Antimony selenide thin films prepared by thermal evaporation of antimony and selenium elemental sources

Abstract

Antimony selenide thin films of 140–190 nm in thickness, with different compositions, were prepared at room temperature by using the dual-thermal evaporation technique. The compositional, structural, morphological and optical studies were performed in the as-deposited state and after annealing at 250°C for an hour, using many analytical techniques. The carrier-type, resistivity, carrier concentration and mobility of the annealed films were determined. The obtained thin films, regardless of their composition, exhibited a smooth surface with a high absorption coefficient of greater than 10⁵ cm⁻¹. The optical band gap of the near-stoichiometric film was 1.56  eV, which decreased to 1.32 eV after annealing. However, higher values were determined with increasing Se elements in the films, but these values did not exceed 1.77 eV. This study shows that antimony selenide thin films can be obtained at room temperature using a thermal evaporation technique, and the properties of the films can be improved with annealing.

KEYWORDS:

• Dual-thermal evaporation
• thin films
• Sb–Se
• annealing temperature

1. Introduction

Thin-film solar cells rely on inorganic absorbers such as copper indium gallium selenide Cu(In,Ga)Se₂ and Cadmium telluride (CdTe) to achieve high device stability and efficiency [1–3]. However, the high cost of gallium and indium, as well as cadmium’s high toxicity, may preclude further developments of Cu(In,Ga)Se₂ and CdTe solar cells. In recent years, perovskite-based solar cells have taken the lead role as their conversion efficiency has the highest reported value of 28%, superseding Cu(In,Ga)Se₂ and CdTe [4]. However, the low stability of the absorber layer and the high toxicity of lead could prevent the commercialization of these solar cells [5]. Hence, it is critical to find a new eco-friendly and earth-abundant material for thin-film solar cells.

Antimony selenide (Sb₂Se₃) is an earth-abundant non-toxic material that has a high absorption coefficient (greater than 10⁵ cm⁻¹), a direct bandgap of (1.1–1.3) eV and a long carrier lifetime (around 60 ns) [6–9]. These good properties make antimony selenide an extremely appropriate absorber layer for thin-film solar cells. According to the theoretical studies, the conversion efficiency of Sb₂Se₃ can reach 30% [10]. Experimentally, the conversion efficiency of a Sb₂Se₃ solar cell rises quickly from 1.9% to 7.6%, though it has been investigated for only a few years [8,11–13]. Consequently, antimony selenide has grown in importance among researchers worldwide.

Many methods have been used to prepare Sb₂Se₃ thin films, such as e-beam evaporation [14], electrodeposition [15,16] and chemical bath deposition [17,18]. Indeed, many studies of Sb₂Se₃ thin-film solar cells based on solution methods are low-cost methods [19–28]. However, these methods employ toxic chemicals and solvents or easily introduce impurities into the film as a result of conducting the processes in a non-vacuum environment. Vacuum-based methods are reliable and simple and have been used in the industrial production of CdTe and Cu(In,Ga)Se₂ solar cells [11,29]. Additionally, Sb₂Se₃ has a high vapour pressure [30] and low melting point [21] and prefers easy thermal evaporation over magnetron sputtering [5,31]. Some groups have deposited Sb₂Se₃ thin films by thermal evaporation at high substrate temperatures, using commercially available Sb₂Se₃ powder as a source material [1,11]. Other researchers prepared the films by co-evaporation and selenization of the metal stack grown with sputtering [32,33]. In both methods, a large amount of Se powder should be used. Especially with increasing substrate temperature because of the high sublimation rate of Sb₂Se₃ from the substrate [32,34]. In this paper, we prepare antimony selenide thin films with different compositions directly using the dual-thermal evaporation technique at room temperature. The prepared samples were annealed at 250°C for an hour. The detailed characterization by means of EDX, XRD, Raman spectroscopy, AFM and UV/VIS/NIR spectroscopy, 4-point probe and Hall system has been discussed in the present study.

