Solvent Engineering and Molecular Doping Synergistically Boost CsPbIBr₂ Solar Cell Efficiency

Abstract

Perovskite solar cells have garnered significant attention due to their outstanding optoelectronic properties, ease of fabrication, and cost-effectiveness, making them a promising candidate for next-generation photovoltaic technologies. However, CsPbIBr₂-based perovskites currently face critical challenges regarding their limited efficiency and relatively poor long-term stability, hindering their broader commercial applications. In this study, we systematically investigated the morphological effects induced by different solvents, including dimethylformamide (DMF), N-methyl-2-pyrrolidone (NMP), and dimethyl sulfoxide (DMSO), on the formation and characteristics of lead bromide (PbBr₂) complexes. Further optimization was achieved through the innovative incorporation of trimesoyl chloride (TMC) doping into the perovskite precursor solution. The optimized precursor solution was subsequently processed using a spin-coating and annealing method, resulting in high-quality CsPbIBr₂ perovskite thin films with improved morphological and optoelectronic properties. The experimental results demonstrated a remarkable enhancement in power conversion efficiency (PCE), with an increase from an initial value of 6.2% up to 10.2%. Furthermore, the optimized CsPbIBr₂ solar cells exhibited excellent stability, maintaining over 80% of their initial efficiency after continuous aging for 250 h in ambient air conditions. This study presents an effective strategy for the controlled morphological and compositional engineering of wide-bandgap perovskite materials, providing a significant step forward in the advancement of perovskite photovoltaic technology.

1. Introduction

The escalating global energy crisis, characterized by rapidly increasing energy consumption, the depletion of fossil fuels, and severe environmental pollution, has made the transition toward sustainable and renewable energy sources an urgent global priority [1,2,3,4,5,6,7]. Solar energy emerges as one of the most promising renewable energy alternatives, offering abundant availability, inexhaustibility, and minimal environmental impact compared to conventional fossil-based energy generation [8,9]. Photovoltaic (PV) technology, which converts solar energy directly into electricity, is considered an essential component in addressing global energy demands and mitigating climate change. Among the diverse photovoltaic technologies available, perovskite solar cells (PSCs) have recently attracted significant scientific and industrial attention due to their superior optoelectronic properties, simple fabrication processes, low cost, and rapidly improving power conversion efficiencies (PCEs) [10,11]. In particular, all-inorganic perovskites, such as CsPbIBr₂, are promising due to their excellent thermal stability and potential resistance against moisture degradation compared to organic–inorganic hybrid counterparts [12,13]. Despite these advantages, CsPbIBr₂-based perovskite solar cells still suffer from limited photovoltaic performance, typically lower efficiencies, and insufficient long-term stability in ambient conditions, hindering their large-scale commercialization and practical applications [14,15,16].

To overcome these limitations, significant research efforts have focused on optimizing perovskite film quality, morphology, and compositional uniformity [17,18,19]. Solvent engineering has emerged as an effective strategy to modulate crystallization kinetics, control grain size and orientation, reduce defect density, and ultimately enhance photovoltaic performance [20,21,22]. Specifically, polar solvents such as dimethylformamide (DMF), N-methyl-2-pyrrolidone (NMP), and dimethyl sulfoxide (DMSO) exhibit unique interactions with perovskite precursors, forming intermediate complexes with lead bromide (PbBr₂) that significantly influence the resultant perovskite morphology. However, detailed comparative studies investigating the effects of these solvents on PbBr₂ complexation and resultant perovskite film properties remain relatively scarce. In addition to solvent engineering, molecular doping techniques have emerged as another promising strategy for defect passivation and film quality improvement in perovskite photovoltaics [23,24,25]. Recent research has indicated that incorporating functional molecular additives can effectively suppress defect states, improve carrier transport, and enhance device stability under operational conditions [26,27]. Nevertheless, systematic investigations combining solvent engineering with molecular doping strategies, particularly for CsPbIBr₂ perovskites, are lacking, and their synergistic effects on film morphology, efficiency, and device stability are yet to be comprehensively elucidated.

In this study, we systematically investigate the role of solvent choice—specifically DMF, NMP, and DMSO—in controlling PbBr₂ complexation and CsPbIBr₂ perovskite morphology. Additionally, we introduce trimesoyl chloride (TMC) as an innovative molecular dopant into the perovskite precursor solution to further optimize film quality. Through a carefully controlled spin-coating and annealing fabrication process, we demonstrate that our integrated solvent engineering and doping strategy notably enhances photovoltaic performance, increasing the CsPbIBr₂ solar cell efficiency from an initial 6.2% to 10.2%. Furthermore, our optimized devices exhibit remarkable ambient stability, retaining over 80% of their initial efficiency after 250 h of continuous exposure to ambient air conditions. These findings provide valuable insights into the morphological and compositional engineering of inorganic perovskite solar cells, potentially accelerating the development of efficient, stable, and commercially viable photovoltaic devices.

