CdIn2S4/In(OH)3/NiCr-LDH Multi-Interface Heterostructure Photocatalyst for Enhanced Photocatalytic H2 Evolution and Cr(VI) Reduction

The development of highly active and stable photocatalysts, an effective way to remediate environment pollution and alleviate energy shortages, remains a challenging issue. In this work, a CdIn2S4/In(OH)3 nanocomposite was deposited in-situ on NiCr-LDH nanosheets by a simple hydrothermal method, and the obtained CdIn2S4/In(OH)3/NiCr-LDH heterostructure photocatalysts with multiple intimate-contact interfaces exhibited better photocatalytic activity. The photocatalytic H2 evolution rate of CdIn2S4/In(OH)3/NiCr-LDH increased to 10.9 and 58.7 times that of the counterparts CdIn2S4 and NiCr-LDH, respectively. Moreover, the photocatalytic removal efficiency of Cr(VI) increased from 6% for NiCr-LDH and 75% for CdIn2S4 to 97% for CdIn2S4/In(OH)3/NiCr-LDH. The enhanced photocatalytic performance was attributed to the formation of multi-interfaces with strong interfacial interactions and staggered band alignments, which offered multiple pathways for carrier migration, thus promoting the separation efficiency of photo-excited electrons and holes. This study demonstrates a facile method to fabricate inexpensive and efficient heterostructure photocatalysts for solving environmental problems.


Introduction
Against the background of the fossil energy shortage and water pollution by heavy metals and organic contaminants, it is essential to develop new technologies for environmental remediation and the production of green and renewable energy. Photocatalysis has drawn significant interest because of its ability to produce chemical energy, decompose organic pollutants, and detoxify heavy metal contamination by utilizing solar energy [1][2][3][4][5]. The performance of this environmentally-friendly technique mainly relies on the ability of photo-excited electrons (e − ) and holes (h + ) in a semiconductor photocatalyst to migrate to the surface and participate in a series of photocatalytic redox reactions. For instance, photo-generated electrons in a semiconductor with negative enough conduction band (CB) potential can react with H 2 O to produce H 2 , a promising substitute for traditional fossil fuels [1,2,6]. Hexavalent chromium ions (Cr(VI)), a serious threat to human health due to their solubility, acute toxicity and high carcinogenicity, can also be reduced to the nontoxic Cr(III) by photo-excited electrons [7,8].
To develop highly efficient and cost-effective photocatalysts, various semiconductors have been investigated during the past few decades, including metal oxides, metal sulfides, (oxy)nitrides, etc. [9][10][11]. Unfortunately, no satisfactory semiconductor photocatalyst has so far been obtained for practical applications. One of the bottleneck problems restricting the performance of single component semiconductor photocatalyst is the swift recombination of photo-generated e − and h + pairs. Fabrication of heterostructure photocatalysts is a promising strategy since this method makes it feasible to spatially separate photo-excited electrons and holes, and thus achieve higher photocatalytic efficiency than the component semiconductors [12][13][14][15][16][17][18][19][20]. For example, Jiang and coworkers fabricated CsPbBr 3 /Bi 2 WO 3 heterostructure photocatalysts by a simple electronic self-assembly process, and successfully improved the CO 2 photoreduction activity to more than 12-fold that of pristine CsPbBr 3 [21]. Han et al. constructed ZnIn 2 S 4 /BiVO 4 photocatalysts by assembling ZnIn 2 S 4 nanosheets on BiVO 4 decahedrons, and increased the photocatalytic CO yield by 9 times in comparison with pristine ZnIn 2 S 4 [22].
Meanwhile, charge-balancing anions A n− and water molecules are sandwiched between these positively charged 2D hydroxide layers. The band gap energy (E g ) and conduction band/valence band potentials can be modulated by varying M 2+ and M 3+ , and the charge density of 2D hydroxide layers depends on the atomic content of M 3+ (0.2 ≤ x ≤ 0.33). LDHs seem to be a good candidate for the development of heterostructure photocatalytic systems, and various LDH-involved composite photocatalysts with enhanced performance have been reported, such as MoS 2 /CoAl-LDH [28], CdS/NiCo-LDH [29], MoS 2 /NiFe-LDH [30], Cu 2 O/NiFe-LDH [31], and g-C 3 N 4 /ZnAl-LDH [32].
