Requirements for increased operating efficiencies mean that future generations of aero-engines will need to operate at temperatures beyond the capabilities of current nickel-base superalloys. As a result, new alloy compositions for turbine disc applications are being developed. Optimising these alloy compositions requires balancing directly competing requirements. Increased Cr contents are needed to provide environmental resistance and increased concentrations of other refractory metals to improve solid solution strengthening. However, these elements compromise the alloyâs long-term microstructural stability by promoting the formation of topologically close-packed (TCP) phases, which are deleterious to alloy performance. High $\gamma \prime$ volume fractions, which are needed to provide high-temperature strength, exacerbate the problem by increasing the concentration of these elements in the $\gamma$ phase. Therefore, an understanding of TCP formation and the compositional limits of stability is vital in the design of new alloys.

This thesis presents a combination of fundamental studies of TCP phase formation in model alloys and microstructural assessment of the thermal stability of developmental alloy compositions. Knowledge of the effect of individual elements on thermal stability is important to enable the development of optimised alloy compositions. As a result, the first fundamental study investigated the effect of Co content on thermal stability. An unexpected transition in $\sigma$ precipitation behaviour after 500 hours at 800°C was observed between 12 and 16 at.% Co. It is proposed that this behaviour may be due to the effect of Co on the $\gamma$/$\gamma \prime$ partitioning behaviour of other elements. Preliminary results from further fundamental studies investigating the effect of the Mo/W ratio and B content on thermal stability are also presented. Decreasing the Mo/W ratio was found to reduce the quantity of $\sigma$ precipitation and promote the precipitation of a W-rich phase. B additions were found to promote the precipitation of the M$3$B$2$ phase.

Thermodynamic predictions are frequently used to inform alloy design as an alternative to time-consuming and costly experiments. However, the accuracy of solvus temperature predictions for TCP phases has not been thoroughly considered. In this work, it was found that differential scanning calorimetry could be used as a means of measuring $\sigma$ solvus temperature in a series of alloys designed to be sufficiently unstable with respect to $\sigma$ precipitation. Comparison of experimental results with thermodynamic solvus temperature predictions revealed a significant underprediction of the $\sigma$ solvus temperatures for all of the studied alloys. This can inform our use of such predictions during alloy design.

The ability to quantify the amount of TCP precipitation that occurs is extremely important when assessing the thermal stability of alloys. A new method was applied to the problem of TCP quantification, involving synchrotron X-ray diffraction of solid aged samples. This was an attempt to avoid some of the problems identified with the commonly used quantification method, which involves electrolytic extraction of minor phases, and assess the accuracy of the results produced by this method. Samples of a currently used commercial alloy, RR1000, were investigated following ageing for up to 5000 hours at 800°C, revealing the evolution of phases at this temperature. The presence of extremely low quantities of minor phases was successfully detected in the solid samples using this method. However, these quantities were too low for this to be a reliable method of quantification for commercial alloys.

In parallel with these fundamental and technique-based studies, the thermal stability of a number of candidate alloys, which were developed during the design of a next-generation disc alloy by Rolls-Royce, was assessed. The alloys were characterised following a variety of thermal exposure temperatures and durations, which were determined by industrial needs at the time. Various minor phases were identified depending on the alloy compositions, including the TCP phases, $\sigma$ and $\mu$, as well as MC and M$\mathrm{<mrow\; data-mjx-texclass="ORD">23</mrow>}$C$6$ carbides and M$3$B$2$ borides.