Quantum genetic algorithm assisted high speed and precision pump-probe thermoreflectance characterization of micro-/nano-structures

The rapid and precise extraction of thermophysical properties remains an enormous challenge in pump-probe thermoreflectance, and quantum algorithms present significant potential for addressing thermal transport problems. This paper introduces an innovative application of the quantum genetic algorithm to the fitting of experimental data from pump-probe thermoreflectance. Initially, a physical model based on the thermal transmission matrix is developed to compute the heat transport of multilayer micro-/nano-structures. The quantum genetic algorithm is then employed for multiparameter optimization in the extraction of thermophysical properties and tested on three samples. Both simulation and experimental results demonstrate that the quantum genetic algorithm assisted pump-probe thermoreflectance, capable of extracting thermophysical properties with high precision in just one minute, is approximately 30 times faster than the conventional method and 10 times faster than the genetic algorithm. Furthermore, measurement sensitivity and algorithmic randomness are analyzed. The reasons behind the phenomenon of equivalent fitting results and the indication of non-convex optimization are also presented.