1. Introduction
The excessive emission of CO
2 has caused great harm to the environment and human beings. CO
2 is the main greenhouse gas that causes global warming, which has brought about a series of environmental problems, including melting ice caps and glaciers, rising sea levels, changes in precipitation patterns, more extreme weather events, and biodiversity loss [
1,
2]. The increase in CO
2 concentration in the atmosphere has also led to a decrease in the pH value of the ocean, resulting in ocean acidification. This has damaged marine ecosystems and the ability of coral reefs to survive, affecting fisheries and other industries that rely on ocean resources. Moreover, human health is also affected by high concentrations of CO
2 in the atmosphere. Exposure to high concentrations of CO
2 can cause respiratory problems, such as asthma and other respiratory infections. To address the issue of excessive CO
2 emissions, it is necessary to take measures to reduce emissions. This includes using clean energy sources, such as solar, wind, and nuclear energy, instead of coal and other fossil fuels that emit large amounts of CO
2 [
3]. In addition, planting trees and other carbon-absorbing plants can help absorb some of the excess CO
2 in the atmosphere. Finally, policies should be implemented to encourage more sustainable practices and reduce emissions from vehicles and other sources. The burning of fossil fuels, such as coal, oil, and natural gas has produced a large amount of greenhouse gas emission, especially CO
2, which has caused global climate change and environmental problems [
4,
5]. To solve these serious problems, developing clean and renewable energy sources, such as solar energy, is of great importance for human beings [
6]. As a new type of technology, the photocatalytic reduction of CO
2 into hydrocarbons has drawn a lot of attention, as it converts the greenhouse gas, CO
2, into renewable energy via solar light irradiation, solving the energy crisis and greenhouse effect simultaneously [
7,
8,
9].
One of the most commonly used photocatalysts for CO2 reduction is titanium dioxide (TiO2), which has a high photocatalytic activity and stability under UV light. The photocatalytic reduction process typically involves two main steps: (1) the absorption of light by the photocatalyst, resulting in the generation of electron–hole pairs, and (2) the transfer of the electrons to a reducing agent, such as H2 or CO, to generate the desired product. In the first step, the photocatalyst absorbs light energy, typically in the form of visible or ultraviolet light, and generates excited electrons. These electrons are transferred to the surface of the photocatalyst, where they can participate in reduction reactions. The excited electrons reduce the CO2 molecule, which is adsorbed on the photocatalyst surface. This reduction process involves the addition of electrons to the CO2 molecule, reducing it to a lower-energy species. One common product of this reduction process is methane (CH4), but other compounds such as methanol or formic acid can also be produced. The overall process of photocatalytic reduction of CO2 involves several steps, including the absorption of light by the photocatalyst, the generation of excited electrons, the reduction of CO2, and the formation of the final product. The mechanism is highly efficient and can convert CO2 into useful compounds without the need for external reducing agents. This process has attracted significant interest for its potential to convert waste CO2 into valuable products while reducing the environmental impact of industrial processes.
The main advantage of photocatalytic reduction is that it can convert CO
2 into useful chemicals without the need for high temperatures or pressures. In addition, photocatalytic reduction can be achieved using solar energy, making it a sustainable and environmentally friendly process. However, there are still some challenges in implementing photocatalytic reduction on a large scale. These include improving the quantum efficiency of the process, increasing the selectivity for the desired product, and reducing the cost of the catalyst [
4,
5,
9]. Future research efforts will focus on overcoming these challenges to make photocatalytic reduction a more practical and efficient method for converting CO
2 into useful chemicals.
