Multiobjective‐Optimization MoS2/Cd‐ZnIn2S4/CdS Composites Prepared by In Situ Structure‐Tailored Technique for High‐Efficiency Hydrogen Generation

Photocatalytic water splitting into hydrogen production provides a new avenue to produce clean chemical fuels. However, developing high‐efficiency photocatalytic materials still remains a challenge till now. Herein, multiobjective‐optimization MoS2/Cd‐ZnIn2S4/CdS (MS/CZIS/CS) composites are successfully constructed by an in situ structure‐tailored technique. Benefiting from the synergistic feature integrating sulfur vacancy, II‐type CZIS/CS heterojunction and Schottky‐type MS/CS heterojunction, such composites not only effectively steer photogenerated carrier transfer but also markedly expedite surface reaction kinetics for hydrogen reduction reaction. As a result, an optimal hydrogen evolution rate of 11.49 mmol g−1 h−1 is achieved over the MS/CZIS/CS catalysts, which is approximately 4.79 times higher than that of pristine ZIS (2.40 mmol g−1 h−1). This work provides some new inspirations for the steering of carrier transfer and the design of multiobjective‐optimization photocatalysts with high efficiency.

photocatalyst.However, just as a coin has two sides, each approach inevitably has its disadvantages.For example, the doping of heteroatom would unavoidably change the atomic structure of ZIS and possibly form some recombination centers; [18] cocatalysis could potentially weaken the light absorption due to the "light shielding" effect and reduce proton reduction reaction sites; [19] while inexpedient construction of heterojunction would also result in stacking interfaces, high carrier recombination, and diminished active sites.
Very recently, multiobjective-optimization system has inspired wide attention to effectively improve the photocatalytic performance of ZIS. [20]Such approach takes full advantage of the synergistic effect and merit of each optimization method.Majhi et al. first synthesized tetragonal β-Bi 2 O 3 nanoplates through a hydrothermal method, and then prepared dual Z-scheme Bi 2 S 3 /β-Bi 2 O 3 /ZIS nanomaterials by a facile reflux route. [21]onsequently, such ternary composites achieved fast electron channelization, enhanced charge carrier separation, and prolonged lifetime, finally resulting in excellent visible-light activity.Nevertheless, such coupled systems still suffer from unsatisfactory photocatalytic efficiency, complex multistep processes, long processing periods, and high costs.That is, realizing the synergistic effect of multiobjective-optimization methods and maintaining the high hydrogen evolution efficiency still remain a great challenge.
Herein, MS/CZIS/CS composites were successfully constructed by an in situ structure-tailored technique (Scheme 1).The multiobjective-optimization composites integrated sulfur vacancy (S v ), II-type CZIS/CS heterojunction, and Schottky-type MS/CS heterojunction.Specially, the S v tended to narrow the bandgap, promote the light harvesting, and boost the carrier separation; II-type CZIS/CS heterojunction was helpful for the electron/hole separation; Schottky-type MS/CS heterojunction was good for the charge transfer, hydrogen reduction reaction kinetics, and surface charge utilization.As a result, compared to pristine ZIS, the multiobjective-optimization MS/CZIS/CS composites exhibited an approximately 4.79-fold enhancement of photoactivity.

