Construction of S-Scheme CuS/Bi5O7I Heterojunction for Boosted Photocatalytic Disinfection with Visible Light Exposure

In this paper, a novel S-scheme CuS/Bi5O7I heterojunction was successfully constructed using a two-step approach comprising the alkaline hydrothermal method and the adsorption–deposition method, and it consisted of Bi5O7I microrods with CuS particles covering the surface. The photocatalytic antibacterial effects on Escherichia coli (E. coli) were systematically examined with visible light exposure. The results suggested that the 3%-CuS/Bi5O7I composite showed the optimal antibacterial activity, completely inactivating E. coli (5 × 108 cfu/mL) in 180 min of irradiation. Moreover, the bacterial inactivation process was scientifically described. •O2− and h+ were the major active species for the inactivation of the bacteria. In the early stages, SOD and CAT initiated the protection system to avoid the oxidative destruction of the active species. Unfortunately, the antioxidant protection system was overwhelmed thereafter, which led to the destruction of the cell membrane, as evidenced by the microstructure changes in E. coli cells. Subsequently, the leakage of intracellular components including K+, proteins, and DNA resulted in the unavoidable death of E. coli. Due to the construction of the S-scheme heterojunction, the CuS/Bi5O7I composite displayed the boosted visible light harvesting, the high-efficiency separation of photogenerated electrons and holes, and a great redox capacity, contributing to an outstanding photocatalytic disinfection performance. This work offers a new opportunity for S-scheme Bi5O7I-based heterojunctions with potential application in water disinfection.


