2005 Progress Report: Highly Time-Resolved Source Apportionment Techniques for Organic Aerosols Using the Aerodyne Aerosol Mass Spectrometer

EPA Grant Number: R832161
Title: Highly Time-Resolved Source Apportionment Techniques for Organic Aerosols Using the Aerodyne Aerosol Mass Spectrometer
Investigators: Jimenez, Jose-Luis , Hannigan, Michael P. , Schauer, James J. , Zhang, Qi
Institution: University of Colorado at Boulder , University of Wisconsin - Madison
EPA Project Officer: Chung, Serena
Project Period: December 1, 2004 through November 30, 2007 (Extended to November 30, 2008)
Project Period Covered by this Report: December 1, 2004 through November 30, 2005
Project Amount: $450,000
RFA: Source Apportionment of Particulate Matter (2004) RFA Text |  Recipients Lists
Research Category: Particulate Matter , Air Quality and Air Toxics , Air


The overall objective of this research project is to develop, validate, and apply fine particulate matter (PM) source apportionment techniques for measurements made with the Aerodyne Aerosol Mass Spectrometer (AMS). The AMS is the only current real-time instrument that provides quantitative size-resolved organic aerosol data with a time resolution of a few minutes. Prior results indicated that AMS organic aerosol data are sufficiently specific to address the critical need for source apportionment of organic aerosols with very high time-resolution. We expect a number of instruments being developed to improve organic detection specificity via chemical ionization and/or photoionization. This project focuses on AMS data but will provide the foundation for using other such data for source apportionment.

Various approaches for apportioning AMS organic are being investigated, including: (1) custom techniques that take advantage of our understanding of the data; (2) standard multivariate receptor models (e.g., UNMIX and positive matrix factorization [PMF]); and (3) advanced data mining techniques currently being developed by the University of Wisconsin (under National Science Foundation funding). All techniques will be tested with synthetic AMS data with several overlapping sources. We then will apply these methods to the well characterized AMS datasets from the Pittsburgh, New York City, and Houston U.S. Environmental Protection Agency (EPA) Supersites. A new field campaign will be carried out (using three AMSs) with two objectives, which are to: (1) compare with the well-established chemical mass balance (CMB) method from collocated organic molecular marker data; and (2) demonstrate the improvements in organic source apportionment from three new techniques designed to improve the sensitivity and selectivity of the AMS for organic aerosols: a time-of-flight mass spectrometer (ToF-MS; replacing the quadrupole used in the standard AMS), low temperature vaporization, and thermal denuding. The primary result of this project will be to demonstrate and validate source apportionment of organic aerosols with very high time-resolution using AMS data. The results of this project can have a rapid and broad impact, because the techniques developed here can be applied to datasets acquired by many researchers around the world, including the more than 30 research groups with an AMS. In addition, these techniques and algorithms also will provide the foundation for source apportionment using data from emerging and future quantitative AMSs.

Progress Summary:

This project has two major components: (1) the development and application of receptor model techniques to AMS data; and (2) the field deployment of several new techniques aimed at obtaining more chemically specific information from the AMS for source apportionment purposes. Our work to date is meeting successfully the original goals of the project, as summarized below for each of the two major components.

Progress in Receptor Model Development and Application

During Year 1 of this project, a new data analysis technique has been developed to deconvolve and quantify hydrocarbon-like and oxygenated organic aerosol (HOA and OOA, respectively) using highly time resolved organic mass spectral data obtained with AMS. This technique was applied successfully to the AMS data acquired at the EPA Pittsburgh Supersite in September 2002. The mass concentrations, temporal variations, size distributions, and mass spectra of HOA and OOA have been resolved, and a detailed analysis of this information yields valuable insights into the sources and processes of atmospheric organic aerosols. We are in the process of applying this technique to more than 30 highly time-resolved AMS datasets acquired from 12 urban locations and 11 rural and remote sites representative of high elevation, forested, pristine, and continentally influenced marine atmospheres. Most of the sites are located in the Northern Hemisphere’s midlatitudes. Preliminary results on these analyses have been presented at major national and international conferences.

