THE EARLY STAGES OF THE PHASE-SEPARATION DYNAMICS IN POLYDISPERSE POLYMER BLENDS

Citation
C. Huang et Mo. Delacruz, THE EARLY STAGES OF THE PHASE-SEPARATION DYNAMICS IN POLYDISPERSE POLYMER BLENDS, Macromolecules, 27(15), 1994, pp. 4231-4241
Citations number
21
Categorie Soggetti
Polymer Sciences
Journal title
ISSN journal
00249297
Volume
27
Issue
15
Year of publication
1994
Pages
4231 - 4241
Database
ISI
SICI code
0024-9297(1994)27:15<4231:TESOTP>2.0.ZU;2-G
Abstract
The thermodynamics and the dynamics of incompatible polydisperse polym er blends are analyzed. The free energy is constructed following the F lory-Huggins approach, where the degree of incompatibility is characte rized by the Flory interaction parameter chi. The Cahn-Hillard approxi mation is used to analyze the early stages of spinodal decomposition d ynamics of a polymer blend quenched into the unstable region. A blend of polydisperse A polymers with the Schulz-Flory distribution and mono disperse B polymers is analyzed by treating polymer A as a one-, two-, and three-component system with a weight-average degree of polymeriza tion and a polydispersity index, which we refer to as two-, three-, an d four-component models, respectively. The thermodynamics and the dyna mics of incompatible monodisperse A-monodisperse B polymer blends are consistent no matter which model is used. When polymer A is polydisper se, however, [S(k,t) - S(k,0)]/S(k,0), where S(k,t) is the characteris tic structure function, is definitely different in the three different models due to kinetic effects. The differences are dependent on the f unctional form of the Onsager coefficients. For wavevector-independent Onsager coefficients, the reduced wavevector for which [S(k,t) - S(k, 0)]/S(k,0) is a maximum, k(peak) is always equal to 1/square-root 2 i n the two-component model, while k(peak) increases as x increases in the three- and four-component models. While for wavevector-dependent O nsager coefficients, k(peak) decreases as chi increases in the three different component models. As chi --> infinity, the difference in k(p eak) between two- and three-component models and between three- and f our-component models is 0.05 and 0.02, respectively, independent of th e weight-average degree of polymerization when the polydispersity inde x of polymer A is equal to 2.0. When the polydispersity index of polym er A is reduced to 1.5, the difference in k(peak) becomes 0.04 and 0. 01, respectively.