Formation of secondary organic material from gaseous precursors in wood smoke.

The apportionment of the contribution of wood smoke emitted particles to the total concentration of particulate matter in a region has been greatly aided by the development of new analytical methods. These analytical methods quantitatively determine organic marker compounds unique to wood combustion such as levoglucosan and dehydroabietic acid. These markers have generally been determined in 24-hour averaged samples. We have developed an instrument based on the collection of particles on an inert filter, desorption of the organic material in an inert atmosphere with subsequent GC separation and MS detection of the desorbed compounds. The GC-MS Organic Aerosol Monitor (OAM) instrument has been used in three field studies. An unexpected finding from these studies was the quantification of the contribution of secondary organic aerosols from gases present in wood smoke in addition to primary wood smoke emitted particles. The identification of this secondary material was made possible by the collection of hourly averaged data that allowed for the time patterns of black carbon, organic material, and wood smoke marker compounds to be included and compared in a Positive Matrix Factorization (PMF) analysis. Most of the organic markers associated with wood smoke (levoglucosan, stearic acid and dehydroabietic acid) are associated with primary wood smoke emissions, but a fraction of the levoglucosan and stearic acid are also associated with secondary organic material formed from gaseous precursors in wood smoke. Additionally, this secondary material was shown to be present in each in of the three urban area where wood smoke burning occurs. There is a need for additional studies to better understand the contribution of secondary particulate formation from both urban and wildfires.Implications: This paper presents results from three field studies which show that in addition to the formation of primary particulate matter from the combustion of wood smoke and secondary particulate matter is also formed from the gaseous compounds emitted with the wood smoke. This material is identified in the studies of wood combustion reported here by the identification and quantification of specific organic marker compounds related to wood combustion and is shown to and represents a contributor nearly as large as the primary emitted material and better quantifying the impact of wood combustion on airborne fine particulate matter.

Apportionment of PM2.5 adjacent to the I-710 Harbor Freeway in Long Beach, CA.

During August and September 2012, a study was conducted to determine the sources of PM2.5 adjacent to the I-710 Long Beach Freeway. The site is directly affected by the emissions from heavy diesel traffic flowing from major container ports about 10 km south of the sampling site. Hourly average data were obtained for particulate species including PM2.5, black carbon and UV absorbing carbon, EC, fine particulate nonvolatile and semi-volatile organic material (NVOM and SVOM), sulfate, nitrate, chloride, ammonium ion, and Na ion, and for related factors including O3, CO, NOX, SO2, and total traffic flow on the I-710. A total of 520 hourly averaged data sets with 15 measured variables were analyzed by EPA-PMF v5.0. The data were best described by a 10-factor solution. Based on the composition and diurnal patterns of the factors, they were assigned to three diesel-related factors (two of which appeared to represent traffic from the ports and one general freeway diesel factor), a light-duty, spark-ignition vehicle-related factor, three secondary factors (one of which was associated with O3 formation processes), and three factors dominated by sulfate, SO2, and chloride, respectively. The diurnal patterns for these last three factors are strongly correlated. Meteorological and refinery upset data indicate that they are associated with emissions from a nearby refinery. The results of the PMF analysis were combined with nephelometer light scattering, corrected for coarse particle scattering and estimated aerosol water content in a multilinear regression analysis to identify visibility degradation sources. Major contributors were the aerosol water content, and the secondary PMF factors associated with either Nitrate and NVOM or NVOM and SVOM. The use of hourly average data made possible the identification of factors associated with gasoline vehicle emissions and both port and non-port diesel emissions.Implications: Hourly averaged data were obtained for PM2.5, its components and factors related to primary emissions and the formation of secondary material at a near freeway sampling location adjacent to the I-710 freeway just south of the Long Beach Boulevard entrance and 10 km north of the Ports of Long Beach and Los Angeles. The major objective of the study was to determine the impact of traffic from the ports at the monitoring site. This manuscript reports on the PMF analysis of the data set. Factors related to both diesel traffic originating from the ports and diesel traffic from non-port origins were identified. The diesel traffic originating from the ports was responsible for 9% of the total traffic and 95% of the BC measured at the sampling site. The non-port diesel traffic was responsible for 15% of the total traffic and 5% of the BC. While the Port 1 diesel traffic coming from the ports contributed a large fraction of the BC, this source contributed only 2% of the CO and 5% of the NOX at the sampling site. The impact of these traffic sources on light scattering was also small. Analysis of sources of sulfate and SO2 at the sampling site indicated that these species did not come from port activities of ships at or approaching the port, but rather from upset flare events at a nearby oil refinery.

Use of a gas chromatography-mass spectrometry organic aerosol monitor for in-field detection of fine particulate organic compounds in source apportionment.

