Final Report: Application of Thermal Desorption GCMS (TD-GCMS) for the Analysis of Polar and Non-Polar Semi-Volatile and Particle-Phase Molecular Markers for Source Attribution

EPA Grant Number: R831088
Title: Application of Thermal Desorption GCMS (TD-GCMS) for the Analysis of Polar and Non-Polar Semi-Volatile and Particle-Phase Molecular Markers for Source Attribution
Investigators: Schauer, James J. , Sheesley, Rebecca J. , Simoneit, Bernd R.T.
Institution: University of Wisconsin - Madison
EPA Project Officer: Chung, Serena
Project Period: January 1, 2004 through December 31, 2006 (Extended to December 31, 2007)
Project Amount: $449,687
RFA: Measurement, Modeling, and Analysis Methods for Airborne Carbonaceous Fine Particulate Matter (PM2.5) (2003) RFA Text |  Recipients Lists
Research Category: Air , Air Quality and Air Toxics , Particulate Matter

Objective:

The overall goal of the proposed project is to fully develop, validate, and employ a cost-effective thermal desorption gas chromatography mass spectrometry (TD-GC-MS) technique for the analysis of semi-volatile and particle-phase organic compounds, which can be applied to both atmospheric and source samples. Four project objectives were defined in the original proposal:

  1. Standardize and validate the TD-GC-MS analyses of particulate matter samples that employ direct silylation derivatization compatible with both polar and non-polar compounds.
  2. Configure a low cost particle-phase and semi-volatile organic compound sampling system that is compatible with TD-GC-MS analysis for molecular marker speciation, and can be used for high frequency, low sample volume sampling networks.
  3. Employ the new TD-GC-MS analysis for the organic compound speciation of the remaining 530 daily PM2.5 samples collected during the 2 year St. Louis Supersite project (100 samples per year are being analyzed as part of the Supersite project by the Project PI).
  4. Collect and analyze by these TD-GC-MS techniques new sets of 4 hour semi-volatile and particle-phase organic compound samples at a network of sites during air pollution episodes that can support air quality models currently being developed at UC-Davis.

Although the overall goal of the project was fulfilled for particle-phase organic compound analysis during the course of funding, the specific project objectives have been modified since the onset of the project. The technologies for semi-volatile ambient sampling, including XAD-coated quartz fiber filters, did not become readily available during the course of the project and the direct silylation method required more development time than originally anticipated. The focus of the detailed project objectives shifted to the following:

  1. Standardize and validate 3 different methods for the TD-GCMS analysis of particulate matter samples: non-polar, methylation, and silylation.
  2. Employ the standardized non-polar TD-GCMS analysis for the organic compound speciation of the remaining 300 daily PM2.5 samples collected during Year 1 of the 2 year St. Louis Supersite project (this includes duplicate analysis as 100 samples per year were analyzed as part of the Supersite project by the project PI).
  3. Analyze a set of every 6-hour PM2.5 samples collected in Riverside, CA as part of the SOAR project (July-Aug 2005) using the newly developed in-situ methylation TD-GCMS method
  4. Analyze a set of every 6 hour PM2.5 samples collected in Fresno, CA (Feb 2007) using the newly developed in-situ silylation TD-GCMS method. This data set has been supplemented with non-polar organic speciation data from solvent extraction GCMS and will be included in air quality modeling efforts by UC-Davis.

