Universal Conductance Distributions in the Crossover between Diffusive and Localization Reg.pdf

a r X i v : c o n d - m a t / 0 1 0 7 3 9 5 v 1 [ c o n d - m a t .d i s - n n ] 1 9 J u l 2 0 0 1 Universal Conductance Distributions in the Crossover between Diffusive and Localization Regimes A. Garc??a-Mart??n and J.J. Sa?enz Departamento de F??sica de la Materia Condensada and Instituto de Ciencia de Materiales “Nicola?s Cabrera”, Universidad Auto?noma de Madrid, E-28049 Madrid, Spain. (February 1, 2008) The full distribution of the conductance P (G) in quasi-one-dimensional wires with rough sur- faces is analyzed from the diffusive to the localization regime. In the crossover region, where the statistics is dominated by only one or two eigenchannels, the numerically obtained P (G) is found to be independent of the details of the system with the average conductance 〈G〉 as the only scaling parameter. For 〈G〉 e2/h, P (G) is given by an essentially “one-sided” log-normal distribution. In contrast, for e2/h 〈G〉 ≤ 2e2/h, the shape of P (G) remarkable agrees with those predicted by random matrix theory for two fluctuating transmission eigenchannels. PACS numbers: 72.15.Rn, 72.10.Fk, 42.25.Dd, 71.30.+h The large conductance fluctuations occurring in meso- scopic samples have attracted large attention during the last decade [1,2]. The magnitude of the fluctuations (～ e2/h) is almost independent of the mean value of the conductance 〈G〉 and the system size (‘universal conduc- tance fluctuations’) [3]. This holds for wires or waveg- uides in the diffusive regime where the length of the wire L is much larger than the mean free path ? but still much smaller than the localization length ξ. As the length of wire L approaches the localization length ξ, the conductance fluctuations become of the same order as the averaged conductance. The averaged values are then not enough to describe the transport properties and the knowledge of the conductance probability distribution is of primary interest. While for a single-mode wire, where ξ ≈ ?, the proba- bility distribution