Ballistic electrons move through very thin wires much like a bullet through the barrel of a gun – they are constrained to move in one direction and encounter little or no resistance along the way. If the wire is only a few atoms thick, its ability to conduct electrons – its conductance – becomes quantized as an integer (N) multiple of the conductance of a single electron. This is because the energies of the electrons in the wire are constrained to narrow “bands” and N corresponds to the number of bands that cross the Fermi energy level, where conduction occurs.

In 2005, Evgeny Tsymbal and colleagues at the University of Nebraska predicted that the number N could be changed by applying a magnetic field to a very thin wire made out of a metallic magnet. In such materials the conduction electrons are magnetic and an applied field should shift the position of the energy bands relative to the Fermi energy – thereby changing N. Since the conductance of the wire is proportional to N, the researchers predicted that a stepwise change in conductance (and also resistance) should be seen. They dubbed this effect “ballistic anisotropic magnetoresistance” (BAMR) – “anisotropic” because the effect is dependent upon the relative orientation of the magnetic field and the direction of conduction.

Now, Bernard Doudin at the Institute of the Physics and Chemistry of Materials in Strasbourg and colleagues at the University of Nebraska have seen BAMR in a number of different atomic-scale wires made from the magnetic metal cobalt. In one sample, the researchers measured a change in conductance consistent with N going from 6 to 7 as the field direction was changed. N was also observed to jump by twos and fours in other samples. The researchers say that this range of responses is related to atomic-scale differences in the structures of the nanowire samples and can be explained by Tsymbal’s BAMR theory.

BAMR could someday be exploited to create extremely small heads for reading data stored on magnetic disks and other media. In theory, this could push magnetic storage densities to the atomic limit. However, Doudin warns that the effect’s sensitivity to structural differences would mean that devices would have to be fabricated with atomic-scale accuracy – something that cannot be done today.