Some comments on the determination of microcystin toxins in waters by high-performance liquid chromatography

Abstract

This paper describes some of the problems that may arise during the analysis of microcystin toxins and nodularin using the method described by Lawton, Edwards and Codd (Analyst, 1994, 119, 1525). Difficulty may be caused by the appearance of additional peaks in the HPLC chromatograms as a result of leaching of material from the C₁₈ trifunctional (C_18t) solid-phase extraction cartridges, and co-elution of other organic compounds in the water makes it difficult to identify toxins using their characteristic UV spectra, expecially at low toxin concentrations. Samples containing chlorine must be treated with sodium thiosulfate since the presence of chlorine results in low toxin recoveries, especially of nodularin. The published method specifies the use of sodium sulfite and acidification with trifluoroacetic acid for the removal of chlorine, but application of this procedure has been shown to produce inconsistent toxin recoveries, and zero percentage recoveries were recorded in 71% of the analyses performed with treated waters having a microcystin LR concentration of 0.4 µg l^–1. The use of microcystin LR as an ‘internal standard’ to adjust the recovery of other microcystin toxins is at best qualitative.