2. Experimental details

Antimony selenide thin films were fabricated utilizing a homemade dual-thermal evaporation technique, following the methods explained in reference [35] for the deposition of Cu(InGa)Se₂ thin films. Two boats are used, with selenium (Se: 99.99%) powder placed in boat 1 (applying a current of 50 A) and antimony (Sb: 99.99%) shot (1–2 mm) placed in boat 2 (applying a current of about 100 A). The name of the supplier of the chemicals is SIGMA-ALDRICH. In this context, one should note that the used approach in this study to deposit antimony selenide thin films requires a separation of the two boats. This separation is necessary to vary the evaporation temperature. The boats were heated instantly to deposit the films on the glass substrates at room temperature. Thin films with different compositions were prepared by weighing Se and Sb powders in different amounts. All films were grown at a high vacuum of 10⁻⁵ mbar. The deposited thin films were then annealed in a tube furnace in a nitrogen atmosphere at 250°C for an hour.

The compositional, structural, morphological and optical studies were performed in the as-deposited films and also after annealing using different techniques. EDX, which is attached to a desktop scanning electron microscope (JEOL- JCM 6000), was used to determine the chemical composition. The crystal structure of the films was studied with X-ray diffraction (Shimadzu XRD-6000), using a CuK_α radiation source produced at an X-ray tube current and voltage of 30 mA and 40 kV, respectively. Also, Raman spectroscopy (Bruker Senterra) was employed to study the structure of the prepared thin films, using an excitation laser with a wavelength of 532 nm (5 mW). The surface morphology was studied by atomic force microscopy (AFM) (Veeco CP-II) in non-contact mode, utilizing Si tips at a 1 Hz scan rate. A UV–VIS–NIR spectrophotometer (Shimadzu) was used at room temperature to investigate the optical properties of the prepared samples with a resolution of 0.1 nm. A stylus profilometer (Dektak150) was used to determine the thicknesses of the films. Also, the electrical resistivity of the films was investigated by a computerized 4-point probe system using a Keithley 2420 instrument. The carrier-type, carrier concentration and mobility were measured using a Hall system at room temperature (Ecopia HMS 3000 Hall measurement system).

3. Results and discussion

Five samples with various compositions were prepared by means of a dual-thermal evaporation technique. The EDX spectra of the deposited thin films are shown in Figure 1(a). The samples were named as SS-0.6, SS-1, SS-1.5, SS-2 and SS-3, depending on the atomic ratio of the Se/Sb as shown in Table 1. The sample SS-1.5 shows a near-stoichiometric Sb₂Se₃ thin film as it illustrates atomic ratios of Se/Sb = 1.48. The samples SS-2 and SS-3 reveal Sb-poor thin films with different atomic concentration ratios of Se/Sb, which are 2.04 and 3.09, respectively. However, the sample SS-0.6 reveals an Sb-rich thin film with an atomic ratio of Se/Sb of 0.61. Moreover, nearly equal concentrations of the elements are revealed in SS-1, with a 1.11 atomic ratio of Se/Sb.

Figure 1. EDX spectra of the prepared thin films. (a) as-deposited thin films and (b) after annealing

Table 1. Elemental compositions of as-deposited thin films.

	Atomic composition
Samples	Se %	Sb %	Se/Sb
SS-0.6	37.75	62.25	0.61
SS-1	52.56	47.44	1.11
SS-1.5 (Near-stoichiometric)	59.74	40.26	1.48
SS-2	67.13	32.87	2.04
SS-3	75.53	24.47	3.09

The EDX spectra of the films after annealing are illustrated in Figure 1(b). The annealing process slightly decreased the atomic percentage of Se for all films, as shown in Table 2. This may be because of a weak bond of these atoms to the sample surface.

Table 2. Elemental compositions of thin films after annealing.

