Nickel titanium, also known as nitinol, is a metal alloy of nickel and titanium, where the two elements are present in roughly equal atomic percentages. Nitinol alloys are named according to the weight percentage of nickel; e.g., nitinol 55 and nitinol 60.
Nitinol alloys exhibit two related and unique properties: the shape memory effect and superelasticity (also called pseudoelasticity). Shape memory is the ability of nitinol to undergo deformation at one temperature, stay in its deformed shape when the external force is removed, then recover its original, undeformed shape upon heating above its "transformation temperature". Superelasticity is the ability for the metal to undergo large deformations and immediately return to its undeformed shape upon removal of the external load. Nitinol can undergo elastic deformations 10 to 30 times larger than alternative metals. Whether nitinol behaves with shape memory effect or superelasticity depends on whether it is above its transformation temperature during the action. Nitinol displays the shape memory effect when it is colder than its transformation temperature, and superelasticity when warmer.
The word "nitinol" is derived from its composition and its place of discovery, Nickel (Ni) - Titanium (Ti) - Naval Ordnance Laboratory (NOL).
History The shape-memory effect was discovered in 1932, when Swedish chemist Arne Ölander observed the property in gold–cadmium alloys. The same effect was observed in Cu-Zn (brass) in the early 1950s.
In the 1950s, William J. Buehler was tasked with finding alloys with high resistance to fatigue under high temperatures to be used in missile nose cones at the Naval Ordnance Laboratory in 1959.
In 1961, they presented a sample at a laboratory management meeting. The sample, folded up like an accordion, was passed around and flexed by the participants. One of them applied heat from his pipe lighter to the sample and the accordion-shaped strip contracted and retook its previous shape.
One year later, the Naval Ordinance Lab announced the successful creation of nitinol. In 1962, Frederick E. Wang joined the team at NOL and discovered the crystal structure responsible for nitinol's properties.
While potential applications for nitinol were realized immediately, practical efforts to commercialize the alloy took place in the 1980s, largely due to the difficulty of melting, processing and machining the alloy.
Mechanism
Nitinol's unusual properties are derived from martensitic transformation, a reversible solid-state phase transformation between two different martensite crystal phases, requiring of mechanical stress.
At high temperatures, nitinol assumes an interpenetrating, cubic structure referred to as austenite (also known as the parent phase). At low temperatures, nitinol spontaneously transforms to a more complicated monoclinic crystal structure known as martensite (daughter phase). Four temperatures are associated to the austenite-to-martensite and martensite-to-austenite transformations. Starting from full austenite, martensite begins to form as the alloy cools to the martensite start temperature, or Ms. The temperature at which the transformation is complete is the martensite finish temperature, or Mf. When the alloy is fully martensite and is subjected to heating, austenite starts to form at the austenite start temperature, As, and finishes at the austenite finish temperature, Af. The cooling/heating cycle displays thermal hysteresis. The hysteresis width depends on the precise nitinol composition and processing. Its typical value requires a temperature range spanning about that can be reduced or amplified by alloying and processing.
Crucial to nitinol properties are two key phase transformation aspects. First is that the transformation is "reversible", meaning that heating above the transformation temperature reverts the crystal structure to the simpler austenite phase. Second, the transformation in both directions is instantaneous.
Martensite's crystal structure (known as a monoclinic, or B19' structure) has the unique ability to undergo limited deformation in some ways without breaking atomic bonds. This type of deformation is known as twinning, which consists of the rearrangement of atomic planes without causing slip, or permanent deformation. It is able to undergo about 6–8% strain in this manner. When martensite reverts to austenite by heating, the original austenitic structure is restored, regardless of whether the martensite phase was deformed. Thus the shape of the high temperature austenite phase is "remembered," even though the alloy is deformed at a lower temperature..
Pressure can be produced by preventing the reversion of deformed martensite to austenite—from to more than . One reason that nitinol works so hard to return to its original shape is that it is an ordered intermetallic compound. In an ordinary alloy, the constituents are randomly positioned in the crystal lattice; in an ordered intermetallic compound, the atoms (in this case, nickel and titanium) occupy specific locations in the lattice. The fact that nitinol is an intermetallic is largely responsible for the complexity in fabricating devices made from the alloy.
