Journal of Hazardous Materials 341 (2018) 365–372

Contents lists available at ScienceDirect

Journal of Hazardous Materials journal homepage: www.elsevier.com/locate/jhazmat

Research Paper

Effect of lead speciation on its oral bioaccessibility in surface dust and soil of electronic-wastes recycling sites Takashi Fujimori a,b,∗,1 , Masaya Taniguchi b,1 , Tetsuro Agusa c,2 , Kenji Shiota b , Masaki Takaoka a,b , Aya Yoshida d , Atsushi Terazono d , Florencio C. Ballesteros Jr. e , Hidetaka Takigami d a

Department of Global Ecology, Graduate School of Global Environmental Studies, Kyoto University, Katsura, Nisikyo-ku, Kyoto, 615-8540, Japan Department of Environmental Engineering, Graduate School of Engineering, Kyoto University, Katsura, Nisikyo-ku, Kyoto, 615-8540, Japan c Center for Marine Environmental Studies, Ehime University, Bunkyo-cho 2-5, Matsuyama, Ehime 790-8577, Japan d Center for Material Cycles and Waste Management Research, National Institute for Environmental Studies (NIES), 16-2 Onogawa, Tsukuba, 305-8506, Ibaraki, Japan e Environmental Engineering Graduate Program, University of the Philippines Diliman, Quezon City, 1101, Philippines b

h i g h l i g h t s

g r a p h i c a l

a b s t r a c t

• Bioaccessibility of Pb in e-waste sur• • • •

face matrices (dust and soil) was assessed. PBET test was used as in vitro bioaccessibility assay. Pb speciation was determined using X-ray absorption spectroscopy. Specific Pb bioaccessibility and species were identified. Influence of Pb species on bioaccessibility was discussed statistically.

a r t i c l e

i n f o

Article history: Received 19 May 2017 Received in revised form 27 July 2017 Accepted 29 July 2017 Available online 4 August 2017 Keywords: Lead Electronic-waste recycling

a b s t r a c t We measured bioaccessible lead (Pb) in simulated gastrointestinal fluids containing Pb-contaminated soil or dust from electronic waste (e-waste) recycling sites to assess the risk of Pb ingestion. The physiologically based extraction test (PBET) was used as in vitro bioaccessibility assay. Pb speciation was determined using X-ray absorption spectroscopy. The total Pb concentrations in dusts (n = 8) and soils (n = 4) were in the range of 1630–131,000 and 239–7800 mg/kg, respectively. Metallic Pb, a common component of ewaste, was ubiquitous in the samples. We also found Pb adsorbed onto goethite and as oxides and carbonate, implying soil mixing and weathering influences. Pb phosphate and organic species were only found in the soil samples, suggesting that formation was soil-specific. We identified other Pb compounds in

∗ Corresponding author at: Department of Global Ecology, Graduate School of Global Environmental Studies, Kyoto University, Katsura, Nisikyo-ku, Kyoto, 615-8540, Japan. E-mail address: [email protected] (T. Fujimori). 1 These authors equally contributed to this work. 2 Current address: Department of Environmental Resources, Graduate School of Environmental and Symbiotic Sciences, Prefectural University of Kumamoto, 3-1-100, Tsukide, Higashi-ku, Kumamoto, 862-8502, Japan. http://dx.doi.org/10.1016/j.jhazmat.2017.07.066 0304-3894/© 2017 Elsevier B.V. All rights reserved.

366 Surface matrices Bioaccessibility Speciation

T. Fujimori et al. / Journal of Hazardous Materials 341 (2018) 365–372

several samples, including Pb silicate, Pb chromate, and Pb(II) hydrogen phosphate. A correlation analysis indicated that metallic Pb decreased bioaccessibility in the stomach, while a Pb speciation analysis revealed a low bioaccessibility for Pb phosphates and high bioaccessibility for organic Pb species. The health risk based on bioaccessible Pb was estimated to be much lower than that of total Pb due to the lower concentrations. © 2017 Elsevier B.V. All rights reserved.

1. Introduction Lead (Pb) pollution from electronic waste (e-waste) recycling poses serious risks to environmental and human health, as cathode ray tubes (CRT), circuit boards, and other electronics often contain Pb. Concentrations of Pb in surface soil [1–8] and dust [1,7–9] collected from e-waste recycling sites in developing countries exceed environmental standards and Pb has been reported in the blood of children [10], placenta [11], urine [12], scalp hair [13], and neonate umbilical cord blood and meconium [14] collected near e-waste recycling sites. Risk assessments based on hazard quotients (HQ) have indicated that Pb is the most hazardous element among metals and metalloids from e-waste dismantling and scrapping activities [8,9]. Recently, relationship between the Pb accumulation in humans and ingestion of surface matrices in e-waste recycling site was also reported [15,16]. Especially, soil was one of main contributing factors to the blood Pb level among groundwater, soil, rice, corn, chicken, and pork in the e-waste dismantling site [15]. However, no systemic studies have investigated the bioaccessibility of ingested Pb from e-waste. A survey on the bioaccessibility of Pb from indoor house environmental material provided the framework and techniques to study Pb from e-waste recycling. Simple, rapid, and inexpensive in vitro bioaccessibility assays are useful for estimating the relative bioavailability of Pb from incidental dust and soil ingestion. By measuring Pb solubilized from a solid matrix into simulated human digestive fluids, the potential amount of Pb available for absorption into the systemic circulation can be estimated. A gastrointestinal extraction system typically comprises several compartments maintained at body temperature (37 ◦ C) with controlled pHs, exposure times, and added biogenic compounds. The physiologically based extraction test (PBET) is a common in vitro bioaccessibility assay that has been used to assess the bioaccessibility of Pb in contaminated soil [17–20] and house dust [21]. It begins with a 1-h incubation in the stomach compartment, followed by a 4-h incubation in the small intestine compartment [17]. The PBET has been shown to be well correlated with the results of available feeding studies using soil Pb [17,22,23]. However, the in vitro PBET has not been applied to heavy metals in dust or soil matrices derived from e-waste recycling facilities. The chemical forms of Pb that used in electronic production were quite different with those in soil samples, which was reported to influence Pb’s bioaccessibility [24,25]. In 2011, the Canadian House Dust Study (CHDS) revealed a significant correlation between bioaccessibility derived from X-ray absorption spectroscopy (XAS) speciation and that measured using simulated gastric extraction, indicating the utility of analyzing Pb speciation with XAS to predict bioaccessibility in house dust [25]. Dust and soil from e-waste recycling sites likely contain different Pb species from those found in house dust [26], because Pb is used in electronics as specific species. For example, CRT glass contains Pb oxides and Pb silicates, while metallic Pb is common in circuit boards and other electronics. Additionally, Pb in dust and soil exposed to air could change chemical form due to e-waste processes, weathering effects, and biogeochemical reactions. For example, it was reported that weathering altered the chemical form and bioaccessibility of zinc in house dust [27].

