Abstract
We have performed a transmission electron microscopy study, using weak beam imaging, of the interface dislocation arrays that form initially at the (001) Ni–Cu interface during coherency loss. Interface dislocations were absent in the 2.5 nm Ni/100 nm Cu bilayers, but were present in the 3.0 nm Ni samples, indicating that the critical Ni film thickness for coherency loss is between 2.5 and 3 nm. The key features of the interface dislocation structure at the onset of coherency loss are: (i) the majority of interface dislocations are 60° dislocations, presumably formed by glide of threading dislocations in the coherently stressed Ni layer, and have Burgers vector in the {111} glide plane; (ii) the interface contained approximately 5% Lomer edge dislocations, with Burgers vector in the {001} interface plane, and an occasional Shockley partial dislocation and (iii) isolated segments of interface dislocations terminating at the surface are regularly observed. Possible mechanisms that lead to these dislocation configurations at the interface are discussed. This experimental study shows that near the critical thickness, accumulation of interface dislocations occurs in a somewhat stochastic fashion with favourable regions where coherency is first lost.
Acknowledgements
This research was funded by DOE, Office of Science, Office of Basic Energy Sciences.