3.1. Characterization of Photocatalysts
X-ray diffraction (XRD) was used to analyze and confirm the crystal structure and phase composition of the existing catalyst (
Figure 1a). As clearly shown in the figure, the XRD pattern of the Cd/Er-BMO sample is very similar to the reference pattern JCPDS No. 21-0102 of the original BMO; the diffraction peak is sharp, indicating that the crystallinity of the prepared sample is good [
25]. In the Cd/Er-BMO sample, the 2θ diffraction angles are 11.0°, 27.2°, 28.3°, 32.5°, 33.2°, 36.1°, 46.7°, 47.1°, 55.6°, and 56.4°, respectively, corresponding to the PDF card of γ-Bi
2MoO
6 (JCPDS No. 21-0102) of (020), (140), (131), (200), (060), (151), (202), (260), (133), and (082). Among them, the most significant diffraction peak of the sample was observed at the (131) crystal plane. Upon the addition of different concentrations of Cd
2+, the peak at (131) continued to change but no other diffraction peaks were generated, further illustrating the successful doping of Cd
2+ into Er-BMO without producing other substances [
20]. Meanwhile, the resolution of the diffraction peak of the sample was high, indicating that the crystallinity was good. In particular, note that under different doping ratios, almost no additional diffraction peaks were revealed in the XRD images of Cd/Er-BMO, indicating that the crystal structure of Er-BMO was not changed after the introduction of Cd
2+. The results showed that compared with Er-BMO, the (131) diffraction peak of Cd/Er-BMO gradually increased with the increase in Cd
2+ doping concentration, and the highest peak appeared in the Cd/Er-BMO sample when Cd was 6%. The peak intensity increased and the shape of the diffraction peak was clear, indicating that the formation of nanocrystals was good. However, the subsequent peak drop can be explained by the excessive Cd
2+ doping concentration inhibiting the growth of crystal size, resulting in a smaller crystal size [
26].
Figure 1b shows that the Cd/Er-BMO (131) diffraction peak shifts to larger diffraction angles with increasing Cd
2+ doping concentration and reaches a maximum of 28.42° at 6% Cd/Er-BMO. The shift of the diffraction peak reflects the contraction of the crystal and the reduction in the crystal plane spacing [
27]. This is because the small ionic radius element replaces the large ionic radius element, because the Bi
3+ ionic radius (1.08 Å) is larger than the Cd
2+ ionic radius (0.97 Å). The subsequent shift to the left is due to lattice distortion caused by the high Cd
2+ doping concentration [
18].
The morphology and particle size of the synthesized samples were observed and studied using scanning electron microscopy (SEM). No particles, cracks, or surface roughness changes were observed in Er-BMO (
Figure 1c) and 6% Cd/Er-BMO (
Figure 1d). Compared with Er-BMO materials, the size of the 6% Cd/Er-BMO nanosheets was significantly reduced, changing from the original sheet shape to the needle shape. The emergence of the needle-like structure indicated that the addition of Cd
2+ resulted in a significant increase in the specific surface area of the intrinsic material and an increase in the contact area for the reaction. The appearance of the needle-like structure may be attributable to the doping of Cd
2+ ions, which affected the growth of Er-BMO [
28].
The microstructure and lattice planes of Er-BMO and 6% Cd/Er-BMO were measured using transmission electron microscopy (TEM) and high-resolution transmission electron microscopy (HRTEM) (
Figure 1e,f) to investigate the microstructure and properties of the materials. Both the Er-BMO and 6% Cd/Er-BMO samples have complete morphology, changing from an irregular flake structure to a needle-like structure. It is worth noting that the size of 6% Cd/Er-BMO was smaller than that of Er-BMO, which is consistent with the SEM analysis. HRTEM images showed the lattice features of Er-BMO (
Figure 1e) and 6% Cd/Er-BMO (
Figure 1f). The lattice fringes of Er-BMO and Cd/Er-BMO were clear, corresponding to the (131) and (200) planes of the orthorhombic crystal of γ-Bi
2MoO
6, respectively, indicating that there was no change in the orthorhombic phase before and after doping. The TEM and HRTEM images confirmed that the doping of Cd
2+ had no significant effect on the crystal lattice and did not produce any other crystalline phase, which is consistent with the XRD results mentioned above [
23].
