All the reagents (analytical grade) were purchased and used as received from Sinopharm Chemical Reagent Co., Ltd. Throughout the study, all solutions were prepared with ultrapure water (18.2 MΩ⋅cm).

4 mmol CH₃CSNH₂, 2.0 mmol In(NO₃)₃·H₂O, and 2.0 mmol Cu(NO₃)₃·3H₂O were dissolved in 40 ml anhydrous ethanol, and stirred with a magnetic stirrer for 60 min to generate a homogenous dark brown suspension. The solution was then transferred into a 100 ml Teflon-lined stainless steel autoclave and heated at 180°C for 2 h. After cooling to ambient temperature naturally, the precipitate was centrifuged, washed alternately with ethanol and water, and dried at 80°C for 24 h.

An X-ray diffractometer (XRD, PANalytical B.V.) was used to determine the phase of the as-obtained product. Field emission scanning electron microscope (FE-SEM, Japan Hitachi LTD. S4800) and transmission electron microscopy (TEM, Japanese electronics JEM-2100PLU) were used to characterize the products’ microstructure. The point of zero charge (pH_pzc) was measured by pH drift method. X–ray photo-electron spectroscopy (XPS) was performed on a PHI 5000C ESCA System. Using a UV–Vis spectrometer (UV-2450, Shimadzu, Japan) equipped with an integrating sphere attachment and BaSO₄ as a reflectance standard, UV–Vis diffuse reflectance spectrum of the dry-pressed disk sample was acquired.

The photocatalytic performance of the synthesized CuInS₂ was determined by monitoring the simultaneous photocatalytic reduction of Cr(VI) and degradation of CAP in aqueous solution under visible-light illumination. The photocatalytic experimental setup (PL-02, Beijing Precise Technology Co., Ltd.) contains a Xe arc lamp (1000 W) with a 400 nm cutoff filter, a set of cylindrical quartz reactors (80 ml), and a cold trap to keep the temperature of reaction solution constant. By dissolving K₂Cr₂O₇ and CAP in ultrapure water or diluting the stock solution with ultrapure water, varied concentrations of Cr(VI) and CAP solutions were obtained. The pH of the solution was adjusted to the anticipant value using a concentrated solution of NaOH or H₂SO₄. A set of tests were performed to investigate the photocatalytic Cr(VI) reduction and CAP degradation over CuInS₂ at various solution pHs and Cr(VI)/CAP ratios. Cr(VI) concentration was determined by the modified N, N-diphenylcarbazide spectrophotometry method (Li et al., 2021). The concentration of CAP in the solution was determined using high-resolution liquid chromatography (HPLC, Accela, Thermo Scientific, United States) equipped with an XB-C18 column (4.6250 mm, 5 m, Yuexu, China) and a UV detector. The mobile phase consisted of a 50: 50 mixture of acid aqueous solution (0.1% acetic acid) and acetonitrile. The chromatograms of CAP were achieved using a flow rate of 1.0 ml min⁻¹, an injection volume of 10 μl, and 277 nm as UV detection wavelength. The column was maintained at a temperature of 30°C. For each time measurement, approximately 4 ml of the aqueous solution was withdrawn from the cylindrical quartz reactors and filtered through a 0.45 nm filter to get rid of catalysts.

The photocatalytic Cr(VI) reduction and CAP degradation over CuInS₂ under a variety of operating conditions were studied by pseudo-first-order kinetics, which was expressed as Eq. 1 (Mao et al., 2022): In ( C 0 C ) = Kt (1)

Here, C₀ represents the concentration of Cr(VI) or CAP following the adsorption-desorption equilibrium, C represents the concentration of Cr(VI) or CAP at the irradiation time of t min, t represents the irradiation time (min), and k represents the apparent rate constant (min⁻¹).

XRD was used to determine the crystalline structure of our CuInS₂ product. As illustrated in Figure 1, there appear the diffraction peaks at 2θ = 28.1°, 32.2°, 47.1°, and 55.1°, which are respectively indexed to the (112), (200), (204), and (116) crystal planes of chalcopyrite structure CuInS₂ (JCPDS card No. 85–1575). The weak and wide diffraction peaks in the XRD pattern suggests that our CuInS₂ product has poor crystallization.

SEM and TEM were used to investigate the morphology and size of our produced CuInS₂. As illustrated in Figures 2A–C, the CuInS₂ sample exhibits a sheet stacking structure with dimensions ranging from 200 to 5,000 nm. The flake structure of CuInS₂ allows for the exposure of more active sites, which is highly advantageous for photocatalytic reactions. The EDS spectrum (Figure 2D) demonstrated that the prepared sample comprised the Cu, In, and S elements with a Cu, In, and S atomic ratio of around 1: 1: 2, confirming the formation of CuInS₂.

