Fabrication of Noble-Metal-Free Mo2C/CdIn2S4 Heterojunction Composites with Elevated Carrier Separation for Photocatalytic Hydrogen Production

Molybdenum-based cocatalyst being used to construct heterojunctions for efficient photocatalytic H2 production is a promising research hotspot. In this work, CdIn2S4 was successfully closely supported on bulk Mo2C via the hydrothermal method. Based on their matching band structures, they formed a Type Ⅰ heterojunction after the combination of Mo2C (1.1 eV, −0.27 V, 0.83 V) and CdIn2S4 (2.3 eV, −0.74 V, 1.56 V). A series of characterizations proved that the heterojunction composite had higher charge separation efficiency compared to a single compound. Meanwhile, Mo2C in heterojunction could act as an active site for hydrogen production. The photocatalytic H2 production activity of the heterojunction composites was significantly improved, and the maximum activity was up to 1178.32 μmmol h−1 g−1 for 5Mo2C/CdIn2S4 composites. 5Mo2C/CdIn2S4 heterojunction composites possess excellent durability in three cycles (loss of 6%). Additionally, the mechanism of increased activity for composites was also investigated. This study provides a guide to designing noble-metal-free photocatalyst for highly efficient photocatalytic H2 evolution.


Introduction
With the increasingly serious energy crisis, H 2 energy is attracting more and more attention due to its renewable, clean, high energy density, and so on [1][2][3]. As a feasible method, photocatalytic H 2 production from water has been a research hotspot among all the methods of H 2 production [4][5][6]. Proverbially, developing stable and high-efficiency visible light photocatalyst is always a core challenge for photocatalysis technology [7,8].
To date, numerous semiconductors-including sulfides, metal oxides, and nitrides-have been extensively exploited for photocatalytic H 2 production [9][10][11]. Among the developed photocatalysts, CdIn 2 S 4 (CIS) is of great interest [12][13][14]. Nevertheless, for bare CIS, the low carrier separation efficiency and lack of active site still need to be urgently solved. To address those issues, various methods have been carried out to obtain elevated photocatalytic activity of CdIn 2 S 4 -based photocatalysts, such as tuning the morphologies [15], constructing heterojunctions [16,17], doping metal or nonmetal elements [18,19], and so on. He et al. [20] prepared ultra-thin CdIn 2 S 4 nanosheets to acquire efficient photocatalytic activity. Chen et al. [21] designed CdIn 2 S 4 /TiO 2 Z-scheme heterojunction with high carrier separation efficiency for photocatalytic H 2 production. Yu et al. [17] reported that PdS-loaded ZnIn 2 S 4 /CdIn 2 S 4 flower-like microspheres had significantly improved photocatalytic activity and high-efficiency stability under aqueous Na 2 SO 3 and Na 2 S solution. Guo et al. [12] prepared CdIn 2 S 4 /CNFs/Co 4 S 3 nanofiber networks with efficient charge separation and adequate active sites for H 2 production using solar-driven water splitting. Additionally, cocatalysts are critical for booting carrier separation efficiency and causing the active site of H 2 production to increase the activity of photocatalysts. Proverbially,

Results and Discussion
The morphologies of all samples were obtained via scanning electron microscopy (SEM), transmission electron microscopy (TEM), and high-resolution transmission electron microscopy (HRTEM). Figure 1A showed that CdIn 2 S 4 had nanoparticle structures with sizes of 25-180 nm, and Figure 1B displayed that Mo 2 C had a bulk structure about 1-2 µm in size. Figure 2A,B also showed that CdIn 2 S 4 and Mo 2 C had nanoparticle structures and bulk structures, which confirmed the results of scanning electron microscopy. Figure 1C clearly revealed that CdIn 2 S 4 was deposited on the surface of Mo 2 C. Simultaneously, compared to the pure molybdenum carbide, Figure 2C shows that the massive molybdenum carbide edge had many CdIn 2 S 4 particles, which also confirmed the results of the scanning electron microscopy. The results of