In recent years, perovskite solar cells have emerged as a promising third-generation photovoltaic technology due to their high power conversion efficiencies and low-cost fabrication processes. Among them, all-inorganic perovskite solar cells, particularly those based on CsPbBr3, have garnered significant attention for their exceptional thermal and environmental stability compared to their organic-inorganic hybrid counterparts. However, the multi-step spin-coating method commonly used for CsPbBr3 perovskite film deposition often results in poor film quality, including low phase purity, small grain sizes, and high defect densities, which limit device performance. Moreover, the use of methanol as a solvent for CsBr poses environmental and health concerns. In this study, we introduce a novel interface engineering strategy using sodium polyacrylate (PAAS) to enhance the efficiency and stability of CsPbBr3-based perovskite solar cells. By replacing methanol with water as a green solvent for CsBr and employing a two-step spin-coating method, we successfully fabricated high-quality CsPbBr3 films. The incorporation of a PAAS interlayer between the perovskite and carbon electrode not only passivates interface defects but also improves energy level alignment, leading to suppressed non-radiative recombination and enhanced charge carrier transport. Through comprehensive characterization, including X-ray diffraction (XRD), atomic force microscopy (AFM), scanning electron microscopy (SEM), and various spectroscopic techniques, we demonstrate that PAAS modification significantly improves film morphology, increases grain size, and reduces surface roughness. Optimized devices achieve a champion power conversion efficiency (PCE) of 9.90% with an open-circuit voltage (VOC) of 1.58 V, short-circuit current density (JSC) of 8.09 mA·cm−2, and fill factor (FF) of 77.21%, outperforming control devices. Furthermore, the PAAS-modified perovskite solar cells exhibit superior stability and reproducibility, making this approach a viable pathway for advancing all-inorganic perovskite photovoltaics.
Perovskite solar cells have revolutionized the field of photovoltaics with their rapid efficiency improvements, reaching certified PCEs over 26% within a short span. The general formula ABX3 for perovskite materials, where A is a monovalent cation (e.g., Cs+, MA+, FA+), B is a divalent metal ion (e.g., Pb2+, Sn2+), and X is a halide anion (e.g., I−, Br−, Cl−), allows for tunable optoelectronic properties. Despite the high efficiencies of organic-inorganic hybrid perovskite solar cells, their instability under ambient conditions, primarily due to the volatile organic cations, hinders commercial deployment. In contrast, all-inorganic perovskites like CsPbBr3 offer remarkable robustness against heat, moisture, and light, coupled with a wide bandgap (~2.3 eV) that enables high open-circuit voltages. However, the fabrication of CsPbBr3 films often involves multiple cycles of CsBr solution spin-coating on PbBr2 layers, leading to incomplete conversion and the presence of undesirable phases like CsPb2Br5. This process not only complicates manufacturing but also introduces numerous defects at grain boundaries and interfaces, promoting charge carrier recombination and degrading device performance. To address these challenges, interface engineering has been employed as an effective strategy to passivate defects, enhance film quality, and improve charge extraction in perovskite solar cells.
In this work, we focus on utilizing PAAS, a water-soluble polymer, as an interface modifier for CsPbBr3 perovskite solar cells. PAAS contains carboxylate groups that can coordinate with undercoordinated Pb2+ ions on the perovskite surface, thereby reducing defect states. Additionally, its hydrophilic nature may facilitate the formation of a protective layer that mitigates moisture ingress while maintaining efficient charge transport. We systematically investigate the impact of PAAS concentration on film properties and device performance, identifying an optimal condition that maximizes PCE. Our findings are supported by a series of experiments, including structural, morphological, optical, and electrical characterizations. We also explore the mechanisms behind the performance enhancement, such as reduced defect density, improved charge carrier dynamics, and better energy level alignment. The results underscore the potential of polymer-based interface engineering in developing efficient and stable all-inorganic perovskite solar cells for practical applications.
