The equilibrium melting points of random ethylene-octene copolymers: A test of the Flory and Sanchez-Eby theories

Abstract

Ethylene/l-octene copolymers produced with metallocene catalysts are believed to have a homogeneous comonomer content with respect to molecular weight. Two series of copolymers of different molecular weights with a 1-octene content ranging from 0 to 39 branches per 1000 carbon atoms were studied. The influence of branch content on structure and melting behavior as well as on isothermal and nonisothermal bulk crystallization was studied. In this article, the equilibrium melting temperatures of ethylene/l-octene random copolymers is the focus. The principal techniques used were thermal analysis and small-angle X-ray scattering. The use of Hoffman-Weeks plots to obtain the equilibrium melting temperatures of ethylene/l-octene random copolymers resulted in nonsensical high values of the equilibrium melting point or showed behavior parallel to the T m = T c line, resulting in no intercept and, hence, an infinite equilibrium melting point. The equilibrium melting temperatures of linear polyethylenes and homogeneous ethylene/ l-octene random copolymers were determined as a function of molecular weight and branch content via Thompson-Gibbs plots involving lamellar thickness data obtained from small-angle X-ray scattering. This systematic study made possible the evaluation of two equilibrium melting temperature depression equations for olefin-type random copolymers, the Flory equation and the Sanchez-Eby equation, as a function of defect content and molecular weight. The range over which the two equations could be applied depended on the defect content after correction for the effect of molecular weight on the equilibrium melting temperature. The equilibrium melting temperature, T m 0 (n, p B ), of the ethylene/l-octene random copolymers was a function of the molecular weight and defect content for low defect contents (p B ≤ 1.0%). T m 0 (n, p B ) was a weak function of molecular weight and a strong function of the defect content at a high defect content (p B ≥ 1.0%). The Flory copolymer equation could predict T m 0 (n, p B ) at p B ≤ 1.0% when corrections for the effect of molecular weight were made. The Sanchez-Eby uniform inclusion model could predict T m 0 (n, p B ) at a high defect content (1.6%≤p B ≤ 2.0%). We conclude that some defects were included in the crystalline phase and that the excess free energies (18-37 kJ/mol) estimated in this study were within the theoretical range.