Summary/Accomplishments (Outputs/Outcomes):

  1. Standardize and validate 3 different methods for the TD-GCMS analysis of particulate matter samples: non-polar, methylation, and silylation. Three different TD-GCMS methods were developed and validated by the UW-Madison research group. Development of the non-polar method was initiated prior to the onset of this STAR project, but the method was standardized and validated during the course of the STAR project. The two in-situ derivatization TD-GCMS techniques, methylation and silylation, were fully developed during the course of the project.
  2. Employ the standardized non-polar TD-GCMS analysis for the organic compound speciation of the remaining 300 daily PM2.5 samples collected during Year 1 of the 2 year St. Louis Supersite project (this includes duplicate analysis as 100 samples per year were analyzed as part of the Supersite project by the project PI). The non-polar TD-GCMS method was employed to analyze 300 samples from the St. Louis Supersite (May 2001-April 2002). Included in this set was the analysis of 2 months of samples which had been previously analyzed by solvent extraction GCMS by UW-Madison research group. The first month of duplicate analysis was used for internal, initial method development, while the second month of duplicate analysis was used for a formal intercomparison between the solvent extraction and TD-GCMS methods. A paper is in press which details the method validation, including this intercomparison, and discusses the results of the full year study of non-polar organic tracers including hopanes, steranes, PAH and alkanes (Sheesley, R.J., Schauer, J.J., Meiritz, M., DeMinter, J.T., Bae, M.S., Turner, J.R. Daily variation in particle-phase source tracers in an urban atmosphere. Aerosol Science and Technology, 2007, 41, 981-993.). The abstract for this paper has been included below.
  3. One full year of daily 24-hour fine particulate matter samples collected in East St. Louis, IL at the EPA funded St. Louis Midwest Supersite were analyzed for organic carbon (OC), elemental carbon (EC) and non-polar organic tracers including polycyclic aromatic hydrocarbons (PAH(s)), hopanes, and alkanes. Two different analytical methods were used for analysis, solvent extraction gas chromatography/mass spectrometry (GCMS) and thermal desorption GCMS (TD-GCMS). The TD-GCMS method was equivalent to the solvent extraction GCMS method for key molecular markers. Select PAH(s) and alkanes were found to have extreme events within the annual study, which had daily 24-hour concentrations that were 10 to 22 times higher than the annual average daily concentration. The OC and EC maxima were only 3 to 5 times higher than the annual average. To further assess the impact of point sources and to evaluate the compatibility of the two organic speciation methods, the six potential every sixth day annual averages were calculated and compared. The extreme concentration days were large enough, in the case of benzo[a]pyrene, to make every sixth day analysis not representative of the true annual average even with events greater than the 99th percentile of the annual distribution removed. The final analysis calculated day of the week averages for select representative organic tracers and revealed that gasoline motor vehicle tracers and OC do not have a distinct day of the week trend. EC, believed to be largely impacted by diesel exhaust, had a midweek concentration peak. These trends cannot necessarily be extrapolated to other urban areas or other events.

    In addition to the yearlong daily trend study at the St. Louis Midwest Supersite, the non-polar TD-GCMS method was also applied to a personal exposure study at a trucking terminal in St. Louis, MO. A paper has been accepted to the Journal of Exposure Science and Environmental Epidemiology (Sheesley, R.J., Schauer, J.J., Garshick, E., Laden, F., Smith, T.J., Blicharz, D., DeMinter, J.T. Tracking personal exposure to particulate diesel exhaust in a diesel freight terminal using organic tracer analysis, 2008, in press). The abstract has been included below.

    Personal exposure to particle-phase molecular markers was measured at a trucking terminal in St. Louis, MO as part of a larger epidemiologic project aimed at assessing carbonaceous fine particulate matter (PM) exposure in this occupational setting. The integration of parallel personal exposure, ambient worksite area and ambient urban background (St. Louis Supersite) measurements provided a unique opportunity to track the work-related exposure to carbonaceous fine PM in a freight terminal. The data was used to test the proposed personal exposure model in this occupational setting:

