Relating the structure, properties, and activities of nanostructured SrTiO₃ and SrO–(SrTiO₃)_n (n = 1 and 2) for photocatalytic hydrogen evolution

Abstract

This study focuses on relating the structure of perovskite oxides with their properties and activities and provides a comparative study of the three members of the Sr–Ti–O system for photocatalytic hydrogen evolution. The three oxides focused on in this study are based on perovskite structure viz. SrTiO₃ and SrO–(SrTiO₃)_n (n = 1 and 2). We have successfully synthesized these three oxides through a methodology that combined the polymeric citrate precursor method with the hydrothermal method. Their crystal structure, morphology, and optical properties (absorption and photoluminescence) were systematically explored. SrTiO₃ belonged to a class of cubic perovskite while Sr₂TiO₄ (n = 1) and Sr₃Ti₂O₇ (n = 2) belonged to layered Ruddlesden–Popper based perovskite oxides. We observed the cube-shaped morphology for nanostructured SrTiO₃ and layered morphology for Ruddlesden–Popper based oxides, Sr₂TiO₄ and Sr₃Ti₂O₇. The photocatalytic hydrogen evolution performance of these nanostructured oxides was investigated. Amongst the three nanostructured oxides, the maximum amount of hydrogen was evolved with Sr₃Ti₂O₇ as the photocatalyst. These results were supported by photoluminescence, time-resolved photoluminescence, and photoelectrochemical studies.

---

1. Introduction

Production of hydrogen energy via the utilization of solar energy is a sustainable and effective solution to the world's energy crisis and environmental problems. In this regard, photocatalytic splitting of water into H₂ and O₂ over a semiconducting catalyst has been recognized as an effective strategy to develop a sustainable energy structure.^1,2 The efficiency of a photocatalytic reaction can be enhanced by modulating several factors such as crystal structure,³ electronic structure⁴ (i.e. position of the conduction band and valence band edge), the presence of vacancies (anion or cation)⁵ and controlling the morphology.^6,7

Over the past few decades, several semiconductor oxides such as TiO₂, ZnO, AgGaO₂, and BiVO₄ have been discovered as efficient catalysts for photocatalytic hydrogen evolution.^8–14 Among all the developed photocatalysts, perovskite structures i.e. ABO₃ have gained much popularity because of their composition which can be easily modulated at A and B sites.^15–17 Among the various perovskite semiconductor oxides, SrTiO₃ is widely studied as a photocatalyst^18–20 for its outstanding structural stability and compositional flexibility. Ruddlesden Popper-based layered perovskites, with a general formula of AO–(ABO₃)_n, (which can be also written as (A_n+1BO_3n+1) or A′₂[A_n−1B_nO_3n+1]) have attracted much attention in the field of photocatalytic water splitting due to their physical properties which can be modified by intercalation and ion-exchange.^21,22 These structures consist of a large number of active sites such as the B site and the AO layer site to facilitate the reaction. Several metal ions can be doped^23,24 at the A and B sites in these structures which modulates the bandgap of the oxide towards the visible region thereby enhancing their photocatalytic activity.

Fig. 1 shows the crystal structure of SrTiO₃ and SrO–(SrTiO₃)_n (n = 1 and 2). SrTiO₃ belongs to the class of perovskite that has a cubic structure composed of corner shared TiO₆ octahedra with Sr, present in the holes formed from the cuboctahedron symmetry. When n = 1 and 2, the structure belongs to a class of oxide known as Ruddlesden–Popper phases. The two-lower symmetry 2D-Ruddlesden–Popper oxides, Sr₂TiO₄ (n = 1) and Sr₃Ti₂O₇ (n = 2) have body-centered tetragonal symmetry. These structures are composed of stacked nSrTiO₃ perovskite layers separated by a SrO rock-salt type layer, as shown in Fig. 1.

Fig. 1 Schematic showcasing the crystal structure of SrTiO₃ and SrO–(SrTiO₃)_n for n = 1 (Sr₂TiO₄) and n = 2 (Sr₃Ti₂O₇).

Here, we discuss a comparative study of the performance of nanostructured SrTiO₃ and SrO–(SrTiO₃)_nviz. Sr₂TiO₄ (n = 1) and Sr₃Ti₂O₇ (n = 2) towards photocatalytic hydrogen evolution reactions. Out of these three perovskite oxides, Sr₂TiO₄ and Sr₃Ti₂O₇ belong to I4/mmm, tetragonal symmetry whereas SrTiO₃ belongs to the Pm3̄m space group. The photocatalytic property of these nanostructured oxides is investigated through hydrogen evolution reactions under UV-visible light irradiation. Photoluminescence, time-resolved photoluminescence, and photoelectrochemical studies have been carried out to see the role of the perovskite structure in influencing photocatalytic activity. To the best of our knowledge, there are no reports on a comparative study of the three kinds of structures on photocatalytic hydrogen evolution performance. However, there is one report²⁵ where the ratio of Sr/Ti in SrTiO₃ was varied and their photocatalytic HER was studied for different ratios from 1.00 to 1.25. The photocatalytic hydrogen evolution activity of these three perovskite oxides has been studied separately.^26–28 In this study, we have tried to highlight that the choice of structure (crystallographic), amongst the cubic and layered perovskites (showcased here with the Sr–Ti–O system), could be an efficient way for the development of a catalyst for a hydrogen evolution reaction with improved performance.

