The Congestion Mitigation and Air Quality Improvement Program: Assessing 10 Years of Experience -- Special Report 264 (2002)

Ozone is formed by the reaction of atomic and molecular oxygen. The only significant oxygen atom production in the troposphere is from photodissociation of nitrogen dioxide (NO₂) into nitric oxide (NO) and oxygen atoms, Reaction 1. The oxygen atoms react with molecular oxygen to produce O₃, Reaction 2. When nitrogen oxides are present, O₃ reacts rapidly with NO to regenerate NO₂, Reaction 3. The first and third reactions occur rapidly, establishing a steady-state equilibrium ozone concentration, which is proportional to the NO₂/NO ratio. Because these reactions only recycle O₃ and NO_x, they are insufficient, by themselves, to create excessive ozone levels.

When volatile organic compounds (VOCs) are present, their oxidation produces the hydroperoxy radical (HO₂) and organic peroxy radicals (RO₂), which react with NO to form NO₂ without destruction of ozone, thereby allowing ozone to accumulate. For the majority of VOCs emitted from anthropogenic and natural sources, the reaction with the hydroxyl radical (HO) initiates the oxidation sequence. However, there is a competition between VOCs and NO_x for the HO radicals. VOCs are consumed in the sequence of ozone formation, while both NO_x and HO and HO₂ radicals act as catalysts. Termination occurs when HO₂ combines to form hydrogen peroxide (H₂O₂) or by reaction of HO with NO₂ to form HNO₃. O₃ production is related to the number of NO to NO₂ conversions effected by VOCs and their decomposition products over the entire photooxidation cycle. The ozone production efficiency (OPE) is defined as the number of O₃ molecules produced per NO_x molecule emitted. This parameter is relevant for the development of regional ozone-mitigation strategies because it provides an indication of the reduction in O₃ that might be expected for a given reduction in regional NO_x emissions. OPE also provides a basis for weighting NO_x emission reductions in the

calculations of cost-effectiveness of control measures (i.e., cost of the control measure in dollars divided by the expected emission reductions in tons).

Estimates of OPE, ranging from 7 to 10, were initially derived on the basis of linear relationships between O₃ and the oxidation products of NO_x at rural sites (Trainer et al. 1993). Chin et al. (1994) derived a lower limit for OPE of 1.7 and argued that the earlier estimates overstated OPE because NO_x is removed from the atmosphere more rapidly than is O₃. More recent studies involving direct, airborne measurements within power plant and urban plumes (Ryerson et al. 1998) and regional analyses of rural O₃ monitoring data (Kasibhatla et al. 1998) yield OPE values in the range of one to three molecules of O₃ per molecule of NO_x.

The concentration of NO_x and VOC/NO_x ratios are the two main factors affecting the OPE. At low VOC/NO_x ratios, HO reacts predominantly with NO₂ to form HNO₃, removing radicals and NO_x from the photochemical cycle and retarding O₃ formation. Under these conditions, a decrease in NO_x concentration favors O₃ formation (ozone formation is hydrocarbon limited). High VOC/NO_x ratios favor HO reaction with VOCs that generate new radicals that accelerate O₃ production. However, at a sufficiently low concentration of NO_x, or a sufficiently high VOC/NO_x ratio, a further decrease in NO_x favors peroxy-peroxy reactions, which retard O₃ formation by removing free radicals from the system (ozone formation is NO_x limited). At a given level of VOC, there exists a NO_x mixing ratio at which a maximum amount of ozone is produced. This optimum VOC/NO_x ratio depends on the reactivity of HO to the particular mix of VOCs present. Because NO_x is removed faster than hydrocarbons, VOC/NO_x ratios tend to increase during transport, and ozone formation can change from hydrocarbon limited in the urban core to NO_x limited in downwind suburban and rural locations. Accordingly, NO_x reductions could lead to higher peak 1-hour average O₃ levels in the urban locations that are currently hydrocarbon limited, but to lower 8-hour average O₃ levels in downwind locations. Ozone formation is complex, and a thorough understanding of the response of ozone levels to specific changes in VOC or NO_x emissions is the fundamental prerequisite to developing cost-effective ozone abatement strategies.

The gaseous precursors of most particulate sulfates and nitrates are SO₂ and NO_x, respectively. Ambient concentrations of sulfate and nitrate are not necessarily proportional to quantities of emissions because the rates at which they form may be limited by factors other than the concentration of the precursor gases. The majority of secondary sulfates are found as a combination of sulfuric acid, ammonium bisulfate, and ammonium sulfate. The majority of secondary nitrates in PM₁₀ are found as ammonium nitrate, though a portion of the nitrate is also found in the coarse particle fraction, usually in association with sodium (this is presumed to be sodium nitrate derived from the reaction of nitric acid with the sodium chloride in sea salt).

