Winter observations of ClNO2 in northern China: Spatiotemporal variability and insights into daytime peaks

This paper compares the formation of ClNO2 and its impact on the tropospheric radical budget at 3 ground sites in China during winter and summer over an extended measurement period. It is important in that it shows in places subject to fresh emissions of NO, that ClNO2 formation can be even more important in the summer and during the daytime when compared to its formation during the winter in the same places.


49
Cl is a potent atmospheric oxidant that reacts analogously to hydroxyl radicals (OH) 50 with hydrocarbons (Wang et al., 2019b). Cl is highly reactive toward alkanes, with the 51 rate constants of its reactions with alkanes being approximately 10-200 times greater 52 than some of the OH + VOCs reactions (Atkinson and Arey, 2003;Burkholder et al., 53 2015). Consequently, Cl enhances the production of ROx (= OH + HO2 + RO2) via 54 Reactions R1-R4, which promotes O3 formation by converting nitric oxide (NO) to 55 nitrogen dioxide (NO2) (Reactions R3 and R5). Cl also consumes O3 via Reaction R8. 56 The net effect of the reactivity of Cl is typically the depletion of O3 in the stratosphere 57 (Molina and Rowland, 1974) and an increase in O3 production in the polluted 58 troposphere (Riedel et al., 2014;Xue et al., 2015). Field observations of ClNO2 were first reported in the marine boundary layer off the 93 coast of the Houston-Galveston area in the USA (Osthoff et al., 2008). Subsequent 94 studies demonstrated the worldwide ubiquity of ClNO2 and confirmed its significant 95 role in photochemistry (Thornton et al., 2010;Mielke et al., 2011;Phillips et al., 2012;96 Edwards et al., 2013; Bannan et al., 2015;Wild et al., 2016;Wang et al., 2016; 97 Bannan et al., 2019;Eger et al., 2019). The role of ClNO2 in the radical budget could 98 be more important than that of OH in winter, because OH production is reduced in 99 winter owing to lower concentrations of O3 and H2O vapor in this season (Haskins et  previous studies (Tham et al., 2016;Wang et al., 2017c;Xia et al., 2019), and a brief 144 introduction is given here. few industrial facilities. When the prevailing wind originates from the north (i.e., 164 remote mountainous regions), the site is upwind from the Beijing downtown area and 165 thus is less polluted. However, when the wind originates from the south, the site 166 receives pollutants from Beijing's urban areas in the NCP (Xia et al., 2019).

168
Mt. Tai is located approximately 40 km south of Jinan City (population: 8.9 million) 169 and 15 km north of Tai'an City (population: 5.6 million) (Wen et al., 2018).  However, considering the low temperature (4.6 ± 6.3 ℃) observed on top of Mt. Tai, 174 this study considered the observation period as winter to early spring.  (Tham et al., 2016;Wang et al., 2017c;Xia et al., 2019). The primary ions used 181 in the Q-CIMS were iodide (I -) and its water clusters, which were generated using CH3I 182 with an inline ionizer ( 210 Po). The iodide adducts, namely IN2O5and IClNO2 -, were 183 then detected by the mass spectrometer. An example of the mass spectrum is shown in 184 Fig. S2. The isotopic ratios of I 35 ClNO2and I 37 ClNO2in the ambient data were used 185 to confirm the identity of ClNO2 (Fig. S3). Gas-phase mixtures of NO2 and O3 produced 186 N2O5 in a dynamic gas calibrator (Sabio Instruments) for N2O5 calibration. The 187 synthetic N2O5 was converted to ClNO2 by passage through a humidified NaCl slurry 188 for ClNO2 calibration. On-site calibrations were performed every 1-2 days, and 189 background detections of N2O5 and ClNO2 were conducted every day by passing 190 ambient air through glass wool. The dependence of the N2O5 sensitivities (normalized 191 to the signal of I(H2O) -) on ambient RH was tested and used to calibrate the N2O5 data 192 (Fig. S4). The normalized sensitivity of N2O5 is the signal ratio of I(N2O5)to I(H2O) -

