1 Introduction

In recent years, many research groups have been working on tin based chalcogenide materials for optoelectronic device applications. Among them, SnS and SnSe are promising light absorbing materials used for low cost thin film solar cell applications owing to their earth abundance, less toxicity of constituent elements and favourable physical properties. Both these materials showed high optical absorption coefficient (~ 105 cm−1) and tunable optical band gap energy (1–2 eV) [1,2,3,4,5,6]. Moreover, both SnS and SnSe are p-type semiconductors exhibiting orthorhombic crystal structure with optical band gaps of 1.3 eV and 1.0 eV respectively [7,8,9]. Though they are efficient materials for solar energy conversion, however, the reported conversion efficiencies are low (SnS—4.63% and SnSe—≤ 1%) [10, 11]. Particularly, the reported theoretical conversion efficiency is 33% for SnS based solar cells, that is very much higher than that achieved at present [12]. Moreover, pure sulfide and selenide based solar cells showed less conversion efficiency as in the case of quaternary materials like CZTS (9.2%) and CZTSe (11.6%) than mixed sulfo-selenide based solar cells, CZTSSe (12.6%) [13]. In this context, tin sulfoselenide (SnSSe) has received more attention due to its adequate environmental stability and possibility to alter the optical and electrical properties according to the amount of S and Se content in the films. SnSSe is a p-type semiconductor, possessing the properties of both SnS and SnSe.

As per the literature survey, the reports available on the synthesis and characterization of SnSSe are meagre. Till now, few chemical and physical deposition techniques have been employed to prepare SnS1−xSex in different forms such as single crystals, nanocrystals, nanosheets and thin films. Table 1 presents the list of research articles reported in literature on p-type SnS1−xSex and the parameters investigated, including few reports on solar cell parameters. It can be seen from the table that although few studies have been reported mainly on the structure and optical properties of SnS1−xSex films deposited by different methods, the research on material properties of SnS1−xSex films lack coherence in presentation of results mapping onto the deposition conditions. Hence, a systematic and detailed investigation on the material properties in relation to the growth conditions are highly essential to optimize the deposition parameters for the development of efficient SnS1−xSex based thin film solar cells.

Table 1 Literature survey and material properties of p-type SnS1−xSex phase

The present work deals with the deposition of SnS1−xSex thin films by thermal co-evaportion technique, which is a suitable approach to prepare compound or alloy materials in thin film form. As per literature survey, no reports are available on thermally co-evaporated SnS1−xSex films. In vacuum techniques, substrate temperature is one of the important parameters to be optimized to grow high quality films. Hence, the aim of this work is to investigate the effect of substrate temperature on the composition, structure, morphology, topography, optical and photosensitivity of thermally co-evaporated SnS1−xSex films and optimize the substrate temperature to obtain good quality films.

2 Experimental and characterization details

2.1 Preparation of SnS1−xSex thin films

Thermal co-evaporation technique (HHV model BC 300 box coater) was employed to deposit SnS1−xSex thin films using SnS (Alfa Aesar, 99.5%) and elemental Se (Sigma Aldrich, 99.99%) as source materials. SnS1−xSex layers were deposited onto ultrasonically cleaned and preheated soda lime glass substrates through co-evaporation at a vacuum of 5 × 10−5 mbar at varied substrate temperatures (Ts) ranging from 200 to 350 °C. The deposition temperature was chosen in this range because the films grown at temperatures < 200 °C were of porous in nature and also not well adherent to the substrate surface. Further, the films formed above 350 °C were highly sulphur/selenium deficient owing to its high vapour pressure at such temperatures and the layers were highly reflecting in visual appearance. So, the substrate temperature was chosen in the range of 200–350 °C. Moreover, a circular type radiant heater of 1 kW was used to heat the substrate holder to required temperature. A K-type thermocouple was placed near the substrate holder to measure the temperature and a proportional integral derivative (PID) controller was used to control the substrate holder temperature. The other deposition parameters such as source to substrate distance and evaporation rate were kept constant as 13 cm and 20 Å/s respectively for all depositions, while the thickness of the as-deposited films was maintained approximately as 1 μm, measured by quartz crystal thickness monitor (model CTM-200).