Introduction
Drinking-water safety is a continuing global concern. Human health can be affected by pathogenic microorganisms in drinking water. Millions of people die every year from waterborne infectious diseases. Currently, chemical oxidation technology is used for drinkingwater treatment to inactivate microorganisms. Chemical oxidation disinfectants are mainly chlorine-containing preparations, such as bleaching powder, liquid chlorine, chloramine, and chlorine dioxide. Most chlorine-containing disinfectants can produce carcinogenic and mutagenic by-products during application. Ozone can also be applied to inactivate the microorganisms in drinking water, but it requires on-site preparation of expensive equipment. In 1985, Matsunaga et al. discovered that TiO 2 loaded with Pt could inactivate Escherichia coli (E. coli) with light exposure, which introduced photocatalytic technology to the antibacterial field [1]. In the last decade, photocatalytic disinfection has attracted great attention from researchers with regard to water treatment [2,3]. Visible light-responsive materials are hot spots in the photocatalytic field for utilizing solar energy efficiently.
Bi-based photocatalysts, such as Bi 2 WO 6 , BiVO 4 , Bi 2 MoO 6 , BiOBr, and BiOI, have been extensively reported with visible light-driven photocatalytic performance. In addition, Bi 5 O 7 I with a layered crystal structure similar to that of BiOI from the [Bi 2 O 2 ] 2+ and double I layers has been discovered with a band gap of about 2.6 eV as a type of bismuth-rich  [27]. All the peaks are consistent with the standard card of orthorhombic Bi5O7I (JCPDS 40-0548) [28].  [27]. All the peaks are consistent with the standard card of orthorhombic Bi 5 O 7 I (JCPDS 40-0548) [28].  [25]. After coupling Bi 5 O 7 I with CuS, the characteristic peaks of CuS cannot be identified in the 3%-CuS/Bi 5 O 7 I, as a result of the low amount of CuS.
The element composition of the 3%-CuS/Bi 5 O 7 I composite was demonstrated using XPS. In Figure S1 (see Supplementary Materials), the co-existence of Bi, O, I, Cu, and S elements can be detected in the survey spectra. In the Bi 4f high-resolution XPS spectrum in Figure 2b, two peaks at 159.6 and 164.9 eV are in line with the typical binding energies of Bi 4f 7/2 and Bi 4f 5/2 , respectively [29]. In Figure 2c, two symmetric peaks were created by fitting the asymmetric O 1s XPS spectrum. The peak at 530.4 eV corresponds to oxygen in the Bi-O of Bi 5 O 7 I, and the peak at 532.0 eV is ascribed to the absorbed -OH group or H 2 O [15,30]. The two peaks of I 3d at 619.5 and 631.0 eV (Figure 2d) are assigned to I 3d 5/2 and I 3d 3/2 , respectively [27]. Moreover, the corresponding peaks of Cu 2p 3/2 and Cu 2p 1/2 can be seen at 932.6 and 952.4 eV (Figure 2e), respectively, relative to the Cu(II) species in CuS [31]. A broad peak of S 2s at 225. 7 eV (Figure 2f) can be associated with the sulfide ions of CuS. In addition, the change in the electron density can be determined by examining the shift of the binding energy in XPS [32,33]. Compared with pure Bi 5 O 7 I, the Bi 4f, O 1s and I 3d peaks of the 3%-CuS/Bi 5 O 7 I composite shifted to a higher binding energy slightly, implying the decrease in the electron density. However, the binding energy of the S 2s and Cu 2p peaks in the 3%-CuS/Bi 5 O 7 I composite show a negative shift compared to pure CuS, demonstrating an increase in the electron density. The XPS results indicate that the electrons could migrate from Bi 5 O 7 I to CuS after the formation of the CuS/Bi 5 O 7 I heterojunction.
Molecules 2023, 28, x FOR PEER REVIEW 5 of 16 after the introduction of CuS, and this can be contributed to an improved photocatalytic disinfection performance.  As exhibited in Figure 3a, the optical characteristics of the materials were determined using a UV-Vis diffuse reflectance spectrometer. The absorption edge of Bi 5 O 7 I is at about 450 nm and the corresponding band gap energy is 2.56 eV, calculated using the Kubelka-Munk theory ( Figure 3b). As for pure CuS, the outstanding visible light absorption could be identified, and the corresponding band gap energy is 1.83 eV, which is displayed in Figure 3c. The absorption intensity of the CuS/Bi 5 O 7 I composite is enhanced and unlike pure Bi 5 O 7 I, a red-shift absorption edge could be observed. The results indicated that the CuS particles deposited on the Bi 5 O 7 I microrods are in favor of the visible light harvesting. The Mott-Schottky test was performed to study the semiconductor type, as well as the band-edge potentials, of Bi 5 O 7 I and CuS. The M-S plot of Bi 5 O 7 I with a positive slope of the linear portion can be seen in Figure 3d, indicating that Bi 5 O 7 I is an n-type semiconductor. As for CuS in Figure 3e, it is a p-type semiconductor as a result of the negative slope in the linear portion. The flat band potentials (E fb ) of Bi 5 O 7 I and CuS are valued to be −0.92 and 1.67 eV (vs. Ag/AgCl), respectively. Based on E NHE = E Ag/AgCl + 0.197 [18], the E fb values of Bi 5 O 7 I and CuS can be transformed to be −0.72 and 1.87 eV (vs. NHE), respectively. The CB potential (E CB ) of the n-type semiconductor is 0.1 eV more negative than E fb [34,35], so the E CB value of Bi 5 O 7 I is −0.82 eV. For p-type semiconductor, the VB potential (E VB ) is 0.1 eV more positive than E fb , and the E VB value of CuS is 1.97 eV [36]. On account of E CB = E VB − E g , the E VB value of Bi 5 O 7 I and the E CB value of CuS are 1.74 and 0.14 eV, respectively.
The separation efficiency of the photoinduced carriers could be evaluated by examining the PL spectra, transient photocurrent response and EIS. Generally, lower photoinduced carrier recombination rate leads to a decreased PL emission intensity. As depicted in Figure 4a, pure Bi 5 O 7 I exhibited a remarkable fluorescence emission from 350 nm to 600 nm. For the CuS/Bi 5 O 7 I composites, the PL intensity was diminished, suggesting that the recombination of the photoinduced carriers was inhibited. Furthermore, it was noticeable that the PL intensity of 3%-CuS/Bi 5 O 7 I was the lowest. As illustrated in Figure 4b, the photocurrent intensity of 3%-CuS/Bi 5 O 7 I is about 0.28 µA/cm 2 , which is higher than that of Bi 5 O 7 I, suggesting that efficient electron-hole separation was achieved after the introduction of CuS. In addition, 3%-CuS/Bi 5 O 7 I exhibited a smaller arc radius than Bi 5 O 7 I (Figure 4c), which is helpful for transmitting and separating photogenerated carriers. In summary, the CuS/Bi 5 O 7 I composite could separate the photoinduced carriers with a high efficiency after the introduction of CuS, and this can be contributed to an improved photocatalytic disinfection performance.

