杏吧原创

St Louis, Missouri

SUPERCONDUCTIVITY is an impressive quantum trick. But you needn鈥檛 be a
magician to pull it off鈥攁 very good refrigerator will do. Just cool down
an ordinary slab of tin or lead to an icy few degrees above absolute zero, and
electrons, as if charmed, will suddenly pass through it effortlessly, like
ghosts through walls, without encountering the slightest flicker of electrical
resistance.

The trick works at far higher temperatures too, in any of the strange ceramic
materials known as high-temperature superconductors. When first discovered
around 1985, these materials seemed to offer a direct path to technological
nirvana. Just around the corner, finally, were electrical transmission lines and
microchips that would use virtually no energy, and cheap, powerful
superconducting magnets for enormous particle accelerators, amazingly efficient
motors and medical imaging devices.

And yet with few exceptions, superconductors remain in the research labs. The
optimism (and hype) of the mid-1980s has itself cooled into a sober recognition
that superconductivity presents a serious intellectual challenge.

When a superconductor leaves the safety of the lab and goes forth into the
real world, it faces practical demands鈥攊t needs to carry large electrical
currents. When these currents flow, they churn up powerful magnetic fields.
Trouble is, these fields circle back and slip inside the material like
saboteurs, destroying its superconductivity. In an instant, a superconductor
turns itself into a normal conductor, or even a useless insulator. It is this
bitter conflict between superconductivity and self-inflicted magnetism that has
condemned high-temperature superconductors to underachievement.

The good news is that at least everyone now agrees that magnetism is a tricky
beast. Researchers have worked out how magnetic fields slip stealthily past the
boundaries of superconductors, and how they spoil the show once inside, by
forming thousands of microscopic tornadoes of swirling electric currents. Like
their larger atmospheric cousins, these tiny whirlwinds cause terrific problems
when they move about. And it is only by learning to control their motion that
researchers are beginning to broker a peaceful coexistence between
superconductivity and magnetism, and to allow the magical technologies of
superconductors finally to become real.

Battle lines

The German physicist Walther Meissner first revealed the physical enmity
between superconductivity and magnetic fields in 1933. Depending on the
prevailing conditions, either side can win the battle. Superconductivity beats a
weak magnetic field every time, rudely expelling it from the interior of a
superconductor. This is the 鈥淢eissner Effect鈥�
(see Diagram). But a
strong magnetic field can turn the tables, penetrate the material and quell its
superconductivity.

At intermediate field strengths, however, things become more complex, and
there is a stalemate between the two: unable to dominate the superconductor as a
whole, the field sends exploratory fingers inside, and establishes control only
within a set of tiny regions. The fingers form thin tubes running from one side
of the superconductor to the other. Outside each tube, there is no magnetic
field and the material remains superconducting. But within each, the
superconductivity is destroyed.

Physicists call the tubes 鈥渧ortices鈥� because in the core of each swirls a
tornado of electrical current. And as the strength of the magnetic field
increases, so do the number and density of the vortices.

Because electrical currents can travel easily through the superconducting
space between vortices, manoeuvring like cross-country skiers through a forest,
magnetic fields don鈥檛 necessarily kill off superconductivity completely. But
even this partial invasion of its territory is enough to render a superconductor
useless for commercial purposes. For the flowing currents鈥� magnetic fields push
on the tornadoes and move them about. The resulting motion dissipates electrical
energy, producing resistance. In the 5-tesla field of a superconducting magnet,
for example, even a good high-temperature superconductor becomes so highly
resistive that it fails to outperform ordinary copper at any temperature above
30 K.

So it is the movement of invading magnetic fingers that causes the trouble.
If they could be held motionless, they wouldn鈥檛 suck energy out of the flowing
current. But doing it isn鈥檛 so easy. To work out how to stick them down,
researchers have been studying their basic tactical habits.

