Aerosol volume

Table II. The measured decay of the aerosol volume ( vol) is fitted...

The particle light scattering coefficient has been continuously measured at this location since 1976. Measurements of the particle size distribution have been made daily since 1978, providing the data base necessary to assess the variability of the normalized aerosol volume distribution. 

It is seen that the distribution is bimodal, with the coarse mode dominating the aerosol volume concentrations. The 1979 average volume concentration of aerosol less than 10 ym diameter was 32.4 ymVcm. From its large standard deviation, it is clear that the coarse particle mode exhibited considerable variation throughout the year. Records show that high coarse mode volume concentrations accompanied moderate-to-high wind speeds. The coarse material was very likely wind-blown dust of crustal composition. 

The aerosol scattering coefficient distribution was calculated from the aerosol volume distribution, using the method described in Friedlander (] ] ). The resultant distribution is plotted in Figure 3. The contribution of the fine aerosol to visibility degradation at China Lake is seen in this figure. 

Figure 2. Normalized aerosol volume distribution, China Lake, CA (1979 average)—average of 254 measurements. The error bars are standard deviations. The distribution is normalized with respect to total aerosol volume concentration of particles less than 10 lOn in diameter.

Because the fine aerosol was found to be responsible for the bulk of light scattering at China Lake, this mode was examined to see if its normalized distribution remained constant throughout 1979. Figure 4 shows the 1979 average aerosol volume distribution at China Lake normalized with respect to the total volume of particles smaller than 2 ym. The error bars represent standard deviations in the 254 measurements. The particle volume distribution in the fine mode is seen to preserve its shape rather well. Over half the fine particle volume is due to particles of less than 0.3 ym diameter. 

On the average, the requirements for application of the statistical technique to filter data were met. Analysis of the 254 measured particle size distributions in 1979 indicates that the fine aerosol volume distribution preserved its shape. The measured sulfur mass distribution followed that of the total submicron volume. By difference, it was assumed that the organics did the same. The low relative humidity at China Lake minimized the formation of aqueous solutions due to water condensation on the particles. Therefore, it is expected that the statistical technique can be used with some success with the China Lake filter data. 

It was found that the requirements were satisfied for application of the linear regression technique to species mass concentrations in a multicomponent aerosol. The results of 254 particle size distributions measured at China Lake in 1979 indicate that the normalized fine aerosol volume distribution remained approximately constant. The agreement between the calculated and measrued fine particle scattering coefficients was excellent. The measured aerosol sulfur mass distribution usually followed the total distribution for particles less than 1 ym. It was assumed that organic aerosol also followed the total submicron distribution. 

This is supported by the correlation between the total aerosol volume of particles with diameters in the 0.1- to -fjbva range and the experimentally determined values of hsp obtained using a nephelometer in many studies (e.g., Fig. 9.24). The slopes of lines such as that in Fig. 9.24, however, depend critically on the nature and history of the air mass and can vary by more than a factor of 10 from clean, nonurban air to highly polluted air in the vicinity of sources. For example, Sverdrup and Whitby (1980a) have shown that the ratio of submicron aerosol volume to hsp, which corresponds to the slope of the line in Fig. 9.24, varies from 5 to 80, depending on the nature of the air mass. In addition, the correlation between bsp and fine particles is usually not as clear-cut as seen in Fig. 9.24. 

Sverdrup, G. M., and K. T. Whitby, The Variation of the Aerosol Volume to Light-Scattering Coefficient, Adr. Environ. Sci. Technoi., 10, 539-558 (1980b). 

Figure 9 shows LUT-retrieved seasonal averages of aerosol volume density V as a function of altitude and latitude from winter 1990 to fall 1994. Tropical values of aerosol volume near 25 km span nearly two orders of magnitude from three months before to three months after the June 1991 eruption of Mt. Pinatubo. Retrieved values of V range from approximately 0.06 pm2cm 3 near 25 km in spring 1991 to 8 pm2cm 3 in fall 1991... 

