Environment International 94 (2016) 69–75

Contents lists available at ScienceDirect

Environment International

j ourna l homepage: www.e lsev ie r .com/ locate /env int

Influence of pollution control on lead inhalation bioaccessibility in PM2.5:A case study of 2014 Youth Olympic Games in Nanjing

Shi-Wei Li a, Hong-Bo Li a, Jun Luo a, Hui-Ming Li a, Xin Qian a, Miao-Miao Liu a, Jun Bi a,Xin-Yi Cui a,⁎, Lena Q. Ma a,b,⁎⁎a State Key Laboratory of Pollution Control and Resource Reuse, School of the Environment, Nanjing University, Nanjing 210046, People's Republic of Chinab Soil and Water Science Department, University of Florida, Gainesville, FL 32611, United States

⁎ Corresponding author.⁎⁎ Correspondence to: L.Q. Ma, State Key Laboratory ofReuse, School of the Environment, Nanjing University, Naof China.

E-mail addresses: lizzycui@nju.edu.cn (X.-Y. Cui), lqm

http://dx.doi.org/10.1016/j.envint.2016.05.0100160-4120/Published by Elsevier Ltd.

a b s t r a c t

a r t i c l e i n f o

Article history:Received 31 March 2016Received in revised form 10 May 2016Accepted 10 May 2016Available online xxxx

Pollution controls were implemented to improve the air quality for the 2014 Youth Olympic Games (YOG) inNanjing. To investigate the influence of pollution control on Pb inhalation bioaccessibility in PM2.5, sampleswere collected before, during, and after YOG. The objectives were to identify Pb sources in PM2.5 using stable iso-tope fingerprinting technique and compare Pb inhalation bioaccessibility in PM2.5 using two simulated lungfluids. While artificial lysosomal fluid (ALF) simulates interstitial fluid at pH 7.4, Gamble's solution simulatesfluid in alveolar macrophages at pH 4.5. The Pb concentration in PM2.5 samples during YOG (88.2 ng m−3) was44–48% lower than that in non-YOG samples. Based on stable Pb isotope ratios, Pb in YOG samples was mainlyfrom coal combustion while Pb in non-YOG samples was from coal combustion and smelting activities. WhilePb bioaccessibility in YOG samples was lower than those in non-YOG samples (59–79% vs. 55–87%) by ALF, itwas higher than those in non-YOG samples (11–29% vs. 5.3–21%) based on Gamble's solution, attributing tothe lower pH and organic acids in ALF. Different Pb bioaccessibility in PM2.5 between samples resulted fromchanges in Pb species due to pollution control. PbSO4 was the main Pb species in PM2.5 from coal combustion,which was less soluble in ALF than PbO from smelting activities, but more soluble in Gamble's solution. Thisstudy showed it is important to consider Pb bioaccessibility during pollution control as source control not onlyreduced Pb contamination in PM2.5 but also influenced Pb bioaccessibility.

Published by Elsevier Ltd.

Keywords:PbInhalation bioaccessibilityPM2.5

Stable isotopeGamble's solutionCoal combustion

1. Introduction

Due to its rapid industrialization, China has been experiencing se-vere haze events, with PM2.5 as a main contributor (Guo et al., 2014;Huang et al., 2014). An increasing body of evidences has shown the cor-relation between exposure to PM2.5 and many diseases including lungcancer (Cakmak et al., 2014; Richmond-Bryant et al., 2014). Lead (Pb)is one of the most enriched metals in PM2.5 and has attracted much at-tention during the past few decades (Sun et al., 2006). Numerous stud-ies have investigated Pb levels in PM2.5, and significant correlationbetween PM2.5 Pb levels and blood Pb levels in children has been ob-served (Liang et al., 2010; Richmond-Bryant et al., 2014). These resultssuggest that PM2.5 inhalation is an important Pb exposure pathway forhumans. It is therefore imperative to assess human health risks throughPb exposure in PM2.5 via inhalation pathway.

Evidence shows that not all Pb in airborne particles can be absorbedinto the systemic circulation (Boisa et al., 2014; Wiseman and Zereini,

Pollution Control and Resourcenjing 210046, People's Republic

a@ufl.edu (L.Q. Ma).

2014). Accurate assessment of Pb exposure via PM2.5 inhalation there-fore requires themeasurement of its bioavailability. Although in vivo as-says using animals are more accurate to measure Pb inhalationbioavailability, they are still in the development phase. Consequently,in vitro assays using simulated lung fluid have been developed as theyare simple and practical to use. There are two common artificial lungfluids, i.e., Gamble's solution and artificial lysosomal fluid (ALF), whichsimulate two different processes after PM2.5 particles are inhaled intolungs. They have been used to measure metal bioaccessibility in PM2.5.For example, inhalation bioaccessibility of platinum group elements inPM of Germany was measured using the two methods (Zereini et al.,2012). Similarly,Wiseman and Zereini (2014) reported Pb bioaccessibi-lity in PM2.5 at 84% (77–91%) by ALF and 4.0% (1.0–9.0%) by Gamble'ssolution. However, compared to the large number of studies investigat-ing Pb levels in PM2.5, there remains a knowledge gap regarding Pb in-halation bioaccessibility in PM2.5.

It is well known that Pb speciation in soil is amain factor influencingits Pb bioaccessibility (Rasmussen et al., 2011; Smith et al., 2011). Differ-ent sources including industrial emission, vehicle exhaust, coal combus-tion and suspended soil particles contribute to Pb in PM2.5

(Charlesworth et al., 2011; Cheng and Hu, 2010). For example, PM2.5

from smelting activities are more enriched with PbO due to Pb vapor

70 S.-W. Li et al. / Environment International 94 (2016) 69–75

oxidization in the air (Batonneau et al., 2004) whereas more PbSO4 isaccumulated in PM2.5 from coal combustion due to the sulfate enrich-ment in coal (Shah et al., 2009). Therefore, it can be expected that Pb in-halation bioaccessibility in PM2.5 from different sources may varygreatly. The influence of Pb sources on its oral bioaccessibility in soilshas been extensively studied (Bannon et al., 2009; Smith et al., 2011;Cao et al., 2003; Hardison Jr. et al., 2004). However, the influence of Pbsource on its inhalation bioaccessibility in PM2.5 has not beenwell eluci-dated. This knowledge gap is mainly due to the difficulty to identify thePb sources in PM2.5 by conventional approaches including chemicalmass balance, positive matrix factorization models and enrichment fac-tor (Liang et al., 2016). In recent years, fingerprinting based on stable Pbisotope ratios has been successfully used to determine Pb sources in var-ious environmental media including airborne PM (Widory et al., 2010).Here we hypothesized that the influence of contamination sources onPb bioaccessibility in PM2.5 can be effectively examined by couplingthe stable Pb isotope ratio technique with simulated lung fluidextraction.

