No abstract

No abstract

The phase diagram of the infinite-range model of spin-glasses exhibits two mixed phases. In these mixed phases, ferromagnetism and spin-glass order coexist, due to freezing of the transverse degrees of freedom or replica symmetry breaking. This may help to interpret a number of recent experimental findings, e.g. , in AuFe.PACS numbers: 75.50.Kj, 05.50.+q Many disordered magnetic materials have been systematically investigated as a function of the concentration of the constituents. In the phase diagrams of these materials, one often finds a spin-glass phase for some range of concentration and a phase with long-range order (ferromagnetic or antiferromagnetic) for other concentrations. ' " Recently, a number of experiments have been performed to study the border region between the spin-glass and ferromagnetic phases, and many interesting data are now available. Unfortunately these data are most often interpreted, qualitatively'"'" and even quantitatively, "'~' with the help of a theory" which, though of great merit in its days, is now obsolete (it has been proven to be incorrect, " at low temperatures, as a solution of its model). Moreover, these theoretical predictions are inadequate because they are concerned with Ising spins, whereas experimental spine are Heisenberg-like (isotropic vector spins). The purpose of this Letter is to draw attention to recent studies, "" which should provide a more satisfactory theoretical basis, and to present new results concerning the nature of two mixed phases, with both spin-glass and ferromagnetic characters. The existence of such intermediate phases may help explain a number of phenomena already observed experimentally, but hitherto ambiguously interpreted.In the spirit of a mean-field theory, we consider the famous infinite-range model" for N classical vector spins S,. Each S; has m components S;" (p. =1, . .. , m) satisfying, for convenience, the following normalization condition: They interact via independent random interactions J;, . distributed according to the following law: P(Z ) = -~e xp --J 2wj 2 (2) so that (J, , ), = J,/N and (J, .~) , =1/N, where ( ),denotes an average over the bond disorder, that is, over P(J,~). In the presence of an external magnetic field H applied along the p, = 1 direction, the Hamiltonian of the model reads where the sum over (ij ) denotes a summation over the N(N -1)/2 distinct pairs of sites.In the limit of large and positive J" the interactions are mainly ferromagnetic and the system exhibits a ferromagnetic phase. For J0=0, the interactions are random in sign and the ordering is of spin-glass type. The border region between ferromagnetic and spin-glass phases occurs for Jo =1. However, it has been shown" how the properties of the model for J, g0 can be simply derived from the case J, = 0, which we therefore first consider.

A probability distribution has been proposed recently by one of us as an order parameter for spin glasses. We show that this probability depends on the particular realization of the couplings even in the thermodynamic limit, and we study its distribution. We also show that the space of states has an ultrametric topology

No abstract

No abstract

2014 Une classification des défauts dans les milieux ordonnés est présentée. Fondée sur des concepts purement topologiques, elle fournit une distinction entre défauts élémentaires (topologiquement stables) et défauts composés. Cette classification systématique contient des résultats dérivés antérieurement de manière empirique et permet de nouvelles prédictions. Elle révèle une liaison entre la nature des défauts dans une phase ordonnée et les phénomènes critiques à la transition de phase. Abstract. 2014 A classification of defects in ordered media is presented. Based on purely topological concepts, it provides a distinction between elementary (topologically stable) and compound defects. This systematic classification recovers some previous empirically derived results and allows new predictions. It exhibits a striking connection between the nature of defects in an ordered phase and the critical phenomena at the phase transition.

The characteristic dissipation process for conventional superfluid flow is phase slippage: motion of quantized vortices in response.to the Magnus force, which allows finite chemical potential differences to occur. Topological considerations and actual construction are used to show that in liquid 3 He-^4, textures with vorticity but no vortex core can easily be constructed, so that dissipation of superfluid flow can occur by motion of textures alone without true vortex lines, dissipation occurring via the Cross viscosity for motions off.Usually dissipative relaxation of the order parameter of a broken-symmetry condensed system occurs by motion of order-parameter singularities. For instance, magnetic hysteresis involves the motion of domain walls, slip of solids that of dislocations, and self-diffusion that of vacancies or interstitials. These are 2-, 1-, and 0-dimensional "order-parameter singularities." All are characterized by a "core" of atomic dimension where the order parameter departs substantially from its equilibrium value.One of the clearest examples of this general rule is phase slippage in superconductors (flux flow and creep) and in liquid helium II: The only way in which these superfluides in bulk form can sustain a gradient of chemical potential, and thus flow dissipatively, is by the continual motion of quantized vortex lines transverse to that gradient. The controlling equation is 1