New g-C₃N₄/GO/MoS₂ composites as efficient photocatalyst for photocathodic protection of 304 stainless steel

With the rapid development of industry, energy shortage and environmental pollution have become the vital problems faced by human beings (Gao et al. 2018; He et al. 2018; Zhou et al. 2019). The utilization of solar energy has become a good solution to these serious problems. Therefore, the semiconductor photocatalyst has attracted increasing attention, and includes CO₂ photoreduction (Hashemizadeh et al. 2018; Xu et al. 2018), hydrogen generation (Mo et al. 2019), pollutant degradation (Xie et al. 2018), and photocathodic protection (PCP) (Guan et al. 2018b). Stainless steel materials can withstand high temperatures, oxidation resistance, and have superior mechanical properties, so they are widely used in industry, especially in environmental protection equipment, including delivery pipes, valves, water treatment equipment, and circulating cooling water systems (Ryan et al. 2002). However, stainless steel still has the risk of corrosion. Many factors induce corrosion in the water treatment system, including the microbial community in the sewage, low pH, and high salt concentration (Cui et al. 2016). Once the stainless steel device fails, it will cause huge economic losses. Therefore, it is very important to propose a reasonable anti-corrosion method (Liu et al. 2014; Chen et al. 2019). Compared to many metal anticorrosion methods, such as corrosion inhibitors (Fernandes et al. 2019; Sadeghi Erami et al. 2019), coating (Kiosidou et al. 2018; Rodriguez et al. 2018), and cathodic protection of sacrificial anodes (Zhang et al. 2012; Guan et al. 2018b), PCP is an economical and environmental method with promising application prospects. Normally, a semiconductor is coated on the protected metal surface or acts as an anode through wires connected to the protected metal (Ren et al. 2016). Electrons of the semiconductor in the valence band (VB) excited by light, migrate into the conduction band (CB), and the corresponding holes generate in the VB (Jing et al. 2019). At the interface between semiconductor thin film and the solution, the holes (h⁺) migrate to the surface of semiconductor particles and oxidize the electron donor (such as H₂O, OH⁻), while electrons (e⁻) migrate to the protected metal. As a result, the electron density on the surface of the protected metal increases and the corrosion potential shifts negatively so that the metal entered the thermodynamic and thermal stable region to achieve the purpose of cathodic protection. Therefore, in PCP, sufficiently negative CB potential is the postulate (Sun et al. 2015). TiO₂ coatings are the most widely used photocatalyst in PCP. But the band gap of TiO₂ is 3.2 eV, which limits its application in visible light range (Guan et al. 2018a). Many semiconductor materials have been developed used as photocatalysts in PCP including metal oxides (ZnO (Liu et al. 2019b), WO₃ (Ma et al. 2019), Fe₂O₃ (Liu et al. 2019a), et al.), metal sulfides (CdS (Vamvasakis et al. 2018), MoS₂ (Kong et al. 2018), et al.) and polymeric (g-C₃N₄ (Xu et al. 2019)). In recent years, g-C₃N₄, as a metal-free photocatalyst, has sparked extensive attention in photocatalysis due to its versatile properties, such as low cost, appropriate band gap (2.7 eV) for visible light absorption, good thermal stability, and optical properties (Wang et al. 2008; Ong et al. 2016; Wang et al. 2018b). However, the high recombination rate of photo-induced electrons and hole pairs is still a barrier for photocatalytic efficiency. A feasible way is constructing a heterojunction to overcome these disadvantages (Jiao et al. 2019). Heterojunction refers to the recombination of two semiconductors with matching band structures, one as an acceptor for electrons and the other as an acceptor for holes. When illuminated, the photoelectrons in the CB of one semiconductor will be transferred to the CB of another semiconductor to improve the efficiency of carrier transfer and separation (Li et al. 2020). Constructing g-C₃N₄ and other semiconductors (such as CdS, FeOCl, WO₃ and Ag₃PO₄) into a heterojunction framework can not only retain the advantages of each component separately, expand the photoresponse range, but also reduce the recombination of photogenerated electrons and holes, thereby improving photocatalytic performance (Katsumata et al. 2013; Du et al. 2021; Nguyen et al. 2021; You et al. 2021). It has been reported that the formation of rGO/Fe₂O₃/g-C₃N₄ nanocomposites exhibited superior photocatalytic ability by enhancing charge migration and separation (Shanavas et al. 2019).

