Embed Size (px)
Citation preview
lable at ScienceDirect
Environmental Pollution 239 (2018) 384e391
Contents lists avai
Environmental Pollution
journal homepage: www.elsevier .com/locate/envpol
Speciation, bioaccessibility and potential risk of chromium in Amazonforest soils*
Leo J.D. Moreira a, b, Evandro B. da Silva c, Maurício P.F. Fontes d, Xue Liu a, **,Lena Q. Ma a, c, *
a Research Center for Soil Contamination and Remediation, Southwest Forestry University, Kunming, 650224, Chinab Federal Rural University of Amazon, C. P 3017, Parauapebas, Par�a, Brazilc Soil and Water Science Department, University of Florida, Gainesville, FL, 32611, USAd Department of Soils, Federal University of Viçosa, 36570-900, Viçosa, Minas Gerais, Brazil
a r t i c l e i n f o
Article history:Received 6 March 2018Received in revised form5 April 2018Accepted 5 April 2018Available online 16 April 2018
Keywords:SpeciationParticle sizeBioavailabilityBioaccessibilityHexavalent CrRisk assessment
* This paper has been recommended for acceptanc* Corresponding author. Research Center for Soil C
tion, Southwest Forestry University, Kunming, 65022** Corresponding author.
E-mail addresses: [email protected] (E.B. da Silva), [email protected] (M.P.F. Fontes(X. Liu), [email protected] (L.Q. Ma).
https://doi.org/10.1016/j.envpol.2018.04.0250269-7491/Published by Elsevier Ltd.
a b s t r a c t
Even though the Amazon region is widely studied, there is still a gap regarding Cr exposure and its risk tohuman health. The objectives of this study were to 1) determine Cr concentrations in seven chemicalfractions and 6 particle sizes in Amazon soils, 2) quantify hexavalent Cr (CrVI) concentrations using analkaline extraction, 3) determine the oral and lung bioaccessible Cr, and 4) assess Cr exposure risks basedon total and bioaccessible Cr in soils. The total Cr in both A (0e20 cm) and B (80e100 cm) horizons washigh at 2346 and 1864mg kg�1. However, sequential extraction indicated that available Cr fraction waslow compared to total Cr, with Cr in the residual fraction being the highest (74e76%). There was littledifference in total Cr concentrations among particle sizes. Hexavalent Cr concentration was also low,averaging 0.72 and 2.05mg kg�1 in A and B horizon. In addition, both gastrointestinal (21e22mg kg�1)and lung (0.95e1.25mg kg�1) bioaccessible Cr were low (<1.2%). The low bioavailability of soil Cr and itsuniform distribution in different particle sizes indicated that Cr was probably of geogenic origin.Exposure based on total Cr resulted in daily intake> the oral reference dose for children, but not whenusing CrVI or bioaccessible Cr. The data indicated that it is important to consider both Cr speciation andbioaccessibility when evaluating risk from Cr in Amazon soils.
Published by Elsevier Ltd.
1. Introduction
Chromium (Cr) occurs naturally in soils and it is mostly from theparental material, averaging 54mg kg�1 worldwide (Kabata-Pendias, 2011). However, Cr concentrations in soils from ultra-mafic rocks can be as high as 60,000mg kg�1 (Alloway, 2013).Chromium can present as trivalent (CrIII) and hexavalent form(CrVI), with CrVI being of the most concern. Trivalent Cr is oftenpresent as Cr(OH)3 and Cr2O3, which can be adsorbed onto soilsurfaces or form complexes with organic matter, thus presentinglow availability (da Silva et al., 2018b; Apte et al., 2006). Besides,
e by Joerg Rinklebe.ontamination and Remedia-4, China.
(L.J.D. Moreira), ebsilva@ufl.), [email protected]
CrIII is an essential nutrient for humans and animals (Eastmondet al., 2008). Oxidation of CrIII to CrVI in soils can occur in thepresence of O2 and MnO2 or due to human activity (Fendorf et al.,1992; Gress et al., 2015). Hexavalent Cr occurs as anionic species(CrO4
2�, HCrO4� and Cr2O7
�2) and is poorly adsorbed onto soils,thereby presenting high availability (Ko�zuh et al., 2000). Due to itssimilar structure to phosphate and sulfate, CrVI is toxic and causesdiseases including cancer (Costa, 2003).
Exposure to Cr causes nausea, fever, headache, and respiratorydistress, with high exposure leading to serious health problemincluding lung cancer (Wilbur, 2000). Risk assessment is oftenbased on total metal concentration (Cox et al., 2013). However, onlya fraction of the metal is available to be absorbed by oral, respira-tory or skin contact (Li et al., 2016). Various assays have beenemployed to evaluate metal bioavailability by simulating digestiveor respiratory processes (Juhasz et al., 2007; Huang et al., 2016). Thesimple bioaccessibility extraction test (SBET) and Gamble's solutionprotocols are the most employed in bioavailability studies for
L.J.D. Moreira et al. / Environmental Pollution 239 (2018) 384e391 385
human risk assessment studies, which have been applied to soiland dust (De Miguel et al., 2012; Coufalík et al., 2016; Mendozaet al., 2017).
