Revealing the roles of light and heat for enhanced interfacial charge transfer in photocatalytic conversion of methanol to formaldehyde coupled with hydrogen evolution

Abstract

Photothermal catalytic methanol (CH₃OH) dehydrogenation is a sustainable pathway for the co-production of hydrogen (H₂) and formaldehyde (HCHO) under mild reaction conditions. However, this process is hindered by the inefficient interfacial electron transfer at the semiconductor–cocatalyst (SC) interface, which is exacerbated by charge leakage. Using TiO₂ as a model, we systematically investigated the individual effects of irradiation intensity, wavelegth, and temperature on interfacial electron-transfer dynamics. Our findings reveal that increasing the photon flux at a constant temperature effectively reduces the SC barrier and accelerates the electron-transfer kinetics, with little impact on the overall charge-transfer duration. Although short-wavelength light exhibits limited penetration in aqueous media, reducing the wavelength enhances SC interfacial electron-transfer. Conversely, raising the temperature under a fixed photon flux not only lowers the SC barrier but also markedly shortens the total charge transfer time. To address the temperature-induced charge leakage, Sn was doped into the TiO₂ lattice. The optimised Pt/3% Sn–TiO₂ catalyst, employed in a fixed-bed reactor designed to overcome poor UV penetration, achieves an H₂ production rate of 82.86 mmol g⁻¹ h⁻¹ under 30 W UV illumination (254 nm) at 80 °C, representing a 4.08-fold enhancement over Pt/TiO₂. In a 20 h stability test, the system maintained average production rates of 88.12 mmol g⁻¹ h⁻¹ for H₂ and 84.13 mmol g⁻¹ h⁻¹ for HCHO, achieving an apparent quantum efficiency (AQY) of 40.02%. Under 320 W UV irradiation, the catalyst sustained 413.5 mmol g⁻¹ h⁻¹ H₂ generation over 6 h with an AQY of 25.16%. This work presents a promising strategy for designing high-performance catalysts by enhancing SC interfacial electron transfer, thereby significantly improving the photothermal catalytic efficiency.

---

Green foundation

1. Photocatalytic hydrogen evolution coupled with organic synthesis offers a promising green chemical pathway. However, its application is limited by low efficiency. Herein, we conducted a kinetic analysis of the effects of irradiation intensity and temperature on interfacial electron transfer during the reaction, aiming to establish a novel theoretical basis for enhancing the efficiency of integrated photocatalytic H₂ production–organic synthesis systems.

2. Our research focuses on improving the efficiency of photocatalytic systems to facilitate their practical implementation. Using this technology, we successfully mineralised low-concentration benzene gas (∼10 mg m⁻³ in air) at a flow rate of 100 L min⁻¹, with emissions meeting stringent regulatory standards.

3. This study provides new insights into interfacial charge transfer under photothermal synergistic effects, providing theoretical guidance for designing advanced photothermal catalysts.

Introduction

With the growing global focus on climate change and carbon emissions, hydrogen (H₂) energy has emerged as a promising alternative to traditional fossil fuels owing to its remarkable energy density, environmental sustainability, and geographical adaptability.^1–4 Recently, integrated strategies combining photocatalytic H₂ production with the oxidation of organic molecules have attracted significant attention for their potential to address key challenges in sustainable energy and chemical synthesis.^5–7 This dual-function approach facilitates the concurrent reduction of protons to produce H₂ using photogenerated electrons and the oxidation of organic molecules to generate value-added chemicals via photogenerated holes.⁸ Consequently, it offers a promising solution to mitigate the high energy consumption and environmental pollution associated with conventional organic chemical manufacturing processes.^9–11

As an example, formaldehyde (HCHO) remains indispensable in polymer manufacturing, textile production, and agricultural applications, despite its well-documented occupational health risks.^12,13 Currently, industrial HCHO production predominantly relies on two thermocatalytic processes: (1) methanol (CH₃OH) partial oxidation over Cu (Ag or Fe₂O₃–MoO₃) catalysts at 200–300 °C (CH₃OH (g) + 1/2 O₂ (g) → HCHO (g) + H₂O (g), ΔH° 298 = −157.0 kJ mol⁻¹), and (2) methane (CH₄) oxidation using Fe₂O₃ (or Cr₂O₃) at 400–800 °C (CH₄ (g) + O₂ (g) → HCHO (g) + H₂O (g), ΔH° 298 = −275.8 kJ mol⁻¹).^14–16 Although these methods are widely adopted, they suffer from intrinsic limitations, such as the inability to produce H₂ simultaneously. Photocatalytic methanol dehydrogenation (CH₃OH (g) → HCHO (g) + H₂ (g), ΔH° 298 = +85.1 kJ mol⁻¹) thus presents a thermodynamically challenging yet promising alternative for concurrent H₂ and HCHO production under mild conditions.¹⁷ However, similar to other photocatalytic H₂ generation systems coupled with organic oxidation reactions, this approach is constrained by relatively low efficiency, which presents significant barriers to its large-scale industrial application.^18,19

For water splitting, the slow oxygen-evolution reaction is often identified as the primary rate-determining step.^6,20 However, even when easily oxidisable organic molecules such as lactic acid (LA) or CH₃OH are used as hole scavengers and efficient reduction cocatalysts such as Pt, Au, Pd, Ag, or MoS₂ are employed, the H₂-production efficiency remains unsatisfactory (Fig. 1a).^21–24 In such cases, the reduction half-reaction becomes the key bottleneck limiting the photocatalytic efficiency (Fig. S1). Previous studies have indicated that poor interfacial electron transfer between the semiconductor and cocatalyst (SC) is one of the main reasons.^25–27 Typically, electron transfer from the semiconductor to the cocatalyst proceeds via two primary mechanisms: thermionic emission and charge recombination (Fig. 1b).^25,28 In thermionic emission, the high work function of the cocatalyst creates a significant Schottky barrier upon contact with the semiconductor, severely impeding photogenerated electron transfer.^29,30 Meanwhile, in the charge recombination process, excessive surface states not only exacerbate charge leakage but also serve as recombination centres for photogenerated electrons and holes, leading to ineffective charge separation.

Fig. 1 Schematic of the mechanism for photocatalytic H₂ production coupled with the oxidation of organic molecules (a), and the two pathways for SC interfacial electron transfer: thermionic emission and charge recombination (b).

To address the issue of insufficient SC interfacial electron transfer caused by high barriers, we doped tin (Sn) doping into the titanium dioxide (TiO₂) in a controlled manner. This modification enhances semiconductor electron concentration, reduces the SC barriers, and minimises charge leakage. These improvements establish a strong positive correlation between Pt/Sn–TiO₂ and temperature, which is crucial for enhancing photothermal catalytic efficiency. Under mild conditions, Pt/3% Sn–TiO₂ exhibits efficient photothermal catalytic CH₃OH dehydrogenation for HCHO production. Specifically, within 20 h at 80 °C under a 30 W low-pressure mercury lamp, the average H₂ and HCHO yields were 88.12 and 84.13 mmol g⁻¹ h⁻¹, respectively, with an average apparent quantum efficiency (AQY) of 40.02%. Using a 320 W low-pressure mercury lamp, the average H₂ production rate reached 413.5 mmol g⁻¹ h⁻¹ within 6 h at 80 °C, with an average AQY of 25.16%. This work provides new insights into the catalyst design and coupling strategy for photocatalytic H₂ production via organic molecule oxidation.

Experimental

Reagents and chemicals

All reagents, including tetrabutyl titanate (Ti(C₄H₉O)₄, analytical reagent (AR)), crystalline tin(IV) chloride (SnCl₄·5H₂O, AR), dihydrogen hexachloroplatinate hexahydrate (H₂PtCl₆·6H₂O, AR), methanol (CH₃OH, AR), ethanol (C₂H₅OH, AR), LA (CH₃CH(OH)COOH, AR), sodium hydroxide (NaOH, AR), acetic acid (CH₃COOH, AR), and sodium sulfate (Na₂SO₄, AR) were purchased from Sinopharm Chemical Reagent. Acetylacetone (CH₃COCH₂COCH₃, AR), ammonium acetate (CH₃COONH₄, AR), and formaldehyde (36–38 wt% HCHO + 10–15 wt% CH₃OH, AR) were purchased from Energy Chemicals (Anqing, Anhui, China). All reagents were used as received without further purification. Deionised water was used during the experiments.

