Distribution of mercury species and mercury isotope ratios in soils and river suspended matter of a mercury mining area

Carluvy Baptista-Salazar *a, Holger Hintelmann b and Harald Biester a
aInstitut für Geoökologie, Abt. Umweltgeochemie, Technische Universität Braunschweig, Langer Kamp 19c, Braunschweig, 38106, Germany. E-mail: c.baptista-salazar@tu-braunschweig.de
bChemistry Department, Trent University, 1600 West Bank Drive, Peterborough, K9L0G2, Ontario, Canada

Received 22nd September 2017 , Accepted 12th January 2018

First published on 1st February 2018


Mercury (Hg) released by mining activities can be dispersed in the environment, where it is subject to species transformations. Hg isotope ratios have been used to track sources in Hg contaminated areas, although it is unclear to what extent variations in δ-values are attributed to distinct Hg species. Hg was mined as Hg sulphide (cinnabar) in Idrija, Slovenia for centuries. Sediments are loaded with mining-residues (cinnabar and calcine), whereas contaminated soils mainly contain Hg bound to natural organic matter (NOM-Hg) related to atmospheric Hg deposition. Hg released from soils and sediments is transported as suspended matter (SM) in the Idrijca river to the Gulf of Trieste (GT), Italy. We determine Hg isotope ratios in river SM, sediments and soils from the Idrijca-catchment to decipher the Hg isotope ratio variability related to Hg species distribution in different grain-size fractions. δ202Hg values of SM collected from tributaries corresponded to those found in soils ranging from −2.58 to 0.19‰ and from −2.27 to −0.88‰, respectively. Speciation measurements reveal that fine fractions (0.45–20 μm) are dominated by NOM-Hg, while larger fractions contain more cinnabar. More negative δ202Hg values were related to higher proportions of NOM-Hg, which are predominant in soils and SM. Rain events increase SM-loads in the river, mainly due to resuspension of coarse grain-size fractions of bottom sediments bearing larger proportions of cinnabar, which leads to more positive δ202Hg values. The large magnitude of variation in δ202Hg and the smaller magnitude of variation in Δ199Hg (−0.37 to 0.09‰) are likely related to fractionation during ore roasting. Soil samples with high NOM-Hg content show more negative δ202Hg values and larger variation of Δ199Hg. More negative δ202Hg values in GT sediments were rather linked to distant sedimentation of soil derived NOM-Hg than to sedimentation of autochthonous marine material. Heterogeneity in the Idrija ore and ore processing likely produce large variations in the Hg isotopic composition of cinnabar and released metallic Hg, which complicate the differentiation of Hg sources. Combining Hg isotope measurements with solid phase Hg speciation reveals that Hg isotope ratios rather indicate different Hg species and are not necessarily symptomatic for Hg pollution sources.



Environmental significance

Tracing dispersion and speciation of Hg in contaminated areas is crucial to evaluate environmental risk potential. We show a novel combination of solid phase Hg speciation and Hg isotope analysis to track Hg released from a Hg mining area. Our results show that Hg isotope ratios in contaminated soils and sediments depend on Hg species (organic matter bound Hg versus cinnabar ore and calcine). Hg species and Hg isotope ratio distribution are related to different grain-size fractions, and strongly influenced by hydrology. Thus, the use of Hg isotope analysis to identify Hg pollution sources and pathways should be combined with solid phase Hg speciation to distinguish isotopic fingerprints of Hg sources from Hg species and related Hg species transformation processes.

1 Introduction

Mercury (Hg) is a globally distributed pollutant1 that has raised critical concern due to its capacity to form monomethylmercury (MMHg), which is a neurotoxin with potential for biomagnification in the food chain.1–3 Chemical transformations of Hg are related to prevailing Hg species in environmental compartments,4 whereas the pooling and distribution of Hg species depend on natural and anthropogenic emission, and re-emission processes.5 Hg mining activities are considered a significant source of contamination6 generating residues and Hg species that may be different to those of the mined ore.

In the past decade, Hg isotope ratio measurements have been used to fingerprint Hg sources,7–10 and also as a tool for monitoring species transformations and their underlying chemical reactions.11 Large-magnitude variations in Hg isotopic composition can be caused by mass dependent (MDF) or mass independent (MIF) fractionation.12,13 For instance, processes associated with MDF include adsorption,14 mineral precipitation,15 desorption,16 mineral dissolution,14 volatilization,17 methylation and demethylation,18,19 and photoreduction.20 MIF is often associated with photochemical reduction of Hg, photodemethylation of MMHg3,21 and redox reactions.15,22 The first two processes produce larger MIF as a result of the magnetic isotope effect,23 conversely to the nuclear volume effect that causes smaller MIF.24 Typically, MDF is expressed as the ratio 202Hg/198Hg and reported as δ202Hg using the δ-notation,25 while MIF is often expressed as Δ199Hg, reporting the difference between measured δ199Hg and the predicted δ-values based on MDF alone.20

Also, Hg isotope ratios have been used to identify Hg sources of contamination, mixing of sources and fractionation processes in mining regions impacted by gold mining, coal deposits and Hg mines.7,9,26 Cinnabar is the main Hg-bearing ore mineral,27 and it has been mined worldwide. Particularly, intensive Hg mining activities in Idrija, Slovenia left a polluted environment that is extended to the regional scale.28 Cinnabar was continuously mined for centuries in Idrija, and has caused contamination of the aquatic system due to the dumping of mining residues, and the contamination of soils by atmospheric Hg deposition attributed to huge emissions of gaseous elemental mercury (GEM) from ore roasting.29,30 Consequently, river sediments are loaded with cinnabar and calcines derived from ore roasting residues,30,31 whereas soils contain mainly Hg bound to natural organic matter (NOM-Hg).32,33 Hg derived from Idrija soils is transported through the Idrijca–Soča river system to the Gulf of Trieste (GT), Italy, mainly as suspended matter (SM). The distribution and relative abundance of the main Hg species, i.e., cinnabar and NOM-Hg, are grain-size dependent.34 The largest input of Hg to the GT is attributed to the suspended load of the riverine system (Idrijca–Soča),35 and therefore to changes in the hydrological regime. This controls not only the amount of suspended load transported to the GT,34,36 but also the distribution of SM grain-size fractions, and the proportions of Hg species in the marine environment. Foucher et al.9 showed that cinnabar (Idrija ore) exhibits a δ202Hg value from −0.38 to −0.14‰, and that a similar Hg isotopic fingerprint is observed in sediments from the Idrijca–Soča river system including its river mouth at the GT. In contrast, marine bottom sediments from the GT and the Adriatic Sea displayed more negative δ202Hg values, which were assumed to reflect the natural background of the Adriatic Sea.9 There, the variation in the distribution of δ202Hg values was attributed to the dilution of Hg derived from the mining area with marine sediments. However, the NOM-Hg derived from soils was not considered to contribute to the observed Hg isotope ratios, and the relationship between Hg isotope ratios and solid phase Hg species was not examined. In this study, we investigated the variation of the Hg isotope composition in SM, soils and sediments and their relationship with solid phase Hg species.34 We analyzed the Hg isotope ratios of (i) SM transported in the Idrijca river and its tributaries, (ii) soil collected from slopes adjacent to tributaries and (iii) from river sediments, with the aim of exploring the degree to which the variability in the Hg isotopic composition of SM, soils and sediments is linked to the proportion of different Hg species.

