INTRODUCTION

Zinc oxide with band gap of about 3.3 eV is low cost environmentally friendly material, especially polar (002) plane in the ZnO crystal structure give high hydroxyl radical (\(^{\bullet }{\text{OH}}\)) decomposition [1], which may contribute to improved photocatalytic efficiency, handling toxic wastes. So, it is one of promising candidates that are focused on exploiting photocatalytic properties besides TiO2 material [2]. However, because of the wide band gap of ZnO, its absorption wavelength falls in the ultraviolet region of the solar spectrum, and this is also a disadvantage because UV only accounts for about 4–5% of the solar spectrum. Besides, the fast recombination rate of photogenerated charge carriers is also a significant factor affecting the photocatalytic performance of ZnO [3]. Therefore, tuning the band gap as well as controlling charge separation and recombination rates in ZnO are key factors for improving the photocatalytic performances. At this point, there have been many different research directions such as doping ZnO with metal elements [35], non-metal elements [6, 7] or combining ZnO with other narrow band gap semiconductors [816]. According to Morasae Samadi et al. [17], although the doping materials helped to narrow the band gap of ZnO, increased the absorption of visible light and improved the photocatalytic efficiency, it led to the formation of defects on the surface or inside the structure of ZnO. These defects act as traps for the photogenerated electron–hole pairs, reducing their mobility, thereby increasing the recombination of photogenerated electron–hole pairs.

Another approach is to combine ZnO with a narrow band gap semiconductor with suitable aligned energy band to form optimal heterostructures for improving photocarriers separation and photocatalytic respond in visible region [10, 14, 15]. ZnO has been combined with semiconductors such as CdS [810], WO3 [11, 12], CuO [1315], Cu2ZnSnS4 (CZTS) [16] and showed enhanced photocatalytic performance under the visible light. Among them, CZTS has many outstanding advantages, including its composition of only inexpensive, environmentally friendly materials and a narrow band gap of about 1.6 eV, good absorption of visible to near-infrared light [18]. In addition, the II type band alignment between ZnO and CZTS is suitable for charge separation process, improving photocatalytic performance [16]. Thus, the fabrication of CZTS/ZnO heterostructures both expands absorption capacity of the solar spectrum, and is suitable for photocarrier separation. Especially, the low cost and environmental friendliness of the components in ZnO and CZTS films combined with simple fabrication procedure will bring high economic efficiency in practical applications.

CZTS thin films have been fabricated by different deposition techniques. To ensure simplicity, low cost, ease to control and non-toxicity, the recent studies are focusing on chemical methods. From this perspective, sol–gel coating technique has been proved to be an attractive method for preparing CZTS thin film. CZTS is a quaternary compound semiconductor, it is easy to form secondary phases during film deposition. The photocatalytic activity of CZTS thin films is strongly related to these secondary phases. The presence of secondary phases in CZTS thin film mainly leads to increase the recombination of photogenerated charge carriers, therefore the photocatalytic performance decreased [19].

