Low-temperature properties of some disordered systems from the statistical properties of nearly degenerate two-level excitations
arXiv:cond-mat/0407289 · doi:10.1140/epjb/e2004-00351-2
Abstract
The thermal fluctuations that exist at very low temperature in disordered systems are often attributed to the existence of some two-level excitations. In this paper, we revisit this question via the explicit studies of the following 1D models (i) a particle in 1D random potentials (ii) the random field Ising chain with continuous disorder distribution. In both cases, we define precisely the `two-level' excitations and their statistical properties, and we show that their contributions to various observables are in full agreement at low temperature with the the rigorous results obtained independently. The statistical properties of these two-level excitations moreover yield simple identities at order in temperature for some generating functions of thermal cumulants. For the random-field Ising chain, in the regime where the Imry-Ma length is large, we obtain that the specific heat is dominated by small non-universal excitations, that depend on the details of the disorder distribution, whereas the magnetic susceptibility and the Edwards-Anderson order parameter are dominated by universal large excitations, whose statistical properties only depend on the variance of the initial disorder via the Imry-Ma length.
19 pages, 4 figures
Cited by in corpus (9)
- Strong disorder RG approach of random systems
- Near-extreme statistics of Brownian motion
- Universal Order and Gap Statistics of Critical Branching Brownian Motion
- Exact Statistics of the Gap and Time Interval Between the First Two Maxima of Random Walks
- Near-extreme eigenvalues and the first gap of Hermitian random matrices
- Statistics of low energy excitations for the directed polymer in a random medium ()
- The KPZ equation in a half space with flat initial condition and the unbinding of a directed polymer from an attractive wall
- Freezing transition of the random bond RNA model: statistical properties of the pairing weights
- Statistics of shocks in a toy model with heavy tails