"Heterogeneous" reactions in the context of atmospheric chemistry are defined as those between gases and either solids or liquids. Heterogeneous reactions have been suggested for decades as being important in the atmosphere. However, historically, forays into this area by a variety of scientists usually concluded in a rapid retreat, due to highly variable results that were (and, unfortunately, continue to be) common. The recognition of the importance of condensed phases in the oxidation of S02 to sulfuric acid (see Chapter 8.C.3) and the discovery of the dramatic Antarctic ozone "hole" (see Chapter I2.C.4) reinforced the key role of heterogeneous reactions, and the atmospheric chemistry community has tackled this area anew. As a result, while we are still in the "dark ages" in terms of understanding kinetics and mechanisms of these processes on a molecular level, a great deal of progress has been made in the past decade.
There are many different types of surfaces available for reactions in the atmosphere. In the stratosphere, these include ice crystals, some containing nitric acid, liquid sulfuric acid-water mixtures, and ternary solutions of nitric and sulfuric acids and water. In the troposphere, liquid particles containing sulfate, nitrate, organics, trace metals, and carbon are common. Sea salt particles dominate in marine areas. In addition, there are large episodic sources of particles emitted directly into both the troposphere and stratosphere, such as rocket exhausts where particles containing carbon soot, alumina, and metal oxides can be emitted in large quantities.
Before we describe some of the common techniques used to study the kinetics and mechanisms of heterogeneous reactions, a few words regarding the difficulties in this area are appropriate. To put these in perspective, consider first the current state of understanding of gas-phase kinetics. There are a number of both absolute and relative rate techniques available for studying gas-phase reactions, and the methodologies for preparing reactants and measuring products are generally quite well developed. As a result, agreement to within ~ 15% on gas phase reaction rate constants measured using different techniques and by different laboratories is now common. In addition, understanding of the reactions at a molecular level is generally good.
In contrast, in most heterogeneous reactions, we really do not even understand what one of the reactants, the surface, looks like on a molecular level; i.e., the condensed-phase molecule and its environment that the incoming gaseous reactant encounters is not well characterized. An example of our incomplete understanding of the nature of surfaces is controversy about effective surface areas for ice and whether ice surfaces prepared in the laboratory for example, are porous or not (e.g., see Keyser et al., 1993; and Hanson and Ravishankara, 1993a).
In addition, how the surface changes during reaction and how such changes affect both the reactivity and mechanism are unclear. One well-recognized phenomenon is that of saturation of the surface, i.e., complete reaction of the surface species so that the reaction comes to a halt. However, there are other potentially important aspects as well. For example, it is well known in surface science that adsorbates can lead to restructuring of the surface (Somorjai, 1994), yet this phenomenon has yet to be demonstrated on surfaces of atmospheric interest. This is in part due to the incompatibility between atmospheric conditions (air at pressures up to 1 atm and containing significant amounts of water vapor) and the ultrahigh-vacuum conditions typical of surface science studies. However, marriage of these two fields will be ultimately needed for a complete understanding of heterogeneous processes in the atmosphere, and progress is being made with some systems (e.g., Hemminger, 1999).
Different terms and symbols have been used in the literature to differentiate reversible, physical uptake from irreversible uptake via chemical reactions. Addi tional confusion arises from the fact that the observed uptake of a species from the gas phase is usually a net uptake affected by a number of factors such as changes in the surface during the uptake (e.g., due to saturation) and reevaporation into the gas phase due to limited solubility of the species. However, the most common terminology now in use is the following:
Surface reaction probability (yrxn) is the net fraction of gas-condensed phase collisions that leads to the irreversible uptake of the gas due to chemical reaction. The symbol -yrxn (or sometimes 0) is most commonly used for reaction probabilities.
Mass accommodation coefficient (a) is the fraction of gas-condensed phase collisions that result in uptake of the gas by the condensed phase:
Number of gas molecules taken up by the surface a = -.
Number of gas-surface collisions
This is the probability that a molecule that strikes the surface will cross the interface into the condensed phase. It does not represent the net uptake; i.e., it does not include the reverse effect of reevaporation from the condensed phase into the gas phase, a is also sometimes referred to as a sticking coefficient for uptake on solid surfaces.
Net collisional uptake probability (yncl) is the net rate of uptake of the gas normalized to the rate of gas-surface collisions and since this is what is measured, is also often referred to as ymcas
The net loss of the gaseous species in the presence of a condensed phase having known volume and surface area is what is typically measured in laboratory studies. To obtain information on the fundamental processes contributing to this net measured uptake, all of the processes shown schematically in Fig. 5.12 must be taken into account: diffusion of the gas to the surface, uptake, diffusion in the liquid phase, and reaction either in the bulk or at the interface itself.