2.2 Characterization details

The structural properties of as-deposited SnS1−xSex films was analysed by Ultima-IV X-ray diffractometer in grazing incidence diffraction (GIXRD) geometry at 1 degree of incident X-rays with Cu Kα radiation source (λ = 1.5406 Å). Diffraction angle in the range, 10–70° was used to identify the crystal planes present in the films and to evaluate other structural parameters. The structural phases present in the films were analysed by Raman spectroscopy using Nanofinder HE confocal Raman spectrometer performed at room temperature in the back scattering configuration, a solid state laser of wavelength 532 nm was used as a light source. Optical power incident on the samples was reduced down to 60 μW to avoid their damage. The × 100 objective lens (NA = 0.95) provided size of excitation spot of about 0.7 μm. Raman spectra were obtained with a spectral resolution better than 3.0 cm−1. The surface morphology and the chemical composition of the as-deposited films were investigated using scanning electron microscope (SEM) (S-4800, Hitachi) attached with energy dispersive X-ray spectroscopy (EDS) (Oxford Instruments, Inca Penta FET × 3). The surface topography of the films were analyzed using atomic force microscopy (AFM) (Solver Nano, NT-MDT). The AFM instrument was used in semi-contact mode with a scanning probe of 10 nm tip radius at a resonance frequency of 227 kHz. Photon RT spectrophotometer (Essent Optics) was used to analyze the optical properties of the layers using unpolarized light at room temperature. Moreover, the photoconductivity of the films was measured at a constant temperature of 307 K under LED illumination with excitation wavelength of 465 nm.

3 Results and discussion

Thermally co-evaporated SnS1−xSex films were appeared to be pale bluish grey in colour, uniform, pinhole free and the scratch tape test revealed that the layers were well adherent to the substrate surface.

3.1 EDS analysis

The elemental composition analysis of as-deposited films was carried out using energy dispersive X-ray spectroscopy (EDS). The EDS analysis revealed the presence of Sn, S and Se as constituent elements in all the as-deposited films. A typical EDS spectrum of SnS1−xSex thin films deposited at Ts = 300 °C was shown in Fig. 1. The atomic percentages of constituent elements in the as-deposited films and the variation in Sn/(S + Se) ratio with substrate temperature are listed in Table 2. It is observed from the table that increase of substrate temperature made the films Sn-rich and Sn/(S + Se) ratio increased with Ts. The variations in Sn/(S + Se) ratio might be due to variations in the vapour pressures of constituent elements. With raise of substrate temperature, S and Se contents were decreased due to re-evaporation, which is a common phenomenon that occurs in thermal methods at higher temperatures. This kind of re-evaporation of chalcogenides (Se/S) from the film surface was reported by various researchers in the literature on different chalcogenide materials [33,34,35,36].

Fig. 1
figure 1

EDS spectrum of SnS1−xSex thin films deposited at Ts = 300 °C

Table 2 Atomic percentages of SnS1−xSex thin films deposited at different substrate temperatures

3.2 Structural properties

Figure 2 shows the GIXRD patterns of SnS1−xSex thin films deposited at different substrate temperatures that vary in the range, 200–350 °C. All the films exhibited an intense peak at 2θ = 31.7° that corresponds to the (111) plane of the orthorhombic crystal structure. The other peaks present in the films were identical to the planes reported for the ternary SnS1−xSex (x = 0.5) crystal phase available in the JCPDS card No. 48-1225. Further, all the peaks observed in this work are similar to those reported by Han et al. for SnS1−xSex solid solutions [37]. This confirmed the formation of ternary chalcogenide SnS1−xSex phase in the films grown in this work.

