Applicability of headspace solid-phase microextraction to the determination of multi-class pesticides in waters.

The applicability of headspace solid-phase microextraction (HS-SPME) to pesticide determination in water samples was demonstrated by evaluating the effects of temperature on the extraction of the pesticides. The evaluations were performed using an automated system with a heating module. The 174 pesticides that are detectable with gas chromatograph were selected objectively and impartially based on their physical properties: vapor pressure and partition coefficient between octanol and water. Of the 174 pesticides, 158 (90% of tested) were extracted with a polyacrylate-coated fiber between 30 and 100 degrees C and were determined with gas chromatograph-mass spectrometry. The extraction-temperature profiles of the 158 extracted pesticides were obtained to evaluate the effects of temperature on the extraction of pesticides. The pesticides were classified into four groups according to the shape of their extraction-temperature profiles. The line of demarcation between extractable pesticides and non-extractable pesticides could be drawn in the physical property diagram (a double logarithmic plot of their vapor pressure and partition coefficient between octanol and water). The plot also revealed relationships between classified extraction features and their physical properties. The new method for multi residue screening in which the analytes were categorized into sub-groups based on extraction temperature was developed. In order to evaluate the quantitivity of the developed method, the 45 pesticides were chosen among the pesticides that are typically monitored in waters. Linear response data for 40 of the 45 was obtained in the concentration range below 5 microg/l with correlation coefficients ranging between 0.979 and 0.999. The other five pesticides had poor responses. Relative standard deviations at the concentration of the lowest standard solution for each calibration curve of the pesticides ranged from 3.6 to 18%. The value of 0.01 microg/l in the limits of detection for 17 pesticides was achieved only under the approximate conditions for screening, not under the individually optimized conditions for each pesticide. Recoveries of tested pesticides in actual matrices were essentially in agreement with those obtained by solid-phase extraction.