high-resolution transmission electron microscopy were analyzed to further confirm the formation of MS heterojunction composites. Figure 2D,E represented that the lattice fringe spacings of (3 1 1) planes (CdIn 2 S 4 ) and (1 0 1) planes (Mo 2 C) were 0.33 nm and 0.21 nm, respectively. More significantly, Figure 2F possessed two lattice fringes of 0.33 nm and 0.21 nm assigned to the (3 1 1) planes of CdIn 2 S 4 [31] and the (1 0 1) planes of Mo 2 C [27], respectively, which demonstrated the simultaneous presence of CdIn 2 S 4 and Mo 2 C in the MS heterojunction composites. In a word, the aforementioned results corroborated the successful synthesis of the heterojunction composites.  In Figure 3, the optical properties of CdIn 2 S 4 , Mo 2 C, and the 5MS hybrid were discreetly studied via UV-Vis diffuse reflectance spectra. As expected, CdIn 2 S 4 possessed an evident edge at about 575 nm that was derived from its transition of the band [15]. Bare Mo 2 C was black and revealed a broad and strong visible light absorption capacity from 300 nm to 800 nm. Interestingly, it was found that the absorption edge of heterojunction composites extended to longer wavelength regions compared to bare CdIn 2 S 4 . Meanwhile, the visible light absorption intensity of heterojunction composites was visibly boosted at 500-800 nm with increasing Mo 2 C content. The stronger visible light absorption capacity plays a pivotal role in achieving solar energy conversion. Based on our previous research [27], the band gaps (E g ) of Mo 2 C and CdIn 2 S 4 were 1.1 eV and 2.3 eV, respectively, which conform to the literature values well [29,31]. It can be clearly seen that the band gap of MS heterojunction composites is reduced compared to CdIn 2 S 4 in Figure S1. As shown in Figure 3B, at a frequency of 1 kHz, Mott-Schottky (M-S) tests were carried out to analyze the type of semiconductors and the charge transfer process. The straight lines of Mo 2 C and CdIn 2 S 4 both have positive slopes, which suggests that they are intrinsic n-type characteristics. Additionally, the conduction band potential values are confirmed on the basis of extrapolation of the straight line toward the x-axis. As everyone knows, the flat band potential is around equal to the conduction band potential for semiconductors with intrinsic n-type characteristics. In line with the intercept of the M-S plots ( Figure 3B,C), the conduction band potentials (E CB ) of Mo 2 C and CdIn 2 S 4 are and −0.27 V and −0.74 V, respectively. Obviously, the conduction band potential of Mo 2 C is more positive than the conduction band potential of CdIn 2 S 4 , and this difference in energy level will help drive electron transfer from the conduction band potential of CdIn 2 S 4 to the conduction band potential of Mo 2 C for photocatalytic hydrogen production. Therefore, on the basis of the following formula, the valence bands (E VB ) of Mo 2 C and CdIn 2 S 4 are 0.83 V and 1.56 V, respectively. The phase of prepared samples is authenticated and discussed via X-ray diffraction patterns (XRD), and the results are illustrated in Figure 4. All as-prepared samples possessed sharp diffraction peaks, indicating excellent crystallinity. The patterns of CdIn 2 S 4 and Mo 2 C were confirmed to be cubic (JCPDF: 27-0060) and hexagonal (JCPDF: 35-0787) structures, respectively [26,32]. The peaks around 2θ of 14.  [27]. The peak intensity of Mo 2 C significantly increased along with the elevation of the Mo 2 C content, which was advantageous for photocatalytic hydrogen production and had been studied in previous reports. The X-ray diffraction patterns (XRD) results demonstrated that MS composites were successfully fabricated, which is consistent with scanning electron microscopy (SEM), transmission electron microscopy (TEM), and high-resolution transmission electron microscopy (HRTEM) results. Subsequently, we also utilized X-ray photoelectron spectroscopy (XPS) to further analyze and study the composition of the photocatalysts as well as the strong electronic interactions between CdIn2S4 and Mo2C. Figure 5A