Fig. 2
figure 2

The GIXRD patterns of SnS1−xSex thin films prepared at different substrate temperatures

Figure 2 shows the effect of change in substrate temperature on the intensity and sharpness of the diffraction peaks, which in turn gives information on crystallinity of the films. At Ts = 200 °C, a small SnSe2 peak was observed at 2θ = 14.6° along with SnS1−xSex phase. This was probably due to the insufficient thermal energy required to form the ternary phase at such temperatures. However, with an increase of substrate temperature to 250 °C, the SnSe2 peak disappeared and only SnS1−xSex phase was observed. At Ts = 300 °C, the intensity of peaks corresponding to the ternary phase was increased and became sharp, representing homogeneity and good crystallinity of the films. This indicated that Se and SnS phases reacted together at such temperatures to form the ternary phase with larger crystallite size. For further increase of temperature to 350 °C, SnS peaks were observed as secondary phase along with SnS1−xSex phase. This might be due to either dissociation of part of the ternary phase into stable binary SnS phase by re-evaporating Se or desorption of Se atoms at such higher temperatures. Such presence of secondary phases were also observed by Banotra et al. for SnS0.4Se0.6 alloy films annealed at higher temperatures [31].

The inter-planar spacing (d) between the crystal planes of as-grown SnS1−xSex layers was calculated for the (111) plane using Bragg’s diffraction law (relation 1), where θ is the diffraction angle. The unit cell dimensions of orthorhombic structure, a, b and c values were evaluated using (101), (011) and (111) planes by following the relation (2). The unit cell volume (V) of the films was calculated using relation (3).

$$d_{hkl} = \frac{\lambda }{2\sin \theta }$$
(1)
$$\frac{1}{{d^{2} }} = \frac{{h^{2} }}{{a^{2} }} + \frac{{k^{2} }}{{b^{2} }} + \frac{{l^{2} }}{{c^{2} }}$$
(2)
$$V = abc$$
(3)

All the calculated lattice parameters of SnS1−xSex films as a function of substrate temperature are listed in Table 3. It was noticed that the lattice parameters varied with increase of substrate temperature. A similar type of variation in lattice parameters in thermally evaporated SnS0.4Se0.6 films was reported by Banotra et al. [31].

Table 3 The lattice parameters of SnS1−xSex thin films

The microstructural parameters of SnS1−xSex films such as crystallite size (D) and lattice strain (ε) were calculated in the present work using Williamson–Hall (W–H) method. The W–H analysis is a simplified integral breadth method, where both size and strain induces broadening of X-ray peaks [36]. Moreover, it is more suitable method to determine crystallite size than the Debye–Scherrer method, because of the consideration of strain developed in the films also for the GIXRD peak broadening (β). The W–H equation in uniform deformation model (UDM), where the strain is assumed to be uniform in all crystallographic directions, is given by the relation (4) [38].

$$\beta \cos \theta = \frac{k\lambda }{D} + 4\varepsilon \sin \theta$$
(4)

Figure 3a–d shows the W–H plots of as-deposited SnS1−xSex films.

Fig. 3
figure 3

W-H plots of SnS1−xSex films deposited at: a Ts = 200 °C, b Ts = 250 °C, c Ts = 300 °C and d Ts = 350 °C

Seven most intensive peaks have been used to obtain crystal parameters by W–H approach. The estimated values of average crystallite size (D) and lattice strain (ε) from the W–H analysis of SnS1−xSex films with respect to change of substrate temperature were shown in Fig. 4a, b. It is observed from Fig. 4a that the crystallite size is increased with increase of Ts and reached a maximum of 22 nm at Ts = 300 °C and then decreased with further rise in substrate temperature due to re-evaporation of S/Se from the film surface and/or reduction in film—substrate surface interaction as observed in case of In2S3 films [39]. Figure 4b demonstrates that the strain in the films is compressive with negative slope at Ts = 200 °C and then increased to positive slope indicating tensile strain in the films at higher Ts. Further, it can be noted from Fig. 4b that the lattice strain increases with the increase of substrate temperature, in contrary to the general trend observed in different thin films. This can be explained as follows. With increase of Ts, the Se-content in the layers was increased because of the re-evaporation of S from film surface, owing to its high vapour pressure, while the film surface is exposed to continuous influx of Sn, S and Se atoms from the sources. This resulted in the occupation of vacant S sites by Se atoms. As Se ionic radius (1.98 Å) is higher than S ionic radius (1.84 Å), replacement of S vacancies by Se atoms led to lattice distortion and increased unit cell volume (as noted from Table 3), causing lattice strain that increases with increase of Ts up to 300 °C. However, the reduced lattice strain at Ts = 350 °C might be due to re-evaporation of both S and Se from the film surface, causing less distortion in the lattice.