2. Result and Discussion

To investigate the impact of solvent-mediated complexation on the morphology and optoelectronic properties of perovskite films, three distinct PbBr₂–solvent complexes were systematically studied. Figure 1a–d show the photographs and corresponding molecular structures of pristine PbBr₂ and PbBr₂ complexes formed with the solvents dimethyl sulfoxide (DMSO), dimethylformamide (DMF), and N-methyl-2-pyrrolidone (NMP). The formation of PbBr₂–solvent complexes was confirmed by Fourier-transform infrared spectroscopy (FTIR), as shown in Figure 1e. Characteristic vibrational peaks appear for each complex, indicating successful coordination between PbBr₂ and the respective solvent molecules. For PbBr₂·DMSO, a prominent S=O stretching vibration is observed at 1024 cm⁻¹, confirming the interaction between DMSO and Pb²⁺. The PbBr₂·DMF complex exhibits a strong C=O stretching vibration at 1672 cm⁻¹, while PbBr₂·NMP shows an amide I stretching band at 1662 cm⁻¹, both of which are characteristic of coordination between Pb²⁺ and the carbonyl or amide groups. These solvent-dependent vibrational features verify complex formation, which significantly influences the crystallization dynamics of the perovskite films. The X-ray diffraction (XRD) patterns of the PbBr₂–solvent complexes (Figure 1f) show distinct crystallographic differences depending on the solvent environment. The PbBr₂·DMSO complex presents a broad peak centered at 12.2° (2θ), indicating poor crystallinity. In contrast, PbBr₂·DMF exhibits a slightly sharper peak at 12.7°, and PbBr₂·NMP reveals a well-defined peak at 13.1°, corresponding to an enhanced long-range order and crystalline packing. These differences highlight the role of solvent–PbBr₂ interactions in modulating the structure of intermediate complexes, which ultimately affect perovskite film formation. Figure 1g presents the XRD patterns of the resulting CsPbIBr₂ perovskite films derived from each complex after spin-coating and annealing. The PbBr₂·NMP-derived perovskite exhibits significantly enhanced crystallinity with sharper, more intense diffraction peaks, suggesting superior crystal quality compared to those derived from DMSO and DMF complexes. This observation emphasizes the critical role of solvent–PbBr₂ interactions in regulating perovskite film morphology and crystallinity. Additionally, the optical absorption spectra of CsPbIBr₂ films fabricated using these different solvent intermediates are displayed in Figure 1h. It is clearly observed that CsPbIBr₂·NMP-derived perovskite films exhibit the highest absorption intensity across the visible spectrum, which is indicative of improved optical quality. This enhanced absorption capability is likely associated with improved photovoltaic performance, suggesting that NMP-based complexation facilitates perovskite crystallization, grain size optimization, and film uniformity. XRD patterns clearly illustrate the fact that the perovskite films derived from PbBr₂·NMP exhibit sharper and more intense diffraction peaks, indicating improved crystallinity and larger grain sizes compared to films derived from PbBr₂·DMF and PbBr₂·DMSO. Scanning electron microscopy (SEM) morphological analysis (Figure S1) showed that NMP-based composites resulted in films with larger particles and better morphological uniformity.