Photocatalytic Disinfection Activity
The antibacterial performances were determined by observing the inactivation of E. coli with visible light exposure. As presented in Figure 5a, a light control experiment without photocatalysts was conducted to exclude the impact of light on E. coli. It was apparent that the survival rate of the bacteria did not noticeably reduce after the light control experiment, indicating that the effect of the visible light on the E. coli was insignificant. When the E. coli solution and photocatalysts were mixed and exposed to visible light, all the synthesized samples exhibited antibacterial performance. In particular, the antibacterial performances of the CuS/Bi5O7I composites were superior to those of CuS and Bi5O7I. Furthermore, the loading amounts of CuS had an impact on the photocatalytic antibacterial rates of the CuS/Bi5O7I composites. The photocatalytic antibacterial performance first boosted with an increase in CuS in the CuS/Bi5O7I composites until the mass ratio reached 3 wt%. However, it gradually declined with a further increase in the CuS amount, demonstrating the critical role of CuS in adjusting the antibacterial performance of CuS/Bi5O7I composites. The results of the antibacterial activities in the dark are displayed in Figure  5b. The antibacterial performances of the synthesized samples in the dark were

Photocatalytic Disinfection Activity
The antibacterial performances were determined by observing the inactivation of E. coli with visible light exposure. As presented in Figure 5a, a light control experiment without photocatalysts was conducted to exclude the impact of light on E. coli. It was apparent that the survival rate of the bacteria did not noticeably reduce after the light control experiment, indicating that the effect of the visible light on the E. coli was insignificant. When the E. coli solution and photocatalysts were mixed and exposed to visible light, all the synthesized samples exhibited antibacterial performance. In particular, the antibacterial performances of the CuS/Bi 5 O 7 I composites were superior to those of CuS and Bi 5 O 7 I. Furthermore, the loading amounts of CuS had an impact on the photocatalytic antibacterial rates of the CuS/Bi 5 O 7 I composites. The photocatalytic antibacterial performance first boosted with an increase in CuS in the CuS/Bi 5 O 7 I composites until the mass ratio reached 3 wt%. However, it gradually declined with a further increase in the CuS amount, demonstrating the critical role of CuS in adjusting the antibacterial performance of CuS/Bi 5 O 7 I composites. The results of the antibacterial activities in the dark are displayed in Figure 5b. The antibacterial performances of the synthesized samples in the dark were inadequate compared to the performances of those exposed to visible light, demonstrating the production of synergistic effects with photocatalysts and visible light on the inactivation of bacteria.
As detected in Figure 5a, the 3%-CuS/Bi 5 O 7 I composite displayed the optimal performance and could inactivate all the E. coli within 180 min. To further evaluate the inactivation procedure of E. coli with the 3%-CuS/Bi 5 O 7 I composite, fluorescence microscopy was performed with stained E. coli. The E. coli was stained with PI and SYTO9 simultaneously. PI can penetrate through the broken cytomembranes and interact with DNA to label dead cells with red fluorescence. SYTO9 can be applied to label live cells with a green fluorescence because it penetrates intact cytomembranes and interacts with DNA [37]. According to Figure 5c, all the cells showed green fluorescence before the light irradiation, suggesting the presence of live bacteria with an intact cytomembranes. After prolonging the light exposure, the green fluorescence reduced progressively, while the red fluorescence increased. After 180 min of visible light exposure, only red fluorescence could be observed, implying that all the cytomembranes were damaged. cells with red fluorescence. SYTO9 can be applied to label live cells with a green fluorescence because it penetrates intact cytomembranes and interacts with DNA [37]. According to Figure 5c, all the cells showed green fluorescence before the light irradiation, suggesting the presence of live bacteria with an intact cytomembranes. After prolonging the light exposure, the green fluorescence reduced progressively, while the red fluorescence increased. After 180 min of visible light exposure, only red fluorescence could be observed, implying that all the cytomembranes were damaged.

