Within the ε-expansion scheme, we argue that the upper critical dimension for the continuous transition (if it exists) in the dilute limit of branched polymers should be du=4. However, Monte-Carlo simulations in d=2, 3 and 4 for a single branched polymer on a finite lattice show evidence of a first-order transition at K=K0 and not a continuous transition which we observe for a linear polymer. Here, K represents bond activity. Our extensive simulation in d=2 is compared with previously known results. The existence of a first-order transition at K0 is not inconsistent with power-law singularities as [Formula: see text], which have been observed by various workers. However, for K<K0, the self-similarity which is a prerequisite of a fractal (i.e., a critical) object is only approximate. At K=K0 the branched polymer is a regular, compact object and not a fractal. In constrast, a linear polymer is a fractal at the transition, as is expected and as our simulation also suggests. Thus, our work shows that the order of the thermodynamic limit N→∞ and K→K0 is important, provided the transition remains first-order in the thermodynamic limit. In previous works, one considers [Formula: see text] after the limit N→∞ has already been taken, whereas we are interested in K→K0, followed by N→∞.
Read full abstract