Inspired by the above benefits, we attempt in this work to construct a CdIn 2 S 4 /In(OH) 3 /NiCr-LDH ternary heterostructure photocatalyst. Considering the fact that CdIn 2 S 4 is a semiconductor photocatalyst with strong visible-light absorption (E g = 2.20 eV) and CB potential more negative than H + /H 2 potential [50], and NiCr-LDH also possesses strong visible-light absorption due to the uniform distribution of multivalent Cr ions within the 2D hydroxide layers and high oxidation potential [51], it seems to be a reasonable route to develop heterostructure photocatalysts involving CdIn 2 S 4 and NiCr-LDH. Moreover, the abundant hydroxyl groups on the surface of NiCr-LDH make it convenient to in-situ growth of In(OH) 3 , and thus it is feasible to deposit CdIn 2 S 4 and In(OH) 3 nanocomposite on NiCr-LDH by one-step hydrothermal growth. The as-prepared multicomponent photocatalytic systems are demonstrated to be effective and recyclable for photocatalytic H 2 production and Cr(VI) reduction. The enhanced photocatalytic performance of CdIn 2 S 4 /In(OH) 3 /NiCr-LDH are attributed to boosted separation and migration of charge carriers due to staggered band alignment, and multiple intimate-contact interfaces. This work demonstrates a simple strategy to construct LDH-based ternary photocatalysts, which generally involve two or more complicated synthesis procedures and are relatively time-consuming. The results of this work could be beneficial for further construction of other multi-component photocatalytic systems for environmental and energy applications. Moreover, NiCr-LDH provides a good platform for the fabrication of heterostructure photocatalysts. To date, there are very limited research reports on NiCr-LDH-based composites in the field of photocatalysis.

Synthesis of NiCr-LDH
Ni(NO 3 ) 2 6H 2 O (1.8 mmol), Cr(NO 3 ) 3 9H 2 O (0.6 mmol) and urea (8 mmol) were dissolved in 60 mL deionized (DI) water, which was then transferred into a 100 mL Teflonlined stainless steel autoclave and heated at T = 120, 150, 180 and 190 • C for 10 h. After being naturally cooled, the precipitates were collected by centrifugation, thorough washed with DI water, and drying at 60 • C in vacuum overnight. The obtained NiCr-LDH samples were labeled as NC-T, where T stands for the temperature of hydrothermal growth.

Evaluation of Photocatalytic Performance
Photocatalytic H 2 production. The experiments were carried out on a Lab Solar system (Labsolar-III AG, Perfectlight Technology, Co. Ltd., Beijing, China). Briefly, 50 mg as-prepared photocatalysts were dispersed in 80 mL solution of Na 2 S (0.35 M) and Na 2 SO 3 (0.25 M) in a 300 mL quartz reactor. The suspension was vacuumed and stirred in the dark for 30 min, and then exposed to light irradiation emitted from a 300 W Xenon lamp (PLS-SXE 300, PerfectLight Technology, Co. Ltd., Beijing, China) positioned above the reactor ( Figure S1). The photocatalytic generated H 2 from water splitting was quantitatively analyzed online every hour by a gas chromatography (GC2014, TCD, 5 Å molecular sieve, Shimadzu, Japan). The apparent quantum yield (AQY) was measured under similar condi- Cr(VI) photoreduction. The experiments were carried out on a multi-tube photocatalytic reaction system (JOYN-GHX-AC, Qiaoyue Electronics, Shanghai, China). A total of 30 mg as-prepared photocatalysts were dispersed in Cr(VI) solution (50 mL, 50 ppm) in 80 mL quartz tubes, and stirred in dark for 30 min. Then, a 400 W Mercury light bulb equipped with cutoff filters (λ ≥ 400 nm) was turned on to shine light on the suspension. At fixed time intervals, 4 mL of suspension was sampled and centrifuged to remove the residual catalysts. Finally, the Cr(VI) content was monitored by comparing the change in absorbance on a Shimadzu UV-V is 3600 spectrometer (Shimadzu, Japan). To test the stability, HP-180 was washed after each run and Cr(VI) solution was refilled.

Characterization of Photocatalysts
The structure and morphology of as-prepared photocatalysts were investigated by XRD, SEM, and TEM analysis. Figure 1 showed the recorded XRD curves of the CdIn 2 S 4 /In(OH) 3 /NiCr-LDH composite photocatalysts HP-120, HP-150, HP-180 and HP-190 together with NC-180 and CIS-4. Figure S2a  A plausible explanation might be that the abundant hydroxyl groups on the surface of NiCr-LDH can bond with In 3+ ions during the hydrothermal growth and thus promote the formation of In(OH) 3 . Figure S2c showed the XRD curves of CdIn 2 S 4 /In(OH) 3 /NiCr-LDH prepared with increasing amounts of TAA. Compared to HP-180, it can be found that the Nanomaterials 2021, 11, 3122 5 of 17 characteristic peak of In(OH) 3 becomes weaker in HP-180-1, and disappears in HP-180-2, indicating that the amounts of TAA could affect the relative content of CdIn 2 S 4 and In(OH) 3 . Figure S2d displayed the XRD patterns of CdIn 2 S 4 /In(OH) 3 /NiCr-LDH synthesized by using different amounts of NiCr-LDH. HP-180 and HP-180-4 exhibited similar diffraction patterns. For HP-180-3, the (003) diffraction peak of NiCr-LDH can hardly be observed, which might be due to reduced amount of NiCr-LDH (0.5 times that of HP-180). In addition, characteristic peaks associated with CdIn 2 S 4 became more pronounced compared to HP-180.