Fig. 4
figure 4

Variation of a crystallite size and b lattice strain with substrate temperatures

3.3 Raman analysis

Figure 5 shows the Raman modes obtained for SnS1−xSex films grown at different substrate temperatures, given in the range of 50–400 cm−1. The deconvolution of peaks done by the Lorentz approximation showed the presence of various optical phonon modes inherent to SnS and SnSe phases in the spectra. It was observed that the Raman modes in the regions of 106-110 cm−1, 133 cm−1, 147/150 cm−1 correspond to SnSe, and those present at 119 cm−1, 185 cm−1 to SnSe2. Further, the relative intensities of these modes decreases with increase of substrate temperature. Similarly, the modes observed in the regions of 90 cm−1, 123 cm−1, 174/176 cm−1, 202 cm−1 are attributed to SnS, where the relative intensities are also decreases with increase of Ts, while the modes present at 95/97 cm−1, 160 cm−1, 191 cm−1, 220 cm−1 are related to SnS became more intense. In addition, with increase of the substrate temperature, a decrease in the width of modes appeared at 95/97 cm−1, 191 cm−1, 220 cm−1 is observed. Another intense mode observed at 183/185 cm−1 can be matched with both SnSe2 and SnS. Further, it should be noted that in the region of 305–312 cm−1, few broad peaks are observed, which are related to Sn2S3 and SnS2 that are not observed in the GIXRD patterns. However, their intensity is insignificant, which indicates marginal presence of such phases in the films. Further, it could be seen from Fig. 5 that initially all the characteristic Raman modes of both SnS and SnSe are not clearly appeared, rather a mixture of different peaks appeared in the form of a broad peaks in the spectra. This appearance of broad peaks might be an indication of the mixed phase of SnS and SnSe. However, with increase of substrate temperature (with decrease of S content in the films as noted from Table 2), there is a clear appearance of Raman modes of SnS and SnSe with a symmetric shape. A similar trend of the appearance of broad peaks in SnS1−xSex (0 ≤ x ≤ 1) alloyed nanosheets was reported by Wei Gao et al. [29], who attributed this to the overlap of both SnS and SnSe peaks. Thomas Schnabel et al. [40] also observed a similar behaviour in Cu2ZnGeSxSe4-x films due to mixed sulfoselenide composition. Moreover, the intensity of Raman modes of SnS1−xSex films was increased with increase of substrate temperature and became sharp, indicating good crystallinity of the films grown at higher temperatures and these results were in agreement with the GIXRD data discussed earlier. Table 4 shows the observed Raman modes of SnS1−xSex films and the reported Raman modes of binary compounds of Sn and S/Se.

Fig. 5
figure 5

Raman spectra of SnS1−xSex films deposited at different substrate temperatures

Table 4 Observed Raman modes of SnS1−xSex films deposited at different substrate temperatures

