Info

u

FIGURE 9.32 Some evidence from field studies along the coast of Washington State on April 22, 1991, both for new particle formation and for condensation on preexisting nuclei: (a) particle surface area for particles with D > 0.02 ¡j.m; (b) ratio of number of particles with D 0.015 ¡xm to those with D > 0.015 jum; (c) number concentration with diameters 0.02-0.024 /¿m (adapted from Covert et at., 1992).

FIGURE 9.32 Some evidence from field studies along the coast of Washington State on April 22, 1991, both for new particle formation and for condensation on preexisting nuclei: (a) particle surface area for particles with D > 0.02 ¡j.m; (b) ratio of number of particles with D 0.015 ¡xm to those with D > 0.015 jum; (c) number concentration with diameters 0.02-0.024 /¿m (adapted from Covert et at., 1992).

number of fine particles was observed, with bursts of nucleation occurring periodically (Flagan et al., 1991). This was attributed to the heterogeneous condensation on the seed particles initially, but at too slow a rate to remove the low-volatility products. As the latter accumulate, homogeneous condensation occurred, forming a burst of new particles. Condensation on these new particle surfaces then occurs until they coagulate with the seed particles and the process begins anew.

For detailed discussions of the quantitative treatment of such condensation processes in the atmosphere, the reader is referred to articles by Pandis et al. (1995) and Kerminen and Wexler (1995).

Coagulation refers to the formation of a single particle via collision and sticking of two smaller particles. Small particles undergo relatively rapid Brownian motion, which leads to sufficient particle-particle collisions to cause such coagulation. The coagulation of smaller particles with much larger ones is similar to condensation of a gas on the larger particle and acts primarily to reduce the number of small particles, adding relatively little to the mass or size of the larger particles. Hence the larger mode will not show significant growth by such a mechanism. As expected, the rate of such processes depends on the diameter of the large particle, how rapidly the smaller particle is carried to it (i.e., the diffusion of the smaller particle), and the concentrations of the particles.

So-called self-coagulation, where the particles are approximately the same size, can, however, lead to changes in the size distribution of the aerosol particles. As one might expect, the rate of this process is a strong function of the particle concentration as well as the particle size.

Table 9.10 shows an estimate (Pandis et al., 1995) of the time scales for coagulation of smaller particles onto larger ones characteristic of various types of air masses. For comparison, typical time scales for condensation, dry deposition of the particles, and transport are also shown. (For discussions of dry deposition of particles, see, for example, Slinn (1982, 1993), Arimoto et al. (1987), and Main and Friedlander (1990)). As expected, condensation is fast, but coagulation is also significant on these time scales in some situations.

2. Reactions of Gases at Particle Surfaces

There are some well-known examples of reactions of gases with solids at the interface that are potentially important in the atmosphere. For example, the reactions of 03 with adsorbed solid polycyclic aromatic hydrocarbons (PAH) are discussed in Chapter 10. Another example is the reaction of NaCl and NaBr in sea salt particles with gaseous oxides of nitrogen such as HN03 (see Chapter 6.J):

The replacement of chloride in sea salt particles by nitrate as well as by sulfate has been observed in many measurements in coastal areas using bulk filter samples (e.g., see Section C.l and Chapter 6.J.2b) and, more recently, in single particles (e.g., Murphy et al., 1997; Gard et al., 1998). Even in these relatively simple cases, however, the nature of the reactions at the interface and how they contribute to particle growth and or transformations is not clear. For example, in the reaction of NaCl with HN03 and other oxides of nitrogen such as N02, the initial reaction forms one or more unique surface nitrate species which do not alter the particle morphology significantly (Vogt and Finlayson-Pitts, 1994; Vogt et al., 1996; Allen et al., 1996). However, if exposed to water vapor even well below the deliquescence point and then dried, this surface nitrate reorganizes into small microcrystallites of NaN03 attached to the original salt surface (see Fig. 11.63 in Chapter 11). This generates very small particles of NaN03 and may be responsible for the observation of small particles in the marine boundary layer which are almost completely devoid of chloride (e.g., Mouri and Okada, 1993).

A further complication is the recent indication that even small amounts of strongly bound surface-adsorbed water may play a critical, indeed determining, role in the interaction of gases with surfaces traditionally thought to be solids. For example, in the NaCl-HN03 reaction, there is evidence that the reaction even in laboratory vacuum systems occurs on sites holding adsorbed water. As a result, the surface does not become saturated as one would expect for a solid surface, since the underlying reactant salt continues to dissolve in the surface water (Beichert and Finlayson-Pitts, 1996).

Another example of reactions at interfaces that is only now being recognized, due to the lack of suitable experimental techniques in the past, is that of species such as S02 and NOz at liquid interfaces. As discussed in Chapters 7 and 8, there is increasing evidence that the reactions of such species at the air-water interface can be fast relative to that in the bulk and may have unique reaction mechanisms compared to those in the bulk or gas phases. Given the paucity of data on such processes at the present time, they are generally not included in present models of aerosol growth. How

TABLE 9.10 Typical Time Scales for Various Aerosol Fates"

Type of air mass

TABLE 9.10 Typical Time Scales for Various Aerosol Fates"

Type of air mass

Fate

Urban

Remote marine

Free troposphere

Nonurban continental

Condensation

0.01-1 h

1-10 h

2-20 h

0.5-20 h

Coagulation of 0.03-jU.m-particles with larger particles

0.1-2 days

10-30 days

~50 days

1-5 days

Deposition 0.03-/im particles 0.3-/^,m particles

0.5-10 days ~1 month

0.5-10 days ~1 month

~1 month ~1 month

Transport

2-5 days

1-2 weeks

3 days to 2 weeks

1-2 weeks

ever, as our understanding of such reactions at interfaces expands, their implications for the growth of aerosols in the troposphere will need to be critically assessed.

Was this article helpful?

0 0

Post a comment