A triple-functional Co₂SnO₄-enabled S-scheme heterojunction with photothermal promotion for efficient solar-driven hydrogen evolution

Abstract

Photocatalytic hydrogen production offers a sustainable route for solar energy conversion, yet its efficiency is hindered by rapid charge recombination, limited light absorption and slow reaction kinetics. In this study, we introduce a strategically engineered Co₂SnO₄@ZnIn₂S₄ S-scheme heterojunction photocatalyst that synergistically enhances charge separation, photothermal conversion and catalytic activity to overcome these challenges. Using a simple solvent self-assembly method, we develop a heterostructure in which Co₂SnO₄ nanoparticles serve as a versatile component by: (1) extending light absorption into the near-infrared (800–1400 nm) range for efficient solar-to-thermal conversion, (2) creating an S-scheme charge transfer pathway that enhances electron–hole separation while retaining high redox potentials, and (3) providing numerous active sites to accelerate proton reduction kinetics. The optimized photocatalyst achieves an impressive hydrogen evolution rate of 12.56 mmol g⁻¹ h⁻¹ and an apparent quantum efficiency of 12.96% at 420 nm. Comprehensive experimental and theoretical analyses validate the S-scheme charge transfer mechanism and quantify the photothermal and catalytic contributions to reaction enhancement. This work not only demonstrates a highly efficient photocatalytic system but also provides critical insights into designing multifunctional heterostructures for solar fuel generation, paving the way for practical renewable energy applications.

---

Xianqiang Xiong

Xianqiang Xiong is currently an Associate Professor at Taizhou University. He received his BS in Applied Chemistry from Central South University in 2013 and his PhD in Chemistry from Zhejiang University in 2018. His research focuses on energy-related small molecule conversion and environmental pollutant remediation, primarily through photocatalytic and photoelectrocatalytic approaches. Dr Xiong's work aims to develop efficient catalytic materials and systems for sustainable energy and environmental applications.

---

Introduction

Photocatalytic hydrogen production is increasingly recognized as a key approach for sustainably harnessing solar energy to generate clean fuel sources.^1–4 This technique uses semiconductor materials to facilitate water splitting, converting abundant solar energy into hydrogen, which is a storable form of chemical energy known for its carbon-free nature and high energy density.^5,6 However, the practical application of photocatalytic hydrogen production faces several challenges. A major issue is the short lifetime of photogenerated charge carriers, which rapidly recombine on femtosecond to nanosecond timescales, outpacing the slower kinetics of catalytic reactions such as water reduction and oxidation.^7–9 This discrepancy in timescales limits the availability of charge carriers for redox processes, resulting in low solar-to-hydrogen conversion efficiencies.^10,11 These fundamental challenges hinder the scalability and commercial feasibility of photocatalytic systems, underscoring the urgent need for innovative strategies to extend charge carrier lifetimes, accelerate interfacial charge transfer, and improve light absorption in order to fully harness the potential of this promising technology.^12–14

As a prominent ternary sulfide semiconductor, the two-dimensional (2D) layered ZnIn₂S₄ has recently attracted significant attention in the field of photocatalytic hydrogen evolution.^15–20 This is primarily due to its favorable properties, including low toxicity, a tunable bandgap ranging from 2.1 to 2.6 eV, a distinctive electronic structure and exceptional photochemical stability.^21,22 These attributes position ZnIn₂S₄ as a promising candidate for solar-driven hydrogen production. However, its practical utility is significantly hindered by several intrinsic limitations, including the rapid recombination of photogenerated electron–hole pairs, suboptimal light absorption efficiency, and inefficient charge separation and transport processes.²³ These issues pose significant barriers to fully realizing the photocatalytic potential of ZnIn₂S₄. To address these challenges, numerous research efforts have focused on enhancing ZnIn₂S₄-based photocatalysts through various strategies, including morphology control,²⁴ cation exchange,²⁵ elemental doping,²⁶ defect engineering,²⁷ co-catalyst loading²⁸ and heterojunction construction.²⁹ Among these, the development of S-scheme heterojunctions stands out as particularly promising. S-scheme heterojunctions typically consist of an oxidative-type semiconductor with a higher valence band (VB) potential and a reductive-type semiconductor with a higher conduction band (CB) potential.^30,31 When semiconductors with different Fermi levels interact, electron transfer occurs at the interface until Fermi level equilibration, resulting in an internal electric field (IEF) and band bending. Upon optical excitation, the IEF and band bending promote the selective recombination of low-energy photogenerated electrons and holes, while conserving their high-energy counterparts for redox reactions.³² This electron transfer mechanism aligns with thermodynamic and kinetic principles and has been substantiated in numerous recent studies. Several ZnIn₂S₄-based S-scheme heterojunctions, including TpBD COF@ZnIn₂S₄,³³ ZnIn₂S₄/ZnSe³⁴ and NiCo₂S₄@C/ZnIn₂S₄,³⁵ have demonstrated enhanced photocatalytic hydrogen evolution activities. Despite these advancements, the potential impact of external stimuli, such as thermal, magnetic, acoustic and electrical fields, on modulating photocatalytic reactions remains relatively unexplored. These external factors could potentially enhance charge separation, modify reaction kinetics and improve the overall catalytic efficiency.

In recent years, the incorporation of photothermal effects into photocatalytic systems has emerged as an innovative and transformative strategy to enhance both solar energy utilization and catalytic performance.³⁶ The photothermal effect, which involves the conversion of near-infrared (NIR) light (ranging from 800 to 2500 nm and constituting approximately 53% of the solar spectrum) into thermal energy, offers a unique opportunity to overcome the limitations associated with traditional photocatalysts, such as ZnIn₂S₄.³⁷ By harnessing the photothermal effect, the localized temperature of the photocatalyst can be significantly increased. This elevation of temperature accelerates reaction kinetics, enhances charge carrier mobility and facilitates the desorption of reaction intermediates, all of which are essential for improving photocatalytic hydrogen evolution.³⁸ Recent studies have highlighted that integrating ZnIn₂S₄ with photothermal materials, such as MoS₂,³⁹ CoN⁴⁰ and CdTe,⁴¹ can synergistically enhance its photocatalytic capabilities. Consequently, using NIR-induced photothermal effects to enhance hydrogen evolution in ZnIn₂S₄-based photocatalytic composite systems shows substantial promise for advancing solar energy utilization.

