Shape memory alloys

Improving the high cycle
fatigue behaviour of
shape memory alloys
PhD Candidate: Freek Sluis
Department: ASM
Section: Structural Integrity & Composites
Supervisor: R.C. Alderliesten, H.E.N. Bersee
Promoter: R. Benedictus
Start date: 01-06-2012
Funding: STW
To make wind energy more attractive, the price per energy
yield should become more competitive with fossil fuels.
The trend for offshore wind turbines has been to increase
the rotor size so that the yield per turbine is maximised.
The smart rotor principle, shown in figure 1, can reduce
the fatigue loading on turbine blades, which in turn can
allow for larger blades. Previous STW and EU projects have
demonstrated that the fatigue loading can be lowered by
as much as 30 %. This translates, with appropriate scaling,
to a power increase of 6 to 8 % [1].
Figure 1: Overlay of smart wind prototype with neutral position and
maximum positions (both positive and negative).
Smart rotor principle
Aerospace Engineering
The smart rotor principle consists of shape memory alloy
wires embedded in the trailing edge of the wind turbine
profile. Because a contractive force can be generated in
the wires the trailing edge of the wind turbine blade can be
moved, through which the aerodynamics can be
influenced. Figure 2 shows the trailing edge device with
channels for the shape memory wires.
Using wires at both the top and bottom will allow for
actuation in both directions.
Current limitations of shape memory alloys
One of the main issues with the application of shape
memory alloys is functional degradation. As can be seen in
figure 4 the response decreases heavily over time.
It has been proposed that partial transformations will
mitigate degradation significantly by skipping the last part
of transformation. The grains that transform latest need
the most energy for transformation and in the process
cause the most damage.
Figure 5: Jump in electrical resistivity
as a function of temperature [4].
 Degradation mechanisms
After transformation can be controlled accurately, focus will
shift towards the damage mechanisms that cause
functional degradation. A test program will be planned in
which partial transformation will be compared to each
other and to full transformations. Microscopy will be used
extensively during this part of the research.
Important during the comparison of different partial
transformation cycles is the trade-off between functionality
and durability.
 Influence on operation
Information on the most sensitive parameters during
partial transformation cycling will have to be acquired.
Higher frequencies, for example will reduce the time
available for cooling and thermal gradients will arise in the
wires. Locally the extend of transformation can be severely
influenced by thermal gradients which can have severe
effects on performance.
Figure 2: Trailing edge flap with
channels in which shape memory
alloy wires can be imbedded.
Shape memory alloys
Shape memory alloys are a peculiar type of metals that can
show contraction upon heating. At room temperature these
metals are in the martensite phase, while at higher
temperatures the austenite phase is preferred. In figure 3
the transformations can be seen as a function of
temperature. The specific material being used is a nickeltitanium alloy which can display elongations of up to 8 %
at room temperature, which can be recovered completely
upon heating.
Placing the wires in the actuator with a 4 % pre-strain will
allow contractive forces to be generated through heating.
The wires will prefer an austenite structure which makes
the wire contract through which the trailing edge flap can
be actuated.
Figure 4: Functional degradation of
shape memory alloys over time [3].
Research question
How can the high cycle fatigue performance of shape
memory alloys be improved by the application of
partial transformation cycles?
Breakdown of research
 Control of partial transformations
The research is broken down into three parts. The main
issue for the first part will be how to define and control
partial transformations. As can be seen in figure 5, there is
a distinct jump in electrical resistivity around the
transformation. By monitoring the resistivity accurately the
extend of transformation can be controlled.
By investigating how the system will react to changes in
parameters such as frequency an important connection is
made between research and actual application. Making this
connection will ensure that this research will be put to
good use in industry and prevent this research from being
a purely academic exercise.
Having just past the Go – No Go meeting means time is
currently spend on designing and constructing a test setup
which closely resembles the conditions in the proposed
application. Designing the setup is done with close
cooperation between lab technicians and supervisory team
since this will influence the results gathered throughout the
A first publication on control of partial transformations is
expected before the end of 2013.
Figure 3: Transformation energies required
during transformation as measured during
differential scanning calorimetry [2].
[1] – G.A.M. van Kuik, J.W. van Wingerden, M. Verhaegen, H. Bijl, A.H. van Zuijlen, and H.E.N. Bersee. Design technology
for reliable smart rotors to make large scale wind turbines more economically attractive, 2011.
[2] – K. Tsuchiya. Mechanisms and properties of shape memory effect and superelasticity in alloys and other materials: a
practical guide, chapter 1, pages 3-14. Woodhouse Publishing Limited, 2011.
[3] – Y. Furuya, H. Shimada, M. Matsumoto, and T. Honma. Fatigue and degradation of shape memory effect in Ti-Ni wire.
In K. Otsuka and K. Shimizu, editors, MRS International Meeting on Advanced Materials, Volume 9, pages 269-274.
Materials Research Society, 1989
[4] – Data sheet Saes Getters, issued 2009, regarding smartflex wires.

similar documents