Spin-Coated Kesterite Based CZTS (Cu₂ZnSnS₄) Thin Films: Structural, Optical, and Compositional Analysis for Photovoltaic Applications

1 Introduction

Kesterite Cu2ZnSnS4 (CZTS) thin films have emerged as promising candidates for earth-abundant, non-toxic solar cells due to their tunable direct bandgap (~1.5 eV), high absorption coefficient (α ≥ 104 cm−1), and low-cost fabrication potential [1,2,3]. Unlike other absorber materials, such as CIGS and CdTe, CZTS is composed of earth-abundant, non-toxic elements, making it economically viable and environmentally friendly [4]. However, despite these benefits, the power conversion efficiency of CZTS-based solar cells remains considerably below their theoretical limit, mainly due to inherent flaws, cation disorder, band tailing, and the creation of secondary phases. These difficulties create deep-level trap states, increase non-radiative recombination, and impair charge transport, therefore constraining device performance [5,6].

CZTS thin films typically crystallize in the kesterite structure, where sulfur anions tetrahedrally coordinate the metal cations (Cu, Zn, Sn) [7,8,9]. The high sulfur content can promote secondary phase formation, posing challenges for achieving single-phase films [10]. Controlling film composition and thickness is therefore critical for optimizing structural, optical, and electronic properties. Although several studies have explored CZTS stoichiometry, annealing conditions, and precursor types, systematic investigations of thickness-dependent structural and optical properties for spin-coated CZTS films are limited [11,12]. Film thickness influences crystallinity, grain size, defect density, bandgap, and Urbach energy, which collectively determine photovoltaic performance. In particular, an optimal thickness is necessary to balance light absorption and carrier collection, since excessively thin films suffer from inadequate absorption while overly thick films cause recombination losses and structural defects. Therefore, precise thickness control is a vital aspect in achieving high-quality CZTS absorber layers.

In this work, spin-coated CZTS thin films with 5–20 layers (~400–1000 nm) were fabricated and systematically analyzed. Microstructural parameters (crystallite size, dislocation density, micro-strain) were extracted from XRD, SEM was used to examine grain morphology, and optical properties including bandgap and Urbach energy were studied. Comparison with recent high-impact literature demonstrates the potential of these films for high-efficiency CZTS solar cells, providing a foundation for future device fabrication and interface optimization [13,14].

2 Materials and Method

To synthesize CZTS thin-film absorber, 40 mL of 2-methoxy ethanol (CH3O·CH2·CH2OH) was used to dissolve copper-chloride dihydrate (CuCl2·2H2O), zinc-chloride (ZnCl2), stannous chloride (SnCl2·2H2O), and thiourea (SC(NH2)2). The concentrations of these compounds were 0.24 M, 0.12 M, 0.08 M, and 0.96 M, correspondingly. The mixture was stirred on a magnetic hotplate at 55°C for 10 min. Subsequently, 7 drops of ethanolamine (C2H7NO) were added as a stabilizer, and stirring continued for an additional 50 min at 55°C to yield a homogeneous, transparent yellow sol-gel solution.

Absorber layers were deposited on SLG substrate using a spin coater (SPS SPIN 150) at a final rotation speed of 3000 rpm for a total of 27.5 s, divided into two sequential steps: 15.5 s and 12.0 s. The rotation speed ramped up at a rate of 500 rpm/s. After placing the substrate on the chuck and centering it, 5 drops of solution were dispensed at the center. The substrate was then dried at 140°C for 10 min and allowed to cool for 5 min to form a single layer (Fig. 1). This procedure was repeated 5, 10, 15, and 20 times to obtain film thicknesses of 459 nm, 608 nm, 822 nm, and 1005 nm, respectively. Finally, the films were annealed at 450°C for 40 min under N2 atmosphere.

Figure 1: Schematic of Spin-coating process for thin film absorber layers.

The film thickness was estimated using a stylus profilometer, a technique widely employed for its high precision in quantifying the thickness of thin films, coatings, and surface layers, which is displayed in Table 1.

