10th International Aerosol Conference September 2 - September 7, 2018 America's Center Convention Complex St. Louis, Missouri, USA
Abstract View
Application of Fisher Ratio and Principal Component Analysis for Identification of Unique Features in Complex Combustion-Emission Samples
CHRISTOS STAMATIS, Lindsay Hatch, William Lichtenberg, Georgios Karavalakis, Patrick Roth, Jiacheng Yang, Kelley Barsanti, University of California, Riverside
Abstract Number: 1571 Working Group: Combustion
Abstract In this work the application of chemometric data analyses is explored for identifying similarities and differences in the chemical composition of combustion emissions that may be linked to the formation of secondary organic aerosol (SOA). Emissions from gasoline combustion have a significant impact on the environment both locally and globally through the production of fine particulate matter (PM). U.S. gasoline blends contain hundreds of compounds, which through incomplete combustion can produce as many products. Many of these products can serve as precursors to SOA, a major mass fraction of fine PM in the atmosphere. Thus, there is motivation for understanding the linkages between fuel composition, combustion emissions, and SOA formation. Advanced instrumental techniques have been critical for increasing the number of compounds identified in complex combustion samples, including those that are most likely to serve as SOA precursors; however, the large data output from such techniques and the complexity of combustion samples, make data processing challenging and time consuming. Chemometric data analysis techniques are increasingly being used to enhance the information that can be obtained through data processing and analysis. As a complement to traditional data analysis, these approaches can lead to improved understanding and model representation of complex processes, including oxidation of gasoline combustion emissions to form SOA.
A MATLAB-based algorithm was developed for preparing and statistically analyzing two-dimensional gas chromatography with time-of-flight mass spectrometry (GC x GC-TOFMS) data. The algorithm performs the following operations: 1. aligns GC x GC-TOFMS chromatograms, 2. calculates Fisher ratios, and 3. performs principal component analysis (PCA) on compounds selected using the Fisher ratios. This algorithm has been applied to gasoline combustion samples collected onto dual-bed absorbent tubes and analyzed using GC x GC-TOFMS. Eight different fuels, with varying aromatic and ethanol content, were tested on a gasoline direct injection vehicle over the LA92 driving cycle; emissions over the entire driving cycle were diluted using a constant volume sampler (CVS) and injected into a mobile chamber where they were oxidized and SOA formation was measured. The Fisher ratio and principle component analyses were performed on samples collected from the CVS and the mobile chamber (prior to oxidation). Details of the algorithm and results of the chemometric analyses will be presented, including implications for better predicting the linkages between fuel composition, combustion emissions, and SOA production.