Wallace J.M., Hobbs P.V. Atmospheric Science. An Introductory Survey

Подождите немного. Документ загружается.

174 Atmospheric Chemistry

that is

For, N

N  2, N

 1.00  10

3

and K  1.40 

15

1

■

Improvements in visibility that frequently follow

periods of precipitation are due, in large part, to the

removal (i.e., scavenging) of particles by precipi-

tation. It is estimated that, on a global scale, precipi-

tation processes account for about 80–90% of the

mass of particles removed from the atmosphere.

Prior to the formation of precipitation, some of the

particles in the air serve as nuclei upon which cloud

particles (water and ice) can form (see Chapter 6).

As cloud particles grow, aerosols tend to be driven to

their surfaces by the diffusion field associated with

the flux of water vapor to the growing cloud particles

(called the diffusiophoretic force). Aerosol particles

less than 0.1



m are collected most efficiently by

diffusiophoresis. As precipitation particles fall

through the air, they collect aerosol particles greater

than 2



m in diameter with reasonable efficiency

by impaction. Aerosols are also removed by

impaction onto obstacles on the Earth’s surface

(such as newly washed automobiles).

The terminal fall speeds of particles 1



m diame-

ter are sufficiently large that gravitational settling (i.e.,

dry deposition) is important as a removal process. For

example, the fall speeds of particles 1 and 10



m in

diameter are 3  10

5

and 3  10

3

1

respectively. It is estimated that 10–20% of the

t 

2  1

(1.40  10

15

)10

 7140 s  1.98 h

t 

 1



Kt

mass of particles removed from the atmosphere is by

dry fallout.

5.4.5 Concentrations and Size

Distributions

One of the oldest and most convenient techniques

(which in various forms is still in widespread use) for

measuring the concentrations of particles in the air

is the Aitken

nucleus counter. In this instrument,

saturated air is expanded rapidly so that it becomes

supersaturated by several hundred percent with

respect to water (see Exercise 5.15). At these high

supersaturations, water condenses onto virtually all of

the particles in the air to form a cloud of small water

droplets. The concentration of droplets in the cloud

(which is close to the concentration of particles) can

be determined by allowing the droplets to settle out

onto a substrate, where they can be counted under a

microscope, or automatically by optical techniques.

The concentration of particles measured with an

Aitken nucleus counter is referred to as the Aitken

(or condensation) nucleus (CN) count.

Condensation nucleus counts near the Earth’s

surface vary widely from one location to another and

can also fluctuate by more than an order of mag-

nitude with time at any one site. Generally, they

range from average concentrations of 10

3

over the oceans, to 10

3

over rural land areas,

to 10

3

or greater in urban polluted air. These

observations, together with the fact that CN counts

decline with increasing altitude, support the view that

land is an important source of atmospheric aerosol,

with human and industrial activities being particu-

larly prolific sources.

Atmospheric aerosol particles range in size from

10

4



m to tens of micrometers. The averages of

numerous measurements of particle number distribu-

tions in continental, marine, and urban polluted air are

shown in Fig. 5.9. The measurements are plotted in the

form of a number distribution in which the ordinate

[dNd(log D)] and the abscissa (D) are plotted on log-

arithmic scales, where dN is the number concentration

of particles with diameters between D and D  dD.

John Aitken (1839–1919) Scottish physicist, although originally an apprentice marine engineer. In addition to his pioneering work on

atmospheric aerosol, he investigated cyclones, color, and color sensations.

Note that if the ordinate were linear in N (which it is not in Fig. 5.9), the concentration of particles within a diameter interval

d(log D) would be equal to the area under the curve in this diameter interval.

P732951-Ch05.qxd 12/09/2005 09:05 PM Page 174

5.4 Tropospheric Aerosols 175

Several conclusions can be drawn from the results

shown in Fig. 5.9:

•The concentrations of particles fall off very

rapidly as they increase in size. Therefore, the

total number concentration (i.e., the Aitken or

CN count) is dominated by particles with

diameters 0.2



m, which are therefore

referred to as Aitken nuclei or condensation

nuclei.

•Those portions of the number distribution curves

that are straight lines in Fig. 5.9 can be

represented by an expression of the form

or, taking antilogs,

(5.31)

where C is a constant related to the

concentration of the particles and 



is the

slope of the number distribution curve. The

value of



generally lies between 2 and 4.

Continental aerosol particles with diameters

larger than 0.2



m follow (5.31) quite closely

with



 3. A size distribution with



 3 is

called a Junge

distribution.

