arXiv:cond-mat/9408064 · Physical review. B, Condensed matter
Tunneling and quantum noise in one-dimensional Luttinger liquids
Claudio Chamon, Denise E. Freed, Xiao-Gang Wen
Abstract
We study nonequilibrium noise in the transmission current through barriers in one-dimensional Luttinger liquids and in the tunneling current between edges of fractional quantum Hall liquids. The distribution of tunneling events through narrow barriers can be described by a Coulomb gas lying in the time axis along a Keldysh (or nonequilibrium) contour. We show that the charges tend to reorganize as a dipole gas,which we use to describe the tunneling statistics. The dipole-gas picture allows us to have a unified description of the low-frequency shot noise and the high-frequency Josephson noise. The correlation between the charges within a dipole (intradipole) contributes to the high-frequency Josephson noise, which has an algebraic singularity at \ensuremath{\omega}=${\mathit{e}}^{\mathrm{*}}$V/\ensuremath{\Elzxh}, whereas the correlations between dipoles (interdipole) are responsible for the low-frequency noise. We show that an independent or noninteracting dipole approximation gives a Poisson distribution for the locations of the dipole centers of mass, which gives a flat noise spectrum at low frequencies and corresponds to uncorrelated shot noise. Including interdipole interactions gives an additional 1/${\mathit{t}}^{2}$ correlation between the tunneling events that results in an \ensuremath{\Vert}\ensuremath{\omega}\ensuremath{\Vert} singularity in the noise spectrum. We present a diagrammatic technique to calculate the correlations in perturbation theory, and show that contributions from terms of order higher than the dipole-dipole interaction should only affect the strength of the \ensuremath{\Vert}\ensuremath{\omega}\ensuremath{\Vert} singularity, but its form should remain \ensuremath{\sim}\ensuremath{\Vert}\ensuremath{\omega}\ensuremath{\Vert} to all orders in perturbation theory. A counting argument also suggests that the leading algebraic singularity at ${\mathrm{\ensuremath{\omega}}}_{\mathit{J}}$ should be \ensuremath{\propto}\ensuremath{\Vert}\ensuremath{\omega}-${\mathrm{\ensuremath{\omega}}}_{\mathit{J}}$${\mathrm{\ensuremath{\Vert}}}^{2\mathit{g}\mathrm{\ensuremath{-}}1}$ to all orders in perturbation theory.