Quasi-ohmic electron extraction in perovskite solar cells through controlled burstein–moss shifts in rare-earth doped SnO₂ ETLs

Abstract

Perovskite solar cells (PSCs) have rapidly evolved into next-generation photovoltaic devices because of their fascinating power conversion efficiencies and low manufacturing costs. However, achieving peak performance requires precise engineering of the electron transport layer (ETL) to optimise charge extraction and suppress recombination. Doping rare-earth (RE) into SnO₂ is one potential way to enhance ETL properties and has been extensively studied¹. In this study, we utilize SCAPS-1D simulations to investigate the impact of rare-earth (RE) doping (La, Ce, and Eu) in SnO₂ ETLs within a FAPbI₃-based PSC architecture. While variations in thickness and donor density of the perovskite layer were examined, our results reveal that device performance is primarily governed by bandgap engineering induced by the Burstein–Moss effect. This doping-induced bandgap widening shifts the conduction band edge, facilitating a quasi-ohmic contact and superior band alignment with the FAPbI₃ absorber. At high defect densities and thicker absorber layer, however, a performance crossover was observed, highlighting the sensitivity of optimized interfaces to severe Fermi-level pinning. Motivated by this findings, we systematically optimized the perovskite absorber thickness and defect density (Nt), identifying a critical sweet spot at 0.6 μm and 10¹⁴ cm^-3, respectively, where photon harvesting and bulk recombination are balanced. Notably, La-doped SnO₂ exhibited the highest tolerance to bulk defects, maintaining superior Voc and FF through enhanced extraction kinetics that outrun trap-assisted recombination. Under optimized conditions, the La-doped device emerged as the champion configuration, significantly outperforming pristine device and marginally better than Ce-, and Eu-doped variants. These findings provide a fundamental framework for leveraging RE-doping to engineer interfacial energy levels, offering a clear pathway for the development of high-efficiency, defect-tolerant perovskite photovoltaics.

Supplementary Information

The online version contains supplementary material available at 10.1038/s41598-026-51783-6.

Introduction

Lead halide perovskites are semiconductor materials with the formula of ABX₃, where A denotes a monovalent cation (for example, methylammonium (MA⁺); formamidinium (FA⁺); or cesium (Cs⁺)), B is typically a divalent metal cation (Pb²⁺ or Sn²⁺), and X represents a halide anion (I⁻, Br⁻, or Cl⁻). These material exhibits exceptional optoelectronic properties; including strong optical absorption, ambipolar charge transport, long carrier diffusion lengths, and high charge mobility. Moreover, they can be synthesized through low temperature, solution-based processes, making them highly cost-effective photovoltaic cell². Perovskite Solar Cells (PSCs) evolved from the dye-sensitized solar cell concepts, where an organic dye molecule is replaced by hybrid perovskite materials, like MAPbBr₃ or MAPbI₃, achieving an initial power conversion efficiency (PCE) around 3.8%³. However, the use of liquid electrolyte led to serious stability issues due to perovskite degradation. Kim et al. achieved a stable PSCs with significantly higher efficiency by replacing the liquid electrolyte with a solid-state hole transport layer (spiro-OMeTAD) and significantly improving the stability and the efficiency⁴. Since then, PSCs have been progressing rapidly in solar cell research.

In a typical solar cell, the perovskite absorber layer (PVK) is sandwiched between a reflecting or opaque back electrode (such as Au or Ag) and a transparent conducting front electrode (often ITO or FTO) in a typical PSC structure. Specialized electron transport layers (ETLs) and hole transport layers (HTLs) are incorporated to efficiently extract charge carriers, while minimizing recombination. To improve contact between the perovskite and transport layers, a mesoporous scaffold are incorporated to increase the efficiency from single-digit values to over 26%⁵, placing them on par with silicon-based solar cells. Perovskite solar cells have exhibited remarkable progress in recent years, with certified single-junction efficiencies now approaching ~ 27% under standard test conditions⁶.Their tunable bandgap, excellent absorption coefficient, and mechanical flexibility further enable emerging applications like portable electronics, tandem solar cell and building integrated photovoltaic system^7–9.