The Amazon region presents remarkable features including highbiodiversity and forest density, extensive rivers, and upland andlowland soils. Ultissols and oxisols are the main soils in the region,which tend to have low metal concentration. However, some soilsin the region from mafic and ultramafic rocks exhibit high metalconcentrations (de Souza et al., 2017). During a highway con-struction in the Amazon forest, Brazil, symptoms such as fever,headache, body weakness and fainting were reported by localworkers. Those symptoms can be associated with various sources,but probably including toxicity from heavy metals. Several studiesreported Hg contamination in the Amazon basin due to its use ingold mining, thus it is the main metal studied in the region(Castilhos et al., 2015; Faial et al., 2015).
However, there is a gap regarding exposure to other metalsincluding Cr and its risk to human health in the Amazon region. Infact, high Cr concentration (142mg kg�1) was reported in theAmazon basin area, being much greater than the backgroundconcentration of 44.8mg kg�1 (dos Santos and Alleoni, 2013). Thus,the objectives of this study was to 1) determine Cr concentrationsin seven chemical fractions and 6 particle sizes in Amazon soils, 2)quantify hexavalent Cr (CrVI) using an alkaline extraction, 3)determine the oral and lung bioaccessible Cr, and 4) assess Crexposure risks based on total and bioaccessible Cr in Amazon soils.
2. Materials and methods
2.1. Soil characterization
In October of 2013, aleatory Oxisol soil samples were collectedfrom a pristine area in the Amazon Forest, Brazil (0�20021.3500N;66�3808.9200W) (Fig. 1). Five samples of each representative horizon(A - 0e20 and B - 80e100 cm) were collected and homogenized.Soil samples were air-dried and sieved through a 2mm sieve and
Fig. 1. Soils sampling location in t
kept in closed container at room temperature.Soil characterization was performed following Donagema et al.
(2011). Briefly, pH was determined in water using a 1:2.5 soil:so-lution ratio. The cation exchange capacity (CEC) was calculated bysumming exchangeable ions including Ca, Mg, Na, K, H and Al.Organic carbon (OC) content was determined by the Walkley-Blackmethod (Burt, 2004). Particle size distribution was determinedusing the pipette method with NaOH solution being the dispersingagent (Gee and Bauder, 1986). Sand fraction was separated bypassing the soil through a 0.2mm mesh sieve, while the silt andclay fractions were separated by sedimentation.
Concentrations of As, Cd, Co, Cr, Cu, Mn, Ni, Pb, Se and Zn weredetermined by digesting the soil using the USEPA Method 3050Bfollowing da Silva et al. (2018a). Briefly, 1 g of soil sample wassuspended in 15mL of 1:1 nitric acid:water solution and heated at105 �C for 6 h. After cooling, 1mL of 30% H2O2 was added anddigested for an additional 30min before bringing samples to a50mL volume with double DI water. Metal concentrations wereanalyzed using inductively coupled plasma mass spectrometry(ICP-MS NexIon 300x, PerkinElmer Corp., Norwalk, CT).
2.2. Chemical and size fractionation of soil Cr
Soil Cr was separated into 7 fractions following sequentialextraction method of Silveira et al. (2006). They included: (E1)soluble-exchangeable, (S2) surface adsorbed, (O3) organic matter,(M4) Mn oxides, (A5) poor crystalline Fe oxides, (C6) crystalline Feoxides, and (R7) residual fractions. Briefly, 1 g of soil was extractedusing: 15mL 0.1M CaCl2, 30mL 1 M NaOAC at pH 5, 5mL NaOCl atpH 8.5, 30mL 0.05M NH2OH/HCl at pH 2, 30 mL 0.2 M oxalic acidþ0.2 M NH4 oxalate at pH 3, 40mL 6M HCl and HNO3-HCl digestion.After each extraction, samples were centrifuged at 2000 rpm for10min and then filtered using Q2 filter paper (Fisher). Chromiumconcentration was determined by ICP-MS.
Soil Cr was also separated into 6 particle sizes including <2000,425e2000, 250e425, 150e250, 105e150 and< 105 mm. Chromium
he Amazon Rainforest, Brazil.
Table 2Human exposure and risk assessment variables values from USEPA (2001).
