Measurements of the diverse fractions of the particulate matter (PM) contribute significantly
to a better understanding of atmospheric aerosol. As an important fraction of PM, organic aerosol
(OA) is a mixture of thousands of organic compounds with different physical and chemical
properties. Molecular characterization of OA has long been a challenge. Fortunately, it benefits
much from the recent rapid development of analytical instrumentation used for laboratory setting
as well as online measurements in the field. There has been sustaining research interest on the
identification and quantification of secondary organic aerosol (SOA) tracers for the source
apportionment of PM and OA. This thesis work presents a comprehensive measurement of organic
aerosol on a molecular level: from offline measurement (offline GC-MS) to online measurement
(Thermal Desorption Aerosol Gas Chromatography, (TAG)), and from unit-mass-resolution to
high-mass-resolution (Orbitrap-MS). Measurements are conducted at two economic hubs in China:
the Pearl River Delta (PRD) region and the Yangtze River Delta (YRD) region, where PM
pollution is a big concern. The application of primary and secondary organic tracers into the source
apportionment model is subsequently examined. Major findings of each chapter are given below:
Chapter II reports work exploring the role of specific secondary organic molecular markers
(MM) in terms of tracking SOA and secondary organic carbon (SOC) from precursors perspective.
Laboratory offline quantification is carried out to obtain ambient concentrations for a set of
secondary organic tracers, derived from biogenic volatile organic compounds (VOCs) (i.e.,
isoprene, α-pinene, and β-caryophyllene), anthropogenic VOCs (i.e., naphthalene and toluene),
and secondary biomass burning (BB). These tracers are used as inputs in positive matrix
factorization (PMF) analysis, which successfully resolves six secondary related factors: secondary
sulfate formation processes, secondary nitrate formation processes, secondary BB formation
processes, SOA_I with collective contributions from naphthalene, ɑ-pinene, and β-caryophyllene,
SOA_II derived from isoprene, and SOA_III with collective contributions from ɑ-pinene and
toluene. Among the six secondary factors, four could not be resolved without precursor-specific
tracers, accentuating the utility of indicative tracers to refine SOA sources. In addition, we confirm
the consistency of SOC estimation between the tracer-based method and PMF output. Nevertheless,
additional tracers are still in need to further split the mixed factors resolved by PMF and fill the
gap of SOC underestimation by the tracer-based method.
Chapter III presents laboratory analysis of OA extracts using Orbitrap mass spectrometry
(MS) coupled with a soft electrospray ionization (ESI) source in positive (ESI+) and negative (ESI-)
modes. Ambient aerosol collected from a suburban site in Hong Kong and biomass burning source
aerosol are subjected to detailed characterization. Three extraction solvents are used and thus three
groups of organics are isolated, i.e., water-extracted (HUmic-LIke Substances, HULIS), methanol-extracted, and DCM-extracted organics. Benefiting from the remarkable resolving power and mass
accuracy, thousands of peaks are identified and successfully assigned with unambiguous molecular
formulas (with elements of C, H, O, N, and S in ESI- and C, H, O, and N in ESI+). In addition,
compounds with aromatic index (AI) larger than 0.5 are supposed to bear an aromatic moiety.
They are of preferential interest and further investigated using a self-developed multidimensional
Kendrick Mass Defect (KMD) framework. In this way, the distributions of the oxygenated organic
aromatic compounds in ambient and biomass burning source aerosol are systematically studied.
Chapter IV reports online bi-hourly measurements of organic tracers for three weeks (9
November – 3 December 2018) in Shanghai, deploying a commercial TAG system. Five episodes
with high concentrations of fine PM are observed and investigated. Compared with non-episode
hours, episodes are characterized by distinct mass increments of secondary species, revealing that
secondary organic and inorganic formation processes contribute significantly to the severe air
quality deterioration during the episodes in urban Shanghai. In addition, using the SOA tracers
measured by TAG, we reconstitute the organic matter (OM) concentration, among which the
contribution of toluene (tracer for vehicle exhaust) derived SOA surpasses 63%. The reconstituted
OM accounts for over 30% of the total OM measured by the Aerodyne aerosol mass spectrometer
(AMS). The results highlight the higher influence of anthropogenic emissions over biogenic
emissions during pollution episodes and call for urgent control strategies on local anthropogenic
emissions.
Chapter V documents the deployment of the TAG instrument for a six-week long (8 March –
19 April 2019) field campaign at the HKUST air quality research supersite. Real-time PM
2.5
concentration, OC/EC, water-soluble ions, and elements are measured in parallel. The average
concentrations of PM
2.5 and the three major ions (sulfate, nitrate, and ammonium) are 17.4 ± 9.2,
4.5 ± 2.2, 1.6 ± 0.8, and 1.7± 0.8 µg/m
3, respectively. A group of 25 polar organic compounds are
detected by the TAG system including five sub-categories: carboxylic acids, aromatic acids, fatty
acids, saccharides, and SOA tracers. Many of them are specific source indicators and can be
incorporated into PMF as high-time-resolution source markers. Then a more detailed view of
diurnal contributions of major sources and atmospheric processes to PM can be obtained.
Post a Comment