1 Introduction
The conversion of fuel into solar energy has greatest potential for achieving the goal of a low-carbon economy [1–6]. In particular, photocatalytic cracking of water is a primary measure adopted to generate clean, efficient, green, and storable hydrogen energy [3,7–14]. However, in many semiconductor photocatalysts, due to their limitations of low activity, poor stability, fast recombination of photogenerated carriers, and narrow light absorption range, the unmodified single catalysts are not ideal for photocatalytic hydrogen evolution [15–20].
Among the many candidate photocatalysts, CdS has been widely used in various photocatalytic reactions because of its suitable band gap and low-cost synthesis [21]. Moreover, the ingenious coupling of hybrid metal sulfides can expand the absorption band of the solar spectrum and exhibit appropriate redox potential [22]. However, CdS suffers from the shortcomings of photoetching, instability, and annihilation of electron−hole pairs [1,23]. To overcome these, a reasonable photocatalyst design is required. Among many methods, such as heterojunction construction, surface modification, and defect engineering, co-catalyst loading is an effective method to address this problem and achieve efficient hydrogen evolution [24–26]. Therefore, in this study, we combine CdS with a co-catalyst to overcome the demerits of CdS. Certain precious metals can act as effective co-catalysts because of their good conductivity and high electron density. However, their high cost limits their wide application [27,28]. Therefore, it is critical to identify non-precious metal co-catalysts such as sulfides, phosphates, transition metal oxides and carbon-based materials, which have been widely used [29,30]. Among them, the transition metal oxide ZnCo2O4 has attracted much attention because of its large specific surface area, suitable electrochemical performance, environment-friendliness and morphological diversity. The synthesis strategies are different, and the prepared ZnCo2O4 includes nano-blocks, nano-microsphere, nanosheets and nanowires [31–33]. Because of its various forms, ZnCo2O4 has been widely used in electrochemistry and capacitor electrode materials. However, it is rarely used in photocatalytic hydrogen production.
To improve the photocatalytic hydrogen evolution performance of CdS, this study constructs a composite photocatalyst ZnCo2O4/CdS by using ZnCo2O4 as a co-catalyst. The appropriate interface between ZnCo2O4 and CdS accelerates the electron transfer, improves the light absorption efficiency, and thus, enhances the hydrogen evolution performance. Thus, this study is expected to contribute to the design of co-catalysts.
2 Experimental
2.1 Synthesis of ZnCo2O4 catalyst
First, 0.58 g of Co(NO3)2·6H2O, 0.29 g of Zn(NO3)2·6H2O, 0.07 g of NH4F, and 0.30 g of CO(NH2)2 were dispersed into 35 mL of deionized water. After stirring in a magnetic agitator for 1 h, the mixed solution was transferred to a 50 mL autoclave lined with Teflon at 120 °C and maintained at this temperature for 5 h. At the end of the reaction, the products were collected, washed with deionized water and ethanol three times, and then dried. Finally, the dried sample was calcined in air at 400 °C for 2 h at a rate of 5 °C∙min–1. The black powder obtained by calcination was named ZCO.
2.2 Synthesis of CdS catalyst
First, 3.68 g of Cd(NO3)2 was added to 50 mL of ethylenediamine and stirred for 1 h. Then, 2.85 g of thiourea was added to the solution and stirred for another 1 h. Next, the solution was transferred to a 100 mL autoclave lined with polytetrafluoroethylene under a reaction temperature of 160 °C and reaction time of 24 h. After the reaction, the product was washed with ethanol and deionized water, and then dried. The finally obtained yellow sample was CdS.
2.3 Preparation of composite materials
The composite catalyst was prepared using a simple physical mixing method. First, 0.005 g of ZnCo2O4 was added to a beaker containing 0.1 g of CdS and 30 mL of alcohol. After ultrasonic treatment for 10 min, the solution was stirred and dried in a water bath. The composite catalyst was named 5%-ZCOCS. Under the same conditions, the composite catalysts with the proportions of 10%-ZCOCS, 15%-ZCOCS, and 20%-ZCOCS were synthesized.
