Technology of fabrication of C d S x T e 1-x solid solution on silicon substrate

. Heterojunction between Si and CdSxTe1-x have been obtained by the method of vacuum deposition of powders of cadmium sulfide and cadmium telluride on the surface of monocrystalline silicon. The optimal temperature regime for the growth of the CdSxTe1-x solid solution on the silicon surface has been determined. The values of the crystal lattice constant and the thickness of the CdSxTe1-x solid solution at the interface of the n/Si – n/ CdSxTe1-x heterostructure are calculated.


Introduction
Obtaining a heterojunction based on silicon (Si) [1][2] and A 2 B 6 [3][4] and A 3 B 5 semiconductor compounds is of scientific and practical interest. This would allow more efficient use of the potential capabilities of the silicon with A 2 B 6 and A 3 B 5 semiconductor compounds in solid-state electronics, photovoltaics and photovoltaic energy. In this work certain design issues in the technology of obtaining an efficient heterojunction based on Si and cadmium telluride CdTe without surface states are considered. However, it is known [5] that the crystal lattice constant of silicon (α = 5.43Ǻ) and cadmium telluride CdTe (α = 6.477Ǻ) differ considerably, therefore, in order to create an effective heterojunction between these semiconductor materials, it is necessary to use an intermediate semiconductor material that would match their crystal lattice constants. Such a material can be a CdSxTe1-x solid solution (SS), which is continuous and its lattice constant changes from α = 5.84Ǻ in cadmium sulfide to α = 6.477Ǻ in cadmium telluride. In this solid solution, one can find a composition, in which the lattice constant would correspond to the lattice constant of silicon. For this purpose, CdSxTe1-x solid solutions of various compositions were formed on the silicon surface.
For this purpose, CdS and CdTe powders in a weight ratio of 10:1 were placed in a quartz crucible and sprayed onto the silicon surface in a vacuum of 10 -5 torr. In this case, the source -the crucible was heated in the aisles to the temperatures of 950 0 C -1100 0 C, and the substrate (Si) was at a temperature of 150-250 0 C.

Methods
We created n/Si -n/CdSxTe1-x heterojunctions by sputtering indium in vacuum of Torr 5 10  with a thickness of ~ 400-500Å on the surface of high-resistance n-type films with a resistivity of sm     . The obtained heterostructures have the most acceptable parameters: no-load voltage and short-circuit current at a temperature of tS = 400 0 C. The n/Si -n/CdSxTe1-x structures obtained at tS =465 0 C have electrophysical characteristics, in particular volt-ampere characteristics, as in n-i-n-structures with a long base, where d/L = 14-55 (d is a thickness of the i-layer, L is the diffusion length of minority current carriers).
With an increase in temperature tS, the thickness of the CdSxTe1-x layer increases, and the output parameters of the n/Si -n/CdSxTe1-x heterostructures deteriorate. Therefore, in this structure, studies were carried out at tS = 400 0 C and 465 0 C.
The phase composition of the CdSxTe1-x solid solution was investigated using the photoelectric method. The spectral distribution of photosensitivity was measured in the valve mode on a ZMR-3 monochromator at room temperature. The radiation source was a DKSSh-1000 xenon lamp with a luminous flux of 53000 lm, brightness up to 120 Mcd/m 2 with a central light spot operating in the mode of the minimum allowable power. The radiation is calibrated in absolute units using a thermoelement with an RTE-9 quartz window. The DKSSh-1000 lamp has a continuous spectrum in the UV and visible regions, and powerful radiation lines in the near IR region (800-820 nm). In this method, it is of great importance to establish the sizes of the entrance and exit slits and the resolution of the monochromator, which make it possible to correctly measure the parameters of the sample. The performed estimate showed [6] that the slit width should be ~ 10 µm. However, in the experiment, a wider slit was set up in order to ensure sufficient photosensitivity of the sample under study, while retaining the possibility of studying subtle phenomena. In this case, the resolution of the monochromator does not exceed 3% in the investigated region of the spectrum.