The specific surface area is an important parameter in the photocatalytic degradation process that affects the performance and efficiency of the catalyst. A larger specific surface area can increase the active sites, improve the adsorption capacity, and increase light absorption. The specific surface area (BET) of Er-BMO and 6% Cd/Er-BMO was determined using N
2 adsorption/desorption isotherms. The pore size distribution was determined using the Barrett-Joyner-Halenda method (BJH). As shown in
Figure 2a,b, in the relative pressure range of 0.8 to 1.0, the isotherms of the two samples showed typical type IV characteristics, namely, the adsorption amount was caused by capillary condensation [
29] and formed a hysteresis loop in the medium and high relative pressure region. The figure shows that the 6% Cd/Er-BMO crystals had a stronger adsorption/desorption curve.
Figure 2b reveals that the pore size distribution of the sample was between 2 and 50 nm. This pore size is between micropores and macropores, which belong to a mesoporous structure. Mesoporous materials usually have a high specific surface area and good fluid transport properties. The specific surface area and average pore size of the tested samples are shown in the insert table in
Figure 2. When doping with Cd
2+, the specific surface area of Er-BMO increased from 8.414 to 12.638 m
2·g
−1, indicating that the doping of Cd
2+ ions resulted in a larger specific surface area of Er-BMO, as well as more active sites. Meanwhile, the increase in active sites equates to more catalytic active centres, which can effectively improve the rate and efficiency of the catalytic reaction. This is consistent with the above SEM analysis results [
30]. In addition, the study also showed that the average pore size of Cd/Er-BMO was better than that of Er-BMO, which resulted in a larger number of active surface sites [
31].
An energy spectrum (EDS) element diagram of 6% Cd/Er-BMO (
Figure 1g–k) shows that the Cd, Er, Bi, Mo, and O elements are uniformly distributed, which means that Cd
2+ ions are successfully doped into Er-BMO.
Analyzing the chemical state of the doped sample benefits the study of the bonding of elements. As such, XPS analysis was used to monitor the valence state and elemental composition of Er-BMO and Cd/Er-BMO. Prior to the analysis of the data, C 1s was used for calibration at 284.8 eV (
Figure 2c) given its stability, universality, clear peaks, and simple electronic structure. It can be observed that the characteristic peaks of the C, Cd, Er, Bi, Mo, and O atoms appeared in the XPS spectrum of 6% Cd/Er-BMO. Among them, the presence of the C element is attributed to the surface contamination of carbon materials (
Figure 2d) [
32], and the peak at 288.4 eV may be attributable to C-N or C-(N)
3 in the aromatic lattice [
33]. As shown in
Figure 2e, Cd 3d is located at 405.0 and 411.6 eV, corresponding to Cd 3d
5/2 and Cd 3d
3/2 of Cd
2+, respectively. Two strong peaks at 158.9 and 164.3 eV were observed in the Bi 4f XPS spectrum of 6% Cd/Er-BMO, which corresponds to Bi 4f
7/2 and Bi 4f
5/2, respectively (
Figure 2f), indicating that the valence state of Bi in 6% Cd/Er-BMO is +3 [
18]. Similarly, Mo 3d corresponded to the two peaks of Mo 3d
5/2 and Mo 3d
3/2 that were located at 232.2 and 235.3 eV, respectively (
Figure 2g), indicating that Mo exists only in the Mo
6+ oxidation state in 6% Cd/Er-BMO [
34]. Two strong peaks at 170.2 and 169.0 eV were shown in the spectrum of Er 4d, corresponding to Er 4d
3/2 and Er 4d
5/2, respectively (
Figure 2h). The O 1s peaks of Er-BMO and 6% Cd/Er-BMO are shown in
Figure 2i. The O 1s peak was fitted using three Gaussian superposition peaks, corresponding to 530.0, 531.2, and 532.7 eV, respectively. The peak at 530 corresponds to lattice oxide species, while the peak at 531.2 corresponds to surface hydroxyl groups and the peak at 532.6 corresponds to weakly adsorbed water [
35]. The comparison result regarding the peaks of Er-BMO and Cd/Er-BMO indicated that doping Cd
2+ has no effect on the valence state of Er-BMO. Cd/Er-BMO has similar peaks to Er-BMO, indicating that their crystal structures are similar. The reduction in electron density weakens the electron shielding effect, thereby affecting the binding energy and bonding environment. After doping, the Gaussian peak of the Er-BMO sample shifts toward the high bond energy direction, indicating that doping with Cd
2+ will change the chemical environment [
36], replacing Bi
3+ or Er
3+ to form ionic bonds with oxygen atoms, or Cd
2+ is inserted into the Er-BMO lattice, resulting in a decrease in the electron density of Bi
3+, Mo
6+, and O
2− [
37]. This also proves that Cd is successfully incorporated into the Er-BMO lattice [
38].