Figure 9. Zonal/seasonal averages of aerosol volume density V as a function of altitude and latitude. Dotted lines mark the tropopause plus 2 kilometers. Upward-pointing triangles on the x-axis mark the eruptions of Kelut (Feb. 1990 - 8 S), Knatubo (Jun. 1991 - 15 N) and Hudson (Aug. 1991 - 46 S).

Figure 9 Ratio of tertiary to primary aerosol volume as a function of drop size for different sample uptake rates. A Cetac microconcentric nebulizer (MCN) was used in a double-pass spray chamber. Other concentric nebulizers behave similarly. (From Ref. 422.)...

The aerosol volume fraction. , is closely related to the mass concentration, which is usually determined by nitration. We assume the filter is ideal, removing all particles larger than single molecules. Then... 

Figure 11.4 Evolution of the moments of the size distribution function for the aerosols shown in Fig. 11.3, The peak in the number distribution probably results when formation by homogeneous nudeaiiun is balanced by coagulation. Total aerosol volume increases with time as gas-tO partide conversion takes place. Total. surface area, A, increases at first and then approaches an approximately constant value, due probably toa balance between growth and coagulation (Husarand Whitby, 1973). The results should be compared with Pig. 11.2.

Thus the volume distribution function is constant over the particle size range where the power law exponent p = —4. Can the constant volume distribution for p = —4 be compatible with the bimodal volume distribution that covers much the same particle size range The power law and bimodal volume distributions are equivalent only as a very rough approximation. Most of the aerosol volume is present in the accumulation... 

Figure 13.5 Aerosoi volume diiitributions for the Navujo (Page, AZ) power plant plume on 7/10/79. This is an example of a fow-humidity (3 to 20% RH) high-solar-radiation environment. The aerosol volume exce.ss over background Ls associated primarily with particles smaller than 0.3 /tm. This is be.si explained by homogcncuu. i gas-phase reactions that form a condensable product and not by aerosol-phase reactions. (After Wilson and McMurry. 1981.)...

This form is quite different from the diffusional growth expression (13.7) in its dependence on particle size in this case, larger particles grow faster than smaller ones. McMurry et ai. (1981) analyzed data for several power plant plumes (Fig. 13.6a). They found that both diffusion to the particles and droplet phase reaction contributed to plume aerosol growth (Fig. 13.6b). However, the droplet-phase reactions accounted for less than 20% of total aerosol volume growth. 

A series of 24 /w-xylene photoxidation experiments (blacklamp irradiation) were performed at 300K. Reacted hydrocarbon mass concentrations ranged from 149 to 1660 pg m, NOx concentrations ranged from 9 to 587 ppb, and HC NOx ratios (ppbC/ppb NOx) ranged from 1.3 to nearly 50. All experiments were performed until wall loss corrected aerosol volume plateaued reached a constant stable value. 

It is obvious that a similar size distribution function can be given for the surface, volume and mass of aerosol particles. Thus, e.g. the volume concentration (aerosol volume per unit volume of air) distributes according to particle radius in the following way ... 

Cantrell, B. K., and K. T. Whitby (1978). Aerosol size distributions and aerosol volume formation for a coal-fired power plant plume. Atmos. Environ. 12, 323-333. 

Another weakness of the power-law distribution is that, even if it is a reasonable approximation of the number distribution for Dp > 0.1 pm, it cannot provide a reasonable approximation of the typical volume distributions (Figure 8.9). Atmospheric aerosol volume distributions always have multiple modes for Dp > 0.1 pm, and (he power-law volume distribution that we calculated above is also a monotonic function (continuously decreasing for a > 3 and increasing for a < 3). For the distribution of Figure 8.9, the power-law distribution grossly overpredicts the volume of the submicrometer particles and seriously underpredicts the volume of the coarse aerosol. Therefore use of a fitted power-law distribution for the calculation of aerosol properties that depend on powers of the diameter (e.g., optical properties, condensation rates) should be avoided. 

FIGURE 8.13 Aerosol volume distributions next to a source (freeway) and for average urban conditions. 

FIGURE 8.16 Measured marine aerosol volume distributions and a model distribution used to represent average conditions. 

---