Introduction

CZTS (Copper Zinc Tin Sulfide) shows excellent photovoltaic properties with optical absorption coefficient (> 104 cm− 1)1. About 90% of incident photons in the sun spectrum can be absorbed by it. CZTS exhibits a band-gap modulation in between of 1.4–1.5 eV2,3,4,5. Due to its excellent optical properties, many researchers found CZTS as a promising alternate absorber layer and used it to develop thin-film solar cell systems6. The abundance, long-life stability, and non-toxic elements of CZTS make it an appropriate material for developing subsequent generation thin-film-based solar cells7. According to (Wang et al.8, performance (power conversion efficiency) of CZTS solar cells currently stated as 12.6% in AM 1.5 spectrum. However, it possible to achieve as high as 32.2%, that is significantly higher than the efficiency reported by Shockley9(Hossain et al.10. Making a high performed CZTS solar cell lot of works carried out both analytically and experimentally so far11,12,13. Conversely, PCE of this kind PV device is comparatively low as CIGS, CdTe, and organic-inorganic metal halide perovskites due to a number of variables including defect chemistry. There are several publications in the literature that addresses how to achieve the high-efficiency. Higher efficiency can be obtained by changing absorber layer of kesterite composition with the intention to occur alongside buffer layer14,15. Tajima et al. reported that, open circuit voltage will increase by adding a CZTS2 layer more concentrated in carriers than the CZTS1 layer15,16. Based on these experimental results, a theoretical investigation was performed by using CdS as a buffer layer. They presented ACZTS/CZTS (p+) bilayer solar cell with silver, which has a maximum efficiency of 17.59% Saha and Alam17. Abderrezek et al. reported that using CdS and ZnS buffer layers with a stacked BSF layer will increase the efficiency. A maximum efficiency of 15.3% was reported in this work by Abderrezek et. al18. In another simulation-based study, it was shown how to use a bilayer absorber material to boost the performance of the solar cell by Mazumder et al.19. Based on these reported works it is crucial to improve the performance of the solar cell using bilayer absorber material or a BSF layer. Within this endeavor, we have put forward a method that is grounded on ZnMgO/CZTS embedded with a BSF layer to achieve maximum performance. A number of materials are used as buffer layer for CZTS solar cell, not limited to ZnS, CdS, ZTO etc20. –15. However, in comparison with the traditional buffer layer CdS, ZnMgO exhibits greater radiation resistance and environmental compatibility as addressed by Hu et al.21. In addition, it was shown that the ZnO window layer and ZnMgO buffer layer provide a good lattice match, which improves the structural stability of the device and makes it suitable for use in solar cells. As ZnMgO has a controlling optical bandgap that is larger than that of CdS and helps to increase the efficiency of solar cells, using it as buffer layer reduce absorption loss.

The device is modeled here with the SOLAR CELLAPS-1D tool, and a detailed investigation has been conducted into effect of multiple optical and structural factors on efficiency of CZTS - based solar cells. This study models and investigates a novel device architecture (ZnO: Al/i-ZnO/ZnMgO/CZTS1/CZTS2/Substrate) of a CZTS-based solar cell. Here, the work was simulated, modeled, and then validated with an extra absorber layer (BSF) to obtain an enhanced performance in terms of cell efficiency. The device has been optimized by considering several factors, including defect density, thickness, bandgap, operating temperature and various shunt and series resistance combinations.

Importance of ZnMgO buffer layer

ZnMgO is a ternary compound with a tunable bandgap, making it suitable for solar cells and PV devices22,23,24. Recently, it was discovered that, in some thin-film solar cell structures, ZnMgO alloy can offer a nontoxic and earth-abundant alternative buffer layer to toxic cadmium sulfide (CdS)25,26,27. Additionally, many solar cell structures based on ZnMgO demonstrate an ability to improve cell efficiency, thus enhancing their utility in the future28,29. Bandgap of hexagonal wurtzite MgxZn1−xO buffer layer will be given as30:

$${E_g}\left( x \right){\text{ }}={\text{ }} - 1.3{\text{ }}{x^2}+{\text{ }}4.8x+2.8$$
(1)

Equation (1) says that bandgap of MgxZn1−xO increases with increasing more fraction of Mg (x), which is validated by experimental data up to x = 35% as this range is consistent with the stability of the wurtzite MgZnO configuration. It is obvious from Table 1, the lattice parameter a increases, while c decreases with x in wurtzite structure of MgxZn1−xO.