Our research group investigated Pb levels in dust and soil from e-waste recycling sites around Metro Manila, the Philippines, in 2010 [8]. We found that total Pb in dust from e-waste recycling sites had the highest level of chronic non-cancer toxicity risk and the Pb contamination originated from e-waste recycling activities. However, we did not investigate the chemical forms and bioaccessibility of Pb at that time. It is important to determine Pb bioaccessibility in surface soil and dust from e-waste recycling in order to conduct appropriate risk assessments. Additionally, the relationship between the chemical forms and bioaccessibility of Pb will improve our understanding of the characteristics of e-waste recycling activities. In this study, we measured the bioaccessibility of Pb in e-waste surface matrices (i.e., dust and soil from the previously collected samples from the Philippines [8]) using an in vitro PBET assay, and examined bioaccessible Pb in simulated gastrointestinal fluids (i.e., the stomach and small intestine) to assess the risk of Pb ingestion. Pb speciation was determined using XAS. Combining these results from each sample with a dataset of total concentrations, bioaccessibility, and speciation of Pb, we examined the relationship between bioaccessibility and speciation to clarify the characteristics of Pb behavior in each gastrointestinal compartment.

2. Materials and method 2.1. Surface dust and soil from e-waste recycling In 2010, we collected soil and dust samples from e-waste recycling sites in northern and southern Metro Manila: Caloocan (North, Metro Manila) and the provinces of Cavite and Laguna. We visited two formal sites in Cavite and Laguna and three informal sites in Caloocan (North) and Cavite. Previously, we measured total Pb concentrations in dust (150–130,000 mg/kg) and soil (23–7800 mg/kg) [8]. Based on the Pb concentration and type of e-waste recycling site, we selected 12 surface matrices (dust, n = 8 named D1–D8; soil, n = 4 named S1–S4) from these samples to assess Pb bioaccessibility and analyze the chemical forms of Pb. Table S1 lists the details of the soil and dust samples and provides a description of the sampling sites. According to a recent review by Chi et al. [28], e-waste recycling sites can be categorized as ‘formal’ or ‘informal.’ Formal sites are operated by approved companies that deal with a large amount of used products from affiliated clients and comply with environmental laws and regulations. Products at formal sites mainly consist of plastics, traditional forms of e-waste (e.g., CRTs, refrigerators, circuit boards, and wire cables), and more recent forms of e-waste (liquid crystal displays, and solar panels). In contrast, informal sites are illegal and have a small number of workers. Informal e-waste commonly consists of traditional forms of e-waste such as CRTs, circuit boards, and wire cables, etc. In the formal sector, Pb tends to accumulate in indoor floor dust without dilution effects, such as rain, wind, and soil mixing. At informal sites, there is generally an open-air environment in the workshop. In this study, we categorized sample types into only two groups, dust and soil, because the number of samples was limited to compare their differences by formal and informal (especially, for soil samples).

