Density-Dependent Spin Polarization in Ultra-Low-Disorder Quantum Wires

Abstract

There is controversy as to whether a one-dimensional (1D) electron gas can spin polarize in the absence of a magnetic field. Together with a simple model, we present conductance measurements on ultra-low-disorder quantum wires supportive of a spin polarization at B=0. A spin energy gap is indicated by the presence of a feature in the range (0.5-0.7)x2e(2)/h in conductance data. Importantly, it appears that the spin gap is not constant but a function of the electron density. Data obtained using a bias spectroscopy technique are consistent with the spin gap widening further as the Fermi level is increased.

Anomalous spin-dependent behavior of one-dimensional subbands

“…Our observation that spin-down, but not spinup levels, drop rapidly in energy as they populate, implies that the spin gap must open almost instantaneously to as much as 0.5 meV, as the first subband is populated. This is not in agreement with the model in reference [10] where it is assumed that a spin gap opens gradually and linearly with increasing gate-voltage. Our results also give insight into the temperature (T ) dependence of the riser up to the 0.7 Structure.…”

“…Previous work [2,9,10,11,12] provides strong evidence that the 0.7 Structure is related to the opening of an energy gap between spin-up and spin-down energy levels as they populate. Our observation that spin-down, but not spinup levels, drop rapidly in energy as they populate, implies that the spin gap must open almost instantaneously to as much as 0.5 meV, as the first subband is populated.…”

Conductance quantization and zero bias peak in a gated quantum wire

“…Studies in ultraclean 0.5 and 1 mm quantum wires showed an evolution of the 0.7 feature toward 0.5 with increasing 1D carrier density, but no differential conductance peak [8]. While the 0.7 ''knee'' is a robust feature for a variety of structures, the amplitude of the zero bias peak may be highly device specific.…”

“…Theories invoking both a density-dependent spin energy gap [8] and a dynamical, Kondo-like mechanism [7,13] have been proposed. Evidence for the spin gap comes from conductance measurements on long, very high mobility wires in which the 1D carrier density (n 1D ) could be varied continuously [8]. There, it was found that increasing n 1D caused the 0.7 feature to migrate toward 0.5 quanta, indicative of a spin-split plateau.…”

“…In AlGaAs/GaAs quantum point contacts (QPC) conductance plateaus occur at integral values of 2e 2 /h corresponding to an integral number of electric subbands carrying the transport current [1,2]. While integral conductance quantization can be understood from a single-electron picture [3][4][5], the widely observed 0.7 conductance feature observed in QPCs appears to involve manyelectron physics [6][7][8][9][10][11][12][13]. In this paper, we report conductance measurements on a 3 mm long quantum wire with seven cross channel gates.…”