achieved stoichiometric water splitting under visible light in 2014 by using a PtO_x cocatalyst loaded on WO₃ as the O₂ generation photocatalyst in combination with Rh-doped titanate as the H₂ evolution photocatalyst.⁴³ Tsuji et al. used a similar O₂ generation photocatalyst in combination with Ru/SrTiO₃:Rh as the H₂ evolution catalyst and a stable shuttle redox mediator of modified polyformaldehyde to construct two-step photocatalytic water-splitting systems.⁴⁸

2.2.3. Pt oxide cocatalysts on other photocatalysts. In 2012, Ma et al. examined nonsacrificial water oxidation under visible light in an aqueous NaIO₃ solution over WO₃ catalysts modified with Pt oxide cocatalysts. The X-ray photoelectron spectral results indicated that the Pt species in the optimal sample are in the form of metallic Pt and Pt²⁺ oxide, without Pt⁴⁺ species. PtO_x/WO₃ demonstrates enhanced activity to oxidize water molecules, with an efficiency of 14% at 420 nm. Ma et al. found that PtO_x cocatalysts can trap photogenerated carriers to reduce IO₃⁻ ions and oxidize water via electrochemical analyses.⁴⁹ Song et al. prepared Ba₂InTaO₆ combined with 1 wt% PtO_x as a cocatalyst via a solid-state reaction in a mixed atmosphere of H₂ and Ar, which achieves an improved H₂ generation rate of 16.0 μmol h⁻¹ g⁻¹ in an aqueous system. The X-ray photoelectron spectral analysis shows that the Pt species in the cocatalyst is PtO without metallic Pt.⁵⁰ Fu fabricated a novel heterostructure of PtO nanowires grown on ZnO, and the selectivity of the arrays can be tuned by a photocontrolled method. In addition, the enhanced activity of the PtO nanowires on ZnO may be due to the improved kinetics compared to those of the unmodified catalyst.⁵¹

Jiang et al. reported g-C₃N₄ modified with PtO, which is synthesized via a typical photodeposition method, for the H₂ evolution reaction. When a PtO cocatalyst was utilized, the H₂ generation rate increased from 1.6 to 11.5 μmol h⁻¹, which demonstrated that PtO has an excellent ability to serve as an electron sink and provide active sites to lower the reaction barrier of H₂ evolution (Fig. 6).⁵² Due to the surface plasmon resonance of Au and the synergetic action of PtO and Au, the HER activity of g-C₃N₄ is enhanced dramatically (Fig. 6a). The HRTEM image confirms the existence of Au and PtO nanoparticles on the g-C₃N₄ (Fig. 6b). The rate of H₂ evolution in Fig. 6c indicates that the PtO lowers the overpotential for hydrogen evolution by acting as an active site. Wang et al. deposited PtO nanodots with high crystallinity on g-C₃N₄. A series of electronic tests revealed that PtO has enhanced conductivity on g-C₃N₄ compared with metallic Pt. The H₂ production rate is more than 260 μmol h⁻¹, which is approximately 50 times higher than that of metallic Pt as a cocatalyst.⁵³ In addition, Pt oxide cocatalysts can also be loaded onto other photocatalysts, such as KCa₂Nb₃O₁₀,⁵⁴ BiVO₄,⁵⁵ and TaON,^41,56,57 to promote the water-splitting reaction. These cocatalysts can even act as efficient electrocatalysts on carbon substrates for H₂ generation.^58,59

Fig. 6 (a) Schematic mechanism of a PtO cocatalyst as a H₂ evolution site, (b) high-resolution transmission electron microscopy image and (c) photocatalytic H₂ generation rates of PtO cocatalysts loaded on a g-C₃N₄ photocatalyst. Reproduced with permission.⁵² Copyright 2016, Elsevier.

2.3. Tailoring with halogen atoms or substrates

2.3.1. Tailoring with halogen atoms. Pt can catalyze H₂ and O₂ recombination to form water, leading to low performance of photocatalytic H₂ evolution.

