Mini-Workshop on ePDF

Ulm University

On the 28th and 29th of March 2018 the University of Ulm hosted a workshop on electron Pair-Distribution Function Analysis (ePDF). This workshop was organized by Tatiana Gorelik and brought together 17 leading experts in ePDF from European and US Universities. Discussions during the two days were organized in a highly interactive format where presentations served as launch pads to exchange views and get inspiration from fellow researchers. This summary aims to inform the interested scientific community about the workshop and its discussions.

The discussions during the workshop fed into concrete agreements on follow up actions. The participants agreed on the need for a standard data processing recipe for ePDF users. As a key building block towards the standardized data processing protocol, a round robin test on a known sample using different instrumentation and different data processing approaches was proposed. The results would reveal any differences in findings that are clearly traceable to the chosen methodology and may be used to justify the necessary of certain processing steps for diffraction data recorded. Finally, the need for future, eventually regular, meetings was underlined.

Detailed Summary of the workshop

At the workshop, discussions were structured into five sessions. The first session focused on specific applications of ePDF. Viktoria Kovács Kis (Institute for Technical Physics and Materials Science, Budapest, Hungary) presented a wide spectrum of mineralogical applications, specifically pointing out the appearance of stray scattering in SAED patterns, making them less favorable for ePDF analysis compared to nano-beam diffraction mode. Konstantin Borisenko (University of Oxford, UK) demonstrated the power of the ePDF method by direct evidence of oxide-layers formation on intermetallic glass films. A software package for the data analysis eRDF Analyser was presented making a smooth transition to the second session with a more practical focus on various software packages available for ePDF analysis. The applications and the demonstration of ePDF suite (Partha Pratim Das, Alejandro Gomez Perez) were followed by a demo of SUePDF by Cheuk-Wai Tai (Stockholm Universit, Sweden). Finally, the demonstration of the package ProcessDiffraction for ePDF analysis was given by Janos Labar (the Institute for Technical Physics and Materials Science, Budapest, Hungary). Workshop participants appreciated in particular the opportunity to compare these packages in a concise format and better understand their relative strengths and characteristics.

The third session comprised a series of special topics. Xiaoke Mu (KIT, Karlsruhe, Germany) presented his work on laterally resolved ePDF supported by multivariate statistical analysis. Apart from this presentation, Xiaoke brought up the fundamental topic of multiple scattering effect on ePDF data and made a strong contribution on the scattering intensity redistribution in amorphous materials due to the multiple scattering. Christopher Wright (Columbia University, US) shared with the group new algorithms being developed for ePDF suite. Finally, the effect of inelastic scattering and its appearance in PDFs as well as practical examples and visions on preferred orientation treatment were addressed by Tatiana Gorelik (Ulm, Germany).

An interactive discussion session originally planned in the program (fourth session) was integrated into the workshop as discussions were rich on all items and participants felt this to be the most useful format.

The second day of the workshop was devoted to the quantitative treatment of ePDF data in the form of PDF refinement (fifth session). Reinhard Neder (Erlangen, Germany) gave an introduction to refinement of PDF data followed by an example of an ePDF refinement separating the structural and instrumental parameters in DISCUS including a deep explanation of the meaning of the parameters, their correlations and their effect in X-ray and neutron PDFs. A particular focus was given to possible false interpretation of low-r artefacts due to the Qmax clipping of the scattering data.

An introductory lecture notes on ePDF can be found here.

Group photo of the ePDF workshop