Formation of secondary organic aerosol from nitrate radical oxidation of phenolic VOCs: Implications for nitration mechanisms and brown carbon formation

Highlights

• NO₃ oxidation of phenolic VOCs forms products other than nitrophenolic compounds.

• SOA isomers are resolved and identified by the IMS-TOF.

• SOA derived from phenolic VOC + NO₃ absorb light strongly.

Abstract

Volatile phenolic derivatives are substantially emitted from biomass burning and produced from photochemistry of atmospheric aromatic volatile organic compounds (VOCs). Oxidation of phenolic VOCs at night by nitrate radicals (NO₃∙) may represent a significant source of secondary organic aerosols (SOA) and brown carbon (BrC) formation in the atmosphere. In this study, NO₃∙ oxidation of five phenolic derivatives, including phenol, catechol, 3-methylcatechol, 4-methylcatechol and guaiacol are investigated in laboratory experiments. The SOA constituents from the NO₃∙ oxidation were analyzed using electrospray ionization ion mobility spectrometry time-of-flight mass spectrometry, which allows for characterization and identification of isomers in the oxidation products. Through these analyses, several classes of nitro-containing products in addition to the well-known nitrophenol compounds were observed, including: (1) the nitrophenol type of products with additional hydroxyl functional groups; (2) non-aromatic/ring-opening nitro-products with lower double bond equivalence; (3) phenol and catechol products from the C₇ phenolic VOCs with carbon-containing substitutions; and (4) nitrated diphenyl ether dimers. The present work indicates that new products from previously unrecognized pathways are formed during NO₃∙ oxidation of phenolic VOCs and may contribute an important portion of the SOA. Some of these products were also observed in ambient aerosols during biomass burning. We suggest that the ubiquity of the nitrophenol type of products in the SOA derived from phenolic VOC + NO₃∙ are responsible for the strong light absorption measured in this study. Therefore, elucidation of these pathways will be critical for understanding the nighttime oxidation and BrC formation mechanisms.

Introduction

Wildfires are becoming increasingly severe globally and biomass burning events have been shown to strongly influence tropospheric chemistry, the climate, and human health (Crutzen and Andreae, 1990; Pope and Dockery, 2006). Namely, volatile organic compounds (VOCs) and nitrogen oxides (NO_x) emitted from biomass burning undergo various oxidation reactions in the troposphere, substantially impacting the tropospheric ozone (O₃) budget (Arnold et al., 2015). Further, secondary organic aerosols (SOA) from these oxidation reactions can affect the Earth's radiative balance by scattering and absorbing solar radiation in the troposphere (Hobbs et al., 1997). Phenolic derivatives have been shown to make up an important portion of the VOC emissions during biomass burning events (Hatch et al., 2015). They are also key intermediates from photooxidation of aromatic hydrocarbons (also abundant from biomass burning emissions) in the atmosphere (Bloss et al., 2005). The major phenolic VOCs have total emission factors ranging from 0.10 to 0.64 g kg⁻¹ from the combustion of fuels native to the western United States (Hatch et al., 2015; Fine et al., 2004; Oros and Simoneit, 2001). Moreover, their SOA formation potential is among the highest due to their relatively large SOA yields (Hatch et al., 2017). These SOA are often found to be light absorbing (i.e., forming brown carbon, BrC), which further enhance their climate impacts (Lin et al., 2018, 2017, 2015).

There have been extensive studies on the oxidation of phenolic VOCs and most of them have focused on hydroxyl radical (∙OH) oxidation (Finewax et al., 2019, 2018; Kroflič et al., 2015; Lauraguais et al., 2016; Nakao et al., 2011; Pang et al., 2019; Pereira et al., 2015; Schwantes et al., 2017; Vidović et al., 2020). In these studies, phenolic VOCs such as phenol, catechol, 3-methylcatechol (3 MC), 4-methylcatechol (4 MC), and guaiacol have been shown to produce SOA in high yields ranging from 25 to 145% from ∙OH-initiated chemistry (Coeur-Tourneur et al., 2010; Finewax et al., 2018; Lauraguais et al., 2014; Nakao et al., 2011; Pereira et al., 2015; Schwantes et al., 2017; Vidović et al., 2020; Yee et al., 2013). The ∙OH-oxidation of phenolic VOCs in the presence of nitrogen oxides (NO_x) were also found to produce large amounts of BrC, due to the substantial formation of nitrophenol products (Bluvshtein et al., 2017; Fleming et al., 2020; Lin et al., 2018, 2017, 2015). These products, such as nitrophenols and nitrocatechols, have been reported from laboratory studies and field measurements (Bluvshtein et al., 2017; Finewax et al., 2018; Fleming et al., 2020; Lin et al., 2018, 2017, 2015). The phenolic VOCs could also be oxidized by nitrate radicals (NO₃∙) which, however, has been understudied. The NO₃∙ oxidation could be an important reaction pathway of phenolic VOCs during nighttime, relevant to both biomass burning events and urban atmosphere (Rollins et al., 2012). BrC formation has also been recently reported during NO₃∙ oxidation of tar aerosols from wood pyrolysis which largely contain phenolic compounds (Li et al., 2020a). Despite their significance, the understanding of NO₃∙ oxidation mechanisms of individual phenolic VOCs and the SOA formation in prior studies have been very limited. Often, nitro-phenolic compounds were the only major products reported in the SOA composition (Bolzacchini et al., 2001; Finewax et al., 2018; Frka et al., 2016; Kitanovski et al., 2014; Vidović et al., 2020, 2018; Yuan et al., 2016). Other products such as those with additional functional groups and oligomers have not been systematically examined, despite that such low-volatility compounds could be important SOA constituents under ambient conditions. It is unclear whether and to what extent these types of products form during NO₃∙ oxidation of phenolic VOCs.

The major SOA constituents of these reactions and the underlying mechanisms remain a challenge to uncover, largely due to the limitations in detection and identification of the particle-phase products. Previously, a variety of analytical techniques have been implemented to characterize the phenolic VOC-derived SOA such as gas chromatography mass spectrometry (GC-MS), liquid chromatography mass spectrometry (LC-MS), chemical ionization mass spectrometry (CIMS), Fourier transform infrared spectroscopy (FTIR), and UV–Vis diode array detectors (Bolzacchini et al., 2001; Finewax et al., 2018; Frka et al., 2016; Kitanovski et al., 2014; Kroflič et al., 2015; Li et al., 2020a; Vidović et al., 2018; Yuan et al., 2016; Zaytsev et al., 2019). However, most of the instruments listed are incapable of unambiguously characterizing the SOA constituents, especially on the isomer level. In this study, we performed NO₃∙ oxidation experiments in a continuous flow stirred tank reactor (CFSTR) of five phenolic VOCs, including phenol, catechol, 3 MC, 4 MC, and guaiacol (structures shown in Fig. 1). We use electrospray ionization (ESI) coupled to an ion mobility spectrometry time-of-flight mass spectrometer (IMS-TOF) for offline analysis of the SOA from these experiments. The IMS-TOF allows for separation, characterization, and sometimes, identification of structural isomers. Additionally, a UV–Vis spectrometer is utilized for analysis of the light absorption of the SOA materials. With these analytical instruments, we comprehensively identify the major products from these reactions and the underlying mechanisms as well as characterize their optical properties.