To suppress this undesired back reaction, Wang has proven the significant ability of Pt modified with fluorine ions to inhibit the back reaction. The H₂ generation rate of fluorine-modified Pt cocatalysts can be 2.3 times that of unmodified Pt cocatalysts in photocatalytic reactions because fluorine occupies the adsorption sites to prevent adsorption of H₂ on metallic Pt cocatalysts.⁶⁰ In addition, Wang and coworkers have further studied the suppression of H₂ combination with O₂ by halogen elements on Pt-based cocatalysts for photocatalytic water splitting by detailed theoretical and experimental analysis (Fig. 7a–e). Remarkably, simulations of the adsorption energy of H₂ on modified Pt-based cocatalysts were carried out, and the results show the following order of H₂ adsorption energy: F-modified < Cl-modified < I-modified < Br-modified < unmodified cocatalysts. Then they synthesized halogen/Pt–TiO₂ and confirmed the existence of halogen in photocatalysts via high-angle annular dark field scanning transmission electron microscopy (Fig. 7f–i). The suppression of H₂ oxidation follows the same trend (Fig. 7j).⁶¹ Recently, Li et al. successfully inhibited the recombination of O₂ and H₂ by the addition of hemin chloride; the hemin captures O₂ generated from water and transfers it away from the surface of the photocatalyst, and the Cl⁻ ions further inhibit the back-recombination of O₂ and H₂.⁶²

Fig. 7 H₂ adsorption (a–e), scanning transmission electron microscopy images (f–i) and photocatalytic H₂ generation rates (j) of Pt cocatalysts modified using different halogen elements. Reproduced with permission.⁶¹ Copyright 2018, Elsevier.

In a supported metal catalyst, the metal atoms can be adsorbed on the surface of the support and undergo charge transfer with the support. During this process, the electronic structures of the metal catalysts will be tuned, presenting enhanced catalytic activity. In particular, Pt atoms as catalysts tend to adsorb onto the bridge formed by two oxygen atoms in anatase TiO₂, and according to the simulations, this composite structure could increase the activity of the adsorption surface.^63,64 Xing and coworkers prepared a Pt^δ+ cocatalyst on a TiO₂ photocatalyst using a leaching method. In this study, these authors investigated the valence state effects of Pt-based cocatalysts on H₂ generation and demonstrated that the photocatalytic H₂ generation rates can be sensitive to the valence states of Pt but that the size effects are negligible at several nanometers. According to simulations of the H adsorption energy of Pt with different valence states, the poor performance of metallic Pt as a cocatalyst may be due to its too strong H adsorption energy, thus decreasing the H₂ generation activity.⁶⁵ Further, Xing synthesized isolated Pt atoms stabilized by anchoring onto TiO₂, and X-ray absorption fine structure spectra were utilized to observe the valence states of Pt in different samples in this study (Fig. 8). The Fourier transformed spectra of Pt L₃-edge EXAFS of 0.2 Pt/TiO₂ reflect that the Pt atoms in the photocatalyst exist as Pt^δ+, in which the isolated Pt atom is combined with oxygen atoms in the TiO₂ substrate (Fig. 8a and b). The H₂ production rate of the sample was measured to be 169.6 mmol h⁻¹, which is the best performance among the controlled samples⁶⁶ (Fig. 8c). The theoretical calculations demonstrate that the performance of photocatalytic H₂ evolution on the TiO₂ photocatalyst can be optimized through introducing single-atom Pt, and due to the decreased H adsorption energy, the catalytic activity of the isolated Pt atom could be even better than that of the metallic Pt deposited on TiO₂ (Fig. 8d–f).

Fig. 8 (a) Scanning transmission electron microscopy image of a single Pt atom as a cocatalyst. (b) The single Pt atom is combined with oxygen atoms in a TiO₂ substrate as detected by X-ray absorption fine structure spectroscopy. (c) Photocatalytic H₂ generation rates and turnover frequencies of different valence states of Pt-based cocatalysts. (d–f) Simulations of the density of states of bare TiO₂ and single Pt atom-modified TiO₂. Reproduced with permission.⁶⁶ Copyright 2014, Wiley-VCH.