Catalyst preparation

The TiO₂ and Sn-doped TiO₂ powders were prepared using a hydrothermal method (Fig. 2). In a typical procedure, 1.2312 g (with a Sn/Ti molar ratio of 3%) of SnCl₄·5H₂O was dissolved in 13.4 mL CH₃COOH within a 100 mL polytetrafluoroethylene autoclave. Subsequently, 40 mL of Ti(C₄H₉O)₄ was added to the solution. After magnetically stirring the solution for 30 min, the autoclave was sealed and hydrothermally treated at 220 °C for 12 h. The resulting precipitates were collected, rinsed repeatedly with distilled water and ethanol, and dried at 60 °C for 12 h. Finally, the dried powders were ground into fine powders and calcined in air at 500 °C for 60 min. To prepare Sn–TiO₂ powders with varying doping concentrations (e.g., 0%, 1%, 3%, and 5%), the same protocol was followed, adjusting the amount of SnCl₄·5H₂O accordingly.

Fig. 2 Schematic of the catalyst preparation process.

Pt/TiO₂ and Pt/Sn–TiO₂ powders were fabricated using the photo-deposition method (Fig. 2). For this process, 0.3 g of (Sn–)TiO₂ was placed in a quartz reactor and suspended in an aqueous solution (20 vol% CH₃OH + 30 μL 0.1 g mL⁻¹ H₂PtCl₆) via sonication. The suspension was degassed with high-purity argon (Ar, 99.999%) for 15 min to eliminate dissolved oxygen and then exposed to continuous irradiation from a 300 W UV high-pressure lamp (model: CEM-M500, Beijing Mircoenerg, Beijing, China) for 2 h. Throughout irradiation, an Ar gas atmosphere was maintained under magnetic stirring conditions. Afterward, the resulting powder was isolated by centrifugation, washed multiple times with pure water, and dried at 60 °C for 12 h.

Characterisation

X-ray diffraction (XRD) patterns were collected using a Shimadzu XRD-6100 Lab X-ray diffractometer equipped with a Cu Kα X-ray source (wavelength λ = 1.5405 Å). X-ray photoelectron spectroscopy (XPS) and ultraviolet photoelectron spectroscopy (UPS) measurements were performed using a Thermo ESCALAB 250Xi spectrometer, employing an Al Kα X-ray source (photon energy = 1486.6 eV), with binding energies calibrated against the C_1s peak at 284.6 eV. UV–vis diffuse reflectance spectra were recorded on a PerkinElmer Lambda 950 UV/VIS/NIR Spectrometer (Waltham, Massachusetts, USA), using BaSO₄ as the reference material. Morphological characterisation was conducted using a JEOL JSM-7900F field-emission scanning electron microscope coupled with an energy dispersive X-ray spectroscopy (EDX, Bruker Flat Quad 5060F). High-resolution transmission electron microscopy images and elements mapping were obtained using a JEOL JEM-2010 microscope equipped with EDX (JED-2300T, JEOL). Irradiation intensity was measured using an optical power meter (MC-PM100B, Beijing Merry Change Technology, Beijing, China). Photoelectrochemical measurements were performed using a CHI 660E electrochemical workstation (Shanghai Chenhua Instruments, China). Gas chromatographic analysis of H₂ was conducted using a Fuli 9790 Plus gas chromatograph (Fuli Instruments, China).

The amount of HCHO produced was determined using the acetylacetone spectrophotometric method.^17,22,31,32 Specifically, the procedure involved adding 15 mL of chromogenic reagent solution (comprising 2 mL of CH₃COCH₂COCH₃, 100 g of CH₃COONH₄, and 15 of mL CH₃COOH, with the pH adjusted to 6.0 using NaOH) to a 50 mL volumetric flask. Subsequently, 20 μL of the reaction solution obtained after CH₃OH catalytic oxidation was introduced into the flask. The solution was diluted with distilled water to a final volume of 50 mL. Afterward, the flask was placed in a drying oven maintained at 60 °C for 1 h. Finally, the HCHO concentration was quantified by measuring the UV–vis absorbance at 414 nm (Rayleigh UV1801, Beijing Beifen-Ruili Analytical Instrument (Group), Beijing, China) based on pre-established calibration curves. Additionally, gas chromatography–mass spectrometry (GC–MS, Agilent 5977C GC/MSD, Santa Clara, California, USA) and high-performance liquid chromatography (HPLC, PolyPak H column, Fuli L75, Fuli Instruments, China) were employed to analyse the products formed from CH₃OH after a 12 h reaction period in a fixed-bed reactor.

The band gap structures of TiO₂ and Sn–TiO₂ were characterised using UPS and UV–vis spectroscopy. The work function (Φ) was determined from the secondary electron cutoff energy (E_SECE) measured using UPS under He-Iα excitation (hν = 21.22 eV):^33,34

Φ = hν − E_SECE (1)

For pristine TiO₂, the UPS spectrum (applied bias: −5 V) yielded E_SECE = 17.79 eV (Fig. S8), resulting in a work function of Φ_TiO2 = 21.22–17.79 = 3.43 eV, corresponding to the position of the Fermi energy level (E_F). Additionally, the valence band (VB) level was extracted from the UPS data and identified at 2.53 eV below the Fermi level. The optical bandgap (E_g) was derived from the UV–vis data using the Tauc equation for indirect transitions:^35,36

(αhν)² = A(hν − E_g) (2)

where α is the absorption coefficient, A is a proportionality constant, and hν is the photon energy. TiO₂ exhibits an E_g of 3.25 eV. By integrating the E_g, Φ, and Fermi level data, the positions of the VB and conduction band (CB) of TiO₂ can be further constructed. The methodology for constructing the energy band diagram of Sn–TiO₂ closely parallels that for TiO₂.

The photocatalytic CH₃OH reaction was analysed using in situ diffuse reflectance infrared Fourier-transform spectroscopy (DRIFTS) with a Bruker INVENIO S spectrometer (Karlsruhe, Germany). The experimental setup included a Harrick diffuse reflectance in situ cell and an MCT detector. Before the analysis, the empty reaction chamber was heated to 200 °C and maintained at that temperature for 0.5 h under an air atmosphere to eliminate residual impurities, followed by cooling to room temperature. The Pt/3% Sn–TiO₂ photocatalyst was introduced into the chamber, which was subsequently sealed tightly. The system was purged with Ar gas at a flow rate of 30 mL min⁻¹ for 15 min, followed by heating to 150 °C and holding at this temperature for 0.5 h to remove any impurity gases adsorbed on the catalyst surface. After cooling the system to 50 °C, Ar was used as the carrier gas to introduce the vapour of a 20 vol% aqueous CH₃OH solution into the reaction chamber via a bubbler at a flow rate of 30 mL min⁻¹ for 15 min, ensuring complete adsorption saturation on the catalyst surface. The chamber was sealed again, and the dark adsorption state was recorded as the background spectrum. A 3 W LED UV lamp (365 nm) was used to irradiate the sample through the observation window, and time-resolved IR absorbance spectra were continuously collected at predefined intervals over a 1 h period to monitor the reaction progress. A UV lamp was continuously illuminated throughout the experiment.

(Photo-)electrochemistry measurement

Photoelectrochemical measurements were carried out using a three-electrode configuration consisting of a glassy carbon working electrode modified with the catalysts for Mott–Schottky analysis or a fluorine-doped tin-oxide (FTO) glass working electrode coated with the catalyst (15 Ω sq⁻¹, active area: 0.5 cm²) for photocurrent characterisation, carbon cloth counter electrode, and Ag/AgCl reference electrode. To mitigate charge recombination at the FTO interface, a 20 μL titanium-based alcoholic precursor solution (10 mL ethanol + 1.485 mL Triton X-100 + 0.102 mL tetrabutyl titanate) was deposited onto pre-cleaned FTO substrates. The coated substrates were subsequently dried under ambient conditions for 5 h and thermally annealed at 500 °C for 1 h in air to form a compact layer. Afterward, a colloidal dispersion containing either 10 mg of (Sn–)TiO₂ or Pt/(Sn–)TiO₂ nanoparticles dispersed in 0.5 mL ethanol with 200 μL of 5 wt% Nafion solution was dropped onto the compact layer, resulting in four distinct electrode configurations: TiO₂|FTO, Sn–TiO₂|FTO, Pt/TiO₂|FTO, and Pt/Sn–TiO₂|FTO.

During the measurement, 15 vol% LA (or 0.5 M Na₂SO₄) aqueous solution was used for open-circuit potential decay (OCP) and voltammetry measurements, with illumination provided by a 300 W high-pressure lamp (model: CEM-M500, Beijing Mircoenergy). Additionally, a 0.5 M Na₂SO₄ aqueous solution without any hole extractors was used as the electrolyte for the Mott–Schottky analysis. The donor density (N_d) concentration was calculated as:^37,38

Here, C_SC² represents the capacitance of the semiconductor space-charge region, which is influenced by several factors, including the active surface area (A), dielectric constant (ε), vacuum permittivity (ε₀), electronic charge (e), applied potential (U), flat-band potential (U_fb), Boltzmann constant (k_B), and temperature (T).

