Abstract: | In the calculation of molecular weight averages from GPC chromatograms and in the correction for zone broadening effects, it is assumed that the retention volumes of the individual species are not affected by the presence of the other components in the sample (linear fractionation process). This assumption was tested by the computer comparisons of the chromatograms of mixtures of narrow molecular weight distribution polystyrene standards and the envelopes of the individual chromatograms of the components of the mixtures. Disagreements between experimental and computer synthesized chromatograms were interpreted as non-linear fractionation effects. The investigation covered a range of molecular weights (1.8 × 106 to 2100), molecular weight distributions (binary and seven component mixtures), and sample loadings (2 to 70 mg). Accurate normalizations of raw GPC chromatograms and calculations of molecular weight averages were made possible through an automated GPC apparatus using a high-speed computer for data acquisition and data reduction. It was found that some nonlinear fractionation effects (large disagreement between chromatograms of the mixture and the computed envelopes of the individual chromatograms of the components of the mixture) became more pronounced as sample loading was increased and as the molecular weight distribution of the sample narrows. The nonlinear effect was also found to be sensitive to the molecular weight of the sample. |