SC I ENCE ADVANCES | R E S EARCH ART I C L E

ENV IRONMENTAL SC I ENCE

1Multiphase Chemistry Department Max Planck Institute for Chemistry Mainz 55128Germany 2State Key Joint Laboratory of Environment Simulation andPollutionControlSchool of Environment Tsinghua University Beijing 100084 China 3College of Engi-neering University of Iowa Iowa City IA 52242 USA 4Center for Global and RegionalEnvironmental Research University of Iowa Iowa City IA 52242 USA 5Center for EarthSystem Science Tsinghua University Beijing 100084 China 6Institute forEnvironmental and Climate Research Jinan University Guangzhou 511443 ChinaThese authors contributed equally to this workdaggerCorresponding author Email yafangchengmpicde (YC) hekbtsinghuaeducn (KH) uposchlmpicde (UP) hsumpicde (HS)

Cheng et al Sci Adv 20162 e1601530 21 December 2016

2016 copy The Authors

some rights reserved

exclusive licensee

American Association

for the Advancement

of Science Distributed

under a Creative

Commons Attribution

License 40 (CC BY)

Do

Reactive nitrogen chemistry in aerosol water as asource of sulfate during haze events in ChinaYafang Cheng1dagger Guangjie Zheng12 Chao Wei1 Qing Mu1 Bo Zheng2 Zhibin Wang1

Meng Gao34 Qiang Zhang5 Kebin He2dagger Gregory Carmichael34 Ulrich Poumlschl1dagger Hang Su61dagger

Fine-particle pollution associated with winter haze threatens the health of more than 400 million people in the NorthChina Plain Sulfate is a major component of fine haze particles Record sulfate concentrations of up to ~300 mg mminus3

were observed during the January 2013 winter haze event in Beijing State-of-the-art air quality models that rely onsulfate production mechanisms requiring photochemical oxidants cannot predict these high levels because of theweak photochemistry activity during haze events We find that the missing source of sulfate and particulate mattercan be explained by reactive nitrogen chemistry in aerosol water The aerosol water serves as a reactor where thealkaline aerosol components trap SO2 which is oxidized by NO2 to form sulfate whereby high reaction rates are sus-tained by the high neutralizing capacity of the atmosphere in northern China This mechanism is self-amplifying be-cause higher aerosol mass concentration corresponds to higher aerosol water content leading to faster sulfateproduction and more severe haze pollution

wnloa

on June 27 2019httpadvancessciencem

agorgded from

INTRODUCTIONPersistent haze shrouding Beijing and theNorth China Plain (NCP) dur-ing cold winter periods threatens the health of ~400million people livingin a region of ~300000 km2 Characteristic features of the winter haze innorthern China include stagnant meteorological conditions with lowmixing heights high relative humidity (RH) large emissions of primaryair pollutants and fast production of secondary inorganic aerosols espe-cially sulfate (see section M1) (1ndash5) Analyzing surface-based observa-tions at a site in Beijing during January 2013 (see section M2) andusing concentration ratios of sulfate to sulfur dioxide ([SO4

2minus][SO2])as a proxy for the sulfate production rate (5) we find that sulfate pro-duction increases as PM25 (particulate matter with a diameter of lessthan 25 mm) levels increase (Fig 1A) Ratios are six times higher duringthe most polluted periods (PM25 gt 300 mg mminus3) than during clean tomoderately polluted conditions (ratios of 01 when PM25 lt 50 mg m

minus3)Traditional air qualitymodels however fail to capture this key feature

of NCP winter haze events even after accounting for aerosol-radiation-meteorology feedback (see sections M2 and M3) (6ndash8) The chemicalmechanisms used in thesemodels usually comprise gas-phase oxidationof sulfur dioxide by OH radicals and aqueous-phase reaction pathwaysin cloud water involving H2O2 and O3 resulting in sulfate productionrates that scale with the intensity of solar ultraviolet (UV) radiation(9 10) During NCP haze days UV radiation is significantly reducedbecause of the aerosol dimming effect resulting in a decrease of mostoxidant concentrations (5) Figure 1B shows that the midday O3 valuesdrop from ~22 parts per billion (ppb) under clean conditions to ~1 ppbduring the haze period (and also lose their typical diurnal variation figS1) The reduced oxidant levels and increased sulfate production suggestthe existence of amissing sulfate production pathway Even after consid-

ering the gas phase and cloudfog chemistry there is still a large gapbetween modeled and observed sulfate (Fig 1C) Adding an apparentheterogeneous process with sulfate production rates that scale withaerosol surface area and RH can greatly improve model predictions(see sectionsM3 toM5) (7) but the chemicalmechanism of themissingsulfate production pathway has not yet been identified

RESULTS AND DISCUSSIONWe find that reactive nitrogen chemistry in aerosol water can explainthe missing source of sulfate in NCP winter haze Aerosol water is akey component of atmospheric aerosols which serves as a mediumthat enables aqueous-phase reactions (11ndash13) The aerosol water con-tent (AWC) in Beijing calculated using measurements of RH andaerosol composition and the ISORROPIA-II thermodynamic equilib-rium model (see section M6) (14ndash16) is well correlated with themissing sulfate content the difference between measured and modeledsulfate (D[SO4

2minus]) (Fig 1C) (see sections M2 to M4) suggesting itsinvolvement in the sulfate production Note that because of the salt-induced freezing point depression (17) aerosol water will not freezefor a winter temperature of ~271 K in Beijing

Taking the impact of mass transfer and ion strength into accountwe make a conservative estimation of sulfate production rate for dif-ferent reactions in the aerosol water under relevant atmospheric tracespecies concentration conditions (see sections M4 and M7 to M9) andfind NO2 to be the most important oxidant in Beijing during hazeperiods (Fig 2B) In the presence of aerosol water gas-phase NO2

can partition into the condensed phase react with SO2 dissolved inthe aqueous phase and produce sulfate as well as nitrite (R1) (18)

2 NO2 ethaqTHORN thorn HSO3 ethaqTHORN thornH2O ethaqTHORN rarr 3 HthornethaqTHORN

thorn 2 NO2 ethaqTHORN thorn SO4

2ethaqTHORN ethR1THORN

Under heavy haze conditions (PM25ge 300 mgmminus3) the sulfate pro-duction rates of the NO2 reaction pathway (R1) are ~1 to 7 mg m

minus3 hminus1much higher than the reaction rates involving other important aqueousoxidants such as O3 and H2O2 According to Zheng et al (7) an ad-ditional sulfate production of ~3 mg mminus3 hminus1 is needed to explain the

1 of 11

SC I ENCE ADVANCES | R E S EARCH ART I C L E

on June 27 2019httpadvancessciencem

agorgD

ownloaded from

observations in the severe winter haze periods of Beijing (see sectionM9) which falls right into the range of production rates from the chem-ical reaction mechanism we proposed (R1) As illustrated in Fig 2 sul-fate production in aerosol water under haze conditions differs from thatin cloud droplets where the major oxidation pathways are reactionswith H2O2 and O3 and NO2 plays only a minor role (12 19) Thustraditional air quality models usually include only the H2O2 and O3 re-action pathways of sulfate production in the aqueous phase whereas theNO2 reaction pathway is neglected (7 20)

The AWC is typically three to five orders of magnitude lower thanthe water content of cloud or fog (21) On this basis how can the NO2

reaction pathway become important in such tiny amounts of waterThe increased importance is due to the relatively high aerosol pH andelevated NO2 concentrations in Beijing and the NCP during hazeperiods As shown in Fig 2 aqueous oxidation rates of S(IV) byNO2 and O3 are strongly pH-dependent High pH increases the sol-ubility and the effective Henryrsquos law constant of SO2 pulling moreSO2 into the aerosol water and thus increasing the reaction rate WhenpH increases by one unit the reaction rates increase by one and twoorders of magnitude for NO2 and O3 respectively The H2O2 reactiondoes not show a strong dependence because high pH reduces its reac-tion rate coefficient which offsets the effect of increased solubilityCompared with North America and Europe aerosols in the NCPare more neutralized (22) as shown by a higher cation-to-anion ratio

Cheng et al Sci Adv 20162 e1601530 21 December 2016

(Fig 1D) This neutralized feature is also well documented and is thereason that acid rain rarely occurs in northern China (see sectionM10) (22) Using the ISORROPIA-II thermodynamic equilibriummodel (see section M6) (14ndash16) and in situ aerosol bulk compositionmeasurements we obtain average pH values of 54 to 62 for aerosolwater under NCP haze conditions (see sections M6 and M9) Similarcalculations based on size-segregated aerosol composition measure-ments even show a higher effective pH and sulfate production rates(fig S2)

Elevated NO2 is another key factor that leads to fast sulfate forma-tion Substantial amounts of NO2 come from direct emission of NOx

(= NO + NO2) Although the NO2-to-NOx ratios were reduced be-cause of weak photochemistry during the haze event the stagnantweather trapped more NO2 near the surface resulting in elevatedNO2 concentrations that were on average three times higher thanthose under clean conditions (~66 ppb Fig 1B) These periods ofhighest NO2 levels occurred when the concentrations of other photo-chemical oxidants that can produce sulfate (H2O2 O3 and OH) werelow (Fig 1B and fig S1) Changes in pH and precursor concentrationstogether lead to the transition from an H2O2-dominated regime ofaqueous sulfate production in cloud droplets to an NO2-dominatedregime of aqueous sulfate production in haze (Fig 2) Earlier studieshad already suggested that the NO2 reaction pathway may contributeto sulfate formation in fogs under polluted conditions (23 24) but

80

60

40

20

0

NO

2 (ppb)

0ndash50

50ndash1

00

100ndash

300

gt300

PM25 (microg mndash3 )

35

30

25

20

15

10

5

0

Mid

day

O3 (

ppb)