A plausible explanation might be that the abundant hydroxyl groups on th NiCr-LDH can bond with In 3+ ions during the hydrothermal growth and thus formation of In(OH)3. Figure S2c showed the XRD curves of CdIn2S4/In(OH prepared with increasing amounts of TAA. Compared to HP-180, it can be fo characteristic peak of In(OH)3 becomes weaker in HP-180-1, and disappears indicating that the amounts of TAA could affect the relative content of In(OH)3. Figure S2d displayed the XRD patterns of CdIn2S4/In(OH)3/NiCr-L sized by using different amounts of NiCr-LDH. HP-180 and HP-180-4 exhi diffraction patterns. For HP-180-3, the (003) diffraction peak of NiCr-LDH c observed, which might be due to reduced amount of NiCr-LDH (0.5 times tha In addition, characteristic peaks associated with CdIn2S4 became more pron pared to HP-180.  Figure 2d showed SEM EDX mapping. It revealed t Cr and Ni elements are homogeneously distributed in HP-180. O is detected analyzed area but its distribution is less even. This result was confirmed by vation. In the HRTEM image in Figure 3c and Figure S3b, the lattice spacing and 0.37 nm correspond to (311), (220) and (006) crystal planes of CdIn2S4, NiCr-LDH, respectively. In Figure 3d   In comparison, the HP-180 photocatalytic system revealed an interwoven-sheet structure composed of nanoflakes. Figure 2d showed SEM EDX mapping. It revealed that Cd, In, S, Cr and Ni elements are homogeneously distributed in HP-180. O is detected in the whole analyzed area but its distribution is less even. This result was confirmed by TEM observation. In the HRTEM image in Figure 3c and Figure S3b, the lattice spacing of 0.33, 0.28 and 0.37 nm correspond to (311), (220) and (006) crystal planes of CdIn 2 S 4 , In(OH) 3 and NiCr-LDH, respectively. In Figure 3d, the lattice spacing of 0.21 and 0.40 nm correspond to (511) and (200) crystal planes of CdIn 2 S 4 and In(OH) 3 , respectively. In Figure S3b, the lattice spacing 0.26 nm corresponds to (012) crystal plane of NiCr-LDH. The HRTEM observation suggested the formation of a composite photocatalyst with multiple intimate-contact interfaces, which could facilitate the transfer and separation of charge carriers across the heterojunction.   XPS measurements were carried out to analyze the surface elemental composition and chemical states of the as-prepared photocatalysts. As shown in the survey spectra ( Figure S4), CIS-4 was composed of Cd, In, and S, while NC-180 was composed of Ni, Cr, C and O elements. For HP-180, the measured XPS spectrum showed the signals of Cd, In, S, Ni, Cr, C, and O elements, suggesting the hybridization of CIS-4 and NC-180. Figure 4 showed the high-resolution spectra of Cd, In, S, Ni, Cr and O. The Cd 3d spectrum of CIS-4 exhibited two main peaks at 405.3 and 412.0 eV, which are associated with Cd 3d 5/2 and Cd 3d 3/2 of Cd 2+ , respectively [50,52]. After coupling with NiCr-LDH, the binding energies of Cd 3d 5/2 and Cd 3d 3/2 of HP-180 were decreased to 405.1 and 411.8 eV, respectively. The In 3d spectrum of CIS-4 had two peaks with binding energies of 444.9 and 452.4 eV, corresponding to In 3d 3/2 and 3d 5/2 of In 3+ , respectively [41,52]. For HP-180, the In 3d 5/2 peak was negatively shifted to 444.8 eV. As shown in Figure 4c, the 161.6 and 162.8 eV peaks could be assigned to S 2p 3/2 and S 2p 1/2 , respectively [40]. No obvious shift was observed in the S 2p spectrum of HP-180. The Ni 2p signal of pure NiCr-LDH can be deconvoluted into doublet peaks centered at 855.7 and 873.4 eV associated with Ni 2p 3/2 and Ni 2p 1/2 of Ni 2+ oxidation state, and peaks at 861.8 and 879.5eV originated from Ni 3+ state [53,54], indicating the co-existence of Ni 2+ and N 3+ ions. In the Ni 2p XPS spectrum of HP-180, the binding energy peaks were positively shifted to 856.3, 862.3, 873.8 and 880.0 eV, respectively. The Cr 2p spectrum of pure NiCr-LDH revealed two characteristic peaks at 577.3 and 587.0 eV, which can be attributed to Cr 3+ in LDH lattice [55], while Cr 2p peaks of HP-180 were positively shifted to 577.6 and 587.3 eV, respectively. The binding energy of O 1s was increased from 531.3 eV for NiCr-LDH to 531.7 eV for HP-180. The decreased binding energies of Cd 3d, and In 3d, and increased binding energies of Ni 2p, Cr 2p and O 1s in CdIn 2 S 4 /In(OH) 3 /NiCr-LDH indicated that there was intimate interfacial contact and strong chemical interaction among CdIn 2 S 4 , In(OH) 3 and NiCr-LDH, which may suppress electron-hole recombination. Combining the characterization results of XRD, SEM, TEM, and XPS analysis, CdIn 2 S 4 /In(OH) 3 /NiCr-LDH heterostructure photocatalysts with strong interfacial interaction were obtained, which could promote spatial charge separation and lead to improved photocatalytic performance. XPS measurements were carried