The Becker-Coppens treatment of extinction is applied to the polarized neutron technique. Analytical expressions for the extinction correction are derived for plate-like crystals. The formalism is applied to data collected on yttrium iron garnet. Measurements were performed on crystals of various thicknesses using wavelengths in the range 0.5-1.1 Å. The correction is shown to be quite adequate even for severe extinction, with a preference for Lorentzian shape of { sigma}, the mean diffracting power, though some systematic deviations are present. The method is shown to be very sensitive to a given model for extinction, mainly because of the independence from any scale factor. It is demonstrated that the flipping ratio R can be out of the range [Rkinematic, Rdynamic] and that such a situation implies the simultaneous presence of primary and secondary extinction and one can have R = Rdynamic for a crystal far from dynamical behaviour. Application of a recent dynamical test proposed by Kato [Acta Cryst. (1976), A32, 453-466] concerning the validity of the physical assumption of the model shows that in neutron refinement for most reflexions the situation is encouraging. Estimation of mosaic spread by γ-ray diffraction leads to a fair agreement with the refined values.
No takes yet. Share an insight, caveat, or question.
Bonnet et al. (1976) studied this question.