Stratospheric Pollution and Ozone Depletion

Tun-Li Shen, Paul J. Woo!dridge, and Mario J. Molina

Dept. of Earth, Atmospheric and Planetary Sciences,

and Department of Chemistry

Massachusetts Institute of Technology

Cambridge, MA 02139

L Introduction

Formation and destruction of ozone in the stratosphere

2.1. Chapman's mechanism

2.2. HOx chemistry

2.3. NOx chemistry

2.4. ClO, chemistry

2.5. Coupling of HO ,/NO x/CIOx reactions 2.6. SO, chemistry

High-latitude ozone !oss and heterogeneous chemistry

3.1. Free radical chemistry of the polar stratosphere

3.2. Polar stratospheric clouds (PSCs) and sulfate aerosols

3.3. Reactions involving PSCs and sulfate aerosols

Field observations and computer models

4.1. Measurement techniques and results

4.2. Observations of the polar stratosphere

4.3. Modeling and assessment

The future of stratosphere ozone

5.1. Determining the lifetimes of halocarbons in the atmosphere

5.2. Ozone depletion from CFC substitutes - the Ozone Depletion Potential 5.3. Other potential threats: supersonic aircraft, etc.

5.4. International regulations on ozone depieting compounds

1. Introduction

The lowest !ayers of the atmosphere, as defined by temperature structure, are the

troposphere (0- -10 km), the stratosphere (-10-50 km), and the mesosphere (50-150 km). An important feature of the stratosphere is that it contains about 90% of the total amount

of ozone in the atmosphere, the significance of which is its strong absorption of ultravio!et

radiation. The absorption is essentia!ly complete between 200 and 290 nm and less strong

in the 290 to 330 nm region. The heat from this absorption of solar radiation by ozone

causes the temperature to increase with altitude, as illustrated in Figure 1. This "inverted"

temperature profile is largely responsible for the dynamic stability of the stratosphere

toward vertical mixing, in contrast to the rapid mixing of the troposphere.

This chapter focuses on the link between the amount of ozone in the stratosphere and

changes in its chemical composition due to anthropogenic sources, i.e. stratospheric pollution.

2. Formation and destruction of ozone in the stratosphere

2.1. Chapman's mechanism

The formation of ozone in the stratosphere is initiated by photodissociation of

molecular oxygen by solar radiation at wavelengths shorter than 242 nm, within the

Herzberg continuum (200-220 nm) and the Schumann-Runge band (185-200 nm) of

oxygen's absorption spectrum:

02+hv(<242nm)-'0+0 (1)

The oxygen atoms released from reaction (1) rapidly combine with oxygen molecules to form ozone:

0+ 02±M->03+M

(2)

where M is N2 or 02. Photodissociation generates atomic oxygen with unit quantum yield:

03+hv(<300nm)-> 0+02 (3)

2

This is the primary source of atomic oxygen in the stratosphere. The net result of reactions

(2) and (3) is the conversion of solar energy to heat; ozone is not destroyed in this

process. However, ozone is very reactive and can be destroyed by various other processes

such as by reaction with oxygen atoms:

0+03—*202 (4)

which converts "odd" oxygen (defined as the sum of ozone and atomic oxygen) back to

"even" oxygen, 02. Normal ozone abundances peak in the 6-8 ppmv (part-per-million-by-

volume) range at an altitude around 20-25 km. Colunm amounts, i.e. vertical integrais,

typically vary from 290 to 310 Dobson units (1 Dobson Unit=10 3 atm cm2.7x10 16 molecules cm-2) on a giobaily averaged basis.

The aboye four steps form the model proposed by Sidney Chapman in 1930. For

twenty years this simple model, involving only oxygen species, appeared sufiicient to

explain the balance between the production and the destruction of ozone in the

stratosphere. It is of interest to note that in the dark, such as during the polar night, there

should be no production or destruction according to this mechanism.

Refinements in measurements revealed that ozone abundances were noticeably smaller

than those predicted by Chapman's reactions. In the 1950s and 1960s other ozone

destruction pathways were proposed, based on the photochemistry of atmospheric water

and the influence of the reactive radicais on the distribution of odd oxygen in the

atmosphere. More recently the importance of nitrogen oxides {Crutzen, 1970; Johnston,

19711 and chlorine compounds [Stolarski and Cicerone, 1974; Molina and Rowland,

1974; Rowland and Molina, 19751 has become apparent, and of great concern due to the

anthropogenic perturbations in the concentrations of these chemicals in the stratosphere.

The main ozone destruction processes to be added to Chapman's reactions can be

considered as catalytic cycles of the form

x+03—x0+02

x0+0—x+02

net: O + 03 - 202

where the free radical X can be H, OH, NO, Cl, or Br. Note that X is not consumed by

these two reactions and that each cycle leads to the destruction of two odd oxygen

species. Other free radical species are less important for ozone destruction due to either

3

low abundances, endothermic reactions, or rapid transformation to nonreactive forms (e.g.

fluorine species to the strongly bound HF). The relative importance of these catalytic

cycles is determined by the concentrations of the active radicais and the reaction kinetics.

Discussed next are the atmospheric photooxidation pathways of source gas, such as

H20, CH4, N20 and the CFCs, to yield the radicais involved the catalytic cycles; together

with the interconversion of these active radicais to "reservoir" species, which are those

that do not participate directly in ozone destruction reactions.

2.2. HO chemistry

HOx refers to the family of water-based free radicais H, OH, and H02. Water in the

stratosphere is scarce: at the temperature of the tropopause, see Figure 1, its low vapor

pressure results in very little being transported up from below, a "freeze-drying" or "coid-

trap" effect. The water mixing ratio is at most -5-6 ppmv (of the order of 1000 times less

than in the troposphere), with roughly a haif coming from the multistep oxidation of C144:

CH4 + 202 -+ -> 2H20 + CO2

HO is produced predominantly by reactions between water or methane and

electronically excited oxygen atoms generated by the photodissociation of ozone, reaction

(3'):

03 + hv ( < 300 nm) - O('D) + 02 (3') O('D) + H20 -* 20H (5)

O(1D) + CH4 - OH + CH3

The main HO catalytic cycle responsible for the removal of ozone is:

0H+O3—HO2+O2 (6)

H02+0—OH+02 (7)

net: 0 +03 - 202

The relative importance of this and the several other HO x catalytic cycles which can be

thought of depends on the altitude under consideration. For example, in the upper

4

stratosphere, where the abundances of O and H are relatively high, the foliowing cycle

destroys odd oxygen:

OH+O—*H+02

H+02+M—>H02+M

H02+0—OH+02 (7)

net: 20—O2

On the other hand, in the lower stratosphere, the foliowing catalytic cycle, in which O

atoms do not participate, becomes important:

OH+O3—HO2+O2 (6)

H02 +03 - OH + 202 (10)

net: 203 -* 302

The efficiencies of the aboye catalytic cycles are strongly affected by the 011I1102 ratio. In

addition, both OH and H02 also play critical roles by interacting with species in the NO

and ClO families (see §2.5).

An additional HO reservoir species, besides H 20, is hydrogen peroxide; it affects the

concentrations of OH and H02 through the foliowing reactions:

H02 + H02 --> H202 +02

H202 + hv—> OH + OH

H202 + OH -> H20 + H02

Hydrogen peroxide concentrations in the stratosphere are quite small, as it is easily

photolyzed.

2.3. NO chemistry

The sum of NO and NO 2 is referred to as NO N. Transport of NO to the stratosphere

from below is negligible due to the short residence time for these species in the

troposphere (ca. 1 day). The chief natural source of NO in the stratosphere is N20, which

is produced by biological processes in soil and is essentially inert in the troposphere.

5

Minor sources of stratospheric NO include galactic cosmic rays and solar proton events

[Crutzen, 1979. In addition, as discussed in §5.3, direct injection of NO into the

stratosphere by proposed fleets of high-altitude aircrafi is a potential future source.

About 95% of N20 in the stratosphere is destroyed by photolysis:

N20 + hv—> N2 + O('D)

and the remainder reacts with O('D):

0(1D)+N20 —2NO

0(1D)+N20 —N2+O2

In the upper stratosphere the most important cycle controlling ozone leveis is the

foliowing:

NO+O3—NO2+O2 (11) NO2+0—NO+02 (12)

net: 03+0—*202

An interesting aspect of nitrogen oxide chemistry is the diurnal, seasonal, and

latitudinal behavior of the interconversion between NO and the reservoir species N205,

which is determined by reactions whose relative importance depends on the available

sunlight:

nighttime reactions- NO + 03 - NO2 + 02 (11)

NO2+03—NO3+02

NO2+NO3+M—*N205+M

daytime reactions- NO2 + hv(<390 nm) -+ NO + 0

N205 + hv—. NO2 + NO3

NO3+hv—NO+02

NO3+hv-+NO2+0

Thus NO increases at sunrise and decreases foliowing sunset. Observations of the

interconversion of the reactive nitrogen species provide useful checks for model

simulations of atmospheric chemistry.

2.4. ClO,. chemistry

The following catalytic cycle involving chiorine species is very efficient in destroying ozone:

Cl+O3-ClO+O2 (19)

ClO+O-Cl+O2 (20)

net: 03+0 - 202

The ClO, cycle is initiated when Cl atoms are released by photodissociation of chlorinated

compounds, and can be interrupted by the formation of reservoir species, mainly HC1, C1ONO2 and HOC1. Atomic chiorine may abstract hydrogen from CI-L, H2, etc. to form HC1:

Cl+CH4-HCl+CH3

The recombination of C1O and NO2 radicais forms C1ONO2, and the reaction of CIO and H02 forms HOCL (see §2.5 below). As a reservoir species, HOC1, though, is generaily Iess important due to its rapid photolysis.

HCI from the oceans or from voicanic eruptions is a very minor source of chlorine in

the stratosphere, because the water that is always present with these sources very

effectively dissolves the HC1 and returns it to the surface in ram [see e.g., Tabazadeh and Turco, 1993]. Rather, surface emissions of CH3C1 comprise the chief natural source of

chiorine to the stratosphere. Anthropogenic sources, namely manufactured

chlorofluorocarbons (CFCs), provide a larger source, which poses a threat to the ozone layer [Molina and Rowland, 19741.

CFCs are used as refrigerants, solvents, aerosol propellants, biowing agents for plastic

foams, etc. These chemicais are very inert: they have no troposphere sinks and are not

water soluble, and thus are mixed rapidly throughout the troposphere and gradually

through the stratosphere, where eventually they photodissociate, as depicted in Figure 2. CFCs need not risc aboye most of the atmospheric 02 and 03 because they can be

7

photodissociated by wavelengths which penetrate to lower altitudes, in the 185-2 10 nm

spectral window (between strong absorptions of 02 to shorter wavelengths, and of 03 to

longer). As examples consider the two most prominent CFCs that reach the stratosphere,

CFC-1 1 (CFC13) and CFC-12 (CF2C12):

CFC13 + hv -* CFC12+ Cl

CF2C12 + hv -* CF2CI + Cl

Subsequent reactions of the CFCI2 and CF2C1 radicais lead to the rapid release of the

remaining chiorine atoms, which then initiate the catalytic cycles.

A bromine cycie also exists, analogous to that involving reactions 19 and 20. As with

chiorine, there are natural sources of stratospheric bromine (mainly methyl bromide, CH3Br) as well as man-made: the "Halons" CF 3Br, CF2C1Br, etc., used in fire extinguishers; and CH3Br, used as a fumigant. Bromine species are present in the

stratosphere in small amounts, relative to chiorine species, but molecule for molecule they

are much more effective in destroying ozone, because the bromine reservoirs are

significantiy less stable (HBr forms much more slowly than HC1; also, BrONO2 photolyzes

more readily than C1ONO2; etc.).

2.5. The coupling of H01N0/Ci0

The importance of a family of species and its cycles depends on the abundance of the

active radicais that initiate the chain reactions, which in turn depends on the amount held

in reservoir form and the likeiihood of radical regeneration. The segregation of reactions

of various species into individual catalytic cycies is a useful tool for understanding the

nature of the ozone destruction processes. Note, however, that the choice of the cycles is

arbitrary; the concentration of any given species is dependent on all of the reactions in

which it participates. An important set of reactions is that which couples the different

radical families, examples of which are discussed below.

The most important reactions coup!ing H0 and N0 are

H02+N0-0H+NO2 OH + NO2 + M - HNO3 + M

HNO3+hv-0H+NO2 (2)

O2+MH02NO2

(24)

02

0202 02O2

(25)

hv (26)+02

02NO2+°22

Reactior stroflY affeCts the paio

of 0 and 02, and henCe the relative

O catalY contrbut s of the vañOUS 1

'° cycles tied-UP hroUgh reaCtiofl

The e ctiveness of the N0 cycle is reduced whefl 02 is

22, a key ctiOfl 0

p1ing the Ox and Ox cycleS. ReaCtiofl 24 alsO ties-UP NO2; the

reverse of iese tWo reactiofl5 is the release of 02 by photolY' i.e., reaCti0 23 and

25. Reacti0 that

oupie IlOx and C10 are

fl20 + Cl

(27)

(28)

02 + ClO HOC1 +02

As can be seen, an increaSe of Ofl has oppOSite effects on the Ox and C1Ox cataliC

cycles. As 0 increaSes, Cl is conVe O

Cl by reacti0fl (27), and the impaCt of C1Ox

is enhace On the other hand, O transfOS O2 into its rese0 speCies

03 by

reaCtiofl (22), ecreaSg the effeCt ofOx n ozone depleti0 Reactiofl5 that play key roles in the interaCtiofl ofOx ad ClO cycles are

CIO + O O2 + Cl

ClO + 1-102 + M 2 Cl0O + M

C1OO2 0

The ClOx catal ccle is influenC most by re actiofl (29) in the upper

st 0rat05Pe and

by reaCti0 (30) in the lower strat0sPe. As mentioned aboye, the formatiofl of CIO2

via reaCtiofl (30) provides a temP0ra reseo for CIO.

2.6 S0 x chemistry

OCS and so2 are belieV to be the prima sources of stratO5P

sulr. Beiflg

releaS mainlY by vañOUS jolOgiCai

proCeSSes OCS wch is ve long-

lived in the trop05P' provideS a conti OUS s

ource of su1t in the stratoSP re, whete continuous

ly

it is photo1Y5 The second suir sourCe, SO2, is impoaflt when inected direCtlY into

the stratosPh by major volca epti0fl5. The role of sulftr hotoce5t has been

9

investigated in sorne detail foliowing the eruptions of El Chichon in 1982 and Mt.

Pinatubo in 1991, and the increases in the stratospheric sulfur loading (more than an order

of rnagnitude) foliowing those major eruptions have been closely monitored along with the

various chemical perturbations and ozone changes.

In the stratosphere SO 2 first reacts with OH to begin its oxidation to H 2 SO4, which

proceeds via the following scheme [Stockwell and Calvert, 1983]:

S02 +OH+M—HS03 +M HS03 +02 - H02 + S03

S03 +H20—*--+H2 SO4

Note that no HO radicals are consumed in this process. The detailed mechanisrn of

reaction 34 is not yet known, but it is clear that the extremely hygroscopic product,

H2 SO4, combines with water vapor to forrn surfate aerosol particles, which play a rnajor

role in stratospheric chemistry by providing surfaces for heterogeneous reactions, as well

as by being involved in the formation of polar stratosphere clouds. These play a crucial role in the near complete seasonal destruction of ozone which occurs over Antarctica.

3. High-latitude ozone loss and heterogeneous chemistry

Before 1985 there was scant evidence of a long-term decline in stratospheric ozone

levels. However, the discovery of the Antarctic "ozone hole" in 1985 changed this: ozone

column measurements showed a dramatic decline of the rnonthly mean value in October at

Halley Bay from 300 to 350 DU (Dobson Units) in the mid-1970s, to values lower than

200 DU [Farman et al., 19851. Furthermore, ozone vertical profiles from bailoon

measurements indicated that the ozone depletion was occurring at altitudes from about 10

to 20 km [Hofhiann et al., 1987], see Figure 3. Based on the steady increase of

tropospheric and stratospheric halocarbons, Farman et al. suggested a possible link

between the growth of active chlorine in the stratosphere (released by CFCs) and the

ozone losses. The discovery of the ozone hole was surprising not only because of its

magnitude but also its location: based only on gas phase chemical models of the

stratosphere, it was anticipated that chlorine-initiated ozone depletion would occur

predominantly at middle and lower latitudes, and at altitudes between 35 and 45 km {see,

e.g., NRC, 19821, not in the lower polar stratosphere.

