Formation of condensable organic vapors from anthropogenic and 1 biogenic VOCs is strongly perturbed by NO x in eastern China 2

Oxygenated organic molecules (OOMs) are the crucial intermediates linking volatile organic compounds (VOCs) to secondary organic aerosol (SOA) in the atmosphere, but comprehensive understandings of the characteristics of OOMs and their formations 29 from VOCs are still missing. Ambient observations of OOMs using recently developed mass of the total signals are by nitrogen- containing OOMs, as nitrates formed through radicals 24% of total signals, significant presence multiple C 5 H 10 O 8 N and C 5 H 9 O 10 N 3 Additionally, the distribution of OOMs concentration on carbon number confirm their 45 precursors driven by AVOCs mixed with enhanced BVOCs during summer. Our results highlight the decisive role of NO x on OOMs formation in densely populated areas, and encourage more studies on the dramatic interactions between anthropogenic and biogenic emissions.


145
The nitrate CI-APi-TOF (Aerodyne Research Inc. and Tofwerk AG), combining a were generated in the sheath flow by exposing air-containing nitric acid to a 153 PhotoIonizer X-Ray (Model L9491, Hamamatsu, Japan). Detailed description of the 154 instrument has been described elsewhere (Junninen et al., 2010;Jokinen et al., 2012). 155 The data were acquired at 1 Hz time resolution and analyzed with a tofTools package 156 (version 6.11) based on MATLAB (Math Works Inc.). Due to the diversity and unknown 157 molecular structures of oxygenated organic compounds, standards for OOMs 158 measurable by the nitrate CI-APi-TOF are still lacking. Like other studies have done 159 (Kirkby et al., 2016;Trostl et al., 2016;Stolzenburg et al., 2018), an empirical method  is comparable to that of sulfuric acid with nitrate clusters (Ehn et al., 2014;Hyttinen et 175 al., 2015), yet the collision frequency of some moderately oxygenated molecules with 176 nitrate clusters is relatively slower. Therefore, calibration by this method leads to a 177 lower limit estimate of OOMs concentrations (Ehn et al., 2014;Trostl et al., 2016), but 178 the accurate quantification of OOMs is not the main concern of this study and the errors  volatile organic compounds (VOCs) (Wayne et al., 1991). The elevated mixing level of 260 total aromatic hydrocarbons is one of the main characteristics of the atmosphere in 261 densely populated areas, in addition to which there should be many alkanes and alkenes 262 which cannot be observed by PTR-ToF-MS (Fu et al., 2013;Xu et al., 2017). In the 263 daytime with strong photochemical reaction (J(O 1 D)> 1×10 -5 s -1 ), we instead observed 264 higher concentrations of isoprene than total aromatics ( Fig.S1(a)). The complex 265 mixtures of anthropogenic and biogenic VOCs can be oxidized through a variety of 266 pathways to produce OOMs, of which some low volatile components will condense 267 into particles, forming organic aerosol. The concentrations of OOMs with mass-to-268 charge ratio (m/z) below 360 Th are usually higher than 10 6 molecules cm -3 , and some 269 can even reach up to 10 7 -10 8 molecules cm -3 . Clustered peaks on the spectra of OOMs Th.