O3 NO214

12

10

08

06

04

02

00

[SO

42ndash

] [S

O2]

0ndash50

50ndash1

00

100ndash

300

gt300

PM25 (microg mndash3 )

300

250

200

150

100

50

0SO

42ndash (

obs

- m

odel

microg

mndash3

)

12008004000

Aerosol droplet solution (microg mndash3)

4002000

AWC (microg mndash3)

Europe

East Asia

North America

24201612080400

Anions (micromol mndash3)

24

20

16

12

08

04

00

Cat

ions

(microm

ol m

ndash3)

A B

C D

Fig 1 Characteristic featuresof amajor hazeevent inBeijingChina (January2013) (A) Sulfatesulfur dioxide ratio ([SO42minus][SO2]) and (B)middayozone (O3) concentrations

(1000 to 1500 local time) and nitrogen dioxide (NO2) concentrations at different fine-particle (PM25) concentration levels (C) Correlation of the unexplained sulfate concentration(DSO4

2minus = observation minusmodel) (see sections M2 and M3) with aerosol droplet solution and AWC (color-coded) (D) Anion-to-cation ratio in PM25 as observed in Beijing (solidsymbols) and in other studies in northern China (open symbols) (table S1) compared to the characteristic ratios reported for aircraft measurements in the Arctic of outflow fromEast Asia Europe and North America (solid red blue and green lines here only NH4

+ NO3minus and SO4

2minus were considered) (87)

2 of 11

SC I ENCE ADVANCES | R E S EARCH ART I C L E

on June 27 2019httpadvancessciencem

agorgD

ownloaded from

during the Beijing haze event of January 2013 fog was not observedand sulfate production by NO2 occurred in aerosol water instead

Sulfate production rates from the NO2 reaction pathway (R1)calculated on the basis of measurement data (hourly concentrationsof NO2 SO2 PM25 and RH sections M2 and M4) show a positivedependence on the PM25 concentration varying from 001 mg mminus3 hminus1

under relatively clean conditions to nearly 10 mg mminus3 hminus1 in the mostpolluted periods (pink circles in Fig 3) These reaction rates can accountfor the systematic underprediction of models and explain the largemissing source of sulfate in the Beijing haze (black diamonds in Fig 3)Under clean conditions the OH reaction (green crosses in Fig 3) dom-inates the oxidation pathways of SO2 As particle concentrations increaseandmore sulfate is produced the photochemistry slows down leading toless OH and sulfate production via this pathway decreases From thisaspect the OH reaction has a negative feedback which is self-bufferedagainst heavy pollution As PM25 concentrations andRH increase simul-taneously [a special feature of haze events in NCP (5)] the OH reactionbecomes weaker whereas the aqueous-phase reaction of NO2 starts toplay a more important role For the January 2013 conditions at PM25

~100 to 200 mg mminus3 the aqueous-phase reaction becomes the dominantoxidation pathway (Fig 3) In contrast to the OH reaction the NO2 re-action shows a positive feedback mechanism where higher particle mat-ter levels lead tomore aerosol water which accelerates sulfate productionand further increases the aerosol concentration This positive feedbackintensifies the PM25 levels during haze periods resulting in a series ofrecord-breaking pollution events

The NO2 reaction with SO2 in aerosol water produces not only sul-fate but also nitrite (R1) which may undergo subsequent oxidation ordisproportionation reactions forming nitrate This is consistent with thehigh nitrate concentrations observed during the haze event which are of

Cheng et al Sci Adv 20162 e1601530 21 December 2016

similar magnitude to the sulfate concentrations (up to ~160 mgmminus3) (5)and have also not yet been explained by air quality models (7) Depend-ing on the pH value of the haze droplets nitrite can also form nitrousacid (HONO) and undergo reversible partitioning with the gas phase(25 26) Moreover the release of HONO removes H+ which may helpto sustain the droplet acidity and efficient sulfate production (27 28)For the same haze periods record high aerosol nitrite concentrations ofup to ~12 mg mminus3 were observed in Shandong northern China withnitrite concentrations well correlated with NOx at high RH (gt50)(25) The nitrite-to-HONOmolar ratio of ~3 in Shandong is a hundredtimes higher than the ratio observed during a pollution event inNanjing Yangtze River Delta of China where only trace amounts ofnitrite were detected in the aerosol phase (up to ~04 mg mminus3) andthe aerosol water there was more acidic (pH ~4) (24 25) This furtherconfirms the more neutralized feature of aerosol particle compositionand haze in northern China

Our study unfolds a new and more comprehensive conceptualmodel of sulfate formation in NCP haze events including not onlythe traditional OH H2O2 and O3 reaction pathways in atmosphericgas phase and cloud chemistry but also the NO2 reaction pathway inaerosol water (Fig 4) The NO2 pathway may not be limited to winterhaze because it may also be important at night and during fog events inpolluted regions with high boundary layer concentrations of PM25 andNO2 and elevatedRH (23 24) The importance ofmultiphase chemistryholds for awide range of aerosol pHWhen aerosols becomemore acid-ic the sulfate production can bemaintained at a high rate through TMIreactions (Fig 2B) The important role of aerosol pH in the multiphasereaction pathwayhighlights the need to better understand the sources ofammonia and alkaline aerosol components fromnatural and anthropo-genic emissions (soil dust seawater agriculture energy industrial and

8765432pH

10ndash7

10ndash6

10ndash5

10ndash4

10ndash3

10ndash2

10ndash1

100

101

102

103

Beijing haze

H2O2 O3 TMI NO2

8765432pH

10ndash5

10ndash4

10ndash3

10ndash2

10ndash1

100

101

102

103

104

105

d [S

O4

2ndash]

dt

(microg

mndash3

hndash1

)

Cloud droplets

A B

Fig 2 Aqueous-phase sulfate production by sulfur dioxide oxidation undercharacteristic conditions Sulfate production rates for (A) cloud droplets and (B) Beijinghaze plotted against pH value Light bluendash and gray-shaded areas indicate characteristicpH ranges for cloud water under clean to moderately polluted conditions and aerosolwater during severe haze episodes in Beijing respectively The colored lines representsulfate production rates calculated for different aqueous-phase reaction pathways withoxidants hydrogen peroxide (H2O2) ozone (O3) transition metal ions (TMIs) and nitro-gen dioxide (NO2) Characteristic reactant concentrations and model calculations forclouds andhaze are taken fromthe literature andobservations and specified inMaterialsand Methods (see sections M7 to M9) (12 21)

10ndash3

10ndash2

10ndash1

100

101

d [S

O42ndash

] d

t (micro

g m

ndash3 h

ndash1)

8006004002000

PM25 (microg mndash3)

NO2 reactionOH reactionMissing source

Fig 3 Importance of the NO2 reaction pathway for sulfate production in theBeijing haze (January 2013) Sulfate production rates calculated for the aqueous-phase NO2 reaction pathway (pH 58) and the gas-phase OH reaction pathwaycompared to the missing source of sulfate Crosses and circles represent hourlyproduction rates calculated on the basis of measurement data and the diamondsrepresent the average missing source (arithmetic mean plusmn SD) (see sections M3 toM5) (7) The pink-shaded area shows the maximum and minimum sulfate produc-tion rates by the NO2 reaction pathway bounded by the aerosol water pH rangingfrom 54 to 62 during haze periods

3 of 11

SC I ENCE ADVANCES | R E S EARCH ART I C L E

on June 27 2019httpadvancessciencem

agorgD

ownloaded from

traffic) Furthermore reactive nitrogen chemistry in aerosol watermight also play a role in nitrate and secondary organic aerosol produc-tion during haze days when photochemistry is reduced

Our results reveal the complex nature of haze pollution events inChina where NOx is not only a precursor for nitrate but also an im-portant oxidant for sulfate formation Thus reductions of NOx emis-sions are expected to reduce nitrate sulfate and PM25 much morethan anticipated by traditional air quality models A large decreasein PM25 has already been observed in relation to traffic and energycontrol measures during the Beijing Olympic Games in 2008 and oth-er events in the NCP Heavy haze conditions with high pollutant con-centration levels and largely neutralized aerosol water are key featuresof atmospheric chemistry in the NCP These features will need to beconsidered in future air quality and pollutant emission control strate-gies in northern China and perhaps also in other regions

MATERIALS AND METHODSM1 The January 2013 haze in BeijingThe severe haze episode in January 2013 is one of theworst atmosphericpollution events ever recorded in China (2 9) In Beijing the daily fine-particle (PM25) concentration reached up to 400 mg m

minus3 exceeding theWorld Health Organization guideline value by 16 times The weak EastAsian winter monsoon which resulted in weakened surface winds andthe anomalous southerly winds was responsible for the haze events (4)The southerly winds transportedmorewater vapor from the sea to east-ern China The anomalous high-pressure system at 500 hPa suppressed

Cheng et al Sci Adv 20162 e1601530 21 December 2016

convection Thus the air in January was more stagnant trapping moreair pollutants andwater vapor near the surface The high PM25 concen-tration reduced the solar radiation and atmospheric photochemistryresulting in decreases in the concentration of photochemical productssuch as OH and O3 (fig S1)

A major feature of PM25 pollution during this haze event was thelarge contribution from secondary species including inorganic (mainlysulfate nitrate and ammonium) and organic species (2 5) Howevercontribution from secondary inorganic species (such as sulfate and ni-trate) showed an increasing trend with increasing pollution levelswhereas contribution from organics decreased (5)

The fast production of sulfate however cannot be reproducedby model simulations which have implemented aerosol-meteorology-radiation feedback and a state-of-the-art chemicalmechanism (6 7) thatis the gas-phase oxidation by OH (29 30) and aqueous-phase oxidationin clouds by H2O2 and O3 (12 21) Stabilized Criegee intermediates(sCIs) were also suggested as an oxidant of SO2 but contribute minorH2SO4 production compared with the conventional OH reaction in themidday (31ndash33) Further analysis has shown that the model simulationcan be improved by introducing an apparent heterogeneous process witha reaction rate coefficient scaled with RH (see section M3) (7)

