Aluminum Cation Doping in Ruddlesden-Popper Sr₂TiO₄ Enables High-Performance Photocatalytic Hydrogen Evolution

Abstract

Hydrogen (H₂) is regarded as a promising and renewable energy carrier to achieve a sustainable future. Among the various H₂ production routes, photocatalytic water splitting has received particular interest; it strongly relies on the optical and structural properties of photocatalysts such as their sunlight absorption capabilities, carrier transport properties, and amount of oxygen vacancy. Perovskite oxides have been widely investigated as photocatalysts for photocatalytic water splitting to produce H₂ because of their distinct optical properties, tunable band gaps and excellent compositional/structural flexibility. Herein, an aluminum cation (Al³⁺) doping strategy is developed to enhance the photocatalytic performance of Ruddlesden-Popper (RP) Sr₂TiO₄ perovskite oxides for photocatalytic H₂ production. After optimizing the Al³⁺ substitution concentration, Sr₂Ti_0.9Al_0.1O₄ exhibits a superior H₂ evolution rate of 331 μmol h⁻¹ g⁻¹, which is ~3 times better than that of Sr₂TiO₄ under full-range light illumination, due to its enhanced light harvesting capabilities, facilitated charge transfer, and tailored band structure. This work presents a simple and useful Al³⁺ cation doping strategy to boost the photocatalytic performance of RP-phase perovskites for solar water splitting.

1. Introduction

The energy crisis, greenhouse effect, and environmental issues caused by the overuse of fossil fuels have received increased attention recently. Thus, it is crucial to develop renewable and sustainable energy resources (e.g., wind, solar) and relevant technologies [1,2,3]. Solar energy is regarded as the most attractive renewable energy due to its abundance and clean nature; it can also be utilized by various routes [4,5,6,7,8,9]. Among them, photovoltaics and photocatalysis for the conversion of solar energy to fuels and electric power have received considerable attention for their efficient utilization of sunlight energy [10,11,12,13,14,15]. Principally, semiconductors used in photovoltaics directly transform solar energy to electricity, while semiconductors in photocatalysis enable the conversion from sunlight energy to chemical energy by means of photoinduced charge carriers for wastewater treatment and photocatalytic water splitting [10,12,13,16,17,18,19,20]. More importantly, photocatalysis is crucial in wastewater treatment and water splitting for hydrogen (H₂) production to address worldwide energy demands and environmental pollution [21,22,23,24]. H₂ is regarded as a highly promising energy carrier because clean water is only combustion product, while photocatalytic water splitting is considered one of the most promising methods of H₂ generation, and the photocatalytic efficiency is determined by the photocatalysts [25,26,27,28]. Consequently, it is crucial to develop cost-effective and earth-abundant photocatalysts with advanced water-splitting abilities [29,30,31].

At present, various types of photocatalysts have been designed and developed for photocatalytic water splitting, such as metal oxides, metal sulfides, carbon nitrides, etc. [32,33,34,35]. Simple metal oxides (e.g., TiO₂ and ZnO) have been extensively investigated in relation to photocatalytic water splitting, showing relatively high photocatalytic activity [36,37]. Nevertheless, the performance of the reported simple oxide-based photocatalysts is significantly limited by their fixed atomic environment, wide band gap, and inferior stability, which limit their photocatalytic efficiency [38,39]. As a result, complex metal oxides are thought to be more applicable as high-performance photocatalysts because of their rich and tunable atomic environments, which exhibit easily tailored optical, chemical, and structural properties [40,41,42,43,44]. In particular, perovskite oxides, a significant class of complex oxides, have received increasing attention in relation to photocatalytic H₂ evolution reaction (HER) due to their easily tailored band gaps/structures, cost effectiveness, compositional/structural flexibility, and superior stability [12,40,41,42]. Additionally, the physical, optical, and electronic properties of conventional perovskites oxides (ABO₃) with three-dimensional structures can be effectively tuned by reducing the dimensions [12,40]. Notably, two-dimensional Ruddlesden-Popper (RP) perovskite oxides have gained increasing attention on account of their superior photocatalytic performance and unique chemical, structural, electronic, and optical features, as well as the fact that they play a promoting role in enhancing the separation and transportation of photo-induced charge carriers [45,46]. Sr₂TiO₄, a typical RP-phase perovskite, has been employed as a photocatalyst for water splitting and dye degradation, but exhibited inferior visible-light-driven photocatalytic performance because of the wide band gap of ~3.5 eV [46].