The photocatalytic performance of materials is closely related to the recombination of electrons and holes as the photoexcitation, generation, separation, and recombination of electron-hole pairs have a significant impact on the efficiency of the photocatalytic reaction. As such, the electron-hole-related properties are analyzed by measuring the UV-visible diffuse reflectance spectrum, transient photocurrent response, and photoluminescence spectrum to understand the optical properties of the material in one step. As shown in
Figure 3a, the UV spectrum of Er-BMO has a band gap absorption edge near 471.23 nm. Compared with Er-BMO, upon the increase in the Cd
2+ doping ratio, the light absorption range of Cd/Er-BMO in the visible light region gradually undergoes a slight red shift, indicating that its light absorption capacity continues to increase and its photocatalytic efficiency continues to improve. The red shift phenomenon stops at 6% Cd/Er-BMO (493.30–513.19 nm). Upon the continued increase in the Cd
2+ concentration, a slight blue shift (513.19–484.11 nm) gradually occurs. The red shift indicates that Cd
2+ doping can broaden the light absorption range of Er-BMO and enhance visible light absorption, which leads to the generation of more electrons and holes and, eventually, may improve the photocatalytic reaction efficiency [
39]. The subsequent blue shift can be explained by the increased Cd
2+ doping concentration. An excessive exposure resulted in a change in the Er-BMO structure. In addition, new absorption peaks were found at 481.8 and 614.8 nm, which can be attributed to the transition of erbium ions 4F
3/2, 5/2 and 4F
9/2 from the 4I
15/2 ground state to the excited state, further confirming that Er
3+ can provide the energy upconversion centre and promote the transfer of electrons in the system [
38]. By plotting the relationship between (αhν)
2 and photon energy (hν), the band gap energy (Eg) of the Cd
2+-doped Er-BMO photocatalyst can be calculated; this is shown in
Figure 3b. In the figure, α, h, and ν are the absorption coefficient, Planck constant, and optical frequency, respectively. As can be seen from
Figure 3b, the band gaps of Er-BMO, 2% Cd/Er-BMO, 4% Cd/Er-BMO, 6% Cd/Er-BMO, 8% Cd/Er-BMO, and 10% Cd/Er-BMO are 2.81, 2.72, 2.63, 2.60, 2.74, and 2.78 eV, respectively. The energy band of the Cd
2+-doped Er-BMO photocatalyst is reduced in line with the trend in adsorption edge, and the minimum energy band is displayed at 6% Cd/Er-BMO. The smaller the band, the narrower the band width. The narrow band width can improve the photocatalytic efficiency. Therefore, 6% Cd/Er-BMO showed the largest photocatalytic activity [
26].
The transient photocurrent method is a widely used characterization technique to detect charge carrier photogeneration and extraction kinetics in optoelectronic devices. In order to illustrate the carrier separation efficiency, the response speed [
40] and the calculated transient photocurrent response is shown in
Figure 3c. As shown in
Figure 3c, the magnitude of the photoresponse intensity was 6% Cd/Er-BMO > 2% Cd/Er-BMO > 10% Cd/Er-BMO > 8% Cd/Er-BMO > 4% Cd/Er-BMO > Er-BMO, among which the photoresponse intensity of 6% Cd/Er-BMO was the largest. This proves that the content of electron-hole pairs generated by its photoexcitation is the highest. Compared with that of Er-BMO, the photoresponse intensity of Cd/Er-BMO increased to varying degrees, indicating that Er-BMO crystals can generate more electron-hole pairs under photoexcitation when doped with Cd
2+, thereby further promoting the photocatalytic reaction. PL is a non-destructive and non-contact optical method for probing the electronic structure of materials. When light hits the sample, it begins a process called photoexcitation, in which the light is absorbed and excess energy is given to the material [
41]. In addition, photoluminescence (PL) spectroscopy can be used to further analyze the separation and transport of photogenerated charges in materials and to further study the recombination efficiency of photocarriers in photocatalysis [
42]. This is the spectrum of the intensity or energy distribution of light with different wavelengths formed by the recombination of electrons and holes in the quasi-equilibrium state of the material. The higher the peak value, the higher the electron-hole recombination efficiency of the material and the worse the photocatalytic performance [
43]. The PL spectra of Er-BMO and Cd/Er-BMO are shown in
Figure 3d. The typical response peaks are identified in the emission spectra of Er-BMO and Cd/Er-BMO. Compared with the strong response peak of Er-BMO, Cd-BMO shows a relatively weak peak (the intensity of 10% Cd/Er-BMO is the highest), which may be attributable to an excessive doping concentration leading to the photoinduced recombination between electrons and holes and the subsequent increase in the PL intensity [
44]. This indicated that the appropriate amount of Cd
2+ doping can promote the transfer of electrons and holes. In summary, Cd/Er-BMO can effectively promote the generation of photogenerated carriers, which inhibit and improve the recombination and separation of electron-hole pairs by doping Cd ions in Er-BMO. In particular, 6% Cd/Er-BMO showed a significantly improved electron excitation and transfer ability and revealed an excellent photocatalytic degradation performance.
