Effective mapping of spin-1 chains onto integrable fermionic models. A study of string and Néel correlation functions

We derive the dominant contribution to the large-distance decay laws of correlation functions towards their asymptotic limits for a spin chain model that exhibits both Haldane and Neel phases in its ground-state phase diagram. The analytic results are obtained by means of an approximate mapping between a spin-1 anisotropic Hamiltonian onto a fermionic model of noninteracting Bogoliubov quasiparticles related in turn (via Jordan–Wigner transformation) to the XY spin-1/2 chain in a transverse field. This approach allows us to express the spin-1 string operators in terms of fermionic operators so that the dominant contribution to the string correlators at large distances can be computed using the technique of Toeplitz determinants. As expected, we find long-range string order both in the longitudinal and in the transverse channel in the Haldane phase, while in the Neel phase only the longitudinal order survives. In this way, the long-range string order can be explicitly related to the components of the magnetization of the XY model. Moreover, apart from the critical line, where the decay is algebraic, we find that in the gapped phases the decay is governed by an exponential tail multiplied by power-law factors. As regards the usual two points correlation functions, we show that the longitudinal one behaves in a 'dual' fashion with respect to the transverse string correlator, namely both the asymptotic values and the decay laws exchange when the transition line is crossed. For the transverse spin–spin correlator, we always find a finite characteristic length which is an unexpected feature at the critical point. The results of this analysis prove some conjectures put forward in the past. We also comment briefly on the entanglement features of the original system versus those of the effective model. The goodness of the approximation and the analytical predictions are checked versus density-matrix renormalization group calculations.