Like atomic-level bricklayers, I. Božović from Brookhaven National Laboratory, G. Dubuis from Ecole Polytechnique Federale de Lausanne, and their colleagues are using a precise atom-by-atom layering technique to fabricate an ultrathin transistor-like field effect device to study the conditions that turn insulating materials into high-temperature superconductors. The technical break-through, which is described in the April 28th issue of Nature (DOI: 10.1038/nature09998; p. 458), could lead to advances in understanding high-temperature superconductivity.
“Understanding exactly what happens when a normally insulating copper-oxide material transitions from the insulating to the superconducting state is one of the great mysteries of modern physics,” said Božović, lead author on the study.
One way to explore the transition is to apply an external electric field to increase or decrease the level of doping and see how this affects the ability of the material to carry current. But to do this in copper-oxide (cuprate) superconductors, one needs extremely thin films of perfectly uniform composition—and electric fields measuring more than 109 V m–1.
Božović’s group has employed molecular beam epitaxy (MBE) to uniquely create such perfect superconducting thin films one atomic layer at a time, with precise control of each layer’s thickness. Recently, they have shown that in such MBE-created films, even a single cuprate layer can exhibit undiminished high-temperature superconductivity.
Now, the researchers have applied the same technique to build ultrathin superconducting field effect devices that allow them to achieve charge separation, and thus electric field strength, for these critical studies.
These devices are similar to the field-effect transistors (FETs) that are the basis of modern electronics, in which a semiconducting material transports electrical current from the source to a drain electrode. FETs are controlled by a gate, positioned above the source–drain channel—separated by a thin insulator—which switches the device on or off when a particular gate voltage is applied to it.
But because no known insulator can withstand the high fields required to induce superconductivity in the cuprates, the standard FET scheme does not work for high-temperature superconductor FETs. Instead, the researchers used electrolytes to separate the charges.
In this setup, when an external voltage is applied, the electrolyte’s positively charged ions travel to the negative electrode and the negatively charged ions travel to the positive electrode. But when the ions reach the electrodes, they abruptly stop, as though they have hit a brick wall. The electrode “walls” carry an equal amount of opposite charge, and the electric field between these two oppositely charged layers can exceed the 109 V m−1 goal.
The result is a field effect device in which the critical temperature of a prototype high-temperature superconductor compound (lanthanum-strontium-copper-oxide) can be tuned by as much as 30 K, which is about 80% percent of its maximal value—almost 10 times more than the previous record.
The researchers have now used this enhanced device to study some of the basic physics of high-temperature superconductivity.
One key finding is that as the density of mobile charge carriers is increased, their cuprate film transitions from insulating to superconducting behavior when the film sheet resistance reaches 6.45 kΩ. This is exactly equal to the Planck quantum constant divided by twice the electron charge squared. Both the Planck constant and electron charge are atomic units—the minimum possible quantum of action and of electric charge, respectively, established after the advent of quantum mechanics early in the last century.
“It is striking to see a signature of clearly quantum-mechanical behavior in a macroscopic sample (up to millimeter scale) and at a relatively high temperature,” Božović said. Most people associate quantum mechanics with characteristic behavior of atoms and molecules.
This result also carries another surprising message. While it has been known for many years that electrons are paired in the superconducting state, the findings imply that they also form pairs (although localized and immobile) in the insulating state, unlike in any other known material. That sets the researchers on a more focused search for what gets these immobilized pairs moving when the transition to superconductivity occurs.
Superconducting FETs might also have direct practical applications. Semiconductor-based FETs are power-hungry, particularly when packed very densely to increase their speed. In contrast, superconductors operate with no resistance or energy loss. Here, the atomically thin layer construction is in fact advantageous—it enhances the ability to control superconductivity using an external electric field.