•The number distributions of particles shown in

Fig. 5.9 confirm CN measurements, which

indicate that the total concentrations of particles

are, on average, greatest in urban polluted air

and least in marine air.

•The concentrations of particles with diameters

2



m (giant particles) are, on average, rather

similar in continental, marine, and urban

polluted air.

Further insights into atmospheric particle size

distributions can be obtained by plotting particle

surface area distributions or particle volume distribu-

tions. In a surface area distribution the ordinate is

dSd(log D) and the abscissa is D ploted on a loga-

d(log D)

 CD

b

log

d(log D)

 const  b log D

rithmic scale, where S is the total surface area of

particles with diameters between D and D  dD.In

a volume distribution the ordinate is dVd(log D)

and the abscissa is D plotted on a logarithmic scale,

where V is the total volume of particles with diame-

ters between D and D  dD.

If particle densities are independent of their diam-

eter, the mass concentration dM of particles with

diameters between D and D  dD is proportional to

dN. If the number distribution follows (5.31) with



 3, D

dN  d(log D), therefore, dMd(log D)

has a fixed value. In this case, the particles in each

logarithmic increment of diameter contribute equally

to the total mass concentration of the aerosol. It fol-

lows from this result that, although particles with

diameters from 0.2 to 2



m (so-called large particles)

are present in much higher number concentrations

than giant particles, the large and giant particles in

Christian E. Junge (1912–1996) German meteorologist. Carried out pioneering studies of tropospheric and stratospheric aerosols

and trace gases. His book Atmospheric Chemistry and Radioactivity (Academic Press, 1963) initiated the modern era of atmospheric

chemistry research.

Diameter D (µ m)

–3

–2

–1

–3

–1

–2

–4

–5

–6

∝ D

–3

d(log D)

dN/d(log D) (cm

–3

)

Aitken

nuclei

(or CN)

Large

particles

Giant

particles

Fig. 5.9 Number distributions of tropospheric particles

obtained from averaging many measurements in continental

(red), marine (blue), and urban polluted (black) air. Also

plotted is Eq. (5.31) with



 3, with the line (dashed) dis-

placed from the other curves for the sake of clarity.

P732951-Ch05.qxd 12/09/2005 09:05 PM Page 175

176 Atmospheric Chemistry

continental air make similar contributions to the

total mass of the aerosol. However, despite their rela-

tively large number concentration, the CN contribute

only 10–20% to the total mass of the atmospheric

aerosol. This is because, as seen in Fig. 5.9, the CN do

not increase in concentration with decreasing size as

rapidly as indicated by (5.31).

Small fluctuations in the slope of a particle number

distribution about values of 2 and 3 (i.e.,



 2

and 3, respectively) appear as local maxima and min-

ima in the surface and volume distribution plots,

respectively (see Exercise 5.16). Shown in Fig. 5.10

are particle surface area and particle volume distri-

bution plots for continental and urban polluted air

in Denver, Colorado. These curves show far more

structure, in the form of maxima and minima, than

the number distribution plots shown in Fig. 5.9. The

maxima and minima in the curves in Fig. 5.10 are

associated with sources and sinks, respectively, of the

aerosol. The prominent maxima in the surface and

volume distribution plots in the particle size range

0.2–2



m diameter is due primarily to the growth

of the CN by coagulation into this size range,

together with particles left behind when cloud

droplets evaporate. Since the sinks for particles

0.2–2



m diameter are weak, particles in this size

range tend to accumulate in the atmosphere: hence,

the peak in the particle surface area and volume dis-

tribution plots for particles between 0.2 and 2



(a)

(b)

Diameter D (µ

600

400

200

–2

–1

dS/d(log D) (µ m

–3

)

Diameter D (µ m)

–2

–1

dV/d(log D) (µ m

–3

)

Fig. 5.10 (a) Typical particle surface area and (b) particle

volume distribution plots in urban-polluted air (black line)

and cleaner continental air (solid red line).

diameter is referred to as the accumulation mode.

Another peak, called the coarse particle mode, occurs

in the particle surface and volume distribution plots

for particles with D 1



m. This mode is attributable

to dusts and industrial processes, which produce fly

ash and other large particles, sea salt, and some bio-

logical materials.

Another common way of subdividing atmosphere

particles by size is to call particles with D 0.01



ultrafine, those with 0.01



m  D 2.5



m fine par-

ticles, and those with D 2.5



m coarse particles.

5.4.6 Residence Times

Figure 5.11 shows estimates of the residence times of

particles in the atmosphere as a function of their size.