2 Experimental section

2.1 Site area and sample collection

The mining activities of the Idrija Hg mine resulted in large-scale contamination of soils and sediments in the area from Idrija to the GT. The area is drained by the Idrijca–Soča river system, which is characterized by high discharge events transporting large amounts of SM to the marine environment.36,37 Sampling procedures were previously described in Baptista-Salazar et al.34 Briefly, SM, soils and fine-grained creek sediments (FGS) were collected and fractionated in the laboratory according to grain-size. SM samples (n = 11) were obtained after a two-step filtration sequence of river water (15 to 140 L) during low flow conditions, and during rain events. Samples were collected at the Idrijca river (ID-samples, n = 4) and its tributaries (T-samples, n = 7), and the total particulate fraction (>0.45 μm) was used for further analysis. Soil samples (n = 6) were collected from hillslope top-soils located adjacent to Idrijca's tributaries. Three grain-size fractions were selected and analyzed, a fine particulate fraction (0.45–20 μm), a fine fraction (20–63 μm) and a fine-coarser fraction (63–90 μm), respectively. FGS samples (n = 10) were collected at the main Idrijca–Soča river system (ID-samples and SO-sample, respectively) and its tributaries (T-samples), and then the 0.45–20 μm fraction and the 20–63 μm fraction were analyzed. All samples (Fig. S1) were collected based on the distance to the mine and the importance of soil catchments as a supply of SM to the aquatic system.

2.2 Analytical methods

2.2.1 Total mercury analysis, mercury speciation and isolation of humic substances. The total Hg content was analyzed by means of a thermal-desorption technique followed by pre-concentration on a gold trap and Hg detection by cold vapor atomic absorption spectrometry (CVAAS) (DMA-80, Milestone, EPA Method 7473). Hg species and binding forms were determined by thermo-desorption-CVAAS.38 Hg is thermally reduced to elemental Hg (Hg(0)), released at specific temperatures, transported by nitrogen as a carrier gas (300 mL min−1) and continuously detected at 253.7 nm. Hg thermo-desorption graphs are obtained, and comparison with the reference material allows the identification of Hg species. Estimation of the proportions of Hg species in the samples is detailed in Baptista et al.34 Humic substances were isolated from 2 soil samples using a standard extraction method.39,40
2.2.2 Hg isotope ratio determination. Samples for analysis of Hg isotope ratios were digested using concentrated HNO3/H2SO4 (7[thin space (1/6-em)]:[thin space (1/6-em)]3)9,41 and cinnabar ore samples using concentrated HNO3/HCl (1[thin space (1/6-em)]:[thin space (1/6-em)]3).9,42 Between 20 and 200 mg of dried sample material was digested with 5 mL of the acid mixture in open glass vessels on a hot plate at ∼120 °C for 6–8 h. After cooling, the samples were diluted to 30 mL with MilliQ water (18 MΩ). The sample mass used for digestion was individually adapted to reach a final Hg concentration of ∼10 ng mL−1 for optimum performance and accuracy of the Hg isotope ratio measurements.25

Hg isotopic ratios were determined by multicollector-inductively coupled plasma mass spectrometry (MC-ICP/MS; Neptune Thermo Fisher Scientific, Germany). Digested samples were introduced using a continuous-flow cold-vapor generation system.25 The standard–sample–standard bracketing approach using a NIST SRM 3133 Hg solution was applied to correct for instrumental mass bias. The isotope composition of Hg was acquired by monitoring 198Hg, 199Hg, 200Hg, 201Hg and 202Hg isotopes.25 Isotope ratio variations were expressed in units of per mil (‰) and reported in standard δ-notation relative to NIST SRM 3133 for MDF (1) and MIF (2):20

 
δxxxHg = 1000 × ([(xxxHg/198Hg)sample/(xxxHg/198Hg)NIST 3133] − 1)xxx = 199, 200, 201 or 202(1)
 
ΔxxxHg = δxxx − (δ202Hg × Axxx) A199 = 0.2520, A200 = 0.5024, A201 = 0.7520(2)

The accuracy of Hg isotope ratio analysis was verified using an UM-Almadén Hg solution as a secondary reference material (δ202Hg measured mean and 1 SD: −0.57 ± 0.08‰, n = 38). The completeness of the digestions was ensured by comparing the Hg concentration obtained here with the Hg concentration reported in Baptista-Salazar et al.,34 for the same samples. Reference materials BCR-580 and PACS-2 (estuarine sediment and marine sediment, respectively) were also analyzed, and the measured Hg isotopic ratios agreed with published values43 (Table S1). Several samples were also repeatedly digested and analyzed to estimate sample homogeneity. The uncertainties of δ202Hg values varied between 0.05 and 0.13‰, and from 0.04 to 0.19 for Δ199Hg. The reported uncertainties (1SD, Table S1) represent the measurement uncertainty of replicate sample digests or the uncertainty of repeated measurements of the same digest. Linear regression, a t-test and a Mann Whitney test were used for comparison between samples.