M2 Sampling location and experimental methodsWe performed aerosol measurements from 1 to 31 January 2013 on theroof of the Environmental Science Building (40deg00prime17PrimeN 116deg19prime34PrimeE~10 m above the ground) on the campus of Tsinghua University anurban background site in Beijing Table S2 summarizes the aerosol-related parameters and experimental methods Concentrations of PM25

and PM10 were measured by an online PM-712 monitor (KimotoElectric Co Ltd) equippedwith aUS Environmental ProtectionAgen-cy PM10 inlet and a PM25 virtual impactor (34) SO4

2minus and other ions inPM25 were measured by an online ACSA-08 monitor (Kimoto ElectricCo Ltd) and the filter-based analysis The PM25 filter samples werecollected from 12 to 24 January by median-volume samplers (Laoying)on prebaked Quartz filters (2500 QAT-UP Pall Corporation) with aflow rate of 100 liters minminus1 (35) The filter samples were analyzed bythe Dionex ion chromatograph (DX-600 for cations and ICS-2000 foranions) (Dionex Corporation) for the concentration of water-solubleinorganic ions including Na+ K+ Ca2+ Mg2+ NH4

+ SO42minus NO3

minusand Clminus Organic carbon (OC) and elementary carbon concentrationsin PM25 were measured by a Sunset Model 4 semicontinuous carbonanalyzer (Beaverton) with a National Institute for Occupational SafetyandHealthndashtype temperature protocol (36) A factor of 16 was adoptedto convert themass ofOC into themass of organics (37 38) Gaseous airpollutants SO2 NO2 and O3 were measured by the AtmosphericEnvironment Monitoring Network (39) The meteorology data weremeasured by theMilos 520Weather Station (VAISALA Inc) More de-tails can be found in the work of Zheng et al (5)

M3 WRF-CMAQ model simulationThe Weather Research and ForecastingmdashCommunity Multiscale AirQuality (WRF-CMAQ) model system was used to determine themissing source of sulfate through comparison with observational dataWRF is a new-generation mesoscale numerical weather predictionsystem designed to serve a wide range of meteorological applications(wwwwrf-modelorg) and CMAQ is a three-dimensional Eulerianatmospheric chemistry and transport modeling system that simulatesmultipollutants (wwwcmascenterorgcmaq) The releases of WRFv351 and CMAQ v501 were used in this study

Cloud chemistry

Photochemistry

Haze chemistrySO2 + NO2 SO4

2minus

SO2 + OH SO42minus

SO2 + H2O2O3 SO42ndash

SO2NOXNH3Dust

Fig 4 Conceptual model of sulfate formation in haze events in NCP Thetraditional OH H2O2 and O3 reaction pathways in atmospheric gas phase andcloud chemistry are included here as well as the NO2 reaction pathway in aerosolwater with elevated pH and NO2 concentrations proposed in this study

4 of 11

SC I ENCE ADVANCES | R E S EARCH ART I C L E

on June 27 2019httpadvancessciencem

agorgD

ownloaded from

The model simulation was configured the same as in the workof Zheng et al (7) as detailed in table S3 To better characterize thestagnant meteorological conditions we applied observational nudg-ing for temperature (T) and RH [above the planetary boundarylayer (PBL)] and wind (within and above the PBL) The surfaceroughness is corrected according to Mass and Ovens (40) byincreasing the friction velocity by 15 times in the PBL schemewhich significantly reduced the high biases in wind and RH simu-lations In general the simulated T RH and wind at the groundsurface agree with observations (7) For the gas-phase reactionswe used the CB05 mechanism with active chlorine chemistry andthe updated toluene mechanism of Whitten et al (41) For theaqueous-phase reactions in clouds we used the updated mecha-nism of the RADM model (20 42) Reactions relevant for the sul-fate formation were discussed in detail in section M4

Here the WRF-CMAQ modeling results were used in thefollowing analysis (i) the modeled sulfate concentration [SO4

2minus] wasused to calculate D[SO4

2minus] the difference between observed andmodeled [SO4

2minus] and (ii) the modeled OH and H2O2 concentrationswere used in the estimation of sulfate formation from the gas-phasereaction of OH with SO2 and the aqueous-phase reaction of H2O2

with HSO3minus

M4 Production of sulfate in WRF-CMAQ modelAccording to current understanding secondary sulfate is producedthrough the oxidation of SO2 In the WRF-CMAQ model oxidationoccurs both in the gas phase and in the cloudfog droplets In the gasphase the major pathway is the OH-initiated reaction (12 43)

OH thorn SO2 thorn M rarr HO2 thorn Sulfate ethM1THORN

in which the second-order kinetic constant can be expressed as

k Teth THORN frac14 k0ethTTHORNfrac12M1thorn k0ethTTHORNfrac12M

kinfinethTTHORN

8lt

9=06Z and Z frac14 1thorn log10

k0ethTTHORNfrac12MkinfinethTTHORN

2( )minus1

ethM2THORN

where [M] is the concentration of N2 and O2 and k0(T) and kinfin(T) rep-resent the low- and high-pressure limiting rate constants respectivelyTheir temperature dependence can be expressed as

k0 Teth THORN frac14 k3000T300

n

and kinfin Teth THORN frac14 k300infinT300

m

ethM3THORN

where k0300 = 33 times 10minus31 cm6moleculeminus2 sminus1 and n = 43 and kinfin

300 =16 times 10minus12 cm3 moleculeminus1 sminus1 and m = 0

In cloudfog droplets the following aqueous reactions have been in-cluded in theWRF-CMAQmodel (7) The expression of their reactionrate as well as the rate coefficients is summarized in table S4

HSO3 thorn H2O2 rarr SO4

2 thorn Hthorn thorn H2O ethM4THORN

SO2 thorn O3 thorn H2O rarr SO42 thorn 2Hthorn thorn O2 ethM5THORN

HSO3 thorn O3 rarr SO4

2 thorn Hthorn thorn O2 ethM6THORN

Cheng et al Sci Adv 20162 e1601530 21 December 2016

SO32 thorn O3rarr SO4

2 thorn O2 ethM7THORN

SO2 thorn H2O thorn 05O2 thorn FeethIIITHORN=MnethIITHORNrarr SO42 thorn 2Hthorn ethM8THORN

HSO3 thorn CH3OOH rarr SO4

2 thorn Hthorn thorn CH3OH ethM9THORN

HSO3 thorn CH3COOOH rarr SO4

2 thorn Hthorn thorn CH3COOH ethM10THORN

M5 Apparent heterogeneous uptake of SO2 on aerosol surfacesTo improve the model simulation for the winter haze events in Beijingand the NCP Zheng et al (7) suggested the use of an apparent heter-ogeneous uptake coefficient (g) of SO2 on aerosol surfaces as a functionof RH (Eq M11) g is defined as the ratio of the number of collisionsthat result in the reaction SO2 (g) + Aerosolrarr SO4

2minus to the theoreticaltotal number of collisions The overall rates of SO2 uptake and sulfateproduction are as fast as if the whole aerosol surface is covered by dust(44) Although the exact mechanism supporting this fast productionrate is still unknown (7) this apparent heterogeneous source of sulfatecan account for the underprediction of modeled sulfate with significantreduction of normalized model bias from minus542 to 63

g frac14 glow 0 le RH le 50glow thorn ethghigh glowTHORN=eth1ndash05THORN ethRH 50THORN 50 lt RH le 100

ethM11THORN

where glow = 2 times 10minus5 and ghigh = 5 times 10minus5With g a sulfate production rate RHg can be determined by Eq M12

RHg frac14 dfrac12SO42

dtfrac14 Rp

Dgthorn 4gn

1

Saerosol SO2frac12 ethM12THORN

where Rp is the radius of aerosol particles Dg is the gas-phase moleculardiffusion coefficient of SO2 n is the mean speed of gaseous SO2 mole-cules and Saerosol is the surface area concentration of aerosol particles

M6 ISORROPIA-II model calculationThe ISORROPIA-II model (15) was used to calculate the AWC andpH The ISORROPIA-II is a thermodynamic equilibrium model thatpredicts the physical state and composition of atmospheric inorganicaerosols It can be used in two modes the reverse mode and the for-ward mode The reverse mode calculated the thermodynamic equilib-rium based on aerosol-phase concentrations whereas the forwardmode relied on both aerosol-phase and gas-phase concentrations(16) Its ability in predicting AWC and pH has been demonstratedby Guo et al (45) and Xu et al (16)

To evaluate the aerosol pH and AWC we performed both reverse-mode and forward-mode model simulations and used their averagesfor further analyses The gaseous NH3 was not measured in our Jan-uary campaign but long-term measurement (46) shows a compactcorrelation between NH3 and NOx concentrations in the winter sea-son of Beijing Accordingly we estimated the NH3 concentration fromthe observed NOx concentration with an empirical equation derivedfrom Meng et al (46) that is NH3 (ppb) = 034 times NOx (ppb) + 063

5 of 11

SC I ENCE ADVANCES | R E S EARCH ART I C L E

on June 27 2019httpadvancessciencem

agorgD

ownloaded from

The contribution of organic compounds to AWC Worg (the massconcentration of aerosol water associated with organics) was esti-mated by the same approach of Guo et al (45)

Worg frac14 OMrorg

sdotrwsdotkorg

eth100=RH 1THORN ethM13THORN

whereOM is themass concentration of organicmatter rw is the densityof water (rw = 10 times 103 kg mminus3) rorg is the density of organics (rorg =14 times 103 kgmminus3) (45) and korg is the hygroscopicity parameter (47) oforganic aerosol compositionsWe adopted a korg of 006 based on pre-vious cloud condensation nuclei measurements in Beijing (48)

