In recent years, the rapid development of perovskite solar cells has garnered significant attention due to their exceptional optoelectronic properties, such as long carrier diffusion lengths, high carrier mobility, and low trap-state densities. The power conversion efficiency (PCE) of perovskite solar cells has surged from an initial 3.8% to an impressive 26.8%, highlighting their potential as a leading photovoltaic technology. Among various types, all-inorganic perovskite solar cells, particularly those based on CsPbI3, have emerged as a focal point of research owing to their superior thermal stability compared to organic-inorganic hybrid counterparts. The electron transport layer (ETL) is a critical component in perovskite solar cells, playing a vital role in extracting and transporting photogenerated electrons while suppressing charge recombination. Zinc oxide (ZnO) has become one of the most prominent ETL materials due to its relatively wide bandgap, high electron mobility, and compatibility with low-temperature processing, making it suitable for flexible device applications. However, while extensive studies have focused on ZnO-based ETLs in organic-inorganic hybrid perovskite solar cells, research on their application in all-inorganic systems remains limited. This study aims to systematically investigate the influence of ZnO ETLs, fabricated under varying precursor concentrations, annealing times, and temperatures, on the morphological, structural, and optoelectronic properties of all-inorganic CsPbI3 perovskite layers and the overall performance of perovskite solar cells.
The fabrication of ZnO ETLs was carried out using a colloidal spin-coating method. Precursor solutions of zinc acetate dihydrate (Zn(CH3COO)2·2H2O) in ethoxyethanol, with concentrations of 0.25, 0.50, and 0.75 mol·L−1, were prepared by dissolving the solute in ethanolamine and stirring at 60°C for 30 min. The solutions were then spin-coated onto cleaned FTO substrates at 1500 rpm for 10 s and 3000 rpm for 20 s, followed by annealing under different conditions (e.g., 250–350°C for 1–5 min). The resulting ZnO films were characterized using scanning electron microscopy (SEM), X-ray diffraction (XRD), and ultraviolet-visible (UV-Vis) spectroscopy to assess their morphology, crystallinity, and optical properties. For the all-inorganic perovskite layer, a precursor solution of HPbI3, PbI2, and CsI in a DMF/DMSO (9:1 v/v) mixture was spin-coated onto the ZnO ETLs and annealed at 120°C for 30 s. The perovskite solar cells were assembled in a planar structure: FTO/ZnO/CsPbI3/Spiro-MeOTAD/Ag. The hole transport layer was deposited by spin-coating Spiro-MeOTAD solution, and a 100 nm thick Ag electrode was thermally evaporated under high vacuum. The devices were evaluated under AM 1.5G illumination to measure current density-voltage (J-V) characteristics, while stability tests were conducted in a nitrogen atmosphere with controlled humidity.
The morphology of the ZnO ETLs was highly dependent on the fabrication parameters. SEM images revealed that as the precursor concentration increased from 0.25 to 0.75 mol·L−1, the ZnO nanoparticle layers developed surface gaps and irregularities, with the highest concentration resulting in larger voids and non-uniform coverage. This was attributed to insufficient nucleation and growth under suboptimal annealing conditions. When the annealing time was extended from 1 to 5 min at 250°C, the amorphous regions on the ZnO surface diminished, and particle size increased, but agglomeration occurred, leading to rough films. Optimizing the annealing temperature to 300°C for 5 min produced ZnO layers with minimal pores, uniform particle distribution, and reduced agglomeration. Further increasing the temperature to 350°C caused partial melting and团聚 of nanoparticles, degrading film quality. XRD analysis confirmed the formation of pure ZnO phases, with diffraction peaks at 31.7°, 34.3°, and 36.2° corresponding to the (100), (002), and (101) crystal planes, respectively. The intensity of these peaks increased with higher precursor concentrations and longer annealing times, indicating improved crystallinity. However, at 350°C, peak intensity decreased due to thermal-induced aggregation.