Variable Values
Adults Children
Average Daily Dose - ADD (mg kg d�1) e e
Concentration - C (mg kg�1) e e
Ingestion rate - IR (mg d�1) 100 200Exposure frequency - EF (d y�1) 279 279Exposure duration - ED (y) 24 4Body weight - BW (kg) 70 16Average time e ATAT¼ no carcinogenic effects (ED x 365 d) 1460 1460AT¼ carcinogenic effects (70 y x 365 d) 25,550 25,550
L.J.D. Moreira et al. / Environmental Pollution 239 (2018) 384e391386
concentration was analyzed by graphite furnace atomic absorptionspectrophotometry (GFAAS; Varian AA240Z, Victoria, Australia).
2.3. Hexavalent Cr
The modified alkaline digestion method 3060A (de Oliveiraet al., 2015) was used to determine CrVI concentration in particlesizes< 2000, 425e2000, 250e425, 150e250, 105e150and< 105 mm. Briefly, alkaline solution was prepared using 0.28MNaCO3 and 0.5M NaOH (pH> 12). In 1 g of soil was added 40mL ofthe alkaline solution. To suppress CrIII oxidation to CrVI, ~400mg ofMgCl and 1M phosphate buffer (0.5MK2HPO and 0.5 KH2PO4)were added. Samples were heated to 90e95 �C for 1 h, stirringevery 15min. After that, samples were centrifuged for 10min at3000 rpm and then filtered using Q2 filter paper (Fisher). Blanksand spike solution of CrIII and CrVI were used as quality control andthey were all within the limits. Cr concentrations were analyzedusing GFAAS.
2.4. Oral and lung bioavailability
In vitro assays are widely used to measure metal bioavailabilityin soils using simulated human gastrointestinal and lung fluids (Liet al., 2016; Juhasz et al., 2007). The simple bioaccessibilityextraction test (SBET) and Gamble's solution protocols wereselected (Coufalík et al., 2016; Mendoza et al., 2017).
Chromium oral bioavailability was determined using SBET test(USEPA, 2012), which uses 0.4M glycine to simulate gastric solutionat pH¼ 1.5 adjusted with concentrated HCl. Briefly, in 0.5 g of soilwas added 50mL of gastric solution and placed in water bath(37 �C) with shaking (60 rpm) for 1 h. Then, samples were centri-fuged (4000 rpm) and filtered using Q2 filter paper (Fisher). Tosimulate accidental ingestion by hand-to-mouth pathway, onlyparticle sizes <250 mm were tested, including 150e250, 105e150and< 105 mm.
Chromium lung bioavailability was determined followingMidander et al. (2007). Gamble's solution was prepared by mixingthe reagents in proper order to prevent precipitation (Table 1), then50mL of Gamble's solution was added in a 50mL vial containing0.5 g of soil. In dark conditions, samples were placed in water bath(37 �C± 2) with shaking (25 rpm) during 1 h. After that, sampleswere centrifuged (4000 rpm) and filtered using Q2 filter paper(Fisher). Tested particle sizes included 53e105 mm and <53 mm andCr concentrations were determined by GFAAS.
2.5. Risk assessment
Risk Assessment based on average daily dose (ADD) was con-ducted to identify potential effects on humans health due toexposure to soil Cr (Cachada et al., 2016). The ADDwas calculated toevaluate the potential risks for adults and children based on soil
Table 1Gamble's solution extraction composition (mg L�1) and addition order.
Addition order Compounds
1 Magnesium Chloride - MgCl2 Sodium chloride - NaCl3 Potassium chloride - KCl4 Disodium hydrogen phosphat5 Sodium sulphate - NaSO4
6 Calcium chloride dihydrate - C7 Sodium acetate - C2H3O2Na8 Sodium hydrogen carbonate -9 Sodium citrate dihydrate e C6
ingestion using Equation (1). The variables and equations followedUSEPA (2001).
ADD ¼ C x IR x EF x EDBW x AT
(1)
Where ADD (mg kg d�1)¼ average daily dose; C (mg kg�1)¼ con-centration; IR (mg d�1)¼ ingestion rate; EF (d y�1)¼ exposurefrequency; ED (y)¼ exposure duration; BW (kg)¼ body weight,and AT¼average time, with no carcinogenic effects (ED x 365 d)and carcinogenic effects (70 y x 365 d). The specific values used foradults and children for each variable are presented in Table 2.
3. Results and discussions
3.1. Soil characterization
Soil characterization and heavy metal concentrations are shownin Table 3. The soil was classified as an Oxisol, having low CEC(0.17e0.57 cmol dm�3), OC content (2.53e3.68%) and pH(4.45e4.84), typical of the Amazon region soils (Quesada et al.,2010). The soil has a clayey texture, including hematite, goethiteand, gibbsite minerals besides kaolinite and quartz minerals (datanot shown).
