In-situ synchrotron x-ray diffraction texture analysis of tensile deformation of nanocrystalline superelastic NiTi wire at various temperatures

According to the state-of-the-art view, superelastic deformation of NiTi wires at room temperature proceeds via stress induced martensitic transformation from B2 cubic austenite to B19'monoclinic martensite. With increasing test temperature, the stress induced martensitic transformation is substituted by plastic deformation of austenite at martensite desist temperature M-D. However, there are many unsolved problems with this widely accepted view. What are the texture and martensite variant microstructure in stress induced martensite and do they depend on test temperature? Does the austenite transform to martensite completely within the transformation plateau range? How the superelasticity changes into plastic deformation of austenite with increasing temperature - is it stepwise or gradual change? How the wire deforms plastically at various temperatures? Does plastic deformation occur in austenite or in martensite, via dislocation slip or deformation twinning? Are the deformation/transformation processes in nanocrystalline NiTi wires same as in large grain polycrystals? We have addressed these long standing but unsolved questions by performing series of in-situ synchrotron x-ray diffraction experiments on superelastic nanocrystalline NiTi wire subjected to tensile tests at 20, 90 and 150 degrees C until fracture supplemented by post mortem TEM analysis of lattice defects created by the tensile deformation. It was found that, in case of conventional superelasticity at 20 degrees C, austenite transformed almost completely to stress induced martensite within the transformation plateau range. The stress induced martensite displayed a sharp two fibre texture reflecting its (001) compound twinned microstructure. Stress induced martensite transformed back to the parent austenite without leaving any significant unrecovered strains and lattice defects in the austenitic microstructure. When loaded further into the plastic deformation range, this martensite deformed via combination of (20-1) and (100) deformation twinning and kinking assisted by [1001(001) dislocation slip in martensite. Recoverability of tensile strains on unloading and heating remained surprisingly large (similar to 10%) up to wire fracture at 62% strain. The superelasticity at elevated temperatures 90 degrees C (150 degrees C) was found to be very different. The austenite transformed into a mixture of phases containing 40% (10%) volume fraction of stress induced martensite within the transformation plateau range. The stress induced martensite displayed four fibre texture, which further evolved with increasing strain. The original < 111 > fibre texture of austenite evolved with increasing strain towards random orientation distribution. The recoverability of tensile strains on unloading and heating sharply decreased with increasing temperature. Two alternative deformation mechanisms are proposed to explain these changes. The first mechanism assumes that martensite stress induced at elevated temperatures appears in a form of thin internally twinned CVP martensite plates surrounded by austenite deforming via dislocation slip. Requirement for strain compatibility at habit plane interfaces affects selection of martensite variants under stress and texture. The second mechanism is based on the idea that martensite stress induced at elevated temperatures immediately deforms plastically and undergoes reverse martensitic transformation to austenite leaving behind unrecovered plastic strain, slip dislocations and {114} austenite twins in the austenitic microstructure of the wire. Since the second mechanism explains the experimental observations better, it is considered to be more realistic.