Concurrent Design of Alloy Compositions of CZTSSe and CdZnS Using SCAPS Simulation: Potential Routes to Overcoming V_OC Deficit

Abstract

Solar energy is the most used renewable energy source. CZTSSe uses earth-abundant elements and has promising optoelectronic properties, resulting in becoming a viable alternative to thin film PV. This work provides design guidelines for CZTSSe-based solar cells, where CZTSSe has a tunable affinity and energy gap. The analysis is based on incorporating a ternary compound material to serve as an electron transport material (ETM). In this regard, CdZnS is a potential candidate that can be utilized as an electron transport layer whose affinity and energy gap can be tuned to adjust the band alignment at the ETL/CZTSSe interface. In order to design a high-efficiency solar cell, one has to tune both the ETL and absorber layers to have a suitable conduction band offset (CBO), thereby minimizing the non-radiative recombination which, in turn, boosts the power conversion efficiency (PCE). Thus, in our presented simulation study, we provide a codesign of alloy compositions of both the CZTSSe photoactive layer and the CdZnS ETL using SCAPS-1D simulation. It is found that using the codesign of alloy compositions of the ternary compound ETL and the absorber enhances the PCE by about 2% and, more importantly, overcomes the main issue in CZTSSe which is its open-circuit voltage (V_OC) deficit. Furthermore, upon optimizing the thickness and doping of both the ETL and absorber layer, as well as the bulk defect of the absorber layer, a PCE of 17.16% is attained in this study, while the calibrated PCE based on a previously published experimental work was 12.30%.

1. Introduction

To address the challenges posed by climate change, it is imperative to focus on the development of renewable energy sources, particularly solar photovoltaic (PV) energy [1]. PV energy has the capacity to transform an unlimited supply of sunlight into electrical energy. Many efforts have been established to achieve high power-conversion efficiency (PCE) PV cells with lower costs including silicon and other thin film materials [2,3,4,5,6,7,8]. Thin film solar cells are a type of PV cell that are made by depositing extremely thin layers of PV materials onto a substrate. These materials can be made from a variety of substances, including silicon, copper indium gallium selenide (CIGS), cadmium telluride (CdTe), and organic photovoltaics (OPVs). Thin film solar cells are a promising technology for generating clean, renewable energy. They offer several advantages over traditional silicon solar cells, including lower material costs, lighter weight, and greater flexibility. In addition, they can be produced using less energy than silicon solar cells, making them more environmentally friendly. Despite their potential benefits, thin film solar cells are still a relatively new technology and are not yet widely used on a large scale. However, ongoing research and development efforts are focused on improving the efficiency and durability of these cells, which could help make them a viable alternative to traditional solar cells in the future. Of the many thin-film materials used in solar cell fabrication that have been developed for use in low-processing cost solar cells, Cu₂ZnSn(S_xSe_1-x)₄ (CZTSSe) has drawn substantial concern from the perspective of PV technologies. CZTSSe utilizes Earth-abundant and non-toxic elements and has beneficial band gap energies (E_g) in the range 1.0–1.5 eV besides their high absorption coefficients (>10⁴ cm⁻³), in addition to its p-type conductivity nature [9,10,11].

However, compared with Si, CdTe, and CIGS, the PCE of CZTSSe still has lower records. Experimental studies report efficiencies of 12.70% with double emitters produced by solution techniques [12] and 12.30% [13] fabricated via sputtering deposition procedures. Meanwhile, the theoretically limiting efficiency is about 32.80% [14]. The low efficiency can be ascribed to the larger open circuit voltage (V_OC) deficit. This deficit can result from recombination losses at the p-n heterojunction interface, an improper choice of the conduction band offset (CBO) between the photoactive layer and the electron transport layer (ETL), un-passivated interface defects, and/or low carrier lifetimes [11,15,16,17,18]. There are a lot of both experimental and theoretical efforts that have been undertaken to push the PCE of CZTSSe by mainly reducing the V_OC deficit [19,20,21,22,23]. In this regard, Yan et al. reported a technique of utilizing two thin films that have distinctive doping levels [19]. Kim et al. presented an experimental study in which they achieved passivation of both front and back interfaces, thereby reducing the recombination, which results in enhancing the PCE [20]. By substituting Ag, which has a relatively large atomic radius, into the host CZTSSe by utilizing aqueous spray pyrolysis, an efficiency of 11.83% was achieved [21]. Further, in [22], the authors managed to control and adjust the ionic movement of Na⁺, which is incorporated to passivate trap states, and they addressed an improved performance of the fabricated solar cell. Furthermore, to improve the quality of CZTSSe-based cells, sulfurization at reduced temperatures has been investigated, and its useful impact was reflected in a higher PCE of this treatment [23].

