Introduction

With the high-speed growth of industrial technology, metal corrosion has become a global problem [1, 2]. Metal corrosion not only shortens the service life of the equipment but also causes huge economic losses, even catastrophic safety accidents and environmental problems. Specifically, Q235 carbon steel (CS) is prone to severe corrosion in NaCl solution [3]. Photocathodic protection is an environmentally friendly and cost-saving technology with great application potential for metal anti-corrosion [4], which uses clean solar energy unlike conventional anti-corrosion technology. In addition, semiconductor photoanode materials are not consumed like traditional sacrificial anodes. This technology uses semiconductor materials (TiO2 [5], g-C3N4 [6], ZnO [7, 8], SrTiO3 [9]) to harvest solar photons and convert light energy into electricity to drive reduction chemical reactions efficiently, thereby effectively alleviating metal corrosion.

TiO2 has been widely used as a photoelectrode material for photocathodic protection due to its wide application range to catalyze various redox reactions, as well as its low cost, non-toxicity, and high chemical and photochemical stability [10,11,12]. However, individual TiO2 materials can only be induced by UV light as a result of their broad bandgap (3.0 eV for rutile, 3.2 eV for anatase). In addition, the available photoelectrons are reduced due to the rapid recombination of photogenerated carriers. To overcome the above disadvantages, numerous approaches have been proposed to heighten the photocathodic protection ability of TiO2-based photoelectrodes. These approaches include surface modification [13], designing highly ordered TiO2 nanotubes (NTs) [14], doping with metals or non-metals [15,16,17,18], and constructing heterojunctions [19,20,21]. More specifically, constructing heterojunctions by combining with other materials has proven to be a valid method for improving the photoelectrochemical properties of TiO2. The materials used for this strategy include metal oxides (In2O3 [22], MoO3 [23], Bi2O3 [24], WO3 [25, 26], RuO2 [27, 28]), metal sulfides and selenides (Ag2S [29], Bi2S3 [30], Ag2Se [31]), graphene [32,33,34], Co(OH)2 [35], and ZnFeAl-layered double hydroxides [36].

In addition, ternary and quaternary chalcogenides and selenium compounds, such as Cu2AgInS4 [37], Cu2AgInSe4 [38], and Cu2ZnSnSe4 [39], have received considerable attention in recent years due to their good photostability, strong absorption in the visible light range, and good electron transport properties. These materials have shown high photocatalytic activity to improve the photoconversion efficiency of quantum dot-sensitized solar cells. Stannum indium sulfide (SnIn4S8) is a ternary chalcogenide semiconductor [40] that shows promising applications in heavy metal reduction and the photocatalytic degradation of organics and pharmaceutical wastewater due to its good chemical stability and strong absorption of visible light [41, 42]. The physical and chemical ability of nanomaterials is mainly determined by their size, structure, and morphology. SnIn4S8 nanosheets have a large specific surface area, which may facilitate visible-light absorption [43] and reduce film resistance due to the fast charge transfer between the nanosheet and the electrolyte [44]. In addition, SnIn4S8 nanosheets have a relatively negative conduction band, which is beneficial to providing photocathodic protection for metals with a negative self-corrosion potential. Therefore, SnIn4S8 nanosheet-modified TiO2 NTs may show improved photoelectrochemical and photocathodic protection performance. However, there are only a few reports on the fabrication and cathodic protection application of SnIn4S8 nanosheet/TiO2 nanotube composites.

In the present study, a SnIn4S8 nanosheet/TiO2 NT heterojunction film was synthesized through a solvothermal reaction and subsequent electrochemical anodic oxidation method. The photocathodic protection performance and mechanism of SnIn4S8/TiO2 composites for Q235 CS were systematically studied.

Methods

Fabrication of SnIn4S8/TiO2 Photoelectrodes

TiO2 NTs were prepared on a titanium sheet (1 cm × 4 cm) by an electrochemical anodic oxidation process. The titanium sheet was cleaned after chemical polishing and anodized under 30 V for 30 min using ethylene glycol and 0.45 wt% NH4F and 8 wt% H2O as an electrolyte solution. A platinum sheet was used as the cathode. The sample was rinsed with deionized water and dried at 60 °C. TiO2 NTs were then acquired after heat treatment under 450 °C for 1.5 h.