Loading noble metals, such as Au and Pt, effectively improve the activity of photocatalysts. Recent studies have shown that MoS₂ can be applied as a substitution for noble metals in the integration of co-catalysts. Sun et al. (Liu et al. 2018) fabricated MoS₂ quantum dots on g-C₃N₄ and found that the composites markedly improved H₂ evolution activities. The enhancement of photocatalytic H₂ evolution was attributed to the formation of the heterojunction interface. The layered structure of MoS₂ is similar to graphene and can match with that of g-C₃N₄. Furthermore, researchers have confirmed the heterojunction formation between g-C₃N₄ and MoS₂ (Zhang et al. 2019b). Therefore, the g-C₃N₄/MoS₂ heterojunction promotes the migration and separation of photo-induced charge. GO has been introduced as an excellent electron conductive material that could promote the transport of photo-induced electrons and hole pairs in most ternary heterojunctions. Yet, as far as we know, there has been no previous report on g-C₃N₄/GO/MoS₂ as a ternary photocatalyst in PCP.

Based on the above issues, we constructed g-C₃N₄/GO/MoS₂ ternary semiconductor heterojunctions through a facile hydrothermal method, trying to develop a green and economical photocatalyst with good photocatalysis performance. The 304 SS was used to investigate the PCP performance of g-C₃N₄/GO/MoS₂. The ternary heterojunction composites exhibited high PCP performance. A possible mechanism of the catalyst promoting PCP was proposed.

The model of fluorine-doped tin oxide (FTO) conductive glass used in this experiment was FTO-P003, which was purchased from Zhuhai Kaivo Optoelectronic Technology Co., Ltd. The composition of 304 SS (wt%) is as follows: Fe (71.87), C (0.07), Cr (17.0), Ni (8.0), Mn (2.0), Si (1.0), P (0.035), and S (0.03). All the chemical reagents used in the experiments were analytically pure and were not further purified in the process of use.

Here, 5 g of melamine was calcinated (heating rate: 2 °C/min) at 550 °C for 3 h in a tube furnace in nitrogen. Afterwards, the resulting bulk g-C₃N₄ was ground into powder in a mortar. Then, 0.5 g of g-C₃N₄ powder was calcinated (heating rate: 5 °C/min) at 500 °C for 3 h in a tube furnace in air. Finally, the corresponding white layer of g-C₃N₄ product was collected (Papailias et al. 2018).

Graphene oxide (denoted as GO) was prepared through oxidation of the natural graphite powder by a modified Hummers' method (Marinoiu et al. 2017, 2020). In detail, the mixture of 3.0 g graphite powder, 3.0 g P₂O_5, and 3.0 g K₂S₂O₈ was added into 50 mL of sulphuric acid (18.4 M). Next, the solution was heated to 80 °C and kept stirring for 5 h. The mixture was slowly diluted with 300 mL deionized (DI) water. Then the pre-oxidized graphite powder was obtained by filtration. In a next step, the pre-oxidized graphite power was put into 50 mL of sulphuric acid (18.4 M); 15 g KMnO₄ powder was slowly put under ice batches during stirring. After that, they were heated to 35 °C and kept for 2 h under stirring. After cooling down to room temperature, 20 mL of 30% H₂O₂ was added dropwise. Following that, the color of the mixture turned to golden yellow. Finally, the mixture was washed with diluted HCl solution (v:v = 1:10) by centrifugation. The as-obtained brown product was washed with DI water until the pH value was about 7.0.