3.4 SEM analysis

Figure 6a–d shows the SEM pictures of SnS1−xSex layers deposited at different substrate temperatures. The SEM pictures clearly revealed a change in the shape and size of the grains with increase of substrate temperature. Also the grains are uniformly distributed on the substrate surface. Initially, the films formed at Ts = 200 °C showed grains that are in bunches of cereal shape. Although the shape of cereal appeared big, it contained large number of smaller grains. With increase of substrate temperature to 250 °C, smaller grains in an individual cereal slowly grown into bigger size, resembling long-type rice grains. With further increase of temperature to 300 °C, the rice shaped grains grew horizontally along the width so that the grain bulges and turned into flake shape. In general, the increase in substrate temperature can influence the nucleation density that leads to the growth of nuclei into bigger size due to more surface diffusion of ad-atoms similar to Ostwald’s ripening that occurs in solid solutions. With further increase of Ts to 350 °C, the edges of the flake shaped grains start melting and the adjacent grains coalescence together that leads to an increase in the flake size, indicating the recrystallization of grains at such temperatures. Also some grains remain in the same shape and size similar to that appeared at Ts = 300 °C. Further, the changes observed in the surface morphology of the films might be associated with the change in the composition of the films with substrate temperature. However, the layers formed at Ts = 300 °C showed uniform distribution of grains of similar shape and size on the substrate than the layers grown at other temperatures.

Fig. 6
figure 6

SEM images of SnS1−xSex films deposited at: a Ts = 200 °C, b Ts = 250 °C, c Ts = 300 °C and d Ts = 350 °C

3.5 AFM analysis

Figure 7a–d shows the 2D AFM images of as-deposited SnS1−xSex films. The images were taken over an area of 5 μm × 5 μm. The images indicate that all the films were continuous. The grains were grown in different orientations, indicating the polycrystalline nature of the films. It can be seen from the pictures that the surface roughness was highly influenced by the substrate temperature. The various surface topographical parameters from the AFM data are also evaluated. Films grown at Ts = 200 °C showed higher surface roughness than films deposited at other temperatures. This might be due to the difference in the ad-atom incorporation kinetics because of the insufficient thermal energy at such low temperatures, which leads to high average surface roughness. Because of this, the incoming flux of atoms attach to the existing nuclei so that more hill-like structures have formed on the substrate surface at lower growth temperatures. Therefore, the films exhibited higher surface roughness at such temperatures. With increase of substrate temperature, the surface roughness was reduced with increased grain size. This might be because of the coalescence of smaller grains at higher temperatures, which results in lateral film formation on the substrate surface leading to larger grains [48].

Fig. 7
figure 7

2D AFM images and corresponding grain boundary images of SnS1−xSex films deposited at: a Ts = 200 °C, b Ts = 250 °C, c Ts = 300 °C and d Ts = 350 °C

The surface roughness parameters such as average roughness (Ra), root mean square roughness (Rq), skewness (Rsk) and kurtosis (Rku) were estimated from the AFM data using NT-MDT spectrum instruments software and image analysis P9 processing module, while the average grain size (d) values were obtained using watershed method. Figure 8 shows the variation in roughness parameters with the substrate temperature of SnS1−xSex films. The average roughness and the RMS roughness of the as-deposited SnS1−xSex films were decreased and varied in the range, 36–22 nm and 45–27 nm respectively with increase of Ts, indicating growth of finest particles on the substrate at higher Ts values [50]. Further, the average grain size of the films was increased with increase of Ts as was observed in the SEM analysis. The skewness in the films was found to be positive and varied in the range, 0.5–0.2, indicating less distortion of peaks. This suggests a change from slightly asymmetric distribution of heights to symmetric distribution at higher Ts values [51]. In addition, kurtosis values were found to be > 3, implying the leptokurtic nature of the films with the appearance of highly intense and sharp peaks [52].

