Theory of site-disordered magnets

TL;DR

This paper develops a field theory for site-disordered magnetic systems like spin glasses, revealing phase transitions, percolation behavior, and novel spin glass phases near critical points.

Contribution

It formulates a comprehensive field theory for realistic disordered magnets, connecting various models and identifying new phases near critical points.

Findings

Identifies a percolation transition in the model.

Shows T_g scales with concentration c for RKKY couplings.

Discovers new spin glass phases near special critical points.

Abstract

In realistic spinglasses, such as CuMn, AuFe and EuSrS, magnetic atoms are located at random positions. Their couplings are determined by their relative positions. For such systems a field theory is formulated. In certain limits it reduces to the Hopfield model, the Sherrington-Kirkpatrick model, and the Viana-Bray model. The model has a percolation transition, while for RKKY couplings the ``concentration scaling'' T_g proportional to c occurs. Within the Gaussian approximation the Ginzburg-Landau expansion is considered in the clusterglass phase, that is to say, for not too small concentrations. Near special points, the prefactor of the cubic term, or the one of the replica-symmetry- breaking quartic term, may go through zero. Around such points new spin glass phases are found.