Rasmiah S. Almufarija,
Muazma Jamilb,
M. Yasir Alib,
M. D. Alshahranic,
Salhah Hamed Alrefaeed,
Mohamed Abdelsabour Fahmyef,
Islam Ragabg,
A. R. Abd-Elwahedh,
Adnan Ali*b and
Arslan Ashfaq*i
aDepartment of Chemistry, College of Science, Princess Nourah bint Abdulrahman University, P. O. Box 84428, Riyadh 11671, Saudi Arabia
bDepartment of Physics, Government College University, Faisalabad, 38000, Pakistan. E-mail: adnnan_1982@yahoo.com
cDepartment of Physics, College of Science, University of Bisha, P. O. Box 551, Bisha 61922, Saudi Arabia
dDepartment of Chemistry, College of Science, Taibah University, Yanbu-30799, Madinah, Saudi Arabia
eAdham University College, Umm Al-Qura University, Adham 28653, Makkah, Saudi Arabia
fFaculty of Computers and Informatics, Suez Canal University, New Campus, 41522 Ismailia, Egypt
gDepartment of Chemistry, College of Science, Qassim University, 51452 Buraidah, Saudi Arabia
hDepartment of Physics, College of Science, Qassim University, Buraydah 51452, Saudi Arabia
iDepartment of Physics, Emerson University Multan, 60000, Pakistan. E-mail: arslan.ashfaq@eum.edu.pk
First published on 20th May 2025
This study investigates a tin-based perovskite solar cell (PSC) incorporating an inorganic hole transport layer, examined through simulations with the SCAPS simulator. The chosen CuZnSn(Se1−xSx) compound emerges as a promising candidate for the hole transport layer, allowing for a tunable band gap via adjustments to the S/(S + Se) ratio. The band gap varies from 0.95 eV for Cu2ZnSnSe4 to 1.5 eV for Cu2ZnSnS4, achieved through strategic valence band offset engineering at the MAPbI3/CuZnSn(Se1−xSx) interface. However, achieving an optimal Valence Band Offset (VBO) at MASnI3/CuZnSn(Se1−xSx) remains challenging yet crucial for realizing high-performance Perovskite Solar Cells. The device efficiency is systematically optimized by manipulating the S content, resulting in a noteworthy Power Conversion Efficiency of 18.29%. Furthermore, it is uncovered that a carefully selected VBO (0.22 eV) is achieved with the CZTSe0.4S0.6 hole transport layer, contributing significantly to the improved performance of the PSC. These findings underscore the importance of precise engineering in achieving optimal device properties for advanced solar energy conversion applications.
Compared to more known technologies such as CdTe, CIGS, and silicon solar cells, this photovoltaic (PV) breakthrough has been achieved rather quickly. Characteristics such as an appropriate band gap, high absorption coefficient, ambipolar charge-carrier transport, small exciton binding energy, long diffusion length, defects tolerance, low fabrication cost, and low effective carrier masses are responsible for PSCs success.4,5
Despite the remarkable accomplishments of PSCs, challenges remain for their broader adoption. Lead (Pb), a toxic element, risks human health and the natural environment.6 Additionally, MAPbI3 experiences reduced performance due to poor stability when exposed to moisture and sunlight.7 To address these issues, various nontoxic metals such as copper, bismuth, germanium, antimony, and tin have been explored as substitutes for Pb in PSCs.8,9 Sn has been identified as the most promising substitution for improving PSC performance.