A study was conducted on the Brigham Young University campus during January and February 2015 to identify winter-time sources of fine particulate material in Utah Valley, Utah. Fine particulate mass and components and related gas-phase species were all measured on an hourly averaged basis. Light scattering was also measured during the study. Included in the sampling was the first-time source apportionment application of a new monitoring instrument for the measurement of fine particulate organic marker compounds on an hourly averaged basis. Organic marker compounds measured included levoglucosan, dehydroabietic acid, stearic acid, pyrene, and anthracene. A total of 248 hourly averaged data sets were available for a positive matrix factorization (PMF) analysis of sources of both primary and secondary fine particulate material. A total of nine factors were identified. The presence of wood smoke emissions was associated with levoglucosan, dehydroabietic acid, and pyrene markers. Fine particulate secondary nitrate, secondary organic material, and wood smoke accounted for 90% of the fine particulate material. Fine particle light scattering was dominated by sources associated with wood smoke and secondary ammonium nitrate with associated modeled fine particulate water. IMPLICATIONS: The identification of sources and secondary formation pathways leading to observed levels of PM2.5 (particulate matter with an aerodynmaic diameter <2.5 µm) is important in making regulatory decisions on pollution control. The use of organic marker compounds in this assessment has proven useful; however, data obtained on a daily, or longer, sampling schedule limit the value of the information because diurnal changes associated with emissions and secondary aerosol formation cannot be identified. A new instrument, the gas chromtography-mass spectrometry (GC-MS) organic aerosol monitor, allows for the determination on these compounds on an hourly averaged basis. The demonstrated potential value of hourly averaged data in a source apportionment analysis indicates that significant improvement in the data used for making regulatory decisions is possible.

Modifications to the sunset laboratory carbon aerosol monitor for the simultaneous measurement of PM2.5 nonvolatile and semi-volatile carbonaceous material.

Semi-volatile organic carbonaceous material (SVOC) in fine particles is not reliably measured with conventional semicontinuous carbon monitors because semi-volatile carbonaceous material is lost from the collection media during sample collection. Two modifications of a Sunset Laboratory carbon aerosol monitor allowing for the determination of semi-volatile fine particulate organic material are described. Collocated conventional and modified instruments were operated simultaneously using a common inlet. Comparisons were made with integrated PC-BOSS data for quartz filter retained nonvolatile organic carbon (NVOC) and elemental carbon (EC), SVOC, and total carbon (TC = SVOC + NVOC + EC) and good agreement was observed between TC concentrations during studies conducted in Rubidoux, CA. Precision of the comparison was sigma = +/-1.5 microg-C/m3 (+/-8%). On the basis of experiments performed with the modified Sunset monitor, a dual-oven Sunset monitor was developed and extensively tested in Lindon, UT; Riverside, CA; and in environmental exposure chambers. The precision for the measurement of TC with the dual-oven instrument was sigma = +/-1.4 microg-C/m3 (+/-13%).

Measurement of fine particulate matter nonvolatile and semi-volatile organic material with the Sunset Laboratory Carbon Aerosol Monitor.

Semi-volatile organic material (SVOM) in fine particles is not reliably measured with conventional semicontinuous carbon monitors because SVOM is lost from the collection media during sample collection. We have modified a Sunset Laboratory Carbon Aerosol Monitor to allow for the determination of SVOM. In a conventional Sunset monitor, gas-phase organic compounds are removed in the sampled airstream by a diffusion denuder employing charcoal-impregnated cellulose filter (CIF) surfaces. Subsequently, particles are collected on a quartz filter and the instrument then determines both the organic carbon and elemental carbon fractions of the aerosol using a thermal/optical method. However, some of the SVOM is lost from the filter during collection, and therefore is not determined. Because the interfering gas-phase organic compounds are removed before aerosol collection, the SVOM can be determined by filtering the particles at the instrument inlet and then replacing the quartz filter in the monitor with a charcoal-impregnated glass fiber filter (CIG), which retains the SVOM lost from particles collected on the inlet filter. The resulting collected SVOM is then determined in the analysis step by measurement of the carbonaceous material thermally evolved from the CIG filter. This concept was tested during field studies in February 2003 in Lindon, UT, and in July 2003 in Rubidoux, CA. The results obtained were validated by comparison with Particle Concentrator-Brigham Young University Organic Sampling System (PC-BOSS) results. The sum of nonvolatile organic material determined with a conventional Sunset monitor and SVOM determined with the modified Sunset monitor agree with the PC-BOSS results. Linear regression analysis of total carbon concentrations determined by the PC-BOSS and the Sunset resulted in a zero-intercept slope of 0.99 +/- 0.02 (R2 = 0.92) and a precision of sigma = +/- 1.5 microg C/m3 (8%).