Winnipeg, Canada
MIX TWO metals together and they blend to form an alloy, right? Well, not always. A new group of materials with outstanding properties actually depends on the fact that the metals it is made of don鈥檛 mix. When molten mixtures of these stand-offish metals cool down, blobs of one metal form inside the other. Stretch the materials and the blobs turn into ultrafine filaments which give the new materials their unique strength and heat resistance. So unusual are these materials that no-one knows what to call them.
No one that is except Alan Russell and Scott Chumbley of the US Department of Energy鈥檚 Ames Laboratory in Iowa, who have spent the past six years investigating the metal-metal composites, as they describe them. 鈥淚 would like to say we had a grand cosmic vision,鈥� says Russell, an associate scientist at Ames and assistant professor at Iowa State University. 鈥淏ut really much of what has happened has been due to good luck.鈥� Because their metal-metal composites are stronger and lighter than most competing engineering materials, Russell and Chumbley鈥檚 good fortune is attracting the interest of companies that make everything from jumbo jets to artificial hips.
The materials consist of a matrix metal, which makes up the bulk of the composite, with thin filaments or ribbons of another metal running through it. This structure combines the benefits of two engineering mainstays-alloys and composites. Alloys, such as bronze or steel, are mixtures of metals with other metals or elements such as carbon. The end product is often stronger, more resistant to corrosion or more flexible than the pure metals. Composites are fabricated by reinforcing a matrix material, such as a metal or plastic, with fibres made from other materials, including glass or silicon carbide. Composites tend to be strong, stiff and lightweight (see 鈥淐ars that grow on trees鈥�, New 杏吧原创, 1 February, p 36).
Advertisement
The researchers have shown that metal-metal composites are incredibly strong for their weight, very resistant to heat, and flexible enough to be processed into a variety of shapes. 鈥淟ight yet strong is the key [selling point],鈥� says Chumbley, a metallurgist at Ames and associate professor in materials science at Iowa State University. The materials also score points over composite materials such as aluminium reinforced with silicon carbide because they are much easier to work with. 鈥淭here is one huge headache associated with using aluminium-silicon carbide for normal applications,鈥� says Russell. 鈥淚t is very difficult to machine.鈥� Not so with the new materials, which are less abrasive and much better at dissipating heat produced by cutting or drilling.
During the past two years, Russell has been experimenting with aluminium as the matrix. The result, he says, has been 鈥渋nstant gratification鈥�. While commercially available aluminium alloys have a tensile strength-a measure of the load they can carry-of between 400 and 675 megapascals, Russell鈥檚 aluminium-based materials are already touching the 1200 megapascals mark. 鈥淎 cable of this material 45 millimetres thick could support the weight of an adult blue whale,鈥� he says. That鈥檚 as much as 190 tonnes. Not bad for materials originally discovered by accident.
The run of good luck began in the mid 1970s when Joze Bevk of Harvard University was trying to produce a superconducting wire from an alloy of niobium and tin, Nb3Sn. This alloy is too brittle to be drawn out into a wire. So Bevk mixed copper and niobium, made it into a wire, and passed the wire through molten tin, allowing tin atoms to diffuse through the copper before reacting with the niobium.
Fortunately, along the way Bevk stopped to examine the properties of the copper-niobium intermediate. Before drawing, the strength was 鈥渘ot very impressive-certainly nothing to write home about,鈥� says Russell. But drawing it into a wire gave a dramatic and unexpected increase in its tensile strength. Anyone who didn鈥檛 know better would have expected the strength of the final product to have been somewhere between that of copper, at 500 megapascals, and niobium鈥檚 1400 megapascals. In fact it was actually more than 2000 megapascals.
Tiny fibres, mighty muscles
This sudden jump in strength comes from copper and niobium鈥檚 mutual dislike for each other. In Bevk鈥檚 material, the niobium atoms clustered together and produced small blobs of niobium dotted throughout the copper matrix. As the metal was drawn into a wire, the niobium blobs were stretched out into very fine filaments. What the blobs gained in length, they lost in thickness, shrinking from about 5 micrometres to just 10 nanometres wide. These tiny filaments, just 30 atoms across, were the extra muscles of the new material.
