Solution Chemistry and Aging of Perovskite Precursor Solutions

The efficiencies of perovskite solar cells (PSCs) have increased dramatically over the past 15 years, with a record power conversion efficiency (PCE) of 25.7% achieved in 2022, and device stability is continuously improving. Due to these recent developments, more and more attention is being devoted to the scalability of PSCs. One of the attractive properties of PSCs is they can be solution processed, making them compatible with many roll-to-roll processing techniques, commonly used in industrial manufacturing. In order to ensure reliable performances of devices made from solution processing, perovskite precursor ink shelf life must be accurately measured and extended. We also note that by understanding the
intermediate complexes which form within solution, we can improve the crystallisation and subsequent quality of perovskite films.

This work aims to find the usable lifetime of several perovskite precursor solutions, particularly for inks which commonly produce high performing PSCs devices. Specifically, we identify a particular degradation mechanism in a stoichiometric CsFAMAPb(IxBr1−x)3 ink over a period of several months. We find that the main degradation pathway is due to a reaction between the organic cation, methylammonium (MA+), and formamidinium iodide (FAI), which leads to the formation of non-perovskite polytypes in perovskite films. Interestingly, we find that solvent choice also has an effect on this reaction. Solutions dissolved in DMSO show a slower rate of reaction than those dissolved in a DMF/DMSO blend. We also explore the affect of precursor age on device stability and find that a photo-induced degradation mechanism occurs in devices made from aged inks. Importantly however, we find that we can delay all these effects, for several weeks at least, by storing precursor solutions at low temperatures (∼ 4◦C). Following on from this work, we investigate the solution stability of methylammonium-free perovskite solutions, namely a CsFAPbI3 precursor ink. We find that devices made from this ink perform consistently for up to 6 weeks after ink
creation. We believe this confirms the conclusions of the earlier work, that the reaction between FAI and MA+ is the most significant degradation pathway in these mixed-cation systems.

Finally, we analyse the methods used to study colloid sizes within PSC solutions. We find that results from dynamic light scattering (DLS) measurements on these inks can be useful - but sometimes misleading. We therefore expand on how DLS data is presented in order to accurately represent the colloid size distribution, emphasizing that volume-weighted data should always be presented over intensity-weighted data for PSC solutions. Additionally, we find that small-angle neutron scattering (SANS) and spin-echo SANS can be very useful in confirming particle size measurements and solute volume % within these inks. Excitingly, we also find that the SANS data can be modelled in such a way that not only size. but composition of the intermediate complexes within PSC solutions can be probed. We use this model to study the evolution of intermediate complexes within a Pb-excess CsFAMAPb(IxBr1−x)3 solution and a MAPbI3 solution. We here find that a noticeable change in MAPbI3 film morphology correlates with a change in composition and size of the MAPbI3 colloids. Studying these intermediate complexes could unlock key information about the crystallisation mechanics of the perovskite film.