A Method for Estimating the Bipolar Charge Distribution Variation on Aerosol Particles with Atmospheric Conditions

Abstract:

Many commonly used instruments that measure the aerosol particle number size distribution, such as Scanning Mobility Particle Sizer, are based on the following principle: The particle sample is brought to a steady-state charge distribution in a bipolar aerosol charger. The particles are then segregated according to their electrical mobilities using differential mobility analyzer, DMA. Finally, the concentration of the particles is measured using a condensation particle counter, CPC. To estimate the particle size distribution, the concentration data are then inverted using an algorithm that takes into account the steady-state charge distribution and the performance characteristics of the DMA and CPC. Considerable effort has gone into the characterization of the instruments used in these measurements. The charge distribution remains the greatest source of uncertainty in the mobility based size distribution measurements. The charge distribution depends, at least, on the properties of the ion and particle (radius, density, relative permittivity and number of charges), concentrations of negative and positive ions, temperature and pressure. With these values given, the collision frequencies of ions and particles can be modeled to determine the charge distribution, but that can be very time consuming. Instead, the charge distribution is usually estimated using a simple parameterization of the results of one such model. The collision process can, however, be described using only two dimensionless parameters, namely the diffusive Knudsen number, Kn_D, and the ratio of electric potential energy to thermal energy, Ψ_E. A given pair of Kn_D and Ψ_E may describe multiple collision conditions, but it defines a single value of dimensionless flux coefficient, H. This allows us to tabulate the values of Kn_D, Ψ_E and H, so that H can be determined by interpolation for conditions corresponding to the measurements. The charge distribution can readily be calculated from the interpolated values of H. This procedure is particularly useful in the case of airborne measurements, since pressure and temperature variations with altitude result in a large deviations of the charge distribution from the commonly used parameterization.