Nondestructive evaluation of thermal-process-induced change of Sn/Cu/Ni plating thin film stacks based on EDX measurement and Bayesian inference

Abstract A new nondestructive evaluation scheme for layer structures of thin film stack has been demonstrated using energy-dispersive X-ray spectroscopy and Bayesian inference. As an example, differences in the layer structure in a Sn/Cu/Ni plating film stack before and after applying a thermal process have been nondestructively detected in this paper. Using a Bayesian inference framework, it is possible to naturally select a layer structure model which fits the experimental data and has minimum complexity. This new method makes it possible to evaluate changes in the depth profile over time at the same location on the sample, which is useful for understanding phenomena in various product fabrication and deterioration processes.