In addition to the work on custom source apportionment techniques just described, we are pursuing the application of two other types of techniques to AMS data:

(1) Graduate students from the Jimenez and Hannigan group (Ingrid Ulbrich and Greg Brinkman) have developed a procedure to automate the running of the PMF and other source apportionment models and the detailed display and analysis of their results. A synthetic data option allows investigators using synthetic data to include the synthetic source profiles and time series on the model output factor profiles and time series to quickly observe the results. So far PMF has been implemented. The code using Igor software automatically generates the most useful plots to observe the results of the model and diagnostic statistics that will help investigators determine the best input parameters to the model. This tool currently is being tested with the Pittsburgh Supersite AMS (PA) and Pittsburgh AMS synthetic (PAsyn) datasets. PMF is being run for a range of the potential input parameters, storing the important results for each run. The user then can observe these results by varying the input parameters graphically.

(2) The application of advanced data mining techniques developed at the University of Wisconsin to AMS data is starting, focusing first on the PA and PAsyn datasets, which also are being used to characterize the other techniques.

Progress in Acquiring and Analyzing More Chemically Specific Field Data

Two major accomplishments of Year 1 were the SOAR-1 and SOAR-2 (Study of Organic Aerosol in Riverside, phases 1 & 2) field experiments. These studies were organized by ProfessorsJimenez and Ziemann of the University of California (UC)-Riverside, with participation from ProfessorSchauer and Dr. Hannigan, to characterize organic aerosols with a variety of techniques. They were carried out at UC-Riverside, which is located in a polluted region downwind of Los Angeles that is heavily impacted by both primary and secondary aerosol. Although it was originally intended that this study would include only the UC-Riverside, the University of Colorado, and the University of Wisconsin groups, many other research groups became interested in making measurements with a variety of complementary methods during this period. As a result, during the SOAR-1 study (July 15–August 15, 2005), approximately 60 scientists from 17 universities, research institutes, and companies participated in what is probably the most complete analysis of organic aerosols performed to date. During SOAR-2 (November 1-24, 2005), about 20 scientists from 8 groups participated. More detailed information on the participants and measurements can be found at: http://cires.colorado.edu/jimenez-group/Field/Riverside05/.

During SOAR-1, the Jimenez group deployed four new instrument combinations for organic aerosol analysis:

  1. A thermal denuder system was used in front of an AMS and a scanning mobility particle sizer to characterize the coupled chemistry-volatility profiles.
  2. A new time-of-flight aerosol mass spectrometer (ToF-AMS) was used to characterize the full-size-composition space with high time resolution.
  3. A new high-resolution time-of-flight aerosol mass spectrometer (HR-ToF-AMS) was deployed in the field for the first time to provide additional insight on the aerosol chemical composition by allowing the direct determination of the elemental composition of every peak in the spectrum of the AMS.
  4. A variable-temperature AMS vaporizer was used some in both the ToF-AMS and HR-ToF-AMS to provide additional insight on the thermal properties of the organic aerosol.

These systems were used to continuously analyze ambient particles, and to analyze secondary organic aerosol (SOA) formed in five reactions [pentadecane + OH/NOx, α-pinene + O3, 3-methyl-2-butenal + OH/NOx, toluene + OH/NOx, gasoline + OH/NOx] carried out in the environmental chamber in the Ziemann laboratory. During SOAR-2, the Jimenez group deployed the HR-ToF-AMS and the thermal denuder for ambient sampling for 3 weeks and carried out several additional chamber experiments.

During SOAR-1, Professor Schauer’s group: (1) collected 24-hour and 5-8 hour samples, which are starting to be analyzed by gas chromatography-mass spectrometry (GC-MS) and other advanced techniques (high resolution GC-MS and liquid chromatography-mass spectrometry (LC-MS)); and (2) deployed a real-time elemental carbon/organic carbon (EC/OC) analyzer that provided continuous data for most of the project. Dr. Hannigan’s group deployed a Micro Orifice Uniform Deposit Impactor (MOUDI) during SOAR-1. Although an AMS provides real time analysis of aerosol size and composition, the MOUDI impactor requires post-sampling mass concentration and ion species analyses. During SOAR-1, a MOUDI and an AMS were run simultaneously. Gravimetric and ionic species analyses of the MOUDI collection surfaces, or substrates, are being performed using a precision balance and an ion chromatograph, respectively. These data then will be compared to the real time data generated by the AMS. Comparing the mass concentration and particle size distribution results of these measurement tools should provide a greater understanding of their operating characteristics and biases of the different chemical characterization tools.

