Isotopic characterization of nitrogen oxides (NOx), nitrous acid (HONO), and nitrate (NO3(p)) from laboratory biomass burning during FIREX

1. Department of Earth, Environmental and Planetary Sciences, and Institute at Brown for Environment and Society, Brown University, Providence, RI, USA 2. Institute for the Study of Earth, Ocean and Space, University of New Hampshire, Durham, NH, USA 3. Department of Chemistry, University of Montana, Missoula, USA 4. Chemical Sciences Division, NOAA Earth System Research Laboratory, Boulder, CO, USA 5. Cooperative Institute for Research in Environmental Sciences, University of Colorado, Boulder, CO, USA 6. Department of Chemistry, University of Colorado, Boulder, CO, USA a. Now at: Environmental Defense Fund, Boston, MA, USA b. Now at: Department of Chemistry, University of Colorado, Boulder, CO, USA c. Department of Civil and Environmental Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA

Jaffe and Briggs, 2012; Yokelson et al., 1996). Additionally, the temporal evolution of 86 HONO in BB plumes varies greatly in different fires and relative contributions from 87 direct emission versus NO 2 conversion to HONO remains unclear. For instance, 88 significant concentrations of HONO and correlation between HONO and NO 2 have been 89 observed in aged plumes, indicating the importance of heterogeneous conversion of NO 2 -90 to-HONO on BB aerosols (Nie et al., 2015). By contrast, no evidence was found for 91 secondary HONO formation in a BB plume during the Southeast Nexus Experiment 92 (Neuman et al., 2016). It is important to constrain HONO directly emitted from BB 93 compared to HONO formed during plume aging. This would reduce uncertainties 94 associated with the total HONO budget and increase our understanding of HONO 95 impacts on O 3 and secondary aerosol formation downwind of BB regions. 96 97 In an effort to better understand reactive nitrogen emissions and chemistry, especially for 98 HONO, new techniques have been developed to analyze the isotopic composition of 99 various species. Stable isotopes provide a unique approach of characterizing and tracking 100 various sources and chemistry for a species of interest (Hastings et al., 2013 used to track gaseous NO x from a variety of major sources including emissions from 108 biomass burning (Fibiger and Hastings, 2016), vehicles (Miller et al., 2017), and 109 agricultural soils (Miller et al., 2018). Using this method, Fibiger and Hastings (2016) 110 systematically investigated BB δ 15 N-NO x from different types of biomass from around 111 the world in a controlled environment during the fourth Fire Lab at Missoula Experiment 112 (FLAME-4). NO x emissions collected both immediately from the BB source and 1-2 113 hours after the burn in a closed environment ranged from -7 to +12‰, and primarily 114 depended on the δ 15 N of the biomass itself. BB emitted HONO isotopic composition has 115 never been measured before. Our recently developed method for HONO isotopic 116 composition analysis (Chai and Hastings, 2018) enables us to not only characterize δ 15 N 117 and δ 18 O of HONO, but also explore the connection between δ 15 N-NO x and δ 15 N-HONO.

119
The Fire Influence on Regional to Global Environments and Air Quality (FIREX-AQ) 120 investigates the influence of fires in the western U.S. on climate and air quality, via an 121 intensive, multi-platform, campaign. chemiluminescence NO/NO x analyzer, which is described in supplemental information.

222
The NO x measurement verified the concentration of the NO x collected for isotopic 223 analysis, shown in Table S3 and Fig. S1. 224 HONO and HNO 3 concentrations were measured using the University of New 225 Hampshire's dual mist chamber/ion chromatograph system (Scheuer et al., 2003) with the 226 sampling inlet placed right next to that of the ADS. The dual channel IC system is custom 227 built using primarily Dionex analytical components. Briefly, automated syringe pumps 228 are used to move samples and standard solutions in a closed system, which minimizes 229 potential contamination. A concentrator column and 5 ml injections were used to improve 230 sensitivity. Eluents are purged and maintained under a pressurized helium atmosphere. 231 Background signal is minimized using electronic suppression (Dionex-ASRS). The 232 chromatography columns and detectors are maintained at 40 °C to minimize baseline 233 drifting. A tri-fluoro-acetate tracer spiked into the ultra-clean sampling water is used as 234 an internal tracer of sample solution volume, which can decrease due to evaporation in During the experiments, two mist chambers were operated to collect gas samples in 243 parallel, each with an integration interval of 5 minutes. One channel of the IC was 244 utilized for concentration measurement; in the other, the mist chamber's solution was 245 transferred into a sample bottle using the syringe pump, and the collected solution was 246 brought to Brown University for isotopic analysis of HNO 3 if sufficient amount (10-20 247 nmol) was collected for each sample.

