ARTICLE pubs.acs.org/est
Hg Speciation and Stable Isotope Signatures in Human Hair As a Tracer for Dietary and Occupational Exposure to Mercury Laure Laffont,*,†,‡,§ Jeroen E. Sonke,*,†,|| Laurence Maurice,†,^ Selma Luna Monrroy,# Jaime Chincheros,# David Amouroux,r and Philippe Behra‡,§ †
)
Observatoire Midi-Pyrenees, Laboratoire de Geosciences Environnement Toulouse, Universite Paul Sabatier Toulouse III, 14 avenue Edouard Belin, 31400 Toulouse, France ‡ LCA (Laboratoire de Chimie AgroIndustrielle), INPT, Universite de Toulouse, ENSIACET, 4 allee Emile Monso, F-31030 Toulouse Cedex 4, France § LCA (Laboratoire de Chimie AgroIndustrielle), INRA, F-31030 Toulouse Cedex 4, France GET, CNRS, F-31400 Toulouse, France ^ GET, IRD, F-31400 Toulouse, France # Laboratorio de Calidad Ambiental, Instituto de Ecología, Universidad Mayor de San Andres (UMSA), Cota Cota Calle 27, La Paz, Bolivia r LCABIE-IPREM UMR 5254, CNRS, Universite de Pau et des Pays de l’Adour, Helioparc Pau Pyrenees 2, Avenue Pierre Angot, F-64053 PAU Cedex 9, France
bS Supporting Information ABSTRACT: Exposure of humans and wildlife to various inorganic and organometallic forms of mercury (Hg) may induce adverse health effects. While human populations in developed countries are mainly exposed to marine fish monomethylmercury (MMHg), this is not necessarily the case for developing countries and diverse indigenous people. Identification of Hg exposure sources from biomonitor media such as urine or hair would be useful in combating exposure. Here we report on the Hg stable isotope signatures and Hg speciation in human hair across different gold miner, indigenous and urban populations in Bolivia and France. We found evidence for both massdependent isotope fractionation (MDF) and mass-independent isotope fractionation (MIF) in all hair samples. Three limiting cases of dominant exposure to inorganic Hg (IHg), freshwater fish MMHg, and marine fish MMHg sources are used to define approximate Hg isotope source signatures. Knowing the source signatures, we then estimated Hg exposure sources for the Bolivian gold miner populations. Modeled IHg levels in hair correspond well to measured IHg concentrations (R = 0.9), demonstrating that IHg exposure sources to gold miners can be monitored in hair samples following either its chemical speciation or isotopic composition. Different MMHg and inorganic exposure levels among gold miners appear to correspond to living and working conditions, including proximity to small towns, and artisanal vs large scale mining activity.
’ INTRODUCTION The biogeochemical cycle and impact of mercury (Hg) in the environment has been studied for many decades. Depending on concentration and speciation, Hg can induce serious health effects in wildlife and humans. In the aquatic environment, Hg is mainly present as Hg(0), Hg(II) complexes, and monomethylmercury (MMHg) complexes. MMHg is the result of methylation of inorganic Hg (IHg) by bacteria and/or abiotic reactions in aquatic systems.1 Because the MMHg absorption rate usually exceeds both excretion and degradation rates this species is bioaccumulated and bioamplified in the foodchain.24 Excessive exposure to MMHg in humans affects neurological functions and causes trembling, eyesight problems, and coordination r 2011 American Chemical Society
disorders.4 Contamination by IHg, mainly Hg(0) vapor, affects the urinary system and the neurological center and can provoke gingivitis, kidney problems, cancers, trembling, and neuropsychiatric troubles.5 Hg and MMHg contamination has been extensively studied in exposed populations such as fish-eating communities and workers who handle Hg(0).6 Hg(0) contamination is usually assessed by total Hg concentration in corrected creatinine in urine, while Received: July 8, 2011 Accepted: October 18, 2011 Revised: October 11, 2011 Published: October 18, 2011 9910
dx.doi.org/10.1021/es202353m | Environ. Sci. Technol. 2011, 45, 9910–9916
Environmental Science & Technology MMHg exposure is evaluated by total Hg concentration analysis in hair and/or blood.7,8 If humans undergo Hg(0) exposure and dietary Hg(II) and MMHg intake, hair can contain IHg as well as MMHg.9 Human Hg exposure in developed countries is predominantly due to consumption of marine food containing MMHg.6,10 However, numerous populations face alternative exposure pathways, including Hg from dental amalgams, urban air pollution, occupational exposure such as gold mining, or more recently observed dietary exposure from rice.11,12 There is a substantial interest in developing a tracer tool that may distinguish and quantify the different sources of Hg exposure. The use of Hg stable isotope signatures to trace processes and sources of Hg in the environment is emerging as a powerful tool.1319 Hg has seven stable isotopes: 196Hg, 198Hg, 199Hg, 200 Hg, 201Hg, 202Hg, and 204Hg which can be fractionated during physical, chemical, and biological processes such as methylation/ demethylation, vaporization/condensation, or oxidation/reduction. Two types of Hg stable isotope fractionation have been documented: 1. mass-dependent fractionation (MDF) and 2. massindependent fractionation (MIF) of the odd isotopes 199Hg and 201 Hg, possibly as a result of nuclear volume effect (NVE)20,21 and/or magnetic isotope effects (MIE).22 Significant MIF of the even 200Hg isotope has also been recently observed in precipitations.23 Aquatic photoreduction of organic matter bound Hg(II) and photodegradation of MMHg(II) are two reactions known to induce MIF by MIE.13,24 In previous studies we have examined the hair Hg isotope signatures of indigenous people feeding on the same fish resource.17 MMHg absorption by the human organism was shown to induce MDF of +2% for δ202Hg and insignificant MIF. We suggested that paired MDF and MIF signatures in human hair may provide insight into the dominant MMHg exposure pathway to humans. Using a novel coupled chromatographymass spectrometry technique Epov et al. have also shown that IHg and MMHg in human hair potentially carry different MDF and MIF signatures.25 In this study, we compared Hg stable isotope signatures in hair from four different populations, three living in the Bolivian Amazon basin and one living in France: 1. gold miners from the eastern Bolivian cordillera that are exposed to elemental Hg(0) and consume little to no fish; 2. gold miners from the Bolivian piedmont that are exposed to elemental Hg(0) as well as fresh water and imported marine fish MMHg; 3. indigenous people from the Bolivian floodplain that are quite exclusively exposed to local fresh water fish MMHg; and 4. Europeans who are mainly exposed to marine fish MMHg. Our main objective is to investigate whether different Hg exposure pathways and/or sources can be resolved by Hg stable isotope and Hg speciation analysis of human hair.