M7 Kinetics of mass transportFor multiphase reactions the overall reaction rate depends not only onthe rate of chemical reactions but also on the mass transport in differentmedium and across the interface To account for the effects of masstransport we adopted the formulationof a standard resistancemodel (12)

1RHaq

frac14 1Raq

thorn 1Jaqlim

ethM14THORN

where RHaq is the sulfate production rate Raq is the aqueous-phase reac-tion rate and Jaqlim is the limitingmass transfer rate For the oxidation ofS(IV) by a given oxidant Oxi (12)

Raq frac14 ethk0frac12SO2sdotH2O thorn k1frac12HSO3 thorn k2frac12SO3

2THORNfrac12Oxi ethM15THORN

where [SO2H2O] [HSO3minus] [SO3

2minus] and [Oxi] are the respective aqueous-phase concentrations and k0 k1 and k2 are the corresponding second-order reaction rate coefficients as detailed in table S4 The aqueous [Oxi]is assumed to be in equilibrium with its gas-phase concentration and canbe determined by Henryrsquos law (12)

frac12X frac14 pinfinethXTHORNsdotHethXTHORN ethM16THORN

where pinfin(X) (atm) is the partial pressure of speciesX in the bulk gas phaseand H (M atmminus1) is the effective Henryrsquos constant (table S5)

The limiting mass transfer rate Jaq (M sminus1) is calculated by EqsM17 and M18 (12)

Jaqlim frac14 minfJaqethSO2THORN JaqethOxiTHORNg ethM17THORN

JaqethXTHORN frac14 kMTethXTHORNsdotpinfinethXTHORNsdotHethXTHORN ethM18THORN

where X refers to SO2 or the oxidant Oxi such as O3 H2O2 and NO2The mass transfer rate coefficient kMT (s

minus1) can be calculated by (12)

kMT Xeth THORN frac14 Rp2

3Dgthorn 4Rp

3an

1

ethM19THORN

where Rp is the aerosol radius and Rp2

3Dgand 4Rp

3an are the continuumregime resistance and the free-molecular (or kinetic) regime resistance

Cheng et al Sci Adv 20162 e1601530 21 December 2016

respectively Dg is the gas-phase molecular diffusion coefficient and nis the mean molecular speed of X a is the mass accommodation co-efficient of X on the droplet surface which accounts for imperfectsticking of impinging molecules to the surface and we adopted liter-ature values of 011 023 20 times 10minus3 (12) and 20 times 10minus4 (49) for SO2H2O2 O3 and NO2 respectively Aqueous-phase mass transfer can beignored for the size range considered here (Dp le 25 mm) (12) Anequivalent Rp of 015 and 15 mm was assumed for aerosols and clouddroplets respectively

M8 Influences of ionic strength on aqueoussulfate-producing reactionsAerosol liquid water constitutes an aqueous electrolyte that can be ex-tremely concentrated with high ionic strength (I) up to 100 M (50)This highly concentrated chemical environment will affect the inclina-tion of a species to participate in the aqueous-phase chemical reac-tions reflected by an apparent reaction rate constant (k) differentfrom that in an ideal solution (21)

The influence of I on k is complicated and is not yet fully under-stood According to current understanding I affects k through itsintegrated effects on the activity coefficient (a) of reactants andproducts for most reactions (51) For example for a reaction wherereactants A and B form an activated complex (AB) which thenquickly decomposes into a final product P (A + B rarr (AB) rarr P)the k-I dependence can be described as (51)

logk

kIfrac140frac14 log aA thorn log aB log aethABTHORN ethM20THORN

where kI=0 refers to the kinetic constant at I of 0 M Theories used topredict ai at given I varied with the ranges of I and the species naturethat is being an ion or a neutral species as summarized in table S6

Major aqueous sulfate-producing reactions considered here includeS(IV) oxidation byH2O2O3 TMI+O2 andNO2 (Fig 2) Among thesereactions the influence of I has been studied experimentally with I rang-ing up to ~5M for H2O2 and slightly higher than 1M for O3 and TMI +O2 The observed k-I relationships of these reactions agree with theore-tical predictions summarized in table S7 (52ndash56) As shown in fig S3withincreasing ionic strength the rate constant forH2O2decreases firstwhen Iis below~1Mand begins to increase once I goes above ~1M (52 57) ForO3 the rate constant is positively related to I For TMI + O2 [here onlyFe(III) andMn(II) are considered as effective catalyzing TMIs (58 59)]the rate constant decreases significantly with increasing I even withoutconsidering the sulfate inhabitation effect (referring to the effect that theformation of the sulfate-TMI complex would reduce catalytically activeTMI concentrations) (55 60)

Currently however no k-I relationship was reported for NO2whereas some plausible estimation can bemade on the basis of the prin-ciple of theories discussed above Two kinds of mechanisms have beensuggested for theNO2-S(IV) reactions One is the oxygen-atom transferreaction (61)

2NO2 thorn SO23 harr ethO2N SO3 NO2THORN2 ethM21aTHORN

ethO2N SO3 NO2THORN2 thornOH harrHO SO3 ethNO2THORN2

3

ethM21bTHORN

6 of 11

SC I ENCE ADVANCES | R E S EARCH ART I C L E

on June 27 2019httpadvancessciencem

agorgD

ownloaded from

HO SO3 ethNO2THORN2

3thorn OH rarr 2NO2

thorn SO24 thornH2O

ethM21cTHORN

in which the first reaction (Eq M21a) is the rate-controlling stepThe other mechanism is an electron transfer reaction (62ndash64)

NO2 thorn SO23 rarr NO2

thorn SO3bull ethM22THORN

followed by the sulfur auto-oxidation processes (12) The first reaction(Eq M22) is the rate-controlling step because follow-up reactions withradicals are very fast

For both mechanisms the rate-controlling step of the NO2-S(IV)reaction is a reaction of an ion with a neutral molecule Accordingto Herrmann (51) the k-I relation for this type of reaction shouldfollow

logk

kIfrac140frac14 bI ethM23THORN

where b is the kinetic salting coefficient According to Kameoka andPigford (65) b is expected to be positive Given a positive b in-creasing I will lead to an increase in k as shown by the red dashedcurve in fig S3 (we take an arbitrary value of 05 Mminus1 for b which isnot determined from experiments and is used only for the purposeof illustration)

During the severe Beijing hazes (when PM25 is higher than300 mg mminus3) ionic strength in aerosol liquid water ranged from 13to 43 M as predicted by the ISORROPIA-II model (see section M6)(15) Direct extrapolation of the observedpredicted k-I relationship(fig S3) into such high ranges of ionic strength may not be appropriateThus the rate constants are taken as for diluted solution althoughbased on current observations and theories (table S7 and fig S3) thistreatment will lead to conservative estimations of the real sulfate pro-duction rates of the NO2 O3 and H2O2 pathways in Fig 2 but anoverestimation of the TMI + O2 pathway

M9 Data used in Fig 2Detailed descriptions on how to derive the sulfate production rate fordifferent aqueous-phase oxidation pathways of SO2 (that is by H2O2O3 TMI + O2 and NO2) in Fig 2 can be found in the study of Seinfeldand Pandis (12) The aqueous-phase reactions involved are listed as Re-actions R1 and M4 to M8 The rate expressions and rate coefficientsand the constants that we used for calculating the apparent Henryrsquosconstant are summarized in tables S4 and S5 respectively

In Fig 2 the ldquocloud dropletsrdquo scenario is taken from the work ofSeinfeld and Pandis (12) and Herrmann et al (21) [SO2(g)] = 5 ppb[NO2(g)] = 1 ppb [H2O2(g)] = 1 ppb [O3(g)] = 50 ppb [Fe(III)] =03 mM [Mn(II)] = 003 mM liquid water content = 01 g mminus3 andcloud droplet radius Rp = 15 mm In this scenario the temperature Tis taken to be the same as that used in the ldquoBeijing hazerdquo scenariodescribed below

The ldquoBeijing hazerdquo scenario was taken according to the measure-ment data during themost polluted haze periods (PM25 gt 300 mgm

minus3)The average values were used in our calculation [SO2(g)] = 40 ppb[NO2(g)] = 66 ppb [H2O2(g)] = 001 ppb [O3(g)] = 1 ppb AWC =300 mg mminus3 aerosol droplet radius Rp = 015 mm and T = 271 K Theconcentrations of [Fe(III)] and [Mn(II)] are pH-dependent (fig S4)

Cheng et al Sci Adv 20162 e1601530 21 December 2016

The pH dependence is mostly due to the precipitation equilibrium ofFe(OH)3 and Mn(OH)2

FeethIIITHORNfrac12 frac14 Ksp FeethOHTHORN3frac12OH3 and frac12MnethIITHORNsat frac14 Ksp MnethOHTHORN2

frac12OH2

ethM24THORN

where Ksp FeethOHTHORN3 and Ksp MnethOHTHORN2 are the precipitation constants ofFe(OH)3 and Mn(OH)2 respectively (66) When all Fe(OH)3 andMn(OH)2 are dissolved further decrease of pH will not increase[Fe(III)] and [Mn(II)] resulting in a plateau at low pH (fig S4) The totalsoluble Fe and Mn are estimated to be 18 and 42 ng mminus3 respectivelybased on data in the literature and observations in Beijing (2 59 66ndash70)

The severe haze events observed in Beijing are a regional phenom-enon (5) Because Beijing is located in the northwestern edge of thepolluted area the air pollution in cities south of Beijing is even moresevere NO2 concentrations in Beijing are comparable to the southerncities whereas SO2 concentrations in the latter are typically two tofour times higher than those in Beijing (fig S5) Thus the sulfate pro-duction rate from the NO2 pathway that we predicted with the Beijingdata in Fig 2 is a conservative estimation for the contribution of thispathway to the sulfate formation in the whole NCP