    Personal exposure = urban background + work site background + personal activity

    To accurately assess the impact of PM emission sources, particularly motor vehicle exhaust, elemental and organic carbon (OCEC) analysis and nonpolar organic molecular marker analysis by thermal desorption gas chromatography/mass spectrometry (TD GCMS) were conducted on all of the PM samples. EC has been used as a tracer for diesel exhaust in urban areas; however, the emission profile for diesel exhaust is dependent upon the operating conditions of the vehicle and can vary considerably within a fleet. Hopanes, steranes, polycyclic aromatic hydrocarbons and alkanes were measured by TD GCMS. Hopanes are source-specific organic molecular markers for lubricating oil present in motor vehicle exhaust. The concentrations of OC, EC and the organic tracers were averaged to obtain average profiles to assess differences in the personal, work site area and urban background samples and were also correlated individually by sample time to evaluate the exposure model presented above. Finally, a chemical mass balance model was used to apportion the motor vehicle and cigarette smoke components of the measured OC and EC for the average personal exposure, worksite area and urban background samples.

  4. Analyze a set of every 6 hour PM2.5 samples collected in Riverside, CA as part of the SOAR project (July-Aug 2005) using the newly developed in-situ methylation TD-GCMS method. The SOAR samples were collected in Riverside, CA in the summer of 2005 with 24 hour and higher time-resolved 6-hour samples. Samples were collected on 90mm quartz fiber filters at a flow rate of 92 lpm. A single 1.45 cm2 punch was used for this analysis, which conserved most of the filter for future analysis. The initial analysis used the 24-hour samples and provided consistent nonpolar data with a volatility-limited list of polar acids. It is often helpful to compare results roughly with previous studies in the same region. The hopanes are in the same range as measured in the LA area by Fine et al (Fine et al., 2004) and the reported acids (methyl phthalic and azelaic) are in the same range as reported in a central California study (Rinehart et al., 2006). This data illustrates that the nonpolar compounds continue to be measured in a rigorous manner even when combined with the diazomethane derivatization. Additionally, the day of the week information has been included in the x-axis to illustrate the potential utility of this data in assessing weekday/weekend trends; although the data is reported in ambient concentration and has not yet been normalized by organic carbon concentration.
  5. Additional developmental work was done with the diazomethane method to improve recovery and quantification of aliphatic diacids, which are higher volatility. The GC method was modified to enhance detection of these more volatile compounds. One week of 6 hour SOAR samples have been analyzed using the improved diazomethane method to focus on time of day fluctuations in the full list of polar acids and nonpolar compounds. The 24-hour SOAR samples quantified azelaic and sebacic acids (nonanedioc and decanedioic acids) while the modified method allowed quantification of aliphatic diacids with molecular weights as low as adipic acid (hexanedioic acid). Measurement of acidic compounds such as the aliphatic and aromatic diacids combined with nonpolar tracers such as hopanes are essential to understanding the sources and secondary processing occurring in ambient particulate matter. The results of the SOAR study will be published as part of a method development paper on the diazomethane TD-GCMS method and are also being used as part of an intercomparison study with in-situ GCMS analysis collocated at the SOAR site.

    Figures A and B show the results of the high time resolution SOAR data for aromatic acids. A distinct diurnal trend is noted. A manuscript is in preparation for submission for publication that covers this work and should be submitted during the spring of 2008.

  6. Analyze a set of every 6 hour PM2.5 samples collected in Fresno, CA (Feb 2007) using the newly developed in-situ silylation TD-GCMS method. This data set has been supplemented with non-polar organic speciation data from solvent extraction GCMS and will be included in air quality modeling efforts by UC-Davis.
  7. The silylation TD GCMS method has made significant progress in Year 3. It had been mentioned in previous progress reports that the method was not ideal for large projects due to instrument wear and tear resulting from high inputs of the silylation reagent. Much time was devoted to method development to limit the amount of reagent in the system, including dilution of the reagent and increased instrument cleaning and maintenance. Additionally, the Markes TD unit has been modified and updated with cooperation from Markes, which has made it more durable for high temperature particulate matter analysis and conducive to repeated silylation analysis. The silylation method is being developed as a separate stand-alone method for the analysis of select polar compounds including levoglucosan, sterols, monoglycerides, and simple carbohydrates.

