Experimental uncertainties of three-dimensional pair distribution function investigations exemplified on the diffuse scattering from a tris-tert-butyl-1,3,5-benzene tricarboxamide single crystal

Diffuse scattering from a substitutionally disordered tris-tert-butyl-1,3,5-benz­ene tricarboxamide single crystal is analyzed with the three-dimensional difference pair distribution function (3D-ΔPDF) method. The real structure of the crystal is shown to consist of infinite polar molecular stacks along the c axis, which are laterally packed in a hexagonal fashion. The orientation of the stacks is disordered, but neighboring stacks strongly prefer antiparallel arrangements. Quantitative orientational pair correlation coefficients are determined for all lateral pairs separated by less than 100 Å. A careful analysis of the factors influencing the accuracy of the 3D-ΔPDF refinement is presented. It is shown that the effect of statistical errors is small compared to systematic errors coming from diffraction geometry distortions, reciprocal space resolution or incompletely corrected background. Various strategies for identifying and decreasing systematic errors are discussed. The impact of the systematic errors on the uncertainty of the results is not just specific for 3D-ΔPDF investigations but also relevant for other quantitative diffuse scattering modeling techniques.