Investigation of Active Species
The experimental trapping of active species was carried out in the photocatalytic disinfection process with the 3%-CuS/Bi5O7I. Ammonium oxalate (AO), isopropanol (IPA) and p-benzoquinone (BQ) were designated as the h + , •OH and •O2 − scavengers, respectively. First, the toxic effects of the scavengers on the E. coli cells were explored. As can be seen in Figure S2, there was no discernible decline in the survival rate of E. coli, suggesting that the toxic effects of scavengers could be negligible. It is noticeable in Figure 6a that the  The experimental trapping of active species was carried out in the photocatalytic disinfection process with the 3%-CuS/Bi 5 O 7 I. Ammonium oxalate (AO), isopropanol (IPA) and p-benzoquinone (BQ) were designated as the h + , •OH and •O 2 − scavengers, respectively. First, the toxic effects of the scavengers on the E. coli cells were explored. As can be seen in Figure S2, there was no discernible decline in the survival rate of E. coli, suggesting that the toxic effects of scavengers could be negligible. It is noticeable in Figure 6a that the photocatalytic antibacterial activities of 3%-CuS/Bi 5  CuS/Bi5O7I. To further explore the production of •O2 − by Bi5O7I and 3%-CuS/Bi5O7I, nitroblue tetrazolium (NBT) was taken as an •O2 − probe agent. It is widely acknowledged that •O2 − can react with NBT to produce formazan, implying that a decline in the representative absorption peak intensity of NBT at 259 nm can be observed with the generation of •O2 − [38,39]. According to Figure 6b,c, the peak intensity of NBT for 3%-CuS/Bi5O7I decreased more notably than that for Bi5O7I, suggesting that more •O2 − could be produced by 3%-CuS/Bi5O7I.

Activities of the Antioxidant Enzymes
CAT and SOD are two valued antioxidant enzymes in the E. coli cells, and they can protect the cell from the stress of active species by transforming them into water and oxygen [40,41]. As illustrated in Figure 7a,b, the activities of CAT and SOD with the 3%-CuS/Bi5O7I increased within the first 60 min of irradiation, indicating a defensive behavior whereby the oxidative damage of active species was resisted. However, reductions in SOD and CAT activities were detected because the overwhelmed defense capabilities of SOD and CAT were overwhelmed by the excessive attack and continuous accumulation of active groups. The collapse of the antioxidant protection system was accountable for the subsequent oxidative injury to the membrane of the E. coli cells.

Microstructure Changes of E. coli Cell
An evaluation of microstructure changes is helpful to understand the apoptosis process of E. coli. SEM was selected to reveal the microstructure changes in the E. coli cells

Activities of the Antioxidant Enzymes
CAT and SOD are two valued antioxidant enzymes in the E. coli cells, and they can protect the cell from the stress of active species by transforming them into water and oxygen [40,41]. As illustrated in Figure 7a,b, the activities of CAT and SOD with the 3%-CuS/Bi 5 O 7 I increased within the first 60 min of irradiation, indicating a defensive behavior whereby the oxidative damage of active species was resisted. However, reductions in SOD and CAT activities were detected because the overwhelmed defense capabilities of SOD and CAT were overwhelmed by the excessive attack and continuous accumulation of active groups. The collapse of the antioxidant protection system was accountable for the subsequent oxidative injury to the membrane of the E. coli cells.
CuS/Bi5O7I. To further explore the production of •O2 by Bi5O7I and 3%-CuS/Bi5O7I, nitroblue tetrazolium (NBT) was taken as an •O2 − probe agent. It is widely acknowledged that •O2 − can react with NBT to produce formazan, implying that a decline in the representative absorption peak intensity of NBT at 259 nm can be observed with the generation of •O2 − [38,39]. According to Figure 6b,c, the peak intensity of NBT for 3%-CuS/Bi5O7I decreased more notably than that for Bi5O7I, suggesting that more •O2 − could be produced by 3%-CuS/Bi5O7I.

Activities of the Antioxidant Enzymes
CAT and SOD are two valued antioxidant enzymes in the E. coli cells, and they can protect the cell from the stress of active species by transforming them into water and oxygen [40,41]. As illustrated in Figure 7a,b, the activities of CAT and SOD with the 3%-CuS/Bi5O7I increased within the first 60 min of irradiation, indicating a defensive behavior whereby the oxidative damage of active species was resisted. However, reductions in SOD and CAT activities were detected because the overwhelmed defense capabilities of SOD and CAT were overwhelmed by the excessive attack and continuous accumulation of active groups. The collapse of the antioxidant protection system was accountable for the subsequent oxidative injury to the membrane of the E. coli cells.