Overall, for both A and B horizons, Cr (2346 and 1864mg kg�1),Ni (62 and 63mg kg�1) and Zn (14.7 and 16.4mg kg�1) concen-trations were higher than the Amazon region background con-centrations (44.8, 2.1 and 3.0mg kg�1 for Cr, Ni and Zn) (dos Santosand Alleoni, 2013). Difference can be attributed to soil parentalmaterial (ferruginous laterite crust), which is rich in heavy metalsand tend to enrichmetals during weathering process (Garnier et al.,2013).
3.2. Soil Cr fractionation
According to Shaheen and Rinklebe (2014), metal bioavailabilityvaries with different fractions, decreasing in the order of soluble-exchangeable (E1)> surface adsorbed (S2)>Mn oxides¼ poor
Concentration (mg L�1)
956019298
e - Na2HPO4 12663
aCl2 $ 2H2O 368574
NaHCO3 2604H5Na3O7 $ 2H2O 97
Table 3Chemical and physical properties Amazon soil (n¼ 3).
Horizons
A (0e20 cm) B (80e100 cm)
CEC cmolc dm�3 0.57 0.17OM % 3.68 2.53Sand 50 50Silt 8 10Clay 42 40
pH 4.45 4.84
V mg kg�1 105 102Mn 207 193Cr 2346 1864Co 3.67 3.41Ni 62.8 63.5Cu 6.55 6.14Zn 14.7 16.4As <0.01 0.01Se 0.56 0.41Cd 0.02 0.02Pb 3.03 2.98
Fig. 2. Chromium fractionation in A and B horizons in an Amazon soil. Numbers insidebars represent Cr percentage from each fraction (n¼ 3).
L.J.D. Moreira et al. / Environmental Pollution 239 (2018) 384e391 387
crystalline Fe oxides (M4 and A5)> organic matter (O3)> crystal-line Fe oxides (C6)> residual (R7) fraction. The data showed that74e76% of the Cr was in the residual fraction, with only 24e26%being in the non-residual fraction for both A and B horizons (Fig. 2).Soluble-exchangeable and surface adsorbed fractions (E1 þ S2)were<1.5mg kg�1 for both horizons, contributing< 0.5% to total Cr.Those fractions are available and easily leached out, especially intropical regions. This is because metals in these fractions are likelyfrom anthropogenic contribution, therefore being more available(Shaheen and Rinklebe, 2014). Geogenic metals in E1 þ S2 fractionusually present low bioavailability than anthropogenic sources. Thelow values of E1 þ S2 fractions indicated limited anthropogenicimpact in this soil (Sierra et al., 2007).
Fractions O3, M4 and A5 are potential available, being affectedby soil pH and Eh (Shaheen and Rinklebe, 2014; Ma and Rao, 1997).The sum of M4 þ A5 fractions corresponded to 2.0e2.1% of total Cr(40e46 mg kg�1) (Fig. 2). Despite of the low percentage, thosefractions can be potentially available, causing risk to human health(Agnieszka and Barbara, 2012). Furthermore, Mn oxides (M4 frac-tion) can oxidize CrIII to CrVI, increasing its availability and therebythe risk to human health (Garnier et al., 2013). In both horizons,fractions associated with organic matter (O3; 123 and157mg kg�1), crystalline Fe oxides (C6; 403 and 273mg kg�1) andresidual (R7; 1630 and 1493mg kg�1) presented the highest con-centrations. The high Cr in the residual fraction (R7) indicated thatCr had low availability in soil.
3.3. Total Cr and CrVI in different particles sizes
Chromium concentrations can vary with particle sizes, so it isimportant to understand Cr distribution among different sizes toassess its potential risk. Fine particles have great capacity to holdmetals due to their large specific surface area. Besides, fine particlesare also easily transported by water or dispersed as dust, thuspresenting greater risks to the environment (Gong et al., 2014). Inthe Amazon soil, however, there was little difference in Cr con-centrations among six particle sizes for both horizons, averaging2020mg kg�1 for all particles sizes excluding 0.425e2mm size in Ahorizon (2700mg kg�1; Fig. 3AB). It was much greater than thebackground concentration for Amazon soils (44.8mg kg�1),exceeding the Brazilian Investigation Value of 400mg kg�1
(CONAMA, 2009). The even distribution of Cr concentrationsamong 6 particle size fraction was another indication that soil Crwas of geogenic origin. Otherwise, more metals tend to accumulatein finer particles due to their large surface areas (Gasparatos, 2013).