SnIn4S8/TiO2 NTs photoelectrodes were fabricated by depositing SnIn4S8 nanosheets on the surface of the TiO2 NTs through a simple solvothermal process (Fig. 1a). Typically, 0.05 mmol SnCl4·5H2O (0.0175 g), 0.2 mmol InCl3 (0.0587 g), and 0.5 mmol thioacetamide (0.0375 g) were added to 80 mL of absolute ethanol and stirred until the solution was homogeneous. The above solution and titanium sheet with fabricated TiO2 NTs were placed on the bottom of a 100 mL Teflon-lined stainless steel autoclave at 180 °C for 3–12 h. Then, the sample was cleaned repeatedly with absolute ethanol and dried at 70 °C for 4 h. The synthesized composites were marked as 3 h SnIn4S8/TiO2, 6 h SnIn4S8/TiO2, 9 h SnIn4S8/TiO2, 12 h SnIn4S8/TiO2, respectively. For comparison, SnIn4S8 nanosheets were prepared on the titanium sheet using the same procedure.

Fig. 1
figure 1

a Schematic illustration of the synthetic process for the preparation of the SnIn4S8/TiO2 composite film; schematic illustration of the experimental devices for photoelectrochemical measurements including b OCP variations and c photocurrent density variations

Full size image

Characterization

The morphology, microstructure, and elemental composition of the fabricated photoelectrodes were studied using a Hitachi SU8220 scanning electron microscope (SEM), high-resolution transmission electron microscope (HRTEM, JEOL JEM-2010), and energy dispersive spectrometer (EDS), respectively. The crystal structures of the prepared photoelectrodes were evaluated by X-ray diffraction (XRD) (D8-advance, Bruker AXS Co.) with Cu Kα irradiation at a wavelength of 0.15406 nm. The surface composition and chemical states were evaluated by X-ray photoelectron spectroscopy (XPS, ESCALAB 250XI, Thermo Scientific Co.) using Al Kα radiation. The optical characteristics were analyzed using a UV–Vis diffuse reflectance spectrophotometer (Hitachi UH4150). The photoluminescence (PL) spectra of the prepared photoelectrodes were measured on an FLS980 Series fluorescence spectrometer. Fourier transform infrared (FTIR) analyses were conducted on a Varian Scimitar 1000 spectrophotometer.

Photoelectrochemical Tests

Photoelectrochemical tests were performed at room temperature on an electrochemical workstation (CHI760E, Chenhua Instrument Shanghai Co., Ltd.). The visible light source was a 300 W xenon light (PLS-SXE 300C, Perfect light Beijing Co., Ltd.) with a 420 nm cut-off filter. A Q235 CS electrode acted as the metal sample for testing, which was made via embedding a Q235 CS sample (1 cm × 1 cm × 1 cm) in an epoxy resin. The composition is as follows: 0.18% C; 0.28% Si; 0.035% S; 0.04% P; 0.55% Mn, and 98.915% Fe.

Figure 1b shows the schematic illustration of the experimental device for the measurement of the variations of the open circuit potential (OCP) of the Q235 CS coupled to the prepared photoelectrodes. The device consists of a photoelectrolytic cell and a corrosion cell, which were connected by a Nafion membrane. The prepared photoelectrodes were placed in the photoelectrolytic cell containing 0.1 M Na2S and 0.2 M NaOH solution, and the Q235 CS was set in the corrosion cell. The Q235 CS linked to the photoelectrodes via an external Cu wire was employed as the working electrode, while a saturated calomel electrode (SCE) and a platinum sheet were utilized as the reference electrode and the contrast electrode, respectively. The photoelectrodes were shined by visible light through a quartz window. The Tafel curves were tested from − 0.25 and + 0.25 V relative to the OCP at a sweep rate of 0.5 mV s−1. The electrochemical impedance spectroscopy (EIS) spectra were tested in a frequency range from 105 to 10−2 Hz. The amplitude of the AC signal was 10 mV. Figure 1c displays the schematic illustration of the experimental device for the photocurrent density variation under intermittent visible light illumination. The photoelectrodes served as the working electrode, and Q235 CS was linked to the ground by a Cu wire. The contrast and reference electrode were short-circuited by a thin wire.

