Metallic components of traffic-induced urban aerosol, their spatial variation, and source apportionment

Metallic components of traffic-induced urban aerosol, their spatial variation, and source apportionment
of 14
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
  Environ Monit Assess (2010) 168:561–574DOI 10.1007/s10661-009-1134-z Metallic components of traffic-induced urban aerosol,their spatial variation, and source apportionment Sandeep Kar · Jyoti Prakash Maity · Alok Chandra Samal · Subhas Chandra Santra Received: 12 February 2009 / Accepted: 27 July 2009 / Published online: 14 August 2009© Springer Science + Business Media B.V. 2009 Abstract This study proposes a practical methodto estimate elemental composition and distrib-ution in order to attribute source and quantifyimpacts of aerosol particles at an urban regionin Kolkata, India. Twelve-hour total particulateswere collected in winter (2005–2006) and an-alyzed by energy-dispersive X-ray fluorescencetechnique to determine multi-elemental composi-tion, especially trace metals. The aerosols consistof various elements including K, Ca, V, Cr, Mn,Fe, Co, Ni, Cu, Zn, Se, and Pb which exhibit sig-nificant concentration at various sites (  p < 0 . 05 ).The concentration of different metallic elementswere found in the order of Zn > Pb > Ni > Cu > Cr > Co. Statistical multivariate analysis andcorrelation matrix analyses were performed forfactor identification and consequent source ap-portionment. Contour profiles demonstrate spa-tialvariationofelementalcompositionsindicatingpossible source contribution along with meteoro-logical influences. Spatial differences were clearlymost significant for Zn, Ni, Pb, and Cu reflecting S. Kar ( B ) · J. P. Maity · A. C. Samal · S. C. SantraDepartment of Environmental Science, Universityof Kalyani, Kalyani 741235, West Bengal, Indiae-mail: sandeepk6@rediffmail.comS. Kar · J. P. MaityDepartment of Earth Science, National Cheng KungUniversity, 1 University Road, Tainan 70101, Taiwan the importance of anthropogenicinputs, primarilyfrom automobile sources. Keywords Atmospheric particulates · Trace metals · Factor analysis · Anthropogenic sources Introduction Trace element concentrations in particulate aero-sols are of immense importance to toxicological,environmental, and occupational health studies.While particulate mass concentrations can bespatially homogeneous, certain toxic particulatecomponents may be unevenly distributed srci-nating from specific sources, creating a potentialfor exposure of large populations to health-relevant particulate components (Krudysz et al.2008). Assessing population exposure to particu-lates thus requires evaluation of the community-scale spatial variability representing differentelemental particulates as well as toxicologicallyrelevant components. Concentrations of trace ele-ments associated with aerosol particles have beenreported in remote, marine background and con-tinental/urban influenced areas throughout theworld (Beavington et al.2004; Al-Momani et al. 2005; Hueglin et al.2005). Rocher et al.(2004) reported the occurrence of heavy metals and ma- jor elements in atmospheric deposition indicating  562 Environ Monit Assess (2010) 168:561–574 atmospheric fluxes of major elements duringcold seasons, due to residential heating, whileheavy metals, whose major sources have constantemission fluxes, exhibit steady atmospheric loadsthroughout the year. Among the trace elementsof airborne particulate matter, Al, Pb, Zn, As,Cd, Ni, and Cr are commonly known heavy met-als, which cause adverse effects on the environ-ment and human health due to their toxicity (Debet al.2002; Fang et al.2006). They occur naturally in a wide range of minerals but their distributionintheenvironmentislargelyduetoemissionfromfossil fuel combustion, industrial activities, andwidespread use of pigments, pesticides, and otherhuman activities (O’Neill1990; Dara1997; Tsai et al.2003;KimOanhetal.2006).Automobiletraffic is considered one of the main reasons for particu-late pollution (Kleeman and Cass1998; Cyrys etal.2003; Park and Kim2005). In developing countries, however, few studieshave been conducted to investigate the chemicalcomposition even of the total suspended partic-ulates (TSP). In India, very little has been doneto characterize the trace elements in particulateaerosols or in their residue. Moreover, most of the reported results were focused on the moredeveloped areas of south-western India and othermajor cities (Negi et al.1988; Tripathi et al.1991). Pandey et al.