3 Results and discussion

3.1 Hg speciation in suspended matter (SM), soils and fine-grained creek sediments (FGS)

The distribution of Hg species in the area affected by Hg mining activities in Idrija depends largely on contamination pathways, with atmospheric Hg deposition and dispersion of mining residues being the prominent Hg sources.28,31 Hg species related to atmospheric deposition usually dominate the finer grain-size fractions of soils, sediments and SM, while ore related Hg species such as cinnabar are associated with coarser grain-size fractions.32,34 As a consequence, the dispersion of Hg species in the Idrijca/Soča river system is closely related to changes in hydrological conditions.34,44

NOM-Hg and cinnabar are the main Hg species found in the particulate fraction (0.45–20 μm) of SM, soils and FGS samples. Quantification of peak areas of NOM-Hg34 ranged from 67.4 to 100% in SM, from 77.2 to 99.5% in soils, and from 12.6 to 99.9% in FGS samples (Table S1). Samples collected close to the mining area are enriched in cinnabar, which decreases with distance to the mine relative to NOM-Hg (Fig. 1). Differences in the distribution of Hg species in FGS and SM are attributed to hydrological separation of grain-size fractions.33 Finer grain-size fractions bearing mostly NOM-Hg are more easily released and mobilized than cinnabar.45 In general, there is a similar distribution of NOM-Hg in the particulate fraction in SM and soil samples indicating soils as the main source of the mobile Hg species in the Idrijca river and its tributaries, rather than the riverbank or bottom sediments as suggested in earlier studies.36,44


image file: c7em00443e-f1.tif
Fig. 1 Distribution of Hg species (natural organic matter-bound Hg (NOM-Hg) and cinnabar) and δ202Hg values of fine grained creek sediments (FGS-A, 0.45–20 μm and FGS-B, 20–63 μm) of the Idrijca river (ID-samples), Soča river (SO-sample) and its tributaries (T-samples), soil samples (A, 0.45–20 μm, B, 20–63 μm, and C, 63–90 μm), and suspended matter (>0.45 μm) collected during low flow conditions (SMlow-flow), and after rain events (SMrain). Locations of selected ancient roasting sites are shown (from Bavec et al.54 and Gosar and Čar55).

The proportion of cinnabar slightly increases in coarser grain-size fractions of FGS samples (Fig. 1 and Table S1), soil samples (20–90 μm) (Table S1), and also in SM collected during rain events (Fig. 1 and Table S1)34 compared to finer fractions (0.45–20 μm). In contrast to SM from the tributaries, SM samples collected from the Idrijca river show an increase of cinnabar downstream (Fig. 1). The dominance of cinnabar in the river is attributed to the re-suspension of bottom and riverbank sediments along the Idrijca river known to contain larger proportions of cinnabar,30 in contrast to the tributaries, where erosion of NOM-Hg is the main source of Hg in SM.34

Moreover, finer grain-size fractions of SM containing mostly NOM-Hg are transported almost entirely to the GT, while coarser fractions containing mainly cinnabar remain in the fluvial sediments, and in the river mouth. In contrast to cinnabar, the continuous transport of NOM-Hg is a potential threat for aquatic systems34 due to its greater availability for methylation reactions,46e.g., by photochemical reactions involving humic substances.4 This may explain the comparatively high MMHg concentrations observed in marine sediments and fish distant from the coast of the GT,47 and adds to MMHg formed by methylation of Hg deposited into the water column from the atmosphere.48,49

3.2 Hg isotopic composition of the source and distribution of Hg isotope ratios in suspended matter (SM), soils and fine-grained creek sediments (FGS)

3.2.1 Isotopic composition of the Hg source. The δ202Hg values of cinnabar ores worldwide show a large variation in Hg isotopic composition.27,50–52 Those variations (primary MDF) are inherent to the processes related to the ore formation (e.g., Hg bearing mineralized rocks, composition of host rocks and hydrothermal fluids).50 The Idrija deposit exhibits different ore zones including concordant and discordant mineralization of cinnabar, elemental Hg, and minor amounts of Hg bearing pyrite and meta-cinnabar.53 The Hg-hosting rocks are strongly disturbed by folds and thrusts, and include black limestones, shales, sandstones, and conglomerates. In addition, different sulfur sources (magmatic and evaporitic, seawater sulfate, sedimentary pyrite and organic sulfur) were reported for the Idrija ore, which exhibits a large variation in δ34S-cinnabar.53 Primary MDF results in natural variation of δ202Hg values of cinnabar throughout the ore body. Therefore, it is debatable if a single δ202Hg value is appropriate to track Hg sources. The known δ202Hg values of the cinnabar ore from Idrija range from −0.38 to −0.14‰, and this was used as a reference to track the Hg contamination from the Idrija Hg mine.9

However, this narrow range may be insufficient for characterization, given the inherent heterogeneity of the ore body. Gray et al.50 found high variability in δ202Hg values of cinnabar in Almadén, Spain in relation to the host rock. While Almadén cinnabar ores show a mean δ-value of −0.82‰, the δ202Hg values of cinnabar in pyrite (−1.73‰) and in limestone (−1.37‰) were considerably more negative.50 Here, the δ202Hg value of a mixed cinnabar rock sample (quartzite, limestone and rocks containing pyrite) ranged from −1.64 to −1.54‰ (Table S1). These values are more positive (from 1.68 to 2.02‰) than the previous range reported for the Idrija cinnabar ore (−0.38 to −0.14‰).9 On the other hand, two samples of Hg(0) produced by cinnabar roasting in Idrija show a general and non-uniform enrichment in light Hg isotopes (Table S1) with δ202Hg of −0.36 and −1.12‰.

In addition to varying Hg isotope ratios in the ore, the ore processing technique used in the Idrija mine varied throughout the centuries,56,57 which led to Hg losses and releases to all environmental compartments during the operational period of the mine. Ore roasting generally leads to more negative δ202Hg in the produced metallic Hg while the remaining cinnabar gets isotopically more positive compared to the original cinnabar ore.50 However, recoveries were usually incomplete and up to 40% of the Hg(0) escaped to the atmosphere.58,59 It has been assumed that condensed bulk GEM shows the Hg isotopic composition of the parent cinnabar,52,60 in contrast to the GEM lost to the atmosphere which shows a different isotopic composition,61 probably enriched in light isotopes. Thus, residual calcines and GEM may show large variations in Hg isotopic composition relative to the unprocessed cinnabar and condensed Hg(0).61 For instance, calcines from the Wanshan mining areas in China and from the two largest Hg mining districts in the U.S show an enrichment in heavier isotopes of 0.80‰ and 3.24‰ in comparison with the δ202Hg of the original ore, −0.74 and −1.72‰, respectively.27,52 On the other hand, metacinnabar presumably forms during roasting,62 resulting in a depletion of light isotopes, and more positive δ202Hg values compared to those of cinnabar, which agrees with the δ202Hg value of 0.23 ± 0.16‰ reported in Foucher et al.9

The GEM lost to the atmosphere could be either subject to long distance transport or to dry or wet regional direct atmospheric deposition or litterfall to soils.63 There, it is strongly bound to organic matter, a process known for causing an additional Hg isotopic fraction with further enrichment in light Hg isotopes.64 Accordingly, δ202Hg values of NOM-Hg derived from atmospheric deposition of GEM in soils of the Idrija area are likely to be more negative than those of ores, mining residues or gangue material. The Idrija Hg mine as a local Hg source with different pollution pathways produced more than one Hg species with varying Hg isotope ratios, so that the contribution of all of them needs to be taken into account when characterizing end-members. The mixing from different sources may complicate the definition of individual source signatures which may be further altered by natural species transformation processes.65

