Design and Simulation of GaAsN Based Solar Cell with AlGaAs blocking layer for Harvesting Visible to Near-infrared Light

In the present study, the performance parameters of GaAsN dilute nitride-based semiconductor solar cell with and without AlGaAs blocking layers have been investigated in detail by Solar Cell Capacitance Simulator in one dimensional software program ( SCAPS-1D ). The thickness of absorber, buffer, and blocking layers are varied to achieve the improvement of open circuit voltage, short circuit current, fill factor, efficiency and also to optimize the device structure. The impact of doping and defect densities on the solar cell performance parameters have been analyzed minutely inside the absorber, buffer, and blocking layers. The solar cell thermal stability parameters are also investigated in the temperature region from 273K to 373K. The efficiency of 43.90% and 40.05% are obtained from the proposed solar cells with and without AlGaAs blocking layer, respectively. The present findings may provide insightful approach for fabricating feasible, cost effective, and efficient dilute nitride solar cell.


Introduction
Dilute-nitride based solar cell structures have demonstrated much attention due to their potentiality to increase efficiency by adjusting band gap of alloy-based materials [1].
GaAs/GaAsN structure has been designed for the application of space and large-scale power plant. The improvement of the efficiency of this structure is very much necessary for widening the application in different fields [2]. Solar cells are made of a single material (Si), and singlejunction (GaAs, CdTe, CuInGaSe) to compound materials, such as perovskite, dye-sensitized, inorganic, quantum dot, quantum well, and dilute nitride based solar cells [3]. Among them alloy based dilute nitride GaAsN compound semiconductor materials solar cell has unique properties.
The alloys of GaAs-GaN provide the opportunity to fabricate GaAsN which recently attract the attention because of their negative and large band bowing (from -7 to -40 eV) largely dependent on compositional properties. The offset of conduction band (> 300 meV) is due to the size and electro-negativity difference among N, Ga and As atoms [4,5]. The new sub-bands of energy can be constructed in the region of the lower energy of the conduction band by the addition of N inside the GaAs host materials which can be described by the band anti-crossing model. The structural properties of the new sub-band can be controlled properly by changing the amount of the incorporated N atoms inside the host GaAs. The new sub-band widens spectral response up to the infrared range, which contributes to raise the overall performance parameters of the cell [5][6][7].
The investigation about optical and morphological properties of GaAsN compound solar cell has been performed by different research groups [8][9][10][11][12]. The theoretical maximum efficiency of about 30.10% and 29.00% has been reported for optimized GaInP/GaAs/GaInNAs and GaInNAs solar cell [13]. The maximum conversion efficiency has been determined to be 24.94%, when a hetero-structure configuration of p-GaAs/p-GaAsN/n-GaAs is employed n-GaAs as buffer layer, [14]. Another numerical study has achieved efficiency of 15.9% for the device structure of n + -GaAs/n + -GaAsN/p-GaAsN [15]. The reported efficiency is much lower than the expected value in dilute nitride solar cell. The introduction of various defects in GaAsN during the incorporation of N atoms that reduces the number of generated photoelectrons causes this low efficient. The scattering of alloy and non-homogeneity of N atoms reduces the electron mobility and minority carrier lifetimes with enhanced nonradiative recombination of the flowing photo-generated carriers towards the electrodes and restricts to increase performance of GaAsN/GaAs solar cell [7,11]. The solar cell with AlGaAs blocking layer can significantly increases the open circuit voltage and consequently increase the efficiency [16]. In order to optimize the performance parameters of GaAsN cell, it is very much essential to investigate doping as well as quantum efficiency (QE), defect density, and the variation of different layer's thickness of the cell structure in presence of the AlGaAs blocking layer.
The solar cells are illuminated under 100 mW/cm 2 (1sun) with global air mas AM 1.5 G solar spectrum at operating temperature 300 K, considering ideal condition for the series (R s ) and shunt (R sh ) resistances. The simulation parameters are adopted from previously studied research works as presented in Table 1.
3 Results and Discussion:

Effect of thickness on PV parameters
The effect of solar cell performance with respect to different layer thickness in Cell 1 and Cell 2 are shown in Figs. 3.  respectively, for Cell 2 ( Fig. 1(b)). It is observed from the results that due to the variation of the thickness of the layer p + -AlGaAs, there is no change of the solar cell output parameters (Fig.   3(c)), while V oc , FF and enhances owing to the rise of the thickness of the layer n-AlGaAs in Cell 2 (Fig. 3(d)). The improvement of these performance parameters may be due to the change of offset voltage (band alignment) with the increase of n-AlGaAs layer thickness [27][28][29].

