Page 109

5
Atmospheric Chemistry of Ozone and Its Precursors

Introduction

Large quantities of chemical compounds are emitted into the atmosphere as a result of anthropogenic and biogenic activities. These emissions lead to a complex spectrum of chemical and physical processes that result in such diverse effects as photochemical air pollution (including the formation of ozone in urban, suburban, and rural air masses), acid deposition, long-range transport of chemicals, stratospheric ozone depletion, and accumulation of greenhouse gases. Over the past 15 to 20 years, many laboratory and ambient atmospheric studies have investigated the physical and chemical processes of the atmosphere. Because these processes are complex, computer models are often used to elucidate and predict the effects of anthropogenic and biogenic emissions—and of changes in these emissions—on the chemistry of the atmosphere.

In this chapter, the gas-phase chemistry of the relatively unpolluted, methane-dominated troposphere is summarized, and the additional complexities of the chemistry of polluted atmospheres are discussed. The tropospheric chemistry of organic compounds of anthropogenic and biogenic origin, respectively, is discussed in detail, and the calculated tropospheric lifetimes of these compounds are presented. The formulation and testing of chemical mechanisms for use in urban and regional airshed computer models are discussed briefly, and the reactivities of organic compounds with respect to ozone formation, as calculated using these models, are discussed.



The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 109
Page 109 5 Atmospheric Chemistry of Ozone and Its Precursors Introduction Large quantities of chemical compounds are emitted into the atmosphere as a result of anthropogenic and biogenic activities. These emissions lead to a complex spectrum of chemical and physical processes that result in such diverse effects as photochemical air pollution (including the formation of ozone in urban, suburban, and rural air masses), acid deposition, long-range transport of chemicals, stratospheric ozone depletion, and accumulation of greenhouse gases. Over the past 15 to 20 years, many laboratory and ambient atmospheric studies have investigated the physical and chemical processes of the atmosphere. Because these processes are complex, computer models are often used to elucidate and predict the effects of anthropogenic and biogenic emissions—and of changes in these emissions—on the chemistry of the atmosphere. In this chapter, the gas-phase chemistry of the relatively unpolluted, methane-dominated troposphere is summarized, and the additional complexities of the chemistry of polluted atmospheres are discussed. The tropospheric chemistry of organic compounds of anthropogenic and biogenic origin, respectively, is discussed in detail, and the calculated tropospheric lifetimes of these compounds are presented. The formulation and testing of chemical mechanisms for use in urban and regional airshed computer models are discussed briefly, and the reactivities of organic compounds with respect to ozone formation, as calculated using these models, are discussed.

OCR for page 109
Page 110 General Schemes of Tropospheric Chemistry Ozone is present in the natural, unpolluted troposphere, and its tropospheric column density is approximately 10% of the total atmospheric (troposphere + stratosphere) ozone column density. (Logan, 1985; Brühl and Crutzen, 1989; Fishman et al., 1990). The ozone present in the stratosphere absorbs short-wavelength radiation ( nm (nanometers or 10-9 meters)) from the sun and allows only those wavelengths nm to penetrate into the troposphere (Peterson, 1976; Demerjian et al., 1980). The sources of ozone in the natural troposphere are downward transport from the stratosphere and in situ photochemical production. Losses result from photochemical processes and from deposition and destruction at the earth's surface. The rates of downward transport, production, and losses are estimated to be of the same order of magnitude (Logan, 1985). The ozone present in the troposphere is important in the atmospheric chemistry because the OH radical is generated from the photolysis of ozone at wavelengths <319 nm (Levy, 1971; DeMore et al., 1990). The formation of OH radicals leads to cycles of reactions that result in the photochemical degradation of organic compounds of anthropogenic and biogenic origin, the enhanced formation of ozone, and the atmospheric formation of acidic compounds (see, for example, Heicklen et al., 1969; Stedman et al., 1970; Finlayson-Pitts and Pitts, 1986; WMO, 1986). The generation of the OH radical from ozone is shown in the following reactions: Energy from solar radiation is represented by hv, the product of Planck's constant, h, and the frequency, v, of the electromagnetic wave of solar radiation. O(1D) is an excited oxygen atom, and M is an inert compound, such as N2 or O2. O(3P) is a ground state oxygen atom. The chemistry of the clean, unpolluted troposphere is dominated by the chemistry of methane (CH4) and

