The solid-state decomposition of austenite can lead to a non-random distribution of product crystals. Methods of the quantitative characterisation of this texture are extremely advanced, and there is a deep understanding of the relationship between the texture and macroscopic properties. There remain, however, important barriers to the complete calculation of texture, some of which have not been documented in the published literature. It is the purpose of this review to focus on the difficulties in order to set the scene for further progress. The advent of electron back scatter diffraction has led to an explosion of papers on microscopic aspects of crystal orientations; some of the issues relating to this technique are also described.KEY WORDS: transformation texture; crystallography; steel; martensite; bainite. Consider first the spatial resolution. Using a step size of 2-4 nm might suggest a resolution of that order, but Fig. 1 illustrates the difficulties. The boundaries in this image appear convoluted whereas they are not in reality. The grain marked with an arrow is suggested to have a size of about 8 nm 24) but it is questionable whether this is a real grain or an artefact given the size of the surrounding black regions which are crystallographically unresolved pixels. The grain boundaries themselves appear ragged, with a roughness which is comparable to the suggested 8 nm grain. Such roughness is not likely to be real since and thin film deposition does not normally lead to a bimodal distribution in grain size.
1517Consider now the problem of angular resolution when dealing with steels. The influence of imperfections in causing a spread in orientation well beyond the stated angular resolution of the EBSD instrument is illustrated in Fig. 2(a). Instrumental resolution should be of the order of 0.5°b ut the illustrated spread in the orientation of the single grain of austenite is closer to 10°. There will then be a corresponding spread in the orientations of bainite, as illustrated in Fig. 2(b), even if there is a strictly fixed orientation relationship between the parent (g) and product (a) phases. Products which form by displacive transformations will contain excess dislocations leading to a spread in diffracted intensity. Furthermore, the contribution to the spread from different crystals of bainite or martensite will not be identical given the nonuniform distribution of dislocations, plastic accommodation and the fact that the stress state of the austenite changes as more plates of martensite form. 25) There are many publications in the literature which suggest that there is a spread in the g/a orientation relationship rather than one established by the crystallography of the phases. It is then reasonable to question whether there is a true spread in orientation relationship or one which is caused by a neglect of the fact that the austenite is not a perfect crystal. The measurements implying spread have in some cases been deduced from observations on meteorites, which cool incredibly slowly as they evo...