During the last decades, various semiconductors have been investigated as potential candidate photocatalysts for efficient and practical CO
2 reduction, such as TiO
2 [
6], ZnGa
2O
4 [
4], LaPO
4 [
9], WO
3 [
10], ZnS [
11,
12], and bismuthates [
13,
14]. As one of the most investigated photocatalysts, TiO
2 has a wide range of applications, including environmental pollution control, energy conversion, and organic wastewater treatment. The photocatalytic activity of TiO
2 is based on its ability to absorb ultraviolet light and generate electron–hole pairs, which are responsible for the activation of oxygen and the reduction of electron acceptors. The holes generated in the valence band of TiO
2 can react with water to produce hydroxyl radicals (•OH), while the electrons in the conduction band can reduce the adsorbed species on the catalyst surface. TiO
2 photocatalysis has been widely used for environmental purification, including the decomposition of organic pollutants in air and water. In addition, TiO
2 photocatalysis can also be used for energy conversion, such as in solar cells and artificial photosynthesis, to convert solar energy into chemical energy. The performance of TiO
2 photocatalysts is influenced by factors such as crystal structure, particle size, and surface structure. Nanostructured TiO
2 with a smaller particle size has a higher specific surface area and more active sites, resulting in higher photocatalytic activity. In addition, the surface structure and morphology of the catalyst can also affect its performance by providing more active sites or faster electron transfer kinetics. Despite its wide range of applications, TiO
2 photocatalysis still faces challenges such as its low activity in the visible light region and its sensitivity to water quality and environmental conditions [
6,
15,
16]. Future research should focus on overcoming these limitations to further improve the efficiency and practicality of TiO
2 photocatalysis in various applications.
Up to now, many efforts have been devoted to improving the photocatalytic activity of semiconductors on CO
2 reduction, such as doping or modification with foreign elements, composition with other semiconductors, and the adjustment of nanostructures [
11,
17,
18,
19]. However, most of these photocatalysts exhibit poor photocatalytic performance for practical application, because of the large band gap and the rapid recombination of charge carriers [
20,
21]. It is still necessary to further improve photocatalytic performance by developing a new type of photocatalyst with strong redox ability, a visible response, and high photocatalytic performance.
Zn
2SnO
4, also known as ZTO, is a ternary oxide semiconductor material with a cubic crystal structure in which zinc and tin exist in a tetrahedral coordination form and oxygen exists in a bidentate coordination form. The band gap of Zn
2SnO
4 material at room temperature is about 3.6 eV, which has good insulating properties and chemical stability. In addition, Zn
2SnO
4 also has high electrical conductivity, fast electron transfer, high chemical sensitivity, and excellent visible light absorption. Zn
2SnO
4 has wide application value in many fields. Zn
2SnO
4 is a versatile material with potential for a wide range of applications. Its unique physical and chemical properties make it widely applicable in energy, environmental monitoring, and fire prevention. With the continuous advancement of science and technology, the application areas of Zn
2SnO
4 will continue to expand, bringing more contributions to the development of human society [
18,
22,
23,
24].
Zn
2SnO
4 has been widely investigated as a photocatalyst, owing to its high electric conductivity, high electron mobility, stability, and charming optical properties [
18,
22,
23,
24]. Wang et al. composited ZTO with SnO
x by a one-pot hydrothermal method [
23]. The composited photocatalysts exhibit improved photocatalytic performance for the degradation of methyl orange (MO) and gaseous benzene. Shi and Dai synthesized highly ordered Zn
2SnO
4 nanotube arrays by using ZnO arrays as templates [
25]. These ZTO arrays exhibit enhanced photocatalytic activity. Julio et al. synthesized high crystalline Zn
2SnO
4 nanostructures using a hydrothermal method [
26]. They modified the Zn
2SnO
4 nanostructure with Pt to improve the photocatalytic performance of H
2 production. Ren et al. synthesized Zn
2SnO
4 with Mo doping and oxygen vacancies with high photocatalytic activity on HCHO oxidation [
27]. Mo doping and the introduction of oxygen vacancies would extend the visible response, promote the separation of charge carriers, and induce new reactive sites for the ZTO-based photocatalysts. Hong et al. composited ZTO with ZIF-8, which exhibits significant photocatalytic performance on Methylene Blue (MB) degradation [
22]. Our previous work also reported a simple synthesis of Sn
4+ ions self-doped Zn
2SnO
4 nanoplates [
18]. The Sn
4+ ions would induce a new doping energy level below the conduction band of ZTO. The visible response is extended, and charge carriers are separated. As a result, the photocatalytic activity on the CO
2 reduction of Sn
4+ ions self-doped Zn
2SnO
4 nanoplates is significantly improved. However, the detailed influence of the doped elements on the band structure and photocatalytic performance of ZTO-based photocatalysts still needs further investigation.