Results and discussion
One of the important issues is the study of the composition of the CdSxTe1-x solid solution. Figures 1a and 1b show the spectral distributions of the photosensitivity of n/Sin/CdSxTe1-x heterostructures obtained for structures grown at substrate temperatures tS = 400 and 465 0 C. In the spectral distribution of photosensitivity between the intrinsic absorption edges of CdS and CdTe for the n/Si -n/CdSxTe1-x heterostructure obtained at a temperature of tS = 400 0 С, there are clear peaks λ1 max = 0.735 μm and λ2 max = 0.930 μm (Fig. 1a). The spectral distribution of the photosensitivity of the n/Si -n/CdSxTe1-x heterostructure obtained at a temperature of tS = 465 0 С has several peaks at λ1 max = 0.664 μm, λ2 max = 0.692 μm, λ3 max = 0.725 μm, λ4 max = 0.750 μm, λ5 max = 0.788 μm, λ6 max = 0.814 μm, λ7 max = 0.870 μm, and λ8 max = 0.905 μm (Fig. 1b). Based on the relationship between the composition of the solid solution and the band gap Eg(х) [7], the composition of the solid solution CdSxTe1-x was determined for the found values of Eg. The determination of the value of the fundamental absorption edge λc.f.a, the band gap Eg, and the composition х of SS for all detected peaks in the photosensitivity spectrum at tS = 400 0 C and 465 0 C are given in Table 1. The values of the lattice constant а0 (х) for the detected solid solutions CdSxTe1-x, which are estimated by the empirical formula а0 (х) [A] = 0,6477−0,0657х [3], are also given in Table 1. It should be noted that when determining the composition of CdSxTe1-x solid solutions with Eg = 1.44 eV using the empirical formula Eg [eV] = 1.74x 2 -1.01x + 1.51 [8], the value for х is obtained in the form of two positive values. For other values of Eg of SS, only one positive value for х is obtained. Therefore, in Table 1, for solid solutions with Eg = 1.44 eV, two values of the composition are given, and for other Eg, one value of the composition х.
The solid solution with the composition х = 0.8 is closer in composition to the CdS layers, and the solid solution with the composition х = 0.28 is closer to the CdTe layers. The estimate shows that the value of а0 (x) for Si and SS with a composition of х = 0.8 differ by 8%, and а0 (x) for CdS and SS films with x = 0.8 differ by only 2%.
Distribution of chemical elements over the surface of obtained layers is investigated. Analyzes were performed on a Jeol -JXA -8900 microanalytical complex using an EMF LINK ISIS (energy -dispersive spectrometer); error ± 2.0%. Shooting conditions: V = 20 kV, I = 10 nA. Standards: native Cd, Te and Si, for S -synthetic FeS. The measurement results and the microphotograph are shown in Fig. 2 a) and b). As can be seen from Fig. 2b, the emission intensity of the secondary electrons knocked out from the surface layers of the elements of cadmium and sulfide is maximum. This means that the surface of the films mainly consists of cadmium sulfide. The next layer consists of a solid solution of cadmium sulfide and cadmium telluride, then a layer of cadmium telluride and a solid solution of cadmium telluride and silicon (in table 2). Table 2. Quantitative analysis of chemical components of surfeces of (n/Si-CdSxTe1-x-n/CdS) layers.  ) and b) show microphotographs taken from the cleavage of the layers, as well as the dependences of the distribution of chemical elements on the thickness of the layers. The measurement was carried out at several points. The results of studies of the dependence of chemical elements on thickness show that in all measured directions the distribution of chemical elements is almost the same and the average spread is no more than ~ 5%. Figure 3b also shows the dependences of the layers over the thickness of the films. The top-most microphotograph corresponds to the silicon substrate. The following microphotographs show the distributions of the chemical elements of sulfide, cadmium and telluride in the layers. From the figure it is also possible to estimate the respective thicknesses of each composition. For example, the thickness of cadmium sulfide is approximately one third of the thickness of cadmium telluride (the bump is the third micrograph from the top). The plateau corresponds to the cadmium telluride layer (from the bottom, the second microphotograph on the left side after the crest). It is also possible to estimate the approximate thickness of solid solutions from the figure (decreases in the plot).

Conclusion
Analyzes of the results obtained show that a solid solution between silicon and cadmium telluride, as well as cadmium telluride and cadmium sulfide is formed with a thickness of ~ up to 2 μm and ~ up to 1 μm, respectively (Fig. 3). According to the authors of [5], on the basis of А 2 В 6 compounds in the А 2 -В 6 systems, regions of solid solutions (homogeneity regions) are formed, the length of which can be much larger than that of the А 3 В 5 compounds. The composition of solid solutions based on А 2 В 6 compounds can be controlled by setting the conditions for their preparation or processing. Physical and physicochemical properties of solid solutions change with a change in composition, and the nature of the change can be both linear and as well as more complex.