out to analyze the surface elemental compositio and chemical states of the as-prepared photocatalysts. As shown in the survey spect ( Figure S4 [50,52]. After coupling with NiCr-LDH, the binding energi of Cd 3d5/2 and Cd 3d3/2 of HP-180 were decreased to 405.1 and 411.8 eV, respectively. Th In 3d spectrum of CIS-4 had two peaks with binding energies of 444.9 and 452.4 eV, co responding to In 3d3/2 and 3d5/2 of In 3+ , respectively [41,52]. For HP-180, the In 3d5/2 pea was negatively shifted to 444.8 eV. As shown in Figure 4c, the 161.6 and 162.8 eV pea could be assigned to S 2p3/2 and S 2p1/2, respectively [40]. No obvious shift was observe in the S 2p spectrum of HP-180. The Ni 2p signal of pure NiCr-LDH can be deconvolute into doublet peaks centered at 855.7 and 873.4 eV associated with Ni 2p3/2 and Ni 2p1/2 Ni 2+ oxidation state, and peaks at 861.8 and 879.5eV originated from Ni 3+ state [53,54], i dicating the co-existence of Ni 2+ and N 3+ ions. In the Ni 2p XPS spectrum of HP-180, th binding energy peaks were positively shifted to 856.3, 862.3, 873.8 and 880.0 eV, respe tively. The Cr 2p spectrum of pure NiCr-LDH revealed two characteristic peaks at 577 and 587.0 eV, which can be attributed to Cr 3+ in LDH lattice [55], while Cr 2p peaks of H Since the solar light response regime also plays an important role in photocatalytic efficiency, UV-Vis diffuse reflectance spectra were measured to investigate the absorption ability of the as-prepared photocatalysts. As shown in Figure 5a, CIS-4 exhibited strong absorption until the wavelength reached around 570 nm. NiCr-LDH could respond to a wider solar spectrum, with two broad absorption bands in the visible region between 330-480 nm and 490-850 nm, which are associated with the d-d transitions of metal ions distributed in the 2D hydroxide layers of LDH [51,56]. The CdIn 2 S 4 /In(OH) 3 /NiCr-LDH heterostructure photocatalyst not only had strong light absorption up to 530 nm, similar to that of CdIn 2 S 4 , but also exhibited an absorption band in the region of 530-850 nm, which was consistent with NiCr-LDH. The enhanced optical absorption would help to utilize solar energy and generate more photo-generated charges, thereby enhancing the photocatalytic activity. As shown in Figure 5b,c, the band gap energies (E g ) of CdIn 2 S 4 and NiCr-LDH were extracted via αhν = A hν − E g n/2 and equal to 2.06 and 2.46 eV, respectively. It was noted that n = 4 (indirect band gap) and n = 1 (direct band gap) were used for CdIn 2 S 4 and NiCr-LDH, respectively. Figure S5a Nanomaterials 2021, 11, x FOR PEER REVIEW 8 strong chemical interaction among CdIn2S4, In(OH)3 and NiCr-LDH, which may sup electron-hole recombination. Combining the characterization results of XRD, SEM, and XPS analysis, CdIn2S4/In(OH)3/NiCr-LDH heterostructure photocatalysts with s interfacial interaction were obtained, which could promote spatial charge separation lead to improved photocatalytic performance. Since the solar light response regime also plays an important role in photocat efficiency, UV-Vis diffuse reflectance spectra were measured to investigate the absor ability of the as-prepared photocatalysts. As shown in Figure 5a, CIS-4 exhibited s absorption until the wavelength reached around 570 nm. NiCr-LDH could respond wider solar spectrum, with two broad absorption bands in the visible region between 480 nm and 490-850 nm, which are associated with the d-d transitions of metal ion tributed in the 2D hydroxide layers of LDH [51,56]. The CdIn2S4/In(OH)3/NiCr-LDH erostructure photocatalyst not only had strong light absorption up to 530 nm, simi that of CdIn2S4, but also exhibited an absorption band in the region of 530-850 nm, w was consistent with NiCr-LDH. The enhanced optical absorption would help to u solar energy and generate more photo-generated charges, thereby enhancing the p catalytic activity. As shown in Figure 5b,c, the band gap energies (Eg) of CdIn2S4 and N LDH were extracted via αhν A hν E / and equal to 2.06 and 2.46 eV, respect It was noted that n = 4 (indirect band gap) and n = 1 (direct band gap) were use CdIn2S4 and NiCr-LDH, respectively. Figure S5a

Photocatalytic Activity
The photocatalytic H2 production activities from water splitting over as-prepared photocatalysts were measured under a 300 W Xe lamp irradiation. Figure 6 showed the results of HP-120, HP-150, HP-180 and HP-190, which were prepared by using the same amounts of Cd(NO3)2•4H2O, In(NO3)3•4.5H2O, thioacetamide and NiCr-LDH, except the growth temperature of NiCr-LDH was increased from 120 to 190 °C. For comparison, H2 evolution of the sample prepared under similar conditions as HP-180 without the introduction of NiCr-LDH (CIS-4) and pure NiCr-LDH (NC-180) were also plotted. As expected, the heterostructure photocatalysts HP-150, HP-180 and HP-190 exhibited better photocatalytic H2 production performance. In particular, HP-180 presented the highest H2 evolution rate of 1093 μmol•g−1•h−1, which is 10.9 and 58.7 times the counterparts of CIS-4 and NC-180, respectively. For HP-180, the calculated AQY is 1.7% at 420 nm. Table S1 lists the photocatalytic H2 evolution of LDH-based heterostructures. It can be found that the AQY of HP-180 is slightly higher than the reported value for ZnS/ZnIn-LDH (AQY=1.3% at 415 nm) [57]. Although the AQY of HP-180 is lower than the reported val-

