3506B24 Final Report
Download 11.49 Mb.
|
- Navigate this page:
- Flaming kg/burn Smoldering kg/burn
- Ft Gordon Ft Benning Ft Gordon
- 6.3 Secondary Organic Aerosol (SOA) Forming Potential
- Table 22 and Figure 28
- Toluene Ethylbenzene m+p-Xylene
- 6.4 CMB Source Apportionment of Ambient VOC at OLC
Table 20 compares emission estimates of major aromatics averaged over the various burns and burn stages, i.e. 4 flaming and 3 smoldering cases at Fort Gordon, and 5 each for Fort Benning, with the average daily aromatics emissions from mobile sources in the two residing Counties. Even under the relatively conservative assumptions, the magnitude of the prescribed burn aromatics emissions seems comparable to the mobile source emissions, especially the Fort Gordon burns tend to clearly exceed the Fort Benning emissions, both in the flaming and smoldering stages, indicating the highest potential for emissions reduction. Thus, it seems most beneficial to direct future focus towards the investigation of possible modifications of the existing burn protocol at Fort Gordon, targeting the management of both the burn conduct and post-burn smoldering stages. Table 20: Estimated emissions of aromatic compounds from flaming and smoldering phases of prescribed burns in comparison with mobile sources in Richmond and Muscogee Counties, applying year 2000 source profiles. Note that the mobile emissions list (m+p)-xylene as one lumped entity.
6.3 Secondary Organic Aerosol (SOA) Forming Potential Like forest fires, or industrial emissions from oil refineries, chemical plants, pulp and paper industries, and vehicular emissions, prescribed burning causes smoke. A large fraction of this smoke is primary organic aerosol directly emitted into the atmosphere. The organic portion is the least understood component of PM2.5. However, PM2.5 is not only made up of such primary emissions; a secondary aerosol fraction resulting from the reaction of VOC in the atmosphere is being recognized to be an important contributor. This secondary organic aerosol (SOA) fraction is extremely complex as the precursor VOC can originate from many different sources. It has been estimated that in the L.A. Basin, e.g. SOA can make up to 80 % of the observed organic particulate carbon under peak photochemical smog conditions, typically coinciding with periods of non-attainment of ozone and PM2.5 [Turpin & Huntzicker, 1995]. A detailed understanding of SOA formation in the atmosphere is essential to characterize the chemical composition of ambient organic aerosols, to accurately incorporate such processes in air quality models, and to be able to attribute the ambient organic aerosol mass to the appropriate man-made and natural sources allowing the development of adequate control strategies, e.g. in urban environments. While aromatic hydrocarbons and biogenic terpenes are major contributors to SOA in the atmosphere, these two compound classes are not solely responsible for SOA formation. Currently, in fact, other potentially important contributors to SOA formation are subject of intense scientific investigation. In addition, there is a large amount of semi-volatile organic material (direct emissions and photochemical oxidation products of volatile VOC emissions) that has the potential to move into the aerosol phase as climate or atmospheric chemistry undergoes subtle changes. The mass of such material is so large in comparison to amounts of material currently in the aerosol phase, that impacts on PM2.5 are potentially significant. Thus, it is important to identify all such SOA precursors, their aerosol-forming potential, and their sources. Because of the complexity of SOA reaction pathways, the vast number of products formed by photochemical oxidation of primary aerosol, and the costly analytical methods required for speciation, indirect methods for quantitative assessment of SOA have been developed and become very useful. One such method utilizes Fractional Aerosol Coefficients (FAC), developed and first published by Grosjean and Seinfeld [1989]. The FAC approach for determining SOA yield is based on measurements of the total aerosol formed in smog chamber reactions of a specific precursor species and a specific oxidant. Since the reaction mechanism is not known, the kinetics and reaction rate constants are also not known. The smog chamber data are, therefore, used to empirically derive the reaction stoichiometry, that is, to determine the amount of condensed matter formed per gram of reactant, which is the FAC or fractional aerosol yield. The FAC can be expressed on a molar, mass or carbon concentration basis [Grosjean and Seinfeld, 1989]. This dimensionless ratio of mass concentration is defined by Grosjean [1992] as: FAC = aerosol from VOC (gm-3) / initial VOC (gm-3) With this definition, and knowing the VOC emission rate and the fraction of VOC that has reacted in the atmosphere, the amount of aerosol formed from each VOC can be calculated as: [Aerosol] produced = [VOC] emitted * [%-VOC] reacted * (FAC) The FAC is a very crude first order approximation to SOA formation P(SOA)! It summarizes the complicated oxidation-condensation processes that govern SOA formation into one constant for each precursor VOC species. SOA can be formed by parent VOC species which have a carbon chain greater than six but generally less than ten (6 Dookwah [2003] applied the FAC method to the aromatic species found in the emissions profile of the mobile sources in Muscogee and Richmond counties mentioned in section 6.2.2 above. For comparison, the unit mass emissions EVOC from the prescribed burns were approximated as outlined above. With the fraction of VOC reacted (FR), the secondary organic aerosol forming potential hence becomes P(SOA) = EVOC * FR * FAC FR is calculated assuming first order principles from reaction with OH, midday [OH] of 106 molec cm-3, and a 5 hour reaction period; FR = 1-exp(-kt). Table 21 lists the OH rate constants (k) [Atkinson, 1990], the fraction of molecules reacted with OH (FR), and the Fractional Aerosol Coefficients (FAC) [Grosjean, 1992] for the aromatic species investigated here. Table 21: OH rate constants (k) from Atkinson [1990], the fraction of VOC species reacted with OH (FR) during a 5 hour period from first order principles, assuming midday [OH] = 106 molec cm-3., and the Fractional Aerosol Coefficients (FAC) from Grosjean [1992] for the most important aromatic species investigated here.