Table 1 Lattice mismatch of various layers21,28,30.
Full size table

When the epitaxy layer has a different lattice parameter than the substrate, the epitaxy layer is deformed either in tension or in compression by biaxial (εxx and εyy) and uniaxial (εzz) deformations as per the following Eqs3132,.

  • Biaxial in the growth plan:

$${\varepsilon _{xx}}={\varepsilon _{yy}}=\frac{{{a_s} - {a_\varepsilon }}}{{{a_e}}}={\varepsilon _{//}}$$
(2)

  • Uniaxial in the direction of growth:

$${\varepsilon _{zz}}={\varepsilon _{yy}}=\frac{{{c_s} - {c_e}}}{{{c_e}}}~={\varepsilon _{ \bot ~}}~~~~$$
(3)

Where ε is deformation or mismatch. as/e and cs/e are lattice parameters of the substrate/epitaxial layer for a wurtzite system. As per the Eqs. (2) and (3), Mg concentration affects the lattice mismatch (ε// and ε⊥) at ZnO/MgZnO and MgZnO/CZTS interfaces evident in Table 1.

Table 2 Simulation parameters of the proposed solar cell33,34,35,36,37.
Full size table

The close lattice parameters between ZnO and MgZnO resulted in the low strain at the ZnO/MgZnO interface. For low Mg concentrations i.e. x ≤ 20%, the biaxial and uniaxial deformations at the ZnO/MgZnO are approximately zero, therefore x = 20% is considered for the present work. In contrast, large mismatch of lattice at MgZnO/CZTS interface produces dislocations that directly affect the defect density. Therefore, it is imperative to find an excellent partner for MgZnO to ensure structural stability38,39. Possibly more research is needed to find a close absorber with these interesting buffer layers.

Theoretical formulation and simulation

The proposed work has been simulated using simulator tool SCAPS 1-D to study the outcomes of the device40. The simulator uses basic semiconductor equations such as electron transport, continuity, hole transport and Poisson Eq6. The solar cell schematic view of basic solar power cell without BSF layer is depicted in Fig. 1(a), while device proposed, having architecture of ZnO: Al/i-ZnO/ZnMgO/CZTS1/CZTS2 with a BSF layer is represented in Fig. 1(b). Equivalent EBD (energy-band diagram) plotted in Fig. 1(c). It has been found that in regulating transportation of carriers through metallic contacts, band offset plays a vital role. By carefully controlling the band offset, the performance of devices can be significantly enhanced, leading to reduced leakage current, improved efficiency of charge separation, and enhanced carrier collection (Jha et al., 2020). The degree of the band offset is resolute by divergence in absorber and buffer layers electron affinity. Difference results in formation of spike-like structures for positive band offsets and cliff-like structures for negative band offsets as shown in Fig. 1(c)42. The following are the equations for valence and conduction bands offset, respectively.

(Prabhu et al., 2021):

$$~\Delta ~{E_C}\left( {CBO} \right){\text{ }}=~{\chi ^{CZTS}} - ~{\chi ^{ZnMgO}}~~~~~\left( {CZTS - buffer} \right)$$
(4)

Thus, the orientation of valence band specifies as follows.

$$\Delta ~{E_V}\left( {VBO} \right){\text{ }}={\text{ }}(~{\chi ^{CZTS}}+{\text{ }}{E_g}^{{CZTS}}){\text{ }} - ~(~{\chi ^{ZnMgO}}+{\text{ }}{E_g}^{{ZnMgO}})$$
(5)

So

$$\Delta ~{E_g}~=~\Delta ~{E_C}~\left( {CBO} \right)+~\Delta ~{E_V}~\left( {VBO} \right)$$
(6)

Where electron affinity is the χ and bandgap energy is the Eg.