2. Experimental

2.1 Materials and methods

Strontium nitrate [(Sr(NO₃)₂), 99%], ethylene glycol (EG) (99%), methanol (ACS grade), citric acid anhydrous (98%), and sodium sulfate (99%) were purchased from Merck. Sodium sulfide flakes were purchased from CDH fine chemicals. Titanium tetraisopropoxide (TTIP) (97%) and sodium sulfite (98%) were purchased from Sigma-Aldrich.

For the synthesis of SrTiO₃, TTIP (12 mmol) was added to a mixture of ethylene glycol (25 mL) and methanol (10 mL). To this aqueous solution of citric acid in 5 mL water and solid Sr(NO₃)₂ (12 mmol) were added. The amount of citric acid taken was equivalent to that of TTIP. The resultant mixture was heated at 130 °C for 4 hours to form a single-phase transparent solution. Afterward, the pH of this solution was adjusted to 13 by adding 5 M NaOH (5 mL). The resulting solution was transferred into a Teflon vessel followed by hydrothermal treatment at 200 °C for 48 h. After the Teflon vessel was cooled down to room temperature, the obtained gel was vacuum dried at 200 °C for 16 h. The products were washed several times with glacial acetic acid (to remove carbonate present as an impurity in the sample), deionized (DI) water, and ethanol, and then dried overnight at 70 °C. SrTiO₃ was finally synthesized by calcining the dried powder at 750 °C for 5 h.

For the synthesis of Sr₂TiO₄ and Sr₃Ti₂O₇, the exact procedure was followed as for SrTiO₃ except for the ratio of Sr and Ti precursor and heating temperature. For Sr₂TiO₄, the ratio of Sr and Ti precursor was kept at 2 : 1 (12 mmol of Sr(NO₃)₂ and 6 mmol of TTIP) and calcined first at 650 °C and then at 1000 °C for 12 h, whereas for Sr₃Ti₂O₇ the ratio was kept at 3 : 2 (9 mmol of Sr(NO₃)₂ and 6 mmol of TTIP) and calcined first at 650 °C and then at 950 °C for 6 h.

2.2 Characterization

Phase identification and crystal structure of the samples were analyzed using powder X-ray diffraction (PXRD, Bruker D8 Advance Eco) with Cu Kα as an X-ray source (λ = 0.15406 nm). The step size of 0.02 with a time of 0.3 s per step was used for data collection. Crystallite size (Scherrer) was calculated using TOPAS v5 software by using the peak positioned at a two-theta value of 32.5°, 31.4°, and 46.6° for SrTiO₃, Sr₂TiO₄, and Sr₃Ti₂O₇ respectively. To obtain information about the morphology of the sample, transmission electron microscopy studies were carried out on a JEOL, JEM-2100 Transmission Electron Microscope (TEM), used at an operating voltage of 200 kV. The samples for the TEM study were prepared by dispersing the powder samples in ethanol and drop-casting them on a carbon-coated copper grid. Raman study was carried out on a WI Tec's Raman microscope, α 300 R. The studies were carried out on powder samples taken on a glass slide. Diffuse Reflectance Spectra were collected on a UV-visible spectrophotometer, Shimadzu UV-2600, in a wavelength range of 200–800 nm with barium sulfate as the reference material. Kubelka–Munk's (K–M) equation was used to calculate the bandgap of the materials from the reflectance spectra. XPS (X-Ray Photoelectron Spectroscopy) studies were carried out on a Thermo Scientific's K-alpha X-ray Photoelectron Spectrometer (XPS) System with the following settings: 0.05 eV step, 1 s time per step, and 5 cycles, Source Al k-alpha-1486 eV. The surface area of the samples was determined using nitrogen adsorption–desorption isotherms with Quanta Chrome Model Q2. Photoluminescence and time-resolved photoluminescence spectra were obtained from Horiba's TCSPC (Time Correlated Single Photon Counting) at an excitation wavelength of 380 nm and 340 nm respectively. Time-resolved photoluminescence decay curves were fitted by using a double logarithmic decay equation which is given below:

y = A₁e^(−x/τ₁) + A₂e^(−x/τ₂) + y₀ (1)

The average lifetime was obtained from the equation given below:

2.2.1 Photocatalytic hydrogen evolution reaction (HER) and photoelectrochemical studies. A top irradiation quartz reactor of capacity 140 mL was used for the photocatalytic reaction for the hydrogen evolution. In a typical experiment, 40 mg of the catalyst were dispersed in a 40 mL aqueous solution containing 0.35 M Na₂SO₃ and 0.25 M Na₂S through ultrasonication. The study was carried out in the absence of any metal co-catalyst to observe the direct influence of the nature of the crystal structure on the photocatalytic performance of Sr–Ti–O based oxides. The solution was purged with N₂ gas for 30 minutes to eliminate the dissolved oxygen. A 450 W Xe lamp was used as a source of light. The photon flux of the Xe lamp was obtained from Ray virtual radiation actinometer, Newport, Model 91150V. The gas components from the reactor were analyzed using gas chromatography (GC, PerkinElmer Clarus 680) with a thermal conductivity detector at an interval of 1 h. The apparent quantum efficiency was calculated using the equation given below:where n is the amount of hydrogen evolved; N_A is Avogadro's constant (6.022 × 10²³ mol⁻¹); h is Planck's constant (6.63 × 10⁻³⁴ J s); c is the speed of the light (3 × 10¹⁰ cm s⁻¹); P is the power density of the incident light (181 × 10⁻³ W cm⁻²); S is the irradiation area (12.6 cm²); λ is the representative wavelength of the incident light (using the radiation spectrum of the Xenon lamp, 390 nm (390 × 10⁻⁷ cm)); and t is the time duration of the incident light (18000 s).

