Problems Associated with SEC Measurements of Polymers Among Laboratories: Summary on Cooperative Determination of Molecular Weight Averages of Polystyrenes by SEC in Japan

How much does band broadening adulterate results deduced from chromatograms measured by size-exclusion chromatography really?

A number of polystyrene samples and standards were measured with several column combinations which differed in their extent of band broadening, sigma(BB). Comparison of the derived chain length distributions showed in some cases good agreement even despite strong distinctions in the determined sigma(BB) values. The calculated number and weight averages of the samples were almost identical in most cases for the column combinations used. Furthermore, the points of inflection of the standards and of multimodale distributions composed of narrow (Poisson like) peaks in microemulsion were examined. When conferred with the theoretical values which were calculated for assumed Poisson distributions the respective deviations from the "true" ones were as high as 10% (almost reaching 20% in unfavorable cases). Simple (correction) procedures were tested in order to obtain the actual average values as well as the effective points of inflection of narrow distributions.

Evaluation of size-exclusion chromatography and size-exclusion electrochromatography calibration curves

Size-exclusion chromatography (SEC) and size-exclusion electrochromatography (SEEC) are chromatographic techniques used to determine molecular mass (weight) distributions (MWD) of polymers. One important step in the data treatment to derive MWD parameters is the modelling of the calibration curves. The calibration curves applied in SEC and SEEC are generally not linear. In this study the modelling of calibration curves is being examined. Different polynomial models have been evaluated and compared, not only for model fit but also for their predictive properties. It was found that sometimes a straight line and sometimes a third-order polynomial model were best. The best model across the effective range (also called linear range) is not always found to be a straight line. The SEEC curves were found to have considerably higher prediction errors than the SEC ones. Reduction of the number of calibration standards to five or six did not greatly affect the predictive properties of the calibration curves, neither in SEC nor in SEEC.