Publication date: 17th February 2025
The crystallisation dynamics of Sn-containing perovskites occur on significantly faster timescales, posing challenges for their synthesis[1]. Despite the similarity in perovskite formability between Pb and Sn perovskites according to the Bartel tolerance factor, Sn-based compositions more readily form 3D, corner-sharing architectures[2]. Moreover, intermediate phases - such as solvates and hexagonal polytypes – that are often observed in Pb-based systems are typically absent during the processing of Sn-containing perovskites. In this work, we investigate the crystallisation kinetics of blade-coated, pure-Pb and mixed PbSn perovskite solutions using in situ GIWAXS. Whilst typical crystalline intermediates are observed for pure Pb compositions, none are observed for PbSn compositions apart from high Cs compositions. This indicates stronger Sn-halide bonding networks and an inherent preference for corner-sharing architectures in Sn-containing compositions. Reactive force-field machine learning simulations further support this, revealing a suppression of non-corner-sharing Sn phases. To overcome this, we introduce a larger cation; DMA+ into the lattice to modulate crystallisation[3], promoting a transient 2H polytype that facilitates formation of the desired, final 3D perovskite phase. This leads to the growth of larger crystal grains, demonstrating a viable pathway to control the crystallisation of Sn-containing perovskites through engineered intermediate phases.
We acknowledge Diamond Light Source for measurements taken on the i07 beamline during sessions si30612-1, si33462-1 and si36427-1, and beamline scientist Dr. Jonathan Rawle for assistance with these measurements.