’ EXPERIMENTAL SECTION Study Area and Sampling. The study area is located in the Bolivian Amazon along the Beni River basin (South America; Figure SI-S1). More information about the study area is given in Supporting Information. Hair of gold miners were sampled at four mines: one native gold mine at Yani (Figure SI-S1, 15350 S 68340 W, 3695 m altitude) and four alluvial gold mines along the Tipuani, Mapiri, and Kaka rivers: Tomachi (15270 S 67470 W, 533 m alt.), San Juanito (15280 S 67480 W, 661 m alt.), Suraqui (15270 S 67510 W, 661 m alt.), and Huacahuilo (15300 S 67530 W, 431 m alt.). At the Yani site, due to the cold climate, miners burn the HgAu amalgam indoors at home exposing
ARTICLE
themselves and their families to Hg(0) vapor. Alluvial gold miners at lower altitudes typically burn HgAu amalgam outside protecting themselves from the Hg vapor transported by the wind. Hair samples from a European control group were taken in a typical European urban area (Toulouse, France). Despite the potential for external Hg contamination of hair, no washing method was applied as there exists no consensus on its effectiveness.26,27 Individuals are weakly exposed to IHg exposure mainly from dental amalgams and urban conditions. Three liquid Hg(0) samples (B06, G01, G02) used by gold miners in the Beni basin were purchased and collected in a gold shop in Guanay. Analytical Methods. Total Hg in hair (THg) was determined by a combustion-atomic absorption method (Milestone, DMA80 analyzer). MMHg and IHg concentrations were measured by GC-ICPMS with the species-specific double isotope dilution method at LCABIE-IPREM (Pau, France). Hg stable isotopic compositions were determined by cold vapor-MC-ICPMS (Thermo-Finnigan Neptune) at the Observatoire Midi-Pyrenees (Toulouse, France). Details on analytical procedures and quality assurance are given in the Supporting Information and elsewhere.17,28 Isotopic compositions are reported in the δ notation relative to SRM NIST 3133 (δ202Hg = [202/198Hgsample/202/198HgNIST3133 1]103 %). MIF is expressed using Δ199Hg and Δ201Hg notation.29
’ RESULTS THg in gold miner and European subject’s hair ranges from 167 ng.g1 to 2349 ng.g1 and from 261 ng.g1 and 1890 ng.g1 respectively. MMHg concentrations of native and alluvial gold miner hair range from 117 ng.g1 to 458 ng.g1 and from 73 ng.g1 to 997 ng.g1 respectively corresponding to a MMHg fraction ranging from 9% to 40% for native gold miner hair and 4% to 86% for alluvial gold miner hair (Table SI-S1). All native gold miner hair samples present negative δ202Hg which range from 0.87 to 0.15%, while alluvial gold miner hair display a large variation of δ202Hg from 0.43 to +2.55%. Hg MIF is observed in the miner populations with anomalies for both odd isotopes 199Hg and 201Hg ranging from +0.11 to +1.18% for Δ199Hg and from +0.19 to +0.99% for Δ201Hg. Hair from the European control group shows positive δ202Hg ranging from +1.81 to +3.23% and positive anomalies Δ199Hg ranging from +0.85 to +1.40% and Δ201 Hg from +0.67 to +1.33%. Liquid Hg(0) used by Bolivian gold miners has an average δ202Hg of 0.28 ( 0.29% (1SD, n = 3), and its anomalies are not significant relative to external reproducibility 0.01 ( 0.04% for Δ199Hg and 0.02 ( 0.06% for Δ201Hg (1SD, n = 3) (Table SI-S1). Average Hg isotope signatures for liquid Hg(0) based on published values and this study will be used in the following: δ202Hg of 0.39 ( 0.37% and Δ201Hg = 0.01 ( 0.04% (1SD, n = 7, see Table SI-S2).13,30,31 ’ DISCUSSION The potential of Hg stable isotopes as a source tracer in human exposure studies depends on whether a given source isotope signature is modified or not during the multiple physicochemical transformations that link a source to a receptor such as hair. Hg has the advantage that it presents at least three different isotope signatures, δ202Hg (MDF) and Δ199Hg and Δ201Hg (MIF) that can potentially act as tracers and that carry different information on Hg sources and speciation processes. In the following, we 9911
dx.doi.org/10.1021/es202353m |Environ. Sci. Technol. 2011, 45, 9910–9916
Environmental Science & Technology
Figure 1. MIF anomaly Δ201Hg (%) plotted as a function of δ202Hg (%) for European hair samples collected at Toulouse (France, Europe), sea bird eggs from the Gulf of Alaska and Southern Bering Strait,35 marine fish from the Gulf of Mexico34 and the San Francisco Bay,36 and certified reference materials BCR-464 tuna fish from the Mediterranean Sea37 and NRC DORM-2 dogfish.13 Error bars represent external reproducibility (2SD).
