Influence of the Thickness of a Layer of Potassium Fluoride Incorporated in the CIGS / CdS Interface on the Macroscopic Electrical Parameters of the Solar Cell

In this work, the heterojunction composed of a n-type ZnO transparent conductive oxide (OTC) layer, a n-type CdS buffer layer and a absorber layer based Cu (In, Ga)Se2 p doped is studied under the influence of a KF layer placed in the CIGS/CdS interface. This study was done by varying the thickness of KF using thin-film simulation software named SCAPS-1D. The presence of KF for a doping of the CIGS absorber of 1016cm-3 improves strongly the electrical parameters that are the Vco, the Jcc the FF, the maximum power and the conversion efficiency of the solar cell Ƞ. However, a decrease of the FF and the Jcc is noticed when the thickness of the KF is greater than 30nm causing a deterioration of the performances of the cell.


Introduction
During the past three decades, CIGS-based solar cells have been widely studied and significant progresses have been realized leading to an increasing development of this technology.Various record yields have been reported in the literature in recent years: EMPA: 20.4 % ( Chirilă, A. and al., 2013), ZSW: 20.8% (Jackson, P. and al., 2014), Solar Frontier: 20.9% (Kushia, K., 2014), Solibro 21.0% (Herrmann, D. et al., 2014), 21.7% (Jackson, P. and al., 2015) and 22.7% (Jackson, P. and al., 2016).The use of potassium fluoride allowed the optimization of the efficiency of the solar cell.Studies on the beneficial effect of potassium fluoride on the generation and recombination of carriers in this type of cell have already been published (Niane, D. and al., & Niane, D. and al., 2018).In this work, a numerical simulation study of the influence of the thickness of the potassium fluoride layer is made on the macroscopic electrical parameters of the solar cell.

Numerical Modelling
Most of the studies on CIGS-based solar cells are experimental, to the detriment of theoretical research which is still insufficient.However, the calculation remains a simple and economical method and it saves us the effort and the time to optimize the parameters of the solar cells in thin layers.In this paper, the study of the influence of potassium fluoride on the macroscopic electrical parameters of the CIGS-based solar cell is done with the SCAPS-1D software.It is a software that resolves the fundamental semiconductor equations (Burgelman, M. andal., 2000 &Burgelman, M. andal., 2013).
Where ε is the dielectric constant, φ the electrostatic potential, n and p the concentration of the free carriers,  and  are the densities of the ionized acceptors and donors, J n and J p the current densities of electrons and holes, R the rate of recombination and G the generation rate of electron-hole.
The KF layer used in this work at a band gap of 1.5eV taken from the experimental data provided in O'Bryan and Skinner's work, cited by Mott and Gurney, its electronic affinity 4.14eV and its dielectric constant 6.05eV (Mott, N. F., & Gurney, R. W. 1964).The KF layer is placed between the CIGS absorbing layer and the CdS buffer layer.The structure of the n-ZnO/n-CdS/p-CIGS heterojunction configured in the software is shown in Figure 1.The basic parameters of the four cell layers used in the simulation are shown in Table 1.

Results and Discussions
The most important parameter of a solar cell is its efficiency ƞ.It is defined as the ratio of the maximum electrical power to the power delivered by the cell of the power density irradiated by the light source.The parameters characteristic of the cells which are usually used in to describe the performance of the cell more are open-circuit voltage (Vco), the density of short-circuit current (Jcc) and the factor of form (FF).
The current-voltage density (J-V) characteristics of solar cells under illumination can be described by the phenomenological equation of the diode (Archer, M. D., 2001).
where J is the current density, V is the applied bias voltage, J 0 is a pre-factor named saturation current density, q is the charge of the electron, Rs and Rp are the series and parallel resistors respectively, A is the ideal factor of the diode, kT is the Boltzmann constant multiplied by the absolute temperature, and Jph is the current generated by light.
The equation of the diode (4) contains photocurrents of independent polarization, it is ideal for example when the diffusion lengths of the minority carriers are very small.In the absence of series resistance and shunts (ie Rs = 0 and Rp = ∞), the equation of the diode is simplified: Thus, the following expressions for the short circuit current Jcc (V = 0) and the open circuit voltage Vco (Jcc = 0) are obtained: The efficiency of the cell is calculated from the ratio of the maximal power density delivered by the cell (JmVm) to the irradiated power density Ps: where the form factor FF is defined as: Figure 2 shows the characteristic J (V) in the case with and without presence of KF.Simulations are made taking 15nm of KF and the doping rate of the CIGS was set at N b =10 16 cm -3 at room temperature 25℃.The table confirms that the solar cell in the presence of KF gives better performance.The efficiency of the cell went from 18.5762% for a cell without KF to 22.6609% for a cell with KF.We estimate the increase at 4.18%.This could be attributed to the beneficial effects of KF on the parameters whose efficiency depends.In what follows, we will study the effect of the KF layer thickness on the electrical macroscopic parameters of the cells that are generally used to describe the performance of the cell.