	Atomic composition
Samples	Se %	Sb %	Se/Sb
SS-0.6	31.85	68.15	0.47
SS-1	50.62	49.38	1.03
SS-1.5 (Near-stoichiometric)	58.05	41.95	1.38
SS-2	65.76	34.24	1.92
SS-3	71.79	28.21	2.54

XRD and Raman spectroscopy were employed to characterise the structural properties of the fabricated thin films before and after annealing. As illustrated in Figure 2, the XRD diffraction peaks of the as-deposited samples SS-1, SS-1.5, SS-2 and SS-3 match the available XRD pattern for Sb₂Se₃ (JCPDS 15-0861) [11,36]. Mainly, for SS-1, a broad peak (130) can be detected with a small intensity, which indicates inadequate crystallization. However, for the three samples SS-1.5, SS-2 and SS-3; (130) and (212) peaks are observed. In addition, (211) and (240) peaks are seen with small intensities in the near-stoichiometric sample SS-1.5, whereas SS-0.6 showed no peak, which means an amorphous film. The Raman spectra of these deposited thin films are shown in Figure 3. No peak is detected for sample SS-0.6. However, the characteristic peaks for SS-1, SS-1.5 and SS-3 samples agree with previous studies reported in the literature. The bands arising at about 83, 175, 192 and 188 cm⁻¹ are due to vibrational modes of the Sb₂Se₃ phase [14,27,36,37]. The Raman spectra of the SS-1.5 near-stoichiometric thin film reveal a peak at around 192 cm⁻¹; this is generally assigned to the Se–Sb stretching vibration mode. Also, the peak observed at 188 cm⁻¹ for the SS-3 sample appears because of the stretching of the Se–Sb–Se bonds.

Figure 2. XRD patterns of deposited thin films with different atomic ratios (Se/Sb).

Figure 3. Raman spectra of deposited thin films with different atomic ratios (Se/Sb).

The crystallinity of the prepared samples was improved with annealing, as shown in Figure 4. The XRD diffraction peaks of all samples after annealing match with the available XRD pattern for Sb₂Se₃ (JCPDS 15-0861) [11,36].

Figure 4. XRD diffraction peaks of all samples after annealing.

This result was confirmed with Raman spectroscopy, as shown in Figure 5. The detected peaks for all annealed samples agree with the reported previous studies. The detected peaks at about 184, 187, 155, 190 and 192 cm⁻¹ are generally attributed to vibrational modes of the Sb₂Se₃ phase [6,11,37,38].

Figure 5. Raman spectra of the samples after annealing.

The crystallite size (D) of the annealed thin films was calculated using Scherrer's equation: $D=0.9\lambda /\mathrm{\beta cos}\theta$where λ is the wavelength of X-rays (1.54 Å for Cu-Kα radiation), θ is Bragg’s angle and β is the FWHM of the (120) XRD peak in radians. The crystallite size of the samples varies between 151.67 and 154.20 Å for all samples except SS-3, as shown in Table 3. The crystal grain diameter of the Se-extra film, SS-3, is about 240 Å may be due to the increase of Se in the film.

Table 3. Crystallite size of the annealed thin films.

Samples	Crystallite size D (Å)
SS-0.6	153.68
SS-1	152.15
SS-1.5	154.20
SS-2	151.67
SS-3	241.645

The surface morphology of the samples was investigated by AFM, as shown in Figures 6 and 7. All the deposited films display a smooth surface with a root mean square roughness of only a few nm. Also, it can be seen that all surfaces are composed of small grains that are distributed over the surfaces. However, annealing led to improvements in the morphology of the surfaces, as shown in Figure 7.

Figure 6. AFM images of deposited thin films with different atomic ratios (Se/Sb).

Figure 7. AFM images of the samples after annealing.

The optical properties of the deposited and annealed thin films were examined by a UV–VIS–NIR spectrometer. The absorption spectra of the deposited thin films are shown in Figures 8 and 9, respectively. Generally, absorbance reveals good values for all films in the visible region before and after annealing. Before annealing, the samples SS-0.6 and SS-1 clearly have a higher value than other deposited samples. No big change was seen in these samples after annealing. However, annealing led to improved absorbance values in the samples SS-1.5, SS-2 and SS-3.

Figure 8. Absorbance spectra of deposited thin films with different atomic ratios (Se/Sb).

Figure 9. Absorbance spectra of prepared thin films after annealing.

