| Abstract Scope |
PFM is ubiquitous for studying piezoelectrics. But to precisely determine piezocoefficients—especially for polycrystalline piezoelectrics—advanced manufacturer-modes, custom approaches, or expensive new systems are necessary. Furthermore, no matter the sophistication, uncertainties are superimposed due to tip-sample convolution associated with the local surface topography. Two broadly applicable approaches are presented to address these widespread challenges. First: carefully considering the mechanics of the AFM cantilever to improve measurement precision, corrects the measured piezoresponse for polycrystalline ceramics. Results improve up to an order of magnitude. Second: calculating, recognizing, and weighting pixels according to surface geometry parameters, improves automated assessment of hundreds of images—not just for PFM, but also for any AFM results. Removing suspect pixels obviously substantially improves local and statistical analyses of such ‘big [AFM] data.’ Both of these advances are important for microscopists, and also for modelling, fabrication, and device design which benefits from more and better analysis of local materials properties. |