In our work, the CZTS/ZnO thin film heterostructures were prepared to improve photocatalytic activity in visible light, thus pure kesterite CZTS phase was required. CZTS thin films without post annealing sulfurization were prepared by sol–gel method. In this approach, Hyungjin Park et al. [20] and Kezhi Zhang et al. [21] fabricated CZTS thin films at different annealing temperatures above 500°C in N2 atmosphere, the structural and optical properties were investigated. Ahmed Ziti et al. [22] also changed annealing temperatures at 300, 325, and 350°C in air atmosphere in preparing CZTS thin films. Recently, Pratyay Amrit et al. [23] deposited CZTS thin films using the method of spin coating of different layers and concentrations of precursor solutions, then CZTS thin films were annealed at 250°C in N2 atmosphere. And Afrina Sharmin et al. [24] changed annealing temperatures from 300–500°C in air atmosphere. Comparing the works described above, in our experiment, CZTS thin films were investigated with different copper ratios in the precursor solutions. The Cu-poor and Zn-rich thin films have been shown to promote good charge carrier flow which favor the photocatalytic process [25]. However, the secondary phases are easy to form in this composition. Therefore, it is very important to control the composition of CZTS thin films in order to obtain pure phase kesterite. For ZnO/CZTS thin film heterostructure, to the best of our knowledge, there are few studies of ZnO/CZTS thin film heterostructures for photocatalysis. For example, work of Zubair et al. [16] prepared ZnO/CZTS heterostructures in the form of composites. And these heterostructures is used to convert CO2 to CH4 under simulated solar light with the high efficiency. In this work, for the first time, ZnO/CZTS heterostructures in the form of thin film were fabricated by a simple chemical method on a glass substrate with good structural and optical quality. In order to reduce the lattice mismatch between ZnO and CZTS in the fabrication of the heterostructure, before depositing the ZnO thin films onto CZTS thin films, the dip-coating of the seed layers was carried out to assist the growth of ZnO thin films. The results showed that with two seed layers, the ZnO/CZTS thin film heterostructure provides an impressive degradation efficiency of methylene blue (MB) solution under visible light irradiation up to 92.8%, 10.2 and 2.9 times greater than that of pristine ZnO and CZTS thin films, respectively.

EXPERIMENTAL

The experimental processing for fabricating the heterostructures was shown in Fig. 1. Firstly, the glass substrates were cleaned and the CZTS thin films were prepared according to the procedures described in our previous work [26]. Here, we fixed withdraw speed of 4 cm/min, annealing temperature of 450°C. The different molar ratios of Cu : Zn : Sn : S of 1.4 : 1 : 1 : 8, 1.6 : 1 : 1 : 8, 1.8 : 1 : 1 : 8 and 2 : 1 : 1 : 8 for fabricating CZTS thin films (labelled C_1.4, C_1.6, C_1.8 and C_2, respectively). After finding optimal composition for the growth of CZTS thin film, ZnO seed layers were coated on films via dip-coating method. This procedure was presented in our previous works [27]. In this experiment, we changed numbers (1, 2 and 3) of dip-coating ZnO seed layers. Then, the obtained samples were immersed into an equal molar mixture 1 : 1 of Zn(NO3)2·6H2O and methenamine (C6H12N4), and deposited at 70°C for 1 h. Finally, the obtained ZnO/CZTS thin film heterostructures were taken out of the solutions and rinsed with distilled water and dried in air overnight. It noted that ZnO/CZTS thin film heterostructures without and with changing in numbers (1, 2, and 3) of dip-coating ZnO seed layers are labelled ZC0, ZC1, ZC2, and ZC3, respectively.

Fig. 1.
figure 1

Schematic illustrating the development of ZnO/CZTS thin film heterostructures.

The structural and optical properties of ZnO/CZTS thin film heterostructures were investigated using X-ray diffraction measurements (XRD; D2 PHARSER; Bruker) with \({\text{Cu}}{{K}_{{{{\alpha }_{1}}}}}\) radiation (wavelength = 1.54060 Å), room temperature Raman spectrum in the backscattering configuration (Labram 300; Horiba JOBIN YVON; excitation wavelength = 632.8 nm), scanning electron microscopy (SEM; Hitachi 4700), the energy-dispersive X-ray (EDX) analysis (Horiba H-7593) and UV-Vis spectroscopy (Halo RB-10).

The photocatalysis property of ZnO/CZTS thin film heterostructures was evaluated through the degradation of MB dye aqueous solution under a 30W compact lamp with λ = 400–650 nm as a visible light source. The sample (area 25 × 25 mm2) was dipped into 10 mL of 10  ppm MB solution under irradiation for 150 min. The absorbance at 662 nm of MB solutions was measured every 30 min with visible light using UV-Vis spectroscopy (Halo RB-10).

