Diurnal variations of water-soluble ions in PM2.5 in Shanghai

J Shanghai Univ (Engl Ed), 2010, 14(4): 235 -240Digital Object Identifier(DOI): 10.1007/s11741-010-0636-2Diurnal variations of water-soluble ions in PM2.5 in ShanghaiFENG Jia-liang (冯加良)，GUAN Jing-jing (管晶晶), . GU Ze-ping (顾泽平)School of Environmental and Chemical Engineering, Shanghai University, Shanghai 200444, P. R. China(Communicated by JIAO Zheng)OShanghai University and Springer-Verlag Berlin Heidelberg 2010Abstract Thirty-six daily time interval PM2.5 samples were collected in diferent seasonal dates in urban Shanghai, and theconcentrations of four anions (C1 , NOg, so2- , C202- ) and five cations (NHt, Na+, k+, Ca2+, Mg2+) were analyzed withion chromatography. Sulfate, nitrate and ammonium were found to be the dominant species, accounting for about 80% ofthe total ions. The daily nitrate to sulfate mass ratio ranged from 0.31 to 0.82, indicating that coal combustion was still themain pollution source in Shanghai. The equivalent ratio of ammonium to sum of nitrate and sulfate showed fixed diurnalvariation pattern in all the sampling days with higher values in the nighttime, suggesting that fine particles in the night weremore neutralized. The oxalate to sulfate ratio was lower in the winter sampling days than that in hotter summer and autumnsampling days. Oxalate was significantly correlated with sulfate in winter sampling days, but not in the summer and autumn,suggesting that the formation mechanism of oxalate and sulfate was similar in winter, however different in hot days.Keywords PM2.5, water -oluble ions, diurnal variation, ShanghaiIntroductiongreatly in recent years in east Asia due to the increasingFine particles (less than 2.5 μm in diameter, alsoemissions of anthropogenic pollutants and the decreas-called as PM2.5) in the atmosphere are of more anding concentration of coarse particlesl8.more concern because they are hazardous to humanShanghai is the biggest industrial and commercialhealth4l. Fine particles can also scatter sunlight andcity in China. The rapid economic progress in the lastcause visibility degradation and global climate changel2].several decades has laid great pressure on the atmo-Water-soluble ionic species are a major part of thespheric environment. Great efforts have been made toaerosol mass, accounting for about half of the PM2.5 incontrol air pollution in China by the government in themany casesl3- 4. Water soluble species are hygroscopiclast two decades. The atmospheric loading of total sus-and can be the cloud condensation nuclei (CCN)5.pended particles (TSP) and inhalable particles (PM1o)Study on the composition and formation mechanism andhave been decreased since the 1990s. However the oC-sources of water soluble species is important to the un-currence frequency of haze or low visibility is increas-derstanding and controlling of urban haze and globaling in the recent years and is of more and more con-climate change.cern. As water-soluble ions in fine particles are of greatSulfate (S0?-)， nitrate (NOz) and ammoniumimportance in solar radiation scattering and visibility(NH+) are the most abundant species among water-degradation, their compositions and possible formationsoluble ions. Atmospheric sulfate and ammonium mainlymechanisms and sources have been widely studied inexist in fine particles and sulfate is mainly associatedShanghai in recent years based on integrated sampleswith ammonium3). Nitrate may be found in both fineof 12 48 h4.9-11I. Unfortunately, the formation mecha-particles and coarse particles6. Gaseous nitric acidnism is still not clear, partly due to the similarity of the(HNO3) can react with sea-salt particles to form coarseintegrated daily samples collected in different days of themode sodium nitrate. While nitrate formed by reac-same season. Studies on samples with higher time reso-tion of HNO3 with NH3 is in fne particles, NHNO3 islution is believed to give better insight into the sourcesvolatile and likely dissociates to gaseous HNO3 and NH3of the