Sulfur dioxide changes to particulate sulfate through gas- and aqueous-phase transformation pathways. In the gas-phase pathway, sulfur dioxide reacts with hydroxyl radicals in the atmosphere to form hydrogen sulfite. This species rapidly reacts with oxygen and small amounts of water vapor to become sulfuric acid gas. Sulfuric acid gas has a low vapor pressure. It condenses on existing particles and nucleates at high relative humidities to form a sulfuric acid droplet or, in the presence of ammonia gas, becomes neutralized as ammonium bisulfate or ammonium sulfate. When fogs or clouds are present, sulfur dioxide can be dissolved in a droplet, where it experiences aqueous reactions that are much faster than gas-phase reactions. If ozone and hydrogen peroxide are dissolved in the droplet, the sulfur dioxide is quickly oxidized to sulfuric acid. If ammonia is also dissolved in the droplet, the sulfuric acid is neutralized to ammonium sulfate. The major pathway to nitric acid is reaction with hydroxyl radicals. Nitric acid leaves the atmosphere fairly rapidly, but in the presence of ammonia it is neutralized to particulate ammonium nitrate.

Sulfur dioxide to particulate sulfate and nitrogen oxide to particulate nitrate reactions compete with each other for available hydroxyl radicals and ammonia. Ammonia is preferentially scavenged by sulfate to form ammonium sulfate and ammonium bisulfate, and the amount of ammonium nitrate formed is only significant when the total ammonia exceeds the sulfate by a factor of two or more on a mole basis. In an ammonia-limited environment, reducing ammonium sulfate concentrations by one molecule would

increase ammonium nitrate concentrations by two molecules. This implies that reducing SO₂ emissions might actually result in ammonium nitrate increases that exceed the reductions in ammonium sulfate where the availability of ammonia is limited.

While the mechanisms and pathways for inorganic secondary particles are fairly well known, those for secondary organic aerosols are not well understood. Hundreds of precursors are involved in these reactions, and the rates at which these particles form are highly dependent on the concentrations of other pollutants and meteorological variables. Organic compounds present in the gas phase undergo atmospheric transformation through reactions with reactive gaseous species such as OH radicals, NO₃ radicals, or O₃. Secondary organic compounds in particulate matter include aliphatic acids, aromatic acids, nitro aromatics, carbonyls, esters, phenols, and aliphatic nitrates (Grosjean 1992; Grosjean and Seinfeld 1989). However, these compounds can also be present in primary emissions [see, for example, Rogge (1991)]; thus they are not unique tracers for atmospheric transformation processes. Particles are formed when gaseous reaction products achieve concentrations that exceed their saturation concentrations. Fraction conversion factors, based on experimental data taken in smog chamber experiments, relate the aerosol products of selected precursors to the original quantities of those precursors. Applying these factors to chemically speciated emission inventories provides an approximate estimate of the equivalent emissions of secondary organic particles. Grosjean (1992) shows that these equivalent emissions are comparable with primary emissions from other carbon-containing sources, such as motor vehicle exhaust in the Los Angeles area. While this empirical model provides an order-of-magnitude estimate of the VOC impacts on PM₁₀, quantitative estimates are very imprecise.

Chin, M., D. J. Jacob, J. W. Munger, D. D. Parrish, and B. G. Doddridge. 1994. Relationship of Ozone and Carbon Monoxide over North America. Journal of Geophysical Research, Vol. 99, pp. 14,565–14,573.

Grosjean, D., and J. H. Seinfeld. 1989. Parameterization of the Formation Potential of Secondary Organic Aerosols. Atmospheric Environment, No. 23, pp. 1733–1747.

Grosjean, D. 1992. In-Situ Organic Aerosol Formation During a Smog Episode: Estimated Production and Chemical Functionality. Atmospheric Environment, Vol. 26A, pp. 953–963.

Kasibhatla, P., W. L. Chameides, R. D. Saylor, and D. Olerud. 1998. Relationships Between Regional Ozone Pollution and Emissions of Nitrogen Oxides in Eastern United States. Journal of Geophysical Research, Vol. 103, pp. 22,663–22,669.

Rogge, W. F., L. M. Hildemann, M. A. Mazurek, and G. R. Cass. 1991. Sources of Fine Organic Aerosols. 1: Charbroilers and Meat Cooking Operations. Environmental Science and Technology, Vol. 25, pp. 1112–1125.

Ryerson, T. B., M. Trainer, M. P. Buhr, G. Frost, P. S. Goldan, J. S. Holloway, G. Hubler, B. T. Jobson, W. C. Kuster, S. A. McKeen, D. D. Parrish, J. M. Roberts, D. M. Sueper, J. Williams, and F. C. Fehsenfeld. 1998. Emission Lifetimes and Ozone Formation in Power Plant Plumes. Journal of Geophysical Research, Vol. 103, pp. 22,415–22,423.

Trainer, M., D. D. Parrish, M. P. Buhr, R. B. Norton, F. C. Fehsenfeld, K. G. Anlauf, J. W. Botenheim, Y. Z. Yang, H. A. Weibe, J. M. Roberts, R. L. Tanner, L. Newman, V. C. Bowersox, J. F. Meagher, K. J. Olszyna, M. O. Rodgers, T. Wang, H. Berresheim, K. L. Demerjian, and U. K. Roychowdhury. 1993. Correlation of Ozone with NO_x in Photochemically Aged Air. Journal of Geophysical Research, Vol. 98, pp. 2,917–2,925.