Shortly after Bardeen, Cooper and Schrieffer published their theory of superconductivity in 1957, the physicist William Little proposed that two electrons can attract each other, not via phononic-based attraction but via repulsion from other electrons. He predicted that such electron pairing could be much stronger than in conventional superconductors.

To test this new form of attraction, he imagined a 1D conducting organic chain that contains an array of polarizable side chains. Each polarizer has a single electron that can hop between a site close to the chain and a site further away from it. Thanks to Coulomb repulsion, an electron travelling down the chain polarizes the side chains, which in turn attracts another electron in the chain.

Although researchers have been trying to synthesize such a system in the lab, all attempts at observing excitonic attractions between electrons have failed – until now that is.

A building block of Little’s model

A team led by Shahal Ilani of the Weizmann Institute in Rehovot, Israel, has now constructed a building block of Little’s model, consisting of a two-site system and a polarizer. The researchers show that when the polarizer interacts strongly with the system, the electrons in it start to attract each other.

Ilani and colleagues created their system and polarizer within two separate carbon nanotubes. Both nanotubes were placed on their own microchip and perpendicular to each other, one above the other inside a scanning probe microscope. The researchers fabricated their set up using a technique that they recently developed to suspend each nanotube between two metallic contacts and above an array of gates.

After cooling the apparatus down to around 10 mK, they then moved the nanotubes closer together until they were roughly 100 nm apart. In this way, they could observe how the polarizer affects the electrons in the system, and more importantly, confirm whether or not they attract each other.

Controlling individual electrons in carbon nanotubes

One of the main advantages of the experiment is that the carbon nanotubes employed are electronically pristine and free from impurities. This means that the researchers can control the individual electrons in them using electrical forces.

“The concept behind our experiment, which we report in Nature doi:10.1038/nature18639, is to separate the electrons into two groups,” explains team member Ilanit Shapir. “Some form the system in which they become attracting and some form a surrounding medium that mediates the interaction between the system electrons and forms the ‘glue’ that makes them attract each other.”

The surrounding medium causes an electron in the system to appear as if it has a positive charge, which in turn attracts another electron to it, she tells

Proving that the electrons attract each other

Thanks to electrometers built inside the nanotubes, the researchers are able to determine the interaction between the system electrons, and thanks to these measurements, they are indeed able to prove that the electrons attract each other.

“We show that when we properly set an electron medium near the system, the electrons can only enter the system as a pair – that is, bound together by their attractive interaction, adds Shapir. “We also demonstrate how the attraction between the electrons depends on the surrounding medium and how we can tune this interaction from being repulsive to attractive.”

The team, which includes researchers from the Freie Universitat Berlin in Germany, Harvard University in the US and the Technical University of Denmark, says that it would now be interesting to use the building block made of two electrons that attract each other to create, from the bottom-up, an engineered superconductor with specific properties. “This may allow us to control, for example, the temperature at which the superconducting state can exist in a material, possibly even bringing it up to room temperature in the future.”

Takis Kontos of the Laboratoire Pierre Aigrain at the Ecole Normale Supériere in Paris, France, who was not involved in the work, told that the experiment is “really impressive” and “very elegant”. “Using a unique experimental setup, the authors' system is the closest implementation of Little's old idea yet.”