Functional doping is reported to be a useful and simple strategy to improve the photocatalytic performance of RP-phase Sr₂TiO₄ photocatalysts for solar water splitting by tuning the band structure/gap, charge transfer capability, and oxygen vacancy concentrations [47,48,49]. For instance, Xiao et al. reported a synergistic strategy by combining silver (Ag⁺) cation doping and reducing pretreatment to enhance the photocatalytic activity of Sr₂TiO₄ [47]. The optimized r-Ag_0.05Sr_1.95TiO₄, with a suitable amount of Ag⁺ cation substitution and reduced pretreatment, displayed a superior H₂ generation rate of 541 μmol h⁻¹ g⁻¹ under visible light illumination due to the enlarged specific surface area, the optimized band gap value, the enhanced light-harvesting capability, the presence of more surface oxygen vacancies, and reduced carrier recombination. In another work, Han et al. employed a fluorine (F⁻) anion doping strategy to boost the photocatalytic performance of Sr₂TiO₄ for H₂ production [48]. Sr₂TiO_3.97F_0.03 with an optimized amount of F⁻ doping displayed outstanding H₂ production performance (282 μmol h⁻¹ g⁻¹) under full-range light irradiation, 44% higher than that of pristine Sr₂TiO₄, because of the increased amount of surface defect, the more strongly negative position of the conduction band, the reduced carrier recombination and well-balanced band gap value, the specific surface area, and the grain size. Nevertheless, the above-mentioned Ag⁺ and F⁻ doped Sr₂TiO₄ suffer from low doping concentrations of foreign ions (≤5 mol %), which may limit the improvement of the photocatalytic performance of Sr₂TiO₄. Meanwhile, metal cation doping with higher oxidation states (e.g., Rh⁴⁺) into RP-phase Sr₂TiO₄ competitively consumed photo-generated electrons, and played a detrimental role in the photocatalytic activity [50,51]. To solve the above-mentioned problems, we report a simple and useful strategy to increase the photocatalytic HER performance of Sr₂TiO₄ using a doping aluminum (Al³⁺) cation with a lower oxidation state and a higher doping concentration in the B-site of RP-phase Sr₂TiO₄.

Herein, we report a simple and useful strategy to increase the photocatalytic HER performance of Sr₂TiO₄ by doping an Al³⁺ cation into the B-site with higher doping concentrations. After optimizing the Al³⁺ doping concentration, Sr₂Ti_0.9Al_0.1O₄ demonstrated an increased photocatalytic H₂ production rate of 331 μmol h⁻¹ g⁻¹ under full-range light illumination (λ ≥ 250 nm), which was nearly three times better than that of Sr₂TiO₄, due to the reduced particle sizes, suppressed band gaps, stronger light harvesting capabilities, and improved separation and transfer capabilities of the photoinduced charge carriers. This study presents a new and efficient RP-type photocatalyst for high-efficiency water splitting, and provides insights that will enable the acceleration of large-scale applications of photocatalytic water splitting.