Particles with diameter 0.01



m have residence

times 1 day; the major removal mechanisms for

particles in this size range are diffusion to cloud

particles and coagulation. Particles 20



m diameter

also have residence times 1 day, but they are

removed by sedimentation, impaction onto surfaces,

and precipitation scavenging. In contrast, particles

with diameters 0.2–2



m have strong sources (from

the coagulation of Aitken nuclei and particles left

behind by the evaporation of cloud droplets) but

weak sinks. Consequently, these particles have rela-

tively long residence times, reaching several hundred

days in the upper troposphere, but precipitation

scavenging and impaction reduce these residence

times to a few tens of days in the middle and lower

troposphere. It is for this reason that the accumula-

tion mode is in the size range of these particles (e.g.,

see Fig. 5.10).

5.5 Air Pollution

In urban and industrialized locations, anthropogenic

emissions can become so large that the concentra-

tions of various undesirable chemical species (pollu-

tants) cause significant deterioration in air quality

and visibility and can pose threats to human health.

Severe air pollution episodes, and associated visibility

reduction, occur when the rates of emissions or forma-

tion of pollutants greatly exceed the rates at which

the pollutants are dispersed by winds and vertical

transport or are removed from the atmosphere by

chemical reactions or deposition. Severe air pollution

episodes tend to occur in association with extended

intervals of light winds and strong static stability (see

Section 3.6 and Box 3.4).

P732951-Ch05.qxd 12/09/2005 09:05 PM Page 176

5.5 Air Pollution 177

5.5.1 Some Sources of Pollutants

Combustion (in power plants, smelters, automobiles,

and of wood, vegetation, etc.) is the largest source of

air pollutants. On a global scale, fossil-fuel combus-

tion is the major source of CO, CO

,NO

, and SO

Many other pollutants are released into the air by

combustion. For example, about 15% of the total

emissions of hydrocarbons are from anthropogenic

sources, most notably the burning of hydrocarbon

compounds (oil, natural gas, coal, and wood). Ideal

(or complete) combustion (or oxidation) of a hydro-

carbon fuel yields only CO

and H

O. However, for a

given quantity of fuel, a precise amount of oxygen is

required for complete combustion, and this ideal

combination of fuel and oxygen is rarely achieved.

Exercise 5.7 Determine the ratio of the mass of

dry air to the mass of isooctane (C

)—called the

air–fuel ratio—for ideal combustion.

Solution: The balanced chemical equation for the

ideal or complete combustion of C

can be writ-

ten as

Therefore, for complete combustion, 1 mol of C

reacts with 12.5 mol of O

or, because the molecular

weights of C

and O

are 114 and 32, respectively,

114 g of C

reacts with 12.5  32 or 400 g of O

We now need to calculate what mass of air contains

400 g of oxygen. The amount of oxygen in air by

volume (or by number of molecules) is 20.95% (see

Table 5.1). Because the apparent molecular weight of

dry air is 28.97, the amount of oxygen in air by mass

is 20.95  3228.97  23%. Therefore, the mass of

air containing 400 g of oxygen is 4000.23  1700 g.

Hence, for complete combustion, 114 g of C

reacts with 1700 g of air. Therefore, the air–fuel

ratio for ideal combustion is 1700114  15. ■

 12.5O

: 8CO

 9H

600

500

400

300

200

100

–3

–2

–1

Diameter D (µ m)

dS/d(log D) (µ m

–3

)

Desig-

nation

Aitken nuclei

Large

particles

Giant particles

Sources

Sinks

Resi-

dence

time

Combustion

Windblown

dusts

Gas-to-particle

conversions

Fly ash,

sea-salt, pollens

Coagulation of

Aitken nuclei

Giant particles

from

industries

Cloud droplet

evaporation

Coagulation

Capture by

cloud particles

Precipitation

scavenging

Dry fallout

Less than an hour

in polluted air

or in clouds

Days to

weeks

Hours to

days

Minutes to hours

Fig. 5.11 Schematic curves of particle surface area distributions for urban polluted air (black line), continental air (red line),

and marine air (blue line). Shown below the curves are the principal sources and sinks of atmospheric particles and estimates of

their mean residence times in the troposphere. [Adapted from Atmos. Environ. 9, W. G. N. Slinn, Atmospheric aerosol particles in

surface-level air, 763, copyright (1975), with permission from Elsevier.]

