Abstract
This thesis presents work carried out to develop the understanding of microstructural evolution and the
corresponding macroscopic creep that occurs in nickel superalloys at gas-turbine operating conditions.
In-situ time-of-flight (TOF) neutron diffraction creep experiments were performed in order to measure
the change in lattice d-spacing of both
[gamma] and [gamma]' in the CMSX-4 single crystal nickel superalloy. The
loading responses of both phases are distinct. The d-spacing evolution of
[gamma] and [gamma]'shows markedly
different behaviour in the primary and tertiary creep regimes, suggesting different deformation mechanisms.
The lattice strain evolution is interpreted in light of current dislocation theories.
It is generally assumed that at gas-turbine operating temperatures, [gamma]coarsens according - R [is propotional to] 3[root]t,
where -R is the mean radius and t is time. Heat-treatments were performed on samples of multimodal
Ni115 to investigate this assumption. Electron microscopy was used to analyze the samples post heat-treatment,
and the frequency distribution of radii was calculated. It is shown that a transient period
can exist for thousands of hours, and the above coarsening rate is not valid. An existing LSW-based
model is further developed to model the coarsening kinetics of a superalloy in real time and real radii
for the first time, and model predictions are compared to experiment.
The creep properties of different [gamma] distributions in the Ni115 nickel superalloy produced by heat-treatment
were examined. At the stresses and temperatures employed it is shown that particle bypass
cannot occur by cutting or bowing and so presumably occurs by a climb-glide motion. The Dyson creep
model is a microstructure based climb-glide bypass model for unimodal distributions. It is developed
further to account for bimodal distributions and the predictions compared to experiment. The fine [gamma], when present, controls dislocation motion, seen in both experiment and model predictions.