2. Materials and Methods

Sr₂Ti_1−xAl_xO₄ (x = 0, 0.05, 0.1, and 0.15) with various Al³⁺ doping concentrations was prepared by the typical solid-state reaction route, using SrCO₃, TiO₂, and Al₂O₃ with different molar ratios as the precursors; these were mixed well in ethanol by 60-min high-energy ball milling. After drying, the mixed precursors were further annealed in air at 1000 °C for 10 h to obtain the final photocatalysts. The phase structures, microstructures, and chemical states of various elements in the as-prepared photocatalysts were analyzed by X-ray diffraction (XRD, Bruker D8 Advance), field emission scanning electron microscopy (FESEM, Hitachi S4800), and with an XPS spectrometer (Thermo ESCALAB 250), respectively. The energy dispersive X-ray spectrometer (EDX) was used to investigate the elemental compositions of various samples. The Autosorb-iQ3 instrument was employed to measure the Brunauer–Emmett–Teller (BET) surface areas of various samples. The ultraviolet-visible (UV-vis) spectrophotometer (Lambda 750S) and FL 6500 photoluminescence (PL) spectrometer were employed to measure the diffuse reflectance spectra and fluorescence emission of the as-prepared photocatalysts, respectively. The photocatalytic performance of various photocatalysts was determined using a photocatalytic performance evaluation instrument (CEL-PAEMD8, China Education Au-light); detailed information regarding this process can be found in our previous work [48].

A three-electrode system was used to investigate the photoelectrochemical properties of photocatalysts in a 0.1 M Na₃PO₄/Na₂HPO₄ aqueous solution (pH = 7.9) under full-range light irradiation on a CHI760E workstation. The photocatalyst powder-based slurries were spin-coated on F⁻ doped tin oxide (FTO) glass to serve as the working electrode [48], while Ag/AgCl and Pt functioned as the reference and counter electrodes, respectively. Cyclic voltammetry of 50 cycles at a scan rate of 200 mV s⁻¹ was used to activate the working electrode. After that, the linear sweep voltammetry (LSV) curves were obtained at a scan rate of 10 mV s⁻¹ and a voltage range of −1.6–0 V. Nyquist plots of various photocatalysts were obtained using electrochemical impedance spectroscopy (EIS) at −1.2 V under light irradiation with frequencies in the range of 0.1–10⁵ Hz, using the CHI760E workstation.

3. Results and Discussion

Generally, a high calcination temperature (≥1000 °C) is required for the synthesis of Sr₂TiO₄ using the solid-state reaction route, due to the high energy barrier of SrO layer intercalation into the SrTiO₃ with an ABO₃ structure [45]. Figure 1 shows the XRD patterns of various Sr₂TiO₄ samples with different Al³⁺ doping amounts calcined at 1000 °C; the main diffraction peaks of the four samples are consistent with the peaks of RP-type Sr₂TiO₄ (JCPDS No. 00-039-1471), suggesting that Al³⁺ doping had no obvious impact on the phase structure of Sr₂TiO₄. It was found that the main peaks (31° ≤ 2θ ≤ 33°) of the Sr₂Ti_{1 − x}Al_xO₄ samples exhibited shifts to higher 2θ values, as compared with the pristine Sr₂TiO₄ in Figure 1, implying that the Al³⁺ cation with a smaller ionic radius (0.535 Å) than that of Ti⁴⁺ (0.605 Å) and Ti³⁺ (0.67 Å) cations was successfully doped in the lattice of RP-phase Sr₂TiO₄ with a lattice contraction [52,53]. Nevertheless, when the doping ratio of the Al³⁺ cation was higher than 0.1, an Sr₄Ti₃O₁₀ (JCPDS: No. 00-022-1444) impurity phase appeared in the Sr₂Ti_0.85Al_0.15O₄ sample, which may have a detrimental effect on photocatalytic activity [54].

The microstructures, particle sizes, and elemental distributions of Al³⁺-doped Sr₂TiO₄ were investigated using SEM and EDX elemental mapping, which showed a significant effect on the photocatalytic performance. SEM images and the corresponding grain size histograms are provided in Figure 2a–d. It seems that the particles of various Sr₂TiO₄-based photocatalysts exhibited irregular shapes, with particle size distributions in the range of 150–300 nm. Although the Al³⁺ doping efficiently decreased the particle sizes of the pristine Sr₂TiO₄ photocatalyst, the more serious particle agglomeration inevitably reduced the BET specific surface areas, as shown in Figures S1 and S2 and Table S1. Furthermore, the corresponding EDX elemental mapping image in Figure 3 illustrates the well-distributed Sr, Ti, and O elements in pristine Sr₂TiO₄ and Sr, Ti, Al, and O elements in Sr₂Ti_0.9Al_0.1O₄, further confirming that Al³⁺ cations were successfully doped in the lattice of Sr₂TiO₄.