Impact of thickness on the quantum efficiency
The effect of GaAsN layer thickness on quantum efficiency (QE) for solar Cell 1 and Cell 2 are demonstrated in Fig. 4. For achieving maximum efficiency as well as reducing the materials cost, the optimum thickness of p-GaAsN absorber layer is fixed at 2 m and 1.5 m for these proposed solar cells Cell 1 and Cell 2, respectively.

Influence of doping concentration on PV parameters
In order to investigate the impact of doping density, the acceptor density N A in p + -GaAs and p ++ -GaAs layers is varied from 5.0×10 14 to 1.0×10 20 cm -3 for Cell 1 and Cell 2, respectively as depicted in Figs. 5 (a) and 5(b). With increasing N A , the values of J sc initially increases for Cell 1 and remains almost constant from N A = 1.0×10 16 cm -3 up to N A = 1.0×10 20 cm -3 ( Fig. 5(a)), while no change of the value of J sc observed for the Cell 2 ( Fig. 5(b)). In contrast, the value of V oc rises from 832 mV to 1054 mV and 1058 mV to 1105 mV for Cell 1 and Cell 2, respectively which results in the enhancement of the . The reason for the increase of the J sc and V oc in Cell 1 is due to the enhancement of the reverse saturation current and electrical conductivity [15,34]. At the value of large N A , recombination of electron-hole pair reduces the number of carrier at the electrode which results in constant J sc [30,34]. As shown in Fig. 5 (c), there is no effect of V oc , J sc , FF, and  on acceptor density in the p + -AlGaAs layer. The value of V oc and  increases abruptly as the donor density N D reaches at 10 19 cm -3 in n-AlGaAs as illustrated in Fig. 5(d). The similar tendency was observed in the previous studies of [16,35].

Impact of defect density on PV parameters
The impact of total density of defect on the photovoltaic output parameters for the different layers of both the solar cells (Cell 1 and Cell 2) are depicted in Fig. 6. Here the 'broken lines' indicate the data for Cell 1 whereas, the solid lines for Cell 2.
As shown in Fig. 6(a), the values of V oc , J sc and  start to decrease at the total defect density of 5.0×10 18 cm -3 while the value of FF shows increasing tendency from 5.0×10 17 cm -3 for the p + -GaAs layer in Cell 1. On the contrary, all the solar cell performance parameters (V oc , J sc , FF and ) of p ++ -GaAs, p-AlGaAs and n-AlGaAs layers in Cell 2 remain constant with changing the total defect density as indicated in Figs. 6(b), 6(c) and 6(d), respectively. A similar results was found in the previous report [12,36]. The value of the total defect density of p-GaAsN layer is varied from 1.0×10 12 to 5.0×10 18 cm -3 and the significant impact of the total defect density on the PV parameters has been represented in Figs. 6 (e) and 6(f  [12,36] . With increasing defect density of the layer n-GaAs in Cell 1, the value of solar cell performance parameters (V oc , J sc , FF and ) decreases due to the reduction of the free carriers as well as electrical conductivity, however no variation is observed of the output parameters of n + -GaAs layer in Cell 2 as depicted in Figs. 6 (g) and 6(h). A similar tendency was found in the previous report of [37].

Effect of temperature on performance parameters:
Considering The semiconductor materials bandgap, E g decreases with rising in temperature due to reduction of the bonding energy and the electron-hole recombination processes enhances. Thus, the number electron-hole also reduces due to temperature rise which maintains constant J sc for Cell1 and Cell 2 over the range of temperature that are generated due to reduction of E g [37,38]. The FF decreases for the combined effect of the V oc , J sc , and .

Enhancement of performance parameter p-GaAsN solar cell
The electrical characteristics (J-V) curve of the designed p-GaAsN dilute nitride solar cells with and without AlGaAs blocking layer has been demonstrated in Fig. 8 Table 2 and Table 3, respectively.     [13,14,22,28]. The value of FF and J sc obtained from the numerical study is also higher than that in most of the previous reported values. The present study provides the insight for the guideline to obtain outmost efficiency of 40.05% and 43.90% for GaASN solar cell without and with AlGaAs blocking layer.

Conclusions
Dilute nitride solar cells of p-GaAsN absorber layer, have been designed to exploit absorption of light in the wide range and investigated in detail. The numerical study has been done by SCAPS-1D simulator to analyze the impact of layer thickness, defect and doping density, and temperature variation on the output parameters such as J sc , V oc , FF and . The numerical simulated output indicates that the solar cell containing AlGaAs blocking layers (Cell 2) exhibits an overall efficiency of 43.90% with doping density of 1.0×10 16