The electrochemical measurements were performed in 0.5 M Na2SO4 aqueous solution (the electrolyte) using a typical three-electrode configuration. The Ag/AgCl, a platinum plate and the pristine ZnO, CZTS thin films and ZnO/CZTS thin film heterostructures grown on FTO substrates were used as reference electrode, counter electrode, and working electrode, respectively. The LED lamp of 80 mW/cm2 with = 380–700 nm was used as the light source for measuring photocurrent.

RESULTS AND DISCUSSION

The Effect of Different Cu Ratios on the Phase Pure Kesterite CZTS Thin Films Using the Sol–Gel Dip Coating Technique without Sulfurization

The crystal structure of CZTS thin films with different Cu ratios is investigated using XRD diffraction patterns in Fig. 2. From Fig. 2, we can see that all samples exhibited diffraction peaks at 2θ = 28.45°, 47.34°, and 56.21° corresponding to the (112), (220) and (116) lattice planes of kesterite CZTS structure, respectively [The PDF 04-017-3032]. Besides, the results also show the presence of secondary phases with different Cu ratios. Specifically, the sample C_2 corresponding to the Cu highest ratios of 2, it is exhibited secondary phase Cu2 – xS. This can be explained due to loss of Sn and S atoms in annealing process, Cu atoms move freely in the lattice and combine with the S atoms to form secondary phase Cu2 – xS [28].

Fig. 2.
figure 2

XRD patterns of CZTS thin films with different Cu ratios.

Therefore, the presence of secondary phase Cu2 ‒ xS was due to the high amount of copper in CZTS thin films. As expected, when the Cu ratios decrease, corresponding to samples C_1.8, C_1.6 and C_1.4, secondary phase Cu2 – xS disappear. The sample C_1.8 corresponds to the ratio of 1.8 : 1 : 1 : 8 (Cu : Zn : Sn : S) exhibited phase pure kesterite structure with strong preferred orientation along (112) plane. It is indicated that CZTS thin films show a good crystal quality. However, it is noted that when the Cu ratios continue to decrease, the secondary phase SnS2 is observed in the CZTS thin films (Samples C_1.6 and C_1.4). The presence of the secondary phase SnS2 becomes stronger at the low Cu ratios. The cause for the formation of SnS2 can be due to high Sn content in CZTS thin films. The composition of elements present in CZTS thin films can be determined through the EDX spectra in Fig. 3 and Table 1.

Fig. 3.
figure 3

EDX spectra of samples (a) C_1.4, (b) C_1.6, (c) C_1.8 and (d) C_2.

Table 1. The elemental composition ratios from EDX spectra

The results shown in Table 1 indicate that the atomic percentage of Sn of the sample C_1.4 presents the highest value compared to that of other samples. The Sn elements combined to S evaporated during the annealing process to form the secondary phase SnS2 as shown in the XRD patterns (Fig. 2). On the other hand, the EDX spectra in Fig. 3 also indicated Cu-poor and Zn-rich CZTS thin films. The morphology of CZTS thin films with different Cu ratios were observed using SEM surface images (Fig. 4). As the Cu ratios increase, the particle size in film tends to increase and the film surface becomes more uniform. But, at the Cu ratio of 2 (Sample C_2), the film surface is no longer uniform.

Fig. 4.
figure 4

SEM surface images of of samples (a) C_1.4, (b) C_1.6, (c) C_1.8 and (d) C_2.

The optical property of CZTS thin films with different Cu ratios were investigated by UV-Vis spectra (Fig. 5). From Fig. 5, as increasing the ratio of Cu from 1.4 to 2, the CZTS thin films absorption increase slightly, otherwise, the band gap of films tends to decrease. This shift in band gap may be due to a change in the degree of pd hybridization of between Cu-d and S-p levels [29]. Besides, the presence of secondary phases Cu2 – xS and SnS2 as in shown in the XRD patterns (Fig. 2) were also responsible for shifting the band gap.

Fig. 5.
figure 5

(a) The plot of \({{\left( {\alpha h\nu } \right)}^{2}}\) versus hν, (b) the optical band gap determination of CZTS thin films with different Cu ratios ((i), (ii), (iii) and (iv)).

