DYNAMIC ALGORITHM FOR LOOP-ERASED SELF-AVOIDING RANDOM-WALKS IN 2-DIMENSIONS AND 3-DIMENSIONS

Monte Carlo simulation of loop-erased self-avoiding random walks in two and three dimensions (2D and 3D) was performed, using an algorithm different from that of Guttmann and Bursill. We obtained 2ν=1.571 in 2D and 2ν=1.230 in 3D. The 3D result compares most favorably with that of Guttmann and Bursill [J. Stat. Phys. 59, 1 (1990)]. Lawler's conjecture [J. Stat. Phys. 50, 91 (1988)] that loop-erased self-avoiding random walks are in a different universality class than self-avoiding random walks is strengthened. © 1995 The American Physical Society.