Quantification of amine functional groups and their influence on OM/OC in the IMPROVE network. (January 2018)

• Record Type: Journal Article
• Authors: Kamruzzaman, Mohammed; Takahama, Satoshi; Dillner, Ann M.
• Is Part Of: Atmospheric environment. Volume 172(2018)
• Page Start: 124
• Page End: 132
• Publication Date: 2018-01
• Subjects: Fourier transform-infrared spectroscopy -- FTIR -- Partial least squares regression -- Particulate matter -- Organic matter -- Organic carbon; Air -- Pollution -- Periodicals; Air -- Pollution -- Meteorological aspects -- Periodicals; 551.51
• DOI: 10.1016/j.atmosenv.2017.10.053
• Languages: English
• ISSNs: 1352-2310

Abstract

Abstract: Recently, we developed a method using FT-IR spectroscopy coupled with partial least squares (PLS) regression to measure the four most abundant organic functional groups, aliphatic C-H, alcohol OH, carboxylic acid OH and carbonyl C=O, in atmospheric particulate matter. These functional groups are summed to estimate organic matter (OM) while the carbon from the functional groups is summed to estimate organic carbon (OC). With this method, OM and OM/OC can be estimated for each sample rather than relying on one assumed value to convert OC measurements to OM. This study continues the development of the FT-IR and PLS method for estimating OM and OM/OC by including the amine functional group. Amines are ubiquitous in the atmosphere and come from motor vehicle exhaust, animal husbandry, biomass burning, and vegetation among other sources. In this study, calibration standards for amines are produced by aerosolizing individual amine compounds and collecting them on PTFE filters using an IMPROVE sampler, thereby mimicking the filter media and collection geometry of ambient standards. The moles of amine functional group on each standard and a narrow range of amine-specific wavenumbers in the FT-IR spectra (wavenumber range 1 550–1 500 cm −1 ) are used to develop a PLS calibration model. The PLS model is validated using three methods: prediction of a set of laboratory standards not included in the model, a peak height analysis and a PLS model with a broader wavenumber range. The model is then applied to the ambient samples collected throughout 2013 from 16 IMPROVE sites in the USA. Urban sites have higher amine concentrations than most rural sites, but amine functional groups account for a lower fraction of OM at urban sites. Amine concentrations, contributions to OM and seasonality vary by site and sample. Amine has a small impact on the annual average OM/OC for urban sites, but for some rural sites including amine in the OM/OC calculations increased OM/OC by 0.1 or more. Highlights: FT-IR spectra collected on ambient particulate matter samples from IMPROVE. PLSR model based on laboratory standards was developed for amine functional group. Organic matter to organic carbon ratios calculated for IMPROVE samples. OM/OC was calculated using amine and C-H, OH, COH and C=O. Functional group contribution to OM/OC varied by sample and site.