According to previous studies [25, 30], the presence of secondary phases mainly leads to increase recombination rate of photogenerated charge carriers, thus the result showed decreased rates of organic matter decomposition. Based on the above analysis results, in our work, the CZTS thin films with ratios of Cu : Zn : Sn : S of 1.8 : 1 : 1 : 8 exhibited CZTS kesterite structure and good crystallinity. This is noted that these CZTS films are suitable for photocatalytic applications. The raman spectrum in Fig. 6a further clarified absence of the secondary phases. The result from Fig. 6a shows a strong peak at 336 cm–1 of the A1 vibration mode from the kesterite phase in CZTS thin film [31]. And the EDX mapping in Fig. 6b indicate a non-uniform distribution of Cu, Zn, Sn, S elements in CZTS thin film.

Fig. 6.
figure 6

(a) Raman spectra, (b) EDX mapping of CZTS thin film with ratio of Cu : Zn : Sn : S ~ 1.8 : 1 : 1 : 8.

The Effect of ZnO Seed Layer on Structural, Optical and Photocatalytic Properties of ZnO/CZTS Thin Film Heterostructures

After determining the optimal condition for obtaining phase pure kesterite CZTS thin films, ZnO thin films were grown on these CZTS thin films to form heterostructures.

In order to evaluate light harvesting capability of all obtained heterostructures, UV-Vis absorption spectra were measured, as shown in Fig. 7a. The pristine ZnO thin film absorbs strongly in the UV region (around 380 nm), while the pristine CZTS thin film absorbs strongly in the wavelength range 500–800 nm. Besides, the bandgap of pristine ZnO and CZTS thin film of 3.26 and 1.47 eV are determined by using Tauc plot method [26], respectively (Fig. 7b). Hence, for pristine ZnO and CZTS thin films, the light absorption capacity is limited in the specific wavelength region of each film. However, for the ZnO/CZTS thin film heterostructures, it is observed that thin film heterstructures have a wide light absorption range from 380 to 800 nm and their absorbance increase significantly. When the number of seed layers increased, an increase in the number of absorbing particles [32], then, ZnO/CZTS thin film heterostructures exhibit a strong absorption. This result also plays an important role for improving the photocatalytic activity under visible light of ZnO/CZTS thin film heterostructures compared with pristine ZnO and CZTS thin films.

Fig. 7.
figure 7

(a) Absorption spectra of pristine ZnO (i), CZTS (ii) thin films and different ZnO/CZTS thin film heterostructures (ZC0 (iii), ZC1 (iv), ZC2 (v) and ZC3 (vi)), (b) the plot of \({{\left( {\alpha h\nu } \right)}^{2}}\) versus hν of pristine CZTS and ZnO thin films, respectively.

Figure 8 shows the XRD patterns of ZnO/CZTS thin film heterostructures with different seed layers. In addition to the diffraction peaks of the kesterite CZTS structure, we also observe (002), (101) and (103) diffraction peaks correspond to the wurzite structure of ZnO [The PDF 04-003-2106] all ZnO/CZTS thin film heterostructures. It is clearly seen that crystal quality of ZnO thin film in heterstructures has a gradual improvement when the number of seed layers increases. According to Yue Wang et al. [33], as the number of seed layers increases corresponding to increase in thickness, the crystal grain size and orientation of ZnO seed layer are improved, thus supporting the growth of ZnO thin films. On the other hand, some previous works indicated that the lattice-mismatch between ZnO and CZTS is about 16% [34, 35], it also explains why the crystal quality of ZnO thin film in the case of growth directly on CZTS thin film without seed layer is low.

Fig. 8.
figure 8

(a) XRD patterns, (b) Raman spectra of pristine ZnO, CZTS thin films and ZnO/CZTS thin film heterostructures.