The operation of PV cells can be better understood with a numerical analysis, which is also useful for enhancing the efficiency and structure of PV cells. The investigation of the effects of crucial physical parameters, such as the mobility and lifetime of electrons and holes, doping density, affinity, and band gap as well as defects of materials, which are time-consuming processes when being altered experimentally, is made possible by numerical modeling [24,25]. Simulation tools save the effort, time, and money of the researchers. Solar cell devices could be simulated using any numerical program that can resolve the fundamental semiconductor equations. Numerous numerical simulators, including wxAMPS, SILVACO, COMSOL, and SCAPS, have been utilized to investigate the performances of various solar cells, especially thin-film devices. Regarding the simulation efforts applied for CZTSSe-based cells, one can find some useful studies in the literature [11,26,27,28]. In [11], the influence of many key factors, such as the composition ratio, absorber thickness, and defects, as well as temperature have been investigated, and the efficiency has been enhanced where the calibrated cell was based on an experimental work while the simulation analysis was performed using SCAPS software. Further, an investigation of different possible defects has been provided in [26], revealing an optimum PCE of 19.06%. The performance of the CZTSSe solar cell under the influence of different ETLs has been numerically explored [27]. Moreover, a CZTSSe solar cell structure has been proposed and simulated using two different HTLs (Cu₂O and MoTe₂) [28].

According to previous studies, the significance of the alloy codesign as the sole design of the absorber parameters may not be sufficient to optimize the cell performance. Instead, the ETL should be simultaneously designed with the absorber to meet both a superior absorption and the band alignment at the same time. Regarding CZTSSe-based solar cells, the band alignment issue between the CZTSSe and the ETL layers has not been thoroughly investigated, especially for different compositions of the CZTSSe absorber layer. In addition, tunable electron affinity ETL candidates have not been investigated with CZTSSe. Consequently, in this work, the ETL is simultaneously codesigned with the absorber by tuning the elemental compositions of both the ETL and absorber compounds to achieve the optimum band alignment between them, thereby achieving higher PCEs.

In this work, we provide a proposed solar cell that is based on the device configuration ZnO:Al/i-ZnO/CdZnS/CZTSSe/Mo, where CdZnS serves as an ETL, while Mo serves as an HTL. First, we calibrate this modeling methodology implemented in the simulator versus the experimental structure whose ETL is CdS [29]. Then, the alloy composition codesign concept is applied to improve the efficiency of the cell. In the presented methodology, the electron affinity and energy bandgap of the Cu₂ZnSn(S_xSe_1-x)₄ absorber can be engineered by controlling the relative composition (x) of S (and thus Se). On the other hand, the electron affinity and the bandgap of the compound ETL (Cd_yZn_1-yS) are controlled via tuning the composition (y). After obtaining the optimum x and y compositions that achieve the maximum PCE, some steps are conducted to optimize the main device parameters. The impacts of the ETL’s thickness and doping are analyzed to assess their influence on cell performance, followed by an optimization of the ETL’s thickness and doping concentration. The thickness and doping concentration of the absorber layer are then optimized. In addition, the bulk defect density inside the photoactive absorber is investigated. The rest of this paper is organized as follows: Section 2 presents the used materials, methods, and simulation techniques; Section 3 discusses the results; finally, Section 4 concludes the work of this paper.