displays the full XPS survey of 5MS composites, and the full XPS survey explains the existence of Mo, Cd, C, S, and In elements in the ternary composites, which shows that the heterojunction composites are prepared successfully. In Figure 5B, for bare CdIn2S4, the peaks at 412.2 eV and 405.5 eV correspond to Cd 3d3/2 and Cd 3d5/2, respectively, and In 3d3/2 and In 3d5/2, respectively, which are the characteristic peaks of Cd 2+ and In 3+ in CdIn2S4. In Figure 5C, for bare CdIn2S4, the peaks at 452.5 eV and 444.9 eV correspond to In 3d3/2 and In 3d5/2, respectively, which are the characteristic peaks of In 3+ in CdIn2S4. In Figure 5B, for 5MS composites, the peaks at 412.1 eV and 405.3 eV correspond to Cd 3d3/2 and Cd 3d5/2, respectively, which are the characteristic peaks of Cd 2+ in the ternary composites. In Figure 5C, for 5MS composites, the peaks at 452.3 eV and 444.8 eV correspond to In 3d3/2 and In 3d5/2, respectively, which are the characteristic peaks of In 3+ in the ternary composites. In Figure 5D, the S 2p data of CdIn2S4 centered at binding energies of 162.9 eV and 161.7 eV match with S 2p1/2 and S 2p3/2, respectively. In Figure 5D, the S 2p data of 5MS heterojunction composites centered at binding energies of 162.8 eV and 161.6 eV match with S 2p1/2 and S 2p3/2, respectively. Nevertheless, a distinct peak at about 169.0 eV could be seen, distributing SO4 2-resulting from hydrothermal processes in Figure 5D. For Mo 3d spectra of bare Mo2C ( Figure 5E), there were three peaks of Mo 3d in Mo2C situate at 233.0 eV, 228.7 eV, and 236.1 eV, corresponding to Mo 3d3/2, Mo 3d5/2, and Mo-O band, respectively. It can also be seen that Mo mainly exists in three forms (Mo 3d5/2 and 227.7 eV, Mo 3d3/2 and 232.5 eV, Mo-O and 235.7 eV) in 5MS composites in Figure 5E. Furthermore, in Figure 5E, there is a distinct peak of 225.9 eV in 5MS composites, which can be attributed to S 2s, which is also consistent with our previous research and what others have reported. In Figure 5F, for C 1s spectra of Mo2C, in addition to the carbon standard peak (284.8 eV), there are two peaks located at 286.4 eV and 288.9 eV, assigned to C-O and C=O, respectively. The binding energies of C-O and C=O in 5MS composites are 286.8 eV and 288.9 eV, respectively, in Subsequently, we also utilized X-ray photoelectron spectroscopy (XPS) to further analyze and study the composition of the photocatalysts as well as the strong electronic interactions between CdIn 2 S 4 and Mo 2 C. Figure 5A displays the full XPS survey of 5MS composites, and the full XPS survey explains the existence of Mo, Cd, C, S, and In elements in the ternary composites, which shows that the heterojunction composites are prepared successfully. In Figure 5B, for bare CdIn 2 S 4 , the peaks at 412.2 eV and 405.5 eV correspond to Cd 3d 3/2 and Cd 3d 5/2 , respectively, and In 3d 3/2 and In 3d 5/2 , respectively, which are the characteristic peaks of Cd 2+ and In 3+ in CdIn 2 S 4 . In Figure 5C, for bare CdIn 2 S 4 , the peaks at 452.5 eV and 444.9 eV correspond to In 3d 3/2 and In 3d 5/2 , respectively, which are the characteristic peaks of In 3+ in CdIn 2 S 4 . In Figure 5B, for 5MS composites, the peaks at 412.1 eV and 405.3 eV correspond to Cd 3d 3/2 and Cd 3d 5/2 , respectively, which are the characteristic peaks of Cd 2+ in the ternary composites. In Figure 5C, for 5MS composites, the peaks at 452.3 eV and 444.8 eV correspond to In 3d 3/2 and In 3d 5/2 , respectively, which are the characteristic peaks of In 3+ in the ternary composites. In Figure 5D, the S 2p data of CdIn 2 S 4 centered at binding energies of 162.9 eV and 161.7 eV match with S 2p 1/2 and S 2p 3/2 , respectively. In Figure 5D, the S 2p data of 5MS heterojunction composites centered at binding energies of 162.8 eV and 161.6 eV match with S 2p 1/2 and S 2p 3/2 , respectively. Nevertheless, a distinct peak at about 169.0 eV could be seen, distributing SO 4 2− resulting from hydrothermal processes in Figure 5D. For Mo 3d spectra of bare Mo 2 C ( Figure 5E), there were three peaks of Mo 3d in Mo 2 C situate