3 Results and Discussion

3.1 Structural Properties

3.1.1 X-Ray Diffraction (XRD) Analysis

The crystalline structure of the sol-gel-processed CZTS absorbers was examined using an X-ray diffractometer (GBC, EMMA) operated at 35 kV and 28 mA, with Cu-Kα radiation (λ = 1.5406 Å). The diffraction patterns were recorded over a 2θ range of 20°–80°. The XRD patterns (Fig. 2) confirm that all films are polycrystalline tetragonal kesterite CZTS, with dominant peaks at 2θ~28.54°, 47.45°, and 56.26° corresponding to the (112), (220), and (312) planes. These peak positions are in good agreement with the standard ICDD data (JCPDS No. 26-0575), indicating the formation of phase-pure kesterite CZTS.

Figure 2: XRD patterns for CZTS absorber layers.

The narrow full width at half maximum (FWHM) values and the sharper diffraction peaks indicate enhanced crystallinity and improved microstructural quality of the films. Lower FWHM values and reduced dislocation density suggest fewer lattice defects, which are critical for absorber layer performance in photovoltaic applications [15]. Grain size plays a vital role in determining the performance of CZTS solar cells, as larger grains reduce grain boundary recombination and facilitate efficient charge transport [12]. Among the studied samples, the 10-layer film with a thickness of approximately 608 nm exhibits the sharpest diffraction peak and the lowest FWHM, indicating superior crystalline quality and larger crystallite size.

To evaluate the size of the crystallite (D), the Debye-Scherrer formula was applied [16].

The X-ray wavelength, denoted by λ , is equivalent to 0.15406 nm. Whereas β represents the FWHM of the diffraction peak at 2 θ , while θ corresponds to the Bragg diffraction angle.

The 10-layer CZTS absorber exhibits the narrowest FWHM (0.53°) and the largest crystallite size (15.47 nm), reflecting improved crystallinity (Table 2). The dislocation density ( δ ), is used to determine how many dislocations there are in a crystalline material per unit volume [17]:

In nanostructured materials, internal lattice strain is often induced by crystal dislocations, which disrupt the ideal arrangement of atoms within the lattice. Based on the isotropic strain model of Williamson-Hall (W-HISM), lattice strain and crystallite size are considered to be independent parameters [18]. However, the anisotropic Williamson-Hall model based on uniform deformation energy density (W-HUDEDM) suggests that as crystallite size decreases, the increase in volume defects at grain boundaries, due to enhanced strain energy concentration. The dislocation density of the present CZTS thin films varies from 4.17849 × 1015 to 1.48 × 1016 lines/m2, depending on film thickness. These values are comparable to those reported for solution-processed CZTS thin films. For instance, dislocation densities on the order of 1015–1016 lines/m2 have been reported for sol-gel and spin-coated CZTS absorber layers [19,20]. The comparable defect density indicates that the films prepared in this study possess microstructural quality consistent with well-crystallized CZTS absorbers, which is beneficial for efficient charge transport and reduced recombination losses. As shown in Table 2, the 10-layer CZTS thin film exhibits the lowest dislocation density ( δ = 4.17849 × 1015 lines/m2), along with a corresponding reduction in lattice strain. This reduced defect density indicates fewer lattice imperfections and a more organized crystalline structure, which is beneficial for efficient charge transport and minimized recombination losses in photovoltaic applications [20]. To keep things simple, the thin film absorber’s strain, ε , is calculated as:

The variation of crystallite size, dislocation density, and lattice strain clearly indicates that the 10-layer CZTS film possesses the most optimized microstructural properties. The reduced defect density and minimal strain at this thickness can be attributed to improved nucleation and grain growth processes, resulting in enhanced crystallinity and superior structural quality [21,22]. As shown in Fig. 3, lower FWHM values correspond to larger crystallite size and lower dislocation density, confirming improved crystal quality.

Figure 3: Variation of FWHM (β), crystallite size and dislocation density with film thickness.

It should be noted that the crystallite size calculated from XRD represents coherently diffracting domain size, which is generally smaller than the surface grain size observed in SEM micrographs. The SEM micrographs indicate that the films consist of agglomerated grains formed by the joining of multiple nanocrystallites, which explains the apparent difference between the XRD-derived crystallite size (8–15 nm) and the larger surface features.