Nanjing is amega city in eastern Chinawith an area of 6587 km2 andpopulation of 8.2 million. It hosted the 2nd Summer Youth OlympicGames (YOG) during August 16th–28th, 2014. Considering the poorair quality inNanjing, the local government implemented pollution con-trols during the YOG. Approximately 2630 construction sites wereclosed, and heavy-industry factories such as iron and steel industriesand petrochemical enterprises were required to reduce manufacturingby 20%. Vehicles with high emissions such as trucks, engineered vehi-cles, and vehicle van transporting hazardous materials were banned inthe city. In addition, 22 nearby cities were asked to cooperate with Nan-jing to close industries with high pollution emission during the YOG(Ding et al., 2015; Pan et al., 2015). Significant improvement in air qual-ity was expected with the pollution control, so were the changes in Pbcontamination source in airborne PM during the YOG.

Therefore, the pollution control during the YOG provided us aunique opportunity to study the influences of Pb sources on its bioacces-sibility in PM2.5. To this end, PM2.5 samples were collected before, dur-ing, and after the YOG in Nanjing. The overall objective of this studywas to investigate the influence of pollution control on Pb inhalationbioaccessibility in PM2.5 by 1) identifying Pb sources using the stableisotope ratio fingerprinting technique, and 2) measuring Pb inhalationbioaccessibility using two in vitromethods (ALF andGamble's solution).The results from this study should provide us useful information for pol-icy decision regarding pollution control and risk assessment for inhala-tion exposure to air-born Pb.

2. Materials and methods

2.1. Sampling of PM2.5

Particulate matter (PM2.5) sample was collected for ~12 h using ahigh volume air sampler (Model TE6070, Tisch Environmental Inc.) ata flow rate of 1.13 m3 min−1 with quartz microfiber filters (Whatman,203 mm × 254 mm). The sampling device was placed on the buildingroof (~30 m height) of the School of the Environment of Nanjing Uni-versity on Xianlin campus. The campus is located in northeastern Nan-jing, ~20 and 5 km away from industrial zone and Qixia lead–zincmines.

A total of 32 PM2.5 samples were collected before (10 samples, June1st–July 31st), during (9 samples, August 1st–August 31st), and after(13 samples, September 1st–October 20th) YOG in 2014. Since thelocal government started source controls in early August, August sam-ples were considered as the YOG period. Before sampling, quartzmicro-fiber filters were heated at 500 °C for 5 h in a muffle furnace, andequilibrated in a desiccator for 48 h before being weighed using an an-alytical balance (Denver SI-234). After sampling, quartz microfiber fil-ters were equilibrated in a desiccator before being weighed for PM2.5

mass. The PM2.5 mass was determined as the difference between the

filter weights before and after sampling. The filters were encased in tin-foil and sealed in plastic bags and stored at−20 °C until analysis.

2.2. Total Pb concentration in PM2.5

Quantification of Pb in PM2.5 was performed by digesting quartz mi-crofiber filters using ultra-pure concentrated HNO3 and 30% H2O2 ac-cording to USEPA Method 3050B with slight modifications. Briefly, 1/16 of quartz microfiber filters were cut into pieces of b2 mm2 usingacid-cleaned ceramic knife. The filters were digested with 10 mL con-centrated HNO3 (1:1) by immersing quartz filters in concentratedHNO3 so all PM2.5 particles were in contact with HNO3. Since PM2.5

and filter were digested together, the digestion process shouldmeasureall the Pb on the filters. The digestion was conducted on a HotBlock di-gestion system (Environmental Express, USA) at 105 °C for 5 h, and then10 mL concentrated HNO3 was added. After concentrated HNO3 wasevaporated to near dryness and cooled down, 1–2 mL 30% H2O2 wasadded. The digestion was continued to reach ~1 mL solution, whichwas diluted to 50 mL with Milli-Q water, and filtered through 0.22 μmfilter before analysis. Concentrations of Pb in the solution were deter-mined by inductively coupled plasma mass spectrometer (ICP-MS,NexIONTM300X, Perkin Elmer, USA) with detection limit of0.008 μg L−1. Standard Reference Material (SRM) D056–540 was per-formed for QA/QC during digestion process. The Pb recovery was of99.4 ± 5.78%, which is within the ±10% recommended by USEPA(2013).

2.3. Stable isotope ratios in PM2.5

Lead stable isotope ratios of 207/206Pb and 208/206Pb in digestion solu-tionwere determined using ICP-MS. Instrument parameters were set as190 sweeps/reading, 1 reading/replicate, and 10 replicates/sample solu-tion. Dwell time of 40 ms was set for 204Pb and 25 ms for 206Pb, 207Pb,and 208Pb. Prior to analysis, Pb concentration in solution was dilutedto ~15 μg L−1 using 0.1 M high-purity HNO3. Standard reference mate-rial (SRM) NIST 981 with concentration of 15 μg L−1 was measuredevery 5 samples to obtain ratio correction factors to compensate formass discrimination. The analytical precision for samples was generallyb0.5% for 207/206Pb and 208/206Pb.Measured 204/206Pb (0.0590±0.0001),207/206Pb (0.9149 ± 0.0024), and 208/206Pb (2.1688 ± 0.0029) in theSRM NIST 981 (n = 10) were in good agreement with the certifiedvalues of 0.0590, 0.9146, and 2.1681.

2.4. Pb bioaccessibility in PM2.5

Two artificial lung fluids including artificial lysosomal fluid (ALF)and Gamble's solution were employed to measure Pb inhalation bioac-cessibility in PM2.5 (Boisa et al., 2014; Colombo et al., 2008; Zereini etal., 2012). Detailed compositions of the lung fluids can be found in sup-port information as Table S1. The extraction procedure was performedaccording to Zereini et al. (2012). Briefly, 1/16 of quartz microfiber fil-ters (b2 mm2) were placed in 50 mL high density polyethylene tubescontaining 30 mL of fluid. Based on the PM2.5 mass, the solid and solu-tion ratio was 1:2400 to 1:14,000, which was in the range of 1:500 to1:50,000 suggested by Julien et al. (2011). The tubes were put in an in-cubator at 37 °C in dark. Samples were shaken for 10min every 4 h on ashaker at 50 rpm. Once inhaled into lungs, part of PM2.5 can be quicklydissolved in the interstitial fluid, but 10–15% of the initially depositedPM2.5 can be present in lungs after 1 day and their clearance in humanbronchial tree may last for several weeks (Hofmann and Asgharian,2003; Lippmann et al., 1980). So the PM2.5 particles deposited in lungare subject to both short and long retention with most studies using1 day as the short term (Julien et al., 2011). For the long term, time var-iations of 4–30 days were reported (Boisa et al., 2014; Zereini et al.,2012). In our preliminary test, some fungi appeared in lung fluid after15-day extraction. To avoid the analysis uncertainty caused by fungi,

