Jul 17, 2020

### Structural properties

The microscopic morphology of the BC and ZnO/ZnS@BC was analysed by SEM (Fig. 2). Figure 2a,b shows the abundant circular pore structure of the BC without loaded particulate matter on the surface. Compared with BC, the surface of the ZnO/ZnS@BC (Fig. 2c,d) was loaded with abundant white particles that were irregular circles and tetragonal crystals with diameters ranging from tens to hundreds of nanometres. The particles were successfully supported on the surface of the BC.

XPS analysis was used to characterise Zn 2p, S 2p, O 1s, and C 1s to investigate the element and chemical valence states of carbon-based composite photocatalysts. Figure 3 shows the full spectrum of the carbon-based composite photocatalyst, which contained the characteristic peaks of Zn 2p, S 2p, C 1s, and O 1s, indicating that the carbon-based composite photocatalyst possessed Zn, S, O, and C elements. The peaks near 1,021.75 and 162.7 eV were the binding energies of Zn2p and S2p1/2, respectively, indicating the presence of Zn–S bonds in the catalyst. This was consistent with the value of the Zn–S bond in other reports29, 30. There was an O 1s peak near 532.0 eV. According to other studies, the surface O 1s peak was at 530.0 eV in pure ZnO31, 32. Compared to other literature, ZnO/ZnS exhibited a larger O 1s peak area near 532.0 eV and lattice oxygen intensity, which could be attributed to the aggregation of ZnS and ZnO26. Therefore, the heterojunction of ZnS and ZnO likely existed in the catalyst. Combined with SEM results analysis, the chemical composition of the white particles should be ZnO and ZnS. The heterojunction of ZnS and ZnO was loaded onto the BC.

Figure 4 shows the TEM image of the BC and ZnO/ZnS@BC and reveals detailed information about the microstructure of photocatalyst. There are no dark spots on the surface of the BC Fig. 4a,b, but darker spots appear on the surface of the BC in Fig. 4c,d. According to the XPS results, the darker spots might be ZnO and ZnS. Additionally, the grain size of the ZnO/ZnS composite was approximately 2 nm, and the ZnO/ZnS had an irregular shape.

EDS was an effective method to analyze the chemical composition of nanocomposites33. The microstructures and element distribution of ZnO/ZnS@BC photocatalyst were further studied by SEM and EDS mapping (Fig. 5a–f), respectively. The Fig. 5b indicated that the sample contains C, O, Zn and S elements, and Fig. 5c–f confirmed that C (red), O (green), Zn (purple) and S (yellow) elements were uniformly distributed in the heterojunction. Therefore, ZnO/ZnS and BC were intimately combined.

### Degradation of NOF under different conditions

Feasibility experiments were performed to investigate the effectiveness of the ZnO/ZnS@BC in degrading NOF under UV-light. Figure 6 shows the removal rate of NOF under various experimental conditions, where C is the concentration of NOF remaining in the solution after irradiation time t, and Co is the initial concentration of NOF. The ZnO/ZnS@BC and UV-light system removed 95% of the NOF, whereas the dark ZnO/ZnS@BC system removed 45% of the NOF, suggesting that the NOF was removed by adsorption. Under UV-light without the ZnO/ZnS@BC system, 20% of the NOF was degraded after 3 h, indicating that the NOF was not degraded by UV-light. The results showed that the NOF removal rate was highest in the ZnO/ZnS@BC and UV-light system. These results indicated that NOF removal is a synergistic process. BC possessed abundant adsorption sites and quickly attracted NOF to the surface. ZnO/ZnS had strong catalytic capability and quickly degraded the NOF. The presence of BC effectively prevented the aggregation of ZnO/ZnS.

### Kinetic study of NOF degradation by a ZnO/ZnS@BC and UV-light system

Kinetics help understand the mechanisms of pollutant degradation. Recent research reports that the kinetics of the photocatalytic degradation of aqueous pollutants follows the pseudo-first-order kinetic model34, and the integration of the pseudo-first-order kinetic model is the following equation:

$${ln}frac{{{text{C}}_{{0}} }}{{{text{C}}_{{text{t}}} }} = {text{k}}_{0} {text{t}}$$

where C0 is the initial concentration of NOF, Ct is the concentration of NOF at time t, k0 is the pseudo-first-order reaction rate constant (min−1), and t is the reaction time (min). The reaction rate constant (k) is calculated from the slope of a plot of ln (C0/Ct) versus (t).

