Inquiry Regarding PDFgetX3 – Multiple Phase Analysis

[Maximiliano Jesús Moreno Zapata]

Dear Professor Simon Billinge,

I hope this message finds you well.

My name is Maximiliano Zapata, and I am currently working on atomic pair distribution function (PDF) analysis for materials containing more than one crystalline phase, specifically different polymorphs of ZrO2. I have been using PDFgetX3 (v2.2.0) to process experimental X-ray diffraction data into PDFs.

While the software works very well for single-phase systems, I have encountered conceptual and practical difficulties when attempting to process data from samples containing multiple phases. In particular, I would like to ask:

Does PDFgetX3 support the direct incorporation of more than one phase with distinct compositions or densities within a single .cfg file?

If not, what would be the recommended workflow for properly accounting for multi-phase mixtures (e.g., monoclinic, tetragonal, and cubic ZrO2) during the PDF transformation step?

Is it acceptable to use a weighted-average composition and density for approximate treatment? Or would this compromise the fidelity of the resulting G(r)?

I am a beginner in this area, but highly interested in research involving X-ray diffraction and scattering techniques. I would be most grateful for any guidance or references you could provide regarding this topic.

Thank you very much for your time and for making such powerful tools available to the scientific community.

Best regards,

Maximiliano J. M. Zapata.

[Andrew Yang]

Hi Maximiliano, welcome to the PDF community!

The composition is used when going from the I(Q) to the S(Q). In this step, the intensity curve is normalized by the atomic form factors, though only the average and second moment of these quantities are used. As such, you should use the weighted-average composition as you suggested. You should not need the atomic density when computing the PDF from experimental I(Q) using PDFgetx3.

Now let's say you want to now determine the structures in your PDF. Here, you should use fitting software like PDFgui or diffpy-CMI (recommended if you know Python). In diffpy-CMI, you can fit a PDF to multiple structural models (e.g. phases) at once. This is done after obtaining the PDF. You can find examples for two-phase fits here (https://www.diffpy.org/diffpy.srfit/examples.html).

-Ay

On Thu, Oct 9, 2025 at 10:42 AM Maximiliano Jesús Moreno Zapata <maximilia...@ufes.br> wrote:

Dear Professor Simon Billinge, I hope this message finds you well. My name is Maximiliano Zapata, and I am currently working on atomic pair distribution function (PDF) analysis for materials containing more than one crystalline phase, specifically

ZjQcmQRYFpfptBannerStart

This Message Is From an External Sender

This message came from outside your organization.

 

ZjQcmQRYFpfptBannerEnd

--
You received this message because you are subscribed to the Google Groups "diffpy-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to diffpy-users...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/diffpy-users/846eb55e-d49f-4872-a23d-5644623b02e7n%40googlegroups.com.

[Max Moreno]

Hi Ay, thank you very much for the warm welcome and for the clear explanation!

Your clarification about the composition makes perfect sense. I will proceed using the weighted-average composition when going from I(Q) to S(Q), and it’s good to know that the atomic density is not required when processing experimental data in PDFgetX3.

Regarding structure determination, I’ll take your suggestion and work with PDFgui first, and then explore diffpy-CMI for multi-phase fits — especially since my samples may contain more than one crystalline component. The link you shared for two-phase examples is very helpful.

Thanks again for your guidance. I really appreciate your support!

Best,
Maximiliano