Short-Circuit Current Density (Jcc)
In Figure 3, the short-circuit density is represented as a function of the thickness of the KF layer.The profile of the short-circuit current density shows an increase as a function of the thickness of the KF layer within the photocell.This increase is more pronounced when the thickness is between 0 and 10nm.Then, this maximum value of Jcc becomes constant when the thickness is between 10nm and 25nm and finally beyond 25nm a slight decrease in the value of Jcc is noticed.Improvement of Jcc with the presence of the KF layer is caused by passivation of the CIGS/CdS interface.This results in a high probability that carriers can be collected in the cell.

Open Circuit Voltage
The effect of the thickness of the KF layer on the open circuit voltage is shown in Figure 4.

Maximal Power
The Figure 5 illustrates the effect of the thickness of KF on the maximum power of the solar cell.
We note an increase of the power with the increase of the thickness of KF up to an optimal zone between 8nm and 25nm.Then, the power decreases gradually when the thickness exceeds 25nm.This decrease is due to the losses of current generated by the solar cell.

Form Factor
The profile of the form factor as a function of the thickness of the KF layer is shown in Figure 6.We find that the form factor is influenced by the increase of the KF layer in the CIGS/CdS interface.Indeed, an improvement of the form factor is observable up to a thickness of KF of around 10nm corresponding to a maximum value of 83.18%, which then follows a deterioration of this parameter.The deterioration of the cell form factor is related to the loss factors noted in the short-circuit current density and the maximum power while its improvement can be attributed to the inter-diffusion of K in the CIGS.

Cell Efficiency
Figure 7 illustrates the shape of the efficiency of the cell ƞ as a function of the thickness of the KF layer.
We note an increase in the conversion efficiency of the cell with the thickness of the KF layer.However, a gradual decrease of ƞ is noticed when the amount of KF exceeds 25nm.The additional gain in efficiency can be attributed to the improved passivation of the CIGS/CdS interface which is caused by a lowering of the valence band maximum compared to the Fermi energy level and on the other hand by the increase of Vco and FF.Moreover, we can say in accordance with the yields found in the literature that the optimal thickness of KF must be taken between 10nm and 30nm not to damage the solar cell.

External Quantum Efficiency
The external quantum efficiency noted (QE), represents the ratio of the number of charges collected on the number of incident photons as a function of the wavelength.These measurements are used to characterize the photocurrent of cells and determine the origin of losses Jcc, his expression is given by (Niane, D. and al., 2016): Whit J(λ) the current collected at one hundred wavelength λ, q the elementary charge, and φ(λ) the photon flux at the chosen wavelength.
The quantum yield depends on several properties of the material such as the absorption coefficient, the junction depth, the thickness of the space charge zone, the lifetime of the carriers and their mobility but also the surface recombination.The quantum efficiency of a solar cell as a function of the wavelength of the incident light is called the spectral response.Figure 8 shows the variation of the quantum efficiency as a function of the wavelength of a solar cell with KF and another cell without KF.From 300 to 400nm and from 400 to 500nm corresponding respectively to the absorption zone of the ZnO window layer and that of the buffer layer CdS, we see clearly that the presence of KF in the CIGS/CdS interface results a more important quantum efficiency.Beyond 550nm corresponding to the absorption zone of the absorbing layer CIGS, the external quantum efficiency continues to increase with the presence of KF reaching a maximum corresponding to 98.5%, but nevertheless decreases with increasing lengths wave.This can be explained by the fact that the CIGS has a direct gap, the absorption in the heterojunction is greater in the base, and it improves when the KF layer is put in the interface CIGS/CdS.Indeed, the presence of the KF layer reduces the width of the band gap and multiplies the phonons.The mechanism of production of the minority carriers will dominate the recombination mechanism in this wavelength range, thus leading to the increase of the current density, hence the quantum efficiency.However, a reversal of the curves is noted for wavelengths greater than 1000nm.

Conclusion
Potassium fluoride has a very important role in the passivation of the CIGS/CdS interface.In this work, we studied the influence of the KF content on the parameters of a n-ZnO/n-CdS/p-CIGS heterojunction solar cell.The conversion efficiency of the cell increased from 18.5762% for a cell without KF to 22.6609% for a cell with KF.However, increasing the thickness of the KF layer beyond 30nm showed a deterioration of parameters such as form factor and current density as well as conversion efficiency.This suggests that the optimal range for a better performance of the cell must be between [10nm, 30nm].

Figure 1 .
Figure 1.Structure of a heterojunction solar cell n-ZnO/n-CdS/p-CIGS

Figure 2 .
Figure 2. Characteristic J(V) Without KF and With KF

Figure 3 .
Figure 3. Short-circuit current density as a function of the thickness of the KF layer

Figure 4 .
Figure 4. Open circuit voltage as a function of the thickness of the KF layer

Figure 5 .
Figure 5. Maximum power as a function of the thickness of the KF layer

Figure 6 .
Figure 6.Form factor as a function of the thickness of the KF layer

Figure 7 .
Figure 7. Efficiency of the cell ƞ as a function of the thickness of the KF layer

Figure 8 .
Figure 8. External quantum efficiency as a function of the wavelength (λ) of a solar cell with KF and another cell without KF

Table 1 .
The basic parameters of the different layers of the cell used in the simulation: (a) and (d) denote the acceptor and donor states respectively

Table 2
compared the different values obtained from one cell with KF and another cell without KF.

Table 2 .
Comparison of the electrical parameters of a cell without KF and another cell with KF