New Sb2Se3-based solar cell for achieving high efficiency theoretical modeling

In this paper, we presented a numerical study of a CdS/Sb2Se3 mono junction solar cell (SC) using the SC Capacitive Simulator (SCAPS-1D). We validated an experimental work using a variety of Sb2Se3 experimental parameters, and the results showed excellent agreement between numerical and experimental J-V curves, yielding a PCE of 7.54%.To continue, we analyzed the impact of Sb2Se3 thin layer thickness, charge carrier concentration, bulk defect density, and interface defect (CdS/Sb2Se3) on solar cell characteristics. With the optimum Sb2Se3 layer thickness of 1.2 µm, carrier concentration of 1015 cm−3, bulk defect of 1013 cm−3, and CdS/Sb2Se3 interface defect densities of 1010 cm−2, we were able to attain an efficiency of 16.62%, Jsc = 35.38 mA/cm2, Voc = 0.66 V, and FF = 70.33%. Finally, we investigated the insertion effect of n-GaAs (ETL) and P+-CuO HTL (BSF) on Sb2Se3 solar cell efficiency. The novel ITO/n-CdS/n-GaAs/p-Sb2Se3/p+-CuO HTL/Au heterostructure achieved a huge efficiency of 19.60%.


Introduction
Recently, it has become clear that we must use energy transformation to improve the quality of life and increase productivity by providing access to renewable energy, which is a critical aspect of socioeconomic growth and development. In light of this, solar cell (SC) systems and thin films have received significant scientific attention and have proven commercially successful. PV materials are engineered to meet challenges such as high energy conservation, competitive prices, easy fabrication processes, and long-term longevity and stability. Several types of solar cells (SCs), including CdTe (Ahmed et al. 2020), the kesterite family (Bouarissa et al. 2021), CIGS (Biplab et al. 2020), perovskite (Jannat et al. 2021), and the family of antimony chalcogenide binary compounds (Sb 2 X 3 ) (Dong et al. 2021), have been extensively researched in the literature due to their optimum gap, strong absorption coefficients in the visible spectrum range as well as excellent power conversion efficiency (PCE). In this context, chalcogenide antimony selenide Sb 2 Se 3 (orthorhombic structure) has been recognized, as a potential SC due to its poor toxicity, low cost, earthly abundance, high electrical conductivity, strong absorption coefficient (> 10 5 cm −1 ), and appropriate energy band gap (1.03 eV indirect and 1.17 eV direct), which is close to the optimal Shockey-Queisser value (Dong et al. 2021). Although, the highest Sb 2 Se 3 thin layer SCs PCE with a CdS/Sb 2 Se 3 superstrate and a CdS/TiO 2 /Sb 2 Se 3 substrate configuration are currently 7.6% (Wen et al. 2018) and 9.2% (Spalatu et al. 2021) respectively. This experimental efficiency remains lower than that of the other semiconductor SCs. Nevertheless, the open-circuit voltage (V oc ) of the Sb 2 Se 3 SC is undoubtedly small, with values ranging from 0.3 to 0.5 V attributed to bulk recombination leakage, interfaces, and back contact recombination loss, implying a large space for approaching its theoretical thermodynamic limit (0.9 V for an E g of 1.2 eV) (Liang et al. 2020). To date, in order to fabricate a good CdS/Sb 2 Se 3 device, different film deposition methods have been utilized to improve their quality and electronic properties, such as thermal evaporation (Cang et al. 2020), vapor transporting deposition (VTD) (Tao et al. 2019), magnetron sputtering , and solution processing (Zhou et al. 2014). Using interdiffusion layers (ETL) such as TiO 2 (Spalatu et al. 2021) at the CdS/Sb 2 Se 3 interface provides one of the opportunities to eliminate the diffusion of Se and Sb to CdS, reducing interface defect formation and improving the Sb 2 Se 3 -based SC.