The hydrogen evolution (HE) behaviour of Pt/TiO₂ and Pt/Sn–TiO₂ under varying light intensities were investigated in an aqueous LA solution system (Fig. 3). The experimental procedure was as follows. First, 10 mg of the catalyst was dispersed into a solution consisting of 10 mL of LA and 60 mL of deionised water. The resulting mixture was then purged with high-purity Ar gas and stirred for 15 min to ensure complete deoxygenation. Subsequently, the light source was activated and, continuous stirring and Ar gas purging were maintained throughout the H₂ production test. The gas flow rate was controlled by adjusting the valve, and the flow rate of the outlet gas mixture (Ar and H₂) was measured using a bubble flowmeter. Simultaneously, the H₂ concentration in the mixed gas was periodically analysed using gas chromatography. The reaction temperature was precisely maintained using an external circulating water bath.

Fig. 3 Photothermal catalytic H₂-production device.

The AQY of Pt/TiO₂ and Pt/Sn–TiO₂ under varying solar light intensities can be determined by integrating the wavelength range from the UV region to 382 nm (band gap: 3.25 eV), as defined by the cut-off edge obtained from UV–vis spectroscopy and normalized to the standard solar light spectrum (ASTM G173-03 (2020)).^39,40 Under one standard solar irradiation intensity (100 mW cm⁻², AM 1.5G), the photon flux density (N₁) that (Sn–)TiO₂ can absorb is 1.204 × 10²⁰ photons per m² per s. Using this value, the AQY can be subsequently calculated.^41–43

Here, n_H2 represents the number of evolved H₂ molecules (mol s⁻¹), N_A denotes the Avogadro constant (6.02 × 10²³ mol⁻¹), N₁ is the number of incident photons (m⁻² s⁻¹), and A is the effective area of the reactor (m²).

Taking Pt/TiO₂ as an example, under standard solar light intensity (100 mW cm⁻², AM 1.5G) at 20 °C, its H₂ production rate is 3.258 mmol h⁻¹ g⁻¹ (based on 10 mg of catalyst). Given that the reactor area was 0.001963 m² (diameter: 5 cm), the AQY was calculated as follows:

The calculation process for AQY under other irradiation intensity conditions follows was similar, involving multiplication of the photon concentration by a corresponding factor. For example, at 300 mW cm⁻² (AM 1.5G), the number of effective incident photons was 3N₁.

Photothermal oxidation of CH₃OH was conducted in a custom-designed tubular fixed-bed reactor.^44,45 A slurry consisting of 0.2 g of (Sn–)TiO₂ dispersed in 5 mL of anhydrous ethanol was sonicated for 15 min and uniformly brush-coated onto a low-pressure mercury lamp (Philips TUV 30 W/G30, 254 nm, Poland). The catalyst-coated lamp was air-dried for 2 h to form a homogeneous thin film before being assembled into the reactor.

For Pt deposition, a 1 wt% Pt cocatalyst was photo-deposited by irradiating 20 μL of a cycle of 1 wt% H₂PtCl₆ aqueous solution under 254 nm UV light (30 W, effective power: 12 W × 0.9) for 2 h. Before irradiation, the reactor was purged with Ar gas for 30 min to remove dissolved oxygen, and an inert atmosphere was maintained throughout the process by a continuous flow of Ar gas. The residual solvent was subsequently removed by purging with dry compressed air at room temperature for 1 h. The reactor was then reconfigured with 1 L of 20 vol% CH₃OH aqueous solution. The pre-treatment step involved circulating deoxygenated CH₃OH at 47 mL min⁻¹ for 30 min using a peristaltic pump. Under continuous Ar bubbling and irradiation, the reaction temperature was stabilised using a heating tape. H₂ evolution was quantified hourly via gas chromatography, and the HCHO concentration was measured hourly using the acetylacetone spectrophotometric method.^22,46 The AQY was determined using the following equations:^45,47,48

Here, n_H2 represents the number of evolved H₂ molecules (mol s⁻¹), N is the number of incident photons, N_A denotes Avogadro's constant (6.02 × 10²³ mol⁻¹), h is Planck's constant (6.62 × 10⁻³⁴ J s), c is the speed of light (3 × 10⁸ m s⁻¹), P signifies the effective power of the UV lamp used in the experiment (12 W), η indicates the effective utilisation rate (0.9), and λ represents the wavelength (254 nm). By substituting the aforementioned constants into the formula, we obtain

As an example, the AQY of Pt/TiO₂ at 20 °C (the n_H2 was 9.889 mmol g⁻¹ h⁻¹, 0.2 g catalyst) was calculated to be 4.81%.

Results and discussion

Composition and morphological analysis

The XRD patterns of undoped TiO₂ and Sn-doped TiO₂ with varying Sn contents (1%, 3%, and 5%) are shown in Fig. 4a. All the samples displayed a pure anatase phase (JCPDS 21-1272), with no SnO₂ diffraction peaks detected. A high-resolution scan of the (101) crystal plane (inset in Fig. 4a) indicates that as the Sn doping level increased, the diffraction peak shifted progressively to lower angles. This phenomenon can be attributed to the substitution of Ti⁴⁺ ions by larger Sn⁴⁺ ions (ionic radius: 69 pm vs. 60.5 pm for Ti⁴⁺),^49,50 which results in lattice expansion within the TiO₂ structure.^51,52 These results align with those obtained from high-resolution TEM (Fig. S2). Furthermore, SEM analysis combined with EDX data (Fig. S3) and elemental mapping (Fig. S2g) for both and TiO₂ and 3% Sn-doped TiO₂ (optimal catalytic performance) verified that the materials retained their nanoparticulate morphology after doping, with Sn uniformly distributed throughout the catalyst at a content consistent with the stoichiometric ratio. Collectively, these findings confirmed the successful incorporation of Sn into the TiO₂ lattice.

Fig. 4 XRD patterns of TiO₂ doped with varying amounts of Sn obtained at a scanning rate of 5° min⁻¹. The high-resolution images of the (101) crystal plane (inset) were obtained at a reduced scanning rate of 1° min⁻¹ (a). In panels b, c, and d, the high-resolution XPS spectra of Ti 2p, O 1s (O_L: lattice oxygen; OH: hydroxyl oxygen; OA: adsorbed oxygen),^56,57 and Sn 3d are presented for both pure TiO₂ and 3% Sn-doped TiO₂.

Furthermore, the XPS patterns showed a blue shift in the Ti 2p peak after Sn doping, accompanied by enhanced hydroxyl oxygen signals and the emergence of a Sn 3d signal. These spectral changes indicate that Sn doping significantly alters the coordination environments of Ti and O.^50,53 Specifically, the replacement of balanced Ti–O–Ti bonds with Sn–O–Ti bonds, driven by the higher electronegativity of Sn (1.96) than that of Ti (1.54),⁵³ induces significant alterations in the local electron density. Such modifications may affect the photoelectric properties of the material. Consistent with previous reports, Sn⁴⁺ doping enhances carrier mobility in TiO₂, suppresses recombination rates, and improves the OCP.^54,55

(Photo-)electrochemistry properties

To investigate the influence of irradiation intensity, wavelength, temperature, and Sn content on the interfacial electron transfer behaviour, we employed the OCP method to measure the interfacial electron-transfer time constant (τ_sc) between the SC (Pt). The underlying principle relies on comparing the time constant with (τ_OCP-dual) and without (τ_OCP) the cocatalyst. Specifically, in the absence of Pt, the electron transfer pathway follows a single semiconductor–solution route, whereas in the presence of Pt, two parallel pathways emerge: semiconductor–solution and semiconductor–cocatalyst–solution.^25,26 This change in the time constant can be used to quantitatively assess the kinetic characteristics of the interfacial electron transfer process.²⁵

Here, τ_OCP-dual and τ_OCP can be obtained by monitoring the OCP decay:^58,59

where the k_B is the Boltzmann constant, T is the temperature, q is the elementary charge, and E is the potential at time, t.

OCP behaviours under various irradiation conditions

Numerous studies have established that increasing the irradiation intensity or decreasing the wavelength can enhance photocatalytic properties.^{21,43,60–62} This study investigated the effect of light intensity and wavelength on interfacial electron-transfer kinetics. Using TiO₂ and Pt/TiO₂ as representative examples, we analysed the OCP behaviour under varying light intensities (Fig. 5) and wavelengths (Fig. S5) in an Ar-bubbled aqueous LA solution. As shown in Fig. 5a and b, the equilibrium photovoltage of the TiO₂ electrodes exhibited a negative potential, steadily shifting in the negative direction with increasing irradiation intensity, regardless of Pt modification. This phenomenon arises from an imbalance in photogenerated carrier consumption, where the oxidation half-reaction rate significantly exceeds the reduction half-reaction rate (Fig. S1).