OCR for page 109
Page 111 its degradation products, formaldehyde (HCHO) and carbon monoxide (CO) (see, for example, Levy, 1972; Crutzen, 1973; Fishman and Crutzen, 1977; Logan et al., 1981). Tropospheric Methane Oxidation Cycle A sequence of reactions (Ravishankara, 1988; Atkinson et al., 1989a; Atkinson, 1990b) starts with the reaction of the OH radical with methane The tropospheric lifetime of methane, , is controlled by reaction with the OH radical, where is the rate constant for the reaction of the OH radical with methane and [OH] is the OH radical concentration. It should be noted that depends on temperature, and hence on altitude, and that the OH radical concentration is temporally and spatially dependent. The lifetime of methane in the troposphere is long enough that a diurnally and annually averaged concentration of global tropospheric OH radical can be used to calculate the lifetime of methane (and of other similarly long-lived trace species). Based on methylchloroform (CH3CCl3) emissions and atmospheric budgets and an equation analogous to Equation 5.5, Prinn et al. (1987) derived a tropospheric lifetime for methylchloroform of 6.3 years and a globally averaged tropospheric OH radical concentration of 7.7 × 105 molecule/cm3. From this OH radical concentration, the methane lifetime is calculated to be approximately 12 years (Vaghjiani and Ravishankara, 1991). For organic compounds that react more rapidly with the OH radical and have much shorter lifetimes ( year), the temporal and spatial variations of the OH radical concentrations need to be considered in the calculation of tropospheric lifetimes.

OCR for page 109
Page 112 Under tropospheric conditions, the methyl radical rapidly, and solely, adds oxygen to form the methyl peroxy radical (CH3O2): which can then react with nitric oxide (NO), nitrogen dioxide (NO2), hydroperoxyl radical (HO2), and organic peroxy radicals (RO2) Methyl peroxynitrate, CH3OONO2, thermally dissociates back to the reactants with a lifetime of methyl peroxynitrate with respect to thermal decomposition of ˜1 sec at room temperature and atmospheric pressure, which increases to ˜2 days for the temperature and pressure conditions in the upper troposphere (Atkinson et al., 1989a; Atkinson, 1990b). Because the reactions of the CH3O2 radical with NO and NO2 have comparable rate constants for the temperatures and pressures encountered in the troposphere (Atkinson, 1990a), methyl peroxynitrate can act as a temporary reservoir of NO2 and CH3O2 radicals in the upper troposphere. The reaction of the methylperoxy radical with NO will dominate over reaction with the HO2 radical for tropospheric NO mixing ratios equal to or great-

OCR for page 109
Page 113 er than approximately 10-30 parts per trillion (ppt) (Logan et al., 1981). However, in the clean, unpolluted lower troposphere, NO mixing ratios are generally <30 ppt (see, for example, Kley et al., 1981; Logan, 1983; Ridley et al., 1987, 1989; Drummond et al., 1988; and Chapter 8), and under these conditions the HO2 radical reaction to form methyl hydroperoxide, CH3OOH, is important. The subsequent reactions of CH3OOH under tropospheric conditions are photolysis and reaction with the OH radical (Ravishankara, 1988; Atkinson, 1989, 1990a, b) These two processes are comparable in importance, and they reform the CH3O and CH3O2 radicals. Wet deposition of methyl hydroperoxide and its incorporation into cloud, fog, and rain water also could be important (Hell-pointner and Gäb, 1989). For a discussion of cloud chemistry, see, for example, Chameides (1984), Jacob (1986), Jacob et al. (1989), and Pandis and Seinfeld (1989). The sole loss process for the methoxy radical in the clean troposphere is through reaction with oxygen to generate formaldehyde (Atkinson et al., 1989a; DeMore et al., 1990) The HO2 radical can lead to the regeneration of the chain-carrying OH radical by reaction with NO or react with peroxy (RO2) radicals (including HO2) or ozone

