A theory of pseudoelasticity is developed within the framework of the phenomenological Ginzburg-Landau Scheme. The system with ℈stripe℉-type ferroelastic domain structure has been discussed using Transfer-Integral method, which directly gives the equillibrium density of domain boundaries and the response of domain pattern against the external stress. The results are applied to the case of Pb3(PO4)2, an ionic ferroelastic material which shows pseudoelasticity and shape memory effect. It is suggested that in this particular case, random stress field is mainly responsible for pseudoelastic behavior. The observed stressstrain curve has been analyzed quantitatively from this standpoint. The origin of the two-way shape memory effect has also been inferred qualitatively from the same stand point.

An unusual yield phenomenon is observed in a heavily worked NiTiFe alloy, which leads to snap action shape memory and superelasticity. The snap action occurs at extremely high velocities and is accompanied by a loud cracking sound. The cause of the instability is attributed to penning, in which martensite growth is halted by having to undergo a reorientation at low angle cell boundaries, which then lowers the resolved shear stress on the martensite plate. Once free of the cell boundary, the plate grows and its strain energy is reduced so that it is able to penetrate subsequent boundaries more easily. This results in a diffuse yield drop, and the said snap action motion.

This paper summarizes our recent study of NiAI+Fe+B alloy ribbons containing 4 to 20% Fe and doped with 300 wt ppm B. Alloy ribbons were successfully fabricated by rapid solidification via melt spinning. The alloys with 8% Fe and 34% Al-equivalent [=Al%+(Fe%)/2] showed the best bend ductility at room temperature. The ribbons exhibited a reversible martensite (body-centered orthorombic structure) to B2 transformation as evidenced by DSC, X-ray and TEM studies. The shape-memory effect, as characterized by measuring the recovery of bend angles with temperature, is sensitive to alloy composition, with the best recovery observed in SMA-15 (B-doped Ni-27% Al-14% Fe). Annealing at 600°C causes aging embrittlement; in particular, in alloys containing 12% Fe.

Cu-Zn-Al shape memory alloys may offer good thermomechanical properties for industrial applications. However, it is important to control transformation temperatures,the aging and the fatigue behaviour.

We compare in this paper the effects of two thermomechanical treatments on industrial springs:

“Direct” treatment consists in homogenization in β phase, annealing and training.

“Inverse” treatment consists in homogenization in β phase, training and annealing.

A correlation between electron microscopy observations and D.S.C. experiments shows that heterogeneous ordered state (β+y2+β3) in Cu25%wtZn9%wtAl is an essential parameter influencing mechanical hysteresis.

An Fe-33Ni-12Co-5Ti (wt.%) alloy was subjected to aging treatments to determine the effect of ausaging (at 700 °C for 5, 10 and 30 hr) on the martensitic transformation and shape memory effect. The Ms temperatures were determined by electrical resistance measurements and microstructures were examined by TEM. The shape memory was studied by bending specimens in liquid nitrogen and found to be associated with a low Ms thermoelastic thin plate martensite.

Evaluation of the shape memory characteristics under thermal and stress cycles is indispensable to ensure the reliability of shape memory alloys used in nuclear power reactors. Then the effect of γ↔E cyclic transformation on the shape memory characteristics was investigated in Fe-14Mn-6Si-9Cr-5Ni (wt%) alloy. ε martensites were stress-induced by pulling specimens at room temperature, and they were reverse transformed by heating under various stresses. The change of surface relief during reverse transformation was observed by a hightemperature optical microscope equipped with a tensile machine. The recovery stress was measured by this apparatus after the “yclic transformation” was repeated various times. The reverse-transformed microstructures were observed by transmission electron microscopy. As a result, the amount of residual ε martensite increased with increasing stress applied during reverse transformation. As the “cyclic transformation” was repeated, the recovery stress remarkably increased at first, and then gradually decreased. This rapid increase of recovery stress is attributed to not only the increase of elongation when specimen is deformed but also the increase of stacking faults which act as nucleation sites of martensite. The decrease of recovery stress is attributed to the decrease of elongation.

