# Automated elucidation of crystal and electronic structures in boron nitride from X-ray absorption spectra using uniform manifold approximation and projection

**Authors:** Reika Hasegawa, Arpita Varadwaj, Alexandre Lira Foggiatto, Masahito Niibe, Takahiro Yamazaki, Masafumi Horio, Yasunobu Ando, Takahiro Kondo, Iwao Matsuda, Masato Kotsugi

PMC · DOI: 10.1038/s41598-025-18580-z · Scientific Reports · 2025-11-10

## TL;DR

This paper introduces an automated method using UMAP to analyze XAS data, enabling precise classification of crystal and electronic structures in boron nitride.

## Contribution

The novel contribution is demonstrating that UMAP outperforms PCA and MDS in analyzing XAS spectra for structural and electronic characterization.

## Key findings

- UMAP effectively captures key features of XAS spectra better than PCA and MDS.
- UMAP enables precise classification of BN spectra by crystal system, defect type, and structural dimensionality.
- Correlating UMAP embeddings with electronic states reveals nontrivial electronic properties.

## Abstract

X-ray absorption spectroscopy (XAS) provides valuable information on the structure and electronic states of materials. Its spectral profiles are complex, influenced by crystal structure and defects, and require extensive expertise for conventional analysis. This study presents an automated XAS analysis approach using various dimension reduction methods to characterize crystal and electronic structure. We show that, Uniform Manifold Approximation and Projection (UMAP), outperforms traditional methods such as Principal Component Analysis (PCA) and Multidimensional Scaling (MDS) in capturing the key features of complex spectra. Applying UMAP to simulated XAS spectra of boron nitride (BN), high-dimensional data can be precisely classified by crystal system, defect type, and structural dimensionality. By correlating UMAP-derived embeddings with electronic states, we identify nontrivial electronic properties. The successful application of this model to experimental data demonstrates its potential for autonomous structural identification, offering new possibilities for data-driven materials design and advancing novel material development.

The online version contains supplementary material available at 10.1038/s41598-025-18580-z.

## Full-text entities

- **Chemicals:** BN (MESH:C017282)

## Full text

_Full body text omitted from this summary view._ Fetch the complete paper as Markdown: https://tomesphere.com/paper/PMC12603293/full.md

## Figures

7 figures with captions in the complete paper: https://tomesphere.com/paper/PMC12603293/full.md

## References

2 references — full list in the complete paper: https://tomesphere.com/paper/PMC12603293/full.md

---
Source: https://tomesphere.com/paper/PMC12603293