In 2017, Jiang et al. synthesized a photocatalyst comprising Pt loaded on anatase TiO₂ with postmodification by a hydrothermal method. The modified in situ preparation process to load Pt cocatalysts shows an enhanced rate compared to other conventional processes and is a favored method to form positively charged Pt cocatalysts by wet impregnation of TiO₂ photocatalysts in a H₂PtCl₆ solution. The enhanced performance for photocatalytic H₂ generation can be attributed to the existence of Pt^δ+ as a cocatalyst in this system.⁶⁷ In the same year, Ou et al. successfully prepared g-C₃N₄ nanosheets with single Pt atoms loaded through a wet impregnation method. These nanosheets present the best performance for photocatalytic H₂ generation. The X-ray photoelectron spectra reveal that the valence state of Pt is Pt^δ+. Regarding the location of the single Pt atoms in g-C₃N₄, the conduction band of the host photocatalyst can be downshifted, and the photogenerated carriers can migrate faster. The Pt species in the form of Pt^δ+ show the capacity to improve the H₂ generation performance.⁶⁸ Recently, Xue and coworkers reported a highly active and durable single Pt atom as a cocatalyst supported on g-C₃N₄ for photocatalytic H₂ generation. The X-ray photoelectron spectral analysis shows that there exists more deficient-state Pt^δ+ in the reported sample than in compared systems, leading to the most excellent activity of 458.1 mmol h⁻¹ mg_Pt⁻¹.⁶⁹

3. Summary and outlook

Suitable cocatalysts are of great importance for obtaining high-efficiency photocatalysts for water splitting. Pt-based cocatalysts can trap photogenerated carriers and act as active sites to lower the activation barriers of water-splitting reactions, thus serving as the most efficient cocatalysts for photocatalytic H₂ evolution. In this review, the positively charged Pt-based cocatalysts for photocatalytic water splitting that have been proposed in recent years have been discussed. Benefitting from the unique characteristics of different valence states of Pt, such as the optimized adsorption of H atoms and suppression capacity for the H₂ oxidation reaction, Pt-based cocatalysts with a positive valence state have been considered ideal reaction sites on semiconductor-based photocatalysts, including metal oxides, sulfides, nitrides and their composites. The valence states of Pt-based cocatalysts can be tailored using oxygen atoms, halogen atoms and the surface atoms of loading substrates. Moreover, the Pt valence states can be rationally designed via ligand-assisted chemical reduction, hydrothermal, template and wet impregnation methods. With tuned valence states, the back reaction of Pt-based cocatalysts can be suppressed, and the forward H₂ evolution reaction can be further promoted.

Although significant advances have been made in Pt-based cocatalysts in the field of water splitting, further work is urgently needed. Structural changes are always involved in small Pt-based cocatalysts during photocatalytic reactions, and a lack of precise control of atomic structures and valence states will undoubtedly decrease the water splitting performance. Therefore, the fabrication of Pt-based cocatalysts with target structures on the nanoscale or even the atomic level as well as the valence states must be achieved, and high-quality transient operando techniques to track the structural changes of cocatalysts during water splitting must be developed. The loading location of cocatalysts has shown its superiority in promoting photocatalytic activity for overall water splitting, but the detailed mechanism is not yet understood, and an in-depth investigation combining experimental studies with corresponding simulations should be carried out. Moreover, most reported Pt-based cocatalysts are aimed at the half reaction of H₂ evolution, in which sacrificial reagents are commonly required. The final goal to produce H₂ as a solar fuel should be realized by the fabrication of high-efficiency photocatalysts that split pure water into H₂ and O₂ without added sacrificial reagents. Thus, positively charged Pt-based cocatalysts that can inhibit H₂ oxidation are highly desired for overall water splitting.