Transition metal oxides have emerged as promising candidates for enhancing ZnIn₂S₄-based photocatalytic systems due to their exceptional light absorption, efficient charge transport properties and superior structural stability.⁴² Among these materials, spinel-type Co₂SnO₄ is particularly notable for its unique combination of advantageous physicochemical properties:^43,44 (1) a narrow bandgap that allows for broad-spectrum light absorption in the near-infrared region (800–1400 nm), (2) high electron mobility that enables rapid charge transfer, and (3) a robust spinel structure with abundant exposed metal sites, ideal for catalytic reactions. These intrinsic properties, along with its chemical durability and cost-effectiveness, position Co₂SnO₄ as a versatile functional material in energy applications, including solar cells and electrocatalysis. Building on these attributes, we propose that integrating Co₂SnO₄ with ZnIn₂S₄ could create a multifaceted photocatalytic system that addresses key challenges in photocatalytic hydrogen evolution. First, forming an S-scheme heterojunction takes advantage of their complementary band structures—ZnIn₂S₄ (CB: −0.79 to −1.20 V; VB: 1.47–1.70 V) as the oxidation semiconductor and Co₂SnO₄ (CB: −0.52 to −1.36 V; VB: 0.25–1.10 V) as the reduction semiconductor, generating an internal electric field for efficient charge separation while retaining high redox potential. Second, the broad-spectrum absorption of Co₂SnO₄, particularly in the NIR region, enables photothermal conversion, generating localized heating that enhances light absorption and accelerates reaction kinetics. Third, Co₂SnO₄ acts as an active co-catalyst, with its spinel structure providing ample active sites that significantly enhance proton reduction kinetics at the interface. However, the strategic integration of these materials to simultaneously utilize S-scheme charge transfer, photothermal conversion and catalytic enhancement remains unexplored, warranting further investigation.

Herein, we demonstrate a simple solvent self-assembly strategy for the rational construction of Co₂SnO₄@ZnIn₂S₄ S-scheme heterojunction photocatalysts with integrated photothermal functionality. Systematic characterization reveals that the incorporated Co₂SnO₄ nanoparticles serve multiple critical functions: (1) forming an interfacial electric field that drives S-scheme charge transfer, as confirmed by combined experimental studies and DFT calculations; (2) extending light absorption to the NIR region for efficient solar-to-thermal conversion, which raises local reaction temperatures and reduces activation barriers; and (3) providing abundant catalytically active sites that accelerate proton reduction kinetics. Photothermal measurements quantitatively demonstrate the temperature-dependent enhancement of photocatalytic activity. The optimized heterostructure achieves exceptional hydrogen evolution performance (12.56 mmol g⁻¹ h⁻¹) and an apparent quantum efficiency of 12.96% at 420 nm, marking a significant advancement over current ZnIn₂S₄-based systems. Through comprehensive mechanistic investigations, we elucidate the synergistic interplay between the S-scheme charge transfer pathway and photothermal effects, offering new fundamental insights into the design of high-efficiency solar-driven photocatalysts.

Results and discussion

Synthetic and characterization

The Co₂SnO₄@ZnIn₂S₄ heterojunction photocatalysts were synthesized using a simple solvent self-assembly approach. This method involved the controlled mixing of Co₂SnO₄ nanoparticles with ZnIn₂S₄ nanosheets in aqueous solution, followed by filtration and washing (Fig. 1a). As shown in Fig. S1,† the zeta potential of ZnIn₂S₄ was measured to be −13.90 mV, while Co₂SnO₄ exhibited a positive value of +7.30 mV under the same conditions. This significant contrast in surface charge confirms the presence of strong electrostatic attraction between the two components, which serves as the primary driving force for their self-assembly into heterojunctions. This method enables the precise integration of Co₂SnO₄ nanoparticles (5–30 wt%) with ZnIn₂S₄ microflowers while preserving the structural integrity of both components. X-ray diffraction (XRD) analysis was used to verify the crystalline phases and structural purity of the synthesized materials. As shown in Fig. 1b, the diffraction patterns of the precursors and final products exhibit excellent agreement with reference standards: Co₂Sn(OH)₄ (JCPDS no. 13-356, Fig. S2†), spinel-type Co₂SnO₄ (JCPDS no. 01-1137),⁴⁵ and hexagonal ZnIn₂S₄ (JCPDS no. 65-2023).⁴⁶ The progressive enhancement of the characteristic Co₂SnO₄ peak at 36.7° (222 plane) with increasing Co₂SnO₄ content confirms the tunable composition of the heterostructures. Importantly, the preservation of all fundamental diffraction peaks without band shifting or the appearance of new phases demonstrates that our synthetic approach maintains the intrinsic crystal structures of both components while enabling their effective integration into heterojunctions.

Fig. 1 (a) Schematic diagram of the preparation processes of the Co₂SnO₄@ZnIn₂S₄ heterojunction; (b) XRD patterns and (c) N₂ adsorption–desorption isotherms of Co₂SnO₄, ZnIn₂S₄ and the x%-Co₂SnO₄@ZnIn₂S₄ heterojunction.

The textural properties of the photocatalysts were systematically investigated using N₂ physisorption measurements, as surface characteristics play a crucial role in photocatalytic performance. Fig. 1c and S3† present the nitrogen adsorption–desorption isotherms and corresponding pore size distribution curves for Co₂SnO₄, ZnIn₂S₄ and the 20%-Co₂SnO₄@ZnIn₂S₄ heterojunction, with quantitative parameters summarized in Table S1.† All samples exhibit type IV isotherms with distinct H3 hysteresis loops, characteristic of mesoporous materials with slit-shaped pores.⁴⁷ The parent ZnIn₂S₄ demonstrates the highest surface area (102.36 m² g⁻¹) and pore volume (0.462 cm³ g⁻¹), while the 20%-Co₂SnO₄@ZnIn₂S₄ composite shows intermediate values (94.59 m² g⁻¹ and 0.443 cm³ g⁻¹, respectively). Notably, the composite exhibits an increased average pore diameter (16.46 nm) compared to both individual components (14.71 nm for Co₂SnO₄ and 15.71 nm for ZnIn₂S₄). This textural evolution suggests that the Co₂SnO₄ nanoparticles preferentially decorate the surface of ZnIn₂S₄ microflowers, partially blocking smaller pores while creating additional mesoporous channels at the heterojunction interfaces. The well-preserved mesoporosity in the heterostructure, coupled with its substantial surface area, provides sufficient active sites for photocatalytic reactions while facilitating the mass transport of reactants and products.

Advanced electron microscopy techniques were employed to elucidate the nanoscale architecture and interfacial characteristics of the synthesized photocatalysts. As revealed by SEM analysis (Fig. 2a), pristine ZnIn₂S₄ exhibits a highly porous, three-dimensional microflower morphology composed of interconnected ultrathin nanosheets, providing an ideal framework for light harvesting and mass transport. In contrast, Co₂SnO₄ nanoparticles exhibit an aggregated morphology with irregular shapes and sizes (Fig. 2b). The 20%-Co₂SnO₄@ZnIn₂S₄ composite preserves the structural integrity of the ZnIn₂S₄ microflowers while demonstrating uniform decoration of Co₂SnO₄ nanoparticles across the nanosheet surfaces, as clearly shown in both SEM and TEM images (Fig. 2c and d).

Fig. 2 SEM images of (a) ZnIn₂S₄, (b) Co₂SnO₄ and (c) the 20%-Co₂SnO₄@ZnIn₂S₄ composite; (d) TEM, (e) HRTEM, (f) SAED and (g–m) elemental mapping images of the 20%-Co₂SnO₄@ZnIn₂S₄ composite.

