Global estimation of thermal parameters from a picoseconds thermoreflectometry experiment

A method for identifying thermal properties of a stack constituted from a metallic thin layer deposited on a silicon substrate is proposed. Data are acquired using the picoseconds time resolved pump-probe technique. A Bayesian technique based on the Monte Carlo Markov Chain is implemented in order to identify simultaneously the thermal conductivity of the layer, the thermal resistance at the interface between the layer and the substrate and the extension of the heat source at the initial time. It is demonstrated that, despite to the correlation between sensitivity functions on the investigated time range, each parameter is accurately identified.