MASnI3, with its narrow band gap of 1.3 eV compared to MAPbI3, exhibits a similar valency. The slight reduction in the radius of Sn2+ (1.35 Å) compared to Pb2+ (1.49 Å) allows for the replacement of Pb2+ with Sn2+ while maintaining the perovskite structure.10 The Goldschmidt tolerance factor and octahedral factor play a critical role in assessing the structural stability and feasibility of forming perovskite phases, thereby serving as essential criteria for the rational design and selection of high-performance perovskite materials in solar cell applications.11 Sn2+ substitution adheres to coordination, causes minimal lattice constant perturbation, and maintains ionic size and charge balance.12,13 Tin-based, environmentally friendly perovskite materials hold the potential to enhance device performance and stability. However, tin-based PSCs face challenges, including sensitivity to oxygen leading to rapid oxidation of Sn2+ into Sn4+, as well as the self-doping effect and rapid crystallization rate during solution preparation, resulting in current losses and reduced power conversion efficiency (PCE).14,15
Numerous techniques have emerged to enhance the performance of MASnI3-based devices. For instance, adopting a n–i–p device configuration with mesoscopic TiO2 and Spiro-OMeTAD as the hole transport material has achieved a PCE of 6.4%.16 In 2019, MASnI3 achieved a PCE of 7.19% through a normal device by employing cation exchange approaches.17 The proper selection of electron transport materials (ETM) and hole transport materials (HTM) is crucial for improving the efficiency, reproducibility, and stability of solar cell. However, the use of organic materials such as Spiro-OMeTAD, PATT, and PEDOT:PSS poses challenges due to potential degradation and high costs, hindering the commercialization of perovskites.18
Inorganic copper-based HTLs, including CuSbS2 and CuSCN, have been investigated for their suitable band gap and long-term stability.19 However, the performance of devices using these materials falls short compared to their organic counterparts. To further enhance the efficiency of PSCs with in-organic HTLs, it is crucial to identify other materials with appropriate energy band positions, inherent properties, excellent chemical stability, and high conductivity.20
Copper-based materials possess chemical stability, high conductivity, and hole transport mobility, making them suitable candidates as HTLs in PSCs. Cu2ZnSnS4 is an environmentally friendly, nontoxic, and abundant material with a high absorption coefficient of 105 cm−1, an heightened energy band gap of 1.5 eV, and excellent stability, making it viable for use in low price devices. While the maximum achieved PCE of CZTS is 9.6%, the alloying of Cu2ZnSn(Se1−xSx)4 compound has demonstrated superior performance, reaching 12.6% efficiency with a tunable band gap.21,22 Notably, CZTSe1−xSx has not only been explored as a potential light absorber in thin-film solar cells but has also shown promise as an HTL in PSCs, achieving an efficiency of 22.77%.23 CZTSe1−xSx can be synthesized using a nanoparticle ink method, eliminating the need for high-temperature sulfuration and enabling its use at lower temperatures. A study in 2016 investigated the band level alignment and valence band position influence of CZTS and CZTSe nanoparticles ink as HTL on the performance of PSCs.24
This manuscript introduces the novel concept of utilizing CZTSe1−xSx as an HTL in MASnI3 PSCs through Solar Cell Capacitance Simulator (SCAPS), marking the first time this device modeling approach has been considered. The traditional structure of PSCs is examined and discussed, focusing on the valence band offset between the perovskite light absorber and CZTSe1−xSx, as well as the band gap of the HTL controlled through composition engineering of the CZTSe1−xSx compound. The effects of these factors on the IV characteristics and efficiency parameters are thoroughly investigated, leading to an optimized power conversion efficiency achieved with a specific S composition. At CZTSe0.4S0.6 composition, a notable efficiency of 17.29% is attained. Further optimization of carrier concentration, defect density, and diffusion length of tin-based perovskite in CZTSe0.4S0.6 results in an enhanced PCE of 17.34%.
The simulation of the device is conducted using SCAPS 3.310, a software developed by the Department of Electronics and Information Systems (ELIS) at the University of Gent.25 The simulation is performed under 1 Sun illumination with an incident power density of 1 kW m−2 at a temperature of 300 K. SCAPS utilizes well-established equations from literature,26 including the continuity equation, Poisson's equation, and electron/hole transport equations, to model various recombination mechanisms for solar cell simulation.
