Piezoelectric-like effect manipulation of polymeric piezoelectrets through machine-learning-assisted pore morphology analysis and porous structural design

Polymeric porous piezoelectrets offer advantages for flexible electronics because of their mechanical compliance and low density. Their piezoelectric-like response originates from charge accumulation at pore interfaces and deformation of pores and wrinkles. Therefore, enhancing charge density and precisely manipulating pore shape are critical for enhancing electromechanical performance. Addressing the challenges of irregular pore morphologies, this work applies a machine-learning-assisted pore morphology analysis workflow to segment, quantify, and classify pore geometries, subsequently guiding a fabrication strategy that adjusts pore irregularity and aspect ratio to enhance the equivalent piezoelectric coefficient. This methodology provides a systematic framework for quantitative microstructure design, offering practical guidance for enhancing the piezoelectric-like response and tailoring the mechanical properties of porous piezoelectrets.