Index of refraction at 532 nm

1.39 + 0.03

1.42 + 0.04

1.32 + 0.01

1.43 + 0.04

Gobbi et al., 1998). Although the relevant optical properties in the infrared have been investigated in laboratory studies (e.g., Toon et al., 1994; Richwine et al., 1995), they have not been probed for atmospheric particles. However, such studies clearly also have the potential to shed light on the composition of PSCs.

Knowing whether the PSCs are solid or liquid solutions is important because, as we shall see in the following section, the reaction probabilities for various reactions on PSCs can differ, depending on the nature of the particle. As a result, chlorine activation and ozone destruction are sensitive to this as well. For example, N205 hydrolysis is much faster in a liquid ternary solution than on NAT (see Table 12.5), and the C10N02 + H20 and HOC1 + HC1 reactions are also faster. Chlorine activation is therefore faster on ternary solution PSCs (Ravishankara and Hanson, 1996; Borrmann et al., 1997b; Del Negro et al., 1997).

For a review of PSCs, see the article by Peter (1997).

b. Uptake of HCl into PSCs

Given the preceding discussion of the nature of PSCs and stratospheric aerosols, it is clear that the uptake and subsequent reactions of HCl with C10N02, HOC1, and N205 both on solid surfaces and into liquid solutions consisting of various combinations of HN03, HzO, and H2S04 must be considered. We first discuss the uptake of HCl onto ice surfaces, which is relevant to Type II PSCs, and then uptake into solutions that are thought to be representative of aerosols and Type I PSCs.

For the heterogeneous reactions of HCl on PSCs and aerosols to be important, there must be mechanisms to continuously provide HCl to the surface. This could occur, for example, if HCl is sufficiently soluble in ice and if it diffuses at a sufficient rate from the bulk to the surface. However, the solubility and diffusion rates have been shown to be sufficiently small that these processes are not expected to be important under stratospheric conditions (see Wolff and Mulvaney, 1991; Dominé et al., 1994; and Thibert and Dominé, 1997).

However, HCl has been shown in a number of studies to be taken up by ice and NAT surfaces, with the amount depending on a number of factors including temperature and the partial pressure of HCl in the gas phase (e.g., see summary in DeMore et al., 1997). The amount of HCl that can be taken up has been shown to correspond to a significant fraction of a monolayer. While the formation of hydrates such as HCl • 6H20 has been observed in laboratory systems (e.g., see Koehler et al., 1993; Chu et al., 1993; Graham and Roberts, 1994, 1995; and Banham et al., 1996), consideration of the phase equilibria under stratospheric conditions suggests that these will not be important at the low HCl partial pressures and higher temperatures of the stratosphere (Wooldridge et al., 1995).

One of the interesting chemical aspects of the heterogeneous chemistry of HCl is why its reactions on ice surfaces are so much more efficient than in the gas phase. A compelling explanation is that HCl ionizes on the solid surfaces, so that the reaction does not involve covalently bound HCl, but rather, the chloride ion. This is consistent with the fact that chloride ions react very rapidly in the gas phase with the relevant species such as C10N02 (Haas et al., 1994) and with the observation that chloride ions from NaCl undergo analogous reactions at room temperature with C10N02 and N205 (Finlayson-Pitts et al., 1989; Livingston and Finlayson-Pitts, 1991; Finlayson-Pitts, 1993).

There is infrared evidence for the ionization of HCl on ice (Horn et al., 1992; Delzeit et al., 1993; Banham et al., 1996; Koch et al., 1997) and molecular dynamics simulations also support this view (Robertson and Clary, 1995; Gertner and Hynes, 1996). In the simulations, HCl becomes incorporated into the ice via hydrogen bonding between the chlorine of HCl and a hydrogen of a surface water or between the hydrogen of HCl and the oxygen of a surface water as depicted in Fig. f 2.24. George and co-workers (Haynes et al., 1992) have shown that under stratospheric conditions, the ice surface is very dynamic, with continuous, rapid evaporation of water molecules from the surface and recondensation. At temperatures of 180-210 K, the rate of water condensation and evaporation corresponds to 10—103 monolayers per second. Thus as HCl is taken

HCl hydrogen-bonded to proton Condensation acceptor water °f water n o HCl

^^cPopp molecules ionization

6 Ice surface

FIGURE 12.24 Schematic of the incorporation of HCl from the gas phase onto the surface of ice via hydrogen bonding, followed by condensation of water and ionization of the HCl (adapted from Gertner and Hynes, 1996).

up at the surface and ionizes, it can also be, in effect, "buried" as surface water molecules evaporate and recondense on top of it (Fig. 12.24).

