Abstract

Spontaneous solar-driven water splitting to generate H2 with no pollution discharge is an ideal H2 generation approach. However, its efficiency remains far from real application owing to the poor lightharvesting and ultrafast charge recombination of photocatalysts. To address these issues, herein, we employed a novel but simple chemical bath deposition (CBD) method to construct CdS/SrTiO3 nanodots-on- nanocubes at room temperature (ca. 25
C). The as-synthesized nanohybrids not only expand light absorption from ultraviolet (UV) to visible light but also significantly retard charge recombination owing to the well-defined heterostructure formation. As a result, the CdS/SrTiO3 exhibits high photocatalytic performance with H2 evolution rate of 1322 lmol g
1 h
1 , which is 2.8 and 12.2 times higher than that of pristine CdS and SrTiO3, respectively. This work provides a universal approach for the heterostructure construction, and inspired by this, higher efficient photocatalysts for H2 evolution may be developed in the near future.

1. Introduction

The crisis of energy and environmental pollution significantly curbs society progress. Solar-driven water splitting to generate clear H2 energy, as one of the solar energy utilization approaches [1–10], has the potential to address those issues. The past few decades have witnessed unabated growth of research on this project, and various materials have been developed as photocatalysts, including oxide, nitride and chalcogenide [11–14]. Among them, SrTiO3 is a good candidate owing to its nontoxic, low-cost, chemical and optical stability, and proper conduction band edge for photocatalytic H2 evolution reaction (HER) [15–19]. However, the low efficiency, stemming from poor light-harvesting and ultrafast charge recombination, still impedes its application. As a typical wide-band-gap semiconductor, the light absorption threshold of SrTiO3 is limited into the near-UV region. However, the UV light accounts for only 4% of solar energy, and thus renders most of the sun’s irradiation energy being wasted. To enhance the light utilization, doping was widely employed, and numerous ions (e. g. Rh, Ru, Ir, Cr and La) were employed [20–23]. Doping did widen light response scope to some extent, but it usually introduces new recombination centers [20], which significantly counteracts its positive effect. Besides, optimizing doping content usually needs extensive experiments which are time-consuming and energyintensive.

Sensitization, another strategy to enhance light utilization, has been widely applied in photocatalysis of wide-band-gap semiconductors [24–27]. Organic dyes, as sensitizers, have been extensively studied [25,26]. But its low-stability and noble metal containment severely hinders its application in mass. Another type of sensitization materials of relatively narrow-band-gap semiconductors, such as Cu2O, CdS and MoS2, have been widely employed [28–30]. Among them, CdS has attracted intensive interests due to its optimal band gap (ca. 2.4 eV) for efficient light utilization, suitable band position for the HER, low expensive and relatively simple synthesis route. Various CdS based sensitization-type photocatalysts, such as CdS/TiO2, CdS/ZnS, CdS/ZnO [27–29,31,32], have been studied. But sensitizing SrTiO3 has received less attention and further boosting the photocatalytic performance is still pursued [33,34]. In addition, the semiconductor-sensitization needs the intimate interfaces formation between sensitizers and sensitized semiconductors to favor charge separation. In view of the previous reports, approaches of hydrothermal, calcination were often used to fabricate sensitization-type photocatalysts [27,28,31,33]. But there still exists potential to reduce time and energy consumption.

Herein, we first develop the CBD method at room temperature to synthesize CdS/SrTiO3 with an intimate interface between CdS and SrTiO3. This novel but simple approach significantly saves time (see Table S1). Besides, the as-synthesized nanohybrids can enhance light utilization to the visible light region, and the welldefined interfaces for the heterostructure benefit the photogenerated charge separation and transportation, which significantly retard charge recombination, and thus more electrons involving in reducing H+ into H2. As a result, the optimized sample of CdS/ SrTiO3 shows high performance for HER with H2 evolution rate of 1322 lmol g
1 h
1 under simulated solar light illumination, which is 2.8 and 12.2 times higher than that of CdS and SrTiO3, respectively.

2. Experimental

2.1. Materials Sr(OH)28H2O (purity = 99%), tetrabutyl titanate [Ti(OBu)4] (purity = 99%), CdCl2 (purity = 99.99%), and CS(NH2)2 (purity = 99%) were purchased from Alfa Aesar chemical co., USA. Triethylene glycol (TEG) (purity = 99%), Polyvinyl pyrrolidone (PVP) (K30), NH4Cl (purity  99.5%), methanol (purity = 99.9%), ethanol (purity = 99.8%), acetic acid (purity = 99%), and NH3H2O (28.0–30.0% NH3 basis) were obtained from Sinopharm Chemical Reagent Co., Ltd. (Beijing, China). All the reagents were used directly without purification. Deionized water with a resistivity of 18.2 MO cm, produced by using a Milli-Q apparatus (Millipore), was used in all the experiments.

