You are here:
HIGH VOLUME INJECTION FOR GCMS ANALYSIS OF PARTICULATE ORGANIC SPECIES IN AMBIENT AIR
Citation:
TURLINGTON, J., D. A. OLSON, J. VOLCKENS, L. STOCKBURGER, AND S. R. MCDOW. HIGH VOLUME INJECTION FOR GCMS ANALYSIS OF PARTICULATE ORGANIC SPECIES IN AMBIENT AIR. Presented at American Association of Aerosol Research 2005, Atlanta, GA, February 07 - 11, 2005.
Impact/Purpose:
1) Develop a TOA protocol that meets the basic assumptions of the method and peer review of the method, 2) understand the role of sampling, analysis, and blank subtraction on the differences between the IMPROVE and STN OC and EC results 3) Evaluate the use of light absorbance methods to estimate EC on human exposure samples, 4) develop a sensitive analytical method suitable for analysis of STN and micro-environmental samples for organic molecular markers, and 5) develop an optimum list of organic molecular markers for source apportionment of particulate matter.
Description:
Detection of organic species in ambient particulate matter typically requires large air sample volumes, frequently achieved by grouping samples into monthly composites. Decreasing the volume of air sample required would allow shorter collection times and more convenient sample collection, potentially resulting in more distinctive source mixtures and expansion of source apportionment capabilities to micro-environmental applications. Most previous efforts for GCMS analysis of airborne particulate matter used splitless injection, which limits sensitivity by permitting injection of only a small fraction of the total extract.
The Human Exposure and Atmospheric Sciences Division recently developed Organic Aerosol Laboratory has completed characterization of a high volume injection method with increases analytical sensitivity by allowing injection of a considerably larger fraction of the extract. Samples are extracted under pressure using a Dionex Accelerated Solvent Extractor in 1:1:1 pentane:dichloromethane:acetone for measurement of both polar and non-polar compounds. Extracts are concentrated and introduced into an Agilent 6890 GC with a 5973 MS Detector by injection of 100 microliters with a Gerstel Programmable Temperature Vaporization Injector. Reproducibility of five point calibration curves run in triplicate have been demonstrated for eleven n-alkanes, three hopanes, and seven polycyclic aromatic hydrocarbons. Linear correlation coefficients for individual calibration curves for hopanes and polycyclic aromatic hydrocarbons in all cases were r2 > 0.999 and for n-alkanes in all cases were r2 > 0.998. Relative standard deviations in response factors without exception were under 3% for concentrations greater than 150 picograms per microliter under 20% for concentrations between 5 and 150 picograms per microliter. Although high signal-to-noise ratios were observed at concentrations as low as 1.5 picograms per microliter, precision between calibration curves was poor. These results indicate a capability of determining concentrations with good precision for all compounds at concentrations as low as 50 picograms per cubic meter even for a 24-hour 10 liter per minute sample.
Although this work was reviewed by EPA and approved for publication, it may not reflect official Agency policy.