Photoelectrochemical studies were performed on a PGSTAT-30 (Autolab) electrochemical workstation using a standard three-electrode system consisting of Ag/AgCl (3 M KCl) as the reference electrode, platinum wire as the counter electrode, and Sr–Ti–O samples deposited on a glass substrate coated with fluorine-doped tin oxide (FTO) as the working electrode. For the synthesis of the working electrode, the catalyst was drop cast on a glass substrate coated with FTO with an area of 1 cm × 1 cm. For the preparation of the catalyst ink, 10 mg of the catalyst was dispersed in 200 μL of isopropyl alcohol containing 10 μL of Nafion resin solution through the ultra-sonication method. Here, 0.1 M Na₂SO₄ (pH = 7) was taken as an electrolyte, and saturated with argon for 30 minutes to remove dissolved oxygen. A 350 W Xe lamp was used as the source of irradiation. LSV (linear sweep voltammetry) curves were obtained in the range from 0 to −1 V vs. Ag/AgCl at a scan rate of 10 mV s⁻¹ under the light. Electrochemical impedance spectroscopy (EIS) was carried out at −0.35 V vs. Ag/AgCl in the frequency range from 0.1 Hz to 100 kHz under light.

3. Results and discussion

The powder X-ray diffraction (PXRD) pattern of the as-prepared nanostructured oxides is shown in Fig. 2. No impurity peaks were observed in the PXRD pattern indicating the formation of a single phase. The pattern for the perovskite SrTiO₃ was indexed based on cubic symmetry (PDF 00-035-0734, space group, Pm3̄m). The other two Ruddlesden–Popper layered structures, Sr₂TiO₄ (n = 1) and Sr₃Ti₂O₇ (n = 2) were indexed based on tetragonal symmetry with the space group, I4/mmm (JCPDF 00-039-1471 and JCPDF 01-078-2479 respectively). The crystallite sizes for all three oxides viz. SrTiO₃, Sr₂TiO₄, and Sr₃Ti₂O₇ were found to be 63 nm, 39 nm, and 42 nm respectively.

Fig. 2 PXRD pattern for SrTiO₃, Sr₂TiO₄, and Sr₃Ti₂O₇.

Raman spectra of the three nanostructured oxides i.e. SrTiO₃, Sr₂TiO₄, and Sr₃Ti₂O₇ are shown in Fig. 3. The Raman spectra obtained for the perovskite SrTiO₃ matched well with the previous reports.^29,30 The spectra consist of a low-frequency band present at 77 cm⁻¹ which was assigned to doubly degenerate modes, E_g. Second order Raman bands were also observed between 200–400 cm⁻¹ and 600–800 cm⁻¹. No first-order bands were observed in the Raman spectra, as expected for the cubic structure. According to the report by Nilsen et al.²⁹ second-order Raman scattering is due to the creation and destruction of two phonons which can originate from anywhere in the Brillouin zone. The authors observed a second-order band or overtone for SrTiO₃ at 369 cm⁻¹ which was attributed to the combination of various bands including the TO₄–TA, TO₄–TO_l, and 2TO₂ bands whereas the band at 684 cm⁻¹ was assigned to the 2TO₃ overtone. It has been previously reported³¹ that there are four Raman active modes, A_1g, and E_g, observed for layered Ruddlesden–Popper oxides with n = 1, Sr₂TiO₄. The Raman bands present at 121, 203, 182, and 571 cm⁻¹ were assigned to the E_g, A_1g, E_g, and A_1g modes respectively (Fig. 3). The broader band observed between 400–450 cm⁻¹ and around 700 cm⁻¹ indicates the presence of a second-order band or defect-induced excitations which may originate at the oxygen sublattice. The Raman spectra of Sr₃Ti₂O₇ (Fig. 3) show bands at 178, 198, 274, 500, and 633 cm⁻¹ which can be assigned to the A_1g, E_g, E_g, A_1g, and A_1g modes respectively which is consistent with a previous report.³² A defect-induced excitation was also observed at 92 cm⁻¹. The band corresponding to the A_1g mode was observed to be more intense in both Sr₂TiO₄ and Sr₃Ti₂O₇, which could be due to symmetric stretching of the oxygen lattice.³³

Fig. 3 Raman spectra of SrTiO₃, Sr₂TiO₄, and Sr₃Ti₂O₇.

