Bayesian Estimation of Thermal Properties Using Periodically Pulsed Photothermal Radiometry: A Focus on Interfacial Thermal Resistances between Layers

Abstract

Pulsed periodic photothermal radiometry (PPTR) allows the investigation of multilayered samples with layer thicknesses of a few hundred nanometers over a duration of a few nanoseconds. The link between theoretical calculations and experimentation is made using a Bayesian approach based on the Metropolis–Hastings algorithm. The accuracy of the PPTR method requires precise calibration to provide an accurate proper emission measurement. Furthermore, the normalization time needs to be optimized through sensitivity analysis. Without this sensitivity analysis, it is impossible to simultaneously estimate the interfacial thermal resistances in a multilayered sample. The method is tested on multilayered samples composed of platinum (Pt), titanium nitride (TiN), silicon nitride (SiN), and silicon (Si). The parameters identified are interfacial thermal resistances between Pt–TiN and TiN–SiN.