Table 1 presents fundamental parameters obtained from experimental and published data for the HTL, ETL, and FTO, as documented in sources. To represent the optical behavior of each layer accurately, the absorption coefficients for all active layers were incorporated. The absorption coefficient of MASnI3 was modeled using the Tauc relation with a pre-factor Aα = 105, while the absorption profiles of TiO2 and CZT(Se1−xSx) were extracted from reported experimental literature.27,30,31 The effective density states for valence and conduction bands is established at 1.8 × 1019 and 2.2 × 1018 cm−3, respectively, with the exception of the ETL.32 The thermal velocities of holes and electrons are set at 1 × 107 cm s−1. For accurate representation, MASnI3 is assigned an electron carrier density of 2.0 × 1016 cm−3.
Parameters | FTO | TiO2 | MASnI3 | Spiro-OMeTAD | CZTSe | CZTS |
---|---|---|---|---|---|---|
εr | 9 | 10 | 8.2 | 3.0 | 10 | 10 |
χ (eV) | 4 | 4 | 4.17 | 2.45 | 4.35 | 4.5 |
Eg (eV) | 3.5 | 3.2 | 1.3 | 3.0 | 0.95 | 1.5 |
Nc (cm−3) | 2.2 × 1018 | 2 × 1017 | 1.0 × 1018 | 2.2 × 1018 | 2.2 × 1018 | 2.2 × 1018 |
Nv (cm−3) | 1.8 × 1019 | 6 × 1017 | 1.0 × 1018 | 1.8 × 1019 | 1.8 × 1019 | 1.8 × 1019 |
μn (cm−2 V−1 s−1) | 20 | 100 | 1.6 | 2 × 10−4 | 60 | 100 |
μp (cm−2 V−1 s−1) | 10 | 25 | 1.6 | 2 × 10−4 | 20 | 25 |
Na (cm−3) | 0 | 0 | 0 | 2 × 1018 | 5 × 1016 | 5 × 1016 |
Nd (cm−3) | 2 × 1019 | 2 × 1019 | 2 × 1016 | 10 | 10 | 10 |
ve (cm s−1) | 1 × 107 | 1 × 107 | 1 × 107 | 1 × 107 | 1 × 107 | 1 × 107 |
vh (cm s−1) | 1 × 107 | 1 × 107 | 1 × 107 | 1 × 107 | 1 × 107 | 1 × 107 |
d (nm) | 50 | 50 | 450 | 200 | 200 | 200 |
Nt (cm−3) | 1 × 1015 | 1 × 1014 | 2.5 × 1013 | 1 × 1014 | 1 × 1014 | 1 × 1014 |
To account for interface recombination, two Interface Defect Layers (IDLs) are introduced between TiO2/perovskite and MASnI3/HTL. These IDLs share identical parameters with MASnI3 for consistency.
To achieve a carrier lifetime of 1 ns in the absorber layer, a defect density of 2.5 × 1013 cm−3 is presumed, aligning with the theoretical range of 1 ns to 4 ns. Table 2 compiles information on interface defects and absorber defects. Front optical filtration employs Transmission Solar Glass, and the back contact metal work function is set to 5.3 eV.33
Parameters | MASnI3 | TiO2/IDL | IDL2/HTL |
---|---|---|---|
Defect type | Neutral | Neutral | Neutral |
Capture cross section for electrons/holes (/cm2) | 2 × 10−14 | 1 × 10−18 | 1 × 10−18 |
2 × 10−14 | 1 × 10−17 | 1 × 10−19 | |
Energetic distribution | Gaussian | Single | Single |
Energy level w.r.t Ev (above Ev, eV) | 0.65 | 0.07 | 0.32 |
Characteristics energy/eV | 0.1 | — | — |
Total density/cm−3 | Variable | 1 × 109 | 1 × 109 |
The band gap (Eg) and electron affinity (χ) of the CZTSe1−xSx HTL are adjusted according to the sulfur content (x). Previous studies indicated that the Eg values for CZTS (x = 1) and CZTSe (x = 0) were 1.5 and 0.95 eV, respectively. The determination of Eg and χ for CZTSe1−xSx at various sulfur concentrations (x) is accomplished using equations from the literature.28
In addition to the reference cell, the study explores CZTSe and CZTS as HTLs, keeping all other parameters constant except for those related to the HTL. The simulated J–V properties with CZTSe HTL show a Voc of 0.52 V, Jsc of 39.88 mA cm−2, FF of 66.28%, and an efficiency of 13.95%. Notably, all efficiency parameters in this simulation are lower than those with Spiro-OMeTAD HTL in tin-based perovskite solar cells, yet these results align with previously reported experimental outcomes. The simulations are conducted with a thickness of 200 nm and a total defect density of 5 × 1016 cm−3 for both CZTSe and CZTS.
