Production of HONO from heterogeneous uptake of NO 2 on 1 illuminated TiO 2 aerosols measured by Photo-2 Fragmentation Laser Induced Fluorescence 3

11 The rate of production of HONO from illuminated TiO2 aerosols in the presence of NO2 was 12 measured using an aerosol flow tube coupled to a photo-fragmentation laser induced 13 fluorescence detection apparatus. The reactive uptake coefficient of NO2 to form HONO, 14 γNO2→HONO, was determined for NO2 mixing ratios in the range 34 – 400 ppb, with γNO2→HONO 15 spanning the range (9.97 ± 3.52)  10 to (1.26 ± 0.17)  10 at a relative humidity of 15 ± 1 16 % and for a lamp photon flux of (1.63 ± 0.09) × 10 photons cm s (integrated between 290 17 and 400 nm), which is similar to values of ambient actinic flux at midday. γNO2→HONO 18 increased as a function of NO2 mixing ratio at low NO2 before peaking at (1.26 ± 0.17) × 10 19 at 51 ppb NO2 and then sharply decreasing at higher NO2 mixing ratios, rather than levelling 20 off which would be indicative of surface saturation. The dependence of HONO production on 21 relative humidity was also investigated, with a peak in production of HONO from TiO2 aerosol 22 surfaces found at ~25 % RH. Possible mechanisms consistent with the observed trends in both 23 the HONO production and reactive uptake coefficient were investigated using a zero24 dimensional kinetic box model. The modelling studies supported a mechanism for HONO 25 production on the aerosol surface involving two molecules of NO2, as well as a surface HONO 26 loss mechanism which is dependent upon NO2. In a separate experiment, significant production 27 of HONO was observed from illumination of mixed nitrate/TiO2 aerosols in the absence of 28 NO2. However, no statistically significant production of HONO was seen from the illumination 29 https://doi.org/10.5194/acp-2020-1216 Preprint. Discussion started: 1 December 2020 c © Author(s) 2020. CC BY 4.0 License.

surfaces found at ~25 % RH. Possible mechanisms consistent with the observed trends in both 23 the HONO production and reactive uptake coefficient were investigated using a zero-24 dimensional kinetic box model. The modelling studies supported a mechanism for HONO 25 production on the aerosol surface involving two molecules of NO2, as well as a surface HONO 26 loss mechanism which is dependent upon NO2. In a separate experiment, significant production 27 of HONO was observed from illumination of mixed nitrate/TiO2 aerosols in the absence of 28 NO2. However, no statistically significant production of HONO was seen from the illumination 29 https://doi.org/10.5194/acp-2020-1216 Preprint. Discussion started: 1 December 2020 c Author(s) 2020. CC BY 4.0 License. 90 % of the primary production of OH averaged over the day (Slater et al., 2020). Oxidation 45 by OH radicals is the dominant removal mechanism for many tropospheric trace gases, such as 46 tropospheric methane, as well as the formation of secondary species, including tropospheric 47 ozone (Levy, 1971), nitric and sulphuric acids which condense to form aerosols, and secondary 48 organic aerosols. Understanding the formation of HONO in highly polluted environments is 49 crucial to fully understand both the concentration and distribution of key atmospheric radical 50 species, as well as secondary products in the gas and aerosol phases associated with climate 51 change and poor air quality. 52

Introduction
