We explore data reduction and correction steps and processed data reproducibility in the emerging single crystal total scattering based technique of three-dimensional differential atomic pair distribution function (3D-$Delta$PDF) analysis. All steps from sample measurement to data-processing are outlined in detail using a CuIr$_2$S$_4$ example crystal studied in a setup equipped with a high-energy x-ray beam and a flat panel area detector. Computational overhead as it pertains to data-sampling and the associated data processing steps is also discussed. Various aspects of the final 3D-$Delta$PDF reproducibility are explicitly tested by varying data-processing order and included steps, and by carrying out a crystal-to-crystal data comparison. We identify situations in which the 3D-$Delta$PDF is robust, and caution against a few particular cases which can lead to inconsistent 3D-$Delta$PDFs. Although not all the approaches applied here-in will be valid across all systems, and a more in-depth analysis of some of the effects of the data processing steps may still needed, the methods collected here-in represent the start of a more systematic discussion about data processing and corrections in this field.
تحميل البحث