For the calculation of the absorption coefficient (α) of the films, we used the following formula: (1) $\alpha =\frac{1}{d}\text{ln}\left\{\frac{{(1-R)}^2}{2T}+{\left[\frac{{(1-R)}^4}{4T^2}+R^2\right]}^{1/2}\right\}$(1) where T is the optical transmission, R is the reflectance and d is the thickness of the prepared film [4]. We found that all prepared films, before and after annealing, exhibit a high absorption coefficient of greater than 10⁵ cm⁻¹, which matches the reported value in the literature [7,9]. To estimate the optical energy gap (E_g), we used the following formula for direct band gap materials: (2) $(\mathrm{\alpha h\upsilon }{)}^2=A(\mathrm{h\upsilon }-E_g)$(2) where A is a constant and υ is the light frequency. We plotted (αhν)² versus hν, as shown in Figures 10 and 11. The E_g values of the sample SS-1.5 were 1.56 eV for the as-deposited film and 1.32 eV after annealing. The reduction of energy gap after annealing has been also observed by Tang et al. [39]. Obviously, this value of around 1.3 eV agrees well with the reported values for the stoichiometric Sb₂Se₃ [14,27]. The E_g values were 1.16 and 1.42 eV for as-deposited films in SS-0.6 and SS-1 samples, respectively. These values changed after annealing to 1.25 and 1.16 eV. On the other hand, SS-2 and SS-3 samples illustrate E_g values of 1.72 and 1.77 eV for as-deposited films, which decreased after annealing to 1.45 and 1.50 eV, respectively. Generally, increasing Se elements in the films leads to an increase in E_g, compared to the stoichiometric Sb₂Se₃. This observation matches the result reported by Tiwari et al., where increasing Se leads to an increase in E_g [14].

Figure 10. The plot of (αhν)² versus hν of deposited thin films with different atomic ratios (Se/Sb).

Figure 11. The plot of (αhν)² versus hν of all thin films after annealing.

The resistivity, carrier concentration, carrier-type and mobility of the annealed thin films were listed in Table 4. Clearly, increasing Se elements in the films leads to an increase in the resistivity regardless of the carriers-type, which shows the higher values with the sample SS-2 and SS-3. These samples illustrate the lower values of carrier concentration in comparison to other films. However, the variation in the mobility values is small and SS-1.5 shows the highest value of mobility.

Table 4. Resistivity, carrier concentration and mobility of the annealed thin films.

Samples	Resistivity (Ω cm)	Carrier concentration (cm^−3)	Mobility (cm^2 s^−1 v^−1)	Hall coefficient (Cm^3/c)
SS-0.6	0.0018	3.3 × 10^18	3.35 × 10^2	1.54
SS-1	0.0125	1.87 × 10^17	2.8 × 10^2	−6 × 10^2
SS-1.5 (Near-stoichiometric)	0.106	2.27 × 10^18	8.1 × 10^2	1.25 × 10^1
SS-2	2.9 × 10^5	8.4 × 10^13	1.97 × 10^1	−1.3 × 10^5
SS-3	3.9 × 10^5	2.3 × 10^12	5.57 × 10^2	−3.96 × 10^5

4. Conclusion

In this work, the dual-thermal evaporation technique has been successfully used to deposit antimony selenide thin films at room temperature. Films with different compositions have been prepared to study the effect of the composition on their properties. The obtained thin films, before and after annealing, exhibited a smooth surface with a high absorption coefficient of greater than 10⁵ cm⁻¹, regardless of their composition. The band gap of the near-stoichiometric film was 1.56 eV, which was reduced to 1.32 eV after annealing. However, higher values were defined with increasing Se elements in the films, but these values do not go over 1.77 eV. Electrical measurements show that increasing Se elements in the films leads to an increase in the resistivity and reduction in carrier concentration regardless of the carrier type. However, the variation in the mobility values is small and SS-1.5 shows the highest value of mobility. This study shows that antimony selenide thin films can be obtained at room temperature using a thermal evaporation technique, and the properties of the films can be improved with annealing.

Acknowledgements

The authors extend their appreciation to the Deputyship forResearch & Innovation, Ministry of Education in Saudi Arabia for funding this research work through the projectnumber 442/101. Also, the authors would like to extend their appreciation to Taibah University for its supervisionsupport.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Additional information

Funding

The authors extend their appreciation to the Deputyship for Research & Innovation, Ministry of Education in Saudi Arabia for funding this research work through the project number 442/101 .