LAUSR

Abstract The front face photothermal radiometry technique has been improved in order to estimate the thermal conductivity of thin films with better accuracy compared to existing techniques. The experimental procedure is based on the front face response to a nanoseconds laser pulse repeated periodically at high frequency, i. e., a Dirac comb waveform. Averaging the thermal response by considering thousands successive pulses allows improving largely the signal noise ratio. The unknown thermal properties and related experimental parameters are identified by minimizing the gap between the measured signal and the theoretical response that accounts with the pulse waveform, the repetition frequency and the detector transfer function. Minimization is first achieved by implementing first a simplex technique that gives an initial set of values to start the Metropolis–Hastings algorithm in a second step. Application of the proposed methodology is done considering amorphous GeTe film deposited on a Si wafer. It is shown that this experimental method as well as the implementation of the Bayes minimization technique allows to identify the thin film intrinsic thermal conductivity with high accuracy considering some uncertainty on the other parameters assumed to be known.