The high resolution XPS spectra of strontium (Sr(3d), Fig. S1a–c, ESI†), oxygen (O(1s), Fig. S2a–c, ESI†) and titanium, (Ti(2p), Fig. S3a–c, ESI†), was obtained for SrTiO₃ Sr₂TiO₄ and Sr₃Ti₂O₇. Peaks at binding energies of 133.7 eV (135.5 eV), 133.9 eV (135.5 eV) and 133.6 eV (135.4 eV) were observed in the Sr 3d_5/2 (3d_3/2) spectra of SrTiO₃ (Fig. S1a, ESI†), Sr₂TiO₄ (Fig. S1b, ESI†) and Sr₃Ti₂O₇ (Fig. S1c, ESI†) respectively. Fig. S2a–c (ESI†) show the high-resolution O 1s spectra for SrTiO₃, Sr₂TiO₄, and Sr₃Ti₂O₇. The peaks were fitted with two Gaussian peaks with binding energies at 529.7 and 531.2 eV for SrTiO₃, 529.3 and 531.4 eV for Sr₂TiO₄ and 529.8 and 531.77 eV for Sr₃Ti₂O₇. The peak at lower energy can be attributed to the metal–oxygen bond i.e. the presence of O²⁻ ions in the crystal structure whereas the peak at higher energy can be related to the oxygen vacancies. Fig. S3a–c (ESI†) shows the high-resolution spectra of Ti(2p) for SrTiO₃, Sr₂TiO₄, and Sr₃Ti₂O₇ respectively. The presence of Ti³⁺ was observed along with the Ti⁴⁺ ion after fitting of the peaks. The peaks centered at 458.3 eV (464.2 eV), 458.1 eV (453.8 eV) and 458.5 eV (464.1 eV) were observed for the Ti⁴⁺ 2p_3/2 (2p_1/2) spectra of SrTiO₃, Sr₂TiO₄ and Sr₃Ti₂O₇ respectively. Ti³⁺ 2p_3/2 (2p_1/2) spectra of SrTiO₃, Sr₂TiO₄ and Sr₃Ti₂O₇ were obtained at 456.4 eV (462.8 eV), 456.8 eV (462.1 eV) and 456.6 eV (461.9 eV) respectively. Thus the presence of peaks corresponding to Ti³⁺ also confirms the presence of an oxygen vacancy in the lattice. Based on the area under the peak, the order for the ratio of oxygen vacancies : M–O and Ti³⁺:Ti⁴⁺ was found to be Sr₂TiO₄ > Sr₃Ti₂O₇ > SrTiO₃.

TEM studies for SrTiO₃ showed nanocubes with a size of 80–100 nm (Fig. 4a). The HRTEM image (Fig. 4b) shows lattice fringes with a spacing of 0.281 nm, corresponding to the (110) plane. A rectangular sheet-like morphology was observed for both Ruddlesden–Popper based oxides i.e. Sr₂TiO₄ (Fig. 4c) and Sr₃Ti₂O₇ (Fig. 4e). The size along one dimension of these sheets was observed to be ∼250 nm and ∼350 nm for Sr₂TiO₄ and Sr₃Ti₂O₇ respectively. The lattice fringes corresponding to a spacing of 0.276 nm for Sr₂TiO₄ (Fig. 4d) and 0.273 nm for Sr₃Ti₂O₇ (Fig. 4f) were observed which corresponded to the (110) plane. The BET surface area of all the samples viz. SrTiO₃, Sr₂TiO₄, and Sr₃Ti₂O₇ were observed to be 15 m² g⁻¹, 6.2 m² g⁻¹, and 14 m² g⁻¹ respectively.

Fig. 4 TEM images of (a) SrTiO₃, (c) Sr₂TiO₄, and (e) Sr₃Ti₂O₇. HRTEM images of (b) SrTiO₃, (d) Sr₂TiO₄, and (f) Sr₃Ti₂O₇. The inset shows the corresponding reduced FFT of the HRTEM image.

The bandgap of the oxides was calculated using the Tauc equation (eqn (4)).

where E_g is the bandgap of the semiconductor materials; h is Planck's constant, υ is the frequency of the light, α is the absorption coefficient; n represents the type of transition (n = ½ stands for direct transition whereas n = 2 is used for indirect transition). It is known that the oxides viz. SrTiO₃, Sr₂TiO₄ and Sr₃Ti₂O₇ possess an indirect bandgap.³⁴ The indirect bandgap of the nanostructured SrTiO₃, Sr₂TiO₄ and Sr₃Ti₂O₇ was found to be 3.26 eV, 3.16 eV, and 3.11 eV respectively (Fig. 5a). A small change in the bandgap was observed with the change in the value of ‘n’ in SrO–(SrTiO₃)_n, which was consistent with the report by Chen et al.³⁵ wherein the authors also observed, through theoretical calculations, that the bandgap is affected by a small value with a slight change in the value of ‘n’ (n concerning AO(ABO₃)_n) in Mn-based Ruddlesden Popper-based perovskites. To calculate the band energy positions, XPS valence band spectra were recorded for SrTiO₃, Sr₂TiO₄, and Sr₃Ti₂O₇ (Fig. S4a–c, ESI†). The position of the valence band maxima was observed to be 2.01, 1.47, and 1.16 eV for SrTiO₃, Sr₂TiO₄, and Sr₃Ti₂O₇ respectively. From the value of the band gap obtained from the Tauc plot and the valence band maxima obtained from the valence band spectra using XPS, the conduction band minima for SrTiO₃, Sr₂TiO₄, and Sr₃Ti₂O₇ were found to be −1.25, −1.69, and −1.95 eV (Fig. 5b). The conduction band positions for Sr₂TiO₄ and Sr₃Ti₂O₇ were observed to be more negative than SrTiO₃ indicating that the Ruddlesden–Popper phases are likely to exhibit superior catalytic activity towards hydrogen evolution compared with SrTiO₃.³⁶

Fig. 5 (a) Tauc plot and (b) schematic showcasing the energy diagram for SrTiO₃, Sr₂TiO₄, and Sr₃Ti₂O₇.