The high-resolution TEM image (Fig. 2e) provides atomic-scale evidence of the heterojunction formation, revealing distinct lattice fringes with spacings of 0.499 nm and 0.261 nm corresponding to the (111) and (311) planes of spinel Co₂SnO₄, respectively, along with a 0.293 nm spacing attributed to the (104) plane of hexagonal ZnIn₂S₄. The crystalline nature is further confirmed by selected area electron diffraction (SAED, Fig. 2f), which shows well-defined diffraction rings indexed to the (104) and (102) planes of ZnIn₂S₄ and the (311) plane of Co₂SnO₄. Energy-dispersive X-ray spectroscopy (EDS) elemental mapping (Fig. 2g–m) reveals the homogeneous distribution of all constituent elements (Zn, In, S, Co, Sn, and O) throughout the heterostructure, with Co and Sn signals precisely colocalized with the nanoparticle positions. The observed morphology, where Co₂SnO₄ nanoparticles are anchored onto the hierarchical ZnIn₂S₄ framework, creates an ideal architecture for charge transfer while maintaining excellent light penetration and reactant accessibility throughout the photocatalyst.

Photocatalytic performance

The photocatalytic hydrogen evolution activity of the synthesized materials was systematically evaluated under full-spectrum irradiation (300 W Xe lamp) with 20 wt% triethanolamine (TEOA) as the sacrificial agent, maintaining a constant reaction temperature of 6 °C through water circulation. As shown in Fig. 3a and S4,† the 20%-Co₂SnO₄@ZnIn₂S₄ heterojunction exhibited exceptional photocatalytic activity, achieving a hydrogen evolution rate of 12.56 mmol g⁻¹ h⁻¹, which represents a 7.61-fold enhancement compared to pristine ZnIn₂S₄ (1.65 mmol g⁻¹ h⁻¹). The composition-activity relationship revealed an optimal loading of 20% Co₂SnO₄, with both lower (10%) and higher (30%) loadings resulting in diminished performance (Fig. 3a).

Fig. 3 (a) Photocatalytic H₂ evolution rate of Co₂SnO₄, ZnIn₂S₄ and x%-Co₂SnO₄@ZnIn₂S₄; (b) average hydrogen generation rate of 20%-Co₂SnO₄@ZnIn₂S₄ with various sacrificial agents; (c) AQE of 20%-Co₂SnO₄@ZnIn₂S₄ at 420 nm; (d) recycling performance of H₂ evolution over the 20%-Co₂SnO₄@ZnIn₂S₄ heterojunction; EPR spectra of (e) DMPO-˙OH and (f) DMPO-˙O₂⁻.

This volcano-type trend arises from two competing factors: insufficient Co₂SnO₄ coverage leads to incomplete heterojunction formation, while excessive loading results in nanoparticle aggregation, both of which compromise charge separation efficiency. Furthermore, the preparation method significantly influences performance, with the solvent self-assembly approach yielding superior results compared to physical mixing (Fig. S5†), highlighting the critical importance of intimate interfacial contact for effective charge transfer. Notably, the 20%-Co₂SnO₄@ZnIn₂S₄ system exhibited comparable performance to 1%-Pt@ZnIn₂S₄ and significantly outperformed 1%-Au@ZnIn₂S₄ (Fig. S6†), highlighting the potential of Co₂SnO₄ as a cost-effective alternative to noble metal cocatalysts. This exceptional activity is further reflected in the apparent quantum efficiency (AQE) of 12.96% at 420 nm (Fig. 3c), surpassing most reported ZnIn₂S₄-based systems (Table S2†).

The choice of the sacrificial agent significantly influenced the photocatalytic performance (Fig. 3b and S7†). TEOA emerged as the most effective donor, yielding 12.56 mmol g⁻¹ h⁻¹, significantly higher than lactic acid (8.83), adipic acid (3.86), Na₂S/Na₂SO₃ (3.78), ethylene glycol (3.42), or methanol (<1 mmol g⁻¹ h⁻¹). This hierarchy correlates with the electron-donating capacity and adsorption strength of each reagent on the catalyst surface, with TEOA's alkaline nature likely enhancing its interaction with the photocatalyst.⁴⁸ Long-term stability tests demonstrated excellent durability, with only a 3.38% loss in activity after five consecutive 2 h cycles (10 h total, Fig. 3d). Interestingly, the activity increased in cycles 2–3, possibly due to surface restructuring or intermediate accumulation that enhances charge transfer. The catalyst demonstrates excellent structural and chemical stability, as evidenced by nearly identical SEM and XPS spectra before and after the reaction (Fig. S8 and S9†). However, the XRD pattern (Fig. S10†) revealed a slight intensity increase for the ZnIn₂S₄ (102) peak at 30.2° after the reaction. This selective enhancement may originate from light-induced crystallinity improvement or surface reconstruction during testing, as commonly observed in sulfide photocatalysts under operational conditions.⁴⁹ Importantly, no new phases emerged, confirming the structural integrity of the Co₂SnO₄@ZnIn₂S₄ heterojunction throughout the stability test.

Electron paramagnetic resonance (EPR) spectroscopy provided direct evidence of enhanced charge separation in the heterostructure system. Under light irradiation, distinct signals corresponding to hydroxyl radicals (˙OH) and superoxide radicals (˙O₂⁻) were observed for ZnIn₂S₄ and 20%-Co₂SnO₄@ZnIn₂S₄, while no radical signals were detected in the dark (Fig. 3e and f). This confirms the photo-induced nature of the radical generation processes. The 20%-Co₂SnO₄@ZnIn₂S₄ heterojunction exhibited stronger radical signals, highlighting two key advantages of the heterostructure design: (1) significantly enhanced generation and separation of photo-induced charge carriers, and (2) more efficient utilization of these carriers in redox reactions. The strong ˙O₂⁻ signal confirms effective electron transfer to adsorbed O₂, while the pronounced ˙OH signal indicates efficient hole accumulation at the appropriate energy levels for water oxidation. The simultaneous enhancement of both ˙OH and reductive ˙O₂⁻ radical generation strongly suggests that the heterojunction not only promotes charge separation but also preserves the strong redox potentials of both components, which is a defining feature of S-scheme systems.⁵⁰ This dual enhancement explains the superior photocatalytic activity observed in hydrogen evolution experiments.

Elucidation of the S-scheme charge transfer mechanism

To establish the fundamental band structure required for S-scheme heterojunction formation, we first conducted systematic Mott–Schottky measurements (Fig. 4a). The characteristic positive slopes confirmed the n-type nature of both semiconductors, with flat-band potentials of −1.09 V (ZnIn₂S₄) and −1.36 V (Co₂SnO₄) vs. Ag/AgCl.⁵¹ Following the well-established relationship for n-type semiconductors, where the conduction band minimum (E_CB) typically lies ∼0.2 eV below the flat-band potential (E_fb), we determined the E_CB positions to be −1.29 V and −1.56 V for ZnIn₂S₄ and Co₂SnO₄, respectively. Complementary Tauc plot analysis (Fig. S11†) yielded bandgap values of 2.37 eV (ZnIn₂S₄) and 1.44 eV (Co₂SnO₄), which enabled calculation of VB positions at 1.08 V and −0.12 V. This staggered type-II alignment, where the bands of Co₂SnO₄ are positioned at higher energies than the corresponding bands of ZnIn₂S₄, satisfies the essential electronic structure requirements for the S-scheme operation.

