Search

International audience; With over 8 million inhabitants and 4 million motor vehicles on the streets, Tehran is one of the most crowded and polluted cities in the Middle East. Frequent exceedances of national daily PM2.5 limit have been reported in this city during the last decade, yet, the chemical composition and sources of fine particles are poorly determined. In the present study, 24-hour PM2.5 samples were collected at two urban sites during two separate campaigns, a one-year period from 2014 to 2015 and another three-month period at the beginning of 2017. Concentrations of organic carbon (OC), elemental carbon (EC), inorganic ions, trace metals and specific organic molecular markers were measured by chemical analysis of filter samples. The dominant mass components were organic matter (OM), sulfate and EC. With a 20% water-soluble organic carbon (WSOC) fraction, the predominance of primary anthropogenic sources (i.e. fossil fuel combustion) was anticipated. A positive matrix factorization (PMF) analysis using the ME-2 (Multilinear Engine-2) solver was then applied to this dataset. 5 factors were identified by Marker-PMF, named as traffic exhaust (TE), biomass burning (BB), industries (Ind.), nitrate-rich and sulfate-rich. Another 4 factors were identified by Metal-PMF, including, dust, vehicles (traffic non-exhaust, TNE), industries (Ind.) and heavy fuel combustion (HFC). Traffic exhaust was the dominant source with 44.5% contribution to total quantified PM2.5 mass. Sulfate-rich (24.2%) and nitrate-rich (18.4%) factors were the next major contributing sources. Dust (4.4%) and biomass burning (6.7%) also had small contributions while the total share of all other factors was < 2%. Investigating the correlations of different factors between the two sampling sites showed that traffic emissions and biomass burning were local, whereas dust, heavy fuel combustion and industrial sources were regional. Results of this study indicate that gas- and particle-phase pollutants emitted from fossil fuel combustion (mobile and stationary) are the principal origin of both primary and secondary fine aerosols in Tehran.

The aerosol mass spectrometer (AMS), combined with statistical methods such as positive matrix factorization (PMF), has greatly advanced the quantification of primary organic aerosol (POA) sources and total secondary organic aerosol (SOA) mass. However, the use of thermal vaporization and electron ionization yields extensive thermal decomposition and ionization-induced fragmentation, which destroy chemical information needed for SOA source apportionment. The recently developed extractive electrospray ionization time-of-flight mass spectrometer (EESI-TOF) provides mass spectra of the organic aerosol fraction with a linear response to mass and no thermal decomposition or ionization-induced fragmentation. However, the costs and operational requirements of online instruments make their use impractical for long-term or spatially dense monitoring applications. This challenge was overcome for AMS measurements by measuring re-nebulized water extracts from ambient filter samples. Here, we apply the same strategy for EESI-TOF measurements of 1 year of 24-hour filter samples collected approximately every 4th day throughout 2013 at the NABEL monitoring station at Zurich-Kaserne, an urban site. The nebulized water extracts were measured simultaneously with an AMS. The application of positive matrix factorization (PMF) to EESI-TOF spectra resolved seven factors, which describe water-soluble OA: less and more aged biomass burning aerosol (LABBEESI and MABBEESI, respectively), cigarette smoke-related organic aerosol (CS-OAEESI), primary biological organic aerosol (PBOAEESI), biogenic secondary organic aerosol (BSOAEESI), and a summer mixed oxygenated organic aerosol (SMOAEESI) factor. Seasonal trends and relative contributions of the EESI-TOF OA sources were compared with AMS source apportionment factors, measured water-soluble ions, cellulose, and meteorological data. Cluster analysis was utilized to identify key factor-specific ions based on PMF. Both LABB and MABB contribute strongly during winter. LABB is distinguished by very high signals from C6H10O5 (levoglucosan and isomers) and C8H12O6, whereas MABB is characterized by a large number of CxHyOz and CxHyOzN species two distinct populations: one with low H : C and high O : C, and the other with high H : C and low O : C. Two oxygenated summertime SOA sources were attributed to terpene-derived biogenic SOA, a major summertime aerosol source in Central Europe. Furthermore, a primary biological organic aerosol factor was identified, which was dominated by plant-derived fatty acids and correlated with free cellulose. The CS-OA factor contained a high contribution of nicotine and high abundance of organic nitrate ions with low m/z.

In this study, we implement a numerical model to predict secondary organic aerosol (SOA) formation from semi- and intermediate-volatility organic compounds emitted from in-use gasoline and diesel vehicles. The model is formulated based on the volatility basis set (VBS) approach, and it accounts for OH oxidation of unspeciated low-volatility organics, which are classified by their volatility. This model incorporates SOA formation data from smog chamber and emission measurements of vehicle exhaust in a Hybrid framework to calculate the contribution of both traditional and non-traditional SOA precursors to total SOA formation observed in photo-oxidation experiments. Emission and SOA formation data were acquired from a series of experiments conducted at the Center for Atmospheric Particle Studies, Carnegie Mellon University on different gasoline and diesel vehicles. Instead of assigning surrogate compounds, the source-specific mass yield for non-traditional SOA precursors are calculated directly from the experiments. The present performance of the model is in good agreement with its previous application to aircraft exhaust. Based on the model predictions, a set of average VBS mass yields for each class of vehicles is presented. The obtained yield matrix is able to reproduce the observed SOA concentrations in the smog chamber within the measurement uncertainty.

This study investigates the secondary organic aerosol (SOA) formation potential of carburetor motorcycles exhaust. This type of two-wheeler is a popular means of transport in many Asian cities. A volatility-based numerical model was employed to predict SOA formation from a fleet of motorcycles in Tehran, capital of Iran. The fleet was a combination of four-stroke, gasoline-powered motorcycles with different engine displacement volumes. Total hydrocarbon (THC) emission factors of all motorcycles were previously measured in a chassis dynamometer laboratory according to cold start Euro-3 emissions certification test procedures. Due to incomplete combustion and lack of control on exhaust emissions, unburned fuel was assumed to be a good surrogate for the exhaust of carburetor motorcycles, regarding SOA formation. 150 cc engine and 200 cc engine motorcycles had the highest SOA formation potential, under atmospheric oxidant concentration, while 125 cc engine motorcycles had the highest SOA emission factor (travel- and fuel-based). It was found out that SOA emission factor of 125 cc engine motorcycles could increase up to 20%, three to five years after production. Average SOA formation from carburetor motorcycles in the present study was 4 times higher than Euro-4 passenger cars and 20 times higher than direct emission of particles from Euro-2 motorcycles and on (according to EMEP/EEA levels for PM2.5 emission factors). Carburetor motorcycles with 180 cc engine volume in the present study, had the lowest SOA formation potential.