In the proposed device, gold (Au) and copper (Cu) are employed as frontside and backside contacts19. For conducting layer Aluminum-doped zinc oxide (ZnO: Al) is used34,35,37. By acting as high-resistive layer, i-ZnO (intrinsic zinc oxide) lowers leakage current36. Highly doped CZTS2, two materials viz. ZnMgO, lightly doped CZTS1 are proposed as BSF layer, buffer layer and absorber layer respectively to improve device performance. Very thin or very thick buffer layer cannot be chosen. Because, thin buffer layer leads to narrow depletion region resulting poor Voc, whereas thick buffer layer absorbs more light resulting reduced overall efficiency. Henceforth, thickness of buffer layer and absorber layer optimization is very much essential subsequently, thickness been altered from 20 nm to 100 nm of window and buffer layers and their performance on the device has been studied. Series and shunt resistance initial values are kept at 1.97 Ω-cm2,1424.03 Ω-cm2 respectively (Tajima et al., 2015) to study the performance of the device. By AM 1.5 global radiation of 100 mWcm− 2 device is irradiated. Details of parameters, recombination model, and the defect specifications used in this simulation are presented in Table-2, −3, and − 4, respectively. The equivalent circuit of the solar cell shown including resistive losses is shown in Fig. 2. The governing equation of the single diode equivalent circuit model is given by43,44:

$$I={I_{ph}} - {I_0}\left( {{{\exp }^{\frac{{q(V+I{R_S})}}{{nKT}}}} - 1} \right) - \frac{{(V+I{R_S})}}{{{R_{Sh}}}}$$
(7)

Where V is applied voltage, diode ideality factor denoted by n, and Rsh is shunt resistance and Rs is series resistance45,46,47,48,49,50. We call photocurrent Iph. Reverse saturation current of diode is denoted by I0. Measurement of transmission, reflection, and/or absorption spectra published in the literature was utilized to calculate the absorption coefficient spectrum of the layers used in this investigation, which is shown in Fig. 351,52,53,54,55.

Table 3 Recombination Model16,19.
Full size table
Fig. 1
Fig. 1The alternative text for this image may have been generated using AI.
Full size image

Thin-film composite structure of (a) basic and (b) Our CZTS solar cells. (c) E-B diagram of solar cell with CZTS2 as BSF and ZnMgO as buffer layers.

Fig. 2
Fig. 2The alternative text for this image may have been generated using AI.
Full size image

Equivalent circuit of an ideal solar cell under illumination.

Fig. 3
Fig. 3The alternative text for this image may have been generated using AI.
Full size image

Absorption profile of all the materials used in simulation tool.

Results and discussions

Impact of bandgap of BSF and absorber layers

Figure 4 presents a contour map showing efficiency of solar power cell varies with absorber layer and BSF bandgaps. It is clear from the figure that Voc (open circuit voltage) rises when CZTS1 and CZTS2 bandgaps widen. Certain levels of the CZTS1 bandgap, falls as bigger bandgap blocks photons of lower energy. With respect to bandgap pair of CZTS2 and CZTS1 layer, the fill factor rises linearly. The reduction in VOC observed by incrementing CZTS2 band gap can be assigned to the higher CZTS2 work function. Although efficiency is rising when given that bigger bandgap for CZTS2 and CZTS1, as seen in image, higher band gap prevents photons of lower energy from being captivated. Consequently, Jsc drops. It is high when considering band gaps of CZTS1 1.5 eV and CZTS2 1.4 eV. Here, we can observe that the fill factor is diminishing as the CZTS1 bandgap increases. In order to get the best possible SOLAR CELL performance, we have locked the pair at 1.5 eV for CZTS1 and 1.4 eV for CZTS2. This work took into account CZTS1 and CZTS2 BSF layers bandgap pairs of 1.5 eV and 1.4 eV, in sequence, in order to achieve an appropriate device performance19.In light of this, we were able to achieve a 23.67% efficiency rate in our work.

Table 4 Defect specification for the each layer15,19,37.
Full size table
Fig. 4
Fig. 4The alternative text for this image may have been generated using AI.
Full size image

Contour plot of (a) VOC, (b) JSC, (c) FF, and (d) η as a function of bandgap of CZTS1 and CZTS2 layers.

