The eigenvalues and eigenvectors of the inverse of the susceptibility matrix are discussed for random spin systems. At a phase transition precipitated by the vanishing of the smallest eigenvalue it is shown that the corresponding eigenvector must be extended, and that the density of states must vanish at zero eigenvalue. Above the transition temperature, generalised Griffiths singularities are associated with the existence of localised states with arbitrarily small eigenvalues. For infinite spin dimensionality the eigenvalue problem is equivalent to an Anderson problem with correlated diagonal and off-diagonal disorder.
The eigenvalues and eigenvectors of the inverse of the susceptibility matrix are discussed for random spin systems. At a phase transition precipitated by the vanishing of the smallest eigenvalue it is shown that the corresponding eigenvector must be extended, and that the density of states must vanish at zero eigenvalue. Above the transition temperature, generalised Griffiths singularities are associated with the existence of localised states with arbitrarily small eigenvalues. For infinite spin dimensionality the eigenvalue problem is equivalent to an Anderson problem with correlated diagonal and off-diagonal disorder.
The susceptibility in arbitrary directions, the specific heat, the energy and the magnetization of general Ising-type chains are exactly expressed in terms of the eigenvectors of the transfer matrix corresponding to the Ising-type interaction. These quantities, as a result, can be explicitly obtained for small spin values and generally obtained from the solution of corresponding eigenvalue problem of finite degree. Numerical estimations are easy for arbitrary spin values. This formula includes the spin- S transverse Ising model with vanishing or non-vanishing parallel external field, the Blume-Emery-Griffiths model and the Blume-Capel model, mixed spin and mixed bond periodic systems such as alternating Ising chains or Ising ferrimagnets and also includes Ising models with random structures. Low temperature behaviors of these systems and the crossover to infinite spin systems are also investigated.
Arguments are given that, for random spin systems, the density of states \ensuremath{\rho}(\ensuremath{\mu}) of the inverse of the susceptibility matrix vanishes as \ensuremath{\rho}(\ensuremath{\mu})\ensuremath{\sim}exp(-A/\ensuremath{\mu}), for \ensuremath{\mu}\ensuremath{\rightarrow}0, throughout the ``Griffiths phase.'' The amplitude A vanishes at the onset of magnetic long-range order, and diverges at the transition between ``Griffiths'' and ``paramagnetic'' phases. For an O(m) spin system, with m\ensuremath{\rightarrow}\ensuremath{\infty}, the spin autocorrelation function C(t) is found to have the ``stretched-exponential'' form, lnC(t)\ensuremath{\sim}-(At)1/2, in the Griffiths phase.
We employ singular value decomposition (SVD) to study the eigenvalue spectra of random spin systems. By SVD, eigenvalue spectrum is decomposed into orthonormal modes Wk with weight ${\ensuremath{\lambda}}_{k}$. We show that the scree plot (${\ensuremath{\lambda}}_{k}$ with respect to k) in the ergodic phase contains two branches that both follow power law ${\ensuremath{\lambda}}_{k}\ensuremath{\sim}{k}^{\ensuremath{-}\ensuremath{\alpha}}$ but with different exponents $\ensuremath{\alpha}$. By evaluating Wk, we verified the part of ${\ensuremath{\lambda}}_{k}$ with k>kTh is universal that follows random matrix theory, where kTh is related to the Thouless energy. We further demonstrate that $\ensuremath{\alpha}$ corresponds only to the exponential part of the level spacing distribution while being insensitive to the level repulsion, or equivalently the system's symmetry. Consequently, $\ensuremath{\alpha}$ gives an underestimation for the many-body localization transition point, which suggests a nonergodic behavior that may be attributed to the Griffiths regime.
One of the points of interest in the study of random systems is the nature of the eigenstates, i.e., exponentially localized or extended, and the density of states. To identify the disorder-induced effects, we consider the one-dimensional s=1/2 xy models with random exchange, or random field. Using the transformation to spinless fermions, the problem is reduced to a nonlinear map, providing accurate numerical estimates for the integrated density of states and the exponential localization length. The results clearly reveal: (i) appearance of disorder-induced tails in the integrated density of states at the bottom and top of the spectrum, yielding to a characteristic field dependence of magnetization and susceptibility at zero temperature; and (ii) important differences between random-exchange and random-field models. In the random-exchange case, the middle of the band is found to be extended, while in the random-field case, corresponding to the standard Anderson model, all states are exponentially localized. This difference also affects the temperature dependence of zero-field specific heat and susceptibility as T→0.
Abstract We study disordered systems at and near quantum phase transitions, which are transitions at zero temperature that are driven by quantum instead of thermal fluctuations. Special emphasis is put on the so‐called Griffiths‐McCoy regions surrounding these transitions in magnetic systems, where various susceptibilities are found to be divergent for T → 0 without the system being critical. We present a phenomenological picture for uncorrelated and weakly or strongly correlated disorder and check it for various models. In higher dimensions the phenomena seem to be connected to a percolation transition.
In many-body systems with quenched disorder, dynamical observables can be singular not only at the critical point, but in an extended region of the paramagnetic phase as well. These Griffiths singularities are due to rare regions, which are locally in the ordered phase and contribute to a large susceptibility. Here, we study the geometrical properties of rare regions in the transverse Ising model with dilution or with random couplings and transverse fields. In diluted models, the rare regions are percolation clusters, while in random models the ground state consists of a set of spin clusters, which are calculated by the strong disorder renormalization method. We consider the so called energy cluster, which has the smallest excitation energy and calculate its mass and linear extension in one-, two- and three-dimensions. Both average quantities are found to grow logarithmically with the linear size of the sample. Consequently, the energy clusters are not compact: for the diluted model they are isotropic and tree-like, while for the random model they are quasi-one-dimensional.