New factorisation tools for studying sources of soot, trace metals and organic matter in atmospheric aerosols

Abstract

Ambient aerosol particles, known for their harmful impacts on human health and the atmosphere, contribute to environmental degradation, visibility reduction, and global climate change. These aerosols, which are part of micron and submicron particulate matter, vary in size and are composed of thousands of chemical species, making their identification and quantification complex in different environments. Factorisation serves as a key method in aerosol research, allowing for the source apportionment of complex data sets obtained from aerosol studies. This process breaks down complex data into simpler, more understandable sources, helping in the analysis of aerosol sources and composition. Using factorisation with advanced instruments could represent more detailed and specific characteristics of aerosol particles that have not been detected before, providing greater understanding of their impact on the environment and human health. The thesis investigates the factorisation for the purpose of source apportionment of ambient aerosol particles, focusing on soot, organic aerosol (OA), inorganic aerosol (IA), and trace metals. This research covers their chemical composition and examines various climatological conditions, diurnal variations, and size-resolved particulate matter (PM). Additionally, the impact of the COVID-19 lockdown and the post-COVID-19 period on the concentrations of trace metal pollutants is examined. The study involves both online and offline measurements using advanced instruments like the High-Resolution Soot Particle Aerosol Mass Spectrometer (HR-SP-AMS), High-Resolution Aerosol Mass Spectrometer (AMS), Filter Inlet for Gases and Aerosols (FIGAERO) Chemical Ionisation Mass Spectrometer (CIMS), and Cooper Xact 625i ambient metal analyser for time-resolved and near real-time analysis. On Guy Fawkes Night, 5th November 2014, in Manchester, ambient soot data was collected for factorisation during a highly polluted event. This research achieved the first successful identification of more than two soot sources using the HR-SP-AMS, which no other current black carbon apportionment technique is currently able to do. The HR-SP-AMS also detected various metals such as Titanium (Ti), Caesium (Cs), Ferrous (Fe) and Strontium (Sr), in which Sr was taken as a tracer in fireworks, and the inclusion of Fullerene data was important in identifying multiple soot sources. Correlation analysis with existing Aethalometer, MAAP, and CIMS data provided a quantitative estimate of source contributions to the black carbon budget, demonstrating the HR-SP-AMS's effectiveness in soot source apportionment and firework marker identification. This study represents a significant advancement in the field of air pollution research by demonstrating for the first time that fullerene data, when integrated with PMF and applied to HR-SP-AMS data, can effectively apportion soot into distinct sources during high pollution events. It successfully differentiates between bonfire-related emissions and other urban pollution sources, offering a more deeper understanding of particulate matter's composition and origin. The distinct categorisation into Bonfire Night factors and non-Bonfire Night factors, along with the specific identification of firework’s tracer i.e., Strontium, provides a clearer picture of the contributions of different sources to the overall black carbon levels. The second research study conducted between 1st July 2019 and 17th September 2020, involved the collection of PM10 trace metal data in Manchester using the Xact 625i analyser. Positive Matrix Factorization (PMF) identified four trace metal sources: Non-exhaust emissions, biomass burning, sea salt, and construction dust. Multiple Linear Regression (MLR) analysis revealed significant correlations, with biomass burning influencing PM1 and PM2.5, and construction dust impacting PM10 and PMcoarse. The COVID-19 lockdown analysis showed a significant reduction in PM levels and exhaust em

Date of Award	1 Aug 2024
Original language	English
Awarding Institution	The University of Manchester
Supervisor	James Allan (Supervisor) & Hugh Coe (Supervisor)