Multivariate statistical analysis of atom probe tomography data

Area of Science:

• Materials Science
• Data Analysis
• Spectroscopy

Background:

• Atom Probe Tomography (APT) generates complex, high-dimensional datasets.
• Multivariate statistical analysis offers potential for extracting meaningful information from APT data.
• Principal Component Analysis (PCA) is a powerful technique for dimensionality reduction and pattern recognition.

Purpose of the Study:

• To investigate the application of Principal Component Analysis (PCA) for analyzing spectrum imaging data from Atom Probe Tomography (APT).
• To demonstrate the utility of PCA in identifying and characterizing microstructural features in different materials.
• To discuss the potential and limitations of PCA for quantitative analysis of APT datasets.

Main Methods:

• Application of Principal Component Analysis (PCA), a multivariate statistical method.
• Analysis of two distinct Atom Probe Tomography (APT) datasets.
• Mathematical description of the PCA method for APT data interpretation.

Main Results:

• PCA successfully distinguished matrix and precipitate phases in a Fe-Cr-Al-Ti steel.
• PCA qualitatively identified nm-scale, core-shell Cu-enriched precipitates in irradiated reactor pressure vessel steel.
• The analysis provided a simple and intuitive description of the identified phases and structures.

Conclusions:

• PCA is a valuable tool for the qualitative analysis of APT data, simplifying the interpretation of complex microstructures.
• Further development is needed to fully leverage PCA for quantitative analysis of APT datasets.
• PCA shows promise for advancing materials characterization using APT.