The family of ABX₃ lead halide perovskite materials continues to forge forward with record efficiencies. The ease of manufacture and numerous simple structural modifications make perovskites attractive for both research and for potential scale-up. As the photoactive layer within the device, the surfaces of perovskite materials themselves have provoked considerable interest. Techniques such as X-ray Photoelectron Spectrometry (XPS), Ultraviolet Photoelectron Spectrometry (UPS) provide insight into the surface chemistry and physics. The surface itself is important as if forms the interface with subsequent layers in the device. The surface would also be especially sensitive to degradation, showing evidence of materials failure before bulk analytical techniques.

We have studied methyl ammonium lead triiodide (MAPI) and triple cation films in several architectures utilising both 1) a simple planar FTO(fluorine doped tin oxide)/TiO₂ substrate architecture and 2) the triple mesoporous TiO₂/ZrO₂/carbon substrate favoured by many as a scale-up architecture due to its manufacturing simplicity. We have examined influence of both the manufacturing consistency and the stability of the perovskite films on the surface composition. Quantified surface analyses show three clear defects that vary with manufacturing. The first two offer insight into the quality of the perovskite manufacturing process: 1) The presence of film pinholes in the perovskite layer reveal the chemistry of the underlying n-type selective film, and are an obvious direct measure of coating quality; 2) The presence of excess PbI₂ varies with both age and manufacturing technique, and suggests that the sequential deposition route allows strong control of lead halide to perovskite conversion in the samples studied. 3) A Pb(0) component is observed in some cases suggesting the presence of metallic lead at the surface, which it is speculated may contribute to losses within the photovoltaic cell. Although not omnipresent across the literature, that Pb(0) is present in some analyses is beyond dispute: XPS has previously shown a distinct shift of the classic perovskite Pb(II) peak to a lower binding energy consistent with Pb(0); UPS has shown metallic behaviour at the valence level with electron density continuing up to the Fermi level; and several other techniques have noted presence of Pb(0). In fact, we demonstrate reproducible appearance of Pb(0) on otherwise pristine films and hence suggest it as a useful indirect measure of film quality.