Spray‑Pyrolysis Fabrication of High‑Purity MAPbI3 Perovskite Particles for Thin‑Film Solar Cells

Background

Organometallic halide perovskites—especially those incorporating methylammonium (MA), formamidinium (FA), or cesium (Cs)—are at the forefront of photovoltaic research and are also explored for transistors, LEDs, and radiation sensors. While most work focuses on thin‑film deposition, nanocrystalline perovskite particles offer quantum‑confinement effects and new device architectures. The literature is dominated by all‑inorganic Cs‑based nanocrystals; however, MA‑based particles remain underexplored, particularly in the context of scalable production.

Presentation of the Hypothesis

We propose three distinct synthesis strategies for MAPbI3 particles:

1. Anti‑solvent precipitation – Dropwise addition of a perovskite precursor solution into toluene.
2. Dry‑mix milling – Mechanical blending of MAI and PbI2 powders on a magnetic stirrer at 200 °C.
3. Spray‑pyrolysis – Atomization of the precursor solution with a 0.2 mm nozzle into a two‑stage stainless‑steel heater (275 °C/175 °C). This method exploits solvent evaporation and in‑situ chemical conversion to form crystalline particles.

Figure 1 illustrates the schematic of the three routes.

Testing the Hypotheses

Perovskite precursors (MAI = 158 mg, PbI2 = 420 mg) were dissolved in 1 mL DMSO. The resulting 1 mL solution was processed by each method, and the powders were annealed at 150 °C (anti‑solvent) or collected directly (spray) for characterization.

1. Anti‑solvent Method

Rapid precipitation produced yellow powder after 2 min; continued stirring for 20 min yielded a colloidal dispersion. XRD (Bruker D5005) confirmed perovskite formation with minor impurity peaks.

2. Milling Method

MAI/PbI2 mass ratios of 1:1 and 1:2 were tested at 200 °C. XRD showed dominant perovskite peaks but residual PbI2 signals, indicating incomplete conversion.

3. Spray Method

Using 2 psig air pressure, droplets entered a 10 cm × 30 cm heater. The first stage (275 °C) evaporated solvent; the second stage at 175 °C prevented decomposition, yielding a highly crystalline, impurity‑free powder.

Figure 2 presents the XRD patterns, and Figure 3 shows SEM images revealing micron‑sized, irregular particles; spray‑derived powders exhibit the narrowest size distribution (Figure 4).

Implication of the Hypothesis

Perovskite pastes were formulated by dispersing 20 mg of each powder in 10 µL ethanol and blade‑coated onto mesoporous TiO2 layers on FTO glass. SEM (Figure 6) shows that only the spray‑derived paste forms a continuous, defect‑free film. UV‑Vis spectra (Figure 7) confirm the expected absorption edge at ~750 nm for spray‑derived films.

We fabricated a basic MAPbI3 solar cell: TiO2 / perovskite / Spiro‑OMeTAD / Au. The device exhibited a PCE of 2.05 % (Voc < 0.4 V, Jsc = 6.3 mA cm‑2, FF = 0.8), limited by interfacial charge recombination likely caused by weak particle binding in the film. Improved binders could enhance performance.

Conclusions

Spray‑pyrolysis consistently produces crystalline, impurity‑free MAPbI3 particles with a narrow size distribution, enabling the fabrication of uniform perovskite films and functional solar cells. This scalable route holds promise for large‑area perovskite device manufacturing.