The fabrication of CsPbBr3 perovskite solar cells in this study follows a simplified two-step spin-coating method. First, a compact TiO2 electron transport layer is deposited on fluorine-doped tin oxide (FTO) substrates via spray pyrolysis. Then, a PbBr2 layer is formed by spin-coating a PbBr2 solution in dimethylformamide (DMF). Subsequently, a high-concentration CsBr aqueous solution (250 mg·mL−1) is spin-coated onto the PbBr2 film, followed by thermal annealing to form the CsPbBr3 perovskite layer. This approach eliminates the need for toxic methanol and reduces the number of processing steps compared to conventional multi-step methods. For interface engineering, PAAS solutions at various concentrations (0.5, 1.0, 1.5, and 2.0 mg·mL−1) in a water-isopropanol mixture (1:9 volume ratio) are spin-coated on the perovskite surface before depositing the carbon electrode. The devices are characterized without any hole transport layer, simplifying the structure and reducing costs. The complete device architecture is FTO/TiO2/CsPbBr3/PAAS/carbon, and all measurements are conducted under ambient conditions unless specified otherwise.
To evaluate the effect of PAAS on the crystalline structure of CsPbBr3 films, we performed XRD analysis. The diffraction patterns for pristine and PAAS-modified films are shown in Figure 1a. Characteristic peaks at 15.3°, 21.7°, 30.8°, and 38.1° correspond to the (100), (110), (200), and (211) planes of cubic CsPbBr3, respectively. Notably, the intensity of these peaks increases with PAAS modification, indicating enhanced crystallinity and phase purity. Conversely, peaks associated with CsPb2Br5 impurities at 11.7° and 29.4° diminish, suggesting that PAAS suppresses the formation of secondary phases. The magnification of the XRD pattern around 30° (Figure 1b) clearly shows the strengthening of the CsPbBr3 (200) peak and the weakening of the CsPb2Br5 peak, confirming the role of PAAS in promoting phase-pure perovskite formation. We attribute this to the coordination between carboxylate groups in PAAS and Pb2+ ions, which stabilizes the perovskite structure and reduces defects. The optimal PAAS concentration is found to be 1.5 mg·mL−1, as it yields the highest CsPbBr3 peak intensities.
Surface morphology and roughness of the perovskite films are critical for charge transport and interface contact. AFM measurements in contact mode reveal that PAAS modification significantly reduces the root mean square (RMS) roughness from 66.2 nm for the pristine film to 53.0 nm for the PAAS-modified film (Figures 1c and 1d). The 3D AFM images (Figures 1e and 1f) visually demonstrate the smoother surface after PAAS treatment. This reduction in roughness facilitates better contact with the carbon electrode, minimizing interface recombination and enhancing device performance. Additionally, smoother surfaces are less prone to moisture adsorption, further improving stability.
SEM images provide further insights into the film morphology and grain size distribution. As shown in Figures 2a to 2e, the pristine CsPbBr3 film exhibits small grains with an average size of 767.08 nm, visible grain boundaries, and some pinholes. In contrast, PAAS-modified films show increased grain sizes and more compact packing. At the optimal concentration of 1.5 mg·mL−1, the average grain size reaches 1078.3 nm, with a narrow size distribution and minimal defects (Figure 2e). The corresponding histograms (Figures 2f to 2j) quantify the grain size evolution, confirming that PAAS promotes grain growth and reduces grain boundaries. This morphological improvement is consistent with the enhanced optical and electronic properties observed in subsequent tests.
The successful incorporation of PAAS is verified by infrared spectroscopy (IR). The IR spectrum of PAAS-modified perovskite films (Figure 3b) displays characteristic peaks at 1537 cm−1 (asymmetric stretching of -COO−), 1451 cm−1 (symmetric stretching of -COO−), and 1405 cm−1 (bending vibration of -CH2- groups), which are absent in the pristine film (Figure 3a). These peaks confirm the presence of PAAS on the perovskite surface. UV-visible absorption spectroscopy (Figure 3c) shows that PAAS-modified films have higher absorption intensity and a slight blue shift compared to the pristine film, indicating improved film quality and reduced defect-mediated absorption. This enhancement is attributed to the larger grain sizes and better crystallinity, which promote light harvesting and charge generation.
