Zero-frequency current noise for the double-tunnel-junction Coulomb blockade

Abstract

We compute the zero frequency current noise numerically and in several limits analytically for the coulomb blockade problem consisting of two tunnel junctions connected in series. At low temperatures over a wide range of voltages, capacitances, and resistances it is shown that the noise measures the variance in the number of electrons in the region between the two tunnel junctions. The average current, on the other hand, only measures the mean number of electrons. Thus, the noise provides additional information about transport in these devices which is not available from measuring the current alone. 1992 PACS 72.70.+m , 73.40.Gk

Shot noise in mesoscopic conductors

“…36, right). They compare their experimental data with the theory of Hershfield et al [283] and find quantitative agreement.…”

“…Right: I-V curve (top) and shot noise (bottom) in a very asymmetric quantum dot. Solid curves show the theory of Hershfield et al [283]. Dashed lines in all cases show the Poisson value 2e I and the value e I .…”

“…The situation is, however, different close to the degenerate points (at the center of the step of the staircase), since there are now two charge states available. Hershfield et al [283] find that the Fano factor in the vicinity of the degenerate points is…”

“…35b, upper curve. Hershfield et al [283] also perform extensive numerical simulations of shot noise and show that the structure in the Fano factor disappears in the symmetric limit R 1 ∼ R 2 . Similar results were subsequently obtained by Galperin et al [285], Hanke et al [286], and Hanke, Galperin, and Chao [287], who also investigate the frequency dependence of shot noise.…”

“…For the general case (Coulomb staircase regime), shot noise was (independently) analyzed by Hershfield et al [283] using a master equation. In the plateau regimes, the transport is via the only state with a fixed number of electrons, and the shot noise is Poissonian (up to exponentially small corrections); in particular, for V < e/2C noise is exponentially small.…”