T. Fujimori et al. / Journal of Hazardous Materials 341 (2018) 365–372

Surface soil was sampled by removing an area 30 cm in diameter and a few centimeters deep, which was then manually packaged. We excluded foreign fragments, such as stones, weeds, and waste fragments. Before packaging, the surface soil was homogenized with a shovel. Surface dust was swept gently with a clean broom and sealed in a sample bag. Isolated, dusty areas were selected for sampling at each building and open-air concrete floor. After samples were transported to the laboratory, they were air-dried indoors for about one week. The surface soil and dust samples collected were 30–60 and 10–30 g/sample (air–dry base), respectively. Although we attempted to collect as much dust as possible, the final amount sampled was limited. Then, the soil and dust were sieved through 1- and 2-mm mesh screens and pulverized with a planetary ball mill (PULVERISETTE 6, FRITSCH) and freezer/mill (6870, SPEX SamplePrep), respectively. Finally, the samples were sieved to <150 ␮m for the experimental measurements. Although it was possible to increase the surface area by physical milling and thereby influence the Pb bioaccessibility, it was necessary for the soil and dust samples to have a high homogeneity to analyze the chemical forms of Pb by the XAS technique. In this study, the X-ray irradiated area of the solid sample was very small (1–2 mm2 ) and the fluorescence X-ray spectrum from the area was detected as the average signal derived from the molecular environment surrounding the Pb. 2.2. Bioaccessibility analysis We used the in vitro PBET assay to assess Pb bioaccessibility. Simulated gastrointestinal fluids were prepared as described previously [17,21]. The gastric solution was a mixture of 1.25 g of pepsin A (from porcine stomach mucosa), 0.5 g of sodium malate, 0.5 g of sodium citrate, 420 ␮L of lactic acid, 500 ␮L of acetic acid, and 1 mL of decanol in 1 L of ultrapure water. The pH was adjusted to 2.5 with hydrochloric acid, representing the average gastric state, which has been used to estimate Pb bioavailability [17]. Then, 400 mg of powdered sample was added to 40 mL of gastric solution in a polypropylene bottle. The capped bottle was placed in a shaker in a water bath at 37 ◦ C for 1 h. For analysis, a 2 mL aliquot was pipetted into a centrifuge tube (25 × 90 mm) and centrifuged at 3500 rpm for 10 min. The supernatant was filtered through a 0.45-␮m mixed cellulose ester membrane filter into a flask. The experimental material that passed through the filter was rinsed with ultrapure water, which was added to the flask. To simulate small intestine fluid, the remaining gastric solution was titrated to pH 7.0 with a saturated sodium bicarbonate solution, and 70 mg of bile salt (porcine) and 20 mg of pancreatin (porcine) were added. The capped bottle was incubated at 37 ◦ C for 4 h. A 2-mL aliquot was pipetted into a centrifuge tube and centrifuged and filtered as described above for analysis. Pb in the simulated gastrointestinal fluids was quantified using inductively coupled plasma atomic emission spectrometry (ICP-AES) (Thermo Electron, IRIS Intrepid) using yttrium as the internal standard. The amount of sample used in the PBET test was quite limited because a subsample had been used for measurement of total Pb concentration [8], and another is also needed for XAS measurement in the research. Therefore, we selected five solid samples (D5, S1, S2, S3, and S4, total Pb range 239–131,000 mg/kg) for the PBET test, which was run three times to determine the repeatability of PBET. The average relative standard deviations of bioaccessible Pb were 15% and 29% for the stomach and small intestine phases, respectively. We also performed quality assurance of Pb bioaccessibility test by using standard reference soil containing Pb (NIST, SRM 2711a). As a result of triplicate test, relative standard deviations were 1.1% in stomach phase and 0.33% in small intestine phase. Laboratory controls for each phase were prepared using sequential digestions in the absence of solid samples. Bioaccessible Pb was expressed as absolute (mg/kg) and relative (%

367

of total) values. The recovery ratios of total Pb concentrations using five certified reference materials for soil, sediment, sludge, and dust were acceptable (>80%). Each analysis of total Pb concentration was performed using laboratory control and analytical blank samples [8].

2.3. X-ray absorption spectroscopy Pb speciation was analyzed using Pb L3 -edge X-ray absorption fine structure (XAFS) spectroscopy at BL01B1 in SPring-8 (Hyogo, Japan) [29]. The samples were sealed immediately and XAFS spectra were measured with fluorescence. XAFS analysis was carried out on the exact same sample (i.e., a subsample) that was used for bioaccessibility after it was milled and sieved. Different Pb species are used in a variety of products, such as CRT glass, solder, paint, and plastics. We selected 23 reference species for analysis: metallic lead (Pb), lead(II) oxide (PbO), lead(IV) oxide (PbO2 ), lead(II,IV) oxide (Pb3 O4 ), lead(II) carbonate (PbCO3 ), lead(II) nitrate (Pb(NO3 )2 ), lead sulfide (PbS), lead(II) sulfate (PbSO4 ), lead(II) chloride (PbCl2 ), lead(II) phosphate (Pb3 (PO4 )2 ), lead(II) hydrogen phosphate (PbHPO4 ), lead(II) acetate (Pb(OAc)2 ), lead(II) silicate (PbSiO3 ; alamosite), lead(II) chromate (PbCrO4 ), calcium plumbate (Ca2 PbO4 ), lead tribasic sulfate (3PbO·PbSO4 ·H2 O) (TBLS), hydrocerussite (Pb3 (CO3 )2 (OH)2 ), hydroxylpyromorphite (Pb5 (PO4 )3 OH), chloropyromorphite (Pb5 (PO4 )3 Cl), lead(II) citrate (Pb3 (C6 H5 O7 )2 ), Pb adsorbed onto goethite (Pb-goethite, 350 mmol/kg, pH 6.5), and gibbsite (300 mmol/kg, pH 6.5) (as representative species of Pb associated to iron or aluminum oxyhydroxides), and Pb complexed to humic acid (300 mmol/kg, pH 5). We followed previous methods for adsorption to goethite and gibbsite [26,30,31] and complexation with humic acid [31–33]. The spectra of the reference compounds were measured using transmission mode after making sample disks containing boron nitride. We identified the chemical forms of Pb using k3 -weighted Pb L3 -edge extended XAFS (EXAFS) spectra. Chemical species were characterized with least squares fitting (LSF) using SIXPack software (ver. 0.63) [34]. Reduced 2 was used to evaluate the LSF for the experimental spectra [35,36] according to formula (1): 1  obs fit 2 (i − i ) N−P N

reduced2 =

(1)

i=1

is the ordinate of the EXAFS spectrum measured from where obs i fit

the sample at the ith energy point, i is the ordinate of the fitted EXAFS spectrum, N is the number of data points in the fitted wavenumber k range, and P is the number of fitted components. Reference EXAFS spectra for the LSF are listed in Fig. S1. The goodness of fit was assessed by the LSF based on the similarity of the shape of calculated and sample spectra and the reduced 2 value. We judged the fitting criteria of the reduced 2 values as follows: <0.05 as good, 0.05–0.20 as acceptable, and >0.20 as poor. We used the PCA to determine the number and type of principal components in k-range 3–8 Å−1 , and target transformation to identify the probable Pb species within the set of reference compounds. Details of the PCA and target transformation and fitting of EXAFS spectra are provided in Ressler et al. [37] and Manceau et al. [38], respectively. Based on the results from the PCA (2.0–7.0 Å−1 ), we used a maximum of five standards in the data analysis. No energy shift was allowed for the LSF and a sum constraint of 1 was imposed. The EXAFS fittings ranged from 2 to 6.3, 6.5, or 7.0 Å−1 , depending on data quality.