Considering that regional transport from cities south of Beijing hascontributed to the severe haze episode in January 2013 (5 71) and judgingfrom the air pollution trend shown in fig S5 the polluted air parcels couldtypically have been processed under severe haze conditions in Beijing andin cities south of it formore than 3 days before the peak pollution hour inBeijing Because cloud amounts were low (7 72 73) during the January2013 winter haze periods we assumed that aerosols spent 100 of theirlifetime under noncloud conditions (RH lt 100) (21) Under theseconditions to produce the observed sulfate concentration (~200 mg mminus3)the requiredaverageheterogeneousproduction rate canbedetermined tobe~3mgmminus3 hminus1 This estimation also falls into the ranges of 1 to 7mgmminus3 hminus1

that we predicted for the NO2 pathway in Fig 2

M10 Aerosol acidity in northern ChinaThe neutralized feature of aerosols in East Asia (including Beijing andthe NCP) has been well documented in the literature with high cation-to-anion ratios (45 74ndash86) Aircraft measurements in the Arctic furthersupport this conclusion showing that the aerosols in the Arctic charac-terized as coming from the outflowof EastAsiaweremostly neutralizedwhereas the aerosols transported from North America to the Arcticwere highly acidic (lines in Fig 1D) (87)

The low aerosol acidity in East Asia can be attributed to its highNH3 and mineral dust emissions For example in 2008 the emissionratios of NH3 to 2times SO2 (molar ratio of NH3 emission divided bytwice the SO2 emission) are 037 086 and 104 for North AmericaEurope and East Asia respectively (87) Northern China is expectedto have a higher atmospheric neutralizing capacity than other parts ofEast Asia because of the high NH3 emission in this region Bothsatellite data (88 89) and emission inventories (90) show that northernChina is one of the most NH3-rich areas in East Asia because of itsintense agriculture activities The NCP (that is the cities of Beijing andTianjin and the provinces of Hebei Henan and Shandong) accountsfor 30 to 40 of the total NH3 emissions in China (90) while con-tributing ~20 of the emissions for SO2 and NOx (91)

7 of 11

SC I ENCE ADVANCES | R E S EARCH ART I C L E

on June 27 2019httpadvancessciencem

agorgD

ownloaded from

The high mass fraction of mineral dust (10 to 20) is a distinctcharacteristic of PM25 in the NCP (2 70 92 93) The influence of min-eral aerosols is also higher for northern China than for southern China(94ndash98) Major sources of mineral aerosols include urban fugitive dust(resuspended road dust construction dust etc) and the long-rangetransported Asian dust (99ndash101) The mineral dusts are observed tobe internally mixed with sulfate nitrate and ammonia suggesting theirparticipation in atmospheric processing (102ndash105) Without mineralcomponents (Ca2+ Mg2+ K+ and Na+) aerosol pH in northern Chinamay drop below 56 showing an acidic nature (37)

In addition despite the high emission of acidic gases (SO2 andNOx) rainwater in northern China has average pH values higher than56 (fig S6) suggesting an alkaline tendency in contrast to other areas(for example the United States) (12 22) A pH of 56 is often taken asthe ldquonaturalrdquo acidity of rainwater (water in equilibrium with CO2)which has been considered as the demarcation line of acidic precipi-tation (12) The high pH values in rainwater of northern China arecaused by alkaline aerosol particles (12) which have a large bufferingcapacity to offset the effects of anthropogenic acidity

M11 Contribution from sCIs and NO3 radicalsWe have also investigated the reactions of S(IV) with NO3 radicalsand sCIs both of which show minor contributions (003 and069) compared to our proposed mechanism S(IV) + NO2(aq)Oxidation by NO3 radicalsAn average NO3 radical concentration of ~25 times 10minus3 parts per trillionwas determined by ourmodeling results for the severe pollutionperiodsTaking an effective Henryrsquos constant of 06 M atmminus1 for NO3 and a re-action rate constant of 14 times 109 Mminus1 sminus1 (106) we determined a sulfateproduction rate of 51 times 10minus4 mgmminus3 hminus1 for the NO3 reaction This rateis only 003 compared to the proposed mechanism S(IV) + NO2(aq)at pH 58 (Fig 2) and is thus negligibleOxidation by sCIAccording to Mauldin et al (31) we calculated RsCI the sulfate for-mation rate from the sCI mechanism by

RsCI frac14 ksCIthornSO2 frac12sCIfrac12SO2 ethM25THORN

where [sCI] and [SO2] are the concentrationsof sCI andSO2 respectivelyand the reaction rate coefficient ksCIthornSO2 is 6 times 10minus13 cm3 moleculeminus1 sminus1The concentration of sCI can be determined by the following equation (31)

frac12sCI frac14 YsCIkO3thornalkenefrac12O3frac12alkenetsCI ethM26THORN

where YsCI is the yield of sCI ~05 kO3thornalkene is the corresponding ratecoefficient for the reaction of O3 with individual alkenes and tsCI is thelifetime of the sCI which is 02 s (31) After taking a typical alkeneprofile for the haze period (107) we determined an RsCI of ~069 ofthe proposed S(IV) + NO2(aq) mechanism

SUPPLEMENTARY MATERIALSSupplementary material for this article is available at httpadvancessciencemagorgcgicontentfull212e1601530DC1fig S1 Weakened photochemistry by aerosol dimming effects during January 2013 in Beijingfig S2 Importance of the NO2 reaction pathway for sulfate production in the Beijing haze(January 2013)fig S3 Influence of ionic strength (I) on rate of aqueous sulfate-producing reactions

Cheng et al Sci Adv 20162 e1601530 21 December 2016

fig S4 Estimation of Fe3+ and Mn2+ concentrations as a function of aerosol water pH duringBeijing hazesfig S5 Regional pollution across the NCP during January 2013fig S6 Annual precipitation pH of China in 2013fig S7 The same as Fig 2 but with a lower limit of reaction rate constants reported by Lee andSchwartz (18)table S1 Previously reported concentrations of cations and anions in PM25 during winter forcities in NCP used in Fig 1Dtable S2 Summary of field observation and methods in this studytable S3 Domain configurations and major dynamic and physical options used in WRF v351table S4 Rate expression and rate coefficients of relevant aqueous-phase reactionstable S5 Constants for calculating the apparent Henryrsquos constant (H)table S6 Summary of suggested activity coefficient (a)ndashionic strength (I) dependencetable S7 Influence of ionic strength (I) on rate of aqueous sulfate-producing reactionsReferences (108ndash128)

REFERENCES AND NOTES1 P Brimblecombe The Big Smoke (Methuen 1987)

2 R-J Huang Y Zhang C Bozzetti K-F Ho J-J Cao Y Han K R DaellenbachJ G Slowik S M Platt F Canonaco P Zotter R Wolf S M Pieber E A BrunsM Crippa G Ciarelli A Piazzalunga M Schwikowski G Abbaszade J Schnelle-KreisR Zimmermann Z An S Szidat U Baltensperger I El Haddad A S H PreacutevocirctHigh secondary aerosol contribution to particulate pollution during haze events inChina Nature 514 218ndash222 (2014)

3 S Guo M Hu M L Zamora J Peng D Shang J Zheng Z Du Z Wu M ShaoL Zeng M J Molina R Zhang Elucidating severe urban haze formation in ChinaProc Natl Acad Sci USA 111 17373ndash17378 (2014)

4 R Zhang Q Li R Zhang Meteorological conditions for the persistent severe fog andhaze event over eastern China in January 2013 Sci China Earth Sci 57 26ndash35 (2014)

5 G J Zheng F K Duan H Su Y L Ma Y Cheng B Zheng Q Zhang T HuangT Kimoto D Chang U Poumlschl Y F Cheng K B He Exploring the severe winter hazein Beijing The impact of synoptic weather regional transport and heterogeneousreactions Atmos Chem Phys 15 2969ndash2983 (2015)

6 J Wang J Wang S Wang J Jiang A Ding M Zheng B Zhao D C Wong W ZhouG Zheng L Wang J E Pleim J Hao Impact of aerosolndashmeteorology interactions onfine particle pollution during Chinarsquos severe haze episode in January 2013 Environ ResLett 9 094002 (2014)

7 B Zheng Q Zhang Y Zhang K B He K Wang G J Zheng F K Duan Y L MaT Kimoto Heterogeneous chemistry A mechanism missing in current models to explainsecondary inorganic aerosol formation during the January 2013 haze episode in NorthChina Atmos Chem Phys 14 16731ndash16776 (2014)

8 R Zhang G Wang S Guo M L Zamora Q Ying Y Lin W Wang M Hu Y WangFormation of urban fine particulate matter Chem Rev 115 3803ndash3855 (2015)

9 D H Ehhalt F Rohrer Dependence of the OH concentration on solar UV J GeophysRes 105 3565ndash3571 (2000)

10 F Rohrer H Berresheim Strong correlation between levels of tropospheric hydroxylradicals and solar ultraviolet radiation Nature 442 184ndash187 (2006)

11 C Pilinis J H Seinfeld D Grosjean Water content of atmospheric aerosols AtmosEnviron 23 1601ndash1606 (1989)

12 J H Seinfeld S N Pandis Atmospheric Chemistry and Physics from Air Pollution toClimate Change (Wiley 2006)

13 B Ervens B J Turpin R J Weber Secondary organic aerosol formation in cloud dropletsand aqueous particles (aqSOA) A review of laboratory field and model studies AtmosChem Phys 11 11069ndash11102 (2011)

14 A Nenes S N Pandis C Pilinis Continued development and testing of a newthermodynamic aerosol module for urban and regional air quality modelsAtmos Environ 33 1553ndash1560 (1999)

