Technological developments in EBSD has enabled great improvements in indexing reliability and accuracy [1].   However, some individual phases continue to pose challenges, especially those that present extremely similar Kikuchi patterns to the EBSD camera for different crystallographic orientations.  This phenomenon is called “pseudosymmetry”, as it commonly involves relatively high intensity bands in certain patterns with an apparent higher symmetry than the crystal structure actually possesses.  In many cases, only very slight differences in inter-band angle separate candidate solutions, and only robust and accurate band detection may identify the correct one among them. 

Conventional Hough-based band detection methods are sufficiently accurate for most indexing requirements.  High accuracy Hough transform settings improves band detection accuracy, and is useful in mitigating pseudosymmetric misindexing [2].  However, these settings result in reduced data acquisition speeds, and may not completely eliminate pseudosymmetry problems.

New band detection refinement methods improve EBSD indexing performance for some of the most chronic cases, at reasonable data acquisition speeds.  Higher accuracy band detection is achieved by iteratively comparing the positions of simulated bands with bands in the actual Kikuchi pattern image, using expected versus actual band widths and, accounting for the hyperbolic shape of bands.  In addition to delivering high precision crystallographic orientation data and helping discriminate different phases with similar Kikuchi patterns, this method is sufficiently sensitive to resolve fine differences in inter-band angle to nearly eliminate many cases of pseudosymmetric misindexing. 

An important application is γ-TiAl alloys, which are promising jet engine turbine materials, combining low density with good oxidation and creep resistance.  The high temperature deformation behaviour of these alloys must be better understood before they can widely replace the higher density Ni-base superalloys; For example, an improved knowledge of the fundamentals of crystallographic slip and its interaction with the γ/γ  lamellar variants could be critical.  Pseudosymmetry, however, is a major issue here (Fig. 1a), arising from γ-TiAl’s close tetragonal c:a unit cell parameter ratio of 1.018, giving the Kikuchi patterns it generates a pseudo-cubic configuration, resulting in indexing inaccuracies.  These mistakes show the same misorientations as boundaries between real γ-TiAl lamellae, causing problems in revealing the true microstructure.  Phase discrimination between coexisting γ(TiAl) andα₂(Ti₃Al) phases is also difficult (Fig. 2a).  Application of a new, automated band detection tool and system knowledge of the confronting pseudo-symmetry almost completely eliminate these issues (Fig.’s 1b and 2b) and are used in real-time during data collection, at normal acquisition speeds.

References:

[1] K. Thomsen et al., Royal Microscopy Society EBSD 2014 conference proceedings (2014)

[2] C. Zambaldi et al., J. Appl. Cryst. 42 (2009), p. 1092-1101

Figures:

Figure 1a. Comparison of results using conventional and special band detection methods for γ(TiAl). Orientation map from conventional band detection, exhibiting multiple orientation errors seen as speckled pixels

Figure 1b Orientation map from a dataset using high accuracy band detection with system knowledge of the encountered pseudosymmetry, resulting in almost no indexing errors. No post-acquisition cleaning has been performed on either map. Scale bars are 20um.

Figure 2a. Comparison of band detection methods using phase maps, with red =γ(TiAl), blue = α2(Ti3Al), black = no solution. Conventional band detection results in some phase discrimination mistakes, seen as speckled blue and red pixels.

Figure 2b New band detection method results in nearly no indexing errors, and fewer non-indexed points. Maps are as-acquired, no post-acquisition data cleaning. Scale bars are 5um

« Back to The 16th European Microscopy Congress 2016

EMC Abstracts - https://emc-proceedings.com/abstract/addressing-pseudo-symmetric-misindexing-in-ebsd-analysis-of-gamma-tial-with-high-accuracy-band-detection/