To address efficiency and stability challenges, extensive research efforts have focused on strategies such as interfacial engineering, doping and compositional tuning of transport layers and the perovskite absorbers, defect passivation, bandgap engineering via mixed halides and cation substitution, dimensional engineering using 2D/3D hybrid structures, and device encapsulation¹⁰. Among these approaches, engineering the ETL has emerged as a particularly effective strategy for improving charge extraction and suppressing interfacial recombination. Tin oxide (SnO₂) has gained significant attention as ETL due to its wide bandgap, high electron mobility, optimum energy-level alignment with perovskite absorbers. In addition, SnO₂ -based ETLs exhibit superior chemical stability and enable low-temperature processing, making them suitable for flexible and scalable device fabrication^11,12 SnO₂-based PSCs have demonstrated PCEs as high as 18.1%¹³. Nevertheless, intrinsic limitations such as low conductivity high density of oxygen -vacancy related traps states and energy-level mismatches at the SnO₂/perovskite interface lead to carrier accumulation, interfacial recombination, and device instability. Furthermore, these surface defects hinder perovskite crystallization, resulting in smaller grain sizes and increased grain boundary density, which ultimately degrade device performance^14,15.

In this regard, Rare earth (RE) elemental doping has emerged as a promising strategy to enhance optoelectronic properties and interfacial quality of SnO₂ ETLs. Rare-earth elements, comprising the fifteen lanthanides, along with scandium (Sc) and yttrium(Y), possess unique electronic configurations with partially filled 4f orbitals and accessible 5 d states, enabling them to effectively modulate the electronic and structural properties of host materials when doped. RE ions typically exist in + 2 or + 3 oxidation state and interact strongly with surrounding anions, facilitating defect passivation, lattice stabilization and efficient charge transfer^16–18. Additionally, RE ions exhibit up-conversion and down-conversion due to their unique 4f orbital structures, which enables the conversion of sub-bandgap photons to visible light and enhances the overall light harvesting efficiency¹⁹. Recent studies have demonstrated that, incorporating RE ions such as Sm³⁺, La³⁺, Yb³⁺, Gd³⁺, and Ce³⁺ into SnO₂ greatly increases charge extraction, lowers defect density, and improves interfacial quality. For instance, Sm³⁺-doped SnO₂ ETL has been demonstrated to enhance perovskite crystal quality and passivate surface oxygen vacancies, achieving a PCE up to 24.10% while reducing hysteresis²⁰. La³⁺-modified SnO₂ improves device efficiency from 14.24% to 17.08% by suppressing recombination and enhancing surface morphology²¹. While Yb³⁺ doping has enabled all-air fabrication efficiencies exceeding 17%²².Incorporating Gd³⁺ into the SnO₂ ETL layer has also enhanced the efficiency up to 22.4% by passivating the defect and improving the energy-level alignment²³. Similarly, Ce³⁺ and other RE dopants have demonstrated improved energy alignment, reduced voltage losses, and PCEs greater than 23%. These findings consistently indicate that RE doping stabilizes the perovskite/ETL interface, reduces nonradiative recombination, and modifies the electronic structure of ETL.

