Accurate contact resistance characterization for thermal conductivity measurement with the Heat Flow Meter method

Abstract

A common method to measure the thermal conductivity of low-conductive materials is to impose a known temperature difference across the thickness of the specimen, and to measure the resulting heat flow once steady-state conditions are reached. In particular, international standards, i.e. ASTM C518 and ISO 8301, define the characteristics of the guarded Heat Flow Meter apparatus and its measurement procedure. However, the actual measured quantity is the overall thermal resistance, which is given by the series of the contact resistance between the specimen and the temperature-controlled clamps, and the resistance of the specimen itself. Thus, the contact resistance must be correctly quantified in order to retrieve an accurate measurement. To this end, common practices are either to rely on a database of known contact resistances for material classes, or to use the “double thickness” method, which allows to eliminate the contribution of the contact resistance by carrying out the measurement on two specimens of the same material, but different thickness. While the first method is rather useless for accurate and reliable measurements, especially of unknown or innovative materials, the latter works only if the contact resistances, and therefore the surface finish, are the same for every surface of the specimen set. This paper presents an analysis of the uncertainties associated with the evaluation of the contact resistance carried out on several samples, and proposes a method to reduce such uncertainties, i.e. by inserting elastic thermal pads of known thermal conductivity between the specimen and the instrument. The results of the validation of the method are also shown, with an analysis on the improvement of the measurement accuracy for specimens with high roughness and irregular surfaces, or with conductivities beyond the instrument declared range.