Synthesis of copper-doped indium zinc sulfide nanosheets for high efficiency photocatalytic hydrogen production

2.1. Synthesis of catalyst

2.2. Performance and characterization of samples

X-ray powder diffraction is a characterization method to analyze the existence form and phase structure of each component in materials. In this paper, Cu-Kα ray is used under the condition of voltage 40 k V, in which the scanning range is 5-80°, scanning rate is 5°/min. All powder samples were detected by Lynx Eye array detector (one-dimensional detector). JEOL-2100F transmission microscope was used for Morphology and elemental analysis of the samples with an acceleration voltage of 200 kV. The method of characterization was to add the catalyst sample solution on the ultra-thin carbon film, then dried it fully and characterized it. X-ray photoelectron spectroscopy is used to study and analyze the elements on the surface of materials to determine the relative content, chemical valence state and existence form of each element. The vacuum degree of the analysis chamber is lower than 5.0×10^-7 mBar, the excitation source is Mono Al Kα ray ($h\nu =$ 1486.6e V), the working voltage is 12 k V, and the filament current is 6 mA. The binding energy of all elements was calibrated by C1s (284.8 eV). The atomic force microscope (AFM) studies the thickness of nanosheets by detecting very weak interatomic forces between the surface of the sample under test and a tiny force sensor.

In order to analyze the hydrogen production performance of the photocatalyst samples, the experimental steps were as follows: 50 mg composite photocatalyst was weighed into the reactor, and 100 m L of S2-/SO32- sacrificant solution was added (the composition was 0.25 mol/L Na₂S solution and 0.35mol/L Na₂SO₃ solution). The reaction solution was treated by ultrasonic for 10 min to ensure that the photocatalyst was evenly dispersed in the solution, and then the catalyst could receive light evenly. The Ar is continuously pumped in for 30 min to displace the air in the device. The simulated light source is a xenon lamp with a 300 W power and a filter ($\lambda \ge$420 nm). The resulting gas is passed through a gas chromatograph equipped with a thermal conductivity detector (TCD, 13X column) to capture the amount of hydrogen produced. Samples were collected every 1 min in a cycle.

3.1. Characterization and analysis of surface structure

High resolution transmission electron microscopy (HRTEM) were used to study the morphology of the prepared samples. Fig. 1 shows the morphologies of Cu-ZIS and ZIS. As shown in Fig. 1, both Cu-ZIS and ZIS exhibit smooth nanosheet morphology. The size of the nanosheets was irregularly distributed, which may be due to the fragmentation of the nanosheets caused by ultrasound during sample pretreatment. It is worth noting that the HRTEM images of Cu-Zis and ZIS are very similar, and no obvious CuS or Cu2S phases are observed, indicating that Cu is indeed doped in indium zinc sulfide to further investigate the thickness of the nanosheets, atomic force microscopy (AFM) was carried out. Isolated nanosheets of Cu-ZIS were selected for characterization. By AFM analysis, it is further determined to be a flat nanosheet structure with ultra-thin thickness, as shown in Fig. 2. From the corresponding height section view, it can be seen that the thickness of ultra-thin nanosheet is 6.312 nm, which is about 2.5 times of the thickness of ZnIn₂S₄ single crystal cell (2.47 nm) along [001] direction. It is further confirmed that the structure is two-dimensional ultra-thin nanosheet.

Fig. 1a) TEM image of Cu-ZIS, b) TEM image of ZIS

a)

b)

In order to understand the crystal structure of the catalyst, the samples were studied using X-ray diffractometer (XRD). XRD patterns of Cu-ZIS (black) and ZIS (red) are shown in Fig. 3. As shown in the Fig. 3, three strong characteristic diffraction peaks appear at 21.6°, 27.7° and 47.2°, corresponding to (006), (102) and (110) lattice planes of hexagonal phase ZnIn₂S₄ (JCPDS No. 72-0773). This shows that the catalyst synthesized is indeed indium zinc sulfide. In the XRD of Cu-ZIS, only the characteristic peaks of indium zinc sulfide are observed, and no CuS or Cu₂S are observed, which further proves that Cu is doped in the lattice of indium zinc sulfide, rather than forming other substances. It is worth noting that the XRD peaks of Cu-ZIS and ZIS are almost the same, indicating that the introduction of Cu has no obvious effect on the lattice structure of indium zinc sulfide, which may be due to the fact that the atomic numbers of Cu and Zn are close and the atomic size difference is small. In Cu-doped indium zinc sulfide, Cu does not cause obvious lattice distortion at the position of replacing Zn.