Results and Discussion

Morphologies and Chemical Compositions

Figure 2 reveals the morphologies of the TiO2 and SnIn4S8/TiO2 composite films. As shown in Fig. 2a, b, a well-aligned and highly ordered NTs film was successfully formed on the titanium surface through an electrochemical anodic oxidation process. The average inner diameter of TiO2 NTs was approximately 50 nm, the wall thickness was approximately 25 nm, and the tube length was approximately 1300 nm. It can be observed that numerous curved SnIn4S8 nanosheets were attached to the TiO2 surface after a solvothermal reaction for 6 h (Fig. 2c, d). The nanostructure composed of thin nanosheets and large hierarchical pores was conducive to the improvement of the light harvestability due to multiple light scattering, which may facilitate the charge transfer in the photochemical reaction and enhance the photocathodic protection performance of the SnIn4S8/TiO2 composite.

Fig. 2
figure 2

a Topographic and b cross-sectional SEM images of the TiO2 NT films, c topographic and d cross-sectional SEM images of the SnIn4S8/TiO2 composite film

Full size image

Figure 3 shows the XRD spectra of the synthesized TiO2 and the SnIn4S8/TiO2 composite films. For pure TiO2, except for the characteristic peaks of the titanium substrate, the peaks at 25.3°, 48.1°, and 53.9° were indexed to the anatase (101), (200), and (105) lattice planes, respectively (JCPDS Card No. 21-1272). For the SnIn4S8/TiO2 composites, the XRD patterns of the obtained samples were similar. Compared with pure TiO2, three other diffraction peaks were observed in the XRD spectra of the SnIn4S8/TiO2 composites. The peaks at 27.5°, 28.4°, and 33.0° were assigned to the lattice planes (311), (222), and (400) of tetragonal SnIn4S8, respectively (JCPDS Card No. 42-1305). No peaks corresponding to binary sulfides and oxides were detected, confirming the high purity of the SnIn4S8/TiO2 composites.

Fig. 3
figure 3

XRD spectra of pure TiO2 and the different SnIn4S8/TiO2 composites

Full size image

To further study the microstructures of the SnIn4S8/TiO2 composite, TEM images were analyzed as shown in Fig. 4a, b, confirming that the SnIn4S8 nanosheets were attached to the TiO2 NTs surface. The wall thickness and inner diameter of the TiO2 NTs were approximately 25 and 50 nm, respectively, which was consistent with the SEM results. Figure 4c is an HRTEM image taken at the edge of the surface of the composite in Fig. 4a, which displays a close contact between the TiO2 NTs and the SnIn4S8 nanosheets. The lattice fringe with a spacing of 0.358 nm was assigned to the anatase (101) plane of TiO2 (JCPDS Card No. 21-1272). The lattice fringe with a spacing of 0.268 nm and 0.323 nm was indexed to the tetragonal (400) and (311) planes of SnIn4S8 (JCPDS Card No. 42-1305), respectively. The results confirmed that SnIn4S8 nanosheets were constructed on the TiO2 NTs surface by the solvothermal process. The corresponding EDS spectrum (Fig. 4d) demonstrated that the SnIn4S8/TiO2 composite film was composed of Ti, O, Sn, S, and In elements.

Fig. 4
figure 4

a, b TEM, and c HRTEM images, and d EDS spectrum of the SnIn4S8/TiO2 composite film

Full size image

The surface components and chemical states of SnIn4S8/TiO2 were studied by XPS. From the XPS survey spectrum (Fig. 5a), it was confirmed that Ti, O, Sn, In, S elements were present in the composite, which was consistent with the EDS result. Moreover, a characteristic peak corresponding to the C element was observed, which was caused by exposure to atmospheric pollution. Figure 5b displays the high-resolution XPS spectrum of Ti 2p. The peaks of the binding energy at 458.5 and 464.2 eV were coincident with Ti 2p3/2 and Ti 2p1/2, respectively, which are characteristic peaks of Ti4+ [45]. This indicated that Ti4+ existed in the SnIn4S8/TiO2 composite. Figure 5c shows the peaks at 529.7 and 531.6 eV, which correspond to lattice oxygen and adsorbed oxygen, respectively [46]. The lattice oxygen represented the Ti–O, indicating the existence of TiO2 in the composite. The adsorbed oxygen came from H2O adsorbed on the surface of the composite, suggesting that the surface of the composite was rich in oxygen vacancies. Figure 5d reveals that the peak centered at 495.0 and 486.6 eV belonged to Sn 3d3/2 and Sn 3d5/2, respectively [40], indicating that the valence state of Sn was + 4. As shown in Fig. 5e, the binding energies of 445.2 and 452.8 eV were assigned to In 3d5/2 and In 3d3/2, respectively, suggesting the existence of In3+ in the composite [47]. The peaks at 162.8 and 161.7 eV in Fig. 5f correspond to S 2p1/2 and S 2p3/2, respectively, showing that S element existed mainly in the form of S2− in the SnIn4S8/TiO2 NTs [48]. The above results further indicate that a SnIn4S8/TiO2 NTs heterojunction film was successfully synthesized by the combination of a solvothermal and electrochemical anodic oxidation process, which was in good agreement with the results from SEM, TEM, and XRD analyzes.