Atmospheric concentrations of HONO range from a few pptv in remote clean environments 53 (Reed et al., 2017) to more than 10 ppb in highly polluted areas such as Beijing (Crilley et al.,54 2019). The main gas-phase source of HONO in the troposphere is the reaction of nitric oxide 55 (NO) with the OH radical. HONO has also been shown to be directly emitted from vehicles 56 (Kurtenbach et al., 2001;Li et al., 2008), for which the rate of emission is often estimated as a 57 fraction of known NOx (NO2+NO) emissions. Many heterogeneous HONO sources have also 58 been postulated including the conversion of nitric acid (HNO3) on ground or canopy surfaces 59 (Zhou et al., 2003;George et al., 2005), bacterial production of nitrite on soil surfaces (Su et  It is estimated that between 1604 and 1960 Tg yr -1 of dust particles are emitted into the 74 atmosphere (Ginoux et al., 2001). Titanium dioxide (TiO2) is a photocatalytic compound found 75 in dust particles at mass mixing ratios of between 0.1 and 10 % depending on the location the 76 particles were suspended (Hanisch and Crowley, 2003). When exposed to UV light (λ < 390 77 nm) TiO2 promotes an electron ( − ) from the conduction band to the valence band leaving 78 behind a positively charged hole (ℎ + ) in the valence band (Chen et al., 2012): 79 2 + ℎ → − + ℎ + (R1) which can then lead to both reduction and oxidation reactions of any surface adsorbed gas-80 phase species such as NO2 leading to HONO. 81 In previous studies of the reaction of NO2 on TiO2 aerosol surfaces, HONO was observed as a 82 major gas-phase product ( In areas with high mineral dust loading, such as desert regions, far from anthropogenic sources, 92 NO2 concentrations are typically low. However, when dust is transported to urban areas, this 93 source of HONO may become significant. One study reported that TiO2 composed 0.75-1.58 94 µg m -3 when aerosol loadings were 250-520 µg m -3 over the same time period in southeast 95 Beijing, when air had been transported from the Gobi desert (Schleicher et al., 2010). 96 In this study, the production of HONO on the surface of TiO2 particles in the presence of NO2 97 is investigated as a function of NO2 mixing ratio, aerosol surface area density and relative 98 humidity using an aerosol flow tube system coupled to a photo-fragmentation laser induced 99 fluorescence detector (Boustead, 2019). The uptake coefficient of NO2 to generate HONO is 100 then determined, and a mechanistic interpretation of the experimental observations is 101 presented. The production of HONO directly in the absence of NO2 from the illumination of a 102 mixed sample of nitrate and TiO2 aerosol is also presented. Using a similar apparatus, previous 103 work had showed that TiO2 particles produce OH and HO2 radicals directly under UV 104 illumination (Moon et al., 2019). The atmospheric implications of these results and the role of 105 photo-catalysts for the formation of HONO are also discussed. 106 2 Method 107

108
The production of HONO from illuminated aerosol surfaces is studied using an aerosol flow 109 tube system coupled to a photo-fragmentation laser induced fluorescence (PF-LIF) cell which 110 allows the highly sensitive detection of the OH radical formed through photo-fragmentation of 111 HONO into OH and NO followed by Laser-Induced Fluorescence (LIF) detection at low 112 pressure. The experimental setup used in this investigation is described in detail in (Boustead,113 2019), therefore only a brief description of the setup is given here. A schematic of the 114 experimental setup is shown in Figure 1. 115 https://doi.org/10.5194/acp-2020-1216 Preprint. Discussion started: 1 December 2020 c Author(s) 2020. CC BY 4.0 License. Figure 1. Schematic of the Leeds aerosol flow tube system coupled to a laser-fragmentation laser-induced fluorescence detector for HONO. The paths of the 355 nm (blue) and 308 nm (purple, depicted as travelling out of the page perpendicular to the 355 nm light) light are also shown. CPC: condensation particle counter; DMA: differential mobility analyser; HEPA: high efficiency particle air filter; FAGE: fluorescence assay by gas expansion; MCP: microchannel plate photomultiplier; MFC: mass flow controller; RH/T: relative humidity/ temperature probe; SMPS: scanning mobility particle sizer.