at 233.0 eV, 228.7 eV, and 236.1 eV, corresponding to Mo 3d 3/2 , Mo 3d 5/2 , and Mo-O band, respectively. It can also be seen that Mo mainly exists in three forms (Mo 3d 5/2 and 227.7 eV, Mo 3d 3/2 and 232.5 eV, Mo-O and 235.7 eV) in 5MS composites in Figure 5E. Furthermore, in Figure 5E, there is a distinct peak of 225.9 eV in 5MS composites, which can be attributed to S 2s, which is also consistent with our previous research and what others have reported. In Figure 5F, for C 1s spectra of Mo 2 C, in addition to the carbon standard peak (284.8 eV), there are two peaks located at 286.4 eV and 288.9 eV, assigned to C-O and C=O, respectively. The binding energies of C-O and C=O in 5MS composites are 286.8 eV and 288.9 eV, respectively, in Figure 5F. After the hybridization of CdIn 2 S 4 and Mo 2 C, the binding energies of all elements have shifted slightly compared to pure CdIn 2 S 4 and Mo 2 C, indicating a strong interaction between CdIn 2 S 4 and Mo 2 C, which is very advantageous for photocatalytic hydrogen production [33,34]. The X-ray photoelectron spectroscopy (XPS) results support the X-ray diffraction patterns (XRD), scanning electron microscopy (SEM), transmission electron microscopy (TEM), and high-resolution transmission electron microscopy (HRTEM) results. Figure 5F. After the hybridization of CdIn2S4 and Mo2C, the binding energies of all elements have shifted slightly compared to pure CdIn2S4 and Mo2C, indicating a strong interaction between CdIn2S4 and Mo2C, which is very advantageous for photocatalytic hydrogen production [33,34]. The X-ray photoelectron spectroscopy (XPS) results support the X-ray diffraction patterns (XRD), scanning electron microscopy (SEM), transmission electron microscopy (TEM), and high-resolution transmission electron microscopy (HRTEM) results. The photocatalytic behavior of all samples was assessed via photocatalytic experiment under visible light irradiation (λ ≥ 420 nm) ( Figure 6). CdIn2S4 revealed low photocatalytic activity due to the scarce active site and rapid carrier recombination. Additionally, for Mo2C, there was no photocatalytic activity because of the rapid carrier recombination resulting from a narrow band gap, coinciding well with the results reported [35]. Extraordinarily, after CdIn2S4 was incorporated with Mo2C, the photocatalytic activities of MS composites increased sharply. The above H2 production process also confirms the results of photocatalysis and electrochemistry, that is, after the combination of Mo2C and CdIn2S4, not only was the photogenerated carrier separation efficiency improved, but the H2 production activity was surprisingly increased as well. In Figure 6A, the amount of H2 produced was severally 973.4 μmol/g (1MS), 3681.5 μmol/g (3MS), 6313.6 μmol/g (5MS), and 4596.9 (7MS) μmol/g in 5 h. In Figure 6B, the corresponding average H2 production rate of MS composites was 181.67 μmol h −1 g −1 , 687.08 μmol h −1 g −1 , 1178.32 μmol h −1 g −1 , and 857.94 μmol h −1 g −1 , respectively. Notably, when the content of Mo2C increased, the H2 production rates displayed a volcano-shaped photoactivity trend. The best weight ratio of Mo2C was determined to be 5 wt%, and the photocatalytic activity could reach the maximum value. This obvious improvement might stem from (i) Mo2C acting as active sites for water reduction; (ii) the separation efficiency of MS heterojunction structures being distinctly boosted. However, after the content of Mo2C exceeds 5%, the photocatalytic activity gradually reduced. This is because high Mo2C can not only block visible light absorption of CdIn2S4 but can also serve as the recombination center of carriers, resulting in the suppression of electron-holes separation [30,36]. The above results imply the appropriate Mo2C is significant for optimizing the photocatalytic activity of photocatalyst. Moreover, as shown in Figure 6B, the stability The photocatalytic behavior of all samples was assessed via photocatalytic experiment under visible light irradiation (λ ≥ 420 nm) ( Figure 6). CdIn 2 S 4 revealed low photocatalytic