discuss two limiting cases of dietary Hg exposure to i) exclusive fresh water fish MMHg and ii) dominant marine fish MMHg, before we arrive at the more complex case of artisanal gold miners who are exposed to both IHg and MMHg. We note that contrary to MMHg exposure, hair is regarded as an inappropriate matrix for determining inorganic Hg exposure.32 This, however, is irrelevant in this study as the objective is not to quantify inorganic Hg exposure but to test the inorganic Hg source hair receptor relationship based on Hg isotope signatures. Limiting Cases of MMHg Exposure. In a previous study, by comparing fresh water fish δ202Hg and human hair δ202Hg in a fish eating Amazonian community, we found that MDF associated with human metabolism enriches MMHg in hair δ202Hg by +2% relative to the fish MMHg consumed.17 Contrarily, we did not find evidence for metabolic MIF as fish Δ201Hg was insignificantly different from hair Δ201Hg. In this study, we observed similar relationships between published data on marine fish and hair from the European control group (Figure 1). Since the MMHg concentration is higher in fish tissue than other food sources, we assume that MMHg in the control subject’s hair is introduced mainly via marine fish consumption.33 A Student-ttest shows that average control group hair Δ201Hg is not significantly different (P = 0.33) from the average Δ201Hg of coastal and open ocean biota, based on studies in the Gulf of Mexico,34 the Gulf of Alaska and Southern Bering Strait (copepod and fish consuming marine seabirds),35 the San Francisco Bay,36 certified reference materials BCR-464 tuna fish from the Mediterranean Sea,37 and NRC DORM-2 dogfish.13 A MDF of +2.2 ( 0.8% for δ202Hg is observed between average marine fish and hair from the European control group. These observations are consistent with our previous study: human metabolism enriches hair MMHg in the heavier isotopes by ∼+2% relative to the MMHg food source. In both the fresh water fish and marine fish exposure cases, the δ202Hg signature of the MMHg source is modified and information on the source is partially lost. However, the ∼2% MDF effect appears robust, and a correction can be applied to human hair δ202Hg to estimate
ARTICLE
Figure 2. MIF anomaly Δ201Hg (%) plotted as a function of δ202Hg (%) hair samples. Alluvial gold miner hair is distributed between three source end-members: marine fish MMHg represented by European hair, amazon freshwater fish MMHg represented by indigenous people hair and IHg of gold miner hair with >80% IHg. Error bars represent external reproducibility (2SD).
the average dietary MMHg δ202Hg signature (Figure 1). While a large variation in marine fish Δ201Hg exists, the similarity in average Δ201Hg between marine fish and bird eggs (Δ201Hg =0.83 ( 0.36%, 1SD, n = 112) and the French control (Δ201Hg =0.92 ( 0.20%, 1SD, n = 11) group suggests the absence of metabolic MIF. In the next section additional information is obtained by looking at the relationship between Δ199Hg and Δ201Hg. Relationships between Δ199Hg and Δ201Hg. The relationship between the two MIF signatures Δ199Hg and Δ201Hg is useful to identify the underlying MIF mechanisms, i.e. MIE or NVE. Experimental photochemical Hg MIF, presumably by the MIE, yields Δ199/201Hg relationships between 1.00 and 1.36.13,24 Dark abiotic reduction, equilibrium HgII-thiol complexation, and liquid Hg evaporation provide Δ199/201Hg estimates for the NVE ranging from 1.54 to 2.030,38,39 with a best estimate Δ199/201Hg of 1.61 ( 0.06.38 Linear regression slopes, Δ199/201Hg, for hair samples are summarized in Figure SI-S2. All slopes range between 1.00 and 1.36 indicating that Δ199Hg and Δ201Hg are most likely induced by the MIE. The Δ199/201Hg slope for the French control group is 1.19 ( 0.13 (2SD) and is not significantly different from the slope of 1.19 ( 0.04 (2SD) of marine fish and bird eggs (ANCOVA; P = 0.93).13,3437 Experiments and theoretical considerations suggest that the MIE reflects a photochemical process and not a biological one.40,41 The compatibility between hair Δ199/201Hg slopes and photochemical MIE therefore suggest that the hair MIF signatures were acquired before the MMHg was incorporated in fresh water and marine fish. This corroborates the conservative nature of the Δ199Hg and Δ201Hg parameters in the simple fish/hair source/receptor relationship of the two limiting case studies. Hg Exposure Sources of Bolivian Gold Miners. Gold miner THg is strongly correlated with IHg concentration in hair (R = 0.94, P < 0.0001, n = 23) as also observed by Li et al.9 The relatively limited variation of δ202 Hg and Δ 201 Hg in the European control and Bolivian indigenous groups likely reflects exposure to one main source of MMHg, e.g. fish. Contrarily, the much larger variation in δ202Hg and Δ201Hg in gold miner’s hair may suggest multiple Hg exposure sources (Figure 2 and 9912
dx.doi.org/10.1021/es202353m |Environ. Sci. Technol. 2011, 45, 9910–9916
Environmental Science & Technology
Figure 3. δ202Hg (%) plotted as a function of IHg fraction (%) in gold miner hair samples. A significant linear relationship is observed with R = 0.81. Error bars represent external reproducibility (2SD) for δ202Hg and errors calculated by uncertainties propagation of IHg and MMHg concentrations (2SD) for IHg fraction.
