In situ synthesis of porous ZnO-embedded Zn1 xCdxS CdS heterostructures for enhanced photocatalytic activity
Rong Chen, Kui Li, Xiao-Shu Zhu, Shuai-Lei Xie, Long-Zhang Dong, Shun-Li Li* and Ya-Qian Lan*
The in-situ synthesis of ZnO-embedded Zn1-xCdxS/CdS heterostructure nanocrystals were achieved using a simple surfactant-free hydrothermal route. The morphology and the content of exposed active sites of Zn1-xCdxS/CdS could be facilely modulated by varying the particle size of ZnS precursors, and the heterostructure with the smallest particle size showed the largest photocatalytic activity of 22.99 mmol/h/g in Na2S + Na2SO3 solution. Furthermore, the in-situ embedding of ZnO cocatalyst on the Zn1-xCdxS/CdS heterostructure dramatically improved its photocatalytic activity to 84.17 mmol/h/g, which is 765 times higher than that of CdS prepared by hydrothermal method. More significantly, the ZnO-embedded Zn1-xCdxS/CdS heterostructure gave considerable H2 -production rate even in the absence of hole scavenger, although the bare Zn1-xCdxS/CdS exhibited no hydrogen evolution, and it showed very competitive photocatalytic activity in methanol in comparison with that in Na2S + Na2SO3 solution under the similarly alkaline circumstance. These investigations indicate that the existence of ZnO cocatalyst in the heterostructure could not only effectively improve the photocatalytic activity by suppressing the recombination of charge carriers and acting as photocatalytic reaction sites, but also make it possible to adopt methanol as the sacrificial reagent for sulfide photocatalyst.
Semiconductor nanocrystals have attracted much attention in the areas of photocatalysis,1 and photoelectrochemistry (PEC)2 owing to their excellent photocatalytic and photoelectronic properties.3 However, the limited visible light absorption and rapid recombination of photo-generated charge carriers in semiconductor photocatalysts restrict their practical applications.4 Fabricating the sulfide heterostructure not only facilitates the separation of charge carriers,5 but also improves the visible light absorption due to their narrow band gap.6 However, the preparation of sulfide heterostructures usually involves complex routes by loading one semiconductor on the other one, which needs precise control of synthetic parameters in each step.7 This method not only limits the scale-up synthesis of heterostructure,5 but also induces the formation of interfacial problems.8 Moreover, the surface based characteristic of photocatalytic water splitting reaction makes the content of exposed active sites an important factor in the design of highperformance heterostructure photocatalyst. The organic surfactants were extensively adopted to modulate the morphology of the nanomaterials. Whereas, the surfactants inevitably contaminate the heterostructures, and may substantially impair the performance of the heterostructures.9 Therefore, searching for an efficient surfactant-free hydrothermal method for the in-situ synthesis of sulfide heteorstructure with improved amount of active sites is an important subject from both theoretical and practical viewpoints.
Cocatalysts, such as graphene, Pt, NiO and MoS2 could facilitate the separation of photo-exited holes/electrons pairs, and/or act as H2 -production sites.10 ZnO was widely used as photocatalyst due to its high electron mobility and long photoexited lifetime.11 Notably, it was extensively demonstrated that the oxygen vacancies in ZnO could work as electron traps to accept electrons and inhibit the recombination of charge carriers.12 The oxygen vacancies in ZnO could also act as the favored sites for catalyzing the N-formylation reaction as reported by Zhu’s group.13 Furthermore, it was demonstrated in Zong’s investigation that the ZnO could function as cocatalyst for improving the catalytic activity of combined reforminghygrogenolysis of glycerol.14 These unique features of ZnO make it an excellent cocatalyst for improving the photocatalytic activity of sulfide photocatalysts. On the other hand, it is well known that the sulfide photocatalysts show an excellent photocatalytic activity only in the Na2S + Na2SO3 sacrificial solution, which is toxic and not stable in the air condition.15 Unfortunately, the sulfide photocatalysts give the negligible photocatalytic activity in the widely used hole scavenger, such as methanol.16 If we could fabricate the hetero-nanomaterial composed by ZnO and sulfide heterostructure, we can simultaneously take the advantages of the two components for photocatalytic water splitting.17 This material maybe not only improve the photocatalytic activity of sulfide photocatalysts in the traditional Na2S + Na2SO3 solution, but also make it possible for the utilization of methanol solution as sacrificial reagent for sulfide photocatalysts. Up to date, both the fabrication of fine grained ZnO-embedded sulfide heterostructure and efficient method for improving the photocatalytic activity in methanol sacrificial solution were scarely reported.