Much like ordinary atoms or molecules, the vortices interact with one another
and form orderly or disorderly arrangements depending on their density and on
the temperature of the material. In magnetic fields of a few tesla, and at
temperatures around 73 K, the vortices act like molecules in a liquid鈥攖hey
move about easily and slip past one another like people in a crowd. At colder
temperatures or in weaker fields, the vortices freeze into a crystalline
lattice, with the tubes arranged in regular rows like the trees in an orange
grove. In this case, the vortices form something very much like a solid.

So whether this weird 鈥渧ortex matter鈥� acts like a solid or liquid depends on
the magnetic field strength and the temperature
(see Diagram).
But in either case, even weak electrical currents make the vortices move,
dissipate energy, and create a material with resistivity. In the liquid, free
streaming vortices follow irregular paths through the material, while those in
the solid move through high-temperature superconductors in regular lock-step
fashion.

Subtle flaws

To prohibit the superconductivity-killing antics of the vortices, materials
scientists are trying to pin them down much as a biologist might an unruly
insect specimen. How do you pin a vortex? The key lies in turning to advantage
subtle flaws in the atomic structure of superconductors themselves.

The ceramic materials that form high-temperature superconductors can, with
great care, be grown as perfect crystals, with all their atoms in precise
positions. More usually, however, a sample of such material will contain some
defects鈥攑laces in the crystal where an atom is dislodged from its proper
position, or where some impurity has found its way in. 鈥淒efects鈥� sound like bad
things, but they have good traits as well鈥攖hey tend to latch on to
vortices.

In a material that is sprinkled liberally with defects, if the temperature
and field strength are low enough, the vortices lock firmly onto the defects and
the vortex liquid solidifies into what physicists call a 鈥渧ortex glass鈥�. The
defects hold each vortex at a specific place in the material. In this state, if
a current flows, the vortices are unable to move, and so, they no longer
dissipate energy.

So whereas the flow of vortices in the liquid or solid states of a
defect-free material undermines superconductivity, the vortex glass in a
defect-ridden material is truly superconducting. This isn鈥檛 a complete solution,
however. For in very strong fields and at high temperatures the vortices still
act like a liquid, and move at will. The 鈥渕elting line鈥� is the set of
temperatures and field strengths where the useful vortex glass melts to form the
useless liquid. To make better and more resilient superconductors, researchers
are trying to push this melting line out towards higher temperatures and
magnetic field strengths.

In 1991, Leonardo Civale of IBM Research Laboratories in Yorktown Heights,
New York, hit on one way to do it. When it comes to pinning down vortices,
defects made of single displaced atoms are good. So Civale reckoned that defects
involving more atoms might be even better. To make them, he bombarded a
superconductor with a stream of ions. As each ion tore through the crystal
lattice, it left behind a trail of wreckage鈥攁 long line of jumbled atoms.
Altogether, the ions created a material riddled with lines of destruction called
鈥渃olumnar defects鈥�, each perfectly configured to trap a vortex. 鈥淭he two are
topologically compatible,鈥� says Lia Krusin-Elbaum, Civale鈥檚 colleague. 鈥淏ecause
vortices are long and thin, just like the columnar defects, the defects are able
to pin them all along their lengths.鈥�

What Civale created was a new and more tenacious vortex glass with a melting
line which lay several tesla and tens of Kelvins beyond that for materials with
ordinary single-atom defects.

But even Civale鈥檚 souped-up vortex glass could not avoid dissipation
altogether. Fluctuating temperatures inside superconductors means that vortices
receive鈥� albeit sporadically鈥攖he energy they need to free
themselves. Imagine a plastic bag being blown through a thicket of trees. It may
hang up on a branch for a time, but eventually it works loose and flies off to
the next tree. Vortices, driven by currents, act the same way. Instead of
remaining immobile, they hop from defect to defect under the influence of
currents in a process called 鈥渧ortex creep鈥�.

Is there a way to stop vortex creep? Physicists now believe that the secret
may lie in turning the vortices against each other. They may act like atoms, but
vortices are able to do what point-like atoms cannot鈥� become tangled up
with one another like strings of spaghetti.