activity due to the scarce active site and rapid carrier recombination. Additionally, for Mo 2 C, there was no photocatalytic activity because of the rapid carrier recombination resulting from a narrow band gap, coinciding well with the results reported [35]. Extraordinarily, after CdIn 2 S 4 was incorporated with Mo 2 C, the photocatalytic activities of MS composites increased sharply. The above H 2 production process also confirms the results of photocatalysis and electrochemistry, that is, after the combination of Mo 2 C and CdIn 2 S 4 , not only was the photogenerated carrier separation efficiency improved, but the H 2 production activity was surprisingly increased as well. In Figure 6A, the amount of H 2 produced was severally 973.4 µmol/g (1MS), 3681.5 µmol/g (3MS), 6313.6 µmol/g (5MS), and 4596.9 (7MS) µmol/g in 5 h. In Figure 6B, the corresponding average H 2 production rate of MS composites was 181.67 µmol h −1 g −1 , 687.08 µmol h −1 g −1 , 1178.32 µmol h −1 g −1 , and 857.94 µmol h −1 g −1 , respectively. Notably, when the content of Mo 2 C increased, the H 2 production rates displayed a volcano-shaped photoactivity trend. The best weight ratio of Mo 2 C was determined to be 5 wt%, and the photocatalytic activity could reach the maximum value. This obvious improvement might stem from (i) Mo 2 C acting as active sites for water reduction; (ii) the separation efficiency of MS heterojunction structures being distinctly boosted. However, after the content of Mo 2 C exceeds 5%, the photocatalytic activity gradually reduced. This is because high Mo 2 C can not only block visible light absorption of CdIn 2 S 4 but can also serve as the recombination center of carriers, resulting in the suppression of electron-holes separation [30,36]. The above results imply the appropriate Mo 2 C is significant for optimizing the photocatalytic activity of photocatalyst. Moreover, as shown in Figure 6B, the stability tests of photocatalytic hydrogen production for 5MS composites are also determined. There was a slight drop in hydrogen production per cycle, probably due to the loss of the photocatalyst during filtration and washing. Significantly, in three photocatalytic tests, 5MS composites reveal highly stable photocatalytic performance ( Figure 6C). The comparison of photocatalytic performance between this work and the current work with similar work was in Supplementary Materials Table S2.
Molecules 2023, 28, x FOR PEER REVIEW 7 of 13 tests of photocatalytic hydrogen production for 5MS composites are also determined. There was a slight drop in hydrogen production per cycle, probably due to the loss of the photocatalyst during filtration and washing. Significantly, in three photocatalytic tests, 5MS composites reveal highly stable photocatalytic performance ( Figure 6C). The comparison of photocatalytic performance between this work and the current work with similar work was in Supplementary Materials Table S2. To explore the carrier separation of photocatalysts, the PL and a series of electrochemical tests are obtained and studied (Figure 7). Generally, photoluminescence (PL) spectroscopy is one of the commonly used methods for evaluating the efficiency of photo-generated electron-hole pair charge separation. Due to its intrinsic properties, different photocatalysts have different emission spectra. For photocatalysts, the lower PL emission peak intensity means the lower recombination of photogenerated charge [37]. Figure 7A display that MS composites possess lower peak than bare CdIn2S4, which means that composites have higher carrier separating efficiency after the CdIn2S4 is supported on the surface of Mo2C cocatalyst. Therefore, the addition of Mo2C can increase the carrier separation of the catalyst. Figure 7B reveals the transient photocurrent responses (TPR) of all photocatalysts to investigate the charge transfer under simulated solar irradiation. Under visible light, the photocurrent responses of all photocatalysts were improved. As expected, it can be seen that all MS composites have stronger photocurrent