Chemical characterization and multivariate analysis of atmospheric 7 PM 2 . 5 particles 8 9

Citation details: 1 2 Ravindra, K., Stranger, M., Van Grieken, R., 2008. Chemical characterization and 3 multivariate analysis of atmospheric PM2.5 particles. Journal of Atmospheric Chemistry, 4 59, 199-218. DOI: 10.1007/s10874-008-9102-5 5 6 Chemical characterization and multivariate analysis of atmospheric 7 PM2.5 particles 8 9 Khaiwal Ravindra 1,2* , Marianne Stranger 2,3 and Rene Van Grieken 2 10 11 1 Centre for Atmospheric and Instrumentation Research (CAIR), University of 12 Hertfordshire, Hatfield, AL10 9AB, United Kingdom 13 2 Micro and Trace Analysis Centre, Department of Chemistry, University of Antwerp, 14 Universiteitsplein 1, B-2610 Antwerp, Belgium 15 3 Higher Institute for Product Development, University College Antwerp, 16 Ambtmanstraat 1, B-2000 Antwerp, Belgium 17 18 Abstract 19 20 The new European Council Directive (PE-CONS 3696/07) frames the inhalable 21 (PM10) and fine particles (PM2.5) on priority to chemically characterize these fractions in 22 order to understand their possible relation with health effects. Considering this, PM2.5 was 23 collected during four different seasons to evaluate the relative abundance of bulk 24 elements (Cl, S, Si, Al, Br, Cu, Fe, Ti, Ca, K, Pb, Zn, Ni, Mn, Cr and V) and water 25 soluble ions (F , Cl , NO2 , NO3 , SO4 2, Na + , NH4 + , Ca 2+ and Mg 2+ ) over Menen, a 26 Belgian city near the French border. The air quality over Menen is influenced by 27 industrialized regions on both sides of the border. The most abundant ionic species were 28 NO3 , SO4 2and NH4 + , and they showed distinct seasonal variation. The elevated levels of 29 NO3 during spring and summer were found to be related to the larger availability of the 30 NOx precursor. The various elemental species analyzed were distinguished into crustal 31 and anthropogenic source categories. The dominating elements were S and Cl in the 32 PM2.5 particles. The anthropogenic fraction (e.g. Zn, Pb, and Cu) shows a more scattered 33 abundance. Furthermore, the ions and elemental data were also processed using principal 34 component analysis and cluster analysis to identify their sources and chemistry. These 35 approach identifies anthropogenic (traffic and industrial) emissions as a major source for 36 fine particles. The variations in the natural / anthropogenic fractions of PM2.5 were also 37 found to be a function of meteorological conditions as well as of long-range transport of 38 air masses from the industrialized regions of the continent. 39 40