Error bars in CBED symmetry?

Abstract

Convergent-beam electron diffraction (CBED) is a powerful tool in analytical electron microscopy. Analysis of the patterns may reveal the crystal structure of the specimen. However, if the crystal is not perfect in the region probed by the electron beam, patterns can be degraded and the analysis may be ambiguous. Convergent-beam diffraction patterns are never perfect, yet there has been little discussion, to date, of how far from a perfect symmetry a convergent-beam pattern may be and still be assigned that symmetry. The human eye is acutely sensitive to pattern recognition and it is often easy to spot small deviations from perfect symmetry. The question is, given a pattern that one knows is from a near-perfect crystal, how far can the symmetry deviate from perfection before one must say that a lower symmetry is observed? The aim of this work was to carefully analyze CBED patterns from a number of known materials and observe to what degree the pattern symmetries deviated from perfection. Analysis was carried out both manually and with the aid of a computer. An attempt has been made to establish confidence limits for use in CBED pattern analysis. © 1985.