249
In addition to MC/IC, the HONO mixing ratios were also measured using high time- open path spanned the width of the stack. This facilitates direct measurement across the 265 rising emissions. The optical path length was set to 58 m. The IR spectra resolution was 266 0.67 cm -1 from 600-4000 cm -1 . Pressure and temperature were continuously recorded 267 with a pressure transducer and two temperature sensors respectively, which were placed 268 adjacent to the White cell optical path. They were used for spectral analysis. were available (Fig. 2) intercepts are not significantly different from zero. All data except one fall within 95% 411 prediction interval bounds of the overall fitting (Fig. 3). Therefore, we conclude that the 412 ADS method has high capture efficiency of HONO in the biomass combustion 413 environment, which assures the accuracy of the isotopic composition analysis and 414 applicability of this method for field-based biomass combustion research. 415 416 3.3 Isotopic composition of HONO and NO x from burning different biomass 417 418 δ 15 N of NO x and HONO emitted from burning various biomass types in this study ranged 419 from -4.3 ‰ to +7.0‰ and -5.3 to +5.8‰, respectively (Table 1). There is no direct 420 dependence of δ 15 N on concentration of either HONO or NO x (Fig. S2). In Fig. 4 and litter, as well as between Engelmann spruce mixture and duff.

438
Our δ 15 N-NO x range falls well within the range (-7‰ to +12‰) found in the FLAME-4 439 experiment (Fibiger and Hastings, 2016). The FLAME-4 study investigated NO x 440 emissions from burning a relatively large range of vegetation biomass from all over the 441 world, and found a linear relationship (Eq. (5)), indicating that 83% of the variation of 442 δ 15 N-NO x is explained by δ 15 N-biomass. The biomass types burned in this work focused 443 on vegetation in the western U.S., and differ greatly from that in FLAME-4, with 444 Ponderosa pine being the only common biomass between the two studies. Specifically, 445 the δ 15 N-biomass range (-4.2‰ to +0.9‰) for this work is much narrower than that of the 446 FLAME-4 experiment (-8‰ to +8‰ including duff and shrub, and the compositions vary slightly amongst each burn. 458 Therefore, the δ 15 N of a particular biomass mixture is mass weighted according to its 459 composition contribution from each part (Table S1) are listed in Table S3). Linear regressions between δ 15 N-HONO and δ 15 N-biomass, as 469 well as that between δ 15 N-NO x and δ 15 N-biomass, show that both δ 15 N-HONO and δ 15 N-470 NO x increase with δ 15 N-biomass in general (Fig. S3) δ 15 N-biomass from this work with those from the FLAME-4 study (Fibiger and Hastings,475 2016) results in a significant linear correlation (Eq. (6)) and is shown in Fig. 5. Despite 476 differences in burned biomass types between the two studies, our δ 15 N-NO x reasonably 477 overlap with the FLAME-4 results within our δ 15 N-biomass range. The relationship 478 between δ 15 N-NO x and δ 15 N-biomass (Eq. (6)) for the combined data highly reproduces 479 that obtained solely from FLAME-4 study (Eq. (5)) and confirms the dependence of reduced mechanism via sensitivity analysis. From these works, we extract major 509 pathways (R1-R11) that are likely responsible for fast gas-phase inter-conversion 510 between NO x and HONO within the combustion system. They found that whether HONO 511 is preferably converted from NO or NO 2 in series during nitrogen transformation 512 (referred to as nitrogen flow) critically depends on temperature. Although our studied fuels are more complicated in composition than a model system 530 involving no more than a few starting species, results from the above studies provide 531 fundamental underpinnings for biomass combustion. Also note that heterogeneous 532 chemistry after these species were emitted was not considered here as the residence time 533 of the fresh plume in our study was ~5 seconds, which is of the same magnitude as that 534 predicted in the nitrogen flow analysis (Houshfar et al., 2012). Kinetic isotope effects 535 (KIE) of these reactions have not been characterized; so only a semi-quantitative 536 prediction is presented here. At low temperatures, R1-R5 are all H-abstraction reactions 537 involving loose transition states that have significant activation energy; a primary KIE is 538 expected for such conditions and leads to 15 N depletion in the product (HONO) ( offsetting the depletion that arose from R1-R5. Consequently, the overall isotope effect of 543 R1-R6 would lead to δ 15 N-HONO < δ 15 N-NO x by a small difference, consistent with our 544 results (Fig. 4). On the other hand, the KIE for the reactions R7-R11 at higher 545 temperatures (> 850 °C) is expected to enrich 15 N in HONO relative to NO x (Chai and 546 Dibble, 2014), leading to an opposite isotope effect to that predicted at lower 547 temperatures. 548 Temperatures of the biomass combustion process span a large range involving different 549 processes including preheating, drying, distillation, pyrolysis, gasification (aka "glowing 550 combustion") and oxidation in turbulent diffusion flames at a range of temperatures 551 associated with changing flame dynamics (Yokelson et al., 1996). Despite this 552 complexity, our measured slight 15 N enrichment in NO x compared to HONO (Table 1, 553 Fig. 4) suggests that the reactions R1-R6 played a more important role than R7-R11 in 554 HONO formation during the FIREX Fire Lab experiments. 555 556