An initial question was whether the Antarctic "ozone hole" is merely a natural

phenomenon, only never before noticed. Overwhelming evidence, however, has since

10

pointed to anthropogenic emissions, specifically CFCs, as the cause [see, e.g., Solomon,

1990; Anderson et al., 1991; WMO, 19921. Early theories put forth to explain the origin

of the Antarctic ozone hole included as principal causes solar cycles, atmospheric

dynamics, and chemistry. Callis and Natarajan [1986] suggested that the decline in ozone

could be related to NO generated in the upper stratosphere by solar uy radiation, in

connection with the 11 -year solar cycle. This turned out to be inconsistent with the

observations that ozone depletion takes place in the lower, rather than upper polar

stratosphere; and that the concentration of NO there is remarkably low. The central idea

of the dynamics theory [e.g., Tung et al., 19861 was that, upon first sunrise, warming of

the Antarctic stratosphere leads to a net upward lifting of ozone-poor air from the

troposphere or lower stratosphere, leading to the springtime decline of ozone. Verification

of this theory would have come from the observations of upward air flow; however, tracer

(e.g. CFCs, CH4 and N20) studies revealed a strong downward flux within the polar

vortex, where ozone depletion is most severe. The chemical theory, as discussed below, is

an extension of the known radical-initiated catalytic cycles, but with consideration of the

unique conditions of the polar stratosphere.

By 1988 it also became clear that significant ozone depletion was taking place at high

northern latitudes in the winter months—e.g., more than 20 % between 530 and 64°N. The

depletion is not as severe and localized as over Antarctica; but is certainly signiflcant

[WMO, 1990a].

3.1. Free radical chemistry in polar stratosphere

In the polar stratosphere very little ozone is produced, as the large solar zenith angle

(low Sun elevation) results in essentially no photodissociation of oxygen. This also means

that catalytic cycles involving atomic oxygen are ineifective. Several catalytic cycles which

result in the net reaction

203 - 302

and which do not require the presence of 0 atoms have been propo sed since the discovery

of the ozone hole.

11

Solomon et al. [1986] suggested a cycle based on the coupling of the H0 and Cl0

families:

OH + 03 -> H02 + 02 (6)

Cl+03-*ClO+02 (19)

ClO + H02 - HOC1 +02 (28)

HOC1+hv-*OH+Cl

net: 203 + hv-> 302

The importance of this cycle can be estimated from measurements of HOC1. Observations

of this species in the Antarctic stratosphere, however, indicate that this cycle makes only

minor contributions to ozone depletion [see, e.g., Toon and Farmer, 19891.

Another proposal was a catalytic cycle involving bromine [McElroy et al., 19861:

Cl+o3->clo+o2 (19)

Br+03-BrO+O2 (35)

ClO + BrO -* Cl + Br +02 (36)

net: 203 -* 302

Actually, the reaction between ClO and BrO has three channeis:

C10+BrO-+C1+Br+02

ClO + BrO -* BrCI +02

ClO+BrO-+Br+OCIO

Laboratory studies have shown that the first and third channeis are equally fast, and that

the production of BrC1 seems to be a minor channe! [see, e.g., HilIs et al., 1987; Friedi and

Sander 1989; Turnipseed et al., 1991], important only at night—photolysis of this species

occurs rapidly during the day:

BrCl+hv-.Br+Cl (39)

12

A mechanism involving the formation of a ClO dimer was also proposed [Molina and Molina, 19871:

dO + ClO + M -* C1202 + M C1202 + hv - Cl + Cl +02

2[Cl+03 —Cl0+02 ] (19)

net: 20 3 +hv-+302

Like the reaction between BrO and ClO, the reaction between two ClOs has three bimolecular channeis:

ClO+ClO—Cl2 +O2

ClO + ClO - Cl + CiOO

do + do -+ Cl + OCiO

These bimolecular reactions are very siow, and hence of little atmospheric importance. On

the other hand, in the lower polar stratosphere, the termolecular reaction that ieads to the formation of C1202, (40), can occur efficientiy as it is facilitated by the higher pressures and lower temperatures there.

The structure of C1202 formed by this reaction is important. The asymmetric dimer,

CIOC1O, wouid not lead to ozone destruction. In the case of the symmetric dimer

(chiorine peroxide), there are two possible channels for photodissociation:

ClOOCl + hv-+ Cl + ClOO (42) C1000 + hv-+ ClO + dIO (43)

In the dimer cycle, production of chlorine atoms by reaction 42 competes with reaction 43

and the thermal dissociation of C100C1, both of which yield two ClO radicais, which leads

to no ozone depletion. However, thermal dissociation is slower than photolysis at

temperatures below -220 K; ftirthermore, experiments [Cox and Hayman, 1988; and

Molina et al., 19901 indicate that reaction 42 is the major photolysis path. In addition, the structure of C1202 has been shown to have the symmetric form by both theory and experiment [McGrath et al., 1989; Birk et al., 1989; Cheng and Lee, 1989].

13

The product of reactions 36 and 42 cou!d be C100 or C1+0 2; however, the Cl-OO

bond strength is only -5 kcallmole, so that even under polar stratospheric conditions

ClOO rapidly decomposes to yield free Cl atoms.

3.2. Polar stratosphenc c!ouds (PSCs)

As purely gas-phase reactions do not support large concentrations of CIO, the

explanation of the Antarctic ozone hole required a shift in thinking, from just gas phase

chemistry to a more comprehensive picture which includes heterogeneous chemistry-

iinking gas phase and aerosol chemistry.

Compared to the troposphere, the stratosphere is extremely dry and practically

cloudless. Under normal conditions, there is a sparse layer of aerosol at altitudes of 12 to

30 km, as first described by Junge in 1961. This background "sulfate" iayer, made up of

small sulfuric acid drop!ets (typically 75% H2SO4and 25% H 20 at mid-latitudes) of radius roughly 0.1 j.im and number density 1 to 10 cm 3 , is present al! latitudes. The number density and size of particles in the stratosphere are seen to vary widely. Increases by

factors of 10 to 100 in the aerosol mass are observed fo!lowing the injection of SO 2 directly into the stratosphere by major vo!canic eruptions, such as El Chichon in 1982 and

Mt. Pinatubo in 1991, which decay over a period of a few years. A possible role of

increased sulfate aerosols in ozone dep!etion was suggested by Hofmann and Solomon

[1989]. Recent observations by G!eason et al. [1993], and Deshier et al.[1992] support these expectations; the reason is explained in §3.3.

In addition to various ground based observations, Iong-term global trends of

stratospheric particle concentrations are being recorded by satel!ite instruments (SAGE,

SAM), the first of which were !aunched in the late 1970s. Over the polar regions stratospheric aerosols and clouds have been observed by the Stratospheric Aerosol

Measurement II (SAM II) instrument on board the Nimbus 7 satellite. As the general

characteristics of these c!ouds have previously been reviewed elsewhere [e.g., Turco et al.,

1989, Hamili and Toon, 19911, only a brief summary is presented here (see a!so Table 1). Observations, including satellite, lidar, and in situ particle measurements, have

indicated that the nature of the PSCs depends mainly on temperature and water vapor

concentration. PSCs fali into two distinct types: the sma!l (0.5 to 2 tm) class of partic!es

—type 1—are most likely crystais of nitric acid trihydrate (NAT) as their formation and

existence corre!ate with the partial pressures of water and nitric acid over NAT (e.g. 3x 10-' torr H20 and 4x10 -7 torr HNO 3 at 198 K) but are aboye frost point of pure ice [Toon et al.1986; Crutzen and Arno!d 1986; and McElroy et al.1986]. Type II PSC are

14

observed to form only when the temperature drops below the ice frost point (191 K for 3 x

1 0 torr H20) and they grow into significantly larger (up to 1 00tm) crystals.

The current understanding is that there are basically three chemical components

involved in the formation of PSCs: H20, HNO3, and H2SO4. Large amounts of HC1 do

not condense: inspection of a phase diagram of the HCWH 20 system, Figure 4, reveals that

HC1 solutions or hydrates are not stable under polar stratospheric conditions [Molina, 1992].

We discuss next the binary systems H201H2SO4 and H201HNO3, and the ternary

system, H20/F12SO4/HNO3. The H20fH2SO4 system is of interest for the lower

stratosphere at all latitudes. When temperature is aboye the threshold of PSC formation,

stratospheric aerosol particles are mainly composed of supercooled aqueous sulfuric acid

droplets, for example in the mid-latitude lower stratosphere at about -16 km the

temperature is roughly 220 K and the sulfate aerosol particles in equilibrium with 5 ppmv

H20 (-.4x10 Torr) have compositions of 70-75 wt % of H2SO4 (see Figure 5). The

equilibrium composition of these droplets changes with temperature: they absorb water to

become increasingly more dilute—reaching about 40 wt % H2SO4 around 195 K. The

thermodynamicahly stable forms for these aerosols are actually crystahhine hydrates (H2SO4

nH20; n = 1, 2, 3, 4, 6.5, 8); however, sulfuric acid solutions have a strong tendency to

supercool: laboratory and fleid observations indicate that sulfate aerosols remain liquid

under most conditions outside the polar vortices [Zhang et al., 1993a; Toon et al., 19931

As mentioned aboye, type-I PSCs most hikely consist of nitric acid trihydrate.

Laboratory measurements of vapor pressures over NAT [Hanson and Mauersberger

1988a, 1988b] confirmed that under typical lower stratospheric conditions, i.e., 3-5

ppmv H20 and -5-10 ppbv HNO3, NAT is stable at temperatures -5 K higher than the ice frost point, see Figure 6. The slopes from plots of 109Pij,j03 vs. logP 20 over NAT were

found to be -3, as required by the Duhem-Margules equation, considering that the crystal

structure ahiows onhy neghigible variations from the 3:1 stoichiometry. Worsnop et al.

[1993] explored the long-term (days) evolution of the HNO 3IH20 system, finding that, in

addition to NAT, it is possible for HN032H20 (nitric acid dihydrate, NA.D) crystais to

exist under stratospheric conditions if nucleated and if the more thermodynamically stable

NAT is not present. Infrared absorption spectra of amorphous and crystalhine films of

H20/HNO3 mixtures [Ritzhaupt and Devlin, 1991; Tolbert and Middlebrook, 1990;

Middlebrook et al., 1992; Smith et al., 1991; Koehler et al., 1992], as well as for small

NAT and NAD aerosol particles have been reported [Barton, Rowland and Devlin, 1993] and agree with the vapor pressure studies.

15

A number of researchers have carried out measurements on the uptake of nitric acid

by sulfuric acid solutions [see, e.g., Reihs et al., 1990; Van Doren et al., 1991] but only

recently over the range of composition and temperatures expected for the polar

stratosphere [Zhang et al., 1993b]. Figure 7 illustrates the expected composition of

H20/H2SO4/HNO3 liquid droplets in equilibrium with fixed water and nitric acid vapor

partial pressures as a function of temperature. Notice that upon cooling the equilibrium

composition of the liquid changes very rapidly below 195 K to increased water and nitric

acid contents. Solutions with these lower H2SO4 concentrations were observed to

crystallize readily into NAT and H2SO4 hydrates [Molina et al., 19931. Hence, a likely

mechanism for the formation of type 1 PSCs involves incorporation of HNO 3 vapor into

the liquid H20/H2 SO4 droplets, followed by crystallization of NAT in the droplets, and by

subsequent growth of the NAT crystais by absorption of additional amounts of I-1NO3 and

H20 vapors.

3.3. Reactions involving PSCs and sulfate aerosols

It is now well-established that heterogeneous reactions occurring on PSCs play a

central role in the chiorine activation leading to polar ozone depletion [see, e.g., Molina,

1991]. The impact of these heterogeneous reactions is twofo!d: (1) chlorine is

repartitioned from the inert reservoirs HCI and C1ONO2 into the much more photolabile

forms, mainly C12 and HOC1, and (2) NO is removed from the gas phase,

"denoxification", in the lower polar stratosphere through incorporation of nitric acid into

PSCs, and is thus unable to interfere with the chiorine cycle by forming C1ONO2. Various

techniques, e.g. Knudsen ce!l reactors, wall-coated flow tubes and droplet-train flow

tubes, have been used in the last few years to measure the reaction probability per

collision, y, of various chiorine and nitrogen species on surfaces representative of

stratospheric particles: ice, NAT, and sulfiiric acid solutions. An extensive review of these

heterogeneous processes has been presented by Kolb et al. [1993].

The most important chiorine activation reaction in the polar stratosphere is:

C1ONO2 + HC1 -> C12 + NN03 (44)

The net result of this reaction taken together with the gas phase reactions 19 and 30, and

chiorine photo!ysis is the conversion of HC1 and NO to CIO and I-1NO3:

16

C1ONO2 + HC1 -* C12 + HNO2 (44) C12+hv—*2C1 (45)

2(C1 + 03 -* ClO + 02) (19) C!O+NO2+M-+c2loNo2+M (30)

net: HC1 + NO2 + 203 -* CIO + HNO3 + 202

As a gas phase reaction, 44 is extremely slow - having an upper limit of 10-19 cm 3 s-1 [Molina et al., 1985; Demore et al., 1992]. It was noted, however, that one must take

extreme care to exciude water from the measurement apparatus to determine good upper

limits to the homogeneous gas phase rate constant, as water adsorbed on surfaces is very effective in promoting the reaction [Molina et al., 1985; Rowland et al., 1986].

To account for ozone depletion over Antarctica, Solomon et al. [19861 suggested that

reaction 44 is promoted by PSCs. Studies carried out on water-ice and NAT have

indicated that it does proceed with high efficiency, having y>0.2 [Molina et al., 1987;

Moore et al., 1990; Leu et al., 1991; Hanson and Ravishankara, 1991a; Abbatt and Molina, 1 992b]. The product C12 rapidly desorbs, but the nitric acid remains at the surface.

Although it is clear that the reaction occurs on the surfaces of PSC particles, it is unlikely

for both dONO2 and HC1 molecules to simultaneously colude on the same active site. A

plausible mechanism involves incorporation of HC1 or C1ONO2 or both into the surface

layers of the particles, followed by reaction. Studies of the HC1 uptake by ice crystais

under these conditions have shown a much larger surface coverage than could be

explained in terms of intact HC1 molecules interacting with the surface through hydrogen

bonding: instead, the results are consistent with the surface layers relaxing to a liquid-like

configuration, where the HCI is ionically solvated [Molina, 1992; Abbatt et al., 19921.

The reaction between C1ONO2 and 1120, which is also very slow in the gas phase,

has also been studied on ice and NAT surfaces [Molina et al., 1987; Tolbert et al., 1987;

Leu, 1988a; Leu, et al., 1991; Moore et al., 1990; Hanson and Ravishankara, 1991a;

Abbatt and Molina, 1992b]. Here the reactant 1120 is already in the condensed phase:

dONO2 +1120 -* HOCI + HNO3 (46)

and, again, nitric acid remains at the surface. The HOC1 product is easily photolyzed to

yield a free chlorine atom. Furthermore, HOC1 can participate in a subsequent

heterogeneous reaction, which also proceeds rapidly on water-ice and on NAT surface [Hanson and Ravishankara, 1992; Abbatt and Molina, 1 992a]:

17

HCI + HOCI - C12 + HO (47)

The combination of reactions 46 and 47 is equivalent to reaction 44—both convert

chiorine reservoirs to gas phase C12. This suggests that the mechanism of reaction 44

involves the two steps—reactions 46 and 47.

If N205 is abundant, another heterogeneous reaction which may convert HCI to an easily photolyzed gas is:

N205 + HC1 -* C1NO2 + HNO3 (48)

This reaction has also been studied on ice and NAT surfaces [Tolbert et al., 1988b; Leu,

1988b; Hanson and Ravishankara, 1991a].

As mentioned in §2.3, NO is converted to the reservoir species N 205 by reactions 13 and 14, or into the lIN0 3 reservoir by the reaction of NO 2 with OH, reaction 22. Hydrolysis of N205 yields the more stable BIN0 3 :

N205 + H20 -* 2HNO3 (49)

In the gas phase, though, this reaction is negligibly slow [Demore et al., 19921. However,

it takes places very efficient!y, with y0.1, on sulfuric acid solutions throughout the

concentration range of the stratospheric aerosols [Mozurkewich and Calvert, 1988; Van Doren et al., 1991; Hanson and Ravishankara, 1991 b].

The high reaction probability of reaction 49 independent of the sulfiiric acid

concentration is in marked contrast to the behavior of reactions 44 and 46 on sulfiiric acid

solutions. Laboratory studies of reaction 46 indicate that y is strongly dependent on the

composition of the aerosols [Tolbert et al., 1988b; Hanson and Ravishankara, 199 ib],

increasing from 1.9x iø (75 wt % sulfuric acid solution) at 230K to 6.4x10 2 (40 wt % sulftiric acid solution) at 215 K, cf. Table 2. Since the solubility of HC1 in 50-80 wt % H2 SO4 solutions is very low, reaction 44 on sulfate aeroso!s is not important at mid-

latitudes. At higher latitudes, however, where the temperature can drop to 200 K and the

aerosol droplets absorb more water and HCI, reactions 44 and 46 on liquid sulfate aerosols become very important.

The net effect of reaction 49 is to convert catalytically active nitrogen oxides to the

inert reservoir HNO3; and to repartition hydrogen and chlorine species, as a consequence

18

of the strong coupling between the various chemical families (see §2.5). The result is to

decrease the importance of NO in ozone depletion, and to increase it for hydrogen and chlorine species, as shown in Figure 8. Analysis of recent field measurements of nitrogen

and chiorine species indicates that this heterogeneous reaction does play an important role

in the lower stratosphere at mid-latitudes, particularly in the presence of the enhanced

sulfate aerosol concentrations which followed the eruption of Mt. Pinatubo in 1991 [see,

e.g., Fahey et al., 19931: the observations cannot be explained with gas phase chemistry

alone. Hence, the effect of volcanic eruptions on stratospheric ozone is indirect: increased

ozone depletion may occur, but only because chiorine chemistry is enhanced. In the

presence of only natural levels of chlorine the effect is expected to be rather small.

