Origin of the 60 degree and 90 degree dislocations and their role in strain relief in lattice-mismatched heteroepitaxy of fcc materials

TL;DR

This paper investigates the atomic structure and formation mechanisms of 60 and 90 degree dislocations in fcc heteroepitaxy, revealing limitations of the classical Matthews model in explaining strain relief processes.

Contribution

It challenges the traditional Matthews model by analyzing the atomic structure of dislocations, showing that assumptions about strain relief are invalid for certain dislocation types in fcc materials.

Findings

The Matthews model's assumptions are invalid for 60 degree dislocations.

90 degree dislocations' strain relief is valid only when formed by vacancy or interstitial aggregation.

Atomic structure analysis clarifies dislocation roles in strain relief.

Abstract

Strain relief in lattice mismatched heteroepitaxy is mediated by formation and/or propagation of dislocations. Due to their technological significance, the process of strain relief in materials with face-centred cubic (fcc) lattices has been analyzed by several researchers1,2 following the work by Matthews and co-workers in the late 1960s to early 1970s3-6. In the Matthews model, it is assumed that the strain relieved by any misfit dislocation is equal to the edge component of the dislocation burgers vector in the interface plane. This assumption has been used in all subsequent analyses of strain relief in lattice mismatched heteroepitaxy [1,2]. Based upon the known three-dimensional atomic structure of the dislocations in fcc lattices, we show that the assumption is not valid for the 60 degree dislocations that form/expand via the conservative glide process. For compressively (tensilely) strained films the assumption is valid only for the 90 degree dislocations that form/expand via aggregation of vacancies (interstitials).