Minimizing and Characterizing Uncertainties in Neutron Strain Measurements with Special Attention to Grain Size Effects

Abstract

The accurate determination of strain during measurement using neutron diffraction depends on many factors. The statistical uncertainty of the diffraction data is not always the most important contributor to the total uncertainty in the measured strain. Other contributors, such as sample positioning, size and shape of the sampling (gauge) volume and the size and distribution of grains within the sampling volume, often play an important role as well. Grain size issues have been the least studied and their impact is often ignored even though the potential uncertainty contribution can be large. Certain methods such as oscillating the sample during measurement can help in reducing the magnitude of the grain size effect and hence also that of the related uncertainty contribution. A thorough characterization of uncertainties due to grain size effects however, in terms of absolute values that should be added to the statistical peak fitting uncertainties has not yet been implemented. This paper will present an improved method to characterize and estimate absolute uncertainty values due to grain size effects.