Fig. 5 OCP behaviours of various electrodes (TiO₂|FTO, Pt/TiO₂|FTO) investigated in an Ar-bubbled 15 vol% LA aqueous solution under various UV irradiation conditions at room temperature (∼25 °C) (a and b). The estimated time constant for SC interfacial electron transfer under varying light intensities and irradiation wavelengths are presented in (c) and (d), respectively. Based on these results, a schematic diagram the effect of irradiation intensity and wavelength on the SC interface barrier is proposed in (e).

An analysis of the electron-transfer time constants under varying irradiation intensities (Fig. S4) shows that for pristine TiO₂ at 180 mW cm⁻², the average time constant increased by factors of 5.25 and 6.65 compared with those at 88 mW cm⁻² and 35 mW cm⁻², respectively (Fig. S4a). This indicates that a higher irradiation intensity leads to slower electron-transfer kinetics. Although stronger illumination promoted the entry of more electrons into the CB (producing a more negative photovoltage at 180 mW cm⁻² in Fig. 5a), the kinetic limitations at the TiO₂/electrolyte interface hindered efficient electron transfer to the solution. After Pt deposition, the photovoltage remained intensity dependent (Fig. 5b), but the catalytic properties of Pt significantly enhanced the electron transfer efficiency. At 180 mW cm⁻², Pt/TiO₂ shows only ∼1.22 and ∼1.33 higher time constants than those at 88 mW cm⁻² and 35 mW cm⁻² (Fig. S4b), respectively, demonstrating a weakened dependence on the light intensity. Notably, Fig. 5c shows that τ_SC decreases with increasing light intensity. Therefore, while pristine TiO₂ exhibits a strong negative correlation between light intensity and electron transfer kinetics, Pt deposition mitigates this relationship by accelerating SC interfacial transfer, enabling higher reaction rates under high-intensity irradiation conditions.

Furthermore, band-pass filters were employed to isolate specific wavelengths—254 nm (∼2.0 mW cm⁻²), 313 nm (∼2.4 mW cm⁻²), and 365 nm (∼283.5 mW cm⁻²)—as individual light sources for evaluating the OCP behaviour of TiO₂ and Pt/TiO₂ electrodes (Fig. S5a and S5b). The results show that although the light intensity at 313 nm is considerably lower than that at 365 nm, the difference in photovoltage between these wavelengths is relatively small (0.023 V (vs. Ag/AgCl) for TiO₂ and 0.001 V (vs. Ag/AgCl) for Pt/TiO₂). This can be attributed to the relatively small difference in wavelength, which resulted in similar penetration depths. Despite its lower intensity, 313 nm light generates higher-energy excited electrons that overcome the SC barrier more effectively, thereby facilitating charge transfer.

Under equivalent light intensity conditions (vs. 313 nm), a shorter-wavelength (254 nm) can theoretically excite higher-energy electrons, which should overcome the SC barrier. However, owing to its limited penetration capacity (Fig. S6), light traveling through the solution to the electrode surface is significantly attenuated. Consequently, the photovoltage generated under 254 nm illumination was substantially lower than that under 365 nm and 313 nm. Further analysis of the interfacial charge-transfer time constant (Fig. S5c and S5d) revealed that, for both TiO₂ and Pt/TiO₂, despite a 118.1-fold difference in light intensity between 365 and 313 nm, the change in the time constant remained marginal (∼1.57-fold). By contrast, under 254 nm illumination, the charge-transfer time constants for Pt/TiO₂ decreased by a factor of 2.41 compared with that at 313 nm. As illustrated in Fig. 5d, shorter wavelengths correspond to smaller SC interfacial charge-transfer time constants, indicating a higher probability of successful charge transfer. This is because, although fewer short-wavelength photons reach the electrode surface, the excited electrons possess higher energy, allowing them more easily overcome the SC barrier. Based on the schematic illustrating the influence of irradiation intensity and wavelength on the SC barrier (Fig. 5e), both a higher light intensity and shorter wavelength contribute to the excitation of more high-energy electrons, thereby promoting more efficient SC interfacial electron transfer.

OCP behaviours under different temperatures

Fig. 6 compares the effects of temperature on the electron transfer dynamics in the TiO₂ and Pt/TiO₂ systems. Notably, temperature exerted a more pronounced positive influence than irradiation intensity on the electron transfer behaviour, particularly in Pt/TiO₂. As shown in Fig. 6a, the photovoltage in TiO₂ increases with irradiation intensity (Fig. 5a), but decreased with increasing temperature in the LA solution. This indicates that under the same irradiation intensity, elevated temperatures both increase CB electron concentration and accelerate the interfacial electron transfer to the solution (Fig. S7a and S8a).

Fig. 6 OCP behaviours of various electrodes (TiO₂|FTO, Pt/TiO₂|FTO) in an Ar-saturated 15 vol% LA aqueous solution were investigated under the same UV irradiation conditions (180 mW cm⁻²) at varying temperatures (a and b). The insets in (a) and (b) show the photovoltage responses under various temperature conditions; (c) variation trend of the SC time constant with interfacial electron transfer; (d) schematic illustrating the photo-thermal synergy influences of SC interfacial electron transfer.

After Pt deposition, the influence of the temperature on the photovoltage diminished (Fig. 6b), indicating that nearly all the additional electrons generated by the temperature increase were rapidly transferred to Pt via the SC interface and subsequently to the solution (Fig. S8b and S9a). Further analysis of the SC electron transfer time constant reveals that increasing the temperature from 10 °C to 50 °C reduces the time constant by an average factor of 3.51 (Fig. 6c), confirming that electron transfer kinetics at the interface are markedly enhanced with rising temperature (Fig. 6d).

OCP behaviours for TiO₂ and Sn–TiO₂

The findings confirm the critical role of Pt in facilitating electron transfer from the semiconductor to the solution. Temperature also significantly influences the charge transfer dynamics in Pt/TiO₂ systems, although considerable charge leakage persists (Fig. 5a and S6a), presenting a major challenge for synergistic photothermal applications.^63,64 Sn doping effectively suppressed charge leakage (Fig. 7). As shown in Fig. 7a and c, the 1% Sn–TiO₂ electrode exhibited the most negative photovoltage value and the slowest decay rate at the same temperature, indicating that Sn doping effectively suppressed the leakage of photogenerated electrons. This conclusion is further supported by the comparative oxygen reduction reaction (ORR) behaviours of TiO₂ and 1% Sn–TiO₂ (Fig. S9b). After Pt loading, Pt/1% Sn–TiO₂ displayed the most positive photovoltage and the fastest decay rate (Fig. 7b, c and Fig. S8d), indicating that excessive electron accumulation at the semiconductor surface caused by the high SC barrier can be mitigated. Therefore, Sn doping achieves the dual benefit of suppressing photogenerated electron leakage while simultaneously facilitating SC interfacial electron transfer. This significantly enhanced the overall charge-transfer efficiency and photothermal catalytic performance, crucially counteracting the increased charge leakage associated with increasing temperature.

Fig. 7 OCP behaviours of various electrodes (1% Sn–TiO₂|FTO and Pt/1% Sn–TiO₂|FTO) in an Ar-bubbled 15 vol% LA aqueous solution under identical UV irradiation (180 mW cm⁻²) at different temperatures (a and b) are investigated. The insets show the photovoltage recorded for each electrode under identical illumination conditions until equilibrium is achieved. After data aggregation, the results are summarised in (c). The estimated SC time constants are presented in (d). (e) Diagram illustrating the Sn-doping influences of SC interfacial electron transfer.

Analysis of SC interfacial charge transfer time constants (Fig. 7d) reveals that Sn doping enhances the transfer rate by factors of 1.46 (at 10 °C), 3.64 (at 30 °C), and 2.81 (at 50 °C). Combined with the increased electron concentrations observed in the Sn-doped samples (Fig. S7b), these findings indicate that Sn doping not only mitigates surface electron leakage but also enables photogenerated electrons to rapidly traverse the reduced SC potential barrier (Fig. S10), achieving efficient charge separation (Fig. 7e and Fig. S11). Consequently, Pt/1% Sn–TiO₂ delivered significantly enhanced photothermal catalytic HE performance in LA solution.

Voltametric behaviour analysis

To further evaluate the influence of Sn doping on TiO₂ performance, we investigated the HER activity of FTO electrodes modified with four distinct catalysts (TiO₂, Sn–TiO₂, Pt/TiO₂, and Pt/Sn–TiO₂). Electrochemical measurements were performed in an Ar-bubbled 15 vol% LA solution, and photocurrent responses were measured under identical irradiation intensities varying temperatures (Fig. 8). As shown in Fig. 8a, the LSVs revealed that Pt/Sn–TiO₂ exhibited the most positive onset potential, whereas Sn–TiO₂ showed the most negative onset potential. This indicates that Sn doping significantly reduced the interaction between the electrons on the TiO₂ surface and the oxidising species (e.g. H⁺), in contrast to pure TiO₂. However, the introduction of Pt markedly enhances this interaction, which is consistent with the trends observed in the previously discussed OCP measurements.