OCR for page 109
Page 114 The self-reaction of HO2 radicals forms hydrogen peroxide, which, like methyl hydroperoxide, can undergo wet deposition and incorporation into cloud, fog, and rain water. When enough NO is present that the reactions of CH3O2 and HO2 radicals with NO dominate over the reactions of these peroxy radicals with HO2 (or other peroxy radicals) or of HO2 radicals with ozone, then the overall methane photooxidation reaction is given by with methane being degraded to formaldehyde, two molecules of NO being converted to NO2, and the OH radical being regenerated. (An equal sign is used instead of an arrow to indicate that the net overall process shown as Reaction 5.18 represents many individual reactions.) When the reaction of the CH3O2 radical with HO2 dominates, then the oxidation of methane becomes a net sink for OH and HO2 radicals, with approximately The NOx concentrations in the atmospheric boundary layer over continental areas in the northern hemisphere are generally high enough that the reactions

OCR for page 109
Page 115 of RO2 and HO2 peroxy radicals with NO dominate over the reactions of the RO2 and HO2 radicals with HO2 and the reaction of the HO2 radical with ozone. The result is net ozone formation. Only in remote locations such as the mid-Pacific Ocean and portions of the southern hemisphere are the NOx concentrations low enough that the reactions of HO2 radicals with ozone (Reaction 5.16) and other peroxy (RO2) radicals dominate, leading to net ozone removal. Formaldehyde also undergoes reaction in the troposphere by photolysis and reaction with the OH radical (Atkinson et al., 1959a; Atkinson, 1990a) followed by and For HCHO, photolysis dominates over reaction with the OH radical (Atkinson, 1988), and the calculated lower tropospheric lifetime of HCHO due to photolysis and, to a lesser extent, reaction with the OH radical is ˜4 hours at the sun's zenith angle of 0º (Rogers, 1990). The tropospheric removal of CO is by reaction with the OH radical, with a calculated lower tropospheric lifetime of ˜2 months. The tropospheric lifetimes of HCHO and CO are thus both much shorter than that of methane.

OCR for page 109
Page 116 Photochemical Formation of Ozone In the troposphere, ozone formation occurs to any significant extent only from the photolysis of NO2 at-wavelengths <424 rim, when sufficient solar energy is absorbed by NO2 to cause it to photodissociate In the absence of other processes that convert NO to NO2, and assuming steady-state conditions, then and the ozone concentration is linked to the NO2/NO concentration ratio during daylight hours. (Here j1 is the diurnally, seasonally, and latitudinally dependent rate of photolysis of NO2, and k2 is the rate constant for Reaction 5.27.) For an NO2/NO concentration ratio of one, a reasonable mid-day value in the dean lower troposphere, and a temperature of 298 K, the resulting ozone concentration is ˜ 5 × 1011 molecule/cm3 (20 parts per billion (ppb) mixing ratio). As discussed above for the methane oxidation cycle, the presence of volatile organic compounds (VOCs) causes enhanced NO-to-NO2 conversion and hence the production of concentrations of ozone that exceed those encountered in the dean background troposphere (see, for example, Parrish et al., 1986). This is discussed further below. For example, for the OH radical-initiated reaction of methane in the presence of NO given above, the overall reaction is

OCR for page 109
Page 117 This leads to a net reaction of Other Reactions in the Tropospheric Nitrogen Cycle In addition to Reactions 5.25 and 5.27 and the reaction of the HO2 radical with NO to regenerate the OH radical, other tropospherically important reactions involve oxides of nitrogen (Finlayson-Pitts and Pitts, 1986; WMO, 1986; Atkinson et al., 1989a; DeMore et al., 1990). The recombination reactions to form nitrous acid (HONO) and pernitric acid (HOONO2) are of little importance because of the rapid photodissociation of HONO