Microstructure Changes of E. coli Cell
An evaluation of microstructure changes is helpful to understand the apoptosis process of E. coli. SEM was selected to reveal the microstructure changes in the E. coli cells

Microstructure Changes of E. coli Cell
An evaluation of microstructure changes is helpful to understand the apoptosis process of E. coli. SEM was selected to reveal the microstructure changes in the E. coli cells treated with the 3%-CuS/Bi 5 O 7 I and visible light exposure (Figure 8a). The untreated E. coli cells displayed a regular short rod-like shape with blunt rounded ends. After irradiation for 60 min, some E. coli cells exhibited surface depressions (yellow arrows) due to the damage of their cytomembranes and the leakage of cellular components. As the irradiation time increased, more damaged E. coli cells could be detected and the degree of surface depressions increased. Finally, all E. coli cells were destroyed and the collapsed cells were liable to adhesion and aggregation (green box). treated with the 3%-CuS/Bi5O7I and visible light exposure (Figure 8a). The untreated E. coli cells displayed a regular short rod-like shape with blunt rounded ends. After irradiation for 60 min, some E. coli cells exhibited surface depressions (yellow arrows) due to the damage of their cytomembranes and the leakage of cellular components. As the irradiation time increased, more damaged E. coli cells could be detected and the degree of surface depressions increased. Finally, all E. coli cells were destroyed and the collapsed cells were liable to adhesion and aggregation (green box).

Leakage of Intracellular Components
The determination of the cell membrane permeability can contribute to the further understanding of the inactivation process of bacteria. When the active species destroyed the cell membrane, the intracellular components could be leaked into the external environment [42,43]. K + , proteins, and DNA were selected as the representatives of the intracellular components in order to analyze the bacterial cell membrane permeability. It is vitally important for K + to maintain cell osmotic pressure, assist with the synthesis of proteins, and balance alkalinity [41]. In Figure 8b, the leakage of K + progressively increased with the increase in the irradiation time. Furthermore, the extracellular K + concentration with the 3%-CuS/Bi5O7I heterojunction (1.17 mg/mL) was considerably greater than that with CuS (0.33 mg/mL) or Bi5O7I (0.60 mg/mL) after 120 min of irradiation. Similar results

Leakage of Intracellular Components
The determination of the cell membrane permeability can contribute to the further understanding of the inactivation process of bacteria. When the active species destroyed the cell membrane, the intracellular components could be leaked into the external environment [42,43]. K + , proteins, and DNA were selected as the representatives of the intracellular components in order to analyze the bacterial cell membrane permeability. It is vitally important for K + to maintain cell osmotic pressure, assist with the synthesis of proteins, and balance alkalinity [41]. In Figure 8b, the leakage of K + progressively increased with the increase in the irradiation time. Furthermore, the extracellular K + concentration with the 3%-CuS/Bi 5 O 7 I heterojunction (1.17 mg/mL) was considerably greater than that with CuS (0.33 mg/mL) or Bi 5 O 7 I (0.60 mg/mL) after 120 min of irradiation. Similar results were obtained for the extracellular protein content (Figure 8c). Typically, proteins can recover and regrow after the injury via bacterial repair mechanisms [44]. Therefore, this is lethal to the bacteria with the injury and a loss of the nucleic acid [45]. Figure 8d proves that DNA was increasingly released during the photocatalytic disinfection process and the extracellular DNA content with the 3%-CuS/Bi 5 O 7 I heterojunction (38.5 ng/mL) was greater than that with CuS (16.2 ng/mL) or Bi 5 O 7 I (19.5 ng/mL) under illumination for 120 min. The results indicated a change in the cell membrane permeability and the leakage of intracellular components. Furthermore, the extent of the damage for cell membrane with 3%-CuS/Bi 5 O 7 I was more severe than that with CuS or Bi 5 O 7 I. Figure 8e presents an illustration of the proposed inactivation process of E. coli. First, the cell was attacked by the photoinduced active species and the antioxidant enzymes, including SOD and CAT, initiated the protection system to avoid the oxidative damage. With the increasing oxidative stress, the antioxidant protection system became overwhelmed. Then, the accumulated active species destroyed the cell membrane. Eventually, apoptosis was triggered by the release of internal components through the ruptured cell membrane.