air borne fine particles and atmospheric reactions.at low relative humidity (RH) and high temperaturel1.In this study, we analyzed the diurnal variation of theIt was found that the fine mode NH4NO3 increasedionic species in fine particles in Shanghai based on dailyReceived Nov.19, 2009; Revised Dec.30, 2009中国煤化工Project supported by the National Natural Science Foundation of CIthe Shanghai LeadingAcademic Discipline Project (Grant No.S30109)MHCNMHGCorresponding author FENG Jia-liang, Ph D, Assoc Prof, E-mail: fengjiaiangoshu.edu.cn236J Shanghai Univ (Engl Ed), 2010, 14(4): 235 240time interval samples of about 4 h and discussed theC202-) and five cations (NH , Na+ , K+, Ca2+, Mg2+)impacting factors on the composition of ions.were analyzed by a dual channel ion chromatography(Metrobm IC, Switzerland). Anions and cations were1 Experimentsanalyzed by using Metrosep A SUPP 5 column with anA SUPP 5 guard column, and Metrosep C2 column with1.1 Sampling sitea C2 guard column respectively. The eluents were 3.2The sampling site is located in the Yanchang Cam-mmol/L Na2CO3/1.0 mmol/L NaHCO3 for the anionspus of Shanghai University in Zhabei District. The sam-and 4.0 mmol/L tartalic acid/0.75 mmol/[L dipicolinicpler was placed on the rooftop of a nine-storey buildingacid for the cations. The relative standard deviation of(about 30 m above ground). This site is surrounded byeach ion was less than 5% for reproducibility test. Thelow rise office and residential buildings. A road withdetection limits (S/N = 3) were less than 0.05 mg/Lmoderate traffic is about 300 m away to the north, andfor all the measured ions. We focused on the ions ofanother road with heavy traffic lies about 800 m awayNO3，so2-，C202- and NHt, whose concentrationsto the west. The sampling site is representative of thewere relatively high and the background contaminationurban atmospheric environment in Shanghai.. were very low.1.2 SamplingSamples were collected on glass fiber filters (What-2 Results and discussionman, GM-A 20.3 cmx25.4 cm, baked at 450 °C for4 h2.1 Concentrations and compositions ofbefore use) with high-volume PM2.5 samplers (Graseby,ions in PM2.5GMWT 2200), and the flow rate was 1.13 m3/min. Af-Sulfate, nitrate and ammonium are the dominantter sampling, the filters were wrapped with annealedions detected in the PM2.5 samples in Shanghai, ac-aluminum foil and stored in a refrigerator at -20 °C tillcounting for about 80% of the total ions. Concentra-analysis.tions of sulfate, nitrate, ammonium, oxalate and someSix sampling events in different seasons were con-important indices are listed in Table 1. Sulfate had theducted. They were December 18 and 19, 2006, Februaryhighest concentrations on all the sampling days, followed9 and 10, May 30 and 31, August 23 and 24, Novemberby nitrate and ammonium. The high concentrations of8 and 9, 2007 and January 22 and 23, 2008. Samplessulfate during our sampling events (6.99 -39.56 ug/m3)were collected during six daily time intervals: afternoonindicate that the pollution level in Shanghai is still quite(13:30 p.m. to 16:00 p.m.), night fll (16:30 p.m. tohigh because sulfate alone would exceed the US NAAQS19:30 a.m.), overnight (20:00 p.m. to 11:30 p.m.) earlyannual PM2.5 standard of 15 μg/m3. Water solublemorning (0:00 a.m. to 6:00 a.m.), morning (6:30 a.m.ions in the fine particles are mainly formed through sec-to 10:00 a.m.) and midday (10:15 a.m. to 13:20 p.m.).ondary reactions and the formation rate is higher in hot1.3 Sample analysisseasons such as late spring and summer than that in win-Small samples (6.0 cm2 each) punched from the sam-ter because of the temperature effect. Thus the summerpled filters and field blank samples were ultrasonicallydays do not necessarily have lower concentrations thanextracted with 10 mL nano-pure water (resistivity>the winter days. For example, the daily average concen-18 MN/cm). After passing through membrane filterstration of sulfate on May 30-31, 2007 is higher than thatwith pore size of 0.45 um, four anions (C1- , NO3, So2-，on December 18-19, 2007 and January 22-23, 2008.Table 1 Daily average concentrations of ions and sorme indices in PM2.5 in ShanghaiSampling dateDec. 