The Physics and Chemistry of Ozone
The ozone molecule consists of three oxygen atoms that are bound together (triatomic oxygen, or O3). Unlike the form of oxygen that is a major constituent of air (diatomic oxygen, or O2), ozone is a powerful oxidizing agent. Ozone reacts with some gases, such as nitric oxide or NO, and with some surfaces, such as dust particles, leaves, and biological membranes. These reactions can damage living cells, such as those present in the linings of the human lungs. Exposure has been associated with several adverse health effects, such as aggravation of asthma and decreased lung function.
Ozone was first observed in the Los Angeles area in the 1940s. The ozone that the ARB regulates as an air pollutant is mainly produced close to ground (tropospheric ozone), where people live, exercise, and breathe. A layer of ozone high up in the atmosphere, called stratospheric ozone, reduces the amount of ultraviolet light entering the earth’s atmosphere. Without the protection of the stratospheric ozone layer, plant and animal life would be seriously harmed. In this document, ‘ozone’ refers to tropospheric ozone unless otherwise specified.
Most of the ozone in California’s air results from reactions between substances emitted from vehicles, industrial plants, consumer products, and vegetation. These reactions involve volatile organic compounds (VOCs, which the ARB also refers to as reactive organic gases or ROG) and oxides of nitrogen (NOx) in the presence of sunlight. As a photochemical pollutant, ozone is formed only during daylight hours under appropriate conditions, but is destroyed throughout the day and night. Therefore, ozone concentrations vary depending upon both the time of day and the location. Ozone concentrations are higher on hot, sunny, calm days. In metropolitan areas of California, ozone concentrations frequently exceed regulatory standards during the summer.
From the 1950s into the 1970s, California had the highest ozone concentrations in the world, with hourly average concentrations in Los Angeles peaking over 0.5 ppm and frequent “smog alerts”. In the early 1970s, the ARB initiated emission control strategies that provided for concurrent and continuing reductions of both NOx and VOC from mobile sources and, in conjunction with the local air districts, stationary and area sources. Since then, peak ozone concentrations have decreased by more than 60 percent and smog alerts no longer occur in the Los Angeles area, despite more than a 35 percent increase in population and almost a doubling in vehicle miles traveled. However, most Californians still live in areas that do not attain the State’s health-based standard (0.09 ppm for one hour) for ozone in ambient air.
This chapter discusses the processes by which ozone is formed and removed, background ozone, the role of weather, and spatial and temporal variations in ozone concentrations. In addition, this chapter includes discussions of research that the ARB has been conducting in the following areas that affect ozone concentrations: reactivity, weekend/weekday effect, and biogenic emissions. Subsequent sections of this chapter include ARB websites for more information. The ARB also conducts more general research in atmospheric processes that affect air pollution; information is available at http://www.arb.ca.gov/research/apr/past/atmospheric.htm#Projects . For more extensive general information on the physics and chemistry of ozone, the reader is referred to Finlayson-Pitts and Pitts (2000), and Seinfeld and Pandis (1998).
1.1 Formation and Removal of Tropospheric Ozone
The formation of ozone in the troposphere is a complex process involving the reactions of hundreds of precursors. The key elements, as summarized in Finlayson-Pitts and Pitts (2000), and in Seinfeld and Pandis (1998), are discussed below.
1.1.1 Nitrogen Cycle and the Photostationary-State Relationship for Ozone
The formation of ozone in the troposphere results from only one known reaction: addition of atomic oxygen (O) to molecular oxygen (O2) in the presence of a third "body" (M). [M is any "body" with mass, primarily nitrogen or oxygen molecules, but also particles, trace gas molecules, and surfaces of large objects. M absorbs energy from the reaction as heat; without this absorption, the combining of O and O2 into O3 cannot be completed.]
O + O2 + M à O3 + M (1)
The oxygen atoms are produced primarily from photolysis of NO2 by the ultraviolet portion of solar radiation (hn).
NO2 + hn à NO + O (2)
Reaction 3 converts ozone back to oxygen and NO back to NO2, completing the "nitrogen cycle."