But how do tiny microfilaments act as mighty muscles? The answer lies in their ability to stop cracks spreading through the matrix. Metals break because they contain imperfections known as dislocations. These can form during hardening from the molten state or when the material is bent or twisted. 鈥淭he atoms in a metal lie in stacked planes. A dislocation occurs when there is a disruption within this crystal lattice and the planes no longer meet at a certain point,鈥� explains Russell.
Dislocations are weak points in the metal structure where cracks form and propagate until they reach the surface. The niobium microfilaments halt this process in several ways. Dislocations moving through the copper stall when they hit the copper-niobium interface, because the planes of atoms in the two metals are tilted at different angles and have different spacing, so they don鈥檛 line up. But even if they miss, the number of directions in which travelling dislocations can veer off without hitting another filament is very small.
In addition, the energy involved in forming a dislocation in a tiny filament is huge. 鈥淚t is difficult for a dislocation to form in a small crystal,鈥� notes Russell, 鈥渁s all the strain associated with the dislocation has to be taken up by a very few atoms.鈥� He says that the virtually defect-free ultrafine metal whiskers are 鈥渢oo nearly perfect鈥� to form a dislocation, and so are very strong-much stronger in fact than the bulk metal from which they are made. This helps explain why the composites鈥� strengths are greater than the sum of their parts.
Bevk鈥檚 chance finding spawned detailed investigations into other metal combinations at the Ames Laboratory by John Verhoeven in the 1980s and now by Russell and Chumbley. Their first move was to replace copper with titanium as the matrix metal. Titanium is lighter, stronger and more resistant to heat than copper, but it still wasn鈥檛 the perfect matrix. 鈥淚t sort of worked, but it wasn鈥檛 ideal. The material didn鈥檛 deform well,鈥� says Russell. Since then they鈥檝e chopped and changed both the matrix and the filament metals, using magnesium, iron and aluminium in their search for useful materials.
One combination made the record books not for its strength but for its heat resistance. Using a magnesium matrix, and either titanium or niobium as the filament metals, Russell and Chumbley produced bimetallic materials not only a little stronger than off-the-shelf magnesium alloys, but more importantly, capable of withstanding temperatures up to 400掳C without deforming-the most heat-resistant magnesium-based material ever produced.
Strong and stable
The ability of materials to withstand high temperatures for a long time is much sought after by aircraft designers. Jet planes are subject to enormous temperatures generated by air friction. But metal alloys often do not hold up well under these conditions because of a phenomenon known as coarsening. At high temperatures, atoms of each metal in an alloy tend to move around and eventually gather together into spherical clusters-a stable formation because it minimises the surface area of the clump of atoms. But these spheres weaken the material because they disrupt the microstructure of the alloy. If the thin filaments in the metal-metal composites were transformed into spheres, they would lose their effectiveness at stopping dislocations spreading.
The magnesium-titanium and magnesium-niobium composites are resistant to coarsening at 400掳C because the filament materials have such high melting points, Russell believes. The titanium and niobium atoms do not move even at 400掳C. The team鈥檚 aluminium-based composites on the other hand resist coarsening thanks to strong bonds between the filaments and the matrix, which keep the atoms in place. Russell thinks that the metals in his aluminium composites actually react with each other to form intermetallic compounds at the interface between the matrix and the filaments. Intermetallics are strong and stable at high temperatures, and so prevent the atoms in the filaments from diffusing and forming spheres.
Perhaps the only drawback to Russell and Chumbley鈥檚 composites is the amount of processing required in their production, which makes them more expensive than competing materials such as aluminium alloys for aeroplanes and steel for cars. 鈥淵ou don鈥檛 build aeroplanes out of wire,鈥� Russell points out, so the composites must be processed into a suitable form. One method the Ames team has come up with involves chopping up the wire and sorting it into bundles-like packs of cigarettes-which are then fused to make solid sections of the metal-metal composite. The technique works but is labour intensive and so the finished product is more expensive.
Making microfilaments
To get around this problem, Russell and Chumbley have joined forces with aircraft manufacturer McDonnell Douglas, which is interested in using the new materials as airframe skins for future aircraft. Together they have commissioned research into a process called equal-channel angular extrusion, which is currently being carried out by Ted Hartwig in the department of mechanical engineering at Texas A&M University. To start with, the two metals are first mixed as powders and compacted to form an ingot or bar. Hartwig then extrudes this through a right-angle bend, subjecting the material to a high shear stress. Gradually, by running the ingot back and forth through the 90掳 bend, microfilaments of one metal are created inside the matrix of the other, without changing the dimensions of the bar. The composite can be processed into products much more easily in this form.
