Optimization of Absorber and ETM Layer Thickness for Enhanced Tin based Perovskite Solar Cell Performance using SCAPS-1D Software

The methyl ammonium tin iodide ( �� � �� � ��� � ) perovskite nanocrystals have attracted research interest and have become a rising star in the horizon of photovoltaics due to its narrow band gap, wide visible absorption coefficient and environmental friendliness than its lead-based counterpart ( �� � �� � ��� � ). In this article, a tin based perovskite solar cell with Zinc oxide (ZnO) and Copper Oxide (CuO) as electron transport medium (ETM) and hole transport medium (HTM) was proposed and investigated numerically using a Solar Cell Capacitance Simulator (SCAPS) tool. With appropriate parameters, a short-circuit current density (Jsc) of 27.56 ��/�� � , open-circuit voltage (Voc) of 0.82 � , fill factor (FF) of 59.32 % , and power conversion efficiency (PCE) of 13.41 % are obtained for the initial simulation. By varying the thicknesses of the absorber and electron transport layer, the optimum thicknesses were observed at 0.6 �� and 0.3 �� for �� � �� � ��� � and ZnO with corresponding PCEs of 14.36 % and 13.42 % . Upon simulation with optimized parameters, a Jsc of 29. 71 ��/�� � , Voc of 0.83 � , FF of 61.23 % and PCE of 15. 10 % were recorded. These values are superior to those obtained without optimization which means that solar cell performance can be improved to some extent by adjusting the perovskite and electron transport layer and also, �� � �� � ��� � Perovskite solar cell (PSC) is a potential environmentally friendly solar cell with considerable efficiency.


I. INTRODUCTION
rganic-inorganic hybrid perovskite nanocrystals have shown the most promising candidacy for high efficient and low-cost solar cells. Halide perovskites have the formula of [1], where is the cation including , , , etc., is or , and the anion is a halogen ion (usually , , or ). Ideally, it has a cubic crystal structure consisting of a corner-sharing BX6 octahedral network with a cation in the interstices. Halide perovskite nanocrystal has the unique property of weak exciton binding energy [2,3] which means light induced excitons will dissociate into free carriers quickly at room temperature. Perovskite materials are characterized with a long carrier diffusion length and high carrier diffusion velocity [4][5][6].
Reported perovskite covers a wide range of band gap energies, from (1.1 ), (1.6 ), 2.3 ) to (3.1 ) [7]. Furthermore, by fine tuning the composition of cations ( , , , etc.) and anions ( , , etc.), it is also possible to vary the absorption spectrum [8]. The application of perovskite nanocrystals was first introduced into dye sensitized solar cell (DSSC) to replace dye pigment by Kojima et al. [9] which results to a record PCE of 3.80 %. Since from then, several device modification and device engineering were done to achieve a PCE > 25 % recently [10].
is considered the mostly used ETM for PSCs device due to its high performance in solar cells as a result of its proper band gap and high transmittance. However, obtaining good quality film of either compact or mesoporous requires high annealing temperature, which limits its application in solar devices and results to increase in the production cost. Consequently, the electron mobility of perovskite materials is ~7. 5 and that of ranged between 0.1-4.0 . These lower values of electron mobility in may result to shortfall in performance of solar cells [11].
Other alternatives to include [12][13], [14][15], [16][17], [18][19] and [20]. For , and , the band gap is larger, and the conduction band edge is much higher than the conduction band of the perovskite layer which prevents smooth electron injection into both , and . Thus, the excited electron remains in the conduction band of the perovskite layer for a longer time [21]. Among them, zinc oxide has many properties that can be used in PSCs, such as high transmittance in the visible spectra and more importantly, its low cost and much higher electron mobility of 115 − 155 which can potentially improve the electron transport efficiency and reduce the recombination loss as an ETM [4,[22][23][24].
The ETM is used to compensate and balance the difference of hole and electron diffusion lengths [2,4]. In addition, the ETM is a blocking layer that prevents holes from reaching the fluorine-doped tin oxide (FTO) electrode. For high performance solar cells, ETMs should meet the following criteria: (a) good optical transmittance in the visible range, which reduces the optical energy loss; (b) the energy levels of ETMs should match that of perovskite materials, which improve the electron extraction efficiency and block holes; (c) good electron mobility. As a result, the design and materials properties of the ETM are crucial for solar cell performance [2][3][4]25].
Lead based perovskites materials are considered as promising candidates for future-generation photovoltaics owing to their unique optoelectronic properties and very low fabrication cost. Despite its exhibited properties, the presence of toxic lead poses a severe concern regarding their environmental friendliness and practical deployment [26]. Furthermore, the intrinsic band gaps that are generally greater than 1. 5 have prevented the realization of its predicted theoretical value. Based on the above known facts, tin halide perovskites have displayed some properties as alternative candidates to lead based absorber. Tin based have large carrier mobility and strong light absorption coefficients due to its electronic configurations similar to those of lead based. Moreover, the band gaps of tin perovskites can be tailored to ∼1.4 eV, approaching the ideal band gaps for single-junction solar cells [10]. Although the overall PCE of tin base is still much lower than that of lead base at present, the field is witnessing their continued rapid progress, as compared to lead base that have actually reached attained level in their development. In this regard, tin is very promising because it combines the merits of high performance, low cost, and the absence of lead metal.
The properties for the layers (absorber and ETM) make them ( a promising ETM and a promising absorber) for PSCs. Meanwhile, although most physical properties of and are similar, there are also some distinct properties for each material. As a result, studies on based solar cells will enrich the family of PSCs, which will in turn help to improve the performance of PSCs. In this article, we designed and studied perovskite solar cells based on and as ETM and absorber using SCAPS-1D. The effect of varying the thickness of various layers through getting the optimized values were explored systematically. The results show that, thickness of absorber and ETM are essential factors to be considered in a solar cell.