Fig. 8
figure 8

Variation in roughness parameters (average roughness, RMS roughness and average grain size) with substrate temperature of SnS1−xSex films

3.6 Photoconductance

The photoresponse of a material depends on the incident light intensity (L) and temperature of the specimen. The response of as-deposited SnS1−xSex films to incident photons was examined using the photoconductance measurements carried out at a constant temperature. It can be determined by measuring the conductance of the films under light (σl) and dark (σd) conditions following the relation (5) [53],

$${\text{Photoresponse}} = \frac{{\sigma_{\text{l}} - \sigma_{\text{d}} }}{{\sigma_{\text{d}} }}$$
(5)

Figure 9 shows the logarithm of photoresponse of the films with incident light intensity (L). The figure shows that the photoconductance was linearly increased with light intensity, which indicates an increased photo-generated charge carriers in the layers. A maximum photoresponse was observed for the films deposited at Ts = 300 °C. Generally, the photoresponse follows the power law with light intensity as \(L^{\gamma }\), where γ value directly gives the type of carrier recombination either mono (γ = 1) or bimolecular (γ = 0.5) in the specimen [54]. The monomolecular recombination involves an electron in the conduction band recombines with a hole in the valance band via traps (energy levels) within the band gap region, which act as recombination centers. Therefore, this type of recombination is referred as trap assisted recombination process. On the other hand, the bimolecular recombination is a band to band recombination that occurs when an electron in the conduction band directly recombines with a hole in the valence band. In the present case, the γ values evaluated for SnS1−xSex films formed at Ts = 200, 250, 300 and 350 °C were 0.56, 0.40, 0.44 and 0.47 respectively. Thus, the γ values are ~ 0.5 for all the films, representing bimolecular type recombination of charge carriers through band-to-band transitions. Therefore, it is inferred from the above results that SnS1−xSex is a potential candidate for thin film solar cell application.

Fig. 9
figure 9

Photoresponse of SnS1−xSex films with light intensity

3.7 Optical analysis

The optical properties of SnS1−xSex films were measured in the wavelength range, 400–1400 nm. Figure 10 shows the optical absorbance spectra of SnS1−xSex films deposited at different Ts values. The spectra indicates that all the films had ~ 70% of absorbance in the visible region. Using the absorbance data, the absorption coefficient (α) of the films was determined and by assuming the direct allowed band-to-band transition in the films, the optical band gap energy (Eg) of the layers was evaluated by the following Tauc relation [54], where is the incident photon energy and A is a constant.

$$\left( {\alpha h\nu } \right)^{2} = A (h\nu - E_{\text{g}} )$$
(6)
Fig. 10
figure 10

Optical absorbance spectra of SnS1−xSex films

The (α)2 versus plots of SnS1−xSex films are shown in inset of Fig. 10, from which the band gap energy of the layers can be evaluated by extrapolating the linear portion of the plots onto the energy axis. The determined Eg values of SnS1−xSex films were decreased from 1.59 to 1.46 eV with increase of Ts values. The decrease in band gap energy of the films is due to variation in composition leading to the appearance of different phases and improvement in crystalline quality of the films [56].

4 Conclusion

SnS1−xSex alloy films were deposited on glass substrates by co-evaporation of SnS and Se precursors at different substrate temperatures that varied in the range, 200–350 °C. The effect of substrate temperature on composition, structure, surface morphology, topography, photoresponse and optical properties of SnS1−xSex films was investigated. The EDS analysis confirmed that the composition of SnS1−xSex films was varied with substrate temperature. The GIXRD analysis revealed the formation of polycrystalline SnS1−xSex films with (111) plane as preferred orientation exhibiting orthorhombic crystal structure. Further, crystallinity of the films was improved with rise of substrate temperature. Raman analysis also supported the GIXRD results with the appearance of characteristic Raman modes of different phases. The SEM analysis revealed that all the as-deposited films were uniformly coated over the substrate surface and the grains appeared in different shapes as Ts increases. The AFM analysis showed that average roughness of the films was decreased from 36 to 22 nm with increase of substrate temperature. The photoresponse measurements revealed that SnS1−xSex films are very sensitive to light and the recombination of carriers was found to be bimolecular type in the films. From the optical studies, the energy band gap of the layers was decreased from 1.59 to 1.46 eV with increase of Ts. From the above analysis, it is concluded that SnS1−xSex alloy films deposited at Ts = 300 °C showed better properties compared to those layers formed at other temperatures, which can be used as an absorber layer in thin film solar cells.