The influence of the elimination of low-intensity data on the applicability range of R₂ in automated structure evaluations

A description is given of the effect on the residual R₂ caused by a misplacement of a fraction of the atoms in a tentative structure model. New expressions are derived for the space groups P1 and P for moments as functions of the threshold a, below which intensity data are omitted. It turns out that the range in which R₂ acts as a discriminator between correct and incorrect models is drastically limited even by low threshold values. Theory and experiment are shown to be in excellent agreement.