Characterization of Aerosols Using Multiwavelength Multistatic Optical Scattering Data
Document ID: 111
Doctoral Dissertation
North Carolina State University
Physics
Raleigh, North Carolina
Abstract
SNYDER, MICHELLE GRACE. Characterization of Aerosols Using Multiwavelength Multistatic Optical Scattering Data. (Under the direction of C. Russell Philbrick and Hans Hallen.)
The study of aerosols and their effect on climate have direct application to global warming and cloud formation. The technique and instrumentation presented here aims to determine aerosol characteristics, including index of refraction, size distribution and number concentration.
Research is conducted using the multistatic multiwavelength optical scattering in conjunction with the polarization ratio method to extract aerosol characteristics. Scattering intensity is measured by a charge-coupled device imager with a field of view along the profile of co-aligned laser beams capturing scattering angles from 120 to 175 degrees. The scattered wavelengths are then separated by a diffraction grating when imaged. The polarization ratio of measured intensities is compared to theoretical Lorentz-Mie scattering calculations by an inversion process to obtain aerosol characteristics.
A rigorous analysis of the polarization ratio technique is presented as it can be applied to single mode, biomodal and trimodal distributions of aerosols. The effectiveness of this method is explored through calculations of squared error for simulated distributions both with and without noise to determine the applicability of the polarization ratio technique. This analysis includes in-depth studies into each of the five parameters that affect the polarization ratio: the mean radius, the standard width, the concentration, the real index of refraction and the imaginary index of refraction. Additionally, this analysis allows error bars to be set on inverted aerosol parameters and gain insight into the optimum wavelengths for characterization.
The inversion process is also discussed, with focus on convergence and accuracy of multiple methods. Prior research has relied upon a binary genetic algorithm, which is shown to be an improvement over a generic Monte Carlo routine. An enhancement to these is shown with the use of a continuous genetic algorithm, which is successful at resolving single mode and dual mode distributions.
Well-known narrow size distributions and controlled materials are used to verify the methods of inversion and characterize the errors associated with this technique. A time lapse sequence of measurements is taken as the aerosol concentration decays to examine this effect on the polarization ratio. Aerosol experiments are conducted in environmentally controlled chambers at North Carolina State University and at the Johns Hopkins University Applied Physics Lab-oratory. Some measurements included an aerodynamic particle sizer, which is how aerosols are currently characterized using a point measurement. The simultaneous measurement allows comparison of aerosol characteristics determined by point measurements and scattering volume measurements. Both visible and near infrared wavelengths were used, and a preliminary study of a nonspherical bioaerosol was conducted.
Ultimately, the goal of this research is to develop a technique and analysis approach that can be directly applied to the construction of a scattering instrument for aerosol characterization. The instrument system will involve a data acquisition and inversion routine that are applicable in multiple environments and measure a large range of aerosol sizes.
This research is a step toward the end goal and significantly improves the understanding of the polarization ratio, methods of data acquisition and analysis, optimization of the technique by experimental improvements, development of an inversion routine, and touches on the applicability of this techniques to non-spherical aerosols.
| Citation: | M. G. Snyder, "Characterization of Aerosols Using Multiwavelength Multistatic Optical Scattering Data", North Carolina State University, Raleigh, NC, United States, Doctoral Dissertation, December 2011, 176 pages |