Performance evaluation of ZnSnN2 solar cells with Si back surface field using SCAPS-1D: A theoretical study

The earth-abundant semiconductor zinc tin nitride (ZnSnN2) has garnered significant attention as a prospective material in photovoltaic and lighting applications, primarily due to its tunable narrow bandgap and high absorption coefficient. This study focuses on a numerical investigation of ZnSnN2 solar cell structures using the SCAPS 1-D software. The objective is to analyze the influence of various physical and geometrical parameters on solar cell performance. These parameters include the thicknesses of the ZnO window layer, CdS buffer layer, ZnSnN2 absorber layer, and Si back surface field layer (BSF), as well as operating temperature, series and shunt resistances (RS and Rsh), absorber layer defect density, interface defects, and the generation-recombination profile of the n-ZnO:Al/n-CdS/p-ZnSnN2/p-Si/Mo structure. We have evaluated the capabilities of this novel material absorber by investigating its performance across a range of thicknesses. We have started with ultrathin absorber thicknesses and gradually increased them to thicker levels to determine the optimal thickness for achieving high efficiency. Under optimal conditions, a thin solar cell with a thickness (wp) of 1 μm achieved an efficiency (η) of 23.9%. In a practical solar cell operating at room temperature, optimal parameters were observed with a thicker absorber layer (wp = 8 μm) and a BSF width of 0.3 μm. The cell exhibited resistances of Rsh = 106 Ω cm2 and Rs = 1 Ω cm2, along with a low defect density (Nt = 1010 cm−3) in the ZnSnN2 semiconductor. These factors combined to yield an impressive efficiency of 29.5%. Numerous studies on emerging ternary nitride semiconductors (Zn-IV-N2) have highlighted ZnSnN2 as a promising material for thin-film photovoltaics. This compound is appealing due to its abundance, non-toxicity, and cost-effectiveness. Unlike conventional solar cells that depend on rare, toxic, and costly elements, these components are still essential for today's solar cell technology.


Introduction
In recent decades, researchers have highlighted the importance of numerical and computing research through advanced programs that model and simulate devices, enabling the exploration of material properties.It yields financial benefits and stimulates investment in scientific research, particularly for predicting manufacturing and experimental outcomes [1].Among the ternary Zn-IV-N 2 semiconductors, ZnSnN 2 consists of non-toxic, earth-abundant, and cost-effective elements [2].Regarding symmetry, Zn-IV-N 2 semiconductors resemble the wurtzite structure of III-N semiconductors [3][4][5].They exhibit comparable electronic, optical, and polarization properties, including direct band gaps, high optical absorption coefficients, and spontaneous polarization [6][7][8].Recent studies have highlighted In x Ga 1-x N and ZnSnN 2 as promising photovoltaic absorbers within the nitride compounds of III-N and II-IV-N 2 materials.However, the growth of crystalline III-V materials requires expensive and complex epitaxial growth techniques.For example, using indium, gallium, and arsenic in many solar cells is undesirable due to certain elements' scarcity, cost, and toxicity [9][10][11][12][13]. p-SnO/n-ZnSnN 2 and p-Si/n-ZnSnN 2 junctions were recently successfully fabricated for photovoltaic applications [14,15].A buffer layer of Al 2 O 3 was employed in the p-SnO/n-ZnSnN 2 solar cell to enhance its efficiency, which was boosted to 1.54% [15].Furthermore, a PV device based on ZnSnN 2 was also theoretically analyzed, exhibiting a ~23% efficiency without defects [16][17][18].Several forms of ZnSnN 2 films have been synthesized, including polycrystalline and monocrystalline films, on monocrystalline substrates such as (sapphire "Al 2 O 3 ", GaN, Yttria-stabilized Zircona "YSZ", etc.) by radio frequency (RF) sputter deposition [19], Molecular beam epitaxy (MBE) [20], and plasma-assisted vapor-liquid-solid technique (VLS) [21].According to experimental findings, it has been determined that ZnSnN 2 possesses a direct bandgap of approximately 1.7 eV [22].Moreover, theoretical investigations show that the direct bandgap of ZnSnN 2 can vary from 1 to 2 eV depending on the degree of disorder in which it crystallizes [11,23].Experimental results have demonstrated that ZnSnN 2 behaves as an n-type degenerate semiconductor with a high concentration of electrons but low mobility [24].While the issue of p-type doping has been a challenge, we have considered it and utilized ZnSnN 2 as a p-type semiconductor [4,25,26].This study investigated various physical and geometrical parameters affecting solar cells, such as the thicknesses of the ZnO window layer, CdS buffer layer, ZnSnN 2 absorber layer, and Si back surface field layer (BSF), as well as A. Laidouci et al. operating temperature, series and shunt resistances, absorber layer defect density, interface defects, and the generation-recombination profile and their effect on the electrical parameters.Fig. 1(b) shows the proposed design of the thin ZnSnN 2 solar cells.The ZnSnN 2 films used in this study have orthorhombic lattice parameters of a = 0.585 nm, b = 0.676 nm, and c = 0.558 nm [21].To minimize strain, CdS and Si were selected with a low strain of 0.37% and 7.17%, respectively, for ZnSnN 2 .This new structure has been proposed by H. Heriche et al. (n-ZnO:Al/n-CdS/p-CIGS/p-Si/Mo) [27].Additionally, silicon has been utilized as a secondary absorber and a Back Surface Field (BSF) to improve efficiency.On the other hand, an ohmic back contact was created by depositing molybdenum (Mo) on a coated glass substrate.