2. Materials and Methods

2.1. Device Structure and Material Parameters

The arrangement of the investigated CZTSSe solar cell is comprised of a 600 nm thick Mo layer as a rear contact, a 1.8 μm thick CZTSSe thin film, a 50 nm thick CdS buffer ETL, a 50 nm thick intrinsic ZnO layer, a 300 nm thick Al doped ZnO (AZO) film that serves as a front contact. Figure 1 illustrates the structure of the proposed solar cell with the energy diagram of the used materials. A comprehensive list of the physical factors of all materials applied in SCAPS-1D simulation is presented in

Table 1. Material parameters used in SCAPS-1D [29].

Parameter	Material						Unit
	AZO	ZnO(i)	CdS	MoSSe	CZTSSe	CdZnS
Thickness	300	50	50	350	1800	50	nm
Electron affinity	4.40	4.40	4.2	4.14	Var	4.3–3.18	eV
Bandgap	3.3	3.3	2.4	1.1	Var	2.42–3.54	eV
dielectric permittivity	9.000	9.000	10.000	13.6	10	9	relative
CB effective density of states	1 × 1018	1 × 1018	1 × 1018	2.2 × 1018	2.2 × 1018	2.2 × 1018	cm−3
VB effective density of states	1 × 1019	1 × 1019	1 × 1019	1.8 × 1019	1.8 × 1019	1.8 × 1019	cm−3
electron mobility	1 × 102						cm2/Vs
hole mobility	25						cm2/Vs
shallow uniform donor density ND	1 × 1018	1 × 1017	1 × 1018	0	0	1.5 × 1018	cm−3
shallow uniform acceptor density NA	0	0	0	1 × 1016	9.16 × 1016	0	cm−3

. In addition, the main defect parameters of the materials are recorded in

Table 2. Defect parameters used in SCAPS-1D [29].

Parameter	AZO	ZnO (i)	CdS	MoSSe	CZTSSe	Unit
defect type	Donor(0/+)	Donor(0/+)	Acceptor(−/0)	Donor(0/+)	(A)/(D)
capture cross section electrons	5.0 × 10−13	5.0 × 10−15	1.0 × 10−13	1.0 × 10−13	1 × 10−13/1 × 10−14	cm2
capture cross-section holes	1.0 × 10−15	1.0 × 10−13	1.0 × 10−15	1.0 × 10−15	1 × 10−12/9 × 10−17	cm2
energetic distribution	Single
reference for defect energy level Et	Above Ev
energy level with respect to Reference	1.650	1.650	1.200	0.8	0.185, 0.85	eV
Deep Defect Density	1.8 × 1016	5.0 × 1014	6.0 × 1017	4.0 × 1014	2.27 × 1016, 9.7 × 1016	cm−3

. In this analysis, the energy band gap and the electron affinity of both the absorber and electron transport layers are simultaneously tuned. To obtain a wide range of tunable energy parameters, the ternary compound Cd_yZn_1-yS is selected as an ETL. The variation of E_g and χ (in eV) vs. x composition is displayed in Figure 2a for the absorber, while Figure 2b demonstrates the variation of E_g and χ (in eV) vs. y composition in the ETL (Cd_yZn_1-yS) [29]. As can be depicted in the figures, the range of E_g is 2.42 to 3.54 eV and from 1 to 1.5 eV for the ETL and the absorber layer, respectively. Additionally, the range of the electron affinity is 4.3–3.18 eV for the absorber, while it is 4.6–4 eV for the ETL.

2.2. Numerical Simulation Methodology and MODELS

SCAPS-1D is a windows-based, one-dimensional solar cell modeling program. It outperforms other solar device simulator programs, and its simulation findings are promising. Herein, SCAPS-1D was used to numerically simulate device J-V characteristics to assess the proposed cell design. The structured solar cell is subjected to DC and AC analysis. Poisson’s equation, holes’ and electrons’ continuity equations, and the carrier transport equations serve as the foundation of SCAPS-1D. The involved semiconductor equations are formulated as follows:

$$\rho =q\left(p-n-N_A^{-}+N_D^{+}\right)$$

(1)

$$\left(\frac{1}{q}\right)\nabla .J_n-R_n+G_n=\frac{\partial n}{\partial t}$$

(2)

$$-\left(\frac{1}{q}\right)\nabla .J_p-R_p+G_p=\frac{\partial p}{\partial t}$$

(3)

$$J_n=-qn{\mu }_n\nabla E_{Fn}$$

(4)

$$J_p=-qp{\mu }_p\nabla E_{Fp}$$

(5)

