Characterization of MoS2/g-C3N4
The X-ray diffraction (XRD) patterns of the prepared samples are shown in Fig. 2a. The diffraction peak of g-C3N4 at 27.4° corresponds to the (200) crystal plane (JCPDS NO. 87-1522)26. The diffraction peaks of MoS2 at 32.9° and 57.7° correspond to the (100) and (110) crystal planes, respectively (JCPDS NO. 75-1539)27. Moreover, in the MoS2/g-C3N4 heterojunction material, the coexisting (200) crystal planes of g-C3N4 and the (100) and (110) crystal planes of MoS2 are clearly observed. These results confirm that the heterojunction material was successfully synthesized and that the structure of each component remained stable during the preparation and composite processes, providing a basis for heterojunction formation.
(a) XRD diffraction patterns and (b) Raman spectra of various samples.
The Raman spectra of the different samples are shown in Fig. 2b. Pure g-C3N4 exhibits two Raman peaks at 1352 cm− 1 and 1596 cm− 1; these correspond to the characteristic D and G peaks, respectively, of the g-C3N4 graphitic material28. Pure MoS2 displays two peaks at 380.3 cm− 1 and 407.8 cm− 1; these peaks are attributed to the in-plane (E12g) and out-of-plane (A1g) vibrational modes, respectively29. The MoS2/g-C3N4 composite also shows distinct D and G peaks along with two MoS2 bands; these results indicate that the structures of all components within the composite remain intact. These results further confirm the successful synthesis of the composite sample and are consistent with the XRD results. FTIR spectra (Fig. S1) similarly indicate the successful synthesis of MoS2/g-C3N4 composites. Peaks of 749 cm− 1 from MoS2 and 2986 cm− 1 from g-C3N4 can be observed on the composite.
SEM images of pure MoS2(a), g-C3N4 (b), MoS2/g-C3N4 (c) and TEM images of MoS2 (d), g-C3N4 (e), MoS2/g-C3N4 (f).
Scanning electron microscopy (SEM) and transmission electron microscopy (TEM) images of each prepared sample are shown in Fig. 3. Figure 3a shows an SEM image of the prepared pure MoS2; here, the image reveals a microspherical shape that is fluffy, porous, and free of agglomeration. The TEM image of MoS2 is shown in Fig. S4. Figure 3b presents an SEM image of pure g-C3N4, and a lamellar stacking structure is clearly observed. The SEM image of the composite sample MoS2/g-C3N4 is shown in Fig. 3c; here, the layers of lamellar g-C3N4 are attached to the surface of the spherical MoS2. Figure 3d shows a TEM image of pure MoS2, in which the spherical three-dimensional MoS2 appears as a black shaded portion due to its resistance to electron penetration. Figure 3e shows a TEM image of pure g-C3N4; here, the image reveals distinct multilayered lamellar stacking, and this result is consistent with the SEM results. Figure 3f shows the TEM image of the composite sample MoS2/g-C3N4; in this image, the multilayer lamellar g-C3N4 is anchored around the spherical MoS2, and this confirms the successful synthesis of the composite material and the formation of a heterogeneous structure.
(a) Nitrogen adsorption and desorption isotherms of MoS2, g-C3N4 and MoS2/g-C3N4 and (b) pore size distributions.
The nitrogen adsorption/desorption isothermal curves of each prepared material are shown in Fig. 4a. The isotherms of the three materials exhibit inflection points in the low-pressure region. In the high-pressure region, the adsorption and desorption curves do not overlap, displaying evident hysteresis loops, indicative of a clear Type IV curve. The specific surface areas of MoS2, g-C3N4 and MoS2/g-C3N4 are 166.36, 43.69 and 107.96 m2/g, respectively. The pore size distributions of the three samples shown in Fig. 4b reveal that the pore sizes of all three materials are in the range of 2–50 nm; these results indicate a standard mesoporous structure and are consistent with the H1 hysteresis loop observed in the adsorption/desorption isotherms. The pore size distributions of MoS2, g-C3N4, and MoS2/g-C3N4 are 11.26, 16.55, and 12.63 nm, respectively. The composites combine the large specific surface area of MoS2 with the large pore size of g-C3N4 to provide extensive reactive sites for the catalytic reactions.
