Long Chain-Branching by Size Exclusion Chromatography: Effects of Molecular Topology and Other Sources of Error

MEIRA, G.R.; VEGA, J.R.; CLEMENTI, L.A.

Torun

Simposio; 29th. International Symposium on Chromatography (ISC 2012); 2012

This work highlights several problems associated with the determination of the distributions of molar masses (MMD) and number of long branches per molecule (LBD)via Size Exclusion Chromatography (SEC) with molar mass-sensitive detectors; i.e.: a Specific Viscosity (SV) detector and/or a Light Scattering (LS) photometer. At any given hydrodynamic volume, the molar mass increases with the level of branching. Thus, whole distributions of molar masses and branched topologies are instantaneously present in the detector cells, even under perfect fractionation by hydrodynamic volume. For quantitative estimates of the global MMD and LBD, such instantaneous distributions must be narrow. However, this condition has not been verified in branched polyacrylates obtained by radical polymerization, where instantaneous molar mass dispersities greater than 2 were observed at high molar masses [1]. Simulated examples are presented to estimate the errors on the MMD and LBD. The chromatograms of the different chain topologies (with each topology characterized by the number of trifunctional branching points per molecule) were simulated from the knowledge of their individual MMDs, in turn obtained via deterministic and stochastic polymerization models. The applicability of the Zimm-Stockmayer (Z–S) equations [2] onto branched homopolymers obtained by radical polymerization is discussed. Apart from the problems introduced by the chromatographically-complex nature of branched homopolymers, the following additional sources of error are evaluated: a) presence of band broadening; b) uncertainty in data treatment constants; c) low sensitivities of the SV and LS detectors at low molar masses; and d) uncertainties in the MMD tails introduced by the signals ratio. The ultimate aim of the work is to provide recommendations on the expected accuracies of the various possible SEC configurations and associated data processing procedures.