Table SI-S1). Based on interviews with gold miner subjects it appears that MMHg exposure takes place through consumption of fresh water fish from Bolivian rivers and canned marine fish. IHg exposure results from liquid Hg(0) used for gold amalgamation. Significant relationships between δ202Hg and IHg percentage (linear regression R = 0.81 and P < 0.0001, n = 23; Figure 3) and between δ202Hg and IHg levels (linear regression R = 0.67 and P = 0.0004, n = 23) are observed. Contrarily, Δ201Hg does not vary with IHg percentage nor with IHg concentrations in gold miner hair. Understanding these trends requires looking at the miner communities. The Yani Group Andean Cordillera, 3700 m Altitude. Among the four gold miner populations we studied, the Yani miners that work at high elevation in the Bolivian Andes consume little to no fish at all as small streams do not contain edible fish species at these altitudes. The Yani group therefore represent a third limiting case, namely that of near-exclusive exposure to IHg. This is reflected in their hair Hg speciation which is dominated by IHg at 79 ( 13% (1SD, n = 5). The Hg isotopic compositions of Yani miners show negative δ202Hg averaging 0.43 ( 0.28% (1SD, n = 5, Figure 2). The negative δ202Hg corresponds well with the δ202Hg of locally used liquid Hg(0), averaging 0.28 ( 0.29% (1SD, n = 3), but also with the average δ202Hg of liquid Hg(0) from diverse sources (0.39 ( 0.37%, 1SD, n = 7, Table SI-S2). It appears therefore that the δ202Hg (MDF) signature of the inorganic liquid Hg exposure source is incorporated in Yani subject’s hair without net modification of the isotopic composition by MDF. This is in contrast with dietary exposure to fish MMHg, which was shown to be accompanied by ∼2% MDF. Yani gold miner subjects also display significant hair Hg MIF with Δ201Hg from +0.22 to +0.99%. This is somewhat puzzling because neither local nor published liquid Hg(0) sources have significant Δ201Hg. Therefore a MIF inducing process might be active somewhere in between the moment of liquid Hg(0) use and the incorporation of IHg in the hair matrix. Three potential explanations will be discussed: i) MIF and MDF during the liquidvapor transition of amalgam Hg(0), ii) the presence of small amounts of MMHg in Yani hair with high Δ201Hg, and iii) postemission IHg MIF in the local atmospheric environment or at the hair surface. Liquid Hg(0) evaporation into a vacuum has been shown to enrich the vapor in the lighter isotopes, with MDF factor
ARTICLE
103lnαvap‑liq of 6.65% at 22 C and 0.79% at 100 C for δ202Hg30 and an extrapolated zero fractionation factor at temperatures higher than 115 C. Gold miners use a blowtorch with a flame temperature of at least 1000 C to evaporate Hg(0) from the amalgam which likely suppresses MDF entirely. Estrade et al. have also noticed weak Δ199Hg and Δ201Hg anomalies, between +0.04 and +0.08%, independent of evaporation temperature.30 However the amalgam-burning step is a physical state transformation that goes to completion and therefore overall should not induce MDF or MIF of the vapor. Postburning Hg(0) vapor condensation and oxidation at ambient temperatures are more likely to induce isotopic fractionation in the exposure environment. Yani gold miners burn HgAu amalgam inside their living quarters thereby generating near-saturated Hg(0) vapor pressures. Extrapolating Estrade et al.’s Hg(0) evaporation observations to a Hg(0) condensation scenario with an equilibrium or kinetic MIF factor, 103lnαliq‑vap, Δ201Hg of at most 0.1%30 can theoretically produce a residual Hg(0) vapor with extreme isotopic composition, i.e. Δ201Hg of +1% at 90% completion of the condensation reaction. Gold miners may therefore be exposed to residual Hg(0) vapors with a range of positive Δ201Hg. A control experiment was conducted in a Guanay (Bolivia) gold shop where amalgam is frequently burned in the courtyard. A 0.5 mL liquid Hg(0) amalgam was burned with a blowtorch, and the produced Hg(0) vapor was sampled with gold traps. Soil and vegetation samples from the courtyard were also sampled. Isotopic composition was determined according to standard procedures.17,31 The soil and vegetation samples (THg of 55 and 20 μg.g1) as well as the Hg from the gold traps did not present significant MIF (Table SI-S1). This suggests that IHg(0) exposure from HgAu amalgam burning is unlikely to have a pronounced MIF signature. An average of 21% of THg in Yani miner’s hair is in the form of MMHg. If this MMHg originates from a fish source, then it potentially carries substantially positive MIF and MDF signatures. For example, a weight fraction of 30% MMHg with Δ201Hg of +3.3%, and 70% IHg with Δ201Hg of 0% will result in our observed maximum THg Δ201Hg of +1% for Yani individual FY3. Across all Yani miners, a potential fish MMHg Δ201Hg signature would have to range from +0.6 to +7.6%. These signatures are unrealistic compared to average local Amazonian fish Δ201Hg of +0.15% and average marine Δ201Hg of +0.83%. In