4. Field observations and computer modeis

Concerns about various anthropogenic emissions leading to global ozone depletion

prompted research efforts directed at developing an understanding of stratospheric

processes as well as to assess of the impact of chemical perturbations to the atmosphere.

Research in three areas has significantly refined our understanding of the chemistry of the

stratosphere. One area of study involves laboratory measurements of elementary rate

constants and absorption cross-sections of numerous reactions important in the

stratosphere [DeMore et al., 19921. Another area of research focuses on measuring the

concentrations of the important trace gases in the stratosphere. And the third area involves

the development of model simulations of the stratosphere. Advancements these three areas in the last two decades have been impressive.

4. 1 Measurement techniques and results

A large number of measurements on individual species by in situ and remote sensing techniques from ground, aircraft, balloon, and spacecraft have been reported in the last

decade. Intercomparison campaigns using different techniques to provide observations of

the same species at the same time and place have validated the accuracy and precision of

sorne measurements (see, e.g., the special issue ini. Atmos. Chem. 10(2) 1990).

Ozone Although stratospheric ozone concentrations are determined predominantly by

photochemical and dynamical processes, leveis are also influenced by galactic cosmic rays,

solar proton events, and volcanic eruptions, in addition to human activity. Thus to

understand the relationship between the variable natural stratospheric ozone levels and

wei

anthropogenic perturbations, it is essential to estab!ish long-term ozone trends. A number

of techniques have been employed to measure the total ozone column abundances and

vertical ozone profiles. The main instruments for measuring columns abundances include

ground-based Dobson and other filter spectrometers, and satellite-based Backscattered

Ultraviolet (BUV), Solar Backscattered Ultravio!et (SBUV), and the Total Ozone

Mapping Spectrometer (TOMS). Current profile measurement techniques include bailoon-

borne ozonesondes, and also a wide range of satellite-based instruments, including BUV,

SBUV, and the Stratospheric Gas and Aerosol Experiment (SAGE). Figure 9 shows

representative profile and column measurements of ozone abundances.

In the upper stratosphere (-30 km to —SO km) the dominant factors controlling ozone

levels are photochemical processes involving atoms or small molecules, with less

important roles played by reservoir species having short photochemical lifetimes such as

HOC1, C1ONO2, etc. Below 30 km, the chemistries of HO, NO x and CIOx are closely coupled and the concentration of ozone is controlled by both chemical and transport

processes. To assess our understanding of the complex chemistry of ozone and the

catalytic cycles, monitoring of the source gases (e.g., H20, N20, CH4, CFCs) and the key

trace gases in the oxygen family (0, 03), nitrogen family (NO, NO2, NO3, N205, FIN03,

H02NO2), hydrogen family (OH, H02, H202), and chiorine family (dO, HOC1, dONO2,

HC1, halocarbons) is essential. We discuss next observations of key species responsible for

stratosphere ozone depletion.

HOx The key HOy species are H20, OH, H02 and H202. Since water is the primary

source of the HO radicals, measurements of water mixing ratios are important to the

understanding of HO chemistry. Two main iii situ techniques have been used for the

measurement of H2O vapor. The Lyman-a hygrometer monitors the fluorescence emitted

by electronically excited hydroxyl radicals, OH*, produced from dissociating H2O with

Lyman-a (121.6 nm) light [Kley and Stone, 19781. The frost-point hygrometer relies on

the equilibrium between water vapor in the atmosphere and an ice surface at the frost-

point temperature [Masterbrook and Oltmans, 19831. Remote sensing techniques

commonly utilize the strong infrared lines, for example one of the six spectral channels of

the Limb Infrared Monitor of the Stratosphere (LIMS) on board the Nimbus 7 satellite is

centered on emissions of H20 (1370-1560 cm -') [the other five channels are: NO2 (1560-1630 cm- '), 03(926-1141 cm'), HNO3 (844-917 cm - '), CO2 (579-755 cm 1 and 637-673 cm-1 ); see WMO, 1990a].

The most important observations in the HO x family are on the OH and H0 2 radicais. Fluorescence induced by solar flux, resonance lamps, and lasers has been the principal

20

technique for OH measurement. Laser induced fluorescence (LIF) has utilized A-X

electronic transitions of OH (excitation of the (0-1) band at 282 nm or the (0-0) band at

309 nm) [Anderson, 1987]. Two iii situ bailoon borne techniques have been used to

measure H02: one is cryogenic trapping of free radicais on a liquid nitrogen cooled

surface, analyzed subsequent!y by electron spin resonance spectroscopy in the laboratory

[Miheicic, et al, 1978; Helten et al., 1984]. Another technique utilizes the chemical

conversion of H02 to OH by addition of NO, i.e. reaction 21, followed by the detection of

OH via induced fluorescence [Stimpfle et al., 19901. Ground-based detection of H02

aboye 35 km from a rotational emission une near 265.7 GHz has been reported [de Zafra

et al., 19841. Also, vertical H0 2 profiles from about 20 to 50 km have been obtained by

Traub et al. [1990] using a far-infrared spectrometer. Simultaneous UF measurements of OH, H02, H2O, and 0 3 have recently been carried out [Stimpfle et al. 1990; Wennberg et

al 19901, providing a most useful comparison between models and observations. H 202 has

been observed by microwave limb sounding [Waters et al., 1981], ground-based millimeter

wave spectrometry [de Zafra et al., 1985], and by far-infrared limb sounding [Chance et

al., 19911; the first two observations, however, provide only upper limits. In addition, simultaneous measurements of OH, H0 2, and H202 been obtained by Park and Carli [1991] using far-infrared spectroscopy.

Nitrogen Oxides NO, NO2, NO3 and N205 are the principal species here. The NO/NO2

ratio is affected by interconversion within the NO2 family (discussed in §2.3) as well as by

reactions with 03, H02, or ClO (2.3 & 2.5), and thus the comparison between observed

ratios of NO/NO2 and model calculations provides important tests of our understanding of

NO in the stratosphere. The principal technique for in situ detection of NO and NO2 is

based on chemiluminescence—NO is converted to electronically excited (luminescent)

NO2* via the reaction: NO+03+NO2*+02 Measurement of NO2 is based on the same principie except NO2 is first photolyzed to yield NO. Extensive data sets are available from in situ bailoon, aircraft and rocket-borne measurements {Ridley et al, 1987; Kondo et al., 19901. In addition, the sum of NO S, N205 and HNO 3 has been measured in situ by the catalytic conversion of all nitric oxides to NO, followed by detection using NO/0 3 chemiluminesence [Fahey et al., 1989; 1990a]. Other methods, mainly infrared and visible

spectrophotometry, have also used to measure NO, NO2 and NO3. NO 2 in the polar

stratosphere has been measured by high resolution infrared [Farmer et al., 1987a; Coffey

et al., 1989; Mankin et al., 19901 and ultraviolet spectroscopy [Wahner et al., 199b,

1990a; Sanders et al., 19891 .Ground-based observation of NO3 has utilized the strong

absorption near 662 nm with moon and stars as a light source [Solomon et al., 198%, and

21

references therein]. Further techniques for NO and NO 2 measurement include pressure modulated radiometry (using a device which selectively modulates the emission from a gas

by using the absorption unes of the same gas as an optical filter) [Drummond and Jarnot,

1978; Roscoe et al., 1978; 19861, long-path absorption {Louisnard et al., 19831, and the

Bailoon-Borne Laser In-Situ Sensor (BLISS) instrument [Webster, 1987]. The BLISS

instrument employed long path tunable diode laser infrared absorption to monitor species inc!uding: NO (1854 cnn), NO2 (1598 cnn 1 ), 03 (1063 cm'), HNO3 (1333 cm'), and

N20 (1525 cm'). The capability of the BLISS to simultaneously measure chemically

coupled nitrogen species provides one of the best checks of our know!edge of NO

chemistry [Webster et al., 1990]. There have been a number of observations of N205 in

the stratosphere, most of which measured absorption or emission of the strong infrared band at 1240 cm-1 [see, e.g., the review by Roscoe, 19911. Combined with information

from satellite data for NO2 (LIMS, SAGE, and SME-the Solar Mesosphere Explorer) and

for HNO3 (LIMS) this provides a basis for estimating NO

(NOy=N0+NO2+NO3+HNO3+2(N2O5)+H02NO2+c!0NO 2 ; of which HNO3 is the major component in the lower stratosphere). Simultaneous measurements of the important mtrogen species have also been carried out by Abbas et al. [1991], as well as by the

shuttle-borne ATMOS instrument which provided the most complete simultaneous

measurements of the nitrogen family so far; making observations of al! the NO species

including NO, NO2, N205, HNO3, H02NO2, and CIONO2. [Russell et al., 1988; Allen and Delitsky, 1990].

Chiorine species These include the CFC source gases, the reservoirs HC1, C1ONO2 and

HOC1; and the C1O radical. The majority of the CFC vertical profiles have been obtained

from grab-samples (canisters fihled with air at the collection point) from aircraft or balloon

flights and later analyzed by gas chromatography [WMO, 1986]. Several techniques have been used to measure stratospheric CIO: in situ resonance fluorescence, bailoon-borne and ground-based millimeter and microwave spectroscopy. In the in situ resonance fluorescence method ClO, is converted to atomic Cl by the reaction Cl0+NO->Cl-1-NO2,

and a resonance lamp is then used to monitor Cl near 120 nm [Weinstock et al., 1981]. On

the ER-2 aircraft in the Antarctic and Arctic missions (AAOE and AASE) ClO was

converted to Cl which was then detected by atomic resonance fluorescence while ozone

was simultaneousiy detected by uy absorption [Brune et al., 198%, 19901. Ground-based

millimeter-wave spectrometric measurements of a ClO rotational line emission were

reported by, e.g., So!omon et al., [1984]. A!so, de Zafra et al. [1987] used this techniue

to measure ClO vertical profi!es over Antarctica. Waters et al., [1981; 19881 and Stachnik

et al., [1992] used a ba!loon-bome millimeter-wavelength heterodyne spectrometer to

22

measure limb emission from a ClO rotational une near 640 GHz (along with nearby unes

of 03 , HC1, ClO and H02). In addition, ClO (as well as 0 3 , SO2 and H20) has been

measured globally by the microwave limb sounder aboard the UARS launched in

September, 1991 [Waters et al., 19931. Observations of BrO in the polar stratosphere have

been made using near-ultraviolet absorption spectroscopy [see, e.g., Carroll et al., 1989;

Wahner, 19901 and in situ by a similar method to that used for ClO [Brune et al., 1989a;

Toohey et al., 19901.

Monitoring CIONO 2 is important because this species links the stratospheric nitrogen

and chiorine cycles. Measurements of C1ONO 2 utilizing the infrared 1292 cm -1 , 809 cm-1 ,

and 780 cm bands have been made by bailoon-borne solar absorption spectroscopy [see,

e.g., Massie et al., 19871 as well as the ATMOS [Zander et al., 19901 and UARS

instruments [Roche et al., 1993]. The abundances of CIONO2 as well as HCI in the polar

stratosphere have also been recorded by techniques including high resolution infrared

spectrophotometry [e.g., Farmer et al., 1987; Coffey et al., 1989; Toon et al., 1989a,

Mankin et al., 19901. As HC1 is usually the most abundant stable reservoir species for

chlorine radicals, our knowledge of the concentration of this species is important to the understanding of the chiorine budget. The strong infrared absorptions of HC1 and HF have

been utilized by ground-based, bailoon [Farmer et al., 19771, aircraft [Mankin and Coffey,

1983], and ATMOS [Raper et al., 1987] and UARS measurements [Reber, et al, 19931.

Multi-species measuremenis The rather large variability in the concentrations of many

of the key species, and the highly coupled nature of the chemical processes makes it

important to simultaneously monitor as many of the reactive species as possible. Also,

measurements of concentration ratios for certain species has also proved to be useful for

testing models. Por example, the C1O/(HCI+C1ONO 2) ratio is a measure of the fraction of

the chlorine that participates in ozone destruction cycles, i.e., the active vs. reservoir

forms of inorganic chlorine. As another example, simultaneous measurements of HC1 and

HF provide an indication of the relative contributions of natural and anthropogenic

sources to chiorine in the stratosphere. The observed increases in both HCI and HF, as

well as the ratio of their concentrations, which has decreased in a 13 -year period between

1977 and 1990 from about 7 to about 4 [e.g., Rinsland, 19911, points to CFCs as the

source of the increased halogen burden in the stratosphere: this ratio has decreased

because there are no natural sources of HP, while HCI is formed from the decomposition of biogenic CH3 0 as well as CFCs.

Several comprehensive experiments have been initiated in the past few years.

Examples include the space shuttle-borne Atmospheric Trace Molecule Spectroscopy

23

(ATMOS) experiment, the Upper Atmosphere Research Satellite (UARS), the Airborne

Arctic Stratospheric Expedition (AASE), and the Airborne Antarctic Ozone Experiments

(AAOE). Equipped with a high resolution Fourier transform near- and middle-infrared

spectrometer, the ATMOS instrument is capable of monitoring more than 40 major and

minor gaseous species from altitudes aboye -16 km [WMO 1986; Farmer et al., 1987b].

The primary goals of UARS complement ATMOS: besides monitoring 15 important

species in the upper atmosphere belonging to the HO, NO x and ClO families {WMO

19861, it also measures solar irradiance, particle energy deposition, temperatures, and

wind flelds. In addition to observations of ozone levels, instruments on board the Nimbus-

7 satellite have provided other important measurements: LIMS (HNO3, NO2, 03, H20,

temperature), SAMS (C1-14, N20, temperature), SBUV/TOMS (ozone, solar flux), and SME (ozone, NO2, aerosols, solar flux). The AAOE [see J. Geophys. Res. 94, 11179-11737 (part 1); 16437-16857 (part 2)] and AASE [see Geophys. Res. Lelt., 17, 3 13-564]

campaigns, using two aircraft (ER-2 and DC-8) platforms, were designed to understand

the perturbed chemistry and rapid ozone loss in the high-latitude lower stratosphere. The

instruments were capable of making in situ measurements of important trace gases including ClO, BrO, OC1O, dONO2, NO, NO2, NON, HNO3, H20, 03, N20, CH4, CO, CFCs, particulates, as well as meteorological parameters.

4.2. Observations of the polar stratosphere

The meteorological structure which facilitates rapid ozone loss in the stratosphere over

Antarctica, i.e. the "ozone hole", is the polar vortex, which is a strong circumpolar wind

pattern that forms over the pole in the dark winter [see, e.g., Schoeberl and Hartmann,

1991]. Over the south pole it is nearly circular and approximately the size of the Antarctic

continent, and the chemically perturbed region is roughly coincident with this vortex. In

the northern hemisphere the different land mass distribution and tall mountains leads to a

less intense and less symmetric vortex. Field studies of vertical profiles of long-lived

tracers (e.g. CFCs and N20) provide evidence of a strong downward flux inside the vortex. The vortex structure, coupled with the very cold temperature, sets the scene for

the perturbed chemistry in the lower polar stratosphere that leads to severe, localized, ozone depletion.

Field campaigns, organized soon after the discovery of Antarctic ozone hole, have

monitored concentrations of the key species that participate in the various catalytic cycles

over Antarctic and the Arctic. These expeditions included: the National Ozone

Expeditions (NOZE 1 & II) in 1986 and 1987, the Airborne Antarctic Ozone Experiment

24

(AAOE) in 1987, the Airborne Arctic Stratosphere Expeditions (AASE 1 & II) in 1989

and 1992. Very strong evidence linking ozone loss to active chiorine radicals comes from

the fleid observations along with laboratory measurements and modeling. The

observational techniques were chiefly those mentioned aboye, though in sorne cases were

adapted to the special conditions of the polar stratosphere and the sampling platforms.

Sorne results are briefly summarized next:

Ozone and ClO As all the important polar ozone destroying catalytic cycles involve dO, it is a key species to monitor along with 0 3 . Measurements of these two species provided striking evidence linking ClO levels to ozone losses, as illustrated in Figure 11

Near 20 km, abundances of ClO are —1 to 2 ppbv in both the Antarctic and the Arctic

vortices, which is orders of magnitude higher than concentrations in the unperturbed rnid-

latitude stratosphere, revealing almost complete conversion of chiorine species to reactive

forms within the chemically perturbed regions. The high levels of CIO at these altitudes

and the anticorrelation with 0 3 are consistent with the ClO dirner catalytic mechanism,

since the rate of ClO-dimer formation, reaction 40, increases quadratically with [ClO]

[Anderson et al., 1991]. Recent satellite observations of ClO (UARS) have also shown

extrernely high leveis in portions of both the northem and southern polar winter vortices,

particularly in regions where the temperature drops below the threshold for PSC formation [Waters et al., 1993].

