The project’s objectives aim to materialize an original idea which has been
the background of this project, consisting in the obtainment of Shape
Memory Alloys (SMAs) able to cumulate the advantages of the structural
properties of the main alloy systems of this group ( Ti-Ni, Cu-Zn, Cu-Al ) in
order to get a new SMA with two way shape memory effect (TWSME)
allowing to memorize two hot shapes and two cold shapes. This alloy could
revolution the applications at the level of the present state of art of
technique, in the field of precision robotics, where the use of shape memory
actuators experience a series of disadvantages, such as: 1) low response
rate; 2) the necessity to attach a bias element in order to reset the cold
shape; 3) the development of maximum force at the beginning of heating and
not at its end, as it is required by most of applications [1]. Under these
circumstances, the contributions to scientific knowledge development
should be detailed at fundamental and applicative level.
From a fundamental point of view the research aims to accomplish a
deepened study on the complex phenomena that accompany internal
changes specific to shape memory multifunctional materials. In most of the
studies performed on the martensitic transformation from SMAs, reported in
literature, the high temperature phase (which by analogy to the martensitic
transformation produced in steels is called austenite) transforms to
martensite, in an irreversible and diffusionless manner, both during cooling
(when the transformation is eventually preceded by a series of order-
disorder transitions as well by a premartensitic transformation) and during
mechanical loading [2]. Actually, with increase of temperature or of applied
stress, sensible increases of the diffusion coefficient occur, in such a way
that one cannot talk about a completely diffusionless transformation. Atomic
diffusion exists but doesn’t occur on long distance – as, for instance, in
gases liquids or crystalline solids heated up to the proximity of the solidus
curve – but only on tens to thousands interatomic distances (migration). In
addition, the transformation isn’t entirely reversible since a certain amount
of energy is irreversible changed to internal friction, which explains the
presence of transformation hysteresis. Therefore martensitic
transformation in SMAs experiences hysteresis, isn’t perfectly reversible
and diffusionless but these phenomena are noticeable only after tens of
millions of cycles, providing heating-cooling temperatures and loading-
unloading stresses are kept within strict limits, located between the finish
transformation temperatures (Af and Mf) and between the limits of plastic
irreversible yield of austenite and stress induced martensite. In other words,
a SMA witch has, for instance, a maximum recovery strain of 8 % will be able
to develop, within an application with high fatigue life (107 cycles),
recoverable strains of maximum 3 %. Due to diffusion augmentation and
mostly to the accumulation of diffusion effects, concomitantly with
martensitic transformation the precipitations of Ni4Ti3, at Ti-Ni based SMAS
or of α-phase (fcc) and/ or γ-phase (complex cubic) at Cu-Zn-Al based and
Cu-Al-Ni based SMAs take place. One of the elements of originality in the
present project consists in analyzing precipitation phenomena in a
correlative way, for the three above mentioned alloy systems, in order to
observe the influence of the formation of the corresponding phases (which
involves atomic migration, i.e. occurrence of concentration gradients at
intraganular level) on the reversibility and the critical transformation points
and on the stress state balance between the precipitates and the martensite
or austenite matrices. This approach has already been reported in literature
in connection to Ti-Ni based SMAs, where it was shown that the change of
the chemical composition of austenite in the proximity of precipitates
causes an intragranular segregation which finally leads to the occurrence of
two direct martensitic transformations: one in the initial austenite which didn’
t experience changes of chemical composition and the other in the proximity
of the precipitates [3]. It is however considered that α-phase and γ-phase
from Cu-based SMAs also influence the corresponding martensitic
transformations and implicitly shape memory phenomena that these
transformations are determining, because these precipitated phases have
different compositions as compared to martensite [4]. From this point of
view it is mentioned that the present research team has recently reported a
series of observations related to the formation of stress induced martensite,
in austenitic or martensitic matrices, as an effect of the formation of hard γ-
phase, in the case of Cu-Al-Ni based SMAs [5]. Thus a three- or four-phase
structure may occur while martensite has a multivariant morphology (up to
24 martensite plate variants) [6]. The contribution to scientific knowledge
development, aimed by this project, is based on the study of the interactions
between the 3 or 4 phases that take part to the transformation, between the
martensite plate variants and between applied loads and long period
stacking order (LPSO) of martensite. LPSO can be determined by means of
X ray diffraction and transmission electron microscopy (TEM). The former
investigation method allows identifying the close packed atomic planes
which are oriented parallel to the surface of analyzed specimen, determining
interatomic distances between these planes and calculating the
microdeformations of crystalline lattice (by means of specialized software
such as POWDER CELL). The latter investigation method enables the
visualization of atom packing within unit cells, by means of electron on
selected area of diffraction (SAD).
