Restoration of the original depth distribution from experimental SIMS profile using the depth resolution function in framework of RMR model

Abstract

In this paper, the problem of depth profiling analysis of nanoscale heterostructures containing doped delta layers and quantum wells using the SIMS method is considered. Based on computer simulation data and previously obtained experimental data, we demonstrated that the RMR model most accurately and completely describes the redistribution of the analyzed element in ultrathin layers that occurs during ion sputtering. A comparative analysis of the surface roughness–ion mixing–recoil implantation (RMR) model with MRI (mixing-roughness-information depth) and UDS (up-and-down slope) models proposed by Hoffman and Dowsett, respectively, was performed. It was shown that the introduction into the MRI model of a parameter describing some layer of constant thickness, in which the components of the analyzed layer and matrix elements are uniformly mixed, is not quite justified. It is concluded that during depth profiling of a monoatomic layer, the center of mass of this layer shifts away from the surface, as predicted by the RMR model, rather than toward the surface, as predicted by the MRI model. It is found that preferential sputtering does not affect the experimental depth distribution of elements obtained by the SIMS method.