Modelling of microstructure sensitive short crack growth in Ni single crystals
File(s)
Author(s)
Karamitros, Vasileios
Type
Thesis
Abstract
Ni single crystal turbine blades have been extensively used in aerospace applications due to their high resistance to fatigue fracture under extreme loading and temperature conditions. The key to the high resistance of Ni single crystals at high temperature is the γ-γ' microstructure and the hardening of the γ' precipitates with temperature increase. However, research done so far has not fully addressed the effect of the microstructure to the fatigue life and predicted fatigue life across length scales. For this reason, microstructurally sensitive short crack growth has been investigated in Ni-based superalloy single crystals using an energy-based method to predict the crack path and growth rate across length scales. Short crack growth studied in homogenised γ-γ' Ni-based superalloy single crystal has achieved a good correlation of the predicted crack growth path and rate with experimental crack growth data. The broad applicability of this physics-based methodology was investigated at microstructural scale by developing Ni-based superalloy single crystal models that explicitly represent the γ and γ' phases. This enabled further research on the nature of crack path in γ-γ' microstructures and the role of crystallography, microstructural properties and the loading conditions on the crack path and the crack growth rate, with the crack paths and growth rates predicted at microstructural scale to achieve a good comparison with crack growth experiments at nanoscale. Therefore, the ability of modelling crack growth in γ-γ' microstructures has driven fundamental research on the physics of microstructurally short crack growth in Ni single crystals, the computation of Ni single crystal fatigue failure limits based on the stored energy per cycle at macroscale defining a critical stored energy (Gc) required for failure and the prediction of both short and long crack growth over thousands of loading cycles compared to extensive experimental data.
Version
Open Access
Date Issued
2022-06
Date Awarded
2022-11
Copyright Statement
Creative Commons Attribution NonCommercial Licence
License URL
Advisor
Dunne, Fionn
MacLachlan, Duncan
Sponsor
Royal Society (Great Britain)
Grant Number
RSIF17004
Publisher Department
Materials
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)