Researchers in the US and Japan say they have observed spin superfluidity and very long distance spin transport in an antiferromagnetic insulator made from graphene for the first time. If confirmed, the new result could bring nearly dissipation-less spin-transport devices, which could be used in information processing and storage applications, a step closer to reality.
Spintronics is a technology that makes use of the spin magnetic moment of the electron and it could be used to make devices that are smaller and more energy efficient than conventional electronics. Individual electron spins – which can point up or down – could also be used to store and transfer information in quantum computers.
Practical spintronics devices have proven to be very difficult to make, however. This is because electron spin does not travel very far in most materials, which means that information being carried by the spins is quickly lost. The main culprit here is the “spin-orbit interaction”: as electrons travel through a material, the relative motions of the positively charged atoms create magnetic fields that have the effect of rotating the electron’s spin.
Researchers have recently started to look into transporting spin current in antiferromagnetic insulators (AFMIs). The energy band gap in these materials prohibits such spin-orbit interactions while supporting pure spin current. More importantly still, an effect called “spin superfluidity” has been predicted to exist in them.
“Spin superfluidity is the coherent spin supercurrent that allows for dissipation-less transport, similar to the flow of electrons or Cooper pairs without resistance (as in a superconductor) or of atoms (as in a superfluid),” explain Petr Stepanov, Jeanie Lau and Marc Bockrath of the University of California at Riverside and Ohio State University, who led this research effort. “Much progress has been made here, but the best experimental evidence for such an effect so far has been limited to thermally excited magnons (spin waves) in oxide-based AFMIs. These still suffer from short spin decay lengths of around 0.2 to 10 nm though.”
Enter graphene
“Interestingly, when a strong magnetic field is applied to an undoped (charge-neutral) piece of monolayer graphene it becomes an AFMI containing opposite spin polarizations (that is, spins pointing in opposite directions) on alternate carbon atoms in the material’s hexagonal lattice,” says Lau. “In our work, we used an all-electrical circuit originally proposed by So Takei and colleagues of Queens College, City University of New York, in 2016, to realize the first robust, long-distance spin transport through this AFMI.”
Working with Allan MacDonald’s team at the University of Texas, Austin, Roger Lake’s group at the University of California, Riverside, Dmitry Smirnov at the National High Magnetic Field Lab, and Takashi Taniguchi’s group at NIMS, Japan, Lau and colleagues used graphene in the quantum Hall regime. This occurs when charge carriers like electrons are confined to a 2D plane, as they are in graphene, and subjected to a perpendicular magnetic field in the Z-direction. To make their measurements, the researchers contacted spin injecting and detecting leads to quantum Hall edge states, which lie adjacent to the antiferromagnetic region in the material. They then applied a voltage between these spin-up and spin-down states.
“We measured non-local voltage signals across a 5-micron long AFMI region, a distance that is 104 to 105 times longer than previously measured spin current decay lengths. In control experiments, the signal disappears when the filter regions of the leads are tuned away from these edge states.
“Among the possible transport mechanisms to explain this effect, our data are most consistent with spin superfluidity in the so-called Néel texture of the AFMI that allows for dissipation-less transport of pure spin current,” she tells Physics World.
“This is a new field, and we hope that our experiments are the first of many,” says Lau. “There are still many questions that need to be answered following our results. For example, how efficient is our spin injection technique? And could we observe similar spin transport in bilayer graphene? This would theoretically allow us to control spin directions because electrons residing on the top and bottom layers in this material have opposite spin polarizations. By using an electric field, we could ‘persuade’ the electrons to reside on one of the two layers and so switch their spin directions.”
Full details of the research are reported in Nature Physics 10.1038/s41567-018-0161-5.