II. MODELING AND SIMULATION
The modeling of the solar cell was done using SCAPS [27][28][29]. The software is based on the basic equations of the semiconductor: hole (1) and electrons (2) continuity equations together with Poisson equation (3) as follows [30]: Where ( ) denotes ionized acceptor-like doping concentration and ( ) denotes ionized donor-like doping concentration. ( ), ( ), ( ), ( ) refer to trapped holes, free electrons and free holes respectively, is the direction along the thickness, D is diffusion coefficient, is generation rate, is electric field, is electron charge, permittivity and is electrostatic potential. By obtaining the solution of the above equations, outputs such as the recombination profile, current voltage characteristics, spectral response and band diagram can be gotten. The device structure of the simulated PSCs is considered with layer configuration as shown in Fig. 1. All the data used in the simulation is as summarized in Tables I and II [14,30,[31][32][33][34][35].    . In the simulation studies, the influence of shunt resistance and series resistance were ignored due to power loses by providing an alternate current path for the incident current generated. Front and back contact work function are 4.0 (FTO) and 4.47 (silver), respectively. For the simulation under illumination the standard AM 1.5 spectrum is used and the cell operating temperature is set at 300 . In our study, we optimized the device configuration as follows: The parameters of FTO, and are kept unchanged and the thickness of 3 3 3 is varied to get the optimum performance parameters. The same procedure is adopted for optimizing the thickness of used in this study by keeping the parameters of FTO, 3 3 3 and unchanged.

A. Modeled PSC, Energy level diagram and absorption coefficient
The PSC and band structure of the tin based perovskite solar cell obtained with simulated parameters are shown in Fig. 1 Fig. 1(b)). The values of ∆ and ∆ are beneficial for carriers in the modeled PSC.
The absorption coefficient of , and (SCAPS data file) is shown in Fig. 1(c) and (d), which shows the fraction of light lost due to scattering and absorption per unit distance of the penetration medium. It is seen from Fig.  1(c) that the absorption coefficient decreases with the increase in the visible wavelength range. This indicates that the fraction of light lost due to scattering and absorbance decreases. It can also be noted from Fig. 1(d) that the value of absorption coefficient decreases with the visible region. With respect to the photon energy (see Fig. 1(d)), we observed an increase in the absorption coefficient. The increase in the absorption coefficient indicates the scattering loss of light while travelling through the medium with high absorption.

B. Analysis for initial device
With these initial parameters from Tables I and II, we studied the current density-voltage (J-V) and quantum efficiency-wavelength ( − ) characteristic of the cell (see Fig. 2 [36], certifying that the device simulation is valid and the input parameters that have been set are close to those for a real device. However, looking at the quantum measurement, the QE also increase with photon energy increase (see Fig. 2(c)). The QE covers the entire visible spectrum and reaches a broad absorption maximum > 80% from 380 to 980 (see Fig. 2(b)) which is in agreement with similar studies [32]. The sweeping at the visible and near IR region of the QE curve is beneficial to the light absorption at the various wavelengths.