Our simulation reveals that CZTSe, as an HTL, demonstrates a power conversion efficiency comparable to Spiro-OMeTAD in tin-based PSCs. This similarity is attributed to the approximately aligned band structures of MASnI3/Spiro-OMeTAD and MASnI3/CZTSe interfaces.
The CZTSe alloys possess the capability of a tunable energy band gap. Therefore, employing a configuration engineering method allows for the investigation of the CZTSe1−xSx compound and the optimization of band alignment at interfaces, consequently influencing PCE in tin-based devices. The changes in performance of the IV parameters with different sulfur concentrations in CZTSe1−xSx as the Hole Transport Layer (HTL) is depicted in Fig. 3.
![]() | ||
Fig. 3 The variation in the performance of the IV parameters with the different S concentration in CZTSe1−xSx as HTL in structure. |
Tin-based PSCs with CZTSe as the HTL exhibit relatively lower performance IV parameters. With the addition of a small fraction of S to CZTSe, Jsc experiences a slight decrease, followed by a more substantial decrease. As the S concentration increases up to approximately 0.6, Voc, Jsc, and η show an increase, while FF initially decreases before starting to increase again, as shown in Fig. 3(c). Beyond x = 0.7, a significant decrease in all parameters is observed. Therefore, optimized performance is observed at x = 0.6.
At the adjusted of the sulfur content, the tin-based PSCs demonstrate a Voc of 0.79 V, Jsc of 28.20 mA cm−2, FF of 76.71%, and PCE of 17.29%. This efficiency is in correspondence with Spiro-OMeTAD-based lead-free PSCs.
The S/(S + Se) ratio serves to control the band gap (Eg) and electron affinity (χ) of the CZTSe1−xSx Hole Transport Layer (HTL). Through composition engineering, the Conduction Band Offset (CBO) and Valence Band Offset (VBO) at the interface of MASnI3/CZTSe1−xSx can be practically manipulated. The band diagram of MASnI3 with CZTSe1−xSx HTL is presented in Fig. 4(a). To facilitate the hole charge transportation from the tin-based absorber layer to the HTL, the valence energy band level of the HTL is usually slightly lower than the conduction band level of MASnI3, leading to a reduction in the Built-in Potential (Vbi). It is related that a small negative VBO can result in a low Voc. CZTSe shows a −0.17 eV VBO and the lowest Vbi (−0.35 V), which corresponds to the smallest Voc in Fig. 3(a).
![]() | ||
Fig. 4 (a) Energy band diagram of tin based PSCs with CZTSe1−xSx HTL (b) MASnI3/HTL valence band offset. |
The VBO of CZTSe1−xSx consistently shifts to deeper energy levels, resulting in values of VBO and Vbi transitioning from negative to positive with the addition of sulfur concentration in CZTSe, as indicated in Fig. 4b and 5. A positive (+ive) VBO is most favorable for reducing charge carrier recombination.36 The complete valence energy band alignment at the MASnI3/CZTSe1−xSx interface is illustrated in Fig. 4(a and b) with a distinctly vertical transfer. The fluctuation in the valence energy band position of CZTSe1−xSx alters the Vbi in the absorber layer.
The Vbi is defined as and the VBO = χ(CZTSe1−xSx) + Eg(CZTSe1−xSx) − χ(MASnI3) − Eg(MASnI3). The values of VBO and Vbi are attained and depicted in Fig. 5. CZTS shows the maximum VBO (0.53 eV) and Vbi (0.89 V) as the VBO becomes deeper with increasing sulfur content, leading to an increase in Vbi. In our study, tin-based PSCs with CZTSe0.6S0.4 or CZTSe0.4S0.6 HTLs show a +ive VBO of 0.08 eV and 0.22 eV, respectively. From Fig. 5, it can be inferred that tin-based PSC with CZTSe0.4S0.6 HTL demonstrates sufficient Vbi, which significantly participate to the larger Voc examined in Fig. 3(a).