(1998) reported concentrations of  atmospheric particulate trace elements like F, As,Be, Cu, Zn, Mn, Ni, Pb, Ti, and V at Bhilai,India, an urban industrial location. Enrichmentcalculations show high enrichment of Zn, Pb, Cu,Cd,As,Be,andClinthatarea.Asimilarstudyhasbeen carried out on trace elemental compositionof atmospheric aerosols (PM 10 ) at a semi-urbansite, Tirupati, southern peninsular India (ChandraMouli et al.2006). Kulshrestha et al.(2001) re- ported chemical composition and association of size-differentiated aerosols at a suburban site ina semi-arid tract of India. Mahajan et al.(1996) also measured elemental composition of particu-late matter and precipitation from a remote back-ground site in India.Although earlier studies found differences inelemental concentrations between sites and at-tributed these variations to local and regionalsources, there is a need to determine the spatialvariability of chemically speciated particulates.The primary aim of this study was to determinespatial variation in particulate aerosols, especiallytoxic elements. Secondarily, we determine therelative impact of multiple sources, includingdiesel- and gasoline-dominated roadways, scat-teredindustrialemissions,andotherlocalsources.Concurrent measurements were conducted at 16distinct sites within a highly urbanized region.Sampling sites were selected based on the antici-patedimpactofvaryingsourcesandheterogeneityin exposure to atmospheric particulates. Resultsfrom spatial distribution analyses coupled withcorrelation analyses are presented to determineinter-site variations in particle-phase elements.This study serves as a preliminary assessmentof possible toxic elements and their respectivesources in this region and their contributions topopulation exposure. Experimental methods Study areaThe study was performed in Kolkata Metropo-lis (22 ◦ 19  N to 22 ◦ 38  N and 88 ◦ 18  E to 88 ◦ 26  E),a highly urbanized region in eastern India, con-centrated along the fringes of the river Ganges.The surrounding area was a flat swampy regionwith a subtropical climate. The annual mean tem-perature was 26.8 ◦ C, and humidity was 66% withmaximum of 94% and minimum of 38% (Kararet al.2006). Particulate air pollution representsa prominent environmental health concern forKolkata city. The air quality of the city becomesworse during winter due to frequent thermal in-version and low wind speed (2–6 m/s) (Kararet al.2006). Though average annual rainfall was1,582 mm, no rainfall was observed during thestudy period. The major contributor of particu-lates is mainly automobile exhaust and scatteredindustrial emission. Industrial activities includethe production of automobiles, ceramic, chemi-cal, pharmaceuticals, food processing, heavy en-gineering, shipbuilding, etc. Several small-scaleindustries like rubber, plastic, printing, textile,metallurgical, and metal workshops are scatteredmainly in the north and central part of the region.Acontrolsite(suburbanarea)wasselectedhaving  Environ Monit Assess (2010) 168:561–574 563 lesstrafficandindustrialemissionsinasamephys-iography and climatic location for comparativeassessment (Kalyani, West Bengal, India).SamplingSamplingsiteswereselectedalongdifferentmajorstreets in an attempt to achieve maximum cover-age of highly populated residential areas, traffic-congested areas, and industrial areas within thestudy area (Fig.1). Sampling was carried out inwinterof2005–2006(DecemberandJanuary)dur-ing the peak traffic hours of the day for 12 h (8:00to 20:00 h). Air samples were collected at a heightof about 6 m above ground level using a low vol-ume air sampler (Envirotech 4600). Atmosphericaerosols were collected using the principle of filtering a known volume of air through a glass-fiber filter paper (Nucleopore 0.45 μ  m, diameter25 mm) of known weight at an average speedof 1.8 to 