The detailed fabrication process of CsPbIBr₂ perovskite solar cells and the molecular doping methodology are illustrated in Figure 2a. Initially, a carefully prepared precursor solution, consisting of CsI- and PbBr₂-based solvent complexes (PbBr₂·DMSO, PbBr₂·DMF, or PbBr₂·NMP), was uniformly deposited onto the substrate by spin-coating. The use of solvent complexes facilitates controlled crystallization kinetics and improved film morphology by modulating the intermediate states formed during the deposition stage. Following the initial precursor deposition, a second spin-coating step introduced TMC as a dopant into the precursor layer. This doping stage enables controlled nucleation and promotes homogeneous crystallization during subsequent thermal annealing. The optimized annealing procedure effectively transforms the precursor into high-quality CsPbIBr₂ perovskite thin films characterized by uniform surface morphology, large grain sizes (about 200 nm), and reduced grain boundaries, which are attributes critical for enhancing photovoltaic performance. Figure 2b provides a detailed illustration of the chemical structure and passivation mechanism of TMC within the CsPbIBr₂ perovskite lattice. The TMC molecule possesses three reactive acyl chloride groups symmetrically arranged around a central benzene ring, which serve as effective binding sites for defect passivation. Specifically, these functional groups interact strongly with under-coordinated lead ions (Pb²⁺) present at grain boundaries or surface defects within the perovskite films. Through the robust Lewis acid-base coordination interactions, the TMC molecules effectively neutralize deep-level defect states, minimizing carrier recombination pathways and significantly enhancing charge-carrier lifetimes. Furthermore, the inclusion of TMC molecules not only reduces intrinsic defect densities but also promotes improved interfacial adhesion and stability of perovskite grains. As illustrated, TMC molecules preferentially anchor at grain boundaries, bridging adjacent perovskite crystals. Such bridging interactions contribute to grain-boundary stabilization and reduce ion migration pathways, thus substantially improving both optoelectronic performance and the long-term stability of solar cells.

To elucidate the effect of the TMC doping concentration on the morphology of CsPbIBr₂ perovskite films, SEM analysis was conducted, as presented in Figure 3a–e. The films were prepared with varying TMC concentrations: pristine (undoped), 1 mg·mL⁻¹, 3 mg·mL⁻¹, 5 mg·mL⁻¹, and 10 mg·mL⁻¹. The pristine perovskite film (Figure 3a) exhibits a small grain size (about 210 nm) and a considerable grain-boundary density, indicative of suboptimal crystallization. Upon the introduction of TMC at a relatively low doping concentration (1 mg·mL⁻¹, Figure 3b), an observable improvement in grain uniformity occurs, yet grain boundaries remain distinct. Increasing the doping concentration to 3 mg·mL⁻¹ (Figure 3c) markedly enhances the perovskite film quality, resulting in significantly larger grain sizes (about 400 nm), reduced grain boundaries, and a highly uniform morphology, indicating optimal crystallization and film quality. However, when further increasing TMC concentrations beyond the optimal level, such as to 5 mg·mL⁻¹ (Figure 3d) and 10 mg·mL⁻¹ (Figure 3e), perovskite morphology deteriorates, exhibiting irregular, non-uniform grains with increased surface roughness and noticeable secondary phases. This degradation in morphology at higher doping concentrations may result from excessive molecular aggregation or disruption in the perovskite lattice formation process. Contact angle measurements (as shown by the insets of Figure 3a–e) reveal a corresponding trend with doping concentrations. The pristine film exhibits a relatively low contact angle (27.9°), indicative of poor surface hydrophobicity. The incorporation of TMC progressively enhances hydrophobicity, reaching a maximum contact angle of 67.8° at 3 mg·mL⁻¹ doping, which is consistent with a well-passivated surface with reduced defect density. At higher doping levels, the contact angles slightly decrease again, confirming the morphological deterioration seen in SEM analysis. The statistical analysis of grain sizes (Figure 3f) clearly highlights the optimal grain size and uniformity achieved at a 3 mg·mL⁻¹ doping concentration, quantitatively confirming SEM observations. The optimized grain size distribution at this concentration aligns well with improved photovoltaic performance, as larger and more uniform grains typically correlate with reduced defect densities and enhanced charge transport properties. To further investigate the influence of the TMC doping concentration on crystallinity and optical properties, XRD, UV–Vis absorption, and photoluminescence (PL) measurements were conducted for CsPbIBr₂ perovskite films at various doping levels. Figure S2a presents the X-ray diffraction (XRD) patterns of CsPbIBr₂ perovskite films prepared at different TMC doping concentrations (pristine, 1, 3, 5, and 10 mg·mL⁻¹). It is clear that at a 3 mg·mL⁻¹ doping concentration, diffraction peaks become sharper and more intense, suggesting enhanced crystallinity and improved film quality. At doping concentrations higher than 3 mg·mL⁻¹, the crystallinity begins to decline, accompanied by broader and weaker peaks, indicating the formation of structural imperfections or secondary phases. The UV–Vis absorption spectra in Figure S2b show that the film doped at a 3 mg·mL⁻¹ TMC concentration exhibits significantly enhanced optical absorption compared to pristine and other doping levels. This indicates optimized film morphology and reduced defect density at this doping concentration, facilitating superior light-harvesting efficiency. Excessive doping (>3 mg·mL⁻¹) adversely affects the film quality, reducing the absorption capability. Correspondingly, Figure S2c displays the steady-state photoluminescence (PL) spectra of perovskite films prepared with different TMC doping concentrations. The film doped at 3 mg·mL⁻¹ exhibits the strongest PL intensity, demonstrating the effective passivation of non-radiative recombination centers and enhanced carrier lifetime. Lower or higher doping concentrations exhibit decreased PL intensities due to incomplete defect passivation or detrimental molecular aggregation effects.