Mechanism of Improved Photocatalytic Activity with CuS/Bi 5 O 7 I Heterojunction
The mechanism of the improved photocatalytic effects of the CuS/Bi 5 O 7 I heterojunction on E. coli inactivation was suggested, and it is depicted in Figure 9.  [36,46]. According to Figure 9a, the E F level of Bi 5 O 7 I is higher than that of CuS. When Bi 5 O 7 I was combined with CuS, the electrons migrated from Bi 5 O 7 I with a higher E F to CuS with a smaller E F , resulting in the formation of IEF [36,47], which can be seen in Figure 9b. The IEF can cause the accumulation or depletion of free charge carriers near the surface and result in the foundation of a well-developed heterojunction interface. The energy-band edges of Bi 5 O 7 I are bent upward toward the interface continuously, and those of CuS are bent downward towards the interface [48,49]. Eventually, the E F levels of CuS and Bi 5 O 7 I are aligned at the same level.
With visible light exposure, Bi 5 O 7 I and CuS can be excited simultaneously. For both semiconductors, photogenerated e − transferred from VB to CB, leaving h + on VB. As shown in Figure S3, the CB potential of Bi 5 O 7 I is (−0.82 eV) is more negative than that of CuS (0.14 eV). As a result, e − in the CB of Bi 5 O 7 I could transfer to that of CuS. Comparably, the h + in the VB of CuS could shift to that of Bi 5 O 7 I. Nevertheless, it is impossible for the gathered e − in the CB of CuS to react with O 2 and generate •O 2 − due to the CB potential of CuS (0.14 eV) being more positive than that of O 2 /•O 2 − (−0.33 eV vs. NHE) [50]. Accordingly, the typical type II mechanism of charge transfer is obviously contrary to the results of the active species, and it is unsuitable for the CuS/Bi 5 O 7 I heterojunction. Based on the aforementioned analysis, an S-scheme heterojunction structure of CuS/Bi 5 O 7 I was suggested and the conceivable mechanism is depicted in Figure 9c. The downward-bending band permits e − to flow out easily while inhibiting h + . In contrast, h + can move along the upwardbending band, while e − cannot [51]. Due to the cooperative action of Coulomb attraction, band bending and IEF, the e − from the CB of CuS may facilely recombine with the h + from the VB of Bi 5 O 7 I [22,52,53]. The CB potential of Bi 5 O 7 I (−0.82 eV) is more negative than that of O 2 /•O 2 − (−0.33 eV vs. NHE), and the remaining e − is competent for reducing O 2 to •O 2 − . In the current study, the generated •O 2 − and the h + from the VB of CuS destroyed the E. coli cell membrane, leading to the release of cellular components. In this manner, the efficient separation of photoinduced carriers and an enhanced redox capacity can be achieved, which contribute notably improving the photocatalytic disinfection performance.

Preparation of Photocatalysts
The details of the used chemicals and reagents can be found in the supporting information. The alkaline hydrothermal method was applied to prepare Bi5O7I. As is typical, 2 mmol Bi(NO3)3·5H2O and 2 mmol KI were dispersed in 30 mL distilled water by stirring, respectively. The KI solution was then added drop by drop to the Bi(NO3)3 solution. After continuous stirring for 1 h, the pH value of the above solution was adjusted to 13 by adding 3 mol L −1 NaOH. Next, the solution was poured into an autoclave and heated at 160 °C for 12 h. After washing with ethanol and distilled water, a white precipitate was obtained. Lastly, the product was dried and collected.
The adsorption-deposition method was adopted to synthesize CuS/Bi5O7I composites. First, a solution of 500 mg Bi5O7I was prepared in 120 mL deionized water, and the desired amounts of 0.1 mol L −1 Cu(CH3COO)2 solution were added dropwise. The mixture was sonicated and stirred for 1 h. Subsequently, 0.1 mol L −1 K2S solution with the same volume of Cu(CH3COO)2 solution was mixed with the above mixture under constant stirring for 2 h. Eventually, the products were washed and dried. Based on the above process, the composites were fabricated and marked as x-CuS/Bi5O7I, with x representing the mass ratio of CuS to Bi5O7I (0.5%, 1%, 3%, and 5%). Correspondingly, pure CuS was prepared