Preparation and Structural Characterization of MS/CZIS/CS Composites
Here, tetragonal CdMoO 4 (CMO) nanoparticles with 0.5-2 μm in diameter were preferred as an important structure-tailored body prepared by a hydrothermal method (Figure S1, Supporting Information).Subsequently, the in situ decomposition of CMO can promote the instantaneous formation of S v , II-type CZIS/CS heterojunction and Schottky-type MS/CS heterojunction.To exemplify this, the obtained samples were denoted as CMO-x/ ZIS (x = 0.5, 1, 2, 3, 4, 5, 6, 20 at%), where x represented the molar percentage of CMO to ZIS.The crystalline structure of as-prepared samples was determined by powder X-ray diffraction (PXRD) technique.As shown in Figure 1A, all the diffraction peaks of pristine ZIS can be ascribed to hexagonal ZIS (JCPDS No. 72-0773). [22]The diffraction peaks at 21.6°, 27.7°, and 47.2°correspond to the (006), (102), and (110) crystal planes, respectively.Moreover, the diffraction signals of CMO-x/ZIS exhibit the same characteristic peaks with pristine ZIS.No characteristic peaks of CMO are detected, even in CMO-20/ZIS composite (Figure S2A, Supporting Information), demonstrating the phase transition or complete decomposition of CMO under the water-bath condition.Nevertheless, the typical diffraction peak at 47.2°corresponding to (110) crystal plane progressively shifts to lower 2θ angle (the inset in Figure 1A and S2B, Supporting Information), indicating the possible doped heteroatoms with larger ionic radius in the crystal lattice of ZIS. [23]As depicted in Figure 1B and S3, Supporting Information, the (110) crystal configuration mainly lies in [Zn-S] 4 tetrahedral coordination consisting of Zn and S atoms.Namely, Zn atoms are potentially replaced by some heteroatoms.
Subsequently, the XPS of high-resolution Mo 3d indicates that the two peaks at 235.4 eV (Mo 3d 3/2 ) and 232.3 eV (Mo 3d 5/2 ) obviously shift to lower binding energies at 233.0 and 229.9 eV, indicating that Mo 6þ valence state is completely evolved to Mo 4þ valence state during the water-bath process. [24,25]This result further confirms the complete decomposition of CMO during the water-bath process (Figure 1C).Specially, a new peak at 226.3 eV (S 2s) attributed to Mo-S bonds confirms the formation of MS in CMO-3/ZIS composite. [26]Subsequently, the ionic difference was calculated to further analyze the possible heteroatom.29] Combined with the EPR spectra, Cd heteroatom with larger electronegativity and ionic radius results in sulfur vacancy in CMO-3/ZIS (Figure 1D). [30]The result further confirms the successful incorporation of Cd heteroatom in the crystal lattice of ZIS.Additionally, excessive Cd 2þ ions were also investigated by XRD in detail.It can be seen that most diffraction peaks match well with cubic CS (JCPDS No. 80-0019) when only CMO and thioacetamide (TAA) are used during the water-bath process, implying the possible formation of CS (Figure S4, Supporting Information).Specially, no obvious diffraction peaks corresponding to CS and MS are observed in CMO-3/ZIS, possibly resulting from the low content, ultrafine size, and amorphous feature. [31,32]o further elucidate the composition and chemical state, CMO, ZIS, and CMO-3/ZIS samples were analyzed by XPS technique.Clearly, the XPS survey spectra of CMO-3/ZIS verify the presence of Zn, In, S, Cd, Mo, and O elements (Figure S5, Supporting Information), in well agreement with the nominal composition.[35] Additionally, the corresponding Zn 2p, In 3d, and S 2p peaks clearly shift to higher binding energies, implying the strong interaction between different components.Figure 1E shows two strong symmetrical characteristic peaks at 405.2 and 411.9 eV, corresponding to spin-orbit splitting of Cd 2þ 3d 3/2 and Cd 2þ 3d 5/2 . [36]The high-resolution O ls XPS spectrum in pristine CMO indicates that two peaks at 530.3 eV and 531.8 eV correspond to lattice oxygen and absorbed O-containing species (Figure 1F). [37]While the two peaks shift to higher binding energies at 531.8 and 533.2 eV in CMO-3/ZIS sample, assigning to absorbed O-containing species and surface H 2 O, further confirming the complete decomposition of CMO.
Figure 2A,B shows the high-resolution transmission electron microscopy (HRTEM) images of as-prepared CMO-3/ZIS.Obviously, the intimate contact between ZIS, CS, and MS can be found (Figure S7A, Supporting Information).Moreover, the interplanar spacing of 0.32 nm can be ascribed to the (102) crystal plane of hexagonal ZIS, [38] the lattice fringe spacings of 0.34 and 0.22 nm match well with (111) crystal plane of cubic CS [39] and (103) crystal plane of hexagonal MS. [40] Similarly, Figure S7B, Supporting Information, also confirms the coexistence of CS and MS.Field emission scanning electron microscope (FESEM) image displays that ZIS exhibits flower-like hierarchical nanowalls composed of numerous cross-linked nanosheets (Figure 2C). [41]Coupling MS and CS with CZIS, the morphology of CMO-x/ZIS gradually tends to become irregular (Figure 2D and S8, Supporting Information).Additionally, energy dispersive spectrometer (EDS) mappings confirm the homogeneous distribution of Zn, In, S, Cd, Mo, and O elements (Figure 2E and Table S2, Supporting Information).Specially, the molar ratio of Zn:In:S is close to the theoretical value of 1:2:4, indicating that the main phase of ZIS is stable.In a word, the above results confirm that ternary MS/CZIS/CS composites are synthesized successfully.