BrO and OCiO Significant enhancements in BrO compared to mid-latitudes have also

been observed over both Antarctica and the Arctic. The diurnal behavior shows that BrO (similar to ClO) can combine with NO 2 to fonn reservoir species BrONO 2 during nighttime. Since the BrO+ClO reaction is believed the only source of OCiO, (see §3.1), an

enhancement of its abundance as measured by ground-based and air-borne UY

spectrometers [Solomon et al., 1989a, 1989c; Wahner et al., 1989a; Schiller et al., 1990]

indicates that the ClO-BrO cycle indeed plays a significant role in the chernistry lower

polar stratosphere. In addition, Solomon et al. [1993] have observed high OCiO

concentrations over Antarctica in the 1992 autumn, when PSCs were unlikely to occur,

supporting the role of sulfate aerosols in chiorine and bromine activation in the coid

stratosphere at high latitudes when the aerosol loading is large, in this case after the Pinatubo eruption.

Halogen reservoirs On!y a few percent of the inorganic chiorine is in free radical form

at mid-latitudes (see §2.5). In contrast, the sustained high levels of ClO within the polar

25

vortex, along with the significant reductions in the column abundances of HCI, C1ONO2

and NO2 clearly demonstrate the extent of the perturbation to chiorine chemistry: up to

70-80 % of the chiorine burden can be in catalytically active form.

Another indicator of perturbed chiorine chemistry in the polar stratosphere is the

HCIIHF ratio. Observations indicate that in the chemically perturbed region of Antarctic

vortex the ratio of HCIIFIF is reduced from the current mid-latitude value of -4 to 5 to

near one, corresponding to a conversion of -80% of the HC1 to other chiorine species

[Coffey et al., 19891. In the Arctic, the ratio is reduced to -2 within the vortex,

corresponding to 60% conversion [Mankin et al., 19901. Gas-phase chemistry alone

cannot account for this: heterogeneous reactions need to be taken into account.

Nitrogen species Measurements of N20 and reactive nitrogen, NO N, reveal a marked decrease of NO inside the polar vortex. Decreases in N 20 levels should be accompanied

by increases in NO (with a linear negative correlation), as the a principal source of NO is N20 [Fahey et al., 1990b,c]. Using this relationship, the difference between the expected

and measured NO concentrations can be used to estimate the extent of denitrification

(condensation of nitric acid into particulates, followed by sedimentation) in the polar

stratosphere. For example, inside the Antarctic vortex, instead of a NO value of 10 ppbv calculated from the observed N 20 value, a much lower value of 1-4 ppbv was observed.

This denitrification being explainable in terms of the heterogeneous reactions coupled with sedimentation of particles containing HNO 3 . This and the liberation of chlorine from its

reservoirs is clearly evident from measurements of the various species versus latitude; see

Figure 12 [Toon et al., 1989b]. Inert species such as HF show normal behavior while the

concentrations of those species involved in the reactions discussed here are markedly perturbed.

4.3. Modeling and assessment

The simultaneous actions of radiative, dynamical, and chemical processes affect the

concentrations of trace species in the stratosphere; therefore, a realistic theoretical model

must consider all of the aboye. According to the level of sophistication, one-, two-, and

three-dimensional models have been employed. One-dimensional (altitude only) models

coupling chemical and transport processes, have been used extensively in the past as

explorative tools for understanding the effects of anthropogenic perturbations to

stratosphere, as they may include very detailed chemistry. Two-dimensional models seek

to predict latitudinal and seasonal variations, and may include feedback between radiation,

26

dynamics, and photochemistry. Finaily, three-dimensional modeis which aspire to rely !ess

extensively on the pararnetrization of various transport processes are being ffirther

developed as more computer power and observational data becornes available.

The physical and chemical processes that determine the concentration for any

atmospheric chemical species are represented by the continuity equation, in which the

number density of a general species i changes with time as

Jni = - L. + V

dt

where P. and L are the chemical and photochemical production and loss terms, and V•Ø. is

the flux divergence, which accounts for transport. Typically this system of equations,

which may involve 100 or more chemical and photochemical reactions, is integrated

numerically using an irnplicit finite difierence representation. Under sorne circumstances

these equations can be simplified; e.g., at steady-state the time derivative is equal to zero;

for short-lived species chemistry is much faster than transport, so the flux divergence for

such species can be neglected; etc. For example, in the winter polar stratosphere vertical

transport within the vortex as well as horizontal transport in and out of the vortex is

siower than the characteristic time for ozone depletion reactions. Thus, the rate of

chemical ozone loss can be approximated by a sirnplified continuity equation in terms of

the two catalytic cycles found to be most important, considering also the fact that the production of 03 is negligible:

d[03]/dt = -2{k40[001[00][M] + k36[ClO][Br0]}

Using this approach with rneasured values of [dO], Anderson et al. [1991] showed that

the rapid ozone loss could be explained, with the first terrn being responsible for as much

as 75 % of the total Antarctic ozone loss. Notice that the loss processes are quadratic in

[X0] and thus the ozone destruction rate increases nonlinearly with increases in stratospheric loading of chiorine and bromine.

A typical set of mid-latitude model results is shown in Figure 10, taken from McElroy

and Salawitch [1989]. In general, the results are in fairly good agreement with the

observations, taking atmospheric variability into account. However, sorne questions

rernain. For example, models have underestimated ozone concentrations aboye 35 km by as much as 30 to 50% (the "40-km ozone problern"). That problem is particularly striking since in the chemistry-dominated upper stratosphere the concentration of ozone is

27

controlled by the well-established photochemical processes represented by various

catalytic cycles [see, e.g., Toumi et al. 1991; Eluszkiewicz and Allen, 19931.

S. The future of stratospheric ozone

Understanding the long term impact of halogenated chemicais on the atmosphere is

important because of the very long atmospheric lifetimes of these compounds.

Investigations of the effects of the CFCs and potential CFC substitutes such as

hydrochlorofluocarbons (HCFCs) on stratospheric ozone have been initiated in the last

few years. Here we discuss the atmospheric fates of halocarbons and how relative 'Ozone

Depletion Potentials' (ODPs) of halocarbons are determined.

5. 1. Determining the lifetimes of halocarbons in the atmosphere

A halocarbon's atmospheric lifetime, specifically in relation to the time required for

diffusion into the stratosphere, determines its influence on stratospheric ozone depletion.

Substitute compounds such as HCFCs have been introduced with the intent to reduce the

amount of chiorine reaching the stratosphere; examples include HCFC-134a (CF3CFH2)

as a substitute for CFC-12 (CF2C!2); and HCFC-141b (CH3CFC12), HCFC-123

(CF3CFC12), and HCFC-22 (CHF2C1) as substitutes for CFC- 11 (CFC13) [Manzer, 1990].

Unlike the CFCs, the HCFCs are destroyed to a large extent in the troposphere by

reaction with OH radicais, and thus have shorter atmospheric lifetimes. In order to

elucidate the atmospheric fate of these new chemicais, knowledge of the reaction rates and

of tropospheric OH leveis are needed. The chemical lifetime due to reaction with OH, TOHI

can be written as

1 =

kOH+hain {]

where kOH+halo is the rate constant for the reaction with OH radicais, and [OH] is a

average tropospheric OH concentration. Increasing efforts are being directed at measuring

and predicting the reaction rates between OH and halocarbons and the subsequent reaction

products; see Table 3.

As one can see, the accuracy of lifetime estimates hinges upon our knowledge not

only of the rate constants, but also the average concentration of OH radicais in' the

troposphere. Direct measurements of tropospheric OH concentrations are extremely

28

difficult; however, in recent years there has been significant progress: techniques using a

14C-tracer method, laser-induced fluorescence, as well as long-path absorption

spectroscopy and chemical ionization mass spectroscopy have been employed to measure

local tropospheric OH concentrations [see, e.g., Mount and Eisele, 19921. An indirect

method to determine the global mean OH concentrations compares measured leveis of

methyl chloroform (CH3CC1 3), which has no natural sources and whose main removal

processes is the reaction with OH, with expected concentrations based on known

industrial emissions [Prinn et al., 1992]. The estimated average concentration was found

to be [OH]8.7x 10 molecule cm-3 (or niIIxIO 5 if a small loss rate to the ocean is

included) using this method.

5.2. Ozone depletion from CFC substitutes - the Ozone Depletion Potential

The ozone depletion potentia! (ODP) of a halocarbon x is usually defined as the

steady state ozone destruction that results from each mass unit of the particular species x,

relative to that of CFC-1 1 [Wuebbles, 19831:

ODP(x) = ¿03 (x)

L03 (CFC -11)

As a relative measure, the ODP does not predict absolute ozone losses but shows the

expected effects of releasing one molecule versus another one into the atmosphere. The

primary factor that determines a compound's ODP is its tropospheric lifetime, which, as

discussed aboye, for a HCFC is determined by its reaction rate with hydroxyl. Some

example reaction rate constants and the derived atmospheric lifetimes and ODPs for

several halo carbons are shown in Table 3.

While substitution of hydrogen containing halocarbons for CFCs in general will result

in less chlorine being delivered to the stratosphere, long-term massive use could still have

deleterious effects. Thus, such compounds are thought of as transitional substances. As

steady-state models are inappropriate for predicting their impact on ozone in the near

future, time dependent ODPs may need to be considered for compounds such as the HCFCs [Solomon and Albritton, 19921.

As discussed earlier, bromine species are one to two orders of magnitude more

efficient at destroying ozone than those of chiorine, on a molecule-per-molecule basis, as

is reflected by the very large ODPs of the Halons (cf. Table 3). Similarly, despite' the

relatively short tropospheric lifetime of methyl bromide (CH 3Br), this species has a sizable

29

ODP of about 0.6, and hence its industrial production might be regulated—or even phased

out—in the future [see, e.g., WMO, 19921.

5.3. Other potential threats: supersonic aircraft, etc.

Human activity impacts the ozone layer not only by pollutants emitted at the surface

which enter stratosphere by transport processes (e.g. CFCs), but may also by the release

of pol!utants (e.g. NO N, HC1, H20, etc.) emitted directly into stratosphere by high-altitude

aircraft, rockets, etc. Potential ozone depletion caused by the engine emissions of

projected fleets of supersonic aircrafi has attracted attention since the early 1970s, focusing mostly on the NO combustion by product. When only gas-phase chemistry was

considered, modeis indicated that the release of NO near 20 km would deplete ozone

through catalytic cycies–reactions 11 and 12; whereas release around 10-12 km (such as

the current subsonic fleet) may lead to a slight increase of ozone, the latter promoted by

the following reactions involving NO (the "smog" reactions), which are initiated by the

reaction between OH radical and hydrocarbons (RH):

OH+RH–H2O+R

R+O2 –*R02

R02+ NO -* RO + NO2

NO2 +hv – NØ+O

0+02 +M-03 +M

Since the fate of NO2 is altitude dependent, there exists a crossover point aboye which

increased NO leads to ozone depletion (where reaction 12, NO2+0—>NO+0 2, is favored) and beiow which ozone is produced (where NO 2 photolysis is favored). However, the ozone changes are sensitive to the balance of competing catalytic cycles as well as

atmospheric circulation. The calculated impact of a fleet of supersonic aircraft, now

referred to as "high speed civil transports" (HSCTs) is being currently reexamined [e.g.,

Johnston et al., 19891. Recent indications are that the impact of HSCTs is highly

dependent on heterogeneous chemistry, especially the N205+H 20.–>2HNO 3 reaction on sulfate aero sois: the predicted effects of NO emissions are smaller when this reaction is taken in to account. On the other hand, H 20 and suiflir emissions are now also important considerations in the evaluation of the impact of these proposed aircraft [HSRP/AESA, 1993],

30

Space shuttle and other rocket launches, particularly those using solid fuel boosters

having chlorine compounds as fuel components, have been considered as possible sources

of ozone depletion, as they emit chiorine directly into the stratosphere. The chlorine

burden due to these sources, however, is hundreds of times smaller than that due to the

CFCs. On the other hand, the effects due to other emissions—mainly particulates such as alumina, etc.—have not yet been quantified [WMO, 1992].

5.4 International regulations on ozone depieting compounds

The first international agreement limiting CFCs, the Montreal Protocol on Substances

That Deplete the Ozone Layer, was approved on 16 September, 1987, while the AAOE mission was only starting to gather the data that would firmly link CFCs to the formation

of the "ozone hole". It provided for a staged control of five CFCs and three bromine-

containing halocarbons, to ultimate!y freeze consumption of the bromine compounds at 1986 levels and to cut CFC consumption to 50 % of 1986 leveis by the year 2000. It was quickly recognized that the restrictions did not go far enough (see Figure 13), leading to amendments negotiated in London in 1990 and in Copenhagen in 1992. Even with ftill compliance to stringent international regulation, the global ozone declines will not be reversed in the near future [see, e.g., Prather and Watson, 19901 and the "ozone hole" is expected to develop annually over Antarctica for several decades to come.

0-9

31

References:

Abbas, M. M., Kunde, V. G., Brasunas, J. C. Herman, J. R., and Massie, S. T. (1991) "Nighttime reactive nitrogen measurements from stratospheric infrared thermal emission observations." J. Geophys. Res., 96, 10885-10897.

Abbatt, J. P. D., Beyer, K. D., Fucaloro, A. F., McMahon, J. R., Wooldridge, P. J., Zhang, R., and Molina, M. J. (1992) "Interactions of HC1 vapor with water ice: Stratospheric implications." J. Geophys. Res., 97, 15819-1582.

Abbatt, J. P. D., and Molina, M. J. (1992a). "The heterogeneous reaction HOC1 + HC1 -+ C12 + HO on ice and nitric acid trihydrate: reaction probabilities and stratospheric implications." Geophys. Res. Lelt., 19, 461-464.

Abbatt, J. P. D., and Molina, M. J. (1992b). "Heterogeneous interactions of dONO2 and HCI on nitric acid trihydrate at 202 K." J. Phys. Chem., 96, 7674-7679.

Allen, M., and De!itsky, M. L. (1990). "Stratosphere NO, NO2, and N205: A comparison of model results with space!ab 3 atmospheric trace molecule spectroscopy (ATMOS) measurements." J. Geophys. Res., 95, 14077-14082.

Anderson, J. G. (1987). "Free radicals in the earth's atmosphere: Their measurements and interpretation." Ann. Rey. Phys. Chem., 38, 489-520.

Anderson, J. G., Brune, W. H., and Proffitt, M. H. (1989). "Ozone destruction by chiorine radicais within the Antarctic vortex: The spatial and temporal evolution of C1O-03 anticorrelation based on in situ ER-2 data." J. Geophys. Res., 94, 11465-1 1479.

Anderson, J. G., Toohey, D. W., and Brune, W. H. (1991). "Free radicais within the Antarctic vortex: The role of CFCs in Antarctic ozone loss." Science, 251, 39-46.

Barton, N., Rowland, B., and Devlin, J. P. (1993). "Infrared spectra of large acid hydrate clusters: Formation conditions of submicron particles of HN0 3 21120 and HN0 3 3H20." J. Phys. Chem., 97, 5848-585 1.

Birk, M., Friedi, R. R., Cohen, E. A., Pickett, H. M., and Sander, S. P. (1989). "The

rotational spectrum and structure of chlorine peroxide." J. Chem. Phys., 91, 6588-6597.

Brune, W. H., Anderson, J. G., and Chan, K. R. (1989a). "In situ observations of BrO in the Antarctic: ER-2 aircraft results from 540 S to 720 S latitude." J. Geophys. Res., 94, 16639-16648.

32

Brune, W. H,, Anderson, J. G., and Chan, K. R. (1989b). "In situ observations of ClO in the Antarctic: ER-2 aircraft results from 540 S to 720 S latitude." J. Geophys. Res., 94, 16649-16663.

Brune, W. H., Toohey, D. W., Anderson, J. G., and Chan, K. R. (1990). "In situ observations of ClO in the Arctic stratosphere: ER-2 aircraft results from 590 N to 800 N latitude." Geophys. Res. Lett., 17, 505-508.

Callis, L. B., and Natarajan, M. (1986). "The Antarctic ozone minimum: Relationship to odd nitrogen, odd chiorine, the final warming, and the 1 1-year solar cycle." J. Geophys. Res., 91, 10771-

Carpenter, C. D. and Lehrman, A. (1925). "The solid-liquid temperature relation in the ternary system 1 —12 SO4—HNO 3-1420 and its relation to the ternary system N205—H20—S03 ." AIChE, 3 5-73.

Carroll, M. A., R. W. Sanders, Solomon, S., and Schmeltekopf, A. L. (1989). "Visible and near-ultraviolet spectroscopy at McMurdo station, Antarctica 6. Observations of BrO." J. Geophys. Res., 94, 16633-16638.

Chance, K. V., Johnson, D. G., Traub, W. A., and Jucks, K. W. (1991). "Measurements of the stratospheric hydrogen peroxide concentration profile using far infrared thermal emission spectroscopy." Geophys. Res. Lett., 18, 1003-1006.

Cheng, B.-M., and Lee, Y. P. (1989). "Production and trapping of gaseous dimeric ClO: The infrared spectrum of chiorine peroxide (ClOOCl) in solid argon." J. Chem. Phys., 90, 5930-5935.

Coffey, M. T., and Mankin, W. G., and Goidman, A. (1989). "Airborne measurements of stratospheric constituents over Antarctica in the austral spring 1987, 2, Halogen and nitrogen trace gases." J. Geophys. Res., 94, 16597-16613.

