Breakthrough of polymers in interactive liquid chromatography.

Two separate peaks are observed for narrow polymer standards in both isocratic and gradient HPLC. One peak appears around the solvent front (the "solvent-plug peak" or "breakthrough peak"), whereas the second peak is retained significantly--or even highly. Although the effect has been observed many times before, it has never been rigorously explained. In this paper we provide a detailed explanation for the breakthrough peak. The two completely separate peaks are demonstrated not to represent to different fractions of the sample (e.g., the low- and high-molecular-mass parts of the distribution). Both peaks are representative of the entire polymeric sample for narrow polymer standard. Because the amount of the polymer in the breakthrough peak may vary, the quantitative analysis of the polymers by LC is jeopardized. The effects of the sample solvent, the (initial) mobile phase composition, the injection volume, the injected sample concentration, the column temperature, and the analyte structure and molecular mass on the breakthrough peak were investigated in LC experiments involving standards of polystyrene and poly(methyl methacrylate). Three necessary and sufficient conditionsare suggested for the breakthrough phenomenon to be observed. Recommendations to avoid the breakthrough phenomenon are given, culminating in a structured method for selecting the best possible sample solvents.