Finite element modelling of plastic instability in irradiated steels (PhD)

While the steels used for the vessel in current nuclear power plants receive only limited amounts of neutrons per unit surface (fluence or dose) during their lifetime, austenitic and ferritic-martensitic steels in future nuclear reactors are expected to receive doses in excess of 100 dpa. Under these conditions, the mechanical properties of the materials are known to degrade radically. In particular, at high dose a dramatic increase of the yield strength and reduction of the subsequent engineering strain-hardening rate occur. This effect is observed in all nuclear steels: bainitic, ferritic/martensitic and austenitic.

At the microscopic level, this type of macroscopic behaviour is generally associated with the appearance, under loading, of narrow channels, or clear bands, crossing the grains. These regions appear to be depleted of irradiation produced defects and are therefore subjected to much larger plastic strain than the rest of the material. This is thought to be one of the reasons for the appearance of plastic instabilities. While fine scale descriptions (molecular dynamics or discrete dislocation dynamics, DDD) should be used for the understanding of fundamental mechanisms of clear band appearance, larger scale (phenomenological or average) representations can be used to model the channelling effect on the average mechanical properties.

The objective of this proposal is to investigate the use of finite element continuum models as possible part of a multi-scale effort to treat irradiation effects, specifically to treat the problem of radiation-induced plastic instability. This work will thus focus on answering the following question: does the appearance of the channelling effect explain the evolution of the average mechanical behaviour of irradiated metals ?

The work will proceed by investigating the ability of phenomenological models to properly represent the average mechanical effects of irradiation based on postulated microstructural descriptions of the dislocation channels. As such, the currently available tools would allow non-evolving microstructural configurations to be investigated, i.e. configurations in which the dislocation boundaries may evolve in their properties, but are fixed in space. In a second stage, the goal will be to describe a postulated spatial evolution of the dislocation channels boundaries (to obtain via computational investigations at finer scales) and to couple this description with phenomenological plasticity descriptions. This will allow the need for incorporating additional modelling features, such as crystal plasticity, to be investigated.