Collective impact of thickness of BSF and absorber layers

Here, CZTS1 absorber layer and CZTS2 BSF layer thickness has been evaluated, relation to solar cell performance metrics open circuit voltage (Voc), short circuit current (Jsc), fill factor (FF), and efficiency (η), respectively. CZTS1 and CZTS2 layers thickness can be adjusted from 0.2 μm to 1 μm and 0.04 μm to 0.14 μm, accordingly, as seen in Fig. 5. It is seen that with an increase in CZTS1 thickness beyond 0.3 μm the Voc doesn’t change much. But in the case of Jsc, FF, and efficiency also for a particular thickness of CZTS1 with an increase in CZTS2 thickness, the value remains constant. Nonetheless, for a fixed CZTS1 thickness, the figure shows that the cell efficiency, fill factor (FF), and short-circuit current density (Jsc) increase with increasing CZTS1 thickness. In contrast, when the CZTS2 thickness exceeds 0.04–0.05 μm, the open-circuit voltage (Voc) remains nearly constant as the CZTS1 thickness is increased from 0.2 to 0.8 μm. The carriers can be formed as we raise the thickness. Thus, as thickness grows, so does the Voc. The increased thickness also leads to an increase in SRH (Shockley-Reed-Hall) recombination, due to which lowers fill factor. On other hand, when CZTS1 thickness grows, so do efficiency and Jsc. To achieve better solar cell performance, we must select an optimal value for CZTS1 and CZTS2 thickness. Considering every single one of these factors, we have decided on 0.07 μm for CZTS2 and 0.8 μm for CZTS1.

Fig. 5
Fig. 5The alternative text for this image may have been generated using AI.
Full size image

Contour plot of (a) VOC, (b) JSC, (c) FF, and (d) η as a function of the thickness of CZTS1 and CZTS2 layers.

Absorber layer doping impact on the cell

In this study, the effects of changing the acceptor dopant concentration on efficiency of a solar cell were analysed as depicted in Fig. 6 (a). Device outcome metrics, namely Jsc, FF (fill factor) and η (efficiency) and, remain constant till dopant concentration of 1016 cm− 3as depicted in Fig. 6 (d). Increase of the electric field in space charge region, when doping concentration is incremented from 1012 cm− 3 to 1018 cm− 3 credited to increase in Voc, which credited to, leading to decline of free carrier recombination36. However, increment in bulk process of recombination decreases minority carrier lifetime, resulting in a lower JSC. As can be seen in the figure, as doping density increments from 1012 cm− 3 to 1018 cm− 3, JSC declines. Fill factor changes steadily with solar cell doping concentration to 1016 cm− 3, but decreases there after due to the weak grain boundaries of n-type in the microcrystalline layer. This weakens device efficiency because of incremented recombination rate. This study found that controlling the grain boundaries is crucial to prevent the efficiency from dropping as the carrier concentration increases.

Fig. 6
Fig. 6The alternative text for this image may have been generated using AI.
Full size image

Current vs. Voltage curves with varying (a) acceptor doping conc. (Na) (b) total defect density (NDD) (c) back contact work function ϕB respectively. Corresponding performance parameters with (d) Na (e) NDD & (f) ϕB.

The efficiency increases slightly from 23.5% to 23.7% for doping concentrations of 1012 cm− 3 to 1016 cm− 3 respectively. However, further increment in doping concentration from 1016 cm− 3, efficiency declines. Therefore, this study, doping concentration of 1016 cm− 3 was considered the best for solar cell’ performance.

