Development of chemometric models to classify solid-state U materials by micro-Raman spectroscopy

Area of Science:

• Analytical Chemistry
• Materials Science
• Spectroscopy

Background:

• Monitoring the peaceful use of nuclear material requires accurate identification of uranium particles in environmental samples.
• Distinguishing various uranium compounds and differentiating them from matrix materials is crucial for nuclear forensics and industrial applications.

Purpose of the Study:

• To develop a chemometric model for classifying four distinct uranium compounds using Raman spectroscopy.
• To assess the model's ability to differentiate uranium particles from matrix materials and outliers.

Main Methods:

• Development of a Soft Independent Modeling of Class Analogy (SIMCA) library for Raman spectral data.
• Utilizing both unsupervised and supervised chemometric models to account for spectral variability.
• Classification of α-U3O8, UO2, UO2(NO3)2·6H2O (UNH), and UO2O2·4H2O (studtite) particles.

Main Results:

• A supervised SIMCA model demonstrated good sensitivity and high specificity for classifying uranium compounds.
• Spectral variability related to particle size, hydration, and oxide phase was observed within classes.
• The method successfully distinguished uranium materials from matrix particulates like flint clay.

Conclusions:

• Raman spectroscopy combined with chemometrics provides a reliable method for distinguishing uranium materials.
• The developed approach offers a rapid, non-destructive technique for characterizing uranium compounds in environmental and forensic contexts.