INTERACTION BETWEEN g-PHASE PRECIPITATES AND MARTENSITE IN CU-ZN-AL ALLOYS
J. Pons and E. Cesari
Departament de Física, Universitat de les Illes Balears,
Ctra. de Valldemossa, km 7.5, E-07071 Palma de Mallorca, Spain
One way of strengthening the Cu-based shape memory alloys (and approaching to the performances of Ni-Ti alloys) is through the introduction of g-phase precipitates in the b phase matrix. The thermal treatments suitable to generate distributions of precipitates of different sizes and densities are well known [1-3]. However, the presence of precipitates produces changes in the martensitic transformation temperatures and the hysteresis accompanying the forward and reverse transformation [1,4,5]. A systematic study on this subject has been carried out in the recent last years, specially in Cu-Zn-Al single crystals with compositions in the vicinity of 16%at Al -15%at Zn, having an electron-to-atom ratio of 1.48 [4-10].
TEM observations show that the martensite plates can absorb
completely the small precipitates (which are coherent with the b-phase matrix) found during its growth.
Nevertheless, strong stresses arise around the precipitates due
to the shape change accompanying the transformation (the hole
left in the b matrix to accommodate
the precipitate is deformed by the martensitic transformation).
The difficult accommodation between martensite and precipitates
requires an extra elastic energy, which is responsible for the
observed shift of the transformation to lower temperatures and a
slight increase of hysteresis [4,5]. The effects on the
transformation temperatures increase with the precipitate size.
The accommodation is not completely elastic, but plastic
deformation of the surrounding martensite occurs, which is
evidenced by the observation by means of TEM and HREM of
dislocations around the precipitates [6-10]. When the
precipitates have a big size, they can not be completely absorbed
by the growing martensite plates, because the deformation that it
would be necessary to accommodate the deformation around the
precipitate becomes too large. In its turn, a complex array of
small martensite plates form in between the precipitates. This
change of the martensite microstructure takes place when the
precipitates have a size of about 100-200 nm (which is coincident
with the lose of coherency with the b-phase
matrix) and is accompanied by a shift of the transformation to
higher temperatures and a big increase of transformation
hysteresis [4,5]. In this stage, the transformation hysteresis is
not dependent on the precipitate size, but on the density of
precipitates, i.e. the distance between them, which restricts the
size of the martensite plates. The g
precipitates have also effects regarding the thermal or
pseudoelastic cycling, i.e., the samples containing a dispersion
of small and coherent precipitates have a better reproducibility
of transformation temperatures during thermal cycling [7] and a
faster induction of the two-way shape memory effect by
thermomechanical or pseudoelastic cycling [8].