The g-C₃N₄/GO/MoS₂ composites were achieved via a hydrothermal method. The amount of GO added was 15 wt%, and the mass ratio of g-C₃N₄ to MoS₂ was 1: 1. In a typical synthesis procedure, 0.1613 g of Na₂MoO₄⋅2H₂O and 0.2504 g of C₂H₅NS were dissolved into 60 mL DI water under magnetic stirring for 1 h. Subsequently, 2.13 g GO aqueous sol with 0.05 g g-C₃N₄ was dissolved into 20 mL DI water. The above two types of mixture solution were interfused completely. After that, the corresponding mixture solution was transferred into a 100 mL Teflon-lined autoclave and maintained at 180 °C for 24 h. Finally, the resulting dark precipitate denoting g-C₃N₄/GO (15 wt%)/MoS₂ was collected and fully washed with DI water and ethanol (Hu et al. 2018). Figure S1 in Supplementary Information is the illustration of the synthetic process for g-C₃N₄/GO (15 wt%)/MoS₂.

The crystal structure of g-C₃N₄, MoS_2, and g-C₃N₄/GO (15 wt%)/MoS₂ was detected using a Bruker-D8 X-ray diffractometer (XRD, Bruker, Germany) with CuKα radiation. X-ray photoelectron spectroscopy (XPS) was performed using a photoelectron spectrometer (Kalpha, ThermoFisher, USA). Characterization of the morphology of g-C₃N₄, MoS_2, and g-C₃N₄/GO (15 wt%)/MoS₂ was performed using a supra 55 field emission scanning electron microscope (FE-SEM, Zeiss, Germany). Transmission electronic microscopy (TEM) images and element mapping of materials were analyzed using a JMT-2100F electron microscope (JEOL, Japan), equipped with a Super-X, energy-dispersive spectrometer (EDS, Bruker Germany). A UV-2550 spectrophotometer (SHIMADZU, Japan) was used to conduct ultraviolet-visible diffuse reflectance absorption spectra (UV-vis) of the samples. The photoluminescence properties of samples were determined using a RF-5301PC spectrofluorophotometer (Shimadzu, Japan) with an excitation wavelength of 400 nm.

The electrochemical experiment was carried out with a set of self-made electrochemical testing equipments, the schematic diagram of which is shown in Figure S2.

Two beakers acted as photochemical cell and electrochemical cell. The electrolyte in the photochemical cell was 0.1 M Na₂S and 0.2 M NaOH, and the electrolyte in the electrochemical cell was 3.5 wt% NaCl. The two beakers were connected by a salt bridge.

In the electrochemical cell, the Pt sheet was used as the counter electrode, the calomel electrode was used as the reference electrode, and the 304 SS electrode coupled with the photoelectrode acted as the working electrode. The light source was a 300 W xenon lamp (PLS-SXE300) with a light intensity of 520 MW/cm², which shone directly on the back of the light electrode. The distance between the light source and the light electrode was 5 cm (Zhang et al. 2014).

Preparation of the salt bridge: 3 g agar powder was dissolved in 100 mL DI water and heated to 90 °C until completely dissolved, followed by the addition of 34.2 g potassium chloride. Subsequently, the above solution was transferred to a U-shaped tube with a rubber dropper and cooled to room temperature.

Preparation of the 304 SS electrode: the 304 SS was cut into small pieces of 10 mm × 10 mm × 3 mm and polished step by step with sandpaper. A copper wire was welded to the back of the 304 SS sheet, followed by sealing with epoxy resin, and then polished evenly with 2,000 mesh SiC papers, so that the exposed surface (working surface) area of the 304 SS sheet was 10 mm × 10 mm. Finally, the above electrode was washed in ethanol with ultrasound for 30 min, then rinsed with DI water, and dried in a drying oven (room temperature) for 24 h for later use.

Preparation of the photoelectrode: the FTO was cut into small pieces of 10 mm × 20 mm and put into the cleaning solution (10 mL DI water + 10 mL isopropyl alcohol + 10 mL acetone) with ultrasonic vibration for 30 min, then rinsed with DI water, followed by sealing the resulting FTO surface with tape, leaving only a 10 mm × 10 mm area as a working surface. Afterwards, 20 mg of photocatalyst, 20 μL naphthol, and 1 mL of ethanol were mixed in a centrifuge tube. After ultrasonic (100 W/40 kHz) bombardment for 30 min, the corresponding slurry was collected. Finally, 20 μL of the slurry was smeared onto the working surface of the FTO as evenly as possible to ensure the same sample mass on the FTO surface (about 0.4 mg/cm²). After natural air drying, the obtained photoelectrode was heated to 100 °C for 2 h.

