the R-phase in a polycrystalline 50Ti48Ni2Fe shape memory alloy has been here investigated by X-Ray Powder counter Diffractometry (XRPD). With the aim to determine the R-phase lattice parameters accurately, a n original XRPD calibration method, specifically apt for shape memory alloys, has been developed. The intrinsic aberrations that affect the XRPD have been takeninto account by evaluating theXRPD spectrum of the high temperature b.c.c. phase. The correction function A(28) has been calculated and applied to the R-phase diffraction peaks. A full indexing of the X-ray spectrum of the R-phase, not achievable by raw experimental data, has thus been obtained and the lattice parameters calculated on a set of reflections in un-aberrated angular positions with a n accuracy of = 1 part in 10000. The calibration method here developed can be used also in the XRPD study of crystalline materials that undergo a first solid-solid phase transition.
l. INTRODUCTIONThe shape memory alloys (SMA) rely their unusual mechanical properties (shape memory and pseudoelasticity) on the thermoelastic martensitic transformation (TMT), a first-order solidsolid, diffusionless, athermal transformation. In NiTi alloys, critically depending upon composition andlor thermo-mechanical treatment, a rhombohedral phase (R-phase) can appear stable between the high temperature b.c.c phase (B2-phase) and the low temperature lnonoclinic phase (M-phase). Since the first detection of electric transport anomalies in the premartensitic region [l], the R-phase in NiTi alloys has been widely investigated. Initially considered a "pre-martensitic" effect and later assessed as an independent phase from the martensitic one, the R-phase does exhibit precursory effects during the B2+R phase transformation: the softening of some elastic constants [2], the appearance of diffuse "extraspots" not in the exact 113 positions in the electronic diffraction patterns [3,4], and the very large minimum in the electrical resistance, have been extensively studied. The break-through was given by [5] investigating a 50Ti47Ni3Fe where the B2+R transformation is well separated from the R+M one, on the temperature scale. Theoretical models to explain the origin of the R-phase transformation and the related precursory effects, have been proposed [6-81. The thermoelastic nature of the B2-R transformation and the process of R-phase variants self-accomodation have been assessed . Moreover, the shape memory effect and the pseudoelasticity related to the B2++Rhas beenillvestigated [9,11-121 as well. Concerning tothe R-phase crystal structure itself, many investigations were also accomplished [10,13-141, Article published online by EDP Sciences and available at http://dx