2.2 m 3 air per minute by using a Wolf blower. After sampling, the filters were carefullyremoved from the filter cartridges and placed ina Tarson Petri dish which was kept at room tem-perature. Due to the hygroscopic nature of thefilter, filters were carefully equilibrated for 10 minin a desiccator, both before and after sampling, toeliminate the effect of humidity.Elemental analysis using EDXRFThe elemental profile of particulates depositedon filter paper was directly analyzed by energy-dispersive X-ray fluorescence (EDXRF). This is Fig. 1 Map of the studyarea showing location of the sampling sites inKolkata  564 Environ Monit Assess (2010) 168:561–574 a non-destructive, multi-elemental technique bywhich the elements from sodium to uranium canbe measured both quantitatively and qualitativelyup to ppb level (Vecchi et al.1994; Jayasekeraet al.2004). This technique involves bombarding the samples by X-rays from an X-ray tube, re-sulting in emission of characteristic X-rays. TheenergyoftheemittedX-raysrepresentselementalsignatures whereas their intensities are a measureof the concentrations. A Jordan Valley EX-3600EDXRF system was used in the present study.This consists of an X-ray generator (X-ray tubewith Rh anode and its power unit) and a Si(Li)detector (with resolution of 143 eV at 5.9 keV) forthe detection of the emitted characteristic X-rays.The X-rays from the sample reach the detectorthrough a thin (0.00125 mm) Be window. Eachsamplewasirradiatedfor1,200satfixedtubecon-dition. This system utilized an automated samplechamber where ten samples can be mounted ata time. The quantitative analysis was carried outonline, by the built-in ExWin software. In orderto validate the data, high purity multi-elementstandard (SRM 1643d) obtained from NIST wasused as reference material for quantitative analy-sis. The data presented for a reference sample(SRM 1643d) in Table1shows that the precisionobtained for most of the elements was < 10%RSD with comparable accuracy (nearing 20%RSDorbetter).Backgroundchecksweremadeonthe filtering medium and the handling procedureby analysis of the unexposed filters in a similar Table 1 Semi-quantitative data for the elements in SRM1643d (ppb) by EDXRFElement SRM 1643dMeasured value Certified valueK 2.22 2.35Ca 29.36 31.04V 27.82 35.1Cr 16.34 18.53Mn 33.04 37.66Fe 84.47 91.2Co 25.36 25.0Ni 58.79 58.1Cu 18.7 20.5Zn 67.46 72.48Se 0.076 0.09Pb 19.47 18.15 manner. Intrinsic trace element concentrationthus deduced was appropriately incorporated inthe result.Statistical analysisExperimentaldataobtainedwereusedforstatisti-cal analysis without any transformation. The dataset was subjected to principal component analysisand correlation study in order to assess the rele-vant sources. The factors were extracted by theprinciple component method and were varimaxrotated, i.e., the extracted factors are rotated sothat the variance of the factor loadings becomes amaximum. Both factor loadings and factor scoreswere calculated. Factor loadings represented thecorrelation between variables and factors. Factorscores represent the influence of factors at indi-vidual cases (sampling sites). Factor analysis wasprocessed with the computer program SPSS 15. Results and discussion Trace element concentrationsTable2presents the descriptive statistics of thetrace elemental composition of atmospheric par-ticulates. Twelve elements including heavy met-als, namely K, Ca, V, Cr, Mn, Fe, Co, Ni, Cu,Zn, Se, and Pb, were measured by EDXRF tech-nique. Results of this study exhibited conspicu-ously high levels of Zn, Cu, Ni, Pb, Mn, Fe, Cr,and Co with respect to the control site, where thetraceelementsandheavymetalshavesignificantlylower values. The trend of heavy metal levelswas found in increasing order of Zn > Pb > Ni > Cu > Cr > Co. The concentration of Zn showedthe maximum value of 344.5 ng/m 3 whereas Coshowed the lowest concentration of 4.53 ng/m 3 (Table2). The other four metals like Pb, Ni,Cu, and Cr were showing moderate values of 43.96, 20.58, 17.36, and 9.63 ng/m 3 , respectively.Therefore, it was clear that the concentration of these heavy metals were in the increasing orderof Zn > Pb > Ni > Cu > Cr > Co. Other urbanelementslikeV,Mn,andFeshowedaveragevalueof 18.97, 29.43, and 117.56 ng/m 3 , respectively.Our data was quite similar to earlier reports on  Environ Monit Assess (2010) 168:561–574 565 Table 2 Descriptivestatistics of concentration(ng/m 3 ) of particulatesand trace elements inaerosol of Kolkata( n = 48 ) TSP  total suspendedparticulates ( μ  g/m 3 ) Elements Mean SD Min. Max. Skewness Kurtosis Control siteTSP 362.4 83.6 274.3 422.7 0.18 − 0 . 