FIGURE 14.39 CCN concentrations measured during a cloud event at Palmer Station, Antarctica, as a function of percent supersaturation (%S). The two lines represent particles that grew to sizes greater than 0.3 and 0.5 ¡xm, respectively (from Saxena, 1996).

was carried out during a burst of CCN production due to cloud processing (vide infra).

The relationship between the number concentration (AO of CCN and the supersaturation (S) is often expressed in the form N = CSk, where C and k are empirical coefficients characteristic of the particular air mass (Pruppacher and Klett, 1997). However, alternate forms such as

where B and k are empirical coefficients and N{) is the number concentration of CCN at infinite supersaturation, have been suggested to match data from laboratory studies where the aerosol composition is relatively simple (Ji and Shaw, 1998).

As discussed earlier, if organics congregate at the air-water interface of particles and act as surfactants, they can lead to a reduced Kelvin effect and hence activation at lower supersaturations. In addition, if they dissolve, they will also contribute to Raoult's law effects of the dissolved species. As discussed by Shulman et al. (f 996), the dissolution of organics in particles may lead to modifed Köhler curves having two maxima instead of one. If the second maximum is at a higher supersaturation than the first, droplets could become partially activated. They also suggest that with the distribution of chemical compositions and particle sizes in the atmosphere, the region beyond the maxima could be relatively flat in shape, rather than having a steep negative slope. This would result in the formation of cloud droplets with radii corresponding to some characteristic metastable size.

The effects of surfactants on particles is discussed in detail in Chapter 9.C.2b. In some cases, the interaction of organic surfactants with particles has been observed to enhance water uptake and hence the cloud nucleating properties of the particle, whereas in others, it inhibits water uptake. An example of the former case is a study by Cruz and Pandis (1998), who coated particles of (NH4)2S04 with the C5 dicarboxylic acid, glutaric acid. A coating of glutaric acid on ammonium sulfate increased the size of the particle and increased its cloud nucleating properties, but in a manner consistent with that expected from Köhler curves for a two-component solution assuming the organic does not alter the surface tension. On the other hand, coating with the water-insoluble dioctyl phthalate did not alter the activation of the inorganic salt. Similarly, Kotzick et al. (1997) and Weingartner et al. (1997) reported that oxidation of carbon or diesel soot particles by 03 increased their CCN activation properties due to the formation of polar surface groups.

As discussed in detail in Chapter 9, Saxena et al. (1995) have measured the hygroscopic behavior of par ticles in the Los Angeles area (i.e., urban) and in the Grand Canyon, Arizona (i.e., nonurban). They found that organics in the urban particles inhibit water uptake, i.e., are hydrophobic in nature, whereas those in the nonurban particles are hydrophilic, i.e., increase water uptake. It may be that these differences are due to the formation of smaller, more oxidized, water-soluble organics during long-range transport to the nonurban site. Saxena et al. (1995) suggest that hydrophobic organics in the urban aerosol may form a surfactant film on the particles that inhibits water uptake. For example, difunctional acids such as oxalic and adipic acids have been shown to slow the rate of evaporation of water from droplets once they are sufficiently concentrated by the initial evaporation of water (Shulman et al., 1997). At any rate, as might be expected, the effects of organics on water uptake for real atmospheric particles clearly can be negative or positive, depending on their particular composition.

As expected from the earlier discussion of the Köhler curves, not all particles act as CCN. For example, only about 15-20% of the Aiken nuclei (see Chapter 9.A.2) in a marine air mass off the coast of Washington state acted as CCN at f% supersaturation (Hegg et al., 1991b). Similarly, in a marine air mass in Puerto Rico, between 24 and 70% of the particles measured at 0.5% supersaturation before cloud formation led to cloud droplet formation (Novakov et al., 1994).

Gillani, Leaitch, and co-workers (1995) carried out a detailed study of the fraction of accumulation mode particles (diameters from 0.f7 to 2.07 /¿m) that led to cloud droplet formation in continental stratiform clouds near Syracuse, New York. When the air mass was relatively clean, essentially all of the particles were activated to form cloud droplets in the cloud interior and the number concentration of cloud droplets increased linearly with the particle concentration. However, when the air mass was more polluted, the fraction of particles that were activated in the cloud interior was significantly smaller than one. This is illustrated by Fig. 14.40, which shows the variation of this fraction (F) as a function of the total particle concentration, NUtV In the most polluted air masses (as measured by large values of ATl()t), the fraction of particles activated was 0.28 + 0.08, whereas in the least polluted, it was as high as 0.96 + 0.05. The reason for this is likely that in the more polluted air masses, the higher number of particles provided a larger sink for water vapor, decreasing the extent of supersaturation.

In short, while anthropogenically produced particles can act as cloud condensation nuclei, only a fraction of them actually do so. This fraction can be close to one

Was this article helpful?

0 0

Post a comment