In this work, Co and Ni, two iron family elements, were selected to be weaved into the lattice of Zn2SnO4 to investigate the influence of foreign elements on the band structure and photocatalytic performance of Zn2SnO4 samples. The photocatalytic reduction of CO2 into CH4 was applied to evaluate the photocatalytic performance of the as-prepared catalysts. It was revealed that the lattice Zn2+ ions were replaced by the Co3+ or Ni3+ ions in the substitutional mode. The doped Co3+ or Ni3+ ions would induce a new doping energy level in the band gap of Zn2SnO4. As a result, the light response is enhanced, the charge carriers are separated, and the photocatalytic activity is improved for the Ni-doped Zn2SnO4 and Co-doped Zn2SnO4 samples. The photocatalytic mechanism of the enhanced CO2 reduction is also discussed.
3. Results and Discussion
To investigate the crystal structure of the ZTO-based samples, the XRD patterns of ZTO, Co-ZTO, and Ni-ZTO are shown in
Figure 1A. All three samples show strong and sharp XRD peaks at 2θ = 17.7°, 29.1°, 34.3°, 35.9°, 41.7°, 51.7°, 55.1°, and 66.4°, corresponding to the (111), (220), (311), (222), (400), (422), (511), and (440) planes of Zn
2SnO
4 (JCPDS. No. 24-1470). There is no other diffraction peak related to SnO
2, ZnO, NiO, or CoO observed in
Figure 1A.
Figure 1B shows the enlargement of the (311) plane for ZTO, Ni-ZTO, and Co-ZTO around 34.3°. Compared with the pure ZTO samples, the peak position of the (311) plane for Co-ZTO and Ni-ZTO shifted to a larger diffraction angle. In addition, the lattice parameters and cell volume also decreased for Co-ZTO and Ni-ZTO, in comparison with the pure ZTO samples. This implies that the introduced Co and Ni atoms were weaved into the lattice of Zn
2SnO
4.
It is known that the ionic radius of Zn
2+ ions and Sn
4+ ions is 74 pm and 71 pm, respectively. As iron family elements, the valence of Co and Ni atoms usually exist as +2 or +3 in metal oxide. The ionic radius of Co
2+ and Ni
2+ is 74 pm and 72 pm. The ionic radius of Co
3+ and Ni
3+ is 63 pm and 62 pm. According to the doping theory for metal oxide, it is difficult for the Ni and Co atoms to be incorporated into the Zn
2SnO
4’s lattice in the interstitial mode, as the ionic radius of Ni and Co is close to the ionic radius of Zn and Sn. Thus, it becomes reasonable to deduce that the Co and Ni ions are doped in the Zn
2SnO
4’s lattice in the substitutional mode. Moreover, if the lattice’s Zn
2+ or Sn
4+ ions are replaced by Co
2+ or Ni
2+ ions, the lattice parameters and cell volume of Zn
2SnO
4 should remain unchanged, owing to the similar ionic radius. If the lattice’s Zn
2+ or Sn
4+ ions are replaced by Co
3+ or Ni
3+ ions, the lattice parameters and cell volume of Zn
2SnO
4 should be reduced, as the ionic radius of the Co
3+ ions (63 pm) and Ni
3+ ions (62 pm) is much smaller than that of the Zn
2+ ions (74 pm) and Sn
4+ ions (71 pm). According to experimental data in
Figure 1B and
Table 1, it is reasonable to deduce that the Co
3+ and Ni
3+ ions are doped in the Zn
2SnO
4’s lattice in the substitutional mode. To further investigate the chemical states of the Co
3+ and Ni
3+ ions in Zn
2SnO
4, the XPS characterization is applied in the following section.