than CdIn2S4, effectively implying charge transfer for composites [38]. The order of photocurrent intensity is: CdIn2S4 < 1MS < 3MS < 7MS < 5MS. The orders are in keeping with the photocatalytic H2 production results. The higher photocurrent indicates faster and more efficient emigration from CB of CdIn2S4 to CB of Mo2C, followed by a reduction reaction. In addition, the photocurrent result is in good agreement with results of PL above and EIS below. The electrochemical impedance spectroscopy (EIS) was measured and recorded on the open circuit potential under visible light. The EIS Nyquist plot is also used to understand the charge transfer of photocatalysts [39]. As shown in Figure 7C, compared to pure CdIn2S4, the semicircle curves of EIS for the MS heterojunction composites are shown, which indicates that photoinduced carriers are tardy recombination and affect migration with the addition of Mo2C. The order of radius is: 5MS < 7MS < 3MS < 1MS < CdIn2S4, matching with the photocurrent results. The illustration in Figure 7C is a model circuit. Moreover, 5MS composites have the smallest arcs compared to other samples. This result implied 5MS composites' presence faster than electron migration and interface electron migration resistance. In summary, the PL and electrochemical results of all samples illustrate that Mo2C could speed up charge separation efficiency to obtain many more free electrons for photocatalytic hydrogen production. To explore the carrier separation of photocatalysts, the PL and a series of electrochemical tests are obtained and studied (Figure 7). Generally, photoluminescence (PL) spectroscopy is one of the commonly used methods for evaluating the efficiency of photogenerated electron-hole pair charge separation. Due to its intrinsic properties, different photocatalysts have different emission spectra. For photocatalysts, the lower PL emission peak intensity means the lower recombination of photogenerated charge [37]. Figure 7A display that MS composites possess lower peak than bare CdIn 2 S 4 , which means that composites have higher carrier separating efficiency after the CdIn 2 S 4 is supported on the surface of Mo 2 C cocatalyst. Therefore, the addition of Mo 2 C can increase the carrier separation of the catalyst. Figure 7B reveals the transient photocurrent responses (TPR) of all photocatalysts to investigate the charge transfer under simulated solar irradiation. Under visible light, the photocurrent responses of all photocatalysts were improved. As expected, it can be seen that all MS composites have stronger photocurrent than CdIn 2 S 4 , effectively implying charge transfer for composites [38]. The order of photocurrent intensity is: CdIn 2 S 4 < 1MS < 3MS < 7MS < 5MS. The orders are in keeping with the photocatalytic H 2 production results. The higher photocurrent indicates faster and more efficient e − migration from CB of CdIn 2 S 4 to CB of Mo 2 C, followed by a reduction reaction. In addition, the photocurrent result is in good agreement with results of PL above and EIS below. The electrochemical impedance spectroscopy (EIS) was measured and recorded on the open circuit potential under visible light. The EIS Nyquist plot is also used to understand the charge transfer of photocatalysts [39]. As shown in Figure 7C, compared to pure CdIn 2 S 4 , the semicircle curves of EIS for the MS heterojunction composites are shown, which indicates that photoinduced carriers are tardy recombination and affect migration with the addition of Mo 2 C. The order of radius is: 5MS < 7MS < 3MS < 1MS < CdIn 2 S 4 , matching with the photocurrent results. The illustration in Figure 7C is a model circuit. Moreover, 5MS composites have the smallest arcs compared to other samples. This result implied 5MS composites' presence faster than electron migration and interface electron migration resistance. In summary, the PL and electrochemical results of all samples illustrate that Mo 2 C could speed up charge separation efficiency to obtain many more free electrons for photocatalytic hydrogen production.