Materials and methods
The energy band diagram for the new thin ZnSnN 2 SCs has been obtained using SCAPS-1D software, as depicted in Fig. 2 (a).Fig. 1 (b) shows the proposed design of the thin ZnSnN 2 photovoltaic cell.This section outlines the physical models utilized in the study, where SCAPS-1D employs a numerical method called Finite Difference Method (FDM) to solve semiconductor equations, Poisson's equation, and continuity equations, which are represented by equations ( 1)-(3) [28].For carriers, the Poisson equation is as follows: where Ψ is the electrostatic potential, n is the electron concentration, p is the hole concentration, ε r is the relative permittivity, e is the electrical charge, N d is the donor concentration, ε 0 is the vacuum permittivity, N a is the acceptor concentration, ρ n, and ρ p are electrons and holes distribution.The continuity equations for carriers are: where G(x) is the charge generation rate, and R(x) is the charge recombination rate.
According to transport theory, the drift-diffusion current can be mathematically expressed by the following equations ( 4) and ( 5): J p = D p dp dx where μ n and μ p are electron and hole mobilities, J n and J p are electron and hole current densities, D n and D p are electron and hole diffusion coefficients, respectively.The EQE (External Quantum Efficiency) is determined by integrating the following expression (6) [29]: where J ph (λ) is the total photocurrent density, and F(λ) is the photon flux of the incident spectrum.
In the single-diode model, the current is typically expressed by the following equation ( 7) [30]: where I L is the load current, R s is the series resistance, R sh is the shunt resistance, I 0 is the reverse saturation current, and I is the output current.It is also possible to represent I 0 as follows in equation ( 8) [30,31]: where A is the p-n junction area, L n and L p are electron and hole diffusion lengths, and n i intrinsic carrier concentration, respectively.The V oc is mathematically represented by the following equation ( 9) [30]: where T is temperature, and K is Boltzmann's constant, respectively.The empirical fill factor can be accurately expressed by equation (10) as follows [31,32]: where v oc is normalized V oc (v oc = V oc /V t ), and V t is the thermal voltage (V t = kT/q).Varshni's law is an empirical equation that quantifies the relationship between temperature and bandgap energy in semiconductors.It is given by equation (11) as follows [17]: where α, β, and Eg (0 K) are material constants.
Hence, the valence band alignment is as follows: where E g is the bandgap energy and χ is the electron affinity.
The built-in potential (V bi ) of an n-p junction device, considering variations in doping density, is determined by the following equation ( 14) [45]: In solar cells, the V oc is typically lower than V bi because it is affected by factors such as recombination losses and series resistance [45].The value of the built-in potential in an n-p junction can also be obtained from the intercept of the Mott-Schottky plot, which is determined using the following equation (15) [45]: where C represents the space charge capacitance per unit area, and N represents the carrier concentration.

