MARSHALLS SIGN RULE AND DENSITY-MATRIX RENORMALIZATION-GROUP ACCELERATION

In applications of White's density-matrix renormalization-group (DMRG) algorithm, computation time is dominated by the diagonalization of la rge sparse Hamiltonians by iterative diagonalization algorithms, whose convergence can be decisively accelerated by the usage of good start vectors. In this paper I show how, using the Marshall sign rule, in a wide class of antiferromagnetic models the number of diagonalization i terations can be reduced below 10, sometimes down to 2, accelerating t he DMRG by an order of magnitude. This acceleration, applicable during the growth of long chains, complements the acceleration procedure pro posed by White. To illustrate the feasibility of the approach, I show how it performs if applied to the calculation of the Haldane gap for S =2.