To demonstrate the practical effectiveness of our solvent engineering and molecular doping strategy, we evaluated the photovoltaic performance of optimized CsPbIBr₂ perovskite solar cells (denoted as “Target”) in comparison to control devices (“Control”). The J–V curves presented in Figure 4a clearly indicate significant improvement in photovoltaic parameters for the optimized device. Specifically, the optimized solar cell (“Target”) achieved a notably higher PCE of 10.70%, accompanied by an enhanced open-circuit voltage (Voc) of 1.10 V, a short-circuit current density (Jsc) of 12.67 mA·cm⁻², and fill factor (FF) of 76.77%. In contrast, the control device showed a comparatively lower PCE (8.07%), Voc (1.07 V), Jsc (12.09 mA·cm⁻²), and FF (62.38%). This marked performance enhancement can be directly attributed to the improved perovskite film morphology, reduced defect density, and enhanced carrier extraction resulting from optimized molecular doping. The statistical analysis of multiple device measurements (Figure 4b) further confirms the reproducibility and reliability of performance improvements in the optimized devices, indicating the consistently higher and narrower distribution of PCE values compared to the control group. To elucidate the underlying mechanism for the enhanced Voc observed, we analyzed the dependence of Voc on incident illumination intensity (Figure 4c). The slope derived from the linear fit for the Target device (1.14 KBT/q) is closer to the ideal diode characteristic (KBT/q), which is significantly lower than that of the control device (1.28 KBT/q). This clearly indicates suppressed trap-assisted recombination in optimized devices, confirming effective defect passivation through the molecular doping approach. Finally, electron-only devices were measured to examine the trap-state density (Figure 4d). The optimized device shows a remarkably lower trap-filled limit voltage (V_TFL = 0.11 V) compared to the control device (0.20 V), suggesting a substantial reduction in trap density within the perovskite layer. This directly explains the improved charge transport, reduced recombination, and significantly enhanced photovoltaic efficiency. To assess the operational stability of the devices under continuous illumination, steady-state power output measurements were performed at their respective maximum power point voltages. As shown in Figure S3, the TMC-doped device achieved a stabilized power conversion efficiency of 9.81%, maintaining this value with minimal fluctuation over a 40 s period. In contrast, the undoped control device stabilized at a lower efficiency of 5.66%, exhibiting slight decay over time. This clear performance difference underscores the beneficial role of TMC doping in suppressing non-radiative recombination and enhancing the steady-state output of CsPbIBr₂ solar cells under operating conditions.

The operational stability of the Target device compared to the control device was systematically evaluated under ambient conditions (relative humidity: ~30%, temperature: ~25 °C), as depicted in Figure 5. Both device groups were encapsulated to minimize extrinsic degradation effects. The optimized device exhibited outstanding stability, retaining approximately 80% of its initial PCE after 250 h of continuous storage in ambient conditions. In sharp contrast, the control device underwent a pronounced decline in performance, retaining only about 40% of its initial efficiency within the same timeframe. This remarkable improvement in stability can be attributed to the effective defect passivation and suppression of ion migration provided by the molecular doping strategy with TMC. By significantly reducing surface and grain boundary defects, the optimized film morphology substantially minimizes the degradation pathways and prolongs the operational lifetime of the perovskite devices.

3. Conclusions

In summary, we systematically investigated the influence of solvent engineering (DMSO, DMF, and NMP) and molecular doping with TMC on the morphology, optoelectronic properties, and photovoltaic performance of CsPbIBr₂ perovskite solar cells. By optimizing PbBr₂–solvent complex formation, we identified NMP as the most effective solvent to enhance perovskite film crystallinity and optical absorption. Furthermore, introducing TMC doping at an optimal concentration (3 mg·mL⁻¹) significantly improved perovskite film quality, reduced trap densities, and enhanced device performance, increasing the PCE from an initial 8.07% to 10.70%. The optimized devices also exhibited excellent operational stability, retaining approximately 80% of their initial efficiency after 250 h under ambient conditions. This study demonstrates a practical and effective pathway toward realizing efficient and stable inorganic perovskite solar cells, providing valuable insights into morphological control and defect passivation strategies essential for future commercialization efforts.