2.4 Photocatalytic hydrogen evolution experiments
The experiment of photocatalytic hydrogen evolution was conducted in a closed glass bottle by using a 5 W white-light multi-channel photocatalytic reaction system. First, we added 10 mg of the catalyst and 30 mL of 10% lactic acid to the glass bottle, stirred the solution for 5 min, and then removed air from the glass bottle with nitrogen. After that, 0.5 mL of the gas was extracted from the glass bottle every hour and detected and analyzed by gas chromatography. The photocatalytic hydrogen evolution stability test was conducted every 4 h, with a total of four cycles.
2.5 Characterization
The crystal structure of the sample was analyzed by powder X-ray diffraction (XRD) under Cu-Kα radiation. The morphology of the sample was analyzed using a scanning electron microscope (SEM, Zeiss Evo 10) and a transmission electron microscope (TEM, Tecnai G2-TF30). The elemental composition and chemical state of the sample were analyzed by X-ray photoelectron spectroscopy (XPS, ESCALAB 250Xi). The adsorption and desorption isotherms of nitrogen were measured using ASAP 2020 M. The results were analyzed by the Brunauer–Emmett–Teller (BET) equation. The Barrett–Joyner–Halenda (BJH) model was used to analyze the pore size distribution curve. Taking BaSO4 as the reference, the UV–Visible diffuse reflectance spectrum (UV–Vis DRS) of the sample was measured using a UV-2550 spectrometer. The photoluminescence (PL) spectra and time-resolved fluorescence (TRPL) spectra were obtained using FluoroMAX-4.
3 Results and discussion
3.1 XRD analysis
The crystal structure of the catalyst is analyzed by XRD. In Fig.1, the diffraction peaks at 31.26°, 36.71°, 44.77°, 59.26° and 65.15° belong to the (220), (311), (400), (511), and (440) crystal faces of ZnCo2O4 (JCPDS No. 23-1390), respectively. The characteristic diffraction peaks of CdS (JCPDS No. 41-1049) are observed at 24.83°, 26.58°, 28.21°, 36.71°, 43.68°, 47.82°, and 51.85°, corresponding to their (100), (002), (101), (102), (110), (103), and (112) crystal planes. In addition to the evident diffraction peaks, the small diffraction peaks also correspond well with the standard card, indicating that CdS has suitable crystallinity. The XRD diagram of the composite catalysts is also shown, where the diffraction peaks of CdS and ZnCo2O4 appear in the diffraction peak of the composite catalyst. There is no new diffraction peak, which indicates that the composite is of high purity and is successfully prepared by CdS and ZnCo2O4.
3.2 SEM and TEM analyses
The morphology and structure of the catalyst are characterized using SEM and TEM, respectively (as shown in Fig.2). As shown in Fig.2(a), CdS exhibits a rod-like structure. Further zooming-in the figure, clearly shows that the surface of the rod-like structure of CdS is relatively smooth and different in length (Fig.2(b)). Fig.2(c) shows ZnCo2O4 nanospheres formed by the accumulation of rods, and Fig.2(d) shows the SEM image of the composite sample. The large ZnCo2O4 nanospheres adsorb many CdS nanorods.
The TEM image is used to further understand the sample’s microstructure, as shown in Fig.2(e, f). Under the TEM, large ZnCo2O4 nanospheres are observed to be tightly packed black parts, while CdS appears as surrounding dispersed nanorods. Fig.2(f) shows the image obtained by magnifying the edges of the contact between the two. In the enlarged image, CdS appears to be smooth, while the rods that from the ZnCo2O4 nanospheres are not smooth. Fig.2(g) indicates a HRTEM image, which shows the lattice stripes of CdS and ZnCo2O4, where the lattice fringe spacing of 0.34 nm belongs to the (002) plane of CdS, and that of 0.24 nm corresponds to the (311) plane of ZnCo2O4. In addition, the energy-dispersive X-ray (EDX) images of Fig.2(h) show the presence of Zn, Co, S, Cd, and O elements in the composite sample. The successful preparation of the composite sample is further illustrated.