In addition to total Cr, distributions of toxic CrVI among differentsizes are shown in Fig. 3CD. Average CrVI concentration was 0.72and 2.05mg kg�1 for A and B horizons. The lower CrVI content inthe surface horizon might be due to microbial activities and thepresence of OC, reducing CrVI to CrIII (Rajapaksha et al., 2013). Incomparison, CrVI in Amazon soil was lower than Glasgow soils at6.4mg kg�1 (Broadway et al., 2010), whichwere contaminatedwithchromite ore processing residue. Besides, it was lower than ultra-mafic soils in Brazil at 446mg kg�1 (Garnier et al., 2009). In the Ahorizon, there was no relation between particle size and CrVIconcentration, however, in the B horizon, CrVI concentrationsdecreased as particle size decreased (Fig. 3CD). Moreover, CrVIconcentrations in 0.105e0.150mm fraction were 1.2 and1.5mg kg�1 for A and B horizons. This fraction had high availability,which can be dispersed by water and dust, which can be easilyingested or inhaled. Although, only one fraction showed high CrVIconcentration in the A horizon, it still poses risk to health human ashuman contacts with surface soil the most.
Hexavalent CrVI in different particles sizes accounted for0.02e0.06% and 0.01e0.14% for A and B horizons. Although being
Fig. 3. Total Cr (AB; mg kg�1) and hexavalent Cr (CD; mg kg�1) distribution within particles sizes of the A and B horizons in an Amazon soil. Numbers inside bars represent CrVIpercentage. Different letters indicate statistic difference at p < 0.05 (n¼ 3).
L.J.D. Moreira et al. / Environmental Pollution 239 (2018) 384e391388
<0.2%, it was at 0.72e2mg kg�1, which is susceptible to leaching.Becquer et al. (2003) showed that CrVI was less adsorbed into soilthan CrIII, hence presenting high availability with potential envi-ronmental consequences.
3.4. Cr oral and lung bioaccessibility and risk assessment
Besides total Cr and Cr speciation in different size fractions, oralbioaccessible Cr based on simulated gastrointestinal assay wasdetermined, which averaged 21e22mg kg�1 (Fig. 4AB). The resultswere higher than uncontaminated soils (3.39e8.84mg kg�1), butsimilar to urban soils (29.0e400mg kg�1) and road soils (up to~30mg kg�1) (Cox et al., 2013; Okorie et al., 2011; Sialelli et al.,2011). Even at low bioaccessibility of 1.4% for A horizon, it repre-sented 27mg kg�1 Cr, which still presents potential risk to humans.In both horizons, bioaccessible Cr concentration increased as par-ticle size decreased (Fig. 4AB). Particles <105 mm showed higherbioaccessible Cr at 27mg kg�1. Similar results were observed inurban soils from Spain and Italy (Okorie et al., 2011). High Cr con-centration in fine particles in the A horizon can cause health issuesas they are easily dispersed and digested by the gastric system(Madrid et al., 2008).
Besides oral bioaccessibility, we also measured lung bio-accessibility of Cr in the soil, averaging 1.25 and 0.95mg kg�1 for Aand B horizon, much lower than oral bioaccessible Cr at21e22mg kg�1 (Fig. 4). Lung bioaccessible Cr from contaminatedsoils was 17 times greater than those in Broadway et al. (2010). Thelow value indicated that the soil probably was not contaminated by
human activities. Low lung bioaccessible Cr can be explained by thelow solubility of crystalline Fe oxides and residual fractions sincelung fluid simulates solution pH at 7.2. Besides, Cr from soluble andexchangeable phases were low, consistent with soil fractionationdata (Fig. 2).
It is known that particle deposition in the respiratory tract de-pends on particle size. While finer particles (<4 mm) are able topenetrate deeper into the alveolar region of the respiratory system,particles of 10e100 mm are inhalable and tend to be deposited inthe upper respiratory tract (Colombo et al., 2008). Similar to oralbioaccessibility data, lung bioaccessible Cr was higher in the Ahorizon, indicating higher risk with surface soils. In addition, bio-accessible Cr increased as particle size decreased (Fig. 4CD). Thedata suggested that the finest fraction of surface soil showed thehighest risk with both high oral and lung bioaccessible Cr.
Based on USEPA (2001), reference dose (RfD) is an estimateddaily exposure to humans that is likely not to cause risk during alifetime exposure. The potential negative effects by exposure toAmazon soil with high Cr concentration from geogenic source wereassessed. The ADD was calculated based on total (2025 and2022mg kg�1), oral bioaccessible (27 and 26mg kg�1) and CrVI (1.2and 3.37mg kg�1) concentrations in both A and B horizons (Fig. 5).As expected, the ADD based on total Cr concentration was muchhigher than oral bioaccessible Cr or CrVI. This difference was ex-pected due to the much greater value of total Cr. Other studies alsonoticed similar results (Gress et al., 2014; Hamad et al., 2014). OnlyADD for children based on total Cr was greater than the oral RfD at3� 10�3mg kg�1 d�1 based on non-carcinogenic risks (Fig. 5BD).