116
All experiments were conducted at room temperature (295 ± 3 K) using nitrogen (BOC, 99.998 117 %) or air (BOC, 21 ± 0.5 % O2) as the carrier gas. A humidified flow of aerosols, ~ 6 lpm (total 118 residence time of 104 s in the flow tube), was introduced through an inlet at the rear of the 119 aerosol flow tube (Quartz, 100 cm long, 5.75 cm ID) which was covered by a black box to 120 eliminate the presence of room light during experiments. A 15 W UV lamp (XX-15LW Bench 121 Lamp, λpeak=365 nm) was situated on the outside of the flow tube to illuminate aerosols and 122 promote the production of HONO (half the length of the flow tube was illuminated leading to 123 an illumination time of 52 s). The concentration of HONO is measured by PF-LIF with 124 sampling from the end of the flow tube via a protruding turret containing a 1 mm diameter 125 pinhole, through which the gas exiting the flow tube was drawn into the detection cell at 5 lpm. 126 The detection cell was kept at low pressure, ~ 1.5 Torr, using a rotary pump (Edwards, E1M80) 127 in combination with a roots blower (Edwards, EH1200). All gas flows in the experiment were 128 controlled using mass flow controllers (MKS and Brooks). The relative humidity (RH) and 129 temperature of the aerosol flow was measured using a probe (Rotronics HC2-S, accuracy ±1 130 https://doi.org/10.5194/acp-2020-1216 Preprint. then passed through a neutraliser to apply a known charge distribution and reduce loss of 140 aerosols to the walls. After the neutraliser the aerosol flow was mixed with both a dry and a 141 humidified N2 flow (controlled by MFCs) to regulate the relative humidity of the system by 142 changing the ratio of dry to humid nitrogen flows. A conditioning tube was then used to allow 143 for equilibration of water vapour adsorption and re-evaporation to and from the aerosol surfaces 144 for the chosen RH, which was controlled within the range ~10-70 % RH. A portion of the 145 aerosol flow was then passed through a high efficiency particle filter (HEPA) fitted with a 146 bypass loop and bellows valve allowing control of the aerosol number concentration entering 147 the aerosol flow tube. Previous studies (George et al., 2013;Boustead, 2019) have shown the 148 loss of aerosol to the walls of the flow tube to be negligible. Aerosol size distributions were 149 measured for aerosols exiting the flow tube using a scanning mobility particle sizer (SMPS, 150 TSI 3081) and a condensation particle counter (CPC, TSI 3775) which was calibrated using 151 latex beads. Any aerosol surface area not counted due to the upper diameter range of the 152 combined SMPS/CPC (14.6 -661.2 nm, sheath flow of 3 lpm, instrumental particle counting 153 error of 10-20 %) was corrected for during analysis by assuming a lognormal distribution, 154 which was verified for TiO2 aerosols generated in this manner (Matthews et al., 2014). 155 However, the majority of aerosols, >90 %, had diameters in the range that could be directly 156 detected. Previously, we had determined by imaging the aerosols using a scanning electron 157 microscope (SEM) that the particles were spherical (Moon et al., 2018). In addition to the 158 experiments with single-component TiO2, mixed ammonium nitrate/TiO2 and single-159 component ammonium nitrate aerosols were also generated using the atomiser for 160 investigations of HONO production from nitrate aerosols without NO2 present. An example of 161 an aerosol size distribution from this work for single-component ammonium nitrate aerosols, 162 mixed ammonium nitrate/TiO2 and single-component TiO2 aerosols is shown in Figure 2. 163 https://doi.org/10.5194/acp-2020-1216 Preprint. Discussion started: 1 December 2020 c Author(s) 2020. CC BY 4.0 License.

Figure 2
Typical aerosol surface area distribution for pure ammonium nitrate aerosols (green) and pure TiO2 aerosols (red) and 50:50 mixed nitrate/TiO2 aerosols (purple) measured after the flow tube.