Results and discussion
Throughout this study, in all simulations, we used ambient temperature, one sun AM1.5G, and considered the flat-band conditions on the front contact.Fig. 1 (b) depicts the thin ZnSnN 2 /Si solar cell's structure proposed by H. Heriche et al. [27].This innovative design replaces the conventional copper indium gallium diselenide (CIGS) material with ZnSnN 2 , an earth-abundant semiconductor.This substitution's main objective is to avoid using toxic and expensive elements, namely indium, selenium, and gallium.This substitution addresses the environmental concerns and cost implications associated with these materials.
Moreover, the objective of this study extends beyond the implementation of ultrathin solar cells; it also delves into the exploration of the potential benefits that a thick absorber layer can offer in achieving high efficiency.As depicted in Fig. 2 (a), the magnitude of the band offset in a buffer-absorber and absorber-BSF layers system is evident.It is crucial to note that the difference in electron affinity between the respective layers intricately determines this magnitude.The band offset refers to the energy difference between adjacent materials' valence and conduction bands in a heterojunction.Moreover, the statement explains that the variation in electron affinity between the layers leads to the emergence of distinct structural characteristics within the layer system.Specifically, positive band offsets give rise to spike-like formations, while negative band offsets result in cliff-like configurations.These descriptions effectively visually represent the band offset's impact on the layer system's structure [43,44].Developing a cliff-like band structure is a remarkable feature of the ZnSnN 2 /CdS interface.The conduction band offset (CBO) of − 0.1 eV suggests that the conduction band energy levels of ZnSnN 2 are slightly lower than those of CdS at the interface.This CBO facilitates electron transfer from CdS to ZnSnN 2 , crucial for efficient charge separation in various optoelectronic devices.
Additionally, the valence band offset (VBO) of − 1 eV indicates an upward shift in energy levels for the valence band of ZnSnN 2 compared to CdS.This VBO promotes hole transfer from ZnSnN 2 to CdS, enhancing charge carrier separation.When both band offsets are negative, it increases carrier recombination, which in turn causes a decrease in V oc [44].These observations provide valuable  insights into the band alignment and electronic structure of the ZnSnN 2 /CdS interface, which can be exploited to design and optimize novel semiconductor heterostructures.At the Si/ZnSnN 2 interface, an intriguing cliff-like band structure is observed.The CBO of − 0.42 eV indicates a high downward shift in the energy levels of Si compared to ZnSnN 2 .This CBO facilitates the transfer of electrons from ZnSnN 2 to Si, enabling efficient electron injection or collection in devices such as photovoltaic cells.
On the other hand, positive band offsets may restrict electron transport, leading to increased electron trapping due to potential barriers in the device, which lowers the cell's J sc and PCE (%) [44].The valence band offset (VBO) of − 0.43 eV reveals an upward shift in the valence band energy levels of ZnSnN 2 relative to Si.This VBO promotes hole transfer from Si to ZnSnN 2 , facilitating the separation and transport of positive charge carriers.The C-V characteristics, Mott-Schottky plot, and the obtained results are depicted in Fig. 2 (b,c).Based on our comprehensive analysis, the Mott-Schottky plot reveals an extracted built-in potential value of approximately 1.3 V.The frequency used in the simulation is 1 MHz.As illustrated in Fig. 2 (b,c), the capacitance value remained constant when the voltage ranged from 0 to 1 V.However, the capacitance underwent a significant and rapid change beyond this threshold.

Impact of window layer and buffer layer thicknesses on photovoltaic performance
The window and buffer layers play a significant role in solar cell performance.To improve performance, ZnO window layer and CdS buffer layer thicknesses have been varied from 20 to 200 nm and 10-100 nm, respectively.Fig. 3 illustrates the impact of window layer and buffer layer thicknesses on the electrical parameters of the ZnSnN 2 -based solar cell at ambient temperature.It can be observed that the variation in the thicknesses of the ZnO window layer and CdS buffer layer affects the J SC (Fig. 3 A. Laidouci et al. and PCE (Fig. 3(d)) of the solar cell.In the case of the J SC (Fig. 3(a)), the results showed a maximum value of over 21.70 mA/cm 2 when the window thickness was ≤60 nm and the buffer thickness was ≥50 nm.For the V oc (Fig. 3(b)), the results indicated a maximum value of over 1.238 V when the window thickness was ≤80 nm and the buffer thickness was ≥60 nm.Regarding the FF (Fig. 3(c)), the results demonstrated a maximum value of over 89.86% when the buffer thickness was ≤60 nm, regardless of the window thickness.Our findings suggest that a maximum efficiency of 23.9% can be achieved with a ZnO window layer thickness of approximately 50 nm and a CdS buffer layer thickness of around 50 nm.It is important to note that an excessively thin layer can lead to leakage current, while an overly thick layer may result in a low carrier separation rate [46].Thus, window and buffer layers can impact photovoltaic cell efficiency, stability, and durability, and optimizing them can lead to improved photovoltaic performance.