368

T. Fujimori et al. / Journal of Hazardous Materials 341 (2018) 365–372

Table 1 Total and bioaccessible Pb concentrations in the stomach and small intestine phases using various dust and soil samples from e-waste recycling sites. Type

Sample

Dust

D1 D2 D3 D4 D5 D6 D7 D8

Soil

S1 S2 S3 S4

a b c

Total Pb concentration (mg/kg)c

Pb concentration (mg/kg)

Bioaccessibility (%)

Stomach

Small intestine

Stomach

Small intestine

233 286 983 1040 2650 2120 710 1340

350 18.4 428 383 994 394 36.8 20.8

1.3 0.81 8.5 12 2.0 51 23 82

2.0 0.052 3.7 4.3 0.76 9.5 1.2 1.3

17700 35400 11500 9010 131000 4130 3060 1630

57.3 868 1130 <10a

<10a 108 15.1 <10a

24 19 14 <1.5b

<4.2b 2.4 0.19 <1.5b

239 4580 7800 657

Under limit of quantification value. Calculated by using limit of quantification value. Total Pb concentration was referred from our previous study [8].

Fig. 1. Absolute (upper) and relative (lower) stomach and intestinal Pb bioaccessibilities in dust and soil samples from e-waste recycling sites. Error bars denote standard deviation of triplicate measurements. Blue and red colors indicate the stomach phase dataset in dust and soil samples, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

3. Results and discussion 3.1. Lead bioaccessibility Total Pb concentration differed among sample matrices. However, the order differed for the bioaccessible Pb concentration, indicating that Pb bioaccessibility in gastrointestinal fluids depended on the sample matrix. Pb was most bioaccessible in the stomach phase, except in one dust sample (D1). The stomach-phase Pb concentrations were 233–2650 mg/kg in dust and <10–1130 mg/kg in soil (Table 1 and Fig. 1). Dust samples had higher bioaccessible Pb concentrations in the stomach phase, which has also been reported in house dust samples in several other studies [21,25]. Pb concentrations in the small intestine phase were relatively lower than those in the stomach phase (dusts 18.4–994 mg/kg, soils <10–108 mg/kg) (Table 1 and Fig. 1). By proportion, Pb bioaccessibility in the stomach phase was greatest for

dust samples D6 (51%) and D8 (82.2%) compared to all other sample types (other dusts 0.8–23%, soils <1.5–24%) (Table 1 and Fig. 1). Pb bioaccessibility in the small intestine phase accounted for <10% of all samples (dusts 0.1–9.5%, soils <1.5–2.4%) (Table 1 and Fig. 1). In contrast to the bioaccessible Pb concentrations, almost all dust samples had a lower Pb bioaccessibility in the gastrointestinal phase from house dust samples [21,25]. This finding suggests that the ratio of insoluble Pb species in the dust matrix derived from the e-waste recycling facility was higher than that from house dust. Other than the chemical form of Pb, Pb bioaccessibility was influenced by various parameters such as pH, particle size, soil organic matter (SOM), and iron (Fe) concentrations [22,39]. Although pH, particle size, and SOM were outside the scope of this study, there is a need to survey these parameters in the future. Iron concentrations were measured in our previous study [8] and were in the range of 9500–143,000 mg/kg among soil and dust samples. No significant correlation was found between the Fe concentration and bioaccessible Pb concentration (n = 11 for the stomach phase and n = 10 for the small intestine phase). The adsorption of Pb2+ on Fe oxide surfaces has been well studied [39] and the presence of iron oxide can result in Pb becoming less bioaccessible. In contrast, other researchers have reported no significant effect of metallic Fe and Fe chloride on the reduction of Pb bioaccessibility in contaminated soil [40]. We could not analyze the chemical form of Fe in our samples, but there is a possible relationship between the chemical forms of Fe and Pb bioaccessibility.

3.2. Lead speciation The Pb L3 -edge EXAFS spectra differed among the original solid samples (Fig. 2), suggesting that Pb in surface matrices existed as multiple Pb species derived from different Pb sources, such as ewaste and other material. From the LSF analysis, there were clear differences in the chemical forms of Pb between dust and soil matrices. Dust metallic Pb was the predominant species in most dust samples (44–71% in D1, D2, D4, D5, and D7) (Table 2); however, in samples D3 and D8, metallic Pb was the second most predominant species (23% and 21%, respectively). Metallic Pb is ubiquitous in e-waste from popular electronics. Activities related to transporting, categorizing, dismantling, and stripping were performed in the formal e-waste recycling sites, and metallic Pb-rich particles could have formed during these activities and contaminated dust. Pbgoethite was prevalent in samples D2, D3, D4, and D5 (15–29%) (Table 2). In general, e-waste contains iron, as do building gates at the formal sites, which would be used throughout the work-

T. Fujimori et al. / Journal of Hazardous Materials 341 (2018) 365–372

369

Table 2 Lead speciation in sample solids (soils and dusts) from e-waste recycling sites analyzed with an LSF of Pb L3 -edge EXAFS spectra. Type