164
As HONO is not directly detectable via LIF, it was necessary to fragment the HONO produced 165 into OH and NO (Liao et al., 2006), with detection of OH via LIF. A 355 nm photolysis laser 166 (Spectron Laser Systems, SL803) with a pulse repetition frequency (PRF) of 10 Hz and pulse 167 duration ~ 10 ns was used to fragment HONO into OH. This fragmentation wavelength was 168 chosen as HONO has a strong absorption peak at ~ 355 nm leading to the breakage of the HO-169 NO bond to form NO and OH in their electronic ground states (Shan et al., 1989). A Nd:YAG 170 pumped dye probe laser (JDSU Q201-HD, Q-series, Sirah Cobra Stretch) with a PRF of 5000 171 Hz, was used for the detection of OH via the fluorescence assay by gas expansion (FAGE) 172 technique which employs the expansion of gas through a small pinhole into the detection cell. 173 The OH radical was measured using on-resonance detection by LIF via the excitation of the 174 A 2 Σ + (ν′ = 0) ← Χ 2 Πi (ν″ = 0) Q1(2) transition at 308 nm (Heard, 2006 ensure the wavelength of the probe laser remained tuned to the peak of the OH transition at 179 308 nm. OH measurements are taken both before and after each photolysis laser pulse allowing 180 measurement of any OH already present in the gas flow to be determined as a background 181 signal for subtraction. The OH generated from HONO photolysis was measured promptly 182 (~800 ns) after the 355 nm pulse to maximise sensitivity to OH before it was spatially diluted 183 away from the measurement region (Boustead, 2019). Offline measurements, with the probe 184 laser wavelength moved away from the OH transition (by 0.02 nm), were taken to allow the 185 signal generated from detector dark counts and scattered laser light to be measured and 186 subtracted from the online signal. To determine an absolute value of the HONO concentration, 187 [HONO], a calibration was performed, in order to convert from the HONO signal, SHONO, using 188 SHONO = CHONO [HONO], as described fully in (Boustead, 2019 (1) where [H2O] is the concentration of water vapour in the humidified gas flow, 2 is the 196 absorption cross section of H2O at 185 nm (7.14 × 10 -20 cm 2 molecule -1 (Cantrell et al., 1997), 197 is the quantum yield of OH for the photo-dissociation of H2O at 185 nm (=1), is the 198 lamp flux and is the irradiation time (the product of which is determined using ozone 199 actinometry (Boustead, 2019). 200 A typical value of the calibration factor was CHONO = (3.63 ± 0.51) × 10 -9 counts mW -1 for N2, 201 leading to a calculated limit of detection of 12 ppt for a 50 s averaging period and a signal-to-202 noise ratio (SNR) of 1 (Boustead, 2019). The typical error in the HONO concentration was 203 15% at 1σ, determined by the error in the calibration. 204 https://doi.org/10.5194/acp-2020-1216 Preprint. Discussion started: 1 December 2020 c Author(s) 2020. CC BY 4.0 License.

Experimental procedure and data analysis
Any HONO seen without the presence of aerosol was therefore due to HONO impurities in the 219 N2 or NO2 gas, the dark production of HONO from the flow tube walls or from the production 220 of HONO from the illuminated reactor walls, which may include production from TiO2 Once aerosols were introduced into the flow tube system a period of ~ 30 min was allowed for 233 equilibration and the measured aerosol surface area density to stabilise. In general, the relative 234 https://doi.org/10.5194/acp-2020-1216 Preprint. Discussion started: 1 December 2020 c Author(s) 2020. CC BY 4.0 License. humidity of the system was kept constant at RH ~ 15 % for all experiments investigating 235 HONO production as a function of NO2 mixing ratio over the range 34 -400 ppb. In a number 236 of experiments, however, RH was varied in the range ~12-37 %. 237 The mixing ratio of NO2 entering the flow tube was calculated using the concentration of the 238 NO2 in the cylinder and the degree of dilution. The NO2 mixing ratio within the cylinder was 239 determined using a commercial instrument based on UV-Vis absorption spectroscopy (Thermo 240 Fisher 42TL, limit of detection 50 pptv, precision 25 pptv) For each individual experiment, the 241 mixing ratio of NO2 was kept constant (within the range 34 -400 ppb) and the aerosol surface 242 area density was varied from zero up to a maximum of 0.04 m 2 m -3 . In order to obtain the 243 HONO produced from illuminated aerosol surfaces in the flow tube for a given mixing ratio of 244 NO2. As well as subtraction of any background HONO, a correction must be made for any loss 245 of HONO owing to its photolysis occurring within the flow tube. 246 In order to determine the rate of photolysis of HONO, the rate of photolysis of NO2 was first 247 determined using chemical actinometry, and the known spectral output of the lamp and the 248 literature values of the absorption cross-sections and photo-dissociation quantum yields for 249 NO2 and HONO were used to determine the rate of photolysis of HONO. When just flowing 250 NO2 in the flow tube, the loss of NO2 within the illuminated region is determined only by 251 photolysis and is given by: 252 where j(NO2) is the photolysis frequency of NO2 for the lamp used in these experiments. From 253 the measured loss of NO2 in the illuminated region, and with knowledge of the residence time, 254 the photolysis frequency, j(NO2), was determined to be (6.43 ± 0.30) × 10 -3 s -1 for the set of 255 experiments using one lamp to illuminate the flow tube. j(NO2) is given by: 256 where 1 and 2 represent the range of wavelengths over which the lamp emits, and and 257 are the wavelength-dependent absorption-cross section and photo-dissociation quantum yield 258 of NO2, respectively, and F is the flux of the lamp at a given wavelength. The flux of the lamp, 259 the spectral intensity of which was measured using a Spectral Radiometer (Ocean Optics QE-260 Pro 500) as a function of wavelength, is shown in Figure 3.