    In-situ silylation TD-GCMS method. A Markes International Thermal Desorption Unit (Model M-10140) (Foster City, CA, USA) coupled with an Agilent Technologies 5973 GC-MS was used for the sample analysis.  A single filter punch (1.0 cm2) was used in the analysis which had 1-7 μg OC. Glassware was baked at 550°C and solvent rinsed before use. Forty μl of 1:5 diluted BSTFA silylation reagent (N,O-bis(trimehtylsilyl)trifluoroacetamide plus trimethylchlorosilane from Supelco) was added to a clean vial which had a needle fitted into the cap to suspend the filter punch. The filter sample was placed on the needle and then spiked with an isotopically labeled internal standard which contains C13 –labeled levoglucosan and deuterated cholesterol. After the vial was closed the filter was wetted with the diluted reagent and then the vial was heated at 70°C for 1 hour. The filter punch was not allowed to come to dryness. Finally, the filter was removed from the vial and inserted into a glass desorption tube and placed into the autosampler. The sample tube was ramped to 360°C over 20 minutes to desorb the compounds of interest from the filter. A high boiler quartz focusing trap designed for analysis of n-hexane to n-tetracontane semi-volatile organic compounds (0°C) was used to focus the sample before the temperature was ramped to 360°C again and desorbed onto the GC column; this focusing step concentrated the analytes desorbed from the filter into a smaller volume of vapor which improved detection in the GCMS.

    A 4-point calibration curve was run at the start of each sample set. The quantification standard included Levoglucosan, glucose, sucrose, fructose and 5 sterols that were used to quantify the corresponding compounds within the filter samples. The quantification standards were spiked onto a blank filter punch and the solvent was allowed to evaporate before continuing with the derivatization procedure detailed above; this was done to more closely parallel the particulate matter samples.

    Method validation included the analysis of duplicate ambient samples, matrix spikes (standard spiked onto ambient samples to assess recovery) and check standards (standards run between calibration curves to assess the stability of the instrument). The method validation results indicate that the analysis is consistent and reproducible, particularly for measurement of levoglucosan, which was the primary motivation for this method.

    Fresno study. Samples collected in Fresno, CA during the winter 2007 season were analyzed by the silylation method. Four, 6 hour samples were collected each day using a 92 lpm medium volume sampler equipped with a 90mm quartz fiber filter. The simple carbohydrate concentrations were quite low for this sampling set, but illustrate the low detection limits that can be achieved with the method (Figure C). Levoglucosan concentrations were at the other extreme; these numbers indicate the unique condition of very high wood smoke contribution in the San Joaquin valley in the winter time (Figure 1D).

    Finally, an intercomparison study was conducted for the entire Fresno sample set, with parallel analysis of levoglucosan by solvent extraction GCMS (n=40). This intercomparison parallels previous validation efforts for the non-polar TD-GCMS.

    In summary, the in-situ silylation TD-GCMS method has been proven to be consistent, reproducible and compatible with the traditional solvent extraction GCMS method. Accelerated instrument wear and tear continues to be the primary obstacle, which prevents this method from being more widely applicable. However, the method has been shown to be robust for the analysis of smaller sample sets with very low mass requirements (only 1-7 μg of OC per sample).

    A manuscript covering this work in preparation and should be submitted in spring 2008.

Figure A. Ambient concentrations of aromatic acids measured by methylation-TD-GCMS in 6 hour PM2.5 samples collected during the SOAR.

Figure A. Ambient concentrations of aromatic acids measured by methylation-TD-GCMS in 6 hour PM2.5 samples collected during the SOAR. 2005 campaign.

Figure B. Ambient concentrations of aromatic acids measured by methylation-TD-GCMS in 6 hour PM2.5 samples collected during the SOAR.