X-ray photoelectron spectroscopy (XPS) results confirm the successful synthesis of MoS2/g-C3N4 composites (Fig. S3). The characteristic peaks of Mo 3d, C 1s, N 1s and S 2p orbitals observed in the composites proved the effective combination of the two components. The XPS spectra in the Mo 3d region, two different peaks were observed at the BE of 228.5 and 231.8 eV. In the XPS spectra of the S 2p region, two distinct main peaks are observed at BE of 161.6 and 162.7 eV. Notably, the binding energy shifts of Mo 3d at 231.8 eV and 228.5 eV indicate that the electronic structure of MoS2 changes significantly when interacting with g-C3N4, which suggests that there is a strong chemical bond between the two materials30. The spectrum of the C 1s region of the pure g-C3N4, MoS2/g-C3N4 sample exhibit two major peaks at BEs of 284.5 eV (attributed to reference C atoms) and 285.9 eV (attributed to C sp2-hybridized atoms of the terminal C(N)3 structure in tri-s-triazine aromatic units)31. The high-resolution XPS spectrum of the pristine g-C3N4 N1s region shows two peaks located at BEs of 398.7 eV and 400.8 eV, which can be indexed to the nitrogen atoms in C=N-C and -NH2 bonds, respectively32.
The results presented in Fig. S6 highlight important insights into the physicochemical properties of MoS2 and g-C3N4, as well as their composite MoS2/g-C3N4. Zeta potential measurements (Fig. S6a) indicate that MoS2/g-C3N4 composites present a distinctive waveforms, suggesting the existence of synergistic interactions between the components, which may enhance their electrochemical properties. In the corresponding hydrodynamic diameter analysis (Fig. S6b), the average diameter of MoS2 (1087.58 nm) is significantly larger than that of the composite (429.93 nm) and g-C3N4 (583.37 nm). The narrower distribution observed in the composites indicates improved dispersion, which contributes to better contact with the bacterial fluid in practical applications.
Photocatalytic antibacterial properties of MoS2/g-C3N4
Firstly, the look antimicrobial function of the composites with different ratios was done by pre-experiment as shown in Fig. S2, the results showed that MoS2/g-C3N4 (3:1) has more toxicity than MoS2/g-C3N4 (3:2) under dark condition for 30 min. Therefore, MoS2/g-C3N4 (3:2) was chosen for the subsequent experiments. The photocatalytic bacterial inactivation activity of the synthesized MoS2/g-C3N4 heterostructures was evaluated using E. coli and S. aureus. Notably, almost no bacterial inactivation was observed in the dark (dark control) or under light irradiation without photocatalysts (light control) for either E. coli or S. aureus. The bare C3N4 exhibited negligible bactericidal efficiency under white LED light irradiation. Impressively, the addition of MoS2/g-C3N4 or MoS2, both of which have broader visible light absorption, significantly enhanced photocatalytic antibacterial activity, especially the MoS2/g-C3N4 heterostructure. As shown in Fig. 5a-b, in the spread plate assays, no visible colonies were observed in the MoS2/g-C3N4 group after 30 min of irradiation under LED light. The improved photocatalytic performance compared with that of the pure MoS2 group indicates that the formation of heterogeneous structures promotes photocatalysis. Moreover, the antibacterial rate towards E. coli (Fig. 6a) was slightly greater than that towards S. aureus (Fig. 6b) because of the differences in cell wall structure.
Photocatalytic performance of MoS2/g-C3N4 against E. coli (a) and S. aureus (b).
The durability of a photocatalyst across multiple antimicrobial cycles is crucial for practical water disinfection. As shown in Fig. 6c, E. coli was completely inactivated within three antimicrobial cycles, each lasting 30 min. Although the antimicrobial resistance slightly decreased in the fourth and fifth cycles, it remained significant. These results highlight the high stability and robustness of MoS2/g-C3N4 in the photocatalytic inactivation of bacteria.
Photocatalytic antibacterial activity of MoS2/g-C3N4 heterostructure under white LED light irradiation and control experiments. The inactivation performances against E. coli (a) and S. aureus (b) and the corresponding spread plate results of E. coli and S. aureus, respectively. (c) Recycling experiments of photocatalytic inactivation of E. coli over MoS2/g-C3N4. (d) Photocatalytic antibacterial activity of MoS2/g-C3N4 in the presence of different scavengers under white LED irradiation.
