Investigating the NO-dependent photooxidation of limonene by OH and O3 using the atmosphere simulation chamber SAPHIR

Yat Sing Pang,Andreas Wahner,Franz Rohrer,Andreas Hofzumahaus,Robert Wegener,Hendrik Fuchs,Astrid Kiendler-Scharr,David Reimer,Martin Kaminski,Xin Li,Anna Novelli,Hans-Peter Dorn,Birger Bohn,Changmin Cho,Sascha Nehr,Philip Carlsson,Ralf Tillmann,Ismail-Hakki Acir,Anna Lutz

doi:10.5194/egusphere-egu21-11130

Abstract

&lt;p&gt;Limonene is the fourth-most abundant monoterpene in the atmosphere, which upon oxidation leads to the formation of secondary organic aerosol (SOA) and thereby influences climate and air quality.&lt;/p&gt;&lt;p&gt;In this study, the oxidation of limonene by OH at different atmospherically relevant NO and HO&lt;sub&gt;2&lt;/sub&gt; levels (NO: 0.1 &amp;#8211; 10 ppb; HO&lt;sub&gt;2&lt;/sub&gt;: 20 ppt) was investigated in simulation experiments in the SAPHIR chamber at Forschungszentrum J&amp;#252;lich. The analysis focuses on comparing measured radical concentrations (RO&lt;sub&gt;2&lt;/sub&gt;, HO&lt;sub&gt;2&lt;/sub&gt;, OH) and OH reactivity (k&lt;sub&gt;OH&lt;/sub&gt;) with modeled values calculated using the Master Chemical Mechanism (MCM) version 3.3.1.&lt;/p&gt;&lt;p&gt;At high and medium NO concentrations, RO&lt;sub&gt;2&lt;/sub&gt; is expected to quickly react with NO. An HO&lt;sub&gt;2&lt;/sub&gt; radical is produced during the process that can be converted back to an OH radical by another reaction with NO. Consistently, for experiments conducted at medium NO levels (~0.5 ppb, RO&lt;sub&gt;2&lt;/sub&gt; lifetime ~10 s), simulated RO&lt;sub&gt;2&lt;/sub&gt;, HO&lt;sub&gt;2&lt;/sub&gt;, and OH agree with observations within the measurement uncertainties, if the OH reactivity of oxidation products is correctly described.&lt;/p&gt;&lt;p&gt;At lower NO concentrations, the regeneration of HO&lt;sub&gt;2&lt;/sub&gt; in the RO&lt;sub&gt;2&lt;/sub&gt; + NO reaction is slow and the reaction of RO&lt;sub&gt;2&lt;/sub&gt; with HO&lt;sub&gt;2&lt;/sub&gt; gains importance in forming peroxides. However, simulation results show a large discrepancy between calculated radical concentrations and measurements at low NO levels (&lt;0.1 ppb, RO&lt;sub&gt;2&lt;/sub&gt; lifetime ~ 100 s). Simulated RO&lt;sub&gt;2&lt;/sub&gt; concentrations are found to be overestimated by a factor of three; simulated HO&lt;sub&gt;2&lt;/sub&gt; concentrations are underestimated by 50 %; simulated OH concentrations are underestimated by about 35%, even if k&lt;sub&gt;OH&lt;/sub&gt; is correctly described. This suggests that there could be additional RO&lt;sub&gt;2&lt;/sub&gt; reaction pathways that regenerate HO&lt;sub&gt;2&lt;/sub&gt; and OH radicals become important, but they are not taken into account in the MCM model.&lt;/p&gt;

Full Text