15 C Fountoukis A Nenes ISORROPIA II A computationally efficient thermodynamicequilibrium model for K+-Ca2+-Mg2+-NH4

+-Na+-SO42minus-NO3

minus-Clminus-H2O aerosolsAtmos Chem Phys 7 4639ndash4659 (2007)

16 L Xu H Guo C M Boyd M Klein A Bougiatioti K M Cerully J R HiteG Isaacman-VanWertz N M Kreisberg C Knote K Olson A Koss A H GoldsteinS V Hering J de Gouw K Baumann S-H Lee A Nenes R J Weber N Lee NgEffects of anthropogenic emissions on aerosol formation from isoprene andmonoterpenes in the southeastern United States Proc Natl Acad Sci USA 112 37ndash42(2015)

17 T Koop B Luo A Tsias T Peter Water activity as the determinant for homogeneous icenucleation in aqueous solutions Nature 406 611ndash614 (2000)

18 Y N Lee S E Schwartz Precipitation Scavenging Dry Deposition and Resuspension(Elsevier 1983) pp 453ndash470

8 of 11

SC I ENCE ADVANCES | R E S EARCH ART I C L E

on June 27 2019httpadvancessciencem

agorgD

ownloaded from

19 X Huang Y Song C Zhao M Li T Zhu Q Zhang X Zhang Pathways of sulfateenhancement by natural and anthropogenic mineral aerosols in China J Geophys Res119 14165ndash14179 (2014)

20 C J Walcek G R Taylor A theoretical method for computing vertical distributions ofacidity and sulfate production within cumulus clouds J Atmos Sci 43 339ndash355 (1986)

21 H Herrmann T Schaefer A Tilgner S A Styler C Weller M Teich T Otto A CheTropospheric aqueous-phase chemistry Kinetics mechanisms and its coupling to achanging gas phase Chem Rev 115 4259ndash4334 (2015)

22 J Cao X Tie W F Dabberdt T Jie Z Zhao Z An Z Shen On the potential high aciddeposition in northeastern China J Geophys Res 118 4834ndash4846 (2013)

23 S N Pandis J H Seinfeld Mathematical modeling of acid deposition due to radiationfog J Geophys Res 94 12911ndash12923 (1989)

24 Y Xie A Ding W Nie H Mao X Qi X Huang Z Xu V-M Kerminen T PetaumljaumlX Chi A Virkula M Boy L Xue J Guo J Sun X Yang M Kulmala C Fu Enhancedsulfate formation by nitrogen dioxide Implications from in-situ observations at theSORPES Station J Geophys Res 120 12679ndash12694 (2015)

25 L Wang L Wen C Xu J Chen X Wang L Yang W Wang X Yang X Sui L YaoQ Zhang HONO and its potential source particulate nitrite at an urban site in NorthChina during the cold season Sci Total Environ 538 93ndash101 (2015)

26 H Su Y Cheng R Oswald T Behrendt I Trebs F X Meixner M O Andreae P ChengY Zhang U Poumlschl Soil nitrite as a source of atmospheric HONO and OH radicalsScience 333 1616ndash1618 (2011)

27 A Laskin D J Gaspar W Wang S W Hunt J P Cowin S D Colson B J Finlayson-PittsReactions at interfaces as a source of sulfate formation in sea-salt particles Science301 340ndash344 (2003)

28 B Alexander R J Park D J Jacob Q B Li R M Yantosca J Savarino C C W LeeM H Thiemens Sulfate formation in sea-salt aerosols Constraints from oxygenisotopes J Geophys Res 110 D10307 (2005)

29 W R Stockwell J G Calvert The mechanism of the HO-SO2 reaction Atmos Environ17 2231ndash2235 (1983)

30 M A Blitz K J Hughes M J Pilling Determination of the high-pressure limiting ratecoefficient and the enthalpy of reaction for OH+SO2 J Phys Chem A 107 1971ndash1978 (2003)

31 R L Mauldin III T Berndt M Sipilauml P Paasonen T Petaumljauml S Kim T Kurteacuten F StratmannV-M Kerminen M Kulmala A new atmospherically relevant oxidant of sulphur dioxideNature 488 193ndash196 (2012)

32 S Kim A Guenther B Lefer J Flynn R Griffin A P Rutter L Gong B K Cevik Potentialrole of stabilized Criegee radicals in sulfuric acid production in a high biogenic VOCenvironment Environ Sci Technol 49 3383ndash3391 (2015)

33 M Boy D Mogensen S Smolander L Zhou T Nieminen P Paasonen C Plass-DuumllmerM Sipilauml T Petaumljauml L Mauldin H Berresheim M Kulmala Oxidation of SO2 by stabilizedCriegee intermediate (sCI) radicals as a crucial source for atmospheric sulfuric acidconcentrations Atmos Chem Phys 13 3865ndash3879 (2013)

34 N Kaneyasu S Yamamoto K Sato A Takami M Hayashi K Hara K KawamotoT Okuda S Hatakeyama Impact of long-range transport of aerosols on the PM25

composition at a major metropolitan area in the northern Kyushu area of Japan AtmosEnviron 97 416ndash425 (2014)

35 Q Zhang F Duan K He Y Ma H Li T Kimoto A Zheng Organic nitrogen in PM25 inBeijing Front Environ Sci Eng 9 1004ndash1014 (2015)

36 G J Zheng Y Cheng K B He F K Duan Y L Ma A newly identified calculationdiscrepancy of the Sunset semi-continuous carbon analyzer Atmos Meas Tech 71969ndash1977 (2014)

37 J K Zhang Y Sun Z R Liu D S Ji B Hu Q Liu Y S Wang Characterization ofsubmicron aerosols during a month of serious pollution in Beijing 2013 Atmos ChemPhys 14 2887ndash2903 (2014)

38 L Xing T Fu J J Cao S C Lee G H Wang K F Ho M Cheng C You T J WangSeasonal and spatial variability of the OMOC mass ratios and high regional correlationbetween oxalic acid and zinc in Chinese urban organic aerosols Atmos Chem Phys 134307ndash4318 (2013)

39 G Tang Y Wang X Li D Ji S Hsu X Gao Spatial-temporal variations in surface ozonein Northern China as observed during 2009ndash2010 and possible implications for futureair quality control strategies Atmos Chem Phys 12 2757ndash2776 (2012)

40 C Mass D Ovens WRF model physics Progress problems and perhaps somesolutions the 11th WRF Usersrsquo Workshop 21ndash25 June NCAR Center Green Campushttpwww2mmmucareduwrfusersworkshopsWS2010presentationssession 44-1_WRFworkshop2010Finalpdf (2010) [accessed December 2015]

41 G Z Whitten G Heo Y Kimura E McDonald-Buller D T Allen W P L CarterG Yarwood A new condensed toluene mechanism for Carbon Bond CB05-TU AtmosEnviron 44 5346ndash5355 (2010)

42 J S Chang R A Brost I S A Isaksen S Madronich P Middleton W R StockwellC J Walcek A three-dimensional Eulerian acid deposition model Physical concepts andformulation J Geophys Res 92 14681ndash14700 (1987)

Cheng et al Sci Adv 20162 e1601530 21 December 2016

43 NASAJet Propulsion Laboratory Chemical Kinetics and Photochemical Data for Use inAtmospheric Studies (NASAJet Propulsion Laboratory 2010)

44 Y Zhang G R Carmichael The role of mineral aerosol in tropospheric chemistry in EastAsiamdashA model study J Appl Meteorol 38 354ndash366 (1999)

45 H Guo L Xu A Bougiatioti K Cerully S Capps J Hite A Carlton S Lee M BerginN Ng A Nenes R Weber Fine-particle water and pH in the southeastern United StatesAtmos Chem Phys 15 5211ndash5228 (2015)

46 Z Y Meng W L Lin X M Jiang P Yan Y Wang Y M Zhang X F Jia X L YuCharacteristics of atmospheric ammonia over Beijing China Atmos Chem Phys 116139ndash6151 (2011)

47 M D Petters S M Kreidenweis A single parameter representation of hygroscopicgrowth and cloud condensation nucleus activity Atmos Chem Phys 7 1961ndash1971 (2007)

48 S S Gunthe D Rose H Su R M Garland P Achtert A Nowak A WiedensohlerM Kuwata N Takegawa Y Kondo M Hu M Shao T Zhu M O Andreae U PoumlschlCloud condensation nuclei (CCN) from fresh and aged air pollution in the megacityregion of Beijing Atmos Chem Phys 11 11023ndash11039 (2011)

49 D J Jacob Heterogeneous chemistry and tropospheric ozone Atmos Environ 342131ndash2159 (2000)

50 Y Cheng H Su T Koop E Mikhailov U Poumlschl Size dependence of phase transitions inaerosol nanoparticles Nat Commun 6 5923 (2015)

51 H Herrmann Kinetics of aqueous phase reactions relevant for atmospheric chemistryChem Rev 103 4691ndash4716 (2003)

52 F Maaszlig H Elias K J Wannowius Kinetics of the oxidation of hydrogen sulfite byhydrogen peroxide in aqueous solution Ionic strength effects and temperaturedependence Atmos Environ 33 4413ndash4419 (1999)

53 H G Maahs Kinetics and mechanism of the oxidation of S(IV) by ozone in aqueoussolution with particular reference to SO2 conversion in nonurban tropospheric cloudsJ Geophys Res 88 10721ndash10732 (1983)

54 J Lagrange C Pallares P Lagrange Electrolyte effects on aqueous atmosphericoxidation of sulphur dioxide by ozone J Geophys Res 99 14595ndash14600 (1994)

55 L R Martin M W Hill The iron catalyzed oxidation of sulfur Reconciliation of theliterature rates Atmos Environ 21 1487ndash1490 (1987)