18-19, 2006 Feb. 09 10, 2007 May 30-31, 2007 Aug. 23 -24, 2007 Nov. 08 09, 2007 Jan. 22 -23, 2008Mean temperature/9C10.0011.0025.0031.0020.005.00NOz /ug.m-38.7426.776.0613.49sO2- /ug.m-312.8339.5619.5611.5416.36C202- /ugm -30.140.370.320.270.350.19NHf /ugm-37.1719.791.574.969.14NO3 /So2-0.68.680.310.630.82NH]1/([NOz1+[So])*0.980.880.620.49.0.770.91(NHt]/[SO2 ]1.491.33中国煤化工000C202- /SO20.010.02Note: *INHJ], (NOg1 and (so2-] are in unit of equivalent nano moles per (MHCNMHG-J Shanghai Univ (Eng! Ed), 2010, 14(4): 235 240237The concentrations of nitrate and ammonium in theevaporation of particulate nitrate under high ambientPM2.5 samples collected in cold days in Shanghai aretemperaturel7.obviously higher than that in hot summer days (see Ta-The SO- mass ratio at our sampling site shows anble 1). It is in accordance with the previous report onthe ions in PM2.5 in Shanghail4l and might be caused byincreasing tendency in the nighttime, from the eveningthe shift of the gas-particle of the semi-volatile ammo-to the next early morning, for example, the ratio in-nium nitrate under different ambient temperatures. Thecreases from 0.50 during the overnight (20:00 p.m.-11:30fine-mode ammonium nitrate concentrations are closelyp.m.) to 0.57 in the early morning (0:00 a.m. -6:00 a.m.)related to gas phase NHg and HNO3 concentrations viaand to 0.62 in the morning (6:20 a.m. -10:00 a.m.) onDecember 18, 2006, and from 0.21 to 0.35 on August 23，the following equation:2007. It should be caused by the shift of the gas particleNH4NO3(aerosol)→HNO3(g) + NH3(g).phase of nitrate to the solid- particle phase with the de~At high temperature, most of the ammonium ni-crease of ambient temperature. The higher SOt- ratiotrate will dissociate and exist as gaseous nitric acid andin the morning should be caused by the higher emis-ammonial7. The observed seasonal change of the ammo-sion rate of nitrogen oxides from engine exhaust duringnium and nitrate in the fine particles are in accordancethe morning rush hours. Thus the formation of nitratewith the seasonal change of ambient temperature. Sim-would be higher.ilar results are also reported by other researchers(4,9Though the NO2- ratio has diferent values in difer-Obvious diurnal variations for the concentrations ofionic species can be found in some of the sampling days,ent sampling days and different daily time intervals, thefor example, the highest concentration of the total ionssum of nitrate and sulfate are significantly correlatedon December 18-19， 2006 is about two times of thewith ammonium in all of our sampling days (see Fig.1),lowest concentration, however the diurnal variations forsuggesting that they are mainly combined with ammo-[NH1the summer and autumn sampling days are relativelynium. From theNOJI+ISO Iratio (see Table 1), wesmall. Generally speaking, the diurnal variations ofcould see that nitrate and sulfate are almost fully neu-the ionic species are much smaller compared with thattralized by ammonium in winter, suggesting tbat theof the primary pollutants such as polycyclic aromaticfine particles in winter are neutral. The nitrate and sul-hydrocarbonsl12. As many of the ionic species such asfate is not totally neutralized. In summer and autumn,nitrate, sulfate and ammonium are from secondary reac-suggesting that either the fine particles are acidic or parttions, the concentrations of these ions will be affected byof the sulfate and nitrate is combined with other speciesthe formation rate besides the meteorological conditionssuch as calciuin. From the equivalent ratio of Sotsuch as temperature and wind speed. Though the loweratmospheric mixing height in the nighttime will cause(see Table 1), ammonia in summer could not fully neu-the accumulation of pollutants, the formation rate oftralize sulfuric acid to (NH4)2SO4. Thus part of thethe secondary species such as nitrate and sulfate in thesulfate