Dependence of the Dresselhaus spin-orbit interaction on the quantum well width

Abstract

We measured the Dresselhaus spin-orbit interaction coefficient β 1 for (001)-grown GaAs/Al 0.3 Ga 0.7 As quantum wells for six different well widths w between 6 and 30 nm. The varying size quantization of the electron wave vector z-component k 2 z ∼ (π/w) 2 influences β 1 = −γ k 2 z linearly. The value of the bulk Dresselhaus coefficient γ = (−11 ± 2) eVÅ 3 was determined. We discuss the absolute sign of the Landé g factors and the effective momentum scattering times.

All-optical evaluation of spin-orbit interaction based on diffusive spin motion in a two-dimensional electron gas

“…Values for D s can be determined from the spatially resolved expansion of the spin polarization. Using D s ¼ 0.0078 m 2 /s, 18 we obtain b ¼ 1.19 Â 10 À13 and a ¼ À1.6 Â 10 À14 eVm. The signs of b and a are determined from the direction of B R and B D with respect to B ex .…”

“…As there is no contribution of v d at x ¼ 0 lm, the g factor is obtained directly from X, jgj ¼ 0.34, which is consistent with the previous results. 17,18 To analyze this spatial modulation of X, the X SOI evaluated is plotted as a function of x and y in Fig. 3(b).…”

“…Each sample contains 15 equivalent QWs, and the carrier densities are 1.2, 1.3, and 1.7 Â 10 15 m À2 for the 10-, 20-, and 30-nm QWs, respectively. 17,18 The Dresselhaus SO coefficient, Rashba SO coefficient, a, is controlled by the Si d-doping profile in the AlGaAs barrier layer and its absolute value is chosen to be smaller than b. In the TRKR measurement, a mode-locked Ti:sapphire laser generates 2-ps-long optical pulses with 79.2 MHz repetition rate.…”

Driving Spin and Charge in Quantum Wells by Surface Acoustic Waves

“…The carrier density in the surfing regime, ρ 0 (r, ϕ, t), was already determined in Sec. IV, with X(x, t) given in (19); according to Eq. (15) the carrier density along the y axis for an initial Gaussian distribution with standard deviation y 0 reads…”

“…For H so we consider general linear-in-momentum couplings, which arise in 2DEGs because of broken structural (Rashba 15 ) or bulk (Dresselhaus 16 ) inversion symmetry, or of strain; 17 linear couplings are dominant with respect to cubic ones in a wide range of parameters. 18,19 Linear-in-momentum couplings can be written in terms of a non-Abelian vector potential A, 20-23 which for spin 1/2 carriers becomes a SU (2) field with three components in the Pauli matrices basis (a = x, y, z), and two components in real space (i = x, y):…”