Deconvolution and Quantification of Hydrocarbon-like and Oxygenated Organic Aerosols Based on Aerosol Mass Spectrometry

详细信息    查看全文

• 作者：Qi Zhang ; M. Rami Alfarra ; Douglas R. Worsnop ; James D. Allan ; Hugh Coe ; Manjula R. Canagaratna ; and Jose L. Jimenez
• 刊名：Environmental Science & Technology
• 出版年：2005
• 出版时间：July 1, 2005
• 年：2005
• 卷：39
• 期：13
• 页码：4938 - 4952
• 全文大小：1597K
• 年卷期：v.39,no.13(July 1, 2005)
• ISSN：1520-5851

文摘

A new technique has been developed to deconvolve andquantify the mass concentrations of hydrocarbon-like andoxygenated organic aerosols (HOA and OOA) usinghighly time-resolved organic mass spectra obtained withan Aerodyne Aerosol Mass Spectrometer (AMS). Thistechnique involves a series of multivariate linear regressionsthat use mass-to-charge ratios (m/z's) 57 (mostly C₄H₉⁺)and 44 (mostly CO₂⁺)-the identified AMS mass spectraltracers for HOA and OOA, respectively-as the initial principalcomponents. Two algorithms have been developed:algorithm 1 is based solely on m/z 44 and m/z 57, andalgorithm 2 is an iterative procedure expanded from algorithm1. This technique was applied to the AMS organicaerosol data acquired at the EPA Pittsburgh Supersiteduring September 2002. The reconstructed organicconcentrations (= HOA + OOA) agree well with themeasured values (r² = 0.997, slope = 0.998), and thereconstructed organic data matrix (size = 3199 time steps× 300 m/z's) explains 99% of the variance in the measuredtime series. In addition, the extracted mass spectrum of HOAshows high similarity to those of diesel exhaust, lubricatingoil, and freshly emitted traffic aerosols observed inurban areas, while the spectrum of OOA closely resemblesthose of aged organic aerosols sampled in rural areasand also shows similarity with the spectrum of fulvic acid-a humic-like substance that is ubiquitous in the environmentand has previously been used as an analogue to representpolyacid components found in highly processed and oxidizedatmospheric organic aerosols. There is evidence for thepresence of a third component, although its contribution tothe total organic signal appears to be small in thisstudy. The most important result is that m/z 44 and m/z 57are reliable AMS mass spectral "markers" that providethe "first guess" for algorithm 2 which allows the quantitativedescription of the organic aerosol concentration andmass spectra over a period of 16 days in a major urbanarea and allows the extraction of mass spectra of OOA andHOA that can be interpreted chemically. These findingsindicate the potential of performing organic sourceapportionment on the basis of total particle mass, ratherthan on the basis of organic tracer compounds that contributea small fraction of this mass.