Photocatalytic Hydrogen Evolution Performance
To verify the relationship between multiobjective-optimization structure and photoactivity, the hydrogen evolution activities of ZIS and CMO-x/ZIS catalysts were evaluated in a Pt-free cocatalytic system upon visible-light irradiation.Clearly, the continuous hydrogen production for all samples increases linearly with extending the irradiation time, revealing the excellent stability of photocatalytic water splitting (Figure 3A).Moreover, CMO-x/ZIS catalysts unsurprisingly exhibit higher hydrogen evolution activity.Therein, the CMO-3/ZIS catalyst displays the highest hydrogen evolution activity, with a hydrogen production of 1.5 mL within 3 h.The calculated hydrogen evolution rate shows that, as shown in Figure 3B, pristine ZIS exhibits a poor photoactivity (2.40 mmol g À1 h À1 ).Notably, the MS/CZIS/CS composites show extraordinary hydrogen evolution performance.Strikingly, a maximum hydrogen evolution rate of 11.49 mmol g À1 h À1 is observed over CMO-3/ZIS catalyst, which is approximately 4.79 times higher than that of pristine ZIS.GThe apparent quantum efficiency (AQE) values of pristine ZIS and CMO-3/ZIS were tested at monochromatic wavelength irradiation.As presented in Figure 3C and Table S3, Supporting Information, the corresponding AQE is consistent with the light absorption.Moreover, CMO-3/ZIS affords an AQE value of 6.17% at a monochromatic light irradiation at 420 nm, which is obviously higher than that of pristine ZIS (AQE = 1.77%).Nevertheless, excessive Cd doing, overloading MS cocatalysis, and massive heterointerface also decrease the hydrogen evolution activity due to lower active sites, higher light shading, and more recombination centers.Additionally, the photocatalytic stability of CMO-3/ZIS was evaluated by cycling experiments.As shown in Figure 3D, the hydrogen production after four cycles shows little deterioration, indicating that CMO-3/ZIS has excellent photostability.Besides, PXRD results also indicate that the peak shift of pristine ZIS can be clearly found after photocatalytic reaction, implying the serious photocorrosion of pristine ZIS (Figure S9A, Supporting Information).While all the diffraction peaks of CMO-3/ZIS remain unchanged after the photocatalytic reaction, further verifying the high photostability (Figure S9B, Supporting Information).

Photoelectric Dynamics Behaviors
The optical properties of as-prepared photocatalysts were investigated by ultraviolet-visible diffuse reflectance spectrum (UV-vis-DRS) spectrophotometer.As shown in Figure 4A and S10, Supporting Information, the absorption edges of ZIS, CS, and MS are approximately 527, 571, and 938 nm, respectively.Excitingly, the light absorption of CMO-x/ZIS is evidently enhanced in the range of 500-800 nm, which is consistent with the optical color of different samples (Figure S11, Supporting Information).Moreover, an obvious redshift for absorption edge can be observed in visible-light bands.Usually, the higher light absorption and redshift of absorption edge are beneficial to the photocatalytic activity.Subsequently, the bandgaps of all samples can be obtained by the Kubelka-Munk function.The bandgaps of ZIS, CS, and MS are determined to be 2.35, 2.18, and 1.32 eV, respectively (Figure S12, Supporting Information).Furthermore, CMO-x/ZIS composites have narrower bandgaps than that of pristine ZIS (Figure 4B and Table S4, Supporting Information).Usually, the narrow bandgap has a considerably positive influence on the carrier excitation and photocatalytic activity.