The XPS technique was employed to investigate the influence of Al³⁺ doping on the chemical states of various elements in Sr₂TiO₄, with results shown in Figure 4 and Figure S2. As depicted in Figure 4a, all samples exhibited two main peaks at binding energy (BE) positions of around 458.0 and 464.0 eV in Ti 2p XPS spectra, corresponding to Ti 2p_3/2 and Ti 2p_1/2, respectively [55]. As compared with pristine Sr₂TiO₄, the Ti XPS peaks of the Al³⁺ doped samples shifted to the lower BE positions, demonstrating the formation of Ti³⁺ cations induced by the Al³⁺ doping. In addition, O 1s XPS spectra were mainly composed of four peaks at BEs of around 530.1, 531.1, 532.0, and 533.2 eV, representing the lattice oxygen, ${\mathrm{O}}_2^{2-}$/O⁻ species, the surface-absorbed hydroxyl groups/oxygen (-OH/O₂), and absorbed carbonates/water on the surface (CO₃²⁻/H₂O), respectively [56]. In fact, the content of surface oxygen vacancy as a crucial factor in determining the photocatalytic performance was closely tied to the ${\mathrm{O}}_2^{2-}$/O⁻ species [57]. It can be seen that the number of ${\mathrm{O}}_2^{2-}$/O⁻ species for Sr₂Ti_0.9Al_0.1O₄ was much higher than those of the other four samples, implying the highest amount of surface oxygen vacancies. Based on the Al 2p XPS spectra in Figure 4c, the intensity of the typical peak for Al at a BE of around 74 eV was gradually increased with the increase in the Al³⁺ doping concentration from 0.05 to 0.10, suggesting that the Al³⁺ cations of suitable amounts were successfully doped into the lattice of Sr₂TiO₄.

Besides the phase structure and morphology, the light-harvesting capability of the photocatalyst is another decisive factor for photocatalytic HER activity [47]. UV-vis spectra were used to identify the impact of Al³⁺ substitution amounts on the sunlight-harvesting capabilities of Sr₂TiO₄. As depicted in Figure 5a, all samples exhibited strong light absorption in the ultraviolet (UV) region. Moreover, the increase in the Al³⁺ doping concentration from 0 to 0.10 in Sr₂TiO₄ gradually led to improved light absorption capabilities at higher wavelengths because of the reduced band gaps. In addition, Sr₂Ti_0.9Al_0.1O₄ exhibited a stronger absorbance intensity and light harvesting capability than Sr₂Ti_0.85Al_0.15O₄ at wavelengths larger than 375 nm. Although Sr₂Ti_0.85Al_0.15O₄ exhibited stronger absorbance intensity at wavelengths from 340–375 nm, the inferior photocatalytic performance of Sr₂Ti_0.85Al_0.15O₄, compared to that of Sr₂Ti_0.9Al_0.1O₄, was mainly attributed to the reduced specific surface area, wider band gap, larger charge transfer resistance and the existence of Sr₄Ti₃O₁₀ impurity. The Kubelka–Munk transformation displayed in Figure 5b was employed to obtain the band gap energies (E_g) of various Sr₂TiO₄ photocatalysts. Sr₂TiO₄ displayed an E_g of 3.32 eV, agreeing well with the reported E_g for Sr₂TiO₄ in previous investigations [58]. For Al³⁺-doped Sr₂TiO₄, decreased E_g values of 3.20, 3.06, and 3.16 eV were observed for Sr₂Ti_0.95Al_0.05O₄, Sr₂Ti_0.9Al_0.1O₄, and Sr₂Ti_0.85Al_0.15O₄, respectively, as displayed in Figure 5b and Table S1; these results suggest that the Al³⁺ doping is effective in narrowing the band gap of Sr₂TiO₄. Notably, Sr₂Ti_0.9Al_0.1O₄ exhibited the lowest E_g with much-enhanced light-harvesting capability, which was beneficial for improving the photocatalytic activity.