Influence of absorber layer thickness on photovoltaic characteristics
Fig. 4 displays the ultrathin solar cell structure's quantum efficiency (QE) (ZnSnN 2 /Si) with varying ZnSnN 2 layer thicknesses ranging from 100 nm to 1 μm.The simulation results demonstrate a significant improvement in the quantum efficiency (QE) with increased absorber thickness of ZnSnN 2 .This enhancement is observed across a wavelength range of 300 nm-826 nm, where 826 nm corresponds to the energy bandgap of ZnSnN 2 .However, no light will be absorbed beyond this wavelength due to insufficient photon energy.The improved quantum efficiency (QE) observed can be attributed to the enhanced absorption of photons across a wide range of wavelengths, specifically from 300 nm to 826 nm.This heightened absorption leads to a greater rate of carrier generation, increasing QE.The phenomenon can be explained by the increasing thickness of the absorber layer, allowing for the absorption of a larger number of photons and consequently generating a greater number of electron-hole pairs.This ultimately enhances the overall quantum efficiency [42].In addition, the insertion of silicon as the Back Surface Field layer (BSF) significantly enhances the solar cell's performance by creating an electric field on the rear face.This electric field serves to lower the surface recombination velocity (SRV), which refers to the rate at which minority carriers (electrons or holes) recombine with opposite charge carriers at the surface of the solar cell.By incorporating silicon as the BSF layer, a potential barrier impedes the migration of minority carriers toward the surface.This barrier reduces the chances of recombination occurring at the surface, thereby minimizing the loss of carriers and improving the overall electrical characteristics of the solar cell.The reduction in surface recombination improves the efficiency of the solar cell by increasing the lifetime of the minority carriers and maximizing their chances of reaching the p-n junction, where they can contribute to the generation of electric current.The silicon BSF layer enhances the solar cell's overall performance and power output by minimizing the recombination losses at the surface [47].Fig. 5 (a and b), at ambient temperature, w ZnO = 50 nm, w CdS = 50 nm, and without defect in p-type ZnSnN 2 , we show the variation of current-voltage characteristics J(V) of the studied ZnSnN 2 solar cell and the output power for different thickness w p (ZnSnN 2 ) of our proposed structure.A thinner absorbing layer causes the lower photocurrent, and a too-thick absorber layer increases series resistance, material consumption, and, therefore, the price per unit of generated power.According to our results, increasing absorber thickness gradually leads to a growing shape of the curves.
The results show that the variation of w p from 0.1 μm to 1 μm leads to increased improvement in the solar cell's performance.The enhancement of J sc and the efficiency is because of the absorbed photons that contribute to the generation of carriers.According to the results shown in Table 4, Fig. 5 (a and b), and Fig. 6 (a and b), a significant improvement has been noted in the yield from 9.96% for 0.1 μm to 23.9% for 1 μm (Maximum power and efficiency are identical under AM1.5G spectrum conditions).An increment of J sc from 09.77 mA/cm 2 for 0.1 μm to 21.61 mA/cm 2 for 1 μm and V oc from 1.1569 V for 0.1 μm to 1.2379 V for 1 μm.However, the increase in FF is not significant.It is also possible that the additional thickness of the absorber layer would result in a higher V oc .Thus, increasing the number of charge carriers in a solar cell can cause an increase in the built-in potential across its p-n junction.The built-in potential is the potential difference across the junction without any external bias or sunlight and is influenced by the doping levels and energy bandgap of the materials used in the cell.A higher built-in potential results in a higher open circuit voltage.Hence, as the absorber    thickness increases, more electron-hole pairs are generated, leading to a higher concentration of charge carriers.This results in an increase in the built-in potential and subsequently raises the open circuit voltage of the solar cell [48].