In addition, a $15,000 supplement to this grant funded Professor Weber of Georgia Tech for deploying two organic aerosol analysis techniques during SOAR-1: (1) real-time water-soluble organic carbon (WSOC) analyzer and (2) filter collection followed by postanalysis of the acidic/neutral/basic fractions of the organic aerosol. Both instruments successfully were deployed during the field study, and analysis of the filter samples is ongoing.

Other groups that participated in SOAR at no cost to this grant, and whose data for organic aerosols will be intercompared with the above methods, include the group of Professor Ziemann at UC-Riverside (thermal desorption particle beam mass spectrometer), the group of ProfessorEatough at Brigham Young University (new EC/OC analyzer including a measurement of semi-volatile OC, filter dynamics measurement system tapered element oscillating microbalance, and several other instruments), the group of Professor Goldstein at UC-Berkeley (thermal desorption aerosol GC-MS [TAG] and several gas-phase instruments), the group of Dr. Doug Worsnop at Aerodyne Research (two soft ionization ToF-AMSs), the group of Professor Sioutas at the University of Southern California (versatile aerosol concentration enrichment system aerosol concentrator), the group of Suzanne Hering at Aerosol Dynamics (TAG and water condensation particle counter), the group of Professor Kim Prather at UC-San Diego (three aerosol time-of-flight mass spectrometers and several other gas-phase and particle instruments), the group of Professors Janet Arey and Roger Atkinson at UC-Riverside (filter + GC-MS analysis for polyaromatic hydrocarbons [PAHs], Nitro-PAHs, and PAH reaction products), the group of Dr. Dennis Fitz at UC-Riverside (evaluation of PM bulk sampling artifacts, under a separate EPA Science To Achieve Results grant), the group of Phil Hopke at Clarkson University (sampler for chemical analysis of oligomers), the group of Mark Thiemens at UC-San Diego (sulfate and nitrate isotope measurement), the group of Suzanne Paulson at UC-Los Angeles (oxidants in particles), the group of Rafael Villalobos-Pietrini at the National Autonomous University of Mexico (PAH analysis), and the group of Professor Seinfeld/Flagan (ambient sampling at Caltech with a ToF-AMS and several other instruments that provide a spatial point of comparison to Riverside).

Preliminary data and insights were exchanged during a data analysis meeting held during the 2005 American Association for Aerosol Research Conference in Austin, Texas, with participation from 10 of these groups. This rich combined dataset, including many state-of-the-art techniques for organic aerosol analysis, will provide insights on the relationship between the techniques and the nature and transformations of organic aerosols in urban areas. Its analysis is expected to provide improved data for understanding organic aerosol formation mechanisms and sources and for use in source apportionment and airshed modeling (e.g., Professors Griffin and Dabdub of the University of New Hampshire and UC-Irvine recently have submitted a proposal to carry out airshed modeling of the SOAR-1 period). Such data and models can aid in the evaluation of the effects of fine particulate matter on human health and the environment, and can be used to develop more accurate air pollution models. This will allow for more efficient targeting of aerosol sources in pollution control strategies.

Future Activities:

Future Work in Receptor Model Development and Application

The plans for Year 2 in the area of receptor model development and application are to: (1)further develop and validate new custom data analysis algorithms that allows the deconvolution and quantification of more than two (i.e., HOA and OOA) physically meaningful components using the unit mass resolution electron impact AMS data; and (2) apply the HOA/OOA deconvolution technique and potentially the new technique(s) to multiple urban, rural, and remote locations, including three EPA Supersites: Pittsburgh, New York City, and Houston; and (3) evaluate the results from the newly developed PMF interface to ambient and synthetic AMS data. A manuscript is planned for submission by the end of the year describing this work. The implementation of UNMIX and/or other source apportionment models into this framework will be pursued in Year 3 of the project. In addition, two manuscripts are being prepared for submission. One paper will present the development and validation of the new data analysis algorithm that allows the deconvolution and quantification of four types of organic aerosols using the Houston EPA Supersite data: urban combustion-related hydrocarbon-like, highly oxygenated OOA, biomass burning aerosols, and fluorinated organic aerosols. A second paper will present the major findings and comparisons of results regarding the characteristics, sources, and processes of organic aerosols in more than 25 locations around the Northern Hemisphere.

Future Work in Acquiring and Analyzing More Chemically-Specific Field Data

After the successful completion of the SOAR-1 and SOAR-2 field experiments, the focus of this area has shifted to the analysis of the enormous database collected, which is proceeding in collaboration with several of the groups involved. In particular, the results from source apportionment using the ToF-AMS data will be compared with those from CMB analysis of the GC-MS data from filter samples, and the receptor models described above will be applied to these datasets.