Figure B. Ambient concentrations of aromatic acids measured by methylation-TD-GCMS in 6 hour PM2.5 samples collected during the SOAR. 2005 campaign.

Figure C. Filter blank subtracted ambient concentrations of sugars measured in Fresno, CA in 2007.

Figure C. Filter blank subtracted ambient concentrations of sugars measured in Fresno, CA in 2007.

Figure D. Filter blank subtracted ambient concentration of Levoglucosan measured in Fresno, CA in 2007.

Figure D. Filter blank subtracted ambient concentration of Levoglucosan measured in Fresno, CA in 2007.

References:

Fine, P. M., B. Chakrabarti, M. Krudysz, J. J. Schauer and C. Sioutas (2004). "Diurnal variations of individual organic compound constituents of ultrafine and accumulation mode particulate matter in the Los Angeles basin." Environmental Science & Technology 38(5): 1296-1304.

Rinehart, L. R., E. M. Fujita, J. C. Chow, K. Magliano and B. Zielinska (2006). "Spatial distribution of PM2.5 associated organic compounds in central California." Atmospheric Environment 40(2): 290-303.


Journal Articles on this Report : 3 Displayed | Download in RIS Format

Other project views: All 9 publications 3 publications in selected types All 3 journal articles
Type Citation Project Document Sources
Journal Article Sheesley RJ, Schauer JJ, Garshick E, Laden F, Smith TJ, Blicharz AP, DeMinter JT. Tracking personal exposure to particulate diesel exhaust in a diesel freight terminal using organic tracer analysis. Journal of Exposure Science & Environmental Epidemiology 2009;19(2):172-186. R831088 (Final)
  • Full-text from PubMed
  • Abstract from PubMed
  • Associated PubMed link
  • Abstract: Nature-Abstract
    Exit
  • Journal Article Sheesley RJ, Deminter JT, Meiritz M, Snyder DC, Schauer JJ. Temporal trends in motor vehicle and secondary organic tracers using in situ methylation thermal desorption GCMS. Environmental Science & Technology 2010;44(24):9398-9404. R831088 (Final)
    R831080 (Final)
    R832161 (Final)
  • Abstract from PubMed
  • Full-text: ES&T-Full Text HTML
    Exit
  • Abstract: ES&T-Abstract
    Exit
  • Other: ES&T-Full Text PDF
    Exit
  • Journal Article Sheesley RJ, Mieritz M, DeMinter JT, Shelton BR, Schauer JJ. Development of an in situ derivatization technique for rapid analysis of levoglucosan and polar compounds in atmospheric organic aerosol. Atmospheric Environment 2015;123(Part A):251-255. R831088 (Final)
    R832276 (Final)
  • Full-text: ScienceDirect-Full Text HTML
    Exit
  • Abstract: ScienceDirect-Abstract
    Exit
  • Other: ScienceDirect-Full Text PDF
    Exit
  • Supplemental Keywords:

    Organic analysis, organic aerosols, thermal desorption GCMS,, RFA, Scientific Discipline, Air, Ecosystem Protection/Environmental Exposure & Risk, particulate matter, Air Quality, air toxics, Environmental Chemistry, climate change, Monitoring/Modeling, Environmental Monitoring, Engineering, Chemistry, & Physics, Environmental Engineering, carbon aerosols, air quality modeling, atmospheric particulate matter, health effects, aerosol particles, atmospheric particles, mass spectrometry, human health effects, PM 2.5, analysis of organic particulate matter, air modeling, air quality models, air sampling, gas chromatography, thermal desorption, carbon particles, air quality model, emissions, particulate matter mass, human exposure, particle phase molecular markers, aersol particles, particle dispersion, aerosol analyzers, measurement methods

    Relevant Websites:

    http://cires.colorado.edu/jimenez-group/Field/Riverside05/ Exit

    Progress and Final Reports:

    Original Abstract
  • 2004 Progress Report
  • 2005 Progress Report
  • 2006 Progress Report