Nickel-titanium B19’ martensite is a strongly plastically anisotropic material with only one available slip system, which is the [100](001) M slip. Despite this, B19’ martensite polycrystals can be homogeneously plastically formed, reaching up to very high plastic strains. The absence of other slip systems is compensated by plastic twinning, in particular by the frequently appearing irreversible (20-1) M twins. However, these twins act on the same (010) M lattice plane as the plastic slip, and thus, do not seem to be a very suitable complement to the slip in terms of the Von Mises criterion. In fact, exactly the same strains as by the (20-1) M twins can be achieved also by the [100](001) M slip itself, and thus, a question arises, whether they can be understood as plastic twins in the conventional sense.

The stress-strain-temperature thermomechanical responses of NiTi shape memory alloys due to B2-B19′ martensitic transformation (MT) should ideally be phase and strain reversible in closed-loop thermomechanical load cycles, where the austenite and martensite phases do not undergo plastic deformation. However, this ideal behavior is only observed when MT occurs under zero or very low externally applied stresses. When MT occurs under higher externally applied stresses, it generates small plastic strains. These strains accumulate whenever MT proceeds under external stress, leading to the accumulation of residual plastic strains, internal stress, and lattice defects during cyclic thermomechanical loads. This accumulation results in the instability of cyclic thermomechanical responses, a phenomenon known as “functional fatigue.”

Great attention has been recently paid to the investigation of plastic deformation in NiTi. Experimental investigations of the mechanical response of NiTi polycrystalline samples within a broad stress-strain-temperature state space have revealed a complex response involving martensitic transformation, reorientation, and plastic deformation processes. The interactions between them result in complex coupled phenomena, such as transformation-induced plasticity, martensite stabilization through plastic deformation, and micro-strain heterogeneity induced by plasticity. Plastic deformation in NiTi not only generates irrecoverable strain at the macroscale, but it also induces substantial strain heterogeneity in the microstructure. This heterogeneity significantly affects the functional properties and may open up new technology pathways for designing sophisticated products. Tailored constitutive models that can reproduce the response in complex loading scenarios can be extremely beneficial.

Compared to conventional engineering materials, NiTi shape memory alloys deform via a wide range of deformation mechanisms owing to the B2⇔B19’ martensitic transformation, including twinning in martensite and plastic deformation by dislocation slip. A detailed understanding of the functional properties of NiTi requires comprehensive knowledge of all deformation processes possibly activated in thermomechanical loads. A stress-temperature diagram (Fig. 1c), constructed from the results of isothermal (Fig. 1a) and isostress (Fig. 1b) tensile tests on superelastic NiTi wire (Fig. 1a,b), provides basic information on the critical stress and temperature conditions at which individual deformation/transformation processes are activated in thermomechanical loads. The σ-T diagram is a very useful tool in NiTi research since it defines stress and temperature conditions under which martensitic transformation occurs and plastic deformation is avoided. Problems arise when multiple deformation processes are activated simultaneously, and one cannot be sure which deformation mechanism is activated. In such cases, in-situ experimental methods (e.g., in- situ electric resistivity, in-situ ultrasonic methods, in-situ x-ray diffraction) are beneficially employed. In this work, we report on the application of in-situ Dynamic Mechanical Analysis (DMA) to detect and distinguish the activation of various deformation/transformation processes during the tensile thermomechanical loading of nanocrystalline NiTi wires, particularly upon isostress heating under a wide range of tensile stresses up to fracture (Fig. 1b,d).

Deformation-transformation mechanisms in NiTi shape memory alloys (SMAs) wires subjected to tension can be investigated using various methods and techniques, often through either isothermal tensile tests or isostress thermal loading. Excluding the R-phase, there are essentially five deformation-transformation processes (elastic deformation, B2 ⇔ B19’ martensitic transformation, martensite reorientation, plastic deformation of martensite, and plastic deformation of austenite) that can be studied under both isothermal and isostress loads.

This paper presents an extension of a well-established constitutive model for NiTi covering both reversible (elastic, martensitic transformation, martensite reorientation) and irreversible (plastic) deformation mechanisms. Besides the inclusion of mechanisms of plastic deformation in both austenitic and martensitic phases in an independent manner, the model also newly captures more complex coupled phenomena of martensitic transformation and plastic deformation, such as transformation-induced plasticity, stabilization of martensite by plastic deformation, or plasticity-induced microstrain heterogeneity leading to functional fatigue. Despite a large number of different mechanisms involved in the model, which is reflected by a considerable number of internal parameters introduced for the description of the evolving microstructure of the material, the model still brings a basic, simple phenomenological understanding of the coupled transformation-plasticity proceeding in NiTi. After successful implementation to FEM software, the model provides new possibilities for simulations of NiTi components' behavior and processing.