Cox, R. A., and Hayman, G. D. (1988). "The stability and photochemistry of dimers of the ClO radical and implications for Antarctic ozone depletion." Nature, 332, 796-800.

Crutzen, P. J. (1970). "The influence of nitrogen oxides on atmospheric ozone content." Quart. J. Roy. Meteoro!. Soc. 96, 320-325.

Crutzen, P. J. (1979). "The role of NO and NO2 in the chemistry of the troposphere and stratosphere." Ann. Rey. Earth Planet. Sci, 7, 443-472.

Crutzen, P. J., and Arnold, F. (1986). "Nitric acid cloud formation in the coid Antarctic stratosphere: a major cause for the springtime ozone hole." Nature, 324, 65 1-655.

33

DeMore, W. B., Sander, S. P., Golden, D. M., Hampson, R. F., Kurylo, M. J., Howard, C. J., Ravishankara, A. R., Kolb, C. E., and Molina, M. J. (1992). "Chemical kinetics and photochemical data for use in the stratospheric modeling;" Evaluation No. 10, JPL Pubi. 92-20. Jet Propulsion Laboratory, Pasadena, CA.

Deshier, T., Adriani, A., Gobbi, G. P., Hofrnann, D. J., Donfrancesco, G. Di,. Johnson, B. J. (1991). "Volcanic aerosol and ozone depletion within the Antarctic polar vortex during the austral spring of 1991." Geophys. Res. Lett., 19, 1819-1822.

de Zafra, R. L., Parrish, A., and Solomon, and P. M, Barrett, J. (1984). "A measurement of stratospheric H02 by ground-based millimeter-wave spectroscopy." J. Geophys. Res., 89, 1321-1326.

de Zafra, R. L., Parrish, A., Barrett, J., and Solomon, P. (1985). "An observed upper limit on stratospheric hydrogen peroxide." J. Geophys. Res., 90, 13087-13090.

de Zafra, R. L., Jaramillo, M., Parrish, A., Solomon, P., Connor, B., and Barrett, J.

(1987). "High concentrations of chlorine monoxide of low altitudes in the Antarctic spring stratosphere: Diurnal variation." Nalure, 328, 408-4 13.

Drummond, J. R., and Jarnot, R. F. (1978). "Infrared measurements of stratospheric composition II. Simultaneous NO and NO2 measurements." Proc. R. Soc. Lond A., 364, 237-254.

Eluszkiewicz, J. and M. Allen, M. (1993). "A global analysis of the ozone deficit in the upper stratosphere and lower mesosphere." J. Geophys. Res. 98, 1069-1082.

Fahey, D. W., Murphy, D. M., Kelly, K. K., Ko, M. K. W., Proffitt, M. H., Eubank, C. S., Ferry, G. C., Loewenstein, M., and Chan, K. R. (1989). "Measurements of nitric oxide and total reactive nitrogen in the Antarctic stratosphere: Observations and chemical implications." J. Geophys. Res., 94, 16665-1668 1.

Fahey, D. W., Kawa, S. R., and Chan, K. R. (1990a). "Nitric oxide measurements in the Arctic winter stratosphere." Geophys. Res. Lelt., 17, 489-492.

Fahey, D. W., Kelly, K. K., Kawa, S. R., Tuck, A. F., Loewenstein, M., Chan, K. R., L. E. Heidt (1990b). "Observations of denitrification and dehydration in the winter polar stratosphere." Nature, 344, 32 1-324.

34

Fahey, D. W., Solomon, S., Kawa, S. R., Loewenstein, M., Podolske, J. R., Strahan, S. E., and Chan, K. R. (1990c). "A diagnostic for denitrification in the winter polar stratospheres." Nature, 345, 698-702.

Fahey, D. W., Kawa, S. R., Woodbridge, E. L., Tin, P., Wilson, J. C., Jonsson, H. H., Dye, J. E., Baumgardner, D., Borrmann, S., Toohey, D. W., Avallone, L. M., Proffitt, M. H., Margitan, J., Loewenstein, M., Podolske, J. R., Salawitch, R. J., Wofsy, S. C., Ko, M. K. W., Anderson, D. E., Schoeberl, M. R., and Chan, K. R. (1993). "In situ measurements constraining the role of sulfate aerosols in mid-latitude ozone depletion." Nature, 363, 509-5 14.

Farmer, C. B., Toon, G. C., Shaper, P. W., Blavier, J. F., and Lowes, L. L. (1987a). "Stratospheric trace gases in the spring 1986 Antarctic atmosphere." Nature, 329, 126-130.

Farmer, C. B. and Raper, O. F. (1977). "The HF:HC1 ratio in the 14-38 km region of the stratosphere." Geophys. Res. Lett., 4, 527-529.

Farmer, C. B., Raper, O. F., and O'Callaghan, F. G. (1987b). "Final report on the first flight of the ATMOS instrument during the spacelab 3 mission, April 29 through May 6, 1985." JPL PubL, 87-32, Jet Propulsion Laboratory, Pasadena, CA.

Farman, J. C., Gardiner, B. G., and Shankin, J. D. (1985). "Large losses of total ozone in Antarctica revea! seasonal ClO/NO interaction." Nature, 315, 207-2 10.

Friedi, R. R., and Sander, S. P. (1989). "Kinetics and products studies of the reaction ClO + BrO using discharge-flow mass spectrometry." J. Phys. Chem., 93, 4756-4764.

Gleason, J. F., Bhartia, P. K., Herman, J. R., McPeters, R., Newman, P., Stolarski, R. S., Flynn, L., Labow, G., Larko, D., Seftor, C., Wellemeyer, C., Komhyr, W. D., Miller, A. J., and Planet, W. (1993). "Record low global ozone in 1992" Science, 260, 523-526.

Hamili, P., and Toon, O. B. (1991). "Polar stratospheric clouds and the ozone hole." Physics Tod.ay, 44, 34-42.

Hanson, D. (1990). "The vapor pressures of supercooled HNO3/1-I20 solutions." Geophys. Res. Lett., 17, 421-423.

Hanson, D. and Mauersberger, K. (1988a). "Vapor pressure of HNO3/H20 solutions at low temperatures." J. Phys. Chem., 92, 6 167-6170.

Hanson, D. and Mauersberger, K. (1988b). "Laboratory studies of the nitric acid trihydrate: Implications for the south polar stratosphere." Geophys. Res. Lelt., 15, 855-858.

35

Hanson, D. R. and. Ravishankara, A. R. (1991 a). "The reaction probabilities of C1ONO2 and N205 on polar stratosphere cloud materials." 1 Geophys. Res., 96, 508 1-5090.

Hanson, D. R. and. Ravishankara, A. R. (199 Ib). "The reaction probabilities of CIONO2 and N205 on 40 to 75% sulfuric acid solutions." J. Geophys. Res. 96, 17307-173 14.

Hanson, D. R. and. Ravishankara, A. R. (1992). "Investigation of reactive and nonreactive processes involving C1ONO2 and HC1 on water and nitric acid doped ice." 1 Phys. chem., 96, 2682-2691.

Helten, M., Patz, W., Trainer, M., Fark, H., Klein, E., and Ehhalt, D. H. (1984). "Measurements of stratospheric H02 and NO2 by matrix isolation and E. S. R. spectroscopy." J. Armos. Chem., 2, 191-202.

Hilis, A. J., Cicerone, R. J., Calvert, J. G. and Birks, J. W. (1987) "Kinetics of the BrO + ClO reactions and implications for stratospheric ozone." Nature, 328, 405-408.

Hofmann, D. J., Harder, J. W., Ro!f, S. R., and Rosen, J. M. (1987). "Balloon-borne observations of the deve!opment and vertical structure of the Antarctic ozone hole in 1986." Nature, 326, 59-62.

Hofinann, D. J., and Solomon, S. (1989). "Ozone depletion through heterogeneous chemistry foliowing the eruption of El Chichon." J. Geophys. Res., 94, 5029-504 1.

HSRP/AESA (1993). 1993 HSRP/AESA Interim Assessment." NASA reference publication, xxxx, National Aeronautics and Space Administration, Washington, D. C.. In press.

Johnston, H. S. (1971). "Reduction of stratospheric ozone by nitrogen oxide catalysis from supersomc transport exhaust." Science, 173, 517-522.

Johnston, H. S., Kinnison, D. E. and Wuebbles, D. J. (1989). "Nitrogen oxides from high-altitude aircraft: An update of potential effects on ozone." J. Geophys. Res., 94, 1635 1-16363.

KJey, D., and Stone, E. J. (1978). "A measurement of water vapor in the stratosphere by photodissociation with Ly(a) (1216 A) light." Rey. Sci. Insrrum. 49, 691-697.

Koehler, B. G., Middlebrook, A. M., and Tolbert, M. (1992). "Characterization of model

polar stratospheric cloud films using Fourier transform infrared spectroscopy and temperature programmed desorption." 1. Geophys. Res., 97, 8065-8074.

36

Kolb, C. E., Worsnop, D. R., Zahniser, Davidovits, P., Keyser, L. F., Leu, M.-T., Molina,

M. J., Hanson, D. R., Ravishankara, A. R., Williams, L. R., and Tolbert, M. R. (1993).

"Laboratory studies of atmospheric heterogeneous chemistry." In J. R. Barker, Ed.,

Current Problems in Atmospheric Chemist,y, in press.

Kondo, Y., Aimedieu, P., Pirre, M., Matthews, W. A., Ramaroson, R., Sheldon, W. R.,

Benbrook, J. R., and Iwata, A. (1990). "Diurnal variation of nitric oxide in the upper stratosphere." J. Geophys. Res., 95, 225 13-22522.

Leu, M. T. (1988a). "Laboratory studies of sticking coefficients and heterogeneous

reactions important in the Antarctic stratosphere." Geophys. Res. Leti., 15, 17-20.

Leu, M. T. (1988b). "Heterogeneous reactions of N205 with H20 and HC1 on ice

surfaces: Implications for Antarctic ozone depletion." Geophys. Res. Leti., 15, 85 1-854.

Leu, M.-T., Moore, S. B. and Keyser, L. F. (1991). "Heterogeneous reactions of chiorine

nitrate and hydrogen chioride on type 1 polar stratosphere clouds." J. Phys. Chem., 95, 7763-7771.

Louisnard, N., Fergant, A., Gramont, L., Lado-Bordowsky, O., Laurent, J., LeBoiteux,

S., and Lemaitre, M. P. (1983). "Infrared absorption spectroscopy applied to stratospheric profiles of minor constituents." J. Geophys. Res. 88, 5365-5376.

Mankin, W. G., and Coffey, M. T. (1983). "Latitudinal distributions and temporal

changes of stratospheric HC1 and HF." J. Geophys. Res., 88, 10776-10784.

Mankin, W. G., Coffey, M. T., Goidman, A., Schoeberl, M. R., Lait, L. R., and Newman,

P. A. (1990). "Airborne measurements of stratospheric constituents over the Arctic in the winter of 1989." Geophys. Res. Lett., 17, 473-476.

Manzer, L. E. (1990) "The CFC-ozone issue: Progress on the development of alternatives to CFCs." Science, 249, 3 1-35.

Massie, S. T., Davidson, J. A., Cantreil, C. A., McDaniel, A. H., Gille, J. C., Kunde, V.

G., Brasunas, J. C., Conrath, B. J., Maguire, W. C., Goidman, A., Abbas, M. M. (1987).

37

"Atmospheric infrared emission of dONO 2 observed by a bailoon-borne Fourier

spectrometer." J. Geophys. Res., 92, 14806-148 14.

Masterbrook, H. J., and Oltmans, S. J., (1983). "Stratospheric water vapor variability for

Washington, DC/Boulder, CO: 1964-82." J. Almos. Sci., 40, 2157-2165.

McElroy, M. B., Salawitch, R. J., Wofsy, S. C. and Logan, J. A. (1986) "Reduction of Antarctic ozone due to synergistic interactions of chiorine and bromine." Nature, 321, 759-762.

McElroy, M. B., and Salawitch, R. J. (1989). "Changing composition of the global stratosphere." Science, 243, 763-770.

McElroy, M. B., Salawitch, R. J., and Minschwaner, K. (1992). "The changing stratosphere." Planet. Space Sci., 40, 373-401.

McGrath, M. P., Clemitshaw, K. C., Rowland, F. S., and Hehre, W. J. (1990). "Structures, relative stabilities, and vibrational spectra of isomers of C1 202 : The role of chiorine oxide dimer in Antarctic ozone depleting mechanisms." J. Phys. Chem., 94, 6 126-6 132.

Middlebrook, A. M., Koehler, B. G., McNeill, L. S., and Tolbert, M. A. (1992). "Formation of model polar stratospheric cloud films." Geophys. Res. Lelt., 24, 24 17-2420.

Mihelcic, D., Ehhalt, D. H., Kulessa, G. F. Klomfass, J., Trianer, M., Schmidt, U. and Rohrs, H. (1978). "Measurements of free radicals in the atmosphere by matrix isolation and electron paramagnetic resonance." Pure Appl. Geophys., 116, 530-53 6.

Molina, M. J., and Rowland, F. S. (1974) "Stratospheric sink for chlorofiuoromethanes: Chiorine atom-catalyzed destruction of ozone." Nature, 249, 810-812.

Molina, L. T., Molina, M. J., Stachnik, R. A. and Tom, R. D. (1985) "An upper limit to the rate of the HC1+C1ONO 2 reaction." J. Phys. Chem., 89, 3779-378 1.

Molina, L. T., and Molina, M. J. (1987). "Production of C1202 from the self-reaction of the ClO radical." J. Phys. chem., 91, 433-436.

Molina, M. J., Tso, T.-L., Molina, L. T. and Wang, F. C.-Y. (1987) "Antarctic stratospheric chemistry of chiorine nitrate, hydrogen chloride, and ice: release of active chiorine." Science, 238, 1253-1257..

Molina, M. J., Colussi, A. J., Molina, L. T., Schindler, R. N. and Tso, T.-L. (1990) "Quantum yield of chlorine-atom formation in the photodissociation of chiorine peroxide (C100C1) at 308 nm." Chem. Phys. Lett., 173, 310-315.

38

Molina, M. J. (1991) "Heterogeneous chemistry on polar stratospheric clouds." Armos. Environ. 25A, 2535-2537.

Molina, M. J. (1992) in CHEMRA J+W VII: "Chemisriy of Ihe atmosphere. The impaci of global change." J. G. Calvert, Ed., Blackwell Sci. Pubi., Oxford.

Molina, M. J., Zhang, R., Wooldridge, P. J., McMahon, J. R., Kim., J. E., Chang, H. Y., and Beyer, K. (1993) "Physical chemistry of the H 2 SO4/HNO3IH20 system: Implication for the formation of polar stratospheric clouds and heterogeneous chemistry." submitted to Science.

Moore, S. B., Keyser, L. F., Leu, M. T., Turco, R. P., and Smith, R. H. (1990). "Heterogeneous reactions on nitric acid trihydrate." Nature, 345, 333-335.

Mount, G. H., and Eisele, F. L. (1992). "An intercomparison of tropospheric OH measurements at Fritz Peak Observatory, Colorado." Science, 256, 1187-1 190.

Mozurkewich, M., and Calvert, J. G. (1988). "Reaction probability of N205 on aqueous aerosols." J. Geophys. Res., 93, 15889-15896.

National Research Council (NRC) (1982) "Causes and effects of stratospheric ozone reduction: An update." National Academy press, Washington D. C.

Park, J. H. and Carli, B. (1991) "Spectroscopic measurements of H02, H202, and OH in the stratosphere." J. Geophys. Res., 96, 22535-22541.

Prather, M. J., McElroy, M. B. and Wofsy, S. C. (1984) "Reductions in ozone at high concentrations of stratospheric halogens." Nature, 312, 227-231.

Prather, M. J., and Watson, R. T. (1990) "Stratospheric ozone depletion and future levels of atmospheric chiorine and bromine." Nature, 344, 729-734.

Prinn, R., Cunnoid, D., Simmonds, P., Alyea, F., Boldi, R., Crawford, A., Fraser, P., Gutzler, D., Hartley, D., Rosen, R., and Rasmussen, R. (1992). "Global average concentration and trend for hydroxyl radicals deduced from ALE/GAGE trichloroethane (methyl chloroform) data for 1978-1990." J. Geophys. Res., 97, 2445-2461.

Raper, O. F., Farmer, C. B., Zander, R. and Park, J. H. (1987) "Infrared Spectroscopic measurements of halogenated sink and reservoir gases in the stratosphere with ATMOS instrument." J. Geophys. Res., 92, 985 1-9858.

Reber, C. A., Trevathen, C. E., McNeaI, R. J., and Luther, M. R. (1993). "The Upper Atmosphere Research Satellite (UARS) mission." J. Geophys. Res., 98, 10643-10647. (pp. 10643-10814 comprise a special section on UARS in this issue)

39

Reihs, C. M., Golden, D. M. and Tolbert, M. A. (1990). "Nitric acid uptake by sulfuric acid solutions under stratospheric conditions: Determination of Henry's !aw solubility." 1 Geophys. Res., 95, 16545-16550.