P732951-Ch05.qxd 12/09/2005 09:05 PM Page 177

178 Atmospheric Chemistry

Since 1981, gasoline-powered internal combustion

engines in the United States have used an oxygen

sensor in the exhaust system and a computer-

controlled fuel flow that keeps the air–fuel ratio

to within a few percent of that needed for ideal

combustion. This system, together with a catalytic

converter that reduces the emissions of particularly

undesirable substances, makes the emissions from

modern automobiles quite small. Older automobiles

tended to run with a fuel-rich mixture (i.e., a mixture

that contains less than the required amount of air for

ideal combustion) and these emit considerable pollu-

tants. For example, suppose the 12.5 mol of O

Exercise 5.7 is reduced to 11.25 mol. Then, 9 mol of

O is produced together with 5.5 mol of CO

and

2.5 mol of CO. In this case, more than one-third of

the carbon in the emissions is the highly poisonous

gas CO. Gas mileage is reduced because the fuel is

not completely burned. However, because the engine

is pumping less air, peak power can actually increase.

Even modern automobiles operate in this “fuel-rich

mode” for almost 1 min after the engine is started,

and for several seconds each time a strong accelera-

tion is required. Therefore, despite the fact that in

most countries automobiles account for only a small

percentage of the total fuel burned, they produce a

large fraction of the CO.

If an automobile has a broken catalyst or does

not have a catalyst, it will emit significant amounts

of unburned (and partially burned) hydrocarbons

(HC), many of which are toxic and carcinogenic.

Measurements of motor vehicle exhausts on high-

ways show that half of the emissions of CO and HC

come from less than 10% of the vehicles, namely

from those that are old or poorly maintained.

Hence the need for regular emission tests of older

automobiles.

The high temperatures associated with combustion

permit oxidation of atmospheric molecular nitrogen

to NO (referred to as thermal NO). At temperatures

below 4500 K, the reactions are

(5.32a)

(5.32b)

(5.32c)

Net: (5.33)

Because of the strong temperature dependence of

reactions (5.32a) and (5.32b), thermal formation of

 N

 M N 2NO  M

2N  2O

N 2NO  O

2O  2N

N 2NO  2N

 M N 2O  M

NO is highly temperature dependent. Under equi-

librium conditions, these reactions produce maxi-

mum NO concentrations at 3500 K (although

equilibrium is not attained in engines). As the

combustion gases cool rapidly to ambient tempera-

tures, the rates of the reverse reactions are reduced

drastically so that the NO concentration is “frozen

in” at the high temperature. An additional strong

source of NO during combustion is the oxidation

of nitrogen-containing compounds in the fuel

(fuel NO).

As shown in Section 5.5.2, NO

emissions from

automobiles play a key role in the formation of pho-

tochemical smog. However, over large geographic

areas, power stations and industries are generally

larger sources of NO

than automobiles.

Exercise 5.8 The balanced chemical equation for

the ideal combustion of a general hydrocarbon fuel

can be written as

(5.34)

If 1 mol of C

is completely burned, show that

mol of (unreacting) nitrogen will be con-

tained in the emissions. Hence, derive an expression

for the total number of moles of gases in the emis-

sions in terms of x and y.

Solution: From the percentages of O

and N

in air

by volume (namely 20.9 and 78) we have for air

(5.35)

Therefore, from (5.35), 1 mol of O

is associated with

3.73 mol of N

in air. From (5.34), mol of

will be associated with 1 mol of C

and

mol of O

burned. Therefore, the total

number of moles of gases in the emissions produced

by the combustion of 1 mol of C

is, from (5.34),

■

Combustion of fuels, particularly coal, dominates

the emissions of sulfur oxides, which are mainly SO

Heavy-metal smelters (e.g., Ni, Cu, Zn, Pb, and Ag)

can be large local sources of SO

There are also lower temperature sources of air

pollutants, e.g., leakages of hydrocarbons from natural

x 

 3.7(x 

3.7(x 

)

3.7(x 

)

number of moles of O

number of moles of N



20.9



3.73

3.7(x 

)





x 



: xCO



P732951-Ch05.qxd 12/09/2005 09:05 PM Page 178

5.5 Air Pollution 179

gas lines, organics from the evaporation of solvents,

and nitrogen compounds from fertilizers.

5.5.2 Smogs

The term smog derives from smoke and fog; it was

originally coined to refer to heavily polluted air that

can form in cities (generally in winter under calm,

stable and moist conditions) due to the emissions of

sulfur dioxide and aerosols from the burning of fossil

fuels (primarily coal and oil). The term is now applied

to all forms of severe air pollution, particularly in

urban areas, that restrict visibility.

a. London (or classical) smog

Prior to the introduction of air pollution abatement

laws in the latter part of the 20th century, many large

cities in Europe and North America regularly suf-

fered from severe smogs. The London smogs were

sufficiently notorious that such pollution became

known as London smog.

