Investigating monoterpene ozonolysis reactions in the mobile DouAir atmospheric simulation chamber: field and laboratory experiments

The chemistry of the atmosphere is usually studied using three different approaches, i.e. field measurements, laboratory studies and chemical model calculations. All three are complementary and powerful means to investigate chemical transformations of pollutants and improve our understanding of the atmosphere. Atmospheric simulation chambers are one of the most direct and critical approaches to mimic and examine chemical transformations under controlled experimental conditions. In combination with box model simulations, they allow assessment of the accuracy of chemical mechanisms implemented in atmospheric models.

During the CERVOLAND field campaign (Characterisation of Emissions and Reactivity of Volatile Organic compounds in the LANDes forest) we deployed a new mobile atmospheric chamber (DouAir) to probe the oxidation of biogenic volatile organic compounds (BVOCs) in real air masses. Biogenic compounds emitted by the surrounding forest (mainly pines - (Maritime pine, Pinus pinaster Ait) were trapped in DouAir and their transformations were probed using state-of-the-art online instrumentation, including PTR-ToF-MS (VOCs), PERCA (peroxy radicals), O₃ and NO_x analysers, and SMPS (aerosols).

The objectives of the present study were to (1) reproduce in the laboratory selected field experiments performed during CERVOLAND, the chemical composition of the air mass being simplified, and (2) compare both the field and laboratory results to 0-D box model simulations using the Master Chemical Mechanisms (MCM). Comparing field observations, laboratory experiments and model simulations provides a critical test of our understanding of atmospheric oxidation processes involving biogenic compounds.

Here, we present ozonolysis experiments of primary biogenic VOCs (mainly monoterpenes) under dark conditions. Initial conditions used for the laboratory experiments were derived from reactant concentrations trapped in DouAir during CERVOLAND. The results show the capability of the model to reproduce oxidation rates of primary VOCs within uncertainty, although the model considerably overestimates measured peroxy radical concentrations. The addition of rapid self- and cross-reactions of monoterpene-derived peroxy radicals in the MCM improves the agreement with the measured peroxy radical concentrations.