Fig. 8 LSV behaviours of four electrodes investigated in an Ar-bubbled 15 vol% LA solution using a conventional three-electrode system (a), along with their photocurrent responses under irradiation from a 300 W UV lamp, measured in either an Ar-bubbled 15 vol% LA solution or an Ar-bubbled 0.5 M Na₂SO₄ aqueous solution (b).

Moreover, under identical temperature conditions in both the LA solution and 0.5 M Na₂SO₄ solution, Sn–TiO₂ showed the lowest photocurrent density, while Pt/Sn–TiO₂ showed the highest (Fig. 8b and Fig. S11). This confirms that Sn doping passivates TiO₂, whereas Pt loading facilitates charge separation. Importantly, for all electrode materials, the photocurrent densities in the LA solution were significantly higher than those in 0.5 M Na₂SO₄ at equivalent temperatures, indicating that LA provides a more favourable environment for the HER. Overall, these findings indicate that Pt/Sn–TiO₂ exhibits superior photothermal catalytic performance for HER.

HE behaviours in LA aqueous solution

These studies confirm that 1% Sn–doped TiO₂ significantly enhances the photothermal catalytic activity of LA by suppressing charge leakage, reducing the SC interface barrier, and accelerating electron transfer. To further investigate the influence of irradiation intensity and temperature on the photothermal synergy effect, we evaluated the HE performance of Pt/TiO₂ and Pt/1% Sn–TiO₂ under varying light intensities and temperatures in a 15 vol% LA aqueous solution (Fig. 9 and Fig. S12).

Fig. 9 Photothermal catalytic HE behaviours of Pt/TiO₂ and Pt/1% Sn–TiO₂ catalysts in a 15 vol% LA aqueous solution under illumination at varying intensities: 100 mW cm⁻² (AM 1.5G, effective photon flux, 2.36 × 10¹⁷ photons per s), 300 mW cm⁻² (AM 1.5G, effective photon flux, 7.09 × 10¹⁷ photons per s), 500 mW cm⁻² (AM 1.5G, effective photon flux, 1.18 × 10¹⁸ photons per s), 1000 mW cm⁻² (AM 1.5G, effective photon flux, 2.36 × 10¹⁸ photons per s) irradiation intensities and their corresponding AQY curves (a–d). Notably, the incident photon flux is calculated by integrating from the UV region to the absorption edge of 382 nm for TiO₂ within the solar spectrum.

The results indicate that, the temperature dependence of Pt/TiO₂ under varying light intensities is not entirely positively correlated (dashed regions, Fig. 9b and c), which can be attributed to changes in the interfacial charge-transfer mechanisms. Under low-intensity irradiation (Fig. 9a and Fig. S12a), the elevated SC barrier restricts electron transfer primarily to thermally enhanced charge recombination.²⁵ Since recombination accelerates with temperature,⁶³ HER efficiency increases accordingly. At intermediate intensities (300 and 500 mW cm⁻²), the increased carrier generation partially lowers the SC barrier. While the overall transfer kinetics remained constrained, the dominant mechanism shifted toward thermionic emission. The associated charge leakage intensifies with temperature (Fig. 6a and S8a), outweighing the recombination benefits and suppressing HE efficiency, which is consistent with the observed OCP characteristics. When the irradiation intensity is further increased (Fig. 9d), the diminished SC barrier shifts the rate-limiting step to the cocatalyst–solution (CS) electrochemical barrier. Considering the temperature-dependent electrocatalytic enhancement caused by Pt (Fig. S9a), elevated temperatures significantly enhanced HER efficiency under high irradiation intensities.

In additional, the temperature-dependent HE behaviours of Pt/TiO₂ at varying wavelengths also was investigated (Fig. S12b). The results showed that under constant temperature and normalised light intensity, both the HE rates and AQY increased with longer wavelengths. Specifically, 365 nm irradiation yielded the highest H₂ production rate at 80 °C, with an AQY of 28.30%, followed by 313 nm (AQY, 3.02%) and 254 nm (AQY, 2.73%). This trend is likely due to the differences in light penetration at different wavelengths (Fig. S6). Over the temperature range of 20–80 °C, the HE rates under 254, 313, and 365 nm increased by factors of 13.49, 3.07, and 1.37, respectively, indicating that shorter wavelengths lead to a more pronounced thermal enhancement effect and stronger photo-thermal synergy, possibly owing to faster SC interfacial charge transfer under short-wavelength irradiation (Fig. 5d and Fig. S5d).

After Sn doping, Sn not only introduced surface states but also effectively suppressed charge leakage (Fig. S7 and S9b). Consequently, under both weak and strong light conditions, the photothermal catalytic HE performance of Pt/1% Sn–TiO₂ mostly surpasses that of Pt/TiO₂ (Fig. 9). Specifically, at light intensities of 50, 100, 300, and 500 mW cm⁻² (AM 1.5G), the HE rate of Pt/Sn–TiO₂ consistently exceeds that of Pt/TiO₂ at the same temperature. Moreover, as the temperature increases, the HE effect of Pt/1% Sn–TiO₂ becomes more pronounced, which can be attributed to the inhibition of charge leakage by Sn doping. Furthermore, under incident light intensities of 300 mW cm⁻² (AM 1.5G) and 500 mW cm⁻² (AM 1.5G), no decrease in the HE rates owing to temperature increase was observed. Instead, a stable positive correlation between the HE rates and temperature was observed. At the same temperature, the higher the incident irradiation intensity, the greater the HE rate. Notably, as the irradiation intensity increased to 1000 mW cm⁻² (AM 1.5G), the gradual reduction of the SC barrier weakened the charge suppression effect caused by Sn doping (Fig. 9d). This explains the only slight increase in the HE rates between of the doped and undoped materials as the temperature increased. Although the HE rate of Pt/1% Sn–TiO₂ reached its maximum value (36.279 mmol g⁻¹ h⁻¹) at 80 °C, the corresponding AQY was the lowest (only 5.14%). This phenomenon can be attributed to light scattering, reduced light absorption, and effective photon loss.^65,66 At higher light intensities, these losses significantly diminish the utilisation efficiency of the effective photons, thereby impacting the quantum efficiency.

Photothermal oxidation of CH₃OH

Inspired by the HE behaviour of Pt/TiO₂ and Pt/1% Sn–TiO₂ in the LA aqueous solution under various conditions, increasing the light flux, reducing the wavelength, and raising the temperature can significantly reduce the SC barrier, thereby enhancing SC interfacial electron transfer. However, owing to the limited penetration depth of short-wavelength light in solution, significant scattering effects, and insufficient photon-absorption efficiency of the photocatalyst under a high light flux, the overall energy utilisation efficiency remains relatively low. To address these limitations, we designed and fabricated a fixed-bed tubular reactor by directly coating the photocatalyst onto the surface of a 30/320 W UV light source (Fig. 10a). This configuration effectively minimised the attenuation of short-wavelength light within the solution, reduced photon scattering, and enhanced the absorption efficiency.⁶⁷ Furthermore, we systematically investigated the influence of varying the Sn doping level on the performance of the TiO₂-based photothermal oxidation of CH₃OH for H₂ and HCHO production (Fig. 10).

Fig. 10 Schematic illustrating the configuration of a self-designed fixed-bed continuous reactor (a). Photothermal dehydrogenation performance of Pt/Sn–TiO₂ with varying Sn doping levels in the conversion of CH₃OH to HCHO under 30/320 W UV lamp irradiation (b, d), along with the corresponding AQY (c). Additionally, the stability of the photothermal catalytic oxidation of CH₃OH was investigated on a Pt/3% Sn–TiO₂ catalyst during a 20 h of cyclic reaction at 80 °C (d). Comparison of the performance achieved using different catalysts for the photocatalytic dehydrogenation of CH₃OH to HCHO in recent years (e). Reaction conditions: Ar-bubbled in 500 mL of 20 vol% CH₃OH aqueous solution, 30 W low-pressure mercury lamp (254 nm, 4.5 mW cm⁻²).