Based on band arrangement and characterization results, the probable photocatalytic mechanism of heterojunction composites is designed and explained to verify the elementary reason for the distinct amendment of the photocatalytic performance of the MS heterojunction photocatalyst (Figure 8). The low H 2 activity of CdIn 2 S 4 may be due to the scarce active site and rapid carrier recombination. In the MS heterojunction, Mo 2 C is an electron capture trap, which can quickly extract photoinduced electrons generated in CdIn 2 S 4 , accelerate the separation of photoinduced electron-hole pairs and reach higher photocatalytic hydrogen production. Clearly, a Type I heterojunction could be formed because the conduction band of Mo 2 C is more positive than that of CdIn 2 S 4 and the valence band of Mo 2 C is more negative than that of CdIn 2 S 4 . The electron of the valence band for Mo 2 C and CdIn 2 S 4 could be stimulated to the conduction band by visible light. Due to excellent electrical conductivity, Mo 2 C can rapidly capture the electrons of CdIn 2 S 4 before the charge recombination, so it appears that the electrons of conduction band for CdIn 2 S 4 transferred to the conduction band of Mo 2 C. Electrons on Mo 2 C could reduce water to hydrogen (H + + e − → 0.5H 2 ). In the public eye, the separation and recombination of charge carrier are two competitive processes. Considering the different transfer rates of e − and h + from CdIn 2 S 4 to Mo 2 C and the excellent metallic conductivity of Mo 2 C, the h + in the VB of CdIn 2 S 4 partially transfers to the valence band of Mo 2 C, and the remaining h + in the valence band of CdIn 2 S 4 and the h + in the valence band of Mo 2 C reacts with lactic acid, thereby reducing the surface charge recombination. The formation of a Type I heterojunction greatly improves the separation rate of carriers, demonstrated clearly by PL and electrochemical results. This obvious improvement might stem from (i) Mo 2 C acting as active sites for water reduction and (ii) the separation efficiency of MS heterojunction structures being distinctly boosted. Based on these results, synergetic modification of carrier separation and active sites result in the remarkably elevated photocatalytic activity of MS heterojunction composites. Based on band arrangement and characterization results, the probable photocatalytic mechanism of heterojunction composites is designed and explained to verify the elementary reason for the distinct amendment of the photocatalytic performance of the MS heterojunction photocatalyst (Figure 8). The low H2 activity of CdIn2S4 may be due to the scarce active site and rapid carrier recombination. In the MS heterojunction, Mo2C is an electron capture trap, which can quickly extract photoinduced electrons generated in CdIn2S4, accelerate the separation of photoinduced electron-hole pairs and reach higher photocatalytic hydrogen production. Clearly, a Type Ⅰ heterojunction could be formed because the conduction band of Mo2C is more positive than that of CdIn2S4 and the valence band of Mo2C is more negative than that of CdIn2S4. The electron of the valence band for Mo2C and CdIn2S4 could be stimulated to the conduction band by visible light. Due to excellent electrical conductivity, Mo2C can rapidly capture the electrons of CdIn2S4 before the charge recombination, so it appears that the electrons of conduction band for CdIn2S4 transferred to the conduction band of Mo2C. Electrons on Mo2C could reduce water to hydrogen (H + + e − → 0.5H2). In the public eye, the separation and recombination of charge carrier are two competitive processes. Considering the different transfer rates of eand h + from CdIn2S4 to Mo2C and the excellent metallic conductivity of Mo2C, the h + in the VB of CdIn2S4 partially transfers to the valence band of Mo2C, and the remaining h + in the valence band of CdIn2S4 and the h + in the valence band of Mo2C reacts with lactic acid, thereby reducing the surface charge recombination. The formation of a Type Ⅰ heterojunction greatly improves the separation rate of carriers, demonstrated clearly by PL and electrochemical results. This obvious improvement might stem from (i) Mo2C acting as active sites for water reduction and (ii) the separation efficiency of MS heterojunction structures being distinctly boosted. Based on these results, synergetic    Synthesis of CdIn 2 S 4: Amounts of 2 mmol In(NO 3 ) 3 ·3H 2 O, 4 mmol thioacetamide, and 1 mmol Cd(NO 3 ) 2 ·4H 2 O were sequentially added to 70 mL deionized water. After stirring for 1 h, turbid liquids were transferred to a Teflon-lined steel autoclave. The resulting reaction was heated at 180 • C for 24 h. After cooling to room temperature, the products were collected via centrifugation, washed with deionized water and ethanol several times, and then dried at 60 • C for 12 h.
Synthesis of Mo 2 C-CdIn 2 S 4: Amounts of 2 mmol In(NO 3 ) 3 ·3H 2 O, 0.1175 g Mo 2 C, 4 mmol thioacetamide, and 1 mmol Cd(NO 3 ) 2 ·4H 2 O were sequentially added in 70 mL deionized water. Then, the turbid liquids were kept at 180 • C for 24 h. After cooling to room temperature, the products were collected via centrifugation, washed with deionized water and ethanol several times, and then dried at 60 • C for 12 h. The heterojunction composites containing 0.2 g of CdIn 2 S 4 with 0.01 g of Mo 2 C were labeled as 5MS photocatalysts. The other Mo 2 C/CdIn 2 S 4 photocatalysts were prepared by introducing different amounts of Mo 2 C into the solution with other reaction parameters fixed. Similarly, the samples of CdIn 2 S 4 with 1, 3, and 7 mass percent Mo 2 C are labeled as 1MS, 3MS, and 7MS, respectively. The additional experiments for optimization of the amounts and conditions were not used in the synthesis process.