The texture of cold rolled and heat treated TiNi sheet has been measured and designated as {110}<110>p. The.anisotropy of the tensile recovery stress, the stress dependence of the transformation temperatures, and the development of the two-way memory effect during thermal fatigue have been measured. The resulting changes of texture after thermal fatigue are presented and explained. The observed effects are explained on the basis of the parent texture and the specific martensite variants which are formed according to the lattice correspondence between austenite and martensite.

CuAlNiCr shape memory alloy ribbons with thickness ranging from 65 to 60 pm and about 15–20 mm width have been produced by melt-spinnig technique. Finer grain and homogenous columnar structure could be examined. Transformation temperatures and grain size of CuAlNiCr alloy increase with increasing heat treatment time. The reversible color change phenomena( red copper color=bronze color ) associated with phase transformation in alloy ribbons was also observed. Microstructural and transformation characteristics of rapidly quenched CuAlNiCr alloys were observed by x-ray diffraction, optical microscope, four-probe electrical resistivity method, scanning and transmission microscope.

Tensil tests for YBaCuO, Bi(Pb)SrCaCuO and TiBaCaCuO show that there always appear ferroelastic loops and shape memory effect due to the phaselike transition at tens degrees above Tc (<150K), which disappear for nonsuperconducting YBaCuO, Y(Pr)BaCuO and CuO. The property of the phaselike transition and the discrepancy on the internal friction peaks near Tc for YBaCuO have been studied and discussed.

As known to all,shape memory behaviour has been observed in a variety of metallic alloys. This phenomenon, however, is not only observed in that field but also in other materials recently, such as in polymers,intermetallic compounds and ceramics, especially in zirconia ceramics.

Swain [1] observed the shape recovery of a bent magnesia-partiallystabilized zirconia (Mg-PSZ) bar upon heating above a certain temperature. The maximum deflection which was nearly completely recovered on heating is about 300 μm.

Chen et al [2] investigated the pseudoelasticity and shape memory effect (SME) in ceria-stabilized tetragonal zirconia polycrystals (Ce- TZP) containing 12 mol% CeO2.Wang et al [3] also observed the SME in Ce-TZP containing 10 mol% CeO2,.Here we would like to introduce some of results conducted recently in our group on SHE in Ce-TZP ceramics.

A peculiar martensite found in thin foils of a Ti5oNi49Fei shape memory alloy has been examined by transmission electron microscopy and electron diffraction, which was entirely different from the R phase and the B19' and B19 martensites widely known so far in the Ti-Ni based alloys. The thin foil martensite is monoclinic with a=0.700nm, b=0.994nm, c=0.632nm and γ =94.4°. The lattice correspondence between the B2 parent phase and the martensite is as follows: <210>B2→ <100>M, <230>B2→ <010>m, <002>B2→ <001>m. The formation conditions and atom positions in the unit cell of the thin foil martensite are discussed.

The reversible transformations between the Austenite (A) and Martensite (M) phases of NITI shape memory wires having a 78°C austenlte finish temperature (950°C annealed) were studied In the cold work and heat treatment ranges between 14 to 62% and 400 to 525°C respectively. The ranges of peak Transformation Temperatures (TI), determined by Differential Scanning Calorimetry (DSC) at a 10°C/min rate, were found to be 56 to 75°C, -28 to 33°C, and 38 to 52°C for the respective high temperature A, low temperature M, and the Intermediate Rhombohedral (R) phases. The degree of cold work and heat treatment had significant effects on the TT of NITI wires. The peak TT of A and M decreases with Increasing cold work. Except for the 14% cold worked wires, the peak TT Increases with Increasing heat treatment temperature for M, and Increases with Increasing heat treatment temperature for A for temperatures higher than 450°C. The peak IT of R Increases with Increasing cold work and decreasing heat treat temperature.

Using MoKα radiation, transmission x-ray diffraction analysis was utilized to determine the phases at room temperature on wires thinned down to 0.05 to 0.01 mm in diameter. The diffraction patterns of body-centered cubic austenite (132) and monodlinic martenslte (B19) for NITi were both Identified. In addition, extra diffraction lines observed for various samples were tentatively assigned to M and the Intermediate R-phase. Depending on the thermal history and the processing conditions, the NITI wires consist of either a pure M, a mixture of A and R, or a mixture of A, R, and M at room temperature.