In the London (or classi-

cal) type of smog, particles swell in size under high

relative humidity and some of the particles serve as

nuclei on which fog droplets form. Sulfur dioxide gas

dissolves in the fog droplets where it is oxidized to

form sulfuric acid (see Section 6.8.5).

In December 1952, cold air moved from the

English Channel and settled over London, producing

a pollution-trapping inversion fog. Over the next

5 days London experienced its worst air pollution

episode. The smog was so thick that people had to

grope their way along the streets, buses crawled

along at a walking pace led by pedestrians with flash-

lights, and indoor events were canceled because the

stage could not be seen.

By the time the smog

had lifted, 4000 people had died of respiratory prob-

lems, and the smog was implicated in an additional

8000 deaths in the months that followed. In the Great

Smog, as it was called, SO

reached peak mixing ratios

of 0.7 ppmv (compared to typical annual mean mix-

ing ratios of 0.1 ppmv in polluted cities with large

coal usage), and the peak particle concentrations

were 1.7 mg m

3

(compared to 0.1 mg m

3

under

more typical urban-polluted conditions). Interest-

ingly, there is no direct evidence that even these high

concentrations of pollutants would, in themselves,

have caused fatalities (see Box 5.4).

After the Great Smog, laws were passed in Britain

and elsewhere banning the use of coal on open fires

for domestic heating and the emissions of black

smoke, and requiring industries to switch to cleaner-

burning fuels. Nevertheless, pollution is still a serious

problem in many cities in Europe and the United

States. Also, many large cities, particularly in devel-

oping countries (e.g., China, India), still suffer from

London-type smogs due to the burning of coal and

wood and to the lack of strict air pollution controls.

Even with stringent pollution controls, increasing

numbers of automobiles in large cities can produce

high concentrations of NO, which can then be con-

verted to NO

(5.36)

The NO

can then be involved in the production of

(see Section 5.5.2b). Because the magnitude of the

rate coefficient for (5.36) increases with decreasing

temperature, the production of NO

by (5.36) can

become significant in cities that have cold winters.

2NO  O

: 2NO

In 1661 the English diarist John Evelyn noted the effects of industrial emissions (and, no doubt, domestic wood burning) on the

health of plants and people. In the 17

century there were days when a plume of smoke half a mile high and 20 miles wide emanated from

London. Evelyn suggested that industries be placed outside of towns and that they be equipped with tall chimneys to disperse the smoke.

What are the pros and cons of this suggestion?

Such events were experienced by one of the authors (PVH) of this book.

Leonardo da Vinci noted, alongside a sketch of the

lung, “dust causes damage.” Some 500 years later

evidence began to mount supporting this conclu-

sion. There appears to be an increasing risk of lung

cancer and mortality with increasing concentra-

tion of the total mass of particulate matter (PM)

in the air, even at concentrations below 0.1 mg m

3

However, because other pollutants, such as SO

5.4 Effects of Particulates on Human Health

Continued on next page

P732951-Ch05.qxd 12/09/2005 09:05 PM Page 179

180 Atmospheric Chemistry

b. Photochemical (or Los Angeles) smog

During the second half of the 20th century emissions

from automobiles became increasingly important as

a source of pollutants in many urban areas. When

subjected to sunlight and stagnant meteorological

conditions, the combination of chemical species in

such strongly polluted urban air can lead to photo-

chemical (or Los Angeles-type) smog.These smogs

are characterized by high concentrations of a large

variety of pollutants, such as nitrogen oxides, O

,CO,

hydrocarbons, aldehydes (and other materials that

are eye irritants), and sometimes sulfuric acid as

well. The chemical reactions that lead to photo-

chemical smog are complex and still not completely

understood. However, some of the major chemical

reactions thought to be involved are described here.

Photochemical smogs are due to the interactions

between a variety of organic pollutants (e.g., hydro-

carbons such as ethylene and butane) and oxides of

nitrogen. The reactions start with (5.15) and (5.16),

which form O

. However, the O

is then depleted by

the rapid reaction (5.17). As shown in Exercise 5.5,

if there were no other reactions (5.15)–(5.17) would

lead to a steady-state concentration of O

given by

(5.18). However, (5.18) predicts O

concentrations in

urban-polluted air of only 0.03 ppmv, whereas typi-

cal values are well above this concentration and can

reach 0.5 ppmv. Therefore, other chemical reactions

must be involved in photochemical smogs. Most

effective are reactions that oxidize NO to NO

with-

out consuming O

, since this would allow net O

production, leading to a buildup of ozone concentra-

tions during the day. The OH radical can initiate a

chain reaction that can act in this way by attacking

the hydrocarbon pollutants in urban air, e.g.,

(5.37)