CHI660D electrochemical workstation was used to measure the electrochemistry performance of 304 SS coupled with different photocatalysts. Electrochemical impedance spectroscopy (EIS) tests were carried out in the frequency between 10⁵ and 10⁻² Hz, and the AC voltage amplitude was 5 mV. Dynamic potential polarization curves (Tafel plot) were tested with a scan rate of 1 mV·s⁻¹. The linear sweep voltammetry (i-V) measurement was obtained from −1.0 V to 0.4 V (vs. SCE) with a scan rate of 0.02 mV·s⁻¹. The time gap between light and dark was 10 s. The CV curves test was performed with a scan rate of 10 mV. The Mott–Schottky curves were tested from −1.0 to 0.0 V (vs. SCE). The AC voltage amplitude was 10 mV, and the test frequency was 1,000 Hz.

The crystal structure of g-C₃N₄, MoS_2, and g-C₃N₄/GO (15 wt%)/MoS₂ was analyzed by XRD. From Figure 1, we can see that at 13.0° and 27.4°, the characteristic diffraction peaks of g-C₃N₄ exhibited corresponding to (100) and (002) planes (JCPDS 87-1526) (Yang et al. 2020). The weak peak at 13.0° was derived from the in-plane structural repeated motif of tri-s-triazine units with an interplanar distance of d = 0.663 nm (Wang et al. 2018c). The dominant peak, which was at 27.4°, was derived from the aromatic ring system stack whose interplanar distance was 0.324 nm (Wang et al. 2018a). As for MoS₂ samples, the two diffraction peaks at 33.3° and 58.8° corresponded well to MoS₂ (JCPDS 37-1492) (Zhang et al. 2019a), which represented the (101) and (110) crystal planes, respectively. The 13.9° diffraction peak corresponding to the (002) crystal plane appeared at 10.3°, which was due to its interlayer expansion effect (Zhang et al. 2020). According to the Bragg equation: 2dsinθ = λ, the interlayer spacing of the (002) plane of the MoS₂ sample could be calculated. The interplanar distance was about 0.86 nm. In g-C₃N₄/GO (15 wt%)/MoS₂ composites, the main performance was the characteristic peak of MoS₂, and the peak height was decreased. This was because the sample contains g-C₃N₄ and GO, which resulted in a decrease in its crystallinity. The reason for the weak characteristic peak of g-C₃N₄ was that the high crystalline MoS₂ grew on the surface of g-C₃N₄ (Fageria et al. 2017). In addition, no characteristic peak of GO was observed in g-C₃N₄/GO (15 wt%)/MoS₂ samples, which was due to the relatively low GO content (Guan et al. 2017).

XPS spectra recorded the surface electronic states and chemical composition of ternary composites. Figure S3a indicated the spectrum of g-C₃N₄/GO (15 wt%)/MoS₂ demonstrated the co-existence of Mo, S, C, O, and N elements. As shown in Figure S3b, the peak at 284.16 eV showed the occurrence of sp² C = C bonds of g-C₃N₄ (Shao et al. 2016). The peak at 285.64 eV can be consistent with sp³ hybridized C atoms (C-(N)₃) (Fu et al. 2018). The peak located at 288.20 eV was ascribed to the sp² carbon N-C = N, which was in the N-containing aromatic rings (Fu et al. 2018). Figure S3c shows N 1s spectrum, the peak at 398.49 eV was attributed to C-N = C groups. The peak at 400.23 eV showed the occurrence of the tertiary nitrogen N-(C)₃ groups (Sun et al. 2020). In addition, the peak at 404.46 eV may be associated with positive charge localization or the charging effects in the heterocycles (Wang et al. 2020). As can be seen in Figure S3d, the peak centered located at 531.53 eV correspond to O 1s orbital (Hu et al. 2018). In Figure S3e, the S 2p peaks, which were located at 161.53 eV and 162.95 eV, were consistent with S 2p_3/2 and S 2p_1/2 of S²⁻, respectively (Guan et al. 2017). In addition, the peak located at 164.23 eV corresponds to the C–S bond formed by replacing sulfur with lattice nitrogen in the g-C₃N₄ framework (Bai et al. 2018). Figure S3f shows the Mo 3d high-resolution XPS spectrum of the g-C₃N₄/GO (15 wt%)/MoS₂ sample. The peaks located at 232.41 eV and 228.89 eV corresponded to Mo 3d_3/2 and Mo 3d_5/2 of Mo⁴⁺ in MoS₂, respectively (Zhou et al. 2020). The peak at 235.65 eV was consistent with the Mo atoms exposed on the surface (Ou et al. 2020). The weak peak at 226.03 eV can be ascribed to S²⁻ and Mo⁴⁺, which is the typical peak for MoS₂ (Bai et al. 2018).