The equations contain several variables and parameters, including J_n and J_p which represent the electron and hole current densities, R and G which denote the recombination and generation rates, n and p which indicate the carrier concentrations, $\rho$ which represent the charge density, and $N_A^{-}$ and $N_D^{+}$ which represent the densities of the acceptors and donors. Finally, E_Fn and E_Fp are the quasi-Fermi levels for electrons and holes.

To investigate and evaluate the performance of a solar cell in the SCAPS-1D software, the following processes were used. First, a validation step of simulated results is performed by calibrating the experimental cell against simulation results. For the given solar cell, the software is executed to find the JV characteristics whether under illumination or dark conditions. Based on the output illuminated current density, the extraction of many metrics can be conducted. These include V_OC, short circuit current density (J_SC), fill factor (FF), external quantum efficiency (EQE), generation and recombination rates, energy band diagrams, etc. These are some examples of the electrical measurements for AC and DC that can be simulated on SCAPS. All of these metrics are calculable and attainable under both dark and light circumstances, with various temperatures and illumination levels [30]. Additionally, SCAPS-1D offers benefits such as intuitive control, the capacity to model up to seven separate films with non-routine measurements (C-V, C-f), and tandems [31].

3. Results and Discussion

In this section, we present all obtained simulation results and a discussion of each finding. Numerical simulation allows several studies to receive an optimized structure of the solar cell. The section is divided into different subsections. Firstly, there are the calibration of this simulation work and the published experimental work to make sure the simulation results meet the practical applications. The obtained current density, voltage curve and main performance parameters, short circuit current density, open circuit voltage, fill factor, and power conversion efficiency, are compared with the published experimental work. Secondly, the codesign concept is applied to the calibrated cell to enhance its output performance parameters. Moreover, the optimization of the basic parameters of the calibrated cell to enhance its performance is introduced. The optimization starts with optimizing the thickness and doping of the electron transport layer. After that, the thickness, the doping, and the bulk defect density of the absorber layer are optimized to enhance the output performance parameters.

3.1. Calibration versus Experimental Work

Measuring calibrating simulation results against experimental data is an essential step in ensuring the accuracy and reliability of computer models. This process involves adjusting the parameters of the simulation model to match the results of real-world experiments, with the aim of minimizing the discrepancy between the two. Notably, experimental data provide a benchmark against which the accuracy of simulation models can be assessed. By comparing the results of the simulation model with experimental data, researchers can identify discrepancies and refine the model to improve its accuracy. In addition, calibration helps to ensure that simulation models are not based on incorrect assumptions or oversimplifications. Thus, this work started by calibrating the simulated solar cell with a published experimental work [29]. As mentioned before, SCAPS is used to analyze the performance metrics: V_OC, J_SC, FF, and PCE. The JV curve is the most significant output metric to measure solar cell performance. As a sequence, the JV curves of the simulated cell and the reported experimental work [29] are illustrated in Figure 3. It can be deduced from Figure 3 that the simulated device using SCAPS-1D agrees with the experiment [29]. By neglecting the series and shunt resistances of the cell in the simulation, the performance metrics of the two cells are given in

Table 3. The output performance factors of the simulation vs. the experimental results [29].

	Voc (V)	Jsc (mA/cm2)	FF (%)	PCE (%)
Measurement	0.521	34.98	67.20	12.30
Simulation	0.524	35.25	67.11	12.41

for the purpose of comparison with the same thickness and doping as in the experimental [29].

