Introduction

41.1. Definition of the aerosol

An aerosol is defined as a dispersed system containing solid or liquid particles suspended in a gas. In our case the gaseous medium is the air in which aerosol particles of different composition and size are suspended.

In more rigorous terms an aerosol or aerocolloidal system exists, if (Hidy and Brock, 1970):

(a) the sedimentation velocity of the particles is small;

(b) inertial effects during particle motions can be neglected (the ratio of inertial forces to viscous forces is small);

(c) the Brownian motion of the particles, due to the thermal agitation of gas molecules, is significant and

(d) the surface of the particles is large compared to their volume.

The physical meaning of the above criteria is as follows. The external force acting on an aerosol is generally gravitation. This means that the lifetime of a particle in the system is determined by its sedimentation velocity. If the particle radius is greater than the mean free path of gas molecules, the vs falling velocity is given by the well-known Stokes equation:

where r and pp are the radius and density of the particle, assumed spherical, // is the dynamic gas viscosity (equal to 1.81 x 10~4 poise at a temperature of 20 °C) while g is the gravitational constant. In the atmosphere, t;s in equation [4.1] depends on altitude above the sea level. Furthermore, the updraft motions in the troposphere make the interpretation of the sedimentation velocity rather complicated. In spite of these problems in the surface air a value of 10 cm s"1 can be accepted with caution as an upper limit, which gives the falling speed of a spherical particle of 30 /tm radius if the density is taken to be unity. We have to emphasize, however, that due to the presence of updrafts, larger particles also can be found in the atmosphere at significant distances from their sources.

The ratio of inertial forces to viscous forces is per dejmitiomm the Reynolds number (Re) of particles. In this way the second criterion can be written in the form:

where p is the air density, while v is the speed of the particle motion caused by some external force. Physically [4.2] means that in a stable system the product of the particle speed and particle size cannot exceed a given value. Thus, under normal atmospheric conditions the speed of motion of a particle of 10 /im radius could not exceed 30 cm s" '. In the case of r = 30 pm the critical velocity is 10 cm s"If the external force arises from the gravitational field, this condition is obviously equivalent to the first criterion.

A very characteristic property of aerosol particles is their Brownian motion. This random motion is a result of the fluctuations in the impact of gas molecules on the particles. It goes without saying that the speed of this motion increases with decreasing size. Generally, Brownian motion is considered significant if the particle radius is smaller than 0.S pm.

Finally, the fourth criterion is satisfied if the particle surface (in cm2) exceeds the particle volume (in cm3) at least a thousand times. For this reason, surface phenomena play an important role in the behaviour of aerocolloidal systems.

The foregoing conditions determine the upper limit of the particle size. The lower limit can be specified in a very simple way. A system is considered an aerosol when the radius of the particles is greater than that of gas molecules. At the same time ma > mg, where mu and mg are the mass of aerosol particles and gas molecules, respectively. Bearing in mind the size of molecules in the air we might define the lower limit to be around 10"7 cm ( = 10~3 pm).

An important consequence of the Brownian motion of aerosol particles is their collision and subsequent coalescence. This so-called coagulation process can be characterized by the particle loss per unit time (Hidy and Brock, 1970):

where N is the number of particles per unit volume, t is the time and D is the diffusion coefficient of particles:

In equation [4.4] k is the Boltzmann constant (1.3803x10 16 erg/degree x x molecule), T is the absolute temperature, A is the Stokes Cunningham correction1, while / is the mean free path of gas molecules2. Thus, the coagulation equation may be written in the following form:

It is concluded on the basis of equation [4.S] that the intensity of the particle loss due to the thermal coagulation is directly proportional to square of the particle concentration, while the coagulation efficiency increases with decreasing particle radius. This means that the coagulation of small particles at a high concentration is a very rapid process. Equation [4.5] is valid only for monodisperse aerosols, i.e. aerosols composed of particles of uniform size. However, the same qualitative conclusion can also be drawn in the case of polydisperse systems.

It should be emphasized that there is no intention here to discuss the dynamics of aerocolloidal systems. For further details the reader is referred to textbooks specialized in the field (e.g. Hidy and Brock, 1970).

With regard to their formation process and size, aerosol particles can be divided into two distinct groups: fine and coarse particles (Whitby, 1978). Fine particles with radius smaller than 0.5-1.0 /tm are formed by condensation and coagulation (see Subsection 4.2.2), while coarse particles arise mostly from surface disintegration (see Subsection 4.2.1). Since this classification provides an explanation of the form of the particle size distribution, we will discuss Whitby's ideas in more detail in Subsection 4.3.2.

Another possibility is to classify particles simply according to their size. In atmospherics physics and chemistry the classification of Junge(1963) is widely used. Junge divided aerosol particles in three groups:

This division is very convenient from the point of view of particle characterization and measurement. Thus, in the range of Aitken particles diffusion effects are significant and particle coagulation is rapid. However, in case of giant particles these phenomena can be neglected and the behaviour of aerosol particles is mostly determined by their sedimentation due to gravitation. The large particles constitute

21 = 6.53 x 10"6 cm, at a temperature of 20 C and a pressure of 760 mmHg.

3 The concentration of these particles is generally measured by means of expansion chambers, the first versions of which were constructed by Aitken.

