For decades, physicists have debated the origins of high-temperature superconductivity—the ability of some materials to carry electricity without resistance at temperatures up to 138 kelvin. Now, new data nix one possible explanation, albeit a less popular one, a team claims. If the finding holds up, it would sever any connection between ordinary superconductors, such as lead and niobium, and the high-temperature materials.
Tiny vibrations called phonons—the essential ingredient in the mechanism behind ordinary superconductors-play no significant role in high-temperature superconductivity, the team claims. "Hopefully, the paper will put an end to this story," says Andrey Chubukov, a theorist at the University of Wisconsin, Madison, who was not involved in the work. Advocates of phonon-based theories of high-temperature superconductivity say they are not persuaded, however.
In an ordinary metal, current meets with resistance as the electrons in it deflect off the vibrating ions in the metal's "crystal lattice." But in a superconductor cooled below its "critical temperature," the electrons form loose "Cooper pairs." Deflecting an electron then requires breaking a pair. At low temperatures, there isn't enough energy around to do that, so the pairs glide through unimpeded. As a result, electric current flows freely through the material—like a milkshake miraculously zipping through a straw with no one sucking on it.
What holds the Cooper pairs together? In an ordinary superconductor, phonons provide the glue. A passing negatively charged electron draws the positively charged ions in the material slightly closer together in its wake, and that concentration of positive charge then pulls along the second electron. The subtle ripples in the crystal lattice are the phonons. But theorists say phonons do not pull hard enough to keep electrons paired at the sky-high temperatures—which are still far below the freezing point of water—achieved in high-temperature superconductors. Instead, many think the glue originates in interactions among the electrons themselves, such as waves of magnetism called spin fluctuations.
Still, some physicists argue that phonons must still play a key indirect role in high-temperature superconductors. And experiments have shown that the phonons and electrons do interact in the compounds.
Now, Claudio Giannetti of the Catholic University of the Sacred Heart in Brescia, Italy, and a dozen colleagues report data that, they say, show that electrons alone tell the whole story. They shined pulses of laser light onto a high-temperature superconductor called bismuth strontium calcium yttrium copper oxide (BSCCO) to study how the material reflects light at various wavelengths, they report today in Science.
The team hit the sample with a one-two punch of laser pulses roughly 100 millionths of a nanosecond, or 100 femtoseconds, long. The first pulse stirred up the electrons in the material; the second pulse measured how much the material's reflectivity had changed. The team was also able to trace the reaction not only in time, but also as a function of the frequency of the reflected light.
The researchers measured the part-in-10,000 changes over a few thousand femtoseconds and then plugged the numbers into a computer model to gauge which processes were most important in carrying energy through the lattice. Electron-electron interactions such as spin fluctuation should carry energy away much faster than phonons do, the researchers argued, making it possible to separate the different contributions.
The ability to study the reflectivity at different wavelengths was key, Giannetti says. That's because the ultrafast electron-electron processes were too fast to observe in the time traces. However, those processes affect the reflectivity at different wavelengths in different ways—100 femtoseconds after the pulse the material was less reflective at longer wavelengths and more reflective at shorter wavelengths. Taken all together, the data show that phonons aren't needed to explain BSCCO's superconductivity, Giannetti says. Electron-electron interactions are strong enough to do the job all by themselves.
Not everyone is convinced. The model the researchers used to fit the data may have been based on unrealistic assumptions, says Christoph Gadermaier, an experimenter at the Jozef Stefan Institute in Ljubljana, Slovenia. For example, he says, it assumed that the electrons, phonons, and electron-electron interactions in BSCCO behaved as if it were in steady "thermal equilibrium." That's a poor assumption that can alter the inferred interaction strengths, he says. And Giannetti's own data suggest that the electron-phonon coupling is still rather strong, Gadermaier says. "For me, phonons definitely play a role," he says.
Alexandre Alexandrov, a theorist at Loughborough University in the United Kingdom and outspoken advocate of phonons, agrees. In his opinion, electrons repel one another too strongly to form Cooper pairs without outside help as Giannetti and his colleagues propose. "They are speaking about something that is impossible," he says.
So is this the end of the debate over phonons in high-temperature superconductors? Probably only for those researchers who already thought it was done.