Otherwise, the crystal quality of CZTS thin films in ZnO/CZTS heterstructures decreased as the number of seed layers increased. Especially, when number of seed layers exceeds two, crystal quality of CZTS thin film decreased markedly. This result is clearly shown through strongly increasing of full-width at half maximum (FWHM) of (112) peak at CZ3 heterostructure. According to previous works on CZTS thin films [3638], the crystalline quality of CZTS thin films is significantly degraded under loss of sulfur which is very volatile at temperatures above 200°C. In our experiment, to ensure good crystal quality of ZnO films, the seed layers are dried at 300°C to evaporate the solvent. It is indicated that sulphur component in CZTS thin films can be decomposed. Furthermore, when the number of seed layers increases, the coating process is repeated several times corresponding to prolonged heat treatment time, as a result, loss of elemental sulfur in the CZTS thin film becomes greater, resulting in decreased crystal quality of the CZTS thin film. It is noteworthy that two seed layers is the best suitable structures for ZnO/CZTS thin film heterstructures.

Raman spectra were used to further clarify crystal structure of ZnO/CZTS heterstructures. The results shown in Fig. 8b demonstrated that except for pristine ZnO thin film, all other samples exhibited the A1 vibration mode from the kesterite phase in CZTS thin films. On the other hand, it can also be found that there are no peaks of Cu3SnS4 (318, 348, and 295 cm–1), Cu2 – xS (457 cm–1) and ZnS (351 cm–1) secondary phases [39]. This result further supports the conclusion that CZTS thin films exhibit a kesterite structure.

In addition, comparing the Raman spectrum of pristine CZTS thin film with those of the ZnO/CZTS thin film heterostructures indicates the presence of peaks at 98 and 437 cm–1 corresponding to E2 symmetries of ZnO [40]. However, it is also noted that the vibration modes of ZnO are not clearly present in CZ0 thin film heterostructure. Again, this can be seen that ZnO deposited on CZTS without the supporting seed layer exhibits poor crystal quality. This result is completely consistent with the results of XRD in Fig. 8a. Consequently, with low-temperature CBD deposition and sol–gel method, ZnO/CZTS thin film heterostructures were successfully prepared with good crystallinity of both the ZnO and the CZTS thin films.

The effect of the number of seed layers on the morphology of ZnO/CZTS thin film heterostructures are also investigated by using SEM images in Fig. 9. It is found that the ZnO thin films distributed over the surface of the CZTS thin films. The CZ2 sample shows the clear hexagonal shape of the ZnO crystals. It also indicates that with CZ3 sample as the number of seed layers exceeds two, the ZnO thin film tend to be more closely packed. Besides, from the cross-sectional SEM images, the thickness of CZTS and ZnO thin films can be determined about 700 nm and about 450–600 nm, respectively, corresponding to different number of seed layers.

Fig. 9.
figure 9

SEM surface (left) and SEM across section images (right) of (a, b) ZC1, (c, d) ZC2 and (e, f) ZC3 heterostructures.

Eventually, the photocatalytic performance of the ZnO/CZTS thin film heterostructures was evaluated by degradation efficiency of MB under visible light irradiation. The results are shown in Figs. 10a–10b. From Fig. 10a, it was found that without a catalyst, there is almost no decomposition under excitation light. Whereas, in the presence of ZnO, CZTS thin films and ZnO/CZTS thin film heterostructures, the degradation efficiency of MB has attained great significance. Specifically, these values of the pristine ZnO and CZTS thin films are about 9.1 and 31.6%, respectively, meanwhile the ZnO/CZTS thin film heterostructures indicated an excellent degradability of 66.1, 77.8, 92.8, and 87.7% corresponding to the ZC0, ZC1, ZC2 and ZC3 thin film heterostructures, respectively.

Fig. 10.
figure 10

(a) The photo-degradation of MB solution under visible light irradiation within 150 minutes, (b) degradation efficiency of MB values, (c) the kapp values of pristine ZnO, CZTS thin films and ZnO/CZTS thin film heterostructures.