OCR for page 109
Page 118 and the thermal decomposition of HOONO2 back to reactants. However, the combination reaction of the OH radical with NO2 is the major gas-phase route to the formation of nitric acid (HNO3), and it is the major homogeneous gas-phase sink for NOx (oxides of nitrogen) in the troposphere. This reaction also serves as a sink for OH and HO2 radicals (odd hydrogen) for NOx mixing ratios ppb, and under these conditions the removal of OH radicals by Reaction 5.35 balances the formation of HOx (oxides of hydrogen) radicals from the photolysis of ozone and HCHO. The major reactions involved in the oxidation of methane in the presence of NOx are diagrammed in Figure 5-1, which emphasizes the chain-cycle nature of this overall reaction process. Ozone also reacts with NO2 to form the nitrate (NO3) radical, Figure 5-1 Major reactions involved in the oxidation  of methane (CH4) in the presence of NOx.

OCR for page 109
Page 119 and the NO3 radical is interconverted with NO2 and dinitrogen pentoxide (N2O5) through the reactions Because NO3 radicals rapidly photolyze (with a photolysis lifetime of ˜5 seconds at a solar zenith angle of 0º) and react rapidly with NO, concentrations of the NO3 radical, and hence of N2O5, remain low during the daytime but can increase during evening and nighttime hours (Platt et al., 1981, 1984; Pitts et al., 1984a). The homogeneous gas-phase reaction of N2O5 with water vapor to form nitric acid is slow enough that only an upper limit can be placed on the rate constant (Atkinson et al., 1989a; Hatakeyama and Leu, 1989), but the wet and dry deposition of N2O5 or of NO3 radicals provides a potentially important nighttime route to the removal of gas-phase NOx and the formation of acid deposi-

OCR for page 109
Page 152 environmental chamber data, it is not clear to what extent even a totally incorrect treatment of the chemistry of a specific class of VOCs will result in incorrect predictions of the airshed model under actual atmospheric conditions. The chemical mechanisms of Gery et al. (1988a) and Carter et al. (1986a) do, however, lead to significantly different predictions of ozone formation at temperatures below ˜298 K for VOC/NOx ratios <10 (Dodge, 1989); the CB-IV mechanism of Gery et al. (1988a) predicts the formation of less ozone at lower temperatures than does the mechanism of Carter et al. (1986a). This discrepancy arises because of different assumptions for the temperature dependence of the ratio of the rate constants for Reactions 5.96 and 5.97, where the acetyl peroxy radical reacts with NO and NO2, respectively. Gery et al. (1988a) assume a temperature dependence in the ratio of the rate constants for Reactions 5.96 and 5.97 of e-5250/T, and Carter et al. (1986a) use a value of the rate constant ratio that is independent of temperature. At room temperature the rate constant ratio for Reactions 5.96 and 5.97 used in both mechanisms is similar. The more sensitive temperature dependence used by Gery et al. (1988a) leads to NOx being sequestered as PAN at lower temperatures and not being available to participate in the formation of ozone. The data of Kirchner et al. (1990) and Tuazon et al. (1991) show that the rate constant ratio for Reactions 5.96 and 5.97 is equal to 2.2, independent of temperature over the range 283-313 K at 740 Torr total pressure of air. This will require revision of the Gery et al. (1988a) CB-IV mechanism and of the Carter et al. (1986a) mechanism.