C. Effect of thickness of Perovskite Absorber nanocrystal
The thickness of the light-absorbing layer plays a critical role in establishing the performance of perovskite solar cells [14,32,34,37]. The variation of the cell performance with the thickness of the absorption layer is shown in Fig. 3(c-f). The thickness of absorber layer was varied from 0.1 to 1.0 . When the thickness of the absorber is too low, the absorption of light is too low and results to low PV parameters. With the increase in thickness from 0.1 to 0.6 , the PCEs of the cells improve significantly and when the thickness exceeds 0. 6 , the PCE of the cell slows down in growth, this is because, if the absorber layer is too thick (above 0.6 ), the photogenerated carriers cannot be collected effectively because they must travel through the absorber to reach the carrier collecting layers before quenching of charge carriers take place [32].
The Voc is constant for devices with a perovskite layer thickness from 0.6 to 1.0 . The main parameter that is negatively affected by the increase of the perovskite layer thickness is the FF, which drops strongly for device thicknesses from 0.1 to 0.5 . When the thickness is increased from 0.6 to 1.0 , the FF improves, as a result, the PCE also increased. This implies that the FF is related to the efficiency of charge extraction that resulted from smaller built in voltage in the thicker devices [37], which means that, increasing thickness increases the photon-capturing ability, which results in an increase in the rate of generation of charge carriers [38].

D. Effect of thickness of the ETM
The Thickness of the ETM significantly affect the performance of a solar cell. Importantly, the selection of the appropriate ETM plays a significant role on the design and implementation of high efficiency perovskite solar cell as the energy band alignment between absorber and ETM layer is a crucial factor for the efficiency improvement of PSCs [12,39,40]. The variation of performance parameters with thicknesses of the ETM is shown in Fig. 4(a-d).
From the results of our simulation, the efficiency decreases slightly from 13.42 % to 13.39 % as thickness is increased from 0.3 to 1.9 . The results here show that, the ETM does  [41,42]. This can be witnessed on the basis of the fact that perovskite material itself could help the generation of charge carriers by photon excitation and ETM layer is just a charge transport layer. Even in the absence of the ETM, the FTO which is an n-type layer will directly have contact with the perovskite layer and then transport the electron without affecting the PCE. Increasing the ETM layer thickness reduces the Jsc of the PSCs by increasing photon absorption and resistance of the cell (see Fig. 5 that shows increase in photon energy with increase QE which is a function of material absorption). The increased QE with increase photon energies can be attributed to the increase in absorption coefficient within the regions and consequence of increased density of localized states in the gap itself due to the rise in new defect states [43] Table III). The FF and Voc are constant for all devices with thickness from 0.3 to 1.9 (see Table III). As a result of the decreasing Jsc, the PCE also decreases with increasing ETM thicknesses at same values recorded for the Jsc which means at those thicknesses, there are no losses or gains which makes the parameters unchanged. At this point, the numbers of photo-generated carriers are equal to the number of absorbed photons. Fig. 6(a-i) describes graphically the solar cell current densities as a function of open circuit voltage.  Fig. 7(a)). When the optimized results are compared with the initial device without optimization, an enhancement of ~1.13 times in PCE, ~1.08 times in Jsc, ~1.01 times in Voc and ~1.03 times in FF. From the simulated parameters, the QE vs.
(wavelength and photon energy) and the energy level diagram for the optimized perovskite solar cell are shown in Fig. 7(b), (c) and (d) respectively. The conduction and valence band offset at / 3 3 3 and / interface were reduced, which can be considered beneficial for the flow of photo-excited charge carriers to prevent losses. The quantum efficiency also shows stronger absorber in the visible region and near IR region.

IV. CONCLUSION
In this paper, a numerical simulation of perovskite solar cells with configuration of / / / was studied using SCAPS-1D tool. Two most important factors (which are absorber and ETM thicknesses) that affect the performance of PSC were investigated. We found that these two factors influence the metrics parameters of the simulated PSCs. The results show that the optimal thickness was 0.6 and the optimal thickness was 0.3 which results to overall PSC with the following photovoltaic performance, Jsc of 29. 71 , Voc of 0.83 , FF of 61.23 % and PCE of 15.10 %.