The influence of various parameters such as initial pH, the ratio of ZnSO4:PS (m:m), and ZnO/ZnS@BC concentration on the kinetics of NOF degradation were investigated. Table 1 presents the values of the kinetic rate constants (k0) related to the various parameters and their regression coefficients R2. The photocatalytic degradation approximately followed the pseudo-first order kinetics (Table 1). The NOF degradation rate constant was highest (0.021 min−1) With a pH of 7, ZnSO4:PS (m:m) ratio of 1:1, and 25 mg ZnO/ZnS@BC.

### Effect of initial pH

The ZnO/ZnS content with ratios of 0.5:1, 1:1, 2:1, 3:1, and 4:1 (ZnSO4:PS (m:m)) in the system were investigated to show the effect of the quantity of ZnO/ZnS on the catalytic activity of ZnO/ZnS@BC. The ZnO/ZnS@BC with a 1:1 ratio had the highest removal capacity in the dark and under UV-light, and the NOF adsorption and degradation rates were 44% and 49%, respectively (Fig. 8). This result reveals that the correct mole ratio of ZnO and ZnS in the compound built more efficient heterojunction nanostructures, and the heterojunctions significantly enhanced the photocatalytic performance, perhaps because the adsorption site and the photocatalytic active site reached an optimal ratio. With the increase of ZnO/ZnS, the removal rate of NOF gradually decreases. The ZnO/ZnS@BC with a 4:1 ratio had the lowest activity with 42% elimination after 3 h. This could be because of the decrease of adsorption sites on the surface of the ZnO/ZnS@BC. Thus, more catalytic active sites would be retained, enhancing the degradation of NOF.

### Effect of different competing ions

The effect of common cations (such as Na+, Fe2+, Cu2+, and Zn2+) and common anions (such as SO42−, NO3, Cl, CO32−, and C6H5O73−) were investigated to study the effect of different ions on the removal of NOF by ZnO/ZnS@BC (results in Fig. 10). The concentration of the anions and cations were 0.1 mM, and the catalytic reaction time was 0.5 h. The influence of cations on the removal was significant (Fig. 10a). Fe2+ slightly promoted the removal effect (the promotion rate was 10% ), and Na+ had almost no impact (the inhibition rate was 2%). Other ions show an inhibitory effect. The inhibition rates of Cu2+ and Zn2+ were 99% and 62%, respectively. The primary reason was that the presence of cations forms a new clathrate with NOF, which was structurally stable and difficult to decompose. All the anions (except CO32−, which slightly improved the removal) acted as inhibitors (Fig. 10b). CO32− ameliorated the pH of the solution, which improved the removal rate of NOF. SO42−, NO3−, and Cl were formed associated with the consumption of the ·OH radical, which caused a decline in the degradation of the substances37. NO3− had the highest inhibition rate of 34%. Additionally, anions might change the electrostatic interaction between NOF and ZnO/ZnS@BC. Therefore, the degradation rate of NOF was different in various ion solutions. The results were consistent with other research39.

### Reusability of the ZnO/ZnS@BC catalyst

Stability was vital for the catalysts and was confirmed by repeating the decomposition processes five times (Fig. 11). After five replicates, the photodegradation efficiency of NOF decreased from 95 to 79%. The ZnO/ZnS@BC catalyst featured high stability and good reusability under ultraviolet light irradiation. The 16% reduction might be a result of the loss of catalyst quality during the recovery process.

Photogenerated h+, ·OH, and superoxide (·O2) were considered the primary active species responsible for the organic compound. In this research, tert-butanol, benzoquinone, and ammonium oxalate were used as scavengers of ·OH, ·O2, and h+, respectively40 (Fig. 12). Approximately 94% of the NOF was degraded within 3 h when there were no quenching agents in the system. However, there was an obvious decline of NOF photodegradation when benzoquinone was added—applied to quench ·O2 in the system, and NOF removal declined by approximately 30%. Thus, the ·O2 was generated in the photocatalytic reaction system when ZnO/ZnS@BC was irradiated by UV-light. This result also proved ·O2 was the predominant active species in the system. In the presence of ammonium oxalate, which quenched h+, the removal of NOF was attenuated, and the removal rate of NOF was significantly reduced by 15%, which implied that h+ was also an important active species for NOF removal. Similarly, tert-butanol (used as a scavenger to quench) suppressed the photocatalytic removal rate by 7%, showing that ∙OH had a slight removal capacity.

Based on the results, each scavenger had a different effect on the photocatalytic degradation of NOF. Benzoquinone had the greatest influence, and tert-butanol had the weakest effect. The decreasing order of contribution of each reactive species on NOF degradation by ZnO/ZnS@BC was as follows: ·O2 > ·OH > h+. This result was similar to research by Xian which used BaTiO3@g-C3N4 to degrade methyl orange under simulated sunlight irradiation41.