It is shown that the phonon dip in the longitudinal phonon branch L[111] in the metals Zr, Hf, Ti, Nb results in an oscillating component in static displacement field around a point defect. The oscillations have the wave vector 2/3(111) and occur in four rod-like shape regions along <111> directions. The value of the oscillations is much smaller then whole atomic displacements. Anomalies of the elastic diffuse scattering intensity from β-phase alloys based on the metals are explained as caused by the scattering on these distorted regions around solutes.

The use of Cu-Zn-Al single crystals (electronic concentration e/a= 1.48) and of a high resolution equipment allows to study the evolution of one or few martensite domains. This kind of study proves the existence of intrinsic thermoelasticity, the differences between different friction processes (interface movements and nucleation) and also the coherence of the hysteresis loops obtained through mechanical stress or temperature. The different evolutions linked to the creation of dislocations or to the change of é precipitates can clearly be seen during the cycling. The observations allow to establish a model of the material behaviour consistent with experimental observation.

Binary or quasi-binary compounds TI-Me (where Me stands for Fe, Co, Ni, Pd. Pt and Au, or Ni with one of the above metals) with the B2 (CsCl) structure are known to undergo martensitic transformations of two types: B2-B19'(B19) and B2-R, with an Intermediate precursor charge-density-wave Incommensurate phase IS. Analysis of both experimental and theoretical data enabled us to establish empirical relations between the transformation temperatures: Ms(B2-B19') and TIs(B2-1S), and fundamental parameters of both electronic and crystal structures of the compounds, such as the lattice parameters, the shift of the d-band in the compound with respect to that in the pure Me, the compound's d-band width, the Ti site-projected d-densities of states at the Fermi level In the compound and BCC Ti, and the temperature of β-β allotropic transformation In pure Ti. This correlation is based on localization of the Me d-bands in the compounds. As a result the Me-Me interaction becomes less important, and the interatomic interactions are controlled by d-d (Ti-Ti) and sp-d (Me-Ti) bonds. The Fermi levels are dominated by the Ti d-states, which control the lattice dynamics as well. The latter is also reflected in closeness of the C' moduli and similar values of the phonon frequencies In Ti and the TiNi compound. Since the intermediate IS phase results from a frozen phonon mode and Is controlled by the electron-phonon interaction, the B2-IS transition temperature can be correlated with the Debye Temperature, θD as: TIS=λθD. Estimates based on experimental data show that for Ti-Ni alloys λ < 1.3. A decrease of λ upon alloying with Fe stabilizes the B2 phase.

MÖssbauer spectroscopy is very sensitive to structural transitions that may occur during a phase change of the type known to arise in the shape memory TiNiFe alloy. In this work we present the results of resistance vs temperature, differential scanning calorimetry (DSC) and Mössbauer spectroscopy (MS) measurements in Ti50Ni47Fe3 sample. The resistance vs temperature curve shows the usual sharp increase associated to the B2 → R transition at Tp = 245 K and then raises until a maximum is attained at ∼ 150 K. The DSC curve shows a peak maximum at the same temperature where the resistivity starts to increase. The room temperature Mössbauer spectrum consists of a single line, indicating that the iron atoms have cubic surroundings. As temperature is lowered, a small asymmetric quadrupole doublet begins to develop. The isomer shift (I.S.) and quadrupole splitting (ΔQ) values of this doublet increase with decreasing temperature until almost constant values are achieved at temperatures ∼ 100 K. We relate the changes in the Mössbauer spectra with previously proposed charge density wave induced premartensitic transition.

