Magnetic control of valley pseudospin: A story of symmetry

Posted 1 April, 2015


Researchers from the University of Washington in Seattle have discovered an interesting method for information storage and manipulation that circumvents many issues in many existing quantum information storage and manipulation schemes. One of the major hurdles in the field of quantum computation architectures lies in the physical connection between the actual qubit and the outside world. A convenient readout method is needed which can carry the qubit information relatively long distance to another point along the computation pathway. In the system discussed here, this comes with the direct optical initialization, manipulation and readout. All that is needed is a very thin sheet of a particular material tungsten-diselenide (WSe2), which lacks inversion symmetry.


Figure 1: In a) the splitting induced by the various contributions from the magnetic moments. (black=spin,green=valley,purple=atomic orbital), with the atomic orbital splitting allowing a measureable difference. b) addressing the separate bands with polarization resolved PL (some results shown in next figure)

Due to the inherent construction of the material used on the atomic level, there exists an asymmetry under inversion of the coordinates, otherwise called inversion symmetry breaking. This inversion symmetry breaking allows a finite orbital contribution to the magnetic moments in the energy degenerate band edges (+/-K) with equal magnitude but opposite sign, which is due to time reversal symmetry. This magnetic moment discussed, which is related to the valley magnetic moment through the Berry curvature, allows for a valley dependent optical selection rule which can be addressed directly and exclusively by left/right circularly polarized light. The magnitude of the effect can be tuned by the external magnetic field B through a direct coupling between the field and the optical magnetic moment. This Zeeman shift in energy is due to the adjustment in the valence band states through m=2 multiplicity in the d-orbitals.

Polarization resolved PL

Figure 2: The population polarization is seen in contrast between the different polarization resolved PL.

The researchers use a method termed polarization-resolved photoluminescence to resolve a magnetic field tuned splitting between the +K and -K valley excitonic resonances. They both excite and detect with a single helicity of light, with the final polarization comparison done between the separate peak positions under the different polarization conditions.

All this is carried out in a Montana Instruments Cryostation at 30K modified with a castle shaped cryostat in which can fit within the bore of a magnet.


Magnetic control of magnetic pseudospin in monolayer WSe2 , Nature Physics 11, 148-152 (2015)

This work should not be considered an endorsement of any product.