Investigation of the optical properties of secondary organic aerosols generated from hydrocarbon ozonolysis

This thesis describes and evaluates a simple scheme by which the refractive index (at 589 nm) of non-absorbing components common to secondary organic aerosols (SOA) may be predicted from molecular formula and density (g / cm3). The QSPR approach described is based on three parameters linked to refractive index – molecular polarizability, the ratio of mass density to molecular weight, and degree of unsaturation. After computing these quantities for a training set of 111 compounds common to atmospheric aerosols, multi-linear regression analysis was conducted to establish a quantitative relationship between the parameters and accepted value of refractive index.
The resulting quantitative relationship can often estimate refractive index to ± 0.01 when averaged across a variety of compound classes. A notable exception is for alcohols for which the model consistently underestimates refractive index. Homogenous internal mixtures can conceivably be addressed through use of either the volume or mole fraction mixing rules commonly used in the aerosol community.
Predicted refractive indices reconstructed from composition data presented in the literature generally agree with previous reports of SOA refractive index. Additionally, the predicted refractive indices lie near measured values we report for SOA generated from vapors of α-pinene and toluene provided the concentration of the precursor hydrocarbon is low.
This method may find use in reconstructing optical scattering of organic aerosols if mass composition data is known. Alternatively, the method described could be incorporated into in models of organic aerosol formation / phase partitioning to better constrain organic aerosol optical properties.