To determine the primary reactive species for photocatalytic bacterial inactivation, the scavenger used by Zhang et al. was selected33. Various scavengers have been introduced into the system: Cr(VI) for electrons (e−), isopropanol for hydroxyl radicals (·OH), sodium oxalate for holes (h+), 4-hydroxy-2,2,6,6-tetramethylpiperidinyloxy (TEMPO) for superoxide radicals (·O2−), and catalase for hydrogen peroxide (H2O2)34,35. As shown in Fig. 6d, the addition of Cr(VI) to capture e− significantly reduced the photocatalytic antibacterial activity, and no antibacterial activity was observed; these results indicated that the enhancement was primarily due to the improved separation efficiency of photogenerated e−_h+. The slight decrease in antibacterial activity upon the addition of isopropanol indicated that ·OH played a minor role. The pronounced inhibition of bacterial inactivation observed with the addition of sodium oxalate, TEMPO, and catalase indicated that h+, ·O2−, and H2O2 were crucial species for photocatalytic antibacterial activity, particularly active H2O2.
As shown in Fig. 7, the red and green fluorescence in the stained cell images indicate dead and live S. aureus, respectively. Without visible light irradiation, the bacteria clearly exhibited green fluorescence with minimal red fluorescence, indicating that many bacteria survived. Upon visible light irradiation, the images displayed red fluorescence, indicating the death of all the Staphylococcus aureus bacteria. Thus, this heterogeneous structure has a photocatalytic sterilization effect.
As shown in Fig. S5, the composite MoS2/g-C3N4 has good biocompatibility, which was achieved by haemolysis experiments, and the haemolysis rates of pure MoS2, g-C3N4 and MoS2/g-C3N4 were all under 3%, which demonstrated its biosafety.
Merged fluorescence-stained images of S. aureus; green and red indicate live and dead S. aureus, respectively.
Photocatalytic activity enhancement mechanism of MoS2/g-C3N4
Figure 8 shows the mechanism underlying the enhanced photocatalytic performance of the synthesized MoS2/g-C3N4 heterostructures. As shown in the left panel of Fig. 8a, the work functions (WFs) of MoS2 and g-C3N4 are calculated from the Ultraviolet Photoelectron Spectroscopy (UPS) data and are 2.74 eV and 5.15 eV, respectively. The work function of MoS2 is much lower, and MoS2 acts as an electron transfer channel to attain efficient charge transfer. The right figure shows the Fermi edge of the sample; here, the valence band top of the sample can be derived, the valence band (VB) of the sample can be derived by the conversion of the vacuum energy level, and the valence bands of MoS2 and g-C3N4 are 1.85 eV and 3.07 eV, respectively. The Mott Schottky curve of the sample is shown in Fig. 8b; thus, MoS2 is an n-type semiconductor, the tangent slope of the M-S curve is positive, and the flat band potential EFB is -0.36 eV, and its conduction band ECB is approximately − 0.36 eV – (0.1–0.3 eV). Additionally, g-C3N4 has a negative M-S curve tangent slope and is a p-type heterojunction, and the flat-band potential EFB is 0.22 eV; this value indicates that the conduction band of g-C3N4 is approximately 0.22 eV + (0.1–0.3 eV). Figure 8c shows that the band gap of g-C3N4 (Eg) of g-C3N4 is 2.62 eV, and its conduction band is 0.45 eV from Eq. (1); these results are consistent with the deduction from the Mott Schottky results. Since MoS2 is a black material, its UV‒vis diffuse reflection is difficult to attain; thus, the approximate position of the conduction band of MoS2 is derived from the M‒S curve to further explore its photocatalytic performance. Under white LED light irradiation, MoS2 generates photogenerated electron‒hole pairs, where electrons are excited from the valence band to the conduction band, and holes remain in the valence band. The electrons migrate from MoS2 to g-C3N4, whereas the holes move from g-C3N4 to MoS2; thus, carrier recombination is prevented, and the photocatalytic activity is significantly enhanced. During the photocatalytic antimicrobial process, ·O2− radicals and H2O2, produced by the reaction of electrons with O2 and H2O, play crucial roles in the sterilization process36, as shown in Eq. (2).
$${\text{E}}_{{{\text{VB}}}} = {\text{E}}_{{{\text{CB}}}} + {\text{ E}}_{{\text{g}}}.$$
(1)
$${\text{e}}^{ – } + {\text{O}}_{{\text{2}}} \to \cdot{\text{O}}_{{\text{2}}} ^{ – } + {\text{H}}_{{\text{2}}} {\text{O}} \to {\text{H}}_{{\text{2}}} {\text{O}}_{{\text{2}}}.$$
(2)
(a) Tauc plots and work functions of MoS2 and g-C3N4, (b) Mott‒Schottky plots, (c) UV‒Vis diffuse reflectance spectra, and (d) proposed mechanism for the photocatalytic bacterial inactivation by MoS2/g-C3N4.