3.2.2 Mass dependent fractionation. The distribution of δ202Hg values (MDF, Fig. S2) in SM, soil and FGS samples in the grain-size fractions from 0.45 to 90 μm is presented in Fig. 1. Data for δ201Hg, δ200Hg and δ199Hg are reported in the ESI (Table S1). In all the samples, δ202Hg values vary between −3.06 and 1.17‰ in the 0.45–90 μm fraction. In the total particulate fraction (>0.45 μm), SM exhibited δ202Hg values from −2.58 to 0.19‰, whereas in soils and FGS δ202Hg values ranged from −2.27 to −0.88‰ and −3.06 to 0.13‰, respectively (Fig. 1). FGS samples and SM collected during rain events, both from the Idrijca river and its tributaries, displayed more positive δ202Hg values, whereas soils, SM collected in tributaries and FGS samples collected at locations distant from the mining area exhibited more negative δ202Hg values. The variation in δ202Hg is similar to that found by Foucher et al.9 in surface sediments collected at the Soča (Slovenia)/Isonzo (Italy) river, and marine sediments from the GT and the Adriatic Sea. In that study, more negative δ202Hg values (−2.53 to −1.19‰) were mainly found in marine sediments distant from the river mouth, while more positive δ202Hg values (−0.48 to −0.13‰) were restricted to the northern part of the GT (near the river mouth), and to river sediments.9

We did not find significant differences in δ202Hg values of SM collected under low flow in the tributaries (δ202Hg = −1.51 ± 1.11‰, p > 0.05), suggesting that the Hg in SM is derived from the same source or related to the same Hg species independent of hydrological conditions. δ202Hg values of SM collected in tributaries (−2.08 ± 0.57‰) are not significantly different from those found in soil (0.45–20 μm, same locations of SM, n = 16, U = 31, p > 0.05), indicating that Hg in SM collected in tributaries is mainly soil derived, and ultimately associated with NOM-Hg.

NOM-Hg is thought to represent Hg associated with humic substances (HS) in soils,41,42 which are a complex mixture of organic constituents of soil organic matter.66 In order to determine the δ202Hg values of NOM-Hg, we isolated HS from two soil samples (0.45–20 μm), T5 and T11. The δ202Hg value of the extracted HS in soil samples T5 and T11 was −1.85‰ and −0.84‰, respectively, which were similar to the values obtained for the original sample prior to extraction (−1.89‰ in T5 and −0.88‰ in T11). Prior to extraction, the proportions of NOM-Hg in samples T5 and T11 had been 97.2% and 96%, respectively,34 and after extraction NOM-Hg in both samples reached up to 99%. The organic carbon content had been 26.2% and 4.26% prior to extraction,34 and increased to 69% in T5 and 57% in T11 after the extraction. The similarity of δ202Hg values between the soil sample and the isolated HS indicates that HS-bound Hg largely determines the Hg isotope composition in these soils. Thus, more negative δ202Hg values in those samples are related to high proportions of this mobile and more bioavailable Hg species, which is also the dominant Hg species in soils and SM collected in tributaries.

In contrast to δ202Hg of SM collected in tributaries of the Idrijca river, the δ202Hg values of SM collected in the Idrijca river become more positive with distance from the mine, which was attributed to resuspension of bottom sediments loaded with a higher proportion of cinnabar and calcines34 (Fig. S3). Erosion of coarser grain-size fractions increases the proportions of cinnabar in the suspended load transported in the Idrijca river system.34 Thus, higher proportions of cinnabar led to more positive δ202Hg values mainly in FGS (r2 = 0.933, p < 0.05), and secondarily in SM (r2 = 0.259, p > 0.05) collected in the Idrijca river. The measured Hg isotopic compositions are a result of the variability of the Hg isotopic composition of cinnabar and calcines, and their occurrence in different grain-size fractions.

The change from more positive to more negative δ202Hg values (Fig. 1 and S3) is related to the increase in proportions of NOM-Hg in soils and SM from tributaries, which increase with distance from the mine. Variations in δ202Hg values related to proportions of cinnabar and NOM-Hg are shown in Fig. 2. Although the entire dataset is lacking significant correlation, some general conclusions regarding the relationship between Hg isotopic composition and Hg species can be drawn based on sample type (soils, SM, and FGS) and location (Idrijca river or tributaries). Based on the Hg species proportions, FGS samples show the largest range of δ202Hg values (Fig. 2), with the most positive values in samples collected close to ancient small-scale roasting sites (T2 and T4, Fig. 1 and 2). These samples are also characterized by high proportions of cinnabar (>70%) or calcine and the highest Hg concentrations (29.3 and 306 mg kg−1, respectively).


image file: c7em00443e-f2.tif
Fig. 2 Percentage of natural-organic-matter-bound Hg (NOM-Hg) and cinnabar vs. δ202Hg of SM collected from the Idrijca river (ID) and its tributaries (T), the Soča river (SO), soils and sediment (FGS) samples. Particular colors represent main rivers and tributaries: ID and SO in black, and T in blue. Also, each type of sample is represented by a particular filling of the symbol: SM as half-solid, FGS as open, and soils as solid symbols. The δ202Hg value of cinnabar is indicated by a red line, and that of metacinnabar by a black line.9 All grain-size fractions (μm) are shown. Samples exhibiting smaller variation of δ202Hg and higher NOM-Hg proportions are shown inside the ellipse.

The occurrence of metacinnabar is supposed to be of minor importance in the Idrija area,53 and previous studies have shown no indication of metacinnabar occurrence.32,34 Therefore, we propose that the more positive δ202Hg values (compared to cinnabar ore) found in our study are related to the roasting process (calcines). Physicochemical transformations of Hg at high temperatures induce significant MDF during and/or after the roasting process.60 Similarly, more positive δ202Hg values in cinnabar bearing soil, SM or FGS samples likely indicate that the residual cinnabar was (at least in part) subject to incomplete roasting and associated GEM losses.

This concept implies that higher proportions of cinnabar lead to more positive δ202Hg values, and that the variability in δ202Hg values is a result of varying proportions of NOM-Hg and cinnabar, respectively. However, this assumption does not readily explain the negative δ202Hg values in some FGS samples (Fig. 2), which are also characterized by high proportions of cinnabar. FGS samples may contain proportions of cinnabar from ore as well as from calcine, which would result in variable δ202Hg values independent of similar proportions of cinnabar. In addition, roasting may produce ash or fine ore particles dominated by lighter isotopes explaining more negative δ202Hg values even if proportions of cinnabar are high. Samples containing >50% of cinnabar exhibit a scatter in the δ202Hg values that is not observed in the samples containing more NOM-Hg (Fig. 2). The occurrence of cinnabar (and calcine) leads to a wide range of δ202Hg values and is attributed to the inherent variability of the ore and species transformation during roasting, whereas NOM-Hg seems to exhibit a smaller range of MDF.