The transient photocurrent response is crucial for analyzing the photogenerated carrier separation and photostability.Clearly, the photocurrent density increases sharply upon the visible-light radiation and then remains stable (Figure S13, Supporting Information), indicating that all samples are very sensitive under visible-light irradiation. [42]To be specific, pristine ZIS shows a photocurrent density of 76.61 mA cm À2 , which is somewhat higher than those of CMO-x/ZIS catalysts.Nevertheless, the deterioration rate of photocurrent density for pristine ZIS reaches 61.23%, revealing its severe photocorrosion (Figure 4C and Table S5, Supporting Information).Interestingly, the deterioration rate significantly declines to 8.91% for CMO-3/ ZIS and 3.31% for CMO-5/ZIS, further confirming that the photostability is markedly improved. [43]esides, charge migration is also a major factor in photocatalytic performance.Electrochemical impedance spectroscopy (EIS) curves suggest that CMO-x/ZIS catalysts with smaller arc radii possess lower interfacial resistance and faster charge transfer (Figure 4D). [44]Photogenerated carrier recombination is considered as another bottleneck problem limiting the photon conversion efficiency.Here, steady-state photoluminescence (PL) spectra were collected to evaluate the radiative recombination process.Obviously, the intensity of CMO-x/ZIS is lower than that of pristine ZIS (Figure 4E), revealing that the carrier recombination is evidently suppressed. [45]Specially, CMO-3/ZIS exhibits the weakest emission peak, which is consistent with the optimal photocatalytic hydrogen production performance.Furthermore, the charge nonradiative recombination (τ 1 ), free exciton interband recombination (τ 2 ), and the general recombination (τ ave ) can be fitted from time-resolved photoluminescence (TRPL) plots over pristine ZIS and CMO-x/ZIS catalysts.As displayed in Figure 4F and Table S6, Supporting Information, CMO-x/ZIS catalysts exhibit a significant increase in τ 1 and τ 2 , indicating an inhibited charge nonradiative recombination and a faster interband charge transfer.That is, the multiobjectiveoptimization system can significantly prolong the charge lifetime and inhibit the photogenerated carrier recombination. [46]he wavelength-dependent surface photovoltage (SPV) spectra track the corresponding absorbance spectra.As shown in Figure 4G, pristine ZIS shows weak signal intensities, indicating a lower surface carrier concentration.On the contrary, remarkable response can be achieved in MS/CZIS/CS composites, which is beneficial to the following protonated reduction. [47]pecially, CMO-3/ZIS displays the largest visible-light-induced photovoltage and the highest charge concentration on the surface.Additionally, linear sweep voltammetry (LSV) curves of pristine ZIS and CMO-x/ZIS demonstrate that CMO-x/ZIS composites exhibit much lower onset overpotentials than pristine ZIS (Figure 4H).The overpotentials at a current density of À0.10 mA cm À2 were À1.41, À1.39, À1.38, and À1.37 V for ZIS, CMO-1/ZIS, CMO-3/ZIS, and CMO-5/ZIS, revealing that less potential barrier for hydrogen evolution can be realized in MS/CZIS/CS system. [48]In a word, the above results and analysis indicate that CMO-x/ZIS catalysts possess enhanced photoelectric dynamics during the photocatalytic process.

Proposed Photocatalytic Hydrogen Evolution Mechanism
To further investigate the carrier concentration and charge transfer path during the photocatalytic process, the band structures of ZIS, CS, and MS were studied according to the Mott-Schottky (M-S) and ultraviolet photoelectron spectroscopy (UPS) measurements.As displayed in Figure 5A-C, ZIS, CS, and MS display positive slopes in the M-S curves, confirming their n-type semiconductor features. [49]The carrier concentration for CMO-3/ZIS is calculated to be 2.9 Â 10 23 cm À3 , which is approximately 1.2 times higher than that of pristine ZIS (Table S7, Supporting Information).The more accumulated charges are beneficial to hydrogen evolution, in well agreement with photocatalytic water splitting performance and SPV result.Additionally, the flat band potentials (E fb ) of ZIS, CS, and MS can be calculated to be À0.24,À0.03, and À0.01 V (vs RHE, pH = 7.0), respectively.Generally, conduction band potential (E CB ) is approximately 0.1 eV more negative than the E fb value for n-type semiconductors. [50]ence, the E CB values of ZIS, CS, and MS are determined to be À0.34,À0.13, and À0.11 eV (vs RHE, pH = 7).Finally, the valence band potentials (E VB ) of ZIS, CS, and MS are calculated to be 2.01, 2.05, and 1.21 eV.