281
The binPMF analysis was performed to characterize the sources or processes of OOMs. not be the optimal PMF solution, it still separates a lot of useful information. We also 292 stress that the urban OOMs mix is unlikely to be a perfect combination of independent, 293 unchanging factors, which is an underlying assumption in the PMF algorithm. As such, 294 there will be no solution which is complete and perfect, but we chose a solution from 295 which we were able to provide us with interesting insights. Details of the PMF 296 diagnostics is provided in section S2 in the supplement ( Fig. S2-S6). For the convenience of discussions, we have grouped these factors based on shared 298 characteristics of certain factors. The following daytime factors are characterized by C6-C9 OOMs ( Fig. 2  The effective DBE of this factor is the largest among all factors (Table 1), with main 326 signals come from compounds with DBE > 2 ( Fig. 2(b)) and consistent with the nature 327 of the oxidation products of aromatics ( Fig. 3(a)). Combined with the correlation with 328 the production rates of OH-initiated primary peroxy radicals (RO2) from aromatics 329 calculated by Eq. (4) (Fig. 4), this factor is supposedly dominated by aromatics-derived 330 OOMs (Aro-OOMs). The Aro-OOMs factor increases from 5:00 LT with a maximum 331 at 10:00 LT and a sub peak around 16:00 LT (Fig. 3(e)), following the diurnal variations 332 of P RO 2 of C7-C10 aromatics ( Fig. 4(b-d)) but poorly correlated with P RO 2 of benzene 333 ( Fig. 4(a)). Furthermore, OOMs with 8 carbon atoms have the highest signal in this 334 factor ( Fig. 2(a) aromatics (P MT−RO 2 > P C 10 Aro−RO 2 ), but aromatics play a more important role in total 341 in this factor since they provide more RO2 in the urban atmosphere ( Fig. 3(f)).

357
The main molecules of the Aro-OOMs factor are summarized in Table S2. The CxH2x- This factor is named due to good correlation with temperature ( Fig. 5), and shows 388 maximum intensity in the afternoon around 15:00 ( Fig. 3(e)). The Temp-related factor 389 is the only one dominated by non-nitrogenous organics ( Fig. 3(b) and (d)), and has the 390 highest effective OSc (Table 1) among all the factors. The CxH2x-4O5 (x=5-11, 391 summarized in Table S3), CxH2x-2O5 (x=5-10), CxH2x-6O5 (x=5-11), and CxH2x-4O6 392 (x=5-10) series are possibly products from RO2 terminated by HO2 (R2a), or closed-393 shell products from RO in reactions R3a or R3b. Temperature starts to rise at 6:00 LT 394 ( Fig. 12(b)), but this factor does not accumulate significantly until after about 10:00 LT  APi-TOF lower. Thus, the total concentration of the Temp-related factor in the gas and 412 aerosol phases was calculated based on gas-particle equilibrium (section S5 in 413 the supplement), and was found to be still temperature dependent (Fig. S7 This factor is dominated by organic nitrates (Fig. 3(c) and (d)  and nor in the forest or rural environments (Yan et al., 2016;Massoli et al., 2018). A 444 reasonable assumption is that these saturated or nearly saturated compounds are the 445 products of aliphatics (including alkanes, alkenes, aliphatic alcohol, etc.) during their 446 oxidation affected intensively by NOx in the urban atmosphere. The Aliph-OOMs factor 447 has a broad afternoon peak lasting from 14:00 to 19:00 LT (Fig. 3(e)), suggesting that 448 the formation of multi-nitrate requires enough OH exposure time.

450
Considering a simple scenario of alkane photo-oxidation under high NOx conditions: 451 the RO2 generated from OH attack is completely terminated by NO ( Fig. 6(a)). The 452 chain-retaining products are CnH2nO (one more carbonyl group than the precursor) and 453 CnH2n+1O3N (one more nitrate group than the precursor), and the re-oxidation of these 454 products is a repetition of the above process which is defined as the basic reaction 455 scheme. The multiple (1st to 3rd) generation products of alkanes summarized in Fig.   456 6(b) are regarded as reference compounds, which we compare OOMs with to 457 investigate other mechanisms that differ from those shown in Fig. 6 As showed in Fig. 6(c), aliphatic OOMs in this factor are mainly the third-generation 472 products followed by the second-generation products, and both of which have one or 473 two oxygen-containing functional groups in addition to the carbonyls and nitrates. It 474 should be noted that the first-generation ( Fig. 6(a)) and basic products ( Fig. 6(b)) here 475 are underestimated due to the low sensitivity of nitrate CI-APi-TOF to these compounds.