In this review, we provide a concise survey of the valence state effects of Pt-based cocatalysts for photocatalytically splitting water into H₂ products. By precisely tailoring the atomic structures and valence states, positively charged Pt-based cocatalysts can determine the direction and efficiency of the water-splitting reaction and thus make promising contributions to photocatalysis.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (21808061, 21838003, 91834301), the Shanghai Scientific and Technological Innovation Project (18JC1410500, 18DZ2252400, 19JC1410400), the “Fundamental Research Funds for the Central Universities” (222201714001, 222201718002), the Shanghai Sailing Program (17YF1402900), the “Chen Guang” project supported by the Shanghai Municipal Education Commission and the Shanghai Education Development Foundation (16CG31).

References

1. A. Fujishima and K. Honda, Nature, 1972, 238, 37–38.
2. X. Chen, S. Shen, L. Guo and S. S. Mao, Chem. Rev., 2010, 110, 6503–6570.
3. S. Chen, T. Takata and K. Domen, Nat. Rev. Mater., 2017, 2, 17050.
4. T. Hisatomi, J. Kubota and K. Domen, Chem. Soc. Rev., 2014, 43, 7520–7535.
5. Z. Wang, C. Li and K. Domen, Chem. Soc. Rev., 2019, 48, 2109–2125.
6. D. Ravelli, D. Dondi, M. Fagnoni and A. Albini, Chem. Soc. Rev., 2009, 38, 1999–2011.
7. Y. Qu and X. Duan, Chem. Soc. Rev., 2013, 42, 2568–2580.
8. B. Mei, K. Han and G. Mul, ACS Catal., 2018, 8, 9154–9164.
9. Y. Moriya, T. Takata and K. Domen, Coord. Chem. Rev., 2013, 257, 1957–1969.
10. Y. Kuang, T. Yamada and K. Domen, Joule, 2017, 1, 290–305.
11. J. Ran, J. Zhang, J. Yu, M. Jaroniec and S. Z. Qiao, Chem. Soc. Rev., 2014, 43, 7787–7812.
12. K. Maeda and K. Domen, J. Phys. Chem. Lett., 2010, 1, 2655–2661.
13. J. Yang, D. Wang, H. Han and C. Li, Acc. Chem. Res., 2013, 46, 1900–1909.
14. C.-F. Fu, X. Wu and J. Yang, Adv. Mater., 2018, 30, 1802106.
15. S. Bai, W. Yin, L. Wang, Z. Li and Y. Xiong, RSC Adv., 2016, 6, 57446–57463.
16. N. Bao, L. Shen, T. Takata and K. Domen, Chem. Mater., 2008, 20, 110–117.
17. J. R. Darwent and G. Porter, J. Chem. Soc., Chem. Commun., 1981, 145–146.
18. H. Yan, J. Yang, G. Ma, G. Wu, X. Zong, Z. Lei, J. Shi and C. Li, J. Catal., 2009, 266, 165–168.
19. D. Y. C. Leung, X. Fu, C. Wang, M. Ni, M. K. H. Leung, X. Wang and X. Fu, ChemSusChem, 2010, 3, 681–694.
20. S. Trasatti, J. Electroanal. Chem. Interfacial Electrochem., 1972, 39, 163–184.
21. A. Kudo and Y. Miseki, Chem. Soc. Rev., 2009, 38, 253–278.
22. R. Subbaraman, D. Tripkovic, D. Strmcnik, K.-C. Chang, M. Uchimura, A. P. Paulikas, V. Stamenkovic and N. M. Markovic, Science, 2011, 334, 1256.
