Unified theory for the scaling of the crossover between strong and weak disorder behaviors of optimal paths and directed/undirected polymers in disordered media

Abstract

In this work we are concerned with the crossover between strong disorder (SD) and weak disorder (WD) behaviors in three well-known problems that involve minimal paths: directed polymers (directed paths with fixed starting point and length), optimal paths (undirected paths with fixed end-to-end/spanning distance) and undirected polymers (undirected paths with fixed starting point and length). We present a unified theoretical framework from which we can easily deduce the scaling of the crossover point of each problem. Our theory is based on the fact that the SD limit behavior of these systems is closely related to the corresponding percolation problem. As a result, the properties of those minimal paths are completely controlled by the so-called red bonds of percolation theory. Our model is first addressed numerically and then approximated by a ``two-term'' approach. This approach provides us with an analytical expression that seems to be reasonably accurate. The results are in perfect agreement with our simulations and with most of the results reported in related works. Our research also lead us to propose this crossover point as a universal measure of the disorder strength in each case. Interestingly, that measure depends on both the statistical properties of the disorder and the topological properties of the network.