476
The multifunctional products of aliphatics are condensable to form SOA (correlation 477 coefficients with SOA showed in Fig. 5). Recent work has showed that autoxidation is 478 more common than previously thought (Wang et al., 2021), and more studies are needed 479 to explore the oxidation mechanisms of anthropogenic aliphatics and to evaluate their The following factors are characterized by C5 OOMs (Fig. 2(a)), of which an isoprene This factor is defined based on its correlation with J(O 1 D) (Fig. 5), having an apparent 503 diurnal cycle with a peak at 12:00 LT (Fig. 7(e)). The major peak of the Photo-related 504 factor is C5H10O8N2 (Fig. 7(a)), most probably generating from double OH attack 505 proceed with double RO2+NO termination (Jenkin et al., 2015). C5H10O8N2 can be also 506 produced in NO3+ isoprene system (Ng et al., 2008;Zhao et al., 2020), whereas in this 507 study, the nocturnal C5H10O8N2 is principally from the Isop-OOMs factor ( Fig. 8(b)) 508 which will be discussed later. Other peaks with nC≤5, like C5H7O7N, C4H7O6N, 509 C5H9O6N, are also likely to be the isoprene products. The total signal of compounds 510 with nC > 5 is not low, although their respective proportions are not as prominent as C5 511 species ( Fig. 7(d)), implying the contribution of other precursors together with isoprene.

512
In addition, the relationship of this factor with isoprene and J(O 1 D) together (Fig. 5) 513 reveals the effect of light-dependent emission of isoprene on it.

515
Ox & SOA-related factor 516 517 The atmospheric oxidation of VOCs produces low-volatile compounds, forming SOA 518 through gas-particle partitioning, and concurrently promotes ozone formation The mass spectra of the Isop-OOMs factor, as its name implies, is exclusively 551 contributed by isoprene-derived compounds (Fig. 7(c)). C5H10N2O8 contributes about 552 30% of the intensity of this factor, and the dominance of C5H10N2O8 was also found in 553 the isoprene nitrates type I factor in Centreville (Massoli et al., 2018). In addition to 554 multi-nitrates (C5H10O7-8N2, C5H8O6-9N2, and C5H9O10N3 summarized in Table. S6), 555 several mononitrate series (C4H7O5-7N, C5H9O4-9N, C5H7O5-8N, and C5H11O5-6N) of 556 this factor are also abundant in the isoprene nitrates type II factor in Centreville 557 (Massoli et al., 2018). Many of isoprene nitrates here have been specially investigated 558 in our previous observations in the YRD (Xu et al., 2021), and have been discovered in 559 other filed measurements (Lee et al., 2016;Massoli et al., 2018) and in many 560 laboratories (Ng et al., 2008;Lambe et al., 2017). Generally, these compounds are 561 second-and third-generation OH oxidation products of isoprene under high-NOx 562 conditions (Wennberg et al., 2018).

564
The diurnal pattern of the Isop-OOMs factor is relatively unclear (Fig. 7(e)), with 565 obvious differences between mean and median values usually caused by plume events.

566
This indicates that isoprene chemistry, usually varying evidently from day (OH-567 initiated) to night (NO3-initiated), is not the driver of this factor. This factor correlates 568 positively with MVK / MACR and SOA (r>0.50, showed in Fig. 5), but not with 569 isoprene and OH. It seems that these isoprene OOMs are produced elsewhere and then 570 transported due to their longer lifetime determined by their relatively high volatility 571 ( Table 1). The Isop-OOMs factor in the continental air masses are more intensive than 572 those in the coastal and YRD air masses (Fig. S8), consistent with the spatial 573 distribution of isoprene emissions (Sindelarova et al., 2014). An archetypal episode 574 affected by continental air masses (August 13 to August 17, 2019) is showed in Fig. 8.

