Study of Composition and Optical Properties of Chemically Deposited Pd-xSb2S3 Thin Films

INTRODUCTION

In recent times, the potentials of antimony sulphide (Sb₂S₃) for use in various optoelectronic and solar thermal devices including solar cell applications have been widely established in the literature [1-4]. Antimony Sulphide is a direct band gap semiconductor material that has been under investigation for use in various device designs over the years and most recently, for application in photovoltaic solar cell devices. For instance, in photovoltaic applications, Sb₂S₃ has been used as the p-type absorber layers in making heterojunction solar cell devices [5] with improved solar conversion efficiency. The use of antimony sulphide thin film for applications in other optoelectronic devices, medicinal use, luminescence materials, and in various industrial uses have been widely reported in the literature [6-14].  Reports by different research groups indicate that antimony sulphide thin films can be grown using variety of different thin film technology including; spray pyrolysis [15-16], pulsed laser ablation deposition technique [17] laser assisted-chemical bath deposition [18], Chemical vapour deposition method [2], spin technique [4], chemical bath deposition technique [6-7], pulsed electro-deposition technique  [19], two-stage process [20], rapid thermal process [21], hydrothermal deposition method [22], sulphidisation method [23], thermal evaporation [24-25], successive ionic layer and reaction (SILAR) technique [26], photochemical deposition method [27], etc.

It is a common knowledge that chemical bath deposition method is a more universally used technique due to the fact that the method offers high quality thin films at low temperature, suitable for depositing large area semiconductor thin films, and has proved to be the simplest and the most economical since the equipment used for deposition are very common and easily affordable. It is understood that  thin film deposition on a substrate can be achieved by either or all of the two steps of nucleation and particle growth. In general, for thin films grown by the solution growth technique, the process of nucleation and film growth are mostly  due to an inter-play of other processes which includes;  simple-ion  cluster  mechanism,  simple  hydroxide  cluster  mechanism,  complex  ion-by-ion  decomposition mechanism,  and  the  complex-cluster  decomposition  mechanism.  The  simple  ion  process could diffuse to the substrate to initiate nucleation and the nucleated layers then  grow by adsorption of ions in the solution  and or nucleation of new crystals. The films formed by the crystals are generally held together by weak forces (Van der Waals  forces). Thus, the chemical bath deposition process  mostly employs a controlled chemical reaction to achieve thin film deposition by precipitation of the desired compound.

According to the literature [26-29], the intrinsic properties of Sb₂S₃ thin films has been modified by introducing different impurities in order increase the versatility of Sb₂S₃–based thin films in different  applications.  The use of palladium impurities to optimise the properties of Sb₂S₃  thin films is relatively rare in the literature hence to the best of our knowledge,  this report is novel in engineering the properties of Sb₂S₃ thin films for enhanced device applications. In the present investigation, the aim of the study is to grow thin films of Sb₂S₃, dope the layers with different concentrations of the dopants, and to characterise the layers using compositional and optical spectroscopy to investigate the composition and the optical constants respectively. This report is a fundamental step toward exploring new pathways for utilization of Sb₂S₃-based thin films in different device designs. 

MATERIALS AND METHODS

Substrate cleaning plays a fundamental role in thin film deposition. The glass slides of dimensions 75 mm by 25 mm by 1mm, were purchased from local suppliers. Prior to the deposition, the glass substrates were cleaned first by a mild detergent solution, then degreased with acetone, and further etched with 5% of hydrochloric acid (HCl) for 60 minutes. Finally, the glass slides were cleaned ultrasonically  by double distilled water and then dried in air. All the source chemicals used for the deposition of Sb₂S₃ thin films were analytical grade, and were obtained from Sigma Aldrich UK through local suppliers and employed directly without further purification. The source of the antimony ions and sulphide ions were antimony trichloride (Sb₂Cl₃) and sodium thiosulphate (Na₂SO₃) respectively.