The whole XPS spectrum of our synthesized CuInS₂ is shown in Figure 3A, which reveals the existence of Cu, In, S 2p, and adventitious C in this sample. The Cu 2p XPS spectrum of our synthesized CuInS₂ is displayed in Figure 3B, which shows sharp peaks with binding energies of 951.48 eV (Cu 2p_1/2) and 931.58 eV (Cu 2p_3/2), respectively. There is no characteristic Cu²⁺ peak at 934.3 eV, indicating that only the Cu⁺ oxidation state is present in our synthesized CuInS₂ (Li et al., 2016; Yue et al., 2017). The In 3 days XPS spectrum in Figure 3C shows two peaks at 452.68 eV and 445.08 eV, which match with In 3d_3/2 and In 3d_5/2, respectively. These binding energy values indicate that In is in the In³⁺ oxidation state (Yang et al., 2015). The S 2p XPS spectrum (Figure 3D) was fitted into two peaks at 163.18 eV (S 2p_1/2) and 161.88 eV (S 2p_3/2), which are attributed to S^2- bonded to In or Cu in CuInS₂ (Liu et al., 2019; Guo et al., 2021).

The UV-Vis diffuse reflectance spectrum of our synthesized CuInS₂ was used to analyze its optical absorption property and band gap energy (E_g). As illustrated in Figure 4A, the CuInS₂ product demonstrates distinct absorption of visible light between 550 and 800 nm. The following formula (Eq. 2) can be used to estimate the E_g of the CuInS₂ product (Liu et al., 2022a): ( α h v ) n = K ( hv- Eg ) (2) where h is the Planck’s constant, α is the absorption coefficient, k is the constant, v is the light frequency, n = 1/2 for an indirect band gap semiconductor, or n = 2 for a direct band gap semiconductor. Since absorbance (A) is directly proportional to the absorbance coefficient (α), the same E_g value can be obtained by replacing α with A. CuInS₂ is a direct band gap semiconductor. As shown in Figure 4B, by projecting the linear part of its (Ahv)² vs. (hv) plot to zero, the E_g value of CuInS₂ is obtained to be 1.52 eV, which is close to the reported values in the literature (Guo et al., 2019; Chumha et al., 2020). This means that the CuInS₂ nanomaterial could be used as a visible-light-driven photocatalyst.

Since the solution pH can modify both the acid-base environment and the existing forms of Cr(VI), it is well established that the solution pH values exert a significant effect on photocatalytic activities (Zhang et al., 2018; Mangiri et al., 2021; Kumar et al., 2022). The effects of varying the initial pH on Cr(VI) reduction are demonstrated in Figures 5A,B According to Figure 5A, the adsorption equilibrium between catalysts and Cr(VI) can be attained after they were mixed for 40 min, because the Cr(VI) concentration did not decrease when the adsorption duration was further raised to 60 min. When the pH value decreased from 7–9 to 5–3, the Cr(VI) adsorption capacity of CuInS₂ increased rapidly. The pH_PZC of prepared CuInS₂ is determined to be 5.68, so in the solutions with pH less than 5.68, the surface charges of CuInS₂ are positive, which is conducive to the adsorption of Cr₂O₇ ²⁻, HCrO₄ ⁻ and CrO₄ ²⁻ (Xu et al., 2020). When the photoreduction process began, the reduction rates of Cr(VI) significantly increased with the lengthening of the irradiation period. In addition, the decrease of the solution pH from 9.01 to 3.00 also resulted in the faster reduction rates of Cr(VI). According to Figure 5B, when the solution pH value is 3.0, the k value (0.0108 min⁻¹) of the photocatalytic Cr(VI) reduction over CuInS₂ is the largest, which is about 15.43 times as that (0.0007 min⁻¹) at pH 9.01. One reason for this is that a lower pH promotes Cr(VI) adsorption on the photocatalyst, hence accelerating photocatalytic conversion. Another possibility is that lowering the pH value of reaction solution raises the chromate reduction potential. For example, one pH unit lower results in a rise in the standard reduction potential by 0.138 V (Zhang et al., 2017). So the E(Cr(VI)/Cr(III)) value increases from 0.24 to 1.06 V as the pH of the solution decreases from 9.00 to 3.00. From the viewpoint of both kinetics and thermodynamics, the photocatalytic Cr(VI) reduction rate would be enhanced in the lower pH solution (Marinho et al., 2016; Zhang et al., 2022).