575
During this period, C5H9O10N3 was almost entirely transported, while C5H10O8N2 has 576 strong in situ photochemical generation, in addition to the source of transport. The following nighttime factors are characterized by C10 OOMs (Fig. 2(a)), which are 589 identified as the oxidation products of monoterpenes. Except for the BVOCs-OOMs I 590 factor ( Fig. 9(a)), the contribution of isoprene-derived OOMs was much lower in these 591 factors. Compared to the above isoprene-related factors, C5H10O8N2 and C5H9O10N3 592 was no longer significantly present in the following factors.

594
BVOCs-OOMs I factor 595 596 The first nighttime factor has its maximum concentration at around 20:00 LT, and 597 decreases to very low value during the day. It is moderately correlated with the 598 production rate of NO3 radical (P NO 3 derived from Eq. 6) at night, and reaches high 599 intensity only under conditions of NO below 1 ppb ( Fig. 10(a)), indicating a chemical process of NO3 radical. The concentration of this factor is mainly from C5 peaks, 601 followed by C6-C10 peaks ( Fig. 9(d)), about 80% of which are ONs (Fig. 2(c)), 602 designating the oxidations of isoprene and monoterpenes by NO3 (BVOCs-OOMs I).

603
In the case of isoprene oxidation, the nitrate groups of C5H9O4-8N, C5H7O5-8N and 604 C4H7O5-6N series (summarized in Table S8) are likely to come from the addition of NO3.

606
These compounds derived from isoprene+NO3 system have been discussed in previous 607 laboratory ( Kwan et al., 2012;Zhao et al., 2020) and ambient data sets (Ayres et al., The second nighttime factor are intense at night and over five times lower during the 627 day. Like the BVOCs-OOMs I factor, this factor has high concentrations when NO is 628 reduced leading to increased NO3 availability (Fig. 10(b)), and about 80% of 629 compounds in this factor are ONs (Fig. 2 (c)). Accordingly, this may also be a factor 630 strongly influenced by NO3. It is dominated by C6-C10 OOMs, among which the highest 631 intensity is at C10 (Fig. 9(d)), coherent with the nature of monoterpene products 632 (BVOCs-OOMs II). This factor has weaker signals at C15 which are plausibly the 633 products of sesquiterpenes but could also be dimmers formed from R4 (monoterpenes 634 + isoprene or monoterpenes + C5 monoterpenes fragments). Compared to the BVOCs 635 OOMs I factor ( Fig. 9(d)), this factor has more large mass molecules (C10) and fewer 636 small mass molecules (C5), resulting in an effective volatility over one order of 637 magnitude lower. A NO3-initiated factor, called the nighttime type-2 factor, has also 638 been discovered in Hyytiä lä , Finland (Yan et al., 2016), but the similar factor we found 639 has a higher proportion of organic nitrates, due to the more abundant NOx here. In terms of fingerprint molecules of this factor (summarized in reported in the experiments of monoterpenes + NO3 system (Nah et al., 2016;Faxon et 657 al., 2018;Takeuchi and Ng, 2019).

659
It is noteworthy that a set of nitrogen-containing radicals, C10H16O6-11N (peak fitting 660 are shown in Fig. S9 abstraction forming a carbon-centered radical (Orlando et al., 2003;Orlando and 669 Tyndall, 2012), (3) the carbon-centered radical can again take up an oxygen molecule 670 and follow the autoxidation route. The C10H16O9N radical is also moderately intense in 671 the BVOCs-OOMs I factor ( Fig. 9(a)), testifying the presence of NO3 chemistry. These 672 C10H16O6-11N radicals are also reported in the CLOUD chamber (Yan et al., 2020). In 673 addition to C10 radicals, a C5 radical, C5H8O5N (peak fitting are shown in Fig. S9), is 674 also found in the BVOCs-OOMs II factor. C5H8O5N are possibly derived from the 675 oxidation of isoprene initiated by NO3, as observed in the laboratory (Zhao et al., 2020).