3.3 XPS analysis
The chemical states of the elements present in the samples are studied by XPS (as shown in Fig.3). In the full spectrum shown in Fig.3(a), all elements of pure CdS and ZnCo2O4 can be observed in the composite sample ZCOCS. Fig.3(b–f) further show the high-resolution XPS spectra of Cd 3d, S 2p, Zn 2p, O 1s, and Co 2p. As shown in Fig.3(b), in the Cd 3d XPS spectrum of CdS, the peaks with binding energies of 404.15 and 410.90 eV correspond to Cd 3d5/2 and Cd 3d3/2, respectively [34]. In addition, the binding energies of 160.01 and 161.19 eV correspond to S 2p3/2 and S 2p1/2, respectively (Fig.3(c)) [35]. Clearly, with the addition of ZnCo2O4, the peaks of elements Cd and S move in the direction of a large binding energy. Fig.3(d) and Fig.3(e) show high-resolution XPS spectra of O and Zn. For Zn, the peaks are Zn 2p3/2 (1020.92 eV) and Zn 2p1/2 (1044.13 eV) [36]. The binding energy at 529.78 eV belongs to the Co–O bond [37]. In addition, the peaks at 531.62 and 533.19 eV are attributed to the surface hydroxyl groups and the O–H species absorbed water on the surface, respectively [38,39]. Moreover, the high-resolution XPS spectrum of Co (Fig.3(f)) presents two sets of double peaks in the spin orbit and two satellite peaks. The first group comprises the binding energies of 779.91 and 781.28 eV, and the second group comprises 794.63 and 796.06 eV, which belong to Co 2p3/2 and Co 2p1/2, respectively. The binding energies of 779.91 and 794.63 eV belong to Co3+, while those of 781.28 and 796.06 eV belong to Co2+ [38,40,41]. The satellite peaks are located at 787.85 and 804.49 eV respectively. Compared to pure CdS and pure ZnCo2O4, the binding energy of each element in the composite sample ZCOCS shifts in varying degrees, which indicates the presence of an interaction force between CdS and ZnCo2O4.
3.4 BET analysis
The specific surface area and pore size distribution of the samples are analyzed through nitrogen adsorption–desorption curves (Fig.4). The nitrogen adsorption–desorption curves of all three samples in Fig.4(a), Fig.4(b) and Fig.4(c) are type IV isotherm and H3 hysteresis ring [42]. Each illustration shows the pore size distribution curves obtained by the BJH method. As shown in the figure, the pore size distribution ranges within 2–50 nm, which belongs to mesoporous materials. According to the results of the BET analysis, the specific surface areas of ZnCo2O4, CdS and 15%-ZCOCS are 49, 35 and 39 m2∙g–1, respectively (Tab.1). The specific surface area of the composite sample is slightly higher than that of the single CdS. An increase in this area can expose more active sites, which is beneficial to the hydrogen evolution reaction. On the other hand, the change in adsorption parameters also indicates the successful preparation of the catalyst.
Tab.1 Adsorption parameters of samples |
| Sample | SBET/ (m2∙g–1) | Pore volume/ (cm3∙g–1) | Average pore size/nm |
|---|---|---|---|
| CdS | 35 | 0.26 | 28.52 |
| ZnCo2O4 | 49 | 0.23 | 16.27 |
| 15%-ZCOCS | 39 | 0.22 | 21.39 |
3.5 Photocatalytic hydrogen evolution
The hydrogen evolution performance of the sample under visible light irradiation (as shown in Fig.5). Fig.5(a) compares the photocatalytic hydrogen evolution properties of CdS, ZnCo2O4 and the composite sample ZCOCS. Due to the rapid recombination of photogenerated carriers, ZnCo2O4 shows very low photocatalytic activity for hydrogen evolution. After the introduction of ZnCo2O4, the hydrogen production capacity of CdS is improved, and the amount of hydrogen evolution can reach 7417.5 μmol∙g–1∙h–1, which is approximately 2.1 times that of CdS. Fig.5(b) compares the hydrogen evolution performance of different proportions of composite catalysts. As shown in the figure, the amount of hydrogen evolution differs with the addition of different loading amount of ZnCo2O4, and when the ratio of ZnCo2O4 to CdS is 20%, the amount of hydrogen evolution decreases. When the radio of ZnCo2O4 to CdS is 15%, the composite sample exhibits the optimal light absorption characteristics and suitable charge separation efficiency. The test of photocatalytic stability is another key point. Under the same conditions, after four times of hydrogen production, the composite catalyst exhibits suitable stability but the hydrogen production decreases a little due to the slight photocorrosion of the sulfide system [1,35]. Fig.5(d) shows the XRD diagram before and after hydrogen production. No miscellaneous peaks are observed in the XRD characteristic peak before and after the reaction, which shows that the structure is relatively stable. Furthermore, we reviewed some related studies on non-precious metal co-catalyst modified CdS for photocatalytic hydrogen evolution and compared them with our study, as shown in Tab.2.
