Dry submicron particles are important in the chemistry of atmospheric aerosols and climate conditions. Although there are data on the role of liquid particles in the atmosphere, the same is not true for dry particles such as soot, mineral aerosols and salts. Background soot in the lower stratosphere has a mean geometric diameter of 0.14 m m. Soot in the upper troposphere has a mean geometric diameter of < 0.2 m m. Solid particles generated in laboratories currently are usually larger than those found in the earth's atmosphere. An available supply of small submicron particles is needed to study atmospheric models in the laboratory.

Fluidized bed generators (FBG) are aerosol generators that use air or other gas pressure to fluidize a powder resulting in the release of dust particles. FBG's currently in use tend to either produce larger particles than needed for atmospheric model studies or are too complex for laboratory use. The authors of this article describe the design and function of a modified fluidized bed aerosol generator to generate high number densities of dry particles using flow rates of <10 L/min. Their FBG is both less expensive, smaller, and more simple in function than most FBG's manufactured commercially.

Figure 1:  Schematic diagram of the fluidized bed aerosol generator with a feed system.  (Thumbnail - Click to enlarge)

Figure 1 above is a diagram of a FBG without a feed system for introduction of extra powder. The bed in the center of the unit contains bronze beads mixed with soot. Bronze beads are used because of their high density (~8g/cm³) (~40m m diameter) compared to that of soot. Soot in this case is carbon black. Dry nitrogen enters from below the bed and converts the bronze bead and soot mixture into an expanded suspended mass much like a liquid. From this comes the term, fluidized. Small light soot particles are caught in the upward nitrogen flow leaving the large soot particles and bronze beads behind in the bed.

The bed itself in this system is transparent polycarbonate tubing (15 cm in length with an ID of 4.76 cm). The figure shows two inlets for the nitrogen feed system at the right. There are also two inlets for the pressure change across the bed and a plugged inlet for an additional feed system between the nitrogen inlets. The bottom of the bed is covered with a mesh screen in turn covered with filters. These keep the bronze/soot mixture from crossing the screen into the air plenum chamber where the dry nitrogen enters. The filters also help to keep the nitrogen flow even throughout the bed and create a pressure decrease across the base of the bed.

The region above the bed is called the freeboard and is where particles separate due to differences in densities. A Po²¹⁰ charge neutralizer is employed above the freeboard because eluted soot particles are charged from mechanical friction. The dry nitrogen sheath flow protects the charge neutralizer from contamination with soot.

Figure 2:  Schematic diagram of the fluidized bed aerosol generator with a feed system  (Thumbnail - Click to enlarge)

Figure 2 above is a diagram of the same fluidized bed generator with the addition of a feed system. In this case, approximately 10 cm of bronze/soot is placed in the bed and the remaining bronze/soot is placed in the reservoir above. The reservoir is connected to the nitrogen flow through a stainless steel tube. The mixture of bronze and soot is added with the aid of a 3 way solenoid . The solenoid keeps the nitrogen flow through the air plenum chamber and bed constant. Because the height of the bed rises as the bronze/soot mixture is added, there is an overflow outlet on the left.

The fluidized bed generator was characterized with a TSI 3924 scanning mobility particle sizer (SMPS). The SMPS consists of a TSI 3071 differential mobility analyzer (DMA) and a TSI 3010 condensation particle counter (CPC). Data were collected with the SMPS software and serial port data acquisition program written in Labview.

The results of changing several parameters were measured with and without the feed system. Without the feed system, adding increasing amounts of soot to the bronze/soot mixture resulted in a parallel increase in the number density of soot; however the final result was below the target of 10⁵ - 10² particles cm^-3. There was also a linear relationship between the number density and increasing nitrogen concentration when the nitrogen flow rate was increased.

With the addition of the feed system there was an increase in the number density. However, number density did depend upon the amount of soot in the bronze/soot mixture as well as the nitrogen flow rate as was the case without the feed system. The size distribution as measured with the SMDS showed a bimodal distribution. One mode was in the supermicrometer diameter range and the other mode in the submicrometer diameter size range. Because the two modes were separate, the larger particles could be removed from the aerosol stream with impaction techniques. Placement of virtual impactors can be seen in Figure 1. When the submicron particles alone are examined, the size distribution is approximately log-normal with a count median diameter between 0.1 and 0.3 m m. Over a 4 hour time period, a number density of >10⁵ particles/cm³ can be produced within 1 standard deviation. Although more than 90% of the mass created by the bed consists of particles with a count median diameter of >1 m m micrometer, this mass makes up only a small fraction of the particle number.

The authors conclude that the goal of the study, generation of a dry, submicrometer aerosol for use in heterogeneous nucleation studies was fulfilled. The fluidized bed generator designed and tested is able to produce large number densities of solid submicrometer particles (10⁵ - 10⁶ particles/cm³) with a log-normal size distribution having a count median diameter of 0.1 to 0.3 m m. This production can be maintained for up to 4 hours. This number density is almost equivalent to that generated by a solution atomizer. The particle size is similar to that of atmospheric aerosols. A small number of large particles (>1 m m) is also generated but was not characterized. The major determinant of the output of the FBG is the ratio of soot to bronze beads in the bed mixture. Output also increased with the soot/bronze feed providing a constant supply of powder. The 3-way solenoid allowed additional control of the output.

By: Susan G. Shami, ScD