An open access, integrated XAS data repository at Diamond Light Source

Giannantonio Cibin,Diego Gianolio,Stephen A Parry,Tom Schoonjans,Oliver Moore,Rachael Draper,Laura A Miller,Alexander Thoma,Claire L Doswell,Abigail Graham

doi:10.1016/j.radphyschem.2019.108479

Abstract

The analysis of reference materials is a fundamental part of the data analysis process, in particular for XAS experiments. The beamline users and more generally the XAS community can greatly benefit from the availability of a reliable and wide base of reference sample spectra, acquired in standard and well-characterized experimental conditions. On B18, the Core EXAFS beamline at the Diamond Light Source, in the past years we have collected a series of XAS data on well characterized compounds. This work constitutes the base for a reference sample database, available as a data analysis tool to the general XAS community. This data repository aims to complement the bare spectroscopic information with characterisation, preparation, provenance, analysis and bibliographic references, so improving the traceability of the deposited information. This integrated approach is the base of success and wide distribution of data repositories in other fields, and we hope it will provide on one side a precious facility for the training of students and researchers new to the technique, and at the same time encourage the discussion of best practices in the data analysis process. The database will be open to the contribution of experimental data from the user community, and will provide bibliographic reference information and access control.

Highlights

X-Ray Absorption Spectroscopy is increasingly used as a scientific and diagnostic tool, as materials’ technology and research is increasingly moving to the investigation of quasi-crystalline, amorphous and biologic materials in real use conditions
Initiatives for supporting the data analysis of XANES and EXAFS data based on deep learning algorithms are being proposed (Zheng et al, 2018)
Data will be by default available and users are expected to upload XAS spectra as files that adhere to the XAS Data Interchange (XDI) format specification (Ravel and Newville 2015; 2016)

Summary

Introduction

X-Ray Absorption Spectroscopy is increasingly used as a scientific and diagnostic tool, as materials’ technology and research is increasingly moving to the investigation of quasi-crystalline, amorphous and biologic materials in real use conditions. The characterisation of nanostructured systems in particular relies on non-crystallographic methods such as diffuse scattering, direct imaging or spectroscopic analysis Among the latter, XAS has the important benefit of providing element-specific electronic and structural information, and if the element and absorption edge studied falls in the hard X-ray regime, in-situ and operando experiments are possible, replicating full-scale process conditions. Initiatives for supporting the data analysis of XANES and EXAFS data based on deep learning algorithms are being proposed (Zheng et al, 2018) This approach relies for quantitative analysis, more than on the ability to accurately model the XAS signal, on the availability of large spectral libraries. There is a clear need for access to well-characterized reference XAS data, acquired on large families of reference compounds, covering multiple absorption edges and elements on most part of the periodic table

Data sources

Data format

Web site

Initial dataset

Conclusions