It is noteworthy that there is some laboratory evidence that HBr, in contrast to HC1, may form a hydrate, HBr • 3H20, under polar stratospheric cloud formation conditions (Chu and Heron, 1995).

Sodeau and co-workers (Sodeau et al., 1995; Koch et al., 1997) have infrared evidence that chlorine nitrate also ionizes on ice at 180 K, forming an intermediate identified as [H2OCl]+ through the initial solvation. Hence heterogeneous reactions on ice may be rapid not only because of the ionization of HC1 but also because of the ionization or partial ionization of C10N02 (Horn et al., 1998). A similar mechanism has been proposed for N2Os hydrolysis on surfaces (Koch et al., 1997). It should be noted, however, as discussed shortly, that Bianco and Hynes (1998) propose, based on ab initio calculations, that the intermediate observed is not [H2OCl]+ but rather solvated HN03.

Molina and co-workers have proposed that the surface layer can be thought of as a "quasi-liquid layer" with significant mobility of the species, particularly in the presence of higher partial pressures of HC1 (Abbatt et al., 1992). Thus the uptake of HC1 can be treated as uptake and solvation in this quasi-liquid layer. The nature of this surface is not well understood, however. Although the existence of a quasi-liquid layer on ice surfaces near the freezing point has been recognized for more than a century, the nature of the ice surface under various conditions even in the absence of other species such as HC1 continues to be the subject of debate (e.g., see Hobbs, 1973; Conklin and Bales, 1993; Knight, 1996a, f996b; Baker and Dash, 1996; Prup-pacher and Klett, 1997; and papers in "Physics and Chemistry of fee," Petrenko et al., 1997).

There is again an analogy to NaCl surfaces at room temperature. Thus when solid NaCl having even small amounts of surface nitrate (formed by reaction with HN03 or N02) is exposed to low pressures of gaseous water, well below the deliquescence points of bulk NaCl and NaN03, a very mobile surface layer is formed; when the water is pumped off, the ions in this mobile liquid layer selectively recrystallize into separate micro-crystallites of NaN03 and NaCl (Vogt and Finlayson-Pitts, f994; Vogt et al., 1996; Allen et al., 1996; Laux et al., 1996).

Because Type t PSCs may consist of NAT under some conditions, uptake of HC1 onto crystalline NAT as well as ice surfaces is of interest. As reviewed by DeMore et al. (1997), the mass accommodation coefficient for HC1 on both ice and NAT at stratospheric temperatures is very large, approaching unity.

HC1 is efficiently absorbed into H2S04-H20 and into HN03-H2S04-H20 solutions, which as discussed earlier, are found in the stratosphere in the form of aerosol particles and Type I PSCs under some conditions (Wolff and Mulvaney, 1991). The solubility of HC1 in these liquid solutions can be expressed in terms of the usual Henry's law constant (Elrod et al., 1995; Abbatt, 1995; Luo et al., 1995; Hanson, 1998). Table 12.4 shows some typical measurements of the Henry's law constants for HC1 in several typical binary and ternary solutions, respectively. Hanson (1998) has shown that the solubility data for HC1 in binary mixtures of H2S04 and water in these and other studies can be fit by the form

^nci = [eo + e\x + e2x2]exp[c0 + (d{) + dix)/T], where x is the mole fraction of H2S04, d() = +6922, d{ = -9800, and the fit parameters c(), e(), e1; and e2 are given by c() = -9.021, e() = +0.363, e, = -2.616, and e2 = +4.995. The Henry's law constants in sulfuric acid-water solutions increase as the temperature decreases and as the dilution of the solution increases. This increase in HC1 solubility as the temperature falls

TABLE 12.4 Some Measured Values of Henry's Law

Constant for HC1 in H2S04 - HzO or H2S04 - HN03 - HzO Solutions at Stratospherically Relevant Temperatures"

TABLE 12.4 Some Measured Values of Henry's Law

Constant for HC1 in H2S04 - HzO or H2S04 - HN03 - HzO Solutions at Stratospherically Relevant Temperatures"

Was this article helpful?

0 0

Post a comment