The morphology of the ternary composites was further characterized by SEM. The g-C₃N₄ exhibited an average of 100 nm thicknesses with a lamellar stackable structure. In the MoS₂/GO composites, MoS₂ was in the form of nanoparticles, and the wrinkle GO was enclosed on the surface of MoS₂ (Figure S4a and 4b). As for the sample, the wrinkles GO wrapped MoS₂ nanoparticles attached to the surface of the g-C₃N₄ sheet layer (Figure S4c). Figure S4d indicated the co-existence of C, N, O, S, and Mo elements in the g-C₃N₄/GO (15 wt%)/MoS₂ sample.

The morphologies of g-C₃N₄/GO (15 wt%)/MoS₂ samples were further observed by TEM. MoS₂ was a petal-shaped nanosheet. MoS₂ nanosheets grew on g-C₃N₄ (Figure S5a). Figure S5b shows the HR-TEM images of g-C₃N₄/GO (15 wt%)/MoS₂. Each petal of MoS₂ nanosheets displayed obvious lattice fringes with a lattice spacing of 0.86 nm, corresponding to the (002) crystal plane of MoS₂. MoS₂ and g-C₃N₄ were closely attached, and the crystal lattices of both could be clearly seen. Figure S5c further demonstrates the co-existence of C, N, O, S, and Mo elements in the g-C₃N₄/GO (15 wt%)/MoS₂ sample.

At an excitation wavelength of 400 nm, the PL spectra of MoS₂, g-C₃N₄/MoS₂ and g-C₃N₄/GO (15 wt%)/MoS₂ composites are displayed in Figure S7. Two emission peaks were observed, which appeared at 450 and 470 nm (Liang et al. 2020). The emission peak of MoS₂ was the strongest, and indicated that the high recombination efficiency of photo-generated carriers. After adding g-C₃N₄, the peak value decreased, indicating that the separation of photo-generated carriers had been promoted. The emission peak of g-C₃N₄/GO (15 wt%)/MoS₂ composites was the weakest, indicating that the addition of GO promoted the transfer of photogenerated electrons and holes. Therefore, the g-C₃N₄/GO (15 wt%)/MoS₂ ternary composites showed the strongest photoelectrochemical activity.

To investigate the photoelectrochemical performance of the composite catalyst, photogenerated current density (I-T) was performed at interval of 50 s/time illumination. As shown in Figure 2, after illumination, the photocurrent density of the samples increased. The greater the increase in photogenerated current density, the higher was the charge separation efficiency of the catalyst. The photocurrent of MoS₂ alone was relatively low, and the photocurrent significantly increased after being combined with g-C₃N₄, which reached 6 μA/cm². g-C₃N₄/GO (15 wt%)/MoS₂ has the best photocurrent response, which can reach 22 μA/cm². This is because the heterostructure of g-C₃N₄/MoS₂ increased the electron-hole separation efficiency, and the addition of GO was more conducive to the photogenerated electron transfer between g-C₃N₄/MoS₂.

