Pick of the Weak Localisation

A Nature Nanotechnology this week with an experiment so easy and simple its crazy to think that it hasn’t been done before (or got into such a high impact journal).  “Direct determination of spin–orbit interaction coefficients and realization of the persistent spin helix symmetry” by Sasaki et al uses a simple technique to directly measure the ratio of Rashba to Dresselhaus SOC (A/B) in zincblende semiconductors using field sweeps and transport measurements alone, as opposed to previous transport and optical measurements which require fitting anylsis that introduces uncertainties.  In zincblende structures, the broken inversion symmetry yields Rashba and Dresselhaus effective magnetic fields that act on the carriers, and when the strengths of the fields are equal a “persisten spin helix” (PSH) state is achieved where there is an invariance with respect to specific rotations of spin and spin-relaxation is surpressed.  The method to measure A/B is based around the varation of weak localisation (WL).  For a wire width << the spin precession length, randomisation of spin is suprssed by 1D confinement, and this fixes the electron momentum direction and thus effective field which means that only WL is observable.  WL manifests itsefls as a positive magnetoconductance, which gives a negative dip in the magnetoconducance as an external field is swept towards zero.


When an external field is applied in the plane of a wire, the degree of spin relaxation depends upon the angle of the total effective field and the external magnetic field; when the fields are not parrallel the uni-directional spin-allignment is broken which induces de-phasing in the time related pairs of electron paths, which leads to quantum interference, which leads to suppression of weak localisation.  On the other hand, when the external and effective fields are parrallel a long spin-relaxation length is preserved, and so the WL (negative magnetoconductance) is enhanced.

 A very simple set up

Experimentally the angle of the combined Rashba and Dresselhaus effective field is measured in a very simple set up.  750nm wide wires of (In,Ga)As are processed.  At various angles, an external field of constant magnitude is applied in the sample plane at a given angle with respect to the crystalline axes, while a second external field is swept perpendicular to the plane, and thus the magneticdue of the magnetoconductance is obtained from the resulting field sweep.  This process is then repeated for various in-plane angles of field.  This yields a polar plot of magnetoconductance amplitude, and the angle of the maximum of the polar plot indicates corresponds to the angle of the combined effective fields.  The ration A/B is then obtained by a simple geometric argument given by the symmetries of the Rashba and Dresselhaus fields for a given current direction.  The authors show that the WL amplitude can be increased by increasing the angle of the in-plane field, and more importantly they gate the sample to tune the value of A, and thus can shift the angle of maximum WL and even achive a PSH state (where the WL amplitude is the same for all in plane magnetic field orientations).




Image: Steve Jurvetson, Some rights reserved