To fix the original "parent shape," the alloy must be held in position and heated to about . This process is usually called shape setting. A second effect, called superelasticity or pseudoelasticity, is also observed in nitinol. This effect is the direct result of the fact that martensite can be formed by applying a stress as well as by cooling. Thus in a certain temperature range, one can apply a stress to austenite, causing martensite to form while at the same time changing shape. In this case, as soon as the stress is removed, the nitinol will spontaneously return to its original shape. In this mode of use, nitinol behaves like a super spring, possessing an elastic range 10 to 30 times greater than that of a normal spring material. There are, however, constraints: the effect is only observed up to about above the Af temperature. This upper limit is referred to as Md, which corresponds to the highest temperature in which it is still possible to stress-induce the formation of martensite. Below Md, martensite formation under load allows superelasticity due to twinning. Above Md, since martensite is no longer formed, the only response to stress is slip of the austenitic microstructure, and thus permanent deformation.
Nitinol is typically composed of approximately 50 to 51% nickel by atomic percent (55 to 56% weight percent). Making small changes in the composition can change the transformation temperature of the alloy significantly. Transformation temperatures in nitinol can be controlled to some extent, where Af temperature ranges from about . Thus, it is common practice to refer to a nitinol formulation as "superelastic" or "austenitic" if Af is lower than a reference temperature, while as "shape memory" or "martensitic" if higher. The reference temperature is usually defined as the room temperature or the human body temperature ().
One often-encountered effect regarding nitinol is the so-called R-phase. The R-phase is another martensitic phase that competes with the martensite phase mentioned above. Because it does not offer the large memory effects of the martensite phase, it is usually of no practical use.
Manufacturing
Nitinol is exceedingly difficult to make, due to the exceptionally tight compositional control required, and the tremendous reactivity of titanium. Every atom of titanium that combines with oxygen or carbon is an atom that is robbed from the NiTi lattice, thus shifting the composition and making the transformation temperature lower.
There are two primary melting methods used today. Vacuum arc remelting (VAR) is done by striking an electrical arc between the raw material and a water-cooled copper strike plate. Melting is done in a high vacuum, and the mold itself is water-cooled copper. Vacuum induction melting (VIM) is done by using alternating magnetic fields to heat the raw materials in a crucible (generally carbon). This is also done in a high vacuum. While both methods have advantages, it has been demonstrated that an industrial state-of-the-art VIM melted material has smaller inclusions than an industrial state-of-the-art VAR one, leading to a higher fatigue resistance. Other research report that VAR employing extreme high-purity raw materials may lead to a reduced number of inclusions and thus to an improved fatigue behavior. Other methods are also used on a boutique scale, including plasma arc melting, induction skull melting, and e-beam melting. Physical vapour deposition is also used on a laboratory scale.
When melting nitinol by induction (or any titanium alloy) it is hard to find adequate material for crucibles. Molten titanium is chemically corrosive, requires extreme temperatures and specific conditions to avoid contamination. For example, if a granite crucible is used, the carbon in granite can bind with titanium resulting in an impure alloy. Another thing to consider is erosion of the crucible walls. Factors that play into the amount of erosion include but are not limited to the movement of the molten nitinol, the crucible's porosity, and the time the molten metal is touching the material. The porosity of a crucible must be balanced so that the crucible has good temperature shock resistance and a short wetting distance, so it does not erode. There are few materials for crucibles that work with titanium casting which is a contributing factor to the difficulty in mass production of nitinol. Experiments are being done to use a coating over a crucible to block chemical interactions so that the base material only must be heat shock resistant and avoid erosion. For example, Y2O3 is being researched as a coating in an Aluminum oxide crucible. The Y2O3 does not chemically interact with the titanium, and the aluminum oxide is heat shock resistant making a suitable combination.
Another method to manufacture nitinol that is less used is reactive sintering. Reactive sintering in a vacuum is used to compress nitinol powder into a shape. The powder is heated up to around about 580 and 650 degrees Celsius via an exothermic chemical reaction and pressed into shape. This method produces a porous surface and because it is in a vacuum the titanium does not have the chance to bond to oxygen. However, the result is not necessarily homogeneous. It can form ni3ti and ti3ni due to the slow heating produced from the reaction. Heating it quickly can improve homogeneity and reduce the size of the pores. Spark plasma sintering was invented in 1980 as a method to make the heating as rapid as possible by sending an electric current through the system. This way the nitinol is purer without ni3ti or ti3ni forming.
Heat treating nitinol is delicate and critical. Aging time and temperature control the precipitation of various Ni-rich phases, and thus controls how much nickel resides in the NiTi lattice; by depleting the matrix of nickel, aging increases the transformation temperature. The combination of heat treatment and cold working is essential in controlling the properties of nitinol products. Nitinol that is made with a focus on shape memory effect is heat treated in a range of 350 degrees Celsius to 450 degrees Celsius. For superelastic Nitinol the temperature is closer to 500 degrees Celsius. If the nickel component is greater than 55.5% then the heat treatment temperature is in the range of 600 to 900 degrees Celsius
Additive Manufacturing