Orientation imaging microscopy (OIM), an electron backscatter diffraction (EBSD) based texture analysis system, has been a key tool in the study and characterization of polycrystalline materials. OIM uses a Hough transform to identify bands in EBSD patterns, and then uses a lookup table to identify these bands and determine the crystal lattice orientation.  Statistically significant datasets on the crystal orientation of a material can be collected in relatively short amounts of time. A number of limitations of OIM, however, have motivated the continued search for a more robust system of EBSD based texture analysis. A new EBSD based texture analysis system that relies on simulated EBSD patterns and the cross-correlation function is presented in this paper. This new system addresses two of the main limitations of OIM; specifically its angular resolution (~0.5 degrees misorientation) and its insensitivity to elastic strain.

Simple, geometrically EBSD patterns are generated and small regions from them are compared with collected EBSD patterns using the cross-correlation function. The cross-correlation function measures small shifts in the Kikuchi bands, which can be used to solve for the elastic strain and crystal lattice orientation of the sample. The patterns are iteratively generated until the solution is converged on. It has been shown that this method can achieve an angular resolution of 0.05 degrees, and can recover the components of the elastic strain tensor to a resolution of 0.001.  Currently, this new method is being used to investigate the tetragonality in the crystal lattice of various steel phases due to carbon content and also to estimate dislocation densities in metals.

Fig 1: A comparison of a simulated EBSD pattern (right) with an EBSD pattern collected from a single crystal silicon sample (left).

Fig 2: left (component of the curvature tensor in degrees/micron), center (component of the elastic strain tensor), left (component of the dislocation density tensor in)

High Resolution OIM