Fig. 5
figure 5

a XPS survey spectrum of the SnIn4S8/TiO2 composite film; high-resolution XPS spectra of b Ti 2p, c O 1s, d Sn 3d, e In 3d and f S 2p

Full size image

To investigate the growth process of the SnIn4S8 nanosheets, the SEM images of the composites fabricated by the solvothermal process for 3, 9, and 12 h were observed. When the reaction time was 3 h, small SnIn4S8 nanosheets appeared on the surface of the TiO2 NTs in the obtained composite (Additional file 1: Fig. S1a, d). After the time reached 6 h, the size and thickness of the nanosheets increased (Fig. 2c, d). However, once the solvothermal reaction time exceeded 9 h, the SnIn4S8 nanosheets completely covered the TiO2 surface and blocked the mouth of the TiO2 NTs, which was harmful to the photogenerated charge separation (Additional file 1: Fig. S1b, c, e, f). The main reactions during the fabrication process are as follows:

$${\text{CH}}_{{3}} {\text{CSNH}}_{{2}} + {\text{H}}_{2} {\text{O}} \to {\text{CH}}_{{3}} {\text{CONH}}_{{2}} { + \text{ H}}_{2} {\text{S}}$$
(1)
$${\text{H}}_{2} {\text{S}} \to {\text{2H}}^{ + } { + \text{ S}}^{{2-}}$$
(2)
$${\text{Sn}}^{{4 + }} { + \text{ 4In}}^{{3 + }} { + \text{ 8S}}^{{2-}} \to {\text{SnIn}}_{{4}} {\text{S}}_{{8}}$$
(3)
$${\text{SnIn}}_{{4}} {\text{S}}_{{8}} { + \text{ TiO}}_{{2}} \to {\text{SnIn}}_{{4}} {\text{S}}_{{8}} {\text{/TiO}}_{{2}}$$
(4)

Optical Characteristics

The optical characteristics of the fabricated photoelectrodes were analyzed by UV–visible diffuse reflectance spectra (Fig. 6a). The light absorption of pure TiO2 was mainly in the UV region, and the absorption edge was approximately 385 nm, which was attributed to the inherent absorption of TiO2. Moreover, the light absorption of TiO2 in the visible region may be ascribed to light scattering caused by cracks or holes in the NTs [49]. The absorption edge of SnIn4S8 and the SnIn4S8/TiO2 composite appeared near 590 and 535 nm, respectively. The bandgap (Eg) values of the photoelectrodes were computed using the following equation [50]:

$$(\alpha {\text{h}}v)^{2} = {\text{A}}({\text{h}}v - E_{{\text{g}}} )$$
(5)

where α, hv, and A imply the absorption coefficient, photon energy, and characteristic constant, respectively. The plot of (αhv)2 versus hv for calculating the bandgap is displayed in Fig. 6b. The Eg value of pure TiO2 was 3.22 eV, which was similar to that of anatase TiO2 (3.2 eV) [51]. The Eg value of SnIn4S8 and the SnIn4S8/TiO2 composites was evaluated to be 2.1 and 2.32 eV, respectively. These results indicated that the absorption capacity of the SnIn4S8/TiO2 composite was enhanced in the visible light region, which is beneficial to the utilization of solar energy and improvement of the photocathodic protection performance.