Tab.2 Comparison of our study with the photocatalytic hydrogen production of CdS modified by other non-precious metals as co-catalysts |
| Composite photocatalyst | Sacrificial agent | Light source | Hydrogen evolution rate/(μmol∙g–1∙h–1) | Ref. |
|---|---|---|---|---|
| Mo2N/CdS | Na2S/Na2SO3 | 300 W Xe Lamp | 970 | [43] |
| NiCo2S4/CdS | Lactic acid | 300 W Xe Lamp | 6850 | [44] |
| Mo2C/CdS | Na2S/Na2SO3 | 300 W Xe Lamp | 1610 | [45] |
| VN/CdS | 10 vol % latic acid | 300 W Xe Lamp | 6240 | [46] |
| Cr0.5Ti0.5N/CdS | 10 vol % latic acid | 300 W Xe Lamp | 2440 | [47] |
| ZnCo2O4/CdS | 10 vol % latic acid | 5W white light | 7417 | This work |
3.6 UV–Vis DRS and PL analyses
The light absorption properties of the samples are evaluated by UV–Vis DRS (as shown in Fig.6). In Fig.6(a), the black catalyst ZnCo2O4 exhibits suitable light absorption properties. The absorption edge of pure CdS nanorods is approximately 543 nm. Compared to CdS, the composite sample 15%-ZCOCS exhibits slightly stronger light absorption intensity and similar absorption band edge. This shows that the introduction of ZnCo2O4 helps to enhance the absorption of visible light. In addition, the bandgap widths of CdS and ZnCo2O4 are estimated by (αhv)1/n = A(hv–Eg). As shown in Fig.6(b), the bandgap width of CdS is 2.32 eV and that of ZnCo2O4 is 1.60 eV [48].
Fig.6(c) shows the PL spectra of CdS, ZnCo2O4 and different composite catalysts excited at the same wavelength. The important information reflected by the PL spectrum is the fluorescence intensity of photogenerated electron–hole pairs recombination. The figure clearly shows that the fluorescence intensity of CdS is very strong, indicating a high recombination rate of electron–hole pairs. The fluorescence intensity of the composite sample with different ratios is lower than that of CdS, indicating that the introduction of ZnCo2O4 can effectively inhibit the recombination of the electron holes, which is more conducive to a good photocatalytic hydrogen evolution reaction. In addition, the TRPL spectra of CdS, ZnCo2O4 and the different ratios of composite catalysts are also studied, as shown in Fig.6(d). The following triple exponential attenuation model is used for fitting:
Meanwhile, the average life expectancy can be expressed as follows:
The fitting results of the average life are shown in the table. The average lifetime of the composite sample is slightly shorter than that of CdS, indicating that the speed of electron transfer between the interfaces is faster. The introduction of ZnCo2O4 promotes the separation of electron−hole pairs.
