My daughter and I saw a neat display today which featured Nitinol (a nickel-titanium alloy). Although it had a number of interesting characteristics, we were most impressed by the fact that it contracts when heated. This, of course, is unlike most (?all) other metals, which expand when heated.
Is there a simple way to explain Nitinol’s response to heating? I’ve done a brief search on the net but the sites I’ve been able to connect to aren’t particularly helpful.
Thanks.
A fairly readable explanation here.
Note that nitinol will not just contract, it will change shape in various manners, depending how you form it when cold. It also can respond to electrical input as well.
You can buy little kits online or elsewhere at science supply stores (sadly, in many areas of the US these have gone the way of tailfins on Cadillacs and the 5-cent Burma Shave). Or even individual wires for about $2 at museums in the US, or for about $20 at the Science Museum in London…
I don’t think Nitinol contracts when heated. Rather, it changes shape. When heated, a long thin wire might change to a shorter, fatter one, yet its net volume expand as any metal would.
But you’re just wrong.
There’s a phase change involved.
I wonder if Uri Geller knows about this stuff 
To clarify for the OP: First of all, Nitinol does not contract when heated, in general. Like any other material, it expands when heated (here [pdf file] are the thermal expansion coefficients for Nitinol; they’re in the same range as, say, steel). It doesn’t help that some Web sites (like this one and this one) repeat the claim that Nitinol contracts when heated. Sorry, just not so.
However, you saw something in a demonstration where it appeared that Nitinol contracted when heated. What you saw in the demonstration was certainly associated with the phase change of Nitinol, from martensite (low temperature) to austenite (high temperature). Here’s another page (a little more layman-friendly, IMO) all about NiTiNOL and phase change. In the phase change, the crystal structure of the material will change. In the case of Nitinol, there’s a substantial change in physical properties associated with the phase change; these changes can be exploited to do some interesting things.
One of these interesting things (which you might have seen) is a Nitinol wire with a bias spring. It works like this: One of the properties that alters with the phase change is the stiffness of the wire. You heat the wire, it changes phase and gets stiffer. So when a Nitinol wire is paired with a bias spring, the spring will stretch the wire farther at low temperatures (less stiff martensitic wire) than at high temperatures (more stiff austesitic wire). So the wire contratcs when heated, but only because the spring stretches the wire out at low temperature.
Another interesting thing which you might have seen is the shape memory effect. If a piece of Nitinol is “trained” into shape at a high temperature and then cooled, the metal will “remember” that shape if it is distorted (this page again has an interesting demo). Deformation of the low temperature martensite looks like permanent plastic deformation, but is really only shifting atomic bonds. When the material is re-heated, it will snap back into shape as the phase change alters the crystal structure. This has been used commercially for pipe couplings: form a coupling at high temperature, cool it, and stretch it out. When re-heated, the coupling will “shrink” back into shape. The same procedure can be used for wires: stretch them out at low temperatures, and they’ll shrink back into shape at high temperatures.
Additional note: there is a volume change of the material associated with the phase change (despite claims to the contrary on some sites). However, this volume change is relatively small: about 0.16%. And, if I’m reading this right, the volume actually increases when Nitinol is heated from martensite to austenite. It’s doubtful that this had anything to do with what you observed, though.
Great job Zut, couldn’t have explained it better myself, and I wrote a Master’s Thesis on the stuff.
A couple other things it is used for:
Actuators (using bias springs), such as one that was on the Mars Rover.
Highly flexible metal (can undergo large deformations without even “apparent” plasticity), eyeglasses, underwire bras.
Dampers (bleeds energy out of an ascillating system because of phase change), Washing machines.
They are using the stuff all over the medical field these days too.
Well, I did a PhD dissertation in the area of smart materials in general and piezoelectrics in particular, so I wound up with a general technical background in all types of smart materials, shape memory alloys included. Not nearly as much as someone who specialized in SMAs, but enough to hack out what you read above.
Hmm… [putting a few things together] flight, you don’t happen to be at the RCOE by any chance, do you? If so, you might be interested to know that a part of my dissertation was on actuation of rotorcraft trailing edge flaps.
zut (and others): Thanks! This is excellent.
What I saw, among other things, was ‘the bias spring’ demonstration. As you can imagine, it looked as if the nitinol was contracting. Now I get it.
And, yes, not only do multiple web sites claim it contracts when heated, so did the display at the Scince Centre.
zut and/or flight, since you’re the resident experts on this stuff, can you tell me how Nitinol holds up under tension or shear?
The reason I ask is that I have eyeglass frames supposedly made of the stuff. A few years ago (only a few months after I bought them), the bridge piece broke (a roughly U-shaped piece of ~2 mm wire holding the nose pads and lens frames). The bridge broke precisely at the bottom of the U, producing two symmetrical J-shaped pieces. Fortunately the frames were under warranty and replaced gratis – the replacements have held up to this day.
The tech at the eyeglass shop remarked that he’d never seen a pair of Nitinol frames fail before, so I’d evidently pushed forward the frontiers of materials science. I don’t know if the broken frames were sent back to the manufacturer for failure analysis, but I’ve always assumed that my first set was out-of-spec or that the broken bit wasn’t really Nitinol but some inferior steel or titanium alloy.
Cool, I am now writing my dissertation on BVI reduction through localized root pitch changes. Most of the previous methods used trailing edge flaps, so my library on them is extensive (oh the joy of reading technical papers!).
The Master’s work was on modelling them for use as dampers, though you may have known that…
That is strange. The reason they use them is precisely because they can take such huge loads with high deformation with failure. I would say it was a manufacturing error, or you put them in a wood chipper. The part you are describing is the part that most commonly breaks (other than the hinges), so I would guess that it was Nitinol.
Note that Nitinol (which is a trademarked name) is the most common SMA, but there are others.
The reason eyeglass frames are made from Nitinol is due to its superelasticity (this is another one of those “interesting things” about Nitinol). Short explanations of this phenomenon can be found here and here. This is getting out of my area of expertise, but, in a nutshell, here’s the explanation: the phase transition from austenite to martensite depends on both temperature and applied stress. Nitinol can be manufactured in a wide range of transition temperatures by slightly altering the constituents in the alloy. If you construct eyeglass frames from an alloy with a transition temperature just above body temp, then bending the frames will induce a phase change: martensite is formed. Because of the shape memory effect, further bending will just shift atomic bonds (refer to links in my post above). Then, when you stop bending, the martensite changes phase back to austenite, and the deformation disappears.
As to why your original frames broke, I can only speculate, just like you, that the alloy was out of spec, or, more likely, it cracked during manufacture in some way. Evidently, fatigue cracking of superelastic Nitinol is not well understood, so perhaps no one can definitively answer your question.
flight: Very interesting. I haven’t been following rotorcraft research recently, but passive damping with SMAs came up last time I talked to the guys in my old lab. And, if you’ve got a library of trailing edge flap actuation papers, there’s probably a couple with my name on them.