The quality of the CsPbI3 perovskite films grown on these ZnO ETLs was significantly influenced by the underlying morphology. SEM images showed that perovskite films on ZnO layers from 0.25 mol·L−1 precursor had large, irregular grains with visible gaps, while those on 0.50 mol·L−1-derived ZnO exhibited smaller, more uniform grains with reduced intergranular spaces. Extending the annealing time of ZnO to 5 min resulted in perovskite films with compact, well-defined crystals and minimal voids. The best perovskite morphology was achieved on ZnO annealed at 300°C for 5 min, featuring dense, pinhole-free films with homogeneous grain size. XRD patterns of the perovskite layers displayed characteristic peaks at 14.2° and 28.3°, assigned to the (110) and (220) planes of γ-CsPbI3, respectively. The sharpness and intensity of these peaks were highest for perovskite films on ZnO ETLs prepared at 0.50 mol·L−1 and 300°C, indicating superior crystallinity. The optical properties of the perovskite films were evaluated using UV-Vis spectroscopy and Tauc plot analysis. The absorption spectra showed that films on optimized ZnO ETLs had higher light absorption in the visible range, which is beneficial for generating more electron-hole pairs. The bandgap energy (Eg) of the perovskite films, calculated from the Tauc plot using the equation:
$$(αhν)^2 = A(hν – E_g)$$
where α is the absorption coefficient, hν is the photon energy, and A is a constant, varied between 1.511 and 1.529 eV depending on the ZnO preparation conditions. This slight variation in Eg was correlated with changes in crystal size and film quality induced by the different ETLs.
Steady-state photoluminescence (PL) and time-resolved photoluminescence (TRPL) measurements were conducted to probe the charge carrier dynamics and defect states in the perovskite films. The PL intensity was highest for films on ZnO ETLs annealed at 300°C, suggesting minimized non-radiative recombination and fewer surface defects. TRPL decay curves were fitted with a bi-exponential function:
$$I(t) = A_1 e^{-t/τ_1} + A_2 e^{-t/τ_2}$$
where I(t) is the fluorescence intensity, τ1 and τ2 are the fast and slow decay lifetimes, and A1 and A2 are their respective amplitudes. The average carrier lifetime (τave) was calculated as:
$$τ_{ave} = \frac{A_1 τ_1^2 + A_2 τ_2^2}{A_1 τ_1 + A_2 τ_2}$$
The results showed that τave increased from 20.12 ns for perovskite on ZnO annealed at 250°C to 39.12 ns for films on ZnO annealed at 300°C, indicating enhanced charge extraction and reduced recombination at the interface. However, at 350°C, τave decreased to 27.50 ns, consistent with increased defect density due to ZnO agglomeration.
The phase stability of the CsPbI3 perovskite films was evaluated by aging them in air with 5–10% relative humidity. Films on ZnO ETLs annealed at 250°C began transitioning from the black γ-phase to the yellow δ-phase within one day, and complete conversion occurred after 30 days. In contrast, films on ZnO annealed at 300°C retained most of the black phase even after 30 days, demonstrating excellent phase stability. XRD analysis confirmed the gradual appearance of δ-phase peaks at 13.1° and 26.2° for films on lower-temperature ZnO, while films on optimized ZnO maintained dominant γ-phase peaks. This enhanced stability is attributed to the improved crystallinity and reduced defects in the perovskite layer when grown on high-quality ZnO ETLs.
