Leveraging Machine Learning for Advanced Nanoscale X-ray Analysis: Unmixing Multicomponent Signals and Enhancing Chemical Quantification

TL;DR

This paper introduces PSNMF, a novel machine learning-based method that improves chemical quantification in nanoscale X-ray analysis by unmixing signals and reducing noise in EDX spectroscopy data.

Contribution

The paper presents PSNMF, a new non-negative matrix factorisation approach tailored for EDX spectral data, enhancing phase identification and noise reduction in nanoscale analysis.

Findings

PSNMF accurately retrieves phase spectral signatures from synthetic and experimental data.

Datasets reconstructed with PSNMF show lower noise and higher fidelity than PCA-based methods.

PSNMF effectively addresses challenges of noisy and overlapping signals in EDX spectroscopy.

Abstract

Energy dispersive X-ray (EDX) spectroscopy in the transmission electron microscope is a key tool for nanomaterials analysis, providing a direct link between spatial and chemical information. However, using it for precisely determining chemical compositions presents challenges of noisy data from low X-ray yields and mixed signals from phases that overlap along the electron beam trajectory. Here, we introduce a novel method, non-negative matrix factorisation based pan-sharpening (PSNMF), to address these limitations. Leveraging the Poisson nature of EDX spectral noise and binning operations, PSNMF retrieves high quality phase spectral and spatial signatures via consecutive factorisations. After validating PSNMF with synthetic datasets of different noise levels, we illustrate its effectiveness on two distinct experimental cases: a nano-mineralogical lamella, and supported catalytic nanoparticles. Not only does PSNMF obtain accurate phase signatures, datasets reconstructed from the outputs have demonstrably lower noise and better fidelity than from the benchmark denoising method of principle component analysis.