Dust

Sample

D1

Percentage of Pb chemical form (%) 44 Pb. Pb-Goethite PbHPO4 20 PbSiO3 22 Pb3 O4 15 3PbO·PbSO4 ·H2 O Pb5 (PO4 )3 OH Pb3 (PO4 )2 PbCrO4 PbO Pb Citrate PbCO3 PbO2 Pb5 (PO4 )3 Cl PbCl2 PbSO4 0.030 Red 2 k range (Å−1 ) 2–6.5

Soil D2

D3

D4

D5

59 23 12

23 15

71 29

57 19

D6

D7

D8

S1

S2

31

21

18 31

23

S3

18

24 (59) 11

14 28 28

(41)

32

S4

30

17 27

20 12 7

19

36

25 26

41

19

0.037 2–7.0

0.037 2–7.0

18

16 21

25 7 31 0.066 2–6.5

0.022 2–6.5

0.071 2–6.5

0.086 2–6.5

0.696 2–6.3

0.088 2–6.5

17 0.100 2–6.5

0.093 2–6.5

0.050 2–6.5

Bold, maximum percentage; (), poor LSF results due to high Red 2 value are used only as references.

day. Iron oxyhydroxides or surrounding soil could have admixed with dust in the worksite, allowing partial adsorption of Pb onto goethite. Supporting this, Pb has been shown to bind to minerals in soil and urban road dust [41]. As well, Pb(II) oxide, Pb(IV) oxide, and Pb(II,IV) oxide were present in D1, D2, D3, D7, and D8, and Pb carbonate was present in D5, indicating weathering of Pb due to gas-solid interactions with atmospheric oxygen and carbon dioxide. Minor Pb species, such as Pb(II) hydrogen phosphate, Pb(II) silicate, hydroxylpyromorphite, and Pb(II) chloride, were identified in several dust samples. In particular, Pb(II) silicate in D1 (22%) and D7 (14%) implicated CRT glass recycling at that formal site in contaminating dust with Pb silicates. Sample D6 was difficult to analyze using LSF because the measured Pb L3 -edge EXAFS spectrum had high noise (Fig. 2). However, it was similar to Pb oxide, especially Pb(II,IV) oxide based on their intervals and amplitudes (Fig. S1). In sample D7, 28% of Pb was TBLS. TBLS is used as an additive in plastics, such as PVC, and that facility stored and recycled plastics; therefore, some of the informal dust Pb contamination was likely from plastics. Soil. We found Pb phosphate (12–36%) and Pb citrate (19–41%) in all four and two of four of the soils, respectively, as shown in Fig. 3. The soil matrix had lower proportions of metallic Pb than the dust matrix, implying that metallic Pb from e-waste changed chemical forms in the soil. For example, previous research has identified Pb phosphate as a reaction product in soil environments using Pb L3 edge EXAFS [42,43]. Furthermore, Pb citrate indicated the presence of organic Pb species [26] in soil. In contrast, the proportion of Pb silicate (27–32% in S1, S3, and S4) was similar between soil and dust samples (Table 2), likely because it was derived from CRT glass and unable to readily react with the soil matrix. Additionally, samples S1 and S3 contained Pb chromate (7% and 26%, respectively) and Pbgoethite was identified in sample S1 (31%). Several of the Pb species, including Pb-goethite and Pb chromate. Soil S4 contained minor Pb species, including Pb(II) oxide and Pb(IV) hydrogen phosphate. 3.3. Influence of Pb speciation on bioaccessibility Fig. 2. Pb L3 -edge EXAFS spectra for the original surface matrices from e-waste recycling sites. The LSF are shown as circles. Solid lines indicate spectra in original solid samples. Blue and red colors indicate fitting results for dust and soil samples, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Because Pb bioaccessibility in each gastrointestinal phase varied greatly within sample type (Fig. 1), we examined possible causes for this large range. Fig. S2 shows the relationship between total and bioaccessible Pb concentrations; the significant negative correlations in the stomach (n = 11, rs = −0.927, p < 0.001) phase obtained from a nonparametric Spearman’s correlation analysis implied that

370

T. Fujimori et al. / Journal of Hazardous Materials 341 (2018) 365–372

Fig. 3. Effect of the sample matrix (dust and soil) on the chemical form of Pb. Ratios of metallic Pb, Pb phosphate, and Pb citrate in solid samples are shown from left to right, respectively. Blue and red circles are dust and soil samples, respectively. ni = no identification by the LSF analysis. Formal and informal indicate dust samples collected from formal and informal e-waste recycling sites, respectively. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 4. Relationship between the proportion of metallic Pb and bioaccessibility in the stomach (n = 9) and small intestine (n = 8) phases.

total Pb concentration and bioaccessibility were partially controlled by the relative abundance of metallic particles in the sample matrices, which have been shown to have low bioaccessibility in the stomach phase [21]. This was supported by the Pb L3 -edge EXAFS analysis. Based on their significant negative correlation, metallic Pb influenced bioaccessibility in the stomach phase (n = 9, rs = −0.720, p < 0.05) (Fig. 4). No correlation was found between bioaccessibility and the proportion of metallic Pb in the small intestine phase (n = 8, rs = −0.132, p = 0.756). Bioaccessibility data below the limit of quantification (S1 and S4 in Table 1) and specific samples where metallic

