SCAPS allows simulation of one-dimensional solar cells [7]. SCAPS software was originally developed to model and simulate the electrical characteristics of heterojunction and thin-film photovoltaic solar cells. It has also been tested on CdTe and CIGS solar cell structures by M. Burgelman et al. [8] [9], the developers. The simulated and measured results were in good agreement with the practical cases. Moreover, several versions have improved its capacities allowing their applicability to crystalline solar cells (Si and GaAs) and amorphous cells (a-Si and Si micro-morph) [10]. In addition, SCAPS has a database containing several types of solar cells, data file, with modifiable parameters (thickness, doping).

The equations that govern the studied model are the equations of Poisson and continuity for electrons and holes.

d 2 φ d x 2 = − q ε 0 ε r ( p − n − N A − + N D + + ρ t q ) (1)

And:

{ 1 q d J x d x = G − R n − 1 q d J p d x = G − R p (2)

In Equation (1) and Equation (2), φ describes electrical potential, q is the unit charge, ε₀ and ε_r are the dielectric constants of vacuum and semiconductor, respectively; n and p are electron and hole density, respectively; N A − and N D + are the density of ionized donors and acceptors, ρ_t the charge density of defects, J_n and J_p the electron and hole current density; G is the generation rate and R_n(p) is the recombination loss.

The transportation of carriers is shown by:

{ J n = − q D n d n d x + q μ n d φ d x J p = − q D p d p d x − q μ p d φ d x (3)

where μ_n(p) is the electron/hole mobility and D_n(p) is the diffusion coefficient.

The generation rate is given by:

G ( λ , x ) = φ 0 ( λ ) e − α ( λ ) ⋅ x (4)

where φ₀ is the incident photon flux, x is a depth and α the absorption coefficient of the active material and λ, the absorbed photon wave length. Cell performance was achieved with scaps under standard conditions (a cell temperature of 25˚C and radiation of 1000 W/m² with an air mass spectrum of 1.5 (AM1.5)).

We aim to optimize the efficiency of our solar cell by modifying the characteristics (mobility, doping) of CdTe with SCAPS. It makes it possible to study and analyze the evolution of characteristics such as Jsc, Voc, FF and therefore the efficiency η as a function of the intrinsic properties of doped CdTe.

The recommended method for introducing a solar cell is to use the graphical interface of SCAPS. In this way, we can interactively introduce all the parameters that will be controlled by SCAPS so as not to have impossible or inconsistent values [7].

We used a thin film solar cell in ITO/CdS/CdTe structure, where CdTe the base layer, CdS the emitter and ITO the antireflecting layer. The front contact is made of aluminum. All the thicknesses of ITO, CdS and CdTe layer were respectively 0.30, 0.08 and 4.00 µm. The parameters of the materials used in this modeling were selected from experimental data in the literature and from the characteristic values (see Table 1) [7] [10] [11].

With the data, we can save the variations of Jsc, Voc, FF and analyze the efficiency η. the J-V characteristic modeled in Figure 1 uses the values in Table 1.

The efficiency of the CdTe/CdS solar cell reached 21.34% similar to that of a practical CdTe based solar cell [11]. This proves that our solar cell model is valid and can be used to perform a simulation analysis of its performance with different parameters of the material. Then, we look for the effect of the evolution of the electrical characteristics (Jsc, Voc, FF and η) as a function of the intrinsic properties of the material (doping, mobility) in order to optimize them and predict their behavior of the photocell.

The efficiency (η) is analyzed as a function of the electrons and holes mobility (Figure 2). The efficiency of the cell decreases with the increase in mobility of the electrons (Figure 2(A)) and increases with the speed of the holes (Figure 2(B)). These results are in agreement with those of the literature. Because a couple of mobility of the charge carriers which is not optimized adds constraints on the thickness of the active layer and increases the losses by recombinations [12].

The conversion efficiency is evaluated based on carrier mobility couples. The yield is maximal and equal to 22.10% for a couple of mobility equal to 100 cm²/Vs for electrons and 25 cm²/Vs to that of the holes (Table 2).

We found that the optimum torque value of carrier mobility is required to obtain maximum efficiency of the cell.

The acceptor density was varied from 10¹² cm⁻³ to 10¹⁹ cm⁻³ in steps of 10 cm⁻³. The decrease in the acceptor density of CdTe greatly reduces the yield (Figure 3). Recombination becomes important with an increase in the density of carriers. Therefore, an increase in density can drop the collection of free electrons generated by the incident radiation on front contact and contribute to lowering the value of Jsc [13]. In our conditions, a density of 10¹⁶ cm⁻³ is the optimum value which would increase the yield.

The density of the effective states was varied from 7.9 × 10¹⁷ cm⁻³ to 22 × 10¹⁷ cm⁻³. An effective state density in the smaller conduction band (CB) gives the best performance [14]. The more the density of the effective states in the CB increases the more the efficiency of the cell decreases (Figure 4). A density of 7.9 × 10¹⁷ cm⁻³ is the best value.

A low electronic affinity of the CdTe absorber indicates a low efficiency from 3.8 eV. For an electronic affinity between 3.8 eV and 3.85 eV, the efficiency equal to 21.37% is maximum and almost constant throughout the range before decreasing after 3.85 eV (Figure 5).

The results obtained are consistent because the differences in electronic affinities and band gap energies cause a discontinuity at the level of the connection of the conduction and valence bands. Therefore, the influence of the electronic affinity of the CdTe layer (CdTe_gap = 1.45 eV vs CdS_gap = 2.5 eV) in the solar cell, acts on the energy difference in the conduction band.

However, for CdS/CdTe cell, the values of ΔEc (χ_CdS − χ_CdTe) lie between 0.20 and 0.25 eV and are found to be independent of the process of cell fabrication [15].