Bereiche | Tage | Auswahl | Suche | Aktualisierungen | Downloads | Hilfe
MM: Fachverband Metall- und Materialphysik
MM 13: Data-driven Materials Science: Big Data and Workflows I
MM 13.3: Vortrag
Dienstag, 10. März 2026, 10:45–11:00, SCH/A251
A Python-based workflow for phase identification and mapping via Raman spectroscopy — •Felix Drechsler1, Mahnaz Mehdizadehlima2, Cameliu Himcinschi1, David Rafaja2, and Jens Kortus1 — 1TU Bergakademie Freiberg, Institute of Theoretical Physics, D-09599 Freiberg, Germany — 2TU Bergakademie Freiberg, Institute of Materials Science, D-09599 Freiberg, Germany
Raman spectroscopy is a powerful tool for identifying phases and compounds. It is highly sensitive to both chemical and structural variations and is particularly attractive due to its minimal sample preparation requirements. This makes it well suited for fast phase identification and the determination of spatial phase distributions, complementing established techniques such as X-ray diffraction, element mapping via X-ray spectroscopy, and phase mapping using electron backscatter diffraction. However, phase-resolved analysis of Raman mappings remains challenging, as it involves processing high-dimensional data matrices and robust identification relies on comparing the complete spectrum with references.
In this talk, we present a Python-based workflow developed to address this challenge. The approach employs multivariate analysis methods and integrates mathematical similarity metrics with reference datasets to enable robust and reproducible phase identification. The proposed workflow provides an efficient way to construct Raman phase maps and offers a valuable tool for material characterization.
Keywords: Raman spectroscopy; Raman imaging; Python; Spinel phases