We therefore first briefly discuss the analysis of systems that involve diffusion in the gas and liquid phases, uptake, and reaction in the bulk liquid or at the interface. Following that, we give a brief description of some of the most common methods used to measure mass accommodation coefficients and reaction kinetics for heterogeneous atmospheric reactions. Included are some new approaches that appear to be especially promising. For a review of this area, see Kolb et al. (1995, 1997).
1. Analysis of Systems with Gas- and Liquid-Phase Diffusion, Mass Accommodation, and Reactions in the Liquid Phase or at the Interface
Both for laboratory measurements and for atmospheric processes, the uptake of a gas into a liquid followed by reaction involves a number of different physical (e.g., diffusion and uptake at the interface) as well as chemical processes. These were depicted in Fig. 5.12. We treat here in more detail the individual steps and how the net uptake of a gas into solution is determined by these steps.
1. Transport of the gas to the surface and the initial interaction. The first step in heterogeneous reactions involving the uptake and reaction of gases into the liquid phase is diffusion of the gas to the interface. At the interface, the gas molecule either bounces off or is taken up at the surface. These steps involve, then, gaseous diffusion, which is determined by the gas-phase diffusion coefficient (Dg) and the gas-surface collision frequency given by kinetic molecular theory.
2. Uptake at the interface, if the gas molecule is taken up at the surface, it enters the interface region and then the bulk. The efficiency of uptake involving crossing the interface is described by the mass accommodation coefficient (a) defined earlier. Molecular-level mechanisms by which gas molecules are taken up into liquids are discussed elsewhere (e.g., see Davidovits et al., 1995; Taylor et al., 1996, 1997; and Nathanson et al., 1996).
3. Diffusion into the bulk. This is determined by the diffusion coefficient in the liquid (D,). Diffusion within the bulk aqueous phase is much slower than gas-phase diffusion and can be rate-limiting under conditions of high reactant concentrations where the rate of the chemical reaction is high. This appears to have been a problem in some experimental studies of some aqueous-phase reactions relevant to the atmosphere where either bulk solutions or large droplets and reactant concentrations higher than atmospheric were used (Freiberg and Schwartz, 1981).
4. Henry's law equilibrium, if there is no reaction in the liquid phase (or it is slow relative to uptake and diffusion), the gas-liquid system eventually comes to equilibrium, which can usually be described by Henry's law discussed earlier. This does not reflect a lack of uptake of the gas at equilibrium but rather equal rates of uptake and evaporation; i.e., it is a dynamic equilibrium (see Problem 12). The equilibrium between the gas-and liquid-phase concentrations is characterized by the Henry's law constant, H (mol L_1 atm~'), where H = [X]/Px.
5. Reaction in the bulk. Reaction can occur in solution close to the surface or throughout the bulk of the liquid phase, depending on the speed of the reaction compared to diffusion. We shall see that whether the reaction occurs close to the surface or throughout the bulk has important implications for the kinetics, since in the former case, the reaction depends on the particle surface area, whereas in the latter it depends on the particle volume.
We shall treat here reactions occurring in the condensed phase as if they are first-order, irreversible reactions with a rate constant k. Of course, this also applies to second-order reactions with the second reactant, B, in great excess, since they are then pseudofirst-order with k' = /c[B].
6. Reactions at the interface. There has been increasing recognition that reactions may also occur at the interface itself. That is, species such as S02, NH3, and organics do not simply cross the interface by physical transport but rather form unique chemical species at the interface (e.g., Donaldson et al., 1995; Allen et al., 1999; Donaldson, 1999; Donaldson and Anderson, 1999). These unique interface species can then react at the surface without actually being taken up into the bulk of the solution. Although relatively little is currently known at a molecular level about such processes, reactions in this "fourth phase" may prove to be very important in atmospheric processes, for example in the generation of HONO in the N02 reaction with water at surfaces (see Chapter 7.B.3b).
The measured, net uptake of a gas into a liquid phase can be related to these various processes, i.e., to Dg, a, D], H, and k. However, concentrations in both the gas and liquid phases as well as the volumes and surface areas available for reaction are quite different in laboratory studies compared to the atmospheric situation. We shall therefore also examine how net gas uptake measured in laboratory studies can be related to uptake under atmospheric conditions.
The physical and chemical processes occurring in a gas-liquid system are often treated in terms of a resistance model described in Box 5.2. As discussed there, the net uptake of gas (ynet) can be treated under some conditions in terms of "conductances," T, normalized to the rate of gas-surface collisions. Individual conductances are associated with gas-phase diffusion to the surface (Tg), mass accommodation across the interface (a), solubility (rsoi), and finally, reaction in the bulk aqueous phase (rrxn). This leads to Eq. (QQ):
Was this article helpful?