From lab to market: Can it be done?
Boston — Think about making wire from a brick. That's what it's like for researchers trying to make useful shapes from recently discovered ceramic superconductors, says Donald M. Smyth, director of the Materials Research Center at Lehigh University in Pennsylvania.
From Bombay and Boston to Paris and Peking, thousands of scientists and engineers are trying to solve the puzzles posed by a new class of materials that lose all resistance to electricity at temperatures that, while still frigid, were thought to be impossibly high only six months ago.
Their efforts fall into three broad areas: working the copper-oxide ceramic materials into useful forms; trying to figure out what makes them work; and looking for ways to push toward less-frigid temperatures.
No one harbors any illusions about the difficulties that lie ahead in trying to take these materials from the lab to the marketplace. Indeed, some scientists don't give them much of a market niche beyond replacing technology already available at colder liquid helium temperatures - encouraging, they say, but not revolutionary.
But given the sheer number of people working on the new superconductors - and the potential economic and scientific bonanzas that may lie ahead - many other researchers are optimistic about the outcome. ``We can't say now where it's all going to end up,'' says Neil Ashcroft, a Cornell University physicist. ``But the pace of discovery is so rapid that within a year you'll have some pretty solid predictions.''
``I do not see any inherent roadblocks'' to practical applications of these new materials, Dr. Smyth says.
From the standpoint of electrical properties, the last of those roadblocks were shoved aside in late May when scientists at IBM's Thomas J. Watson Research Center at Yorktown Heights, N.Y., announced that a carefully prepared thin-film crystal of the new superconducting material carried enough current to be of practical use in electrical circuits. Up to that point, it was already clear that the material was a superconductor and that in principle it could withstand extraordinarily large magnetic fields without losing its resistance-free properties.
The scientists built their crystal - atoms at a time - out of layers of individual grains of material. The grains within each layer were aligned side by side, like so many pancakes. They found that the crystal's current-carrying ability was some 30 times better going ``with the grain'' than against it.
``There's no hype left because of the IBM results,'' says James F. Smith, head of the Center for Materials Science at the Los Alamos National Laboratory in New Mexico.
``I'm predicting two years before we see the first [small-scale] devices,'' says Lehigh's Smyth. By small-scale, he means electronics.
Researchers at IBM and the National Bureau of Standards, among others, have used the new ceramic materials to make rudimentary versions of devices known as SQUIDs, which are extremely sensitive to magnetic fields and can be instruments for areas as diverse as medicine, geology, space research, physics, and defense. IBM's device was made of a thin film about 100 times thinner than a human hair.
Thin films of the ceramic material are likely to find their first application as connecting lines between components on circuit boards. IBM scientists have also found that the copper-oxide materials can be ``painted'' on materials used for making printed circuits, using a high-temperature technique called plasma spraying.
``This means you can start beating heat problems'' that crop up as designers try to squeeze more and more components into smaller and smaller chips, says Dr. Smith. The next step would be connections between components on individual chips. ``Hybrid'' circuits might include a new generation of transistors made from new semiconductor composites that operate best at liquid-nitrogen temperatures - within the limits of the ceramic superconductors, says Harold Weinstock, program manager for superconductivity at the US Air Force Office of Scientific Research.
Additional impetus for these applications has come from Bell Communications Research in New Jersey. Late last month Bellcore scientists announced a technique that would help chipmakers ensure that the ceramics maintain their superconducting properties during fabrication without damaging the other components on a chip in the process.
Paul Chu, who along with his collaborators first announced the discovery of superconductors that would work at liquid-nitrogen temperatures, says, ``It's also important to look at the normal [non-superconducting] properties of these materials.''
The material of choice at the moment is a made from a recipe of copper, oxygen, barium, and the rare-earth metal, yttrium. Dr. Chu says that a pinch less of one ingredient, or a pinch more of another, can in principle yield nearly all the basic building blocks for a new generation of electronic circuits: superconductors; semiconductors, which form the basis of transistors; and insulators, which don't conduct electricity at all.
Much more work remains to be done. For example, it's still not clear if the new superconductors are suitable for applications in microwave and other radio equipment. But things seem to be humming along in electronics when compared with the agonizing progress in the realm of wires and other so-called bulk forms.
If computer chips are the thoroughbreds of electronics, then wires are the draft horses. Wind wire around a metal object, such as an iron bolt, apply a battery to it, and you have an electromagnet. Wire windings are found in motors and generators. Wire made from copper alloys can be bent in a variety of shapes. But wires made from the ceramic superconductors? ``To my knowledge, no one has made a wire that is both flexible and superconducting,'' says John Rowell, assistant vice-president of solid state science and technology research at Bellcore.