Ridley, B. A., McFarlan, M., Schmeltekopf, A. L., Proffitt, M. H., Albritton, D. L., Winkler, R. H., and Thompson, T. L. (1987). "Seasonal differences in the vertical distributions of NO, NO 2, and 0 3 in the stratosphere near 500 N." 1. Geophys. Res., 92, 11919-11929.

Rinsland, C. P., Levine, J. S., Goidman, A., Sze, N. D., Ko, M. K. W., and Johnson, D. W. (1991). "Infrared measurements of HF and HC1 total co!umn abundances aboye Kitt Peak, 1977-1990: seasonal cycles, long-term increases, and comparisons with model calculations." 1 Geophys. Res., 96, 15523-15540.

Ritzhaupt, G., and Dev!in, J. P. (1991). "Infrared spectra of nitric and hydrochloric acid-hydrate thin film." 1 Phys. ('hem., 90, 90-95.

Roche, A. E., Kumer, J. B., Mergenthaler, J. L., Ely, G. A., Uplinger, W. G., Potter, J. F., James, T. C., and Sterritt, L. W. (1993). "the Cryogenic Limb Array Etalon Spectrometer (CLAES) on UARS: Experiment description and performance." 1 Geophys. Res., 98, 10763-10775.

Roscoe, H. K., Kerridge, B. J., Gray, L. J., Wells, R. J., and Pyle, J. A. (1986). "Simultaneous measurements of stratospheric NO and NO 2 and their comparison with model predictions." J. Geophys. Res., 91, 5405-54 19.

Roscoe, H. K. (1991). "Review and revision of measurements of stratospheric N 205 ." 1 Geophys. Res., 96, 10879-10884.

Row!and, F. S., and Molina, M. J. (1975). "Chlorofluoromethanes in the environment." Rey. Geophys. SpacePhys., 13, 1-35.

Rowland, F. S., Sato, H., Khwaja, H., and Elliott, S. M. (1986). "The hydrolysis of chiorine nitrate and its possible atmospheric significance." 1. Phys. Chem., 90, 1985-1988.

Russell, J. M. III, Farmer, C. B., Rinsland, C. P., Zander, R., Froidevaux, L., Toon, G. C., Gao, B., Shaw, J., and Gunson, M. (1988). "Measurements of odd nitrogen compounds in the stratosphere by the ATMOS experiment on spacelab 3." JGeophys. Res., 93, 1718-1736.

Sanders, R. W., Solomon, S., Carroll, M. A., and Schmeltekopf, A. L. (1989). "Visible and near-ultraviolet spectroscopy at McMurdo station, Antarctica 4. Overview and daily measurements of NO2, 03, and OCIO during 1987." 1. Geophys. Res., 94, 11381-11391.

40

Sehilier, C., Wahner, A., Piatt, U., Dom, H.-P., Caujes, J., and Ehhalt, D. (1990). "Near uy atmospheric measurements of column abundances during airbomne Arctic stratospheric expedition, January-Februamy 1989: 2. OCiO observations." Geophys. Res. Lett., 17, 501-504.

Schoeberi, M. R., and Hartmann, D. L. (1991). "The dynamics of the stratospheric polar vortex and its relation to springtime ozone depietions." Science, 251, 46-52.

Smith, R. H., Leu, M.-T., and Keyser, L. F. (1991). "Infrared spectra of solid films formed from vapors containing water and nitric acid." J. Phys. Chem., 95, 5924-5930.

Solomon, P., de Zafra, R., Parrish, A., and Barrett, J. W. (1984). "Diurnal variation of stratosphemic chlorine monoxide: A critical test of chiorine chemistmy in the ozone !ayer." Science, 224, 1210-1214.

Soiomon, S., Garcia, R. R., Sherwood, F., and Wuebbles, D. J. (1986). "On the depletion of Antarctic ozone." Nature, 321, 755-758.

Solomon, S., Mount, G. H., Sanders, R. W., and Schmeltekopf, A. L. (1989a) "Visible spectmoscopy at McMurdo Station, Antarctica, 2. Observation of OC1O." J. Geophys. Res., 92, 8329-8338.

Solomon, 5., Miller, H. L., Smith, J. P., Sandems, R. W., Mount, G. H., Schmeitekopf, A. L., and Noxon, J. F. (1989b). "Atmospheric NO 3 1. Measurement technique and the annuai cycle at 40

0N." J. Geophys. Res., 94, 11041-11048.

Solomon, S., Sanders, R. W., Carroll, M. A., and Schmeltekopf, A. L. (1989c). "Visible and neam-ultraviolet spectroscopy at McMurdo station, Antarctica 5. Obsemvations of the diumnal variations of BrO and OCiO." J. Geophys. Res., 94, 11393-11403.

Solomon, S., Sandems, R. W., Miiler, H. L. Jr. (1990). "Visible and near-uitraviolet spectroscopy at McMumdo station, Antarctica 7. OCiO diurnal photochemistmy and implications fom ozone destruction." J. Geophys. Res., 95, 13807-138 17.

Solomon, S., and Albritton, D. L. (1992). "Time-dependent ozone depletion potentials for short- and iong-term forecasts." Nature, 357, 33-37.

Solomon, 5., Sandems, R. W., Garcia, R. R. and Keys, J. G. (1993) "Increased chlorine dioxide ovem Antarctica caused by voleanic aerosols from Mount Pinatubo." Nature, 363, 245-248.

Stachnik, R. A., Hardy, J. C., Tamsaia, J. A., Watems, J. W., and Erickson, N. R. (1992). "Submiiiimeterwave heterodyne measurements of stratospheric dO, HC1, 0 3 and H02 :

First results." Geophys. Res. LetI., 19, 193 1-1934.

41

Stimpfle, R. M., Wennberg, P. O., Lapson, L. B., and Anderson, J. G. (1990). "Simultaneous, in situ measurements of OH and H0 2 in the stratosphere." Geophys. Res. Lelt., 17, 1905-1908.

Stockwell, W. R., and Calvert, J. G. (1983). "The mechanism of the HO—SO 2 reaction." Almos. Enviro., 17, 2231-2235.

Stolarski, R. S., and Cicerone, R. J. (1974). "Stratospheric chiorine: A possible sink for ozone."J. Can. Chem., 52, 1610-1615.

Stolarski, R., Bojkov, R., Bishop, L., Christos, Z., Staehelin, J., Zawodny, J. (1992). "Measured trends in stratospheric ozone." Science, 256, 342-349.

Tabazadeh, A., and Turco, R. P. (1993). "Stratosphere chiorine injection by volcanic eruptions: HC1 scavenging and implications for ozone." Science, 260, 1082-1086.

Tolbert, M. A., Rossi, M. J., Maihotra, R., and Golden, D. M. (1987). "Reaction of chiorine nitrate with hydrogen chloride and water at Antarctic stratosphere temperature." Science, 238, 1258-1260.

Tolbert, M. A., Rossi, M. 3., and Golden, D. M. (1988a). "Heterogeneous interactions of chiorine nitrate with hydrogen chioride, and nitric acid with sulftiric acid surfaces at stratospheric temperatures." Geophys. Res. Lett., 15, 847-850.

Tolbert, M. A., Rossi, M. J., and Golden, D. M. (1988b). "Antarctic ozone depletion chemistry: Reactions of N205 with H20 and HCI on ice surfaces." Science, 240, 1018- 102 1.

Tolbert, M. A. and Middlebrook, A. M. (1990) "Fourier transform studies of model polar stratospheric cloud surfaces: Growth and evaporation of ice and nitric acid/ice." J. Geophys. Res., 95, 22423-2243 1.

Toohey, D. W., Anderson, J. G., Brune, W. H., and Chan, K. R. (1990). "In situ measurements of BrO in the Arctic stratosphere." Geophys. Res. Lett., 17, 513-516.

Toon, G. C., and Farmer, C. B. (1989). "Detection of HOC1 in the Antarctic stratosphere." Geophys. Res. LeIl., 16, 1375-1377.

Toon, G. C., Farmer, C. B., Lowes, L. L., Schaper, P. W., Blavier, J.-F., and Norton, R. H. (1989a). "Infrared measurements of stratospheric composition over Antarctica during 1986 austral spring." J. Geophys. Res., 94, 11613-11624.

Toon, G. C., Farmer, C. B., Lowes, L. L., Schaper, P. W., Blavier, J. -F., and Norton, R. H. (1 989b). "Infrared aircraft measurements of stratospheric composition over Antarctica during September 1987." J. Geophys. Res., 94, 16571-16596.

42

Toon, O. B., Hamili, P., Turco, R. P., and Pinto, J. (1986). "Condensation of HNO3 and HC1 in the winter polar stratospheres." Geophys. Res. Lelt., 13, 1284-1287.

Toon, O. B., Growell, E., Gary, B., Bait, L., Newman, P., Pueschel, R., Russell, P.,

Sehoberi, M., Toon, G., Traub, W., Valero, F., Selkirk, H., and Jordan, J. (1993).

"Heterogeneous reaction probabilities, solubilities, and physical state of coid volcanic

aerosols." submitted to Science.

Toumi, R., Kerridge, B. J., and Pyle, J. A. (1991). "Highly vibrationally excited oxygen as a potential source of ozone in the upper stratosphere and mesosphere." Nature, 351, 217.

Traub, W. A., Johnson, D. G., and Chance, K. V. (1990). "Stratospheric hydroperoxyl measurements." Science, 247, 446-449.

Tung, K.-K., Ko, M. K. W., Rodriguez, J. M., and Sze, N. D. (1986). "Are Antarctic ozone variations a manifestation of dynamics or chemistry." Nature, 322, 811-814.

Turco, R. P., Toon, O. B., and Hamili, P. (1989). "Heterogeneous physicochemistry of the polar ozone hole." J. Geophys. Res., 94, 16493-16510.

Turnipseed, A. A., Birks, J. W., and Calvert, J. G. (1991). "Kinetics and temperature dependence of the BrO + ClO reaction." 1 Phys. Chem.,95, 43 56-4364.

Van Doren, J. M., Watson, L. R., Davidovits, P., Worsnop, D. R., Zahniser, M. S., and Kolb, C. E. (1991). "Uptake of N205 and HNO3 by aqueous sulfuric acid droplets." 1. Phys. Chem., 95, 1684-1689.

Wahner, A., Jakoubek, R. O, Mount, G. H., Ravishankara, A. R., and Schmeltekopf, A. L. (1 989a). "Remote sensing observations of nighttime OCiO column during the airbonie Antarctic ozone experiments, September 8, 1987." J. Geophys. Res., 94, 11405-1 1411.

Wahner, A., Jakoubek, R. O., Mount, G. H., Ravishankara, A. R., and Schmeltekopf, A. L. (1989b). "Remote sensing observations of daytime column NO2 during the airbome Antarctic ozone experiments August 22 to October 2, 1987." 1. Geophys. Res., 94, 16619-16632.

Wahner, A., Caujes, J., Dom, H.-P., Platt, U., and Sehiller, C. (1990). "Near UV atmospheric absorption measurements of column abundances during airborne Arctic stratospheric expedition, January-February 1989: 1. Techniques and NO2 observations." Geophys. Res. LetI., 17, 497-500.

43

Wahner, A., Caujes, J., Dom, H.-P., Platt, U., and Schiller, C. (1990). "Near UV atmospheric absorption measurements of column abundances during airborne Arctic stratospheric expedition, January-February 1989: 3. BrO observations." Geophys. Res. Leti., 17, 517-520.

Waters, J. W., Hardy, J. C., Jarnor, R. F., and Pickett, H. M. (1981). "Chiorine monoxide radical, ozone, and hydrogen peroxide: Stratosphere measurements by microwave limb sounding." Science, 214, 61-64.

Waters, J. W., Stachnik, R. A., Hardy, J. C., and Jarnor, R. F. (1988). "ClO and 03 stratosphere profiles: Bailoon microwave measurements." Geophys. Res. Lett, 15, 780-783.

Waters, J. W., Froidevaux, Read, W. G., Manney, G. L., Elson, L. S., Flower, D. A., Jarnot, R. F., and Harwood, R. S. (1993). "Stratospheric ClO and ozone from the microwave limb sounder on the upper atmosphere research satellite." Nature, 362, 597-602.

Webster, C. R., and May, R. D. (1987). "Simultaneous in situ measurements and diurnal variations of NO, NO2, 03, jNO2, CH4, H20, and CO2 in the 40- to 26-km region using an open path tunable diode laser spectrometer." J. Geophys. Res., 92, 119 13-1 1950.

Webster, C. R., May, R. D., Toumi, R., and Pyle, J. A. (1990). "Active nitrogen partitioning and the nighttime formation of N205 in the stratosphere: Simultaneous in situ measurements of NO, NO2, HNO3, 03, and N20 using the BLISS diode laser spectrometer." J. Geophys. Res., 95, 1385 1-13866.

Weinstock, E. M., Phillips, M. J., and Anderson, J. G. (1981). "In situ observations of C1O in the stratosphere: A review of recent results." J. Geophys. Res., 86, 7273-7278.

Wennberg, P. O., Stimpfle, R. M., Weinstock, E. M., Dessler, A. E., Lloyd, S. A,. Lapson, L. B., Schwab, J. J., and Anderson, J. G. (1990). "Simultaneous in situ measurements of OH, H02, 03 and H2O: A test of modeled stratospheric HO chemistry." Geophys. Res. Lett., 17, 1909-1912.

World Meteorological Organization (WMO) (1986). "Atmospheric ozone 1985." WMO global ozone research and monitoring project, Rep. no. 16.

World Meteorological Organization (1990a). "Report of International Ozone Trends Panel 1988" WMO global ozone research and monitoring project, Rep. no. 18.

World Meteorological Organization (1990b). "Scientific assessment of stratospheric ozone:1989." WMO global ozone research and monitoring project, Rep. no. 20.

44

World Meteorological Organization (1992). "Scientific assessment of stratospheric ozone: 1991 ." WMO global ozone research and monitoring project, Rep. no. 25.

Worsnop, D. R., Fox, L. E., Zahniser, M. S., and Wofsy, S. C. (1993). "Vapor pressures of solid hydrates of nitric acid: Implications for the polar stratospheric clouds." Science, 259, 71-74.

Wuebbles, D. J. (1983). "Chlorocarbon emission scenarios: potential impact on stratospheric ozone." J. Geophys. Res., 88, 1433-1433.

Zander, R., Gunson, M. R., Foster, J. C., Rinsland, C. P., and Namkung, J. (1990).

"Stratospheric C1ONO2, HCI, and HF concentration profiles derived from Atmospheric

Trace Molecule Spectroscopy Experiment Spacelab 3 observations: an update." J. Geophys. Res., 95, 20519-20525.

Zhang, R., Wooldridge, P. J., Abbatt, J. P. D., Molina, M. J. (1993a). "Physical chemistry of the H2 SO4/1-120 binary system at low temperature: Stratospheric implications." J. Phys. Chem., in press.

Zhang, R., Wooldridge, P. J., Molina, M. J. (1993b). "Vapor pressure measurements for the H2 SO4/HNO 3IH20 and H2 SO4ÍHCl/H20 systems: Incorporation of stratospheric acids into background sulfate aerosols." J. Phys. Chem., in press.

45

Table 1. Properties of polar stratospheric clouds

sulfate aerosol Type 1 PSC Type II PSC

composition 40-80 wt % H2SO4 FIN0 3 •3H20 crystals H20 ice

size 0.01-1 p.m 0.3-3 J.im 1-100 tm

formation temperature 195-240 K <195 K <187 K

Table 2. Reaction probabilities (y) for heterogeneous reactions important

in the stratospherea

Reaction ?NAT

Y ice Y H2SO4solution

C1ONO2 + HCI 0.3 (200 - 202K)

0.3 (200 - 202K) -

C1ONO2 + H20 0.006 (200 -202K)

0.3 (200 - 202K) 0.064 (40 wt %, 218 K)b

HCI + HOC1 0.1 (195 -200K)

0.3 (195 - 200K) -

N205 + HC1

0.003 (200 K)

0.03 (190-220 K)

N205 + H20

0.0006 (200 K)

0.03 (195 -200 K)

0.1(200-230 K)

a. DeMore et aL, 1992

b.logyl.87-0.0747W,for4o<W<75, WisH2SO4wt%.

Table 3. OH rate constants, atmospheric lifetime, and ozone depletion potential (OPD)

of selected halocarbons.a

molecule koH (@ 280K) estimated lifetime (yrs) ODP

CFCs

CFC-11 (CFC13) - 55 1.0

CFC-12 (CF2C12) - 116

CFC-113 (CFCI2CF2C1) - 110 1.07

CFC-1 14 (CF2C1CF2C1) - 220 0.8

CFC-1 15 (CF2C1CF3) - 550 niO.5

HCFCs, etc.