To improve the performance of Sb 2 Se 3 SCs, we propose using the SCAPS-1D to analyze and optimize the ITO/CdS/Sb 2 Se 3 /Au SC characteristics. We then fit and validate Sb 2 Se 3 experimental J-V characteristics using the available experimental parameters, resulting in a strong agreement between experimental and theoretical simulations (PCE = 7.54%, J sc = 29.25 mA/cm 2 , V oc = 0.44 V, and FF = 58.28%), demonstrating that the SCAPS -1D program is perfect program for describing and developing Sb 2 Se 3 -based SC characteristics. Following that, we investigated the effects of Sb 2 Se 3 film width, carrying capacity, defect density, the insertion of n-GaAs as a second buffer, and CuO HTL as a BSF on several recombination losses and efficiency. The novel combination ITO/n-CdS/n-GaAs/p-Sb 2 Se 3 / p + -CuO (HTL)/Au achieved an excellent efficiency of 19.60%, a V oc of 0.73 V, a J sc of 36.38 mA/cm 2 , and a FF of 73.47%, which may encourage the experimental laboratory to produce the same configuration. The SCAPS-1D software developed by Gent University in Belgium (Burgelman and Marlein 2008) was largely used to theoretically describe and analyze SC devices by solving semiconductor equations such as the Poisson, the continuity equations for electrons, the continuity equations for holes (1,2,3), as well as drift and diffusion Eqs. (4, 5).
In this paper, we are interested in the analysis and development of the optoelectronic performance of the Sb 2 Se 3 SC via SCAPS-1D. Note that SCAPS-1D is a simulation tool with seven semiconductor layers of input from which we can compute the effect of several electronic parameters. The physical parameters of the SC device that have x been taken into account in the simulation environment (1.5 air mass spectrums, ambient temperature of 25 °C, and 100 mW/cm 2 sun illuminations) were recorded from experimental measurements and other scientific papers and are shown in Tables 1, 2, 3, and 4.

Theoretical analysis of the ITO/CdS/Sb 2 Se 3 /Au conventionnel solar structure
In this part of the work, we show a numerical investigation and optimization of ITO/CdS/ Sb 2 Se 3 /Au SC through the SCAPS-1D tool. Figure 1 depicts the schema structure and energy band diagram of Sb 2 Se 3 hetero-junction SC. As shown, the ITO was employed as the window thin film, CdS as a buffer film, Sb 2 Se 3 as an absorber, and Au as the back electrode. It is observed from Fig. 1 that CdS/Sb 2 Se 3 has a negative conduction band offset CBO − . This negative sign indicates that the conduction band of CdS is lesser than that of Sb 2 Se 3 , which may be one of the factors for implying the free flow of electrons, hence minimizing short circuit current and thus affecting SC performance.
a. Validation of Sb 2 Se 3 simulated parameters with experimental work In this first subsection, the experimental parameters of the ITO/CdS/Sb 2 Se 3 /Au-based SC reported in the work of Xixing WEN et al., such as Sb 2 Se 3 layer thickness (0.9 µm),  Tables 1, 2, 3, and 4, are collected and fed into SCAPS-1D software. Xixing et al. (Wen et al. 2018) used vapor transport deposition of antimony selenide thin film solar cells at various heating temperatures, pressures, and substrate temperatures to analyze the crystallinity evolution and fabricate high-quality solar cells with a PCE of 7.6%, a V OC of 0.42 V, a J SC of 29.9 mA/cm 2 , and a FF of 60.4%. At 2.10 14 cm −3 in Sb 2 Se 3 defect density, we found a strong agreement between the experimental (Wen et al. 2018) and theoretical J-V curves (see Fig. 2), resulting in a PCE of 7.54%, a V OC of 0.44 V, a J SC of 29.25 mA/cm 2 , and a FF of 58.36%. This finding demonstrates the realistic models and the excellence of software used in this work.
b. Impact of Sb 2 Se 3 thickness and charges concentration on the conventional SC characteristics.
