A Survey and Comparison of Methods for Quantitative Texture Analysis

Numerous methods for measuring and analyzing texture are explored and discussed. Traditional pole figure inversion methods are compared against more recent developments in quantitative texture analysis involving the Rietveld whole pattern refinement method. A modified equal area pole figure collection strategy is compared to the conventional equiangular method using laboratory X-ray diffraction (XRD). The two measurement strategies were found to yield qualitatively similar orientation distribution functions (ODF), and the former set takes about half the time to collect. However, the estimated texture was found to be strongly dependent on the ODF computation parameters used in MTEX. The texture evolution during compression of a high specific strength Mg-base alloy containing long period stacking order (LPSO) phase is studied using synchrotron XRD. The initial texture was found to be much sharper than previously published results, demonstrating the importance of using a “direct method” to compute the OD for this material. The evolution of the texture of the Mg and LPSO phases suggests that the deformation is accommodated almost entirely by twinning in the former case and by dislocation slip, elastic deformation, and possibly some twinning in the latter. The texture of α-uranium foils is measured via X-ray (both Cu- and Mo-Kα radiation) and neutron diffraction, and the results are analyzed with MTEX and MAUD (Rietveld), respectively. The three radiation sources are found to yield slightly different results, based on probing different depths of the foils. Finally, the capability of measuring texture on the Neutron Residual Stress Mapping Facility (NRSF2) and Nanoscale Ordered Materials Diffractometer (NOMAD) instruments at the Oak Ridge National Laboratory (ORNL) is demonstrated by comparing measurements to those obtained via conventional XRD methods. Pole figure inversion in MTEX (for XRD and NRSF2) and Rietveld analysis in MAUD (for NOMAD) are employed for analysis, and the results are quantitatively compared. Reasonable agreement is found across all methods, and the differences between methods are attributed mostly to differences in the analysis procedure. The implications of slight variations in texture, as estimated by the various techniques, are explored through calculations of tensor properties and comparisons with predicted deformation texture evolutions. Finally, suggestions for improving the data collection speed by collecting less data are explored for both instruments.