Tuning dielectric/mechanical performance of polymeric composites for flexible electronics based on electric field-driven configurational entropy change: High throughput simulation and application

High-sensitivity capacitive sensing materials require high dielectric permittivity and low elastic modulus. The development period of polymeric composites was long and costly due to an insufficient understanding of the inorganic filler's paradoxical role in modulating dielectric-mechanical properties and the lack of methods to regulate the filler's distribution states quantitatively. This paper conducts high-throughput Markov chain Monte Carlo and finite element simulations to investigate the effects of 1D filler's orientation angle distribution and self-assembly degree induced by the assisted electric field on the dielectric permittivity and elastic modulus of the composites. General regression models were developed by artificial experience and machine learning to establish the composite structure-property mapping. Furthermore, the relative sensitivity of the composite's capacitive sensing can be improved from 20% to over 40%, and the absolute sensitivity increases by over 6 times by controlling the field strength to make the TiO2w state between orientation and full self-assembly in PDMS. This work provides a general physical and data-driven strategy for the rational design of polymeric composites with multi-objective requirements for flexible electronics.