56 L R Martin M W Hill A F Tai T W Good The iron catalyzed oxidation of sulfur(IV) inaqueous solution Differing effects of organics at high and low pH J Geophys Res 963085ndash3097 (1991)

57 H M Ali M Iedema X-Y Yu J P Cowin Ionic strength dependence of the oxidation ofSO2 by H2O2 in sodium chloride particles Atmos Environ 89 731ndash738 (2014)

58 A Huss Jr P K Lim C A Eckert Oxidation of aqueous sulfur dioxide 1 Homogeneousmanganese(II) and iron(II) catalysis at low pH J Phys Chem 86 4224ndash4228 (1982)

59 B Alexander R J Park D J Jacob S Gong Transition metal-catalyzed oxidation ofatmospheric sulfur Global implications for the sulfur budget J Geophys Res 114D02309 (2009)

60 L R Martin T W Good Catalyzed oxidation of sulfur dioxide in solution The iron-manganese synergism Atmos Environ 25 2395ndash2399 (1991)

61 C L Clifton N Altstein R E Huie Rate constant for the reaction of nitrogen dioxidewith sulfur(IV) over the pH range 53-13 Environ Sci Technol 22 586ndash589 (1988)

62 T Nash The effect of nitrogen dioxide and of some transition metals on the oxidation ofdilute bisulphite solutions Atmos Environ 13 1149ndash1154 (1979)

63 R E Huie The reaction kinetics of NO2 Toxicology 89 193ndash216 (1994)64 G Spindler J Hesper E Bruumlggemann R Dubois T Muumlller H Herrmann Wet

annular denuder measurements of nitrous acid Laboratory study of the artefactreaction of NO2 with S(IV) in aqueous solution and comparison with fieldmeasurements Atmos Environ 37 2643ndash2662 (2003)

65 Y Kameoka R L Pigford Absorption of nitrogen dioxide into water sulfuric acidsodium hydroxide and alkaline sodium sulfite aqueous solutions Ind Eng ChemFundam 16 163ndash169 (1977)

66 T E Graedel C J Weschler Chemistry within aqueous atmospheric aerosols andraindrops Rev Geophys 19 505ndash539 (1981)

67 N Meskhidze W L Chameides A Nenes G Chen Iron mobilization in mineral dust Cananthropogenic SO2 emissions affect ocean productivity Geophys Res Lett 30 2085(2003)

68 F Solmon P Y Chuang N Meskhidze Y Chen Acidic processing of mineral dust ironby anthropogenic compounds over the north Pacific Ocean J Geophys Res 114D02305 (2009)

69 S-C Hsu G T F Wong G-C Gong F-K Shiah Y-T Huang S-J Kao F Tsai S-C C LungF-J Lin I-I Lin C-C Hung C-M Tseng Sources solubility and dry deposition ofaerosol trace elements over the East China Sea Mar Chem 120 116ndash127 (2010)

70 S Tian Y Pan Z Liu T Wen Y Wang Size-resolved aerosol chemical analysis ofextreme haze pollution events during early 2013 in urban Beijing China J HazardMater 279 452ndash460 (2014)

71 Z Wang J Li Z Wang W Yang X Tang B Ge P Yan L Zhu X Chen H Chen W WandJ Li B Liu X Wang W Wand Y Zhao N Lu D Su Modeling study of regional severe hazes

9 of 11

SC I ENCE ADVANCES | R E S EARCH ART I C L E

on June 27 2019httpadvancessciencem

agorgD

ownloaded from

over mid-eastern China in January 2013 and its implications on pollution prevention andcontrol Sci China Earth Sci 57 3ndash13 (2014)

72 B Zhang Y Wang J Hao Simulating aerosolndashradiationndashcloud feedbacks onmeteorology and air quality over eastern China under severe haze conditionsin winterAtmos Chem Phys 15 2387ndash2404 (2015)

73 Y Wang Q Zhang J Jiang W Zhou B Wang K He F Duan Q Zhang S Philip Y XieEnhanced sulfate formation during Chinarsquos severe winter haze episode in January 2013missing from current models J Geophys Res 119 10425ndash10440 (2014)

74 P K Quinn D J Coffman T S Bates T L Miller J E Johnson K Voss E J WeltonC Neusuumlss Dominant aerosol chemical components and their contribution toextinction during the Aerosols99 cruise across the Atlantic J Geophys Res 10620783ndash20809 (2001)

75 R J Park D J Jacob B D Field R M Yantosca M Chin Natural and transboundarypollution influences on sulfate-nitrate-ammonium aerosols in the United StatesImplications for policy J Geophys Res 109 D15204 (2004)

76 S-H Chu PM25 episodes as observed in the speciation trends network Atmos Environ38 5237ndash5246 (2004)

77 T K V Nguyen M D Petters S R Suda H Guo R J Weber A M G Carlton Trends inparticle phase liquid water during the Southern Oxidant and Aerosol Study AtmosChem Phys 14 10911ndash10930 (2014)

78 D Voutsa C Samara E Manoli D Lazarou P Tzoumaka Ionic composition of PM25 aturban sites of northern Greece Secondary inorganic aerosol formation Environ SciPollut Res 21 4995ndash5006 (2014)

79 U Makkonen A Virkkula J Maumlntykenttauml H Hakola P Keronen V Vakkari P P AaltoSemi-continuous gas and inorganic aerosol measurements at a Finnish urban siteComparisons with filters nitrogen in aerosol and gas phases and aerosol acidity AtmosChem Phys 12 5617ndash5631 (2012)

80 V-M Kerminen R Hillamo K Teinilauml T Pakkanen I Allegrini R Sparapani Ion balancesof size-resolved tropospheric aerosol samples Implications for the acidity andatmospheric processing of aerosols Atmos Environ 35 5255ndash5265 (2001)

81 M Sillanpaumlauml R Hillamo S Saarikoski A Frey A Pennanen U Makkonen Z SpolnikR Van Grieken M Braniš B Brunekreef M-C Chalbot T Kuhlbusch J SunyerV-M Kerminen M Kulmala R O Salonen Chemical composition and mass closure ofparticulate matter at six urban sites in Europe Atmos Environ 40 212ndash223 (2006)

82 A Ianniello F Spataro G Esposito I Allegrini M Hu T Zhu Chemical characteristics ofinorganic ammonium salts in PM25 in the atmosphere of Beijing (China) Atmos Chem Phys11 10803ndash10822 (2011)

83 L-x Yang D-c Wang S-h Cheng Z Wang Y Zhou X-h Zhou W-x Wang Influence ofmeteorological conditions and particulate matter on visual range impairment in JinanChina Sci Total Environ 383 164ndash173 (2007)

84 S Long J Zeng Y Li L Bao L Cao K Liu L Xu J Lin W Liu G Wang J Yao C MaY Zhao Characteristics of secondary inorganic aerosol and sulfate species in size-fractionated aerosol particles in Shanghai J Environ Sci 26 1040ndash1051 (2014)

85 S N Behera M Sharma Investigating the potential role of ammonia in ion chemistry offine particulate matter formation for an urban environment Sci Total Environ 4083569ndash3575 (2010)

86 E Stone J Schauer T A Quraishi A Mahmood Chemical characterization and sourceapportionment of fine and coarse particulate matter in Lahore Pakistan Atmos Environ44 1062ndash1070 (2010)

87 J A Fisher D J Jacob Q Wang R Bahreini C C Carouge M J Cubison J E DibbT Diehl J L Jimenez E M Leibensperger Z Lu M B J Meinders H O T PyeP K Quinn S Sharma D G Streets A van Donkelaar R M Yantosca Sourcesdistribution and acidity of sulfatendashammonium aerosol in the Arctic in winterndashspringAtmos Environ 45 7301ndash7318 (2011)

88 L Clarisse C Clerbaux F Dentener D Hurtmans P-F Coheur Global ammoniadistribution derived from infrared satellite observations Nat Geosci 2479ndash483 (2009)

89 A Boynard C Clerbaux L Clarisse S Safieddine M Pommier M Van DammeS Bauduin C Oudot J Hadji-Lazaro D Hurtmans P-F Coheur First simultaneous spacemeasurements of atmospheric pollutants in the boundary layer from IASI A case studyin the North China Plain Geophys Res Lett 41 645ndash651 (2014)

90 X Huang Y Song M Li J Li Q Huo X Cai T Zhu M Hu H Zhang A high-resolutionammonia emission inventory in China Global Biogeochem Cycles 26 GB1030 (2012)

91 Q Zhang D Streets G Carmichael K B He H Huo A Kannari Z Klimont I S ParkS Reddy J S Fu D Chen L Duan Y Lei L T Wang Z L Yao Asian emissions in 2006for the NASA INTEX-B mission Atmos Chem Phys 9 5131ndash5153 (2009)

92 F Yang J Tan Q Zhao Z Du K He Y Ma F Duan G Chen Q Zhao Characteristicsof PM25 speciation in representative megacities and across China Atmos Chem Phys 115207ndash5219 (2011)

93 J-j Cao Q-y Wang J C Chow J G Watson X-x Tie Z-x Shen P Wang Z-s AnImpacts of aerosol compositions on visibility impairment in Xirsquoan China Atmos Environ59 559ndash566 (2012)

Cheng et al Sci Adv 20162 e1601530 21 December 2016

94 T Okuda J Kato J Mori M Tenmoku Y Suda S Tanaka K He Y Ma F Yang X YuF Duan Y Lei Daily concentrations of trace metals in aerosols in Beijing Chinadetermined by using inductively coupled plasma mass spectrometry equipped withlaser ablation analysis and source identification of aerosols Sci Total Environ 330145ndash158 (2004)

95 J Duan J Tan S Wang J Hao F Chai Size distributions and sources of elements inparticulate matter at curbside urban and rural sites in Beijing J Environ Sci 24 87ndash94 (2012)