3.2. Photocatalytic Degradation
In order to explore the photocatalytic pollutant degradation ability of Er-BMO and Cd/Er-BMO with different doping ratios, 5 mg·L−1 SMZ was photocatalytically degraded under simulated visible light radiation under a 400 W long arc xenon lamp. Meanwhile, recycling degradation experiments and free radical capture experiments were conducted to study the stability of bismuth molybdate materials and the free radical degradation mechanism. For the repeatability of the experiment, we tested the irradiation intensity of the catalyst in the experiment, which is 4.72 × 104 Lux (Err: 0.13 × 104 Lux).
Firstly, the degradation efficiency diagram of the Er-BMO and Cd/Er-BMO crystals for the SMZ solution under simulated visible light irradiation to study the photocatalytic degradation performance is shown in
Figure 4a. The relative concentration (C
0 − C/C
0) of the SMZ solution (each point separated by 45 min) was plotted as a function of reaction time. The concentration of pollutant solution was measured using a UV-visible spectrophotometer, and the degradation efficiency under the visible illumination time can be calculated according to the following formula:
As shown in the figure, the adsorption degree of SMZ at 6% Cd/Er-MBO (more than 20%) is large, while the adsorption on Er-MBO can be ignored, which proves that the existence of Cd
2+ is related to the adsorption capacity. This may be related to surface acidity, as the addition of Cd
2+ leads to an increase in the number of oxygen vacancies. The degradation efficiency of Er-BMO for SMZ was only 3.11% within the first 180 min. Other than that, the degradation efficiency of SMZ by BMO was only 4.1% in 210 min, which proved that pure BMO was not active in the degradation process [
18]. After adding Cd
2+, the degradation efficiency was significantly improved with 6% Cd/Er-BMO revealing the most obvious increase, which reached 93.89% degradation efficiency in 180 min. For different contents of Cd/Er-BMO, the order of the degradation efficiencies is 6% Cd/Er-BMO > 8% Cd/Er-BMO > 10% Cd/Er-BMO > 4% Cd/Er-BMO > 2% Cd/Er-BMO. The results fully reflect the good photocatalytic ability of Cd/Er-BMO.
The repeated degradation experiment of 6% Cd/Er-BMO is shown in
Figure 4b. Four repeated photocatalytic experiments were carried out on 6% Cd/Er-BMO under the same reaction conditions to investigate the degradation of SMZ. That is, after the experiment was completed, the sample was recovered and the experiment was repeated for a total of four times. After one experiment, the sample is usually rinsed with water and dried before the next experiment. There will be some losses during the experiment. We will reduce the amount of pollutant solution according to the quality of the remaining samples in the previous proportion to compensate for the losses in repeated experiments. The results showed that the degradation efficiencies of the four experiments were 93.89%, 91.59%, 90.03%, and 89.40%, respectively, indicating its good sustainability, durability, and stability. In addition, the 6% Cd/Er-BMO photocatalyst after the degradation experiment was characterized using XRD and SEM (
Figure 4c,d). A comparison before and after the experiment showed that the peaks shown in the XRD spectrum of 6% Cd/Er-BMO were still consistent. The SEM image suggested the 6% Cd/Er-BMO maintained its morphology, which confirmed its stability and great potential.
At the same time, ICP-MS was used to detect the concentration of ions in the solution after photocatalytic degradation, and the concentration of the specified element was obtained using the correlation function. The test results and related functions are shown in
Table 1. Cd
2+ (0.0835 mg·L
−1) and Bi
3+ (0.4839 mg·L
−1) showed that there was a small amount of ion overflow during photocatalytic degradation, but the photocatalytic efficiency was not affected. It should be noted that the Cd
2+ concentration shown in the table is 0.0835 mL·L
−1, which is lower than the continuous emission concentration of the Chinese government (0.1 mg·L
−1) (ref: GB 8978-1996; China National Comprehensive sewage discharge standard). In summary, our experiments were carried out under prescribed and reasonable conditions, which also indicated that 6% Cd/Er-BMO is an excellent photocatalytic material. Nevertheless, the concentration of 0.0835 mg·L
−1 is higher than the limit for drinking water (0.005 mg·L
−1) and irrigation water (0.01 mg·L
−1), demonstrating that the solution still needs further treatment before it can be used for drinking and irrigation, and that the presence of Cd
2+ in the photocatalyst seems to place some limits on the application of this catalyst.