In view of the aforementioned ideas, we select Zn1- xCdxS/CdS as the sulfide heterostructure to synthesize ZnOembedded fine-grained Zn1-xCdxS/CdS heterostructure using a surfactant-free hydrothermal method based on the following points: i) Zn1-xCdxS/CdS shows improved structure matching and photocatalytic performance due to the tunable parameters, and the excellent photocatalytic activity of Zn1-xCdxS;18 ii) ZnS precursors with different morphologies were prepared by changing the concentrations of NaOH, and the content of exposed active sites of the Zn1-xCdxS/CdS heterostructure could be facilely modulated by the varied particle sizes of ZnS precursors; iii) the in-situ embedding of the well-dispersed ZnO nanocrystals on the Zn1-xCdxS/CdS heterostructure was obtained by the reaction between OH- and the residual Zn2+ derived from the cation exchange of ZnS with Cd2+ .
Herein, the Zn1-xCdxS/CdS heterostructures with modulated morphologies were facilely fabricated through varying the particle sizes of ZnS precursors by the implantation of OH- on their surfaces. The photocatalytic activity of Zn1-xCdxS/CdS heterostructure was found closely dependent on the exposed active sites, and the heterostructure with the smallest particle size shows the maximal H2 -evolution activity of 22.99 mmol/h/g. The in-situ synthesis of ZnO-embedded Zn1- xCdxS/CdS heterostructure was obtained by introducing different contents of NaOH. The ZnO nanocrystal, serving as cocatalyst to separate the charge carriers and acting as H2- production sites, shows tremendous enhancement on the photocatalytic activity of Zn1-xCdxS/CdS heterostructure to 84.17 mmol/h/g, especially in the methanol solution. Notably, the ZnO-embedded Zn1-xCdxS/CdS shows distinct photocatalytic H2 -production of 0.18 mmol/h/g even in the absence of sacrificial agent, although the Zn1-xCdxS/CdS shows no hydrogen production. The content of exposed active sites and the loading of ZnO cocatalyst on the heterojunction show synergistic effect on the photocatalytic H2 -evolution activity.
Cadmium nitrate tetrahydrate (Cd(NO3)2 ·4H2O), zinc chloride (ZnCl2 ), thiourea (CH4N2S) and sodium hydroxide (NaOH) are analytical grade, and used as received without further purification.
Synthesis of ZnS precursors with different particle sizes
The ZnS precursors were prepared using ZnCl2 , CH4N2S as Zn, S sources via a hydrothermal method. NaOH was adopted to modulate the particle size of ZnS. In a typical process, 3.0 mmol ZnCl2 and 6.0 mmol CH4N2S were dissolved in 25 mL distilled water (solution A). Different contents of NaOH with the molar ratios of NaOH/Zn equaling to 0, 1, 2, 4, 8, 10, 20 were dissolved in 5 mL distilled water (solution B). Then the solution B was added into solution A slowly under ultrasound for 0.5 h. The resultant solution was subsequently transferred into 50 mL teflon-lined autoclave and maintained 180 oC for 21 h. The final white products were rinsed three times with distilled water and ethanol respectively, and dried at 60 oC for overnight in a vacuum oven to evaporate the solvent. The obtained ZnS nanocrystals with different particle size were used as precursor for the fabrication of the heterostructure samples.
Synthesis of the heterostructure samples with modulated morphology by the different particle size of ZnS precursors
30 mg ZnS precursors (prepared with the molar ratios of NaOH/Zn = 0, 1, 2, 4, 8, 10, 20) were dissolved in 10 mL deionized water under ultrasound for a few minutes. Suitable content of Cd(NO3)2 ·4H2O (Cd/(Zn+Cd) = 30 at%) were dissolved in deionized water (2.8 mL) and then drop into the aforementioned solution quickly under mild stirring. After several minutes, the obtained solution was transferred into 15 mL autoclave and maintained 140 oC for 12 h. The final products were rinsed with distilled deionized water and ethanol for three times, respectively, and dried at 60 oC overnight in the vacuum oven to evaporate the solvent.
In-situ synthesis of the ZnO cocatalyst
30 mg ZnS with the smallest particle size (prepared with the NaOH/Zn molar ratio of 4) was dissolved in 8 mL deionized water under ultrasound for a few minutes (Solution A). Suitable content of Cd(NO3)2 ·4H2O (Cd/(Zn+Cd) = 30 at%) was dissolved in deionized water (3 mL) (Solution B). Different contents of NaOH with the ratio of NaOH/Cd equaling to 1, 5, 10, 15 and 20 were dissolved in 1 mL distilled water and then dropped into the mixture solutions of A+B quickly under mild stirring. After several minutes, the obtained solution was transferred into 15 mL autoclave and maintained 140 oC for 12 h. The final yellow products were rinsed with distilled deionized water and ethanol for three times respectively, and dried at 60 oC overnight in the vacuum oven to evaporate the solvent.
The powder X-ray diffraction (XRD) patterns were recorded on a D/max 2500 VL/PC diffractometer (Japan) equipped with graphite monochromatized Cu Kα radiation (λ = 1.54060 Å). Corresponding work voltage and current is 40 kV and 100 mA, respectively. The transmission electron microscopy (TEM) and high-resolution TEM (HRTEM) images were recorded on JEOL-2100F apparatus at an accelerating voltage of 200 kV. Surface morphologies of the heterojunction materials were examined by a scanning electron microscope (SEM, JSM- 7600F) at an acceleration voltage of 10 kV. The energydispersive X-ray spectroscopy (EDX) was taken on JSM- 5160LV-Vantage typed energy spectrometer. UV−visible diffused reflectance spectra was recorded using a Cary 5000 UV-Vis spectrometer (Viarian, USA) with BaSO4 as a reflectance standard. The Brunauer−Emmett−Teller (BET) specific surface area (SBET), nitrogen adsorption of the heterojunction samples were analyzed by an Autosorb-iQ adsorption apparatus (Quantachrome instruments, USA). All of the samples were degassed at 90 °C for 3 hours prior to nitrogen adsorption measurements. The SBET was determined by a multipoint BET method using adsorption data in the relative pressure (P/P0 ) range of 0.05 − 0.3. Electrochemical impedance spectra (EIS) measurements were carried out in three-electrode system and recorded over a frequency range of 500 kHz-200 MHz with ac amplitude of 10 mV at 0.5 V. Na2S (0.1 M) and Na2SO3 (0.02 M) mixture solution was used as the supporting electrolyte. EIS data were recorded using Electrochemical workstation (EC-lab, SP-150, VMP3-based instruments, America) under a surface power density of about 0.1 mW/cm2 . Raman spectra of powder samples were obtained on Lab-RAM HR800 with a laser excitation wavelength of 514.5 nm.