Isotopic composition of nitrates collected on particle filters 557
All Nylasorb filter extract solutions showed no detectable NO 3 and NO 2 concentrations, 558 indicating no significant amount of HNO 3 was collected on these filters, which is 559 consistent with the very low concentrations measured by MC/IC (note that low 560 concentration and limited sample volume also preclude further isotopic analysis of HNO 3 561 collected by MC/IC). By contrast, we found 5 out of 20 particulate filter extract solutions 562 had detectable NO 3 concentration that were sufficient (10 nmol N) for isotopic 563 composition analysis (Table 1). δ 15 N and δ 18 O reported here are considered to represent 564 NO 3 -(p). δ 15 N-NO 3 -(p) of the five samples (burns) range from -10.6 to -7.4 ‰, all of 565 which are more 15 N depleted than that of HONO and NO x . In addition, the smaller range 566 of δ 15 N-NO 3 than that of δ 15 N-HONO and δ 15 N-NO x rules out possible transformation of 567 NO x and HONO to nitrate on the filters, which could distort the isotopic composition of 568 NO x and HONO. 569 570 In the FLAME-4 experiments, only one particulate filter had captured NO 3 -(p) above the 571 concentration detection limit, whereas HNO 3 collected on Nylasorb filters from 7 572 experiments were above the concentration detection limit and therefore only δ 15 N-HNO 3 573 (-0.3‰ to 11.2‰) were reported (Fibiger and Hastings, 2016). The contrast with our 574 filter results are likely attributed to different formation mechanisms under different 575 conditions, in addition to variation of fuel types. Of the 7 detectable HNO 3 collections 576 from FLAME-4, 5 represented room burns for which samples were collected from smoke 577 aged for 1-2 hours in the lab, and the sampled HNO 3 was likely a secondary product. By 578 contrast all our observed NO 3 -(p) were in fresh emissions and may have been derived 579 from plant nitrate (Cárdenas-Navarro et al., 1999) and/or combustion reactions. There 580 have been no other studies on δ 15 N of NO 3 -(p) and HNO 3 directly emitted from fresh 581 plumes to the best of our knowledge, so more investigation using both laboratory work 582 (isotope effect) and kinetic modeling will be needed in order to understand formation 583 mechanisms of HNO 3 and NO 3 -(p) in the biomass combustion process and their 584 respective isotope effects. 585 586 In Interestingly, the linear correlation between δ 15 N-HONO and δ 15 N-NO x for the biomass 634 we studied suggests systematic co-production of NO x and HONO occurs during biomass 635 combustion and both of them are released as primary pollutants in fresh smoke. The 636 relationship between δ 15 N-HONO and δ 15 N-NO x likely reflects that HONO was produced 637 to a larger extent at moderate combustion temperatures (< ~800 °C) than higher 638 temperatures on the basis of a simplified mechanism for flow of reactive nitrogen species. 639 However, we note that this relationship is derived from all measured δ 15 N-HONO and 640 δ 15 N-NO x in fires ranging from smoldering to flaming, so is not necessarily 641 representative of a particular combustion condition. Still, it is likely that a compilation 642 over a range of conditions is more useful for potentially distinguishing HONO sources 643 and formation pathways in the environment since it will always be a challenge to assess 644 exact combustion temperatures. Determining these relationships in real wildfire smoke 645 will be essential for better constraint on NO x and HONO budgets, and eventually may 646 improve ozone and secondary aerosol predictions for regional air quality.  concentration measured using MC/IC method for various stack fires (fire numbers are 984 referred to Table 1). 985 986 987 3 surprising and we will be exploring this further in the coming months along with ancillary da collected during the burns by Bob Yokelson (U. Montana) and Jim Roberts (NOAA/ESRL).