5 -- HETEROGENEOUS CHEMISTRY

5.1 Introduction

Heterogeneous chemical reactions are crucial to understanding the chemical balance of the stratosphere. A heterogeneous reaction is a chemical process that involves solid, liquid, and gaseous phases (hence, it is a multiphase process). Heterogeneous chemistry occurs on or in condensed particles (such as liquid water droplets or solid ice particles) that are in contact with gaseous molecules. Such multiphase processes include adsorption (i.e. adhesion of a thin layer of molecules to a surface) or absorption of molecules onto particles. This is followed by chemical reactions on the particle surfaces or within the particles themselves. Prior to 1985, such heterogeneous reactions were thought to be unimportant, but they are now recognized as being quite important in stratospheric chemistry. Indeed, they are the physical mechanism behind the Antarctic ozone hole phenomenon.

Chemical reactions on the surfaces of particles take place via a chain of processes. First, the reactant species must make contact with the particles by kinetic motion. Second, the reactants are absorbed onto the surface of the particle by either physical or chemical bonding, where they then diffuse into the body of the particle or remain on the surface. Third, the reactants meet on (or in) the particle, chemically forming different products. These products will then diffuse into the particle, remain on the surface, or be desorbed (emitted) from the particle.

This complex chain of events is represented as the product of a number of parameters, including the concentrations of the reactants, the total particle surface area in a given volume of air, a so-called "sticking coefficient" (representing the probability that a molecule will actually stick to a particle after a collision), temperature (representing the speed or kinetic energy of the molecules in the air parcel), and the particle type. These parameters allow us to represent heterogeneous reactions in computer mathematical models. This also requires us to include the particle surface area, a difficult task.

5.2 Stratospheric Particles

5.2.1 Sulfate aerosols -- Various particles exist throughout the stratosphere. Their discovery and inclusion within atmospheric models was a great step forward in understanding the chemistry of the stratosphere. Only few types of particles in the stratosphere exist on which gases can either be adsorbed (adhesion to the surface of a particle) or absorbed. The most common particles are stratospheric sulfate aerosols. These submicron sized particles were discovered by Junge et al. (1961) and are typically composed of a solution of sulfuric acid and water, at least in the middle latitudes where the temperatures are warm enough to maintain the particles in a liquid state. The sulfuric acid comes from carbonyl sulfide (COS) and sulfur dioxide (SO2) carried into the stratosphere via tropical lifting by the Brewer-Dobson Circulation, or by direct injection of SO2 into the stratosphere from very explosive volcanic eruptions, such as El Chichon in 1982 or Mount Pinatubo in 1991. Because sulfate aerosols are small, they settle out of the stratosphere at a very slow rate of about 100 meters per year for a spherical aerosol with a radius of 0.1 microns (4 millionths of an inch). Because of this small settling speed, most sulfate aerosols are carried out of the stratosphere by Brewer-Dobson Circulation descent in the higher latitudes (see Chapter 6).

5.2.2 Polar stratospheric clouds (PSCs) -- The second class of particles in the stratosphere on which heterogeneous reactions take place are polar stratospheric clouds (PSCs). These are described in Chapters 11. PSCs are responsible for the Antarctic ozone hole in that they provide the surfaces on which the ozone destroying reactions occur. PSCs are divided into two classes, Type I and Type II. Both form only under extremely cold conditions that are found only occasionally in the winter polar vortex region of the Arctic and Antarctic stratosphere. They tend to form much more frequently in the more isolated, colder Antarctic vortex, hence the existence of the ozone hole over there and not over the Arctic. The temperature at which they form is referred to as the frostpoint. Type II PSCs are better understood than Type I PSCs.

Type II PSCs are water ice particles that form when the temperature falls below 188 K (-85°C). They are relatively large, with a diameter at least 10 microns. Type II PSCs fall rather rapidly at about 1.5 kilometers per day. This is referred to as sedimentation, the settling out of small particles from the atmosphere.

