EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Wavelet and scattering transform descriptors for batch quality assessment of doped hydroxyapatite nanocatalysts: a multiscale feature-extraction framework for X-ray diffraction fingerprints

Fatemeh Shirvani, Farzaneh Moghaddar, Mojtaba Hasanlu

Advanced Manufacturing Research · 2026

DOI: 10.21595/amr.2026.26718 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Advanced Manufacturing Research, 2026. Verfasst von Fatemeh Shirvani und 2 weiteren Autor:innen. Die Arbeit nennt 44 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Heterogeneous nanocatalysts encode compositional and structural information in complex spectroscopic patterns that are difficult to quantify with single-scale analysis. In this work, a wavelet-based multiscale descriptor framework is applied to the X-ray diffraction (XRD) patterns of pure and cation-substituted hydroxyapatite nanocatalysts (Ag-, Co-, and Sr-HAP) previously reported for the selective hydration of aromatic nitriles to amides. The Discrete Wavelet Transform (DWT) and the Wavelet Scattering Transform (WST) are computed for each diffractogram and compared in terms of feature dispersion, translation invariance, and inter-sample discriminability. The DWT captures localized peak-intensity variations but retains a wide dynamic range (variance ≈ 6.2×10 12 ) that is sensitive to amplitude and translation. The WST compresses the descriptor space by roughly ten orders of magnitude (variance ≈ 1.2×10 2 ) while preserving the sample ordering induced by cation substitution and remaining stable under small deformations. These properties yield a compact, mathematically defined, translation-invariant representation of doped-HAP diffraction fingerprints that is directly compatible with downstream chemometric and supervised-learning workflows. Because the descriptors are deterministic and fully reproducible, the framework is proposed as a candidate approach for future batch-to-batch quality assessment during doped-nanocatalyst production – a materials-processing concern relevant to manufacturing quality control – though this application remains untested against actual production-scale batch-variation data. The novelty of the present work lies in the systematic construction and quantitative comparison of DWT and WST descriptors for a family of doped nanocatalysts, providing a reproducible feature-extraction pipeline for structurally similar catalyst systems. The main limitation is the small number of samples analyzed (four catalyst compositions), which restricts the present study to descriptor characterization rather than predictive modelling; extension to larger catalyst libraries is required for full data-driven prediction of catalytic activity.

Zitieren

APA

Shirvani, F., Moghaddar, F., & Hasanlu, M. (2026). Wavelet and scattering transform descriptors for batch quality assessment of doped hydroxyapatite nanocatalysts: a multiscale feature-extraction framework for X-ray diffraction fingerprints. Advanced Manufacturing Research. https://doi.org/10.21595/amr.2026.26718

Vancouver

Shirvani F, Moghaddar F, Hasanlu M. Wavelet and scattering transform descriptors for batch quality assessment of doped hydroxyapatite nanocatalysts: a multiscale feature-extraction framework for X-ray diffraction fingerprints. Advanced Manufacturing Research. 2026. doi:10.21595/amr.2026.26718

BibTeX
@article{Shirvani2026,
  author = {Shirvani, Fatemeh and Moghaddar, Farzaneh and Hasanlu, Mojtaba},
  title = {Wavelet and scattering transform descriptors for batch quality assessment of doped hydroxyapatite nanocatalysts: a multiscale feature-extraction framework for X-ray diffraction fingerprints},
  journal = {Advanced Manufacturing Research},
  year = {2026},
  doi = {10.21595/amr.2026.26718}
}
Kostenlos starten →