OCR for page 109
Page 153 Ozone Formation Potential of Various Vocs VOCs emitted from anthropogenic and biogenic sources react in the troposphere in the presence of NOx and sunlight to lead to the photochemical formation of ozone. For two decades it has been known that VOCs vary widely in the speed with which they react in the troposphere and in the extent to which they promote or inhibit ozone formation (see, for example, Altshuller and Bufalini, 1971, and references therein; Dimitriades, 1974; Pitts et al., 1977). Several reactivity scales have been proposed to define the potential of VOCs to form ozone in the atmosphere, and these have included the maximum amount of ozone generated in irradiated mixtures of a single VOC, NOx, and air for several hours (Wilson and Doyle, 1970; Laity et al., 1973; Dimitriades and Joshi, 1977; Joshi et al., 1982); the rate of NO photooxidation (Heuss and Glasson, 1968; Glasson and Tuesday, 1970a,b, 1971), the rate of ozone formation (Heuss and Glasson, 1968; Winer et al., 1979), the rate of VOC consumption (Heuss and Glasson, 1968), all in irradiated mixtures of VOCs and NOx; and the rate constants for the reaction of the VOC with the OH radical (Darnall et al., 1976; Pitts et al., 1977). Although there are differences among these reactivity scales, the reactivities of VOCs as obtained from these scales are in a general order of increasing reactivity: alkanes and monoalkylbenzenes < 1-alkenes and dialkylbenzenes < trialkylbenzenes and internal alkenes (Hems and Glasson, 1968; Darnall et al., 1976). There are problems associated with basing reactivity scales on smog chamber experiments because of chamber effects (see, for example, Bufalini et al., 1977; Joshi et al., 1982; Carter et al., 1982a), and experimental chamber data are not directly applicable to ambient atmospheric conditions because they generally do not take into account the dilution and continuous input of VOCs and NOx in ambient air (Carter and Atkinson, 1989b). The use of the OH radical reaction rate constant scale does not suffer from these effects (Darnall et al., 1976), but this scale does not take into account the reactions subsequent to the initial OH radical reaction and it ignores other tropospheric loss processes, such as photolysis and reaction with NO3 radicals and ozone. Thus, for example, the formation of photoreactive products, such as formaldehyde, leads to increased overall reactivity with respect to ozone formation, whereas the generation of products such as organic nitrates, which act as sinks for NOx and radical species, leads to a decreased ozone-forming potential (Carter and Atkinson, 1987, 1989b). A useful definition of reactivity is that of incremental reactivity, defined as the amount of ozone formed per unit mount (as carbon) of VOC added to a VOC mixture representative of conditions in urban and rural areas in a given air mass (Dodge, 1984; Carter and Atkinson, 1987, 1989b; Carter, 1991),

OCR for page 109
Page 154 where D[ozone] is the change in the amount of ozone formed as a result of the change in the amount of organic present, D[VOC] (note that Carter and Atkinson [1989b] used the quantity D([ozone] - [NO]) rather than D[ozone] under conditions where the maximum ozone was not attained and NO was not fully consumed). This concept of incremental reactivity corresponds closely to control strategy conditions, in that the effects of reducing the emission of a VOC or group of VOCs, or of replacing a VOC or group of VOCs with other VOCs, on the ozone-forming potential of complex mixture of VOC emissions are simulated. The theoretical and experimental studies of Bufalini and Dodge (1983), Dodge (1984), and Carter and Atkinson (1987, 1989b) show that, in agreement with previous scales, VOCs exhibit wide variations in reactivity with respect to ozone formation. Furthermore, the absolute and relative calculated incremental reactivities of VOCs depend on VOC/NOx ratios (Bufalini and Dodge, 1983; Dodge, 1984; Carter and Atkinson, 1989b) and on the "scenario" used (i.e., the VOC mix to which incremental changes are made and associated physical factors, such as the amount of dilution [Carter and Atkinson, 1989b]). Table 5-4 shows, as an example, the incremental reactivities calculated when selected VOCs are added to an eight-component urban VOC mixture at various VOC/NOx ratios (Carter and Atkinson, 1989b). These data are in general agreement with the earlier modeling study of Dodge (1984) and show that the incremental reactivities of VOCs depend on the particular VOC and vary with the VOC/NOx ratio. In particular, the incremental reactivities are generally independent of, or increase with, the VOC/NOx ratio up to a VOC-/NOx ratio of ˜6-8. At higher ratios, the absolute magnitude of the incremental reactivities decreases with increasing VOC/NOx ratio, becoming dose to zero (or more negative) at high VOC/NOx ratio, where the formation of ozone is NOx-limited and VOC control becomes irrelevant (Carter and Atkinson, 1989b; Sillman et al., 1990b). The observed negative incremental reactivities are due to the presence of NOx or radical sinks in the chemistry of the VOC, with the NOx sinks being most important at high VOC/NOx ratios and the radical sinks most important at low VOC/NOx ratios (Carter and Atkinson, 1989b).