96 G Zhang X Bi L Y Chan X Wang G Sheng J Fu Size-segregated chemicalcharacteristics of aerosol during haze in an urban area of the Pearl River Delta regionChina Urban Clim 4 74ndash84 (2013)

97 Y L Sun Z F Wang P Q Fu T Yang Q Jiang H B Dong J Li J J Jia Aerosolcomposition sources and processes during wintertime in Beijing China Atmos ChemPhys 13 4577ndash4592 (2013)

98 Y Wang G Zhuang X Zhang K Huang C Xu A Tang J Chen Z An The ion chemistryseasonal cycle and sources of PM25 and TSP aerosol in Shanghai Atmos Environ40 2935ndash2952 (2006)

99 J-J Cao Z-X Shen J C Chow J G Watson S-C Lee X-X Tie K-F Ho G-H WangY-M Han Winter and summer PM25 chemical compositions in fourteen Chinese citiesJ Air Waste Manag Assoc 62 1214ndash1226 (2012)

100 H Wang S-C Tan Y Wang C Jiang G-y Shi M-X Zhang H-Z Che A multisourceobservation study of the severe prolonged regional haze episode over eastern China inJanuary 2013 Atmos Environ 89 807ndash815 (2014)

101 X Y Zhang Y Q Wang T Niu X C Zhang S L Gong Y M Zhang J Y Sun J BrandtAtmospheric aerosol compositions in China Spatialtemporal variability chemicalsignature regional haze distribution and comparisons with global aerosols AtmosChem Phys 12 779ndash799 (2012)

102 Y Hu J Lin S Zhang L Kong H Fu J Chen Identification of the typical metal particlesamong haze fog and clear episodes in the Beijing atmosphere Sci Total Environ511 369ndash380 (2015)

103 H Hwang C-U Ro Direct observation of nitrate and sulfate formations from mineraldust and sea-salts using low-Z particle electron probe X-ray microanalysis AtmosEnviron 40 3869ndash3880 (2006)

104 L Li M Li Z Huang W Gao H Nian Z Fu J Gao F Chai Z Zhou Ambient particlecharacterization by single particle aerosol mass spectrometry in an urban area ofBeijing Atmos Environ 94 323ndash331 (2014)

105 Q Yuan W Li S Zhou L Yang J Chi X Sui W Wang Integrated evaluation of aerosolsduring haze-fog episodes at one regional background site in North China Plain Atmos Res156 102ndash110 (2015)

106 Y Rudich R K Talukdar A R Ravishankara Multiphase chemistry of NO3 in the remotetroposphere J Geophys Res 103 16133ndash16143 (1998)

107 S Guo J Tan J Duan Y Ma F Yang K He J Hao Characteristics of atmosphericnon-methane hydrocarbons during haze episode in Beijing China Environ MonitAssess 184 7235ndash7246 (2012)

108 R Zhang J Jing J Tao S-C Hsu G Wang J Cao C S L Lee L Zhu Z Chen Y ZhaoZ Shen Chemical characterization and source apportionment of PM25 in BeijingSeasonal perspective Atmos Chem Phys 13 7053ndash7074 (2013)

109 K He Q Zhao Y Ma F Duan F Yang Z Shi G Chen Spatial and seasonal variability ofPM25 acidity at two Chinese megacities Insights into the formation of secondaryinorganic aerosols Atmos Chem Phys 12 1377ndash1395 (2012)

110 Y Sun G Zhuang Y Wang L Han J Guo M Dan W Zhang Z Wang Z Hao Theair-borne particulate pollution in BeijingmdashConcentration composition distribution andsources Atmos Environ 38 5991ndash6004 (2004)

111 X Gao L Yang S Cheng R Gao Y Zhou L Xue Y Shou J Wang X Wang W Nie P XuW Wang Semi-continuous measurement of water-soluble ions in PM25 in Jinan ChinaTemporal variations and source apportionments Atmos Environ 45 6048ndash6056 (2011)

112 M-D Chou M J Suarez C-H Ho M M-H Yan K-T Lee Parameterizations for cloudoverlapping and shortwave single-scattering properties for use in general circulationand cloud ensemble models J Climate 11 202ndash214 (1998)

113 E J Mlawer S J Taubman P D Brown M J Iacono S A Clough Radiative transfer forinhomogeneous atmospheres RRTM a validated correlated-k model for the longwaveJ Geophys Res 102 16663ndash16682 (1997)

114 A Xiu J E Pleim Development of a land surface model Part I Application in amesoscale meteorological model J Appl Meteorol 40 192ndash209 (2001)

115 J E Pleim A combined local and nonlocal closuremodel for the atmospheric boundary layerPart I Model description and testing J Appl Meteorol Clim 46 1383ndash1395 (2007)

116 J S Kain The KainndashFritsch convective parameterization An update J Appl Meteorol 43170ndash181 (2004)

117 S-Y Hong J-O J Lim The WRF single-moment 6-class microphysics scheme (WSM6)J Korean Meteor Soc 42 129ndash151 (2006)

118 T Ibusuki K Takeuchi Sulfur dioxide oxidation by oxygen catalyzed by mixtures ofmanganese(II) and iron(III) in aqueous solutions at environmental reaction conditionsAtmos Environ 21 1555ndash1560 (1987)

10 of 11

SC I ENCE ADVANCES | R E S EARCH ART I C L E

Dow

n

119 P Debye E Huumlckel The interionic attraction theory of deviations from ideal behavior insolution Z Phys 24 185 (1923)

120 P Debye W McAuley The electric field of the ions and the neutral salt effect Physik Z26 22 (1925)

121 C W Davies T Shedlovsky Ion association J Electrochem Soc 111 85Cndash86C (1964)122 H C Helgeson T H Brown A Nigrini T A Jones Calculation of mass transfer in

geochemical processes involving aqueous solutions Geochim Cosmochim Acta 34569ndash592 (1970)

123 K S Pitzer Electrolyte theorymdashImprovements since Debye and Hueckel Acc Chem Res10 371ndash377 (1977)

124 K S Pitzer Activity Coefficients in Electrolyte Solutions (CRC Press 1991)125 F J Millero J B Hershey G Johnson J-Z Zhang The solubility of SO2 and the

dissociation of H2SO3 in NaCl solutions J Atmos Chem 8 377ndash389 (1989)126 C Bretti F Crea C Foti S Sammartano Solubility and activity coefficients of acidic and

basic nonelectrolytes in aqueous salt solutions 2 Solubility and activity coefficients ofsuberic azelaic and sebacic acids in NaCl(aq) (CH3)4NCl(aq) and (C2H5)4NI(aq) atdifferent ionic strengths and at t = 25degC J Chem Eng Data 51 1660ndash1667 (2006)

127 C Bretti F Crea C De Stefano S Sammartano G Vianelli Some thermodynamicproperties of DL-tyrosine and DL-tryptophan Effect of the ionic medium ionic strengthand temperature on the solubility and acidndashbase properties Fluid Phase Equilib 314185ndash197 (2012)

128 A J Rusumdar ldquoTreatment of non-ideality in the multiphase model SPACCIM andinvestigation of its influence on tropospheric aqueous phase chemistryrdquo thesisBrandenburg University of Technology Cottbus Germany (2013)

Cheng et al Sci Adv 20162 e1601530 21 December 2016

AcknowledgmentsFunding This study was supported by the Max Planck Society (MPG) the National NaturalScience Foundation of China (21107061 21190054) and the EU project BACCHUS (603445)and PEGASOS (265148) YC thanks the MPG Minerva program GZ thanks ChineseScholarship Council for financial support of her study at the Max Planck Institute for ChemistryAuthor contributions KH YC and GZ proposed the initial idea YC and HS designedand led the study GZ YC and HS conducted the data analyses CW QM and BZperformed the model simulation QZ and KH provided the field observation and supportedthe model analyses ZW and MG supported the model analyses YC HS GZ and UPinterpreted the data YC HS UP GZ and GC wrote the manuscript with inputsfrom all coauthors Competing interests The authors declare that they have no competinginterests Data and materials availability All data needed to evaluate the conclusionsin the paper are present in the paper andor the Supplementary Materials Additional datarelated to this paper may be requested from the authors

Original submission 17 December 2015Transferred submission 7 July 2016Accepted 30 November 2016Published 21 December 2016101126sciadv1601530

Citation Y Cheng G Zheng C Wei Q Mu B Zheng Z Wang M Gao Q Zhang K HeG Carmichael U Poumlschl H Su Reactive nitrogen chemistry in aerosol water as a source ofsulfate during haze events in China Sci Adv 2 e1601530 (2016)

loa

11 of 11

on June 27 2019httpadvancessciencem

agorgded from

ChinaReactive nitrogen chemistry in aerosol water as a source of sulfate during haze events in

Carmichael Ulrich Poumlschl and Hang SuYafang Cheng Guangjie Zheng Chao Wei Qing Mu Bo Zheng Zhibin Wang Meng Gao Qiang Zhang Kebin He Gregory

DOI 101126sciadv1601530 (12) e16015302Sci Adv

ARTICLE TOOLS httpadvancessciencemagorgcontent212e1601530

MATERIALSSUPPLEMENTARY httpadvancessciencemagorgcontentsuppl20161219212e1601530DC1

REFERENCES

httpadvancessciencemagorgcontent212e1601530BIBLThis article cites 121 articles 5 of which you can access for free

PERMISSIONS httpwwwsciencemagorghelpreprints-and-permissions

Terms of ServiceUse of this article is subject to the

registered trademark of AAASis aScience Advances Association for the Advancement of Science No claim to original US Government Works The title

York Avenue NW Washington DC 20005 2017 copy The Authors some rights reserved exclusive licensee American (ISSN 2375-2548) is published by the American Association for the Advancement of Science 1200 NewScience Advances

on June 27 2019httpadvancessciencem

agorgD

ownloaded from