1998-02-15
14:15 at CAB D2

Bridging the gap from microscopic to semi-macroscopic modeling of polymers

Kurt Kremer

Max-Planck-Institut f. Polymerforschung, Mainz, Germany

At a first sight it is tempting to simulate a melt of polymers on the atomistic level. Such a simulation should then provide a complete information about the properties of the system. It is almost trivial to show that even with the present high speed computers such a simulation is impossible. On the other hand, it is questionable whether such an attempt should be undertaken at all. It would provide data containing a huge amount of irrelevant information. As it holds for all disordered complex macromolecular materials polymers can be characterized by a hierarchy of different length and time scales. Especially time scales span an extremely wide range. On the microscopic level dynamics is dominated by the local oscillations of bond angles. On the semi-macroscopic level the behavior is dominated by the overall diffusion of the chains or the relaxation of the overall conformation of the objects. These times depend on chain length and temperature can easily span more than 10 decades. In this sense I will discuss first attempts to bridge the gap from microscopic to mesoscopic and thereafter to the semi-macroscopic regime within a simulation scheme. I will describe a mapping procedure to go from microscopic to a mesoscopic description. Thereafter, first steps toward the next level of description are discussed. Starting from the conformations of polymer chains on the coarse grained level, the chains are mapped onto an extended soft particle with only three internal degrees of freedom compared to the 0 (3N) degrees of freedom of the whole object.