O3 + NO à NO2 + O2 (3)
Reactions 1 and 3 are comparatively fast. Therefore, the slower photolysis reaction 2 is usually the rate-limiting reaction for the nitrogen cycle and the reason why ozone is not formed appreciably at night. It is also one of the reasons why ozone concentrations are high during the summer months, when temperatures are high and solar radiation is intense. The cycle time for the three reactions described above is only a few minutes. Ozone accumulates over several hours, depending on emission rates and meteorological conditions. Therefore, the nitrogen cycle operates fast enough to maintain a close approximation to the following photostationary-state equation derived from the above reactions.
[O3]photostationary-state = (k2/k3) x [NO2]/[NO] (the brackets denote concentration)
The ratio of the rate constants for reactions 2 and 3, (k2/k3), is about 1:100. Assuming equilibrium could be reached in the ambient air and assuming typical urban pollution concentrations, a NO2 to NO ratio of 10:1 would be needed to generate about 0.1 ppm of ozone (a violation of the state one-hour ozone standard [0.09 ppm]). In contrast, the NO2 to NO emission ratio is approximately 1:10; therefore, the nitrogen cycle by itself does not generate the high ozone concentrations observed in urban areas. The net effect of the nitrogen cycle is neither to generate nor destroy ozone molecules. Therefore, for ozone to accumulate according to the photostationary-state equation, an additional pathway is needed to convert NO to NO2; one that will not destroy ozone. The photochemical oxidation of VOCs, such as hydrocarbons and aldehydes, provides that pathway.
1.1.2 The VOC Oxidation Cycle
Hydrocarbons and other VOCs are oxidized in the atmosphere by a series of reactions to form carbon monoxide (CO), carbon dioxide (CO2) and water (H2O). Intermediate steps in this overall oxidation process typically involve cyclic stages driven by hydroxyl radical (OH) attack on the parent hydrocarbon, on partially oxidized intermediate compounds, and on other VOCs. The Hydroxyl radical is ever-present in the ambient air; it is formed by photolysis from ozone in the presence of water vapor, and also from nitrous acid, hydrogen peroxide, and other sources. In the sequence shown below, R can be hydrogen or virtually any organic fragment. The oxidation process usually starts with reaction 4, from OH attack on a hydrocarbon or other VOC:
RH + OH à H2O + R (4)
This is followed by reaction with oxygen in the air to generate the peroxy radical (RO2).
R + O2 + M à RO2 + M (5)
The key reaction in the VOC oxidation cycle is the conversion of NO to NO2. This takes place through the fast radical transfer reaction with NO.
RO2 + NO à NO2 + RO (6)
R can also be generated by photolysis, which usually involves only VOCs with molecules containing the carbonyl (C=O) bond. The simplest VOC molecule that contains the carbonyl bond is formaldehyde (HCHO). Because formaldehyde enters into several types of reactions of importance for understanding ozone formation and removal, we will use it to help illustrate these reactions. The oxidation cycle for formaldehyde can be written in the following sequence of reactions.
OH + HCHO à H2O + HCO (7)
HCO + O2 à HO2 + CO (8)
HO2 + NO à NO2 + OH (9)
Hydroperoxyl radical (HO2) is generated by reaction 8, and the hydroxyl radical (consumed in reaction 7) returns in reaction 9 to complete the cycle. In addition, reaction 9 produces the NO2 required for ozone formation, as described above. Also, the carbon monoxide (CO) generated by reaction 8 can react like an organic molecule to yield another hydroperoxyl radical.
OH + CO à H + CO2 (10)
H + O2 + M à HO2 + M (11)
Another component that formaldehyde provides for smog formation is a source of hydrogen radicals.
HCHO + hn à H + HCO (12)
The hydrogen atom (H) and formyl radical (HCO) produced by this photolysis reaction yield two hydroperoxyl radicals via reaction with oxygen, as shown in reactions 8 and 11.
The reactions above comprise the simplest VOC oxidation cycle. Actually, hundreds of VOC species participate in thousands of similar reactions.
1.1.3 The Nitrogen Dioxide and Radical Sink Reaction
Another reaction is central to a basic understanding of ozone formation: the NO2 plus radical sink reaction that forms nitric acid.
