AUTOMATED HEADSPACE ANALYSIS OF FUMIGANTS 1,3-DICHLOROPROPENE AND METHYL ISOTHIOCYANATE ON CHARCOAL SAMPLING TUBES

Charcoal tubes are widely used for collecting organic vapor in the atm osphere, and the measurement is usually completed by analyzing an aliq uot of the solvent phase following solvent extraction, typically with carbon disulfide. However, the sensitivity of this method is limited a nd sometimes too low for monitoring contaminants at trace levels in th e environmental atmosphere. In this study, the potential of static hea dspace analysis techniques was explored on two common fumigants, 1,3-d ichloropropene (1,3-DCP) and methyl isothiocyanate (MITC), on both coc onut- and petroleum-based charcoal sampling tubes, using an automated and programmable headspace sampler. Three important parameters in the headspace analysis, equilibrating temperature and time, and amount of extracting solvent, were optimized individually for each compound-char coal tube combination to achieve maximum sensitivity of GC analysis. H igher stability was observed for both isomers of 1,3-DCP and MITC on p etroleum-based charcoal, and 180 and 190 degrees C, and 5 min were sel ected as the equilibrating temperatures and time, respectively. On coc onut-based charcoal tubes, however, all the compounds were more sensit ive to the temperature, and 160 and 140 degrees C, and 5.0 and 3.0 min were therefore determined as the equilibrating temperatures and times for the 1,3-DCP isomers and MITC, respectively. Reducing solvent volu me from 3 to 1 ml in 9-ml headspace vials improved the sensitivity and 1.0 ml benzyl alcohol was therefore selected for all the compound-cha rcoal tube combinations. Compared to the conventional extraction metho d with CS2, the optimized headspace methods were 10-35 times more sens itive, and equivalently reproducible except for MITC on coconut-based ORBO-32 tubes. Better sensitivity and precision of measurements were c onsistently obtained on petroleum-based charcoal tubes, and the minimu m detection limits were estimated as 0.2 and 0.5 ng per tube for the ( Z)- and (E)-isomers of 1,3-DCP, respectively, and 2.0 ng per tube for MITC. With the automated headspace method, sample preparation was simp lified and sample throughput was greatly enhanced, and up to 200 sampl es could be analyzed on a 24-h basis under the optimum conditions.