Russell hopes the method will eventually produce a lightweight, relatively inexpensive aluminium composite with a strength of between 800 and 900 megapascals. Such a material would be a big hit with aircraft manufacturers, who could then reduce the weight of components without sacrificing their strength. Richard Lederich, a technical specialist with McDonnell Douglas Phantom Works R&D division, believes this goal is 鈥渞easonable and achievable鈥� and thinks that the equal-channel angular extrusion process could be easily scaled up if it proves successful. But cost will remain the crucial factor. 鈥淭en to twenty per cent higher [than today鈥檚 materials] would be acceptable,鈥� says Lederich. Above that, and McDonnell Douglas will lose interest.
Most aircraft manufacturers use a titanium-aluminium-vanadium alloy-which has been around since the 1960s-for outer skins and some machined parts. The mass production of a lighter material becomes economically feasible if its cost is likely to be offset by fuel savings over the life of an aircraft. The researchers have been looking at a titanium-yttrium composite with promising properties, but Lederich says the high cost of yttrium makes this route doubtful. 鈥淲e are driven by affordability and I can鈥檛 see it being cheap enough.鈥� He believes the most hopeful composite is aluminium-titanium, because both metals are cheaper.
Chumbley thinks the major market for the materials is in the aircraft industry. 鈥淲e would like to use these materials in automobiles but that field is so competitive that even a half-cent difference in price can keep a new material out. Our composites may be better, but will that justify a higher price?鈥� Aircraft manufacturing is not quite as cost-sensitive, he says. 鈥淐ustomers should be willing to pay slightly more for lighter yet stronger planes with lower fuel costs.鈥� For the same reason, Chumbley can also see the materials being used to build a space station. As the alloys are light, they would need less thrust to put them in orbit, he says.
Russell points out that there is more to the story than just lighter, stronger alloys. The new composites have 鈥渁 cornucopia of benefits鈥�, he says. 鈥淭he copper-niobium composites have a very nice side effect. As the niobium strands run throughout the axis of the copper matrix, they don鈥檛 affect the conductivity of the wire.鈥� High-strength high-conductivity copper wire is in demand because electric motors are manufactured by winding wire onto a fast-spinning core. Copper wire often breaks, and attempts to strengthen the copper have tended to reduce its conductivity, sometimes by as much as 90 per cent. The copper-niobium composite retains its conductivity while gaining in strength, giving the best of both worlds. NRC of Newton, Massachusetts has decided to scale up the copper-niobium production process and believes that this composite will also find uses in high-power electric switches, because it has better wear resistance than existing metals and alloys.
Medical implants
The Ames team has other corporate partners too. One undisclosed company wants to use metal-metal composite technology to strengthen the gold wire solder junctions used in silicon chip manufacture. These ultrafine gold wires, a mere 20 micrometres across, can fail either because they are not strong enough or because the gold near the weld has been weakened by the heat of ultrasonic welding used in the production process. Chumbley and Russell are hoping to strengthen the gold wires by reinforcing them with a ductile metal.
Another company, medical manufacturer Smith and Nephew Richards of Memphis, Tennessee, is also interested in metal-metal composites. The company believes they may be useful in implants such as artificial hips, giving stronger and more fatigue-resistant devices. The main problem in producing implants, explains Chumbley, is trying to match the mechanical properties of bone, which may be possible with the titanium-based composites. This could prevent the loosening that occurs with implants that are too different from bone.
Meanwhile, Russell and Chumbley continue to explore other potential applications, while at the same time tweaking the combination of metals to see if the materials will spring any more pleasant surprises. In the near future, they intend to switch from an ultrahigh-purity aluminium matrix to a commercial aluminium alloy which will be both stronger and cheaper. Further down the road, Russell would like to produce trimetallic composites, made with filaments of two distinct metals, to see what exciting properties they have. But unlike Bevk before him, he knows exactly what he鈥檚 looking for.
