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

Abstract

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%.

1 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 (Sb₂X₃) (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 Sb₂Se₃ (orthorhombic structure) has been recognized, as a potential SC due to its poor toxicity, low cost, earthly abundance, high electrical conductivity, strong absorption coefficient (> 10⁵ 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 Sb₂Se₃ thin layer SCs PCE with a CdS/Sb₂Se₃ superstrate and a CdS/TiO₂/Sb₂Se₃ 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 Sb₂Se₃ 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/Sb₂Se₃ 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 (Tang et al. 2019), and solution processing (Zhou et al. 2014). Using interdiffusion layers (ETL) such as TiO₂ (Spalatu et al. 2021) at the CdS/Sb₂Se₃ interface provides one of the opportunities to eliminate the diffusion of Se and Sb to CdS, reducing interface defect formation and improving the Sb₂Se₃-based SC.

To improve the performance of Sb₂Se₃ SCs, we propose using the SCAPS-1D to analyze and optimize the ITO/CdS/Sb₂Se₃/Au SC characteristics. We then fit and validate Sb₂Se₃ 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/cm², V_oc = 0.44 V, and FF = 58.28%), demonstrating that the SCAPS -1D program is perfect program for describing and developing Sb₂Se₃-based SC characteristics. Following that, we investigated the effects of Sb₂Se₃ 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-Sb₂Se₃/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/cm², and a FF of 73.47%, which may encourage the experimental laboratory to produce the same configuration.

2 Material parameters and device architecture

2.1 Symbols

Ѱ:

Electrostatic potential.

${\upvarepsilon }^{0}$:

Vacuum permittivity.

${\varepsilon }_{r}$:

Semiconductor permittivity.

n and p:

Free carrier concentrations.

${N}_{d}^{+}$:

Ionized donor density.

${\text{The }}\;{\text{ great}}$:

Acceptor density.

"ρdef":

Defect charge density.

G:

Generation rate.

${j}_{n}$:

Electron current density.

${j}_{p}$:

Hole current density.

q:

Elementary charge.

U_n:

Electrons recombination rate.

U_p:

Holes recombination rate

µ_p:

Holes mobility.

µ_n:

Electrons mobility.

HTL:

Hole Transport Layer

BSF:

Back-surface field.

E_g:

Energy band gap.

χ:

Electron affinity.

ε:

Dielectric permittivity,

NC:

Conduction band states density.

NV:

Valence band states density.

V_thn and V_thp:

E⁻ and p⁺ hole thermal velocity.

ETL:

Electron transport layer.

R_s:

Series resistance.

R_sh:

Shunt resistance

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).

$$ \frac{\partial }{{\partial x}}(\varepsilon ^{0} \varepsilon _{r} \frac{\delta }{{\delta x}}) = - q\left( {p - n + N_{d}^{ + } - N_{A}^{ - } + \frac{{\rho def}}{q}} \right) $$

(1)

$$ - \frac{{\partial j_{n} }}{\partial x} - U_{n} + G = \frac{\partial n}{{\partial t}} $$

(2)

$$ - \frac{{\partial j_{p} }}{\partial x} - U_{p} + G = \frac{\partial p}{{\partial t}} $$

(3)

$$ j_{n} = - \frac{{U_{n} n}}{q}\frac{{\partial E_{Fn} }}{\partial x} $$

(4)

$$ j_{p} = + \frac{{U_{p} p}}{q}\frac{{\partial E_{Fp} }}{\partial x} $$

(5)

In this paper, we are interested in the analysis and development of the optoelectronic performance of the Sb₂Se₃ 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 been taken into account in the simulation environment (1.5 air mass spectrums, ambient temperature of 25 °C, and 100 mW/cm² sun illuminations) were recorded from experimental measurements and other scientific papers and are shown in Tables 1, 2, 3, and 4.