Fig. 4. Oral (AB; mg kg�1) and lung (CD; mg kg�1) bioavailable Cr in different particle sizes of A and B horizons in an Amazon soil. Numbers inside bars represent bioaccessible Cr% inrelation total Cr. Different letters indicate statistic difference at p< 0.05 (n¼ 3).
Fig. 5. Average daily dose (ADD) of Cr ingestion for carcinogenic (AC) and non-carcinogenic (BD) risks of the A horizon and B horizon in an Amazon soil (n¼ 3).
L.J.D. Moreira et al. / Environmental Pollution 239 (2018) 384e391 389
L.J.D. Moreira et al. / Environmental Pollution 239 (2018) 384e391390
The data indicated that the risk from soil Cr was much lower afterconsidering Cr speciation or bioaccessible Cr. As such, it is impor-tant to consider both Cr speciation and bioaccessibility whenevaluating the risk from soil Cr.
4. Conclusion
For the Amazon soils, Cr concentrations at 1864e2346mg kg�1
were much greater than the background concentration in AmazonState soils, most likely due to the parental material. The lowcontribution of available fractions (E1 and S2) and little differencein total Cr concentrations within particles sizes indicated lowanthropogenic impact in the area, i.e., soil Cr is of geogenic origin.CrVI was higher in the subsurface horizon, which might be relatedto soil biota activities due to larger amount of organic matter in thesurface horizon. Besides, in A horizon there was no relation be-tween particle size and CrVI concentration, however, in the B ho-rizon, CrVI concentrations decreased as particle size decreased.Both oral and lung bioaccessible Cr increased as particle sizedecreased in both horizons. ADD based on total Cr concentrationwas much higher than oral bioaccessible Cr or CrVI, with onlychildren value being higher than the RfD based on non-carcinogenic risks. The data indicated that it is important toconsider Cr speciation and bioaccessibility when evaluating the riskfrom soil Cr.
Acknowledgments
Senior author was supported in part by the Brazilian NationalCouncil of Scientific and Technological Development (CNPq) (Pro-cess N� 233939/2014-8), University of Florida and Federal Univer-sity of Viçosa. Second author was supported by CNPq project N�
246758/2012-0.
References
Agnieszka, J., Barbara, G., 2012. Chromium, nickel and vanadium mobility in soilsderived from fluvioglacial sands. J. Hazard. Mater 237, 315e322.
Alloway, B.J., 2013. Heavy Metals in Soils: Trace Metals and Metalloids in Soils andTheir Bioavailability Third Edition XVIII, 614 p.
Apte, A.D., Tare, V., Bose, P., 2006. Extent of oxidation of Cr (III) to Cr (VI) undervarious conditions pertaining to natural environment. J. Hazard. Mater 128,164e174.
Becquer, T., Quantin, C., Sicot, M., Boudot, J.P., 2003. Chromium availability in ul-tramafic soils from. NewCaledonia. Sci. Total Environ. 301, 251e261.
Broadway, A., Cave, M.R., Wragg, J., Fordyce, F.M., Bewley, R.J.F., Graham, M.C.,Ngwenya, B.T., Farmer, J.G., 2010. Determination of the bioaccessibility ofchromium in Glasgow soil and the implications for human health risk assess-ment. Sci. Total Environ. 409, 267e277.
Burt, R., 2004. Soil Survey Laboratory Methods Manual.Cachada, A., da Silva, E.F., Duarte, A.C., Pereira, R., 2016. Risk assessment of urban
soils contamination: the particular case of polycyclic aromatic hydrocarbons.Sci. Total Environ. 551, 271e284.
Castilhos, Z., Rodrigues-Filho, S., Cesar, R., Rodrigues, A.P., Villas-Boas, R., de Jesus, I.,Lima, M., Faial, K., Miranda, A., Brabo, E., et al., 2015. Human exposure and riskassessment associated with mercury contamination in artisanal gold miningareas in the Brazilian Amazon. Environ. Sci. Pollut. Res. 22, 11255e11264.
Colombo, C., Monhemius, A.J., Plant, J.A., 2008. Platinum, palladium and rhodiumrelease from vehicle exhaust catalysts and road dust exposed to simulated lungfluids. Ecotoxicol. Environ. Saf. 71, 722e730.
CONAMA, Conselho Nacional deMeio Ambiente, 2009. Resoluç~ao CONAMA No 420.Di�ario Of. da Uni~ao no 249 2013, pp. 81e84.
Costa, M., 2003. Potential hazards of hexavalent chromate in our drinking water.Toxicol. Appl. Pharmacol. 188, 1e5.
Coufalík, P., Miku�ska, P., Matou�sek, T., Ve�ce\vra, Z., 2016. Determination of thebioaccessible fraction of metals in urban aerosol using simulated lung fluids.Atmos. Environ. 140, 469e475.