Tangled vortices lead to new kinds of vortex matter with improved electrical
properties. Suppose, for example, that a group of vortices could be tied
together in a knot. If one vortex were to pop off the defect that was holding it
down, it would quickly become ensnared by others, which remain anchored
themselves. So in principle, tangling could reduce vortex creep and the
resistivity that goes with it.

One way to tangle vortices is to crisscross Civale鈥檚 columnar defects in a
process called splay. At IBM, Krusin-Elbaum has modified Civale鈥檚 technique.
Instead of high-energy ions, she uses protons which have been accelerated close
to the speed of light to bombard high-temperature superconductors. When they
slam into the material鈥檚 atoms, the protons can split them apart, producing
atomic fragments that rip through the superconductor at random. The result is
columnar defects, but unlike Civale鈥檚, which are parallel, these defects are
crisscrossed, or splayed like toothpicks in a jumbled pile.

In a magnetic field, the splayed defects still trap vortices, but now along
nonparallel lines. And the misalignment between the defects impedes vortex
creep. With parallel defects, creeping occurs when a vortex sends out 鈥渢ongues鈥�
of magnetic field to explore neighbouring columnar defects. When a tongue
encounters an unoccupied column, the vortex slides over like a train changing
tracks. But crisscrossing the defects prevents vortices from completing the
switch. When the tongues encounter the intersection of two defects, they become
stuck. The defects compete for the vortex, and the competition ends in
stalemate.

David Nelson, a vortex matter theorist from Harvard University in Cambridge,
Massachusetts, compares the phenomenon to a pole-vaulter running through a
forest. When all the trees are upright, the vaulter may speed through the forest
by holding her pole vertically. But imagine what would happen after a storm.
With some of the trees blown over, the pole-vaulter has serious problems
negotiating the woods no matter how he holds the pole. Similarly, vortices are
unable to cross the landscape of tilted columns in a splay glass. As a result,
the material can carry currents, in a superconducting magnet or motor for
example, and the vortices won鈥檛 move. The superconductor remains superconducting
and the device it is a part of works like it should.

Working with collaborators at the American Superconductor Company in
Westborough, Massachusetts and Los Alamos National Laboratory in New Mexico,
Krusin-Elbaum has already managed to reduce the resistivity of mercury-based
superconducting tapes nearly a thousand-fold at 5 tesla and 110 K.

But splitting atoms isn鈥檛 the only way to entangle vortices. Mikhail Indenbom
of the Institute for Solid State Physics in Russia and Gianfranco D鈥橝nna of the
Swiss Federal Institute of Technology in Lausanne use alternating currents to
push and pull vortices, and so tangle them up.

Indenbom and D鈥橝nna induce a circulating sinusoidal current in the
superconducting material. When the current is parallel to the vortices, it makes
the vortices kink. The largest of these kinky vortices form near the surface of
the superconductor. When they grow large enough to bump into their neighbours,
an instability forces thousands of vortices to wind together, forming huge
rotating 鈥渢wisters.鈥� It鈥檚 like the ultimate disaster movie in which thousands of
ordinary tornadoes intertwine to form a supertornado. The twisters then move to
the centre of the superconductor, leaving room for new ones to form at the
edge.

Twisters are produced at the ultra-low temperature of 20 K, but Indenbom and
D鈥橝nna reported last year (Nature, vol 385, p 702) that even at far
higher temperatures the twisters can last for hours鈥攅loquent testimony to
the fact that vortices are difficult to cut and, therefore, to untangle. So far,
however, twisters have only been produced in materials with few defects, and no
one knows whether this technique would work in a material with defects. 鈥淭his is
a tantalising, but not completely proven, way of getting entanglement into the
game,鈥� says Nelson.

Still, with ingenious ideas such as splayed defects and vortex entanglement,
researchers are beginning to get an edge in the game, and to bring the troubling
behaviour of magnetic fields into line. One day soon, the real revolution of
high-temperature superconductors may even begin.