As shown in Fig. 10b, under irradiation by a 30 W low-pressure mercury lamp (254 nm, with an effective photon flux of 1.38 × 10¹⁹ photons per s), the HE rates of all catalysts increased with rising temperature. Notably, when the Sn doping level was 3%, the catalyst achieved the highest H₂ production rate, reaching 82.86 mmol g⁻¹ h⁻¹ at 80 °C (below methanol–water azeotrope boiling point, ∼85 °C at 1 atm (ref. 68)), with a corresponding AQY of 40.27% (Fig. 10c), representing a 4.08-fold increase compared with the performance of Pt/TiO₂. Meanwhile, the catalytic efficiency of Pt/3% Sn–TiO₂ within the temperature range under 30 W UV light irradiation can reach up to 29.45–39.36 times that under dark conditions (Fig. S13a). Subsequently, experiments were conducted under higher irradiation intensity conditions (320 W, 254 nm, with an effective photon flux of 1.34 × 10²⁰ photons per s¹) (Fig. S13b). The average HE rate reached up to 413.5 mmol g⁻¹ h⁻¹ at 80 °C within 6 h, with the corresponding average AQY reaching 25.16%. Despite a nearly 10-fold increase in the incident irradiation intensity, the HE efficiency only improved by ∼5.0 times, resulting in a lower AQY compared with that observed under 30 W conditions. This discrepancy can be primarily attributed to the fact that under high irradiation intensities, the SC interfacial electron transfer process is predominantly governed by thermionic emission, which is accompanied by intensified charge leakage during this process.

Given that the AQY was higher at a power of 30 W, we conducted a stability test at 80 °C using the Pt/3% Sn–TiO₂ catalyst (Fig. 10d). Although the AQY decreased after 10 h, during the subsequent 20 h reaction process, the average yields of H₂ and HCHO were 88.12 and 84.13 mmol g⁻¹ h⁻¹, respectively, with an average quantum yield of 40.02%. These results indicate that Pt/3% Sn–TiO₂ has excellent potential for practical applications in photothermal catalytic CH₃OH reforming. Moreover, these production rates rank among the leading values in the field of photocatalytic dehydrogenation of CH₃OH to HCHO (Fig. 10e). Post-reaction characterisation of the catalyst using XRD and TEM (Fig. S14), performed according to literature protocols,^69,70 indicated no significant changes in the phase composition or microstructure, confirming its high structural stability under these reaction conditions.

Mechanistic analysis

To investigate the reaction mechanism of simultaneous HCHO formation during photocatalytic H₂ production from CH₃OH over the Pt/3% Sn–TiO₂ catalyst, in situ DRIFTS analysis was conducted under illumination in an inert atmosphere (Fig. 11a). As shown in Fig. 11b–d, the absorption bands at 3679 cm⁻¹ and 3653 cm⁻¹ are attributed to the hydroxyl stretching vibrations (ν_OH) of CH₃OH and water, respectively.^71,72 The peak at 2948 cm⁻¹ is attributed to the C–H stretching vibration (ν_CH) of methanol,⁷¹ with distinct C–O stretching vibrations (ν_CO) at 1034 cm⁻¹ (gas phase) and 1011 cm⁻¹ (absorbed state).⁷³ Signals detected at 2921 and 2820 cm⁻¹ are assigned to the C–H stretching vibrations (ν_CH) of methoxy (–CH₃O) species,^71,73 while the peaks at 1143 cm⁻¹ and 1055 cm⁻¹ are assigned to the C–O stretching vibrations (ν_CO) of –CH₃O species.⁷³ The absorption bands at 2353 and 2319 cm⁻¹ are characteristic of CO₂,⁷² while at 2046 cm⁻¹ is characteristic of CO.⁷³ Finally, the peaks at 1649, 1560, and 1355 cm⁻¹ are indicative of formate species (ν_HCOO) that accumulate on the catalyst surface.^71,73

Fig. 11 In situ DRIFT spectra of methanol photocatalytic oxidation over Pt/3% Sn–TiO₂ catalyst under UV irradiation (3 W, 365 nm) after subtracting the background obtained under dark adsorption-equilibrium conditions at 50 °C in an Ar atmosphere (a). (b)–(d) Magnified views of the corresponding colour-highlighted regions labelled in (a). Based on the results of the in situ DRIFT and GC–MS analyses, a proposed reaction mechanism for the photocatalytic dehydrogenation of CH₃OH to HCHO over a Pt/3% Sn–TiO₂ catalyst is presented in (e).

Analysis of the results reveals that the characteristic peaks of CH₃OH remained relatively stable throughout the reaction, showing slight attenuation (e.g., 2948 cm⁻¹) over time (Fig. 11c). This can be attributed to the continuous replenishment of CH₃OH in the reaction chamber. During the experiment, Ar gas was introduced into the reaction system by bubbling it through a 20 vol% CH₃OH solution, after which the system was sealed for detection. If the system had been purged with inert gas at this stage to remove the physically adsorbed species, the amount of CH₃OH remaining on the catalyst surface would likely have been insufficient to produce detectable reaction-related signals. In addition, several intermediate products were observed during the reaction, including methoxy (2921, 2820, 1143, and 1055 cm⁻¹) and formate species (1649, 1560, and 1355 cm⁻¹). The gradual increase (e.g., 1143, 1560, and 1355 cm⁻¹) in the peak intensity indicates that the reaction progressed continuously. Moreover, the signals corresponding to the final products CO₂ (2353 and 2319 cm⁻¹) and CO (2046 cm⁻¹) became increasingly pronounced as the reaction proceeded. By contrast, functional groups associated with the target product HCHO were barely detected, indicating that under gas-phase reaction conditions, HCHO is highly susceptible to further photocatalytic oxidation, a finding consistent with previously reported literature.^71,73,74 A long-term photothermal catalytic experiment conducted under liquid-phase conditions (Fig. 10e) revealed that HCHO production declined significantly after reacting for 10 h, potentially owing to its subsequent oxidation.

To further investigate the products of the liquid-phase reaction, the CH₃OH solution obtained after 12 h of reaction was analyzed using GC–MS and HPLC (Fig. S15). The mass spectrum (Fig. S15a) exhibited a weak signal corresponding to the HCHO fragment ion at m/z 29, whereas intense peaks arise at m/z 89, 119, 149, and 179. As reported in the literature,^75–78 these signals are characteristic of oligomeric HCHO fragments, indicating that HCHO exists predominantly in polymerised form in aqueous solution, with only trace amounts present as free molecules. Additional ion peaks at m/z 45 and 59 are tentatively assigned to HCOOH and methyl formate, respectively, consistent with the DRIFTS results and previous reports.⁷⁹ The presence of HCHO was also confirmed by HPLC analysis (Fig. S15b).

Fig. 11e shows the reaction mechanism for the photothermal catalytic dehydrogenation of CH₃OH to HCHO. Upon photoexcitation of TiO₂, the oxidation half-reaction involving photogenerated holes (h⁺) proceeds more rapidly than the reduction half-reaction involving photogenerated electrons (e⁻). This kinetic imbalance led to accumulation of negative charges on the catalyst surface. Although Pt serves as an effective reduction cocatalyst that facilitates electron transfer from the semiconductor to the solution, its efficiency remains insufficient to match the rapid oxidation process (Fig. S1 and S16). To address this imbalance, photothermal synergy leverages the combined effects of irradiation intensity, light wavelength, and temperature to significantly enhance interfacial electron transfer. Specifically, under higher irradiation flux, shorter wavelengths generate higher-energy photoelectrons that can overcome the semiconductor energy barrier more effectively, thereby improving charge separation and migration. Although elevated temperatures may promote some degree of charge leakage, they concurrently accelerate interfacial charge-transfer kinetics. The overall effect is the enhancement of charge-transfer efficiency. Consequently, the catalytic performance demonstrates a moderate positive correlation with temperature.

However, in conventional dispersed-catalyst systems, the limited penetration depth of short-wavelength light restricts photon absorption by the photocatalyst, thereby limiting the overall efficiency. Furthermore, increasing the temperature exacerbates charge leakage, counteracting the benefits of thermal activation and yielding only a marginal improvement in the activity of Pt/TiO₂ with increasing temperature. Sn doping effectively addresses these limitations by suppressing charge leakage, increasing the electron density, reducing the SC barrier, and facilitating electron transfer at the SC interface. Concurrently, the adoption of a fixed-bed reactor mitigates the poor transmittance of short-wavelength light through liquid media. Through synergistic optimisation of both catalyst and reactor designs, Pt/3% Sn–TiO₂ exhibited excellent performance in the photothermal catalytic dehydrogenation of CH₃OH to HCHO (Fig. 10e).