Preparation of Photocatalysts
The details of the used chemicals and reagents can be found in the supporting information. The alkaline hydrothermal method was applied to prepare Bi 5 O 7 I. As is typical, 2 mmol Bi(NO 3 ) 3 ·5H 2 O and 2 mmol KI were dispersed in 30 mL distilled water by stirring, respectively. The KI solution was then added drop by drop to the Bi(NO 3 ) 3 solution. After continuous stirring for 1 h, the pH value of the above solution was adjusted to 13 by adding 3 mol L −1 NaOH. Next, the solution was poured into an autoclave and heated at 160 • C for 12 h. After washing with ethanol and distilled water, a white precipitate was obtained. Lastly, the product was dried and collected.
The adsorption-deposition method was adopted to synthesize CuS/Bi 5 O 7 I composites. First, a solution of 500 mg Bi 5 O 7 I was prepared in 120 mL deionized water, and the desired amounts of 0.1 mol L −1 Cu(CH 3 COO) 2 solution were added dropwise. The mixture was sonicated and stirred for 1 h. Subsequently, 0.1 mol L −1 K 2 S solution with the same volume of Cu(CH 3 COO) 2 solution was mixed with the above mixture under constant stirring for 2 h. Eventually, the products were washed and dried. Based on the above process, the composites were fabricated and marked as x-CuS/Bi 5 O 7 I, with x representing the mass ratio of CuS to Bi 5 O 7 I (0.5%, 1%, 3%, and 5%). Correspondingly, pure CuS was prepared correspondingly without the addition of Bi 5 O 7 I. A schematic diagram of the preparation of the CuS/Bi 5 O 7 I composites is exhibited in Figure 10.
Molecules 2023, 28, x FOR PEER REVIEW 12 of 16 correspondingly without the addition of Bi5O7I. A schematic diagram of the preparation of the CuS/Bi5O7I composites is exhibited in Figure 10.

Characterization
The characterization of the synthesized materials is clarified in the supporting information.

Photoelectrochemical Measurement
Electrochemical tests, such as transient photocurrent response, electrochemical impedance spectroscopy (EIS), and Mott-Schottky (M-S) plots, were studied using a threeelectrode system on a CHI760E electrochemical workstation. The selected electrodes and measurement parameters are described in the supporting information.

Determination of Photocatalytic Disinfection Performance
To evaluate the photocatalytic disinfection performance of the as-prepared materials, E. coli (ATCC 8739) was taken as the model bacterium. All microbiological-related apparatuses and operations should be kept sterile for the bactericidal experiments. A single E. coli colony was taken from a flat plate and inoculated in a Luria-Bertani (LB) liquid medium to culture overnight at 37 °C. Next, 1% of the inoculation amount was transferred to a fresh LB liquid medium and cultured at 37 °C for 3 h. After centrifugation at 8000 rpm for 5 min, the bacteria were collected and dispersed in saline. Typically, the photocatalytic disinfection investigation was conducted with a 100 mL E. coli solution (5 × 10 8 cfu/mL) treated by adding 20 mg photocatalyst. As a light source, a 300 W Xe lamp with a 420 nm cutoff filter (CELHXF300-T3, Beijing Zhongjiaojinyuan Co., Ltd., Beijing, China) was utilized. During the bactericidal reactions, 5 mL suspension was withdrawn and centrifugated at different intervals. The supernatant was further used for the determination of extracellular K + , DNA, and protein concentration. The precipitate was washed three times with a PBS buffer solution (0.1 mol L −1 , pH 7.4) and suspended in a 10 mL PBS buffer solution. The above solution (1 mL) was diluted with sterilized saline and incubated onto an agar plate at 37 °C for 48 h. The plate count method was applied to examine the E. coli cell density at different irradiation time. A certain amount of E. coli solution was diluted with a gradient of 10 −1 and coated on a flat plate containing an LB solid medium, with incubation at 37 °C for 24 h. The number of colonies on the plate was counted. Laser scanning confocal microscopy (CLSM) and SEM were performed to detect the dead/live E. coli

Characterization
The characterization of the synthesized materials is clarified in the supporting information.

Photoelectrochemical Measurement
Electrochemical tests, such as transient photocurrent response, electrochemical impedance spectroscopy (EIS), and Mott-Schottky (M-S) plots, were studied using a three-electrode system on a CHI760E electrochemical workstation. The selected electrodes and measurement parameters are described in the supporting information.

Determination of Photocatalytic Disinfection Performance
To evaluate the photocatalytic disinfection performance of the as-prepared materials, E. coli (ATCC 8739) was taken as the model bacterium. All microbiological-related apparatuses and operations should be kept sterile for the bactericidal experiments. A single E. coli colony was taken from a flat plate and inoculated in a Luria-Bertani (LB) liquid medium to culture overnight at 37 • C. Next, 1% of the inoculation amount was transferred to a fresh LB liquid medium and cultured at 37 • C for 3 h. After centrifugation at 8000 rpm for 5 min, the bacteria were collected and dispersed in saline. Typically, the photocatalytic disinfection investigation was conducted with a 100 mL E. coli solution (5 × 10 8 cfu/mL) treated by adding 20 mg photocatalyst. As a light source, a 300 W Xe lamp with a 420 nm cutoff filter (CELHXF300-T3, Beijing Zhongjiaojinyuan Co., Ltd., Beijing, China) was utilized. During the bactericidal reactions, 5 mL suspension was withdrawn and centrifugated at different intervals. The supernatant was further used for the determination of extracellular K + , DNA, and protein concentration. The precipitate was washed three times with a PBS buffer solution (0.1 mol L −1 , pH 7.4) and suspended in a 10 mL PBS buffer solution. The above solution (1 mL) was diluted with sterilized saline and incubated onto an agar plate at 37 • C for 48 h. The plate count method was applied to examine the E. coli cell density at different irradiation time. A certain amount of E. coli solution was diluted with a gradient of 10 −1 and coated on a flat plate containing an LB solid medium, with incubation at 37 • C for 24 h. The number of colonies on the plate was counted. Laser scanning confocal microscopy (CLSM) and SEM were performed to detect the dead/live E. coli cells and the cell morphology. Furthermore, the superoxide dismutase (SOD) and catalase (CAT) activities of E. coli were measured.