71S.-W. Li et al. / Environment International 94 (2016) 69–75

15-day was selected as the long term, with 1 day as the short to assessbioaccessible Pb in PM2.5. For 1 day extraction, pH was measuredevery 4 h with no significant pH fluctuation being found. For the15 day extraction, pH was checked every day and adjusted using nitricacid and sodium hydroxide. At the end of 1- and 15-days, tubes werecentrifuged at 4000 rpm for 10 min with supernatants filtered through0.22 μmpolypropylenemembrane filters. Concentrated HNO3 at 1%wasadded into the supernatants for sample preservation. Solutions werestored at 4 °C until analysis using ICP–MS.

2.5. Statistical analyses

All experiments were run with two replicates. One-way analysis ofvariance (ANOVA) based on least-significant difference (LSD) was ap-plied using SPSS 10.0 for windows to determine the significant differ-ences in Pb concentrations, Pb bioaccessibility and stable Pb isotoperatios in PM2.5 collected before, during, and after the YOG. All graphswere created using Origin 9.0.

3. Results and discussion

3.1. Changes in PM2.5 and Pb concentrations before, during, and after theYOG

Based on PM2.5 concentrations, our data suggested that source con-trols by local government during the YOG were effective in improvingthe air quality in Nanjing. The PM2.5 concentrations were 83–267, 41–105, and 39–179 μg m−3 in samples collected before, during, and afterYOG (Fig. 1A). The average PM2.5 concentration during the YOG period(61.4 μg m−3, n = 9) was 64% and 47% lower than those before(172 μg m−3, n = 10) and after the YOG (117 μg m−3, n = 13). ThePM2.5 levels were also lower than the average PM2.5 concentration of106 μg m−3 in August 2012 of Nanjing (Shen et al., 2014), indicatingthe effectiveness of the source control at reducing PM2.5 levels in

Fig. 1. PM2.5 concentration (A) and Pb concentration (B) based on air volume before, during, acollected before, during, and after YOG period using artificial lysosomal fluid (ALF) (C) and G75th percentile, solid lines and squares in boxes are the median and mean values, error bars revalues. Means marked with different letters indicate significant difference (p b 0.05). Dash lpresents USEPA standard, grade II 35 μg m−3.

Nanjing. The mean PM2.5 concentration during YOG period was compa-rable to 64.7 μg m−3 during the 2008 Olympic in Beijing (Wang et al.,2009) and 77.0 μg m−3 during the 2010 Asian Games in Guangzhou(Tao et al., 2015). Moreover, the average PM2.5 concentration after theYOGwas 32% lower than those before the YOG, which was partially ex-plained by the changes inmeteorology. Meteorological parameters dur-ing sampling days, including temperature, relative humidity, visibilityand wind speed, were obtained from www.wunderground.com (Fig.S1). The wind speed was negatively correlated with PM2.5 concentra-tions while the relative humidity was positively correlated with PM2.5

concentration (p b 0.05) (Marcazzan et al., 2001). Based on the meteo-rology parameters, the averagewind speed before-YOGwas 1.72m s−1,which was lower than 1.97 m s−1 after-YOG. Similarly, the average rel-ative humidity before-YOG was 88%, which was higher than that after-YOG (75%). The meteorological differences (i.e., higher wind speed andlower relative humidity) may partially explain the lower PM2.5 levels insamples after-YOG than before-YOG.

For samples collected before, during, and after the YOG, the Pb con-centrations in PM2.5 based on air volumewere 42.1–122, 36.9–77.9, and40.5–233 ngm−3, (averaging 88.2, 49.1, and 95.3 ngm−3) (Fig. 1B). Theaverage Pb concentration in YOG samples was 44–48% lower than non-YOG samples and was comparable to that during 2008 Beijing OlympicGames (48.9 ng m−3) (Schleicher et al., 2012). However, though lowerPb concentrations were observed, they were still higher than those incities from Europe and USA during normal days. For example, the aver-age Pb concentrations in PM2.5 were 25 ngm−3 for 20 U.S. cities during2003–2005 (USEPA, 2007) and 5.5–13 ng m−3 for European cities in-cluding Athens, Helsinki, and Frankfurt (Pakkanen et al., 2001;Remoundaki et al., 2013; Wiseman and Zereini, 2014).

At present, there is no standard for Pb concentrations in PM2.5. How-ever, there are limits for total suspended particulates (TSP, b100 μm) inthe air, which are 500 ng m−3 and 150 ng m−3 by China's Ambient AirQuality Standard (CAAQS) and US National Ambient Air Quality Stan-dard (NASQS), while standard for Pb in PM10 is 500 ng m−3 based on

nd after the Youth Olympic Games (YOG) in Nanjing and Pb bioaccessibility (%) in PM2.5

amble's solution (Gam) (D) after 1 day and 15 days extraction. Boxes represent 25th topresent 1st to 99th percentiles, crossing symbols represent the minimum and maximumine in Fig. A presents Chinese National PM2.5 standard, grade II 75 μg m−3 and dot line

72 S.-W. Li et al. / Environment International 94 (2016) 69–75

European directive of 2008/50/EC. To make a comparison between Pbconcentrations in PM2.5 and these standards, we estimated the Pbmass in PM2.5, TSP and PM10. Based on previous studies on air qualitymonitoring in Nanjing from 2010 to 2012, the mean mass ratios ofPM2.5 to PM10 and PM2.5 to TSP were 0.73 and 0.5 (Hu et al., 2012;Shen et al., 2014). In addition, it was assumed that Pb concentrations(mg kg−1) in PM10 and TSP were equal to that in PM2.5 even thoughPb is more enriched in fine particles (Hu et al., 2013). Lead concentra-tions in TSP and PM10 in Nanjingwere estimated by dividing Pb concen-tration in PM2.5 with mass ratios of PM2.5 to PM10 and PM2.5 to TSP. Theestimated average Pb concentrations in PM10 and TSP in YOG sampleswere 67.3 (50.5–107) and 98.2 (73.8–156) ng m−3, while they wereat 126 (57–188) and 184 (83–274) ng m−3 in non-YOG samples. Forthe less stringent limits set by CAAQS (500 ngm−3 in TSP) and Europe-an directive of 2008/50/EC (500 ngm−3 in PM10), Pb concentrations inall samples complied with the values. However, Pb concentrations in11% of YOG samples exceeded the NASQS limit (150 ng m−3) and 41%did in non-YOG samples.