Fig. 2AFM of Cu-ZIS

Fig. 3XRD for Cu-ZIS (black) and ZIS (red)

Fig. 4EDS spectrum for Cu-ZIS

In the initial feed ratio, the ratio of Zn to Cu was 5:1, but under this reaction condition, Zn must react more easily than Cu to form indium zinc sulfide, meaning that the ratio of Cu to Zn in the final product Cu-ZIS should be less than 1:5. Energy dispersive spectroscopy (EDS) characterization was performed to investigate the specific incorporation amount of Cu in the reaction. As shown in the Fig. 4, it should be noted that the peak of Mo element appears in the figure, which is caused by the fact that the base used in the test contains Mo. Therefore, the element content obtained is the result of normalization after removing Mo element. The results show that the atomic ratio of Cu to Zn is about 1:6, and the atomic percentage of Cu is about 2 %, which is consistent with the expectation. In order to further investigate the bonding state of Cu, X-ray photoelectron spectroscopy of Cu-ZIS was performed. As shown in Fig. 5, the 2p orbital diagram of Cu shows an obvious characteristic peak of Cu⁺ or Cu⁰ (933 eV), rather than the characteristic peak of Cu²⁺ (934.5 eV), indicating that Cu exists in the form of Cu⁺ or Cu⁰ in Cu-ZIS. In view of the strong coordination effect of S on Cu, and the most stable state of Cu in CuInS₂ is Cu+, it is speculated that Cu exists in the form of Cu²⁺ initially, and in the hydrothermal process, Cu²⁺ is very likely to be reduced to the most stable form of Cu⁺, rather than Cu⁰.

Based on the above experimental data, we successfully synthesized Cu-doped indium zinc sulfide ultra-thin nanosheets. For comparison, ultra-thin indium zinc sulfide nanosheets have also been synthesized. In Cu-ZIS, the percentage of Cu atoms is about 2 %, which mainly exists in the form of Cu⁺. Because Cu⁺ and Zn²⁺ have the same number of electrons and similar atomic size, the lattice structure of ZnIn₂S₄ does not change significantly after Cu replaces Zn, which indicates that Cu+ can exist stably in ZnIn₂S₄ to some extent.

Fig. 5XPS for Cu-ZIS

Fig. 6The catalytic properties of hydrogen production

3.2. Characterization of catalytic properties of hydrogen production

Evaluation of surface catalytic reactions by photocatalytic hydrogen evolution. The results under visible light irradiation are shown in Fig. 6. In order to determine the performance of the photocatalyst accurately, the water decomposition experiment of Na₂S and Na₂SO₃ as sacrificial agent without chloroplatinic acid as cocatalyst was measured under visible light irradiation ($\lambda \ge$420 nm). In Fig. 6, the hydrogen production rate of pure ZnIn₂S₄ is lower (23.5 μmol·h^-1·g^-1), which may be due to the rapid recombination of electrons and holes on the surface of ZnIn₂S₄. The hydrogen production rate of Cu-ZIS is 50.2 μmol·h^-1·g^-1, which is about twice that of ZIS. The above results indicate that the introduction of Cu heteroatoms can significantly improve the catalytic performance of indium zinc sulfide, which may be due to the fact that the introduction of Cu heteroatoms improves the hole separation efficiency of photogenerated electrons to a certain extent, thus improving the utilization efficiency of visible light, and electrons can effectively reduce H⁺.