Fig. 2: Photolithographic patterning and long-term stability of microprocessed films.

a Optical imaging of a microprocessed 50 μm square before and after thermally quenching, stored under an ambient atmosphere (42%RH). The darker regions indicate the black phase, while the growing lighter parts are areas of the film that have transitioned to the yellow non-perovskite phase. b Micro-Raman characterization of a phase-stable microprocessed area; micro-Raman spectra are recorded inside (black), outside (blue) and on the border (red) of the microprocessed area, with an optical image of this area and corresponding measurement points (circles) inset. The image scale bar represents 50 μm. The labeled vertical lines link the detected modes to their assigned chemical species. c Temporal stability of black thin films stored under ambient conditions (42–48%RH) and processed with the different grid sizes indicated. The trends (solid lines) are modeled using the Johnson-Mehl-Avrami-Kolmogorov (JMAK) model, through Eq. (1). d The average transformation rate constant, k, over five samples as a function of microgrid size, as derived from Avrami plots (see Supplementary Fig. 28). The error bars correspond to the upper and lower bounds of determined values. e Comparison of the long-term phase stability of an as-grown film (red) stored in a glovebox with films installed with a 40 μm grid and stored under dry (black) and ambient, unregulated (blue) environments. Circle markers indicate the average fraction of surviving black-phase squares of five samples each with their upper and lower limits depicted by the solid fill region. The blue broken vertical line indicates where the ambient study was stopped due to moisture damage (Supplementary Fig. 31).