It has been widely reported that the separation, transport, and recombination behaviors of photogenerated charge carriers play crucial roles in the photocatalytic activity of photocatalysts, which can be characterized using the PL technique [59]. The PL spectra of the Sr₂TiO₄ and various Al³⁺ doped Sr₂TiO₄ are displayed in Figure 5c. Under excitation at a wavelength of 380 nm, all of the investigated samples exhibited emission peaks at a wavelength range of 420–550 nm. As compared with Sr₂TiO₄, Sr₂Ti_0.95Al_0.05O₄ and Sr₂Ti_0.9Al_0.1O₄ showed lower PL emission peak intensities, indicating the facilitated separation and transport capabilities of the charge carriers. Nevertheless, Sr₂Ti_0.85Al_0.15O₄ with excessive Al³⁺ doping exhibited a higher PL emission peak intensity than that of pristine Sr₂TiO₄ because of the inhibited separation and transport of photoexcited carriers induced by the existence of the impurity phase in Sr₂Ti_0.85Al_0.15O₄.

The photocatalytic activities of the as-prepared Sr₂Ti_{1 − x}Al_xO₄ (x = 0, 0.05, 0.1, and 0.15) photocatalysts were evaluated by testing the H₂ production amounts in a mixed Na₂SO₃/Na₂S aqueous solution under full-range light irradiation, in which Na₂SO₃ and Na₂S functioned as sacrificial agents to react with the photoinduced holes. Figure 6a displays steady and continuous H₂ production during photocatalytic water splitting under full-range light illumination, and the H₂ production rates of all samples are presented in Figure 6b. Sr₂Ti_1−xAl_xO₄ (x = 0, 0.05, 0.1, and 0.15) exhibited better photocatalytic HER activity than pristine Sr₂TiO₄, demonstrating the effectiveness of Al³⁺ doping in boosting the photocatalytic activity. Notably, Sr₂Ti_0.9Al_0.1O₄ exhibited the maximum H₂ evolution rate (of 331 μmol h⁻¹ g⁻¹) among the three Al³⁺ doped samples, which was nearly three times higher than that of the undoped Sr₂TiO₄ (85 μmol h⁻¹ g⁻¹) due to the enhanced light-harvesting capabilities, suppressed carrier recombination, and reduced E_g value. In addition, Sr₂Ti_0.85Al_0.15O₄ with excessive amounts of Al³⁺ doping exhibited an inferior H₂ generation rate to that of Sr₂Ti_0.9Al_0.1O₄ due to the inhibited separation and transport of photoinduced carriers and the large E_g value. As for the photocatalytic stability of HER, the Sr₂Ti_0.9Al_0.1O₄ photocatalyst exhibited steady and continuous H₂ production for 15 h (Figure 7a). The amount of H₂ produced with Sr₂Ti_0.9Al_0.1O₄ was measured to 82.6 μmol in the first 5 h, and the initial H₂ production rate was 328 μmol h⁻¹ g⁻¹. In addition, the H₂ evolution amount was increased to 219.8 μmol after 15 h of operation and the average H₂ evolution rate was 292 μmol h⁻¹ g⁻¹, which was nearly 90% of the initial value after 15 h of continuous operation (Figure 7a). As compared with the sample before the photocatalytic reaction, the Sr₂Ti_0.9Al_0.1O₄ photocatalyst exhibited no obvious changes in the phase structure and particle sizes after the stability test under full-range light illumination, as depicted in Figure 7b,d,e. As shown in Figure 7c, the cycling performance of Sr₂Ti_0.9Al_0.1O₄ photocatalyst was determined by testing the HER rates in three cycles. The Sr₂Ti_0.9Al_0.1O₄ photocatalyst exhibited a high recyclability without any obvious degradation in the HER rates (~297 μmol h⁻¹ g⁻¹).

