Quantum interference effects and spin-orbit interaction in quasi-one-dimensional wires and rings

Abstract

We study two kinds of quantum interference effects in transport -the Aharonov-Bohm effect and the weak-localization effect -in quasi-one-dimensional wires and rings to address issues concerning the phase-coherence length, spin-orbit scattering, and the Aux cancellation mechanism which is predicted to be present when the elastic mean free path exceeds the sample width. Our devices are fabricated on GaAs/Al"Gal "As and pseudomorphic Ga"Inl As/Al" In, As heterostructure materials and the experiments carried out at 0.4 -20 K temperatures. In the GaAs/Al"Ga~, As devices which exhibit no significant spin-orbit scattering, we were able to extract a phase-coherence length I& from the amplitude of the Aharonov-Bohm magnetoresistance oscillations in different sized rings. We find it to be in agreement with l& deduced from the weak-localization data in parallel wires when the one-dimensional weaklocalization theory including thegux cancellation mechttnism is used to fit the data. We therefore unambiguously establish that the same l& governs the behavior of the two quantum interference phenomena of Aharonov-Bohm oscillations and weak localization, and that the Aux cancellation is in force. In the pseudomorphic Ga"In, "As/Al"In, "As heterostructure devices which exhibit strong spin-orbit interaction effects, l& exceeds the spin-orbit-scattering length at low temperatures.The amplitude of Aharonov-Bohm oscillations can only be explained by introducing reduction factors due to spin-orbit scattering.

Magnetoconductance correction in zinc-blende semiconductor nanowires with spin-orbit coupling

“…In Appendix D, we show that the application of the 1D magnetoconductance formula of Ref. [37] to our nanowire device likewise leads to an unusual trend for l φ . This behavior is not seen in our device with our magnetoconductance model, where the dephasing length remains nearly constant.…”

“…In these works, the magnetoconductance data are analyzed by means of a 1D magnetoconductance formula [37], which does not consider the mesoscopic details of the system. In Appendix D, we use this formula to fit the magnetoconductance data of our device.…”

“…We find good agreement between theory and experiment and reasonable transport parameters. For comparison, we also apply the 1D magnetoconductance formula of Kurdak et al [37], which has been frequently used by other authors [9][10][11]13,[15][16][17][18]. The fitted curves show larger deviations from the experimental observations and an unusual trend of the dephasing length.…”

“…Thereby, we extract spin relaxation and dephasing rates as well as SOC strengths. To complete our study, we compare our results to the frequently used one-dimensional (1D) magnetoconductance formula of Kurdak et al [37] and discuss the discrepancies in the resulting fits and the gained transport parameters [9-11,13,15-18].…”

Direct thermal conductance measurements on suspended monocrystalline nanostructures

“…2͒. [9][10][11] The meandering pattern of the nϩ GaAs conductors posits most of the line length above the suspended reservoir. From the transducers, nϩ leads on top of each bridge provide electrical contact to wirebond pads.…”