3.7 Electrochemical analysis
The photoelectrochemical test is conducted to further understand the charge separation and transfer efficiency in the photocatalysis process (as shown in Fig.7). Fig.7(a) shows the periodic photocurrent response curve of the sample, where the sample shows a good response to the periodicity of switching light. The composite sample 15%-ZCOCS shows good sensitivity and its current density is much higher than those of CdS and ZnCo2O4. This shows that the composite samples can provide more stable carriers and are easy to separate while the separation efficiency of pure samples is poor. In addition, the electrochemical impedance spectra (EIS) of the samples are measured to study the electron transfer efficiency, as shown in Fig.7(b). In the presence of ZnCo2O4, 15%-ZCOCS has a smaller arc radius. The smaller the arc radius, the lower is the transfer resistance, and the faster is the charge transfer. All these indicate that the presence of ZnCo2O4 promotes the separation and transfer of photogenerated carriers. Fig.7(c) shows the linear sweep volt-ampere curve of the sample. In general, a lower overpotential contributes to hydrogen production [49]. The diagram shows that under the same conditions, the composite has a lower hydrogen evolution overpotential, which is more conducive to the occurrence of photocatalytic hydrogen evolution. In addition, the cyclic voltammetry (CV) curves are measured to further describe the electrochemical properties of the samples. Fig.7(d, e) show the CV curves of the sample, and the area of the CV curve represents the redox ability. Furthermore, the CV curves under the different scanning speeds are studied, and the scanning speed is found to be proportional to the ring area of the CV curves [50].
In addition, the flat band potential of the catalyst is determined by conducting the Mott−Schottky test, and the position of the conduction band is determined. As shown in Fig.7(f), compared to the saturated calomel electrode, the flat-band potentials of CdS and ZnCo2O4 are –0.93 and –0.54 V, respectively. From the tangent slope, both catalysts are n-type semiconductors. After conversion to the normal hydrogen electrode, the flat band potentials of CdS and ZnCo2O4 are –0.69 and –0.30 V, respectively. According to the characteristic of the n-type semiconductor, the flat band potential is 0.1–0.2 V higher than the conduction band potential [51,52]. As a result, the conduction band potentials of CdS and ZnCo2O4 are –0.79 and –0.4 V, respectively. According to the relationship between the band gap and the conduction band potential: Eg = EVB – ECB, where the band gaps of CdS and ZnCo2O4 are 2.32 and 1.60 eV, respectively, and their valence band potentials are 1.53 and 1.2 V.
3.8 Possible photocatalytic mechanism
As shown in Fig.8(a, b), we conduct hydroxyl radical capture experiments on this basis.
In the experiment, terephthalic acid was mainly used to scavenge the hydroxyl radicals. The fluorescence intensity indicates that the binding rate of terephthalic acid to hydroxyl radical is high, which indirectly reflects the oxidation ability of the holes. In addition, with time, the system produces more photogenerated electron−hole pairs; thus, the fluorescence intensity shows that the composite catalyst has high oxidation ability [53].
Fig.9 shows the charge transfer mechanism between CdS and ZnCo2O4. In the electrochemical tests, both CdS and ZnCo2O4 are typical n-type semiconductors. Combined with XPS, the binding energy of Cd and S elements in the composite sample is larger than that of the elements in CdS, indicating that the electron cloud density in CdS decreases. In contrast, the electron cloud density in ZnCo2O4 increases. Based on the above analysis results, the possible mechanism of hydrogen evolution is proposed. When exposed to visible light, the excitation produces electron−hole pairs. Because ZnCo2O4 has low impedance, good conductivity and close contact with CdS, the electrons in the conduction band of CdS can be quickly transferred to ZnCo2O4. At the same time, the hydrogen evolution reaction occurs on the ZnCo2O4 surface, and the useless holes in the CdS valence band are consumed by lactic acid, which effectively hinders the recombination of electron holes and makes more electrons participate in the hydrogen evolution reaction. These results show that the photocatalytic activity of the catalyst for hydrogen evolution is improved.
4 Conclusions
ZnCo2O4/CdS composites are synthesized by physical mixing, and their properties are further characterized by SEM, TEM, and XPS. Compared to the single CdS, the photocatalytic hydrogen evolution performance of the co-catalyst ZnCo2O4 modified composite is significantly improved. The photocatalytic hydrogen evolution capacity of 15% ZnCo2O4/CdS reaches 296.7 µmol, which is twice as high as that of the single CdS, and has good stability. Fluorescence shows that the addition of ZnCo2O4 accelerates the charge separation and transfer. In addition, based on the analysis results of various characterizations, the possible mechanism of photocatalytic hydrogen evolution is proposed. The results of this study show that ZnCo2O4 can be used as a good co-catalyst. Meanwhile, the study contributes to the construction of a photocatalyst with high efficiency and low cost.