Primary Sources and Secondary Formation of Organic Aerosols in

Apr 9, 2012 - State Key Joint Laboratory of Environmental Simulation and Pollution Control, College of Environmental Sciences and Engineering, Peking ...
0 downloads 9 Views 3MB Size
Article pubs.acs.org/est

Primary Sources and Secondary Formation of Organic Aerosols in Beijing, China Song Guo,† Min Hu,†,* Qingfeng Guo,† Xin Zhang,† Mei Zheng,† Jun Zheng,‡ Chih Chung Chang,§ James J. Schauer,∥ and Renyi Zhang†,‡ †

State Key Joint Laboratory of Environmental Simulation and Pollution Control, College of Environmental Sciences and Engineering, Peking University, Beijing, China 100871 ‡ Department of Atmospheric Science, Texas A&M University, College Station, Texas 77843, United States § Reearch Center for Environmental Changes, Academia Sinica, Taipei, Taiwan ∥ Environmental Chemistry and Technology Program, Wisconsin State Laboratory of Hygiene, University of Wisconsin-Madison, Madison, Wisconsin 53706, United States S Supporting Information *

ABSTRACT: Ambient aerosol samples were collected at an urban site and an upwind rural site of Beijing during the CAREBEIJING-2008 (Campaigns of Air quality REsearch in BEIJING and surrounding region) summer field campaign. Contributions of primary particles and secondary organic aerosols (SOA) were estimated by chemical mass balance (CMB) modeling and tracer-yield method. The apportioned primary and secondary sources explain 73.8% ± 9.7% and 79.6% ± 10.1% of the measured OC at the urban and rural sites, respectively. Secondary organic carbon (SOC) contributes to 32.5 ± 15.9% of the organic carbon (OC) at the urban site, with 17.4 ± 7.6% from toluene, 9.7 ± 5.4% from isoprene, 5.1 ± 2.0% from α-pinene, and 2.3 ± 1.7% from β-caryophyllene. At the rural site, the secondary sources are responsible for 38.4 ± 14.4% of the OC, with the contributions of 17.3 ± 6.9%, 13.9 ± 9.1%, 5.6 ± 1.9%, and 1.7 ± 1.0% from toluene, isoprene, α-pinene, and β-caryophyllene, respectively. Compared with other regions in the world, SOA in Beijing is less aged, but the concentrations are much higher; between the sites, SOA is more aged and affected by regional transport at the urban site. The high SOA loading in Beijing is probably attributed to the high regional SOC background (∼2 μg m−3). The toluene SOC concentration is high and comparable at the two sites, implying that some anthropogenic components, at least toluene SOA, are widespread in Beijing and represents a major factor in affecting the regional air quality. The aerosol gaseous precursor concentrations and temperature correlate well with SOA, both affecting SOA formation. The significant SOA enhancement with increasing water uptake and acidification indicates that the aqueous-phase reactions are largely responsible SOA formation in Beijing.



apportionment.3−10 Chemical mass balance (CMB) model has been successfully applied to apportion POA in various environments around the world,11−13 so that primary source emissions are relatively well understood. However, the comprehension of secondary sources is more difficult, due to the lack of knowledge on their composition and formation mechanisms.14 The United States Environmental Protection Agency (USEPA) has recently identified several SOA tracers from the oxidation of specific precursors, including isoprene, α-

INTRODUCTION Organic compounds constitute a significant fraction of atmospheric fine particulate matter (PM), and a comprehensive understanding of their impacts on human health, climate, and visibility requires a detailed characterization of their composition and sources.1 However, current knowledge about organic aerosols, including primary organic aerosols (POA) emitted directly from primary sources and secondary organic aerosol (SOA) from the oxidation of volatile organic compounds (VOCs), are still very limited. The most uncertain part is probably the chemical composition and formation mechanisms of SOA.2 Recently, much effort has been made to identify particulate organic tracers which are useful in source identification and © 2012 American Chemical Society

Received: Revised: Accepted: Published: 9846

November 28, 2011 March 5, 2012 April 9, 2012 April 9, 2012 dx.doi.org/10.1021/es2042564 | Environ. Sci. Technol. 2012, 46, 9846−9853

