| Summary: | <p>Intergranular Stress Corrosion Cracking (IGSCC) has been one of the most important
challenges to nuclear power plant operation worldwide. The degradation of alloy 600 (Ni-based
alloy) and its susceptibility to Stress Corrosion Cracking (SCC) when exposed to primary water
environment in pressurized water reactors (PWRs) has been a topic of interest for many
decades among the SCC community. However, the understanding or prediction of SCC
mechanisms remains questionable. To address this, an IGSCC predictive model called ‘Local
Model’ has been developed in the framework of EDF’s CSI project. The main focus of this
research is to investigate IGSCC in alloy 600 and to experimentally validate a predictive model.
Firstly, it is important to know whether stress corrosion cracking occurs in these materials, in
the way presented in the existing experiments and models in the literature. These have shown
a very large scattering for the needed for the crack initiation and crack propagation, indicating
significant model uncertainties.</p>
<p>In this thesis, an experimental and modelling program has been developed for investigating
SCC in alloy 600 for conditions simulating PWR conditions under constant load. The goal of
the experiments was to find a reliable and quantitative determination of crack initiation and
crack propagation for these conditions. For the modelling aspect, the aim was to determine the
critical parameters of SCC by considering the phenomena in the local scale of grain boundaries
and crack tips. The model requirements include the ability to predict the crack initiation and to
model crack growth rate as a function of macroscopic and measurable parameters. This has
been validated by experiment to ensure that the assumptions of the model are realistic enough
to make sure of reliability for reactor safety applications. By doing this, the model allows for a
better understanding of the underlying mechanisms of SCC. As such, it focuses on the
phenomenological approach rather than the empirical approaches in other models.</p>
|
|---|