Pb was not identified (D6 and S3 in Table 2) were not included in the correlation analysis. These results were consistent with the correlation between total Pb concentration and bioaccessibility (Fig. S2). Similarly, a low bioaccessibility of metallic Pb in the stomach phase was reported in the CHDS [25]. Therefore, the ingestion of metallic Pb from e-waste recycling would not increase Pb bioaccessibility in the stomach. Furthermore, we estimated the influences of other chemical forms of Pb by correlation analysis. We analyzed Pb compounds that had been identified more than three times in the LSF analysis of the Pb L3 -edge EXAFS data: Pb-goethite, Pb(IV) hydrogen phosphate, Pb(II) silicate, and Pb(II) phosphate. These compounds exhibited no significant correlations with bioaccessibility. The specific sample matrix also influenced Pb speciation and its bioaccessibility. A nonparametric Mann-Whitney U test showed that the specific dusts derived from the formal e-waste recycling sites (D1–D5, ref. Table S1) contained significantly less bioaccessible metallic Pb than other samples, as shown in Fig. 3 (p < 0.01). Because the informal sites were open-air facilities (Table S1), weathering could have reduced the amount of metallic Pb by changing its chemical form. Additionally, significantly lower Pb bioaccessibility was observed in the stomach phase in dust from formal sites than in the other samples, as shown in Fig. 1 (p < 0.05). Because e-waste recycling activities might have a high metallic Pb content in dust matrix, a relatively low Pb bioaccessibility was found in formal dust. Conversely, this implied that the soil matrix had a relatively higher Pb bioaccessibility, which was derived from highly soluble Pb species such as Pb citrate, Pb oxides, and Pbgoethite [25] contained in the soil matrix (Table 2 and Fig. 3). We considered the effects of other Pb species on its bioaccessibility. In vivo hydroxylpyromorphite formation from solid samples containing Pb phosphate has been observed in a mouse model [44]. Soil samples originally contained Pb phosphate (12–36% in Table 2 and Fig. 3) and a similar formation mechanism might have occurred in the in vitro system. Therefore, these phosphates had low bioaccessibilities through the gastrointestinal fluids. A previous study reported that Pb citrate has a high bioaccessibility in the stomach phase [25]. The Pb citrate percentage (41% in S2, 19% in S3, and unidentified in S4) followed the same trend as Pb bioaccessibility in the stomach phase (19% in S2, 14% in S3, and <1.5% in S4) (Table 1 and Fig. 1). These results suggest that Pb citrate in soil increased Pb bioaccessibility. In contrast, sample S1 was also soil containing no Pb citrate, but it had a higher Pb bioaccessibility (24%) because it contained a lower proportion of Pb phosphate (low bioaccessibil-

T. Fujimori et al. / Journal of Hazardous Materials 341 (2018) 365–372

ity) and was the only soil sample to contain Pb-goethite (highly bioaccessibility) (Table 1). Although future studies should include a larger sample size to ensure precision, we found several associations between Pb speciation and bioaccessibility that help clarify the absorption of Pb ingested from dust or soil into the systemic circulation. 4. Conclusion In this study, we measured the bioaccessibility of Pb in e-waste surface matrices (dust and soil) using the in vitro PBET assay. Pb was most bioaccessible in the stomach phase. Low Pb bioaccessibility in the gastrointestinal phase suggested a high proportion of insoluble Pb species in the dust matrix derived from an e-waste recycling facility. The Pb L3 -edge EXAFS analysis suggested that Pb in surface matrices existed as multiple Pb species derived from different Pb sources, including e-waste and other materials. There were clear differences in the chemical forms of Pb between the dust and soil matrices. Metallic Pb, a common component of e-waste, was ubiquitous in the samples. We also found Pb adsorbed onto goethite and as oxides and carbonate, implying influences from soil mixing and weathering. Pb phosphate and organic species were only found in the soil samples, suggesting that formation was soil-specific. We identified other Pb compounds in several samples, including Pb silicate, Pb chromate, and Pb(II) hydrogen phosphate. A correlation analysis indicated that the bioaccessibility of metallic Pb was low in the stomach phase. Pb speciation analysis revealed a low bioaccessibility for Pb phosphates and a high bioaccessibility for organic Pb species. Although future studies should include a larger sample size to ensure precision, we found several associations between Pb speciation and bioaccessibility that will improve our understanding of the absorption of Pb ingested from dust or soil into the systemic circulation. The health risk based on bioaccessible Pb was estimated to be much lower than that based on total Pb, due to the lower concentrations. Additionally, e-waste dust and soil contain various other heavy metals and elements [8]. Toxicological effects should be assessed in the future using a multi-element matrix. Acknowledgments We thank A. Eguchi (Chiba University) and K. Bekki (National Institute of Public Health) for supporting sampling campaign in the Philippines; K. Takata and M. Hirayama (NIES) for supporting measurement by ICP-AES/MS; K. Chimura (NIES) for preparation of environmental samples; G. Suzuki and N. Uchida-Noda (NIES) for maintenance of sample stock; K. Nitta, T. Ina, K. Kato, and T. Uruga (BL01B1) for support with Pb L3 -edge XANES measurements at SPring-8 (Proposals 2012B1046, 2013B1046, 2015A1904). And, the author T.F. greatly acknowledge the financial support by Grant for Environmental Research Projects from the Steel Foundation for Environmental Protection Technology (No. 13C-29 (2013) and 14C30-23 (2014)) and the support by a Grant-in-Aid for Challenging Exploratory Research from JSPS, Japan (No. 15K12251). Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.jhazmat.2017.07. 066. References [1] K. Brigden, I. Labunska, D. Santillo, M. Allsopp, Recycling of Electronic Wastes in China and India: Workplace and Environmental Contamination, Technical Note 09/2005 (Section 1), Greenpeace International, Greenpeace Research Laboratories, Department of Biological Sciences, University of Exeter, Exeter EX4 4PS, UK, 2005.

