Experimental studies were made of stacking faults in samples which had been prepared by using a range of solidification rates. Shear-type faults were observed on {110} planes in plasma-sprayed polycrystalline specimens that had been grown at a rate of 106cm/h. These were thermally unstable, and disappeared during heat treatment. Condensation faults on (001) planes, which were suggested to be associated with Si loss during processing, were observed in single crystals that had been grown at rates of 1 to 30cm/h. These were thermally stable. Previous experimental evidence for the transformation of a C40 high-temperature phase to a C11b low-temperature phase was re-evaluated and, as a result, the existence of the C40 structure was questioned. Thus, transformation-induced stacking faults, which had originally been described as ¼<111> faults that lay on (110) planes, were reinterpreted as being 1/6[001] condensation faults which lay on (001) planes.

B.K.Kad, K.S.Vecchio, R.J.Asaro, B.P.Bewlay: Philosophical Magazine A, 1995, 72[1], 1-19