Impact of absorber layer defect density

Absorber layer defect density has significant impact on opto-electric characteristics of a material system, particularly for solar cells. Here the outcomes of solar cell is presented by considering defect density from between 1012 and 1018 cm− 3, presented in Fig. 6 (b). Higher defect density lead to an increment in recombination, resulting in greater recombination of photo generated charge carriers in absorber layer, which can decrease solar cell performance. Results from the Fig. 6 (e) indicate that the solar cell’s performance remains relatively constant when defect density is between 1012 and 1014 cm− 3. However, when defect density is incremented to 1018 cm− 3, the performance of solar cell significantly reduced, as seen by the reduction in Voc, Jsc, FF, and efficiency. Jsc is decreasing from 28.766 mA/cm2 to 9.05 mA/cm2, Voc is decreased from 1.05 V to 0.87 V, FF is from 76.60 to 47% and the efficiency is decreasing from 24% to 3.7% when defect density is incremented from 1012 to 1018 cm-3 respectively. Therefore, it is crucial to research and understand the effects of defects in absorber layer to optimize solar cell performance.

Impact of back contact work functions

Back contact work function can have a major effect on efficiency of CZTS solar cells. Here BC work function is altered from 5 to 5.6 eV respectively56,57 to measure the cell outcome as presented in Fig. 6 (c). From Fig. 6 (f), it is clear that as back-contact metal work function grows, efficiency rises as well. As seen in the image, there is an increasing behavior up to 5.4 eV, after which there is no desirable change. Choosing a higher work function for back contact makes ohmic contact with absorber layer which is p-type material However, non-ohmic back contact will behave like a Solar cell Schottky barrier to the PN junction in device. Back contact work function affects energy barrier at interface between back contact and CZTS absorber layer, which in turn impacts efficiency of charge carrier extraction from absorber layer. Low back contact work function can ease effective electron extraction from CZTS absorber layer, which is crucial for resulting higher efficiency in CZTS solar cells. This is because a low work function back contact can reduce the energy barrier for electron extraction and increase the flow of electrons towards the back contact. In contrast, a high work function back contact can create a significant barrier for electron extraction, which can limit the efficiency of the CZTS solar cell. In such cases, the flow of electrons towards the back contact is reduced, resulting in lower efficiency. Therefore, back contact work function is a vital factor to consider in design and optimization of CZTS solar cells, and a low work function back contact is typically preferred for achieving high efficiency. To get the best possible solar cell performance, our model simulated at 5.4 eV for back contact to examine the solar cell’s efficiency.

Study of temperature on cell characteristics

The Varshni relation provides a way to describe semiconductors temperature dependence of bandgap energy. It is provided by the Eqs5859,:

$${E_g}(T)={E_g}(0) - \frac{{\alpha {T^2}}}{{(T+\beta )}}$$
(8)

Where bandgap energy at absolute zero is denoted by Eg (0), α is temperature coefficient of the bandgap, bandgap energy at temperature T is denoted by Eg(T) and β is a material-dependent constant. The Varshni relation is useful for predicting how the semiconductor bandgap energy changes with temperature59. This signifies importance for solar cells because the bandgap energy determines the maximum efficiency of the cell. As the temperature of the solar cell increases, the bandgap energy decreases, which can lower performance of cell. In study, performance of the solar cell was evaluated by alternating temperature from 300 K to 400 K with and without a BSF layer as in Fig. 7. Results was tabulated in Table 5, which likely shows how the efficiency of the solar cell changes with temperature. The Varshni relation could be used to model this temperature dependence of the bandgap energy and predict the corresponding changes in efficiency. Jsc is given as59,

$${J_{SC}}=q\int_{{h\upsilon ={E_g}}}^{\infty } {\frac{{d{N_{ph}}}}{{dh\upsilon }}} d(h\upsilon )$$
(9)

VOC is the extreme voltage available from a solar cell, given as,

$${V_{OC}}=\frac{{KT}}{q}\ln \left( {\frac{{{J_{SC}}}}{{{J_0}}}+1} \right)$$
(10)

The fill factor can be expressed as60,

$$FF=\frac{{{V_{OC}} - \ln ({V_{OC}}+0.72)}}{{{V_{OC}}+1}}$$
(11)

Since the minority carriers are produced thermally, Jo is extremely sensitive to variations in temperature. For p–n junction solar cell, reverse saturation current density, Jo, has characterized as follows:

$$~{J_0}=q\left( {\frac{{{D_n}}}{{{L_n}{N_A}}}+\frac{{{D_p}}}{{{L_p}{N_D}}}} \right)n_{i}^{2}$$
(12)

