By Zeeya Merali

Extracting entangled electrons from superconductors could help to create quantum-computing networks. It might even put the theory of quantum mechanics through one of its toughest tests yet.

Hopes for building a working quantum computer hinge on physicists' ability to intertwine electrons into pairs such that changes made to one instantly affect its partner -- a process called entanglement. Photons can be entangled relatively easily; electrons are much harder. This is a big problem if we ever want to integrate quantum computing with electronic chips, says Szabolcs Csonka at the University of Basel in Switzerland.

One place that entangled electrons can be found is in superconductors. Usually they repel each other, but at the low temperatures seen in superconductors electrons pair up and move around as units. "Entanglement is inherent in these pairs," says team member Christian Schönenberger, also at the University of Basel. But until now, physicists have struggled to extract the entangled electron pairs from the superconductor then split them apart, Schönenberger explains.

The team set up an aluminium superconductor and built an escape route for the electrons using nanowires. Ironically, the key to extracting the electrons is to make it almost impossible for them to cross these wires by applying a high voltage that acts as a barrier, says Schönenberger. The laws of quantum mechanics allow pairs of electrons to 'tunnel' through the barrier only very occasionally -- helping the team to isolate and control the pairs as they slowly break out.

Once a pair has reached the other side of the barrier and is out of the superconductor, the electrons' natural repulsion kicks in and the pair splits apart, says Schönenberger. The nanowire path beyond the superconductor is configured into a junction and the team ensures that the electrons take separate directions when they meet this fork by placing a 'gate' built from a nanocrystal, called a quantum dot, at the head of each path. The quantum dots can be tuned by changing the voltage across them, so that they each attract one electron then spit it out to continue its journey.

"The quantum dots act like turnstiles, letting through only one electron at a time," says team member Lukas Hofstetter. The team has succeeded in producing a stream of entangled electron pairs and splitting them in this manner. The research is published in Nature.

Quantum question

Dan Browne, a quantum physicist at University College London says that the technique has important implications for quantum computing. "For many years it didn't look like we would see many advances for quantum information processing in solid-state [electronic] systems," he says. "This is the first time anyone's demonstrated that a stream of entangled electrons can be created on tap."

Brendon Lovett, an expert on quantum computing at the University of Oxford, UK, believes that the work could help physicists to build a quantum-computing network. Lovett and his colleagues have shown that quantum information can be transferred from an electron's spin to the spin of an atomic nucleus, where it can be stored more effectively -- creating a form of 'quantum memory'.

"Picture a network with a number of small processors at each node made of an atomic nucleus interacting with two or three electrons," says Lovett. The team's new method would provide a clever way to communicate information across the network, by sending entangled electrons between nodes, he explains. "This might be one way to scale-up quantum computers."

Entangled electrons might also help to answer a bigger question that is often taken for granted: Does entanglement really exist?

Most physicists now accept that it does and that members of entangled pairs can communicate with each other instantaneously -- a feature that famously made Einstein bristle. But not everybody agrees and some physicists challenge whether our conventional understanding of quantum mechanics is correct.

Technically, entanglement has not been proved, says Browne. "There are still a couple of loopholes that need to be closed," he says.

One of these loopholes involves whether or not entangled entities display the correlations expected; the results are slightly ambiguous in photon tests, says Browne. "You need very high-quality entanglement and very high efficiency in your measuring devices to know with certainty that the effects you are seeing aren't an artefact of the measuring instrument," he says.

Repeating entanglement experiments with a combination of entangled photons and electrons might remove the last shades of ambiguity, says Browne. "We actually don't have highly efficient light detectors, but electron detectors could be better."