J, zyxwvutsr Phys. Chem. zyxwvuts 1980, 84, zyxwvut 1611-1614 1611 (5) D. Y. H. Pui and B. Y. H. Liu, “TsI Quarterly”, TSI Inc., St. Paul, MN, (6) D. D. Cooke and M. Kerker, Appl. Opt., 14, 734 (1975). (7) B. Y. H. Llu, R. N. Berglund, and J. K. Agarwal, Afm. Environ., 8, (8) K. T. Whitby, D. A. Lundgren, and C. M. Peterson, Int. J. Air Water (9) R. A. Fletcher, G. Mulholland, zyxwvutsrqpo I. Chabay, and D. zyxwvutsr S. Bright, J. Aerosol (10) J. P. Gollub, I. Chabay, and W. H. Flygare, Appl. Opt., 12, 2838 (1973). (11) J. P. Gollub, zyxwvutsrqpon I. Chabay, and W. H. Flygare, J. Chem. phys., 61, 2139 (12) I. Chabay and D. S. Bright, J. ColbMInterface Sci., 63, 304 (1978). (13) C. N. Davlies, Proc. zyxwvutsrqponm R. SOC. London, 57, 18 (1945). (14) H. 2. Cumrnlnsand M. L. Swinney, Prog. ClqD., 8, 135, 155, 157 (1970). (15) S. K. Frekllander “Smoke, Dust and Haze: Fundamentals of Aerosol May/June 1979. 717 (1974,). Poliut., 9, 263 (1965). Sci., in press. (1974). Behavior”, Wiley, New York, 1977, Chapter 2. issue. 3, 381 (1969). (16) R. A. Fletcher, D. S. Brlght, and I. Chabay, J. Phys. Chem., this (17) B. Y. H. Liu, V. A. Marple, and H. Yazdani, Enviorn. Sci. Techno/., (18) R. Chang and E. J. Davis, J. ColioM Interface Sci., 54, 352 (1976). (19) E. J. Davis and A. K. Ray, J. Aerosol Sci., 9, 411 (1978). (20) R. A. Mllllkan, Phys. Rev., 22, l(1923). (21) M. Kerker, “The Scatterlng of Light and Other Electromagnetic Radiation”, Academic Press, New York, 1969. (22) J. V. Dave, “Subroutines for Computing the Parameters of Elec- tromagnetic Radiation Scattered by a Sphere”, Report No. 320-3237, IBM Scientific Center, Palo Alto, CA, May, 1968. (23) J. Happel and H. Brenner, “Low Reynolds Number Hydrodynamics With Special Appllcatlons to Partlculate Media“, Prentlce-Hall, En- glewood Cliffs, N.J., 1965. Low Reyniolds Number Fluid Flow Induced by Settling Aerosol and Detected by the Particle Doppler Shift Spectrometer Robert A. Fletcher,” Davld S. Brlght, and Ilan Chabay Gas and Particulate Science Dlvlsion, Center for Analytical Chemistry. Natlonai Bureau of Standards, Washington D.C. 20234 (Received September 28, 1979) Publlcation costs assisted by the National Bureau of Standards Air flow generated by gravitationally settling 5-15-pm diameter droplets has been observed. The magnitude of the flow phenomenon is derived from measurements of the particle settling velocities by using the particle Doppler shift spectrometer (PDSS). The PDSS can determine particle velocity and size with high accuracy due to the inherent internal calibration characteristic of the instrument. The magnitude of the fluid flow velocity is dependent on the particle number concentration. Experimental evidence verifying the flow pattern is presented. Introduction Each settling particle in a cloud drags some of the air medium with it which effectively increases the settling velocity of neighboring particles. The net effect is to in- crease the apparent settling velocity of a cloud of particles. In a closed system, such as our PDSS optical scattering chamber, the downward flow of the medium, which results from the settling particles, must be compensated by an equal, but oppositely directed flow. If only part of the chamber is fillled with settling particles, an air circulation pattern will ble set up as shown in Figure la. This occurs because the .return flow will be along the path of least resistance which is the path void of particles. If particles homogeneously fill the tube as in Figure lb, there will be no apparent Settling velocity increase due to interaction of the return flow with the aerosol. This description follows that of Happel and Brenner.l An alternative way to describe this particle-induced fluid motion is to examine the net flow in a small sample volume element. For case lb, the net flow will be zero for a small volume element taken anywhere in the chamber. While for case la, if a volume element is taken in the particle stream, a net downward flow will be observed. Likewise, outside the particle column, there will be a upward flow. In both cases, the net flow through any cross-sectional slice of the tube is zero. We present the first experimental measure of settling- aerosol-induced fluid flow in the low Reynolds number regime. The fluid velocity measurements are made by measuring the settling velocity of clouds of 5-15-pm di- ameter dioctyl phthalate (DOP) particles which are gra- This article not subject to U.S. Copyright. vitationally settling in a small closed vessel filled with air. The particle Doppler shift spectrometer (PDSS) is used to detect and measure the particle-induced fluid motion. The fluid motion is observed as a small velocity increment to the particle settling velocity which is calculated from Stokes law. Measurements of such small velocities are primarily due to the internal calibration capability of the PDSS. The PDSS measures particle settling velocity directly. From the gravitational settling velocity, the aerodynamic size can be determined. The PDSS is capable of deter- mining particle diameter to an accuracy of 0.05 pm (Bright et al.)’ as a result of an internal calibrating procedure based on the fact that the light scattering intensity is a rapidly varying function of particle diameter over the size range of 5-15 pm. The Experimental Section contains a description of the PDSS and experimental conditions. In the Results section, the data reduction technique and experimental results are presented. The Discussion section summarizes the work and offers possible applications of the results. Experimental Section A schematic of the experiment is shown in Figure 2. Pure DOP aerosol is generated by a medicinal atomizer modified to generate large (5-15-pm diameter) aerosol. The atomizer is driven by a positive pressure air stream that is precisely flow controlled. Dilution air flowing past the atomizer entrains the aerosol stream and determines the final particle concentration. The aerosol is then passed through a charge neutralizer and then is sampled in a 1.3 Published 1980 by the American Chemical Society