From the measured j(NO2), and with knowledge of  and  for NO2, the flux of the lamp was 262 determined to be (1.63 ± 0.09) × 10 16 photons cm -2 s -1 integrated over the 290 -400 nm 263 wavelength range of the lamp. Using this flux, and the known  and  for HONO over the 264 same wavelength range, j(HONO) was determined to be (1.66 ± 0.10) × 10 -3 s -1 . 265 In the presence of aerosols under illuminated conditions, the rate of heterogeneous removal of 266 NO2 at the aerosol surface to generate HONO is given by: 267 where k is the pseudo-first order rate coefficient for loss of NO2 at the aerosol surface, and 268 which leads to the generation of HONO. The postulated mechanism for HONO production 269 from NO2 is discussed in section 3.3.2 below, but for the definition of k it is assumed to be a 270 first order process for NO2. Integration of equation ((5)   The reactive uptake coefficient of NO2 to generate HONO, 2 → , defined as the 279 probability that upon collision of NO2 with the TiO2 aerosol surface a gas-phase HONO 280 molecule is generated, is given by: 281 where is the mean thermal velocity of NO2, given by = √(8 /( ) with R, T and M as 282 the gas constant, the absolute temperature and the molar mass of NO2, respectively, SA is the 283 aerosol surface area density (m 2 m -3 ) and k is defined as above. Rearrangement of equation ((7)  284 gives: 285 (8)  Rideal or Langmiur Hinshelwood-type mechanism, (iii) subsequent desorption of HONO from 334 the surface, and finally (iv) competitive removal processes for HONO both on the surface and in the gas-phase that are either dependent or independent on the NO2 mixing ratio. The model 336 includes the gas-phase photolysis of NO2 and HONO and the gas phase reactions of both 337 HONO and NO2 with OH and O( 3 P) atoms. 338 To the best of our knowledge the enthalpy of adsorption of NO2 onto a TiO2 surface has not 339 been determined, nor the bimolecular rate coefficients for the chemical steps on the surface 340 shown in Figure 5. Hence, for each of the steps a rate coefficient (s -1 or cm 3 molecule -1 s -1 ) was 341 assigned, as given in Table 1, and with the exception of the experimentally determined j(NO2) 342 https://doi.org/10.5194/acp-2020-1216 Preprint. Discussion started: 1 December 2020 c Author(s) 2020. CC BY 4.0 License. and the calculated j(HONO), and the gas-phase rate coefficients which are known, the rate 343 coefficients were estimated, with the aim of reproducing the experimental NO2 dependence of 344 the HONO production and NO2 reactive uptake coefficient; justification of chosen values is 345 given below. 346

Reactions
Rate coefficient d Model 1

Model 2 and 3
Model 2 only -Eley-Rideal mechanism Model 3 only -Langmuir-Hinshelwood mechanism Common to both Models 2 and 3  The production of HONO on TiO2 aerosol surfaces was measured as a function of the initial 382 NO2 mixing ratio. Figure 6 shows the dependence of the HONO concentration, measured at 383 https://doi.org/10.5194/acp-2020-1216 Preprint. Discussion started: 1 December 2020 c Author(s) 2020. CC BY 4.0 License. the end of the flow tube, on the initial NO2 mixing ratio for an aerosol surface area of (1.6 ± 384 0.8) × 10 -2 m 2 m -3 . A sharp increase in HONO production at a low mixing ratio of NO2 was 385 seen followed by a more gradual reduction in HONO production after a peak production at ~ 386 54 ± 5 ppb NO2. 387  RH values for a fixed aerosol surface area density of (1.59 ± 0.16 × 10 -2 m 2 m -3 ) and at two 389 NO2 mixing ratios, displaying a peak in HONO production between 25 -30 % RH. Above ~ 390 37 % RH, for experiments including single-component TiO2 aerosols, it was found that 391 significant aerosols were lost from the system before entering the flow tube, speculated to be 392 due to loss to the walls of the Teflon lines. As such the RH dependence was only studied up to 393 At the higher initial concentration of NO2 = 170 ppb, the RH dependence showed a similar 419 peak in HONO production between ~25 -30 % RH but less HONO was produced overall, as 420 expected from Figure 6 given the higher NO2. Previous work on the production of HONO from 421 suspended TiO2 aerosols reported a strong RH dependence of the uptake coefficient, , of NO2 422 to form HONO with a peak at ~ 15 % RH and decreasing at larger RH (Gustafsson et al., 2006). 423 The same trend for the NO2 uptake coefficient was observed by Dupart

