Unified theory of double twinning in magnesium confirmed by experiment

Student thesis: Doctoral ThesisDoctor of Philosophy

Abstract

Magnesium alloys find application in a variety of industries, such as the aerospace, automotive and medical industries, due to magnesium’s high strength-to-weight ratio. A number of magnesium alloys contain rare earth elements, which randomise the otherwise strong basal texture of the materials, thus improving their ductility and formability. Recently, rare-earth texture has been obtained in rare-earth-free alloys containing calcium, and the importance of double twins in inducing the nucleation of recrystallised grains with randomised orientations has been shown. Double twinning itself is not a well-understood phenomenon, and the experimental work carried out on magnesium alloys has highlighted the need for further theoretical research on the topic. With the term double twin we indicate a twin that retwins internally for a second time, and in magnesium the most common double twinning sequence is (10¯11)−(10¯12). In this thesis we focus on the structure of double twin to matrix interfaces, whose orientation within the parent grain has not been measured unambiguously so far. The foundations of this work are bicrystallography and the topological theory of interfacial defects, which exploit crystal symmetry to predict the admissible defects in an interface. They are applied in chapter 2 to classical twinning, which in magnesium occurs on the (10¯12), (10¯11) and (10¯1¯3) planes; twinning disconnections accomplishing the twinning shears are derived. In chapter 3 the phenomenological theories of double twinning as a simple shear occurring on an invariant plane in a one-step process are reviewed. These are then contrasted with the topological model, according to which double twinning entails a two-step process where a primary twin is first formed by the nucleation and motion of primary twinning disconnections and internal retwinning is then mediated by secondary twinning disconnections. In particular, it is shown that the predictions for the double twin habit planes of the phenomenological theory may be reproduced in the topological model by requiring that the final double twin to matrix interface be rendered free of misfit. The phenomenological and topological approaches are thus unified, and their differences and similarities are pointed out. In chapter 4, the case of faceted twin interfaces is considered. Facets have been shown to exist in experimental observations of single twin interfaces, and they have been modelled theoretically as accumulations of twinning disconnections; here the topological model is used to predict the emergence of commensurate facets in double twin to matrix interfaces as a result of the combination of primary and secondary twinning disconnections. High-resolution transmission electron microscopy is then used to seek validation of theoretical predictions: a hot rolled rare-earth magnesium alloy is investigated, and (¯1012)−(¯1011) double twins are observed. The commensurate facets envisaged in the topological model are reported to exist in the double twin to matrix interfaces, and are found to cause the interfaces to deviate from the orientation of the primary twinning plane; this is proposed to be the reason for the lack of agreement between measurements of the macroscopic orientation of double twin interfaces in the literature. Moreover, (¯1012)−(¯1011)−(¯1012) triple twins are found in the specimen, and the interface between a triple twin and the matrix is observed for the first time. This interface is found to be fully incommensurate, as predicted by analysis carried out using the topological model.

Date of Award1 Sept 2023
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorAnthony Paxton (Supervisor) & George Booth (Supervisor)

Cite this

'