3.2. Changes in Pb sources in PM2.5 before, during, and after the YOG

In addition to Pb concentrations in PM2.5 based on air volume, themean Pb concentrations were also calculated based on PM2.5 particlemass, which were 514, 850, and 764 mg kg−1 in samples collected be-fore, during, and after YOG (Table 1). The relatively lower Pb mass con-centration in non-YOG PM2.5 samples may attribute to the effect ofsecondary organic aerosol (Huang et al., 2014). There are two mainsources of PM2.5, i.e., primary emitted particles (directly emitted to theatmosphere, like dust from smelting or coal combustion) and secondaryorganic aerosol (formed through gas-to-particle conversion in light-in-duced chemical reaction based on air composition) (Hinds, 1999;Huang et al., 2014). It was reported that the particles from secondary or-ganic aerosol constituted 51–77% of PM2.5 during haze days (Huang etal., 2014). Since haze events frequently occurred during the non-YOGperiod, the relatively larger proportion of secondary organic aerosol inPM2.5 may dilute Pb mass concentration in non-YOG PM2.5 samples.On the other hand, it was possible that Pb in PM2.5 collected during dif-ferent periods have different originations. For example, Pb concentra-tion was 1788 mg kg−1 in coal combustion dust, 6140 mg kg−1 inmetallurgic dust, and 238 mg kg−1 in unleaded gasoline (Tan et al.,2006). Given the different Pb concentrations between non-YOG andYOG PM2.5 samples, it was expected that Pb sources in PM2.5 were dif-ferent for non-YOG and YOG periods.

Table 1Descriptive statistics of Pb concentrations (based on PM2.5 mass and air volume) and sta-ble isotope ratios in PM2.5 collected before, during, and after Youth Olympic Games (YOG)in Nanjing.

Pb (ng m−3) Pb (mg kg−1) 207Pb/206Pb 208Pb/206Pb

Before YOG (n = 10)Min 42.2 410 0.8522 2.0896Max 122 606 0.8588 2.1113Median 90.8 523 0.8549 2.0948Mean 88.2 514 0.8547 2.0969SD 29 65 0.0020 0.0067

During YOG (n = 9)Min 36.9 530 0.8520 2.0875Max 77.9 1332 0.8663 2.1243Median 45.7 835 0.858 2.1076Mean 49.1 850 0.8593 2.1092SD 13.5 249 0.0048 0.0122

After YOG (n = 13)Min 40.5 445 0.8498 2.0821Max 233 1046 0.8576 2.0975Median 87.7 751 0.8539 2.0887Mean 95.3 764 0.8535 2.0896SD 51.3 169 0.0026 0.0050

To further track the changes of Pb sources before, during, and afterYOG, stable Pb isotope ratios in PM2.5 samples were measured (Table 1and Fig. 2). Significant differences in the stable isotope ratios(207Pb/206Pb and 208Pb/206Pb) between YOG and non-YOG sampleswere observed (p b 0.05). Stable Pb isotope ratios in PM2.5 were clearlydistinct from soils around Qixia Pb–Znminerals (Fig. 2), so their contri-bution was excluded. Stable Pb isotope ratios in most of the non-YOGsamples were between those of coal combustion ash and metallurgicdust (Fig. 2), suggesting that both sources contributed to Pb in thesesamples. This finding is consistent with Hu et al. (2014) who showedthat coal emissions and smelting activities were two important sourcesfor Pb in PM2.5 from urban sites of Nanjing. For YOG PM2.5 samples, theisotopic ratios (207/206Pb and 208/206Pb) were elevated, being mainly o-verlapped with those of coal combustion with several points overlap-ping with unleaded gasoline (Fig. 2). During the YOG period, manysmelters in Nanjing were shut down, and it was not surprising that itscontribution to air Pb pollution was reduced in YOG samples (Ding etal., 2015). However, coal combustion power plants were still in opera-tion during the YOG to supply electricity to the city. Therefore, Pb inYOG samples may come from coal combustion and unleaded gasoline(Fig. 2). However, Pb concentration in YOG PM2.5 was 850 mg kg−1,which was much higher than that for unleaded gasoline dust(238 mg kg−1) (Tan et al., 2006). Therefore, unleaded gasoline did notmake much contribution to Pb in YOG PM2.5 samples even thoughsome data points overlapped with the isotopic ratios of unleaded gaso-line dust. Overall, we found that Pb pollution sources changed from coalcombustion and smelting (non-YOG) to coal combustion (during YOG).Therefore, PM2.5 samples during the YOG and non-YOG provided us aunique opportunity to study the effects of Pb sources on Pb inhalationbioaccessibility in PM2.5.

3.3. Pb bioaccessibility in PM2.5 by in vitro assays

Two in vitro assays, artificial lysosomal fluid (ALF) at pH = 4.5 andGamble's solution at pH= 7.4 were used tomeasure Pb bioaccessibilityin PM2.5 (Fig. 1C, D). In addition, two extraction durations (1 and15 days) were used to simulate the short and long term retention ofPM2.5 particles after being inhaled into lungs.

ALF mainly simulates Pb inhalation bioaccessibility in macrophages.After 1 day extraction, Pb bioaccessibility in YOG PM2.5 samples (61 ±4.3%, n = 9) was significantly lower than those in before (66 ± 6.4%,n=10) and after YOG samples (78± 4.6%, n=13). The data indicated

Fig. 2. Ratios of 208Pb/206Pb vs. 207Pb/206Pb in PM2.5 collected before, during, and afterYouth Olympic Games (YOG) in Nanjing. Ranges or individual isotope ratios of differentanthropogenic Pb sources, including coal, leaded and unleaded gasoline (Chen et al.,2005), metallurgy dust (Tan et al., 2006), and lead growth curve (Cumming andRichards, 1975) and soil around lead-zinc mines.

Fig. 3. Comparison of Pb bioaccessibility extracted by Gamble's solution, IBM (inhalationbioaccessibility method), and IBM without organic acids, DPPC, or mucin.

