MONTE-CARLO ALGORITHMS FOR THE ATOMISTIC SIMULATION OF CONDENSED POLYMER PHASES

Significant progress has been made recently in the field of atomistic simulation of polymer melts through the advent of new powerful Monte C arlo methods. This article reviews the state of the art in the area. S ampling the configurational space of a dense polymer system is difficu lt, because of complications introduced by high density and the connec tivity of the chain molecules. We describe how some novel algorithms a ttempt to solve the problem, compare them using a set of stringent per formance criteria and discuss their strengths and their weaknesses, th eir successes and their failures. Although we have still not reached t he stage where realistically long polymeric chains with atomistic deta il can be treated successfully, there is ground for hope. Configuratio n-bias Monte Carlo (CBMC) and its extensions, concerted-rotation (ConR ot)-based algorithms, and hybrid Monte Carlo (HMC) have opened up new possibilities for the investigation of more realistic polymer models t han the ones used hitherto. The field of possible applications is vast : studies of polymers in melts and in solution, prediction of single-p hase thermodynamic properties and phase equilibria, biopolymer modelli ng and, hopefully, the long-time behaviour of macromolecular systems, may soon become tractable with the rapid evolution of novel Monte Carl o methods.