Excitation Emission Matrix Fluorescence Spectroscopy based Sensing of Combustion Generated Particulate Matter

Abstract

Human exposure to particulate matter (PM) is associated with adverse health effects, including cardiopulmonary diseases, neurological diseases, and lung cancer. PM can originate from mobile and industrial combustion sources, forest fires, domestic fuel burning, and other natural (soil dust, sea salt, etc.) and anthropogenic sources. Within the human respiratory tract, PM size determines the region of deposition, residence time, solubility, and tissue uptake. The chemical composition of the particle determines the potential for biochemical reaction with tissues and cells. The effect of PM exposure on health outcomes is a critically important research area. Miniaturized exposure monitoring devices and flexible orthogonal in-situ analysis of samples would greatly enhance the ability to clarify the relationships between PM exposure and health. Real time knowledge of these relationships can lead to improved health risk assessment due to PM exposure, guidance for the management of diseases, and targeted intervention strategies to reduce health risk. This work addresses the need for improved exposure assessment, quantification, and characterization of ultrafine particles in the environment. The anticipated outcome of this research is the development of a real-time analysis tool that will collect airborne nanoparticles and provide in-situ analysis of PM chemical composition—which will determine their toxic potential using spectroscopic techniques. Chemical characterization of PM is critical for the assessment of the hazard associated with human exposure to potentially harmful agents, notably combustion-generated PM. Specifically, polycyclic aromatic hydrocarbons (PAHs) found in combustion PM have been associated with carcinogenic and mutagenic effects. Chapter 1 introduces the key concepts useful for development of the PM analysis tool. In Chapter 2, we analyse PM chemical composition from various sources for (1) organic carbon content, (2) the presence and concentrations of PAHs with low molecular weight (LMW, 126<MW<202), which may exhibit acute toxicity, and (3) higher molecular weight PAHs (HMW, 226<MW<302) which may have carcinogenic impacts. We use combustion-generated PM from both laboratory sources as well as real-world sources (e.g., woodsmoke and diesel exhaust PM). The colder flames result in lower PM yields; however, fraction of PAHs constituting the PM increases significantly. For hotter flames, PM yields are higher with very low fractions of PAHs constituting the PM. In Chapter 3, we applied the Principal Component Regression – Excitation Emission Matrix (PCR-EEM) method to predict LMW and HMW PAH concentrations as well as individual PAH concentrations in combustion generated PM from 16 reference PAHs. The reference PAH concentration are obtained using Gas Chromatography Mass Spectroscopy (GCMS). The models are trained on samples from the UW inverted gravity flame reactor (IGFR) and tested on PM emissions from biomass cookstove combustion and diesel exhaust samples accurately predicts HMW PAH concentrations with R-squared = 0.976 and overestimates LMW PAHs. In Chapter 4, we present the design and evaluation of am electrostatic precipitation based capillary-based sensor that collects fine and ultrafine particulate matter onto the outer surface of a capillary for in-situ spectroscopic analysis. At optimized operating conditions, the capillary collection efficiency is relatively constant at 65 % for all particles in the 0.2 μm - 2 μm range while it drops below 50 % for the larger particles, likely due to a lower charge-to-mass ratio. Fluorescence spectroscopy is used to demonstrate the integration of the capillary collector with in-situ spectroscopic analysis techniques. Automation of fluorescence analysis is required for in-situ spectroscopic analysis of PM. However, solvent extraction as a preprocessing step and the need for benchtop instrumentation hinders development of a miniaturized sensor. In Chapter 5, we present a methodology that eliminates labor-intensive sample preparation and allows to miniaturize the detection platform. We describe the development of a solid phase excitation emission matrix (SP-EEM) analysis method that eliminates pre-processing steps and reduces the cost of the standard liquid phase excitation emission matrix (LP-EEM) analysis (the experimental procedure followed in Chapter 3 to obtain EEMs of combustion generated PM). We evaluated external and internal excitation schemes for their ability to produce the best SP-EEM signatures with lowest scattering and reflection interference. Analysis of woodsmoke and cigarette smoke samples showed good agreement with LP-EEMs. In internal excitation, SP-EEM spectra shows fluorescent interference from PDMS for wavelengths < 250nm. The external excitation EEM spectra are dependent on the incident angle; ranges of 30-40⁰ and 55-65⁰ showed the best results. SP-EEM technique can be used for development of miniaturized sensors for chemical analysis of combustion generated PM. In summary, this thesis: (i) provides a statistical model for determining concentrations of potentially carcinogenic PAHs in combustion generated PM using EEM fluorescence spectroscopy; (ii) includes design and evaluation of a miniaturized collection and fluorescence analysis setup for PM2.5; and (iii) shows the feasibility of SP-EEM analysis for the development of miniaturized sensors for in-situ chemical analysis of combustion generated PM.