Novosselov, IgorSingh, Ishrat2022-07-142022-07-142022-07-142022Singh_washington_0250O_24542.pdfhttp://hdl.handle.net/1773/49086Thesis (Master's)--University of Washington, 2022The presence of particulate matter (PM) in the environment can lead to adverse health impacts, including respiratory, cardiovascular, neurological, and lung cancer. Air pollution has been estimated to cause 4.9 million deaths, and 94% were caused by PM. Sources of combustion-generated PM range include wildfires, residential wood burning, traffic emissions, etc. While the epidemiological link between PM exposure and adverse health effects is clear, there is a lack of information regarding source-specific differences in PM toxicity. Thus, there is a clear need to quantify PM presence in the environment and identify its sources and toxicity. Reactive oxygen species (ROS) have been proposed as one surrogate metric for the toxicity of PM. Excitation emission matrix (EEM) spectroscopy has been well documented as a low-cost, reliable method for analyzing the organic fraction of PM and can be used in source apportionment. In this study, we investigate the correlation between EEM signature and ROS measurements in PM. PM collected from laboratory flame cookstove (natural and forced draft), and wildfire smoke (collected indoor and outdoor) are analyzed by EEM and the dithiothreitol (DTT) assay for ROS. While total integrated fluorescence of EEM does not provide clear patterns of association with ROS, with the implementation of principal component analysis (PCA) and regression of the EEM spectra, we show that specific EEM patterns can be correlated with the ROS measurement. EEM-PCA may be a useful alternative to evaluate the ROS level in combustion-generated aerosols.application/pdfen-USnoneEnvironmental engineeringAtmospheric sciencesMechanical engineeringThesis