23. X. Wang, K. Maeda, A. Thomas, K. Takanabe, G. Xin, J. M. Carlsson, K. Domen and M. Antonietti, Nat. Mater., 2008, 8, 76.
24. J. Xing, W. Q. Fang, H. J. Zhao and H. G. Yang, Chem.–Asian J., 2012, 7, 642–657.
25. M. Yoshida, A. Yamakata, K. Takanabe, J. Kubota, M. Osawa and K. Domen, J. Am. Chem. Soc., 2009, 131, 13218–13219.
26. H. B. Gray, Nat. Chem., 2009, 1, 7.
27. W. R. Grove, The London, Edinburgh, and Dublin Philosophical Magazine and Journal of Science, 1839, 14, 127–130.
28. R. Abe, K. Sayama and H. Arakawa, Chem. Phys. Lett., 2003, 371, 360–364.
29. K. Maeda, K. Teramura, D. Lu, N. Saito, Y. Inoue and K. Domen, Angew. Chem., Int. Ed., 2006, 45, 7806–7809.
30. Y. Hang Li, J. Xing, Z. Jia Chen, Z. Li, F. Tian, L. Rong Zheng, H. Feng Wang, P. Hu, H. Jun Zhao and H. Gui Yang, Nat. Commun., 2013, 4, 2500.
31. Y. H. Li, C. Peng, S. Yang, H. F. Wang and H. G. Yang, J. Catal., 2015, 330, 120–128.
32. Y. H. Li, C. Z. Li and H. G. Yang, J. Mater. Chem. A, 2017, 5, 20631–20634.
33. X. N. Ren, Z. Y. Hu, J. Jin, L. Wu, C. Wang, J. Liu, F. Liu, M. Wu, Y. Li, G. Van Tendeloo and B. L. Su, ACS Appl. Mater. Interfaces, 2017, 9, 29687–29698.
34. N. R. Manwar, A. A. Chilkalwar, K. K. Nanda, Y. S. Chaudhary, J. Subrt, S. S. Rayalu and N. K. Labhsetwar, ACS Sustainable Chem. Eng., 2016, 4, 2323–2332.
35. B. S. Huang, F. Y. Chang and M. Y. Wey, Int. J. Hydrogen Energy, 2010, 35, 7699–7705.
36. J. Xing, H. B. Jiang, J. F. Chen, Y. H. Li, L. Wu, S. Yang, L. R. Zheng, H. F. Wang, P. Hu, H. J. Zhao and H. G. Yang, J. Mater. Chem. A, 2013, 1, 15258–15264.
37. S. K. Parayil, H. S. Kibombo, C. M. Wu, R. Peng, T. Kindle, S. Mishra, S. P. Ahrenkiel, J. Baltrusaitis, N. M. Dimitrijevic, T. Rajh and R. T. Koodali, J. Phys. Chem. C, 2013, 117, 16850–16862.
38. 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.
39. Y. L. Sui, S. B. A. Liu, T. F. Li, Q. X. Liu, T. Jiang, Y. F. Guo and J. L. Luo, J. Catal., 2017, 353, 250–255.
40. E. Talas, Z. Paszti, L. Korecz, A. Domjan, P. Nemeth, G. P. Szijjarto, J. Mihaly and A. Tompos, Catal. Today, 2018, 306, 71–80.
41. R. Abe, M. Higashi and K. Domen, ChemSusChem, 2011, 4, 228–237.
42. G. Ma, S. Chen, Y. Kuang, S. Akiyama, T. Hisatomi, M. Nakabayashi, N. Shibata, M. Katayama, T. Minegishi and K. Domen, J. Phys. Chem. Lett., 2016, 7, 3892–3896.
43. K. Maeda, ACS Appl. Mater. Interfaces, 2014, 6, 2167–2173.
44. K. Maeda, D. L. Lu and K. Domen, ACS Catal., 2013, 3, 1026–1033.
45. Y. Miseki, S. Fujiyoshi, T. Gunji and K. Sayama, Catal. Sci. Technol., 2013, 3, 1750–1756.