To understand the redox potential, UPS test was conducted to analyze the work function, Fermi levels, and interfacial charge transfer. [51]Usually, the difference between vacuum level (E vac ) and Fermi level (E F ) is defined as the work function (Ф) of semiconductor, namely, the formula is Ф = E vac -E F . [52]As shown in Figure 5D-F and S14, Supporting Information, the work functions of ZIS, CS, and MS are estimated to be 4.72, 4.98, and 5.06 eV.Namely, the Fermi levels of ZIS, CS, and MS decrease gradually (Figure 5G).After contact, the three components effectively generate built-in space charge layers due to the difference of E F .Namely, the thermal equilibrium leads to the Fermi level realignment.Consequently, II-type CZIS/CS heterojunction and Schottky-type MS/CS heterojunction can be generated due to the potential difference between different components.That is, as shown in Figure 6A, the photogenerated electrons spontaneously transfer from CZIS to CS at the driving force of the internal electric field due to II-type CZIS/CS heterojunction, and finally transfer to MS at the driving force of the internal electric field due to Schottky-type MS/CS heterojunction.Consequently, the accumulated electrons can be reduced to H 2 on the surface of MS.Meanwhile, the photoexcited holes can be automatically extracted from CS to CZIS and finally scavenged by tri-ethanolamine (TEOA) sacrificial agent.That is, the multiobjective-optimization MS/CZIS/CS composites provide a directional charge transfer pathway, finally boosting the photocatalytic water splitting into hydrogen generation.
To verify abovementioned mechanism, the electron spinresonance (ESR) technique was performed to validate the charge migration process.Here •O 2 À and •OH radicals were detected using DMPO as a radical trapping agent.As shown in Figure 6B, ZIS and CMO-3/ZIS exhibit no response of DMPO-•O 2 À in dark.However, under the visible-light irradiation, the signal intensity of DMPO-•O 2 À gradually increases with extending the irradiated time, suggesting the electrons can be generated in ZIS and CMO-3/ZIS.Importantly, the signal intensity of CMO-3/ZIS is obviously higher than that of ZIS catalyst, confirming the effective charge transfer and sufficient charge utilization.It should be noted that the CB potentials of CS and MS are not enough positive to generate the •O 2 À free radicals.That is, higher carrier separation efficiency of ZIS results in more •O 2 À free radicals due to the synergistic effect of multiobjective optimization.Similarly, there are no peaks of DMPO-•OH for pristine ZIS and CMO-3/ZIS in the dark (Figure 6C).However, four characteristic peaks with a standing ratio of intensity 1:2:2:1 within 10 min visible-light irradiation can be found.Moreover, the signal intensity of CMO-3/ZIS is about four times higher than that of pristine ZIS.The possible cause is that the VB potentials of both ZIS and CS are enough positive to produce •OH free radicals.Consequently, integrating sulfur vacancy, II-type CZIS/CS heterojunction and Schottky-type MS/CS heterojunction, increases the •OH production of ZIS.Specially, the enhanced difference between •O 2À free radicals and •OH free radicals also further confirms the above analysis.Finally, compared with pristine ZIS, the upshift of Zn 2þ 2p and Cd 2þ 3d XPS peaks toward higher binding energy and the downshift of Mo 4þ 3d XPS peak toward lower binding energy in CMO-3/ZIS also confirm the electron transfer path from CZIS to CS, finally, transfer to MS.