The LSV curves of various Sr₂Ti_{1 − x}Al_xO₄ (x = 0, 0.05, 0.1, and 0.15) photocatalysts were measured under full-range light illumination to further verify the enhanced photocatalytic activity of Sr₂Ti_0.9Al_0.1O₄. As depicted in Figure 8a, after the cyclic voltammetry activation, the photocatalysts showed similar onset potential for the cathodic current. In addition, Sr₂Ti_0.9Al_0.1O₄ exhibited larger current densities than those of Sr₂TiO₄, Sr₂Ti_0.95Al_0.05O₄, and Sr₂Ti_0.85Al_0.15O₄, demonstrating the superior photocatalytic HER activity of Sr₂Ti_0.9Al_0.1O₄. Based on the EIS results in Figure 8b, the Sr₂Ti_0.9Al_0.1O₄ photocatalysts exhibited the smallest charge transfer resistance among the four investigated samples due to the suppressed recombination of photogenerated carriers, as evidenced by the PL results, which were consistent with the photocatalytic HER activities of various photocatalysts under irradiation.

Mott–Schottky curves were further tested to measure the energy band positions, including the conduction bands (CB) and the valence bands (VB) of the Sr₂TiO₄ and Sr₂Ti_0.9Al_0.1O₄ photocatalysts. As shown in Figure 8c, the flat band potentials of the photocatalysts can be acquired by extrapolating the Mott–Schottky curves to the X-axis. It can be observed that the flat band positions of Sr₂TiO₄ and Sr₂Ti_0.9Al_0.1O₄ were measured to be −0.37 and −0.68 eV (vs. Ag/AgCl), respectively. Thus, the CB edges of Sr₂TiO₄ and Sr₂Ti_0.9Al_0.1O₄ were calculated to be −0.17 and −0.48 eV (vs. the normal hydrogen electrode, NHE), respectively. Based on the E_g values measured by the Kubelka–Munk transformation, the VB edges of Sr₂TiO₄ and Sr₂Ti_0.9Al_0.1O₄ were calculated to be 3.15 and 2.58 eV (vs. NHE), respectively. Based on the schematic diagrams of the band structures of Sr₂TiO₄ and Sr₂Ti_0.9Al_0.1O₄, as shown in Figure 8d, it seems that the Al³⁺ doping effectively regulated the band positions of Sr₂TiO₄. The CB edge of Sr₂Ti_0.9Al_0.1O₄, which was more strongly negative than that of the pristine Sr₂TiO₄, enabled a much higher driving force of the photoexcited electrons for water reduction, and led to a significant enhancement of the photocatalytic HER performance of Sr₂TiO₄.

4. Conclusions

In conclusion, we successfully fabricated several Sr₂Ti_1−xAl_xO₄ (x = 0, 0.05, 0.1, and 0.15) photocatalysts using a conventional solid-state reaction method. The Sr₂Ti_0.9Al_0.1O₄ photocatalyst exhibited the most outstanding photocatalytic HER activity (331 μmol h⁻¹ g⁻¹) among the three investigated Al³⁺-doped Sr₂TiO₄ photocatalysts under full-range light irradiation, which was nearly three times higher than that of Sr₂TiO₄. Such a remarkable improvement in the photocatalytic performance of Sr₂TiO₄ induced by Al³⁺ doping was attributed to the reduced particle sizes, enhanced light harvesting capability, facilitated charge transfer, and tailored band structure. Furthermore, Sr₂Ti_0.9Al_0.1O₄ also exhibited good stability and recyclability for photocatalytic HER. In sum, this study reports a highly promising catalyst (Sr₂Ti_0.9Al_0.1O₄) for photocatalytic water splitting, which may contribute to the further development of perovskite-based photocatalysts for efficient solar conversion.