Impact of operating temperature on photovoltaic performance
Fig. 7 shows the current density-voltage characteristic J-V for different operating temperatures.Temperature is a key factor that affects the performance of semiconductor devices, especially solar cells.For example, as the temperature increases, the semiconductor lattice expands due to thermal expansion, changing the crystal structure.Moreover, the electron-phonon interaction becomes more significant at high temperatures.These effects reduce the band gap of the semiconductor.As a result, the charge carrier concentration increases, which influences the short-circuit current (J sc ) [49].In a photovoltaic cell, the reverse saturation current is highly dependent on temperature, which makes V oc the most affected parameter [32,50].According to equation ( 8), a diode's reverse saturation current (I 0 ) strongly depends on the concentration of intrinsic charge carriers (~n i 2 ), which are highly affected by temperature and described by diffusion theory [51].The output results are shown in Fig. 8 (a and b) and Table 5.A decrease in V oc is observed with an increase in temperature from 1.2622 V for 280 K to 1.1063 V for 400 K.This behavior of a decrease of open-circuit voltage (V oc ) with rising temperature is caused by an increase in the diode's reverse saturation current (I 0 ), whereas the short-circuit current (J sc ) has slightly increased.An increment in the short-circuit current (J sc ) from 21.60 mA/cm 2 for 280 K to 21.64 mA/cm 2 for 400 K.
Meanwhile, an increase in temperature leads to a decrease in FF.Earlier, it was stated that a drop in V oc is mainly responsible for the drop in FF, while a rise in J sc does not have much effect [31].The maximum efficiency obtained is 24.42% at 280 K when w(ZnSnN 2 ) = 1 μm.

Influence of absorber layer defect density on photovoltaic characteristics
Fig. 9 shows the current density-voltage characteristic J-V for different defect states (N t ) in the ZnSnN 2 semiconductor.Fig. 10 (a   and b) shows the extracted parameters from J-V curves (J sc , V oc , FF, and η) as a function of absorber layer defect density (N t ).All these parameters get decreased when defects are included.In electronic devices, it is common for semiconductor defects to cause electronhole recombination [28].Defects in a material can create localized energy levels within the bandgap, trapping charge carriers and decreasing their lifetime.It can cause current leakage, reducing the solar cell's yield and negatively impacting its electrical characteristics [52].The corresponding efficiencies for 10 10 cm − 3 and 10 17 cm − 3 were about 23.9% and 15.54%, respectively.These values have undergone testing to address the limited availability of experimental data for the p-type ZnSnN 2 semiconductor.The reported total density of defects for the ZnSnN 2 material can be found in the previous literature [4,26,33].The improvement in the recombination process, which causes the annihilation of the charge carriers, is primarily responsible for the decrease in performance with an increase in defect density.A lower defect density results in a higher carrier diffusion length and a lower recombination rate, which improves PV performance.

Influence of defect density at CdS/ZnSnN 2 and ZnSnN 2 /Si interfaces
In this section, we performed a numerical simulation to explore the influence of defect density on the interfaces of CdS/ZnSnN 2 and ZnSnN 2 /Si in ZnSnN 2 solar cells.Structural defects in heterojunction PV devices can result in interfacial defects.Hence, it is crucial to investigate the impact of these defects on the solar output parameters.Our findings indicate that defects at the interface I (Buffer/ Absorber) can increase the likelihood of carrier trapping and R s in the HJT [47].Fig. 11(a) shows the effect of defect density at the CdS/ZnSnN 2 interface on the PV parameters.The neutral defect density at the CdS/ZnSnN 2 interface varies from 10 7 to 10 13 cm − 2 .The Fig. 7. J(V) characteristics for different operating temperatures, where w p (w ZnSnN2 ) = 1 μm.
A. Laidouci et al. corresponding parameters can be found in Table .2.In Fig. 11(a), V oc decreases from 1.2377 to 1.2037 V as the CdS/ZnSnN 2 interface defect density ranges from 10 7 to 10 13 cm − 2 .While J sc remains unchanged as the CdS/ZnSnN 2 interface defect density ranges from 10 7 to 10 13 cm − 2 .The high defect density at the buffer/absorber interface increases the series resistance of the heterojunction solar cell [47].Our results show a slight decrease in the fill factor, dropping from 89.12% to 89.02%, corresponding to a variation in the CdS/ZnSnN 2 interface defect density from 10 7 to 10 13 cm − 2 .The efficiency decreased from 23.9% to 23.15% as the CdS/ZnSnN 2 interface defect density increased from 10 7 to 10 13 cm − 2 .This increase in defect density led to a higher carrier recombination rate [47].    A. Laidouci et al.
parameters of the solar cell remain nearly constant as the defect density increases from 10 7 to 10 13 cm − 2 .Consequently, the defects at interface II (Absorber/BSF) do not significantly impact the photovoltaic performance parameters [47].