Fig. 4 (a) Mott–Schottky (M–S) plots of Co₂SnO₄ and ZnIn₂S₄ at different frequencies; DFT-calculated work functions of (b) Co₂SnO₄ and (c) ZnIn₂S₄; high-resolution XPS and in situ irradiated high-resolution XPS spectra of Co₂SnO₄, ZnIn₂S₄ and 20%-Co₂SnO₄@ZnIn₂S₄ samples: (d) Zn 2p, (e) In 3d, (f) S 2p, (g) Co 2p, (h) Sn 3d, and (i) O 1s.

Having established the band positions, we next investigated the interfacial electronic structure through density functional theory (DFT) calculations (Fig. 4b and c). The calculated work functions (Φ) revealed a substantial 0.23 eV difference between Co₂SnO₄ (6.32 eV) and ZnIn₂S₄ (6.55 eV), corresponding to Fermi level positions of −6.32 eV and −6.55 eV relative to the vacuum level, respectively.⁵² The work functions of ZnIn₂S₄ and Co₂SnO₄ were experimentally determined via ultraviolet photoelectron spectroscopy (UPS, Fig. S12†). The measured Φ values of 6.09 eV for ZnIn₂S₄ and 6.14 eV for Co₂SnO₄ exhibit excellent agreement with the corresponding DFT calculations, further validating the electronic structure alignment between theory and experiment. This significant offset induces spontaneous electron transfer from Co₂SnO₄ to ZnIn₂S₄ upon contact, resulting in three key effects: (i) upward band bending in Co₂SnO₄, (ii) downward band bending in ZnIn₂S₄ and (iii) the formation of a strong IEF.

To rigorously validate the predicted interfacial charge transfer, we performed high-resolution XPS analysis (Fig. 4d–i). An increase in binding energy typically indicates a decrease in electron density, suggesting electron depletion from the atom. As a result, shifts in binding energy provide direct insight into the direction of carrier migration within the photocatalyst. In this study, as shown in Fig. 4d, the high-resolution Zn 2p spectrum of pristine ZnIn₂S₄ can be deconvoluted into two peaks at 1045.12 eV and 1022.08 eV, corresponding to Zn 2p_1/2 and Zn 2p_3/2 of Zn²⁺, respectively. The high-resolution In 3d XPS spectrum (Fig. 4e) reveals the In 3d_3/2 and In 3d_5/2 peaks of In³⁺ at binding energies of 452.63 eV and 445.10 eV, respectively. Fig. 4f shows the S 2p spectrum of ZnIn₂S₄ with two distinct peaks at 163.08 eV and 161.84 eV, corresponding to S 2p_1/2 and S 2p_3/2, confirming the presence of S²⁻ states.⁵³ Meanwhile, Fig. 4g presents the deconvoluted Co 2p spectrum of Co₂SnO₄, with fitted peaks at 795.31 eV and 780.04 eV assigned to Co 2p_1/2 and Co 2p_3/2, along with two satellite peaks at 804.10 eV and 788.69 eV. The high-resolution Sn 3d spectrum of Co₂SnO₄ (Fig. 4h) shows two characteristic peaks at 494.85 eV and 288.7 eV, corresponding to Sn 3d_3/2 and Sn 3d_5/2. Additionally, three distinct peaks at 531.78 eV, 530.37 eV and 529.96 eV appear in the O 1s spectrum (Fig. 4i), corresponding to absorbed water, in the form of hydroxyl groups (OH⁻), oxygen vacancies (O-vacancies) and metal–oxide bonds (Metal–O), respectively. Notably, compared to pristine ZnIn₂S₄, the Zn 2p (1044.65 eV and 1021.54 eV), In 3d (451.68 eV and 444.13 eV), and S 2p (163.03 eV and 161.73 eV) high-resolution XPS spectra of the 20%-Co₂SnO₄@ZnIn₂S₄ heterojunction show a significant negative shift in the absence of light. Moreover, after the integration of Co₂SnO₄, the high-resolution XPS spectra for the 20%-Co₂SnO₄@ZnIn₂S₄ heterojunction, Co 2p (795.39 eV and 780.15 eV), Sn 3d (495.51 eV and 487.11 eV), and O 1s (533.02 eV, 531.58 eV, and 530.00 eV), reveal shifts towards higher binding energies relative to Co₂SnO₄ alone. These shifts provide direct spectroscopic evidence of electron transfer from Co₂SnO₄ to ZnIn₂S₄ in the dark, corroborating the model derived from work function calculations. The direction and magnitude of these shifts align well with the calculated Fermi level offset, indicating the establishment of the IEF prior to illumination.⁵⁴

To dynamically monitor charge transfer under operational conditions, we further employed in situ illuminated X-ray photoelectron spectroscopy (XPS) (Fig. 4d–i). Upon light excitation, a comparison of the XPS spectra for the 20%-Co₂SnO₄@ZnIn₂S₄ heterojunction in the dark reveals distinct shifts: the high-resolution spectra of Zn 2p, In 3d, and S 2p shift towards higher binding energies, while the Co 2p, Sn 3d, and O 1s spectra shift towards lower binding energies. This light-induced inversion provides definitive evidence for the S-scheme mechanism: photogenerated electrons in the CB of ZnIn₂S₄ recombine with holes in the VB of Co₂SnO₄via the IEF channel, while useful electrons and holes accumulate in the CB of Co₂SnO₄ and the VB of ZnIn₂S₄, respectively. The observed shifts agree with the predicted charge flow, highlighting the dynamic nature of the S-scheme under operational conditions.

Surface photovoltage (SPV) spectroscopy provides valuable insights into the charge separation dynamics of the Co₂SnO₄@ZnIn₂S₄ heterostructure. The SPV spectra (Fig. S13†) reveal distinct behaviors for each component: pristine Co₂SnO₄ shows a negligible response due to poor charge separation, whereas ZnIn₂S₄ exhibits a strong characteristic n-type semiconductor signal in the 300–600 nm range. Remarkably, the 20%-Co₂SnO₄@ZnIn₂S₄ composite demonstrates significantly reduced SPV intensity despite its enhanced photocatalytic activity. This apparent paradox can be explained by the S-scheme charge transfer mechanism, where photogenerated electrons in ZnIn₂S₄ recombine with holes in Co₂SnO₄ at their interfacial junction. This process depletes surface holes in ZnIn₂S₄ while accumulating electrons in Co₂SnO₄, resulting in the observed SPV suppression. The SPV results align perfectly with the in situ XPS observations (Fig. 4). Together, these techniques offer compelling experimental evidence for the proposed S-scheme mechanism, explaining the system's exceptional photocatalytic performance through efficient charge separation and the preservation of strong redox potentials.