addition, the presence of fish MMHg in hair should also influence the THg δ202Hg (MDF) signature. Marine fish MMHg has an average δ202Hg of +0.26 ( 0.49% (1SD, n = 112), which if absorbed by the human body will be fractionated by +2% and then diluted with IHg in hair. For the above scenario Yani miners should have developed δ202Hg up to +0.5% which is not the case. However, it has been documented that in vivo demethylation can take place in humans in the brain and liver.42 Based on studies of bacterial and photochemical demethylation we would expect in vivo demethylation of MMHg to produce IHg that is enriched in the lighter isotopes by MDF.13,40 In the absence of sunlight one would not expect photochemical MIF to accompany in vivo demethylation. This process would then leave the MMHg Δ201Hg unchanged from MMHg reactant to demethylation IHg product. While a fish MMHg related origin for the observed Yani miner Δ201Hg signatures is unlikely, we cannot rule out the influence of complex metabolic reactions such as demethylation on hair THg MIF anomalies. Novel speciesspecific Hg stable isotope fingerprinting by GC-MC-ICPMS has been able to detect species-specific IHg and MMHg MDF 9913
dx.doi.org/10.1021/es202353m |Environ. Sci. Technol. 2011, 45, 9910–9916
Environmental Science & Technology
ARTICLE
and MIF signatures in fish and human hair and may be useful to resolve the issue.25 Finally, MIF inducing reactions that occurred in between the moments of amalgam burning Hg(0) emission and human IHg exposure may have generated the Yani miner Δ201Hg signatures. It is not known whether the IHg in Yani miner hair reflects exposure to Hg(0) vapor or ingested forms of Hg(II). It is also not known whether the IHg species in hair was incorporated through metabolism or is present as exogenous IHg contamination by hair surface adsorption and absorption. No attempts were made to desorb potential exogenous IHg as literature studies and washing tests with S-bearing ligands in this study were inconclusive.26,27,32 Near-field Hg(II) deposition measured in precipitation in North American industrialized regions has been recently shown to have positive Δ201Hg signatures up to 0.51%.23 Gratz et al. (2010) and Sonke (2011) have suggested a potential role for in-cloud photochemical processes to generate the positive precipitation Δ201Hg.23,43 It is likely that amalgam burning Hg(0) emissions are partly oxidized in the local atmospheric environment, leading to elevated Hg(II) levels in precipitation that may or may not carry positive Δ201Hg. Similarly, IHg adsorbed on external hair surfaces may be subject to uncharacterized photochemical MIF processes. More work is needed to explore the origin of the Yani miner hair Δ201Hg signatures. In the following, we consider the Yani group as a third limiting case, i.e. that of dominant IHg exposure. Piedmont Alluvial Gold Miners 431661 m Altitude. The hair Hg isotope signatures of gold miners from the Andean piedmont show a large variation of δ202Hg and Δ201Hg (Figure 2). We have correlated these variations with [Hg]T, age of people, way of using Hg, and fish consumption frequency, and no significant relationship was observed. In a MDF vs MIF diagram (Figure 2) piedmont gold miner Hg isotope signatures appear to be bound by three isotopic end members that represent the three limiting exposure cases that were discussed above: 1. marine fish MMHg exposure, represented by the French control group, 2. local Bolivian freshwater fish MMHg exposure reflected by Bolivian indigenous people,17 and 3. dominant IHg exposure as reflected by the Yani gold miners. We therefore suggest that hair THg isotope signatures of piedmont gold miners represent a mixture of IHg exposure from liquid Hg handling and MMHg absorbed via consumption of marine fish and/or Amazonian freshwater fish. The approximate contribution of each source to THg in the piedmont miner’s hair can be estimated by solving the isotope mass balance eqs 1, 2, and 3 δ202 HgT ¼ MMHgM %δ202 HgMMHgM þ MMHg%B δ202 HgMMHgB þ InHg%δ202 HgInHg
ð1Þ
Δ201 HgT ¼ MMHgM %Δ201 HgMMHgM þ MMHg%B Δ201 HgMMHgB þ InHg%Δ201 HgInHg
MMHgM % þ MMHgB % þ InHg% ¼ 100
ð2Þ ð3Þ
where δ202HgT and Δ201HgT are the MDF and MIF isotope signatures of hair THg, δ202HgMMHgM and Δ201HgMMHgM are isotopic signatures of MMHg after ingestion of Hg from marine fish, δ202HgMMHgB and Δ201HgMMHgB are isotopic signatures of MMHg after ingestion of Hg from Bolivian fresh water fish, and δ202HgInHg and Δ201HgInHg are isotopic signatures of IHg
Figure 4. Model estimated concentration (ng.g1) of three predominant Hg species in human hair across six different populations. One SD uncertainties based on the number (n) of individuals are included. The three lefthand side populations, e.g. diverse miners with >80% IHg, Villa Copacabana, and Toulouse, represent the limiting cases of dominant exposure to IHg, Amazon fish MMHg, and marine fish MMHg. Estimated origins and levels of hair Hg species are shown for three gold miner populations.