The performance factors obtained from the experiment and simulation results in Table 3 show a slightly higher J_SC than the experimental results, which is most probably due to a better photo-absorption in the absorber layer. It is worth mentioning that the shading loss in the front metal contact on the top is ignored due to the one-dimensional geometry considered, which will influence it by reducing J_SC [32]. A very small difference between V_OC measures can be noticed, which might result from the multiple reflections from the interfaces, mechanisms which are not included in SCAPS [33]. The fill factor and the power conversion efficiency have a subtle difference of less than 0.1%. The output performance parameters, V_OC, J_SC, FF, and PCE, besides the JV curves of the experimental work [29], and the simulated solar cell show the validity of the used materials parameters and design.

3.2. Codesign and Optimization

Theoretical studies give physical insight with a thorough physical explanation of the output performance. The codesign problem could be solved utilizing practical combinatorial methods, such as with an orthogonal composition gradient in the utilized layers [34]. CuSbCh₂ [35], Cu₂ZnSnCh₄ [36], CIGS [37], and CdTe [38] absorbers or CdS [39], Cu_xZn_1-xS [40], Zn–Ni–Co–O [41], and Mg_xZn_1-xO transport layers are being compositionally graded and investigated using combinatorial methods [42].

In this part of the design, the composition x of the Cu₂ZnSn(S_xSe_1-x)₄ absorber is tuned to adjust the electron affinity and the energy bandgap, and composition y of the Cd_yZn_1-yS ETL is concurrently adjusted to achieve the optimum efficiency. Figure 4a illustrates the PCE variation with the composition x (of the absorber) and composition y (of the ETL). As illustrated in the figure, the optimum PCEs are achieved in two points: point 1 (x = 0.8, y = 0.52) and point 2 (x = 0.7, y = 0.14), with a PCE exceeding 14%. The illuminated JV characteristics of the initial design and the two optimum case cells are illustrated in Figure 4b. The simulation results reveal a boost in the V_OC of the codesign compared to the initial cell. This necessitates the usefulness of the codesign technique in alleviating the V_OC deficit encountered in this type of thin-film solar cell. The performance metrics (V_OC, J_SC, FF, and PCE) of the initial design and the two optimum cases are listed in

Table 4. The output performance parameters of the initial design, and the two optimum cases.

	Voc (V)	Jsc (mA/cm2)	FF (%)	PCE (%)
Initial design	0.524	35.25	67.11	12.41
Case 1 Cd0.86Zn0.14S (x = 0.7)	0.738	26.98	70.64	14.06
Case 2 Cd0.48Zn0.52S (x = 0.8)	0.789	25.01	71.95	14.21

for comparison.

To give a physical insight into the previous results, the issue of band alignment, which is responsible for such results, is highlighted. Notably, reducing the charge recombination at the interface between ETL and the absorber is very important to enhance carrier extraction. To accomplish this, the CBO at the ETL/absorber interface should be 0–0.3 eV [43], where the CBO is given with Equation (6),

$$\mathrm{C}\mathrm{B}\mathrm{O}=({\chi }_{\mathrm{a}\mathrm{b}\mathrm{s}}-{\chi }_{\mathrm{E}\mathrm{T}\mathrm{L}})$$

(6)

where ${\chi }_{\mathrm{a}\mathrm{b}\mathrm{s}}$ and ${\chi }_{\mathrm{E}\mathrm{T}\mathrm{L}}$ are the affinities of the photoactive layer and ETL, respectively. Figure 5 illustrates the CBO dependence on different x and y values. The affinities of the ETL and the absorber with their corresponding energy gaps are listed in

Table 5. Main band parameters of the initial design and the two optimum cases.

	${\mathit{\chi }}_{\mathit{E}\mathit{T}\mathit{L}}$	Eg (ETL)	${\mathit{\chi }}_{\mathit{a}\mathit{b}\mathit{s}}$	Eg (Absorber)	CBO (eV)
Initial design	4.300	2.420	4.462	1.097	0.162
Case 1 Cd0.86Zn0.14S (x = 0.7)	4.253	2.467	4.180	1.329	−0.073
Case 2 Cd0.48Zn0.52S (x = 0.8)	3.945	2.78	4.120	1.384	0.175

for the three scenarios.