Fig. 6
figure 6

UV–visible diffuse reflectance spectra of TiO2, SnIn4S8, and SnIn4S8/TiO2 composite film

Full size image

The PL spectra measurement was used to examine the separation, transfer, and recombination of photogenerated carriers of semiconductor materials [52]. The weaker PL intensity means the lower the recombination rate of photogenerated carriers [53]. The PL intensity of the SnIn4S8/TiO2 composite was lower than that of TiO2 NTs (Additional file 1: Fig. S2), indicating that the sensitization of the SnIn4S8 nanosheets could effectively restrain the recombination of photogenerated carriers in the TiO2 NTs.

Photoelectrochemical Performance

Photocurrent density is a significant parameter used to investigate the photocathodic protection properties of semiconductor photoanodes. The greater the photocurrent density of the metal coupled to the photoanode, the better the photoelectric conversion performance and the photocathodic protection effect of the photoanode [54]. Figure 7a displays the photocurrent density changes of a Q235 CS electrode coupled to pure TiO2 or SnIn4S8/TiO2 NTs photoanodes. Before the lamp was turned on, the photocurrent densities of the metal electrode coupled to the photoanodes were almost zero, indicating no photogenerated electrons were transferred to the steel electrode at this time. When the light was turned on, both pure TiO2 and the SnIn4S8/TiO2 composite photoanodes exhibited a photocathodic protection current for the Q235 CS electrode. This was because photogenerated electrons were carried from the photoanodes to the surface of the steel electrode. The SnIn4S8/TiO2 composite fabricated by the 6 h solvothermal reaction demonstrated the greatest photocurrent density (about 100 μA cm−2), which was approximately 8 times larger than pure TiO2. This indicates that the modification of the SnIn4S8 nanosheets remarkably enhanced the photoelectron separation and transmission efficiency of the TiO2 NTs. The photocurrent densities decreased in the order of 6 h SnIn4S8/TiO2 > 9 h SnIn4S8/TiO2 > 12 h SnIn4S8/TiO2 > 3 h SnIn4S8/TiO2 > TiO2. As the time of the solvothermal reaction increased, the SnIn4S8 nanosheets absorbed more light energy to generate sufficient photoelectrons, thus exhibiting a larger photocurrent. However, when the solvothermal reaction time exceeded 6 h, the increase in the thickness of the nanosheets increased the transfer distance of the photoelectrons in the nanosheets [55, 56]. In addition, the TiO2 NTs cannot absorb light to generate photoelectrons since the oversized SnIn4S8 nanosheets blocked their mouth. This ultimately led to a decrease in photoelectrons transferred to the Ti substrate, which decreased the photocurrent of the composite.

Fig. 7
figure 7

Variations in the a photocurrent density and b potential of the Q235 CS electrode coupled to the prepared TiO2 and different SnIn4S8/TiO2 photoanodes under intermittent visible light irradiation

Full size image

The variation of the potential of the metal coupled to the semiconductor photoanode is another considerable parameter used to investigate the photocathodic protection property of the photoanodes [57]. Photo-excitation of the photoanodes generates electrons, which are transferred to the surface of the steel and reduce its potential, and then protect the steel electrode. Figure 7b demonstrates the OCP changes of Q235 CS linked to TiO2 and SnIn4S8/TiO2 NTs photoanodes. The potentials of the coupled Q235 CS decreased slightly before illumination due to the galvanic effect. After the light was turned on, the photogenerated potentials of the Q235 CS electrode coupled to TiO2 and SnIn4S8/TiO2 NTs had undergone significant negative shifts. The order of the potential drop was 6 h SnIn4S8/TiO2 (0.45 V vs. SCE) > 9 h SnIn4S8/TiO2 (0.36 V vs. SCE) > 3 h SnIn4S8/TiO2 (0.32 V vs. SCE) > 12 h SnIn4S8/TiO2 (0.30 V vs. SCE) > TiO2 (0.18 V vs. SCE). The potentials of Q235 CS linked to the SnIn4S8/TiO2 composites slowly increased after the light was turned off, and the coupled potentials were still much lower than that of bare Q235 CS. This indicated that the composites could continue to protect the steel for some time even in the absence of light due to the slow release of electrons stored in the composites. Therefore, combined with the results of photocurrent densities, it can be concluded that the SnIn4S8/TiO2 composites had better photocathodic protection for Q235 CS than pure TiO2. The optimal photocathodic protection was achieved for the SnIn4S8/TiO2 composite prepared for 6 h. In addition, the 6 h SnIn4S8/TiO2 composite photoanode displayed higher photocathodic protection properties than most photoanodes previously reported (Additional file 1: Table S1). The SnIn4S8/TiO2 composites mentioned in the following sections were obtained for the sample developed by 6 h of reaction time.