Impact of parasitic resistances on photovoltaic performance
Using SCAPS software, losses at the p-n junction could be considered.Additional research is required to explore potential applications of ZnSnN 2 material.Our study focused on the impact of series and shunt resistances on the device.However, further investigation is necessary to fully demonstrate the advantages of this emerging ternary nitride semiconductor in the context of photovoltaic  devices and optoelectronic applications.Fig. 12 shows the variation of R s and R sh of the new thin ZnSnN 2 solar cell and the corresponding change in (a) Jsc, (b) Voc, (c) FF, and (d) η at ambient temperature.As discussed in Ref. [53], high shunt and low series resistance lead to high efficiency.Based on this condition, series, and shunt resistances have been studied to improve performance from 1 to 5 Ω cm 2 and 10 1 -10 6 Ω cm 2 , respectively.We can observe a similarity in the trends of J SC (Fig. 12(a)), V OC (Fig. 12(b)), FF (Fig. 12  (c)), and PCE (Fig. 12(d)).In the case of J SC (Fig. 12 (a)), the results indicate a maximum value of over 21.64 mA/cm 2 when the series resistance (R s ) is approximately 1 Ω cm 2 , and the shunt resistance (R sh ) is greater than or equal to 10 2 Ω cm 2 .Regarding V OC (Fig. 12  (b)), the results show a maximum value of over 1.24 V, regardless of the series resistance value, when the shunt resistance (R sh ) is greater than or equal to 10 2 Ω cm 2 .Similarly, FF (Fig. 12(c)) demonstrates a maximum value of over 88% when the shunt resistance (R sh ) is greater than or equal to 10 3 Ω cm 2 , irrespective of the series resistance value.Fig. 12(d) displays the efficiency (PCE), and it illustrates that the best efficiency of 23.9% is achieved when R s = 1 Ω cm 2 and R sh = 10 6 Ω cm 2 .Thus, achieving an ideal balance between series and shunt resistance is crucial for optimal solar cell performance.

Influence of back surface field (BSF) layer thickness on photovoltaic characteristics
A back surface field layer with a higher doping concentration is added on the back side of a solar cell.This work combines ZnSnN 2 with a thin silicon layer to act as BSF [27,54].It has been demonstrated that incorporating a passivation layer may enhance the efficiency of collecting photogenerated carriers in photovoltaic devices [47].As an element, silicon is highly abundant, beneficial, and cheaper than some III-V materials [55].Under different absorbing thickness layers, the effect of variation in Si (BSF) layer thickness on cell performance is illustrated in Fig. 13.The thickness of the absorber layer varies between 1 μm and 8 μm, while the thickness of the Si A. Laidouci et al. layer varies between 0.1 μm and 0.9 μm.It can be observed that the variation in the thicknesses of the absorber layer and BSF layer affects the J SC (Fig. 13(a)), V OC (Fig. 13(b)), FF (Fig. 13(c)), and PCE (Fig. 13(d)) of the solar cell.It can be seen that the J sc and η increase slightly with increasing Si thickness.In contrast, the absorber layer's thickness strongly affects the cell's performance [56].Shockley-Queisser confirms the better performance of this ZnSnN 2 -based SC, as shown in Fig. 14.For the value of w p (ZnSnN 2 ) = 8 μm, we obtain an efficiency of η 29.5% (∼ 30%) as mentioned in Ref. [57].A high absorption coefficient (α) of ~10 5 cm − 1 made ZnSnN 2 comparable to the III-V, I-II-IV-VI, and I-III-VI 2 semiconductors [6,35,58,59].During p-n junction formation, an electric field is formed at the interface ZnSnN 2 /Si on the rear face to lower surface recombination velocity, thereby improving the solar cells' electrical characteristics [60].As a result of the 0.3 μm width of the back surface field layer and the 8 μm width of the absorber layer, the J sc is 25.50 mA/cm 2 , with a conversion efficiency of 29.5%.On the other hand, in the case of a thin ZnSnN 2 solar cell with a thickness of w p = 1 μm, a P + -P junction is formed by incorporating a BSF layer (p + -Si) between the absorber layer (p-ZnSnN 2 ) and the rear side of the cell.This configuration results in a significant electric field strength of approximately 3.56 MV/cm, as determined through a SCAPS software analysis [61,62].This information is visually represented in Fig. 15.The high electric field in the p + -p junction effectively reflects minority electrons from the back surface.This reflection mechanism helps increase the J sc by reducing the dark current, as discussed in Ref. [63].

