SearcharxivSearch

arXiv subjects

Uzi Hizi

Publications and source records attributed to Uzi Hizi.

2 recordsLinked to original sources

Anharmonic Ground state selection in the pyrochlore antiferromagnet

In the pyrochlore lattice Heisenberg antiferromagnet, for large spin length $S$, the massive classical ground state degeneracy is partly lifted by the zero-point energy of quantum fluctuations at harmonic order in spin-waves. However, there remains an infinite manifold of degenerate collinear ground states, related by a gaugelike symmetry. We have extended the spin-wave calculation to quartic order, assuming a Gaussian variational wavefunction (equivalent to Hartree-Fock approximation). Quartic calculations \emph{do} break the harmonic-order degeneracy of periodic ground states. The form of the effective Hamiltonian describing this splitting, which depends on loops, was fitted numerically and also rationalized analytically. We find a family of states that are still almost degenerate, being split by the term from loops of length 26. We also calculated the anharmonic terms for the checkerboard lattice, and discuss why it (as well as the kagomé lattice) behave differently than the pyrochlore at anharmonic orders.

cond-mat.str-el

Effective Hamiltonians for large-S pyrochlore antiferromagnet

The pyrochlore lattice Heisenberg antiferromagnet has a massive classical ground state degeneracy. We summarize three approximation schemes, valid for large spin length $S$, to capture the (partial) lifting of this degeneracy when zero-point quantum fluctuations are taken into account; all three are related to analytic loop expansions. The first is harmonic order spin waves; at this order, there remains an infinite manifold of degenerate collinear ground states, related by a gauge-like symmetry. The second is anharmonic (quartic order) spin waves, using a self-consistent approximation; the harmonic-order degeneracy is split, but (within numerical precision) some degeneracy may remain, with entropy still of order $L$ in a system of $L^3$ sites. The third is a large-N approximation, a standard and convenient approach for frustrated antiferromagnets; however, the large-N result contradicts the harmonic order at $O(S)$ hence must be wrong (for large $S$).

cond-mat.str-el