A key mechanism underlying these improvements is the substitution of Sn⁺ sites with RE³⁺, which introduces shallow donor states that raise the free electron density close to the conduction band edge²⁴. As the donor concentration increases, the fermi level of the ETL layer shifts upward, filling the lower conduction band states and causing a Burstein–Moss shift (BMS) which manifests as a widening of the optical bandgap and a shift of the absorption edge toward higher photon energies. At moderate doping levels, the formation of shallow donors enhances conductivity and facilitates more efficient electron extraction across the ETL/Perovskite interface. However, when the doping levels exceeds the optimal level point, defect-induced scattering, band tailing, and oxygen vacancy clustering dominate, leading to the formation of recombination centers that deteriorate the device performance²⁵. Similar trade-offs between beneficial shallow donor formation and detrimental over-doping effects have been reported in related oxide semiconductor systems^25–27. Experimental studies have reported the presence of the Burstein–Moss shift in rare-earth-doped SnO₂ systems; however, a comprehensive understanding of the underlying performance enhancement mechanisms, particularly band gap modulation, interfacial band alignment, and recombination behavior remains insufficiently explored through numerical simulation. Optimizing parameters such as the thickness of the perovskite absorber layer is the key because they strongly influence the device performance (e.g., light absorption and charge collection). In addition, adding too much thickness will increase the amount of charge recombination occurring within the bulk of the device, and decrease the open circuit voltage, fill factor and power conversion efficiency (Voc, FF and PCE). Therefore, achieving lower defect densities will be important to achieve higher efficiencies.

In this work, a systematic investigation based on SCAPS-1D is conducted on lanthanum (La), cerium (Ce), and europium (Eu) doped SnO₂ electron transport layers within FAPbI₃-based perovskite solar cells. The study aims to elucidate how rare-earth-induced band structure modifications specifically through the Burstein Moss effect influence charge extraction, recombination dynamics, and overall power conversion efficiency. To establish a comprehensive device framework, this study explicitly correlates the optimization of the perovskite absorber (thickness and defect density) with the rare-earth-induced bandgap modifications in the SnO₂ ETLs. Subsequently, by analysing J–V characteristics and EQE spectra, we evaluate the comparative impact of La-, Ce-, and Eu-doping on interfacial charge transport and bulk recombination behaviour. In particular, the effect of quasi-ohmic electron extraction caused by favorable conduction band alignment in quickly depleting carriers from the absorber bulk and hindering recombination is investigated. This integrated approach offers a robust methodology for identifying the optimal ETL configuration to maximize the performance and defect tolerance of high-efficiency FAPbI₃ perovskite solar cells.

Device structure and simulation parameters

This work employs a theoretical device simulation to investigate the impact of the energy level alignment between the ETL and the perovskite absorber layer. Simulation of PSCs provides an invaluable means to discover much regarding the optoelectronic and physical response of solar cell structures. The Solar Cell Capacitance Simulator, or SCAPS-1D, was used in this work for carrying out electrical simulations and all simulations were performed under standard AM 1.5G solar illumination conditions with a light intensity of 1000 W/m² and with a fixed operating temperature of 300 K^28–32. The device was modelled using a standard n–i–p planar architecture, as shown in Fig. 1, with a layer stack of FTO/(doped or undoped) SnO₂/FAPbI₃/Spiro-OMeTAD/Au, where FTO serves as the transparent conducting oxide (TCO), SnO₂ functions as the electron transport layer (ETL), Spiro-OMeTAD acts as the hole transport layer (HTL), and gold serves as the top electrode.The hole transport layer was fixed to be spiro-OMeTAD throughout the simulation. Subsequently, to investigate the impact of bandgap engineering, rare earth elements (Eu, Ce, and La) are doped into ETL and the thickness and defect density of the perovskite absorber layer are systematically varied to understand the response of bulk characteristics when the barrier height across ETL is modulated. Trap states were modelled using a gaussian energy distribution, with a characteristic energy of 0.1 eV, for all defect levels. The photovoltaic performance was evaluated in terms of short-circuit current density (Jsc), open-circuit voltage (Voc), fill factor (FF), and power conversion efficiency (PCE). The material parameters used in the simulations were established through a two-step calibration strategy. Initially, experimentally reported ranges of key properties such as bandgap, electron affinity, dielectric constant, carrier mobility, and doping density for both pristine and rare-earth-modified SnO₂ were collected from literature based on characterisation techniques including ultraviolet photoelectron spectroscopy, Kelvin probe force microscopy, optical absorption, and electrical conductivity measurements^33–52 These parameters were then systematically tuned within physically realistic limits in SCAPS to obtain consistent energetic alignment and recombination behaviour at the device level. Therefore, the final parameter set represents a calibrated device model rather than an exact description of a specific experimental structure.

