Youngs, Ian John; (2002) Electrical percolation and the design of functional electromagnetic materials. Doctoral thesis (Ph.D.), University College London (United Kingdom).

Preview

Text Electrical_percolation_and_the.pdf Download (42MB) | Preview

Abstract

Composite materials with tailored electromagnetic properties are of significant technological importance for a diverse range of applications, including electromagnetic shielding and human tissue simulants. This thesis investigates the use of critical behaviour near the percolation threshold, as observed in conductor-insulator composites, for the design of improved electromagnetic materials. The work considers a process that is more efficient than other current methods for the analysis of experimental data and ultimately the design of electromagnetically-functional materials. This is based on the use of a Genetic Algorithm. A literature review is presented which leads from the fundamental polarisation and conduction processes in materials to models for the frequency, temperature and filler fraction dependencies of their electromagnetic properties. The largest contributions come from effective medium and percolation theories. Composites containing fillers with a range of conductivity and particulate size (in both the micron and nanometer ranges) have been studied. The extreme range of properties exhibited by these materials challenges the performance limits of current broadband measurement techniques. The experimental results also provide a severe test of effective medium theories. Whilst these continue to be widely used in the literature for conductor- insulator composites, they fail for filler contents above ten percent. Percolation theory is shown to provide a significantly improved fit to the filler fraction and frequency dependence. However, accurate and generalised methods for determining the percolation threshold in real materials are not well established. A trend is observed between the accuracy of a model and the level of analysis required on the composite material under investigation. The question of at what point an empirical study becomes the most efficient design option remains important and unanswered. Nevertheless, the proposed genetic algorithm-based analysis method should enable a significantly shorter lead time for empirical studies, since frequency domain data from as few as two formulations can be used to extract all the parameters required to map the entire frequency and filler fraction domain for a given two component system. Conventional methods rely on a minimum of five (and typically many more) formulations to determine these parameters from dc conductivity measurements.

Download activity - last month

Export as XMLCSVJSON

Download activity - last 12 months

Export as XMLCSVJSON

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item