Alloys based on near-equiatomic NiTi are capable of thermoelastic martensite transformations which give rise to shape memory and superelasticity (pseudoplasticity) effects. In particular, at temperatures above Ms and Af but below Md, NiTi alloys can deform by mechanisms of stressinduced martensite transformation and may display large anelastic strains which involve little or no deformation by slip. Under these conditions, the strain-controlled fatigue resistance of NiTi alloys may be exceptionally high [1]. In contrast, fcc metals like copper will, during strain controlled fatigue, eventuate severe plastic strain localization in the form of persistent slip bands (PSBs), whose behavior at free surfaces is intimately associated with fatigue crack initiation [2]. It is well known that fatigue crack initiation in fcc metals can be delayed by modifying surfacebreaking PSB structure or environment [3,4]. However, any attempt to strengthen the surface region in order to impede surface expression of persistent slip bands must contend with the large subsurface strain localizations enforced by the PSBs, which subject the film to strain cycling at roughly the characteristic PSB shear strain amplitude. Thermoelastic nickel-titanium alloys, capable of pseudoplastic straining, may be uniquely suited to the role of surface protection in situations where fatigue crack initiation is associated with PSB interaction with the external environment. In the present work, thin surface microalloys of NiTi, produced by ion sputtering, have been applied to polycrystalline copper fatigue specimens that were subsequently subjected to both monotonic loading, and fatigue under plastic strain control to produce mature PSB structures in the bulk material. Optical and scanning electron microscopy studies are described which assess the effect of the NiTi surface microalloy on the behavior of PSBs in the near surface region, and the attendant effect of the films on surface behavior of persistent slip bands. Although the results are incomplete with respect to the effects of the film on fatigue crack initiation, it has been shown for the first time that martensite transformation can be stress-induced (at temperatures above Ms) in a thin NiTi film on a plastically deforming substrate. This finding augurs well for the potential of such films to effectively suppress slip band penetration during low cycle fatigue.

Elastic and inelastic neutron-scattering and high-resolution electron-microscopy examinations of the β2 (B2) phase of NixAl(100-x) alloys show a direct relationship between the anomalous changes in lattice dynamical behavior and the evolution of static premartensitic structural configurations as the β2 cools toward its martensitic transformation temperature, Ms. The resulting microstructure is a fine-scale mosaic assembly of nonuniformly distorted and modulated domains, in which {110}<110> shear-plus-shuffle displacements give rise to the {110}B2 micromodulations of ∼1.3 nm wavelength. These displacements are derived from the unusually low energy of the Σ4<ζζ0>–TA2 phonon mode and its anomalous temperaturedependent incomplete softening at ζ = 0 (viz., the elastic constant C) and at ζ ≈ 0.16. These inhomogeneously strained domains (ISDs) are believed to be centered on low strain-amplitude defects. They are viewed as strain embryos of the product 7M(5, 2) martensite but are generally too weak to act as potent nucleation centers. Similar ISD configurations develop at defects with higher strain amplitudes (e.g., dislocations, grain boundaries) and these are the most likely sites for heterogeneous nucleation to occur.

It is well known that mechanical properties of Shape Memory Alloys are strongly dependent upon the test temperature (T) respect to transformation temperatures: the stress-strain curves however hinder the true deformation processes acting.

Electrical resistance(ER), a physical property sensibly affected by electronic structure modifications, traditionally used to follow the growth of thermal martensite, is here investigated to follow the modifications of a NiTi alloy, in an initial single phase structure, under applied stress. ER measurements are here detected with the aim to distinguish different deformation processes at four test temperatures Ti (i=1,..,4): in martensitic phase,either at T1 <Mf or at T2<As within the hysteresis cycle; in parent phase, either at Af<T3 or at Ms<T4 within the hysteresis cycle, where T2 = T4. Results are examined in comparison with previous obtained data and discussed at the light of imprinted deformation.

This paper discussed the influence of cold working and annealing process on R–phase SME in an equiatomic TiNi alloy by means of tensile test, phase transformation, temperature measurement, and shape recovery examination. The results show that the increase in cold deformation of TiNi alloy got the increase in both tensile strength and R–phasc transition (TR–Ms). As R–phase becomes stable, the SME will be improved and the decay of memory effect will also be controlled at fatique test. The increase in annealing temperature got the decrease in R–phase transition range. In that case the stability of R–phase and the SME of the alloy also become bad. When annealing at 600°C, the recrystallization occurs no R–phase is found. Surely the SME of alloy becomes bad.