A capture experiment is an experimental method to detect and study particles, molecules, ions, or reaction intermediates. Free radicals and holes play a key role in many biological and chemical processes. Therefore, studying free radicals and holes plays an important role in understanding these processes. As shown in
Figure 4e, the formation and reaction mechanism of common functional groups and holes during the degradation of SMZ solution were studied through capture experiments. Since hydroxyl radicals (·OH), superoxide radicals (·O
2−), and holes (h
+) are the most active in photocatalysis, 1 mM isopropanol (IPA), 1 mM sodium bicarbonate (NaHCO
3), and nitrogen (N
2) are used as capture agents to capture ·OH, h
+, and ·O
2− functional groups. In addition, nitrogen was continuously introduced through a tube during the experiment to capture superoxide radicals. When IPA, NaHCO
3, and N
2 were added to the experiment, the photocatalytic degradation performance decreased significantly. Compared with the reaction system without the addition of a capture agent, the degradation effect of 6% Cd/Er-BMO decreased from 93.89% to 6.64% after adding IPA as a capture agent during the degradation process, which seriously affected the photocatalytic reaction activity. Secondly, the degradation effect dropped from 93.89% to 26.79% after adding nitrogen. After adding NaHCO
3, the degradation effect dropped from 93.89% to 46.43%. The results indicated that after co-doping, the ·OH, h
+, and ·O
2− radicals are still important active substances in the photocatalytic degradation of organic matter, and they play an important role in the degradation of SMZ. According to previous studies, the addition of IPA in the case of doping with only Cd
2+ did not significantly change the photocatalytic degradation performance [
18], which further confirmed that the doping of Er
3+ benefits the formation of ·OH [
23].
3.3. Photocatalytic Mechanism Analysis
The photocatalytic mechanism refers to the process and principle of photocatalysts inducing chemical reactions under light conditions, the study of which is conducive to understanding the nature of the photocatalytic process, optimizing the design of catalysts, and improving the efficiency of photocatalysis. Semiconductor photoexcitation as a photocatalytic mechanism refers to the band structure of semiconductor particles, which consists of a low-energy valence band (VB) filled with electrons and an empty high-energy conduction band (CB) with a forbidden band between them. When a light source of appropriate energy irradiates the semiconductor, the semiconductor will be activated as a result of the absorption of the energy of the photon, resulting in electron jumping from the valence band to the conduction band, generating electron-hole pairs. The electron-hole pairs separate and migrate to the surface of the semiconductor, and some of them recombine during the migration process, losing their activity. As such, to improve the quantum yield of semiconductor photocatalytic reactions, it is necessary to make the acceptor potential lower than the semiconductor conduction band potential, as well as making the donor potential higher than the semiconductor valence band potential, to effectively inhibit the direct recombination of photogenerated electrons and holes.
For the co-doping system in this paper, when the sample was irradiated with light with energy equal to or greater than the band gap energy, e
− and h
+ are formed and exist in the conduction band and valence band, respectively. As Cd
2+ was doped in Er-BMO, especially 6% Cd/Er-BMO, the band gap was decreased, the generation of electron-hole pairs was promoted with the direct recombination of photogenerated electrons, and the holes were effectively suppressed. As such, a large number of e
− and h
+ were separated to the appropriate position to contact the pollutants, improving the degradation efficiency. ·O
2− was developed when e
− contacts O
2, and then the h
+ and active radical (·O
2− and ·OH) were employed for the SMZ degradation based on the free radical capture experiment results. Note that h
+ is mainly used directly for the degradation experiments, as opposed to using ·OH formed by h
+. Thus, the possible degradation process is as follows: SMZ is hydrolyzed during the reaction to form the product TP 1, then h
+ and ·OH break the S-N bond to form the products TP 2 and TP 3. Finally, all intermediate products further react with h
+ and ·O
2− to form smaller molecules or mineralize to CO
2 and H
2O [
2]. In addition, in order to simplify the picture, we will replace the pollutant in the left half of
Figure 5 with R-H, which will generate H
2O and simple organic matter (R·) during the reaction process. The detailed photodegradation process is as follows: