Effects of NO x and SO 2 on the Secondary Organic Aerosol 1 Formation from Photooxidation of a -pinene and Limonene 2

12 Anthropogenic emissions such as NO x and SO 2 influence the biogenic secondary organic aerosol (SOA) 13 formation, but detailed mechanisms and effects are still elusive. We studied the effects of NO x and SO 2 on the 14 SOA formation from photooxidation of α -pinene and limonene at ambient relevant NO x and SO 2 concentrations 15 (NO x : < 1 ppb to 20 ppb, SO 2 : <0.05 ppb to 15 ppb). In these experiments, monoterpene oxidation was dominated 16 by OH oxidation. We found that SO 2 induced nucleation and enhanced SOA mass formation. NO x strongly 17 suppressed not only new particle formation but also SOA mass yield. However, in the presence of SO 2 which 18 induced high number concentration of particles after oxidation to H 2 SO 4 , the mass yield of SOA at high NO x was 19 comparable to that at low NO x . This indicates that the suppression of SOA yield by NO x was mainly due to the 20 suppressed new particle formation, leading to a lack of particle surface for the organics to condense on. By 21 compensating the suppressing effect on nucleation of NO x , SO 2 also compensated the suppressing effect on SOA 22 yield. Aerosol mass spectrometer data show that increasing NO x enhanced nitrate formation. The majority of the 23 nitrate was organic nitrate (57%-77%), even in low NO x conditions (<~1 ppb). Organic nitrate contributed 7%- 24 26% of total organics assuming a molecular weight of 200 g/mol. SOA from α -pinene photooxidation at high 25 NO x had generally lower hydrogen to carbon ratio (H/C), compared with at low NO x . The NO x dependence of the 26 chemical composition can be attributed to the NO x dependence of the branching ratio of the RO 2 loss reactions, 27 leading to lower fraction of organic hydroperoxide and higher fractions of organic nitrate at high NO x . While 28 NO x suppressed new particle formation and SOA mass formation, SO 2 can compensate such effects, and the 29 combining effect of SO 2 and NO x may have important influence on SOA formation affected by interactions of 30 biogenic volatile organic compounds (VOC) with anthropogenic emissions. yield and aerosol chemical composition 93 were examined. We used ambient relevant NO x and SO 2 concentrations so that the results can shed lights on the 94 mechanisms of interactions of biogenic VOC with anthropogenic emissions in the real atmosphere. by the of SO has implications for SOA formation by


Introduction
lower than the detection limit of NO x analyzer) and "high NO 2 " condition (NO x 800 ppb). Similarly, at low NO x 75 (initial NO <0.3 ppb, α-pinene ~20 ppb), Han et al. (2016) found that the acidity of seed has no significant effect 76 on SOA yield from α-pinene photooxidation. 77 While these studies have provided important insights into the effects of NO x and SO 2 on SOA formation, a 78 number of questions still remain elusive. For example, many studies used very high NO x and SO 2 concentrations 79 (up to several hundreds of ppb), and the effects of NO x and SO 2 at concentrations relevant to ambient 80 anthropogenic-biogenic interactions (sub ppb to several tens of ppb for NO 2 and SO 2 ) are unclear. Moreover, 81 many previous studies on the SOA formation from monoterpene oxidation focus on ozonolysis or do not 82 distinguish the OH oxidation and ozonolysis in photooxidation, and only few studies on OH oxidation have been 83 conducted (Eddingsaas et al., 2012a;Zhao et al., 2015). More importantly, studies that investigated the combined 84 effects of NO x and SO 2 are scarce, although they are often co-emitted from anthropogenic sources. According to 85 previous studies, NO x mainly has a suppressing effect on SOA formation while SO 2 mainly has an enhancing 86 effect. NO x and SO 2 might have counteracting effect or a synergistic effect in SOA formation in the ambient. 87 In this study, we investigated the effects of NO x , SO 2 and their combining effects on SOA formation from the 88 photooxidation of α-pinene and limonene. α-pinene and limonene are two important monoterpenes with high 89 emission rates among monoterpenes (Guenther et al., 2012). OH oxidation dominated over ozonolysis in the 90 monoterpene oxidation in this study as determined by measured OH and O 3 concentrations. The relative 91 contributions of RO 2 loss reactions at low NO x and higher NO x were also quantified using measured HO 2 , RO 2 , 92 and NO concentrations. The effects on new particle formation, SOA yield and aerosol chemical composition 93 were examined. We used ambient relevant NO x and SO 2 concentrations so that the results can shed lights on the 94 mechanisms of interactions of biogenic VOC with anthropogenic emissions in the real atmosphere. 95 2 Experimental 96

Experimental setup and instrumentation 97
The experiments were performed in the SAPHIR chamber (Simulation of Atmospheric PHotochemistry In a large 98 Reaction chamber) at Forschungszentrum Jülich, Germany. The details of the chamber have been described 99 before (Rohrer et al., 2005;Zhao et al., 2015). Briefly, it is a 270 m 3 Teflon chamber using natural sunlight for 100 illumination. It is equipped with a louvre system to switch between light and dark conditions. The physical 101 parameters for chamber running such as temperature and relative humidity were recorded. The solar irradiation 102 was characterized and the photolysis frequency was derived (Bohn et al., 2005;Bohn and Zilken, 2005). 103 Gas and particles phase species were characterized using various instruments. OH, HO 2 and RO 2 concentrations 104 were measured using a laser induced fluorescence (LIF) system with details described by Fuchs  Systems 43i), respectively. More details of these instrumentation are described before (Zhao et al., 2015). 112 The number and size distribution of particles were measured using a condensation particle counter (CPC, 113 TSI, model 3786) and a scanning mobility particle sizer (SMPS, TSI, DMA 3081/CPC 3785). From particle 114 number measurement, the nucleation rate (J 2.5 ) was derived from the number concentration of particles larger than 115 2.5 nm as measured by CPC. Particle chemical composition was measured using a High-Resolution Time-of-116 were also derived. SOA yields were calculated as the ratio of organic aerosol mass formed to the amount of VOC 124 reacted. The concentration of organic aerosol was derived using the total aerosol mass concentration measured by 125 SMPS with a density of 1 g cm -3 (to better compare with previous literature) multiplied by with the mass fraction 126 of organics in total aerosol characterized by AMS. The organic aerosol concentration was corrected for the 127 particle wall loss and dilution loss using the method described in Zhao et al. (2015). The loss of vapor on the wall 128 was not corrected here. 129 In the experiments with added SO 2 , sulfuric acid was formed upon photooxidation. Sulfuric acid was partly 130 neutralized by background ammonia, which was introduced into the chamber mainly due to humidification. The 131 density of the aerosol was derived using the linear mixing of the density of organic aerosol (assuming 1.32 g cm -3 132 from one of our previous studies and the literature ( The SOA formation from α-pinene and limonene photooxidation was investigated at different NO x and SO 2 139 levels. Four types of experiments were done: with neither NO x nor SO 2 added (referred to as "low NO x , low 140 SO 2 "), with only NO x added (~ 20 ppb NO, referred to as "high NO x, low SO 2 "), with only SO 2 added (~15 ppb, 141 referred to as "low NO x , high SO 2 "), and with both NO x and SO 2 added (~20 ppb NO and ~15 ppb SO 2 , referred 142 to as "high NO x , high SO 2 "). For low NO x conditions, background NO concentrations were around 0.05-0.2 ppb, 143 and NO was mainly from the background photolytic process of Teflon chamber wall (Rohrer et al., 2005). For 144 low SO 2 conditions, background SO 2 concentrations were below the detection limit of the SO 2 analyzer (0.05 145 ppb). In some experiments, a lower level of SO 2 (2 ppb, referred to as "moderate SO 2 ") was used to test the effect 146 of SO 2 concentration. An overview of the experiments is shown in roof was opened to start photooxidation. In the experiments with SO 2 , SO 2 was added and the roof was opened to 149 initialize nucleation first and then VOC was added. The particle number concentration caused by SO 2 oxidation 150 typically reached several 10 4 cm -3 (see Fig. 2 high SO 2 cases) and after VOC addition, no further nucleation 151 occurred. The detailed conditions of the experiments are shown in Table S1. The experiments of α-pinene and 152 limonene photooxidation were designed to keep the initial OH reactivity and thus OH loss rate constant so that 153 the OH concentrations of these experiments were more comparable. Therefore, the concentration of limonene 154 was around one-third the concentration of α-pinene due to the higher OH reactivity of limonene. 155 In the photooxidation of VOC, OH and O 3 often co-exist and both contribute to VOC oxidation. In order to 158 study the mechanism, it is helpful to isolate one oxidation pathway from the other. In this study, the reaction rates 159 of OH and ozone with VOC are quantified using measured OH and O 3 concentrations multiplied by rate 160 constants. Typical OH and O 3 concentrations in an experiment were around (1-15)×10 6 molecules cm -3 and 0-50 161 ppb, respectively, depending on the VOC and NO x concentrations added. The relative importance of the reaction 162 of OH and O 3 with monoterpenes in a typical experiment is shown in Fig. S1. The VOC loss was dominated by 163 OH oxidation over ozonolysis. This makes the chemical scheme simple and it is easier to interpret than cases 164 when both OH oxidation and ozonolysis are important. 165

Results and discussion
As mentioned above, RO 2 fate, i.e., the branching of RO 2 loss among different pathways has an important 166 influence on the product distribution and thus on SOA composition, physicochemical properties and yields. RO 2 167 can react with NO, HO 2 , RO 2 , or isomerize. The fate of RO 2 mainly depends on the concentrations of NO, HO 2 168 and RO 2 . Here, the loss rates of RO 2 via different pathways were quantified using the measured HO 2 , NO and 169 S2 and the relative importance of different RO 2 reaction pathways is compared in Fig. 1, which is similar for both 172 α-pinene and limonene oxidation. In the low NO x conditions of this study, RO 2 +NO dominated the RO 2 loss rate 173 in the beginning of an experiment (Fig. 1a) because a trace amount of NO (up to ~0.2 ppb) was formed from the 174 photolysis of HONO produced from a photolytic process on the chamber wall (Rohrer et al., 2005). But later in 175 the experiment, RO 2 +HO 2 contributed a significant fraction (up to ~40 %) to RO 2 loss because of increasing HO 2 176 concentration and decreasing NO concentration. In the high NO x conditions, RO 2 +NO overwhelmingly 177 dominated the RO 2 loss rate (Fig. 1b), and with the decrease of NO in an experiment, the total RO 2 loss rate 178 decreased substantially (Fig. 1b). Since the main products of RO 2 +HO 2 are organic hydroperoxides, more organic 179 hydroperoxide relative to organic nitrate is expected in the low NO x conditions here. The loss rate of RO 2 +RO 2 180 was estimated to be ~10 -4 s -1 using a reaction rate constant of 2.5×10 -13 molecules -1 cm 3 s -1 (Ziemann and 181 Atkinson, 2012). This contribution is negligible compared to other pathways in this study, although the reaction 182 rate constants of RO 2 +RO 2 are highly uncertain and may depend on specific RO 2 (Ziemann and Atkinson, 2012).

Effects of NO x and SO 2 on new particle formation 184
The effects of NO x and SO 2 on new particle formation from α-pinene oxidation are shown in Fig. 2a. In low SO 2 185 conditions, both the total particle number concentration and nucleation rate at high NO x were lower than those at 186 low NO x , indicating NO x suppressed the new particle formation. The suppressing effect of NO x on new particle 187 formation was in agreement with the findings of Wildt et al. (2014). This suppression is considered to be caused 188 by the increased fraction of RO 2 +NO reaction, decreasing the importance of RO 2 +RO 2 permutation reactions. 189 RO 2 +RO 2 reaction products are believed to be involved in the new particle formation ( Similar suppression of new particle formation by NO x and enhancement of new particle formation by 198 SO 2 photooxidation were found for limonene oxidation (Fig. 2b). 199

3.3.1
Effect of NO x 201 Figure 3a shows SOA yield at different NO x for α-pinene oxidation. In order to make different 202 experiments more comparable, the SOA yield is plotted as a function of OH dose instead of reaction time. In low 203 SO 2 conditions, NO x not only suppressed the new particle formation but also suppressed SOA mass yield. 204 Because NO x suppressed new particle formation, the suppression on SOA yield could be attributed to the absence 205 of nucleation and thus the absence of condensational sink or to the decrease of condensable organic materials. We 206 found that when new particle formation was already enhanced by added SO 2 , the SOA yield at high NO x was 207 comparable to that at low NO x and the difference in SOA yield between high NO x and low NO x was much 208 smaller ( Fig. 3a). This finding indicates that NO did not significantly suppress the formation of condensable 209 organic materials, although NO obviously suppressed the formation of products for nucleation. Therefore, we 210 conclude that the suppressing effect of NO x on SOA yield was mainly due to suppressing nucleation, i.e., to the 211 absence of particle surface as condensational sink. 212 For limonene oxidation, similar results of NO x suppressing the particle mass formation have been found 213 in low SO 2 conditions (Fig. 3b). Yet, in high SO 2 conditions, the SOA yield from limonene oxidation at high NO x 214 was still significantly lower than that at low NO x , which is different from the findings for α-pinene SOA. That 215 might be caused by either the larger difference between the volatility of oxidation products formed under 216 different NO x conditions for limonene case compared to α-pinene or by the different ranges of VOC/NO x for α-217 pinene (VOC/NO x = 1 at high NO x , see Table 1)  The different findings in these studies from ours may be attributed to the difference in the VOC oxidation 233 pathways (OH oxidation vs. ozonolysis), VOC and NO x concentration ranges, NO/NO 2 ratio as well as OH 234 concentrations, which all affect SOA yield. Note that even at "high NO x " the NO x concentration in this study was 235 much lower than in many previous studies and the NO x concentration range here was more relevant to the 236 anthropogenic-biogenic interactions in the ambient. 237

238
For both α-pinene and limonene, SO 2 was found to enhance the SOA mass yield, at given NO x levels, 239 especially for the high NO x cases (Fig. 3). The enhancing effect of SO 2 on particle mass formation can be 240 attributed to two reasons. Firstly, SO 2 oxidation induced new particle formation, which provided more surface were acidic with the molar ratio of NH 4 + to SO 4 2around 1.5-1.8. As mentioned above, inducing new particle 247 formation by SO 2 is especially important at high NO x conditions, when nucleation was suppressed by NO x . In 248 addition, we found that the SOA yield in limonene oxidation at a moderate SO 2 level (2 ppb) was comparable to 249 the yield at high SO 2 (15 ppb) when similar particle number concentrations in both cases were formed. Both 250 yields were significantly higher than the yield at low SO 2 (<0.05 ppb, see Fig. S3). This comparison suggests that 251 the effect in enhancing new particle formation by SO 2 seems to be more important. The role of SO 2 on new 252 particle formation is similar to adding seed aerosol on providing particle surface for organics to condense. 253 Artificially added seed aerosol has been shown to enhance SOA formation from α-pinene and β-pinene oxidation 254 anthropogenic-biogenic interactions in the real atmosphere as discussed below when SO 2 and NO x often co-exist 259 in relative high concentrations. 260

3.4
Effects of NO x and SO 2 on SOA chemical composition 261 The effects of NO x and SO 2 on SOA chemical composition were analyzed on the basis of AMS data. We 262 found that NO x enhanced nitrate formation. The ratio of the mass of nitrate to organics was higher at high NO x 263 than at low NO x regardless of the SO 2 level, and similar trends were found for SOA from α-pinene and limonene 264 oxidation (Fig. 4a). Higher nitrate to organics ratios were observed for SOA from limonene at high NO x , which is 265 mainly due to the lower VOC/NO x ratio resulted from the lower concentrations of limonene (7 ppb) compared to 266 α-pinene (20 ppb) (see Table 1 In order to determine the contribution of organic nitrate to total organics, we estimated the molecular Moreover, we found that the contribution of organic nitrate to total organics (calculated using a 290 molecular weight of 200 g/mol for organic nitrate) was higher at high NO x (Fig. 4b), although in some 291 experiments the ratios of NO 2 + to NO + were too noisy to derive a reliable fraction of organic nitrate. This result is 292 consistent with the reaction scheme that at high NO x , almost all RO 2 loss was switched to the reaction with NO, 293 which is expected to enhance the organic nitrate formation. Besides organic nitrate, the ratio of nitrogen to carbon 294 atoms (N/C) was also found to be higher at high NO x (Fig. S4) The chemical composition of organic components of SOA in terms of H/C and O/C ratios at different 298 NO x and SO 2 levels was further compared. For SOA from α-pinene photooxidation, in low SO 2 conditions, no 299 significant difference in H/C and O/C was found between SOA formed at low NO x and at high NO x within the 300 experimental uncertainties (Fig. 5). The variability of H/C and O/C at high NO x is large, mainly due to the low 301 particle mass and small particle size. In high SO 2 conditions, SOA formed at high NO x had the higher O/C and 302 lower H/C, which indicates that SOA components had higher oxidation state. The higher O/C at high NO x than at 303 low NO x is partly due to the higher OH dose at high NO x , although even at same OH dose O/C at high NO x was 304 still slightly higher than at low NO x in high SO 2 conditions. 305 For the SOA formed from limonene photooxidation, no significant difference in the H/C and O/C was 306 found between different NO x and SO 2 conditions (Fig. S5), which is partly due to the low signal resulting from 307 low particle mass and small particle size in high NO x conditions. 308 Due to the high uncertainties for some of the H/C and O/C data, the chemical composition was further 309 analyzed using f 44 and f 43 since f 44 and f 43 are less noisy (Fig. 6). For both α-pinene and limonene, SOA formed at 310 high NO x generally has lower f 43 . Because f 43 generally correlates with H/C in organic aerosol , 311 lower f 43 is indicative of lower H/C, which is consistent with the lower H/C at high NO x observed for SOA from 312 α-pinene oxidation in presence of high concentrations of SO 2 (Fig. 5). The lower f 43 at high NO x was evidenced 313 in the oxidation of α-pinene based on the data in a previous study (Chhabra et al., 2011). The lower H/C and f 43 314 are likely to be related to the reaction pathways. According to the reaction mechanism mentioned above, at low 315 NO x a significant fraction of RO 2 reacted with HO 2 forming hydroperoxide, while at high NO x almost all RO 2 316 reacted with NO forming organic nitrate. Compared with organic nitrate, hydroperoxide have higher H/C ratio. 317 The same mechanism also caused higher organic nitrate fraction at high NO x , as discussed above. 318 Detailed mass spectra of SOA were compared, shown in Fig 7. For α-pinene, in high SO 2 conditions, 319 mass spectra of SOA formed at high NO x generally had higher intensity for CHOgt1 family ions, such as CO 2 + 320 (m/z 44), but lower intensity for CH family ions, such as C 2 H 3 + (m/z 15), C 3 H 3 + (m/z 39) (Fig. 7b) than at low 321 NO x . In low SO 2 conditions, such difference is not apparent (Fig. 7a), partly due to the low signal from AMS for 322 SOA formed at high NO x as discussed above. For both the high SO 2 and low SO 2 cases, mass spectra of SOA at 323 high NO x show higher intensity of CHN1 family ions. This is also consistent with the higher N/C ratio shown (C 4 O + ) than at low NO x (Fig. S6). It seems that overall mass spectra of the SOA from limonene formed at high 327 NO x had higher intensity for CH family ions, but lower intensity for CHO1 family ions than at low NO x . Note 328 that the differences in these m/z were based on the average spectra during the whole reaction period and may not 329 reflect the chemical composition at a certain time. was regulated by varying NO x concentrations. We confirmed that NO x suppressed new particle formation. NO x 334 also suppressed SOA mass yield in the absence of SO 2 . The suppression of SOA yield by NO x was mainly due to 335 the suppressed new particle formation, i.e., absence of sufficient particle surfaces for organic vapor to condense 336 on at high NO x . NO x did not significantly suppress the formation of condensable organics from α-pinene Organic nitrate compounds are estimated to contribute 7-26% of the total organics using an average 351 molecular weight of 200 g/mol for organic nitrate compounds and a higher contribution of organic nitrate was 352 found at high NO x . Generally, SOA formed at high NO x has a lower H/C compared to that at low NO x . The 353 higher contribution of organic nitrate to total organics and lower H/C at high NO x than at low NO x is attributed to 354 the reaction of RO 2 with NO, which produced more organic nitrate relative to organic hydroperoxide formed via 355 the reaction of RO 2 with HO 2 . The different chemical composition of SOA between high and low NO     conditions. A: α-pinene, low SO 2 , B: α-pinene, high SO 2 , C: limonene, low SO 2 , D: limonene, high SO 2 . Note 721 that in the low SO 2 , high NO x condition (panel C), the AMS signal of SOA from limonene oxidation was too low 722 to derive reliable information due to the low particle mass concentration and small particle size. Therefore, the 723 data for high NO x in panel C show an experiment with moderate SO 2 (2 ppb) and high NO x instead.  Figure 7. The difference in the mass spectra of organics of SOA from α-pinene photooxidation between high NO x 726 and low NO x conditions (high NO x -low NO x ). SOA was formed at low SO 2 (a) and high SO 2 (b). The different 727 chemical family of high resolution mass peaks are stacked at each unit mass m/z ("gt1" means greater than 1). 728 The mass spectra were normalized to the total organic signals. Note the log scale of y-axis and only the data with 729 absolute values large than 10 -4 are shown.