(5.38)

(5.39)

The resulting radicals, CH

in (5.37), H in (5.38), and

CO in (5.39), then become involved in reactions

that oxidize NO to NO

and regenerate OH. For

example, CH

from (5.37) can initiate the series of

reactions

(5.40a)

(5.40b)

(5.40c)

(5.40d)

The net results of (5.40), together with (5.37), are

(5.41)

 2O

 2NO : H

O  2NO

 HCHO

 NO : NO

 OH

O  O

: HCHO  HO

 NO : CH

O  NO

 O

: CH

OH  CH

CHO : H

O  CH

OH  CO : H  CO

OH  CH

: H

O  CH

and O

, generally increase with increasing concen-

trations of PM, it is difficult to separate the health

effects of the different pollutants. Weather also

plays a confounding role because of the effects of

relative humidity on particles and of temperature

on humans.

Most statistical studies of the effects of PM on

human health have utilized the total mass of PM

with diameters below 10



m (PM-10). However,

there are several reasons for suspecting that parti-

cles 1



m in diameter have a more deleterious

effect on human health than the larger particles

that dominate PM-10 measurements. Smaller par-

ticles can penetrate deeper into the small airways

of the lungs. There is also evidence that very small

5.4 Continued

particles (0.05



m diameter) have enhanced tox-

icity. A recent examination of lung tissue samples

preserved from the victims of the Great Smog of

London (see Section 5.5.2) has revealed high con-

centrations of small particles and metals such as

lead of the type that derive from coal combustion

and diesel engines. It might not be chance coin-

cidence that 1952 (the year of the Great Smog)

was also the year that London completed a change

from electric trams to diesel buses, which emit

many small toxic particles.

Despite the introduction of air pollution con-

trol strategies, small particles are still prevalent

in significant number concentrations in large

cities.

P732951-Ch05.qxd 12/09/2005 09:05 PM Page 180

5.5 Air Pollution 181

Thus, in this case, the oxidation of CH

results in

the oxidation of NO to NO

without consuming O

Reaction (5.41) produces formaldehyde (HCHO),

which is an eye irritant and a source of HO

[see

(5.26) and (5.27)].

Similarly, the acetyl radical (CH

CO) from (5.39)

is involved in a series of reactions leading to the

methyl radical CH

and the peroxyacetyl radical

(CH

COO

). The methyl radical oxidizes NO by

reactions (5.40), and the peroxyacetyl radical reacts

with nitrogen dioxide

(5.42)

The chemical species on the right-hand side of (5.42)

is the vapor of a colorless and dangerously explosive

liquid called peroxyacetyl nitrate (PAN), which is

an important component of photochemical smogs

and another major eye irritant. Other alkenes oxidize

NO to NO

without consuming O

and regenerate

OH and can do so faster than the aforementioned

reactions.

Shown in Fig. 5.12 are typical variations through

the course of a day in the concentrations of some of

the major components of photochemical smogs in

Los Angeles. Ozone precursors (NO

and hydrocar-

bons) build up during the morning rush hour, and

aldehydes, O

, and PAN peak in the early afternoon.

The role of polycyclic aromatic hydrocarbons

(PAHs) in air pollution and public health was first

realized in the early 1940s following the discovery that

organic extracts of particles from polluted air (e.g.,

COO

 NO

: CH

COO

benzo[a]pyrene) produce cancer in laboratory exper-

iments on animals. PAHs are emitted by diesel and

gasoline engines, coal-fired electric power-generating

plants, biomass burning, and cigarettes. They are pres-

ent in air as volatile particulates and gases. Reactions

initiated by OH in the day and NO

at night convert

gaseous PAHs to nitro-PAH derivatives, which are

responsible for 50% of the mutagenic activity of

respirable airborne particles in southern California.

5.5.3 Regional and Global Pollution

The effects of anthropogenic pollution now extend

to regional and global scales. Europe, Russia, the

northeastern United States, India, and large areas of

southeastern Asia are regularly covered by enormous

palls of polluted air that reduce visibility significantly,

produce acid deposition, soil and erode buildings

and other materials, and have deleterious effects on

human health, animals, and plants.

The fact that pollutants can be transported over

large distances is well illustrated by air pollution

episodes in the Arctic, known as arctic haze,which

can be as severe as those in cities. The pollutants

originate from fossil-fuel combustion, smelting, and

other industrial processes in northern Europe and

Russia. The pollutants are transported to the Arctic

by synoptic-scale flow patterns, primarily from

December to April. Because the arctic atmosphere

is generally stably stratified during this time of

the year, vertical mixing is limited; also, precipitation

is rare so that wet removal processes are weak.

Consequently, the pollutants can be transported over

large distances with relatively little dilution. A major

contributor to arctic haze is SO

, which is converted

to sulfate particles over the long transport distances.

Glacial records show that air pollution in the

Arctic has increased markedly since the 1950s, paral-

leling the increases in SO

and NO

emissions in

Europe. Interestingly, ice cores from Greenland show

unusually high lead concentrations from 500 B.C.

to 300 A.D. This is attributed to Greek and Roman

lead and silver mining and smelting activities, which

apparently polluted large regions of the northern

hemisphere. However, cumulative lead deposits in

the Greenland ice during these eight centuries were

only 15% of those caused by the widespread use

of lead additives in gasoline from 1930 to 1995.

Lead additives to gasoline were eliminated in the

United States in 1986, more than 60 years after their

introduction.

Time of day (hour)

Aldehydes

PAN

Hydrocarbons

Mixing ratio (ppbv)

812

200

400

Fig. 5.12 Typical variations during the course of a day of

some important pollutants in photochemical smogs in Los

Angeles. [Adapted from P. A. Leighton, Photochemistry of Air

Pollution, Academic Press, New York, 1961, p. 273, with

permission of Elsevier.]

P732951-Ch05.qxd 12/09/2005 09:05 PM Page 181

182 Atmospheric Chemistry

Anthropogenic influences are now apparent on a

global scale, as illustrated by the worldwide increase

in CO

concentrations since the industrial revolution

(Fig. 5.13a). Other trace gases (e.g., CH

and N

also show increasing concentrations worldwide over

the past 150 years or so (Figs. 5.13b and 5.13c).

5.6 Tropospheric Chemical

Cycles

The reservoirs of chemical species in the Earth sys-

tem are the solid Earth, the hydrosphere (oceans and

fresh water), the cryosphere (ice and snow), the

biosphere, and the atmosphere. Chemical species can

be transferred between these reservoirs. Because

under steady-state conditions, a chemical species can-

not accumulate indefinitely in any of the reservoirs,

there must be continual cycling of species through

the various reservoirs. This is termed geochemical

cycling.

The geochemical cycling of carbon has been dis-

cussed in some detail in Chapter 2. Here we consider

the tropospheric portions of the geochemical cycles

of two other important chemical species, namely

nitrogen and sulfur. We will be concerned with rela-

tively rapid interchanges involving the atmosphere

and other reservoirs (generally the oceans and the

biosphere). Many aspects of global chemical cycles

are not well understood; therefore, in many cases, the

magnitudes of the sources and sinks given here are

only rough estimates. Also, the relative importance of

the various species is not determined exclusively by

the magnitude of the emissions: atmospheric resi-

dence times must also be taken into account.

5.6.1 The Nitrogen Cycle

Nitrogen gas (N

) constitutes more than 99.99% of

the nitrogen present in the atmosphere, and N

O (an

important greenhouse gas) makes up more than 99%

of the rest of the nitrogen. The other nitrogen species

in the atmosphere are therefore present in only trace

concentrations (see Table 5.1), but are nonetheless

of crucial importance in atmospheric chemistry. For

example, NH

is the only basic gas in the atmosphere.

Therefore, it is solely responsible for neutralizing

acids produced by the oxidation of SO

and NO

;the

ammonium salts of sulfuric and nitric acid so formed

become atmospheric aerosols. Nitric oxide (NO) and

play important roles in both tropospheric and

stratospheric chemistry.

All of the nitrogen-containing gases in the air are

involved in biological nitrogen fixation and denitrifi-

cation. Fixation refers to the reduction and incorpo-

ration of nitrogen from the atmosphere into living

biomass. This transformation is accomplished by bac-

teria that are equipped with a special enzyme system

for this task; the usual product is NH

.The term fixed

nitrogen refers to nitrogen contained in chemical

compounds that can be used by plants and microor-

ganisms. Under aerobic (i.e., oxygen-rich) conditions,

other specialized bacteria can oxidize ammonia to

nitrite and then to nitrate; a process referred to as

nitrification. Most plants use nitrate taken through

their roots to satisfy their nitrogen needs. Some of the

nitrate undergoes bacterial reduction to N

and N

(termed denitrification), which returns fixed nitrogen

from the biosphere to the atmosphere. In this case,

nitrate acts as the oxidizing agent; therefore, denitri-

fication generally occurs under anaerobic conditions,

where oxygen is unavailable. Fixed nitrogen can also

be returned from plants to the atmosphere in the

(ppbv)

O (ppbv)

(ppbv)

a) Carbon dioxide (CO

)

b) Methane (CH

)

c) Nitrous oxide (N

360

340

320

300

280

310

300

290

280

270

1800

1600

1400

1200

1000

800

1800

19001700 2000

Years A.D.

