2012 Theses Doctoral

Monte Carlo Simulations of Powder Diffraction at Time-of-Flight Neutron Sources

Li, Li

Measured powder diffraction patterns contain contributions from the sample and the instrument. Most available data analysis software operates on the measured data to extract sample parameters, however, few programs can take sample parameters and rigorously simulate the expected diffraction profile for a given instrument. In this work Monte Carlo methods, within the framework of McStas software, are used for the simulation of neutron diffraction at the SMARTS (Spectrometer for Materials Research at Temperature and Stress) diffractometer in the Los Alamos Neutron Science Center.

The simulations include all the instrumental components, such as the moderator, guide system, collimator, detector banks and sample. The results of the simulations are in excellent agreement with the experimental data for different ideal powder samples. The simulations also yield information on the line broadening introduced into the diffraction profile as a function of energy and are used to predict the size and strain limit above which line broadening studies cannot be performed on this instrument. Theoretical derivations of line profile analysis are presented to provide an accurate explanation of the formation of diffraction peaks from the powder sample. This thesis demonstrates how rigorous scattering theory can be used to design optimal diffraction instruments.