436
The reactive uptake coefficient, 2 → for NO2→HONO on TiO2 aerosol particles was 437 determined experimentally for 18 different initial NO2 mixing ratios, and is shown in Figure 8. 438 For each initial NO2 mixing ratio, the gradient of the first order rate coefficient for HONO 439 production, k, as a function of aerosol surface area density (e.g. Figure 4)  The HONO production on illuminated TiO2 aerosol surfaces was investigated for each of the 470 mechanisms outlined in Table 1. 471

Model 1 472
Model 1 (see Table 1 and Figure 5), which contains the simplest mechanism, was designed to 473 reproduce the decreasing value of the NO2 uptake coefficient to form HONO, 2 → , with 474 increasing NO2 and also the plateauing at higher NO2 mixing ratios caused by NO2 reaching a 475 maximum surface coverage, as seen by Stemmler et al., (2007). A decrease in the uptake 476 coefficient of NO2, 2 , onto dust aerosol surfaces was also seen in studies where the 477 formation of HONO from NO2 uptake was not directly studied (Ndour et al., 2008;Dupart et 478 al., 2014). The mechanism for Model 1 which is given in Table 1 describes the adsorption of one NO2 molecule to a surface site which then undergoes the reaction which forms HONO, 480 followed by desorption of HONO to the gas-phase, R9-R11. Any representation of the specific 481 chemical processes which convert NO2 to HONO on the surface following the initial photo-482 production of electron-hole pairs in the TiO2 structure (R2) was not included here as the 483 primary focus was to produce the relationship between 2 → and the NO2 mixing ratio. 484 Gustafsson et al., (2006) reported that the measured rate of photo-induced HONO production 485 is 75% that of the rate of NO2 removal, whereas the dark disproportionation reaction (R28) 486 would predict a 50% yield, and hence that the HONO observed in their studies is not simply a 487 photo-enhancement of: 488  at the air-water interface suggested an orientational preference of NO2 on the surface, with both 524 oxygen atoms facing away from the interface which may imply that the asymmetric dimer 525 ONO-NO2 can form directly, meaning the high barrier between the symmetric and asymmetric 526 forms does not need to be overcome (Murdachaew et al., 2013). 527 The energy barrier to isomerisation of symmetric N2O4 in the gas-phase may be reduced due 528 to the interaction with water adsorbed on surfaces. We therefore rule out the dimer in the gas-529 phase adsorbing onto the surface first, and then reacting to form HONO (Varner et al., 2014). 530 An interesting question is whether the first NO2 molecule adsorbed to the surface dimerises via 531 the addition of a gaseous NO2 via an Eley-Rideal (ER) type process, or whether a Langmuir-532 Hinshelwood (LH) type mechanism is operating in which both NO2 molecules are first 533 adsorbed and then diffuse together on the surface forming N2O4. Both ER and LH mechanisms 534 to form the NO2 dimer have been included in the model, denoted as Model 2 and Model 3, 535 respectively. The outputs for Models 2 and 3 (see Table 1 for details of the processes included) 536 for the HONO concentration and 2 → as a function of NO2 are shown in Figure 10   , as a function of the initial NO2 mixing ratio. The mechanisms used for these model runs included a 2:1 stoichiometric relationship between the NO2 adsorbed on the TiO2 aerosol surface and the HONO produced, as well as additional HONO loss reactions which are dependent on NO2, see Table 1 for details. Models 2 and use an Eley-Rideal and Langmuir-Hinshelwood mechanisms, respectively, for the formation of the NO2 dimer on the aerosol surface.