371

[2] M.H. Wong, S.C. Wu, W.J. Deng, X.Z. Yu, Q. Luo, A.O.W. Leung, C.S.C. Wong, W.J. Luksemburg, A.S. Wong, Export of toxic chemicals −a review of the case of uncontrolled electronic-waste recycling, Environ. Pollut. 19 (2007) 131–140. [3] A. Leung, Z.W. Cai, M.H. Wong, Environmental contamination from electronic waste recycling at Guiyu, southeast China, J. Mater. Cycles Waste Manage. 8 (2006) 21–33. [4] N.N. Ha, T. Agusa, K. Ramu, N.P.C. Tu, S. Murata, K.A. Bulbule, P. Parthasaraty, S. Takahashi, A. Subramanian, S. Tanabe, Contamination by trace elements at e-waste recycling sites in Bangalore, India, Chemosphere 76 (2009) 9–15. [5] X. Tang, C. Shen, D. Shi, S.A. Cheema, M.I. Khan, C. Zhang, Y. Chen, Heavy metal and persistent organic compound contamination in soil from Wenling: an emerging e-waste recycling city in Taizhou area China, J. Hazard. Mater. 173 (2010) 653–660. [6] B.N. Lopez, Y.B. Man, Y.G. Zhao, J.S. Zheng, A.O.W. Leung, J. Yao, M.H. Wong, Major pollutants in soils of abandoned agricultural land contaminated by e-waste activities in Hong Kong, Arch. Environ. Contam. Toxicol. 61 (2011) 101–114. [7] J. Li, H. Duan, P. Shi, Heavy metal contamination of surface soil in electronic waste dismantling area: site investigation and source-apportionment analysis, Waste Manage. Res. 29 (2011) 727–738. [8] T. Fujimori, H. Takigami, T. Agusa, A. Eguchi, K. Bekki, A. Yoshida, A. Terazono, F.C. Ballesteros Jr., Impact of metals in surface matrices from formal and informal electronic-waste recycling around Metro Manila the Philippines, and Intra-Asian comparison, J. Hazard. Mater. 221–222 (2012) 139–146. [9] A.O.W. Leung, N.S. Duzgoren-Aydin, K.C. Cheung, M.H. Wong, Heavy metals concentrations of surface dust from e-waste recycling and its human health implications in Southeast China, Environ. Sci. Technol. 42 (2008) 2674–2680. [10] X. Huo, L. Peng, X. Xu, L. Zheng, B. Qiu, Z. Qi, B. Zhang, D. Han, Z. Piao, Elevated blood lead levels of children in Guiyu, an electronic waste recycling town in China, Environ. Health Perspect. 115 (2007) 1113–1117. [11] Y. Guo, X. Huo, Y. Li, K. Wu, J. Liu, J. Huang, G. Zheng, Q. Xiao, H. Yang, Y. Wang, A. Chen, X. Xu, Monitoring of lead cadmium, chromium and nickel in placenta from an e-waste recycling town in China, Sci. Total Environ. 408 (2010) 3113–3117. [12] H. Wang, M. Han, S. Yang, Y. Chen, Q. Liu, S. Ke, Urinary heavy metal levels and relevant factors among people exposed to e-waste dismantling, Environ. Int. 37 (2011) 80–85. [13] T. Wang, J. Fu, Y. Wang, C. Liao, Y. Tao, G. Jiang, Use of scalp hair as indicator of human exposure to heavy metals in an electronic waste recycling area, Environ. Pollut. 157 (2009) 2445–2451. [14] Y. Li, X. Xu, K. Wu, G. Chen, J. Liu, S. Chen, C. Gu, B. Zhang, L. Zheng, M. Zheng, X. Huo, Monitoring of lead load and its effect on neonatal behavioral neurological assessment scores in Guiyu, an electronic waste recycling town in China, J. Environ. Monit. 10 (2008) 1233–1238. [15] Y. Yang, X.S. Lu, D.L. Li, Y.J. Yu, Effects of environmental lead pollution on blood lead and sex hormone levels among occupationally exposed group in an e-waste dismantling area, Biomed. Environ. Sci. 26 (2013) 474–484. [16] Q. Song, J. Li, A systematic review of the human body burden of e-waste exposure in China, Environ. Int. 68 (2014) 82–93. [17] M.V. Ruby, A. Davis, R. Schoof, S. Eberle, C.M. Sellston, Estimation of lead and arsenic bioavailability using a physiologically based extraction test, Environ. Sci. Technol. 30 (1996) 422–430. [18] W.R. Berti, S.D. Cunningham, In-place inactivation of Pb in Pb-contaminated soils, Environ. Sci. Technol. 31 (1997) 1359–1364. [19] S.C. Hamel, K.M. Ellickson, P.J. Lioy, The estimation of the bioaccessibility of heavy metals in soils using artificial biofluids by two novel methods: mass-balance and soil recapture, Sci. Total Environ. 243/244 (1999) 273–283. [20] J. Yang, D.E. Mosby, S.W. Casteel, R.W. Blanchar, Lead immobilization using phosphoric acid in a smelter-contaminated urban soil, Environ. Sci. Technol. 35 (2001) 3553–3559. [21] A. Turner, K.-H. Ip, Bioaccessibility of metals in dust from the indoor environment: application of a physiologically based extraction test, Environ. Sci. Technol. 41 (2007) 7851–7856. [22] M.V. Ruby, R. Schoof, W. Brattin, M. Goldade, G. Post, M. Harnois, E.M. Osby, S.W. Casteel, W. Berti, M. Carpenter, D. Edwards, D. Cragin, W. Chappell, Advances in evaluating the oral bioavailability of inorganics in soil for use in human health risk assessment, Environ. Sci. Technol. 33 (1999) 3697–3705. [23] M.H. Zia, E.E. Codling, K.G. Scheckel, R.L. Chaney, In vitro and in vivo approaches for the measurement of oral bioavailability of lead (Pb) in contaminated soils: a review, Environ. Pollut. 159 (2011) 2320–2327. [24] J. Kumpiene, A. Lagerkvist, C. Maurice, Stabilization of As, Cr Cu, Pb and Zn in soil using amendments − a review, Waste Manage. 28 (2008) 215–225. [25] P.E. Rasmussen, S. Beauchemin, M. Chénier, C. Levesque, L.C.W. MacLean, L. Marro, H. Jones-Otazo, S. Petrovic, L.T. McDonald, H.D. Gardner, Canadian house dust study: lead bioaccessibility and speciation, Environ. Sci. Technol. 47 (2013) 4959–4965. [26] L.C.W. MacLean, S. Beauchemin, P.E. Rasmussen, Lead speciation in house dust from canadian urban homes using EXAFS micro-XRF, and micro-XRD, Environ. Sci. Technol. 45 (2011) 5491–5497. [27] S. Beauchemin, P.E. Rasmussen, T. Mackinnon, M. Chénier, K. Boros, Zinc in house dust: speciation, bioaccessibility, and impact of humidity, Environ. Sci. Technol. 48 (2014) 9022–9029. [28] X. Chi, M. Streicher-Porte, M.Y.L. Wang, M.A. Reuter, Informal electronic waste recycling: a sector review with specialfocus on China, Waste Manage. 31 (2011) 731–742.