Photocatalytic Activity
The photocatalytic H 2 production activities from water splitting over as-prepared photocatalysts were measured under a 300 W Xe lamp irradiation. Figure 6 showed the results of HP-120, HP-150, HP-180 and HP-190, which were prepared by using the same amounts of Cd(NO 3 ) 2 ·4H 2 O, In(NO 3 ) 3 ·4.5H 2 O, thioacetamide and NiCr-LDH, except the growth temperature of NiCr-LDH was increased from 120 to 190 • C. For comparison, H 2 evolution of the sample prepared under similar conditions as HP-180 without the introduction of NiCr-LDH (CIS-4) and pure NiCr-LDH (NC-180) were also plotted. As expected, the heterostructure photocatalysts HP-150, HP-180 and HP-190 exhibited better photocatalytic H 2 production performance. In particular, HP-180 presented the highest H 2 evolution rate of 1093 µmol·g−1·h−1, which is 10.9 and 58.7 times the counterparts of CIS-4 and NC-180, respectively. For HP-180, the calculated AQY is 1.7% at 420 nm. Table S1 lists the photocatalytic H 2 evolution of LDH-based heterostructures. It can be found that the AQY of HP-180 is slightly higher than the reported value for ZnS/ZnIn-LDH (AQY=1.3% at 415 nm) [57]. Although the AQY of HP-180 is lower than the reported values for g-C 3 N 4 @pDA/NiCo-LDH (4.5% at 420 nm) [58] and phosphorylated-NiAl-LDH/g-C 3 N 4 (4.7% at 420 nm) [59], they involved three steps of growth. The simple strategy to obtain ternary heterostructure photocatalysts in this work still has its advantage and could shed light on further design, and easy and cost-effective preparation of new LDH-based multicomponent nanocomposite photocatalysts. It was noted that the weak catalytic activity of HP-120 might be due to the poor crystalline quality of NiCr-LDH grown at 120 • C. To examine the long-term reusable life and stability, HP-180 was reused four consecutive times (Figure 6c). There was no obvious decrease in photocatalytic activity of HP-180, and the H 2 evolution rate remained at 91% that of freshly prepared sample after four consecutive runs. Moreover, there were no observable changes in the XRD pattern of the sample collected after the cycling test in comparison with that of as-prepared sample. We note that we also investigated the influence of the amount of catalysts on H 2 evolution. As shown in Figure S6, 50 mg of HP-180 leads to higher photocatalytic H 2 production. In this work, all H 2 production measurements were conducted using 50 mg photocatalysts.  Figure S7a,b showed H2 production activities of HP-180, HP-180-1 and HP-180-2 wi decreased relative content of In(OH)3/CdIn2S4, which was estimated to be around 2.3, 0.4 and 0, respectively, due to increased amounts of TAA used during synthesis. It can b found that the amount of evolved H2 decreased in the following sequence: HP-180 > H 180-1 > HP-180-2. This indicates that the relative amount of CdIn2S4 to In(OH)3 could affe the photocatalytic activity. Moreover, the stronger H2 evolution activity of HP-180 com pared to HP-180-2 implies that the construction of ternary CdIn2S4/In(OH)3/NiCr-LDH helpful to further improve photocatalytic H2 evolution compared to CdIn2S4/NiCr-LDH Additionally, Figure S7c,d compared the H2 generation over samples prepared by usin  Figure S7a,b showed H 2 production activities of HP-180, HP-180-1 and HP-180-2 with decreased relative content of In(OH) 3 /CdIn 2 S 4 , which was estimated to be around 2.3, 0.45 and 0, respectively, due to increased amounts of TAA used during synthesis. It can be found that the amount of evolved H 2 decreased in the following sequence: HP-180 > HP-180-1 > HP-180-2. This indicates that the relative amount of CdIn 2 S 4 to In(OH) 3 could affect the photocatalytic activity. Moreover, the stronger H 2 evolution activity of HP-180 compared to HP-180-2 implies that the construction of ternary CdIn 2 S 4 /In(OH) 3 /NiCr-LDH is helpful to further improve photocatalytic H 2 evolution compared to CdIn 2 S 4 /NiCr-LDH. Additionally, Figure S7c,d compared the H 2 generation over samples prepared by using 50, 100 and 300 mg of NiCr-LDH, corresponding to HP-180-3, HP-180 and HP-180-4, respectively. The H 2 evolution rate of HP-180 is 9.11 and 7.34 times that of HP-180-3 and HP-180-4.