exposure. Average European δ202HgMMHgM and Δ201HgMMHgM of hair are 2.32 ( 0.43% and +0.92 ( 0.20% (1SD, n = 11) respectively, and average indigenous people δ202HgMMHgB and δ201HgMMHgB of hair are 1.18 ( 0.12% and +0.13 ( 0.04% (1SD, n = 7, data for Villa Copacabana17). They define the two MMHg end members. The isotopic signature of IHg exposure is approximated here by the average isotopic composition of five gold miners with >80% (83 to 96%) IHg fraction and yields a δ202Hg of 0.43 ( 0.28% and Δ201Hg = +0.53 ( 0.32% (1SD, n = 5). The mass balance results are shown in Table SI-S3. It is of interest to compare the model-estimated fractions of IHg and MMHg to the independently measured IHg and MMHg speciation (Figure SI-S3). Significant positive correlations are obtained between the measured and modeled IHg and MMHg fractions (R = 0.9, P < 0.0001 and R = 0.6, P = 0.002 respectively). Modeled speciation fractions are at times 1 as several samples do not lie within the ternary mixing domain of the supposed three Hg exposition sources (Figure 2, Table SI-S3). These outliers were ignored in the regression analysis. Given the complexity of both Hg speciation and stable isotopic analysis, the obtained results are encouraging. Full source appointment by the model is summarized in Figure 4 for the populations under study. Notable differences in local Amazonian fish MMHg exposure and IHg exposure are apparent. Gold miner camps at remote locations and near rivers (Tomachi and San Juanito) show higher exposure to local fish MMHg. A higher organization level of gold exploration activities, i.e. use of heavy equipment (Yani, Tomachi and Suraqui), appears to increase IHg exposure when compared to artisanal mining activities (San Juanito, Huacahuilo). Possibly the quantity of liquid Hg used and the frequency of amalgam burning explains this trend. Our Bolivian case study shows that Hg isotope signatures of human hair provide unique information on the sources of Hg exposure to humans, provided that the different sources are isotopically distinguishable. In the case of MMHg exposure through fish diet the ensemble of MDF and MIF isotope signatures is able to identify fresh water and marine fish MMHg exposure. Isotopic analysis is therefore a complementary tool to speciation analysis for tracing sources of organic Hg species, more specifically. While human metabolism shifts dietary MMHg MDF signatures by approximately +2%, this feature appears to 9914
dx.doi.org/10.1021/es202353m |Environ. Sci. Technol. 2011, 45, 9910–9916
Environmental Science & Technology be robust across two different populations and can be successfully corrected for. IHg MDF signatures do not appear to be fractionated between the exposure source and incorporation in human hair. However, similar to IHg speciation analysis, IHg isotope signature analysis is hampered by our inability to discern between exogenous and endogenous IHg in hair. A detailed study of bodily variation in Hg MDF may help identify where MMHg MDF takes place and why IHg MDF appears absent. Hair MIF signatures associated with liquid Hg(0) exposure of gold miners also appear to be unique yet are less well understood. Isotopic deconvolution of Hg exposure sources agrees fairly well with independent hair Hg speciation analysis. In this exploratory study Hg isotope signatures then allow detection of broad exposure trends for those populations where enough subjects were included. Future work should profit from existing cohort studies to refine Hg isotope signatures as exposure source tracers. We anticipate that much of the hair Hg isotope signature variation within populations results from variation in Hg source signatures, such as shown for the marine environment (Figure 1). Continued cataloging of Hg exposure sources should therefore accompany bioindicator studies.
’ ASSOCIATED CONTENT
bS
Supporting Information. Text, Figures SI-S1SI-S3, and Tables SI-S1SI-S3. This material is available free of charge via the Internet at http://pubs.acs.org.
’ AUTHOR INFORMATION Corresponding Author
*E-mail: laff
[email protected] (L.L.),
[email protected] (J.E.S.).
’ ACKNOWLEDGMENT We would like to thank the INSU-CNRS EC2CO Program (RIMES), the IRD, the Midi-Pyrenees and Aquitaine regions, and ANR-09-JCJC-0035-01 grant for funding and cooperation agreements with Bolivian institutions. L.L. acknowledges the Ministere de l’Enseignement Superieur et de la Recherche in France for her Ph.D. grant. C. Boucayrand, M. Carayon, and C. Causserand at the GET are thanked for helping during sample preparation and experimentation, R. Freydier for assistance on the MC-ICPMS, E. Tessier for assistance on the GC-ICPMS, B. Guerrero for drawing Figure SI-S1, and D. Point and M. Monperrus for fruitful discussions. Three anonymous reviewers helped improve the quality of this contribution. ’ REFERENCES (1) Ullrich, S. M.; Tanton, T. W.; Abdrashitova, S. A. Mercury in the aquatic environment: A review of factors affecting methylation. Crit. Rev. Environ. Sci. Technol. 2001, 31 (3), 241–293. (2) Clarkson, T. W.; Vyas, J. B.; Ballatorl, N. Mechanisms of mercury disposition in the body. Am. J. Ind. Med. 2007, 50 (10), 757–764. (3) Morel, F. M. M.; Kraepiel, A. M. L.; Amyot, M. The chemical cycle and bioaccumulation of mercury. Annu. Rev. Ecol. Syst. 1998, 29, 543–566. (4) IPCS, Methylmercury. In Environmental Health Criteria 101; World Health Organization: Geneva, 1990. (5) IPCS,Inorganic Mercury. In Environmental Health Criteria 118; World Health Organization: Geneva, 1991.