In order to investigate the stability of the SnIn4S8/TiO2 composite, the long term potential variation of a Q235 CS electrode connected to the photoanode under intermittent visible light illumination was analyzed. As displayed in Fig. 8a, the self-corrosion potential of the bare Q235 CS was − 0.58 V versus SCE in 3.5 wt % NaCl solution. When the CS was connected to the SnIn4S8/TiO2 composite, the potential dropped rapidly with visible light irradiation, which may be due to the transfer of photogenerated electrons from the composite to the steel. Under long term visible light irradiation, the potential of steel was stable at − 0.96 V versus SCE. This was only 0.07 V higher than the photogenerated potential obtained from the OCP results, indicating that the SnIn4S8/TiO2 composite had good stability. After the light irradiation was stopped, the potential of Q235 CS increased back to − 0.74 V versus SCE, which was still far from the self-corrosion potential, demonstrating that the composite can provide the steel with continuous protection in the dark state. Moreover, as shown in Fig. 8b, c, the XRD pattern and FTIR spectrum of the SnIn4S8/TiO2 composite after the photoelectrochemical test were consistent with those before the test, respectively, indicating that the composite possesses good photoelectrochemical stability.

Fig. 8
figure 8

a Long-term potential changes of Q235 CS connected to the SnIn4S8/TiO2 composite under intermittent visible light illumination; b XRD patterns of the SnIn4S8/TiO2 composite before and after photoelectrochemical tests; c FTIR spectra of the SnIn4S8/TiO2 composite before and after photoelectrochemical tests

Full size image

The Tafel curves of the pure Q235 CS, the Q235 CS electrode linked to the pure TiO2, and the 6 h SnIn4S8/TiO2 composite in dark conditions and under irradiation were tested to further evaluate the photoelectrochemical performance of the SnIn4S8/TiO2 composites (Fig. 9). Corrview software was utilized to compute the corrosion potential (Ecorr) and corrosion current (icorr), and the results are displayed in Table 1. Under visible light irradiation, after Q235 CS was linked to the pure TiO2 or the SnIn4S8/TiO2 composite electrode, the Ecorr shifted negatively, indicating that photogenerated electrons effectively migrated from the photoelectrode to the steel electrode, thereby providing a photocathodic protection effect for the Q235 CS electrode. The Ecorr of Q235 CS linked to the SnIn4S8/TiO2 composite (− 0.92 V vs. SCE) under irradiation was much lower than that of pure TiO2 (− 0.75 V vs. SCE), showing that the composite was able to provide Q235 CS with better cathodic protection than pure TiO2. The icorr of Q235 CS linked to the SnIn4S8/TiO2 composite notably increased compared with that of bare Q235 CS under light irradiation (Table 1). This is as a result of the polarization of photoinduced electrons accelerating the rate of chemical reactions at the interface [58, 59].

Fig. 9
figure 9

Tafel curves of Q235 CS in 3.5% NaCl solution under different conditions, a bare Q235 CS, b Q235 CS coupled to TiO2 NTs film in the dark, c Q235 CS coupled to SnIn4S8/TiO2 composite film in the dark, d Q235 CS coupled to TiO2 NTs film under visible light irradiation, and e Q235 CS coupled to SnIn4S8/TiO2 composite film under visible light irradiation

Full size image
Table 1 Electrochemical parameters obtained by the Tafel curves of Q235 CS, TiO2 in the dark, SnIn4S8/TiO2 in the dark, TiO2 under illumination, and SnIn4S8/TiO2 under illumination
Full size table

EIS was used to further study the photogenerated carrier separation and transfer process of the SnIn4S8/TiO2 composite film and the corrosion resistance of the Q235 CS electrode. Figure 10a shows the Nyquist plots of Q235 CS, Q235 CS coupled to a TiO2 NT film, and the SnIn4S8/TiO2 composite film under and after visible light irradiation. The fitting circuit for EIS using ZSimpWin software consisted of an Rs(QfRf)(QdlRct) model, as demonstrated in Fig. 10b, where Rs indicates the electrolyte resistance, Qf and Rf indicate the capacitance and resistance of the semiconductor film electrode at high frequencies, respectively, and Qdl and Rct indicate the electric double-layer capacitance and the charge transfer resistance at low frequencies, respectively.

