Size resolved composition of diesel exhaust nanoparticles

Dataset

Description

Knowledge of physico-chemical characteristics of particle emissions from combustion engines is essential for various modelling purposes and environmental analysis. It is of particular interest to obtain emission factors of intermediate-volatility organic compounds (IVOC) and semi-volatile organic compounds (SVOC) which have not been comprehensively reported in the literature due to the limitations of characterisation methods. In the current study, a multi-stage Nano impactor and the two-dimensional gas chromatography (GC×GC) mass spectrometry (MS) technique were used to comprehensively characterise size fractionated particle phase emissions from a light-duty diesel engine based on the particle size, compound groups and carbon number. The number size distributions of particles between 1.2-1000 nm were also investigated. Exhaust gas samples were taken before a diesel oxidation catalyst (DOC), after the DOC and after the DOC combined with a catalysed diesel particulate filter (DPF). In samples taken before the DOC (engine-out), the total particulate SVOC (IVOC+SVOC) emission factor was approximately 105 mg/kgf (which was ~49% of the total particle mass) and the peak concentration of different classes of SVOC was found in the particle size bins close to 100 nm where most of the total particle mass was found. Alkanes, with maximum abundance at C24, were the most dominant class of SVOC in samples taken before and after the aftertreatment devices. Total particulate SVOC emissions were removed with ~75% efficiency using the DOC and by ~92% using the DOC+DPF. Alkanes, cycloalkanes, bicyclics and monoaromatics were all removed by >90% using the DOC+DPF; however, oxygenates were removed by only ~76% presumably due to the oxidation of different species within the aftertreatment system and reappearance as oxygenates. A high concentration of particles was measured in the sub-2.5 nm range. These particles were efficiently removed by the DOC+DPF due to both the loss of semi-volatile constituents and physical filtration.
Date made available2019
PublisherUniversity of Birmingham

Cite this