The composition of Type I PSCs are still not well understood. Indeed, their structure and composition have been the topic of extensive research over the last few years. It is believed that Type I PSCs are composed of a supercooled liquid ternary solution of nitric acid, sulfuric acid, and water ice (HNO3-H2SO4-H2O), as well as frozen nitric acid trihydrates (NAT). The frostpoint for Type I PSCs is 195 K (-78°C). Type I PSCs are much smaller than Type II PSCs, with particle diameters on the order of 1 micron. Their sedimentation rate is consequently much slower, on the order of 10 meters (0.01 km) per day. See Chapter 11 for additional discussion of PSCs.

Many questions remain about the microphysics and chemistry of PSCs. The microphysics questions involve determining the stable forms of PSCs under stratospheric conditions, the solubility of various gases on PSC surfaces, and how PSCs undergo phase changes over time. The chemistry questions relate to reaction rates and sensitivity to variables such as temperature, surface area, and PSC composition. Because of the difficulty of sampling PSCs, all of these areas are currently of tremendous research interest.

5.3 Heterogeneous Reactions

Heterogeneous reactions are extremely important, since they free chlorine and bromine from reservoir species into reactive forms. The importance of these reactions results from the freeing of chlorine from relatively benign chlorine forms into highly reactive forms, and the removal of reactive nitrogen species (NOx) into more stable forms such as nitric acid (HNO3).

There are 5 basic heterogeneous reactions necessary to understand stratospheric chemistry. The five reactions are given below.

Rates for these reactions depend on a number of factors, including particle type (via the sticking coefficient), particle surface area, and temperature. Reaction 4 does not have a temperature dependent sticking coefficient, and is therefore important wherever and whenever we find particles in the stratosphere (e.g., after volcanic eruptions). In fact, the nitrogen chemistry of the middle latitude stratosphere cannot be explained without including reaction 4. Because the other reaction rates increase dramatically in cold temperatures, they are extremely important to an understanding of the ozone budget of the polar stratosphere.

Molecules such as HCl (hydrochloric acid) and ClONO2 (chlorine nitrate), as shown in reaction 1, are mostly nonreactive in their gaseous state, which is why they act as chlorine reservoir species. However, when these molecules are dissolved in liquids such as sulfuric acid-water solutions, they become highly reactive with one another. The absorption of HCl and ClONO2 on a liquid sulfate aerosol particle in the stratosphere results in the release of Cl2 (molecular chlorine) into a gaseous form (desorption) and the retention of a nitric acid (HNO3) molecule with the aerosol particle.

The nonreactive HNO3 formed remains in a solid (frozen) state on the surfaces of the PSCs. As the PSCs undergo sedimentation, the HNO3 is carried out of the stratosphere. This process leads to the removal of nitrogen from the stratosphere in a process called denoxification (see Chapter 11). Since HNO3 photolysis results in the formation of reactive NO2, which in turn reacts with ClO to form the reservoir ClONO2 species, the removal of nitrogen from the stratosphere means that there is more reactive, ozone destroying ClO.

The chlorine species Cl2 and HOCl created in reactions 1, 2, and 5 are short lived. They are quickly photolyzed by sunlight even in visible wavelengths. The Cl2 molecules are photolyzed into two chlorine atoms, which then participate in the ClX catalytic cycles outlined above in sections 4.2.8 and 4.2.9. The HOCl photolysis liberates ClO, another ClX species that destroys ozone. The rate of ClO destruction of ozone is derived in Section 5.3.1.

5.3.1 Destruction of ozone via the ClO-ClO reaction -- We can calculate an expression for the amount of ozone loss by ClO (chlorine monoxide) by making two assumptions. First, let's assume that NOX is not involved in removing ozone from the stratosphere. Second, let's assume that the stratosphere is cold enough to form PSCs, which leads to the conversion of nonreactive chlorine into the reactive form ClO. The ClO dimer, Cl2O2, is created in the termolecular reaction already presented in Section 4.2.9.