NO2 + OH + M à HNO3 + M (13)
The previous discussion can be used to explain the typical pattern of ozone concentrations found in the urban atmosphere. Nitric oxide concentrations are relatively high in the early morning because the free radicals needed to convert the NOx emissions (which are primarily NO) to NO2 are not yet present in sufficient quantities. After sunrise, photolysis of formaldehyde (reaction 12) and other compounds starts the VOC oxidation cycle for the hundreds of organic gases present in the atmosphere. Subsequent NO to NO2 conversion by the peroxy radical (reaction 6) results in NO2 becoming the dominant NOx species. When the NO2 to NO ratio becomes large enough, ozone builds up. In the South Coast Air Basin (Los Angeles area), the highest ozone concentrations are observed in the San Bernardino Mountains, many miles downwind from the highest concentration of emission sources (freeways, power generating facilities, and oil refineries along the coast), because the reactions involving the organic gases are relatively slow. Meanwhile, NO2 concentrations decrease via the sink reaction 13.
Winds disperse and dilute both NOx and ozone. During the day, NOx is also diluted by the diurnal rising of the inversion layer, allowing for more mixing (see section 1.4 for further discussion). For ozone, however, the deepening mixing layer may cause its concentration to decrease on some days and increase on others. Although increased mixing almost always dilutes NOx, the effect of increased mixing on ozone concentrations depends upon whether higher concentrations of ozone are present aloft. Ozone that is trapped above the inversion layer overnight is available to increase the concentrations of ozone generated by the following day's emissions.
During the night, NO and ozone combine to form NO2 and oxygen via reaction 3 until either the NO or ozone is consumed. Nitrous acid or HONO is also present at night in polluted ambient air in California. Nitrous acid is produced from NO2 and water, and is also emitted from various combustion sources. Its levels are low during the day because sunlight breaks it down rapidly. At sunrise, sunlight causes gas-phase HONO to react rapidly to provide NO and OH, two key reactants in the formation of ozone. In this way, they help initiate ozone formation in the morning by being available to react with VOCs as soon as their emissions increase due to an increase in human activity.
Nitric acid (HNO3) was once thought
to be a permanent sink for NOx
radicals. However, nitric acid on surfaces may react with NO to regenerate NO2, which would
increase the ozone-forming potential of NOx
Although VOCs are necessary to generate high concentrations of ozone, NOx emissions can be the determining factor in the peak ozone concentrations observed in many locations (Chameides, 1992; National Research Council, 1991). VOCs are emitted from both natural and anthropogenic sources. Statewide, natural VOC sources dominate, primarily from vegetation. However, in urban and suburban areas, anthropogenic VOC emissions dominate and, in conjunction with anthropogenic NOx emissions, lead to the peak concentrations of ozone observed in urban areas and areas downwind of major urban areas.
The relative balance of VOCs and NOx
particular location helps to determine whether the NOx
behaves as a
net ozone generator or a net ozone inhibitor. When the VOC/ NOx
ratio in the
ambient air is low (NOx
relative to VOC), NOx
inhibit ozone formation. In such cases, the amount of VOCs tends to limit the
amount of ozone formed, and the ozone formation is called "VOC-limited". When
the VOC/ NOx
ratio is high
(VOC is plentiful relative to NOx), NOx
generate ozone. In such cases, the amount of NOx
tends to limit
the amount of ozone formed, and ozone formation is called "NOx
differ substantially by location and time-of-day within a geographic area.
Furthermore, the VOC/ NOx
near the ground might not represent the ratio that prevails in the air above the
ground where most of the tropospheric ozone is generated.
Photochemical reactivity, or reactivity, is a term used in the context of air quality management to describe a VOC's ability to react (participate in photochemical reactions) to form ozone in the atmosphere. Different VOCs react at different rates. The more reactive a VOC, the greater potential it has to form ozone. Examples of the more reactive VOCs in California’s atmosphere include propene, m-xylene, ethene, and formaldehyde. The ARB has helped to pioneer an approach to ozone control that considers the reactivity of each VOC constituent. In California’s urban areas, ozone formation tends to be limited by the availability of VOCs. Therefore, the reactivity-based regulatory approach has been applied in conjunction with reduction of NOx emissions. Reactivity-based regulations promote the control of those VOCs that form ozone most effectively, thereby guiding the affected industries (such as manufacturers of motor vehicle and consumer product formulators that use solvents) to choose the most cost-effective processes and designs to reduce VOC emissions. Further information is available from the ARB website, http://www.arb.ca.gov/research/reactivity/reactivityresearch.htm.