In addition, the prepared photocatalysts were also tested for photocatalytic Cr(VI) reduction. Figure 7a,b showed the photocatalytic reduction efficiency of Cr(VI) over NC-180, CIS-4, HP-180, HP-180-1, HP-180-2 and the mixture of CdIn 2 S 4 , In(OH) 3 and NiCr-LDH. Without the presence of catalysts, there was barely any change in Cr(VI) concentration. The presence of CIS-4, HP-180, HP-180-1 and HP-180-2 photocatalysts resulted in an obvious decrease in Cr(VI) concentration, suggesting that they play an important role in the photocatalytic reduction of Cr(VI). Moreover, the Cr(VI) removal efficiency decreased in the following sequence: HP-180 > HP-180-1 > HP-180-2 > CIS-4 > mixture > NC-180. Among all the tested photocatalysts, HP-180 exhibited the highest photoreduction efficiency at 97% after 120 min illumination, which is higher than CIS-4 at 75% and NC-180 at 6%. The kinetic rate constant of HP-180 is 0.03019 min −1 , which is 1.7 and 74.5 times higher than those of CIS-4 (0.01125 min −1 ) and NC-180 (0.0004 min −1 ). The scavenger tests revealed that electrons played a dominant role in the photocatalytic reduction of Cr(VI) in this work ( Figure S8). Figure 7c displayed the results of stability tests of Cr(VI) reduction over HP-180. It can be found that the removal efficiency remains at 94% that of the as-prepared sample after three consecutive runs, and the slight decrease in efficiency might be caused by loss of catalyst due to washing after each run. Furthermore, there were no observable changes in the XRD pattern of the sample collected after the cycling test. Consequently, the results of the photocatalytic experiments suggested that CdIn 2 S 4 /In(OH) 3 /NiCr-LDH heterostructure photocatalysts are capable of effective H 2 evolution from water splitting and reduction of Cr(VI) with good stability.

Plausible Photocatalytic Mechanism
To investigate the separation and transfer efficiency of photo-generated charge carriers in CdIn 2 S 4 /In(OH) 3 /NiCr-LDH, electrochemical impedance spectroscopy (EIS), transient photocurrent response and photoluminescence (PL) measurements were carried out. Figure S9 showed samples and as-prepared working electrode for EIS and transient photocurrent measurements. Figure 8a showed Nyquist plots of HP-120, HP-150, HP-180 and HP-190 together with CdIn 2 S 4 and NiCr-LDH. It can be found that the semicircles of HP-120, HP-150, HP-180 and HP-190 have smaller diameters than that of the counterparts CdIn 2 S 4 and NiCr-LDH. The semicircles in the EIS spectra can be ascribed to the contribution from the charge transfer resistance (R p ) and the constant phase element (CPE1) at the photocatalyst/electrolyte interface, while the inclined line is associated with the ion diffusion in the electrolyte [60,61]. The corresponding equivalent circuit is shown as the inset of Figure 8a. The extracted R p values are 391.8, 339.6, 683.7, 1665, 1686 and 2356 Ω for HP-190, HP-180, HP-150, HP-120, CdIn 2 S 4 and NiCr-LDH, respectively. It can be found that samples HP-190, HP-180 and HP-150 have smaller R p values than CdIn 2 S 4 and NiCr-LDH, and HP-180 has the smallest R p value, indicating enhanced electron transfer and suppressed charge recombination in HP-190, HP-180 and HP-150, which might be due to the multiple intimate-contact interface in the CdIn 2 S 4 /In(OH) 3 /NiCr-LDH samples. The lower impedance and higher charge transfer efficiency is beneficial for promoting photocatalytic performance, which is consistent with the results of photocatalytic H 2 evolution. As illustrated in Figure 8b, CdIn 2 S 4 /In(OH) 3 /NiCr-LDH exhibited higher photocurrent density response than that of the counterparts CdIn 2 S 4 and NiCr-LDH, indicating improved charge separation efficiency. Figure 8c showed the PL spectra of CdIn 2 S 4 , NiCr-LDH and HP-180. It is known that when a photocatalyst is exposed to light irradiation, electrons are excited from the valence band to the conduction band. After that, photo-generated electrons can recombine with holes to emit light (radiative recombination), trapped by deep energy levels associated with defects in bulk, and migrate to the surface to potentially participate in photocatalytic reactions. HP-180 had a lower emission intensity than CdIn 2 S 4 and NiCr-LDH, implying reduced radiative recombination of electrons and holes in CdIn 2 S 4 /In(OH) 3 /NiCr-LDH, and possibly more photo-generated charge carriers to contribute to photocatalytic reactions. Therefore, the results of the EIS, photocurrent response and PL measurements demonstrated that the construction of CdIn 2 S 4 /In(OH) 3 /NiCr-LDH can accelerate the separation and migration efficiency of photo-generated electrons and holes, which are highly favorable for photocatalytic H 2 evolution and Cr(VI) reduction.