As shown in Figure 6A, the pH value of the solution had little effect on the adsorption of CAP by CuInS₂. However, the photodegradation rates of CAP can be remarkably affected by the pH of the solution. As shown in Figure 6B, when the solution’s starting pH is 3.0, the CAP degradation rate is the fastest at the given irradiation period, and the k value is 2.9 times that of pH 9.01. The rate of CAP degradation reduced as the pH of the solution increased, indicating that the more acidic solution promotes photocatalytic CAP degradation in the mixed solution system of CAP and Cr(VI). One reason for this is that the formed Cr(OH)₃ tends to settle on the surface of the photocatalyst particles in neutral and alkaline solutions, reducing the available active sites of the photocatalysts and inhibiting further photocatalytic degradation of CAP and reduction of Cr(VI) (Wang et al., 2021a). The decrease in Cr(VI) conversion reduces the consumption of photogenerated electrons, causing more recombination of photogenerated electrons and holes and diminishing the photocatalytic oxidative degradation of CAP (Sun et al., 2021).

In the mixed solution of Cr(VI) and CAP, the possibility of simultaneous Cr(VI) reduction and CAP oxidation by the photocatalysis of CuInS₂ was tested. The effects of different initial CAP concentrations on the photocatalytic reduction of 10 mg/L Cr(VI) over CuInS₂ and the corresponding kinetic behaviors were studied. As illustrated in Figure 7A, without CuInS₂, the reduction of Cr(VI) in the Cr(VI)/CAP mixed solution under visible-light irradiation can be neglected. In the mere Cr(VI) solution (without CAP), the photocatalytic reduction of Cr(VI) by CuInS₂ under visible-light irradiation for 120 min removed only 77.5% of Cr(VI). When the CAP/Cr(VI) ratio was 0.5: 1, over 94.3% of Cr(VI) was decreased after 120 min of visible-light irradiation, implying that the presence of CAP might increase Cr(VI) reduction by serving as a photogenerated hole scavenger. Also, it can be observed that a variation in the CAP/Cr(VI) ratio can lead to a change in the Cr(VI) reduction rate. The optimal CAP/Cr(VI) ratio is 0.5: 1, and the k value is about 3.7 times that without CAP (Figure 7B). This might be attributed to that when CAP concentration rises, more CAP or intermediates would be adsorbed on the CuInS₂ surface, potentially covering the catalyst’s active sites (Cherifi et al., 2021).

Besides, the effects of varying the initial Cr(VI) concentration on the degradation of 10 mg/L CAP over CuInS₂ and the corresponding kinetic behaviors were examined. As shown in Figure 8A, CAP cannot be degraded without the presence of CuInS₂ catalyst, and the coexistence of Cr(VI) can effectively improve the degradation rate of CAP in the presence of CuInS₂ catalyst, similar to the effect of CAP on the reduction process of Cr(VI) in the mixed Cr(VI)/CAP solution under visible-light irradiation. The photodegradation kinetic behaviors of CAP over CuInS₂ in the solutions containing different ratios of CAP and Cr(VI) were further explored, and the results are presented in Figure 8B. The k values increase with decreasing the CAP/Cr(VI) ratio from 1: 1 to 1: 2 and subsequently decrease with further decreasing the CAP/Cr(VI) ratio from 1: 2 to 1: 2.5. The optimized CAP/Cr(VI) ratio is 1: 2, with a k value (0.0078 min⁻¹) for CAP degradation is about 1.5 times that of the mere CAP solution [without Cr(VI)]. The above results indicated that there is strong synergy between photocatalytic Cr(VI) reduction and CAP degradation over CuInS₂.

By carrying out the photocatalytic experiments with or without the addition of EDTA-Na₂ (the scavenger for photogenerated holes) as well as in air or N₂ environment, the mechanism of photocatalytic Cr(VI) conversion and CAP degradation over CuInS₂ was investigated. Before the start of the experiments under the N₂ environment, the reaction solution was purged with high-purity (> 99.999%) N₂ for 1 h to eliminate the dissolved O₂, and this process was maintained throughout the photocatalytic process. As shown in Figure 9, under visible-light illumination, the Cr(VI) conversion and CAP degradation over CuInS₂ in the Cr(VI)/CAP mixed solution in the air environment were more efficient than those in the N₂ environment. The suppressing impact of N₂ environment was more noticeable in the case of Cr(VI) reduction, with the Cr(VI) reduction rate dropping from 71.9% in air to 53.3% in N₂ environment. This is because that the interaction of dissolved O₂ with photogenerated electrons can produce superoxide radicals (•O₂) (Wang et al., 2016). •O₂ can be further converted to H₂O₂ or disproportioned to •OH, which affects the Cr(VI) reduction and CAP degradation, respectively. Furthermore, it has been shown that •O₂ is capable of reducing Cr(VI) to Cr(V), hence improving Cr(VI) conversion (Wei et al., 2016; Deng et al., 2017). As a consequence, the rates of Cr(VI) reduction and CAP oxidation in the air environment are higher than those in the N₂ environment. As shown in Figure 9, the conversion proportion of Cr(VI) and the degradation proportion of CAP are 71.9% and 49.6%, respectively, without the addition of EDTA-Na₂ in the air environment under visible-light irradiation for 120 min. After adding EDTA-Na₂, the Cr(VI) conversion rose to 91.5% while the CAP degradation decreased to 32.2%. Because EDTA-Na₂ can efficiently capture photogenerated holes, so enhancing photogenerated charge carrier separation and has a promoting influence on Cr(VI) reduction (Patnaik et al., 2018). On the other hand, the addition of EDTA-Na₂ led to the decline in CAP degradation efficiency, suggesting that CAP degradation was mostly dependent on photogenerated holes (Qu et al., 2020). In the N₂ environment, the improvement in the Cr(VI) conversion and the decrease in the CAP degradation were also observed after adding EDTA-Na₂. Nevertheless, when the reaction was carried out in the N₂ environment, EDTA-Na₂ had a smaller promoting impact on the reduction of Cr(VI), owing to the reduced of O₂/•O₂ ⁻ mediated reduction in the decrease of dissolved O₂ (Wang et al., 2016; Cherifi et al., 2021).