Fig. 1.

. Schematic diagrams of device architecture.

It should be noted that the Burstein–Moss (BM) effect is not explicitly implemented as a separate physical model in SCAPS-1D. In the present study, its influence was incorporated indirectly by adjusting the ETL donor concentration together with corresponding modifications in the effective bandgap and electron affinity. This treatment emulates the carrier-density-induced Fermi-level shift and conduction-band modulation reported for doped SnO₂ transport layers. Hence, the BM effect discussed here should be understood as an effective band-structure calibration approach used to interpret band-alignment-driven performance trends rather than a direct microscopic simulation. The bulk defect states in the perovskite absorber were modelled using a Gaussian distribution centred approximately at the mid-gap energy (~ 0.55 eV from the band edge) and the electron and hole capture cross sections were set to 10^–15 cm². This assumption represents deep trap levels that are widely reported to dominate trap-assisted recombination and voltage losses in hybrid perovskite solar cells^24,51(The table detailing the fixed parameters has been added and provided as Supporting Information (Table S1)). The simulation parameters used for each layer are summarized in Table 1.

Table 1.

The simulated parameters for each layer.

Parameters	Spiro-OMeTAD	Perovskite	SnO2	Eu: SnO2	Ce: SnO2	La: SnO2
Thickness (µm)	0.1	0.5	0.05	0.05	0.05	0.05
Bandgap (eV)	3.2	1.47	4.05	4.03	4.00	3.95
Electron Affinity (eV)	2.1	4	4.50	4.20	4.10	4.00
Dielectric permittivity	3	6.6	9	11	10	10
CB effective density of states (cm− 3)	2.5 × 1018	1.2 × 1019	2.2 × 1019	2.2 × 1019	2.2 × 1019	2.2 × 1019
VB effective density of states (cm− 3)	1.8 × 1019	2.9 × 1018	1.8 × 1018	1.5 × 1018	1.5 × 1018	1.5 × 1018
Electron mobility (cm2/V.s)	2 × 10− 4	2.7	200	200	200	200
Hole mobility (cm/V.s)	2 × 10− 4	1.8	50	50	50	50
Shallow donor density ND (cm− 3)	0	1 × 1016	1 × 1018	1 × 1018	1 × 1018	1 × 1018
Shallow acceptor density NA (cm− 3)	1 × 1014	1 × 1016	0	0	0	0

Results and discussions

SCAPS-1D were performed to systematically investigate the influence of rare-earth (RE) doping in SnO₂ electron transport layer on the band structure, charge transport and overall photovoltaic performance of FAPbI₃-based PSC. To implement bandgap engineering, Eu, Ce, and La dopants were introduced into the lattice, modulating key parameters such as bandgap energy and electron affinity. This incorporation modifies the electronic density of states and reduces defect density, leading to enhanced electron extraction and improved conductivity within the ETL.

The charge transport behaviour of the ETLs is directly reflected in the spatial distribution of the recombination rate across the device thickness. Effective extraction of electrons from the perovskite into the ETL significantly reduces carrier accumulation within the absorber bulk, leading to a suppression of bulk recombination (Fig. 2a). Compared to pure SnO₂, the RE-doped SnO₂ ETLs show reduced recombination rates, suggesting both improved electron selectivity and accelerated extraction kinetics. Specifically, La-doped SnO₂ exhibits the lowest bulk recombination due to its favourable conduction band alignment with the perovskite (Fig. 2b). This quasi-ohmic interface facilitates seamless electron extraction and minimizes carrier pile-up at the junction. In contrast, the less ideal band alignment in Ce- and Eu-doped SnO₂ hinders charge carrier separation, leading to higher carrier accumulation and consequently increased recombination levels. Ultimately, these profiles indicate that La-doping most effectively suppresses bulk recombination, resulting in superior device performance. To further understand the carrier transport and to confirm the Quasi-ohmic electron extraction behaviour, the energy band alignment and Fermi level distribution at the ETL/perovskite interface were analysed (Fig. 3). The simulated band alignment at the ETL/perovskite interface shows a gradual change in the conduction-band offset (CBO) with rare-earth doping of SnO₂. In the device employing undoped SnO₂, the ETL conduction band lies significantly below that of the perovskite, resulting in a large cliff-type offset as ΔEc ≈ − 0.50 eV (Fig. 3a). This strong negative offset enhances carrier accumulation and interfacial recombination losses, as reported in previous studies⁵².