OCR for page 109
Page 155 TABLE 5-4 Calculated Incremental Reactivities of CO and Selected VOCs as a Function of the VOC/NOx Ratio for an Eight Component VOC Mix and Low Dilution Conditions   VOC/NOx, ppbC/ppb Compound 4 8 16 40 Carbon monoxide 0.011 0.022 0.012 0.005 Ethane 0.024 0.041 0.018 0.007 n-Butane 0.10 0.16 0.069 0.019 n-Octane 0.068 0.12 0.027 -0.031 Ethene 0.85 0.90 0.33 0.14 Propene 1.28 1.03 0.39 0.14 trans-2-Butene 1.42 0.97 0.31 0.054 Benzene 0.038 0.033 -0.002 -0.002 Toluene 0.26 0.16 -0.036 -0.051 m-Xylene 0.98 0.63 0.091 -0.025 Formaldehyde 2.42 1.20 0.32 0.051 Acetaldehyde 134 0.83 0.29 0.098 Benzaldehyde -0.11 -0.27 -0.40 -0.40 Methanol 0.12 0.17 0.066 0.029 Ethanol 0.18 0.22 0.065 0.006 Urban mixa 0.41 0.32 0.088 0.011 aEight-component VOC mix used to simulate VOC emissions in an urban area in the calculations. Surrogate composition, in units of ppb compound per ppbC surrogate, was ethene, 0.025; propene, 0.0167; n-butane, 0.0375; n-pentane, 0.0400; isooctane, 0.0188; toluene, 0.0179; m-xylene, 0.0156; formaldehyde, 0.0375; and inert constituents, 0.113. Source: Adapted from Carter and Atkinson (1989b).

OCR for page 109
Page 156 Theoretical studies of the ozone-forming potential of VOCs have investigated the factors that influence these reactivities. There are several approaches to dealing with this topic conceptually, but the computer modeling studies of Atkinson and Carter (1989b) and of Carter (1991) provide a useful framework for discussing the various factors involved. A generalized scheme for the degradation of VOCs in the troposphere in the presence of NOx is The rate of formation of ozone and oxidation of NO are then determined by the rate of formation of RO2 radicals and the number of molecules of NO converted to NO2 per RO2 radical generated. To a first approximation these processes can be dealt with independently, and the ozone-forming potential of a VOC then depends on • The rate at which the organic compound reacts in the troposphere. This reaction rate is equal to the inverse of its lifetime (i.e., its decay rate). The quantity of interest is the fraction of the emitted organic that has reacted Coy whatever route, photolysis, reaction with OH radicals, NO3 radicals, ozone, etc.) during the time being considered. • The reaction mechanism subsequent to the initial reaction(s) of the organic compound. Different aspects of the reaction mechanisms, for example, the VOC/NOx concentration ratio, become important under different conditions. The following features of a reaction mechanism affect the formation of Ozone: • The existence of NOx sinks (low values of g in Reaction 5.100) in the reaction mechanism lead to a lowering of reactivity with respect to ozone formation. Examples include the generation of alkyl nitrates from the reaction of alkyl peroxy radicals with NO (which competes with the pathway to form NO2 and the corresponding alkoxy radical),