73S.-W. Li et al. / Environment International 94 (2016) 69–75

that not only the total Pb concentrations in YOG PM2.5 (49.1 ng m−3)were lower than those before and after YOG samples (88.2 and95.3 ng m−3; Fig. 1B), so were the bioaccessible Pb based on ALF. Thebioaccessible Pb in this study (66–78%)were lower than the 84% report-ed byWiseman and Zereini (2014). This may be due to the heterogeniccompositions of airborne PM. After 15 day extraction, the mean Pb bio-accessibility was significantly (p b 0.05) reduced to 60 ± 3.6% in sam-ples collected before YOG, while those collected during and after YOGslightly reduced to 59± 3.3 and 77± 4.5%with no statistical difference(p = 0.21–0.50). This was consistent with previous study that averagePb bioaccessibility was reduced from 84 to 60% as extraction durationextended from 1 to 30 days (Wiseman and Zereini, 2014). The slight re-duction in Pb bioaccessibility may be attributed to resorption of mobi-lized Pb onto particle surfaces (Zereini et al., 2012).

Gamble's solution differs from ALF as it measures Pb bioaccessibilityafter PM2.5 enters interstitial fluid in the lungs. Based on Gamble's solu-tion, the average Pb bioaccessibility in PM2.5 after 1 day extraction were10, 20, and 11% before, during, and after YOG, with Pb bioaccessibility inYOG samples being significantly higher. After 15 day extraction, thevalues decreased sharply to 5.0, 0.08, and 6.0% (Fig.1D). Besides the re-sorption process (Wiseman and Zereini, 2014; Zereini et al., 2012), thereduction of Pb bioaccessibility in Gamble's solution may be attributedto the formation of insoluble Pb mineral, limiting Pb dissolution duringthe extraction. For example, Pb bioaccessibility in mining waste tailingsand dustswasmeasuredbyGamble's solution after 630 h, insoluble leadphosphate minerals was observed after 24 h through SEM, which isformed on the surface of dust, effectively limiting Pb dissolution(Wragg and Klinck, 2007).Where Pb bioaccessibility in ALFwas notablyhigher than those in Gamble's solution, which was mostly attributed totheir pH difference, i.e., 4.5 vs. 7.4, as Pb is less soluble at higher pH (Li etal., 2014).Wiseman et al. also found that Pb bioaccessibility in PM2.5wasnotably higher in ALF (84%) than that in Gamble's solution (4.0%) after1 day extraction (Wiseman and Zereini, 2014). In gastrointestinalfluid, the same phenomenon was also observed. For example, Pb bioac-cessibility decreased from 46–99 to 22–60%when the extraction pH in-creased from 1.5 to 2.5 in house dust samples (Li et al., 2014). BesidespH, the difference in chemical components in the two lung fluids mayalso influence Pb bioaccessibility. For example, based on Gamble's solu-tion and water extract of four reference materials at same pH, the ex-tractable Pb was significantly higher in Gamble's solution than that inwater (Julien et al., 2011). ALF and Gamble's solution have differentchemical compositions (Table S1), with ALF having more organicacids. Organic acids have been reported to promote Pb dissolutionfrom Pb minerals in dust samples (Debela et al., 2010; Li et al., 2014),which can partially explain the higher Pb bioaccessibility by ALF thanGamble's solution.

Currently, there is no standardmethod tomeasure Pb inhalation bio-accessibility in PM2.5. Due to its lack of organic components that arepresent in human respiratory tract, Gamble's solution has been modi-fied by adding organic acids, proteins, and surfactants (Boisa et al.,2014; Gray et al., 2010; Stebounova et al., 2011). These additional com-ponents were expected to increase Pb dissolution kinetics and hence Pbbioaccessibility. To test this hypothesis, we added organic acids (ascor-bic acid and uric acid), protein (mucin), and surfactant (dipalmitoylphosphatidyl choline-DPPC) to Gamble's solution (Table S1). The mod-ified Gamble's solution was based on Boisa et al. (2014) and detailedcompositions can be found in Table S1. Themethod was used to extractPM2.5 samples before, during, and after YOG (i.e., #1, 13, and 27) for1 day. The Pb bioaccessibility was 23–43%, which was much higherthan 5.3–11% in Gamble's solution (Fig. 3).

To test the effect of individual component, we removed organicacids, mucin, or DPPC from modified Gamble's solution and extractedsamples #1, 13, and 27 again. The Pb bioaccessibility was decreased to1–9, 0.2–28, and 5.6–18%, indicating that these components contributedto the increased Pb bioaccessibility in PM2.5 (Fig. 3). For example, surfac-tant DPPC is a phospholipid secreted by type 2 cells and located in

alveoli of lungs. Previous studies suggested that DPPC in lung fluid in-creased the wettability of PM2.5 to reduce aggregation of PM2.5 and im-proved the contact between PM2.5 and artificial lung fluid (Julien et al.,2011). However, without in vivo data, we cannot determine whichmethod better simulated human lungs. These in vitro methods are de-veloped to simulate the physiological environment in human lungfluids. Therefore, the extractable Pb should be more realistic than totalPb concentrations. In addition, these methods are favorable than invivo tests due to their cost saving and easy operation (Colombo et al.,2008; Wiseman, 2015; Zereini et al., 2012).

Different Pb bioaccessibility in PM2.5 samples collected before, dur-ing, and after YOG may be explained by the changes in Pb contamina-tion sources due to pollution control. Stable isotope ratios illustratedthat Pb in YOG samples was mainly from coal combustion ash, whilePb in non-YOG sampleswas from both smelting activities and coal com-bustion (Fig. 2). We tried to identify Pb speciation in PM2.5 samplesusing X-ray absorption spectroscopy (Fig. S2), but the result was unsuc-cessful due to the interference from the complicated matrix in PM2.5.However, based on Pb isotope ratio and X-ray absorption spectroscopy,Tan et al. (2006) reported that Pb in PM10 samples from coal combus-tion, and iron and steel smelters in Shanghai were mainly present asPbSO4, PbO, and PbCl2. Based on scanning Raman microspectrometryand X-ray photoelectron spectroscopy, Pb in smelter dust was mainlypresent as PbSO4, PbSO4·PbO, PbSO4·4PbO, and PbO (Batonneau et al.,2004), and based on X-ray absorption spectroscopy, PbSO4 was themain Pb species in coal fly ash (Shah et al., 2009). Based on these litera-tures, we expected higher portion of PbO in non-YOG samples and higherportion of PbSO4 in YOG samples. It is reported that PbSO4 has lowersolubility than PbSO4·PbO and PbSO4·4PbO at pH = 4.5 in ALF, but theopposite is observed at pH = 7.4 in Gamble's solution (Sommers andLindsay, 1979). While Na2Ac and NaHCO3 are present in Gamble'ssolution, they are absent in ALF (Table S1). The HCO3

− buffers Gamble'ssolution to slightly alkaline pH (pH= 7.4), making PbSO4 more solublein alkaline solution than PbO in the presence of Ac2−. To test this hypoth-esis, we measured Pb bioaccessibility of PbSO4 and PbO using ALF andGamble's solutions. Based on ALF, the bioaccessible Pb concentrationswere 324 ± 43.2 and 537 ± 27.1 mg L−1 for PbSO4 and PbO, and inGamble's solution they were 364 ± 10.0 and 85.6 ± 6.27 μg L−1. Theresults confirmed that PbSO4 had lower bioaccessibility than PbO in ALF,but the opposite was observed in Gamble's solution (data not shown).

