Solvent annealing has been shown to be an effective and simple post-deposition treatment for perovskite solar cells (PSCs) that does not rely on complex chemical modification on the precursor solution or on the as-deposited films [1]. The ability to improve the efficiency of a PV device through such a simple process is therefore in principle an attractive approach. However, due to a lack of uniformity and reproducibility, this process is not widely used for the production of high-efficiency devices, nor seems likely to be highly relevant for commercial applications.

We have demonstrated a new approach to solvent-based post-treatment of perovskite films that addresses these major issues with reproducibility and uniformity, and have shown the applicability of this approach to large-scale processing of PSCs [2]. This is achieved by controllably passing a solvent aerosol across the film surface within a bespoke reactor. By controlling the aerodynamic processes, a boundary layer forms through which the solvent diffuses towards the film surface, leading to a highly-uniform and controllable exposure to the solvent. This was initially conducted for conventional methylammonium lead iodide (MAPbI₃) films and the aerosol treated films were analysed using several materials characterisation techniques. Analysis of the aerosol treated films indicated grain growth, improvement in microscale uniformity, reduction in lead iodide impurity phases and better perovskite/ HTL alignment. To explore the impact of the aerosol solvent treatment, the treated films were incorporated into PSCs.

Gaining such high level of control of the process has allowed us to not only achieve a power conversion efficiency (PCE) enhancement of p-i-n structured photovoltaic devices to over 20% using MAPbI₃, but also, more significantly, we have achieved large improvements in uniformity of film morphology after treatment over the whole reactor area (~30 cm²), which also translates to reduced PCE variability of devices produced from treated films.  This improved large-area uniformity leads to a much less marked efficiency drop when scaling up device area to 1 cm² compared to untreated films as well as increased stability under continuous maximum-power-point tracking in both N₂ atmosphere and ambient air. This highlights the potential for this technique to be used for commercial PV processing. Furthermore, we also demonstrate the versatility of the technique and its applicability to a wide range of perovskite compositions and cell architectures, including Cs-FA, triple cation, Br-containing, ‘conventional’ (n-i-p), and HTL-free architectures, with efficiency enhancements in all cases.