Environmental Science & Technology

Article

pinene, β-caryophyllene, and toluene from environmental chamber experiments,15−18 and those tracers have been proven to reasonably quantify SOA yields and contributions in ambient PM.19,20 However, due to the complexity of SOA tracer analysis, few SOA tracer measurements have been conducted in the field,13,19−23 especially in China. Beijing is the capital and a major metropolis of China. With the rapid economic development and large energy consumption, Beijing has experienced serious air pollution and became one of the hotspots for PM pollution in the world. Guo et al. reported that PM10 concentration was as high as 171.5 ± 91.4 μg m−3 in summer of 2006, Beijing.24 Given the high aerosol loading in China, understanding of the aerosol composition and sources in polluted environment can effectively reduce uncertainties in simulating aerosol loading in the atmosphere and its various effects on a global scale. In this study, primary and secondary organic tracers were measured in the urban and rural sites of Beijing. The tracerbased techniques were used to apportion the primary and secondary sources of PM organic carbon. To the best of our knowledge, our work represents the first comprehensive study on both primary and secondary sources in China. The factors influencing SOA formation (e.g., temperature, oxidants, and particle acidity) and different aging processes are discussed in this work. Although the influences of those factors have been studied in chamber experiments,15,25−27 limited studies have been conducted in the ambient air and some factors have showed contrary impacts from place to place.28−30 Our work provides direct evidence of the ambient SOA formation mechanism proposed in the laboratory studies. The results of this study will undoubtedly improve our knowledge of SOA chemical characteristics and formation mechanisms especially in a heavily polluted environment and greatly reduce uncertainties of SOA simulations in global models.

The Teflon filters were extracted by deionized water to measure water-soluble inorganic compounds (Na+, K+, Mg2+, Ca2+, NH4+, NO3−, SO42‑ and Cl−) by ion-chromatograph (DIONEX, ICS-2500). One punch (1.45 cm2) was taken from the second channel quartz filter for EC and OC analysis via a thermal-optical method using a Sunset Laboratory-based instrument. The remaining two quartz filters were then extracted and analyzed to determine the chemical composition of particulate organic matters. Some daytime or nighttime samples were combined to ensure that most of the organic compounds could be detected. Day samples were separated from night samples. The organic species were identified and quantified using authentic standards and internal standards. The details of the analytical procedures have been described previously.31 Briefly, the samples were first spiked with a mixture of ketopinic acid (KPA), 26 deuterated compounds and two carbon isotope (13C)-substituted compounds, and then ultrasonically extracted with dichloromethane/methanol (3:1, v/v) solution at room temperature. The extracts were filtered and concentrated using a rotary vacuum evaporator and further condensed to about 0.5−1 mL under a flow of high purity nitrogen. Each extract was split into two fractions, one of which was derivatized with BSTFA (BSTFA/TMCS, 99:1; Supelco) to convert polar organic compounds into trimethylsilanized derivatives. Both the derivatized and underivatized fractions were analyzed using an Agilent GC-MS system (6890 plus GC-5973N MSD) equipped with an Agilent DB-5MS GC column (30 m × 0.25 mm × 0.25 μm). In total 164 organic compounds were quantified, including 14 SOA tracers derived from isoprene, α-pinene, βcaryophyllene, and toluene. Because pure standards were not available for most SOA tracers, quantification was based on KPA, which was quantified using an external calibration curve and the results were comparable with other studies.20 During the sampling periods of fine PM, hourly VOCs concentrations were measured by an automated GC/MS/FID (Varian CP-3800 and Saturn 2200 MS) system, using dualcolumns and dual-detectors to simultaneously analyze both low- and high-boiling point NMHCs with each injection. The analytical system was an upgraded one described by Chang et al.32 Source Apportionment. A chemical mass balance (EPA, version CMB 8.2) receptor model was employed to apportion the primary sources of OC. Five primary sources were chosen in the model, including vegetative detritus,5 biomass burning,8 coal burning,3 noncatalyzed engines and diesel engines.10 Cooking source was not included in this study, because in most samples the cholesterol concentration, historically used as a tracer for cooking,33 was around the detection limit, indicating that cooking source was an insignificant contributor during the study period. However, if cooking or other sources contributed to OC, they would be apportioned as other OC. The criteria for acceptable fitting results included the square regression coefficient of the regression equation R2 > 0.85, the sum of square residual Chi-square value χ2