Investigating the physics of ferroelectric domains and domain walls using novel scanning probe methods

Abstract

Ferroelectric materials offer promising functionalities for future technologies, with their spontaneous polarisation facilitating the creation of memory devices and the unique behaviour often exhibited by their nanoscale domain walls. Scanning Probe Microscopy has been one of the key tools for investigating these materials in recent years. This thesis outlines the use of three novel investigations of domains and domain walls using various scanning probe methods. Kelvin Probe Force Microscopy has been used to reliably map the spatial distribution of the electrostatic potential across conducting domain walls in x-cut lithium niobate. Head-to-head and tail-to-tail domain walls, which exhibit n and p type conduction respectively, have been created between coplanar electrodes. The morphology of the switched domain has been established using Tomographic Piezoresponse Force Microscopy and Transmission Electron Microscopy, which show that the respective charged sections meet subsurface, forming a p-n junction. The domain wall has been modelled as a two-dimensional hemiellipse and its resistive properties have been extracted based on the magnitude of the polar discontinuity at every point along its surface, which reveal a peak at the centre of the interelectrode gap. This is used to rationalise a peak in the measured electric field across the junction, showing that there is no need to consider the physics associated with depletion regions in standard semiconductor p-n junctions, in order to explain the behaviour of those which form in domain walls.

In addition, the fundamental contrast mechanism in the scanning probe technique, Charge Gradient Microscopy has been systematically investigated. Various sets of ferroelectric domain structures were studied: 180° c+/c- domains in periodically poled lithium niobate, 180° in-plane polarised domains in x-cut lithium niobate and a1-a2 and a-c domains in barium titanate. Kelvin Probe Force Microscopy was used to map the surface potential across each set of domains and was compared with the spatial integral of the passive currents mapped using Charge Gradient Microscopy. The results demonstrate that the current signals reflect spatial variations in the local electrostatic potential, rather than in bound charge.

Finally, the domain structure in lead germanate has been investigated in three dimensions, using Tomographic Piezoresponse Force Microscopy. Phase-field modelling shows that the domains in this material bifurcate along two different axes, with saddle point features emerging in the domain wall surface. Three dimensional imaging and subsequent recreation of the domain microstructure has allowed these saddle points to be detected. The lack of electrical conductivity at supposedly charged portions of these domain walls, combined with this mutual domain bifurcation, suggest that the polarisation may rotate to significantly reduce the polar discontinuity at head-to-head and tail-to-tail sections of the domain wall. Hence, there is no need for screening charges which often contribute to conduction in many other uniaxial ferroelectric domain walls which support polar discontinuities.

Date of Award	Dec 2023
Original language	English
Awarding Institution	Queen's University Belfast
Sponsors	Northern Ireland Department for the Economy
Supervisor	Raymond McQuaid (Supervisor) & Marty Gregg (Supervisor)