Based on the theoretical calculations of Shockley and Queisser who reported that the maximum PCE can be achieved for an energy gap of 1.34 eV [44], it is expected that a higher PCE is achieved (with respect to the calibrated cell) when the energy gap ≅ 1.33 eV (which corresponds to x = 0.7). However, the highest PCE is obtained for a slightly higher bandgap of about 1.38 eV (which corresponds to x = 0.8), as indicated in Table 4. This can be attributed to the crucial role of the CBO as illustrated in Figure 6, which shows the energy band diagrams of these two cases compared to the initial structure. Although the initial design reveals a favorable spike-like band offset as Figure 6a signifies, the low bandgap results in a lower PCE. For the case of E_g = 1.33, a CBO of −0.073 eV is obtained (Figure 6b and Table 5), which corresponds to an unfavorable cliff-like band. Conversely, a CBO of 0.175 eV is observed for E_g = 1.38 eV (see Figure 6c and Table 5), signifying a favorable low spike-like band offset. These interesting results show the importance of the alloy codesign as the sole design of the absorber parameters may not be sufficient to optimize the cell performance. Instead, the ETL should be simultaneously designed with the absorber to meet both superior absorption and band alignment at the same time.

Furthermore, the optimization after the alloy composition codesign of the ETL and absorber compositions started with optimizing the thickness and doping of the ETL of the two optimum cases as illustrated in Figure 7. Figure 7a demonstrates that the performance improves at a certain thickness for both cases of composition. Since most of the photons with short wavelength are absorbed near the surface, it is desirable to have a thinner ETL. As can be deduced from Figure 7a, the optimum thickness is around 0.02 µm with PCEs of ≅14.13%, and 14.3% for cases 1 and 2, respectively. Additionally, one can observe in Figure 7b that varying the doping concentration from 10¹⁸ cm⁻³ to 10²⁰ cm⁻³ results in an increase in the PCE. The highest efficiency is achieved approximately in the range 5 × 10¹⁹ cm⁻³ to 10²⁰ cm⁻³ due to an increase in conductivity. The PCE increases to about 14.3% for the two cases, as shown with a slightly higher PCE for x = 0.7.

The influence of the thickness, the doping concentration, and the defect density of the absorber material on the PCE has been investigated as illustrated in Figure 8. The thickness of the photoactive layer is a critical factor since it is responsible for absorbing the incident photons. However, it should not be excessively thick to allow for an effective collection of the photogenerated carriers before recombination. In the simulation, the thickness of the absorber layer varied from 0.3 μm to 2 μm, as depicted in Figure 8a. As can be inferred from the figure, the PCE does not show a significant effect from increasing absorber thickness beyond 1 µm. Further, decreasing the doping of the absorber material makes a gradient doping at the interface between the absorber and the ETL, which raises the diffusion current and enhances the PCE, as can be inferred from Figure 8b. Additionally, the higher doping concentration leads to an improved built-in electric field, facilitating better charge-carrier separation and ultimately enhancing the overall performance of the cell. However, it is important to note that further increasing the doping concentration can result in a higher recombination rate. This phenomenon is not conducive to an increase in the open-circuit voltage and, consequently, the PCE. Therefore, there is a delicate balance to be maintained, ensuring optimal doping concentrations for achieving the desired cell performance. As can be deduced from Figure 8b, the optimum doping level for the absorber layer in both scenarios is $\cong 2.8\times {10}^{15}$ cm⁻³.

Moreover, the bulk defects of the absorber material have a great influence on the PCE. As can be inferred from Figure 8c, there is no significant improvement in the PCE below the defect density of 10¹⁴ cm⁻³. The optimum factors of the previous optimization are listed in

Table 6. The optimum technological factors of the ETL and the absorber layer.

ETL Thickness (µm)	ETL Doping (cm−3)	Absorber Thickness (µm)	Absorber Doping (cm−3)	Absorber Defects (cm−3)
0.02	3.8 × 1019	2.420	2.8 × 1015	1014

. Combining these optimization steps yields a high V_OC of 0.8 V, a J_SC of 28.97 mA/cm², a FF of 73.72%, and a PCE of 17.16%. The JV and EQE curves of the initial and optimum cells are illustrated in Figure 9 for comparison.

