Statistical analysis of a HPLC method for microcystins determination in drinking water

Statistical analysis of high performance liquid chromatography (HPLC) method for the determination of 0.1~1.0 mg·L
^-1
microcystins (MCs) in drinking water was investigated to optimize the instrument operating conditions and reduce the detection limit. The detection limits were obtained by two ways. One through method detection limit (MDL) and the other through detection limit from the calibration curve (DTC). The Cochran test results revealed the homoscedasticity of the HPLC method. The effect of system drift on the calibration curve was noticed by the lack of fit (LOF) test. MDL yielded a less reliable detection limit than DTC. The detection limit of HPLC method for MC-LR was 0.009 mg·L
^-1
under recommended operating conditions.