Optimizing laboratory X-ray diffraction contrast tomography for grain structure characterization of pure iron

Abstract

Laboratory diffraction contrast tomography (LabDCT) is a recently developed technique for 3D nondestructive grain mapping using a conical polychromatic beam from a laboratory-based X-ray source. The effects of experimental parameters, including accelerating voltage, exposure time and number of projections used for reconstruction, on the characterization of the 3D grain structure in an iron sample are quantified. The experiments were conducted using a commercial X-ray tomography system, ZEISS Xradia 520 Versa, equipped with a LabDCT module; and the data analysis was performed using the software package GrainMapper3D, which produces a 3D reconstruction from binarized 2D diffraction patterns. It is found that the exposure time directly affects the background noise level and thus the ability to distinguish weak spots of small grains from the background. With the assistance of forward simulations, it is found that spots from the first three strongest {hkl} families of a large grain can be seen with as few as 30-40 projections, which is sufficient for indexing the crystallographic orientation and resolving the grain shape with a reasonably high accuracy. It is also shown that the electron current is a more important factor than the accelerating voltage to be considered for optimizing the photon numbers with energies in the range of 20-60keV. This energy range is the most important one for diffraction of common metals, e.g. iron and aluminium. Several suggestions for optimizing LabDCT experiments and 3D volume reconstruction are finally provided.