LIGAND-FIELD STABILIZATION IN NICKEL-COMPLEXES THAT EXHIBIT EXTRAORDINARY GLASS-TRANSITION TEMPERATURE ENHANCEMENT

A thermodynamic interpretation of the ligand field stabilization energy appropriate to pseudo-octahedral d8 complexes is employed to estimate the synergistic enhancement of the glass transition temperature in blends of nickel acetate with poly(4-vinylpyridine) (P4VP). The maximum enhancement of T(g) is almost-equal-to 100-degrees-C, which occurs at a metal/polymer-repeat-unit concentration of approximately 1:2 on a molar basis. This suggests that the nickel cation coordinates to two pyridine ligands under favorable conditions. Energetic considerations coupled with the crystal structure of nickel acetate tetrahydrate suggest that pyridine ligands replace weak-base waters of hydration in the coordination sphere of the metal center. "Effective" coordination cross-links form when the pyridine pendant groups reside on different macromolecular chains. Ligand field stabilization energies (LFSE) are calculated for two hydrated nickel acetate model complexes that coordinate to either one or two pyridine groups in a pseudo-octahedral mixed-ligand arrangement. These models simulate the formation and disruption of coordination cross-links, and the LFSE's are consistent with the fact that cross-link dissociation is an endothermic process. The positive quantity R{T(g)(blend) - T(g)(undiluted P4VP)} represents thermal energy that must be supplied to dissociate coordination cross-links. The simple coordination-interaction model equates this thermal energy to the LFSE difference between the two nickel acetate model complexes mentioned above. The concentration dependence of the synergistic T(g) response is formulated via the quadratic x(1 - x) following the Margules symmetric model for nonideal energetics, where x represents the mole fraction of the metal salt. The experimental enhancement of T(g) represents 70% of the prediction based on LFSE's of the pseudo-octahedral model complexes. Reasons for the discrepancy between experiment and prediction are discussed in terms of microstructures not accounted for by the simple model.