Characteristics And Sources Of Dicarboxylic Acids In PM_2.5 In Shanghai

Low–molecular-weight dicarboxylic acids(DCAs) are important components of organic aerosols. Due to their low vapor pressure, high polarity and hygroscopicity,dicarboxylic acids have important impact on the physical and chemical properties of aerosols, and on the health effects of aerosols. Studying the properties of particulate dicarboxylic acids is thus very important to investigate the formation of secondary organic aerosols and haze.In this study, high-volume PM2.5 samplers were used to collect seasonal atmospheric particles at Baoshan of Shanghai from 2013 to 2014, and at Xujiahui,Songjiang, Chongming and Fengxian in July of 2012. The concentrations and seasonal variations of carbonaceous aerosols, inorganic ions and DCAs in the collected particles were analyzed with an thermal/optical carbon analyzer, GC-MS and ion chromatography(IC). The sources and formation ways of DCAs were investigated. Comparison of the IC method and GC-MS method for DCAs was conducted. The following conclusions were reached:1. IC method for DCAs had high efficiency, but can only measure C2-C5 dicarboxylic acids. Though complicated and tedious, GC-MS method for DCAs was sensitive. Solvent of CH3OH/H2O(1:1 v/v) had higher extraction efficiency for DCAs of > C5 than H2 O.2. The average concentrations of OC and EC in PM2.5 at Baoshan were 9.46μg/m3 and 3.61 μg/m3. Higher concentrations of OC and EC were found in winter and spring, while lower in summer and autumn. Good correlation(R2= 0.86) between OC and EC indicated that they were similar sources. Higher OC/EC ratio was in summer than other seasons, and should be caused by the faster formation rate of secondary organics under the higher ambient temperature and stronger solar radiation in summer.3. The total concentrations of water-soluble inorganic ions in PM2.5 at Baoshan in the spring, summer, autumn and winter were 45.12 μg/m3, 20.36 μg/m3, 28.20μg/m3 and 55.89 μg/m3 respectively, and SO42-, NO3-, NH4+ were the dominant species which summed up to 85% of the total inorganic ions. Significant correlations between K+, Cl- and NO3- indicated that biomass burning may be a important source for Cland NO3-.4. The average concentration of DCAs in PM2.5 at Baoshan was 695.7 ng/m3,among-which oxalic acid, averaged at 415.7 ng/m3, was the most predominant individual compound, accounting for 58% of the total DCAs, followed by azelaic acid,succinic acid, malonic acid and glutaric acid. Higher concentrations of DCAs were found in spring and winter than autumn and summer. DCAs accounted for 11% of the mass concentration of SOA, approving the importance of DCAs in SOA.5. The concentrations of DCAs at SAES, SJ, CM and FX were similar and significantly correlated each other, showing the regional character of DCAs in PM2.5in Shanghai. Meanwhile, concentrations of DCAs at the urban site(SAES) were higher than the other sites, indicating the impact of local anthropogenic emission sources.6. The ratio of C3/C4 in summer(1.97) was significantly higher than that in winter(0.73), suggesting the significant contribution of DCAs from photo-chemical oxidation. The average ratio of C6/C9 in PM2.5 at Baoshan was 0.37, without significant seasonal variations, suggesting that kitchen emission and biomass burning were important sources of DCAs in Shanghai. DCAs was significantly correlated with SO42- and NO3-, indicating that DCAs were from similar emission sources or formed via similar ways as SO42- and NO3-, and might be mainly from secondary photo-chemical oxidation. DCAs had also good correlations with K+ and NO3-,primary emission sources like biomass burning and vehicle emission should thus be important contributor to DCAs in PM2.5 in Shanghai.7. Results of PMF analysis showed that DCAs in PM2.5 in Shanghai had a wide range of sources, and the main sources were vehicle emissions(contributed 44% of the total DCAs), secondary photo-chemical reaction(26%), re-suspended dust and long-range transport(25%), whereas biomass combustion had relatively small contribution(5%). More than half of the DCAs of > C6 would be from vehicleemission, while much of the DCAs of < C5 were from secondary formation. Biomass burning had significant contribution to the DCAs of > C6, but inessential to DCAs of< C5.