Evaluation of the New Capture Vaporizer for Aerosol Mass Spectrometers (AMS):...

Hu, W., D. A. Day, P. Campuzano-Jost, B. Nault, T. Park, T. Lee, P. Croteau, M. R. Canagaratna, J. T. Jayne, D. Worsnop, and J. Jimenez-Palacios (2018), Evaluation of the New Capture Vaporizer for Aerosol Mass Spectrometers (AMS): Elemental Composition and Source Apportionment of Organic Aerosols (OA), Anal. Chem., 2, 410−421, doi:10.1021/acsearthspacechem.8b00002.
Abstract: 

To reduce the quantification uncertainty of commercial aerosol mass spectrometers (AMS) and aerosol chemical speciation monitors (ACSM), a new capture vaporizer (CV) was recently built to replace the standard vaporizer (SV). A collection efficiency (CE) ∼ 1 in the CV AMS/ACSM has been demonstrated for ambient aerosols, but the CV also leads to increased thermal decomposition of the analytes because of longer residence time and vaporizer surface contact. This study reports on the performance of the CV for analyzing organic aerosol (OA) elemental composition and source apportionment, using both HR-ToF-AMS and ACSM for the first time. The methodology for obtaining elemental ratios from AMS spectra is updated to account for differences in OA fragmentation between the CV and SV. An artifact CO+ signal is observed for some chemically reduced laboratory compounds. If that signal is included in elemental analysis, the O:C is substantial overestimated, while accurate results are observed if the anomalous CO+ is ignored. The estimated uncertainty of O:C (H:C) of standard organic species with the CV-AMS is 23% (18%). Consistent time series of positive matrix factorization (PMF) factors and their fractions of total OA were found across the CV and SV in the three very different ambient data sets ranging from biogenic- to anthropogenic-dominated, indicating limited loss of source determination information despite the increased fragmentation. In some cases, bootstrap analysis of CV data sets shows higher uncertainty for the apportionment of total oxygenated OA (OOA) into different subtypes than that of the SV data set, which is potentially due to lower signal-to-noise at larger m/z from the increased thermal decomposition in the CV.

PDF of Publication: 
Download from publisher's website.
Research Program: 
Tropospheric Composition Program (TCP)
Mission: 
SOAS
KORUS-AQ