- Ifeffit - millenia.cars.aps.anl.gov

XrayLarch Developer's / Topics in XAS Analysis meeting, Jan 29th
by Matthew Newville 20 Jan '26

20 Jan '26

Hi Folks, We have had nearly regular Zoom meetings on X-ray Larch Development over the past few years, but have not had one in several months. In discussions with Mauro and others, we would like to bring these meetings back and expand the scope to include focused discussions on selected topics and/or presentations of developments in XAS analysis and data science that might be of interest to Larch users and developers. To start, we’ve invited Athar Khodabakhsh from Helmholtz-Zentrum Berlin to speak on Machine Learning work with XAS, similar to a talk given at the recent DataXAS meeting at ETH Zurich (in association with Paul Scherrer Institute). This meeting will be: January 29, at 1500 UTC (4 PM Paris, 9 AM Chicago) Athar Khodabakhsh, Helmholtz-Zentrum Berlin, Convolutional AutoEncoder for Anomaly Detection and Chemical State Identification in XAS To avoid publishing a Zoom link on this mailing list, the link will be available at https://millenia.cars.aps.anl.gov/xraylarch/meetings/ several hours before the meeting begins. If anyone would like to attend or have the Zoom link, please feel free to contact me. Please let me know if you are interested in giving a presentation or have a suggestion for a topic for discussion. Hope to see you there, --Matthew Newville (newville(a)cars.uchicago.edu)

1 0

Larch 2026.1.0
by Matthew Newville 20 Jan '26

20 Jan '26

Hi Folks, Larch 2026.1.0 is now available and will begin to appear for automatic updates. There have been many fixes made since 2025.3.0 was released (in September). A few of the most important recent changes are: 1. The Larch Desktop folder now includes a shortcut for Jupyter Lab, running in the Python environment with Larch installed. This should make it much easier to run a notebook using Larch. It also includes a Terminal useful for running updates. 2. Many more Jupyter and Python examples are now available, including showing how to plot with wxmplot or Bokeh. Using wxmplot gives the same plots as created with Larix, shown in plot windows exterior to the browser, while using Bokeh shows plots embedded in the browser, but with more built-in interactivity than standard matplotlib plots. 3. In Larix, improvements in reading data, normalization, and copying normalization parameters continue. 4. In Larix, all the panels that “build model to fit data” (Feffit, Pre-edge Peaks, general purpose curve-fitting) now do a much better job of using the latest model for a particular data group and switching between these models when you switch groups. That is, if you have completely different Path models (say, for different edges) in one Session, the Feffit Panel now fully and correctly replaces the Feffit model when selecting each data group. 5. Feff Paths in the Feffit Panel now show a table with the Path geometry, including scattering angles. 6. More release information is at https://urldefense.us/v3/__https://github.com/xraypy/xraylarch/releases/tag… As usual, many of these changes came from Mauro Rovezzi. Thanks to all those who helped with code, bug reports, suggestions, or even just encouragement to continue in these challenging times ;) If you have any trouble, questions, or suggestions, please let us know. We’re planning to start up Larch developer Zoom meetings again, and to expand these discussions to other developed methods. I’ll send out a separate message about that in the next day or so. —Matt Newville

2 1

Help installing Demeter from source
by Joe Graves 16 Jan '26

16 Jan '26

Hello everybody, I'm a graduate student and new to XAFS analysis. I'm trying to install Demeter from source on Linux (Fedora 43). The instructions to do so from Bruce Ravel all point toward the github of Ifeffit, which as I'm sure you all know, doesn't exist anymore as ifeffit has been deprecated in favor of Larch. Are there any updated instructions for installing Demeter from source? Alternatively, can I install Demeter from source using the old instructions, but installing Larch when I would otherwise be instructed to install ifeffit? P.S. I would like to add that that Bruce Ravel's links to the ifeffit mailing list are also all broken. I was able to subscribe when I found the following link using a search engine: https://millenia.cars.aps.anl.gov/mailman3/lists/ifeffit.millenia.cars.aps.… Joe Graves PhD Student, Department of Chemistry Graduate Research Assistant, School of Energy Resources Member of the Moreau Lab

3 2

Query: abnormal FT-EXAFS peak at ~1.1A
by Huaming 24 Dec '25

24 Dec '25

Hello! I faced an issue with fitting of FT-EXAFS data (Rh@zeolite) with Artemis. There is one distinct peak located at ~1.1 angstron. The intensity is comparable to the nearby peak of Rh-O scattering. I wonder the peak (~1.1A) is a real peak or just noise. Raw data is enclosed. Another question is why S02 can be negative in data fitting with Artemis. Cheers, Huaming

3 2

There are some issues with fitting FeNi DAC parameters using fdmnes.
by 1000566384＠smail.shnu.edu.cn 20 Dec '25

20 Dec '25

Hello! I am a graduate student. When using Fdmnes to fit FeNi DAC, I found that the fitted peaks do not match the experimental peaks very well. The main issue is that the two transition peaks in the experiment merge into a single peak in the fitted curve. First of all, thank you very much for your help. If you could help me adjust the parameters, I would be happy to send you the experimental data for you to take a look. Best of luck!