4. Environmental implications

Due to poor air quality in China, Chinese government has imple-mented various pollution control measures during YOG. Thus total con-centrations of PM particles and PM-born Pb are expected to decreaseaccordingly. However, little attention has been focused on the associat-ed changes in Pb bioaccessibility in PM particles, which is important for

Fig. 4. Lead concentrations in the livers and kidney of mouse after two days oforopharyngeal aspiration of ~50 mg of PbO or PbSO4 minerals.

74 S.-W. Li et al. / Environment International 94 (2016) 69–75

assessing the health risks associatedwith inhalation of PMparticles. Ourdata showed that, due to the source control during YOG, Pb in PM2.5 hadlower bioaccessibility in alveolar macrophages (simulated by ALF), buthigher Pb bioaccessibility in interstitial lung (simulated by Gamble's so-lution) in YOG samples than non-YOG samples. This highlights the im-portance of not only reducing total Pb load in airborne PM, but alsobioaccessible Pb, which may directly impact human health.

Compared with Pb oral bioaccessibility using simulated gastrointesti-nal solutions, studies on Pb inhalation bioaccessibility using simulatedlung fluid are rather limited. Gamble's solution was the only in vitromethod to measure Pb inhalation bioaccessibility until the developmentof ALF (Beeston et al., 2010; Wiseman and Zereini, 2014). Several studiesshowed that significantly higher amount of Pb in PM2.5 was extracted byALF thanGamble's solution. It is easier for smaller particles like PM2.5 to berespired into distal lung part and phagocytized bymacrophages, which issimulated by ALF. Therefore, given the higher release of Pb in ALF and thepotential of PM2.5 to be phagocytized, the Pb bioaccessibility in PM2.5maybe underestimated if only using Gamble's solution. Furthermore, alveolarmacrophages, once exposed to hazardous elements (such as Pb in PM2.5),can be activated to produce large amount of reactive oxygen species,which are closely related with pulmonary and systemic inflammation(Huang et al., 2009). Consequently, it is expected that Pb bioaccessibilityin ALF (representing the cellular fluid in alveolar macrophages) can pro-vide valuable information for toxicology studies of PM2.5.

As mentioned above, there is no standard in vitro method to measurePb inhalation bioaccessibility in PM2.5, and it is essential to validate invitro method with animal-based in vivo tests. However, there are chal-lenges to conduct in vivo tests to measure Pb inhalation bioavailability.First, after being inhaled, air particles can be deposited in nasal, laryngeal,bronchial airways, and alveolar region with various clearance rates. Theirclearance rates depend on particle size and breathing cycle of animals.The particles in nasal, laryngeal, and bronchial may be cleared relativelyquickly, but in alveolar region they may last several weeks, even severalyears (Hofmann and Asgharian, 2003; Lippmann et al., 1980). It is there-fore difficult to decide the reasonable exposure duration for in vivo tests.

In addition, PM2.5 is often collected on filters, which is difficult to re-move. We tried to remove the particles from filters by sonication andfreeze drying, and then exposed them to mice. We found that ~60% ofthe Pb on filters was lost after dissolving in water during sonication. Inaddition, the aggregation state of PM2.5 particles changed during freezedrying, which induced inaccuracy for in vivo exposure. We also tried toexposemicewith PbO and PbSO4 to see if we can obtain similar trend asin vitro data. Briefly, 100mg of PbO or PbSO4 (b100 μm)were dispersedinto 50 mL saline solution, and exposed by oropharyngeal aspiration tomice (Gavett et al., 2003), which were first fasted for 1 day (Li et al.,2016). After 1 day, the Pb concentrations were measured in mice liverand kidney. Unlike the in vitro result, therewas no significant differencefor PbO and PbSO4 bioavailability (6.9 ± 3.7 vs. 7.9 ± 4.2 μg kg−1 inliver, and 19 ± 12 vs. 9.6 ± 6.7 μg kg−1 in kidney) (Fig. 4). The unsuc-cessful datamay be resulted from the uncertainty of biological biomark-er, exposure duration, or individual difference among mice. At present,exposure chamber may be a promising tool for in vivo inhalation bio-availabilitymeasurement. However, studies using an exposure chamberare at the development phase as no in vivo data have been reported.Nevertheless, these in vitro methods were developed to simulate thephysiological condition in human lung fluid, and bioaccessible Pb maybe more realistic than total Pb concentration for risk assessment. Assuch, even though they may be less accurate than in vivo data, in vitroPb bioaccessibility measurement in PM2.5 is an important step in theright direction to accurately assess the risk of PM2.5 to human health.

Acknowledgements

This work was supported in part by Jiangsu Provincial Natural ScienceFoundation of China (no. BK20130558), Jiangsu Provincial InnovationTeam Program, and Jiangsu Provincial Double Innovation Program.

References

Bannon, D.I., Drexler, J.W., Fent, G.M., Casteel, S.W., Hunter, P.J., Brattin, W.J., Major, M.A.,2009. Evaluation of small arms range soils for metal contamination and lead bioavail-ability. Environ. Sci. Technol. 43, 9071–9076.

Batonneau, Y., Bremard, C., Gengembre, L., Laureyns, J., Le Maguer, A., Le Maguer, D.,Perdrix, E., Sobanska, S., 2004. Speciation of PM10 sources of airborne nonferrousmetals within the 3-km zone of lead/zinc smelters. Environ. Sci. Technol. 38,5281–5289.

Beeston, M.P., van Elteren, J.T., Šelih, V.S., Fairhurst, R., 2010. Characterization of artificiallygenerated PbS aerosols and their use within a respiratory bioaccessibility test. Analyst135, 351–357.

Boisa, N., Elom, N., Dean, J.R., Deary, M.E., Bird, G., Entwistle, J.A., 2014. Development andapplication of an inhalation bioaccessibility method (IBM) for lead in the PM10 sizefraction of soil. Environ. Int. 70, 132–142.