From the above results of degradation efficiency of MB, it can be seen that the pristine ZnO and CZTS thin films have the low efficiency, due to the limited partial absorption at specific wavelengths. On the contrary, owing to the ability to absorb a wide wavelength range, ZnO/CZTS thin film heterostructures have significantly improved the photocatalytic efficiency. Just another reason to explain better photocatalytic efficiency of ZnO/CZTS thin film heterostructures that there is the formation of (type-II) band alignment at the ZnO/CZTS interface which supports the separation of photogenerated e/h+ pairs, reducing the recombination of electron–hole and thus, enhancing degradation efficiency of MB. Schematic diagram of band alignment for ZnO/CZTS heterostructure thin film will be shown in Fig. 11b.

Fig. 11.
figure 11

(a) Transient photocurrent response of pristine ZnO (iii), CZTS (ii) thin films and ZnO/CZTS thin film heterostructure (labelled ZC2) (i), (b) schematic illustration suggestion of the degradation mechanism of MB solution over ZnO/CZTS heterostructures under the visible-light.

Additionally, for ZC0 and ZC1 thin film heterostructures corresponding to without and with one seed layers, it is seen that their degradation efficiency of MB are lower than those of ZC2 and ZC3 thin film heterostructures. This may be due to the seed layers is not good enough to support for the growth of ZnO films, the poor crystal quality with many defects of ZnO thin films, these defects act as charge trapping, resulting in reduced degradation efficiency of MB. In which, ZC2 thin film heterostructure shows the highest degradation efficiency of MB. It implies that in our experiment, ZnO/CZTS heterostructure thin film with two seed layer provides good optical and structural properties to enhanced photocatalytic performance. However, as the number of seed layer exceeds two, corresponding to CZ3 thin film heterostructure, degradation efficiency of MB decreases. It can be explained that the crystal quality of ZnO thin film is significantly increased, leading to decrease in the porosity of film and thus reduced capacity for MB absorption. On the other hand, the crystal quality of the CZTS thin film is strongly degraded, ZnO/CZTS interface appears many defects, and hence photocarrier recombination increases.

Furthermore, in our experiment, we use concentration of MB very small (10 ppm), the plot of lnC0/C versus time represents a straight line with R2 > 0.95 for all samples. Therefore, we can calculate the apparent rate constant (kapp) using pseudo first order photo-degradation kinetics [4143] to evaluate degradation rate of the MB solution: \(\ln \left( {\frac{{{{C}_{0}}}}{C}} \right) = {{k}_{{{\text{app}}}}}t\). Where, C0, Ct are the concentration of MB solution before and after visible light irradiation, respectively, t is irradiation time. The result in Fig. 10c of the ZC2 sample shows the highest degradation rate up to 0.01683 min–1, which corresponds to the highest photocatalytic efficiency shown in Figs. 10a–10b. ZC2 sample gives a decomposition rate 6.5 and 29.2 times higher than that of pristine CZTS and ZnO thin films, respectively.

The transient photocurrent response for multiple on–off cycles of pristine ZnO, CZTS thin films and ZC2 thin film heterostructure are investigated to study the separation and charge transfer efficiency (as shown in Fig. 11a). The results show that the ZC2 thin film heterostructure has higher photocurrent intensity (79.45 µA/cm2) than that of the pristine ZnO (7.13 µA/cm2) and CZTS (31.54 µA/cm2) thin films. It imples more effective charge separation of the photogenerated electron–hole pairs in ZnO/CZTS thin film heterostructure [44]. This also further clarifies the improved photocatalytic activity of ZnO/CZTS thin film heterostructures compared to the pristine ZnO, CZTS thin films (Figs. 10a–10c).