3.1.2 Raman Spectroscopic Analysis

Raman spectroscopy was employed to further verify phase purity and to identify possible secondary phases. Raman spectra were recorded in the wavenumber range of 140–495 cm−1, as shown in Fig. 4. All samples exhibit a prominent Raman peak at approximately 328 cm−1, characteristic of kesterite CZTS, in agreement with reported literature [9,23]. No additional peaks corresponding to secondary phases such as orthorhombic Cu2SnS3, cubic ZnS, SnS2, or Cu2−xS were detected. Notably, the absence of Raman signatures associated with Cu2−xS conductive secondary phases indicates that although the films exhibit slight Cu-rich composition, the formation of detrimental shunting phases was effectively suppressed. This observation confirms that CZTS nanoparticles in a single phase were successfully produced for every solvent [23,24].

Figure 4: Raman spectra of CZTS thin film absorber layers.

Among all samples, the 10-layer film displays the highest Raman peak intensity and sharpness, confirming superior crystallinity and phase purity. Thus, the combined XRD and Raman analyses confirm the formation of well-crystallized, phase-pure polycrystalline CZTS absorber layers.

3.2 Morphological and Compositional Characterization

The surface morphology of CZTS thin films was examined using scanning electron microscopy (SEM, EVO 18, Zeiss), as shown in Fig. 5a–d. The micrographs reveal a tight and linked granular structure, resembling entangled nanoscale networks. The 5-layer film exhibits noticeable gaps and microcracks, while increasing the deposition layers leads to a significant reduction of these structural imperfections. The 10-layer film displays dense, uniform grains with minimal voids, indicating improved film compactness and surface uniformity.

Figure 5: SEM images of CZTS absorber layers: (a) 5 layer, (b) 10 layer, (c) 15 layer, and (d) 20 layer.

This morphological evolution can be explained by the nucleation growth mechanism described by Shinde et al. [25]. Initially, metal ions form nucleation sites on the substrate, followed by sulfur incorporation and CZTS production. Subsequent grain expansion and coalescence lead to bigger crystalline domains, resulting in denser and more compact films with increasing thickness. The morphological enhancement observed in the 10-layer film is consistent with the XRD data, which also reveal better crystallinity and optimized microstructural quality.

The elemental composition of the films was examined using energy-dispersive X-ray spectroscopy (EDX). The related spectra (Fig. 6) and quantitative data presented in Table 3 demonstrate that all samples exhibit slight Cu-rich and Zn-poor compositions compared to the ideal stoichiometric ratio. Such compositional deviation is commonly observed in solution-processed CZTS films and is attributed to the higher mobility of Cu ions and the higher volatility of Zn and S species during annealing [26,27,28].

Figure 6: EDX spectra of CZTS absorber layers for different film thicknesses: (a) 5 layer, (b) 10 layer, (c) 15 layer, and (d) 20 layer.

Although Cu-rich compositions can improve single-phase development, excessive Cu content may lead to the development of highly conductive Cu2−xS phases, which reduce solar cell efficiency by introducing shunting routes. However, Raman analysis shows the absence of such detrimental secondary phases, showing that the Cu-rich state stays within an acceptable range. Among all samples, the 10-layer film exhibits the most balanced stoichiometry, approaching the ideal Cu:Zn:Sn:S ratio of 2:1:1:4, consistent with its superior structural and morphological properties [29].

3.3 Optical Properties

The optical properties of CZTS thin films were examined through the utilization of a UV-visible spectrophotometer operating in the 400–1000 nm wavelength range. Fig. 7 shows the variation of absorption coefficient (α) with photon energy. All samples exhibit strong absorption in the visible region, with absorption coefficients exceeding 104 cm−1, confirming their suitability as absorber layers for photovoltaic applications [24,30].

Figure 7: Absorption coefficient α versus photon energy (h υ ) of CZTS thin films.

The absorbance can be used to compute the absorption coefficient using Lambert-Beer’s law [31], α=2.303Ad(4) where absorbance is denoted by A, d (cm) is the film thickness. In thin films, reflection and interference at the film interfaces may influence optical transmission. However, no noticeable interference fringes were observed in the spectra, and due to the strong absorption nature of CZTS films, Lambert’s law provides a reasonable approximation for estimating the absorption coefficient.

