Adam Keil

Monitoring of toxic compounds in air using a handheld rectilinear ion trap mass spectrometer.

A miniature, handheld mass spectrometer, based on the rectilinear ion trap mass analyzer, has been applied to air monitoring for traces of toxic compounds. The instrument is battery-operated, hand-portable, and rugged. We anticipate its use in public safety, industrial hygiene, and environmental monitoring. Gaseous samples of nine toxic industrial compounds, phosgene, ethylene oxide, sulfur dioxide, acrylonitrile, cyanogen chloride, hydrogen cyanide, acrolein, formaldehyde, and ethyl parathion, were tested. A sorption trap inlet was constructed to serve as the interface between atmosphere and the vacuum chamber of the mass spectrometer. After selective collection of analytes on the sorbent bed, the sorbent tube was evacuated and then heated to desorb analyte into the instrument. Sampling, detection, identification, and quantitation of all compounds were readily achieved in times of less than 2 min, with detection limits ranging from 800 parts per trillion to 3 parts per million depending on the analyte. For all but one analyte, detection limits were well below (3.5-130 times below) permissible exposure limits. A linear dynamic range of 1-2 orders of magnitude was obtained over the concentration ranges studied (sub-ppbv to ppmv) for all analytes.

Chlorine and bromine atom ratios in the springtime Arctic troposphere as determined from measurements of halogenated volatile organic compounds

Received 24 January 2006; revised 19 April 2006; accepted 16 May 2006; published 7 September 2006. [1] The concentrations of a suite of halogenated volatile organic compounds (HVOCs) were measured near Barrow, Alaska, from January to April 2005. The HVOCs are produced from the reaction of bromine and chlorine atoms with ethene and propene. During periods of decreasing ozone concentration, increases in the HVOC concentrations allowed for the calculation of the ratio of bromine to chlorine radical concentrations, based on available kinetic data. We use these concentration data to interrogate the chemistry that results in tropospheric ozone depletion in the Arctic, the possible sources of ozone depleting halogen molecules, and the spatial scale in which ozone depletion occurs. We report calculated halogen atom concentration ratios ([Br]/[Cl]) during partial ozone depletion events. The concentration ratio was observed to range from 80 ± 30 to 990 ± 300 when ozone concentrations were above 15 ppb. These data make it clear that chlorine and bromine atom chemistry is active in the Arctic troposphere beginning at twilight, even absent large-scale ozone depletion, and that the sources of the chlorine atoms are poorly understood.

Facility monitoring of toxic industrial compounds in air using an automated, fieldable, miniature mass spectrometer

Gaseous samples of nine toxic industrial compounds (acrolein, acrylonitrile, carbon disulfide, cyanogen chloride, ethylene oxide, formaldehyde, hydrogen cyanide, phosgene, and sulfur dioxide) were detected, identified, and quantitated using a fully automated, fieldable, miniature mass spectrometer equipped with a glow discharge electron ionization source and a cylindrical ion trap mass analyzer. The instrument was outfitted with a combined direct air leak and dual thermal desorption tube inlet that allowed for continuous sampling of compounds with throughput times of 2 min or less. Most compounds showed a linear response over the concentration ranges studied (sub-parts per billion [ppb] to parts per million [ppm]). Sorbent tube limits of detection (20 ppb to 8 ppm for all analytes) were lower than those reported for the two compounds examined using direct leak (acrylonitrile 16 ppm and phosgene 500 ppb). All limits of detection were below the concentration at which the compound poses an immediate danger to life and health. Sensitivity, probability of true positives, and the false positive rate for each analyte were investigated and described using receiver operating characteristic (ROC) curves. High quality data with low false positive and negative rates are indicative of the good chemical specificity and sensitivity of the instrument. Complex matrices consisting of second-hand smoke, gasoline exhaust, diesel fuel exhaust, and multiple analytes were also studied. Detection limits for analytes generally increased in the mixtures, but analytes were still detected at concentrations as low as 100 ppb.