HCFC-22 (CF2HC1) 3.3 x 1015 15.8 0.055

HCFC-123 (CF3CHC12) 3.0x10 4 1.7 0.02

HCFC-124 (CF3CHFCI) 7.6x10 15 6.9 0.022

HCFC-141b (CH3CFC12) 4,4x10 15 10.8 0.11

HCFC-142b (CH3CF2C1) 2.3x10 15 22.4 0.065

HCFC-225ca(CF3CF2CHCl2) 1 .9x 10-1 4 2.8 0.025 HCFC-225cb(CF 2C1CF2CHFC1) 6.5x10-15 8.0 0.033

CC14 47 1.08

CH3CC13 8.0x10 15 6.1 0.12

Brominated compounds

H-1301(CF3Br) - 66-69 16

H-1211(CF2C1Br) - 19-20

H-1202(CF2Br2) - 4 1.25

H-2402(CF2BrCF2Br) - 22-30

H-1201(CF2HBr) 7.1x10 15 6 nil.4

H-2401(CF3CHFBr) 1.6x10 14 2 0.25

H-2311(CF3CHC1Br) 5.2x10 14 1 0.14

CH3Br 3.3x10 14 1-2 0.6

a. From WMO, 1990b and WMO, 1992.

47

Figure captions

Figure 1, Average temperature profile of air aboye the earth's surface.

Figure 2. Simplified diagram of the atmospheric behavior of chlorofluorocarbons.

Figure 3. Ozone profiles measured in 1987 over McMurdo Station, Antarctica [Hofmann et al., 1987]

Figure 4. HC1 vapor pressure as a function of temperature for the HCIIH 20 system. The dashed unes represent the vapor pressures of liquids whose composition is given in wt % HCI, and the solid unes give the coexistence conditions for the two condensed phases. The dotted unes endose the thermodynamic stability region for the HC! hexahydrate, which only nucleates from liquid solutions at temperatures below 170 K, and hence, is not likely to be formed in the stratosphere.

Figure 5. (a) Liquid-solid phase diagram for the H 2 SO4/H20 system. The composition of the solids is given by the vertical unes. (b) As Figure 4 but for the H2 S041H20 system. Superimposed on these is are unes representing the equilibrium compositions of supercooled liquid droplets in the stratosphere, assuming 3 ppmv of H20 at 100 mb, that is —16 km altitude. F is the ice frost point, and M is where crystalline H 2 SO4 4H20 would melt upon warming under these conditions.

Figure 6a. Nitric acid vapor pressure as a function of temperature for the I-1NO 3/H20 system. The dashed unes represent the vapor pressures of liquids whose compositions are given in wt % HNO 3, and the solid unes represent coexistence for two condensed phases.

Figure 6b. Log of nitric acid partial pressure vs. log of water partial pressure for the HNO3 rn}i20 crystalline hydrates. The dashed unes indicate temperatures, and the solid unes represent coexistence for two condensed phases. Note that the siopes of the dashed unes are -n, and that for a given water vapor pressure NAT (HNO3 •3H20) is stable at —5 K warmer than ice under typical polar stratospheric conditions (shaded region).

Figure 7. Ternary diagram for the H2 SO4IHNO3/H20 system. The dashed unes indicate the eutectics between the crystalline phases [Carpenter and Lehrman, 19251. Superimposed upon this are vapor-liquid equilibrium dilution curves (solid unes) for stratospheric aerosol droplets at 100 mb (-16 km) and at ambient mixing ratios of 5 ppmv H20; and 10 ppbv I-1NO 3 (a), 5 ppbv FIN0 3 (b), and 2.5 ppbv HNO3 (c), as estimated from H2 SO4/HNO 3/H20 vapor pressure data. Also along the dilution unes are the temperatures (dotted unes), the frost point of crystalline HNO 3 •3H20 (i.e., NAT supersaturation,

48

300 1989

• 19901

18811 290 a

.01

o

1 iT •.1 •.•••••..••. \ ¡ I

,__/

280

270

2601 i 1 1 1 1 1 1 1 1 1 1 1 Jan Mar May Jul Sep Nov

aL

u u

CIO MIXING RATIO(pptv) cn O O O

ru

O

o pl

o CJ)-.4J

O

o o o

03 MIXING RATIO (ppbv)

II 1 1 1 1 1 1 I ;: -

-Ç

-

SUMMARY OF LATITUDE VARIATIONS OF STRATOSPHERIC TRACE GASES

6 c'J E o o

COLLAR o k CORE

CINO3 NO2

1,01

1 -

LHCI LHN0(~ 4)

/ 1

CINO3--- HF - - - - -

- - -

NO2

HNO3 (--4)

NOi rHF

HCI OL_ 30

40 50 60

70 80 90

LATITUDE (°S)

0] 1 lO lOO

40

< 3C

11

2C 0.1 lo loo

MIXING RAT1O (ppbv)

fb

OH MIXING RATIO (ppbv) 0.001 0.01 0.1

nE

40

w

1-

li <3C

2(

0.01 0.1 1 lO MIXING RATIO (ppbv excepf OH)

SNAT,1), and the point at which the HNO 3 vapor pressure reaches a supersaturation of 10 with respect to NAT (SNAT=lO). Liquid vapor pressures used for the interpolation were those measured over ternary solutions having 35to 70 wt % H2SO4 [Zhang et al., 19931 and for O % H2SO4 extrapolations of those reported for supercooled HNO 3 (aq) to lower temperatures [Hanson, 1990].

Figure 8. Calculated catalytic cycle fractional contributions to odd oxygen destruction: gas phase only and (b) with heterogeneous hydrolysis of N 205 {McElroy

et al., 19921.

Figure 9a. Average ozone concentration versus altitude measured over Payerne, Switzerland, for three 2-year periods, 1969-1970, 1979-1980, and 1989-1990. Periods were chosen to be approximately during solar maximum and 2 years were used to remove most of any quasi-biennial oscillation effect. [Stolarski et al., 1992].

Figure 9b. Daily global ozone amount (area-weighted 65°S to 65°N) from NOAA-1 1 SBUV/2. The 1992 data are represented by the thick solid line. The 1991 data are represented by the dotted une. The 1990 data are represented by the dashed line. The 1989 data are represented by the thin solid une. [Gleason et al., 19931.

Figure 10. Calculated and observed concentrations for the (a) odd nitrogen fainily and the inorganic chlorine family, (total inorganic chiorine

Cl=HCl+ClONO2+ClO+HOCl) [McElroy and Salawitch, 19891. Data from the ATMOS experiment for 30°N on 30 April and 1 May, 1985 are shown as dotted unes [but with observed C1ONO 2 profiles updated with the results of Zander et al., 19901. Model results are shown by the solid curves. Also shown in (b) is the calculated profile for OH.

Figure 11. September 16, 1987 measurements of chlorine monoxide and ozone carried out during the Airborne Antarctic Ozone Experiment [Anderson et al., 1989].

Figure 12. Variation of trace gases from mid-latitudes to the core of the Antarctic polar vortex during September, 1987. The dashed lines are for the region where there were no measurements. In the "collar" region, the large C1ONO 2 concentrations may be the result of mixing C1O-rich air from inside the vortex with NO 2-rich from outside [Toon et al., 1989b].

Figure 13. Measured (1960-1990) and projected (from 1990 on) atmospheric chiorine loadings with and without international protocols [C&E News, 19931.

49

—80 —60 —40 -aD O ¿U

TEMPERATURE (°C)

1i

Ui cr- (J (1, u.'

o-

IR

70

60

E

• 5C

a ELO

EE.I.Is]

01

E

o\i \Ot

0 C4 T4 LYT C

\cli

CIO

TROPOS

1..

0 100 200

30

20 U)

a) E o

LLJ o 1-

-J

íEi

a U)

w cr- u) u) LLJ cc a-

OZONE PARTIAL PRESSURE (nanobars)

TemperatUre (K)

280 240 200 180 160 -2 10

LIQUID HC13H20 \ \\ \

10 \\ 10 15 20 25ç% Weight HCI

\ 5 \ \

-4 Io

6 H20

ICE' a..

e lO- '

o

- 7 01 ' LO = \

\ t•• ....... polar

108 Stratosphere

10 3.5 4.0 4.5 5.0 5.5 6.0

1000/T (K)

1,Ii

280

260

a,

4-

o a) o- E220 a, -

200

Mole Fraction H2 SO 4 0.05 0.1 0.5 LO

0 20 40 60

iz

Dy

%wt H2SO4

mi

2 . lo

o - (sJ 10

= a-

TEMPERATURE (K)

270 250 230 210 200 190 •1

37.6 Wf% H 2 SO4

45.6 \\\ \\ \

64.5-\ \ \ \\ —PURE ICE 577 \\\

•'\ 67.8—'

4.

4. H2504.6.5H20 .4 \ \ .4

.4

.4 .4

.4 .4

HSOÇ4HO

H2SO4'H20 't 2

.4

-\ F

1 -4 lo

M'

-5 L ¡o 3.5

tetra-/di hydrate

4.0 4.5

lOOO/T(K)

ice/crtá 1 tin hydrates

5.0 5.5

Temperafure (K)

r,or 94fl 220 200

1

'o

o

33

- - 1-...__ ¡ 1

7080 90 °o Weight HNO3

60 -

50 N

40 HNO3 H 2 0 -

30 -

20 HN03 3 H2 0

10

N •\. •

\ o - o- o >

fi) o z 1 -7

I0

"ICE"

•\ Polar

Stratosphere HN

N

NSI

1'

3.5 4 4.5 -

I000/T (K ' )

5.5 T1

o 10

•1-

a)

u,

65

u, a)

o a- o

rn o z 'lo

\E - HNO 3'H20

230K D

2201< ' c - o

\ 210K

205K\\ - '

2001< \ /\

-195K\ /X / \ '

\ ,• .' \ '

,• ' .'ç \ i 1 /

2H20/ S '

/ 1 UICEIS

Ç 1

j I

3

Q

/

EI

l0 IO 10'

H20 Vapor Pressure (torr)

F 7 (7.

ct HNO3

0.7

o

o . 00 -(

3 O O O.5.- (iO •...

0.4

03

0.2

0.1

O

-

\ ••...\\\\ II

ICE '\

N

O

H20 WEIGHT FRACTION H 2SO4 H2SO4--

30

25

¡26

o24

22

20

ki

B r

1 1 / 03+0

1.

/ 1 1

2 1 ¡ 1

fi N 0X \

18 1

10-2 10-1 100 FRACT1ON OF TOTAL LOSS

30

25

26

24

22

20

NO2 -i-o ¡

¡

¡ ¡

/ /

/

/

HOx'\

\ \\

03+O\

\Br 4

- \J • \ /../

-

•

18 1 10 10 1 100

FRACTION OF TOTAL LOSS

cD a-

s1

U3

a-

Feii1

¡Iii]

2

001

25

r' • I 1 e-

-

o o

15E

o o. o-

10<

5

ti

-

—

- 1989to1990'\ '4

1 '4

J - 1979to1980- --9

- - - - — .#d•_ __ - , ---

-

1969to1970 - #1

/1 as

I

/ 1 • 1

• 1 -

1 a •

'4 -

'4

-

-

1 1 1 It tI 1 II 1 1 1 1 • 0 50 100 150

Ozone concentration (nbar)

(5 z o 94 w z

Anfarctic one hole appears

Under 1992 revisions

F1.I:Ii FsI.I']

-J 1 o. o w 1 Q1 CO o

~- o

Antarcfic hole may disappear

F;~

Under original Montreal protocol

.Uney $990 / revisions

YEAR

7

o- o-

2. MOLINA & MOLINA Stratospheric Ozone 25

Chapter 2

Stratospheric Ozone

Mario J. Molina and Luisa T. Molina

Department of Earth and Planetary Sclences, 54-1312, Massachusetts Institute of Technology, Cambrldge, MA 02139

The ozone layer is a veiy important component of the atmosphere which shields the earth's surface from damaging ultraviolet radiation from the sun. Ozone is continuously being generated by the action of solar radiation on atmospheric oxygen, and it is destroyed by catalytic processes involving trace amounts of free radical species such as nitrogen oxides. More than a decade ago the release of chlorofluorocarbons (CFCs) of industrial origin was predicted to lead to stratospheric ozone depletion. Photodecomposition of the CFCs in the stratosphere produces significant amounts of chlorine free radicals, which are ver)' efficient catalysts for the destruction of ozone. Recent observations have established clearly that the rapid decline in ozone over Antarctica in the spring months is indeed caused by man-made chiorine species.

Ozone (03) is continuously being generated in the atmosphere by the action of solar ultraviolet radiation of wavelengths shorter than about 220 nm on molecular oxygen (0 2) to fomi atomic oxygen (0), followed by the recombjnation of 0-atoms with Oz:

02 +hv->0+0 (1)

0+02 ->03. (2)

Ozone is a relatively unstable species which is easily destroyed by various chemical processes. Its peak concentration is several parts of ozone per million parts of air, and it is found primarily in the stratosphere, the region between about 10 and 50 km aboye the Earth's surface.

Ozone absotbs ultraviolet radiation vety efficiently in the wavelength range between 200 and 300 nm, where molecular oxygen and nitrogen are

practically transparerit. Hence, one of the important functions of the stratospheric ozone layer is to shield the surface of the Earth from solar ultraviolet radiation, which is harmful to living organisms. This ultraviolet radiation is known to cause human skin cancer as well as pose a threat to certain crops, forests, and ecological systems. In the process of absorbing this radiation the ozone molecule is destroyed:

03 + hv -> 02 + 0 (3)

This proccss does not lead to net ozone depletion because it is rapidly fol]owed by reaction 2, which regenerates the ozone. Reactions 2 and 3 have, however, another important function, namely the absorption of solar energy, as a result, the temperature increases with altitude, and this inverted temperature profile gives risc to the stratosphere (see Figure 1). In the lower layer, the troposphere, the temperature decreases with altitude and verticai mixing occurs on a relatively short tune scale. In contrast, the stratosphere is very stable towards vertical mixing because of its inverted temperature profile.

The oxygen atoms, formed in the stratosphere predominantly by reaction 3, occasionally react with ozone instead of adding to molecular oxygen:

0 +03->02+02 (4)

Reactions 1 to 4 are known collectively as the Chapman mechanism, (first outlined by Sidney Chapman (1) in 1930. They basically explain how ozone can exist iii the stratosphere in a dynamic balance; it is continuously beirig produced by the action of solar ultraviolet radiation on oxygen molecules and destroyed by several natural chemical processes in the atmosphere.

Catalytic ozone destruction cycles

The concentration of ozone in thc stratosphere is lower than predicted from reactions 1 - 4. This is due to the presence of trace amounts of some reactive species known as free radicals. These species have an odd number of electrons and they can speed up reaction 4 by means of catalytic chain reactions. Nitrogen oxides, NO and NO 2, which are naturally present in the stratosphere at leveis of a few parts per billion (ppb), are the most important catalysts in this respect. The reactions, first suggested by Paul Crutzen (2) and by Harold Johnston (3) in the early 1970's, are as follows:

NO + 03 -> NO2 + 02 (5)

NO2 + O -> NO + 02 (6)

0097-6 156/92fl)483-0024s1J600flJ Net reaction: O + 03 -> 02 + 02. (4) © 1992 American Chemical Society

26 THE SCIENCE OF GLOBAL CHANGE

2. MOLINA & MOLiNA Stratospheric Ozone 27

Ci + 03 -> 00 + 02 (8)

OO+O->C1+O2 (9)

Net reactjon: O + 03 -> 02 + 02. (4)

Dic stratosphere contains, however, only small amounts—a few tenths of a ppb—of chiorine free radicais of natural origin. They are produced by the decomposition of methyl chloride, CH 3CI. The nitrogen oxides (NO and NO2) are more abundant and are produced in the stratosphere by the decomposition of nitrous oxide, N20. Both CH30 and N20 are of biological origin: these compounds, released at the Earth's surface, are sufficiently stable to reach the stratosphere in significant amounts.

The Role of Chlorofiuorocarbons

In 1974, Molina and Rowland (5) suggested that chlorofluorocarboji (CFCs) could provide an important source of chlorine free radicais to the stratosphere and hence would pose a threat to the ozone layer. Dic CFCs are man-made chemicals used as refrigerants, solvents, propellants for spray cans, biowing agents for plastic foam, etc. Dic two most important ones are CFC-11 (CF0 3) and CFC-12 (CF202). Diese compounds are chemically inert and insoluble in water; thus, they are not removed in the lower atniosphere, in contrast to most other gases released to the environment at the Earth's surface. Instead, the CFCs risc into the stratosphere, where they are eventually destroyed by short-wavelength solar ultraviolet radiation of the type that is shielded by the ozone layer. Because diffusion into the stratosphere is very slow, the residence time for the CFCs in the environment is of the order of a century.

The photodecomposition of the CFCs Jeads to the release of chiorine

fi

atoms in the ozone layer. Diese atoms can then participate in reactions 8 and 9, as well as in other chemical and photochemical reactions. A schematic representation of the more important reactions is shown in Figure 2. Dic ClOx catalytic chain mechanism (Reactions 8 and 9) may be interrupted, for example, by reaction of the Cl atom with methane (CH 4) to produce the relatively stable hydrogen chioride molecule (HO); or by reaction of chiorine monoxide (CIO) with NO 2 or H02 to produce CIONO2 (chlorine nitrate) or HOCI (hypochlorous acid):

Cl + CH4 -> HO + CH3 (10)

CiO + NO2 -> OONO2 (11)

00 + H02 -> HOCI + 02. (12)

The chlorine-containing product species (Ha, CIONO 2, HOCI) are "inert resezvoirs" because they are not directiy invoived in ozone depletion; however, they eventually break down by absorbing solar radiation or by reaction with other free radicais, returning chlorine to its catalytically active form. Ozone is formed fastest in the upper stratosphere at tropical latitudes (by reactions 1 and 2), and in those regions a few percent of the chlorine is in its active "free radical" form; the rest is in the "inert reservojr" form (see Figure 3).