The photovoltaic performance of the perovskite solar cells was assessed through J-V measurements under standard illumination conditions. The key parameters, including open-circuit voltage (Voc), short-circuit current density (Jsc), fill factor (FF), and PCE, are summarized in Table 1 for devices fabricated with different ZnO ETLs. The data represent average values from multiple devices to ensure reproducibility.
| ZnO Preparation Conditions | Voc (V) | Jsc (mA·cm−2) | FF (%) | PCE (%) |
|---|---|---|---|---|
| 0.25 mol·L−1, 250°C, 1 min | 0.66 | 9.90 | 44.38 | 2.90 |
| 0.50 mol·L−1, 250°C, 1 min | 0.67 | 10.94 | 46.39 | 3.40 |
| 0.75 mol·L−1, 250°C, 1 min | 0.67 | 10.48 | 44.86 | 3.15 |
| 0.50 mol·L−1, 250°C, 3 min | 0.71 | 10.64 | 61.02 | 4.61 |
| 0.50 mol·L−1, 250°C, 5 min | 0.72 | 11.62 | 62.99 | 5.27 |
| 0.50 mol·L−1, 300°C, 5 min | 0.79 | 12.15 | 66.36 | 6.37 |
| 0.50 mol·L−1, 350°C, 5 min | 0.81 | 10.88 | 66.04 | 5.82 |
The device with ZnO ETL prepared at 0.50 mol·L−1 and annealed at 300°C for 5 min achieved the highest PCE of 6.37%, with a Voc of 0.79 V, Jsc of 12.15 mA·cm−2, and FF of 66.36%. This performance is attributed to the optimal ZnO morphology, which facilitated high-quality perovskite growth, efficient charge transport, and reduced recombination. The J-V curves exhibited minimal hysteresis, indicating good interfacial properties. In contrast, devices with ZnO from higher concentrations or suboptimal annealing showed lower PCE due to poor film coverage and increased defects. The stability of the perovskite solar cells was monitored over 480 hours in a nitrogen environment with 10–15% humidity. The normalized PCE retention is presented in Table 2, showing that the optimized device maintained 79.1% of its initial efficiency, outperforming others which degraded more rapidly.
| ZnO Preparation Conditions | PCE Retention after 480 h (%) |
|---|---|
| 0.25 mol·L−1, 250°C, 1 min | 50.1 |
| 0.50 mol·L−1, 250°C, 1 min | 54.2 |
| 0.75 mol·L−1, 250°C, 1 min | 56.9 |
| 0.50 mol·L−1, 250°C, 3 min | 60.4 |
| 0.50 mol·L−1, 250°C, 5 min | 64.7 |
| 0.50 mol·L−1, 300°C, 5 min | 79.1 |
| 0.50 mol·L−1, 350°C, 5 min | 71.5 |
The enhanced stability of the optimized device is correlated with the improved phase stability of the perovskite layer and the robust ZnO ETL, which effectively mitigated environmental degradation. The relationship between the ZnO ETL properties and the perovskite solar cell performance can be further analyzed using the diode equation:
$$J = J_0 \left( e^{\frac{qV}{nkT}} – 1 \right) – J_{ph}$$
where J is the current density, J0 is the reverse saturation current, q is the electron charge, V is the voltage, n is the ideality factor, k is Boltzmann’s constant, T is the temperature, and Jph is the photocurrent density. The lower n values for devices with optimized ZnO ETLs indicate reduced recombination losses, contributing to higher FF and Voc.
In conclusion, this study demonstrates that the preparation conditions of ZnO ETLs profoundly influence the morphology, crystallinity, and optoelectronic properties of all-inorganic CsPbI3 perovskite films and the overall performance of perovskite solar cells. Optimizing the precursor concentration to 0.50 mol·L−1 and annealing at 300°C for 5 min yielded ZnO layers with uniform morphology and high crystallinity, which in turn promoted the growth of high-quality perovskite films with enhanced phase stability and reduced defect density. The resulting perovskite solar cells achieved a peak PCE of 6.37% and maintained 79.1% of their initial efficiency after 480 hours in a controlled environment. These findings underscore the importance of tailored ETL fabrication for advancing all-inorganic perovskite solar cells, providing a pathway for low-temperature, cost-effective production of efficient and stable photovoltaic devices. Future work will focus on further optimizing the ZnO/perovskite interface through surface passivation and exploring scalable deposition techniques to facilitate the commercialization of perovskite solar cells.