Fig. 2.

(a) Recombination rate as a function of device thickness for pristine and RE-doped SnO₂ (La, Ce, and Eu) electron transport layers and (b) energy levels of each functional layer in the PSCs.

Fig. 3.

Simulated operating-condition energy band alignment at the ETL/perovskite interface for (a) undoped SnO₂, (b) Eu-doped SnO₂, (c) Ce-doped SnO₂, and (d) La-doped SnO₂ electron transport layers.

With Eu and Ce doping as shown in Fig. 3b and c, the conduction band gradually shifts upward, reducing the offset to moderate (ΔEc ≈ − 0.20 eV) and small (ΔEc ≈ − 0.10 eV) cliff alignments, respectively. This near-flat band configuration weakens band bending and supports smoother electron transport across the interface. In contrast, La-doped SnO₂ exhibits an almost matched or slightly higher conduction-band edge, producing a flat to small positive spike as Ec ≈ 0 eV (Fig. 3d). Such alignment promotes quasi-ohmic electron extraction by minimising the transport barrier while suppressing back-carrier recombination, consistent with reported numerical optimisation trends⁵³.

The performance trends relative to increasing perovskite defect density (Nt), as illustrated in Fig. 4, closely mirror the observations from the absorber thickness study. In the SCAPS-1D framework, Nt represents the bulk trap states, and its increase directly triggers trap-assisted (Shockley-Read-Hall) recombination. As the density of these recombination centres rises, carrier lifetimes are significantly curtailed, leading to a systematic decline in Voc, Jsc, FF, and PCE (Figs. 4a–d). At low to moderate Nt values, La-doped SnO₂ consistently outperforms all other variants. This is attributed to its optimal conduction band alignment and superior electron extraction kinetics. By rapidly removing electrons from the perovskite bulk, the La-doped interface effectively “outruns” the recombination process, mitigating the impact of traps and maintaining higher Voc and FF. A noticeable crossover in device performance is observed at very high absorber defect densities (Nt > 10¹⁷ cm⁻³), where the photovoltaic parameters of La-doped devices become slightly lower than those of the pristine SnO₂-based device. In this regime, enhanced trap-assisted recombination significantly reduces carrier lifetime and limits the effective depletion width within the perovskite absorber. The resulting weakening of the built-in electric field lowers the carrier-extraction driving force and increases bulk recombination losses before photogenerated carriers can reach the transport-layer interfaces, and at low defect density (Nt = 10¹⁴ cm⁻³), the field is uniformly distributed across the absorber, indicating efficient drift-assisted transport. Consequently, the performance advantage associated with improved band alignment and carrier selectivity in the doped ETLs becomes less pronounced, leading to the observed convergence and eventual crossover in device efficiency. The corresponding defect-dependent device performance (Figure S1) is provided in the Supporting Information. This behaviour indicates a transition from interface-influenced charge transport to bulk-recombination-limited device operation at elevated defect densities which is consistent with the band alignment shown in the Fig. 3.

Fig. 4.

. Variation of (a) Voc, (b) Jsc, (c) FF, and (d) PCE as a function of defect density of the perovskite layer for the doped and undoped ETL.