The occurrence of cinnabar is grain-size dependent,34 and this explains the differences in δ202Hg values in the FGS samples comparing the 0.45–20 and 20–63 μm grain-size fractions (Fig. 1). Here, the δ202Hg values of FGS samples ID-4 and SO-1 in the fraction 0.45–20 μm are −0.53‰ and −0.96‰, respectively, and proportions of cinnabar are 86.9% and 95.4%, respectively. In contrast, the coarser 20–63 μm fraction in both samples shows more positive δ202Hg values (0.22‰ and 0.12‰), although proportions of cinnabar in both samples are slightly larger (93.8% and 97.7%). The variable Hg isotopic composition for cinnabar and calcines suggests that the increase in δ-values is not necessarily proportional to the cinnabar content. Therefore, similar proportions of cinnabar in different samples show a large range of δ202Hg. The ID-4 FGS sample exhibited a more negative δ202Hg value compared to the related ID-4 SM sample. During rain events, SM resembles eroded FGS, since the energy transport under those hydrological conditions promotes the erosion of larger grain-size fractions, and the resuspension of bottom sediments loaded with both cinnabar from ore and calcines, which results in more positives Hg isotope ratios. This also explains the differences between samples collected during low and high (rain events) flow conditions. The proportion of cinnabar increased from 12.4% to 34.1% in SM samples collected during rain events, which led to ∼0.44‰ more positive δ202Hg values (median values, n = 3). On the other hand, the occurrence of cinnabar often increases the Hg concentration.34 Hg concentration in samples containing >50% of cinnabar ranged from 0.93 to 306 mg kg−1, which is larger than that observed in samples with less cinnabar (<50%, 0.58 to 56.2 mg kg−1). This suggests that the Hg concentration in itself is not the deciding factor defining Hg isotopic composition, and therefore, it is not sufficient to define the extent of contamination of the Hg source itself.

Soils samples containing high proportions of NOM-Hg (>70%) revealed more negative δ202Hg values in all the grain-size fractions (from 0.45 to 90 μm) than the FGS samples which contain mainly cinnabar, underlining strong connection between Hg species and δ202Hg values (Fig. 2). This was particularly observed in SM from tributaries (all with the exception of the one closest to the mine, SM-T2), SM-ID-1 (Idrijca river, upstream of the mining area) and FGS samples collected far away from the mine.

3.2.3 Mass independent fractionation. The distribution of Δ199Hg is shown in Fig. 3, and values for Δ201Hg and Δ200Hg are presented in Table S1. Samples analyzed in this study showed a reasonable spread in Δ199Hg values ranging from −0.37 to 0.09‰. However, as with MDF, there is no obvious relationship between the sample type and MIF. Instead, individual values are controlled by a combination of species and sampling location. For example, none of the sampled soils showed significant difference in Δ199Hg among different grain size fractions. Variations in Hg isotopic composition were mainly found in soil samples, SM from tributaries and FGS from tributaries. Here, the Δ199Hg values ranged from −0.14 to 0.09‰, which is similar to those found in marine sediments (GT and Adriatic, −0.09 to 0.08‰).9 Overall, there appears to be a general trend for increasing Δ199Hg values in samples having more negative δ202Hg values. This is consistent with previous studies analyzing terrestrial samples from Hg mining areas containing mine wastes.52,60 This “opposite trend” of MDF and MIF is characteristic for nuclear volume fractionation, and may be caused by redox reactions during ore roasting (reduction) and secondary re-oxidation of Hg(0).60 Moreover, samples with high proportions of cinnabar and high mercury concentration may exhibit less MIF in comparison with samples dominated by NOM-Hg. This is related to the slow weathering of cinnabar67 that hinders processes controlled by the magnetic isotope effects, which generally results in larger MIF (e.g., photochemical reduction21 and photodegradation,68 and liquid-evaporation69). The isotopic variation of δ202Hg and Δ199Hg among all samples in our study indicates a mixing of cinnabar and NOM-Hg caused by the ore mining.
image file: c7em00443e-f3.tif
Fig. 3 Hg isotopic composition (δ202Hg and Δ199Hg) of suspended matter (SM) collected from the Idrijca river (ID) and its tributaries (T), from the Soča river (SO), of soils, and sediment samples (FGS). Particular colors represent main rivers and tributaries: ID and SO in black, T in blue, and marine samples in green. Also, each type of sample is represented by a particular type and/or filling of the symbol: SM as half-solids, FGS as open symbols, soils as solid symbols, marine sediments as stars, metacinnabar as a cross and cinnabar as a plus.9

3.3 Implications for the distribution of Hg species and Hg isotopes in the GT

The Idrijca–Soča/Isonzo (in Italy) river system continuously exports Hg to the GT, and discharge is prone to be enhanced during heavy rain events, when the sediment load increases.34–36 Thus, the distribution of Hg species and isotopes found in the GT is strongly affected by variations in the hydrological regime. Changes in the transport energy spread grain-size fractions at different distances from the Isonzo river mouth and towards the central parts of the GT, with variable proportions of cinnabar and NOM-Hg.34,47 In addition, the transport of particles in the GT is influenced by wind direction37,70 (NE to ENE),71 tides and the anticlockwise circulation system.37 Rain events promote erosion and resuspension of bottom sediments containing cinnabar, resulting in the transport of coarser grain-size fractions with both higher Hg concentrations and more positive δ202Hg values. Our results would readily explain the more negative δ202Hg values found at distant locations to the shoreline (−2.39 to −1.49‰), as well as confirm the more positive δ202Hg values found in and near the river mouth (−0.38 to −0.14‰) reported by Foucher et al.9 Hence, the observed Hg isotope ratio distributions may alternatively be explained by the spatial distribution of different Hg-species (NOM-Hg and cinnabar) in the GT. The variability of δ202Hg values in the GT, from more positive to more negative in southern areas of the GT and the Adriatic Sea in comparison to the northern part of the GT and the river system, was previously attributed to the dilution of the Hg (cinnabar) derived from Idrija with marine sediments.9 However, Foucher et al.9 analyzed river sediments only, containing mainly cinnabar with higher Hg concentrations (coarser grain-size fractions), while marine sediments with low Hg concentrations were considered as not impacted by the Hg mine, so that δ202Hg values in those samples were assumed to be the natural background of Adriatic Sea sediments. A high contribution of terrestrial sources was reported in the northern and central Adriatic Sea.72 Therefore, the differentiation between the natural Hg background (autochthonous) and terrestrial Hg sources (allochthonous) based on differences in Hg isotope ratios is not straightforward. Exported NOM-Hg may be a significant source of Hg in some areas of the GT. However, since its Hg isotope composition could be very similar to the marine background, the question of source differentiation becomes challenging. Processes in deeper sediment layers, and also in areas outside the river plume may lead to larger MIF. This would explain the enrichment in odd isotopes in Adriatic Sea bottom sediments, which was not observed in samples from the GT9 or in the samples of our study.