Conclusion
In summary, MS/CZIS/CS composites were delicately constructed through an in situ structure-tailored technique and applied as an efficient visible-light-driven water-splitting photocatalyst.Experimental results confirmed that the multiobjective-optimization structure integrated sulfur vacancy, II-type CZIS/CS heterojunction and Schottky-type MS/CS heterojunction.The CMO-3/ZIS composite demonstrated an optimal hydrogen evolution rate of 11.49 mmol g À1 h À1 , which was about 4.79 times that of pristine ZIS (2.40 mmol g À1 h À1 ).The photoelectrochemical characterization revealed that the multiobjective-optimization composites provided less photocurrent deterioration, higher carrier separation, faster charge transfer, slower carrier recombination, lower overpotential of hydrogen evolution, and higher surface charge concentration, finally, promoting the hydrogen evolution reaction.This work explores the synergistic effect between the multiplex modification and opens an inspiration to delicately design high-efficiency and stable photocatalytic for water splitting system.

Experimental Section
Chemicals: All reagents were analytical grade and used without further purification throughout the experiments.The deionized (DI) water was utilized in the experiment.Zinc chloride (ZnCl 2 ), sodium molybdate dihydrate (Na 2 MoO 4 •2H 2 O), and TEOA were purchased from Sinopharm Chemical Reagent Co., Ltd.Indium nitrate hydrate (In(NO 3 ) 3 •6H 2 O), TAA, and absolute ethanol were purchased from Aladdin Chemical Synthesis of CMO Nanomaterials: CMO nanomaterials were synthesized by a hydrothermal method. [53]Briefly, 0.6169 g of Cd(NO 3 ) 2 •4H 2 O was dissolved into 60 mL of DI water with magnetic stirring, and marked solution A. Then, 0.4893 g of Na 2 MoO 4 •2H 2 O was dissolved into 20 mL of DI water with magnetic stirring, and marked solution B. Subsequently, the solution B was slowly dropped into the solution A with stirring for 30 min.Afterward, the mixed solution was transferred into a Teflon-lined stainlesssteel autoclave, heated to 160 °C, and maintained for 12 h.After naturally cooling to room temperature, the white products were washed with DI water and absolute ethanol and then dried at 60 °C overnight to obtain tetragonal CMO.
Synthesis of MS/CZIS/CS Composites: In detail, a certain amount of as-prepared CMO was ultrasonically dispersed into 80 mL DI water, and marked solution A. Then, 0.4089 g ZnCl 2 and 0.4508 g TAA were added into solution A with magnetic stirring.Thereafter, solution B was prepared by dissolving 0.9812 g In(NO 3 ) 3 •6H 2 O into 20 mL DI water.Afterward, solution B was slowly added to solution A. Upon magnetically stirring for 30 min, the mixed solution was heated to 80 °C and maintained for 6 h under water-bath conditions.Finally, the precipitate was collected by centrifugation, washed with DI water and absolute ethanol, and then dried at 60 °C overnight in air.Pristine ZIS was prepared by a similar method without the addition of CMO raw materials.For comparison, cubic CS (JCPDS No. 80-0019) and hexagonal MS (JCPDS No. 75-1539) were synthesized according to previous report (Figure S15 and S16, Supporting Information). [54,55]atalyst Characterization: The phase and crystal structure of as-synthesized samples were verified by PXRD (D8-type, Brucker) with a scanning rate of 5°min À1 in the range of 10°-90°, using Cu Kα as radiation source (λ = 1.5406Å).The morphology and microstructure were observed by FESEM (Sigma 500, Zeiss) operated at 10 kV and HRTEM (Talos F200X, Thermo Fisher) operated at 200 kV.The elemental content and distribution were monitored on an EDS.The surface composition and chemical state were investigated by XPS (JPS-9010 MC, JEOL) equipped with a monochromatic Al Kα X-ray source.All binding energies were referred to adventitious C 1s signal at 284.8 eV.The fitting of the spectra was performed using the XPS Peak 41 software.The UV-vis-DRS was recorded to analyze the light absorption characteristic and bandgap using a PerkinElmer Lambda-950 spectrophotometer.The recorded spectrum ranged from 350 to 850 nm with a resolution of 1 nm.PTFE coating was used as a 100% reflectance standard.The steady-state PL emission spectrum was obtained on a Hitachi F-7000 spectrofluorometer equipped with a 450 W xenon lamp as the excitation light source (λ = 455 nm).The photoemission spectrum was recorded from 460 to 800 nm with 1 nm step.TRPL spectroscopy was recorded on an Edinburgh FLS1000 spectrometer to analyze the charge lifetime.The SPV spectrum was recorded on a PL-SPV1000 to analyze the photoinduced charge behavior.The testing current was fixed as 20 A, and the scanning wavelength ranged from 300 to 800 nm.EPR/ESR spectrum was conducted on a Bruker A300 spectrometer to analyze the crystal defect and active radicals.UPS was performed on a ThermoFischer EscaLab 250 Xi spectrometer to determine the work function and Fermi level, using Helium Iα as the excitation source (hv = 21.22 eV).