Table 1 Back and front grid parameters

Table 2 Optimized parameters that were employed in the ITO/CdS/Sb₂Se₃/CuO HTL/Au hetero-junction device simulation

Table 3 The interfaces parameters utilized in this work

Table 4 The materials' defect input that was used in this simulation

3 Results and discussion

3.1 Theoretical analysis of the ITO/CdS/Sb₂Se₃/Au conventionnel solar structure

In this part of the work, we show a numerical investigation and optimization of ITO/CdS/Sb₂Se₃/Au SC through the SCAPS-1D tool. Figure 1 depicts the schema structure and energy band diagram of Sb₂Se₃ hetero-junction SC. As shown, the ITO was employed as the window thin film, CdS as a buffer film, Sb₂Se₃ as an absorber, and Au as the back electrode. It is observed from Fig. 1 that CdS/Sb₂Se₃ has a negative conduction band offset CBO⁻. This negative sign indicates that the conduction band of CdS is lesser than that of Sb₂Se₃, which may be one of the factors for implying the free flow of electrons, hence minimizing short circuit current and thus affecting SC performance.

Fig. 1

The configuration and band diagram of the Sb₂Se₃ conventional SC

• a. Validation of Sb₂Se₃ simulated parameters with experimental work

In this first subsection, the experimental parameters of the ITO/CdS/Sb₂Se₃/Au-based SC reported in the work of Xixing WEN et al., such as Sb₂Se₃ layer thickness (0.9 µm), carrier concentration (2.10¹⁶ cm⁻³), and interface defect (2.10¹¹ cm⁻²) with other parameters shown in 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/cm², and a FF of 60.4%.

At 2.10¹⁴ cm⁻³ in Sb₂Se₃ 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/cm²_, and a FF of 58.36%. This finding demonstrates the realistic models and the excellence of software used in this work.

Fig. 2

Comparing J-V features of experimental and simulated conventional solar cell structures (see Fig. 1)

• b. Impact of Sb₂Se₃ 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 Sb₂Se₃ experimental model with theoretical model one, we start optimizing the Sb₂Se₃ thickness and carrier density (Na) in the 0.2–1.2 μm and 10¹³–10¹⁸ cm⁻³ ranges, respectively. The results for quantum efficiency (a) and current density (b) versus Sb₂Se₃ 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 Sb₂Se₃ solar cells. Figure 3b also depicts the effect of the Sb₂Se₃ carrier concentration density on the J-V properties. It can be seen that J_sc increases with acceptor concentration to a maximum at 10¹⁵ 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.

Fig. 3

Sb₂Se₃ thickness and acceptor concentration effect on QE a and J–V characteristics b

Figure 4 shows the evaluation and representation of the dual effects of Sb₂Se₃ acceptor concentration and layer depth on ITO/CdS/ Sb₂Se₃ /Au S.C. characteristics (V_oc, J_sc, FF, and PCE). The figure shows how the acceptor density of the Sb₂Se₃ changed the S.C. characteristics. Since at high carrier concentrations > 10¹⁷ cm⁻³, PCE, FF, and open-circuit voltage, all exhibit good values even at low absorber thicknesses, their maximum values are only attained at thicknesses greater than 0.8 μm. Although J_sc behaves differently, the greater long-wavelength photon absorption in this layer can account for the increase in J_sc as absorber thickness increases. However, as acceptor carrier concentration increases, the 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/cm², FF of 75.33%, and V_oc of 0.58 V for thickness and carrier concentration of 1.2 μm and 10¹⁸ cm⁻³, respectively. However, in this section, we are interested in achieving a high J_sc (35 mA/cm²), 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 10¹⁵ cm³ 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.