Figure 11b shows the degradation mechanism of MB solution with ZnO/CZTS thin film heterostructures under visible light irradiation. The energy levels of the conduction band edge (ECB) and valence band edge (EVB) are calculated [41]:

$${{E}_{{{\text{VB}}}}} = \chi - {{E}_{{\text{c}}}} + 0.5{{E}_{{\text{g}}}},$$
(1)
$${{E}_{{{\text{CB}}}}} = {{E}_{{{\text{VB}}}}} - {{E}_{{\text{g}}}}.$$
(2)

Where, χ is the electronegativity of the semiconductor and χ(CZTS) = 4.80 eV [41}, χ(ZnO) = 5.79 eV [45], and Ec is the energy of free electrons on the hydrogen scale (4.5 eV). In our experiment, the optical band gap of ZnO and CZTS are 3.26 and 1.47 eV, respectively.

Under visible excitation light, due to the bandgap of CZTS of 1.47 eV (~700 nm), the electrons in the valence band (VB) of CZTS was excited to the conduction band (CB), leaving a hole in the VB. As a result, electron–hole pairs are generated in CZTS (Eqs. (3)). In the case of CZTS thin film alone, the excited electrons quickly returned and recombined with the holes in the VB. However, with the presence of ZnO thin film, because the conduction band of ZnO is lower than that of CZTS, the photoelectrons generated from CB of CZTS are transferred quickly to CB of ZnO and play the role of reducing O2 to create superoxide radical (\(^{\bullet }{\text{O}}_{2}^{ - }\)) (Eqs. (45)). In contrast, the photoholes from ZnO transfer to CZTS which plays the role in oxidation to generate hydroxyl radical (\(^{\bullet }{\text{OH}}\)) to decompose MB solution (Eqs. (6)(7)). Thus, the recombination of photogenerated electron–hole pairs is reduced, which enhances the degradation efficiency of MB.

$$\begin{gathered} {\text{ZnO/CZTS}} + h\nu \left( {{\text{visible light}}} \right) \\ \to {\text{ZnO/CZTS }}\left( {{{{\text{h}}}^{ + }} + {{{\text{e}}}^{-}}} \right), \\ \end{gathered} $$
(3)
$$\begin{gathered} {\text{ZnO/CZTS }}\left( {{{{\text{h}}}^{ + }} + {{{\text{e}}}^{-}}} \right) \\ \to {\text{ZnO }}\left( {{{{\text{e}}}^{-}}} \right){\text{/CZTS }}\left( {{{{\text{h}}}^{ + }}} \right), \\ \end{gathered} $$
(4)
$${{{\text{O}}}_{2}} + {{{\text{e}}}^{-}} \to \,{{\,}^{\bullet }}{\text{O}}_{2}^{ - },$$
(5)
$${{{\text{h}}}^{ + }} + {{{\text{H}}}_{{\text{2}}}}{{{\text{O}}}^{{}}} \to \,{{\,}^{\bullet }}{\text{OH}},$$
(6)
$${\text{MB}} + {{(}^{\bullet }}{\text{O}}_{2}^{ - }\& \,{{\,}^{\bullet }}{\text{OH}}) \to {\text{C}}{{{\text{O}}}_{2}} + {{{\text{H}}}_{{\text{2}}}}{\text{O}}.$$
(7)

CONCLUSIONS

In this paper, the ZnO/CZTS thin film heterostructures were fabricated to improve photocatalysis in visible light. The phase pure kesterite CZTS thin films were obtained at ratio of 1.8 : 1 : 1 : 8 (Cu : Zn : Sn : S) without sulfurization. By investigating the effect of the number of ZnO seed layers on the crystal structure, optical and photocatalytic properties of the ZnO/CZTS thin film heterostructures, it can be found that two ZnO seed layers support good growth of ZnO thin films on the surface of CZTS thin films, contributing to improving the crystal structure of ZnO thin films and maintaining the crystal quality of CZTS thin films (This result was confirmed by XRD patterns). Moreover, UV-Vis spectra have shown that the combination of ZnO and CZTS layers has extended the light absorption range, in order to better exploit the visible light region. This combination also improved charge carrier separation of the photogenerated electron–hole pairs, thereby contributing to enhancement of photodegradation efficiency of MB solution compared with pristine ZnO and CZTS films.