Influence of Reactant Stoichiometry on Morphology and Structure of CH3NH3PbI3 Perovskite Films

Background

Solar photovoltaics have shifted from silicon and II–VI compounds to organic‑inorganic perovskites (OIP) due to their rapid power‑conversion efficiency (PCE) gains—from ~3.4 % in 2004 to 23.3 % in 2018 (22.6 % certified) [8–10]. OIP’s high absorption, long carrier diffusion lengths, and direct bandgap enable ultra‑thin active layers, reducing cost and material usage [7,18]. However, their intrinsic instability—sensitivity to moisture, oxygen, heat (>100 °C), and UV light—poses a major challenge, as MAI volatilization leaves PbI₂ residues [20–25]. Understanding the precise role of precursor stoichiometry is therefore essential for engineering robust, high‑performance perovskite solar cells.

Despite extensive research on MAPbI₃ crystallization, the influence of non‑stoichiometric precursor ratios on film microstructure and phase evolution remains underexplored. Previous work has shown that a slight MAI excess improves film density and device performance, yet the mechanistic pathways—particularly the emergence of intermediate lead‑halide phases—are not fully resolved [16,32]. This study addresses this gap by combining Raman spectroscopy, X‑ray powder diffraction (XRPD), and scanning electron microscopy (SEM) to trace phase formation and morphological changes across a systematic range of PbI₂ : CH₃NH₃I ratios.

Methods

Synthesis

Lead iodide (PbI₂), methylammonium chloride (CH₃NH₃Cl), and pre‑synthesized methylammonium iodide (MAI) served as precursors. A 2 % molar amount of CH₃NH₃Cl relative to MAI was added to partially substitute iodine with chlorine, enhancing perovskite stability [16,34]. Dimethylformamide (DMF) was the solvent of choice. Precursor solutions were stirred at 70 °C for 1 h, then spin‑coated (1200 rpm, 30 s) onto cleaned glass or FTO/glass substrates in a nitrogen glovebox. Thermal annealing ranged from 70–180 °C for 30 min on a preheated hot plate.

Characterization

SEM imaging (SEC miniSEM SNE 4500 MB) and EDX spectroscopy assessed morphology and composition. XRPD (DRON‑4‑07, CuKα, 40 kW, 18 mA) collected 2Θ = 10–120°, step 0.04°, count 4 s; Rietveld refinement extracted lattice parameters. Raman spectra were recorded with 532 and 671 nm lasers; excitation power was minimized (< 5×10² W/cm²) to prevent photodegradation.

Results and discussion

Solution Chemistry

Raman spectra (Figure 1) reveal that mixing PbI₂ with MAI in DMF induces the formation of lead polyiodide species (e.g., [PbI₃]⁻, [PbI₄]²⁻), evidenced by a shifting Pb–I mode from 114 to 121 cm⁻¹ as MAI concentration increases. This shift correlates with the red‑shift of the optical absorption edge from 2.54 eV (pure PbI₂) to 2.24 eV (1:3 mixture), confirming enhanced iodide coordination. The presence of 2 % CH₃NH₃Cl introduces additional Raman bands (e.g., 953, 997 cm⁻¹) that are negligible in the final perovskite solutions due to the low chlorine content.

Film Morphology

SEM images (Figure 2) show that MAI‑only films form smooth, glass‑like surfaces with minor heterogeneities, whereas PbI₂ films exhibit elongated, wire‑like grains that grow anisotropically at 90 °C. Perovskite films (Figure 3) exhibit pronounced particle anisotropy at a 1:1 ratio—needle‑like grains dominate. Increasing MAI excess to 1:2 yields maple‑leaf‑shaped particles with multiple growth directions; at 1:3 the grains become smaller and the film appears denser, reflecting the suppression of anisotropic growth by excess organic cations.

Phase Evolution

XRPD patterns (Figure 4) confirm that a stoichiometric 1:1 ratio produces a single‑phase MAPbI₃ between 70–80 °C. Higher temperatures (>120 °C) introduce PbI₂ peaks, indicating thermal decomposition per the reaction:

$$\mathrm{CH}_3\mathrm{NH}_3\mathrm{PbI}_3\overset{>80^{\circ}\mathrm{C}}{\to}\mathrm{PbI}_2+\mathrm{CH}_3\mathrm{I}\uparrow+\mathrm{NH}_3\uparrow$$

For a 1:2 ratio, an intermediate (CH₃NH₃)₂PbI₄ forms below 180 °C, which subsequently converts to MAPbI₃ upon further heating. Similarly, the 1:3 ratio first yields (CH₃NH₃)₃PbI₅ and (CH₃NH₃)₂PbI₄ before the final perovskite crystallizes. The orthorhombic, tetragonal, and cubic phases of MAPbI₃ are well documented; in all tested samples the tetragonal phase (I4/mcm) dominates, as evidenced by the split (220)/(004) peaks (Figure 5).

Raman Spectroscopy

Raman spectra (Figure 6) show that films derived from 1:1, 1:2, and 1:3 mixtures possess identical tetragonal perovskite signatures. However, prolonged laser irradiation (200–400 s) or elevated excitation power causes spectral shifts indicative of a transition toward a metastable PbI₂‑rich state, ultimately leading to complete decomposition at higher fluences. This laser‑induced sensitivity underscores the necessity for careful handling during spectroscopic studies.

Additional Raman experiments on 1:3 films annealed up to 180 °C (Figure 7) reveal that the metastable phase is retained at the highest temperature, aligning with the XRPD evidence of perovskite decomposition.

Conclusions

The study demonstrates that tailoring the PbI₂ : MAI ratio in DMF solution allows precise control over perovskite film morphology and phase evolution. A 1:1 stoichiometry yields a stable MAPbI₃ phase at 70–80 °C, whereas higher MAI excess induces intermediate lead‑halide phases that ultimately transform into MAPbI₃ upon annealing. The Pb/I stoichiometry remains constant across all compositions; therefore, variations in device‑relevant properties arise from anisotropic grain growth rather than elemental composition. Raman spectroscopy confirms the susceptibility of perovskite films to laser‑induced degradation, with a metastable transition to PbI₂ preceding complete decomposition.