Where

$$n_{i}^{2}={N_c}{N_v}exp\left( {\frac{{ - {K_g}}}{{{K_T}}}} \right)m_{e}^{{(*3/2)}}m_{h}^{{(*3/2)}}exp\left( {\frac{{ - {E_g}}}{{{E_T}}}} \right)$$
(13)

Where intrinsic carrier density is ni, donor and acceptor atom densities are ND and NA, minority carrier diffusion constants are Dp and Dn in the p and n regions, respectively, and minority carrier diffusion lengths are Lp and Ln in the p and n areas. Boltzmann’s constant is KB, ideality factor is n, and temperature in kelvin is T.

Table 5 Simulated data demonstrating the cell’s performance at various temperatures.
Full size table
Fig. 7
Fig. 7The alternative text for this image may have been generated using AI.
Full size image

Variation of current density with applied voltage considering different temperature values.

In a solar cell, the decline in capacitance while incrementing the temperature is primarily because to a decline in depletion region width at pn junction. When voltage is employed to solar cell, the depletion region widens, reducing the capacity of the cell. At higher temperatures, the increased thermal energy of charge carriers causes an increment in diffusion of charge carriers through depletion region. This diffusion causes a decline in width of depletion region, which in turn declines capacity of solar cell. Therefore, decline in capacitance of a solar cell with incrementing temperature is primarily due to decline in width of depletion region caused by increased diffusion of charge carriers. An increment in resistance and a decline in number of available charge carriers can also contribute to the decrease in capacity as depicted in Fig. 8. From obtained results, it is observed that change in capacitance/area is obtained as function of voltage at different temperature from 300 to 400 K. From − 1.0 to 0 V, there is no significant change in capacitance. However, while incrementing voltage from 0 to 0.6 V, increment in capacitance achieved. Further, degradation in capacitance is occurred at higher voltages.

Fig. 8
Fig. 8The alternative text for this image may have been generated using AI.
Full size image

Capacitance Vs voltage profile of the solar cell with different temperature values.

It can be noticed from Table 5 that while increasing the temperature, decrement in Voc is obtained due to an increment in thermal energy of the electrons in cell, which can lead to higher electron-hole recombination and a decline in bandgap energy. Further, short-circuit current (Isolar cell) of cell also increases via increasing the temperature, but the increase in Isolar cell is generally less than the decrease in Voc. Therefore, decrement in overall device is achieved due to presence of charge carrier recombination with the same increment in temperature.

Impact of shunt (Rsh) and series (Rs) resistances

In a solar cell, shunt resistance and series resistance are two important parameters that can significantly affect the outcomes of cell. In general, reduced series resistance and extensive shunt resistance are desirable for improving the performance of a solar cell60. Series resistance refers to the resistance of the conductive path through which current flows in a solar cell. It encompasses the resistance of metal contacts, resistance of semiconductor material itself, and the resistance of any other elements in the current path. A high series resistance can lead to drop of voltage across cell, reducing its output power and voltage. This, in turn, can lower the efficiency of the cell and reduce its overall performance. Shunt resistance, on the other hand, refers to the resistance between the two terminals of the solar cell, in parallel with the solar cell itself. A high shunt resistance is desirable as it can minimize the leakage current and prevent power loss. A low shunt resistance, on the other hand, can lead to reduction in the efficiency and major loss of power of cell.

Table 6 Performance evaluation using the Rs and Rsh combinations.
Full size table
Fig. 9
Fig. 9The alternative text for this image may have been generated using AI.
Full size image

Variation of current density with applied voltage of solar power cell for various series and shunt resistance pairs.

In summary, a high series resistance can lead to a voltage drop and lower the efficiency of the solar cell, while a low shunt resistance can result in power loss and reduce the overall performance of the cell. Therefore, optimizing both shunt and series resistance is crucial to improve efficiency and performance of the solar cells. A number of series and shunt resistance pairs has been considered to optimize the solar cell performance as shown in Fig. 9. Experimentally obtained shunt resistance and series resistance pairs is shown in Table 6, along with the impact of our device when fed with such pairings. Considering the optimized value of Rs=1.97 ohm.cm2 and Rsh=1424.04 ohm.cm219,61 this work reported an efficiency of 23.67% as depicted in Fig. 10.