in fine particles in summer might be in form ofnight is lower. It is thus not surprising that the diurnalbisulfate or sulfuric acid. The formation of nitrate isvariation of the concentrations of the measured ions isnot favored under acidic conditionsl5] and insufficientnot profound, and no fixed diurnal variation pattern isammonia should also be a reason for the low nitrateconcentration in summer in Shanghai.found in our study.2.2 Sources and formations of so2- andNO3 in PM2.5120Mass ratio CO has been used to indicate the relary- 3.14x-4.06公100R*=0.96tive importance of stationary U8. mobile sources. Coal30 tburning emissions have ratio of less than one while theratio for mobile source is larger than one(13-14. The60 Fdaily average O- ratios of our study are 0.68 to 0.820ton the winter sampling days and 0.31 on the summer20-sampling days (see Table 1), indicating that coal burn-ing is more important than the vehicle emission sources.中国煤化工2530Our result is in accordance with the previously reportedresulsl911, and in agreement with the fact that coal.MHCNMHGis still the main energy source in Shanghai. The lowFig.I Correlation Detween Nni and (NO3 +S02-) invalue of the SO ratio in summer will be caused by thePM2.5 in Shanghai238J Shanghai Univ (Engl Ed), 2010, 14(4): 235 -2402.3 Diurnal variations of theNHt1Moderate positive correlationbetween the[NO3]+|SO万[NHt 1equivalent ratio[NO3 }+[So万ratio and relative humidity is found (seeAs sulfate, nitrate and ammonium are the dominantFig.3), suggesting that sulfate and nitrate in fine par-species in the water-soluble ions. It is usually assumedticles are more neutralized at higher relative humidity.that they control the acidity of the fine particles[16- 17].Ionic species are hygroscopic, fine particles will containmore water under higher relative humidity. The ob-From the diurnal variation of theNO:)+ISO万INHt1; equiva-served positive correlation can be partly explained bylent ratio, we can estimate the change of the extent ofthe dependence of the formation of ammonium nitrateneutralization of sulfate and nitrate by armmonia anon the relative humidity. When ammonium nitrate isthe change of the particle acidity. In all our samplingin solution instead of in solid phase, the dissociation ofevents, the ratio increased from afternoon to early nextammonium nitrate will be less likely to occur and themorning then decrease (see Fig.2), indicating that theequilibrium will shift to the left side (aerosol phase offine particles in the nighttime are more neutralized than(1) in Section 2.1)[19]. Our results suggest that, at giventhat in the daytime. The enmission rate of ammonia istemperature and emission conditions, more ammoniumnormally lower in the nighttime because of the lowernitrate will be formed at higher relative humidity. Fur-temperaturel18]. The nitrate and sulfate would be lessther more, the increased formation of fine-mode ammo-neutralized in the nighttime if other conditions kept un-nium nitrate under high humidity will increase the lightchanged. Thus the diurnal variation of the:NO31+[SO万[NH+ ]scattering and reduce atmospheric visibility.ratio in Shanghai should not be caused by the diurnal1.+ 2006-12-18variation of ammonia. Several other causes can be re-_1.20- + 2007-02-08sponsable for this phenomenon. Firstly, the photochem-6 100 + 2070-030ical formation rate of nitrate and sulfate in the night-0.80-time will be much lower than that in the daytime due运信0.to the lower temperature and low radiation strength.