In the proposed reaction pathway, adsorbed CH₃OH molecules on the catalyst surface initially undergo C–H bond cleavage via hydroxyl radicals (˙OH) generated by h⁺, forming surface-bound –CH₃O and H₂O.^79,80 The –CH₃O species are subsequently dehydrogenated through oxidation by h⁺, yielding coordinated HCHO and releasing atomic H. The H atoms are adsorbed and bridged on adjacent Pt sites, where they combine with other electron-acquired H atoms to form H₂.^81–83 HCHO molecules desorb into the solution and gradually polymerise into oligomeric species as the concentration increases.^75–77 Simultaneously, some adsorbed –CH₃O groups may couple with neighbouring HCHO to form methyl formate.⁸⁴ Adsorbed HCHO may also further oxidise by adjacent ˙OH radicals, resulting in the formation of HCOOH or other by-products.⁸⁵

Conclusions

In summary, this work systematically decouples the effects of irradiation intensity, wavelength and temperature on SC interfacial charge-transfer dynamics during photothermal catalysis, employing Pt/TiO₂ in LA aqueous solution as a model system. The study revealed that under isothermal conditions, increasing the irradiation intensity reduces the SC barrier, enhances electron transfer, and mitigates the detrimental effects associated with high photon flux. Under an equivalent condition, shorter wavelengths correspond to higher-energy photoexcited electrons, which exhibit a superior ability to overcome the SC barrier, thus promoting more efficient interfacial charge transfer—despite the inherent limitation of reduced penetration depth in the solution. Under a constant irradiation intensity, regardless of the Pt loading, a strong positive correlation was observed between the temperature and electron transfer kinetics, and the SC interfacial electron-transfer time constant decreased markedly as the temperature increased. To further minimise charge leakage and amplify the photothermal synergistic effect, an appropriate amount of Sn was doped into TiO₂. The optimised Pt/3% Sn–TiO₂ catalyst exhibits exceptional performance for CH₃OH dehydrogenation. Under 30 W UV irradiation (254 nm) at 80 °C, the average H₂ and HCHO production rates over 20 h reach 88.12 mmol g⁻¹ h⁻¹ (AQY: 40.02%) and 84.13 mmol g⁻¹ h⁻¹, respectively. Under 320 W UV irradiation (254 nm) at 80 °C, the average H₂ production rate within 6 h reaches 413.5 mmol g⁻¹ h⁻¹ (AQY: 25.16%). This study elucidates the roles of light and thermal energy in the SC interfacial electron-transfer behaviour during the photothermal catalytic process from the perspective of electron transfer kinetics, offering new insights into the design of highly efficient photothermal catalysts and the enhancement of photocatalytic performance.

Author contributions

H. K. Xiang: writing – original draft, investigation, funding acquisition, formal analysis, data curation. W. Q. Liu, Z. K. Zeng, G. L. Zhang, B. L. Lin & J. X. Chen: investigation, formal analysis. X. H. Wei & S. B. Zhou: resources. T. Q. Xiong: project administration and review. P. F. Li & L. B. Qian: resources, funding acquisition, and review.

Conflicts of interest

There are no conflicts to declare.

Data availability

The data supporting this article have been included as part of the supplementary information (SI). Supplementary information: DOCX file containing OCP behaviour, HR-TEM and SEM images, estimated time constant, Mott–Schottky curves, LSV curves, UV–vis absorption spectra, UPS spectra, photocurrent curves, photothermal catalytic HE performance data, GC–MS spectrum, and a performance comparison table. See DOI: https://doi.org/10.1039/d5gc03911h.

Original data are available from the authors upon reasonable request.

Acknowledgements

This work was supported by the National Natural Science Foundation of China (no. 12205089, 22402001), the Research and Development Fund of Hubei University of Science and Technology (Grant No. BK202403), the Open Fund of the Ministry of Education Engineering Research Centre for Phosphorus Resource Development and Utilization (LKF 202406), Horizontal Project from Hubei Zhongda Technology Co., Ltd (No. 2025HX028) and Wuhan Yuntian Feiran Technology Co., Ltd (No. 2025HX029), Science and Technology Program of Hubei Province (No. 2024AFB401) and Natural Science Research Project of Anhui Educational Committee (No. 2024AH051132).

During the preparation of this study the authors used AI-assisted technologies [https://ai.youdao.com/] to enhance the readability and language of the manuscript.