2.2. Synthesis of SrTiO3 SrTiO3 nanocubes were synthesized by using a modified method [35]. Namely, 32 mmol Sr(OH)28H2O, 32 mmol Ti(OBu)4, 16 mL NH3H2O and 1.48 g PVP were together added into a flask with 80 mL TEG. After that, the mixture was heated at 160
C for 2 h with strong stirring. Following that, 20 mL acetic acid was added to remove the by-product SrCO3 when the obtained light yellow sol was cooled to room temperature. Further reaction for 30 min, the end-product (white powder) was received after resulting composite being thoroughly washed with water and ethanol, respectively, and dried at 80
C for 6 h in a vacuum oven.

2.3. Preparation of CdS/SrTiO3 0.2 g as-synthesized SrTiO3 was ultrasonically dispersed in water. Following that, 1.5 mmol CdCl2, 3 mmol NH4Cl and 7.5 mmol CS(NH2)2 were together added with strong stirring. After that, NH3H2O was dropwise added to modify the solution pH. Further reaction for a certain time, the suspension was centrifuged and washed with water and ethanol for several times. Then the end-products were obtained after vacuum-oven-drying at 80
C for 6 h. As a control, pure CdS was also synthesized. All the reaction conditions were same with that of CdS/SrTiO3 fabrication except for no SrTiO3 addition and the usage of unpolished reactor which offers a carrier for the CdS nucleation. After reaction for 2 h, the yellow product was gently scraped off from the inside wall of the reactor and thoroughly washed by using water and ethanol. Then the pristine CdS powder was obtained after being dried at 80
C for 6 h in a vacuum oven.

2.4. Evaluation of photocatalytic activities Photocatalytic H2 evolution was performed in a Pyrex glass cell which had a flat, round upside-window with an irradiation area of 38 cm2 for external light incidence. A 300 W Xenon arc lamp with an AM 1.5 cut-off filter (CEL-HXF 300, Beijing China Education Au-light Co., Ltd) was used to simulate the visible light source. 

The illumination intensity was adjusted to 100 mW cm2 . The H2- solar system (Fig. S1) (Beijing China Education Au-light Co., Ltd) with a gas chromatogram (GC), equipped with a thermal conductivity detector (TCD), TDX-01 column and Ar carrier gas, was used to collect and on-line detect evolved H2. 0.02 g of the photocatalyst was suspended in a glass cell with 80 mL of water solution containing 8 mL methanol as hole scavenger. The cell was kept at 5 C by using a circulating water system. Before irradiation, the reaction system was pumped to vacuum. It should be mentioned here, the cycled cold water (marked by red arrow in Fig. S1) can effectively prevent the reaction solution pumped out of the system. The H2 evolution rate was determined by GC. The apparent quantum yield (Ø) was estimated by the following equation: ð Þ¼ % ne np  100 ¼ 2nH2 np  100 ð1Þ np ¼ h hv ¼ I  t  S hv ð2Þ where Ø is the apparent quantum yield, ne is the number of reacted electrons, np is the number of incident photos, nH2 is the number of evolved H2 molecules, h is the total energy of incident photos (J), h is the Planck constant (J s
1 ), m is the frequency of light (Hz), I is the illumination intensity (W m
2 ) determined with a ray virtual radiation actinometer, t is the irradiation time (s), S is the irradiation area (m2 ).

2.5. Characterization Transmission electron microscopy (TEM) images were obtained on a JEM 2100F (JEOL, Japan) operated at 200 kV. Scanning electron microscopy (SEM) images were obtained on a Gemini microscope (Zeiss Ltd., Germany) equipped with an energy dispersive X-ray spectroscopy (EDS). X-ray powder diffraction (XRD) was carried out with a Rigaku D/max-7000 using filtered Cu Ka irradiation. The UV–Visible absorption spectra were recorded with a UV–Visible spectrophotometer (UV-2550, Shimadzu, Japan). The transient photocurrent was measured on a CHI 760 E electrochemical system (Shanghai, China) using Ag-AgCl as reference and Pt as counter electrodes. The working electrode was prepared by dispensing sample suspension in ethanol onto the ITO/glass of fixed area (0.196 cm2 ). The electrolyte is a methanol solution (1.33 M) which was filled in a quartz cell with a side window for external light incidence. Light on and off was controlled by a baffle installed on a stainless steel black box. Inductively coupled plasma atomic emission spectrometry (ICP-AES, ICPE-9000 Shimadzu) was used to measure the Cd content. The zeta-potential was performed by using a Malvern Zetasizer Nano ZSP instrument (Malvern Instruments, Malvern, UK). All the samples were diluted to about 0.05 wt% with certain pH modified by NH3H2O solution, and then put into a disposable folded capillary cell (DTS1060) to measure the zeta potential. During test, the cell was maintained at 25
C. Every datum was received from the average value of three individual measurements.

