We propose a new algorithm which works effectively in global updates in Monte Carlo study. We apply it to the quantum spin chain with next-nearest-neighbor interactions. We observe that Monte Carlo results are in excellent agreement with numerically exact ones obtained by the transfer matrix method.
We present detailed discussion on a new approach we proposed in a previous paper to numerically study quantum spin systems. This method, which we will call re-structuring method hereafter, is based on rearrangement of intermediate states in the path integral formulation. We observed our approach brings remarkable improvement in the negative sign problem when applied to one-dimensional quantum spin-1/2 system with next-nearest-neighbor interactions. In this paper we add some descriptions on our method and show results from analyses by the exact diagonalization and by the transfer matrix method of the system on a small chain. These results also indicate that our method works quite effectively.
We apply the transfer-matrix density matrix renormalization group (DMRG) approach to frustrated quantum spin chains, down to T=0.025K in the thermodynamic limit. The incommensurability problem of exact diagonalization and the negative sign problem of quantum Monte Carlo vanish completely. For illustration, we give results for chains with next-nearest-neighbor frustration and for the delta chain, which has been a testbed for many thermodynamic methods. The DMRG approach proves to be an extremely powerful method for the old problem of the thermodynamics of frustrated systems.
We start this series of lectures on quantum Monte Carlo methods with the simplest model for a strongly correlated system, namely an antiferromagnetic Heisenberg S‐1/2 chain. We will review methods for its simulation starting with the world‐line algorithm and then introducing the loop‐algorithm with global updates. We discuss next a model for doped antiferromagnets, with emphasis on the strong correlation limit that is central for high temperature superconductors and related materials, namely the so‐called t‐J model. An exact canonical transformation for this model that leads to a formulation with separated charge and spin degrees of freedom will be discussed. Results for a single hole in antiferromagnetic chains and planes will be summarized, giving an initial picture of charge dynamics in a quantum antiferromagnet. An additional algorithmic element, namely the determinantal method, will be introduced next, in order to deal with a finite number of charge degrees of freedom. Merging the loop‐ and the determinantal methods leads to a new one, the hybrid‐loop algorithm, that combines the efficiency of the loop‐algorithm for spins with the one of the determinantal method for fermions. Finally we discuss a first application of the hybrid‐loop algorithm, where the spectral function of the t‐J model in one dimension with finite doping is determined. There we can see how charge spin separation shows up in the one‐particle spectral function, and a detailed description can be achieved by comparison with an analitically soluble model, namely the t‐J model with 1/r2 interaction. The results show that in addition to spinons and holons expected in one‐dimensional metals, antiholons with charge Q = 2e, spin S = 0, and twice the mass of the holons are necessary to describe the inverse photoemission spectra at finite doping.