3.1 Photocatalytic H₂ evolution

The photocatalytic hydrogen evolution of nanostructured SrTiO₃ and SrO(SrTiO₃)_n (n = 1 and 2) was studied by observing the amount of hydrogen evolved from an aqueous solution containing 0.25 M Na₂S and 0.35 M Na₂SO₃ (used as the hole-scavenger). The study was carried out in the absence of any metal co-catalyst to see the direct influence of the nature of the crystal structure on the photocatalytic performance of Sr–Ti–O based perovskites and Ruddlesden–Popper based oxides. At first, the experiments were carried out under dark conditions. No H₂ gas was observed in the absence of light. Temporal hydrogen evolution under light is shown in Fig. 6a and b. It was observed that the amount of H₂ evolved follows the order Sr₃Ti₂O₇ > Sr₂TiO₄ > SrTiO₃. Apparent quantum yield was calculated using eqn (3). The quantum yield obtained has been tabulated in Table 1. The recyclability test was performed on Sr₃Ti₂O₇ (exhibiting the evolution of the highest amount of hydrogen amongst the three oxides) to check its reusability. The recyclability test was examined for 12 h and each experimental cycle was carried out for 4 h (Fig. 6c). The oxide was centrifuged and washed with water between each cycle. A slight decrease in the rate of H₂ evolution was observed for three cycles. The stability of the catalyst was checked after the recyclability test of 12 h through PXRD (Fig. S5, ESI†). It was observed that the crystal structure of the catalyst remained unchanged after reusability. No additional peaks corresponding to any impurity were observed suggesting that the oxide was stable after 12 h.

Fig. 6 (a) Plot showing the amount of hydrogen evolved per gram of the nanostructured catalyst (SrTiO₃, Sr₂TiO₄, and Sr₃Ti₂O₇). (b) The rate of hydrogen evolution per gram of the catalyst using nanostructures of SrTiO₃, Sr₂TiO₄, and Sr₃Ti₂O₇ as photocatalysts and (c) the recyclability test of Sr₃Ti₂O₇ for three consecutive cycles.

Table 1 Apparent quantum yield for hydrogen evolution using SrTiO₃, Sr₂TiO₄, and Sr₃Ti₂O₇ nanostructures as photocatalysts

Sample	n (amount of hydrogen gas evolved after 5 h of the reaction) (mol)	Apparent quantum yield (AQY) (%)
SrTiO3	12.23 × 10^−6	0.018
Sr2TiO4 (SrO(SrTiO3))	22.6 × 10^−6	0.03
Sr3Ti2O7 (SrO(SrTiO3)2)	36.5 × 10^−6	0.058

---

To ascertain the reason for the observed trend for the photocatalytic hydrogen evolution, photoluminescence (PL) and time-resolved photoluminescence studies were carried out for the three oxides. Fig. 7 shows the photoluminescence spectra for all the samples. Here, the samples were excited at a wavelength of 380 nm, and the emission spectra were recorded in the range of 395–700 nm. All the samples showed PL emission in the violet-blue region with a peak centered at 414 nm and 436 nm. It has been reported³⁷ that the photoluminescence for SrTiO₃ based oxides arises mostly due to a recombination of the electrons and holes that are trapped in the intermediate states (present within the bandgap). These intermediate states arise as a result of distortion, oxygen vacancies, etc. It has also been reported³⁷ that the emission in the violet-blue region occurs due to the presence of a surface or shallow defects which may arise due to oxygen vacancies. Oxygen vacancies are also known to affect the catalytic behavior of oxides.^5,38 The presence of a controlled concentration of defects (oxygen vacancies) is known to increase the photocatalytic efficiency of SrTiO₃ towards hydrogen evolution.³⁹ The oxygen vacancies act as electron donors which either result in an increased charge transport or a shift in the Fermi level towards the conduction band.³⁹ Such a phenomenon is likely to improve the charge separation behavior of the oxide. In our studies, the presence of defect/oxygen vacancies was also confirmed by Raman and XPS studies. In addition to the presence of defects, it was also observed that the PL emission, corresponding to the defect, decreased in the following order SrTiO₃ > Sr₂TiO₄ > Sr₃Ti₂O₇. With the same excitation energy and no significant changes in the optical bandgap of the oxides, the decrease in the PL emission can be related to a decrease in the recombination of electron and hole pairs giving rise to radiative emission. To further investigate the lifetime of photo-induced charge carriers, time-resolved photoluminescence decay spectra were recorded. The data was fitted using eqn (1) and the average lifetime (τ_avg) was calculated using eqn (2). Fig. 8a–c show the second exponential decay fit of the time-resolved PL of the three oxides. The parameters obtained after fitting are listed in Table 2. The average lifetime was found to follow the order Sr₃Ti₂O₇ > Sr₂TiO₄ > SrTiO₃, for the Sr–Ti–O system. From the values obtained for τ_avg (ns), it can be concluded that the recombination of photo-induced charge carriers is delayed for Ruddlesden–Popper based oxides (Sr₂TiO₄ and Sr₃Ti₂O₇) compared with SrTiO₃. The longer lifetime implies that a large number of photo-induced electrons could reach the surface of the catalysts which would be captured by the H⁺ ions. Thus, the longer lifetime of the photo-induced charge carriers for Sr₃Ti₂O₇ further supports the increased photocatalytic properties of Sr₃Ti₂O₇. The presence of interlayers (SrO layer in our study) results in a reduction of the recombination of photogenerated charge carriers by separation of the electrons and holes thereby enhancing the photocatalytic water splitting reaction.⁴⁰

Fig. 7 Photoluminescence spectra of SrTiO₃, Sr₂TiO₄, and Sr₃Ti₂O₇.