Effect of the RE-doped SnO₂ on the absorber layer thickness

As shown in Fig. 5a, Voc decreases across all devices as the absorber thickness increases. This trend is attributed to increased bulk recombination, as charge carriers must travel longer distances to reach their respective transport layers, increasing the probability of recombination events. Notably, devices with RE-doped SnO₂ exhibit a steeper decline in Voc compared to pristine SnO₂, indicating a higher sensitivity to thickness-induced recombination. The steeper decline in Voc observed for RE-doped SnO₂ devices suggests a higher sensitivity to the carrier collection-recombination competition. While RE-doping optimizes the interface for superior extraction, the resulting high Voc is more susceptible to degradation when the absorber thickness exceeds the effective carrier diffusion length, leading to a more rapid increase in the bulk recombination rate compared to pristine devices. As seen in Fig. 5b, Jsc initially increases with absorber thickness (0.6–0.8 μm) due to enhanced light harvesting and higher photogeneration rates. At the optimal thickness, Eu-doped SnO₂ produces a slightly higher Jsc compared to La- and Ce-doped variants, likely due to its higher electronic conductivity which promotes efficient electron extraction. Beyond this optimal point, however, Jsc begins to decline as bulk recombination becomes the dominant factor.

Fig. 5.

Variation of (a) Voc, (b) Jsc, (c) FF, and (d) PCE as a function of thickness of the perovskite layer for the doped and undoped ETL with perovskite Nt as 10¹⁴ cm⁻³.

The FF consistently decreases with increasing thickness (Fig. 5c). Interestingly, while RE-doped devices exhibit superior FF at lower absorber thicknesses due to reduced interfacial charge-transfer resistance, they show a more pronounced decline as thickness increases, eventually falling below pristine devices. This crossover suggests that while RE-doping optimizes the interface, the resulting high-efficiency architecture is more susceptible to space-charge accumulation and increased internal series resistance when the absorber thickness exceeds the optimal carrier diffusion length. Figure 5d reveals an optimal intermediate thickness of ∼0.6 μm, where the trade-off between absorption enhancement and recombination loss is balanced. While Eu-doping offers localized Jsc advantages, La-doped SnO₂ achieves the highest peak PCE. This confirms that the superior Voc and FF provided by the favourable band alignment of La-doped SnO₂ ultimately govern the overall device efficiency.

After optimizing the thickness and the density of defects of the perovskite absorber layer, we independently optimized the electron transport layer by varying its parameters while maintaining the conditions of the absorber as constant. The optimized configurations were characterized via current density–voltage (J–V) and External Quantum Efficiency (EQE) spectra to elucidate the ETL’s role in charge transport, recombination suppression, and overall efficiency. Figure 6 compares the simulated (a) J–V characteristics and (b) QE spectra of PSCs employing pristine and RE-doped SnO₂ (La, Ce, and Eu) ETLs. As summarized in Table 2, devices incorporating RE-doped ETLs exhibit enhanced Voc, Jsc, and FF compared to the pristine SnO₂ reference, indicating significantly improved charge extraction. Among the dopants, La-doped SnO₂ demonstrates the superior overall photovoltaic performance, followed by Ce- and Eu-doped variants. The superior performance of La-doped SnO₂ is attributed to favourable band alignment and increased donor concentration. These factors induce a controlled Burstein–Moss effect, where the filling of the lower states in the conduction band leads to an upward shift of the conduction band edge and a widening of the optical bandgap. This band engineering suppresses interfacial carrier accumulation and enables quasi-ohmic electron transport at the ETL/perovskite interface. In contrast, while Ce- and Eu-doping also trigger Burstein–Moss band filling, their modifications are less effective. This is likely due to the mixed valence states of Ce and Eu, which can cause conduction band distortion and increased carrier scattering. These localized disruptions increase recombination probability, which is reflected in the lower Voc, reduced FF, and diminished short-wavelength EQE response observed for Ce- and Eu-doped devices. Ultimately, La-doping provides the most effective balance of bandgap engineering and recombination suppression, resulting in the champion photovoltaic performance. The slightly improved performance of La-doped SnO₂ can be attributed to its stable + 3 oxidation state and larger ionic radius, which promote effective substitution and passivation of oxygen-vacancy defects, thereby reducing recombination. In contrast, Ce and Eu exhibit mixed valence states that can introduce localized states and increase carrier scattering. As a result, La doping causes less perturbation to the Sn–O conduction network, leading to more efficient charge extraction.