EIS test is an effective method to analyze and evaluate the PCP performance of samples. Figure 3 shows the EIS Nyquist plots and equivalent circuit diagram of 304 SS coupled with different photoelectrodes in light. The 304 SS coupled with g-C₃N₄/MoS₂ showed an obvious and smaller capacitive reactance arc than pure g-C₃N₄ and MoS₂. After adding GO, two arcs appeared in the low-frequency region. The small semi-circular arc indicated low charge transfer resistance (Wei et al. 2019). As shown in Figure 3, the equivalent circuit as R (CR(QR)) (CR) fitting circuit was adopted to fit the EIS results using the ZSimpWin software, where Cdl and Qf represent the double-layer capacitance and the capacitance of the surface film, respectively. Furthermore, Rs, Rf, Rp, and Rct represent the electrolyte solution resistance, the resistance of oxide films, electrolyte, and charge transfer, respectively. Y and n are constants that are independent of frequency. Detailed fitting data of the impedance spectrum of the material are shown in Table 1. Rct values can be used to characterize the charge transfer resistance of semiconductor materials. The Rct coupled with g-C₃N₄ and MoS₂ was 1,032 Ω•cm² and 1,324 Ω•cm², respectively. The Rct coupled with g-C₃N₄/MoS₂ was 967.9 Ω•cm², indicating that the heterostructure reduced the charge transfer resistance of the composites. After adding GO, the Rct values decreased significantly. The decrease in Rct value indicated the higher photogenerated electrons transfer. The EIS results showed that the g-C₃N₄/GO (15 wt%)/MoS₂ composites transferred a large number of electrons to the surface of 304 SS, resulting in the best photochemical cathodic protection performance.

Figure 4 demonstrates the Tafel plot of 304 SS coupled with kinds of photoelectrodes in light. The corresponding parameters calculated are displayed in Table 2. The potential of 304 SS coupled with g-C₃N₄/MoS₂ was between that of g-C₃N₄ and MoS₂. After adding GO, the corrosion potentials of 304 SS coupled with all three samples significantly shifted negatively. 304 SS coupled with g-C₃N₄/GO showed the most negative corrosion potential. As for the corrosion current, g-C₃N₄/MoS₂ was higher than that of MoS₂ and g-C₃N₄ alone, which is ascribed to the formation of heterojunctions. The increase in the corrosion current increased the electron transfer rate (Qiu et al. 2020). After the introduction of GO, the corrosion currents of 304 SS coupled with g-C₃N₄ and MoS₂ significantly increased. GO promoted the transfer of photo-generated electrons and holes. 304 SS coupled with g-C₃N₄/GO (15 wt%)/MoS₂ revealed the highest current density, which was 77.19 μA•cm⁻². The heterojunction of g-C₃N₄/MoS₂ promoted the separation of photogenerated electrons and holes, and the addition of GO promoted the transfer of photo-generated electrons and holes.

To study the photoelectrochemical performance of the composite photocatalyst further, the photoinduced i-V curves of the composites between −1.0 and 0.4 V were measured under intermittent light. The photoelectrochemical properties of the composites were tested by intermittent illumination of 10 s (Figure 5). The samples generated the zigzag photocurrent response between −1.0 and 0.4 V. Comparing the absolute value of photocurrent of different samples, the photo-generated current density produced by g-C₃N₄/GO (15 wt%)/MoS₂ was the highest. The g-C₃N₄/GO (15 wt%)/MoS₂ showed the best photoelectrochemical properties.

The CV curves of the 304 SS coupled with samples were also tested under light. The results are shown in Figure 6. In this experiment, the CV curves of 304 SS coupled with different photocatalysts were tested at a scan rate of 10 mv. The potential ranges of the samples were approximately −0.2 V to 0.4 V. 304 SS coupled with MoS₂ showed double capacitance characteristics. The cathode and anode peaks of other photocatalysts were not the same (Fageria et al. 2017). According to the Randles–Sevcik equation, the electroactive surface area (ECSA) was proportional to the peak of the cathode and anode (Shi & Zhao 2014). The larger ECSA means more active sites for the electrochemical reaction. The high cathode and anode peak value generated a larger ECSA. In addition, it exhibited better electrochemical performance. In Figure 6, we can see that the 304 SS coupled with g-C₃N₄/GO (15 wt%)/MoS₂ showed the highest peak in the cathode and anode curves. Therefore, the ternary photocatalyst has once again been proved to have the best electrochemical performance.