Plausible Photocatalytic Mechanism
To investigate the separation and transfer efficiency of photo-generated charge ca riers in CdIn2S4/In(OH)3/NiCr-LDH, electrochemical impedance spectroscopy (EIS), tran sient photocurrent response and photoluminescence (PL) measurements were carried ou Figure S9 showed samples and as-prepared working electrode for EIS and transient pho tocurrent measurements. Figure 8a showed Nyquist plots of HP-120, HP-150, HP-180 an HP-190 together with CdIn2S4 and NiCr-LDH. It can be found that the semicircles of HP 120, HP-150, HP-180 and HP-190 have smaller diameters than that of the counterpar CdIn2S4 and NiCr-LDH. The semicircles in the EIS spectra can be ascribed to the contribu tion from the charge transfer resistance (Rp) and the constant phase element (CPE1) at th photocatalyst/electrolyte interface, while the inclined line is associated with the ion diffu sion in the electrolyte [60,61]. The corresponding equivalent circuit is shown as the ins of   (Figure 8d) versus NHE, respectively. The CB potential of undoped n-type semiconductor is typically 0.3 eV more negative than the flat-band potentials [63], and thus the CB potentials of CdIn 2 S 4 , In(OH) 3 , and NiCr-LDH are around −1.43, −0.88 and −0.05 eV versus NHE, respectively. Subtracting the corresponding band gap energies of 2.06, 4.15 [45], and 2.46 eV, the VB potentials of CdIn 2 S 4 , In(OH) 3 , and NiCr-LDH are located at about 0.63, 3.27 and 2.41 eV (vs. NHE), respectively. photocatalytic reactions. HP-180 had a lower emission intensity than CdIn2S4 and NiCr-LDH, implying reduced radiative recombination of electrons and holes in CdIn2S4/In(OH)3/NiCr-LDH, and possibly more photo-generated charge carriers to contribute to photocatalytic reactions. Therefore, the results of the EIS, photocurrent response and PL measurements demonstrated that the construction of CdIn2S4/In(OH)3/NiCr-LDH can accelerate the separation and migration efficiency of photo-generated electrons and holes, which are highly favorable for photocatalytic H2 evolution and Cr(VI) reduction.  (Figure 8d) versus NHE, respectively. The CB potential of undoped n-type semiconductor is typically 0.3 eV more negative than the flatband potentials [63], and thus the CB potentials of CdIn2S4, In(OH)3, and NiCr-LDH are around −1.43, −0.88 and −0.05 eV versus NHE, respectively. Subtracting the corresponding band gap energies of 2.06, 4.15 [45], and 2.46 eV, the VB potentials of CdIn2S4, In(OH)3, and NiCr-LDH are located at about 0.63, 3.27 and 2.41 eV (vs. NHE), respectively.
Based on the above results, a possible mechanism with facilitated charge transfer process and improved photocatalytic activity of CdIn2S4/In(OH)3/NiCr-LDH heterostructure photocatalysts in this work was proposed in Scheme 1. CdIn2S4 and NiCr-LDH with Based on the above results, a possible mechanism with facilitated charge transfer process and improved photocatalytic activity of CdIn 2 S 4 /In(OH) 3 /NiCr-LDH heterostructure photocatalysts in this work was proposed in Scheme 1. CdIn 2 S 4 and NiCr-LDH with strong visible-light absorption can be easily excited under light illumination and engender electrons and holes in the CB and VB, respectively. As the CB of CdIn 2 S 4 (−1.43 eV vs. NHE) is more negative than that of In(OH) 3 (−0.88 eV vs. NHE) and NiCr-LDH (−0.05 eV vs. NHE), the photo-generated electrons in the CB of CdIn 2 S 4 are transferred to the CB of In(OH) 3 and NiCr-LDH, where the electrons can effectively reduce H + to produce H 2 molecules. Meanwhile, the holes accumulated in the VB of CdIn 2 S 4 could be quenched by the sacrificial agent (S 2− , SO 3 2− ). During the photocatalytic reaction, the multi-interfaces with strong interactions play a vital role in the reaction as they ensure an unimpeded pathway for the fast electron transfer. In the photocatalytic reduction of Cr(VI) reaction, the photo-generated electrons can reduce Cr 2 O 7 2− to Cr(III) (0.51 eV vs. NHE) [40]. Regarding the existence of the In 2 S 3 impurity phase, it has been reported by Ma et al. that In 2 S 3 also has strong visible light absorption and similar CB potential as that of CdIn 2 S 4 . Therefore we are not going to discuss charge transfer between CdIn 2 S 4 and the impurity phase of In 2 S 3 here [64].
