It is rigorously proved that at any nonzero temperature, a one- or two-dimensional isotropic spin-
S Heisenberg model with finite-range exchange interaction can be neither ferromagnetic nor antiferromagnetic. The method of proof is capable of excluding a variety of types of ordering in one and two dimensions.
We rigorously prove that in three or more dimensions, the nearest-neighbor, simple-cubic, ferromagnetic, quantum Heisenberg model of spin
S(=21,1,…) has a phase transition at nonzero temperature.
The mean-field Heisenberg ferromagnet is a quantum spin model on the complete graph with isotropic spin-1/2 interactions. This non-commuting Hamiltonian is permutation and
SU(2) invariant, and its Gibbs states undergo an
SU(2) symmetry breaking phase transition at inverse temperature
β=2. We consider the associated Davies generator, a canonical model of open-system thermalization, and prove tight asymptotic estimates for its spectral gap at all noncritical temperatures. For fixed
β<2, the gap as a function of number of qubits
n is
Θ(1), while for fixed
β>2 the gap is
Θ(n−1). The matching upper bound of the spectral gap is witnessed by the total magnetization order parameter, suggesting that the low-temperature (
β>2) slowdown is associated with broken continuous symmetry. Two key ingredients in our approach are a comparison argument, which introduces auxiliary generators to bound dissipation on nontrivial representations of the symmetry groups
SU(2) and
Sn, and a decomposition of the space of observables into spherical tensor operators to reveal a form of monotonicity.
A Green-function method is employed within a random phase approximation to study the one-dimensional quantum Heisenberg model with long-range ferromagnetic interaction proportional to ${r}^{\ensuremath{-}p}.$ We have shown that for
1<p<2 there exists a phase transition at finite temperatures and estimate its critical temperature.
A modified spin-wave theory is applied to the one- and two-dimensional quantum Heisenberg model with long-range ferromagnetic interactions proportional to ${\mathit{r}}^{\mathrm{\ensuremath{-}}\mathit{p}}$[scrH=-1/2${\mathit{tsum}}_{\mathit{i}\mathrm{\ensuremath{\ne}}\mathit{j}}$ (
J0/
rijp)
Si\ensuremath{\cdot}
Sj]. It is shown that for dp2d there exists a phase transition at finite temperatures; the critical temperature is estimated. The susceptibility and the specific heat are obtained at low temperatures. Our results for d=1, p=2, and S=1/2 agree with the exact solution of the Haldane-Shastry model. For d=1, we find a scaling relation \ensuremath{\alpha}+1=${\ensuremath{\gamma}}^{\mathrm{\ensuremath{-}}1}$ when 2p3.