To gain direct spatial insight into the interfacial charge separation dynamics, we performed comprehensive photo-assisted Kelvin probe force microscopy (KPFM) measurements under both dark and illuminated conditions (Fig. 5 and S14†). The KPFM analysis revealed significant differences in surface potential behavior between pristine ZnIn₂S₄ and the 20%-Co₂SnO₄@ZnIn₂S₄ heterostructure, providing compelling evidence for the S-scheme charge transfer mechanism. In the dark, the heterostructure exhibited significantly higher contact potential difference (CPD) values (74.15–128.64 mV) compared to pristine ZnIn₂S₄ (40.44–89.57 mV), as shown in Fig. 5e1 and e2. This substantial positive shift in CPD results from spontaneous electron transfer from Co₂SnO₄ (with its lower work function of 6.32 eV) to ZnIn₂S₄ (work function = 6.55 eV), which is consistent with the formation of characteristic band bending at the interface and the establishment of a strong IEF. Under Xe lamp illumination, both materials exhibited reduced surface potentials, but with markedly different magnitudes. While pristine ZnIn₂S₄ showed a modest potential drop (ΔCPD = 24.6 mV), the heterostructure displayed a much more pronounced response (ΔCPD = 65.3 mV). This nearly 1.7-fold greater potential change in the composite material clearly visualizes the enhanced charge separation efficiency facilitated by the S-scheme configuration.⁵⁵

Fig. 5 (a) Schematic diagram of the in situ KPFM test. (b) Tested IEF (with ZnIn₂S₄ intensity set to “1”) of the 20%-Co₂SnO₄@ZnIn₂S₄ heterojunction. KPFM surface potential images of (c1 and d1) ZnIn₂S₄ and (c2 and d2) 20%-Co₂SnO₄@ZnIn₂S₄ before and after light irradiation. Surface potential profiles along the lines of (e1) ZnIn₂S₄ and (e2) 20%-Co₂SnO₄@ZnIn₂S₄ before and after light irradiation.

The spatial correlation of these potential changes with the structural hierarchy observed in SEM (Fig. 2c), where Co₂SnO₄ nanoparticles uniformly surround ZnIn₂S₄ microflowers, confirms the interfacial nature of the charge separation process. These KPFM observations are in perfect agreement with complementary characterization techniques, including zeta potential measurements (ZnIn₂S₄: −13.9 mV; 20%-Co₂SnO₄@ZnIn₂S₄: −11.7 mV) (Fig. S1†) and Gouy–Chapman model calculations of surface charge density.⁵⁶ Most importantly, these results correlate with the quantified 1.52-fold enhancement in interfacial electric field strength for the heterostructure compared to pristine ZnIn₂S₄ (Fig. 5b). The nanoscale potential mapping provided by KPFM offers direct experimental visualization of the proposed S-scheme mechanism, in which photogenerated electrons in ZnIn₂S₄ efficiently recombine with holes in Co₂SnO₄ through the IEF channel, while beneficial charge carriers accumulate in the corresponding energy bands for redox reactions.

Photothermal confinement effect

The light absorption properties of the photocatalysts were systematically investigated using UV-vis diffuse reflectance spectroscopy (DRS), as shown in Fig. 6a. Pristine ZnIn₂S₄ exhibited an absorption edge around 560 nm, corresponding to its bandgap of 2.37 eV. In contrast, the dark brown Co₂SnO₄ nanoparticles demonstrated broad absorption across the entire UV-vis-NIR spectrum, consistent with their narrow bandgap of 1.44 eV. The x%-Co₂SnO₄@ZnIn₂S₄ heterojunctions showed intermediate absorption edges between 500 and 530 nm, with significantly enhanced absorption in the visible and NIR regions, compared to pure ZnIn₂S₄. This extended absorption spectrum, especially in the NIR region (>700 nm), highlights the composite's potential for efficient photothermal conversion, a key attribute for maximizing solar energy utilization in photocatalytic applications.

Fig. 6 (a) UV-vis DRS of Co₂SnO₄, ZnIn₂S₄, and x%-Co₂SnO₄@ZnIn₂S₄. (b) Photothermal mapping images and timeline of ZnIn₂S₄, Co₂SnO₄, and Co₂SnO₄@ZnIn₂S₄ under UV-vis and UV-vis-NIR irradiation. (c) Infrared photographs of ZnIn₂S₄ and 20%-Co₂SnO₄@ZnIn₂S₄ during photocatalytic H₂ evolution testing under UV-vis and UV-vis-NIR illumination.

To directly assess the photothermal effect, we employed infrared thermography under simulated solar irradiation (Fig. 6b), with detailed temperature changes summarized in Fig. S15.†⁵⁷ The Co₂SnO₄ sample demonstrated exceptional photothermal conversion efficiency, reaching significantly higher temperatures than ZnIn₂S₄ under identical illumination conditions. Most notably, the 20%-Co₂SnO₄@ZnIn₂S₄ heterojunction reached an equilibrium temperature of 49.8 °C after 120 s of UV-vis irradiation, significantly outperforming pristine ZnIn₂S₄ and approaching the photothermal efficiency of pure Co₂SnO₄. When NIR light was incorporated into the illumination spectrum, the heterojunction's temperature further increased to 62.2 °C, demonstrating its efficient utilization of the full solar spectrum. These findings confirm that the composite not only retains Co₂SnO₄'s excellent photothermal properties but also preserves the photocatalytic functionality of ZnIn₂S₄, resulting in a synergistic system where thermal energy enhances catalytic processes.

The influence of photothermal effects on photocatalytic performance was quantitatively evaluated through wavelength-dependent hydrogen evolution experiments (Fig. S16†).⁵⁸ While pristine ZnIn₂S₄ exhibited a minimal response to NIR irradiation (only 7.8% increase in H₂ evolution rate), the 20%-Co₂SnO₄@ZnIn₂S₄ heterojunction demonstrated a remarkable 25.1% enhancement under full-spectrum (UV-vis-NIR) illumination compared to UV-vis alone, achieving an impressive H₂ evolution rate of 12.56 mmol g⁻¹ h⁻¹. This significant improvement can be attributed to several photothermal-enhanced mechanisms: (1) accelerated reaction kinetics at elevated temperatures, (2) enhanced charge carrier mobility, and (3) improved interfacial charge transfer facilitated by the thermal energy. Real-time temperature monitoring during photocatalytic testing (Fig. 6c) further confirmed the superior photothermal conversion efficiency of the heterojunction, showing significantly higher heating rates compared to ZnIn₂S₄. These findings collectively demonstrate that the localized heating generated by Co₂SnO₄ NIR absorption actively contributes to the photocatalytic process rather than merely being a byproduct of light absorption.

