key: cord-0855873-7d2pl9g1
authors: Yang, Jieru; Wang, Shenbo; Zhang, Ruiqin; Yin, Shasha
title: Elevated particle acidity enhanced the sulfate formation during the COVID-19 pandemic in Zhengzhou, China()
date: 2021-12-18
journal: Environ Pollut
DOI: 10.1016/j.envpol.2021.118716
sha: 578b05ca948d8c3e15a8d884514dafe05d065092
doc_id: 855873
cord_uid: 7d2pl9g1

The significant reduction in PM(2.5) mass concentration after the outbreak of COVID-19 provided a unique opportunity further to study the formation mechanism of secondary inorganic aerosols. Hourly data of chemical components in PM(2.5), gaseous pollutants, and meteorological data were obtained from January 1 to 23, 2020 (pre-lockdown) and January 24 to February 17, 2020 (COVID-lockdown) in Zhengzhou, China. Sulfate, nitrate, and ammonium were the main components of PM(2.5) during both the pre-lockdown and COVID-lockdown periods. Compared with the pre-lockdown period, even though the concentration and proportion of nitrate decreased, nitrate was the dominant component in PM(2.5) during the COVID-lockdown period. Moreover, nitrate production was enhanced by the elevated O(3) concentration, which was favorable for the homogeneous and hydrolysis nitrate formation despite the drastic decrease of NO(2). The proportion of sulfate during the COVID-lockdown period was higher than that before. Aqueous-phase reactions of H(2)O(2) and transition metal (TMI) catalyzed oxidations were the major pathways for sulfate formation. During the COVID-lockdown period, TMI-catalyzed oxidation became the dominant pathway for aqueous-phase sulfate formation because the elevated acidity favored the dissolution of TMI. Therefore, the enhanced TMI-catalyzed oxidation affected by the elevated particle acidity dominated the sulfate formation, resulting in the slight increase of sulfate concentration during the COVID-lockdown period in Zhengzhou.

been reported. However, the differences in SNA formation pathways are still unclear, and there is a 100 lack of detailed acidity analysis on sulfate formation after the outbreak of the COVID-19 pandemic. 101

In this study, hourly data, including chemical components of PM2.5, gaseous pollutants, and 102 meteorological variables, were collected by a series of online monitoring instruments from January 103 lockdown (January 24 to February 27, 2020), on the roof of a six-story building (about 20 m above 114 the ground) in Zhengzhou University (34°48′ N; 113°31′ E). Detailed information can be found in 115 (Yang et al., 2020). The site is close to residential areas, surrounded by busy roads, and there is a 116 coal-fired power plant 6 km to the southeast ( Figure S1 of Supplemental Materials). 117

The ambient ion monitor (URG-9000D, Thermal Science, USA) was used to continuously 118 measure water-soluble inorganic ions (NO3 -, SO4 2-, Cl -, Na + , NH4 + , K + , Mg 2+ , and Ca 2+ ) and trace 119 gases (HNO3, HCl, NH3, and HONO). Organic carbon (OC) and elemental carbon (EC) were 120 quantified by NIOSH (National Institute for Occupational Safety and Health) 5040 thermal-optical 121 transmittance method using a semi-continuous carbon analyzer (Model-4, Sunset Laboratory, USA). (1) 143

There are stable and metastable conditions that can simulate the thermodynamic equilibrium of 144 aerosols in the ISORROPIA-II model. Stable state refers to the precipitation of salt in the aerosol 145 when it reaches saturation in the aqueous phase, while metastable refers to the salt always in the 146 aqueous phase without considering its saturated precipitation (Fountoukis and Nenes, 2007) . 147

Considering the high RH level recorded in the sampling period, the aerosol system was operated in 148 metastable conditions. There are two modes in the model. The forward-mode calculations using the 149 total concentrations of gas and aerosol species as inputs are less affected by these measurement accounting for 58-69% and 55-63%, respectively. Moreover, the percentages of SNA in PM2.5 188 increased with the aggravation of pollution, especially the proportions of NO3and SO4 2-189 respectively increased from 25% and 13% during the clean period to 31% and 15% during the heavy 190 pollution period during the COVID-lockdown. In addition, another major component, organic 191 matter (OM) showed a downward trend (Fig. 2) . Therefore, SNA formation played a crucial role in 192 PM2.5 increase even during the COVID-lockdown period in Zhengzhou. 193 Compared with the pre-lockdown period, the concentration and proportion of nitrate to SNA 194 during the COVID-lockdown period decreased by 33% and 6%, respectively, revealing the 195 lockdown had a powerful effect on nitrate reductions. However, nitrate remained the largest 196 composition in PM2.5 during the COVID-lockdown period. On the other hand, the disproportionate 197 drop of nitrate compared to NO2 concentration (60%) implied the improved conversion of NO2 198 during the COVID-lockdown period. Furthermore, the nitrogen oxidation ratios (NOR = NO3 -/ (0.08-0.72) were significantly higher than those during the pre-lockdown period (0.06-0.61) 201 under all pollution levels (Fig. 3a) . Therefore, nitrate formation was enhanced relative to NO2 202 emission reduction during the COVID-lockdown period, which was consistent with the 203 characteristics of Beijing and Shanghai (Chang et al., 2020; Huang et al., 2020). As for sulfate, the 204