Based on these two assumptions, we can write that the production of Cl2O2 by the ClO-ClO termolecular reaction is exactly balanced by the photolytic destruction of Cl2O2. This gives us

The photolysis rate constant is denoted by J, while the ClO-ClO reaction rate constant is k, and [M] is the density of the other molecule (M) involved.

The direct loss of ozone is via the bimolecular reaction Cl+O3 --> ClO+O2. Using this reaction, we can balance the production and loss of ClO via

Here, k2 is the Cl+O3 reaction rate constant. The most dominate Clx species are ClO and Cl2O2, hence we write

In this expression, we used equation A to estimate Cl2O2. We can now solve equation C and substitute into equation B to calculate the ozone loss rate in terms of Clx. Solving the equations, we find that

For small values of [Clx], the concentration of [ClO] is about equal to [Clx]. Using this formula together with the righthand side of equation B, we can calculate the ozone loss rate in the ozone hole using estimates of Clx. Figure 5.28 displays the ozone loss time scale calculated for conditions at 70°S at an altitude of 20 km, corresponding to a pressure level of 50 mb or 50 hPa. Preindustrial values of Clx were smaller than 3 ppbv (3 reactive chlorine molecules per billion molecules of air). Solving for these preindustrial conditions, we find that the time scale for ozone loss was greater than an entire season! Hence, the chlorine activated by PSCs did not cause much ozone loss. Current Clx stratospheric values are greater than 3 ppbv, implying an ozone loss time scale of about 2.5 weeks. This is why the Antarctic ozone hole now develops seasonally, but did not do so in the past.

The assumptions invoked here are reasonable for Antarctic conditions. In the Arctic spring however, temperatures are not cold and persistent enough to keep all of the chlorine in the form of Clx via heterogeneous processes on PSCs. In particular, once ClO is formed from HCl and ClONO2 via a heterogeneous reaction, the ambient NO2 will react to reform ClONO2. This is because denoxification of the Arctic stratosphere does not occur. The paucity of PSCs means that HNO3 is not carried out of the stratosphere via PSC sedimentation, and when sunlight returns, HNO3 is photolyzed by UV light to recreate NOx species. Hence, it cannot be assumed that all of the chlorine is in the reactive Clx form.

5.3.2 Chlorine and nitrogen activation/deactivation -- The heterogeneous reactions (particularly reaction 1) are crucial to ozone loss for two reasons. First, the chlorine is liberated from the HCl and ClONO2 reservoir species (enabling the catalytic reactions), and second, the NO2 is locked up as nonreactive HNO3 and cannot deactivate chlorine. Hence, the chlorine is free to destroy ozone at an extremely efficient rate via the Clx catalytic cycles.

Figure 5.29 illustrates the added pathway for activating chlorine because of the heterogeneous chemistry. This figure is a modification of Figure 5.22. In Figure 5.22, ClONO2 is converted to Cl by slow photolysis, while Figure 5.29 has an additional pathway for converting ClONO2 to Cl by the heterogeneous reaction with HCl on the PSC surface.

Figure 5.30 illustrates how reactive nitrogen is locked away (or deactivated) as HNO3, and therefore is unavailable to deactivate ClO. This figure is a modification of Figure 5.17. In Figure 5.17, ClONO2 is converted to NO3 by slow photolysis, while Figure 5.30 has an additional pathway for converting ClONO2 to HNO3 by the heterogeneous reaction with HCl on the PSC particle (top panel). This pathway exists as long as temperatures are cold enough to form stratospheric particles. The NOx is locked away and cannot deactivate the ClO, and ClO remain at high levels. As soon as the Sun rises, the HNO3 can be photolyzed, producing NO2, which will then react with ClO to reform ClONO2 (bottom panel). Without further heterogeneous reactions, the chlorine in the ClONO2 remains locked away.