Analytical model for neutron diffraction peak shifts due to the surface effect

Residual strains measured by neutron diffraction near sample boundaries can be biased by the surface effect as a result of incomplete filling of the instrumental gauge volume. This effect is manifested as anomalous shifts of diffraction lines, which can be falsely interpreted as a lattice strain unless appropriate data corrections are made. A new analytical model for the surface effect has been developed, which covers a broad variety of instrumental arrangements, including flat mosaic and bent perfect crystal monochromators, narrow slits, and Soller and radial collimators. This model permits the spurious peak shifts to be predicted quantitatively, and also allows the optimum configuration parameters, such as curvature of a focusing monochromator, which lead to suppression of the surface effect, to be calculated. The model has been thoroughly validated by comparisons with Monte Carlo simulations and experiments on a stress-free calibration sample. Predictions of the model proved to be very accurate, often within the interval of experimental errors, which makes it suitable for use in data analysis.