In previous work that investigated HONO production from humic acid aerosols, a saturation 542 effect was seen with HONO production plateauing with increasing NO2 mixing ratio (Stemmler 543 et al., 2007), with the decreasing uptake coefficient, 2 → , with increasing NO2 being 544 attributed to NO2 fully saturating available surface sites. However, the observed decrease of 545 [HONO] at the high NO2 mixing ratios shown in Figure 8 and Figure 10a suggests that 546 additional reactions on the surface may remove HONO and result in the reduction of [HONO] 547 that is measured. As [HONO] decreases with the increase in the NO2 mixing ratio, the removal 548 process should either involve NO2 directly: 549 or involve species made rapidly from NO2 on the surface, such as NO2 + : 550 OH abstraction via NO2 (~133-246 kJ mol -1 ) (Lu et al., 2000). In the gas-phase these reactions 556 are too slow to be important but they could be enhanced on the surface, potentially more so on 557 a photoactive surface such as TiO2. The NO2 dependent loss reaction, kR19 in Table 1 to a plateau, as seen in Model 1 (see Figure 9). In order to observe the model output seen in 561 The photolysis of particulate nitrate was not considered in Models 2 or 3, due to the lack of 574 particulate nitrate in the system at t=0. The gas-to-particle conversion of any HNO3 formed 575 was not considered to be important due to the assumption that most HNO3 formed would For Model 2, which includes the production of HONO via the Eley-Rideal mechanism, in order 579 to reproduce the experimentally observed sharp increase followed by a decrease in both 580 [HONO] and 2 → as a function of increasing NO2 mixing ratio, the modelled rate 581 coefficient for the adsorption of a gas-phase NO2 molecule to another the surface adsorbed 582 NO2 to initially form the symmetric N2O4 dimer, kR12, had to be larger than for the isomerisation 583 step to form HONO and HNO3 via trans-ONO-NO2, kR13. Interestingly, for HONO production 584 via the Langmuir-Hinshelwood mechanism, Model 3, the modelled rate coefficient for the 585 diffusion of one NO2 molecule across the surface to form the dimer with another NO2 molecule, 586 kR14, had to be smaller than for the isomerisation step, kR15, to more closely represent the 587 experimental results for the uptake coefficient. Additionally, in order to reproduce the 588 experimental trend in HONO formation as a function of NO2 mixing ratio, the rate coefficient 589 for the NO2 dependent loss reaction, kR19, had to be larger than the NO2 independent reactions, 590 no gaseous NO2 was added to the gas entering the flow tube. As shown in Figure 11, for the 621 illumination of pure nitrate aerosols, although a small amount of HONO was observed at higher 622 aerosol loadings, no statistically significant production of HONO was seen. 623   Using j(pNO3) = (3.29 ± 0.89) × 10 -4 s -1 , the rate of HONO production from nitrate photolysis 658 at Cape Verde was calculated to be (pNO 3 ) = (4.73 ± 1.01) × 10 -5 s -1 from the mixed 659 nitrate/TiO2 aerosol experiment. Although for pure nitrate aerosol in the absence of TiO2 the 660 data were scattered and the HONO production small (Figure 11), an upper limit estimate of 661 (pNO 3 ) =(1.06 ± 1.15) × 10 -6 s -1 under conditions at Cape Verde could be made using 662 equation (11), as done for rate of HONO production from mixed nitrate/TiO2 aerosols. The 663 atmospheric implications of this will be considered below. 664 4 Implications of HONO production from TiO2 for tropospheric chemistry The net gas-phase production of HONO from Equation (13 was calculated to be -3.8 ppb hr -1 698 (a net loss) as expected due to HONO loss by photolysis peaking at solar noon, suggesting the 699 production of HONO heterogeneously from TiO2 and NO2 (~25 ppt hr -1 ) would have little 700 effect on the overall HONO budget for Beijing summertime at noon. Using Equation (14) this missing HONO production rate for Cape Verde was 34.6 ppt hr -1 , 723 which is within a factor of two of the rate of HONO production (68 ppt hr-1) calculated from 724 nitrate photolysis using our experimental HONO production data for mixed nitrate/TiO2 725 aerosols. These results provide further evidence that particulate nitrate photolysis in the 726 presence of photocatalytic compounds such as TiO2 found in dust could be significant in