1 0

Help for checking my EXAFS fitting
by cbmeherrem＠gmail.com 10 Dec '25

10 Dec '25

Hello, I am a graduate student working on Ge and Sn K-edge XAS measurements. I have performed some EXAFS fitting in Artemis, but I am not fully confident that my analysis is correct since my sample is a bit complex. I was wondering if anyone with expertise in EXAFS fitting might be willing to look over my results. I would be happy to email my fitting files and analyzed data if someone is available to provide guidance. Thank you very much for your time and help. Best Maharram

2 1

Data for Mn metal L-edge XAS
by Ismael Graff 15 Nov '25

15 Nov '25

2 1

Strange low-r peak in Pt foil EXAFS
by FAN Dongxiao 03 Nov '25

03 Nov '25

Dear All, My target is the L3-edge of Pt foil and Pt-contained high-entropy alloy. In the Pt foil L-edge FT-EXAFS (k range: 3-20 A-1), there is a strange peak lower than the first near neighboring peak. This low-r peak makes confused because I did not found in other metal foil spectra. Also this peak cannot be reduced by increasing the Rbkg value. When I decrease the k range, this strange peak disappears with phase shift. In the Pt-contained high-entropy alloy, this strange peak becomes more pronounced than the 1st NN. Can anyone can give me help interpret this strange peak? Best regards, Fan ******************* JASRI, Sping8 *******************

2 1

Shifting FDMNES spectra for comparison between multi-site and single-site runs
by oliel1995＠gmail.com 08 Oct '25

08 Oct '25

Hello everyone, I’m running FDMNES calculations, and I have the following situation: Run A: Multi-site calculation with 3 non-equivalent positions. FDMNES internally applies epsii shifts, and I merge the per-site spectra — the merged spectrum is correct internally, meaning the relative positions between sites are preserved. Run B: Single-site calculation with 1 non-equivalent position. My goal is to compare the merged spectrum from Run A with the single-site spectrum from Run B. The challenge: because FDMNES aligns energies internally per run, the merged spectrum from Run A is not correctly referenced to the same absolute energy scale as Run B. My question: What is the recommended way to shift the merged multi-site spectrum from Run A so that it can be meaningfully compared with a completely different single-site calculation? Is it acceptable to use the position of a peak B, WL, etc, for a rigid shift, and are there best practices for doing this while preserving the internal relative positions of the merged spectrum? Thank you! Sincerely, Olga Olga Filimonova Post-doc researcher - ID26 European Synchrotron Radiation Facility (ESRF), Grenoble, France

1 0

Symposium on Data-Driven Approaches in X-ray Absorption Spectroscopy
by Clark, Adam Hugh 08 Oct '25

08 Oct '25

Dear Colleagues, The upcoming Symposium on Data-Driven Approaches in X-ray Absorption Spectroscopy, is presently accepting abstracts for oral presentations and registration is open. The symposium aims at bringing together researchers in the field of x-ray absorption spectroscopy, machine learning, and computational modelling. This symposium will focus on recent developments in data-driven and machine learning techniques for the interpretation of XANES and EXAFS spectra, enabling new insights into materials structure and dynamics from large complex operando/in situ datasets. The program will feature four themed sessions: 1. Data-Driven Decomposition and Spectral Extraction 2. Machine Learning for XANES Prediction and Analysis 3. Advanced Approaches to EXAFS Analysis 4. Bridging Simulations and Experiments with Machine Learning Date: 5th-6th January 2026 Location: ETH Zurich Hönggerberg campus, Switzerland Abstract submission deadline: 15th October 2025 Symposium Webpage: https://urldefense.us/v3/__https://indico.psi.ch/event/dataxas2026__;!!G_uC… We warmly invite your participation in this symposium and look forward to your contributions and discussions. __________________________________________ Paul Scherrer Institut Adam Hugh Clark WLGA/219 Forschungsstrasse 111 5232 Villigen PSI Schweiz Telefon: +41563104151 E-Mail: adam.clark(a)psi.ch

1 0