Transient methods are widely used to determine the thermal transport properties of dielectrics and other materials. We investigated the limits and potentials of the transient hot-strip (THS) method for the simultaneous determination of the individual thermal conductivity and the thermal diffusivity of layered composites. The experimental data are analysed by solving the related inverse problem of parameter identification which is formulated as an output least-squares problem. The forward problem is governed by the finite-element solution of the heat conduction equation. As expected, classical difficulties of inverse problems such as ill-posedness and, especially, the verified existence of several local minima complicate a unique solution. The proposed identification algorithm considers two strategies to achieve success. First, the transport parameters of the individual layers are determined successively, beginning with the layer next to the heater. And second-which is surprising-the measurement time has to be sufficiently long so that the THS signal contains information about the sample surface. Otherwise, the thermal properties cannot be uniquely determined and the iteration may lead to a local minimum. The identification procedure is widely verified using simulated data. Results for experimental data are also given.
Read full abstract