46. Y. Qi, S. S. Chen, M. R. Li, Q. Ding, Z. Li, J. Y. Cui, B. B. Dong, F. X. Zhang and C. Li, Chem. Sci., 2017, 8, 437–443.
47. Z. Song, T. Hisatomi, S. Chen, Q. Wang, G. Ma, S. Li, X. Zhu, S. Sun and K. Domen, ChemSusChem, 2019, 12, 1906–1910.
48. K. Tsuji, O. Tomita, M. Higashi and R. Abe, ChemSusChem, 2016, 9, 2201–2208.
49. S. S. K. Ma, K. Maeda, R. Abe and K. Domen, Energy Environ. Sci., 2012, 5, 8390–8397.
50. K. Song, J. Yang, P. F. Jiang, W. L. Gao, R. H. Cong and T. Yang, Eur. J. Inorg. Chem., 2015, 5786–5792.
51. S. R. Fu, B. B. Zhang, H. Y. Hu, Y. J. Zhang and Y. P. Bi, Catal. Sci. Technol., 2018, 8, 2789–2793.
52. J. Jiang, J. G. Yu and S. W. Cao, J. Colloid Interface Sci., 2016, 461, 56–63.
53. C. Wang, H. Q. Fan, X. H. Ren, Y. Wen and W. J. Wang, Appl. Surf. Sci., 2018, 462, 423–431.
54. H. Suzuki, O. Tomita, M. Higashi and R. Abe, Catal. Sci. Technol., 2015, 5, 2640–2648.
55. J. Yan, H. Wu, H. Chen, Y. Zhang, F. Zhang and S. F. Liu, Appl. Catal., B, 2016, 191, 130–137.
56. S. Chen, Y. Qi, T. Hisatomi, Q. Ding, T. Asai, Z. Li, S. S. K. Ma, F. Zhang, K. Domen and C. Li, Angew. Chem., Int. Ed., 2015, 54, 8498–8501.
57. H. Hagiwara, M. Nagatomo, C. Seto, S. Ida and T. Ishiahara, Catalysts, 2013, 3, 614–624.
58. X. Cheng, Y. H. Li, L. R. Zheng, Y. Yan, Y. F. Zhang, G. Chen, S. R. Sun and J. J. Zhang, Energy Environ. Sci., 2017, 10, 2450–2458.
59. M. K. Kundu, T. Bhowmik, R. Mishra and S. Barman, ChemSusChem, 2018, 11, 2388–2401.
60. M. Wang, Z. Li, Y. Q. Wu, J. T. Ma and G. X. Lu, J. Catal., 2017, 353, 162–170.
61. M. Wang, W. L. Zhen, B. Tian, J. T. Ma and G. X. Lu, Appl. Catal., B, 2018, 236, 240–252.
62. Z. Li, B. Tian, W. L. Zhen, Y. Q. Wu and G. X. Lu, Appl. Catal., B, 2017, 203, 408–415.
63. C. Jin, Y. Dai, W. Wei, X. C. Ma, M. M. Li and B. B. A. Huang, Appl. Surf. Sci., 2017, 426, 639–646.
64. F. S. Han, Z. H. Zhou, X. H. Zhang, Z. X. Huang, M. T. Li and L. J. Guo, J. Phys. Chem. C, 2018, 122, 2546–2553.
65. J. Xing, Y. H. Li, H. B. Jiang, Y. Wang and H. G. Yang, Int. J. Hydrogen Energy, 2014, 39, 1237–1242.
66. J. Xing, J. F. Chen, Y. H. Li, W. T. Yuan, Y. Zhou, L. R. Zheng, H. F. Wang, P. Hu, Y. Wang, H. J. Zhao, Y. Wang and H. G. Yang, Chem.–Eur. J., 2014, 20, 2138–2144.
67. Z. Jiang, M. A. Isaacs, Z. W. Huang, W. F. Shangguan, Y. F. Deng and A. F. Lee, ChemCatChem, 2017, 9, 4268–4274.
68. M. Ou, S. P. Wan, Q. Zhong, S. L. Zhang and Y. A. Wang, Int. J. Hydrogen Energy, 2017, 42, 27043–27054.
69. Y. Xue, Y. G. Lei, X. Y. Liu, Y. N. Li, W. N. Deng, F. Wang and S. X. Min, New J. Chem., 2018, 42, 14083–14086.

---