Fig. 5.13 Changes in the concentrations of (a) CO

(b) CH

, and (c) N

O over 300 years deduced from analy-

sis of Greenland and Antarctic ice cores. [Adapted from

D. J. Wuebbles et al., “Changes in the chemical composition

of the atmosphere and potential impacts,” in Brasseur, Prinn

and Pszenny, eds, Atmospheric Chemistry in a Changing World,

P732951-Ch05.qxd 12/09/2005 09:05 PM Page 182

5.6 Tropospheric Chemical Cycles 183

form of N

O. Biomass burning returns fixed nitrogen

to the atmosphere as N

,NH

O, and NO

The principal sources and sinks of nitrogen-

containing species in the atmosphere are shown in

Fig. 5.14. Assuming that fixation and denitrification

of N

are in approximate balance, the main atmos-

pheric sources of nitrogen-containing species are

biogenic emissions from the terrestrial and marine

biosphere (NH

O, and NO

), decomposition of

proteins and urea from animals (NH

), biomass

burning and fossil fuel consumption (NO

,NH

and N

), and lightning (NO

). Nitrous oxide (N

is by far the largest reservoir of nitrogen in the

atmosphere and, except for N

, it has a much longer

residence time than other nitrogen species.

The main sinks of nitrogen-containing species

are wet removal by precipitation (NH

and NO

dry deposition (NO

and NH

), and the chem-

ical breakdown of N

O in the stratosphere. Because

anthropogenic sources of NH

O, and NO

(from

fossil fuel consumption, biomass burning, and agri-

cultural nitrate fertilization) are appreciable, human



activities may be causing significant perturbations in

the budgets of these species in the atmosphere.

5.6.2 The Sulfur Cycle

Sulfur species enter the atmosphere primarily in a

chemically reduced state. The most important reduced

sulfur gases are H

S, DMS, COS, and CS

.Their main

natural sources are biogenic reactions in soils, marsh-

land, and plants. Biogenic reactions in the oceans,

due primarily to phytoplankton, are sources of DMS,

COS, and CS

.When reduced sulfur gases are

released into the oxygen-rich atmosphere, they are

oxidized to SO

and then over 65% of the SO

is oxi-

dized to (the remainder of the SO

is removed

by dry deposition). Estimates of the fluxes of these

natural emissions of sulfur gases and their transfor-

mations to SO

and are given in Fig. 5.15.

It can be seen from Fig. 5.15 that DMS dominates

the emissions of sulfur gases from the oceans.

Most

of the sulfides in the air are oxidized rapidly by the

OH radical; therefore, their residence times are only

2

+ NH

[0.6 Tg(N)]

Biomass

burning

[OCEANS]

26 15

Soil emissions

including

fertilizer loss

140

[2 x 10

Tg(N)]

Biomass

burning

Biomass burning

and fossil fuel

consumption and

industrial

processes

Fixation

Denitrification

Dry

removal

of NO

140100

(

Fixation

Denitrification

100

[11 x 10

Tg(N)]

[1300 Tg(N)]

by bacteria in plants,

soils and water

Soil emissions

from nitrification,

denitrification

and fertilizers

Fossil

fuel

combustion

To the stratosphere

[0.15 Tg(N)]

6 12 20

Wet

removal

of NO

Decomposition

of protein,

urea from

domestic (23)

and wild

animals (3)

[OCEANS]

Lightning

Dry

removal

of NO

and

Wet

removal

of NH

and

0.5

OCEANS

CONTINENTS

[9 x 10

Tg(N)]

Fig. 5.14 Principal sources and sinks of nitrogen-containing gases in the atmosphere. Numbers alongside the arrows are esti-

mates of average annual fluxes in Tg(N) per year; various degrees of uncertainty, some quite large, are associated with all of the

fluxes. Numbers in square brackets are total amounts of the species in the atmosphere. [Adapted from P. V. Hobbs, Introduction

to Atmospheric Chemistry, Camb. Univ. Press, 2000, p. 148. Reprinted with the permission of Cambridge University Press.]

An enormous amount of sulfate is ejected into the air from the oceans in the form of sea salt. However, because these are relatively

large particles, they are recycled quickly back to the ocean.Therefore, they do not have a significant effect on the global sulfur cycle.

P732951-Ch05.qxd 12/09/2005 09:05 PM Page 183