Polymer reactors and molecular weight distribution. Part III. Gel permeation chromatography, methods of correcting for imperfect resolution

This paper reports a preliminary evaluation of four methods of correcting for imperfect resolution in gel permeation chromatography. These are methods of Tung, Smith, Hess and Kratz, and Pickett. To evaluate the methods over a range of resolutions, many broad- and narrow-distribution polystyrene samples covering the molecular weight range 1000–500,000 were analyzed with three different column combinations having residence times from 20 to 150 min. The results show the increasing importance of the resolution correction with increasing molecular weight and decreasing residence times. For Tung's method the correction to molecular weight averages ranged from about 4% for 150 min. of residence time to about 12% for 20 min. Residence times as short as 10 or 15 min. should be possible without serious loss of accuracy. Inconsistencies in the high molecular weight results for the methods of Tung and Smith indicate inadequacies in the assumed chromatogram shapes for single molecular species, due to effects of column overloading and species interaction. These effects can be reduced by decreasing the sample concentration. When these effects are not prominent, the methods of Tung and Smith appear to be adequate. Smith's modified method permits the use of various chromatogram shapes for single species to account for overloading and interaction. Computational difficulties at present limit the use of the method of Hess and Kratz. Pickett's method looks promising, especially if concentration dependence can be included in the chromatograms representing the single species. Further evaluation needs to be done to determine which method best handles the region where overloading and species interaction occur.