Photocatalytic Hydrogen Production
The photocatalytic H2 -production experiments were performed via a photocatalytic H2 -production activity evaluation system (CEL-SPH2N, CEAULight, China) in a 300 mL Pyrex flask, and the openings of the flask were sealed with silicone rubber septum. A 300 W xenon arc lamp through a UV-cutoff filter with a wavelength range of 420−1000 nm, which was positioned 13 cm away from the reaction solution, was used as a visible light source to trigger the photocatalytic reaction. The focused intensity on the flask was ∼200 mW·cm−2, which was measured by a FZ-A visible-light radiometer (CEL-SPH2N, CEAULight, China). In a typical photocatalytic H2 -production experiment, 5 mg of the as-prepared photocatalyst was suspended in 100 mL of mixed aqueous solution containing Na2S (0.35 M) and Na2SO3 (0.25 M). Before irradiation, the system was vacuumed for 5 min via the vacuum pump to completely remove the dissolved oxygen and ensure the reactor was in an anaerobic condition. A continuous magnetic stirrer was applied at the bottom of the reactor to keep the photocatalyst particles in suspension during the experiments. H2 content was analyzed by gas chromatography (GC-7900, CEAULight, China). All glasswares were carefully rinsed with distilled water prior to usage.
Results and discussion
Zn1-xCdxS/CdS heterostructures with adjustable morphologies were accessible by varying the particle sizes of ZnS precursors. 19 Scheme 1 illustrates the fabrication route for modulating the morphologies of ZnS precursors, the Zn1- xCdxS/CdS heterostructure and ZnO-embedded Zn1-xCdxS/CdS heterostructure. The ZnS precursors possessing different particle sizes were firstly fabricated, where the hydroxide plays an indispensable role in the formation of the fine-grained ZnS precursor by the repulsive interaction among ZnS nanoparticles.20
The morphologies of Zn1-xCdxS/CdS heterostructures with the optimal content of Cd could be facilely modulated by the different particle sizes of ZnS precursors (Fig. S1).21 The Zn2+ derived from the cation exchange between ZnS and Cd2+ reacted with the superfluous hydroxyl ion and produced [Zn(OH)4] 2-, then immediately transformed into the ZnO,22 and the ZnO-embedded Zn1-xCdxS/CdS heterostructure was formed. The Zn1-xCdxS/CdS heterostructure samples derived from the ZnS precursors with different particle sizes were abbreviated as ZACm (m =0, 1, 2, 4, 10, 20, m equals to the molar ratio of NaOH/Zn in preparation of ZnS). As can be observed from SEM images, the adoption of NaOH leads to the decrease of average diameter of ZnS nanoparticles from 1.5 µm to around 20 nm (Fig. 1a, b). Significantly, the morphologies of Zn1- xCdxS/CdS heterostructure samples were closely dependent on that of ZnS precursors: the heterostructure derived from the ZnS with smaller particle sizes shows the higher porous morphology. (Fig. 1c, d and Fig. S2).
Characterization of the phase structure via XRD patterns (Fig. 1e) indicates that the modulated morphology of heterostructure samples resulting from the different particle sizes of ZnS show tremendous effect on the crystallinity of the heterostructure. Compared with ZAC0, ZAC4 possesses much stronger diffraction peaks corresponding to CdS and Zn1-xCdxS phases due to its smaller particle size, and larger content of heterostructure active sites confirmed by the largest amount of Cd from the electron dispersive X-ray spectrum (EDX) results (Fig. S3). The S, Zn and Cd distribute uniformly in the whole matrix of the nanoparticles (Fig. S4). Effect of the as-prepared heterostructure samples on their UV-Vis diffuse reflection spectra and band structure is displayed in Fig. 1f. Notably, the spectra of the heterostructure samples show two band edges, and the entire diffuse reflection spectra could be divided into region I and II, which are corresponding to absorption in the UV and visible regions, respectively.18 Moreover, the heterostructure samples with smaller particle size possess the larger bandgap (Eg ) in the region II (inset of Fig. 1f), which may be attributed to the quantum effect of CdS.18