Fig.4

Impact of Sb₂Se₃ thick and charges carrier concentration on the PV characteristics of studied heterostructure

• c. Effect of the Sb₂Se₃ bulk defect density and interfacial defect on conventional Sb₂Se₃ solar cell characteristics

The Sb₂Se₃ bulk defect density and interface defects are critical parameters for designing a high-performance CdS/Sb₂Se₃ photovoltaic cell with low parasitic resistance. The most common intrinsic defects in the Sb₂Se₃ 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 10¹⁰ to 10¹⁶ cm⁻³ (bulk defect) and from 10¹⁰ to 10¹⁶ cm⁻² (CdS/Sb₂Se₃ 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 Sb₂Se₃ 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/Sb₂Se₃/Au solar cell with 1.2 μm absorber layer thickness, 10¹⁵ cm⁻³ carrier concentration, 10¹³ cm⁻³ bulk defect density, and 10¹⁰ cm⁻² for CdS/Sb₂Se₃ interface defect density, an optimal efficiency of 16.62%, V_oc of 0.66 V, J_sc of 35.38 mA/cm², 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.

Fig. 5

Impact of Sb₂Se₃ defect on the PV characteristics of studied heterostructure

In the next part of this work, we set the Sb₂Se₃ 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/Sb₂Se₃/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 Sb₂Se₃ 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-Sb₂Se₃) 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 Sb₂Se₃ absorber due to Se and Sb interdiffusion, which is the original cause of the very deficient interface and Sb₂Se₃ absorber layer, 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 Sb₂Se₃ 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.

Fig. 6

Optimal Sb₂Se₃ solar cell structure with n- GaAs ETL and p⁺- CuO HTL layers

• e. Effect of the incorporation of n- GaAs and P + -CuO HTL interlayers on the Sb₂Se₃ solar cell characteristics

To create a dual-buffer-layered Sb₂Se₃ 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/Sb₂Se₃ interface after the addition of n-GaAs (see Fig. 6), indicating that the Sb₂Se₃ 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/Sb₂Se₃/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 photo-generated 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).

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

The effect of n-GaAs ETL and P⁺-CuO HTL interlayer thickness from 20 to 200 nm on the basic parameters of Sb₂Se₃ SCs, including PCE, V_oc, J_sc, and FF, were investigated and shown in Fig. 8.

Fig. 8

Impact of n-GaAs ETL and P⁺- CuO HTL thickness on the PV characteristics of studied heterostructure

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 Sb₂Se₃ 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/cm², V_oc of 0.73 V, and FF of 73.47%. This rise is due to a decrease in charge carrier recombination (Ait Abdelkadir et al. 2022), 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 Sb₂Se₃ 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).

• f. Effect of parasitic resistance on new hetero solar cell characteristics

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.cm²to 10 Ω.cm² 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 Sb₂Se₃ 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 Sb₂Se₃-based SCs. We can see that the outcome of this paper paves the way for higher Sb₂Se₃ SC efficiency.

Fig. 9

the effect of Rs an Rsh parasitic resistance on the new PV hetero solar cell characteristics

Table 5 Comparative study of the current results and those found in the literature

4 Conclusions

In this paper, SCAPS-1D program was used to validate a theoretical model that describes the experimental Sb₂Se₃ solar cell characteristics. We found that several parameters, including Sb₂Se₃ thin layer thickness, charge carrier concentration, and bulk and interface defects, limit the performance of Sb₂Se₃-based solar cells. The analysis of several features revealed the possibility of achieving 16.62% efficiency with 1.2 μm Sb₂Se₃ layer thickness, 10¹⁵ cm⁻³ carrier concentration, 10¹³ cm⁻³ bulk defect, and 10¹⁰ cm⁻² interface defect densities. Following this optimization study, we discovered that inserting n-GaAs (100 nm) at the n-CdS/p-Sb₂Se₃ 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 Sb₂Se₃ 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/Sb₂Se₃/Au new solar cell achieves 19.60% efficiency, J_sc of 36.38 mA/cm², V_oc of 0.73 V, and FF of 73.47%, which will be encouraging to do experimental work on Sb₂Se₃ next-generation cost-efficient thin-film PV.