Our results show that the definition of a natural Hg isotopic composition or background value in an area affected by high Hg input for more than 500 years of Hg mining is difficult. The Hg isotopic composition is naturally shifted by natural processes in soils, aquatic systems and biota, and it is further altered by ore processing. Our study stresses the importance of combining Hg species and Hg isotopic composition to better understand transformations and to determine whether the Hg isotopic composition found in the environment is distinctly different from that of the Hg source of contamination.

3.4 Influence of end-members to predict the isotopic composition of mercury

Binary mixing models of Hg isotopes have been used to trace contamination sources, and also to predict the contributions of different contamination sources.9,10 Typically, these models assumed that both the regional background as well as the contaminant were sufficiently characterized by a single or narrow range of δ202Hg values and Hg concentration. Previous calculations predicted that 90% of the Hg found in fluvial Idrijca–Soča sediments and marine sediments in the northern part of the GT originated from the Idrija region, whereas this contribution was reduced to only 40–50% in other GT sediments.9 Cinnabar was considered as the main source for explaining Hg contamination and the resulting variations in δ202Hg. In contrast, we were not able to fit our data to this simple binary model (r2 = 0.3326, p > 0.05). The measured δ202Hg values in our work were more negative than previously reported values for δ202Hg in the Idrija mining area. While the binary model uses a single δ-value for cinnabar to predict δ-values of impacted environmental samples, we now believe that Hg associated with different cinnabar ores and other ore roasting products is much more varied, and more information is needed to discriminate “pristine” material from δ202Hg values of the entire ore body. Also, some of the Hg isotopic compositions observed in this study seem to be more similar to calcines, which is in agreement with studies conducted in other Hg mining areas.27 There is a wide variation in δ202Hg for Hg deposits and cinnabar mineralized rocks,27,50–52 so that the use of a narrowly defined δ202Hg value or range to characterize the cinnabar ore seems questionable. Furthermore, we did not find a correlation between δ202Hg and 1/Hg (r2 = 0.1759, p > 0.05, Fig. S4). The lack of correlation was not surprising, considering (i) that only 30% of our samples contains cinnabar, (ii) the variability in the Hg isotopic composition of the ore and that caused by ore roasting, (iii) the lower Hg concentrations compared to Foucher et al.,9 and (iv) that SM, FGS and soil samples show differences in both Hg species and Hg concentrations. Hg concentrations may reflect Hg contamination and indicate mixing of sources, but should not be used alone for defining Hg sources and mixing, or to predict Hg isotopic compositions.

The inherent processes related to the ore formation (primary MDF, Hg-bearing mineralized rocks and composition of host rocks) shift the δ202Hg values and result in a continuum of δ202Hg, which would greatly complicate any mixing model. Also, ore processing (secondary MDF) causes additional variation in the Hg isotopic composition of the contamination source. Although, the occurrence of cinnabar in samples affected by mining activities in Idrija clearly leads to more positive δ202Hg values, the isotope composition of the environmental samples will depend on all Hg species.

4 Conclusions

From combined solid phase Hg species and Hg isotope measurements in soils, FGS and SM, we found that the Hg isotopic composition and distribution of Hg species in the Idrija mining area are closely connected. Our data reveal that the distribution of cinnabar and NOM-Hg determines the Hg isotopic composition in the area affected by Hg mining in Idrija. Soils contaminated by atmospheric Hg deposition are characterized by a dominance of NOM-Hg and show more negative δ202Hg values in all grain-size fractions. Sediments dominated by cinnabar ore or calcine show more positive δ202Hg values in coarser grain-size fractions. Our results indicate the need for combining Hg isotope analysis with Hg speciation to differentiate Hg isotopic compositions related to the ore and to Hg species. Different from previous studies,14 the observed variability in the Hg isotopic composition and Hg species further reveals that there is not a defined Hg isotopic composition for the endmember regarding the source of Hg. Our interpretation offers an alternative explanation for the different δ202Hg values observed in sediments of the GT in previous studies. More positive δ202Hg values in the GT river mouth sediments are related to the occurrence of cinnabar and calcines, while more negative δ202Hg values in GT bottom sediments are related to NOM-Hg from soils of the Idrija mining area than to the Adriatic Sea Hg background. Moreover, the distribution and transport of Hg species and Hg isotope ratios are strongly related to grain-size fractions, and therefore dependent on hydrological conditions. Variations observed in δ202Hg values were related to large MDF and small MIF during ore processing and natural variation in the ore body. The occurrence of cinnabar and calcines led to larger variation of δ202Hg values in the samples, in comparison with the proportions of NOM-Hg. This variation is also favored for the erosion of larger grain-size fractions containing higher proportions of cinnabar and calcines. Although, a better interpretation of different δ-values is given by combining Hg isotope analysis and Hg speciation, prediction of Hg isotopic composition results is challenging. Our study suggests that the variation in Hg isotopic signatures is likely given by different Hg species or species transformations rather than Hg sources.

Conflicts of interest

There are no conflicts to declare.

Acknowledgements

This work was funded by the NTH graduate school “GeoFluxes” of the federal state of Lower Saxony, Germany and, financially supported by AG Umweltgeochemie – TU Braunschweig, and the Deutsche Akademische Austauschdienst (DAAD). We are thankful for the logistic support of the Geological Survey of Slovenia, Komunala Idrija and the Idrija Mercury mine, especially to Mateja Rejc, Mateja Gosar, Zmago Bole, Tatjana Dizdarevič and Vesna Miklavcic for their kind help and assistance during sampling. We thank Brian Dimock for his help and support during the Hg isotope analysis at Trent University.