Fluorescence Microscopy Assays of Live/Dead E. coli
To distinguish live and dead cells, SYTO9 and propidium iodide (PI) were chosen as the staining agents. The PI solution (5 µg/mL) and SYTO9 solution (5 µg/mL) were mixed with an equal volume. The above PI/SYTO9 solution was added into the PBS buffer solution with E. coli prepared according to the method described in Section 2.4 and the reaction was conducted in the darkness for 10 min. After centrifugation, the stained E. coli was washed with PBS and observed using CLSM.

Microstructure Observation of E. coli
After washing with PBS in the suspension described in Section 2.4, the E. coli cells were fixed with a 2.5% (v/v) glutaraldehyde solution at 4 • C for 6 h. Subsequently, the E. coli cells were washed with the PBS buffer solution and gradually dehydrated with an ethanol solution (30%, 50%, 70%, 90%, and 100%) for 10 min each time and with tert-butanol for 20 min. After supercritical drying, the microstructure of the E. coli cells was examined using SEM.

Antioxidant Enzymes Assay
The supernatants at various intervals as described in Section 2.4 were employed for CAT and SOD activity assays. The CAT assay kit (A007-1, Jiancheng Biotech, Nanjing, China) and the SOD assay kit (A001-1, Jiancheng Biotech, Nanjing, China) were used according to the instructions.

Leakage of Intracellular Components
The supernatant obtained from the suspension at different irradiation time as described in Section 2.4 was used to study the leakage of K + , proteins, and DNA. ICP-OES (5110VDV, Agilent Technologies Co. Ltd., Palo Alto, CA, USA) was operated to detect the K + released from the E. coli cells. Extracellular protein content was evaluated by BCA protein quantification kit (PA115, Tiangen Biochemical Technology Co. Ltd., Beijing, China). Extracellular DNA content was determined using a NanoDrop One at 260 nm.

Conclusions
This work successfully constructed a novel S-scheme CuS/Bi 5 O 7 I heterojunction successfully using a two-step approach comprising the alkaline hydrothermal method and the adsorption-deposition method. Overall, 3%-CuS/Bi 5 O 7 I composite exhibited the optimal disinfection of E. coli with visible light illumination, completely inactivating E. coli (5 × 10 8 cfu/mL) within 180 min. The active species, including •O 2 − and h + , destroyed the cell membranes of the bacteria. Cellular components, such as K + , protein, and DNA, were released as the permeability of the cell membranes changed. The S-scheme transfer pathway for the photoinduced electrons and holes across the CuS/Bi 5 O 7 I heterojunction was revealed. The high separation efficiency of photoinduced carriers was achieved, as well as an enhanced redox capacity, which contributed to improved photocatalytic disinfection activity. This work offers fresh insights into the design of S-scheme Bi 5 O 7 Ibased heterojunctions for the inactivation of bacteria utilizing inexhaustible solar energy.
Supplementary Materials: The following supporting information can be downloaded at: https:// www.mdpi.com/article/10.3390/molecules28073084/s1, Figure S1: XPS survey spectrum of 3%-CuS/Bi 5 O 7 I; Figure S2: Photocatalytic disinfection efficiency of E. coli with different scavengers in the dark; Figure S3: Type II charge transfer mechanism of CuS/Bi 5 O 7 I composite against E. coli under visible light irradiation; Table S1: Analysis results of EDS.

Institutional Review Board Statement: Not applicable.
Informed Consent Statement: Not applicable.

Data Availability Statement:
The data of the study can be provided by corresponding author upon reasonable request.