Keith Bein - Identification of sources of atmospheric PM at the Pittsburgh Supersite. Part I: Single particle analysis and filter-based Positive Matrix Factorization

Version 1

      Publication Details (including relevant citation   information):

      Pekney, N.J., Bein, K.J., Davidson, C.I., Wexler, A.S., and M.V.   Johnston (2006),

      Identification of sources of atmospheric PM at the Pittsburgh   Supersite. Part I:

      Single particle analysis and filter-based Positive Matrix   Factorization,

      Atmospheric   Environment, 40(S2), S411-S423.

      Abstract:

      During the Pittsburgh Air Quality Study (PAQS), July   2001-September 2002, three collocated instruments analyzed the   composition of ambient particulate matter (PM): (1) A single   particle mass spectrometer, RSMS-3, was deployed to obtain   high-temporal-resolution measurements of single particle size   (> 1.1 μm) and composition which were correlated with   meteorological data to identify sources; (2) PM2.5 and   PM10 were collected on cellulose filters using   high-volume (hi-vol) samplers, followed by microwave-assisted   digestion and analysis by Inductively Coupled Plasma – Mass   Spectrometry (ICP-MS). Positive Matrix Factorization (PMF) was   used to identify possible source categories; and (3) a Micro   Orifice-Uniform Deposit Impactor (MOUDI) obtained   size-distributed samples of PM.  Several days of MOUDI   filters were selected for microwave-assisted digestion and   analysis by ICP-MS.

      In this paper, sources identified using the single particle data   were compared to the PMF results for the hi-vol/ICP-MS   data.  The strengths of each method were combined to   hypothesize the most likely sources of various elements in   ambient PM in Pittsburgh.  In the final results, Mo and Cr   are attributed to local specialty steel facilities; Fe, Mn, Zn,   and K are attributed to a steel mill SE of the monitoring   station; internally mixed Pb-containing particles are attributed   to a major source to the NW; and Ga is attributed to coal   combustion sources to the NW. There is a notable lack of oil   combustion sources.

      The MOUDI data were used to resolve discrepancies between the   single particle and hi-vol/ICP-MS data concerning the detection   of Ti and Se.  The hi vol data showed appreciable Ti and Se   mass, but RSMS-3 was unable to detect significant numbers of   Ti-containing particles because of their large size, while we   hypothesize that the volatility of Se caused it to be distributed   more evenly over all emitted particles such that the amount of Se   in any individual particle is below the limit of detection.

      Address (URL): http://