Cox, S.F., Chelliah, M.C.M., McKinley, J.M., Palmer, S., Ofterdinger, U., Young, M.E.,Cave, M.R., Wragg, J., 2013. The importance of solid-phase distribution on theoral bioaccessibility of Ni and Cr in soils overlying Palaeogene basalt lavas,Northern Ireland. Environ. Geochem. Health 35, 553e567.
de Oliveira, L.M., Lessl, J.T., Gress, J., Tisarum, R., Guilherme, L.R.G., Ma, L.Q., 2015.Chromate and phosphate inhibited each other's uptake and translocation inarsenic hyperaccumulator Pteris vittata L. Environ. Pollut. 197, 240e246.
De Miguel, E., Mingot, J., Chac�on, E., Charlesworth, S., 2012. The relationship be-tween soil geochemistry and the bioaccessibility of trace elements in play-ground soil. Environ. Geochem. Health 34, 677e687.
de Souza, E.S., Texeira, R.A., da Costa, H.S.C., Oliveira, F.J., Melo, L.C.A., Faial, K.,do, C.F., Fernandes, A.R., 2017. Assessment of risk to human health fromsimultaneous exposure to multiple contaminants in an artisanal gold mine inSerra Pelada, Par�a, Brazil. Sci. Total Environ. 576, 683e695.
da Silva, E.B., de Oliveira, L.M., Wilkie, A.C., Liu, Y., Ma, L.Q., 2018a. Arsenic removalfrom As-hyperaccumulator Pteris vittata biomass: Coupling extraction withprecipitation. Chemosphere 193, 288e294. https://doi.org/10.1016/j.chemosphere.2017.10.116.
da Silva, E.B., Li, S., de Oliveira, L.M., Gress, J., Dong, X., Wilkie, A.C., Townsend, T.,Ma, L.Q., 2018b. Metal leachability from coal combustion residuals underdifferent pHs and liquid/solid ratios. J. Hazard. Mater. 341, 66e74. https://doi.org/10.1016/j.jhazmat.2017.07.010.
Donagema, G.K., de Campos, D.V.B., Calderano, S.B., Teixeira, W.G., Viana, J.H.M.,2011. Manual de m�etodos de an�alise de solo. Embrapa Solos-Documentos(INFOTECA-E).
Dos Santos, S.N., Alleoni, L.R.F., 2013. Reference values for heavy metals in soils ofthe Brazilian agricultural frontier in Southwestern Amazonia. Environ. Monit.Assess. 185, 5737e5748.
Eastmond, D.A., MacGregor, J.T., Slesinski, R.S., 2008. Trivalent chromium: assessingthe genotoxic risk of an essential trace element and widely used human andanimal nutritional supplement. Crit. Rev. Toxicol. 38, 173e190.
Faial, K., Deus, R., Deus, S., Neves, R., Jesus, I., Santos, E., Alves, C.N., Brasil, D., 2015.Mercury levels assessment in hair of riverside inhabitants of the Tapajos River,Para State, Amazon, Brazil: fish consumption as a possible route of exposure.J. Trace Elem. Med. Biol. 30, 66e76.
Garnier, J., Quantin, C., Echevarria, G., Becquer, T., 2009. Assessing chromate avail-ability in tropical ultramafic soils using isotopic exchange kinetics. J. soilssediments 9, 468e475.
Garnier, J., Quantin, C., Guimar~aes, E.M., Vantelon, D., Montarg�es-Pelletier, E.,Becquer, T., 2013. Cr (VI) genesis and dynamics in Ferralsols developed fromultramafic rocks: the case of Niquelandia, Brazil. Geoderma 193, 256e264.
Gasparatos, D., 2013. Sequestration of heavy metals from soil with FeeMn con-cretions and nodules. Environ. Chem. Lett. 11, 1e9.
Gee, G.W., Bauder, J.W., 1986. Particle-size analysis. Methods soil anal. Part1dphysical mineral. methods 383e411.
Gong, C., Ma, L., Cheng, H., Liu, Y., Xu, D., Li, B., Liu, F., Ren, Y., Liu, Z., Zhao, C., others,2014. Characterization of the particle size fraction associated heavy metals intropical arable soils from Hainan Island, China. J. Geochem. Explor 139,109e114.
Gress, J., de Oliveira, L.M., da Silva, E.B., Lessl, J.M., Wilson, P.C., Townsend, T.,Ma, L.Q., 2015. Cleaning-induced arsenic mobilization and chromium oxidationfrom CCA-wood deck: potential risk to children. Environ. Int. 82, 35e40.
Gress, J.K., Lessl, J.T., Dong, X., Ma, L.Q., 2014. Assessment of children's exposure toarsenic from CCA-wood staircases at apartment complexes in Florida. Sci. TotalEnviron. 476, 440e446.