Optimized ultrathin ZnSnN 2 solar cells
Table .6displays the photovoltaic parameters that we have obtained based on ultrathin ZnSnN 2 structure solar cells compared with CIGS [64] and CZTS [65] ultrathin structures solar cells, where w p ~ 500 nm.In Table .5, the ZnSnN 2 ultrathin solar cell shows the best performance over CIGS and CZTS, with an efficiency of 20.08%.Fig. 1(a) depicts the thin ZnSnN 2 solar cell's structure without BSF.Fig. 16 shows the J-V characteristics of the thin ZnSnN 2 SC BSF and with a BSF layer in illumination and w p (ZnSnN 2 ) = 1 μm.
Using the parameters listed in Tables 1-3, we obtained the J-V characteristic at room temperature.After inserting BSF, improvement in J sc 19.86-21.63mA/cm 2 and efficiency 19.6% to 23.9% (about Δη = 4.3%) was observed (Table .7).Therefore, inserting the BSF layer reduces recombination at the rear surface, enhancing carrier collection through efficient charge carrier extraction and reducing surface recombination velocity by increasing V oc and FF [66,67].The optimal results were achieved with J sc = 21.63 mA/cm 2 , V oc = 1.24V, FF = 89.1%,and PCE ~24% with an optimal BSF layer thickness of 0.3 μm.

Generation and recombination mechanism
This section analyzes the generation and recombination mechanisms of the thin ZnSnN 2 structure.It can be seen in Fig. 17 that the highest generation rate is achieved for the proposed thin solar cell when w p = 1 μm, w BSF = 1 μm, and N t (ZnSnN 2 ) = 10 10 cm − 3 .The maximum generation rate (9.62 × 10 21 cm − 3 s − 1 ) was found at positions of ~2 μm.This is because the absorption rate of photons at this particular position in the studied cell is higher than that of other positions.Consequently, the maximum generation rate describes the maximum number of electrons generated at a specific location.On the other hand, recombination is the inverse process of generation, during which electrons and holes can recombine and mutually annihilate [68].Depending on the type of recombination, this can involve carriers from the conduction band, the valence band, or intermediate energy levels [66].As previously stated, defects in each layer of the proposed cell also affect electron-hole recombination.The maximum recombination rate for the cell (5.46 × 10 21 cm − 3 s − 1 ) was also observed at positions of 2 μm, similar to the generation rate.

Impact of the front contact and back contact work function on photovoltaic parameters
This section analyzes the effect of the front contact and back contact work function on the performance of thin ZnSnN 2 SCs.With an

Table 7
Comparison between performances of thin solar cells with and without Si BSF layer, where w p (w ZnSnN2 ) = 1 μm.

