Commission on Crystallographic Computing

Pair distribution function: DiffPy

Christopher L. Farrow
Michigan State University, East Lansing, MI, USA

DiffPy is an open-source library of Python software modules for analysis of diffraction data and the study of atomic structure of materials. It contains the publicly released software modules that have been developed by the Diffraction sub-group of the NSF-funded DANSE software project ( The DiffPy library is designed as a modular collection of packages, which can be selected during the installation procedure. Currently, the stable packages perform real-space structure analysis from atomic pair distribution function (PDF) data. This yields information about local and nano-scale structure in materials, which complements crystallographic data that yields the long-range order. In the future the DiffPy library will also contain packages for Rietveld refinement and Complex Modeling, which are still in development. These are currently not released but can be viewed by downloading from the DANSE development subversion repository (see

The DiffPy library has been used to build two end-user applications - PDFfit2 and PDFgui, which are available for download at PDFfit2 is a program for PDF modeling and structure refinement to experimental PDF data. PDFgui is a graphical interface built on PDFfit2 calculation engine. PDFgui has extensive facilities for organization of related fits and for common analysis tasks such as data extraction and plotting.

In the first part of my talk I will describe the PDF method and demonstrate features of PDFgui. In the second part of my talk I will discuss the design and implementation of PDFfit2 and PDFgui. Developing DiffPy packages, SrRietveld and SrFit, will also be discussed in this context.

DANSE, is a software construction project to provide data analysis tools for neutron scattering experiments that will be carried out at the Spallation Neutron Source at Oak Ridge National Laboratory.

These pages are maintained by the Commission Last updated: 15 Oct 2021