Hamad, S.H., Schauer, J.J., Shafer, M.M., Al-Rheem, E.A., Skaar, P.S., Heo, J., Tejedor-Tejedor, I., 2014. Risk assessment of total and bioavailable potentially toxic el-ements (PTEs) in urban soils of BaghdadeIraq. Sci. Total Environ. 494, 39e48.
Huang, X., Betha, R., Tan, L.Y., Balasubramanian, R., 2016. Risk assessment of bio-accessible trace elements in smoke haze aerosols versus urban aerosols usingsimulated lung fluids. Atmos. Environ. 125, 505e511.
Juhasz, A.L., Smith, E., Weber, J., Rees, M., Rofe, A., Kuchel, T., Sansom, L., Naidu, R.,2007. Comparison of in vivo and in vitro methodologies for the assessment ofarsenic bioavailability in contaminated soils. Chemosphere 69, 961e966.
Kabata-Pendias, A., 2011. Trace Elements in Soils and Plants. CRC Press.Ko�zuh, N., �Stupar, J., Gorenc, B., 2000. Reduction and oxidation processes of chro-
mium in soils. Environ. Sci. Technol. 34, 112e119.Li, S.-W., Li, H.-B., Luo, J., Li, H.-M., Qian, X., Liu, M.-M., Bi, J., Cui, X.-Y., Ma, L.Q., 2016.
Influence of pollution control on lead inhalation bioaccessibility in PM 2.5: acase study of 2014 Youth Olympic Games in Nanjing. Environ. Int. 94, 69e75.
Ma, L.Q., Rao, G.N., 1997. Chemical fractionation of cadmium, copper, nickel, andzinc in contaminated soils. J. Environ. Qual. 26, 259e264.
Madrid, F., Biasioli, M., Ajmone-Marsan, F., 2008. Availability and bioaccessibility ofmetals in fine particles of some urban soils. Arch. Environ. Contam. Toxicol. 55,21e32.
Mendoza, C.J., Garrido, R.T., Quilodr�an, R.C., Segovia, C.M., Parada, A.J., 2017. Eval-uation of the bioaccessible gastric and intestinal fractions of heavy metals incontaminated soils by means of a simple bioaccessibility extraction test. Che-mosphere 176, 81e88.
Midander, K., Wallinder, I.O., Leygraf, C., 2007. In vitro studies of copper releasefrom powder particles in synthetic biological media. Environ. Pollut. 145,51e59.
Okorie, A., Entwistle, J., Dean, J.R., 2011. The application of in vitro gastrointestinalextraction to assess oral bioaccessibility of potentially toxic elements from anurban recreational site. Appl. Geochem. 26, 789e796.
Quesada, C.A., Lloyd, J., Schwarz, M., Pati~A$\pm$o, S., Baker, T.R., Czimczik, C.,Fyllas, N.M., Martinelli, L., Nardoto, G.B., Schmerler, J., others, 2010. Variations inchemical and physical properties of Amazon forest soils in relation to theirgenesis. Biogeosciences 7.
Rajapaksha, A.U., Vithanage, M., Ok, Y.S., Oze, C., 2013. Cr (VI) formation related to Cr(III)-muscovite and birnessite interactions in ultramafic environments. Environ.Sci. Technol. 47, 9722e9729.
L.J.D. Moreira et al. / Environmental Pollution 239 (2018) 384e391 391
Shaheen, S.M., Rinklebe, J., 2014. Geochemical fractions of chromium, copper, andzinc and their vertical distribution in floodplain soil profiles along the CentralElbe River, Germany. Geoderma 228, 142e159.
Sialelli, J., Davidson, C.M., Hursthouse, A.S., Ajmone-Marsan, F., 2011. Human bio-accessibility of Cr, Cu, Ni, Pb and Zn in urban soils from the city of Torino. Italy.Environ. Chem. Lett. 9, 197e202.
Sierra, M., Martínez, F.J., Aguilar, J., 2007. Baselines for trace elements and evalua-tion of environmental risk in soils of Almería (SE Spain). Geoderma 139,209e219.
Silveira, M.L., Alleoni, L.R.F., O’connor, G.A., Chang, A.C., 2006. Heavy metal
sequential extraction methodsda modification for tropical soils. Chemosphere64, 1929e1938.
USEPA, 2012. EPA 9200.2e86 April 2012 Standard Operating Procedure for anin Vitro Bioaccessibility Assay for Lead in Soil 1e16.
USEPA, 2001. Risk assessment Guidance for Superfund (RAGS) volume III - Part A:process for conducting Probabilistic risk assessment, Appendix B. Off. Emerg.Remedial Response U.S. Environ. Prot. Agency III, pp. 1e385.
Wilbur, S.B., 2000. Toxicological Profile for Chromium. US Department of Healthand Human Services. Public Health Service (Agency for Toxic Substances andDisease Registry).
