Abstract:
In order to improve jet engine efficiency and performance, manufacturers have been trying over
the last five decades to increase the working temperature of gas turbines. This was achieved by
improving materials performance and component design. The latter technological breakthrough is
known as Thermal Barrier Coating (TBC), which consists of applying a ceramic insulating layer on the
internally cooled parts of the turbine. This technology is now applied in military and civil aircraft
engines, and allows temperature improvement up to 150°C. However, understanding degradation
mechanisms and improvement in manufacturing still remain important activities in turbine
development.
This PhD thesis was founded by a turbine manufacturer, Snecma, with the aim of developing a
new type of high temperature coating. The ceramic topcoat of TBC’s is currently deposited on typical
binary platinum aluminide diffusion coating or NiCoCrAlY overlay, called bondcoat, which stands at the
component/ceramic interface. In this work, a new kind of intermetallic was studied, a ternary
compound of the Ni-Al-Pt system, called α.phase, and a manufacturing route to deposit it as an overlay
coating was developed.
The main result of this thesis is the achievement of a reliable, reproducible, and controlled
manufacturing process of α-phase coatings. This process is based on sputtering multlilayers of pure
metals, followed by the annealing of the layered coating. Produced coatings are thinner than
commercial systems as they are richer in platinum (typically 5 m instead of 70 m), hence the so-called
name of "low mass bondcoat".
Such high temperature intermetallic coatings were characterised during this project (by XRD,
SEM, EDS, FIB and TEM), as well as their isothermal and thermal cycled oxidation behaviour at high
temperature. These systems were topped with a commercial ceramic layer in order to assess their
potential as bondcoats for a full TBC system. Lifetimes are relatively promising, and failure modes,
which will be described and discussed, are very specific compared to state of the art coatings. This
specificity is proven to be due to the non conventional deposition route rather than to the new compound used as a bondcoat.