Catalytic enhancement mechanism

To gain a fundamental understanding of the enhanced hydrogen evolution activity at the atomic level, we performed DFT calculations to explore the catalytic properties of Co₂SnO₄ and ZnIn₂S₄. Based on the Sabatier principle, positive ΔG_H* values suggest a weak affinity for H* adsorption, while negative values indicate a strong interaction with H*, making the detachment of H₂ more challenging. Therefore, a |ΔG_H*| value approaching zero is considered ideal for promoting hydrogen production activity.⁵⁹ Our calculations reveal important differences in the catalytic behavior between the constituent materials. The S sites on pristine ZnIn₂S₄ exhibit a ΔG_H* of −0.69 eV, indicating relatively strong hydrogen binding that may hinder efficient H₂ desorption. In contrast, the Co and Sn sites in Co₂SnO₄ exhibit more favorable ΔG_H* values of 0.13 eV and 0.19 eV, respectively. These nearly thermoneutral adsorption energies, closely adhering to the Sabatier principle, suggest that the Co sites on Co₂SnO₄ provide ideal hydrogen binding characteristics for efficient H₂ evolution.^60,61

Building on these theoretical insights, we conducted experimental electrochemical measurements to validate the predicted catalytic enhancement. Linear sweep voltammetry (LSV) results indicate that Co₂SnO₄ exhibits the highest current density among the three samples (Fig. 7b).⁶² Moreover, the combination of Co₂SnO₄ with ZnIn₂S₄ demonstrates a significantly higher photocurrent density compared to pristine ZnIn₂S₄. This outcome confirms the electrocatalytic effect of Co₂SnO₄ in enhancing hydrogen evolution activity. Electrochemical impedance spectroscopy (EIS) further confirms the enhanced charge transfer kinetics, with Co₂SnO₄ exhibiting the smallest arc radius in the Nyquist plots (Fig. 7c) and the lowest charge transfer resistance (Table S3†). These results collectively demonstrate that the incorporation of Co₂SnO₄ not only improves charge separation but also actively participates in catalyzing the hydrogen evolution reaction.^63,64

Fig. 7 (a) Diagram of hydrogen-adsorption Gibbs free energies for H₂ evolution. (b) Current–voltage curves for proton reduction measured in Na₂SO₄ solution. (c) Electrochemical impedance spectra. Cyclic voltammetry curves of (d) ZnIn₂S₄, (e) Co₂SnO₄ and (f) 20%-Co₂SnO₄@ZnIn₂S₄ film electrodes at different scan rates. (g) Possible mechanism of light-driven H₂ production over ZnIn₂S₄ modified with Co₂SnO₄.

To further elucidate the origin of the enhanced catalytic performance, we characterized the electrochemically active surface area (ECSA) through cyclic voltammetry analysis. The results reveal that the ECSA, quantified by double-layer capacitance (C_dl), increases from 8.08 μF cm⁻² for ZnIn₂S₄ to 9.56 μF cm⁻² for the heterostructure (Fig. 7d–f and S17†). This 18% enhancement in ECSA indicates that the dispersed Co₂SnO₄ nanoparticles create additional catalytically active sites while maintaining accessibility to the ZnIn₂S₄ substrate. The optimal 20% Co₂SnO₄ loading strikes a balance between introducing sufficient active sites and preventing nanoparticle aggregation.

Complementing the structural and electrochemical analyses, we conducted an Arrhenius analysis of temperature-dependent hydrogen evolution rates to investigate the thermodynamic aspects of catalytic enhancement (Fig. S18†). The apparent activation energy (E_a) decreases from 19.1 kJ mol⁻¹ for ZnIn₂S₄ to 16.0 kJ mol⁻¹ for the heterostructure, representing a 16% reduction in the energy barrier. This thermodynamic facilitation, consistent with the DFT predictions, demonstrates how the Co₂SnO₄-induced photothermal effect synergistically enhances the reaction kinetics by promoting H–OH bond activation in adsorbed water molecules.⁶⁵

Performance enhancement mechanism

The remarkable photocatalytic hydrogen evolution activity of the Co₂SnO₄@ZnIn₂S₄ heterostructure arises from a sophisticated interplay of three complementary mechanisms, each addressing fundamental challenges in photocatalysis. Central to this enhancement is the precisely engineered S-scheme charge transfer pathway between Co₂SnO₄ and ZnIn₂S₄. Co₂SnO₄ acts as a reduction photocatalyst with a higher Fermi level, while ZnIn₂S₄ functions as an oxidation photocatalyst with a lower Fermi level. When these materials come into contact in the dark, electrons from Co₂SnO₄ diffuse into ZnIn₂S₄ until their interface Fermi levels equalize. This process causes band bending due to charge reorganization and results in the formation of an interfacial electric field from Co₂SnO₄ to ZnIn₂S₄. Upon exposure to light, both ZnIn₂S₄ and Co₂SnO₄ are excited, generating electrons and holes via photon absorption, with holes remaining in the VB. Influenced by the IEF and coulombic attraction, electrons with weak reducibility in the CB of ZnIn₂S₄ and holes with weak oxidability in the VB of Co₂SnO₄ recombine. This recombination ensures that highly reductive photoelectrons remain in the CB of Co₂SnO₄, where they reduce H⁺ into H₂ in water. Meanwhile, the photogenerated holes in the VB of ZnIn₂S₄ are consumed by a sacrificial reagent, in line with the S-scheme heterojunction electron transfer mechanism. This sophisticated charge management system is clearly demonstrated by several pieces of evidence, including significant quenching of photoluminescence signals, extended charge carrier lifetimes observed in time-resolved spectroscopy (Fig. S19 and Table S4†), and enhanced photocurrent generation (Fig. S20†). The interfacial electric field not only promotes efficient charge separation but also preserves the strong reducing power of electrons in the Co₂SnO₄ CB and the oxidizing capability of holes in the ZnIn₂S₄ VB, thereby creating optimal conditions for photocatalytic reactions.⁶⁶

Beyond the advantages of efficient charge separation, the system also greatly benefits from the exceptional photothermal conversion properties of Co₂SnO₄. This composite effectively captures near-infrared radiation, converting typically wasted low-energy photons into valuable thermal energy. Under full-spectrum illumination, this process increases the local reaction temperature to 62.2 °C. This photothermal effect brings about several key benefits: the increased temperature accelerates charge carrier mobility through thermal activation, lowers the activation energy for hydrogen evolution from 19.1 to 16.0 kJ mol⁻¹, and enhances mass transport properties. The thermal energy input creates a positive feedback loop, where improved charge separation promotes more redox reactions, generating additional heat through exothermic processes, which further enhances the overall system efficiency.

The final critical component in this synergistic system is the exceptional catalytic performance of Co₂SnO₄ for proton reduction. Theoretical calculations suggest that Co₂SnO₄ achieves an almost optimal hydrogen adsorption free energy of 0.13 eV, effectively balancing proton adsorption with the desorption of hydrogen molecules. Electrochemical characterization corroborates this computational prediction, demonstrating a significant increase in current density compared to pristine ZnIn₂S₄, along with a notable decrease in charge transfer resistance at the catalyst–electrolyte interface. The dispersed Co₂SnO₄ nanoparticles not only provide abundant active sites but also create favorable local environments for efficient proton adsorption and subsequent H–H bond formation. This catalytic enhancement, combined with improved charge separation and photothermal effects, results in an exceptional hydrogen evolution rate of 12.56 mmol g⁻¹ h⁻¹.

