Hierarchical ferroelectric composites for energy conversion technologies

Abstract

This Thesis presents several investigations into the effects of porosity on the electromechanical properties of porous freeze-cast ferroelectric ceramics. A combination of modelling and experimental approaches was used to provide insights into processing-microstructure-property relationships in these materials across a range of length scales.

Finite element modelling was employed to explore the hierarchical microstructural features of freeze-cast ferroelectrics that are beneficial for sensing, actuating and energy harvesting. These features include wall thickness, wall defects (i.e., pores within ceramic walls) and pore channel defects (i.e., ceramic grains within pore channels). Thin lamellar walls have been found to promote domain switching due to a lowered stress state. Wall defects were three times more effective than pore channel defects at preventing their neighbouring elements from polarising.

Using insights from the modelling work, adjustments to the classical Rayleigh analysis were proposed. This analysis is an important technique used to separate the intrinsic from the extrinsic contributions to piezoelectric and dielectric properties in ferroelectric materials. However, in porous freeze-cast ferroelectric ceramics, the assumptions for monolithic ceramics no longer hold, leading to these modifications. An adjustment to account for the relative density was implemented to decouple the dependency of the apparent intrinsic contribution on porosity. Furthermore, a median local field correction was proposed to account for the decrease in internal field intensity as the measurement direction misaligned with the orientation of the freeze-cast ceramic walls.

The optimisation of processing parameters for freeze-cast lead zirconate titanate was then outlined. Lower cooling rates produced thicker lamellar ceramic walls with a higher density of wall defects, leading to greater variation in functional properties. The optimised porous freeze-cast ferroelectric composite (25 vol.% solid loading, 4°C/min cooling rate, and 5-hour sintering) exhibited a piezoelectric strain coefficient d33 of 325 pC/N, retaining 86% of the piezoelectric coefficient of dense reference samples. The freeze-cast samples exhibited 2- and 4-fold improvements in the piezoelectric voltage coefficient and the figure of merit for energy harvesting, respectively, compared to their monolithic counterparts.

This Thesis advances the understanding of porous freeze-cast ferroelectric ceramics. By accounting for grain clamping effects and improving the separation of intrinsic and extrinsic contributions, this work contributes more accurate predictive models for advanced ferroelectric composites. It bridges the understanding gap between monolithic ferroelectric ceramics and composites. These findings directly advance the design and optimisation of more complex, reliable and efficient next-generation ferroelectric devices for actuating, sensing and energy harvesting applications.

Date of Award	10 Sept 2025
Original language	English
Awarding Institution	University of Bath
Supervisor	James Roscow (Supervisor), Chris Bowen (Supervisor) & Hamideh Khanbareh (Supervisor)