The overall rate of generation and recombination in tin-based PSCs with CZTSe1−xSx is depicted in Fig. 6. With a thickness of 450 nm for MASnI3, the majority of incident light is absorbed, allowing only a small portion to be transmitted, given its large range of the absorption co-efficient (104–105 cm−3). As seen in Fig. 6(a), carrier generation primarily appears in the MASnI3 layer. The suitable band gap of CZTSe1−xSx in the range of 0.95–1.5 eV impacts the generation rate and recombination rate of the charge carriers for the transmitted light.
When CZTSe1−xSx is employed as the Hole Transport Layer (HTL), maximum charge carrier generation was examined in CZTSe. The charge carrier generation rate decreases as the sulfur content increases, attributed to the widening band gap. The interface of MASnI3/CZTSe1−xSx, a potential barrier hinders the motion of electrons from the HTL to MASnI3 due to the Conduction Band Offset (CBO). Simultaneously, the Vbi in the perovskite layer propels photon-generated majority charge carriers toward the CZTSe1−xSx, facilitating easy recombination with already available minority carriers in the CZTSe1−xSx.37
Hence, the total recombination rate in CZTSe1−xSx is noted to be less than Spiro-OMeTAD. For certain concentrations, the recombination rate is particularly small due to the reduced accessibility of the recombined electrons, driven by the high sulfur concentration and an appropriate Valence Band Offset (VBO). Among the CZTSe1−xSx compositions, CZTSe0.4S 0.6 exhibits a slightly lower total recombination rate.
The defect density (Nt) of MASnI3 is adjusted to 2.5 × 1013 cm−3, aligning with the carrier diffusion length of 0.9 μm based on earlier numerical simulated investigations of prime structure. To investigate the impact of Nt further, we varied Nt from 1012–1017 cm−3 and illustrated the change of I–V curves with Nt in Fig. 7. The cell demonstrates a substantial performance improvement with a decrease in Nt in tin-based perovskite, consistent with simulations of lead-based perovskite.38 A lower Nt results in better photovoltaic performance, achieving Voc of 0.80 V, Jsc of 29.13 mA cm−2, FF of 78.31%, and PCE of 18.29%.
Experimental literature indicates that Sn-based PSCs exhibits favorable charge-transport characteristics. To delve deeper into the effect of Nt on the device, we examine the impact of Nt on the charge carrier diffusion length (L) in Fig. 8, based on and
, where μ represents carrier mobility and is used to determine the diffusion length of carriers. Lower Nt values correspond to longer diffusion lengths (L), which contributes to the enhancement of cell performance.
Considering the influence of Nt and ln(lp), the devices performance parameters are optimum when Nt is as low as 2.491 × 1013 cm−3 (resulting in a ln(lp) of 0.9 μm), and the absorber layer thickness is 450 nm. This substantial performance improvement is attributed to the increased ln(lp) associated with the reduction in Nt. Table 3 provides a comparative summary of MASnI3-based perovskite solar cells reported in the literature along with the current simulated device. In this study, a PCE of 18.29% was achieved for the MASnI3/CuZnSn(Se1−xSx) heterostructure, demonstrating the role of optimized absorber thickness, defect passivation, and favorable band alignment through interface engineering. Notably, an experimental absorption model was employed in our simulation, resulting in more realistic and experimentally relevant efficiency values. In contrast, many reported studies rely on idealized absorption assumptions, often overestimating performance. Although some literature devices show higher PCEs (e.g., 25.05% with CuI HTL, 24.28% using BaTiO3/CuO interfaces), they typically incorporate more complex or less stable components. Our proposed structure offers a cost-effective, stable, and environmentally friendly architecture, providing a promising foundation for experimental fabrication and further optimization of high-efficiency, lead-free perovskite solar cells.
This journal is © The Royal Society of Chemistry 2025 |