Conclusions

This work demonstrates a rational design of a Co₂SnO₄@ZnIn₂S₄ S-scheme heterojunction that synergistically integrates charge separation, photothermal conversion and catalytic enhancement for efficient solar hydrogen production. The heterostructure achieves an exceptional H₂ evolution rate of 12.56 mmol g⁻¹ h⁻¹ by combining three key features: (1) an interfacial electric field enabling efficient charge separation while preserving redox potentials, (2) broad-spectrum light absorption extending into the NIR region for photothermal heating, and (3) catalytically active Co₂SnO₄ sites that accelerate surface reactions. Advanced characterization and DFT calculations reveal that the photothermal effect dynamically enhances charge transfer kinetics while reducing activation barriers. This work establishes a new paradigm in photocatalyst design by demonstrating how precisely engineered thermal and electronic effects can work in concert to overcome fundamental limitations in photocatalysis. It not only provides a high-performance material system but also offers general principles for developing advanced solar fuel technologies.

Data availability

All data are available in the main text or the ESI.†

Author contributions

Kai Li designed and performed the experiments, analyzed the data, and drafted the manuscript. Zhaochao Yan assisted in data analysis and contributed to manuscript revision. Shuhan Sun and Qianmin Fan carried out material characterization studies and validation. Huayue Zhu and Chenglin Wu participated in the data analysis and interpretation. Yanxian Jin and Sónia A. C. Carabineiro provided critical feedback, supervised the project, and revised the manuscript. Ruiqiang Yan and Bingjing He contributed to data curation. Xianqiang Xiong conceived the study, supervised the research, and finalized the manuscript. All authors reviewed and approved the final version of the manuscript.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

The present study received financial support from the “Lingyan” R&D Plan Project of Zhejiang Province (2025C02218). SACC acknowledgesFCT/MCTES (Fundação para a Ciência e Tecnologia and Ministério da Ciência, Tecnologia e Ensino Superior) for the project UID/50006/2023 and https://doi.org/10.54499/CEECINST/00102/2018/CP1567/CT0026.