730
The experimental production of HONO from both illuminated TiO2 aerosols in the presence of 731 NO2 and from mixed nitrate/TiO2 aerosols in the absence of NO2 was observed, with the 732 HONO concentrations measured using photo-fragmentation laser-induced fluorescence 733 spectroscopy. Using experimental data, the reactive uptake of NO2 onto the TiO2 aerosol 734 surface to produce HONO, 2 → , was determined for NO2 mixing ratios ranging from 735 34 to 400 ppb, with a maximum 2 → value of (1.26 ± 0.17) × 10 -4 for single-component 736 TiO2 aerosols observed at 51 ppb NO2, and for a lamp photon flux of (1.65 ± 0.02) × 10 16 737 photons cm -2 s -1 (integrated between 290 and 400 nm). The measured reactive uptake 738 coefficient, dependence of the HONO production mechanism on TiO2 aerosols showed a peak in 749 production between ~25-30 % RH, with lower HONO production at higher NO2 mixing ratios 750 observed for all RHs tested. The increase in HONO production with increasing RH can be 751 attributed to a higher concentration of H2O on the surface increasing its availability for the 752 hydrolysis reaction to give HONO, whereas a decrease in HONO production after RH ~ 30 % 753 could be due to the increased water surface concentration inhibiting the adsorption of NO2. 754 Using the laboratory reactive uptake coefficient for HONO production, 2 → , the rate of 755 production of HONO from illuminated aerosols in Beijing in summer for typical NO2 mixing 756 ratios and aerosol surface areas was found to be similar to that estimated previously for the 757 production of HONO from urban humic acid aerosol surfaces in Europe. 758 In the absence of NO2, significant HONO production from 50:50 mixed nitrate/TiO2 aerosols 759 was measured. Using the experimental HONO concentrations observed, a rate of HONO 760 production from nitrate photolysis was calculated, which was then scaled to the ambient 761 conditions encountered at the Cape Verde Atmospheric Observatory in the tropical marine 762 boundary layer. A HONO production rate of 68 ppt hr -1 for the mixed nitrate/TiO2 aerosol was 763 found for CVAO conditions, similar in magnitude to the missing HONO production rate that 764 had been calculated previously in order to bring modelled HONO concentrations into line with 765 field-measured values at CVAO. These results provide further evidence that aerosol particulate 766 nitrate photolysis may be significant as a source of HONO, and hence NOx, in the remote 767 marine boundary layer, where mixed aerosols containing nitrate and a photo-catalytic species 768 such as TiO2, as found in dust, are present. 769 However, the production of HONO from pure, deliquesced ammonium nitrate aerosols alone 770 could not be definitively confirmed over the range of conditions used in our experiments, 771 suggesting that another component within the aerosol is necessary for HONO production. 772 Future work should be directed towards studying pure nitrate aerosols over a wider range of 773 conditions, for example varying the aerosol pH, and also adding other chemical species into 774 the aerosol which may promote HONO production. 775 Data availability. Data presented in this study can be obtained from authors upon request 776 (d.e.heard@leeds.ac.uk) 777 Competing interests. The authors declare that they have no conflict of interest. 778 https://doi.org/10.5194/acp-2020-1216 Preprint. Discussion started: 1 December 2020 c Author(s) 2020. CC BY 4.0 License.