1.2 Background Ozone Concentrations in
Although little mixing occurs between the troposphere and stratosphere, stratospheric ozone intrusion occasionally causes localized ozone increases, especially at high mountain locations. Most of this intrusion is due to “tropopause folding”, which results from strong storms that draw stratospheric air down into the troposphere. In California, this tends to occur in spring. Because stable, stagnant conditions are necessary to support high ozone concentrations in urban California, this process generally does not contribute significantly to peak ozone concentrations. Stratospheric ozone intrusion is also due to general stratospheric subsidence. On a global basis, California is particularly prone to springtime stratospheric ozone intrusion from this process. However, this process is a relatively minor contributor to surface ozone concentrations in California, especially in the summer when ozone concentrations tend to be highest.
Another process leading to ground-level ozone arises from photochemical reactions involving natural precursors. Plants emit VOCs (see section 1.3), and soil microbes produce NOx that is vented into the air. Small amounts of NOx are also emitted from crops, apparently related to fertilizer application. Natural precursors may react with anthropogenic precursors to produce ozone concentrations that are of ambiguous origin. Where vegetation produces large amounts of VOCs, if anthropogenic NOx is also present, significant amounts of ozone can be produced.
Lightning contributes to the formation of ozone by heating and ionizing the air along the path of the discharge, thus forming the ozone precursor NOx. However, lightning tends to occur when meteorological conditions are not conducive to high ozone concentrations. Wildfires also contribute to ozone formation by producing NOx from combustion, and by distilling VOCs from vegetation. However, wildfires in California are not a major contributor to ozone pollution.
Finally, transport from outside of California contributes to in-state ozone concentrations. Cities in neighboring states and Mexico emit ozone precursors that impact California. In addition, urban plumes can be lofted high enough into the atmosphere to be entrained in global circulation and transported thousands of miles. In particular, ozone due to emissions in Asia, reaches California in springtime. However, this transport is not a major contributor to peak ozone concentrations in California because downward mixing of Asian ozone to the surface is precluded by the strong surface inversion usually present during high ozone episodes. Also, periods of effective long-range transport are generally restricted to spring, while high ozone concentrations due to local sources in California tend to occur in late summer and fall.
1.3 Effect of Vegetation on Ozone Concentrations
California's varied ecosystems interact with emissions related to human activity to influence ozone concentrations. Certain desert species, oaks, and pines emit substantial amounts of highly reactive VOCs, called biogenic emissions. Vegetation can either increase or decrease the ambient ozone concentration as the result of complex processes briefly described below.
Vegetation can reduce ozone concentrations by providing cooling and by removing pollutants. The shade provided by trees lowers ozone concentrations in several ways. It reduces the pollutant emissions from many sources (such as less evaporation of fuel from cooler parked vehicles). By cooling homes and offices, tree shade lowers emissions associated with electricity generation because less power is needed for air conditioning. In addition, cooling reduces the speed of chemical reactions in ambient air that lead to the formation of ozone.
Vegetation can also enhance the removal of ozone through deposition on plant
surfaces. The surfaces of leaves and pine needles allow for deposition of ozone
and NO2. Several different
factors affect pollutant removal, such as how long a parcel of air is in contact
with the leaf, and the total leaf area available for deposition. Also, rain
tends to reduce ambient ozone concentrations by washing out atmospheric gases as
well as gases deposited on leaves and needles.