Fig. 8 Time-resolved photoluminescence decay of (a) SrTiO₃, (b) Sr₂TiO₄ and (c) Sr₃Ti₂O₇.

Table 2 Parameters obtained after exponential fitting of the decay curves for SrTiO₃, Sr₂TiO₄, and Sr₃Ti₂O₇

Sample	A 1	τ 1 (ns)	A 2	τ 2 (ns)	Average lifetime τavg (ns)
SrTiO3	1800	37.7 ± 1.0	213	351 ± 8.6	202
Sr2TiO4	1646	38.0 ± 1.2	198	367 ± 8.8	215
Sr3Ti2O7	1619	40.3 ± 1.0	194	392 ± 10	230

---

To further evaluate the charge transfer behavior of the three oxides in the presence of light, photoelectrochemical studies were carried out. The photocurrent density (Fig. 9a) was found to be the highest for Sr₃Ti₂O₇. The onset potential was observed to be ∼−0.40 V vs. Ag/AgCl, which is nearly the same for all three samples. The overpotential for the hydrogen evolution reaction (HER) for the three oxides was evaluated from the current–voltage curve at −0.1 mA cm⁻²vs. Ag/AgCl (Fig. 9b). The overpotential was found to follow the order SrTiO₃ > Sr₂TiO₄ > Sr₃Ti₂O₇ for the Sr–Ti–O system. Fig. 9c shows Nyquist plots for the three oxides. Based on R_ct (charge transfer resistance), R_s (solution resistance), and a constant phase element with impedance, which is related to the angular frequency of the applied potential, ω (using eqn (5)),

where Z₀ is a constant, and the value of n ranges as follows: 0 < n < 1, , the curve was fit to an equivalent circuit, shown in the inset of Fig. 9c. The parameters obtained after the fitting are tabulated in Table 3. The charge transfer resistance originating as a result of the ionic and electronic resistance across the electrode–electrolyte interface, R_ct, were observed to follow the order: Sr₃Ti₂O₇ < Sr₂TiO₄ < SrTiO₃. Mott–Schottky studies in the presence of light (Fig. 9d) were carried out to calculate the charge carrier density using the following equation:where C is the interfacial capacitance, ε is the dielectric constant of the material (250, 38, and 50 for SrTiO₃ Sr₂TiO₄ and Sr₃Ti₂O₇ respectively),⁴¹ε₀ is the permittivity of vacuum (8.854 × 10⁻¹² F m⁻¹), e is the fundamental charge (1.602 × 10⁻¹⁹ C), A is the area of the electrode (1 cm²), N_d is the charge carrier density, V is the applied potential, k_B is Boltzmann's constant, and T is the temperature. The positive slope of the plot suggests the n-type semiconducting behavior of the oxides. The n-type behavior in SrTiO₃ arises due to the presence of oxygen vacancies.^42,43 The charge carrier density N_d was calculated from the slope using eqn (7).The slope was observed to be 1.07 × 10¹⁰ F⁻² cm⁴ V⁻¹, 0.88 × 10¹⁰ F⁻² cm⁴ V⁻¹ and 0.87 × 10¹⁰ F⁻² cm⁴ V⁻¹ for SrTiO₃ Sr₂TiO₄ and Sr₃Ti₂O₇ respectively. Thus, N_d was calculated to be 5.3 × 10¹⁹ cm⁻³, 4.2 × 10²⁰ cm⁻³ and 3.2 × 10²⁰ cm⁻³ for SrTiO₃ Sr₂TiO₄ and Sr₃Ti₂O₇ respectively. The charge carrier density was found to be in the following order: Sr₂TiO₄ ≈ Sr₃Ti₂O₇ > SrTiO₃ consistent with the photocatalytic performance of the oxide.

Fig. 9 (a) Photocurrent density, (b) overpotential, (c) Nyquist plot and (d) Mott–Schottky plot for SrTiO₃, Sr₂TiO₄, and Sr₃Ti₂O₇.

Table 3 Parameters obtained after fitting the Nyquist plot

Sample	R ct (kΩ)	R s (Ω)	CPE (μmho)	n (in constant phase element)
SrTiO3	20.6	36.5	16.0	0.940
Sr2TiO4	14.5	28.4	12.5	0.973
Sr3Ti2O7	12.3	26.5	12.1	0.974

---

Thus, from the trend observed for the photocatalytic hydrogen evolution there are a few interesting observations. One, the photocatalytic hydrogen evolution increases with the introduction of the SrO layer i.e. when comparing SrTiO₃ with Sr₂TiO₄ (SrO(SrTiO₃)). The SrO layer in Ruddlesden–Popper based layered perovskites⁴⁴ is responsible for the dissociation of water. The oxygen site (apical oxygen) in SrTiO₃ of Ruddlesden–Popper based layered perovskites favors the adsorption of hydrogen enabling its recombination with other adsorbed hydrogen to form H₂. Wei et al.⁴⁵ also observed that insertion of the SrO layer in SrTaO₂N to form Sr₂TaO₃N (a Ruddlesden–Popper based oxynitride) resulting in the improved photocatalytic performance of the oxynitride. Second, with the increase in the SrTiO₃ perovskite unit in Ruddlesden–Popper based layered perovskites i.e. on comparing Sr₂TiO₄ and Sr₃Ti₂O₇, the photocatalytic activity was found to increase for Sr₃Ti₂O₇. For photocatalytic water splitting, the active sites in Ruddlesden–Popper based layered perovskites are the B-site cations (Ti in Sr–Ti–O based systems), AO layers (SrO layer in Sr–Ti–O based systems), and defects arising due to oxygen vacancies. This arrangement of layers is known to suppress the charge carrier recombination, as evident from time-resolved photoluminescence studies, and promote charge carrier transfer, as evident from a decrease in charge transfer resistance.