193
in the presence of 1 pptv of N2O5. The normalized sensitivities and detection limits of 194 the N2O5 and ClNO2 measurements were (0.9-2.2) × 10 -5 Hz/Hz/pptv and 4-7 pptv (3σ 195 in 5 minutes), respectively during the three campaigns. The variation in the sensitivities 196 and detection limits of N2O5 and ClNO2 were small within each campaign (Text S1, 197 Table S1, and Fig. S5). A virtual-impactor design (Peng et al., 2020) was adopted, and 198 the sampling tube was replaced daily to minimize inlet artifacts.  The concentrations of the NO3 -, SO4 2-, and NH4 + measured simultaneously by the 212 MARGA and ACSM were in good agreement, whereas the concentration of Cl -213 measured by the ACSM was slightly lower than that measured by the MARGA, which 214 was possibly due to the significant proportion of refractory chloride, e.g., NaCl, present 215 in the aerosols (Xia et al., 2020). We assumed that the particles sampled by a wide-216 range particle spectrometer (WPS) were spherical in shape and calculated the aerosol 217 surface area density (Sa) and volume density (Va). A parameterization was adopted to   VOCs.
where ki is the rate constant for a specific VOC + NO3 reaction and k NO+NO 3 represents 231 the rate constant for Reaction R11. The ambient concentrations of NO3 were estimated 232 by assuming that NO3 and N2O5 were in dynamic equilibrium.
The loss rates of NO3 due to NO  The loss rate coefficient of N2O5 on the aerosol surface (k(N2O5)) is expressed as where c(N2O5) represents the average molecular velocity of N2O5. The rate constants 240 (k1, ki, and k NO+NO 3 ) and equilibrium constant (K eq ) are calculated as temperature-241 dependent parameters.
242 243 γ(N2O5) and φ(ClNO2) were estimated using steady-state analysis in applicable cases 244 (Brown et al., 2006). This method assumes a steady state of N2O5, which means that the production rate of N2O5 is equal to its loss rate. We adopted the criteria described 246 by Xia et al. (2020) to select the cases, namely low concentrations of NO, an increasing 247 trend of ClNO2 concentrations, and stable air masses. Equation (5) was then established 248 by plotting τ N 2 O 5 -1 × NO 2 × K eq against 0.25 × S a × C N 2 O 5 × NO 2 × K eq , with 249 γ(N2O5) as the slope and k(NO3) as the intercept in the linear regression (Brown et al., 250 2003). Here, the derived γ(N2O5) was accepted when the regression had R 2 > 0.5 and   and were significantly lower than that at Mt. Tai (179 ± 247 pptv). The nocturnal ratio 290 of ClNO2/N2O5 at each site displayed large day-to-day variability, which was positively 291 dependent on the ambient RH ( Fig. S7) and, to a lesser extent, positively correlated  Beijing, and Mt. Tai sites throughout the campaign in winter (this study) and previous 307 summer field studies ( Table 1). The shaded areas indicate the 10th and 90th percentiles.

309
The nocturnal production of ClNO2 was insignificant in Wangdu despite the presence summer study at Mt. Tai were due to emissions from distinct coal-fired power plants, 358 whereas this winter study found that coal burning had less effect on concentrations of 359 ClNO2. The campaign-averaged levels of SO2 and particulate SO4 2were 1.6 ± 1.6 ppbv 360 and 3.6 ± 2.9 μg m -3 , respectively, during the winter observations, which were 361 significantly lower than those observed in the summer campaigns (2.9 ± 3.7 ppbv and 362 14.8 ± 9.0 μg m -3 , respectively). The reduced effect from coal-fired power generation spring compared with that in summer.

417
The thermal decomposition of N2O5 was suppressed in winter and resulted in high 418 ratios of N2O5/NO3 ( Fig. 5a; up to approximately 1000), which favored N2O5 loss over 419 NO3 loss. However, the γ(N2O5) in winter was systematically lower than that in summer 420 (Fig. 5b), which indicated slower N2O5 loss in winter. This result differs from previous In the winter campaigns, high concentrations of ClNO2 were sustained after sunrise.

439
Distinct peaks in ClNO2 concentrations were observed on 3-4 days in each campaign, 440 as shown in Fig. 6. Other daytime cases from the three sites are shown in Fig. S9-11.

441
The validity of the daytime peaks was checked by performing isotopic analysis of 442 in ClNO2 concentrations. For example, the daily maximum rates of jNO2 (1-minute 457 average) for the Wangdu case shown in Fig. 6a (2.5 × 10 -3 s -1 ) was significantly lower 458 than the highest rate observed during this campaign (6.0 × 10 -3 s -1 ). The attenuated solar 459 radiation reduced the photolysis of ClNO2, which allowed it to persist for a longer (e.g., Gao et al., 2005;Zhou et al., 2009;Jiang et al., 2020). 3.4 Impact of daytime ClNO2 on atmospheric oxidation capacity 492 We used the box model (Section 2.5) to show the impact of ClNO2 on photochemical 493 oxidation at the three sites (Fig. 6a-c) (Tham et al., 2016).

501
However, the impact of ClNO2 increased considerably in the cases of daytime-peak 502 concentrations, as shown in Fig. 7 Fig. 7a-c). The winter P(Cl) peak in Wangdu (Fig. 7a, 0.46 ppbv h -1 ) was twice the 506 summer average value (0.24 ppbv h -1 ) (Tham et al., 2016). P(Cl) from other sources 507 (e.g., the HCl + OH reaction) was minor (8.8-14.5 %) during these cases. The relative 508 importance of ClNO2 in primary radical production varied among these sites. ClNO2 509 had a minor contribution in Beijing but became increasingly important in Wangdu and

510
Mt. Tai (Fig. 7b, c). HONO photolysis was the most important source of OH at the two 511 ground sites, whereas O3 was also important at Mt. Tai.

513
The liberated Cl (mostly from ClNO2 photolysis) accounted for 28.5-57.7 % of the 514 daytime (06:00-18:00 LT) oxidation of alkanes, 6.1-13.7 % of that of alkenes, 5.3- The impact of Cl in the NCP is likely larger than the result shown above. Our model 524 calculations considered photolysis of ClNO2 (and HCl + OH) as the source of Cl, but 525 not other photolabile Cl-containing gases. However, in the Wangdu field campaign, we 526 frequently observed elevated daytime concentrations of bromine chloride (BrCl) and 527 molecular chlorine (Cl2), which dominated the Cl production (Peng et al., 2020). In where strong ClNO2 production is expected to occur (Zhang et al., 2017). It is thus 531 highly desirable to measure ClNO2 in the residual layer in future studies to Data availability.

554
The datasets described in this study is available by contacting the corresponding 555 author (cetwang@polyu.edu.hk). Competing interests. 565 The authors declare that they have no conflict of interest.