``There's a big difference between making a three-inch or six-inch wafer and a magnet. The average small magnet has 15 kilometers of wire in it,'' says Donald Capone II, of Argonne National Laboratory in Illinois.
Researchers at AT&T Bell Laboratories in New Jersey have poured the ceramic superconductor's powdered ingredients into a thin metal tube. The tube is then stretched, shaped, and fired so that the core becomes the superconducting ceramic. At Argonne, on the other hand, the powders are mixed with a binding material that holds everything together while the copper-oxide mixture is shaped. Once shaped, the ceramic is fired, burning away the binder and leaving the superconductor.
Another approach is to treat the ingredients like metals instead of ceramics. For example, a team at the Massachusetts Institute of Technology substituted europium for yttrium, then combined it with copper and barium to form an alloy. Once the alloy is shaped, it is combined with oxygen at high temperature to turn it into a superconductor.
Unlike their thin-film relatives, wires made from the ceramic superconductors so far cannot yet carry enough current to make them practical. This is because the grains of material in shapes formed by ceramic techniques are a hodgepodge, unlike the tightly formed ranks of perfectly aligned grains in the crystal IBM produced. The MIT team hopes that the alloy route may lead to more uniformly aligned grains, hence a higher ``current density.''
The goal is to carry 100,000 to 1 million amperes for every square centimeter of cross section. Currently the prototype wires are handling only about 1,000 to 10,000 amps per square centimeter.
Some researchers point out that the solution may lie along avenues other than trying to shape new materials into traditional forms.
``You know, we may have to learn better how to exploit the properties of this material,'' says Robert Doremus, chairman of the materials engineering department at Rensselaer Polytechnic Institute in Troy, N.Y., as he surveys efforts to whip ceramic superconductors into shape.
``Motors and generators are made the way they are because of the qualities of metals and wires,'' he says. Instead of struggling to make the new superconductors conform to old ways of making such devices, it might prove more fruitful to rethink the way those machines are made in light of the ceramics' properties.
``What's holding us back is a lack of scientific understanding'' of the new materials, says Bill R. Appleton, director of the solid-state division of the Oak Ridge National Laboratoy. A correct theory not only would help pinpoint the proper ways of mixing and fabricating the new materials for maximum effect, it also might point the way to superconductors that work at even higher temperatures.
To that end, scientists are probing the materials with X-rays and microwaves, pummeling them with neutrons, and peering at them through electron microscopes in an attempt to feed theorists the information they need to explain why the new superconductors work and what roles the chemical ingredients play in the overall recipe.
Indeed, the pace of experiments is so great that the scientists trying to come up with the theory of how these materials work have been left spinning in the dust.
``There's a plethora of information,'' says Princeton University physicist and Nobel laureate Philip W. Anderson. ``The problem is, how trustworthy is it? What was the quality of the sample? I'd like to see some real old-fashioned, simple measurements on single crystals'' of the new materials. Two related concepts are driving scientists to push for higher-temperature superconductors: widespread use, and reliability.
The first isn't difficult to understand. Shoebox-sized supercomputers aren't likely to crop up in every den or office if they require frequent doses of frosty liquid nitrogen, which boils at 77 Kelvins (about 321 degrees below zero, F.).
And even if nitrogen-cooled devices somehow caught on like wildfire, the superconductors they used would have to retain their resistance-free traits at around 150 K. For some applications, it still might be desirable to cool the new superconductors to liquid-helium temperatures (about 4 K), some researchers say.
``It's very important to get superconductors above 90 K'' if liquid nitgrogen is to be the coolant, says Stanford University's Theodore Gabelle. ``Ninety Kelvins is not enough of a margin above 77 K for reliable technology.''
This is because superconductors are somewhat finicky. Not only must they be cooled to very low temperatures. To be practical, they also must withstand large magnetic fields and high electrical currents. These last two characteristics tend to improve as the temperature falls. ``For most applications you want to cool to about half'' of the temperature at which the material loses its electrical resistance, says Dr. Capone.
The most consistent hints of higher temperatures come from researchers who report seeing evidence of superconductivity at between 225 and 240 K. The problem is that the measurements are often indirect, difficult to reproduce, and often can be attributed to other phenomena that up to a certain point mimic superconductivity. Also, the samples involved are unstable.
Still, say some scientists, where there's smoke, there's fire. ``Superconductivity is marvelous: If it's present in a trace, it will let you find it,'' Dr. Gabelle says.
Second of three articles. Next: US and Japan in hot competition.