Effect of Experimental Variables on Flash Thermal Diffusivity Data Analysis

Abstract

Flash thermal diffusivity data is usually analyzed with a thermal model which assumes axial heat conduction and uniform illumination of the flashed surface. For high accuracy data reduction, it becomes important to bound the errors caused by radial heat flow and by non-uniform laser beam profiles. These effects are examined analytically for a case in which the incident laser beam is confined to a radius smaller than the sample radius. The dependence of the output of an averaging detector on the magnitude of the radial heat transfer coefficient is presented and the linear dependence of radial and axial loss sensitivity coefficients is discussed. From this discussion, we conclude that inclusion of radial loss effects in analysis of the thermal response of multilayer structures is not important unless the radial loss factor is very large. Analytical results are presented for the temperature vs. time response of a two layer composite sample with interfacial thermal resistance and high thermal losses at the sample faces. The use of these results to reduce data for two multilayer samples is presented to show the utility of new data reduction techniques.