Validation and Application of a Microscopic Chemical Imaging System for PM Source Apportionment

Many urban and industrial areas suffer from high levels of coarse particulate matter (PM10) and fine particulate matter (PM2.5). In order to evaluate the health impacts of particulate matter and develop effective pollutant abatement strategies, one needs to know the source contributions to the observed concentrations. The most common approach involves the collection of ambient air samples on filters, laboratory analysis to quantify the chemical composition (e.g., IC, AA, XRF, automated colorimetry, thermal/optical reflectance for OC/EC, and GC/MS for PAHs), and application of receptor modeling methods such as the chemical mass balance (CMB) receptor model and positive matrix factorization (PMF) to determine the source contributions. This approach is expensive and time consuming. An alternative method for physically characterizing and apportioning the sources of ambient PM is the application of microscopic chemical imaging (MCI) to identify and apportion the sources of the ambient particulates. The MCI method involves measuring individual particle fluorescence coupled with morphological data to develop unique source profiles that form the basis of a source identification library. Ambient filter samples are then analyzed using the MCI method and the source attribution is based on the individual particle analysis coupled with identification using the source library. Using the MCI approach, the apportionment of ambient PM to specific sources can be performed in an inexpensive, affordable way and in near real-time, affording results that will provide valuable information to help policy makers and regulators take the necessary steps to abate such pollution and protect human health. In this dissertation a validation and application of the relatively inexpensive, near-real time MCI method for PM source apportionment is described