艺0.40+ 20-1108Secondly, the emission rate of NO and NO2, the pre-0.20+ 200-08-3cursors of nitrate, is lower in the night due to the low1-:0-16.0016.30 2011000-6:00 6:20-10:0010:30-130traffic emission. The amount of nitric acid and sulfu-Time intervalric acid formed in the nighttime should thus be smallerthan that in the daytime. Then they can be easily neu-Fig.2 Diurnal variation ofNOg1+[90万INH1ratio in PM2.s intralized by ammonia. In the daytime, the ammonia inShanghaithe air might be not enough to neutralize all the sulfu-1.40ric acid and nitric acid, and some of the nitric acid will1.20y-0.008x-0.156be absorbed by the mineral particles to form sodium厂1.00R2=0.34nitrate or calcium nitrate. Thirdly, the partitioning of员0.80ammonium nitrate is in favor of the particle phase at善0.lower temperature.There is also another possible cause of the diurnalvariation of the[NOZ1+[SO2-1. (NHt1ratio. It is the vertical0.00405060708090100transport of pollutants. As ammonia is mainly emit-RH/%ted by the near-ground sources such as agriculture andFig.3 Correlation between ambient RH and thelivestock-farming, the fine particles at higher altitude,INOj1+10万[NH+ratio in PM2.5 in Shanghaiwhich is mainly formed by precursors from stationarysources such as power plants, will be more acidic than2.4 Oxalate in PM2.5 in Shanghaithat at lower altitude. During the daytime, the atmo-Oxalate is the dominant dicarboxylic acid in finespheric mixing layer is higher due to the higher tem-particles. Previous studies find that oxalate (C202- )perature and stronger solar radiation. The pollutantsare mainly from secondary reactions in urban areas offrom the higher altitude will mix and reach the nearChina though it can also be emitted directly from vehi-ground layer more easily and make the collected parti-clertions of oxalate in ourcles less neutralized. During the nighttine, the atmo-stud!中国煤化工0.54 ug/m3, and thespheric mixing height is usually low, and the transporta-tion of particles from high altitude to low altitude is notoxalaMYHCNMHGable1)islowerintheeasy or very slow. More detailed study on the verticalwinter sampling days than that in the summer and au-movement of pollutants is needed in the future.tumn sampling days. It is quite interesting to note thatJ Shanghai Univ (Engl Ed), 2010, 14(4): 235 240oxalate is highly correlated with nitrate and sulfate inModerate correlations between oxalate and nitrate canthe winter sampling days, while no significant correla-be found in the sarmpling days in summer and autumn,tion between oxalate and sulfate is found in summer andsuggesting that they are formed in similar ways. Pre-autumn (see Fig.4), suggesting that the formation mech-vious study shows that sulfate in Shanghai in summeranism of oxalate and sulfate is similar in winter, howevershould be formed by in-cloud processesl. Thus oxalatedifferent in the hot days such as summer and autumn.might be formed by gas phase photochemical reactions.60y-77.33x-1.76y= 188-3.5113R2=0.84R2=0.92。品4030o %6言3020-告o0叶%0.0.2 0.3.4.5.60.10.4 0.5Oralate/(ugm-)Oxlate/(ugm~)25 ry=25.29x-2.39、25y 6.63x-10.44R2=0.39R*=0.0820s-E 12o叶105F。。。g。。。O.D30.40.6Oralate/(ugm)Orxlate(ugm~)((dFig.4 Correlations of oxalate with nitraste and sulfate in PM2.s in Shanghai () and (b) were for the winter samples; (c)and (d) for the summer and autumn sarmples)3 ConclusionsReferencesAnalyses of the 36 daily time interval PM2.5 sam-[1] DOCKERY D W, POPE C A. Acute respiratory efects ofples show that sulfate, nitrate and ammonium are theparticulate air pollution {J]. Annual Review of Publicdominant ionic species in urban Shanghai, accountingHealth, 1994, 15: 107-132.for about 80% of the total ions. The daily S2- ratio[2] JACOBSON M Z. Global direct radiative forcingdue to multicomponent anthropogenic and naturalof 0.31 to 0.82 indicate that coal combustion is still theaerosols [J]. Journal of Geophysical Research, 2001, 106main pollution source in Shanghai. Sulfate and nitrate(D2): 1551-1568.are almost totally neutralized by ammonia in the win-ter sampling days, while the fne particles in summer[3] ADAMS P J, SEINFELD J H, KCH D M. Globalconcentrations of tropospheric sulfate, nitrate， andshould be acidic or part of the nitrate and sulfate areammonium aerosol simulated in a general circulationcombined with other species such as sodium or calcium.model [J] Journal of Geophysical Research, 1999, 104TheNO3 1+[SO-]NHIratio are higher in the nighttime than(D11): 13791-13823.in the daytime, and should be caused by the low forma-4]YEB,JIX,YANGH,YAOX,CHANC,CADLEs,tion