42 172.5 ± 37.2K 528.2 154.1 137 740 − 0 . 96 1.46 426 ± 187.2Ca 7,265.9 2,278.1 3,750 12,550 0.78 0.61 3,072 ± 965.7V 18.97 4.98 11.92 27.79 0.28 − 1 . 1 0.85 ± 0.07Cr 9.63 4.28 2.16 15.79 − 0 . 38 − 0 . 94 0.704 ± 0.15Mn 29.43 10.74 11.17 51.86 0.44 0.04 15.61 ± 3.82Fe 117.56 55.81 25.76 212.7 0.02 − 0 . 79 27.4 ± 1.42Co 4.53 1.07 2.17 6.18 − 0 . 58 0.08 0.74 ± 0.06Ni 20.58 7.49 9.57 35.59 0.63 − 0 . 45 3.41 ± 0.27Cu 17.36 6.31 5.65 28.24 − 0 . 1 − 0 . 61 1.04 ± 0.08Zn 344.5 104.1 184.3 520.5 0.11 − 0 . 97 49.42 ± 13.6Se 2.23 0.75 1.27 3.52 0.31 − 1 . 29 0.54 ± 0.2Pb 43.96 8.76 23.57 57.28 − 0 . 55 0.58 3.62 ± 0.38 concentrations of the elements in KMA (Kararet al.2006), and also exhibited quite elevatedvalues compared to those reported for differ-ent urban regions worldwide (Yatin et al.2000; Manoli et al.2002; Athanassiadis and Rao2003; Al-Momani et al.2005). A comparison of particulates and metallic ele-ment concentration of different countries is pre-sented in Table3.Fine and coarse fractions of  particulates, along with total particulates, werestudied in different cities worldwide. In Taichung,Taiwan,themaximumTSPvaluewasmeasuredata traffic junction. This is similar with our findingsof a high TSP value (362.4 μ  g/m 3 ) in a traffic-abundant region. In other areas, fine or coarsefraction ranges from 8.83 to 97 μ  g/m 3 . Amongthe major elements, Fe is the most dominant el-ement (1,306 to 4,800 ng/m 3 ) in aerosol particu-lates in all the countries except in New Zealandand Greece. Among the heavy metals, Zn wasdominant (ranges from 4 to 1,060 ng/m 3 ) withelevatedconcentrationsinTaiwan,Japan,Greece,and Dhaka. Zn, Pb, and Cu being dominant inparticulates showed similarity with our study. Ab-sence of Co (except in Canada) indicates its lowgeneral abundance in particulates, in agreementwith our study where it was the least abundant el-ement.OtherelementslikeCd,Ni,andCrshowedmoderate abundance in all the referred countries.Regarding identification of respective sources andtheirapportionment,researchersfollowedvariousanalyticaltoolstodemonstrateappropriatesourcefactors. Among the factors, road dust resuspen-sion and vehicular traffic are the common sourcesof particulates and metallic elements, along withsome site-specific sources that indicate presenceof particular elements. The source apportionmentfindings of our study are in general agreementwith the majority of reported data, which pointout traffic emissions and traffic-induced road dustresuspension as the major sources of fine urbanaerosolwithsumcontributionsupto80%.Criticaldiscussion and comparison of source factor iden-tification are provided below (Factor analysis of trace elements section).Spatial distributionElemental concentrations revealed by the studyshow significant variation among 16 different sitesin study area. Figure2a, b, and c shows spa-tial distribution of the concentration of differenttrace elements and heavy metals. Concentrationof V shows an increasing trend towards the south-east from the river side. Ca also shows similartrends having higher concentrations in the east-ern and southeastern part. This type of variationwas probably due to aerial transport governedby the wind direction towards southeast whichwas observed in winter season. There was littlevariation in concentration of Co around the studyarea, indicating dispersion from a diffused source.Except for some areas in the northern region,Cr concentration was higher at Beleghata andeastern part of KMA. Presence of tannery indus-try on that portion may be one of the reasonsfor the presence of Cr. Concentration of Cu wasalso higher in the central and southern portions
Similar documents
View more...
Related Search
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!