Evidence for reptation in an entangled polymer melt

Abstract

ENTANGLEMENTS, their nature and their role in the dynamic properties of concentrated polymer solutions and melts are not well understood^1,2. The classical molecular view of entanglements has been one of rope-like intermolecular couplings at a number of points along the length of a molecule; molecules in motion would drag past these couplings, the essential effect being one of enhanced friction^1,3. There has been a growing realisation that this model is inadequate^2,4,5. The essence of the problem, rather, seems to be that of the topological restrictions imposed on the motion of each polymer molecule by its neighbours: movement of a given polymer chain is constrained at the points of entanglement or intersection with adjacent chains². Theoretical treatment of the topological problem is difficult⁶, and has met only with limited success⁵. An interesting proposal regarding the motion of molecules within entangled polymer systems has been put forward by De Gennes^4,7: according to this, the motion of a given polymer molecule is confined within a virtual ‘tube’ defined by the locus of its intersections (or points of ‘entanglement’) with adjacent molecules (Fig. 1). The molecule is constrained to wriggle, snake-like, along its own length, by curvilinear propagation of length defects such as kinks or ‘twists’⁸ along the tube; this mode of motion has been termed reptation⁴ (from reptile). Reptative motion clearly satisfies the central requirement of entangled systems: that of the non-crossability by a given chain of the contours of its adjacent neighbours. In a real polymer melt the topological environment of any given molecule (that is, the virtual ‘tube’ surrounding it) will itself change with time. This is because the adjacent molecules defining it are themselves mobile. If this reorganisation is sufficiently slow then the translational motion of the enclosed molecule will be effectively curvilinear (reptative). Consideration of the problem⁹ suggests that this will be the case in an entangled system. One then expects translational diffusion to be dominated by reptation. There is no direct experimental evidence supporting the physical reality of curvilinear motion in entangled polymer systems. I report here the results of experiments on diffusion within a polyethylene melt critically designed to test the reptation concept.