rate and the shift of the gas-particle partitioning ofCHAN T, MULAWA P. Concentration and chemical com-ammonium nitrate in the nighttime.position of PM2.s in Shanghai for a 1 year period [J].The oxalate to sulfate ratio is lower in the winterAtmospheric Environment, 2003, 37(4): 499- -510.sampling days than that in summer. Meanwhile, 0x-[5] TAYLOR K, PENNER J E. Response of the climate sye-alate is significantly correlated with sulfate and nitratetem to atmospheric aerosols and greenhouse gases [J]in the winter sampling days, but not in the sunmer and中国煤化工autumn sampling days, suggesting that the formation[6M, WExLER A s. Sizemechanism of oxalate and sulfate is similar in winter,MHC N M H Gte, nitrate, and anm-while different in the hot seasons such 88 summer andnium at a costal site in Hong Kong [J]. Atmosphericautumn.Environment, 1999, 33(6): 843- -853.240J Shanghai Univ (Engl Ed), 2010, 14(4): 235-240[7] LEE T, KREIDENWEIS S M, COLLETT J J L. Aerosolern California: Part 1- spatial variations [J]. Journalion characteristics during the big bend regional aerosolof the Air and Waste Management Association, 2000,and visibility observational study [J. Journal of Air and50(1): 2034 -2044.Waste Management Association, 2004, 54(5): 585 -592.[15] SEINFERT J H, PANDIS S N. Atmospheric chemistry8] KIMJ Y, SoNGC H, GHIMY s, WoNJ G, YooN sand physics [M]. New York: Wiley, 1998: 537-539.C, CARMICHAEL G R, Woo J H. An investigation on[16] KERMINEN V M, HILLAMO R, TEINILA K, PAKKANENNH3 emissions and particulate NH -NO3 formation inT, ALLEGRINI I, SPARAPANI R. Ion balances of size-East Asia (J]. Atmospheric Environment, 2006, 40(12):2139- 2150.resolved tropospheric aerosol samples: Implications forthe acidity and atmospberic processing of aerosol [J][9]YAOX,CHANCK,FANGM，CADLES,CHANT,Atmospherie Enviroument, 2001, 35(31): 5255 5265.MULAWAP, HE K, YE B. The water-soluble ionic com-position of PM2.s in Shanghai and Bejing, China [].17] SCHWAB J J, FELTON H D, DEMERJIAN K L. AerosolAtmospheric Environment, 2002, 36(26): 4223-4234.chemical composition in New York state from inte-10] XIu G, ZHANG D, CHEN J, HUANG X, CHEN z, Guograted filter samples: Urban/rural and seasonal con-H, PAN J. Characterization of major water -soluble in-trasts [] Journal of Geophysical Research, 2004,organic ions in size fractionated particulate matters in109(D16), DOI: 10.1029/2003JD004078.Shanghai campus ambient air [J]. Atmospheric Envi-[18] HARPER L A, CATCHPOOLE V R, DAVIS R, WEIR Kronment, 2004, 38(2): 227-236.L. Ammonia volatilization: Soil, plant, and microcli-11] WANG Y, ZHUANG G S, ZHANG X Y, HUANG K, Xumate effects on diurnal and seasonal fuctuations [J].C, TANG A H, CHEN J M, AN z s. The ion chemistry,Agronomy Journal, 1983, 75(2): 212-218.seasonal cycle, and sources of PM2.5 and TSP aerosol in[19] STELSON A w, SEINELD J H. Relative humidity andShanghai [J]. Atmospheric Environment, 2006, 40(16):temperature dependence of the ammonium nitrate dis-2935- 2952.sociation constant [J] Atmospheric Environment, 1982,[12]GuZP,FENGJL,HANWL,LIL,W∪MH,FU16(5): 983 992.J M, SHENG G Y. Diurnal variations of polyeyclic aro-matic hydrocarbons associated with PM2.5 in Shang-20]YAOX,LAUAPS,FANGM,CHANCK,HuM.hai, China [I]. Journal of Environmental Sciences, 2010,Formation and size distribution characteristics of ionic22(3): 386 392.species in atmospheric particulate matter in Beijing,China: 2-dicarboxylic acids (J]. Atmospheric Environ-[13] ARIMOTO R, DUcE R A, SAVOIE D L, PROSPERO Jment, 2003, 37(21): 3001-3007.M, TALBOT R, CULLEN J D, TOMZA U, LEWIs NF, RAY B J. Relationships among aerosol constituents[21] HUANG X F, Hu M, HEL Y, TANG X Y. Chemicalfrom Asia and the North Pacific during Pem-West A [J].characterization of water soluble organic acids in PM2.sJournal of Geophysical Research, 1996, 101(D1): 2011-in Bejing, China [J. Atmospberic Environment, 2005,2023.39(16): 2819- 2827.[14] KIM B M, TEFFERA s, ZELDIN M D. Characterizationof PM2.5 and PM1o in the south coast air basin of south-(Editor HONG Ou)中国煤化工MYHCNMHG