Little Big Wire.

High-temperature superconductivity makes a bid for the power grid

In 1986, scientists discovered a new family of materials that can conduct electricity with absolutely no resistance. Because these so-called superconductors work at much higher temperatures than any previously identified superconductors, the discovery was considered one of the most important of that decade. The findings triggered a wave of euphoria that flooded scientific meetings and the popular press.

At an American Physical Society meeting in New York City that convened just months after the discovery hit the news, thousands of researchers overflowed a huge hall. An all-night session was filled with presentations about the phenomenon of high-temperature superconductivity and speculations about its impact. That event became known as the Woodstock of physics.

Because the materials potentially could carry huge currents resistance-free and because cooling them to their effective temperature is far less onerous a task than it is for their lower-temperature cousins, many scientists turned visionary and predicted a technological revolution.

"We physicists were the darlings of the media for a year," recalls Paul M. Grant of the Electric Power Research Institute (EPRI) in Palo Alto, Calif. At the time, he led the research group at IBM's Almaden Research Center in San Jose, Calif., that determined the atomic structure of one of the materials. "If someone shoves a mike in your face ... you can say some things that, in retrospect, are pretty silly," he says.

To be sure, the revolution has yet to materialize, mainly because the new substances--complex and brittle ceramics--proved to be tougher to tame than the early pundits suspected. But the dream didn't die.

Recently, researchers and designers have started demonstrating full-scale prototypes of electric power cables, motors, transformers, and other equipment made of high-temperature superconducting (HTS) wire. The new devices waste less energy than their conventional counterparts. The energy requirement is less even when taking into account the power to cool them to -196 [degrees] C--or often lower temperatures when strong magnetic fields are present. Moreover, the new superconducting equipment is smaller, lighter, safer, and often more environmentally friendly than conventional devices.

"We're offering a way to make [motors and other electrical equipment] as perfect as they're ever going to get," says James Daley, manager of the Department of Energy's (DOE) superconductivity program, which is based in Washington, D.C.

In Detroit next year, if all goes well on one of the projects sponsored by Daley's office, power will flow through HTS cables to commercial customers for the first time.

The main component of the HTS devices now beginning to prove themselves is wire made with compounds of bismuth, strontium, calcium, copper, and oxygen. These matericals are generically known as BSCCO (pronounced bisco).

However, the high cost of using BSCCO looms as a serious obstacle to its wide use. That cost may come down through high-volume production of the wire and improvements in its performance. Still, even BSCCO manufacturers are looking for less expensive alternatives to the material.

In labs throughout the world, researchers are racing to develop wire from another high-temperature superconductor called yttrium barium copper oxide, or YBCO (pronounced ibco). YBCO wires may cost less than BSCCO ones. They're also expected to carry higher currents than BSCCO wires in the presence of the magnetic fields found in many types of equipment.

If YBCO wire can be made in practical lengths--a challenge that arose as soon as YBCO was discovered in 1987--high-temperature superconductivity may finally offer both cost low enough and performance high enough to become a central technology of the 21st century, say many industry observers. "It's a big leap of faith to go to YBCO," notes Donald U. Gubser of the Naval Research Laboratory in Washington, D.C. But it's a leap that "could lead to tremendous rewards," he adds.

By ordinary human scales of hot and cold, there's nothing really high temperature about high-temperature superconductors. As the materials are cooled from ambient temperature, their superconductivity kicks in at so-called critical temperatures. But those temperatures are high only in comparison with the critical temperatures of the more aptly named low-temperature superconductors.

Those substances, mostly metals or metal alloys, lose all resistivity when chilled to a temperature that liquefies helium, which is -269 [degrees] C. By contrast, the critical temperatures of high-temperature superconductors tend to be above -196 [degrees] C, the temperature at which nitrogen liquifies. That's still profoundly cold, but it's much less of a challenge for refrigeration systems to meet.

Building practical power devices with high-temperature superconductors entailed making them into a suitable wire. That took more than a decade, but finally developers figured out ways to reliably encase the granular, brittle BSCCO in silver and extrude the package into long filaments.

Further rolling and heating coerces the material, on a microscopic level, to align in a manner that permits large currents to flow. The technique, which researchers developed by about 1990, works because BSCCO crystals have a layered structure--"like a deck of cards," says Gregory J.Yurek, a former Massachusetts Institute of Technology metallurgist who co-invented the idea of blending BSCCO with metal. He's now president of HTS-wire maker American Superconductor Corp. in Westborough, Mass.

Rolling shatters the weak bonds between the cardlike flakes of BSCCO, spreading each deck lengthwise by sliding layers past each other. Heating then merges overlapping flakes, creating an extended, continuous path along which current can flow.

Even with this wire-making tactic in hand, there was still a long slog ahead to practical applications. "A lot of hard work has gone into fine-tuning the process incrementally so today you have wire with high enough performance," EPRI's Grant notes. At present, the maximum amount of current per cross-sectional area, or critical current density, of BSCCO--roughly 70,000 amperes per square centimeter (A/[cm.sup.2]) at -196 [degrees] C-- is about three times what it was 4 years ago, he says. Above the critical current density, a superconducting wire starts to resist electric flow.

Because BSCCO wires are made only partially of the pure superconducting material, the actual critical current density of wires reaches only about 15,000 A/[cm.sup.2]. Nonetheless, that's 15 to 150 times copper's maximum current density, which varies dramatically depending on how well a cable or other device containing the metal can dissipate heat.

Because of their high capacity, HTS cables can be so svelte that Detroit's local electric utility, in partnership with DOE and other institutions, is replacing nine underground, oil-cooled copper power cables that serve nearly 14,000 customers with just three liquid-nitrogen-cooled BSCCO cables. By next summer, when the swap should be complete, a mere 900 pounds of BSCCO wire will have taken the place of 25,000 lbs. of copper wire.