Notes and references

1. S. Xiong, R. Tang, D. Gong, Y. Deng, J. Zheng, L. Li, Z. Zhou, L. Yang and L. Su, Chin. J. Catal., 2022, 43, 1719–1748.
2. R. Li, J. Luan, Y. Zhang, L. Jiang, H. Yan, Q. Chi and Z. Yan, Renewable Sustainable Energy Rev., 2024, 206, 114863.
3. P. Ganguly, M. Harb, Z. Cao, L. Cavallo, A. Breen, S. Dervin, D. D. Dionysiou and S. C. Pillai, ACS Energy Lett., 2019, 4, 1687–1709.
4. H. Du, Y. N. Liu, C. C. Shen and A. W. Xu, Chin. J. Catal., 2017, 38, 1295–1306.
5. Y. Fan, X. Hao, N. Yi and Z. Jin, Appl. Catal., B, 2024, 357, 124313.
6. J. Qi, W. Zhang and R. Cao, Adv. Energy Mater., 2018, 8, 1701620.
7. S. K. Cushing, F. Meng, J. Zhang, B. Ding, C. K. Chen, C.-J. Chen, R.-S. Liu, A. D. Bristow, J. Bright, P. Zheng and N. Wu, ACS Catal., 2017, 7, 1742–1748.
8. R.-T. Gao, J. Zhang, T. Nakajima, J. He, X. Liu, X. Zhang, L. Wang and L. Wu, Nat. Commun., 2023, 14, 2640.
9. R. Shen, N. Li, C. Qin, X. Li, P. Zhang, X. Li and J. Tang, Adv. Funct. Mater., 2023, 33, 2301463.
10. K. Peng, Z. Wu, X. Liu, J. Yang and Z. Guan, Small, 2025, 21, 2406035.
11. Q. Wang, T. Hisatomi, Q. Jia, H. Tokudome, M. Zhong, C. Wang, Z. Pan, T. Takata, M. Nakabayashi, N. Shibata, Y. Li, I. D. Sharp, A. Kudo, T. Yamada and K. Domen, Nat. Mater., 2016, 15, 611–615.
12. H. Li, J. Zhang, X. Zhou, Z. Wu and L. Zhang, J. Mater. Sci. Technol., 2025, 231, 1–10.
13. U. Prasad, J. L. Young, J. C. Johnson, D. L. McGott, H. Gu, E. Garfunkel and A. M. Kannan, J. Mater. Chem. A, 2021, 9, 16137–16149.
14. T. Yang, Z. W. Chen, X.-Z. Yue, Q.-C. Liu, S. S. Yi and Y.-F. Zhu, Adv. Funct. Mater., 2024, 34, 2313767.
15. C. Du, B. Yan, Z. Lin and G. Yang, J. Mater. Chem. A, 2020, 8, 207–217.
16. C. Du, B. Yan and G. Yang, Nano Energy, 2020, 76, 105031.
17. Z. Guan, Z. Xu, Q. Li, P. Wang, G. Li and J. Yang, Appl. Catal., B, 2018, 227, 512–518.
18. X. Jiao, Z. Chen, X. Li, Y. Sun, S. Gao, W. Yan, C. Wang, Q. Zhang, Y. Lin, Y. Luo and Y. Xie, J. Am. Chem. Soc., 2017, 139, 7586–7594.
19. M. Yang, X. Q. Zhan, D. L. Ou, L. Wang, L. L. Zhao, H. L. Yang, Z.-Y. Liao, W.-Y. Yang, G.-Z. Ma and H. L. Hou, Rare Met., 2025, 44, 1024–1041.
20. L. Wu, M. Li, B. Zhou, S. Xu, L. Yuan, J. Wei, J. Wang, S. Zou, W. Xie, Y. Qiu, M. Rao, G. Chen, L. Ding and K. Yan, Small, 2023, 19, 2303821.
21. J. Wang, S. Sun, R. Zhou, Y. Li, Z. He, H. Ding, D. Chen and W. Ao, J. Mater. Sci. Technol., 2021, 78, 1–19.
22. W. K. Chong, B. J. Ng, X. Y. Kong, J. Low, H. W. Lee, L. L. Tan and S. P. Chai, Chem Catal., 2025, 5, 101227.
23. E. K. Andreou, I. Vamvasakis, A. Douloumis, G. Kopidakis and G. S. Armatas, ACS Catal., 2024, 14, 14251–14262.
24. M. Sigl, M. Egger, F. Warchomicka, D. Knez, M. Dienstleder, H. Amenitsch, G. Trimmel and T. Rath, J. Mater. Chem. A, 2024, 12, 28965–28974.
25. Y. Li, S. Li, L. Meng and S. Peng, J. Colloid Interface Sci., 2023, 650, 266–274.
26. P. Wang, Z. Shen, Y. Xia, H. Wang, L. Zheng, W. Xi and S. Zhan, Adv. Funct. Mater., 2019, 29, 1807013.
27. D. Yang, J. Liang, L. Luo, R. Deng, G. Li, Q. He and Y. Chen, Chin. Chem. Lett., 2021, 32, 2534–2538.
28. K. Wu, R. Jiang, Y. Zhao, L. Mao, X. Gu, X. Cai and M. Zhu, J. Colloid Interface Sci., 2022, 619, 339–347.
29. W. Ge, J. Song, S. Deng, K. Liu and P. Yang, Sep. Purif. Technol., 2024, 328, 125059.
30. S. Wang, D. Zhang, X. Pu, L. Zhang, D. Zhang and J. Jiang, Small, 2024, 20, 2311504.
31. S. Cao, J. Yu, S. Wageh, A. A. Al-Ghamdi, M. Mousavi, J. B. Ghasemi and F. Xu, J. Mater. Chem. A, 2022, 10, 17174–17184.
32. Y. Wang, M. Liu, C. Wu, J. Gao, M. Li, Z. Xing, Z. Li and W. Zhou, Small, 2022, 18, 2202544.
33. S. Bao, Q. Tan, S. Wang, J. Guo, K. Lv, S. A. C. Carabineiro and L. Wen, Appl. Catal., B, 2023, 330, 122624.
34. Y. Zhong, M. Li, X. Luan, F. Gao, H. Wu, J. Zi and Z. Lian, Appl. Catal., B, 2023, 335, 122859.
35. Z. Li, J. Xu, Z. Liu, X. Liu, S. Xu and Y. Ma, Int. J. Hydrogen Energy, 2023, 48, 3466–3477.
36. R. Xiong, X. Ke, W. Jia, Y. Xiao, B. Cheng and S. Lei, J. Mater. Chem. A, 2023, 11, 2178–2190.
37. H. Luo, H. Zhu, K. Xue, C. Liu, N. Chen, Y. Zhang and L. Yu, J. Energy Chem., 2025, 104, 655–669.
38. K. Geng, P. Shan, J. Shi, Y. Shen, S. Yan, M. Zhao, F. Guo, G. Wang and W. Shi, Chem. Eng. J., 2025, 507, 160561.
39. J. P. Tang, Y. Chen, Z. Y. Wang, Y.-H. Hu, J. H. Wang, L. Bao, Z. Y. Zhao and Y. J. Yuan, ACS Catal., 2025, 15, 265–274.
40. H. Qin, W. Zhang, S. Zhao, Z. Yao, Q. Zheng, P. Zhang, S. Zhang and X. Xu, Chem. Eng. J., 2024, 489, 151213.
41. Z. G. Tai, G. T. Sun, X. H. Zhang, X. B. Yang, T. Wang, Z. Y. Fang, Q. Ye, L. C. Jia and H. Q. Wang, J. Mater. Sci. Technol., 2023, 166, 113–122.
42. Z. Li, Q. Yan, Q. Jiang, Y. Gao, T. Xue, R. Li, Y. Liu and Q. Wang, Appl. Catal., B, 2020, 269, 118827.
43. Y. Qi, N. Du, H. Zhang, P. Wu and D. Yang, J. Power Sources, 2011, 196, 10234–10239.
44. C. Chen, Q. Ru, S. Hu, B. An, X. Song and X. Hou, Electrochim. Acta, 2015, 151, 203–213.
45. X. Guo, Q. Xiao, T. Yang, Y. Liu and Z. Jin, Sep. Purif. Technol., 2023, 325, 124764.
46. Q. Yu, S. Sun, A. R. Puente-Santiago, C. Wu, X. Xiong, Y. Jin and B. Weng, Appl. Catal., B, 2025, 367, 125098.
47. Y. Ai, J. Hu, X. Xiong, S. A. C. Carabineiro, Y. Li, N. Sirotkin, A. Agafonov and K. Lv, Appl. Catal., B, 2024, 353, 124098.
48. J. Li, J. Yao, Q. Yu, X. Zhang, S. A. C. Carabineiro, X. Xiong, C. Wu and K. Lv, Appl. Catal., B, 2024, 351, 123950.
49. Y. Zhang, J. Pan, G. Gong, R. Song, Y. Yuan, M. Li, W. Hu, P. Fan, L. Yuan and L. Wang, Catalysts, 2023, 13, 120.
50. J. Zhu, Q. Bi, Y. Tao, W. Guo, J. Fan, Y. Min and G. Li, Adv. Funct. Mater., 2023, 33, 2213131.
51. Z. Yan, X. Zhang, B. Yu, J. Yao, D. Han, Y. Jin, C. Wu, G. Dai and X. Xiong, Ceram. Int., 2024, 50, 3052–3063.
52. H. Zhu, M. Zhao, J. Zhou, W. Li, H. Wang, Z. Xu, L. Lu, L. Pei, Z. Shi, S. Yan, Z. Li and Z. Zou, Appl. Catal., B, 2018, 234, 100–108.
53. J. Li, X. Zhang, X. Xiong, C. Wu, Y. Jin and K. Lv, Sep. Purif. Technol., 2025, 354, 129444.
54. Y.-H. Wu, Y.-Q. Yan, Y.-X. Deng, W.-Y. Huang, K. Yang and K.-Q. Lu, Chin. J. Catal., 2025, 70, 333–340.
55. H. Zhang, Y. Gao, S. Meng, Z. Wang, P. Wang, Z. Wang, C. Qiu, S. Chen, B. Weng and Y.-M. Zheng, Adv. Sci., 2024, 11, 2400099.
56. Q. Fan, Z. Yan, J. Li, X. Xiong, K. Li, G. Dai, Y. Jin and C. Wu, Appl. Surf. Sci., 2024, 663, 160206.
57. W. Li, G. Zuo, S. Ma, L. Xi, C. Ouyang, M. He, Y. Wang, Q. Ji, S. Yang, W. Zhu, K. Zhang and H. He, Appl. Catal., B, 2025, 365, 124971.
58. Y. Shen, J. Shi, J. Lei, P. Shan, A. Bian, W. Kong, A. Proskurin, M. Wu, C. Lu, G. Wang and W. Shi, Chem. Eng. J., 2025, 508, 161072.
59. S. Sk, R. Madhu, D. S. Gavali, V. Bhasin, R. Thapa, S. N. Jha, D. Bhattacharyya, S. Kundu and U. Pal, J. Mater. Chem. A, 2023, 11, 10309–10318.
60. F. Wu, X. Zhang, G. Li, L. Wang, J. Huang, A. Song, A. Meng and Z. Li, Chem. Eng. J., 2024, 498, 155152.
61. L. Guo, J. Gao, M. Li, Y. Xie, H. Chen, S. Wang, Z. Li, X. Wang and W. Zhou, EcoEnergy, 2023, 1, 437–447.
62. B. Li, W. Guo, X. F. Lu, Y. Hou, Z. Ding and S. Wang, Mater. Rep.: Energy, 2023, 3, 100230.
63. Y. Xin, J. Tian, X. Xiong, C. Wu, S. A. C. Carabineiro, X. Yang, Z. Chen, Y. Xia and Y. Jin, Adv. Mater., 2025, 37, 2417589.
64. W. H. L. Bo-Yuan Gao, J. Electrochem., 2024, 30, 2312111.
65. P. Shan, P. Hao, K. Geng, W. Xue, B. Xiong, J. Hou, F. Guo, Y. Sun and W. Shi, Chem. Eng. J., 2025, 509, 161502.
66. X. Yan, J. H. Dong, J. Y. Zheng, Y. Wu and F. X. Xiao, Chem. Sci., 2024, 15, 2898–2913.

---

Footnotes

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d5ta03030g

‡ These authors contributed equally to this work.

---