Subscribe to the gold package and get unlimited access to Shamra Academy
Register a new userGeneralized Lyapunov exponent of random matrices and universality classes for SPS in 1D Anderson localisation
141
0
0.0
(
0
)
Authors
Christophe Texier
Ask ChatGPT about the research
No Arabic abstract
Products of random matrix products of $mathrm{SL}(2,mathbb{R})$, corresponding to transfer matrices for the one-dimensional Schrodinger equation with a random potential $V$, are studied. I consider both the case where the potential has a finite second moment $langle V^2rangle<infty$ and the case where its distribution presents a power law tail $p(V)sim|V|^{-1-alpha}$ for $0<alpha<2$. I study the generalized Lyapunov exponent of the random matrix product (i.e. the cumulant generating function of the logarithm of the wave function). In the high energy/weak disorder limit, it is shown to be given by a universal formula controlled by a unique scale (single parameter scaling). For $langle V^2rangle<infty$, one recovers Gaussian fluctuations with the variance equal to the mean value: $gamma_2simeqgamma_1$. For $langle V^2rangle=infty$, one finds $gamma_2simeq(2/alpha),gamma_1$ and non Gaussian large deviations, related to the universal limiting behaviour of the conductance distribution $W(g)sim g^{-1+alpha/2}$ for $gto0$.
rate research
Read More
In a recent publication, J. Phys.: Condens. Matt. 14 13777 (2002), Kuzovkov et. al. announced an analytical solution of the two-dimensional Anderson localisation problem via the calculation of a generalised Lyapunov exponent using signal theory. Surprisingly, for certain energies and small disorder strength they observed delocalised states. We study the transmission properties of the same model using well-known transfer matrix methods. Our results disagree with the findings obtained using signal theory. We point to the possible origin of this discrepancy and comment on the general strategy to use a generalised Lyapunov exponent for studying Anderson localisation.
I present a general framework allowing to carry out explicit calculation of the moment generating function of random matrix products $Pi_n=M_nM_{n-1}cdots M_1$, where $M_i$s are i.i.d.. Following Tutubalin [Theor. Probab. Appl. {bf 10}, 15 (1965)], the calculation of the generating function is reduced to finding the largest eigenvalue of a certain transfer operator associated with a family of representations of the group. The formalism is illustrated by considering products of random matrices from the group $mathrm{SL}(2,mathbb{R})$ where explicit calculations are possible. For concreteness, I study in detail transfer matrix products for the one-dimensional Schrodinger equation where the random potential is a Levy noise (derivative of a Levy process). In this case, I obtain a general formula for the variance of $ln||Pi_n||$ and for the variance of $ln|psi(x)|$, where $psi(x)$ is the wavefunction, in terms of a single integral involving the Fourier transform of the invariant density of the matrix product. Finally I discuss the continuum limit of random matrix products (matrices close to the identity ). In particular, I investigate a simple case where the spectral problem providing the generalized Lyapunov exponent can be solved exactly.
We report improved numerical estimates of the critical exponent of the Anderson transition in Andersons model of localization in $d=4$ and $d=5$ dimensions. We also report a new Borel-Pade analysis of existing $epsilon$ expansion results that incorporates the asymptotic behaviour for $dto infty$ and gives better agreement with available numerical results.
We describe a Borel-Pade re-summation of the $beta$-function in the three Wigner-Dyson symmetry classes. Using this approximate $beta$-function we discuss the dimensional dependence of the critical exponent and compare with numerical estimates. We also estimate the lower critical dimension of the symplectic symmetry class.
We report a numerical investigation of the fluctuations of the Lyapunov exponent of a two dimensional non-interacting disordered system. While the ratio of the mean to the variance of the Lyapunov exponent is not constant, as it is in one dimension, its variation is consistent with the single parameter scaling hypothesis.
Log in to be able to interact and post comments
comments
Fetching comments
Sign in to be able to follow your search criteria