Characterization
The crystal structures of pristine Mo 2 C, CdIn 2 S 4 , and Mo 2 C/CdIn 2 S 4 were tested using Cu Ka radiation (I = 1.5406 Å, 40 kV and 40 mA) in X-ray diffraction patterns (XRD, D/Max-RB, Rigaku, Japan). Transmission electron microscopy (TEM), high-resolution transmission electron microscopy (HRTEM) (F-20, FEI, Hillsboro, OR, USA), and scanning electron microscopy (SEM) (S-4800; Hitachi, Tokyo, Japan) images were obtained. UV-Vis diffuse reflectance spectra (DRS) were carried out using a T9S spectrophotometer with BaSO 4 as a reflectance standard. X-ray photoelectron spectroscopy (XPS) analysis was examined with a monochromatic X-ray source manufactured on an X-ray photoelectron spectrometer (Thermo Fisher Scientific K-Alpha, Waltham, MA, USA). The photoluminescence (PL) was observed using a fluorescence spectrophotometer (F-4500, Hitachi, Tokyo, Japan) with a Xe lamp as the excitation light source.

Electrochemical Measurements
The electrochemical impedance spectroscopy (EIS)-photocurrent measurement (TPR) curves of photocatalysts were obtained using a CHI660E electrochemical workstation. The samples, a saturated calomel electrode (SCE), and a Pt wire were employed as the working electrode, reference electrode, and counter electrode, respectively. All electrochemical tests used incandescent lamps under visible light. The aqueous solution of Na 2 SO 4 (0.5 mol L −1 ) served as the electrolyte. In total, 5 mg of photocatalyst was suspended in mixed solution with ethanol and Nafion. The as-prepared samples were dispersed into a circle with a diameter of 6 mm on the bottom-middle of an ITO glass matrix using a micropipette and dried at room temperature.

Photocatalytic H 2 Evolution Test
The photocatalytic H 2 evolution reaction was operated in a typical reaction system according to our previous reports [40][41][42]. The photocatalytic experiment was carried out under visible light (λ ≥ 420 nm). The information of used Xenon lamp and the intensity of the incident radiation entering the photoreactor were added in Supplementary Materials. First, 30 mL of lactic acid/H 2 O solution (10% (v/v) pH = 2.2) was prepared, then 30 mg samples were added in the above solution by ultrasound treatment for 30 min. Before irradiation, the air in reaction container was driven out through the high-purity argon (Ar) gas for 0.5 h. The amount of hydrogen generation was tested via gas chromatography (GC-7920, TCD). Furthermore, stability tests were also carried out to evaluate the stability of MS photocatalysts in three continuous experiments. For the next test, the samples were collected via centrifugation and washing thoroughly with ethanol and water several times and were then dried at 60 • C after every test for H 2 generation.

Conclusions
In this research, the novel noble-metal-free Mo 2 C/CdIn 2 S 4 composites were firstly developed via a facile fabrication process for highly efficient photocatalytic H 2 production. The structure, morphology, and performance of heterojunction composites were analyzed using different characterization techniques. The visible light photocatalytic properties of the Mo 2 C/CdIn 2 S 4 composites were investigated, and the best photocatalytic activity was up to 1178.32 µmmol h −1 g −1 for 5Mo 2 C/CdIn 2 S 4 composites. Notably, the heterojunction composites possessed high stability in three cycles. The remarkable elevated photocatalytic activity may be due to the accelerated separation efficiency and more active sites for photocatalytic water splitting. We believe that Mo 2 C, as noble-metal-free cocatalyst to modify other photocatalysts, has great potential for efficient photocatalytic hydrogen production.
Supplementary Materials: The following supporting information can be downloaded at: https://www.mdpi.com/article/10.3390/molecules28062508/s1, Figure S1: Tacu's curve of MS heterojunction composites; Table S1: The incident radiation intensity entering the photoreactor is shown in the table below; Table S2: Comparison of hydrogen evolution data of Mo2C/CdIn2S4 composites compared with other literature reports.