A 4g of antimony trichloride with the desired amount of acetone was prepared and included in a clean beaker and sealed. A 31 g of sodium thiosulphate powder was dissolved in 125 ml of water and stirred for 15 minutes. The reaction bath contained a 20 ml each of the cation source  Sb+, and anion source S-, and 10 ml of the complexing agent, and then stirred for 10 minutes using a magnetic stirrer. The temperature of the reaction bath was maintained at 298 K and the rotational speed of the stirrer was fixed at 50 revolutions per minute.  The pH of the solution was found to be acidic, typically 5.0. The solution was then distributed into 2 separate beakers with one kept as control, and in the other, a 10 ml of the different concentrations of the dopant were included separately and carefully labelled. The glass substrates were held vertically using a synthetic foam and deposition was allowed to take place for 1 hour. The films were then removed carefully at the end of the deposition time, and washed with distilled water, and then dried.

The film thickness was measured using standard procedure in the current literature as  reported by different authors [6-8], hence the film thickness was measured using  the  gravimetric  method  or  double  weight  method. The film thickness were in the range 185 nm for the as-deposited layers and between 205 nm to 215 nm for the doped layers. The transmittance and absorbance versus wavelength measurements were done using a Unico –UV-2102 PC  spectrophotometer operated  at  normal  incident  of  light  in  the  wavelength  range  of  300  nm  to  1000 nm.

RESULTS AND DISCUSSION

Fig. 1 show the transmittance versus wavelength plots for the as-grown layers and layers doped at the respective concentrations of the palladium impurities. The transmittance exhibited a concentration dependent behaviour in that the transmittances were higher for films grown at concentrations ≤ 0.3 M and decreased by almost 50% for the as-grown films.  This behaviour was attributed to the increase in the film thicknesses at the higher concentrations of the palladium impurities. Such increase can cause an increase in the crystallites sizes in the films grown at the higher concentrations. The increased crystallites size as they tend to approach the bulk crystalline Sb₂S₃ could lead to larger unfilled inter-granular volume so that the absorption per unit thickness is reduced in the films, hence causing the observed phenomena.

Fig. 1 also reveals that a blue shift effect is  induced by the introduction of the palladium impurities on the transmittance spectra. This observation could be attributed to the effect quantum confinement as a result of the reduced particle dimension as the dopants tend to blend with the host atoms. The effect of quantum size effects induced by similar or different parameters on the transmittance properties of chalcogenides thin films has been reported by other research groups  independent of the deposition technique, in the literature  [31-32]..

Fig. 2 show the variation of the absorbance against wavelength plots. The result indicate that the absorbance decreased in the blue region of the electromagnetic spectrum (lower wavelengths) up to the region of the fundamental absorption, and then decreased gradually to minima at the region of lower photon energies (longer wavelengths). In the literature, such observation are common in the absorbance versus wavelength plots for various chalcogenides thin films [30-34]. An important observation in Fig. 2 is that the absorbance were typically lower for the films grown at the higher concentrations, and higher for films doped at concentrations ≤ 0.2 M, and exhibiting maximum values for the as-deposited film. 

     Fig. 3 show the plots of (αhν)² vs hν.  The optical absorption coefficient α , was calculated using the formula contained in the literature  [35-38] as;

                                                   (1)

In equation (1),α retains its meaning, d is the film thickness (in nm for this study), and T is the transmittance in percentage. The optical absorption coefficients data were used to evaluate the energy band gap according to the relation contained in the literature [38-39] as;

                                           (2)

As shown in equation (2),  α retains its meaning, h is  the Planck’s constant, ν is the frequency of the electromagnetic radiation, B is an energy independent constant,  but generally depend on the refractive index and the effective masses of the hole and electron respectively [38], Eg is the energy band gap, and n is an index that determines the nature of the transition exhibited by the materials under investigation. In direct transition,  it is universally accepted that n = 0.5. A close look on Fig.3, clearly reveals that the transition are all direct with values in the range 2.30 eV to 2.48 eV. The values are within the range reported by other research groups [30-32], and also confirms that the films could be used in different optoelectronic applications especially in solar cell devices, lasers, and light emitting diodes (LED). It also explores the possibilty of fabricating Sb₂S₃-based homojunction solar cell devices with improved efficiencies.