The following compares the P(SOA) potential for the most important contributing aromatics from flaming and smoldering sources of prescribed burns with the ones for the mobile sources in the residing counties. Isopropyl-benzene and propyl-benzene were minor contributors with negligible P(SOA) potential. Note that the mobile source profiles of the two counties were almost identical, i.e. the fleet distribution and activities were proportionately identical. Therefore, the relative P(SOA) potential for only one of the two counties is listed in Table 22 and Figure 28 for comparison with the prescribed burn P(SOA) potential profile. Also, the smoldering samples collected on April 29th on both Forts Benning and Gordon are depicted in the figure’s P(SOA) profile but were not considered in any of the average calculations above, because the one taken at Fort Gordon was far too high to be considered a smoldering sample, probably due to stagnant conditions and accumulation of pollutants at the sampling site; and the other one taken at Fort Benning did not meet the criteria for a smoldering sample either, because the pollution level was close to or even lower than the downwind sample from that day. Toluene is estimated to be the major SOA precursor species for the flaming phases of prescribed burns on either installation, accounting for more than 40 % of the total potential atmospheric SOA burden, closely followed by the xylenes. The sum of all xylenes becomes the dominant precursor for SOA from smoldering emissions with potentially 37 % at Fort Benning and 44 % at Fort Gordon. Table 22: Profiles of SOA forming potential for aromatics from flaming and smoldering vs. mobile sources sub-divided into gasoline and diesel powered vehicles.
Figure 28: Relative contributions (profile) of major aromatic compounds emitted by flaming (FL) and smoldering (SM) sources of prescribed burns on Forts Gordon and Benning to the SOA forming potential of the region, compared with the potential from mobile sources of Muscogee county. Note that the mobile sources carry (m+p)-xylene as a lumped entity but is here plotted assuming an 80/20 ratio. It should be noted here, that the application of the SOA forming potential emphasizes the relative contribution of species for certain samples, which basically eliminates the uncertainties of the assumed plume rising velocity and flaming rate for the comparison of flaming samples, since those parameters were assumed to be constant. The comparison of smoldering samples benefits from the same fact. The comparison of the two different burn stages is however highly uncertain due to the uncertainty of the initial assumption going into the calculation, such as the different durations of the burn stages and the different plume rise times. Another uncertainty arises from the possibility of pollutants accumulation during the flaming stage under conditions of insufficient venting, causing an increase of the local background and/or the smoldering sample itself, which seems to have occurred in one instance (on 4/29 at FtG). In general, if the source profile and emissions data of precursor species are known, the SOA formation potential of those species can be estimated utilizing the empirical FAC approach. Quantification of atmospheric SOA concentrations, however, is questionable, since the approach neglects important variables such as timescales involved in SOA formation, transport factors, relative humidity influences, competition between VOC species, synergistic reactions of VOC species and other possibilities that exist in ambient gas mixtures that do not exist in controlled chamber studies. Despite these limitations, FACs can be used to compare the relative importance of VOC sources for SOA formation. This study was limited to the analysis of aromatics, since, for comparison reasons, only mobile emissions species profiles were available at this point. A more comprehensive study will have to follow in future investigations. 6.4 CMB Source Apportionment of Ambient VOC at OLC The Chemical Mass Balance (CMB) applications and validation protocol for PM2.5 and VOCs was prepared by the Desert Research Institute, Reno, NV, for the for U.S. Environmental Protection Agency, Research Triangle Park, NC, in 1998. The CMB model relates the measured VOC concentrations at the receptor to potential sources, and apportions the source contributions following the principal of chemical mass balance. The CMB model is based on the following equation:  Cik = fijSjk i = 1… m, j = 1… n.
where Cik is the observed concentration of species i in sample k, Sjk is the total concentration of material (particles, VOCs, etc) from source j in sample k and fij is the mass fraction of species i in source j (i= 1… m; J= 1...n; k= 1…l). The concentration of each chemical species at the receptor becomes a linear combination of the contributions from each source to the total concentration of each species at the site. Given the chemical composition of the ambient sample, Cik, and the source profiles, fik, the equation can be solved to provide the source contribution Sij by means of multiple regressions. The CMB model works under the following assumptions: 1) compositions of source emissions are constant over the period of ambient and source sampling; 2) chemical species do not react with each other; 3) all sources with a potential for contributing to the receptor have been identified and their emissions have been characterized; 4) the number of sources or source categories is less than or equal to the number of chemical species; 5) the source profiles are linearly independent of each other; and 6) measurement uncertainties are random, uncorrelated, and normally distributed. Here, source profiles from Photochemical Assessment Monitoring Station (PAMS) in the United States were used for chemical mass balance as shown in Table 23. These source profiles can be generally considered representative of larger urban environments in the US, since they were characterized in 1 hour ozone non-attainment areas. They include diesel exhaust (HDDE), gasoline exhaust (LDGE), liquid gasoline (LGAS), evaporated gasoline (EGAS), refinery fugitives (REFG), industrial coating (INCO), primers and enamel (PMEN), and printing (PRNT). Fifty-five (55) compounds were identified for each source profile, and the concentration of each compound was normalized to the sum of 55 compounds as a fraction. Here, it is assumed that isoprene is the only chemical species representing a biogenic source (BIOG), even though other biogenic tracers were measured such as -, and -pinene. From the VOC emissions measurements described above, additional source profiles representing the flaming and smoldering stage, respectively, were created for 28 VOC species. Due to their similarity, with exception of m- and p-ethyl-toluene, and n-propyl-benzene the profiles were statistically indistinguishable, only one profile representing the average flaming emissions (FLAM) was introduced into the CMB. Download 11.49 Mb. Share with your friends:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||