References

  1. W. F. Fitzgerald, D. R. Engstrom, R. P. Mason and E. A. Nater, The case for atmospheric mercury contamination in remote areas, Environ. Sci. Technol., 1998, 32, 1–7 CrossRef CAS.
  2. M. Aastrup, J. Johnson, E. Bringmark, I. Bringmark and Å. Iverfeldt, Occurence and transport of mercury within a small catchment area, Water, Air, Soil Pollut., 1991, 56, 155–167 CrossRef CAS.
  3. J. D. Blum, Applications of stable mercury isotopes to biogeochemistry in Handbook of Environmental Isotope Geochemistry, ed. M. Baskaran, Springer, Berlin Heidelberg, 2012, pp. 229–245 Search PubMed.
  4. F. M. M. Morel, A. M. L. Kraepiel and M. Amyot, The chemical cycle and bioaccumulation of mercury, Annu. Rev. Ecol. Syst., 1998, 29, 543–566 CrossRef.
  5. C. T. Driscoll, R. P. Mason, H. M. Chan, D. J. Jacob and N. Pirrone, Mercury as a global pollutant: sources, pathways, and effects, Environ. Sci. Technol., 2013, 47, 4967–4983 CrossRef CAS PubMed.
  6. UNEP, Global mercury assessment. Sources. Emissions, releases and environmental transport, 2013 Search PubMed.
  7. P. M. Donovan, J. D. Blum, M. B. Singer, M. Marvin-DiPasquale and M. T. K. Tsui, Isotopic composition of inorganic mercury and methylmercury downstream of a historical gold mining region, Environ. Sci. Technol., 2016, 50, 1691–1702 CrossRef CAS PubMed.
  8. X. Feng, D. Foucher, H. Hintelmann, H. Yan, T. He and G. Qiu, Tracing mercury contamination sources in sediments using mercury isotope compositions, Environ. Sci. Technol., 2010, 44, 3363–3368 CrossRef CAS PubMed.
  9. D. Foucher, N. Ogrinc and H. Hintelmann, Tracing mercury contamination from the Idrija mining region (Slovenia) to the Gulf of Trieste using Hg isotope ratio measurements, Environ. Sci. Technol., 2009, 43, 33–39 CrossRef CAS PubMed.
  10. J. G. Wiederhold, U. Skyllberg, A. Drott, M. Jiskra, S. Jonsson, E. Björn, B. Bourdon and R. Kretzschmar, Mercury isotope signatures in contaminated sediments as a tracer for local industrial pollution sources, Environ. Sci. Technol., 2015, 49, 177–185 CrossRef CAS PubMed.
  11. B. A. Bergquist and J. D. Blum, The odds and evens of mercury isotopes: applications of mass-dependent and mass-independent isotope fractionation, Elements, 2009, 5, 353–357 CrossRef CAS.
  12. J. D. Blum, L. S. Sherman and M. W. Johnson, Mercury isotopes in earth and environmental sciences, Annu. Rev. Earth Planet. Sci., 2014, 42, 249–269 CrossRef CAS.
  13. S. Ghosh, Y. Xu, M. Humayun and L. Odom, Mass-independent fractionation of mercury isotopes in the environment, Geochem., Geophys., Geosyst., 2008, 9, Q03004 CrossRef.
  14. H. Hintelmann and R. Harris, Application of multiple stable mercury isotopes to determine the adsorption and desorption dynamics of Hg(II) and MeHg to sediments, Mar. Chem., 2004, 90, 165–173 CrossRef CAS.
  15. C. N. Smith, S. E. Kesler, B. Klaue and J. D. Blum, Mercury isotope fractionation in fossil hydrothermal systems, Geology., 2005, 33, 825–828 CrossRef CAS.
  16. B. A. Bessinger, Use of stable isotopes to identify sources of mercury in sediments: a review and uncertainty analysis, Environ. Forensics, 2014, 15, 265–280 CrossRef CAS.
  17. W. Zheng, D. Foucher and H. Hintelmann, Mercury isotope fractionation during volatilization of Hg(0) from solution into the gas phase, J. Anal. At. Spectrom., 2007, 22, 1097–1104 RSC.
  18. P. Rodríguez-González, V. N. Epov, R. Bridou, E. Tessier, R. Guyoneaud, M. Monperrus and D. Amouroux, 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, 9183–9188 CrossRef PubMed.
  19. H. Hintelmann, R. Douglas Evans and Y. Villeneuve, Measurement of mercury methylation in sediments by using enriched stable mercury isotopes combined with methylmercury determination by gas chromatography-inductively coupled plasma mass spectrometry, J. Anal. At. Spectrom., 1995, 10, 619–624 RSC.
  20. J. D. Blum and B. A. Bergquist, Reporting of variations in the natural isotopic composition of mercury, Anal. Bioanal. Chem., 2007, 388, 353–359 CrossRef CAS PubMed.
  21. W. Zheng and H. Hintelmann, Mercury isotope fractionation during photoreduction in natural water is controlled by its Hg/DOC ratio, Geochim. Cosmochim. Acta, 2009, 73, 6704–6715 CrossRef CAS.
  22. L. S. Sherman, J. D. Blum, K. P. Johnson, G. J. Keeler, J. A. Barres and T. A. Douglas, Mass-independent fractionation of mercury isotopes in Arctic snow driven by sunlight, Nat. Geosci., 2010, 3, 173 CrossRef CAS.
  23. A. L. Buchachenko, Mercury isotope effects in the environmental chemistry and biochemistry of mercury-containing compounds, Russ. Chem. Rev., 2009, 78, 319 CrossRef CAS.
  24. E. A. Schauble, Role of nuclear volume in driving equilibrium stable isotope fractionation of mercury, thallium, and other very heavy elements, Geochim. Cosmochim. Acta, 2007, 71, 2170–2189 CrossRef CAS.
  25. D. Foucher and H. Hintelmann, High-precision measurement of mercury isotope ratios in sediments using cold-vapor generation multi-collector inductively coupled plasma mass spectrometry, Anal. Bioanal. Chem., 2006, 384, 1470–1478 CrossRef CAS PubMed.
  26. G. E. Gehrke, J. D. Blum and M. Marvin-DiPasquale, Sources of mercury to San Francisco Bay surface sediment as revealed by mercury stable isotopes, Geochim. Cosmochim. Acta, 2011, 75, 691–705 CrossRef CAS.
  27. S. J. Stetson, J. E. Gray, R. B. Wanty and D. L. Macalady, Isotopic variability of mercury in ore, mine-waste calcine, and leachates of mine-waste calcine from areas mined for mercury, Environ. Sci. Technol., 2009, 43, 7331–7336 CrossRef CAS PubMed.