The absorber film thickness and charge concentration have a significant impact on the carriers generated when photons are incident on solar cell devices. So, after validating the Sb 2 Se 3 experimental model with theoretical model one, we start optimizing the Sb 2 Se 3 thickness and carrier density (Na) in the 0.2-1.2 μm and 10 13 -10 18 cm −3 ranges, respectively. The results for quantum efficiency (a) and current density (b) versus Sb 2 Se 3 thickness and carrier concentration are shown in Fig. 3a and b, respectively. Figure 3a shows that by increasing the absorber thickness (while keeping the other parameters constant), the quantum efficiency increases and reaches a maximum value at 1.2 μm, which can be explained by collecting the maximum number of photons, resulting in enhanced production of electron-hole pairs. As a result, the cell's output will increase, improving the overall efficiency of the Sb 2 Se 3 solar cells. Figure 3b also depicts the effect of the Sb 2 Se 3 carrier concentration density on the J-V properties. It can be seen that J sc increases with acceptor concentration to a maximum at 10 15 cm −3 and decreases with higher concentrations, which is due to the charges recombination rate and impurity scattering, which increase as acceptor carrier concentration increases, reducing carrier collection at the interface and forcing current to be drastically reduced. Figure 4 shows the evaluation and representation of the dual effects of Sb 2 Se 3 acceptor concentration and layer depth on ITO/CdS/ Sb 2 Se 3 /Au S.C. characteristics (V oc , J sc ,  Impact of Sb 2 Se 3 thick and charges carrier concentration on the PV characteristics of studied heterostructure lifetime of photogenerated electrons shortens, reducing the number of carriers gathered at the interface and thus decreasing J sc (Biplab et al. 2020). As a result, we can see in Fig. 4 that the maximum efficiency was ~ 8.25% with J sc of 18.85 mA/cm 2 , FF of 75.33%, and V oc of 0.58 V for thickness and carrier concentration of 1.2 μm and 10 18 cm −3 , respectively. However, in this section, we are interested in achieving a high J sc (35 mA/cm 2 ), which will result in the creation of more electron pairs and thus higher solar cell efficiency. So, we suggested keeping the carrier concentration density (N a ) at 10 15 cm 3 and the thickness at 1.2 μm as optimal practical values. The low FF and V oc at these optimal values can be resolved by minimizing traps at recombination centers and interface-induced recombination losses caused by bulk depth carrier trap zones, inappropriate energy-level alignment, mismatched lattice at the interface, and dangling bonds at surface interfaces (Dong et al. 2021); the implications of this will be shown in the following sections.
c. Effect of the Sb 2 Se 3 bulk defect density and interfacial defect on conventional Sb 2 Se 3 solar cell characteristics The Sb 2 Se 3 bulk defect density and interface defects are critical parameters for designing a high-performance CdS/Sb 2 Se 3 photovoltaic cell with low parasitic resistance. The most common intrinsic defects in the Sb 2 Se 3 crystal structure are V se , V Sb , Sb i , Se i , Sb Se , and Se Sb (Huang et al. 2019). As a result, we proposed analyzing and optimizing this parameter from 10 10 to 10 16 cm −3 (bulk defect) and from 10 10 to 10 16 cm −2 (CdS/Sb 2 Se 3 interface defect) to minimize higher band bending at the absorber/buffer interface, which is a major impediment to the generated electrons and holes (electrical transport across the junction interface) and bulk charge carrier recombination Figs. 4 and 5 depicts a significant decrease in the three parameters that determine the yield of the Sb 2 Se 3 device as bulk and interface defect increase, owing to an increase in trap-assisted Shockley-Read-Hall (SRH), surface recombination velocity, and reduction lifetime. The J SC and FF decrease because electrons are more likely to be captured and device resistance increases, reducing efficiency. For ITO/CdS/Sb 2 Se 3 /Au solar cell with 1.2 μm absorber layer thickness, 10 15 cm −3 carrier concentration, 10 13 cm −3 bulk defect density, and 10 10 cm −2 for CdS/Sb 2 Se 3 interface defect density, an optimal efficiency of 16.62%, V oc of 0.66 V, J sc of 35.38 mA/cm 2 , and FF of 70.33% was found, which is more promising than the efficiency of the reported article (Cang et al. 2020;Tang et al. 2019;Tao et al. 2019;Zhou et al. 2014). These results provide critical quantitative insights to understand the defect's impact on device performance. In the next part of this work, we set the Sb 2 Se 3 material parameters at their optimal values and discuss the influence of the incorporation of GaAs and CuO HTL interlayers on the device performances.