An overview of theoretical and experimental research on ZnSnN 2 solar cells
The structures based on ZnSnN 2 semiconductor and their corresponding conversion efficiencies are listed in Table .8[15][16][17][18]72,73].Recent theoretical and experimental studies on various ZnSnN 2 solar cells demonstrate the suitability of this material for enhancing efficiency and reducing costs, as shown in Table 8.It has been recognized that developing alternative materials, such as ZnSnN 2 , is of utmost importance and poses a significant challenge.Moreover, the potential applications of ZnSnN 2 extend beyond solar cells and encompass various optoelectronic devices.For instance, it can be employed as a light emitter, offering a stable functionality due to its low mismatch (or low strain) with wurtzite III-N materials, demonstrating exceptional compatibility [4,[74][75][76][77][78].The nomenclature used throughout this paper is listed in Table .9.

Conclusion
This study employed SCAPS-1D software to investigate the electrical properties of ZnSnN 2 solar cells.Numerous factors were analyzed, encompassing the thickness of the window layer, buffer layer, absorber layer, and BSF layer, as well as the operating temperature, parasitic resistances (series and shunt), defect density at the interfaces, and the absorber layer, electric field, contacts work function, and generation-recombination profile.The parameters were optimized under room temperature, one sun AM1.5G, and flat-band conditions on the front contact.The study's results demonstrated that a thin solar cell with a thickness of 1 μm achieved an efficiency of 23.9% under optimal conditions.For a practical solar cell with a thicker absorber at room temperature, the following optimal values were determined: w p = 8 μm, w BSF = 0.3 μm, R sh = 10 6 Ω cm 2 , R s = 1 Ω cm 2 , and a low defect density in the ZnSnN 2 semiconductor (N t = 10 10 cm − 3 ), resulting in a significantly higher efficiency of 29.5%.Compared to conventional thin-film solar cells like CIGS and CZTS, ZnSnN 2 -based structures offer distinct advantages, including a high absorption coefficient (10 5 cm − 1 ) and high efficiency (~30%).Consequently, developing environmentally friendly and cost-effective materials such as ZnSnN 2 is paramount and poses a significant challenge for future advancements in solar technology.However, additional investigations are required to address limitations and explore further factors.This study offers valuable insights into the p-type characteristics of ZnSnN 2 , illuminating an area of research that requires further attention and development.Furthermore, this study emphasizes the significant potential of this material for various optoelectronic devices.Finally, as a promising perspective, we propose the incorporation of ZnSnN 2 in tandem solar cells, in conjunction with Si, to attain high efficiency.This proposition is based on recent developments in the field.

Fig. 3 .
Fig. 3. Contour plot of the effect of the thickness of the ZnO window layer and the thickness of the CdS buffer layer on (a) Jsc, (b) Voc, (c) FF, and (d) η.
Fig. 11(b)  shows the effect of defect density at the ZnSnN 2 /Si interface on the PV parameters.In Fig.11(b), it is evident that the output

Fig. 9 .
Fig. 9. J-V curves of thin ZnSnN 2 solar cell at the varying absorber layer defect densities (N t ).

Fig. 10 .
Fig. 10.(a) Variation of V OC and J SC , (b) Variation of FF (%) and η (%) as a function of defects density in the p-ZnSnN 2 layer.

Fig. 13 .
Fig. 13.Contour plot of the effect of the thickness of the absorber layer and the thickness of the BSF layer on (a) Jsc, (b) Voc, (c) FF, and (d) η.

Fig. 14 .
Fig.14.The detailed balance limit (or Shockley-Queisser limit) of power conversion efficiency as a function of band gap energy for single-junction solar cells[57].

Fig. 16 .
Fig. 16.J-V curves of the new thin ZnSnN 2 SC with and without BSF.

Fig. 17 .
Fig. 17.Variation of carrier recombination and generation rates of the thin ZnSnN 2 solar cell structure.

Fig. 18 .
Fig. 18.Effects of (a) front contact metal work function and (b) back contact work function on PV performance.

Table 1
Input parameters used in the study and their values.

Table 3
Simulation parameters for front and back contacts.

Table 4
Impact of absorber layer thickness on PV cell parameters of ZnSnN 2 at 300 K.

Table 5
Influence of operating temperature on the photovoltaic parameters.

Table 6
PV parameters for ultrathin ZnSnN 2 structure solar cells compared with CIGS and CZTS ultrathin solar cells (w p ~500 nm).

Table 8
Functional characteristics of experimental and simulated ZnSnN 2 Solar cells.