for the fast electron transfer. In the photocatalytic reduction of Cr(VI) reac generated electrons can reduce Cr2O7 2− to Cr(III) (0.51 eV vs. NHE) [40]. existence of the In2S3 impurity phase, it has been reported by Ma et al. tha strong visible light absorption and similar CB potential as that of CdIn2S are not going to discuss charge transfer between CdIn2S4 and the impurit here [64]. Scheme 1. Illustration of the transfer process of the photo-excited charge carriers a tocatalytic mechanism for H2 production and Cr(VI) reduction in the CdIn2S4/In heterostructure photocatalyst.

Conclusions
In summary, CdIn2S4/In(OH)3/NiCr-LDH heterostructure photocata pared by a simple and cost-effective hydrothermal method through one CdIn2S4 and In(OH)3 on NiCr-LDH. The obtained CdIn2S4/In(OH)3/NiCrimproved photocatalytic performance on H2 production and Cr(VI) redu mal CdIn2S4/In(OH)3/NiCr-LDH sample HP-180 has a photocatalytic H2 e 1093 μmol•g −1 •h −1 , which is 10.9 and 58.7 times that of the counterparts CIS respectively. HP-180 showed good stability and the H2 evolution rate re after four consecutive runs. Furthermore, HP-180 also showed enhanced reduction of Cr(VI), and the removal efficiency increased from 75% for C NC-180 to 97% after 120 min illumination. The photocatalytic mechanis gated by combining UV-Vis DRS, EIS, transient photocurrent response, PL calculations. The improved photocatalytic activity is tentatively attributed separation and transfer of photo-excited electrons and holes due to the mu contact interfaces and strong visible-light absorption ability. This work ca further design, and easy and cost-effective preparation of new LDH-base composite photocatalysts for applications in energy conversion and enviro diation. Scheme 1. Illustration of the transfer process of the photo-excited charge carriers and plausible photocatalytic mechanism for H 2 production and Cr(VI) reduction in the CdIn 2 S 4 /In(OH) 3 /NiCr-LDH heterostructure photocatalyst.

Conclusions
In summary, CdIn 2 S 4 /In(OH) 3 /NiCr-LDH heterostructure photocatalysts were prepared by a simple and cost-effective hydrothermal method through one-pot depositing CdIn 2 S 4 and In(OH) 3 on NiCr-LDH. The obtained CdIn 2 S 4 /In(OH) 3 /NiCr-LDH exhibited improved photocatalytic performance on H 2 production and Cr(VI) reduction. The optimal CdIn 2 S 4 /In(OH) 3 /NiCr-LDH sample HP-180 has a photocatalytic H 2 evolution rate of 1093 µmol·g −1 ·h −1 , which is 10.9 and 58.7 times that of the counterparts CIS-4 and NC-180, respectively. HP-180 showed good stability and the H 2 evolution rate remained at 91% after four consecutive runs. Furthermore, HP-180 also showed enhanced photocatalytic reduction of Cr(VI), and the removal efficiency increased from 75% for CIS-4 and 6% for NC-180 to 97% after 120 min illumination. The photocatalytic mechanism was investigated by combining UV-Vis DRS, EIS, transient photocurrent response, PL and theoretical calculations. The improved photocatalytic activity is tentatively attributed to the boosted separation and transfer of photo-excited electrons and holes due to the multiple intimate-contact interfaces and strong visible-light absorption ability. This work can shed light on further design, and easy and cost-effective preparation of new LDH-based ternary nanocomposite photocatalysts for applications in energy conversion and environmental remediation.
Supplementary Materials: The following are available online at https://www.mdpi.com/article/10 .3390/nano11113122/s1, Figure S1: Light intensity versus wavelength of a 300 W Xe lamp (PLS-SXE 300) used in photocatalytic H2 evolution experiments, supplied from Beijing Perfect Light Technology, Co. Ltd., Figure Figure S3: (a) TEM and (b) HRTEM images of HP-180 recorded from a different area, other than that used in Figure 3c,d, Figure S4: XPS survey spectra of CIS-4, NC-180 and HP-180, revealing that the heterostructure catalyst is mainly com-posed of Cd, Cr, In, Ni, C, O and S elements, Figure S5: The band structure (a,b) and partial density of states (PDOS) (c,d) of CIS-4 and NC-180, respectively, Figure S6: Photocatalytic H2 production when 10, 50 and 100 mg HP-180 were used. It can be observed that the photocatalytic activity is related to the amount of catalyst and 50 mg leads to higher H 2 yield, Figure S7

Data Availability Statement:
The data present in this study are available on request from the corresponding author.