3. Results and discussion

The morphologies of as-synthesized pristine SrTiO3 and CdS/ SrTiO3 nanohybrids were investigated through TEM. As shown in Fig. 1a, the pristine SrTiO3 is nanocube with an average particle size of 9.0 nm obtained by counting two hundred SrTiO3 nanocubes (see Fig. S2). After the CBD reaction, small nanodots (Fig. 1b) with diameter of ca. 5 nm domains (Fig. 1c) can be observed on the surface of SrTiO3 nanocubes. Further enlarged image (Fig. 1d) of the nanohybrids indicates the lattice fringes of small dots marked by the black arrow, and nanocube having spacing of 0.36 and 0.28 nm should be assigned to the (1 0 0) and (1 1 0) lattice planes of CdS and SrTiO3, respectively. The intimate contact for the interface denoted by orange dashed line manifests the well-defined heterostructure formation, which is beneficial for the photogenerated charge transportation and separation, and in turn photocatalytic performance enhancement.

he crystal structures of the CdS/SrTiO3 nanohybrids synthesized by CBD reaction of 2 h, as well as pristine SrTiO3 nanocubes, were revealed by using XRD analysis. As shown in Fig. 2a, all the peaks appeared for the pure SrTiO3 can be well indexed to the cubic SrTiO3 (JCPDS No. 74-1296), while for the CdS/SrTiO3 nanohybrids, additional weak diffraction peaks appeared at 26.5
, 28.2
, 34.7
, 43.7
and 51.8
(marked with r) can be assigned to the diffraction from the (0 0 2), (1 0 1), (1 0 2), (1 1 0) and (1 1 2) planes of hexagonal CdS (JCPDS No. 65-3414), corroborating the existence of CdS in the nanohybrids. The weak peaks for CdS may stem from its relatively small grain size, as observed from TEM images mentioned above. The TEM and XRD investigation collaboratively verified the formation of the CdS/SrTiO3 heterostructures. Furthermore, the synthesis mechanism was carefully investigated. As we know, affected by solution environment there are two typical kinds of surfaces for the oxide semiconductors, as shown in Eqs.

Study shows that the oxide surface can be converted from positively to negatively charged with the increase of alkalinity [23,36]. Therefore, the charge characteristic of SrTiO3 dependence on pH was first probed by using a Malvern Zetasizer Nano ZSP instrument. Every sample with fixed pH was repeatedly fabricated and tested for three times. Here, the concentration of the SrTiO3 suspending solution was fixed at 1 g/L being similar to that of involving in CBD reaction. The characterized results, as plotted in Fig. 2b, show that charged SrTiO3 nanocubes become more negative with the increase of pH, and exhibit electronegativity when the pH higher than ca. 4.1. But the potential tends to be stable after pH  10. Obviously, the negatively charged SrTiO3 in alkaline solution facilitates the Cd2+ absorption. However, here, the pH of solution for CBD was determined at 10 in view of the gently discharged S2
for CS(NH2)2 being well for controlling the growth rate of CdS in relatively lower pH alkaline solution. The pH of solution for continuous reaction of 4 h has been tested every 30 min. As shown in Table S2, there is no significant change for the solution during CdS/ SrTiO3 synthesis. The slight decrement could be attributed to the consumption and volatilization of NH3H2O. As reaction proceeding, the color of the suspension gently converted from white to deep yellow as shown in Fig. S3. TEM (Fig. 3) and ICP-AES (Table 1) (Every sample was repeatedly synthesized and tested for five times.) measurements of the products synthesized at different time together indicate the content of CdS increases with the increase of reaction time. Likewise, a same conclusion can also be drawn by further characterization of the element mapping by SEM-EDS of hybrid samples with identical scanning time of 3 min. As shown in Fig. S4, the signals of Cd and S elements become stronger with increasing synthesis time from 1 to 4 h. Therefore, the CdS content in heterostructure can be easily controlled by varying reaction time. This approach is novel and simple, and has potential to be applied in other systems.