The only additional field experiment planned for this project will take place in June 1-10, 2006, at the Missoula, Montana, Biomass Burning Chamber operated by the U.S. Department of Agriculture. The objective of this deployment is to obtain signatures of biomass burning aerosols using the HR-ToF-AMS and the thermal denuder. This type of aerosol is estimated to make major contributions to organic aerosol concentrations worldwide, but profiles of its composition with the HR-ToF-AMS are lacking. In addition, laboratory experiments also will be carried out at the University of Colorado with pure-component organic particles to characterize the contributions of various species and organic functional groups to different peaks in the HR data. The eventual goal, once enough profiles have been assembled, is to perform CMB receptor modeling using HR-ToF-AMS ambient data.

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

Other project views: All 115 publications 31 publications in selected types All 31 journal articles
Type Citation Project Document Sources
Journal Article Takegawa N, Miyakawa T, Kondo Y, Jimenez JL, Zhang Q, Worsnop DR, Fukuda M. Seasonal and diurnal variations of submicron organic aerosol in Tokyo observed using the Aerodyne aerosol mass spectrometer (AMS). Journal of Geophysical Research-Atmospheres 2006;111(D11):D11206. R832161 (2005)
R832161 (2006)
R832161 (2007)
R832161 (Final)
  • Full-text: Wiley-Full-text PDF
  • Abstract: Wiley-Abstract & Full-text HTML
  • Journal Article Zhang Q, Worsnop DR, Canagaratna MR, Jimenez JL. Hydrocarbon-like and oxygenated organic aerosols in Pittsburgh: insights into sources and processes of organic aerosols. Atmospheric Chemistry and Physics 2005;5(12):3289-3311. R832161 (2005)
    R832161 (2006)
    R832161 (2007)
    R832161 (Final)
    R831080 (Final)
  • Full-text: ACP-Full Text PDF
  • Abstract: ACP-Abstract
  • Other: Hal Archives-Full Text PDF
  • Journal Article Zhang Q, Alfarra MR, Worsnop DR, Allan JD, Coe H, Canagaratna MR, Jimenez JL. Deconvolution and quantification of hydrocarbon-like and oxygenated organic aerosols based on aerosol mass spectrometry. Environmental Science & Technology 2005;39(13):4938-4952. R832161 (2005)
    R832161 (2006)
    R832161 (2007)
    R832161 (Final)
    R831080 (Final)
  • Abstract from PubMed
  • Full-text: SUNY Albany-Full Text PDF
  • Abstract: ES&T-Abstract
  • Other: ES&T-Full Text PDF
  • Supplemental Keywords:

    ambient air, tropospheric, air pollution, particulates, environmental chemistry, monitoring, carbonaceous particles, combustion aerosols, source apportionment, primary organic aerosols, secondary organic aerosols, ToF-AMS, thermal denuder, particulate matter mass, particulate organic carbon,, RFA, Scientific Discipline, Air, Ecosystem Protection/Environmental Exposure & Risk, particulate matter, Air Quality, Environmental Chemistry, Monitoring/Modeling, Environmental Monitoring, Environmental Engineering, particulate organic carbon, atmospheric dispersion models, atmospheric measurements, model-based analysis, time resolved apportionment, source apportionment, chemical characteristics, emissions monitoring, environmental measurement, airborne particulate matter, air quality models, air quality model, air sampling, speciation, particulate matter mass, analytical chemistry, aerodyne aerosol mass spectrometry, monitoring of organic particulate matter, modeling studies, chemical transport models, real-time monitoring, aerosol analyzers, chemical speciation sampling, particle size measurement

    Relevant Websites:

    http://cires.colorado.edu/jimenez Exit
    http://cires.colorado.edu/jimenez-group/Field/Riverside05/ Exit
    http://cires.colorado.edu/jimenez-group/AMSsd/ Exit
    http://www.engr.wisc.edu/cee/faculty/schauer_james.html Exit
    http://spot.colorado.edu/~hannigan/ Exit
    http://www.asrc.cestm.albany.edu/qz/ Exit

    Progress and Final Reports:

    Original Abstract
  • 2006 Progress Report
  • 2007 Progress Report
  • Final Report