OCR for page 109
Page 157 the formation of peroxyacetyl nitrate (PAN) or its analogues, and the formation of other organic nitrogen-containing compounds This aspect of the reaction mechanisms becomes important at high VOC/NOx ratios, where the availability of NOx becomes limiting, and the formation of organic nitro compounds competes with the formation of NO2 (with subsequent photolysis to generate ozone). • The generation or loss of radical species can lead to a net formation or loss of OH radicals (d > 1 or < 1, respectively, in Reaction 5.100), which in turn leads to an enhancement or suppression of OH radicals in the entire air mass and hence to an enhancement or suppression of overall reactivity of all chemicals through the effect on the formation rate of RO2 radicals (Reaction 5.99). The effects of radical formation or loss are most important at low VOC/NOx ratios, where the formation of ozone is determined by the rate at which RO2 radicals are formed. The alkenes, including isoprene and the monoterpenes of biogenic origin, react with ozone in addition to reacting with OH and NO3 radicals (Atkinson and Carter, 1984; Atkinson, 1991; Atkinson et al., 1990b; Tables 5-1 and 5-2), and this reaction process can act as a sink for ozone, especially under the low NOx conditions encountered in rural and clean atmospheres. However, because the reactions of ozone with alkenes lead to the generation of radicals

OCR for page 109
Page 158 from biogenic sources will act as ozone sinks unless NOx mixing ratios are < 0.1 ppb. To a first approximation the overall reactivity of VOC towards ozone formation is where The mechanistic reactivity then reflects the presence of radical and NOx sources or sinks in the VOC's reaction mechanism subsequent to the initial loss process. An extreme example is benzaldehyde, whose reaction mechanism subsequent to the initial OH radical reaction results in the loss of radical and NOx sinks, with the overall OH radical reaction having d = 0 (no net OH radical formation in Reaction 5.100) and (b-g) = 1 (consumption of one molecule of NOx per molecule of benzaldehyde reacted in Reaction 5.100). The approach in Chapter 8 to assess the importance of isoprene and other biogenic organic emissions in the formation of ozone in urban-suburban, rural, and remote air masses uses the kinetic reactivities of the VOCs measured in ambient air, and hence is based on the instantaneous rate of formation of RO2 radicals (Reaction 5.99). While the differences in the mechanistic reactivities of the various VOCs are neglected, this simpification is not expected to significantly alter the conclusions drawn in Chapter 8, especially since isoprene has a high positive mechanistic reactivity compared with the reactivities of the alkanes and aromatic VOCs (Table 5-5) that comprise the bulk of anthropogenic emissions observed in ambient air.

OCR for page 109
Page 159 TABLE 5-5 Calculated Incremental Reactivities and Kinetic and Mechanistic Reactivities for CO and Selected VOCs for Maximum Ozone Formation Conditions, Based on Scenarios for 12 Urban Areas in the U.S. Compound   Incremental reactivity, mole O3/mole C Kinetic reactivity, fraction reacted Mechanistic reactivity, mole O3/mole C Carbon monoxide 0.019 0.043 0.45 Methane 0.0025 0.0016 1.6 Ethane 0.030 0.049 0.61 Propane 0.069 0.21 0.34 n-Butane 0.124 0.37 0.34 n-Octane 0.081 0.75 0.107 Ethene 0.77 0.81 0.95 Propene 0.82 0.97 0.85 trans-2-Butene 0.81 0.99 0.82 Benzene 0.023 0.21 0.111 Toluene 0.106 0.64 0.17 m-Xylene 0.50 0.96 0.52 Formaldehyde 1.26 0.97 1.30 Acetaldehyde 0.70 0.92 0.77 Benzaldehyde -0.29 0.95 -0.31 Acetone 0.055 0.058 0.95 Methanol 0.147 0.16 0.93 Ethanol 0.19 0.44 0.42 Isoprene 0.70 1.00 0.70 a-Pinene 0.21 0.99 0.21 Urban mixa 0.28     (Table continued on next page)