ARTICLE
(6) Mergler, D.; Anderson, H. A.; Chan, L. H. M.; Mahaffey, K. R.; Murray, M.; Sakamoto, M.; Stern, A. H. Methylmercury exposure and health effects in humans: A worldwide concern. Ambio 2007, 36 (1), 3–11. (7) Drasch, G.; Wanghofer, E.; Roider, G. Are blood, urine, hair, and muscle valid biomonitors for the internal burden of men with the heavy metals mercury, lead and cadmium? - An investigation on 150 deceased. Trace Elem. Electrolytes 1997, 14 (3), 116–123. (8) Wilhelm, M.; Muller, F.; Idel, H. Biological monitoring of mercury vapour exposure by scalp hair analysis in comparison to blood and urine. Toxicol. Lett. 1996, 88 (13), 221–226. (9) Li, P.; Feng, X. B.; Qiu, G. L.; Shang, L. H.; Li, G. H. Human hair mercury levels in the Wanshan mercury mining area, Guizhou Province, China. Environ. Geochem. Health 2009, 31 (6), 683–691. (10) Sunderland, E. M.; Gobas, F.; Heyes, A.; Branfireun, B. A.; Bayer, A. K.; Cranston, R. E.; Parsons, M. B. Speciation and bioavailability of mercury in well-mixed estuarine sediments. Mar. Chem. 2004, 90 (14), 91–105. (11) Feng, X.; Li, P.; Qiu, G.; Wang, S.; Li, G.; Shang, L.; Meng, B.; Jiang, H.; Bai, W.; Li, Z.; Fu, X. Human Exposure To Methylmercury through Rice Intake in Mercury Mining Areas, Guizhou Province, China. Environ. Sci. Technol. 2007, 42 (1), 326–332. (12) Lee, R.; Middleton, D.; Caldwell, K.; Dearwent, S.; Jones, S.; Lewis, B.; Monteilh, C.; Mortensen, M. E.; Nickle, R.; Orloff, K.; Reger, M.; Risher, J.; Rogers, H. S.; Watters, M. A review of events that expose children to elemental mercury in the United States. Ciencia Saude Coletiva 2010, 15 (2), 585–598. (13) Bergquist, B. A.; Blum, J. D. Mass-dependent and -independent fractionation of Hg isotopes by photoreduction in aquatic systems. Science 2007, 318 (5849), 417–420. (14) Carignan, J.; Estrade, N.; Sonke, J. E.; Donard, O. F. X. Odd Isotope Deficits in Atmospheric Hg Measured in Lichens. Environ. Sci. Technol. 2009, 43 (15), 5660–5664. (15) Das, R.; Salters, V. J. M.; Odom, A. L. A case for in vivo massindependent fractionation of mercury isotopes in fish. Geochem. Geophys. Geosyst. 2009, 10, Q11012. (16) Jackson, T. A.; Whittle, D. M.; Evans, M. S.; Muir, D. C. G. Evidence for mass-independent and mass-dependent fractionation of the stable isotopes of mercury by natural processes in aquatic ecosystems. Appl. Geochem. 2008, 23 (3), 547–571. (17) Laffont, L.; Sonke, J. E.; Maurice, L.; Hintelmann, H.; Pouilly, M.; Sanchez Bacarreza, Y.; Perez, T.; Behra, P. Anomalous Mercury Isotopic Compositions of Fish and Human Hair in the Bolivian Amazon. Environ. Sci. Technol. 2009, 43 (23), 8985–8990. (18) Estrade, N.; Carignan, J.; Donard, O. F. X. Isotope Tracing of Atmospheric Mercury Sources in an Urban Area of Northeastern France. Environ. Sci. Technol. 2010, 44 (16), 6062–6067. (19) Estrade, N.; Carignan, J.; Donard, O. F. X. Tracing and Quantifying Anthropogenic Mercury Sources in Soils of Northern France Using Isotopic Signatures. Environ. Sci. Technol. 2011, 45 (4), 1235–1242. (20) Bigeleisen, J. Nuclear Size and Shape Effects in Chemical Reactions. Isotope Chemistry of the Heavy Elements. J. Am. Chem. Soc. 1996, 118 (15), 3676–3680. (21) Schauble, E. A. Role of nuclear volume in driving equilibrium stable isotope fractionation of mercury, thallium, and other very heavy elements. Geochim. Cosmochim. Acta 2007, 71 (9), 2170–2189. (22) Buchachenko, A. L.; Ivanov, V. L.; Roznyatovskii, V. A.; Artamkina, G. A.; Vorob’ev, A. K.; Ustynyuk, Y. A. Magnetic isotope effect for mercury nuclei in photolysis of bis(p-trifluoromethylbenzyl)mercury. Dokl. Phys. Chem. 2007, 413, 39–41. (23) Gratz, L. E.; Keeler, G. J.; Blum, J. D.; Sherman, L. S. Isotopic Composition and Fractionation of Mercury in Great Lakes Precipitation and Ambient Air. Environ. Sci. Technol. 2010, 44 (20), 7764– 7770. (24) Zheng, W.; Hintelmann, H. Mercury isotope fractionation during photoreduction in natural water is controlled by its Hg/DOC ratio. Geochim. Cosmochim. Acta 2009, 73 (22), 6704–6715. 9915
dx.doi.org/10.1021/es202353m |Environ. Sci. Technol. 2011, 45, 9910–9916
Environmental Science & Technology
ARTICLE
