Variational wave functions for the Heisenberg antiferromagnet are generated by a recursive scheme. Numerical results for the ground-state energy are presented for the one-dimensional case, where they are better than previous variational results.
We present a variational wave function for the ground state of the two-dimensional spin-1/2 Heisenberg antiferromagnet. It is a modification of a Monte Carlo variational wave function proposed by Huse and Elser which gives excellent results for the ground-state energy but breaks the rotational symmetry of the squared staggered magnetization. In their case the x component of the squared staggered magnetization (${\mathit{m}}_{\mathit{x}}^{\mathrm{\ifmmode^\circ\else\textdegree\fi{}}}$)2 is much larger than (${\mathit{m}}_{\mathit{y}}^{\mathrm{\ifmmode^\circ\else\textdegree\fi{}}}$)2 and (${\mathit{m}}_{\mathit{z}}^{\mathrm{\ifmmode^\circ\else\textdegree\fi{}}}$)2, which are the y and z components, respectively. In this paper we show how to restore the symmetry between ${\mathit{m}}_{\mathit{x}}^{\mathrm{\ifmmode^\circ\else\textdegree\fi{}}}$ and ${\mathit{m}}_{\mathit{y}}^{\mathrm{\ifmmode^\circ\else\textdegree\fi{}}}$. However, ${\mathit{m}}_{\mathit{z}}^{\mathrm{\ifmmode^\circ\else\textdegree\fi{}}}$ is still much smaller than ${\mathit{m}}_{\mathit{x}}^{\mathrm{\ifmmode^\circ\else\textdegree\fi{}}}$. We also study the enforcement of the full rotational symmetry in our trial wave function using only two spin interaction terms. Finally, we discuss how errors in Monte Carlo data affect the efficiency of algorithms which determine more systematically the best variational parameters for the ground-state wave function.
We study a variational wave function for the ground state of the two-dimensional S=1∕2 Heisenberg antiferromagnet in the valence bond basis. The expansion coefficients are products of amplitudes h(x,y) for valence bonds connecting spins separated by (x,y) lattice spacings. In contrast to previous studies, in which a functional form for h(x,y) was assumed, we here optimize all the amplitudes for lattices with up to $32\ifmmode\times\else\texttimes\fi{}32$ spins. We use two different schemes for optimizing the amplitudes; a Newton conjugate-gradient method and a stochastic method which requires only the signs of the first derivatives of the energy. The latter method performs significantly better. The energy for large systems deviates by only $\ensuremath{\approx}0.06%$ from its exact value (calculated using unbiased quantum Monte Carlo simulations). The spin correlations are also well reproduced, falling $\ensuremath{\approx}2%$ below the exact ones at long distances (corresponding to an $\ensuremath{\approx}1%$ underestimation of the sublattice magnetization). The amplitudes h(r) for valence bonds of long length r decay as ${r}^{\ensuremath{-}3}$. We also discuss some results for small frustrated lattices.
The ground-state properties of the spin-(1/2 Heisenberg antiferromagnet on a square lattice are studied by using a simple variational wave function that interpolates continuously between the N\'eel state and short-range resonating-valence-bond states. Exact calculations of the variational energy for small systems show that the state with the lowest energy has long-range antiferromagnetic order. The staggered magnetization in this state is approximately 70% of its maximum possible value. The variational estimate of the ground-state energy is substantially lower than the value obtained for the nearest-neighbor resonating-valence-bond wave function.
We study the spin-1/2 Heisenberg antiferromagnet on a square lattice using variational Monte Carlo calculations. We construct a variational wave function which gives accurate ground-state energy and is consistent with sum rules for the dynamical structure function. The spin-wave velocity and the spin-stiffness constant can be calculated using sum rules. We obtain Zc==c/c0, where c0 is the classical spin-wave velocity. This and other ground-state quantities are in good agreement with spin-wave theory and Monte Carlo calculations.
Two-dimensional spin 1/2 antiferromagnetic Heisenberg models are numerically studied using Entanglement Perturbation Theory, where the ground state wave function is described by a product of local matrices defined at every site, and each matrix is optimized variationally to minimize the energy. We first apply this method for the spin 1/2 antiferromagnetic Heisenberg model on the square lattices to reproduce correctly the known ground state energy and the spin structure factors. Then, we study the spin 1/2 antiferromagnetic Heisenberg model on the triangular lattice with spatially anisotropic nearest neighbor couplings, J and J ', to establish the ground state phase diagram as a function of J'/ J.
We have found a new basis, described by three variables, in which the ground-state probability is concentrated in a small subspace. The variables are certain distances d and $\stackrel{\ifmmode \tilde{}\else \~{}\fi{}}{d}$ from the two N\'eel states, and an energy e. This leads to a variational wave function $\ensuremath{\varphi}(d,\stackrel{\ifmmode \tilde{}\else \~{}\fi{}}{d},e)$ which gives excellent spin-spin correlations for the ground state. Most of the probability occurs on a line $d+\stackrel{\ifmmode \tilde{}\else \~{}\fi{}}{d}=\mathrm{const}$ which connects the N\'eel states; surprisingly, $|\ensuremath{\varphi}(d,\stackrel{\ifmmode \tilde{}\else \~{}\fi{}}{d},e)|$ is largest and approximately constant on the central part of this line. Possible applications and generalizations to other models are discussed.