Based on the study, it is also to be noted there are two important synergistic factors that affect the photocatalytic behavior of the oxides: (a) crystal structure and the arrangement of atoms, and (b) the presence of defects (oxygen vacancies). Thus, synergism in the role of the SrO layer and the SrTiO₃ perovskite unit along with the layered morphology, low crystallite size, and presence of defects (oxygen vacancies), is presumed to have resulted in an improved photocatalytic performance for Ruddlesden–Popper based layered perovskite. This improved performance was evident from the reduced overpotential, low charge transfer resistance, and high charge carrier density for Ruddlesden–Popper based layered perovskites. Hence, the appropriate choice of the crystal structure from a series could result in attaining the desired factors required for designing an efficient catalyst for photocatalytic HER.

4. Conclusions

We have successfully synthesized three members of the Sr–Ti–O system viz. SrTiO₃, Sr₂TiO₄, and Sr₃Ti₂O₇ using the combination of a polymeric citrate precursor and a hydrothermal method. SrTiO₃ was observed to crystallize into a cubic structure and possessed cube-like morphology. Sr₂TiO₄ and Sr₃Ti₂O₇ were found to crystallize in a tetragonal crystal structure. Sheet-like morphology was observed for these two oxides. We have demonstrated photocatalytic hydrogen evolution activity for these three nanostructured oxides. All three structures were found to be active photocatalysts under a full range of light irradiation and in the absence of any co-catalyst. We observed that the nanocubes of SrTiO₃ (perovskite cubic crystal structure) were less active for the photocatalytic HER than layered Sr₂TiO₄ and Sr₃Ti₂O₇ (tetragonal, Ruddlesden–popper structure). Synergism of many factors which included the presence of a SrO layer, SrTiO₃ unit, layered morphology, low crystallite size, and defects is attributed to the observed photocatalytic performance of the Ruddlesden–popper structure in comparison to cubic perovskite SrTiO₃. Thus, the findings in our study highlight the influence of the Ruddlesden–Popper crystal structure on the hydrogen evolution efficiency of perovskite oxides. The study opens a strategic approach based on choosing the crystal structure while designing highly efficient catalysts.

Conflicts of interest

There are no conflicts of interest to declare.