DOE estimates that because of electrical resistance and other losses, more than 7 percent of all the power currently generated in the United States is wasted. By converting the nation's power grid to HTS technology, "we can save half the existing losses," Daley claims.

Last July, in a milestone for practical high-temperature superconductivity, another DOE project led by Rockwell Automation in Greenville, S.C., began to test the first HTS motor powerful enough to potentially find wide use in industry. According to DOE, motors rated at 1,000 horsepower or more eat up a large chunk--about 30 percent--of U.S. electric power. The prototype 1,000-horsepower, BSCCO-based motor is currently half the size and weight of a conventional motor but shows about the same efficiency. However, its developers say that they expect to slice its energy losses in half.

Meanwhile, the same consortium is building a 5,000-horsepower motor. And the U. S. Navy has been exploring the possibility of 25,000-33,000-horsepower HTS motors to drive all-electric ships. "Twenty-five-thousand-horsepower motors exist today, but they're so big and heavy that [ships] don't use them," Gubser notes. An HTS motor of that power would tip the scales at about 40,000 lbs., a third the weight of a conventional motor, he estimates.

The green potential of HTS devices is soon to be put to the test by Waukesha (Wisc.) Electric, DOE, and their partners. Early next year, they expect to install a large, BSCCO-based transformer at a utility substation. Because of their low temperatures, HTS transformers don't require the hazardous oils that cool conventional transformers.

The oils, which often contain toxic, flame-retarding compounds called PCBs, sometimes leak into the ground, Daley says. By eliminating the oils, HTS transformers and power cables offer an environment-protection bonus over conventional technology, Daley notes. Moreover, the new transformers could be used in buildings where conventional transformers have been banned because of the risk that their cooling oils will catch fire.

Despite the apparent advantages of high-temperature superconductors, major economic and technical hurdles remain as engineers push the new materials toward the power grid.

BSCCO is costly because it's encased in silver. Currently, BSCCO wire costs 20 to 30 times as much as copper wire does. However, a number of superconductivity veterans predict the price will drop to a tenth of today's figure within the next 3 to 5 years, as long as the predicted demand for HTS cables and devices materializes.

Contributing to that anticipated price drop, supply should also increase dramatically because American Superconductor is building the first large-scale plant for making the wire. By 2002, the company plans to be churning out 10,000 kilometers of it per year. "We expect the cost of production to fall by about a factor of six when we open that plant," says company spokesman John B. Howe.

As BSCCO becomes more abundant, its quality is also likely to continue to improve, Gubser says, so the cost of making devices from the wire may diminish.

In the meantime, wire researchers are lavishing most of their funding and laboratory studies on what they hope will be BSCCO's successor--YBCO. "People believe that YBCO wire can be manufactured with significantly lower cost" than BSCCO wire, Gubser says.

That's because scientists have found ways to make YBCO wire by starting with a strip of inexpensive metal such as nickel. The researchers then coat the strip with a few layers of materials that keep the metal atoms from contaminating the superconductor, and they finally add a thin YBCO topcoat.

Because YBCO has a high critical current density, that last layer can be extremely thin but still enable the wire to carry as much current as BSCCO wire--and possibly more, scientists say. Researchers have measured up to 3 million A/[cm.sup.2] in some YBCO wire samples at -196 [degrees] C, says David C. Larbalestier of the University of Wisconsin-Madison. That's more than 40 times the critical current density of BSCCO (SN: 4/29/95, p. 269).

Yet making YBCO wire is more easily said than done. So far, no one has found a way to fabricate it in practical lengths. The challenge has been to get the YBCO, which lacks the laminar structure of BSCCO, to lie down in a uniform sheet that has crystal lattices lined up nearly perfectly in three dimensions from one grain to the next.

Why must grain boundaries line up so well? Think of superconductivity as a flow of pairs of electrons, called Cooper pairs, says Oak Ridge (Tenn.) National Laboratory physicist John D. Budai (SN: 2/11/95, p. 88; 6/7/97, p. 351). Those electrons "have to dance around each other in a careful way so as to never run into the atoms in the crystal," says Budai, who a decade ago developed one of the main ways of aligning YBCO films. "They can't keep doing that when they hit the boundary between two [misaligned] grains" because there the atomic layout abruptly changes, he says.

By using very expensive fabrication techniques, researchers can now align their YBCO grains to within a 10 [degrees] angle of each other, which allows for adequate currents--but only up to a half-dozen meters. Beyond that, irregularities still wreck the wire's properties.

Interestingly, a sprinkle of calcium where the grains touch could be a solution. In an experiment on just two grains meeting at a 24 [degrees] angle, Jochen Mannhart of the University of Augsburg in Germany and his colleagues found that a dose of calcium could boost current through the intersection sixfold. The calcium may directly modify the border, researchers suspect. Also, because there's a shortage of electrons in calcium ions compared with the yttrium ions that they displace, the calcium effectively injects additional positive charges that can ferry current across the boundary region. The researchers describe their findings in the Sept. 14 NATURE.

In a commentary accompanying that report, Grant calls this calcium doping "the best idea that's come along in the 5 years since coated conductors first emerged."

Yurek says that he doubts such doping will benefit YBCO films because the grains are already much better aligned than those in the German experiment. Budai and Daley also doubt that added calcium is necessary for practical wire-making.

While such uncertainties surrounding YBCO continue to challenge research scientists and engineers, the demonstrations of BSCCO-based devices are convincing utilities and industry that HTS technology works and is reliable. Says Larbalestier, "Technically, the picture is clear. Those things work."

That should make it easier for YBCO-based technology, if and when it becomes ready, to take over what BSCCO has begun.