As the existing states of the Co and Ni ions in Zn
2SnO
4 are closely related to the valence of the Co and Ni ions, the XPS Survey, Co 2p of the Co-ZTO sample, Ni 2p of the Ni-ZTO sample, O 1s, Zn 2p, and Sn 3d spectra for the ZTO, Co-ZTO, and Ni-ZTO samples are plotted in
Figure 2. All the ZTO-based photocatalysts exhibit strong XPS peaks related to the C 1s, O 1s, Sn 3d, and Zn 2p. As shown in
Figure 2B, a weak XPS peak centered at about 779.2 eV is observed for the Co-ZTO samples. This weak Co 2p peak can be ascribed to the Co
3+ ions [
29]. As shown in
Figure 2C, the weak XPS Ni 2p peak at about 855.6 eV can be attributed to the Ni
3+ ions [
30]. Hence, it can be confirmed by the XPS that the introduced Co and Ni atoms exist as +3 valence in this work. Moreover, as the ionic radius of the Co
3+ ions (63 pm) and Ni
3+ ions (62 pm) is much smaller than that of Zn
2+ ions (74 pm) and Sn
4+ ions (71 pm), the lattice’s Zn
2+ or Sn
4+ ions are replaced by Co
3+ or Ni
3+ ions in the substitutional mode, leading to the reduced parameters and cell volume. In
Figure 2D, the lattice’s O 1s peak is at about 530.1 eV, and the surface’s O 1s peak is at about 531.5 eV for all three samples. In
Figure 2E, the Zn 2p
3/2 is at 1020.8 eV for ZTO and 1020.9 eV for Co-ZTO and Ni-ZTO. In
Figure 2F, the Sn 3d
5/2 and Sn 3d
3/2 are at 486.2 eV and 494.7 eV for all ZTO-based samples.
To investigate the morphology of Co-doped Zn
2SnO
4 and Ni-doped Zn
2SnO
4, the TEM and HR-TEM images of Co-ZTO and Ni-ZTO are shown in
Figure 3. As shown in
Figure 3A,B, both the Ni-ZTO and Co-ZTO samples consist of microspheres with an average diameter of 300 nm~500 nm. Moreover, these microspheres are constructed by a series of nanoplates, which are intercrossed with each other by self-assembly. These microsphere structures may leave many pores on the surface, increase the specific surface area, and improve the stability of photocatalysts, eventually benefiting the photocatalytic performance. Moreover, as shown in
Figure 3C, the fringe lattice spacing for Co-ZTO is estimated to be 0.488 nm, corresponding to the (111) plane of Zn
2SnO
4. As shown in
Figure 3D, the lattice spacing for Ni-ZTO is estimated to be 0.260 nm, corresponding to the (311) plane of Zn
2SnO
4. It should be noted that the lattice spacing for Co-ZTO and Ni-ZTO in the HR-TEM images is smaller than that of the standard Zn
2SnO
4 cards. This may be caused by the substitutionally doped Co
3+ and Ni
3+ ions in the Zn
2SnO
4’s lattice. Thus, it can be concluded from the TEM images that both the Co-ZTO and Ni-ZTO samples are composed of microspheres, which consist of nanoplates intercrossed with each other.
To better investigate the morphology of the as-prepared Co-ZTO and Ni-ZTO samples, the EDS elemental mapping of the Zn, Sn, O, Co and Zn, Sn, O, Ni for Co-ZTO, and Ni-ZTO samples are shown in
Figure 4 and
Figure 5, respectively. It is clear in
Figure 4 and
Figure 5 that the ZTO-based samples mainly consist of Zn, Sn, and O elements. In addition, a small amount of Ni and Co elements are randomly dispersed and did not enrich in some points. This also suggests the existence of Co
3+ and Ni
3+ ions doped in the lattice of Zn
2SnO
4 in the substitutional mode.
To obtain physical insight into the band structure of Co
3+-doped Zn
2SnO
4 and Ni
3+-doped Zn
2SnO
4, a theoretical calculation based on the density functional theory calculation (DFT) is carried out, and the corresponding density of the states for ZTO, Co-ZTO, and Ni-ZTO is shown in
Figure 6. Theoretical calculations were performed using the Vienna ab initio software package (VASP.5.4.4). As shown in
Figure 6A, the band gap of pure ZTO is estimated to be about 0.93 eV, which is much smaller than the experimental data. This is because the DFT calculation usually underestimates the band gap of the semiconductors [
31,
32,
33]. As shown in
Figure 6B,C, the valence bands of Co-ZTO and Ni-ZTO are mainly composed of the O 2p states as well as the Sn and Zn states. The conduction bands of Co-ZTO and Ni-ZTO are mainly contributed by the Zn and Sn states, with a small amount of the O 2p states. Moreover, compared with pure ZTO samples, new doping energy levels occurred in the band gaps of Co-ZTO and Ni-ZTO, respectively. The introduced doping energy levels are mainly contributed by the Co/Ni and O elements, suggesting the occurred energy levels originated from the doped Co and Ni ions. Hence, it can be concluded that the introduced Co
3+ and Ni
3+ ions in the substitutional mode would induce new doping energy levels above the valence band of ZTO.