Photoluminescence (PL) and time-resolved photoluminescence (TRPL) measurements are conducted to investigate charge carrier recombination. The PL spectrum (Figure 3d) of the PAAS-modified film exhibits higher intensity and a blue shift, signifying reduced non-radiative recombination due to defect passivation. TRPL decay curves (Figure 3e) are fitted with a bi-exponential function: $$I(t) = A_1 e^{-t/\tau_1} + A_2 e^{-t/\tau_2}$$ where τ1 and τ2 are the fast and slow decay lifetimes, and A1 and A2 are their relative amplitudes. The average carrier lifetime (τave) is calculated as $$\tau_{\text{ave}} = A_1 \tau_1 + A_2 \tau_2$$. The parameters are summarized in Table 1. The PAAS-modified film shows a longer τave (8.256 ns) than the pristine film (6.478 ns), indicating suppressed recombination and improved charge carrier extraction. This aligns with the enhanced device performance.
| Sample | τ1 (ns) | A1 (%) | τ2 (ns) | A2 (%) | τave (ns) |
|---|---|---|---|---|---|
| Pristine | 0.7428 | 75.778 | 7.8946 | 28.87 | 6.478 |
| With PAAS | 0.8732 | 77.266 | 10.0899 | 26.927 | 8.256 |
The photovoltaic performance of devices with different PAAS concentrations is evaluated through J-V measurements under AM 1.5G illumination. The J-V curves are displayed in Figure 3f, and the corresponding parameters are listed in Table 2. The pristine device achieves a PCE of 8.22% with VOC = 1.51 V, JSC = 7.23 mA·cm−2, and FF = 75.19%. With PAAS modification, all parameters improve, reaching a maximum PCE of 9.90% at 1.5 mg·mL−1 PAAS (VOC = 1.58 V, JSC = 8.09 mA·cm−2, FF = 77.21%). This enhancement is attributed to better film quality, reduced recombination, and improved interface contact. However, at higher concentrations (e.g., 2.0 mg·mL−1), performance declines slightly, likely due to excessive PAAS aggregation that disrupts film morphology.
| PAAS Concentration | VOC (V) | JSC (mA·cm−2) | FF (%) | PCE (%) |
|---|---|---|---|---|
| Pristine | 1.51 | 7.23 | 75.19 | 8.22 |
| 0.5 mg·mL−1 | 1.54 | 7.43 | 75.95 | 8.71 |
| 1.0 mg·mL−1 | 1.56 | 7.88 | 76.05 | 9.26 |
| 1.5 mg·mL−1 | 1.58 | 8.09 | 77.21 | 9.90 |
| 2.0 mg·mL−1 | 1.52 | 7.48 | 75.83 | 8.63 |
To quantify the defect density in the perovskite films, we employ the space charge limited current (SCLC) method. The dark J-V curves for electron-only devices (FTO/TiO2/CsPbBr3/PAAS/carbon) are shown in Figure 4a. The trap-filled limit voltage (VTFL) is lower for the PAAS-modified device (1.172 V) than for the pristine device (1.235 V). The defect density (ntrap) is calculated using the formula: $$n_{\text{trap}} = \frac{2 \epsilon_0 \epsilon_r V_{\text{TFL}}}{e d^2}$$ where ε0 is the vacuum permittivity, εr is the relative permittivity of CsPbBr3 (~25), e is the elementary charge, and d is the film thickness (~400 nm). The calculated ntrap values are 1.362 × 1022 cm−3 for the pristine film and 1.292 × 1022 cm−3 for the PAAS-modified film, confirming that PAAS reduces defect states. Lower defect density minimizes non-radiative recombination, leading to higher charge carrier mobility and improved device efficiency.
