Analytical performance testing of an atrazine immunoassay system.

A rigorous performance evaluation of an enzyme immunoassay (EIA) kit for the determination of atrazine in water samples was undertaken. Eleven individual batches of samples containing standards and spiked drinking waters were analysed and precision, bias and limit of detection were measured using statistical analysis. The technique was shown to be capable of achieving performance criteria (a total standard deviation of less than 5% or 2.5 ng, whichever is the greater) demanded of modern analytical systems and achieved a limit of detection of 9.2 ng l-1. There was no statistically significant bias measured for drinking water samples. Interference tests showed that the atrazine immunoassay was not significantly affected in the pH range 4.0-8.0 or by drinking water matrix components (anions, cations and chlorination by-products), even at their maximum allowable concentrations. There was a small extent of cross-reaction with simazine and atrazine degradation products, but given the persistence of atrazine, through its resistance to hydrolysis, breakdown products are likely to be present at much lower concentrations than the parent compound in drinking water. Simazine may potentially be more problematic, so it would be prudent to monitor a proportion of samples for simazine to determine the extent to which this may be contributing to the 'atrazine' measured in drinking water samples using the EIA kit.