Acknowledgements

A. V. thanks INST, Mohali for the fellowship. SV thanks CSIR (01(2943)/18-EMR-II), Govt. of India for funding.

References

1. K. Maeda, K. Teramura, D. Lu, T. Takata, N. Saito, Y. Inoue and K. Domen, Nature, 2006, 440, 295.
2. H. Tong, S. Ouyang, Y. Bi, N. Umezawa, M. Oshikiri and J. Ye, Adv. Mater., 2012, 24, 229–251.
3. Y.-F. Li, Z.-P. Liu, L. Liu and W. Gao, J. Am. Chem. Soc., 2010, 132, 13008–13015.
4. S. Ouyang and J. Ye, J. Am. Chem. Soc., 2011, 133, 7757–7763.
5. A. Kumar and V. Krishnan, Adv. Funct. Mater., 2021, 31, 2009807.
6. N. M. Flores, U. Pal, R. Galeazzi and A. Sandoval, RSC Adv., 2014, 4, 41099–41110.
7. L. Zhang, T. Xu, X. Zhao and Y. Zhu, Appl. Catal., B, 2010, 98, 138–146.
8. Y. Liu, B. Huang, Y. Dai, X. Zhang, X. Qin, M. Jiang and M.-H. Whangbo, Catal. Commun., 2009, 11, 210–213.
9. S. Ouyang, N. Kikugawa, D. Chen, Z. Zou and J. Ye, J. Phys. Chem. C, 2009, 113, 1560–1566.
10. T. A. Kandiel, A. Feldhoff, L. Robben, R. Dillert and D. W. Bahnemann, Chem. Mater., 2010, 22, 2050–2060.
11. T. Li, H. Cai, C. Li, X. Liu and F. Huang, Front. Chem., 2020, 8, 351.
12. S. Chen, D. Huang, P. Xu, W. Xue, L. Lei, M. Cheng, R. Wang, X. Liu and R. Deng, J. Mater. Chem. A, 2020, 8, 2286–2322.
13. J. Cai, A. Cao, Z. Wang, S. Lu, Z. Jiang, X.-Y. Dong, X. Li and S.-Q. Zang, J. Mater. Chem. A, 2021, 9, 13890–13897.
14. S. Kumar, A. Kumar, A. Kumar and V. Krishnan, Catal. Rev., 2020, 62, 346–405.
15. H. Mai, D. Chen, Y. Tachibana, H. Suzuki, R. Abe and R. A. Caruso, Chem. Soc. Rev., 2021, 50, 13692–13729.
16. G. Zhang, G. Liu, L. Wang and J. T. S. Irvine, Chem. Soc. Rev., 2016, 45, 5951–5984.
17. A. Kumar, A. Kumar and V. Krishnan, ACS Catal., 2020, 10, 10253–10315.
18. T. Kimijima, K. Kanie, M. Nakaya and A. Muramatsu, Appl. Catal., B, 2014, 144, 462–467.
19. P.-L. Hsieh, G. Naresh, Y.-S. Huang, C.-W. Tsao, Y.-J. Hsu, L.-J. Chen and M. H. Huang, J. Phys. Chem. C, 2019, 123, 13664–13671.
20. A. Vijay and S. Vaidya, ACS Appl. Nano Mater., 2021, 4, 3406–3415.
21. T. Oshima, T. Yokoi, M. Eguchi and K. Maeda, Dalton Trans., 2017, 46, 10594–10601.
22. K.-i Shimizu, Y. Tsuji, T. Hatamachi, K. Toda, T. Kodama, M. Sato and Y. Kitayama, Phys. Chem. Chem. Phys., 2004, 6, 1064–1069.
23. X. Sun, Y. Xie, F. Wu, H. Chen, M. Lv, S. Ni, G. Liu and X. Xu, Inorg. Chem., 2015, 54, 7445–7453.
24. H. Zhang, S. Ni, Y. Mi and X. Xu, J. Catal., 2018, 359, 112–121.
25. K. Yamada, H. Suzuki, R. Abe and A. Saeki, J. Phys. Chem. Lett., 2019, 10, 1986–1991.
26. H. Jeong, T. Kim, D. Kim and K. Kim, Int. J. Hydrogen Energy, 2006, 31, 1142–1146.
27. X. Sun, Y. Xie, F. Wu, H. Chen, M. Lv, S. Ni, G. Liu and X. Xu, Inorg. Chem., 2015, 54, 7445–7453.
28. Y. Liu, L. Xie, Y. Li, R. Yang, J. Qu, Y. Li and X. Li, J. Power Sources, 2008, 183, 701–707.
29. W. G. Nilsen and J. G. Skinner, J. Chem. Phys., 1968, 48, 2240–2248.
30. S. Banerjee, D.-I. Kim, R. D. Robinson, I. P. Herman, Y. Mao and S. S. Wong, Appl. Phys. Lett., 2006, 89, 223130.
31. R. Viennois, P. Hermet, D. Machon, M. M. Koza, D. Bourgogne, B. Fraisse, A. P. Petrović and D. Maurin, J. Phys. Chem. C, 2020, 124, 27882–27893.
32. A. Dias, J. I. Viegas and R. L. Moreira, J. Alloys Compd., 2017, 725, 77–83.
33. S. Kamba, P. Samoukhina, F. Kadlec, J. Pokorný, J. Petzelt, I. M. Reaney and P. L. Wise, J. Eur. Ceram. Soc., 2003, 23, 2639–2645.
34. C.-H. Lee, N. J. Podraza, Y. Zhu, R. F. Berger, S. Shen, M. Sestak, R. W. Collins, L. F. Kourkoutis, J. A. Mundy, H. Wang, Q. Mao, X. Xi, L. J. Brillson, J. B. Neaton, D. A. Muller and D. G. Schlom, Appl. Phys. Lett., 2013, 102, 122901.
35. A. Chen, X. Zhang, Z. Zhang, S. Yao and Z. Zhou, J. Mater. Chem. A, 2019, 7, 11530–11536.
36. H. Chen and X. Xu, Appl. Catal., B, 2017, 206, 35–43.
37. A. E. Souza, G. T. A. Santos, B. C. Barra, W. D. Macedo, S. R. Teixeira, C. M. Santos, A. M. O. R. Senos, L. Amaral and E. Longo, Cryst. Growth Des., 2012, 12, 5671–5679.
38. T. Zheng, W. Sang, Z. He, Q. Wei, B. Chen, H. Li, C. Cao, R. Huang, X. Yan, B. Pan, S. Zhou and J. Zeng, Nano Lett., 2017, 17, 7968–7973.
39. H. Tan, Z. Zhao, W.-B. Zhu, E. N. Coker, B. Li, M. Zheng, W. Yu, H. Fan and Z. Sun, ACS Appl. Mater. Interfaces, 2014, 6, 19184–19190.
40. Z. Liang, K. Tang, Q. Shao, G. Li, S. Zeng and H. Zheng, J. Solid State Chem., 2008, 181, 964–970.
41. W. Kwestroo and H. A. M. Paping, J. Am. Ceram. Soc., 1959, 42, 292–299.
42. P. Calvani, M. Capizzi, F. Donato, S. Lupi, P. Maselli and D. Peschiaroli, Phys. Rev. B: Condens. Matter Mater. Phys., 1993, 47, 8917–8922.
43. O. N. Tufte and P. W. Chapman, Phys. Rev., 1967, 155, 796–802.
44. Y. Zhu, H. A. Tahini, Z. Hu, J. Dai, Y. Chen, H. Sun, W. Zhou, M. Liu, S. C. Smith, H. Wang and Z. Shao, Nat. Commun., 2019, 10, 149.
45. S. Wei and X. Xu, Appl. Catal., B, 2018, 228, 10–18.

---

Footnote

† Electronic supplementary information (ESI) available. See DOI: https://doi.org/10.1039/d2ma00097k

---