To compare this work with some recent publications in the literature, a group of similar structures has been collected and illustrated in

Table 7. Comparison of different structures based on CZTSSe photovoltaic devices utilizing CdZnS as buffer layers.

Structure	Work	Year	PCE (%)
i-ZnO/ITO/CdZnS/CZTS	[45]	2022	8.50
n-ITO/ZnO/CdZnS/CZTS	[46]	2019	11.20
AZO/ITO/CdZnS/CZTS	[47]	2022	13.23
ITO/i-ZnO/CdZnS/CZTS	[48]	2021	4.88
AZO/i-ZnO/CdS/CZTSSe	[29]	2020	12.30
Case 1 Cd0.86Zn0.14S (x = 0.7)	This work	2023	14.06
Case 2 Cd0.48Zn0.52S (x = 0.8)	This work	2023	14.21

. In [45], the PCE of the CZTSSe solar cell device has been notably increased from 7.4% to 8.5% by adding Zn to the CdS buffer layer. In [46], a study aimed to investigate the effect of three different buffer layers, namely CdS, ZnS, and CdZnS, on the performance parameters of CZTS solar cells through numerical simulations. The simulation results showed that a CdZnS buffer layer achieves high efficiency CZTS solar cells. PCEs utilizing these buffer layers were around 11.20%. In [47], the utilization of a CdZnS buffer layer has resulted in a 13.23% conversion efficiency in CZTS solar cells, which is an improvement over the conventional buffer layer, CdS, that only yielded 12.61%. In [48], compared to the CZTS/CdS device, the CZTS/CdZnS device exhibited an increase in both the conversion efficiency and fill factor. Specifically, the PCE increased from 2.31% to 4.88%.

Finally, the co-design of the elemental compositions of the electron transport layer (ETL) and the absorber represents a new field, and its widespread application across various materials has not been widely explored. One of these few efforts was devoted to the codesign of alloy compositions of CdSe_xTe_{1_x} (as an absorber material) and Mg_yZn_1-yO (as an ETL) [49]. In their experimental study, the authors revealed that the solar cell performance is a strongly coupled function of both x and y compositions. They achieved a PCE of up to 17.7% at specific atomic compositions (x = 0.04 and y = 0.18) [49]. As this field of research is advancing, a need for simulation studies is mandatory to investigate and evaluate new material systems and to provide design guidelines for the corresponding solar cells. This study serves as a significant contribution by shedding light on this emerging field, paving the way for further advancements.

4. Conclusions

Photovoltaics are the most effective solution to the energy demand issue. Utilizing new materials and designs are required to meet the solution to this issue. CZTSSe-based solar cells have a promising future according to their unique set of optical and electrical parameters, in addition to their abundance. CdZnS is a good candidate for traditional ETMs. Using the alloy composition codesign concept improves solar cell performance. In this context, a Cd₀.₄₈Zn₀.₅₂S with an absorber composition of x = 0.8 is found to be the optimum criterion in the proposed cell with a PCE of 14.21%. This choice integrates both a suitable bandgap of 1.38 eV with a favorable low spike band offset. Further, optimizing the ETL thickness and doping yields a PCE exceeding 14.3%. After optimizing the thickness and concentration of the absorber film, a PCE of 15.65% was achieved. The bulk defects in the absorber layer have a substantial influence on the PCE. Decreasing the defect density resulted in increasing the PCE. In the investigated cell, no significant PCE improvement below N_t of 10¹⁴ cm⁻³ has been attained, and a PCE of 17.16% was achieved at this value. This simulation-based study offered the advantage of gaining insight into the optimization of the cell performance of CZTSSe-based devices by enabling a comprehensive exploration of the alloy composition codesign of both the absorber and the ETL to surmount the V_OC deficit of this kind of solar cell which is considered the main issue in developing CZTSSe-based solar cells.