Cakmak, S., Dales, R., Kauri, L.M., Mahmud, M., Van Ryswyk, K., Vanos, J., Liu, L.,Kumarathasan, P., Thomson, E., Vincent, R., 2014. Metal composition of fine particu-late air pollution and acute changes in cardiorespiratory physiology. Environ. Pollut.189, 208–214.

Charlesworth, S., De Miguel, E., Ordóñez, A., 2011. A review of the distribution of particu-late trace elements in urban terrestrial environments and its application to consider-ations of risk. Environ. Geochem. Health 33, 103–123.

Chen, J., Tan, M., Li, Y., Zhang, Y., Lu, W., Tong, Y., Zhang, G., Li, Y., 2005. A lead isotope re-cord of Shanghai atmospheric lead emissions in total suspended particles during theperiod of phasing out of leaded gasoline. Atmos. Environ. 39, 1245–1253.

Cheng, H., Hu, Y., 2010. Lead (Pb) isotopic fingerprinting and its applications in lead pol-lution studies in China: a review. Environ. Pollut. 158, 1134–1146.

Cao, X., Ma, L.Q., Chen, M., Hardison, D.W., Harris, W.G., 2003. Weathering of lead bulletsand their environmental effects at outdoor shooting ranges. J. Environ. Qual. 32,526–534.

Colombo, C., Monhemius, A.J., Plant, J.A., 2008. Platinum, palladium and rhodium releasefrom vehicle exhaust catalysts and road dust exposed to simulated lung fluids.Ecotoxicol. Environ. Saf. 71, 722–730.

Cumming, G., Richards, J., 1975. Ore lead isotope ratios in a continuously changing earth.Earth Planet. Sci. Lett. 28, 155–171.

Debela, F., Arocena, J.M., Thring, R.W., Whitcombe, T., 2010. Organic acid-induced releaseof lead from pyromorphite and its relevance to reclamation of Pb-contaminated soils.Chemosphere 80, 450–456.

Ding, J., van der A, R.J., Mijling, B., Levelt, P.F., Hao, N., 2015. NOx emission estimates dur-ing the 2014 Youth Olympic Games in Nanjing. Atmos. Chem. Phys. Discuss. 15,6337–6372.

Gavett, S.H., Haykal-Coates, N., Highfill, J.W., Ledbetter, A.D., Chen, L.C., Cohen, M.D.,Harkema, J.R., Wagner, J.G., Costa, D.L., 2003.World Trade Center fine particulate mat-ter causes respiratory tract hyperresponsiveness in mice. Environ. Health Perspect.111, 981.

Gray, J.E., Plumlee, G.S., Morman, S.A., Higueras, P.L., Crock, J.G., Lowers, H.A., Witten, M.L.,2010. In vitro studies evaluating leaching of mercury from mine waste calcine usingsimulated human body fluids. Environ. Sci. Technol. 44, 4782–4788.

Guo, S., Hu, M., Zamora, M.L., Peng, J., Shang, D., Zheng, J., Du, Z., Wu, Z., Shao, M., Zeng, L.,Molina, M.J., Zhang, R., 2014. Elucidating severe urban haze formation in China. Proc.Natl. Acad. Sci. 111, 17373–17378.

Hardison Jr, D.W., Ma, L.Q., Luongo, T., Harris, W.G., 2004. Lead contamination in shootingrange soils from abrasion of lead bullets and subsequent weathering. Sci. Total Envi-ron. 328, 175–183.

Hinds, W.C., 1999. Aerosol Technology: Properties. Behavior, and Measurement of Air-borne Particles (2nd).

Hofmann, W., Asgharian, B., 2003. The effect of lung structure on mucociliary clearanceand particle retention in human and rat lungs. Toxicol. Sci. 73, 448–456.

75S.-W. Li et al. / Environment International 94 (2016) 69–75

Hu, X., Zhang, Y., Ding, Z., Wang, T., Lian, H., Sun, Y., Wu, J., 2012. Bioaccessibility andhealth risk of arsenic and heavy metals (Cd, Co, Cr, Cu, Ni, Pb, Zn and Mn) in TSPand PM2.5 in Nanjing, China. Atmos. Environ. 57, 146–152.

Hu, X., Ding, Z., Zhang, Y., Sun, Y., Wu, J., Chen, Y., Lian, H., 2013. Size distribution andsource apportionment of airborne metallic elements in Nanjing, China. Aerosol AirQual. Res. 13, 1796–1806.

Hu, X., Sun, Y., Ding, Z., Zhang, Y., Wu, J., Lian, H., Wang, T., 2014. Lead contamination andtransfer in urban environmental compartments analyzed by lead levels and isotopiccompositions. Environ. Pollut. 187, 42–48.

Huang, Y.C.T., Zhuowei, L., Carter, J.D., Soukup, J.M., Schwartz, D.A., Yang, I.V., 2009. Fineambient particles induce oxidative stress and metal binding genes in human alveolarmacrophages. Am. J. Respir. Cell Mol. Biol. 41, 544–552.

Huang, R.-J., Zhang, Y., Bozzetti, C., Ho, K.-F., Cao, J.-J., Han, Y., Daellenbach, K.R., Slowik,J.G., Platt, S.M., Canonaco, F., Zotter, P., Wolf, R., Pieber, S.M., Bruns, E.A., Crippa, M.,Ciarelli, G., Piazzalunga, A., Schwikowski, M., Abbaszade, G., Schnelle-Kreis, J.,Zimmermann, R., An, Z., Szidat, S., Baltensperger, U., Haddad, I.E., Prevot, A.S.H.,2014. High secondary aerosol contribution to particulate pollution during hazeevents in China. Nature 514, 218–222.

Julien, C., Esperanza, P., Bruno, M., Alleman, L.Y., 2011. Development of an in vitromethodto estimate lung bioaccessibility of metals from atmospheric particles. J. Environ.Monit. 13, 621–630.

Li, H.B., Cui, X.Y., Li, K., Li, J., Juhasz, A.L., Ma, L.Q., 2014. Assessment of in vitro lead bioac-cessibility in house dust and its relationship to in vivo lead relative bioavailability. En-viron. Sci. Technol. 48, 8548–8555.

Li, J., Li, C., Sun, H.-J., Juhasz, A.L., Luo, J., Li, H.-B., Ma, L.Q., 2016. Arsenic relative bioavail-ability in contaminated soils: comparison of animal models, dosing schemes, and bi-ological end points. Environ. Sci. Technol. 50, 453–461.

