A sensitivity study using maximum entropy to interpret SANS data from the Ringhals Unit 3 NPP

Abstract

SANS experiments were performed on a high Ni weld surveillance sample from the Ringhals NPP and the Maximum Entropy method was applied to determine the most probable size distribution of irradiation-induced scattering features. The results were shown to be consistent with atom probe observations. The sensitivity of the data analyses with respect to constraints such as the limited experimentally available Q range was explored. The calculated volume fraction and the mean volume-weighted diameter of the precipitates were found to be relatively insensitive to Q max (the maximum scattering vector) greater than ∼0.40 Å −1. However, use of a lower Q max results in a shift of the size distribution to larger diameters and a reduced particle number density. Simulations demonstrated that the experimentally observed decrease in the A-ratio at higher Q values is consistent with the presence of vacancies or higher Mn contents in smaller features. Importantly, features which are experimentally unresolvable do not add to the apparent volume fraction of the features which are resolved.