     Fig. 4 shows the variation of the extinction coefficient with photon energy. The extinction coefficient was calculated using the relation contained in the literature [7, 16, 18-19, 36, 39]  as;

                                                               (3)

In equation (3), α retains its meaning, λ is the wavelength (300 nm to 1000 nm), and π is a constant. The extinction coefficients increased gradually down to the region of the fundamental absorption and then increased. The plots for the as-deposited layers were also higher compared to that of  the doped layers. As shown in Fig. 4, it was also observed that the films grown at concentrations ≥ 0.3 M exhibited lower values of the extinction coefficient compared to the films doped with palladium impurites of concentrations ≤ 0.2 M. Such behaviour is typical of most thin films including the chalcogenides family. This behaviour observed in Fig. (4) is in agreement with the reports of other research groups [18-19, 39-40] in the literature.

Fig. 5 show the  variation of the optical density with the photon energy for the as-deposited antimony sulphide thin films, and at the various concentrations of the dopants investigated in the study. The optical density was calculated using the relation [40];

                                      (4)

In equation (4), α retains its meaning and d is the film thickness.  The optical density were typically in the range 1.2 to 8.3.

           Fig. 6. show the change imposed on the variation of the energy band gap and film thickness with the different concentrations of the dopants. The values the energy band gap was highest for the as-grown layer (i.e zero palladium impurity), and then increased gradually with an increase in the concentration of the palladium impurities. The film thicknesses also exhibited similar corresponding increase with increase of the dopants.

Figs. 7. show a typical RBS plots, indicating the influence of the dopant impurities on the composition of the layers. The results show that the post-deposition heat treatments of the layers influenced the composition in that the percentage of oxygen was relatively high in all the annealed cases. It is possible that this increased oxidation during the annealing process also contributes significantly to the shift of the energy band gap toward higher values.

Fig. 8. gives the X-ray diffractograms of the layers. The result from the X-ray diffractometry indicates that the layers crystallised in stibnite orthorombic crystal structure, consistent with the International Centre for Diffraction Data (ICDD) Powder Diffraction File- PDF:001-0538. The results show diffraction peaks in all the layers, indicating that they are all polycrystalline. However the effect of the concentrations of the palladium impurities on the X-ray diffractograms are clearly pronounced especially with respect to the degree of texturing.  At the higher concentrations of the dopants, the diffraction peaks belonging to (002), (100) and (211) were increasing consistently with an increase in concentrations of the palladium impurities. The plots (Fig. 8) also show that at concentration of 0.3 M, there was a suppresion of the other diffraction peaks, with the (221) peaks appearing as the most prominent diffraction peaks. Such variation of the diffraction peaks caused by change in concentration and/or other deposition variables have been widely reported by other authors for similar/related chlacogenides in the literature [41-50].

Fig. 9 shows the SEM micrograph of the un-doped layer while Fig. 10 shows the SEM micrograph of the layer that was doped with 0.3 M of the palladium impurities. Physical observation of the micrographs clearly indicate that the increase in the concentration of dopants generally increased the crystallites. The grains show seed-like structures and were uniformly packed. Other authors have observed similar findings in the literature [41-50].

CONCLUSIONS

The influence of palladium impurities at different concentrations on the compositional and optical properties of chemically deposited antimony sulphide thin films is reported. The results show that the presence of the dopants modified the optical properties significantly. In particular, the transmittances of the doped layers were higher compared to the transmittances of the  as-deposited films. The energy band gap were found to be direct in both cases, with values in the range 2.30 eV to 2.38 eV. The extinction coefficient were typically lower for the palladium doped layers. The values of the energy band gap strongly indicate that the films could be used in different device designs including optoelectronic applications especially as absorber layers in solar cell devices.

ACKNOWLEDGEMENTS

The author would like to thank the technical staff of Materials for Energy & Environmental Sustainability, Obafemi Awolowo University, Ile Ife, Nigeria, for performing the characterisation.

Conflict of interest

The authors declare that there are no conflicts of interest regarding the publication of this manuscript.

If you have any problem in viewing images, kindly please download the PDF file.