Photocatalytic Hydrogen Evolution Activity: The photocatalytic water splitting activity was evaluated in a quartz photoreactor with a 300 W Xe-lamp (Perfect Light Co. Ltd., China) as a visible-light source (λ ≥ 420 nm).The light area was approximately 5.28 cm 2 and the irradiated distance was maintained at 10 cm.In a typical process, 2 mg of photocatalyst was dispersed in a mixed solution containing 8 mL of DI water and 2 mL of sacrificial reagent (TEOA).Then, the dispersion was transferred into a 35 mL photoreactor.Prior to light irradiation, the suspension was degassed using high-purity argon gas to completely remove the air.Moreover, the photoreactor was kept at 10 °C through a cooling water system.Subsequently, the suspension was irradiated under visible-light source.At given interval of 0.5 h, 0.1 mL of the resulting gas was extracted from the chamber to qualitatively analyze the gas components using a gas chromatography (GC7900, TIANMEI, China).Cycling experiments were carried out under the same condition to evaluate the stability of photocatalytic activity.All of the photocatalytic experiments were conducted in quadruplicates.Additionally, the wavelength-dependent AQE was measured by means of different monochromatic light filters (420, 450, 475, 500 nm) and calculated according to the following formula: AQE ¼ 2 Â ðthe number of evolved hydrogen moleculesÞ the number of incident photons Â 100% Photoelectrochemical Measurement: The photoelectrochemical measurements, including transient photocurrent response, EIS, LSV, and M-S measurements, were conducted on a CHI 660E electrochemical workstation using 0.1 mol L À1 Na 2 SO 4 solution as the electrolyte.A standard three-electrode configuration was used, consisting of a photocatalyst-supporting FTO working electrode (as-prepared samples), a counter electrode (Pt plate), and a reference electrode (calomel electrode).A 300 W Xe-lamp with a UV cutoff filter (λ ≥ 420 nm) was applied as the visible-light source.To prepare the working electrode, 0.5 mg of photocatalyst was firstly dispersed in 0.5 mL mixed solvent (250 μL EtOH, 250 μL water, and 25 μL Nafion) to form homogeneous catalyst ink (10 mg mL À1 ).Then, 50 μL catalyst ink was drop-coated onto 1 cm 2 FTO electrode.Afterward, the working electrodes were dried naturally.All potentials in this work were given versus reversible hydrogen electrode (RHE) (E RHE = E Hg/HgCl þ 0.059 Â pH þ 0.24).

Figure 3 .
Figure 3. Photocatalytic hydrogen production performance.A) Time-dependent photocatalytic hydrogen production curves and B) hydrogen evolution rates of ZIS and CMO-x/ZIS; C) wavelength-dependent AQE and absorption spectra of ZIS and CMO-3/ZIS; D) cycling curves of photocatalytic hydrogen production in the presence of CMO-3/ZIS.

Figure 5 .
Figure 5. Semiconductor type and energy band structure.M-S curves of A) ZIS, B) CS, and C) MS; UPS spectra of D) ZIS, E) CS, and F) MS; G) schematic energy band arrangement of MS/CZIS/CS catalysts.

Figure 6 .
Figure 6.A) Schematic illustration of the charge dynamics in MS/CZIS/CS system; B,C) DMPO spin-trapping ESR spectra of ZIS and CMO-3/ZIS catalyst in dark and light.