PNNL

Abstract

As particulate matter abatement strategies reduce anthropogenic nitrogen oxide (NOx) emissions, the chemical regime of an urban atmosphere may change to one in which autoxidation becomes increasingly important. Developed countries such as the United States have already experienced large declines in nitrogen oxides because of controls on emissions, and the possibility for increased aerosol exists as autoxidation results in highly oxidized molecules (HOM) which efficiently form particulate matter. Using a-pinene reaction with hydroxyl radicals as an archetype system, a molecular representation of organic aerosol was created and evaluated against laboratory experiments. The applica-tion of parameters constrained by these experiments to the ambient atmosphere indicated autoxidation is becoming more competitive in urban areas of the United States while at the same time HOM production rates are declining due to decreases in oxidant abundance. For locations such as Guangzhou, China, where emissions of NOx generally increased from the 1990s to late 2000s, HOM production was also predicted to increase. These results demonstrate that anthro-pogenic emissions can enhance HOM and the resulting biogenic organic aerosol formation through oxidant abundance. Ambient particulate matter is responsible for a significant fraction of the global burden of disease (Lim et al., 2012) and affects climate (Myhre et al., 2013). The organic portion of ambient aerosol largely forms due to the condensation of low vapor pressure (Jimenez et al., 2009) or highly soluble (Saxena and Hildemann, 1996) gases in the atmosphere . Oxidation of monoterpenes from vegetation, results in significant formation of particle mass (Pandis et al., 1991), and is likely the dominant source of organic aerosol in the southeastern United States (Zhang et al., 2018; Xu et al., 2018). While monoterpenes are included as a secondary organic aerosol (SOA) source in most chemical transport models, many current regional to global scale model parameterizations (Pye et al., 2010, 2017; Koo et al., 2014) lack a mechanistic dependence of monoterpene SOA on NOx and oxidant identity (e.g. OH vs. ozone). This limits the ability of models to predict how changes in emissions will affect ambient concentrations and new particle formation events. Building a mechanistic understanding of formation pathways is critical for being able to determine historic drivers of ambient particulate matter as well as inform predictions of future levels