d. Theoretical Analyzing of ITO/CdS/n-GaAs/Sb 2 Se 3 /CuO HTL/Au new hetero structure n this section, we investigate and analyze the effect of n-GaAs and P + -CuO HTL insertion on the Sb 2 Se 3 SC properties. Figure 6 depicts the new hetero SC schematic configuration and band diagram. According to the band diagram, incorporating a thin n-GaAs layer (100 nm) results in a positive and low conduction band offset (CBO), which aids in the free flow of electrons from the absorber layer (p-Sb 2 Se 3 ) to hybrid buffer layers (n-GaAs/ n-CdS). Furthermore, WILLIAMS et al. (Williams et al. 2020) demonstrated that CdS is unsuitable as a direct transmitter to the Sb 2 Se 3 absorber due to Se and Sb interdiffusion, which is the original cause of the very deficient interface and Sb 2 Se 3 absorber layer, 514 Page 10 of 16  Optimal Sb 2 Se 3 solar cell structure with n-GaAs ETL and p + -CuO HTL layers potentially lowering device performance via interface recombination loss. As a result, using a thin layer of n-GaAs can provide the opportunity to fabricate high Sb 2 Se 3 SC quality with a low interfacial defect. Moreover, the insertion of CuO HTL creates a high potential barrier at back contact, potentially reducing recombination at this interface. e. Effect of the incorporation of n-GaAs and P + -CuO HTL interlayers on the Sb 2 Se 3 solar cell characteristics To create a dual-buffer-layered Sb 2 Se 3 solar cell, a second buffer layer was added to the first buffer layer, and the parameters were altered by adjusting the thickness ratio, as shown in Fig. 8a. A dual buffer layer is created here by combining n-CdS and n-GaAs. The FF and PCE values were found to be higher than in the single-buffer-layer cases. We observed a positive CBO + with an optimum offset of 0-0.4 eV at the n-GaAs/Sb 2 Se 3 interface after the addition of n-GaAs (see Fig. 6), indicating that the Sb 2 Se 3 absorber is in conjunction with the good buffer layer (n-GaAs), which can yield better efficiencies by lowering interface recombination and selective charge collection. However, high recombination of electron minority charge carriers at the metal back contact layer gives the chance to boost the SC efficiency due to the possibility of high impurity doping concentration on the back of the solar cell. This can be accomplished by injecting a higher doping concentration into the back-surface field (BSF) layer than the active absorber layer, creating a high potential barrier that can reflect electrons to the P-N junction space (Abdelkadir et al. 2022b;Ait Abdelkadir et al. 2022;Kaminski et al. 2002).
As shown in Fig. 7, using CuO HTL as a back surface field (ITO/CdS/GaAs/Sb 2 Se 3 / CuO HTL) improves SC quantum efficiency (QE) and current density, which can be explained by the high electric field between the grain boundary and the interior of the grain (Zhou et al. 2014), decreasing carriers at the deep center, and increasing the created electric potential (see Fig. 7c, d). The strong electric field at the interfaces accelerates photogenerated carrier separation at the depletion region, drawing them away from the junction quickly. While the holes pass through the HTL layer and are collected by the rear contact, the electrons travel into the buffer layer. Charge carriers avoid recombination by using band offsets to reach the metal contact (Biplab et al. 2020).
