Information theoretic characterization of frustrated systems

A recently proposed measure of the frustration effect is discussed in detail in the Ising model on the Kagome lattice with nearest-neighbor and next-nearest-neighbor interactions. The measure quantifies the degree of correlation dominated by the frustration effect; it manifests the difference between frustrated and unfrustrated systems, its negative sign indicating the presence of frustration. We study the behavior of the measure for different values of interactions and show that it is responsible for the understanding of phase transitions in frustrated systems. The reentrance phenomenon is discussed in terms of the measure.