Dresden 2026 – wissenschaftliches Programm

Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe

DS: Fachverband Dünne Schichten

DS 20: Poster

DS 20.15: Poster

Donnerstag, 12. März 2026, 18:30–20:30, P2

XRD Data Analysis and Curve Fitting Using Differential Evolution in Python — •Simon Lorch, Tobias Pollenske, and Joachim Wollschläger — Institute of Physics, University of Osnabrück, Barbarastr. 7, 49076 Osnabrück, Germany

Precise evaluation of X-ray diffraction (XRD) data is essential for understanding thin-film and interface structures at the atomic scale. In this contribution, we present a Python-based framework for automated fitting and quantitative analysis of experimental diffraction data using Differential Evolution (DE) as a global optimization strategy. The method minimizes the difference between measured and simulated intensity curves by evolving parameter sets such as layer thickness, density, and interfacial roughness toward the global minimum of an error function. Unlike conventional structure-factor-based approaches, the model deliberately omits detailed atomic scattering factors and instead focuses on the optimization of macroscopic parameters to reproduce the experimental curve shape. This allows a robust and efficient fitting of complex multilayer systems even in cases where structural details or phase information are incomplete. Compared to traditional gradientbased approaches, DE shows superior robustness against local minima and enables reliable fitting of complex multilayer systems with correlated parameters. The implementation allows user-defined boundary conditions, parallel evaluation of population members, and integration of physical models for scattering amplitude and Debye-Waller damping. This open and modular approach shows how evolutionary algorithms can make XRD data analysis more reproducible and accurate.

Keywords: XRD; Differential Evolution; global optimization; curve fitting; thin-film analysis