References

1. S. Chen, T. Takata and K. Domen, Nat. Rev. Mater., 2017, 2, 17050.
2. Y. Wang, A. Vogel, M. Sachs, R. S. Sprick, L. Wilbraham, S. J. A. Moniz, R. Godin, M. A. Zwijnenburg, J. R. Durrant, A. I. Cooper and J. Tang, Nat. Energy, 2019, 4, 746–760.
3. H. Nishiyama, T. Yamada, M. Nakabayashi, Y. Maehara, M. Yamaguchi, Y. Kuromiya, H. Tokudome, S. Akiyama, T. Watanabe, R. Narushima, S. Okunaka, N. Shibata, T. Takata, T. Hisatomi and K. Domen, Nature, 2021, 598, 304–307.
4. W. Li, W. Duan, G. Liao, F. Gao, Y. Wang, R. Cui, J. Zhao and C. Wang, Nat. Commun., 2024, 15, 6763.
5. M.-Y. Qi, M. Conte, M. Anpo, Z.-R. Tang and Y.-J. Xu, Chem. Rev., 2021, 121, 13051–13085.
6. M. Z. Rahman, T. Edvinsson and J. Gascon, Nat. Rev. Chem., 2022, 6, 243–258.
7. C. Huang, R. N. Ci, J. Qiao, X. Z. Wang, K. Feng, B. Chen, C. H. Tung and L. Z. Wu, Angew. Chem., Int. Ed., 2021, 60, 11779–11783.
8. R. Cui, W. Li, W. Duan, Y. Wang, Q. Gao, Y. Huang, X. Tuo, R. Li and W. Ma, Appl. Catal., B, 2025, 378, 125571.
9. D. Friedmann, A. Hakki, H. Kim, W. Choi and D. Bahnemann, Green Chem., 2016, 18, 5391–5411.
10. L. Xiong and J. Tang, Adv. Energy Mater., 2021, 11, 2003216.
11. H.-G. Jin, P.-C. Zhao, Y. Qian, J.-D. Xiao, Z.-S. Chao and H.-L. Jiang, Chem. Soc. Rev., 2024, 53, 9378–9418.
12. T. Salthammer, S. Mentese and R. Marutzky, Chem. Rev., 2010, 110, 2536–2572.
13. Z. Li, A. Ivanenko, X. Meng and Z. Zhang, J. Hazard. Mater., 2019, 380, 120822.
14. G. Garcia, E. Arriola, W.-H. Chen and M. D. De Luna, Energy, 2021, 217, 119384.
15. Y. Xu, C. Wang, X. Li, L. Xiong, T. Zhang, L. Zhang, Q. Zhang, L. Gu, Y. Lan and J. Tang, Nat. Sustainability, 2024, 7, 1171–1181.
16. E. Thomas and R. Philip, Physical Chemistry (2nd Edition), Pearson Education, Inc., Pearson Education Asia Ltd, & China Machine Press, Hongkong & Beijing, 2010.
17. H. Jiao, J. Yang, X. Li, C. Wang and J. Tang, Green Chem., 2022, 24, 8345–8354.
18. L. Sun, X. Yu, L. Tang, W. Wang and Q. Liu, Chin. J. Catal., 2023, 52, 164–175.
19. H. Zhang, P. Sun, X. Fei, X. Wu, Z. Huang, W. Zhong, Q. Gong, Y. Zheng, Q. Zhang, S. Xie, G. Fu and Y. Wang, Nat. Commun., 2024, 15, 4453.
20. G. Righi, J. Plescher, F.-P. Schmidt, R. K. Campen, S. Fabris, A. Knop-Gericke, R. Schlögl, T. E. Jones, D. Teschner and S. Piccinin, Nat. Catal., 2022, 5, 888–899.
21. R. Wang, M. Lv, J. Zhang, F. Wang, X. Wang, N. Wang, G. Yu and Z. Zheng, Chem. Eng. J., 2024, 502, 157981.
22. S. Tian, H. Guo, Q. Zhao, L. Liu, H. Wang, X. Wang, S. Zhang and W. Cui, Appl. Catal., B, 2024, 355, 124201.
23. Z. Gao, T. Montini, J. Mu, N. Luo, E. Fonda, P. Fornasiero and F. Wang, J. Am. Chem. Soc., 2024, 146, 24440–24449.
24. F. Gao, W. Li, W. Duan, G. Liao and C. Wang, Small, 2025, 21, 2503677.
25. H. Xiang, Z. Wang and J. Chen, J. Phys. Chem. Lett., 2021, 12, 7665–7670.
26. Z. Wang, B. Mei and J. Chen, J. Catal., 2022, 408, 270–278.
27. Z. Wang, N. Xue and J. Chen, J. Phys. Chem. C, 2019, 123, 24404–24408.
28. D. Friedmann, C. Mendive and D. Bahnemann, Appl. Catal., B, 2010, 99, 398–406.
29. B. E. Nieuwenhuys, O. G. Van Aardenne and W. M. H. Sachtler, Chem. Phys., 1974, 5, 418–428.
30. S. Trasatti, J. Electroanal. Chem. Interfacial Electrochem., 1972, 39, 163–184.
31. Y. Fan, Y. Jiang, H. Lin, J. Li, Y. Xie, A. Chen, S. Li, D. Han, L. Niu and Z. Tang, Nat. Commun., 2024, 15, 4679.
32. Y. Fan, W. Zhou, X. Qiu, H. Li, Y. Jiang, Z. Sun, D. Han, L. Niu and Z. Tang, Nat. Sustainability, 2021, 4, 509–515.
33. J. E. Whitten, Appl. Surf. Sci. Adv., 2023, 13, 100384.
34. S. C. Ray, D. K. Mishra, A. B. Panda, H. T. Wang, S. Bhattacharya and W. F. Pong, J. Phys. Chem. C, 2022, 126, 8947–8952.
35. F. Wooten, Optical Properties of Solids, ed. F. Wooten, Academic Press, New York and London, 1972.
36. R. A. Smith, Semiconductors, Cambridge University Press, London, New York, Melbourne, 1978.
37. A. Lasia, Electrochemical Impedance Spectroscopy and its Applications, ed. A. Lasia, Springer, New York, 2014.
38. K. Gelderman, L. Lee and S. W. Donne, J. Chem. Educ., 2007, 84, 685–688.
39. ASTM International, Standard Tables for Reference Solar Spectral Irradiances: Direct Normal and Hemispherical on 37° Tilted Surface (ASTM G 173-03), US Standard, 2020.
40. L. Dai and R. N. Jazar, Nonlinear Approaches in Engineering Applications, Springer Cham, 2022.
41. Y. Wang, X.-B. Li, H.-L. Wu, Y. Yang, Z. Liu, L.-P. Zhang, C.-H. Tung and L.-Z. Wu, ACS Sustainable Chem. Eng., 2019, 7, 7286–7293.
42. D. Gunawan, L. Y. Lau, J. A. Yuwono, P. V. Kumar, L. Oppong-Antwi, I. Kuschnerus, S. L. Y. Chang, R. K. Hocking, R. Amal, J. Scott and C. Y. Toe, Chem. Eng. J., 2024, 486, 150215.
43. F. Qin, Y. Kang, X. San, Y.-L. Tang, J. Li, X. Zhang, K. Zhang and G. Liu, J. Am. Chem. Soc., 2025, 147, 12897–12907.
44. H. Xiang, Z. Wang and J. Chen, Chin. J. Struct. Chem., 2022, 41, 2209069–2209073.
45. H. Xiang, J. Cheng, L. Tao, S. Yang, G. Zhang, J. Chen, Y. Wang, S. Zhou, H. Yu, L. Qian and T. Xiong, Chem. Eng. J., 2025, 511, 161968.
46. H. Huang, Y. Jin, Z. Chai, X. Gu, Y. Liang, Q. Li, H. Liu, H. Jiang and D. Xu, Appl. Catal., B, 2019, 257, 117869.
47. F. Xing, R. Zeng, C. Cheng, Q. Liu and C. Huang, Appl. Catal., B, 2022, 306, 121087.
48. R. Li and C. Li, in Advances in Catalysis, ed. and C. Song, Academic Press, Cambridge, San Diego, London, 2017, ch. 1, pp. 1–57.
49. Y. Duan, N. Fu, Q. Liu, Y. Fang, X. Zhou, J. Zhang and Y. Lin, J. Phys. Chem. C, 2012, 116, 8888–8893.
50. S. Mehraz, P. Kongsong, A. Taleb, N. Dokhane and L. Sikong, Sol. Energy Mater. Sol. Cells, 2019, 189, 254–262.
51. X. Bi, G. Du, A. Kalam, D. Sun, Y. Yu, Q. Su, B. Xu and A. G. Al-Sehemi, Chem. Eng. Sci., 2021, 234, 116440.
52. Y. Zhang, L. Zhang, L. Tang, R. Du and B. Zhang, ACS Nano, 2024, 18, 8411–8422.
53. J. Li and H. C. Zeng, J. Am. Chem. Soc., 2007, 129, 15839–15847.
54. J. A. Dorman, J. Weickert, J. B. Reindl, M. Putnik, A. Wisnet, M. Noebels, C. Scheu and L. Schmidt-Mende, J. Phys. Chem. C, 2014, 118, 16672–16679.
55. S.-H. Chen, C.-M. Ho, Y.-H. Chang, K.-M. Lee and M.-C. Wu, Chem. Eng. J., 2022, 445, 136761.
56. H. Xiang, T. Luo, Y. Ji, T. Xiong, L. Qian, S. Yang and H. Wang, Environ. Technol. Innovation, 2024, 36, 103804.
57. A. Wiatrowski, M. Mazur, A. Obstarczyk, D. Wojcieszak, D. Kaczmarek, J. Morgiel and D. Gibson, Coatings, 2018, 8, 412.
58. J. Bisquert, A. Zaban, M. Greenshtein and I. Mora-Seró, J. Am. Chem. Soc., 2004, 126, 13550–13559.
59. D. Yang, Z. Wang and J. Chen, Catal. Sci. Technol., 2022, 12, 3634–3638.
60. Y. Tang, Z.-F. Xu, Y. Sun, C. Wang, Y. Guo, W. Hao, X. Tan, J. Ye and T. Yu, Energy Environ. Sci., 2024, 17, 7882–7894.
61. T. Ge and J. Chen, J. Phys. Chem. Lett., 2024, 15, 1241–1245.
62. M. Wu, Y. Yang, S. Ma, X. Guo, H. Yu, Y. Cao, J. Wu, L. Liu and T. Yao, Sep. Purif. Technol., 2025, 356, 129832.
63. H. Xiang, Z. Wang and J. Chen, J. Chem. Phys., 2021, 154, 221102.
64. H. J. Snaith, L. Schmidt-Mende, M. Grätzel and M. Chiesa, Phys. Rev. B: Condens. Matter Mater. Phys., 2006, 74, 045306.
65. H. Kisch, Angew. Chem., Int. Ed., 2010, 49, 9588–9589.
66. X. Wei, H. Xiang, Y. Xu, Z. Wang and J. Chen, AIChE J., 2023, 69, e17985.
67. J. Chen and H. Xiang, China Patent, 201922088195.0, 2019.
68. J. Rumble, CRC Handbook of Chemistry and Physics (104th Edtion), CRC Press, 2023.
69. D. Kundu, S. Zamindar, S. K. Tudu, S. Maji, A. Hazra, P. Ghorai, S. Samanta, N. C. Murmu and P. Banerjee, ChemCatChem, 2025, e00796.
70. D. Kundu, G. Murmu, S. Zamindar, N. C. Murmu, P. Banerjee and S. Saha, ACS Appl. Energy Mater., 2024, 7, 8423–8435.
71. C. Fu, F. Li, J. Zhang, D. Li, K. Qian, Y. Liu, J. Tang, F. Fan, Q. Zhang, X.-Q. Gong and W. Huang, Angew. Chem., Int. Ed., 2021, 60, 6160–6169.
72. M. El-Roz, M. Kus, P. Cool and F. Thibault-Starzyk, J. Phys. Chem. C, 2012, 116, 13252–13263.
73. G. D. Belletti, E. Colombo, N. Cabana, P. Quaino and S. Collins, Top. Catal., 2022, 65, 915–925.
74. B.-B. Xu, M. Zhou, M. Ye, L.-Y. Yang, H.-F. Wang, X. L. Wang and Y.-F. Yao, J. Am. Chem. Soc., 2021, 143, 10940–10947.
75. J. Kim, B. D. Kay and Z. Dohnálek, J. Phys. Chem. C, 2010, 114, 17017–17022.
76. S. K. Gupta and A. Kumar, in Reaction Engineering of Step Growth Polymerization, ed. S. K. Gupta and A. Kumar, Springer US, Boston, MA, 1987.
77. J. F. Walker, Formaldehyde (Third Edition), Robert E. Krieger Publishing Co., Inc., New York, 1975.
78. Z. Li, Z. Zhang, B. D. Kay and Z. Dohnálek, J. Phys. Chem. C, 2011, 115, 9692–9700.
79. K. R. Phillips, S. C. Jensen, M. Baron, S.-C. Li and C. M. Friend, J. Am. Chem. Soc., 2013, 135, 574–577.
80. Z. Yan, X. Yang, G. Gao, R. Gao, T. Zhang, M. Tian, H. Su and S. Wang, J. Catal., 2022, 411, 31–40.
81. H. Zhang, X. Yang, G. Gao, J. Yan, M. Zhao and H. Su, Catal. Sci. Technol., 2022, 12, 773–785.
82. T. A. Kandiel, R. Dillert, L. Robben and D. W. Bahnemann, Catal. Today, 2011, 161, 196–201.
83. R. Gao, A. Safrany and J. Rabani, Radiat. Phys. Chem., 2002, 65, 599–609.
84. H. Zhang, X. Yang, G. Gao, Y. Liu, M. Zhao, J. Yan, H. Du, X. Xiao and H. Su, Fuel, 2024, 366, 131391.
85. Z. Li, X. Wang, L. Jia and X. Xing, Catal. Commun., 2017, 92, 23–26.

---