April 2013

Smart Materials

Got something to say?

Become a contributor: editor@experimentation-online.co.uk

Smart Materials: Shape Memory Alloys


Shape Memory Alloys (SMA) are truly unique among metals; they have properties that most other metals don’t. For one, they are super elastic, meaning you can bend them far beyond the elastic limit of any other equivalent metal. The truly amazing property though is that SMA undergo a reversible solid state phase change in crystalline structure from martensite to austenite, and vice versa [1]. This phase change happens when the SMA crosses a transition known as a transition temperature. All metals have a transition temperature between demonstrating ductile and brittle characteristics, but SMA take this to a whole different level.

Professor David Grant at the University of Nottingham researches SMA, and the video below is his demonstration of an SMA rod [2].

Click here for video.

The nature of shape memory behaviour

Warning: getimagesize(/home/erudit8/public_html/experimentation/common/imgs/shapememoryalloys_chrismorris_april2013_1.jpg) [function.getimagesize]: failed to open stream: No such file or directory in /home/erudit8/public_html/common/functions.php on line 203

Warning: Division by zero in /home/erudit8/public_html/common/functions.php on line 204

Figure 1 - Left - BCC Structure, Right - FCC Structure [4].

Shape memory behaviour, as previously mentioned, is the result of a reversible phase change between austenite and martensite and vice versa [1]. Austenite and martensite have face-centred cubic (FCC) and body-centred cubic (BCC) crystal structures respectively [3]. BCC materials are generally harder and less malleable than their FCC counterparts [4]. The video shown above demonstrates a super-elastic FCC phase at room temperature and a much less malleable BCC phase when cooled in liquid nitrogen [2]. This is an example of a one-way memory SMA; it remembers a shape at high temperature. Some shape memory alloys possess a two-way memory characteristic, which has the alloy retain a specific shape at both high and low temperatures (either side of the transition) [6]. Figure 1 shows the typical layout of a BCC material (left), which has crystals placed further apart from one another, making it harder for them to move.

Eventually shape memory alloys revert to their original shape after crossing their transition temperature. Usually SMA can be forced into their martensitic phase and placed in position and upon returning to their original position actually perform the intended function. This is particularly important in biomedical applications [5]. This particular property is favourable in specific situations; however the original shape of the alloy can be altered [7]. Shape memory alloys are “trained” by repeatedly hot and cold working the material. Muscle wires, like the rod shown in [2] can be clamped in a particular position in their cold phase, and then when a higher temperature is applied they will maintain the clamped position. Subsequently applying a cold phase temperature will forcibly maintain this shape further. In this way the alloy can also begin to develop two-way memory behaviour [6] [7].

The temperature transition at which this reversible phase change occurs can effectively be tuned by the alloy content of the SMA during manufacture [6]. Figure 2 demonstrates how altering temperature changes the crystal structure of the material between austenite and martensite; the subscript “s” and “f” depict the temperature at which the phase change starts and finishes respectively.

Warning: getimagesize(/home/erudit8/public_html/experimentation/common/imgs/shapememoryalloys_chrismorris_april2013_2.jpg) [function.getimagesize]: failed to open stream: No such file or directory in /home/erudit8/public_html/common/functions.php on line 224

Warning: Division by zero in /home/erudit8/public_html/common/functions.php on line 225

Figure 2 - Schematic of crystal structure of SMA with different temperatures [9].


Shape memory alloys are used in a variety of industries, for example in marine and aerospace applications SMA are used as couplings for metal to metal pieces [1]. This is done by heating a radial SMA piece and expanding it, so that when it shrinks back to its original position over the metal pieces, it clamps them in place with significant radial forces, somewhat analogous to interference fits [1].

Shape memory alloys however are most prolific in the biomedical field. Titanium is widely known as one of very few bio-compatible materials. Most prosthetic joints are made of titanium alloys because the human immune system accepts them. One particularly prevalent example of SMA is shown by David Grant in the video demonstration above [2]. Towards the end of the video a small clip is shown, this clip is used as a bone stitch. The methodology behind it is to drill small holes in two fractured pieces of bone, and then deform the martensitic phase of the clip to fit into each hole. The temperature of the human body then returns this particular SMA to its austenitic phase, which pulls the fractured bone pieces together to allow it to heal.

A wide variety of SMA can be used as biomaterials, for instance Niinomi et al discuss the use of low Young’s modulus titanium SMA for spinal fixation devices [5]. Xiong et al investigated nickel titanium SMA foams to be used to stimulate bone tissue re-growth [8]. SMA foams can stimulate bone regrowth by possessing a specific level of porosity (200-500 micro metres) allowing bone ingrowth.


Shape memory alloys are fascinating materials due to their truly unique behaviour. Most metals typically do not demonstrate such a high level of elasticity, but more importantly the shape memory behaviour of these alloys allows for a highly functional and specific use. For example, the nature of bone damage is never identical - there are infinite ways in which a bone can break - so SMA bone clips can be uniquely used for every situation, possibly removing the need for a cast. The road to recovery could be made a lot shorter and much less restricting with SMA bone clips. The Niionomi et al paper [5] gives a very up to date comparison of current biomedical alloys and shows in particular how the field of SMA in biomedical science is very cutting edge.


[1]. Shape Memory Alloys, available from http://www.azom.com/article.aspx?ArticleID=1744 [accessed 18/3/13].

[2]. Grant, David M. , Shape Memory Alloy Demonstration, University of Nottingham, available from: http://www.youtube.com/watch?v=1rrPv5AlVXg [accessed 15/3/13].

[3]. Arciniegas, M. et al (2007) Study of hardness and wear behaviour of NiTi shape memory alloys, Journal of Alloys and Compounds, 460, p213-219.

[4]. Primary Metallic Crystalline Structures, available from http://goo.gl/TqJgL [accessed 18/3/13].

[5]. Niinomi, M et al (2012) Development of new metallic alloys for biomedical applications, Acta Biomaterialia, 8, p3888-3903.

[6]. Shape Memory Alloys and Their Applications, available from http://www.stanford.edu/~richlin1/sma/sma.html [accessed 19/3/13].

[7]. How shape memory alloys are “trained”, available from http://goo.gl/CMvVC [accessed 20/3/13].

[8]. Xiong, J.Y et al (2007) Titanium-nickel shape memory alloy foams for bone tissue engineering, Journal of the Mechanical Behaviour of Biomedical Materials, 1, p269-273.

[9]. Definition of a Shape Memory Alloy, available from http://goo.gl/wrTQ5 [accessed 21/3/13].


Respond to this article

Got something to say? Be the first to comment.

More from Smart Materials

Smart Materials: Every little thing

Wendy Carr

Wendy introduces the increasingly relevant topic of smart materials.

Similar articles

Smart machines all around you

Jennifer Wood

Can viruses be considered smart?

Intelligence of the unexpected

Victoria Behr

A look at some organisms you might not have previously considered 'smart'

Mucky motherboards

John Allan

Using bacteria as computers...

Rats given infrared 'sixth sense'

Alice Tobin

Scientists have given rats a 'sixth sense'

Artificial intelligence: the next level

Nash Matinyarare

The Blue Brain Project: can scientists build a realistic computer model of the brain?