To help understand the complex mechanisms by which vegetation influences ambient ozone concentrations, the ARB established a “Biogenic Working Group” (BWG). The BWG has developed vegetation maps, leaf biomass databases, emission factors, and a California-specific “biogenic emissions inventory through geographic information systems” (BEIGIS) that has satisfactorily accounted for observed ambient ozone concentrations. The information developed by the BWG will help the ARB to better model ozone formation, and to better determine the relative importance of VOC and NOx control. Additional information is available from the ARB website, http://www.arb.ca.gov/research/ecosys/biogenic/biogenic.htm.
1.4 Role of Weather in Ozone Air Quality
In the troposphere, the air is usually warmest near the ground. Warm air has a tendency to rise and cold air to sink, causing the air to mix, which disperses ground-level pollutants. However, if cooler air gets layered beneath warm air, no mixing occurs -- the air is stable or stagnant. The region in which temperature is so inverted is called an inversion layer. One type of inversion occurs frequently several thousand feet above the ground and limits the vertical dispersion of pollutants during the daytime. Another type of inversion occurs on most evenings very near the ground and limits the vertical dispersion of pollutants to a few hundred feet during the night. Pollutants released within an inversion tend to get trapped there. When the top of the daytime inversion is especially low [in elevation], people can be exposed to high ozone concentrations. Mountain chains, such as those downwind of California’s coastal cities and the Central Valley, help to trap air and enhance the air quality impact of inversions. Cooler air draining into the state’s valleys and ‘air basins’ also enhances inversion formation.
The direction and strength of the wind also affect ozone concentrations. Based on worldwide climate patterns, western coasts at California’s latitude tend to have high-pressure areas over them, especially in summer. By preventing the formation of storms, and by promoting the sinking of very warm air, these high-pressure areas are associated with light winds and temperature inversions, both of which limit dispersion of pollutants.
Because tropospheric ozone forms as a result of reactions involving other pollutants, the highest concentrations tend to occur in the afternoon. The photochemical reactions that create ozone generally require a few hours (see section 1.1) after the emissions of substantial VOC emissions, and are most effective when sunlight is intense and air temperatures are warm. Ozone concentrations in California are usually highest in the summer. The prevailing daytime winds in summer are on-shore, bringing relatively clean air from over the ocean to the immediate coastal areas, but carrying emissions of ozone precursors further inland. With the climatically favored clear skies and temperature inversions that limit the vertical dispersion of pollutants, these emissions are converted into ozone, with the highest concentrations tending to occur at distances a few tens of miles downwind of urban centers (ARB 2002).
During the periods of the year when the sunlight is most intense, much of California experiences a high frequency of inversions, relatively low inversion heights, and low wind and rainfall. As a result, no other State has more days per year with such a high potential for unhealthy ozone concentrations.
Additional information on the effects of weather on air pollution is available from the following textbooks:
1.5 Spatial and Temporal Variations of Ozone Concentrations
Spatial Variations of Ozone Concentrations
Due to the lack of ozone-destroying NO, ozone in rural areas tends to persist at night, rather than declining to the low concentrations (<30 ppb) typical in urban areas and areas downwind of major urban areas, that have plenty of fresh NO emissions. Ratios of peak ozone to average ozone concentrations are typically highest in urban areas and lowest in remote areas (ARB 2002). Within the ground-based inversions that usually persist through the night, ozone concentrations can be very low. In urban areas, emissions of NO near the ground commonly reduce ozone below 30 ppb. In rural areas, however, NO emissions are less prevalent and nighttime ozone may persist well above 30 ppb.
1.5.2 Temporal Variations in Ozone Concentrations
Ambient ozone concentrations tend to vary temporally in phase with human activity patterns, magnifying the resulting adverse health and welfare effects. Ambient ozone concentrations increase during the day when formation rates exceed destruction rates, and decline at night when formation processes are inactive. This diurnal variation in ozone depends on location, with the peaks being very high for relatively brief periods of time (an hour or two duration) in urban areas, and being low with relatively little diurnal variation in remote regions. In urban areas, peak ozone concentrations typically occur in the early afternoon, shortly after solar noon when the sun’s rays are most intense, but persist into the later afternoon, particularly where transport is involved. Thus, the peak urban ozone period of the day can correspond with the time of day when people, especially children, tend to be active outdoors.
