Abstract
A way to extract useful chemical information from the volatile profile provided by a headspace-mass spectrometer (HS-MS) is developed in order to improve sensitivity in HS-MS analysis. The methodology is based on the selection of a narrow window in the volatile profile where the signal-to-noise ratio was maximal by combining the data acquisition time and scan rate. To test this approach, benzene, toluene, ethylbenzene, and p-xylene (BTEX) as well as their mixtures were quantified in drinking waters. Individual hydrocarbons were determined between 1 and 30 μg/L (mean RSD, 4.0% for 10 μg/L) while mixtures were quantified at a microgram per liter level by using the partial least-squares multivariate algorithm with a relative standard prediction error of under 3.5%. These results indicate that the method proposed is useful as a sensitive and selective tool for the determination of BTEX and surpasses other reported HS-MS alternatives. In addition, the proposed methodology can be extended to others that insert analytes from a sample directly into a MS, such as membrane introduction mass spectrometry among others.

This publication has 12 references indexed in Scilit: