Physico-chemical aspects of polyethylene processing in an open mixer 6. Discussion of hydroperoxide formation and decomposition

The work concerns different mechanistic and kinetic aspects of hydroperoxide formation and decomposition. It has been shown previously that initiation in mixers cannot be attributed to the free radicals formed on hydroperoxide decomposition but must result essentially from primary initiation. The discussion in this work showed that this primary initiation should result from mechanical breakdown of the polymer in the mixer. The absence of any significant temperature effect on the initial rate of hydroperoxide formation with PE-LD leads to additional results concerning the initiation and termination reactions. The best explanation involves a model based on shear induced free radical formation depending on polymer melt viscosity. Formation of chain carrying radicals by diffusion out of the cage is also governed by melt viscosity. The oxidation chains initiated this way are terminated essentially by geminate recombination following chemical migration of the free valence according to the "relay race" mechanism. The initial rate of hydroperoxide formation on polyethylene processing has been related to the polymer melt flow by an empirical equation. This equation shows the contrary effects of melt viscosity on mechanically induced chain breaking (enhanced with increasing viscosity) and on diffusion of the free radicals out of the cage (reduced with increasing viscosity). The ratio of the initial rate of hydroperoxide formation to the maximum of the hydroperoxide concentration reached gives access to the rate constant for hydroperoxide decomposition. The activation energies determined this way do neither correspond to conventional monomolecular nor to conventional bimolecular hydroperoxide decomposition. Hence, these conventional reactions do not determine the course of the oxidation of polyethylene in open mixers. The discussion shows that many ideas on mechanisms and kinetics in polymer melts need to be revised. The same should be valid even more so for polymers in the solid phase. (C) 2000 Elsevier Science Ltd. All rights reserved.