We use fermion mean field theory to study possible plaquette ordering in the antiferromagnetic SU(4) Heisenberg model. We find the ground state for both the square and triangular lattices to be the disconnected plaquette state. Our mean field theory gives a first order transition for plaquette ordering for the triangular lattice. Our results suggest a large number of low lying states.
In order to understand the properties of Mott insulators with strong ground state orbital fluctuations, we study the zero temperature properties of the SU(4) spin-orbital model on a square lattice. Exact diagonalizations of finite clusters suggest that the ground state is disordered with a singlet-multiplet gap and possibly low-lying SU(4) singlets in the gap. An interpretation in terms of plaquette SU(4) singlets is proposed. The implications for LiNiO_2 are discussed.
We predict that an external field can induce a spin order in highly frustrated classical Heisenberg magnets. We find analytically stabilization of collinear states by thermal fluctuations at a one-third of the saturation field for kagome and garnet lattices and at a half of the saturation field for pyrochlore and frustrated square lattices. This effect is studied numerically for the frustrated square-lattice antiferromagnet by Monte Carlo simulations for classical spins and by exact diagonalization for $S=1/2$. The field induced collinear states have a spin gap and produce magnetization plateaus.
Quantum electrodynamics in 2+1 dimensions is an effective gauge theory for the so called algebraic quantum liquids. A new type of such a liquid, the algebraic charge liquid, has been proposed recently in the context of deconfined quantum critical points [R. K. Kaul {it et al.}, Nature Physics {bf 4}, 28 (2008)]. In this context, we show by using the renormalization group in $d=4-epsilon$ spacetime dimensions, that a deconfined quantum critical point occurs in a SU(2) system provided the number of Dirac fermion species $N_fgeq 4$. The calculations are done in a representation where the Dirac fermions are given by four-component spinors. The critical exponents are calculated for several values of $N_f$. In particular, for $N_f=4$ and $epsilon=1$ ($d=2+1$) the anomalous dimension of the Neel field is given by $eta_N=1/3$, with a correlation length exponent $ u=1/2$. These values change considerably for $N_f>4$. For instance, for $N_f=6$ we find $eta_Napprox 0.75191$ and $ uapprox 0.66009$. We also investigate the effect of chiral symmetry breaking and analyze the scaling behavior of the chiral holon susceptibility, $G_chi(x)equiv<bar psi(x)psi(x)bar psi(0)psi(0)>$.
Crystal structure of spinel compound CuIr$_{2}$S$_{4}$ was examined by powder X-ray diffraction for the insulating phase below the metal-insulator transition at $T_{MI}$ = 230 K. The superstructure spots are reproduced by considering the displacement of Ir atoms. A model for the ionic ordering of Ir$^{4+}$ and Ir$^{3+}$ with the same number is proposed for the insulating phase on the basis of the structural analysis. The model suggests that the structural change at $T_{MI}$ is driven by the formation Ir$^{4+}$ dimers. In addition, we found that the superstructure spots becomes significantly weak below 60 K, without any significant effects on electric and magnetic properties. Possible mechanism for the transition is discussed.
We present results for the phase diagram of an SU($N$) generalization of the Heisenberg antiferromagnet on a bipartite three-dimensional anisotropic cubic (tetragonal) lattice as a function of $N$ and the lattice anisotropy $gamma$. In the isotropic $gamma=1$ cubic limit, we find a transition from N{e}el to valence bond solid (VBS) between N=9 and N=10. We follow the N{e}el-VBS transition to the limiting cases of $gamma ll 1 $ (weakly coupled layers) and $gamma gg 1$ (weakly coupled chains). Throughout the phase diagram we find a direct first-order transition from N{e}el at small-$N$ to VBS at large-$N$. In the three-dimensional models studied here, we find no evidence for either an intervening spin-liquid photon phase or a continuous transition, even close to the limit $gamma ll 1$ where the isolated layers undergo continuous N{e}el-VBS transitions.