(25) Epov, V. N.; Berail, S.; Jimenez-Moreno, M.; Perrot, V.; Pecheyran, C.; Amouroux, D.; Donard, O. F. X. Approach to Measure Isotopic Ratios in Species Using Multicollector-ICPMS Coupled with Chromatography. Anal. Chem. 2010, 82 (13), 5652–5662. (26) Morton, J.; Carolan, V. A.; Gardiner, P. H. E. Removal of exogenously bound elements from human hair by various washing procedures and determination by inductively coupled plasma mass spectrometry. Anal. Chim. Acta 2002, 455 (1), 23–34. (27) Li, Y.-F.; Chen, C.; Li, B.; Wang, J.; Gao, Y.; Zhao, Y.; Chai, Z. Scalp hair as a biomarker in environmental and occupational mercury exposed populations: Suitable or not? Environ. Res. 2008, 107, 39–44. (28) Monperrus, M.; Gonzalez, P. R.; Amouroux, D.; Alonso, J. I. G.; Donard, O. F. X. Evaluating the potential and limitations of doublespiking species-specific isotope dilution analysis for the accurate quantification of mercury species in different environmental matrices. Anal. Bioanal. Chem. 2008, 390 (2), 655–666. (29) Blum, J. D.; Bergquist, B. A. Reporting of variations in the natural isotopic composition of mercury. Anal. Bioanal. Chem. 2007, 388 (2), 353–359. (30) Estrade, N.; Carignan, J.; Sonke, J. E.; Donard, O. F. X. Mercury isotope fractionation during liquid-vapor evaporation experiments. Geochim. Cosmochim. Acta 2009, 73 (10), 2693–2711. (31) Sonke, J. E.; Zambardi, T.; Toutain, J. P. Indirect gold trap-MCICP-MS coupling for Hg stable isotope analysis using a syringe injection interface. J. Anal. At. Spectrom. 2008, 23 (4), 569–573. (32) Kempson, I. M.; Lombi, E. Hair analysis as a biomonitor for toxicology, disease and health status. Chem. Soc. Rev. 2011, 40 (7), 3915–3940. (33) Larsen, E. H.; Andersen, N. L.; Moller, A.; Petersen, A.; Mortensen, G. K.; Petersen, J. Monitoring the content and intake of trace elements from food in Denmark. Food Addit. Contam. 2002, 19 (1), 33–46. (34) Senn, D. B.; Chesney, E. J.; Blum, J. D.; Bank, M. S.; Maage, A.; Shine, J. P. Stable Isotope (N, C, Hg) Study of Methylmercury Sources and Trophic Transfer in the Northern Gulf of Mexico. Environ. Sci. Technol. 2010, 44 (5), 1630–1637. (35) Point, D.; Sonke, J. E.; Day, R. D.; Rosenau, D. G.; Hobson, K. A.; Vanderpol, S. S.; Moors, A. J.; Pugh, R. S.; Donard, O. F. X.; Becker, P. B. Methylmercury photodegradation influenced by sea ice cover in Arctic marine ecosystems. Nat. Geosci. 2010. (36) Gehrke, G. E.; Blum, J. D.; Slotton, D. G.; Greenfield, B. K. Mercury isotopes link mercury in San Francisco Bay forage fish to surface sediments. Environ. Sci. Technol. 2011. (37) Perrot, V.; Epov, V. N.; Pastukhov, M. V.; Grebenshchikova, V. I.; Zouiten, C.; Sonke, J. E.; Husted, S. r.; Donard, O. F. X.; Amouroux, D. Tracing Sources and Bioaccumulation of Mercury in Fish of Lake Baikal Angara River Using Hg Isotopic Composition. Environ. Sci. Technol. 2010, 44 (21), 8030–8037. (38) Zheng, W.; Hintelmann, H. Nuclear Field Shift Effect in Isotope Fractionation of Mercury during Abiotic Reduction in the Absence of Light. J. Phys. Chem. A 2010, 114 (12), 4238–4245. (39) Wiederhold, J. G.; Cramer, C. J.; Daniel, K.; Infante, I.; Bourdon, B.; Kretzschmar, R. Equilibrium Mercury Isotope Fractionation between Dissolved Hg(II) Species and Thiol-Bound Hg. Environ. Sci. Technol. 2010, 44 (11), 4191–4197. (40) Kritee, K.; Barkay, T.; Blum, J. D. Mass dependent stable isotope fractionation of mercury during mer mediated microbial degradation of monomethylmercury. Geochim. Cosmochim. Acta 2009, 73 (5), 1285–1296. (41) Rodriguez-Gonzalez, P.; Epov, V. N.; Bridou, R.; Tessier, E.; Guyoneaud, R.; Monperrus, M.; Amouroux, D. Species-Specific Stable Isotope Fractionation of Mercury during Hg(II) Methylation by an Anaerobic Bacteria (Desulfobulbus propionicus) under Dark Conditions. Environ. Sci. Technol. 2009, 43 (24), 9183–9188. (42) Clarkson, T. W. The three modern faces of mercury. Environ. Health Perspect. 2002, 110, 11–23. (43) Sonke, J. E. A global model of mass independent mercury stable isotope fractionation. Geochim. Cosmochim. Acta 2011, 75 (16), 4577–4590. 9916
dx.doi.org/10.1021/es202353m |Environ. Sci. Technol. 2011, 45, 9910–9916