proportions of SO4 2were slightly increased, accompanied by elevated sulfur oxidation ratios (SOR 205 = SO4 2-/ (SO4 2-+ SO2)) during the COVID-lockdown period (Fig. 3b) , demonstrating the greater 206 importance of sulfate formation during the COVID-lockdown periods in Zhengzhou. reported that the formation of HNO3 is mainly limited by the oxidants OH and O3 produced by the 232 photochemical reaction of NOx and VOC. In addition, compared to the pre-lockdown period, the 233 concentration of HONO, another main precursor for OH radical, during the COVID-lockdown 234 period decreased with the average values from 4.7 ± 3.9 to 3.0 ± 2.3 μg m -3 (Table 1) the elevated acidity resulted in low rates of O3 oxidation because more S(IV) was in the state of 287 SO2·H2O with a low oxidation rate constant (Xue et al., 2016) . The NO2 oxidation rate constant was 288 negatively related to particle acidity (Wang et al., 2020b) , and thus the production rate of NO2 289 oxidation decreased along with the less NO2 concentration during the lockdown. In Fig.7, during  290 the COVID-lockdown period, the sum of the aqueous-phase sulfate production rates of four routes 291 (TMI + NO2 + O3 + H2O2) is greater than before. As for the total aqueous-phase sulfate production 292 rates during the COVID-lockdown period, the enhanced TMI-catalyzed oxidation reaction rates at 293 daytime can offset the reduction of the other three oxidation pathways and hence lead to the high 294 total aqueous-phase sulfate production rates increasing of 2-179% compared with those during the 295 pre-lockdown period. Therefore, the enhanced TMI-catalyzed oxidation affected by the elevated 296 particle acidity dominated the sulfate formation and resulted in the slight increase of sulfate 297 concentration during the COVID-lockdown period in Zhengzhou. In addition, the elevated particle 298 acidity was also beneficial to the gas-particle partitioning of NH3 into NH4 + and prevented a 299 

To estimate the uncertainties of aqueous sulfate-producing rates by input data, an uncertainty 303 analysis was conducted. Therefore, we made an uncertainty analysis. First, we estimated the pH 304 uncertainty based on sensitivity tests of pH to input data (Fig. S4) TNa, TCl, Mg 2+ , and RH) were adjusted up to within their maximum positive uncertainties, and uncertainties, which represented the AWCmax case; For the AWCmin case, the above factors were set 314 to the opposite case. pHmax and AWCmax cases lead to 4 % and 35% (slope−1) errors, respectively, 315 pHmin, and AWCmin cases can result in approximately 6 % and 26% deviations, respectively in Fig.  316 S8. The uncertainty of H2O2, SO2, and O3 is 10%. Fe and Mn uncertainties were set to be 20%. 317

The details can be found in the Supplement (Text S1). 318

Based on the sensitivity of pH, AWC, Fe, Mn, H2O2, SO2, and O3, the maximum and minimum 319 rate scenarios of daily variation of four sulfate production paths are calculated and illustrated in Fig.  320 S9. Detailed data can be found in Table S3 . TMI-catalyzed oxidation exhibit uncertainties in the 321 range of −74−613%. Considering calculation uncertainties, markedly enhanced TMI-catalyzed 322 oxidation rates are observed during the COVID-lockdown period (Fig. S9) . Therefore, the oxidation 323 pathway of TMI rather than H2O2 was found to contribute greatly to atmospheric sulfate formation 324 during the COVID-lockdown period (Fig. S9) . 325 326

Even though the concentration of NO2 and nitrate decreased, the contribution ratio of nitrate to 328 PM2.5 during the COVID-lockdown period was the largest. Moreover, the nitrate formation was 329 enhanced due to the elevated O3 concentration, which was favorable for the homogeneous and 

The authors declare that they have no known competing financial interests or personal 551 relationships that could have appeared to influence the work reported in this paper. Tables   Table 1 A 

and SNA components and meteorological factors impact on air pollution through 2013-492 2017 in Beijing

High aerosol acidity despite declining 495 atmospheric sulfate concentrations over the past 15 years

Aerosol liquid water 498 driven by anthropogenic inorganic salts: Implying its key role in haze formation over the North 499

COVID-19 impact 502 on the concentration and composition of submicron particulate matter in a typical city of 503

Sulfate formation 505 enhanced by a cocktail of high NOx, SO2, particulate matter, and droplet pH during haze-fog 506 events in megacities in China: An observation-based modeling