The optical bandgap ( E g ) was estimated by plotting α E 2 vs. E , where E = h υ represents the incoming photons’ quantum energy, and α is an absorption coefficient (Abs), as shown in Fig. 8.

Figure 8: Bandgap calculation of CZTS absorber layers: (a) 5 layer, (b) 10 layer, (c) 15 layer, and (d) 20 layer.

The linear extrapolation of these plots yields direct bandgap energies ranging from 1.48 to 1.53 eV (Table 4), which are close to the ideal bandgap (~1.5 eV) for single-junction solar cell applications [32]. The estimated uncertainty in the extracted bandgap values is ±0.02 eV, primarily arising from the linear fitting procedure. The obtained band gap values are consistent with previously reported CZTS thin films, which typically exhibit band gaps in the range of 1.45–1.55 eV depending on deposition technique and composition.

A slight reduction in bandgap with increasing film thickness is observed, which can be attributed to improved crystallinity and reduced quantum confinement effects in thicker films [26,33].

The Urbach energy (EU), which reflects the degree of structural disorder and defect density, was determined using [34]: α=α0expEEU(5) where absorption coefficient is denoted by α , α 0 is a constant and the Urbach energy is represented by E U (as illustrated in Fig. 9).

Figure 9: Dependence of ln ( α ) on incident photon energy for CZTS thin films.

The lowest Urbach energy value (0.57 eV) recorded for the 10-layer film suggests lower defect density, increased crystallinity, and enhanced optoelectronic quality. In contrast, greater Urbach energies (0.67–0.78 eV) suggest increased disorder [35], likely coming from antisite defects, grain boundary states, and compositional changes. Such defects can introduce localized states within the bandgap, enhancing non-radiative recombination and deteriorating photovoltaic performance by reducing open-circuit voltage (Voc) and fill factor (FF) [36,37,38]. Further tuning of deposition and annealing settings is expected to decrease these defect states and boost device performance. The Urbach energy values acquired in this study (0.57–0.78 eV) are analogous to those documented for solution-processed CZTS films. Higher Urbach energies (0.60–0.90 eV) have been documented as a result of structural disorder, but lower values have been attained in high-quality films produced using various methods. For example, Azmi et al. reported Urbach energies of approximately 0.29–0.35 eV, while Moreno et al. observed values below 0.40 eV for improved crystalline CZTS films. These results suggest that more modification of growth conditions could diminish defect states and enhance optoelectronic quality [39,40].

The Moss relation was applied to calculate the refractive index (n) of the CZTS thin film absorber layers analyzed in this study [26]: Egn4=k(6) where k denotes a constant equal to 108 eV.

Fig. 10 illustrates the inverse relationship between refractive index and bandgap. A higher refractive index (~2.5–3.0) combined with an optimal bandgap (~1.5 eV confirms the excellent optical quality of the CZTS absorber layers [36], particularly for the 10-layer film.

Figure 10: Relation between Refractive index (n) with Bandgap ( E g ).

4 Conclusions

Spin-coated CZTS thin films with optimized thickness (~608 nm) exhibit high crystallinity, controlled microstructural properties, and an optimal direct bandgap of 1.50 ± 0.02 eV, indicating strong potential for photovoltaic applications. SEM analysis confirms that grain size increases with film thickness, correlating with XRD-derived crystallite size, while Urbach energy analysis provides insights into defects and compositional inhomogeneity. Although Cu-rich compositions promote single-phase kesterite formation, precise stoichiometry control is essential to suppress secondary Cu2−xS phases that may induce shunting pathways and reduce open-circuit voltage, as widely reported in recent studies [6,41,42].

This work develops a complete thickness-dependent framework for spin-coated CZTS thin films by integrating structural, compositional, morphological, and optical investigations. Future study should focus on full device fabrication, interface engineering, and precision stoichiometry optimization to minimize defect-assisted recombination and boost power conversion efficiency. Overall, the present findings give crucial parameters for constructing high-performance, low-cost, and earth-abundant kesterite-based photovoltaic systems.