The effect of n-GaAs ETL and P + -CuO HTL interlayer thickness from 20 to 200 nm on the basic parameters of Sb 2 Se 3 SCs, including PCE, V oc , J sc , and FF, were investigated and shown in Fig. 8.
It is clearly noticed in Fig. 8a that the FF grows linearly with the thickness of the n-GaAs thin layer, leading to a rise in PCE. This could be attributed to the formation of a proper depletion region, which reduces interface string resistance and enhances carrier collection. However, a very low decrement of J sc was observed, which could be due to the high radiative recombination coefficient that we take into account in this simulation (2.3.10 -9 ), and no significant effect on SC V oc with n-GaAs layer thickness adjustments was observed. The Sb 2 Se 3 SC characteristics are saturated with the increment of the CuO HTL thickness at a high efficiency of 19.60% with J sc of 36.38 mA/cm 2 , V oc of 0.73 V, and FF of 73.47%. This rise is due to a decrease in charge carrier recombination (Ait , which improves carrier gathering and increases SC efficiency. As a result, investigators can be more confident in using n-CdS/n-GaAs hybrid buffer layers with CuO HTL as back contact to achieve maximal Sb 2 Se 3 device performance. Next, we set the n-GaAs layer thickness to 100 nm and began varying the CuO HTL (BSF) layer thickness from 20 to 200 nm (see Fig. 8b). The influence of parasitic resistance on the new hetero SC is also investigated. As illustrated in Fig. 9a, b, augmenting the R s from 0 Ω.cm 2 to 10 Ω.cm 2 causes the J SC and FF to decrease linearly, increasing the SC efficiency inversely to the increase in R sh and thus improving the SC PCE. As a result, for high Sb 2 Se 3 SC efficiency, it is necessary to fabricate this dispositive with low R s and high R sh . Furthermore, we compare the findings of this study to previous studies reported in published research. Table 5 summarizes the comparative studies of current outcomes with some recently published Sb 2 Se 3 -based SCs. We can see that the outcome of this paper paves the way for higher Sb 2 Se 3 SC efficiency.

Conclusions
In this paper, SCAPS-1D program was used to validate a theoretical model that describes the experimental Sb 2 Se 3 solar cell characteristics. We found that several parameters, including Sb 2 Se 3 thin layer thickness, charge carrier concentration, and bulk and interface defects, limit the performance of Sb 2 Se 3 -based solar cells. The analysis of several features Fig. 7 Current density versus potential (J-V) a, Quantum efficiency b, and electric field (c and) of the conventional SC and the optimal one revealed the possibility of achieving 16.62% efficiency with 1.2 μm Sb 2 Se 3 layer thickness, 10 15 cm −3 carrier concentration, 10 13 cm −3 bulk defect, and 10 10 cm −2 interface defect densities. Following this optimization study, we discovered that inserting n-GaAs (100 nm) at the n-CdS/p-Sb 2 Se 3 interface and P + -CuO HTL (100 nm) as a BSF increased the solar cell's efficiency even further. Furthermore, the inserted n-GaAs second buffer layer has been an important role in forming a positive CBO at the interface, allowing electron injection and diffusion from Sb 2 Se 3 to CdS and thus increasing the device's yield. In addition, P + -CuO HTL (BSF) was used to create a high barrier potential at the back contact, which reduces carrier recombination.
Finally, the ITO/CdS/GaAs/Sb 2 Se 3 /Au new solar cell achieves 19.60% efficiency, J sc of 36.38 mA/cm 2 , V oc of 0.73 V, and FF of 73.47%, which will be encouraging to do experimental work on Sb 2 Se 3 next-generation cost-efficient thin-film PV. Acknowledgements The authors would like to thank Dr. Burgelman of Ghent University in Belgium for providing the SCAPS 1D simulation tool, as well as everyone else who contribute significantly to this scientific paper.

Authors' contributions
The study's conception and design were contributed to by all of the authors. Material preparation, data collecting, and analysis were provided by Ph. D student AAA and Professor MS, while   (2021) This work Baig et al. (2020)