Literature review on emissions of semi- and intermediate volatile organic compounds and formation of organic aerosols with focus on the refinery sector
Organic aerosols (OA) are an important component of ambient particulate matter (PM) but their origin and formation is not well understood. Traditionally, OA have been described by two classes: a) primary organic aerosol (POA) and b) secondary organic aerosol (SOA). POA refers to non-volatile organic compounds (VOC) that are emitted directly from the source in the particle phase and do not undergo any atmospheric process other than dilution and deposition, while SOA include all OA formed by oxidation and subsequent condensation of gaseous precursors. Recent studies have shown that this classification of OA is too simple because the semi-volatile nature of emitted POA is not taken into account and that other organic compounds such as condensable particulate matter (CPM) and semi- and intermediate-volatile organic compounds (S/IVOC)1 are also precursors of OA. The organic fraction of CPM contributes to POA (since it forms aerosol without any chemical reaction) but not to SOA. S/IVOC, on the other hand, can play a significant role in the formation of SOA in the atmosphere. The definitions of CPM and S/IVOC partly overlap, since both are defined by their volatility at a certain temperature.
This report reviews the main literature, and provides insights regarding the sources and emissions of S/IVOCs and CPM and the impact of the former on SOA concentrations. It also provides high-level estimations of S/IVOC emissions from the European refinery sector and reviews their importance in comparison to those from other sectors.
The literature review has shown that there are no published S/IVOC emissions measurements from refineries in Europe. However, S/IVOC emissions can be estimated using indicator substances, such as polycyclic aromatic hydrocarbons (PAHs). An indicative estimate of total S/IVOC emissions from all refineries in the EU27+ is 100 t S/IVOC per year, with a range of 10 to 1000 t/year. This is, by comparison, three orders of magnitude lower than the estimates for wood burning in Europe. In addition, the results from modelling simulations using atmospheric chemical and transport models (CTMs) have shown that S/IVOC emissions from refineries contribute only a minor fraction to secondary OA concentrations over Europe. Any uncertainties of modelling OA concentrations could be reduced through undertaking refinery S/IVOC measurements that can help to develop representative sectoral emission inventories and determine the refining contribution to OA more accurately.