Research Projects

Project at a Glance

Title: Understanding primary organic aerosol volatility at atmospherically realistic concentrations for SIP analysis

Principal Investigator / Author(s): Kleeman, Michael

Contractor: UC Davis

Contract Number: 10-313

Research Program Area: Atmospheric Processes

Topic Areas: Modeling, Monitoring


This report explores the volatility of primary organic aerosol (POA) emitted from light duty gasoline-powered vehicles at atmospherically relevant concentrations using state of the science instrumentation. Existing POA partitioning models were tested through analysis of emissions changes that result from perturbations to the dilution air used during vehicle sampling. Measurements were re-interpreted with the application of new theoretical models that can be extended to regional-scale modeling applications.

Emissions from a fleet of 8 gasoline vehicles operated on the UC driving cycle were characterized by (1) thermal-optical measurements of filter-collected organic carbon and elemental carbon, (2) GC-MS analysis of denuder-filter-PUF samples, (3) high resolution Aerosol Mass Spectrometer (HR-AMS) measurements for organic mass and the bulk and elemental compositions of organic species, (4) Time-of-Flight Chemical Ionization Mass Spectrometer (ToF CIMS) measurements of gas-phase concentrations and elemental compositions of carbonyls, alcohols, ketones and organic acids, and (5) multi-wavelength photoacoustic spectrometer (PAS) measurements of black carbon (BC). Vehicle exhaust was diluted to atmospherically relevant concentrations as different features of the dilution air were perturbed: temperature (25-100C), relative humidity (55-85%), and background black carbon particles concentrations (0-255 μg m⁻).

The majority (~75-80%) of the particle phase emissions from the vehicle fleet were categorized as non-volatile elemental carbon (EC) that will not evaporate in the atmosphere. Real-time measurements show that the highest EC emissions occurred during the cold-start portion of the test and/or during periods of hard acceleration. The remaining fraction (~20-25%) of the particle phase emissions was POA that could be broadly classified as a semi-volatile material (similar to motor oil) or an effectively non-volatile material (hypothesized to be fuel combustion products). The POA emissions were generally more volatile during the cold start portion of the driving cycle and less volatile after the engine and exhaust system reached operating temperature. Emissions of semi-volatile and non-volatile organic carbon from different vehicles could not be predicted a-priori. Half of the tested vehicles in the current study had emissions dominated by motor oil while the other half of the vehicles had emissions dominated by fuel combustion products. Further tests are needed using the methodology developed in this report to develop fleet wide characterization of the emissions of both types of POA for use in future regional modeling applications.

The prevalence of carbonyl species in the POA suggests that these species may be basic building blocks that are transformed into non-volatile POA as the vehicle exhaust ages. Carbonyl emissions increased with humidity suggesting an aqueous production pathway. Increased levels of black carbon in the dilution air scavenged carbonyl precursors and reduced total carbonyl production rates. The AMS may categorize the low-volatility POA (possibly associated with carbonyls) as refractory material that is not reported. Emissions of carbonyl oxidation products (acids) peaked during the cold-start portion of the driving cycle and had a strong correlation to NOx emissions.


For questions regarding this research project, including available data and progress status, contact: Heather Choi at (916) 322-3893

Stay involved, sign up with ARB's Research Email Listserver