Fig. 10
figure 10

a Nyquist plots of bare Q235 CS, Q235 CS coupled to TiO2 NTs film, and SnIn4S8/TiO2 composite film under and after visible light irradiation, and b the corresponding equivalent circuit

Full size image

The fitting impedance parameters of EIS using the equivalent circuit are also shown in Table 2. The diameter of the impedance arc of the coupled Q235 CS electrode under visible light was smaller than that of pure Q235 CS, and the Rct values of the coupled Q235 CS electrode decreased significantly. The results indicate that the rate of the electrochemical reaction at the interface between the Q235 CS and the solution increased remarkably, which may be due to the migration of photoelectrons from the photoanode to the Q235 CS [60]. In addition, the Rct value of the SnIn4S8/TiO2 composite was smaller than that of TiO2, which may be because the heterojunction formed by TiO2 and SnIn4S8 was conducive to the separation and migration of photoinduced charges. These results were consistent with those obtained from the Tafel curves (Fig. 9). Furthermore, the diameter of the impedance arc of the SnIn4S8/TiO2 composite under visible light was distinctly smaller than the impedance after the light was turned off, suggesting that more photoelectrons migrated from the SnIn4S8/TiO2 composite to Q235 CS under visible light irradiation. This demonstrates that the SnIn4S8/TiO2 composite could offer Q235 CS effective photocathodic protection in the presence of visible light.

Table 2 Fitting impedance parameters of Nyquist plots using the equivalent circuit in Fig. 10b
Full size table

Mechanism

Figure 11 displays the proposed mechanism for the above-mentioned photocathodic protection of the SnIn4S8/TiO2 composites and the photoelectron separation and transfer process. When the visible light illuminated the SnIn4S8/TiO2 heterojunction, both the SnIn4S8 and TiO2 were excited to generate photoelectrons and holes. The photoelectrons migrated from the conduction band of SnIn4S8 to the conduction band of TiO2 because the potential of the former is much lower than that of TiO2 (Fig. 11a, b). Subsequently, these electrons migrated to the surface of the steel electrode driven by the electric field force because the potential was lower than its self-corrosion potential (Fig. 11c). This process inhibited the corrosion of the Q235 CS as a result of cathodic polarization. Meanwhile, the photogenerated holes moved from the valence band of TiO2 to the valence band of SnIn4S8. After the light was turned off, the stored electrons in the SnIn4S8/TiO2 composites migrated continually to the Q235 CS electrode, which realized the protection of steel in the dark state. In addition, the photogenerated holes reacted with S2− in the electrolyte solution to form polysulfides (S(x)2−) [61], which could accelerate the photogenerated carrier separation.

Fig. 11
figure 11

Proposed mechanisms for the enhanced photocathodic protection performance of the SnIn4S8/TiO2 composite for Q235 CS under visible light illumination

Full size image

Conclusions

With the increasingly prominent energy crisis and environmental pollution caused by metal corrosion, there is an urgent need to develop highly efficient visible-light-responsive semiconductor photoanodes. In this study, a SnIn4S8/TiO2 nanotube photoanode was successfully fabricated via a solvothermal treatment and subsequent electrochemical anodic oxidation method. The results indicated that the nanostructure of the SnIn4S8/TiO2 composite consisted of thin nanosheets and large hierarchical pores, which were conducive to photogenerated carrier separation. The optical characteristic analysis showed that the absorption capacity of the SnIn4S8/TiO2 composite was enhanced in the visible light region. The composite fabricated by a solvothermal reaction for 6 h exhibited the optimal photocathodic protection performance. The photocurrent density of Q235 CS coupled with the 6 h SnIn4S8/TiO2 composite achieved 100 μA cm−2, which was approximately 8 times larger than pure TiO2. The maximum negative shift value of the photoinduced potential of Q235 CS could reach 0.45 V versus SCE. The excellent photocathodic protection effect of the SnIn4S8/TiO2 composite for Q235 CS suggests that the composite is a promising photoelectrode material for the photocathodic protection of metals.