Effect of pretreating samples with boric acid before liquid-chromatographic determination of urinary catecholamines.

Isolation of catecholamines on prepacked gravity-fed ion-exchange columns in conjunction with liquid-chromatographic separation is occasionally subject to disturbances of unknown origin, especially after injection of H3BO3 eluates from the ion-exchange columns onto the chromatography column. Peak spreading, sometimes even peak splitting, has been observed in many laboratories--effects easily confused with those sometimes caused by voided columns or by impurities. To elucidate the origins of these interferences, we performed a combined voltammetric and chromatographic investigation. We show that complexation of the catechol group by borate ions and dissociation of those pH-labile complexes in the course of the chromatographic separation is the cause of those peak distortions. Addition of a small volume of diluted HCI to the borate eluates eliminates such interference and leads to reproducible, easily interpretable chromatographic separations.