Fig. 10
Fig. 10The alternative text for this image may have been generated using AI.
Full size image

Contour plot of (a) VOC, (b) JSC, (c) FF, and (d) η as a function of Rs and Rsh.

Generation and recombination profile

G-R profile of structure is shown in Fig. 11. The generation rate, which is impacted by the material’s thickness and absorption coefficient, is quantity of electrons produced at each site of the device as a result of photon absorption. The link between intensity of the incident light, absorption coefficient, and quantity of electron-hole pairs produced is shown in Eq. (14)62.

$$\:I={I}_{0\:}{e}^{-\alpha\:x}$$
(14)

where I0 is the light intensity at the top surface generation distance inside the material at which the light intensity is being computed, and α is the absorption coefficient, which is commonly expressed in cm−1. Generation at any point in device may be obtained by differentiating the equation and given as :

$$\:G=\alpha\:{N}_{0}{e}^{-\alpha\:x}$$
(15)

Where N0 is the surface photon flux (photons/unit-area/sec).

Due to intensity of incident light, high production rate 1021 (cm−3s−1) of the electron-hole pairs is seen on rear surface of CZTS solar cell. However, at the time when the recombination rate peaks, the recombination velocity (SRV) at the surface is 107 cm/s, (Saeed et al., 2022) which is connected to revelation of defect states that reduce the lifetime and raise recombination rate of carriers and diffusion length of minority carriers.

Fig. 11
Fig. 11The alternative text for this image may have been generated using AI.
Full size image

Generation and recombination rates of photo-induced carriers across the solar cell on illumination.

Optimization of solar cell performance

With a view to optimize performance of device, the simulation was enacted considering the variations of defect density, thickness of the absorber layer and doping concentration, BSF layer and buffer layer of proposed structure. With air mass (AM) 1.5 global radiation of 100 mW/cm2 at T = 300 K device is irradiated, where Rs = 1.97 ohm.cm2 and Rsh = 1424.04 ohm.cm2 indicate the optimum efficiency. The J-V and P-V curves and equivalent Quantum Efficiency (QE) spectrum as function of incident wavelength of light of optimized cell are presented in Fig. 12.

Table 7 Comparison of performance with earlier reported experimental and simulation research.
Full size table
Fig. 12
Fig. 12The alternative text for this image may have been generated using AI.
Full size image

(a) J-V and (b) P-V curves of the optimized solar cell with and without a BSF layer. (c) Quantum efficiency spectra of the optimized solar cell. (d) Comparison of J-V curve with the existing reports.

We were able to acquire the PCE of the suggested cell with more than 23%, optimized values from simulation presented in the figure. Solar power cell with no BSF has a PCE of 19%. Figure 12 (d) illustrates J-V analysis of the previous work with our proposed work. The comparative analysis is shown in Table 7. This work shows that use of novel buffer layer with BSF layer increases the PCE to more than 4% compared to the previously reported result63,64,65,66.

Conclusion

The numerical simulation of the proposed device was carried out in SOLAR CELLAPS 1-D simulation software. The effect of various factors, namely thickness and absorber layer bandgap and BSF layer, doping effect, shunt resistance, series resistance, temperature and work function on device’s performance was studied in this work. Based on our analysis, 1.4 eV for the CZTS2 and 1.5 eV for CZTS1 layer are optimized bandgaps, gives an optimum performance of the solar cell. It was detected that increasing thickness of CZTS1 layer led to an increment in device’s PCE (power conversion efficiency). Additionally, it shows optimal thickness for BSF layer and the absorber layer was 70 nm and 800 nm, respectively. Furthermore, the authors have observed that there was a variation in device performance with metal work function values. Optimal metal work function value for back contact in this work was considered at 5.4 eV. A maximum efficiency of 23.67% was reported when Rs and Rsh were taken as 1.97 Ω-cm2 and 1424.03 Ω-cm2 respectively. Overall, findings of this study are handy for development of cost-effective, advanced, and eco-friendly CZTS solar cells.