372

T. Fujimori et al. / Journal of Hazardous Materials 341 (2018) 365–372

[29] T. Uruga, H. Tanida, Y. Yoneda, K. Takeshita, S. Emura, M. Takahashi, M. Harada, Y. Nishihata, Y. Kubozono, T. Tanaka, T. Yamamoto, H. Maeda, O. Kamishima, Y. Takabayashi, Y. Nakata, H. Kimura, S. Goto, T. Ishikawa, The XAFS beamline BL01B1 at SPring-8, J. Shynchrotron Rad. 6 (1999) 143–145. [30] T.E. Alcacio, D. Hesterberg, J.W. Chou, J.D. Martin, S. Beauchemin, D.E. Sayers, Molecular scale characteristics of Cu(II) bonding in goethite-humate complexes, Geochim. Cosmochim. Acta 65 (2001) 1355–1366. [31] S. Beauchemin, L.C.W. MacLean, P.E. Rasmussen, Lead speciation in indoor dust: a case study to assess old paint contribution in a Canadian urban house, Environ. Geochem. Health 33 (2011) 343–352. [32] K. Xia, W. Bleam, P.A. Helmke, Studies of the nature of Cu2+ and Pb2+ binding sites in soil humic substances using X-ray absorption spectroscopy, Geochim. Cosmochim. Acta 61 (1997) 2211–2221. [33] G. Sarret, A. Manceau, J.L. Hazemann, A. Gomez, M. Mench, EXAFS study of the nature of zinc complexation sites in hurnic substances as a function of Zn concentration, J. Phys. IV France 7 (1997) (C2-799-802). [34] S.M. Webb, SIXpack: a graphical user interface for XAS analysis using IFEFFIT, Phys. Scr. T115 (2005) 1011–1014. [35] A.J. Slowey, S.B. Johnson, J.J. Rytuba, G.E. Brown, Role of organic acids in promoting colloidal transport of mercury from mine tailings, Environ. Sci. Technol. 39 (2005) 7869–7874. [36] T. Fujimori, M. Takaoka, S. Morisawa, Chlorinated aromatic compounds in a thermal process promoted by oxychlorination of ferric chloride, Environ. Sci. Technol. 44 (2010) 1974–1979. [37] T. Ressler, J. Wong, J. Roos, I.L. Smith, Quantitative speciation of Mn-bearing particulates emitted from autos burning (methylcyclopentadienyl)manganese tricarbonyl-added gasolines using XANES spectroscopy, Environ. Sci. Technol. 34 (2000) 950–958.

[38] A. Manceau, M.A. Marcus, N. Tamura, Quantitative speciation of heavy metals in soils and sediments by synchrotron X-ray techniques. Applications of Synchrotron Radiation in Low Temperature, Geochem. Environ. Sci. 49 (2002) 341–428. [39] R. McKenzie, The adsorption of lead and other heavy metals on oxides of manganese and iron, Aust. J. Soil Res. 18 (1980) 61–73. [40] S.L. Brown, I. Clausen, M.A. Chappell, K.G. Scheckel, M. Newville, G.M. Hettiarachchi, High-iron biosolids compost-induced changes in lead and arsenic speciation and bioaccessibility in co-contaminated soils, J. Environ. Qual. 41 (2011) 1612–1622. [41] J.E.S. Barrett, K.G. Taylor, K.A. Hudson-Edwards, J.M. Charnock, Solid-phase speciation of Pb in urban road dust sediment: a XANES and EXAFS study, Environ. Sci. Technol. 44 (2010) 2940–2946. [42] D.H. Moon, J.-W. Park, Y.-Y. Chang, Y.S. Ok, S.S. Lee, M. Ahmad, A. Koutsospyros, J.-H. Park, K. Baek, Immobilization of lead in contaminated firing range soil using biochar, Environ. Sci. Pollut. Res. 20 (2013) 8464–8471. [43] L. Li, K.G. Scheckel, L. Zheng, G. Liu, W. Xing, G. Xiang, Immobilization of lead in soil influenced by soluble phosphate and calcium: lead speciation evidence, J. Environ. Qual. 43 (2014) 468–474. [44] A.L. Juhasz, D. Gancarz, C. Herde, S. McClure, K.G. Scheckel, E. Smith, In situ formation of pyromorphite is not required for the reduction of in vivo Pb relative bioavailability in contaminated soils, Environ. Sci. Technol. 48 (2014) 7002–7009.