To explore the influence of the doped Co and Ni ions on the band structure and light absorption capacity of the Zn
2SnO
4 samples, the UV-vis diffuse reflectance absorption spectra of the pure ZTO, Co-ZTO, and Ni-ZTO samples are shown in
Figure 7A. The pure ZTO samples show strong absorption in the UV region, which can be attributed to the electrons’ transition from the valence band to the conduction band of ZTO. For the Co-ZTO and Ni-ZTO samples, no blue shift or red shift is observed for the absorption edge of the spectra. This suggests that the doped Co
3+ and Ni
3+ ions did not change the band gap of the Zn
2SnO
4 microspheres. Moreover, it is found that the absorption spectra from the UV to the visible region are enhanced, which is caused by the doped Co
3+ and Ni
3+ ions in the substitutional mode. According to the DFT calculation results, the extended absorption can be ascribed to the electrons’ transition from the energy levels of doped Co
3+ and Ni
3+ ions in the substitutional mode to the conduction band of ZTO. In addition, the Tauc plots of ZTO, Co-ZTO, and Ni-ZTO are plotted in
Figure 7B. The band gap for ZTO, Co-ZTO, and Ni-ZTO is estimated to be 3.30 eV. As the doped Co and Ni ions did not change the morphology and nanostructure of the ZTO microspheres, the band gap of Co-ZTO and Ni-ZTO remained unchanged, in comparison with the pure ZTO samples. Therefore, according to the UV-vis absorption spectra, it can be confirmed that new doping energy levels originated from the doped Co
3+ and Ni
3+ ions that occurred in the band gap of ZTO.
The efficient separation and transfer of photogenerated charge carriers is beneficial for enhancing photocatalytic performance.
Figure 7C illustrates the photocurrent response versus time for the ZTO-based samples, which was analyzed to evaluate the separation efficiency of the charge carriers. Notably, the photocurrent response, as observed in
Figure 7C, exhibited a remarkable improvement for Co-ZTO and Ni-ZTO when compared to pure ZTO, during five on/off repetition cycles of the light. This observation strongly indicates an enhanced separation efficiency of the charge carriers in the Co-ZTO and Ni-ZTO samples.
The photocatalytic reduction of CO
2 into CH
4 is performed to investigate the effect of doped Co
3+ and Ni
3+ ions on ZTO samples, shown in
Figure 8. For the pure ZTO samples, only a small amount of CH
4 (1.02 × 10
−7 mol) is yielded after 8 h irradiation. For the Co-ZTO samples, more CH
4 is produced under irradiation. There is about 1.78 × 10
−7 mol of CH
4 detected for 8 hours’ irradiation. The Ni-ZTO samples exhibit the highest photocatalytic activity and about 1.92 × 10
−7 mol of CH
4 is produced, which is almost twice that for pure ZTO. Hence, doping Co or Ni ions into the lattice of ZTO is an effective method to improve the photocatalytic performance of ZTO-based photocatalysts.
Based on the experimental data and theoretical calculation results, the photocatalytic mechanism of ZTO, Co-ZTO, and Ni-ZTO can be explained as follows: under irradiation, only a small number of electrons can be excited from the valence band to the conduction band of ZTO. Hence, the pure ZTO samples exhibit poor photocatalytic activity in the CO2 reduction. After the introduction of Co and Ni ions in the substitutional mode, new doping energy levels occurred above the valence band of ZTO, and the light response from the UV to the visible region was also enhanced. More electrons can be excited from the doping energy levels of Co or Ni ions to the conduction band of ZTO and take part in the photocatalytic reaction. The introduction of Co3+ or Ni3+ ions would also benefit the separation of the charge carriers. The holes would react with H2O to yield H+. The electrons would react with CO2 to generate CO as the immediate product, which would further react with H+ and electrons to generate CH4 as the final product. As a result, more CH4 molecules would be produced for the Co-ZTO and Ni-ZTO samples. It is expected that the photocatalytic activity can be further improved by optimizing the amount of doped Co and Ni ions.