C represented the capacitance of the space charge region. e was the electronic charge. ε was the relative dielectric constant of the semiconductor. ε₀ represented the permittivity for free space. N_D was the charge carrier density. E was the applied potential. E_FB was the flat band potential. T was the temperature. k was the Boltzmann constant (Jing et al. 2019). The flat band potential (E_FB) was determined from the horizontal intercept. As shown in Figure 7, the E_FB of MoS₂ was about −0.93 eV. After adding GO, its E_FB was negatively shifted to −0.99 eV. The E_FB of g-C₃N₄ was about −0.83 eV. After adding GO, its E_FB was negatively shifted to −1.01 eV. The negative shift of E_FB increases the Fermi level of the material, which was more favorable for the transfer of photogenerated electrons (Liu et al. 2018; Ou et al. 2020). Through GO modification, the E_FB of a photocatalyst was negatively shifted, which improved the transfer efficiency of photogenerated electrons and holes, thereby improving the photocathodic protection ability of the composites. The E_FB of g-C₃N₄/MoS₂ was −0.89 eV, which was between the MoS₂ and g-C₃N₄. After GO was compounded with g-C₃N₄/MoS₂, the E_FB was negatively shifted to −1.13 eV. This indicated that g-C₃N₄/GO (15 wt%)/MoS₂ had better photogenerated charge separation efficiency and more excellent photocathodic protection performance.

Based on the results of the aforementioned experiments, the possible mechanism of the catalyst promoting PCP is proposed in Figure 8. Under visible light, the CB of g-C₃N₄ and MoS₂ both generated photogenerated electrons and left the photogenerated holes on the VB. In the g-C₃N₄/GO (15 wt%)/MoS₂ sample, g-C₃N₄ and MoS₂ formed the heterojunction. Photogenerated electrons transferred from the CB of g-C₃N₄ to the CB of MoS₂. The formation of the heterojunction electric field promoted the separation of photogenerated electrons and holes. The addition of GO further promoted the separation and transfer of photogenerated electrons and holes and simultaneously reduced the Fermi level of the composites. The photoelectrons on the surface of the samples transferred to the 304 SS surface through the wire, which made the self-corrosion potential of 304 SS drop sharply and resulted in the cathodic polarization of 304 SS.

In this research, the g-C₃N₄/GO (15 wt%)/MoS₂ composites were successfully synthesized by a facile hydrothermal method. MoS₂ was a petal-shaped nanosheet characterized by SEM and TEM. Wrinkled GO wrapped MoS₂ nanoparticles attached to the surface of the g-C₃N₄ sheet. The EIS results depicted that the Rct value of 304 SS coupled with g-C₃N₄/MoS₂ decreased from 1,032 to 967.9 Ω•cm² compared with g-C₃N₄ alone. Notably, the Rct value of 304 SS coupled with g-C₃N₄/GO (15 wt%)/MoS₂ composites decreased to 35.66 Ω•cm², which was 29 and 37 times lower than that of g-C₃N₄ and MoS₂ alone.

The polarization curve fitting values exhibited that g-C₃N₄/GO (15 wt%)/MoS₂ provided the highest current density (77.19 μA•cm²) for the 304 SS, which was four times higher than that of single g-C₃N₄. The mechanism is that, under light conditions, photo-generated electrons were generated from the CB of g-C₃N₄ and transferred to the VB of MoS₂, and the heterojunction formed between g-C₃N₄ and MoS₂. The formation of a heterojunction electric field promoted the separation of photogenerated electrons and holes. The addition of GO further promoted the separation and transfer of photogenerated electrons and holes and simultaneously reduced the Fermi level of the composites. Therefore, the ternary composite improved the photocathode protection performance for 304 SS. This research is expected to provide new insights to explore high-efficiency photocatalysts and new corrosion protection technologies with potential application prospects.

Research was supported by the Grant from National Natural Science Foundation of China (81572218, 21776172, 52071198), Science and Technology Commission of Shanghai Municipality (Shanghai Municipal Natural Science Foundation, 13ZR1439100) and Shanghai Municipal Bureau of Health (20124303). The research was supported by the Shanghai Engineering Research Center of Energy-Saving in Heat Exchange Systems.

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

All relevant data are included in the paper or its Supplementary Information.