676
Another hypothesis is that C5H8O5N is formed from the fragmentation process of The third nighttime factor (BVOCs-OOMs III) is dominated by nitrogen-containing 690 species with a contribution ratio about 90%, among which dinitrates account for more 691 than 20% (Fig. 2(c)). When grouped by carbon numbers, C10 OOMs have the strongest 692 signal. Unlike the above two NO3-related factors, this factor shows no correlation with 693 PNO3, but has positive correlation with NO, especially during the daytime (Fig. 10(c)).

694
C9H14O, a typical product of NO-affected monoterpenes oxidation (Calogirou et al., 695 1999), is found to be correlated with this factor (Fig. 10(c)). It is reasonable to infer that 696 these organic nitrates may come from terminations of monoterpenes-RO2 by NO. In 697 addition to the elevated intensity during the nighttime, this factor still remains at a 698 relatively high concentration in the morning, which is much higher than that of the two involving OH or NO3 oxidation of monoterpenes proceeding RO2 + NO terminations.

711
Nitrated phenols are of concern, because of their phytotoxicity (Rippen et al., 1987)  NP III factor is less pure than the NP I & II factors (Fig. 11), its time series follows well 719 with C6H4O5N2 ( Fig. 11(f)), implying that this factor is driven by di-nitrated-phenols After performing PMF analysis, over 1000 non-nitro molecules have been identified 738 through HR peaks fitting in each factor. The mean concentration of total non-nitro

740
Ensemble chemical properties of these non-nitro OOMs are summarized in Fig. 12. The 741 number of carbon atoms implies the precursor information of OOMs. C5 OOMs, which 742 principally consist of isoprene products benefited from the high reactivity and intensive 743 emissions of isoprene in summer, are the most abundant (Fig. 12(c)). While C6-C9

744
OOMs are mostly likely formed from the oxidation of AVOCs such as aromatics and 745 aliphatic series in the urban and suburban atmosphere, and as we expected, these 746 AVOCs-derived OOMs account for about 50% of the total signal (Fig. 12(c)). The 747 intensity of OOMs decreases from C7 to C9 determined by the concentration distribution 748 of precursors, but becomes a plateau at C10 (Fig. 12(c) measurements (Yan et al., 2016;Lee et al., 2016;Massoli et al., 2018).

763
OOMs grouped by carbon numbers or nitrogen numbers consistently have absolute 764 high concentrations in the daytime (Fig. 12(a) and (b)), revealing the crucial role of 765 photochemical progress, involving RO2 + NO termination reactions, on OOMs. In 766 addition, The C5-C10 OOMs are enhanced again during 19:00-22:00 LT, and the 767 nighttime peak of C10 OOMs is even higher than its daytime peak ( Fig. 12(a)). The 768 nocturnal C10 OOMs are more intense than C9 OOMs ( Fig. 12(a)), and there are more 769 C10 nitrates than C9 nitrates (Fig. 12(c)). These results show the fate of VOCs degraded 770 by NO3 during the nighttime, which are more important to monoterpenes. In contrast to 771 nitrogen-free OOMs, organic nitrates are enriched through the reactions of BVOCs with 772 NO3 in the early evening ( Fig. 12(b)), as indicated by three BVOCs nighttime chemistry 773 factors.

775
Apart from reflecting the influence of NOx, multi-nitrates also imply the multiple 776 generations of VOCs oxidation, which is evident in the products of isoprene (e.g.,

777
C5H10O8N2 and C5H9O10N3) due to its two carbon-carbon double bonds. As products of 778 mononitrates, multi-nitrates follow mononitrates in diurnal variation, with double peaks 779 initiated by OH and NO3 respectively ( Fig. 12(b)). Considering that the formation of 780 organic nitrate is only a small branch of RO2 + NO termination, the contribution of 781 multi-step oxidation should be larger than that shown in Fig. 12(c). The binPMF analysis, which avoids the uncertainty introduced by high-resolution peak 801 fitting to the input data matrix, was applied to deconvolve the complexity of the data