Further development of the software jGiXa for grazing incidence X-ray fluoresence analysis

Abstract

Combined Grazing incidence X-Ray fluorescence and X-Ray reflectometry analysis, as proposed by de Boer [1], needs a sophisticated software using global optimization algorithms to fit calculated curves for a specific sample-setup-model to measurement results of angle dependent fluorescence signals from ultra thin layers on a reflecting surface (e.g. Si-wafer). The software package jGiXa, developed by Ingerle [2] for this purpose, used a graphical user interface and optimization algorithms based on MatLab code and a Java based calculation of the spectra. This diploma thesis focuses on the further development of the jGiXa software. Therefor a Java based graphical user interface and global optimization algorithm were implemented, allowing jGiXa to work as a Java standalone software. This is advantageous in respect to portability, usability, maintainability, scalability, performance, etc. To further optimize the performance of the software a meta optimization was implemented to find ideal parameters for the global optimization algorithm. Furthermore a new model for intermediate layers was implemented, to allow sample models with a continuous transition between layers. To test the new software, comparison evaluations for different measurements were made.