Since the 1980s, small additions of rhenium have been found to markedly
improve the creep properties of nickel based superalloys. However, the underlying
causes for this effect are not yet fully understood. In this work, a
variety of experimental and modelling techniques have been used to evaluate
possible mechanisms improving creep properties at high temperature.
Ab initio modelling (via the Density Functional Theory) has shown that
rhenium clusters are unstable in the nickel lattice. This observation was
confirmed by Extended X-ray Absorption Fine Structure, which showed
that rhenium atoms are dispersed in the nickel lattice. Atom Probe Tomography
also found no clusters in Ni-Re alloys and in CMSX-4, a commercial
single-crystal superalloy.
Atom Probe Tomography, in conjunction with Phase Field Modelling, was
also used to show that the rhenium enrichment close to the [gamma]/[GAMMA] interfaces
is formed upon cooling from service temperature. Thus means that the
rhenium enrichment observed with the atom probe does not play a role in
the strengthening effect played by rhenium. Finally, ab initio modelling
(via the Density Functional Theory) was used to investigate the effect of
rhenium on the stacking fault energy of nickel to ascertain whether rhenium
content could increase the spacing between Shockley partial dislocations.
The present work shows that rhenium clustering, the rhenium enrichment
close to the [gamma]/[GAMMA] interfaces and the rhenium effect on stacking fault energy
cannot be the mechanisms underlying the rhenium-effect in nickel based
superalloys. It therefore appears that single atoms of rhenium may slow
down dislocation motion in the [gamma] phase of nickel based superalloys. This is
supported by the observation that rhenium is the slowest diffusing element
in nickel and it is therefore expected to slow down all diffusion activated
processes, such as the well-established climb-assisted glide of dislocations
in the [gamma] phase of nickel based superalloys.