OCR for page 109
Page 160 (Table continued from previous page) aAll-city average urban VOC mix. Source: Adapted from Carter (1991) Table 5-5 shows the calculated incremental reactivities and kinetic and mechanistic reactivities for selected VOCs for conditions that simulate those in 12 U.S. cities, with an all-city average VOC mix and the NOx concentrations varied to yield the maximum amount of ozone (Carter, 1991). Summary The general features of the atmospheric chemistry of ozone and its precursors are well understood. The chemistry of the polluted troposphere is considerably more complex than that of a less polluted, methane-dominated troposphere because of the presence of many VOCs of various classes. VOCs in the troposphere are photolyzed and react with OH and NO3 radicals and ozone to form organic peroxy radicals (RO2). Subsequent reactions lead to the conversion of NO to NO2, the generation of OH radicals, and the formation of ozone. The atmospheric chemistry of anthropogenic VOCs, including alkanes, alkenes, and aromatic hydrocarbons, is generally understood. The kinetics of the initial reactions of the majority of anthropogenic VOCs, and the photolysis rates of these VOCs, have been determined experimentally or can be reliably calculated. However, there are many uncertainties concerning the chemistry of aromatic hydrocarbons, carbonyl compounds, and long-chain alkanes and alkenes. Biogenic VOCs, including isoprene and the monoterpenes, are believed to foster episodes of high ozone concentrations in the presence of anthropogenic NOx. These VOCs are highly reactive toward OH and NO3 radicals and ozone. The mechanisms and products of the important reactions of these compounds are not well understood. An essential component of air quality models is the chemical mechanism that describes the series of reactions in the troposphere subsequent to emissions of VOCs and NOx. Chemical mechanisms are constructed using kinetic, mechanistic, and product data and refined by comparison with environmental chamber data. Two mechanisms recently developed for use in urban airshed models (those of Gery et al. [1988a, 1989] and Carter et al. [1986a]) have been tested against a common base of environmental chamber data; hence their close agreement does not guarantee their correctness. VOCs vary widely in the speed with which they react in the troposphere

OCR for page 109
Page 161 and the extent to which they promote or inhibit ozone formation. Several VOC reactivity scales have been proposed. One useful measure is incremental reactivity, defined as the amount of ozone formed per unit amount of VOC added to a VOC mixture representative of conditions in urban and rural areas. The ozone-forming potential of a VOC depends on the rate at which it reacts and on the reaction mechanism subsequent to the initial reaction. VOC reactivities generally decrease with increasing VOC/NOx ratios; at high ratios, ozone formation is NOx-limited and VOC control is irrelevant. The following areas of uncertainty need to be addressed before the role of VOCs in the formation of ozone can be assessed in detail: • The detailed tropospheric degradation reaction mechanisms of the aromatic VOCs are not well understood. In particular, the reactions of the hydroxycyclohexadienyl-type radicals under ambient tropospheric conditions require study. • The tropospheric reaction mechanisms of biogenic VOCs (for example, isoprene and the monoterpenes) must be investigated, and the gas- and aerosol-phase products determined under realistic atmospheric conditions. • Reactions under low-NOx conditions, such that reactions of organic peroxy radicals with HO2 and other RO2 radicals and of HO2 with ozone dominate over the reactions of HO2 and RO2 radicals with NO, require study. In addition to the need for further kinetic and mechanism data, environmental chamber studies carried out at low VOC/NOx ratios, close to rural ambient ratios, would be extremely useful. • The chemistry of long-chain alkanes and long-chain alkenes (for example, the 1-alkenes) needs to be elucidated. • The formation of carbonyl compounds during the atmospheric degradation reactions of VOCs and the atmospheric chemistry of these carbonyl compounds require further study. In particular, there is a need for absorption cross-sections, photodissociation quantum yields, and photodissociation product data (as a function of wavelength) for the carbonyl compounds. • The role of heterogeneous reactions in the chemistry of tropospheric ozone needs to be clarified, expanding on issues raised by Lelieveld and Crutzen (1990). Long-term research (over a period of 5 years or more) is needed to obtain the basic kinetic and product data required for formulation of detailed chemical mechanisms for the atmospheric degradation of anthropogenic and biogenic VOCs. Short- to medium-term programs (2-5 years) are needed to determine the reactivities of VOCs, either singly or as mixtures, with respect to ozone formation. Short-term studies (1-2 years) will be particularly useful

OCR for page 109
Page 162 for assessing the effects on ozone formation of conversion from gasoline to alternative fuels and for quantifying the effects of biogenic VOCs on urban, suburban, and rural ozone.