41.2 Classification and measurement of atmospheric aerosol particles

Aitken3 particles: r<0.1 /im; large particles: 0.1 £ r < 1.0 /tm; giant particles: r^l.O /an.

a transition between the two characteristic ranges mentioned. Since their size is comparable to the wavelength of visible light these particles play a great role in the optical properties of the air. Large and giant particles have a significant inertia which can be utilized for their measurement (see below). Moreover, the gianl particles and significant portion of the large particles can be studied with an optical microscope.

Because of their small radius, the size and size distribution (see Subsection 4.3.2) of Aitken particles may be determined with a diffusion battery. This device is composed by an ensemble of capillary tubes, through which the air is drawn at low velocity. As a result of their Brownian diffusion, the smaller aerosol particles are deposited on the walls of the tubes during the aspiration. This particle loss is a function of the diffusion coelficient and consequently of the size of the particles (see equation [4.4]).

The total number concentration of aerosol particles can be measured with expansion chambers. In these devices the air sampled is humidified and suddenly expanded to produce a significant water vapour supersaturalion. In the supersaturated environment, water condenses on the aerosol particles. The number of droplets formed in this way is equal to the particle concentration. The droplets are generally counted allowing them to settle on a microscope slide or, after calibration, by the extinction of a light beam through the chamber. This type of measurement actually determines the total number concentration of particles. However, since the concentration of Aitken particles is much larger than that of large and giant particles (see later), the result essentially gives the number of particles with radius smaller than 0.1 /xm.

The smaller aerosol particles can be captured from the air for subsequent counting and size measurement by means of so-called thermal precipitators. In these instruments, metal wires are heated to produce a temperature gradient. Aerosol particles move away from the wire in the direction of a cold surface, since the impact of more energetic gas molecules from the heated side gives them a net motion in that direction. The particles captured are studied with an electron microscope. Another possible way to measure Aitken particles is by charging them electrically under well-defined conditions. The charged particles are passed through an electric field and are captured as a result of their electrical mobility (see equation [4.6]). Since size and electrical mobility are related, the size distribution of particles can be deduced. These devices are called electrical mobility analyzers.

There are several methods to detect large particles. Thus, particles can be sludied in situ in the gaseous medium. In single particle optical counters, the particles are illuminated and the light scattered individually by each particle is measured photoelectrically at a given angle. The number of such scattering signals is a measure of the particle number while the amplitude of each signal gives, after suitable calibration, the particle size. In another type of optical device, we detect the light scattered by a cloud of particles in a fixed solid angle. In this case the counter is called an integrating nephelometer.

Generally, the large and giant particles are collected from a given air sample by inertial deposition. Inertial sampling is carried out by placing an obstacle or collector in the air stream as shown in Fig. 21(a). As can be seen the air goes around the obstacle, while the trajectory of particles with a higher inertia than the air molecules deviates from their trajectory, and the particles strike the collector. With increasing gas stream velocity, and with decreasing collector size, the radius of the particles collected decreases." In other words this means that the efficiency of the collection (impaction) increases. The impaction efficiency "is defined as the ratio of the volume of gas cleared of particles by the collection element to the total volume swept out by the collector" (Friedlander, 1977). During aircraft sampling the collector (e.g. a microscopic slide) is exposed directly in the moving air. If the aircraft has a velocity of 200 km hr"1, large particles can be captured with an acceptable efficiency on slides 0.1 mm wide. For giant particles this characteristic size is 1 mm. (b) Impactor Fig. 21

Inertial collection of aerosol particles by an obstacle and by a two-stage impactor

Fig. 21

Inertial collection of aerosol particles by an obstacle and by a two-stage impactor

In the surface air the wind speed is generally not high enough for the direct collection of large particles. However, a suitable relative velocity can be achieved by attaching the collecting slide to the end of a rotating arm. Alternatively, the air can be accelerated by pumping it into or through a tube. In this latter case the stream

4 See Subsection 5.2.1.

velocity can be further increased by passing the air through narrow slits. Such a device is called an impactor; its collection efficiency is an inverse function of the slit size. A great advantage of this sampling procedure is that particles with different dimensions can be separated by means of an impactor containing slits (jets) with different sizes. Fig. 21(b) represents in a schematic way such an instrument of two stages, termed a cascade impactor. As we see the air is sucked gradually through progressively narrower slits. This allows the capture of smaller and smaller particles by the slides placed at an appropriate distance behind the jets.The particles collected are counted and their size is measured by an optical or electron microscope or they may be analyzed chemically (see Subsection 4.4.1).

The different collection procedures are combined by suitable filters consisting of fine fibers or membranes. These filters remove large and giant particles from the air by inertial impact on the fibers (e.g. glass fibers) or around the holes of membranes, while Aitken particles are collected by making use of their Brownian diffusion. If the filter material is electrically charged, electric forces have to be also taken into account. For the microscopic study of aerosol particles, the membrane filters are very suitable. These filters consist of synthetic organic membranes containing holes of approximately cylindrical form. To obtain filters with holes of uniform size, the filter mentioned is bombarded with fission recoil fragments, and the nuclear tracks are then etched out chemically (Nuclepore filters).

Experimental methods for aerosol studies have recently been reviewed by Friedlander (1977). The reader can find further details and references in his book.

41.3 Importance of the study of atmospheric aerosol

We have mentioned several times that the study of aerosol particles contributes to the solution of many problems in atmospheric science. Figure 22 shows some of the branches of atmospheric physics that involve aerosol studies.

 Nomenclature
0 0