In addition to varying during the day, ozone concentrations vary during the week. In the 1960s, the highest ozone concentrations at many urban monitoring sites tended to occur on Thursdays. This pattern was believed to be due to the carryover of ozone and ozone precursors from one day to the next, resulting in an accumulation of ozone during the workweek. In the 1980s, the highest ozone concentrations at many sites tended to occur on Saturdays and the “ozone weekend effect” became a topic of discussion. Since then, the weekend effect has become prevalent at more urban monitoring locations and the peak ozone day of the week has shifted to Sunday. Although ozone concentrations have declined on all days of the week in response to emission controls, they have declined faster on weekdays than on weekends. Thus, the peak ozone period of the week now tends to coincide with the weekend, when more people tend to be outdoors and active than during the week.
The causes of the ozone weekend effect and its implications regarding ozone control strategies have not yet been resolved. Almost all of the available data represent conditions at ground level, where the destruction of ozone by fresh emissions of NO is a major factor controlling ozone concentration. However, most ozone is formed aloft, and the air quality models used to analyze ozone formation have not demonstrated the ability to represent the ozone-forming system aloft with sufficient realism. In addition, several potentially significant photochemical processes are yet to be fully incorporated in simulation models. These deficiencies leave unresolved this fundamental question: does the ozone weekend effect occur because more ozone is formed (aloft) on weekend, because more ozone is destroyed (at the surface) on weekdays, or because ozone formation is more efficient on weekends? More information may be obtained from the ARB website, http://www.arb.ca.gov/aqd/weekendeffect/weekendeffect.htm .
Ozone concentrations also vary seasonally. Ozone concentrations tend to be highest during the summer and early fall months. In areas where the coastal marine layer (cool, moist air) is prevalent during summer, the peak ozone season tends to be in the early fall. Additionally, as air pollution controls have reduced the emissions of ozone precursors and the reactivity of VOCs, ozone concentrations have declined faster during times of the year when temperatures and the amount of sunlight are less than during the summer. Thus, the peak ozone season corresponds with the period of the year when people tend to be most active outdoors.
Also, ozone concentrations can vary from year to year in response to meteorological conditions such as El Niño and other variations in global pressure systems that promote more or less dispersion of emissions than typical. Although peak ozone concentrations vary on a year-to-year basis, peak ozone concentrations in southern California have been declining on a long-term basis, as anthropogenic emissions of VOC and NOx have declined. However, since the advent of the industrial revolution, global background concentrations of ozone appear to be increasing (Finlayson-Pitts and Pitts, 2000). This increase has implications regarding the oxidative capability of the atmosphere and potentially global warming processes (ozone is a strong greenhouse gas but is present at relatively low concentrations). Further discussion of these topics is beyond the scope of this document.
Air Resources Board/Planning and Technical Support Division. 2002. California Ambient Air Quality Data – 1980 – 2001. Sacramento, CA.: December. Data CD Number: PTSD-02-017-CD
Chameides WL, Fehsenfeld F, Rodgers MO, Cardelino C, Martinez J, Parrish D, Lonneman W, Lawson DR, Rasmussen RA, Zimmerman P, Greenberg J, Middleton P, Wang T. 1992. Ozone Precursor Relationships in the Ambient Atmosphere. Journal of Geophysical Research 97:6037-55.
Finlayson-Pitts BJ, Pitts JN. 2000. Chemistry of the Upper and Lower Atmosphere - Theory, Experiments, and Applications. Academic Press, San Diego, CA.
National Research Council. 1991. Rethinking the Ozone Problem in Urban and Regional Air Pollution, National Academy Press, Washington, DC.
Neiburger M, Edinger JG, Bonner WD. 1982. Understanding our Atmospheric Environment. W.H. Freeman & Co., San Francisco, CA.
Seinfeld JH, Pandis SN. 1998. Atmospheric Chemistry and Physics - from Air Pollution to Climate Change. John Wiley and Sons, New York, NY.
Smith TB, Lehrman DE, Knuth WR, Johnson D. 1997. Monitoring in Ozone Transport Corridors. Final report prepared for ARB/RD (contract # 94-316), July.
Environmental Protection Agency
Review of the California Ambient Air Quality Standard for Ozone