Liang, C.-S., Duan, F.-K., He, K.-B., Ma, Y.-L., 2016. Review on recent progress in observa-tions, source identifications and countermeasures of PM2.5. Environ. Int. 86,150–170.

Liang, F., Zhang, G., Tan, M., Yan, C., Li, X., Li, Y., Li, Y., Zhang, Y., Shan, Z., 2010. Lead inchildren's blood is mainly caused by coal-fired ash after phasing out of leaded gaso-line in Shanghai. Environ. Sci. Technol. 44, 4760–4765.

Lippmann, M., Yeates, D., Albert, R., 1980. Deposition, retention, and clearance of inhaledparticles. Br. J. Ind. Med. 37, 337–362.

Marcazzan, G.M., Vaccaro, S., Valli, G., Vecchi, R., 2001. Characterisation of PM10 and PM2.5

particulate matter in the ambient air of Milan (Italy). Atmos. Environ. 35, 4639–4650.Pakkanen, T.A., Loukkola, K., Korhonen, C.H., Aurela, M., Mäkelä, T., Hillamo, R.E., Aarnio,

P., Koskentalo, T., Kousa, A., Maenhaut, W., 2001. Sources and chemical compositionof atmospheric fine and coarse particles in the Helsinki area. Atmos. Environ. 35,5381–5391.

Pan, L., Tang, H., Liu, X., 2015. Review and revelation of air quality assurance work duringthe Nanjing Youth Olympic Games. Environ. Monit. Forewarning 7, 1–3.

Rasmussen, P.E., Beauchemin, S., Chénier, M., Levesque, C., MacLean, L.C., Marro, L., Jones-Otazo, H., Petrovic, S., McDonald, L.T., Gardner, H.D., 2011. Canadian house duststudy: lead bioaccessibility and speciation. Environ. Sci. Technol. 45, 4959–4965.

Remoundaki, E., Kassomenos, P., Mantas, E., Mihalopoulos, N., Tsezos, M., 2013. Composi-tion and mass closure of PM2.5 in urban environment (Athens, Greece). Aerosol AirQual. Res. 13, 72–82.

Richmond-Bryant, J., Meng, Q., Davis, A., Cohen, J., Lu, S.-E., Svendsgaard, D., Brown, J.S.,Tuttle, L., Hubbard, H., Rice, J., 2014. The influence of declining air lead levels onblood lead–air lead slope factors in children. Environ. Health Perspect.

Schleicher, N., Norra, S., Chen, Y., Chai, F., Wang, S., 2012. Efficiency of mitigation mea-sures to reduce particulate air pollution—a case study during the Olympic SummerGames 2008 in Beijing, China. Sci. Total Environ. 427–428, 146–158.

Shah, P., Strezov, V., Nelson, P.F., 2009. X-ray absorption near edge structure spectrometrystudy of nickel and lead speciation in coals and coal combustion products. EnergyFuel 23, 1518–1525.

Shen, G.F., Yuan, S.Y., Xie, Y.N., Xia, S.J., Li, L., Yao, Y.K., Qiao, Y.Z., Zhang, J., Zhao, Q.Y., Ding,A.J., 2014. Ambient levels and temporal variations of PM2.5 and PM10 at a residentialsite in the mega-city, Nanjing, in the western Yangtze River Delta, China. J. Environ.Sci. Health A 49, 171–178.

Smith, E., Kempson, I.M., Juhasz, A.L., Weber, J., Rofe, A., Gancarz, D., Naidu, R., McLaren,R.G., Grafe, M., 2011. In vivo-in vitro and XANES spectroscopy assessments of leadbioavailability in contaminated periurban soils. Environ. Sci. Technol. 45, 6145–6152.

Sommers, L., Lindsay, W., 1979. Effect of pH and redox on predicted heavy metal-chelateequilibria in soils. Soil Sci. Soc. Am. J. 43, 39–47.

Stebounova, L.V., Adamcakova-Dodd, A., Kim, J.S., Park, H., O'Shaughnessy, P.T., Grassian,V.H., Thorne, P.S., 2011. Nanosilver induces minimal lung toxicity or inflammationin a subacute murine inhalation model. Part. Fibre Toxicol. 8, 5.

Sun, Y., Zhuang, G., Zhang, W., Wang, Y., Zhuang, Y., 2006. Characteristics and sources oflead pollution after phasing out leaded gasoline in Beijing. Atmos. Environ. 40,2973–2985.

Tan, M., Zhang, G., Li, X., Zhang, Y., Yue, W., Chen, J., Wang, Y., Li, A., Li, Y., Zhang, Y., 2006.Comprehensive study of lead pollution in Shanghai by multiple techniques. Anal.Chem. 78, 8044–8050.

Tao, J., Zhang, L., Zhang, Z., Huang, R., Wu, Y., Zhang, R., Cao, J., Zhang, Y., 2015. Control ofPM2.5 in Guangzhou during the 16th Asian Games period: implication for hazyweather prevention. Sci. Total Environ. 508, 57–66.

USEPA, 2007. Review of the National Ambient Air Quality Standards for Lead (EPA-452/R-07-013).

USEPA, 2013. Test Methods for Evaluating Solid Wastes SW-846 EPA Method 6020 B: In-organic Elements by ICP-MS. USEPA, Washington, DC.

Wang, W., Primbs, T., Tao, S., Simonich, S.L.M., 2009. Atmospheric particulate matter pol-lution during the 2008 Beijing Olympics. Environ. Sci. Technol. 43, 5314–5320.

Widory, D., Liu, X., Dong, S., 2010. Isotopes as tracers of sources of lead and strontium inaerosols (TSP & PM2.5) in Beijing. Atmos. Environ. 44, 3679–3687.

Wiseman, C.L.S., 2015. Analytical methods for assessing metal bioaccessibility in airborneparticulate matter: a scoping review. Anal. Chim. Acta 877, 9–18.

Wiseman, C.L., Zereini, F., 2014. Characterizing metal (loid) solubility in airborne PM10,PM2.5 and PM1 in Frankfurt, Germany using simulated lung fluids. Atmos. Environ.89, 282–289.

Wragg, J., Klinck, B., 2007. The bioaccessibility of lead from Welsh mine waste using a re-spiratory uptake test. J. Environ. Sci. Health A 42, 1223–1231.

Zereini, F., Wiseman, C.L., Püttmann,W., 2012. In vitro investigations of platinum, palladi-um, and rhodium mobility in urban airborne particulate matter (PM10, PM2.5, andPM1) using simulated lung fluids. Environ. Sci. Technol. 46, 10326–10333.