We postulated the possible mechanisms for the photocatalytic Cr(VI) reduction and CAP oxidation over CuInS₂ as shown in Figure 10, based on the aforesaid results. Under visible-light irradiation, photogenerated electrons (e ⁻ ) and photogenerated holes (h ⁺ ) are produced respectively in the conduction band (CB) and valence band (VB) of CuInS₂ (Eq. 3). Cr(VI) can be reduced by e ⁻ (Eq. 4), while CAP can be oxidized by h ⁺ (Eq. 5). The two simultaneous processes are capable of accelerating the separation of e ⁻ and h ⁺ , resulting in a greater amount of e ⁻ for Cr(VI) reduction and h ⁺ for CAP oxidation (Liu et al., 2022c). In addition, e ⁻ can combine with dissolved O₂ to form •O₂ ⁻ (Eq. 6), and •O₂ ⁻ can reduce Cr(VI) in the presence of H⁺ (Eq. 7) (Xia et al., 2018). Furthermore, •O₂ ⁻ can combine with H⁺ to make H₂O₂ (Eq. 8), which subsequently reacts with e ⁻ to form the powerful oxidizing •OH (Eq. 9). Meantime, CAP may be oxidized by h ⁺ as well as the oxidizing species created, such as •O₂ ⁻, •OH, and H₂O₂ (Eq. 10). CuIn S 2 + h v → CuIn S 2 ( h v B + + e C B − ) (3) Cr ( VI ) + 3 e C B − → Cr ( III ) (4) h V B + + CAP → Degradation   Products (5) O 2 + e C B − → • O 2 − (6) Cr ( VI ) + 2 • O 2 − + 4 H + → Cr ( III ) + 2 H 2 O+ O 2 (7) 2 • O 2 − + 2 H + → O 2 + H 2 O 2 (8) H 2 O 2 + e C B − → • OH+O H - (9) • O 2 - / H 2 O 2 / • ΟΗ+CAP → Degradation   products (10)

The photocatalytic activity and durability of a catalyst are equally significant in practical applications. The photocatalytic endurance of CuInS₂ was tested by performing five successive cycles of Cr(VI) reduction and CAP degradation in the mixed Cr(VI)/CAP solution by the same process as mentioned above, but 40 mg of photocatalyst and 200 ml of mixture were used. When each cycle ended, the photocatalyst was collected, washed and dried at 80°C for 12 h. In each cycle test, a certain amount of original Cr(VI)/CAP mixture was injected to maintain the initial concentration of pollutants. As indicated by Figure 11, both the Cr(VI) reduction rate and the CAP degradation rate decrease only a bit as the cycle number rises. The reduced percentage of Cr (VI) and the degraded percentage of CAP are in turn 71.9% and 49.6% in the first cycle, but still 68.7% and 46.2% in the fifth cycle, respectively. Thus, the CuInS₂ photocatalyst has been shown to have fair reusability for synchronous photocatalytic Cr(VI) conversion and CAP degradation. Figures 12A,B show the XRD patterns and survey XPS spectra of the CuInS₂ before and after the reuse tests. As can be seen from Figures 12A,B, the peak number and location of the CuInS₂ after the reuse tests are virtually identical to those of fresh CuInS₂, showing that the crystal structure, composition and elemental valence of CuInS₂ have little change. Accordingly, CuInS₂ appears to have strong stability and fair reusability, which bodes well for its future use in wastewater treatment.