Fig. 6.

Comparison between (a) J-V Curve and (b) QE Curve of simulated PSCs using different ETLs: SnO₂, Eu-doped SnO₂, Ce-doped SnO₂, La-doped SnO₂.

Table 2.

Comparison of photovoltaic performance parameters between simulated and experimentally reported devices.

Device structure	Voc (V)	Jsc (mA/cm2)	FF (%)	PCE (%)	Reference
FTO/SnO2/FAPbI3	0.84	24.18	81.95	16.78	This work
FTO/SnO2/MAPbI3	1.06	20.67	65	14.24	Experimental21
FTO/La: SnO2/FAPbI3	1.07	24.18	85.20	22.04	This work
FTO/La: SnO2/MAPbI3	1.09	21.77	72	17.08	Experimental21
FTO/Ce: SnO2/FAPbI3	1.07	24.18	85.15	22.02	This work
FTO/Ce: SnO2/FA₁₋ₓMAₓPb(I, Br)₃	1.08	21.98	66.4	15.77	Experimental54
FTO/Eu: SnO2/FAPbI3	1.06	24.18	84.76	21.87	This work
FTO/Eu: SnO2/FA₁₋ₓMAₓPbI₃	1.17	25.62	80.16	24.11	Experimental33

A quantitative consistency check was performed by estimating the short-circuit current density through integration of the EQE spectra under the AM1.5G photon flux. The integrated current values for the pristine and RE-doped SnO₂-based devices were found to be approximately 23.9–24.2 mA cm^− 2, which agrees well with the simulated Jsc value of about 24.18 mA cm^− 2 obtained from the J–V characteristics. The small deviation (< 2%) confirms the self-consistency of the optical generation profile and carrier transport modelling employed in the simulation.

Conclusion

In this study, a systematic numerical investigation was conducted using SCAPS-1D to optimize FAPbI₃-based perovskite solar cells through rare-earth (RE) doping of the SnO₂ electron transport layer. Our findings demonstrate that while various ETL parameters influence performance, the enhancement is primarily driven by bandgap engineering mediated by the Burstein–Moss effect. The incorporation of La, Ce, and Eu dopants induces a controlled upward shift of the conduction band edge, facilitating a quasi-ohmic interface and significantly reducing interfacial recombination. The integration of ETL optimization with absorber layer tuning revealed a critical performance “sweet spot” at an absorber thickness of 0.6 μm and a defect density of 10¹⁴ cm⁻³. Notably, La-doped SnO₂ emerged as the superior ETL configuration, exhibiting not only the highest power conversion efficiency but also a remarkable tolerance to bulk defects. This resilience is attributed to enhanced electron extraction kinetics that effectively outrun trap-assisted recombination. However, the observed performance crossover at extreme defect densities (Nt > 10¹⁶ cm⁻³) underscores the sensitivity of highly optimized interfaces to severe Fermi-level pinning, providing a boundary condition for the efficacy of RE-doping. Ultimately, this research provides a robust design framework for implementing RE-doped SnO₂ in high-efficiency photovoltaics. By precisely tailoring interfacial energy levels and optimizing absorber dimensions, we demonstrate a clear pathway toward developing defect-tolerant, high-performance perovskite solar cells. These results suggest that RE-doping techniques hold significant potential for the next generation of stable and efficient photovoltaic implementations.

Supplementary Information

Below is the link to the electronic supplementary material.

Author contributions

Praveena drafted the manuscript and conducted formal analysis and investigation. Poonam contributed to methodology, Rahul Narasimhan A. supervised the work and contributed to validation and critical manuscript review.

Funding

Open access funding provided by Vellore Institute of Technology.

Data availability

The publication includes data to support the findings. Supporting information can be provided upon request.

Declarations

Competing interests

The authors declare no competing interests.