Journal of Neutron Research

DiffMod is a simulation program for the evolution of a neutron ensemble in a thermal target – moderator – reflector assembly of a pulsed neutron source based on the statistical description of diffusion, scattering, moderation, and absorption processes. The spatial resolution, the energy resolution and the diffusion directions are strongly restricted to achieve calculation times in a realistic moderator – reflector assembly below 1 hour. In comparison with Monte-Carlo simulations describing the geometry and interactions between neutrons and moderator material exactly, we prove that the DiffMod approach can deliver intensities and pulse shapes that are exact within 10% compared to the Monte-Carlo simulations that require much more computing power. In addition, a time-resolved illustration of the spatial distribution of the neutrons at different energy levels is provided.

Null-scattering Ti:Zr alloy (Ti_67.6Zr_32.4) is widely used in neutron scattering for sample containers, pressure cells and gaskets since it does not produce contaminant Bragg reflections from the sample environment. Here we report superconductivity in this material with critical temperature T c = 1.4  K, lower critical field H_c1≈ 100 Oe and upper critical field H_c2 ≈ 4500 Oe. These properties are largely independent of the amount of a small quantity of ferromagnetic iron found in materials of all providers. Our findings bear on the use of Ti:Zr in neutron scattering experiments at applied field and low temperatures.

Data-reduction software used at neutron-scattering facilities around the world, Mantid and Scipp, ignore correlations when propagating uncertainties in arithmetic operations. Normalization terms applied during data-reduction frequently have a lower dimensionality than the quantities being normalized. We show how the lower dimensionality introduces correlations, which the software does not take into account in subsequent data-reduction steps such as histogramming, summation, or fitting. As a consequence, any uncertainties in the normalization terms are strongly suppressed and thus effectively ignored. This can lead to erroneous attribution of significance to deviations that are actually pure noise, or to overestimation of significance in final data-reduction results that are used for further data analysis. We analyze this flaw for a number of different cases as they occur in practice. For the two concrete experiments that are comprised in these case studies the underestimation turns out to be of negligible size. There is however no reason to assume that this generalizes to other measurements at the same or at different neutron-scattering beamlines. We describe and implement a potential solution that yields not only corrected error estimates but also the full variance-covariance matrix of the reduced result with minor additional computational cost.