  28. D. Kocman, P. Vreča, V. Fajon and M. Horvat, Atmospheric distribution and deposition of mercury in the Idrija Hg mine region, Slovenia, Environ. Res., 2011, 111, 1–9 CrossRef CAS PubMed.
  29. T. Dizdarevič, U. Repinc and B. Režun, Monitoring and assessment of mercury pollution in the Idrija mining area – environmental influences of mercury ore processing – case studies selected at Slovenian, Mexican, Hungarian group meeting in Idrija, July 2012 Search PubMed.
  30. M. Gosar, S. Pirc and M. Bidovec, Mercury in the Idrijca river sediments as a reflection of mining and smelting activities of the Idrija mercury mine, J. Geochem. Explor., 1997, 58, 125–131 CrossRef CAS.
  31. H. Biester, M. Gosar and G. Müller, Mercury speciation in tailings of the Idrija mercury mine, J. Geochem. Explor., 1999, 65, 195–204 CrossRef CAS.
  32. H. Biester, M. Gosar and S. Covelli, Mercury speciation in sediments affected by dumped mining residues in the drainage area of the Idrija mercury mine, Slovenia, Environ. Sci. Technol., 2000, 34, 3330–3336 CrossRef CAS.
  33. M. Gosar, R. Šajn and H. Biester, Binding of mercury in soils and attic dust in the Idrija mercury mine area (Slovenia), Sci. Total Environ., 2006, 369, 150–162 CrossRef CAS PubMed.
  34. C. Baptista-Salazar, J.-H. Richard, M. Horf, M. Rejc, M. Gosar and H. Biester, Grain-size dependence of mercury speciation in river suspended matter, sediments and soils in a mercury mining area at varying hydrological conditions, Appl. Geochem., 2017, 81, 132–142 CrossRef CAS.
  35. A. Širca, R. Rajar, R. C. Harris and M. Horvat, Mercury transport and fate in the Gulf of Trieste (Northern Adriatic)—a two-dimensional modelling approach, Environ. Model. Software, 1999, 14, 645–655 CrossRef.
  36. D. Žagar, A. Knap, J. J. Warwick, R. Rajar, M. Horvat and M. Četina, Modelling of mercury transport and transformation processes in the Idrijca and Soča river system, Sci. Total Environ., 2006, 368, 149–163 CrossRef PubMed.
  37. S. Covelli, J. Faganeli, M. Horvat and A. Brambati, Mercury contamination of coastal sediments as the result of long-term cinnabar mining activity (Gulf of Trieste, northern Adriatic sea), Appl. Geochem., 2001, 16, 541–558 CrossRef CAS.
  38. H. Biester and C. Scholz, Determination of mercury binding forms in contaminated soils: mercury pyrolysis versus sequential extractions, Environ. Sci. Technol., 1997, 31, 233–239 CrossRef CAS.
  39. R. S. Swift, Organic matter characterization, Methods Soil Anal. Part 3—Chemical Methods, 1996, vol. 35, pp. 1011–1069 Search PubMed.
  40. Isolation of IHSS Samples|IHSS, http://humic-substances.org/isolation-of-ihss-soil-fulvic-and-humic-acids/, accessed December, 2017.
  41. A. Kuldvere, Decomposition of cinnabar and organomercurials in geological materials with nitric acid-sulphuric acid for the determination of total mercury by cold vapour atomic absorption spectrometry, Analyst, 1990, 115, 559–562 RSC.
  42. R. Fernández-Martínez and M. I. Rucandio, Study of the suitability of HNO3 and HCl as extracting agents of mercury species in soils from cinnabar mines, Anal. Bioanal. Chem., 2005, 381, 1499–1506 CrossRef PubMed.
  43. J. Liu, X. Feng, R. Yin, W. Zhu and Z. Li, Mercury distributions and mercury isotope signatures in sediments of Dongjiang, the Pearl River Delta, China, Chem. Geol., 2011, 287, 81–89 CrossRef CAS.
  44. D. Kocman, T. Kanduč, N. Ogrinc and M. Horvat, Distribution and partitioning of mercury in a river catchment impacted by former mercury mining activity, Biogeochemistry, 2011, 104, 183–201 CrossRef CAS.
  45. K. Gai, T. P. Hoelen, H. Hsu-Kim and G. V. Lowry, Mobility of four common mercury species in model and natural unsaturated soils, Environ. Sci. Technol., 2016, 50, 3342–3351 CrossRef CAS PubMed.
  46. S. Jonsson, U. Skyllberg, M. B. Nilsson, P.-O. Westlund, A. Shchukarev, E. Lundberg and E. Björn, Mercury methylation rates for geochemically relevant HgII species in sediments, Environ. Sci. Technol., 2012, 46, 11653–11659 CrossRef CAS PubMed.
  47. R. Piani, S. Covelli and H. Biester, Mercury contamination in Marano Lagoon (Northern Adriatic sea, Italy): source identification by analyses of Hg phases, Appl. Geochem., 2005, 20, 1546–1559 CrossRef CAS.
  48. J. D. Blum, B. N. Popp, J. C. Drazen, C. A. Choy and M. W. Johnson, Methylmercury production below the mixed layer in the North Pacific Ocean, Nat. Geosci., 2013, 6, 879 CrossRef CAS.
  49. R. F. Lepak, R. Yin, D. P. Krabbenhoft, J. M. Ogorek, J. F. DeWild, T. M. Holsen and J. P. Hurley, Use of stable isotope signatures to determine mercury sources in the Great Lakes, Environ. Sci. Technol. Lett., 2015, 2, 335–341 CrossRef CAS.
  50. J. E. Gray, M. J. Pribil and P. L. Higueras, Mercury isotope fractionation during ore retorting in the Almadén mining district, Spain, Chem. Geol., 2013, 357, 150–157 CrossRef CAS.
  51. R. S. Smith, J. G. Wiederhold, A. D. Jew, G. E. Brown Jr, B. Bourdon and R. Kretzschmar, Stable Hg isotope signatures in creek sediments impacted by a former Hg Mine, Environ. Sci. Technol., 2015, 49, 767–776 CrossRef CAS PubMed.
  52. R. Yin, X. Feng, J. Wang, P. Li, J. Liu, Y. Zhang, J. Chen, L. Zheng and T. Hu, Mercury speciation and mercury isotope fractionation during ore roasting process and their implication to source identification of downstream sediment in the Wanshan mercury mining area, SW China, Chem. Geol., 2013, 336, 72–79 CrossRef CAS.
  53. J. V. Lavric and J. E. Spangenberg, Stable isotope (C, O, S) systematics of the mercury mineralization at Idrija, Slovenia: constraints on fluid source and alteration processes, Miner. Deposita, 2003, 38, 886–899 CrossRef CAS.
  54. Š. Bavec, H. Biester and M. