Solutions and Challenges in Focusing Trap and Spectral Deconvolution for Characterization of Odor Active VOCs in Hot Mix‐Asphalt

ABSTRACT This work shows the measurement of trace of odor‐active compounds in air sampler collect from channeled emissions of Asphalt Mixture Plants by Nalophan bags, including formaldehyde, hydrocarbons, and volatile sulfur compounds by thermal desorption–gas chromatography–mass spectrometry (MS) equipped with an automated, three‐stage preconcentration and water‐management mechanism without liquid cryogen thanks to a dedicated focusing traps. The success of this application has benefited from computational workflow in Non‐Targeted Analysis, applied to the complex raw MS data to obtain qualitative and quantitative matrix information through deconvolution by MassHunter Unknowns Analysis software. Deconvolution is one of the most challenging steps in GC‐MS data processing reconstructing a pure mass spectrum for each component, permitting the identification of co‐eluted volatile organic compounds (VOCs) by both matching the electronic ionization (EI) mass spectra and the Linear Temperature‐Programmed Retention Index with reference values provided by spectral library. The described approach permitted a semi‐untargeted screening of VOCs in complex matrix by single quadrupole‐MS. Several oxygenated VOCs were found with relatively low abundances under the unresolved complex mixtures produced by co‐elution in the range between heptane and dodecane. Samples collected in typical asphalt production conditions produced between 2000 and 4500 deconvolution results, with 300 to 500 library hits considering a minimum match factor of 70/100. Very VOCs resulted chromatographically resolved, while early eluting peaks were qualified by standard injection and single‐ion monitoring acquisition, since EI spectra of such light molecules are characterized by an insufficient number of peaks to achieve a deconvolution under the predominant early elution of carbon dioxide.