Capacitance-voltage (C-V) and Mott-Schottky (M-S) analyses provide insights into charge accumulation and built-in potential. The C-V curves (Figure 4b) show reduced capacitance for PAAS-modified devices, indicating less charge accumulation at the interface and more efficient charge separation. The M-S plots (Figure 4c) derived from C-V data reveal a higher built-in voltage (Vbi) for the PAAS-modified device (1.35 V) compared to the pristine device (1.25 V). This increase in Vbi enhances the driving force for charge carrier separation and collection, contributing to the higher VOC and overall performance of the perovskite solar cell.
Electrochemical impedance spectroscopy (EIS) is conducted to study charge recombination dynamics. The Nyquist plots (Figure 4d) are fitted with an equivalent circuit model (inset), where the recombination resistance (Rrec) corresponds to the semicircle radius. The PAAS-modified device exhibits a larger Rrec (748.2 Ω) than the pristine device (551.2 Ω), indicating suppressed charge recombination and improved charge transport. This is consistent with the longer carrier lifetimes observed in TRPL measurements.
Transient photocurrent (TPC) and transient photovoltage (TPV) measurements further elucidate charge carrier dynamics. The TPC decay (Figure 4e) shows a shorter average extraction time (τavg) for the PAAS-modified device (3.90 μs) compared to the pristine device (4.43 μs), suggesting faster charge extraction. The TPV decay (Figure 4f) reveals a longer recombination lifetime (τavg = 7.061 ms) for the modified device versus the pristine device (6.198 ms), confirming reduced recombination. These results collectively demonstrate that PAAS modification enhances charge separation and collection while mitigating recombination losses.
Water contact angle measurements assess the hydrophilicity of the perovskite films. As shown in Figure 5, the PAAS-modified film has a lower contact angle than the pristine film, indicating increased hydrophilicity due to the presence of PAAS. This hydrophilic layer can attract and immobilize water molecules, preventing them from penetrating the perovskite structure and thus enhancing environmental stability. This property is crucial for the long-term durability of perovskite solar cells under ambient conditions.
The reproducibility and stability of the devices are evaluated through statistical analysis and long-term testing. Figures 6a to 6d present box plots of VOC, JSC, FF, and PCE for 20 independent devices each of pristine and PAAS-modified perovskite solar cells. The PAAS-modified devices show narrower distributions and higher average values, indicating better reproducibility. Additionally, the stability over 30 days (Figures 6e to 6f) demonstrates that PAAS-modified devices retain over 90% of their initial PCE, while pristine devices degrade more rapidly. This enhanced stability is attributed to the defect-passivating and protective effects of PAAS.
In conclusion, we have developed an efficient and stable CsPbBr3 perovskite solar cell by incorporating a PAAS interlayer via interface engineering. The use of water as a solvent for CsBr simplifies the fabrication process and reduces environmental impact. PAAS modification improves film quality by increasing grain size, enhancing crystallinity, and reducing surface roughness. It also passivates interface defects, suppresses non-radiative recombination, and optimizes energy level alignment, leading to significant improvements in photovoltaic parameters. The champion device achieves a PCE of 9.90%, with high VOC, JSC, and FF. Moreover, the PAAS-modified perovskite solar cells exhibit excellent stability and reproducibility, making them suitable for practical applications. This work highlights the potential of polymer-based interface engineering in advancing all-inorganic perovskite photovoltaics and provides a foundation for future developments in high-performance, stable perovskite solar cells.
The success of this study opens avenues for further optimization, such as exploring other functional polymers or combining PAAS with additional passivation strategies. Future work could focus on scaling up the fabrication process and integrating these devices into tandem structures for higher efficiency. Overall, our findings contribute to the ongoing efforts to commercialize perovskite solar cells by addressing key challenges in efficiency and stability.