It was recalled that the construction of the  surface required the application of mechanical constraints in order that the energy of non-equilibrium fault configurations could be calculated. This choice of mechanical constraint was not unique, and it was shown that  surfaces which were constructed by using differing schemes could also differ significantly. The measurement of planar fault energies by using transmission electron microscopic techniques was also critically assessed. It was discovered that the constraining effects of the bounding partial dislocations could significantly modify the local fault energy. It was concluded that experimentally measured fault energies should therefore be expected to differ from atomistically calculated values.

A.H.W.Ngan: Philosophical Magazine Letters, 1995, 72[1], 11-9