In order to estimate the extent of ozone depletion caused by a given release of CFCs, computer modeis of the atmosphere are employed. Diese modeis incorporate information on atmospheric motions and on the rates of over a hundred chemical and photochemical reactions. Dic results of measurements of the various trace species in the atmosphere are then used to test the modeis. Because of the compiexity of atmospheric transport, the calculations were carried out initially with one-dimensional models, averaging the motions and the concentrations of chemical species over latitude and longitude, leaving only their dependency on altitude and time. More recently, two-dimensjonal models have been developed, in which the averaging is over longitude only.

Dic fundamental aspects of the problem are well established: the measured concentrations of the CFCs indicate that they accumulate in the lower atmosphere and that they reach the stratosphere. As expected, chlorine atoms and 00 radicais are found in the stratosphere together with other species such as O, OH, HO 2, NO, NO2, HCI, CIONO2, HOCI, etc. Dic observed concentrations are in reasonabie agreenient with the model predictions if the iimitations of the models, as weil as atmospheric variabiiity, are taken into account.

Obsei-ved Stratospheric Ozone Trends

It is only recently that a decrease in stratospheric ozone Jeveis attributable to the CFCs has been observed. In spite of the relatively large natural

Reactions 5 and 6 constitute a catalytic cycle because the radical NO that attacks 03 is regenerated by the reaction of NO 2 with an 0-atom. The net effect is the removal of one 0 3 molecule and one 0-atom. Thus, although the concentration of NO and NO 2 (or NOx) in the stratosphere is small, each NO molecule can destroy thousands of ozone molecules before being scavenged by a reaction such as the following:

OH + NO2 -> HNO3 (7)

Other important catalysts are the free radicais OH and H 021 produced in the stratosphere by the decompositjon of water vapor.

Chlorine atoms are also veiy efficient ozone destruct ion catalysts, as noted originally by Stolarski and Cicerone (4):

- MESOPAUSE-

MESOSPHERE '

¿

- STRATOPAUSE -

'7 STRATOS Pl-lE RE

Mt ( 1 EVEREST

TROPOPAUSE -

-80 -60 -40 -20 0 20 40

TEMPERATURE (°C)

Figure 1. Average temperatura profile of air aboye the earth's surface.

Figure 2. Schematic representation of the more important reactions and compounds in the stratospheric chemistry of chiorine a 'w and mid-latitudes.

80

70

60 E

50 Li

-. 40

-J 4 30

20

El

0.01

0.1

E

to a: a-

loo

1000

28

E SCIENCE OF GLOBAL CHANGE 2. MOLINA & MOLINA Stratosplzeric Ozone

29

fluctuations in ozone Jeveis, it has been possible to show, by careful examination of the records over the last two decades, that statistically significant changes have occurred in the winter months at high latitudes (6). Furthermore, as first pointed out by Farrnan and co-workers (7), the ozone levels over Antarctica have dropped dramatically in the spring months starting in the early 1980's (see Figure 4). This Antarctic ozone hole was not predicted by earlier models; its cause was not clear until recent years, when laboratory experiments, fleid measurements over Antarctica and model calculations provided veiy strong indications that the ozone loss can indeed be traced to man-made CFCs.

Polar Ozone Chemistry

The high-latitude (polar) stratosphere has several unique characteristics. High-cnergy solar UV radiation is scarce over the poles, thus ozone is not generated there. However, the total ozone column abundance is large in this region because ozone is transportad towards the poles from lower latitudes and higher altitudes. Furthermore, temperatures over the poles are veiy low. The catalytic cycles responsible for ozone destruction are active mainly at higher temperatures and in the presence of abundant solar Uy radiation; thus ozone is predicted to be vexy stable over the poles based on conventional gas phase chemistry, and a chemical explanation of Antarctic ozone depletion requires a different mechanism.

The stratosphere is veiy dry clouds do not form at lower latitudes because the temperature is not low enough. 1-lowever, the stratosphere over Antarctica is distinctive: the temperature can drop to below 9O 0 Celsius during the winter and spring months, leading to the condensation of water vapor and nitric acid vapor, that is, to the formation of ice clouds (polar stratospheric clouds or PSCs).

Several authors (8,9) suggested that PSCs could play a major role in the depletion of ozone over Antarctica by pronioting the release of active chlorine from its reservoir species, mainly by the following reaction:

Ha + dONO2 -> a2 + HNO3. (13)

Laboratory experiments by our group showed that reaction 13 occurs veiy slowly in the gas phase (10). However, in the presence of ice surfaces the reaction proceeds veiy efficiently: the product °2 is immediately released to the gas phase, whereas HNO 3 remains frozen in the ice (11). Other groups also found that this heterogeneous (i.e., multiphase) process occurs efficiently (12,13), and that a similar reaction also occurs with N 205 as a reactant:

HCI + N205 -> aNO2 + HNO3 (14)

Both 02 and CINO2 absorb visible and near ultraviolet radiation, so that they can readily photolyze even with the faint amount of sunlight

CYCLES -

Figure 3. Simplified diagram of the atmospheric behavior of chlorofluorocarbons.

/7 1987

30 15 AUGUST 1987

E ,çç - •1

20 )

1-

1- -

-J

0 0.5 1.0 l.

OZONE PARTIAL PRESSURE (10 4 torr)

Figure 4. Ozone profiles measured in 1987 over Halley Bay, Antarctica, by Farman (20).

30 TITE SCIENCE OF GLOBAL CHANGE

2. MOLINA & MOLINA Stratospheric Ozone 31

available in the early spring over Antarctica. This process releases free chlorine atoms which react rapidly with ozone, producing oxygen molecules and chiorine monoxide (reaction 8).

The mechanism for reaction 13 is not well established yet, but it is likely to proceed through ionic intermediates (11): the Cl atom in chiorine nitrate is slightly electropositive, so that it readily combines with negative chloride ions to produce C12; the HCI on ice is expected to be at least partially ionized. Wc have found that HCI has a veiy high mobility on the ice surface, so that even small amounts of HCi will enable reaction 13 to occur. It is also possible for this reaction to proceed in two steps: the initial step is the reaction of chlorine nitrate with ice; it is followed by the reaction of the product HOC1 with HCI on the ice substrate:

dONO2 + H20 -> HOCI + HNO3 (15)

HOC1 + HCi -> H20 + a2 (16)

Laboratory experiments have shown that reaction 15 occurs on ice in the absence of Hd (11-13); furthermore, the product HOC1 appears on a time scale of minutes, in contrast to C1 2 in reaction 13, which is produced on at most a millisecond time scale (11). Thus, in this mechanism HOCI serves as an mtermediate: if there is enough H on thc ice, HOCI will react with HC1 while still on the ice surface; otherwise the HOCI will desorb, eventually finding an HCi molecule in the ice, perhaps after several adsorption-desorption cycles.

The presence of PSCs also Ieads to the removal of nitrogen oxides (NO and NO2) from the gas phase. As long as there are significant amounts of NO2 it will react with chlorine monoxide (CO) to produce chiorine nitrate (reaction 11). This species subsequent]y reacts with HCi on PSC surfaces to produce nitric acid (reaction 13), which remains in thc condensed phase. Also, nitric acid directly condenses with water to form nitric acid trihydrate particles, hence it is not available to regenerate NO 2 by photochemical processes, as it does when it is in the gas phase.

The catalytic cycle described earlier (reactions 8 and 9) cannot explain the rapid depletion of ozone over the South Pole, because reaction 9 requires free oxygen atoms, which are too scarce in the polar stratosphere to react at any appreciable rate with ao. Several catalytic cycles that do not require oxygen atoms have been suggested as being at work over Antarctica.

In 1986 we proposed a cycle that involves the self reaction of chiorine monoxide radicais, without requiring free oxygen atoms to regenerate the chiorine atoms (14):

32 TRE SCIENCE GLOBAL CHANGE

2. MOLINA & MOLINA Stratospheric Ozone

33 00 + 00 -> 0202 (17)

2(0 + 03 -> 00 + 02) (8)

0202 + hv -> CI + 000 (18)

000 ->Cl+02 (19)

Net reaction: 203 + hv -> 302 (20)

Much has been learned in recent years about the "00 dimer", 0202, produced iii reaction 17. It is actually dichiorine peroxide, 000C1; its geometiy is now well established from submillimeter wave spectroscopy (15). Photolysis of 0000 around 310 sun - the atmospherically important wavelengths - yields chiorine atoms and 000 radicals (16), as given in reaction 18, rather than two 00 radicals, even though CIO-OCI is the weakesi bond (it has a strength of about 17 Kcal/mol (17)). Thermal decomposition of 0000 (the reverse of reaction 17) occurs ver)' fast at room temperature, but more slowly at polar stratosphenc temperatures. Hence, photolysis is the predominant destruction path for CIOOCI in the polar stratosphere and two Cl atoms are produced for each ultraviolet photon absorbed.

McElroy a al. (18) suggested a cycle in which chiorine and bromine are coupled in the destruction of ozone:

00 + BrO -> Cl + Br + 02 (21)

Br + 03 -> BrO + 02 (22)

0+03->00+02 (23)

Net reaction: 203 + hv -> 302 (20)

There are natural sources of brominated hydrocarbons as well as man-made sources, such as the 'halons", which are used in fire extinguishers. Reaction 21 is veiy fast and generates Cl and Br atoms directly; the cycle does not require a photolytic step. Although this cycle occurs with high efficiency, it is less important than the chlorine peroxide cycle because of the much smaller concentrations of bromine compounds in the stratosphere—parts per trillion Vs. paris per billion for the chlorine compounds.

These catalytic cycles proceed efficiently as long as the NO 2 leveis are Iow; otherwise NO2 reacts with ClO(reaction 11), interferingwith reactions 17 and 21. Thus, PSCs are important not on!y because they release chiorine from inert reservoirs, but also because they scavenge nitrogen

oxides from the gas phase, thereby setting the stage for rapid ozone destruction.

Atmospheric Measurements In the Polar Stratosphere

Several key expeditions have been launched in the last few years to measure trace species in the stratosphere over Antarctica as well as over the Arctic. The results provide very convincing evidence for the occurrence of the chemical reactions discussed aboye and demonstrate the critical role played by man-made chiorine in the formation of the Antarctic "ozone hole". (See, for instance, J. Geophys. Res. 94, Nos. D9 and D14, 1989, for a collection of articles on the Antarctic expedition of 1987, and Geophys Res. LetL 17, No. 4, 1990, for the Arctic expedition of 1989). When ozone is being depleted in the spring months over Antarctica, a vezy large fraction of the chlorine is present as the free radical 00 (see Figure 5); NOx levels are veiy low; there are cloud particles and they contain nitrate; etc. Over the Arctic a large fraction of the chlorine is also activated, but ozone depletion is Iess severe because the temperatures are not as low as over Antarctica: the active chiorine remains in contact with ozone only briefly before the Arctic air masses mix with warmer air from lower latitudes. This air also contains NO2, which passivatcs the chiorine.

A detailed analysis of tite atmospheric measurements over Antarctica by Anderson eral. (19) mdicatcs that tite cycle comprising reactions 17 - 19 (the chlorine peroxide cycle) accounts for about 75% of the observed ozone depletion, and reactions 21 - 23 account for the rest. While a clear overail picture of polar ozone depletion is emerging, much remains to be learned. For example, tite physical chemistzy of tite acid ices titaS constitute polar stratospheric clouds needs to be better understood before reiable predictions can be made of future ozone depletion, particularly at northern latitudes, where tite chemical changes are more subtle and occur over a larger geographical area.

Tite Montreal Protocol

Under the auspices of tite United Nations Environment Progranime, many countries--including the industrialized nations of tite world—signed an agreement in 1985 in Vienna to regulase tite production of CFCs. This initial agreement was followed by the Montreal Prorocol in 1987, which called only for a reduction of 50% in tite manufacture of CFCs by the end of the century. In view of tite strength of the scientific evidence linking stratospheric ozone depletion with the release of CFCs, the initial provisions were strengthened in 1990 through the London Amendinents to the protocol: the CFCs will be essentially phased out by the end of the century. Other cornpounds, such as the halons, carbon tetrachloride (Ca 4) and methyl chloroform (C1-1 3CCI3), which were not included in the initial negotiations, will also be regulated. Because of the long residence times of the CFCs in the atmosphere, even if the protocol were fully enforced, ozone depletion would continue well into the next century.

2.50

E 1.75 aa

w z

1.250 o

0.75

0.25

34

TITE SCIENCE OF GLOBAL CHANGE 2. MOLINA & MuLINA Stratospheric Ozone 35

In terms of the uses of CFCs, roughly a third of the amount now produced will be replaced with hydrochlorofluorocarbons (HCFCs) or hydrofluorocarbons (HFCs), which are compounds with chemical and physical properties similar to those of the CFCs, except that their molecules contains hydrogen atoms. They are destroyed predominantly in the troposphere by reaction with the hydroxyl radical (OH), forming water and an organic free radical which rapidly photo-oxidizes to yield water-soluble products. Hence, only a small fraction of their release reaches the stratosphere. Another third of the CFC usage will be replaced by "not in ldnd" compounds; for example, FC-1 13, which is used to clean electronic boards, can be replaced in many instances by soap-and-water-based or terpene-based solvcnts. Finally, the last third of the usage will be dealt with by conservation: current practicas lead to an unnecessarily large release of CFCs to the environment for certain uses such as cleaning solvents or as refrigerants for automobile air conditioning.

The CFC-ozone depletion issue has demonstrated that mankind has the potential to seriously modify the atmosphere on a global scale. We need to learn much more about the environment to prevent its inadvertent deterioration by human activities.

A.

:T'k" 1 i

'4 —

64 66 68 70 72 LATITUOE (degrees south)

Figure S. Measurements of chiorine monoxide byAnderson etaL (19) and of ozone by Proffitt el aL (21) carried out in 1987 during the Airborne Antarctic Ozone Experiment.

Literature Cited

Chapman, S. Mem. R. MeteoroL Soc., 1930, 3, pp. 103-105. Crutzen, P. J. Q. J. R. Meteoro!. Soc., 1970, 96, pp. 320-325. Johnston, H. Science, 1971, 173, pp. 517-522. Stolarski, R. S.; Cicerone, R. J. Can. J. Chem., 1974, 52, pp. 1610-15. Molina, M. J.; Rowland, F. S. Nature, 1974, 249, pp. 810-812.

Report of ¡he International Ozone Trends Panel 1988, Global Ozone Research atId Monitoring Project Report No. 18, World Meteorological Organization, Geneva, Switzerland, 1988. Farman, J.C.; Gardiner, B.G.; Shanklin, J.D. Nature, 1985, 315, pp. 207-210.

S. Solomon, S.; Garcia, R.R.; Rowland, F.S.; Wuebbles, D. J. Nature, 1986, 321, pp. 755-758. Toon, O.B.; Hamili, P.; Turco, R.P., Pinto, J. Geop/zys. Res. Leu, 1986, 13, pp. 1284-1287. Molina, LT.; Molina, MJ.; Stachnik, R.A.; Tom, R.D. J. Phys. Chern., 1985, 89, 3779-3781. Molina, MJ., Tso, T.-L, Molina, LT.; Wang, F.C.-Y. Science, 1987, 238, pp. 1253-1257. Leu, M.T. Geophys. Res. LetI., 1988, 15, pp. 17-20. Tolbert, MA; Rossi, M.J.; Maihotra, R.; Golden, M.D. Science, 1987, 238, pp. 1258-1260. Molina, LT.; Molina, M.J. J. Phys. Chem., 1987, 91, pp. 433-436. Birk, M.; Friedi, R.R.; Cohen, EA; Pickett, H.M. J. Chem. Fhys., 1989, 91, pp. 6588-6597.

Molina, M.J.; Colussi, AJ.; Molina, LT.; Schindler, R.N.; Tso, T.-L Chem. Phys. LetL, 1990, 1973, pp. 310-315. Cox RA; Hayman, G.D. Nature, 1988,322, pp. 796-800. McElroy, M.B.; Salawitch, R.J.; Wofy, S.C.; Logan, JA Na~, 1986, 321, pp. 759-762. Anderson, J.G.; Toohey, D.W.; Brune, W.H. Science, 1991, 251, pp. 39-46. Farman, J.C. New Scientirt, 1987, 12, pp. 50-54. Proffitt, M.H.; Steinkamp, MJ.; Powell, JA; McLaughlin, RJ.; MilIs, OA; Schmeltekopf, A.L; Thompson, T.L; Tuck, A.F.; Tyler, T.; Winker, R.H.; Chan, K.R. J. Geophys. Res., 1989, 94, pp. 16,547-16,555.

RECEIVED October 7, 1991

Reprinted from ACS Symposium Senes No. 483 77,e Sciencc of Global Change David A. Dunnettc and Robert J. O'Bricn, Editors Copynghc © 1992 by the American Chemical Society Reprinted by permission of the copyright owner

.0 a 0.

o u o

)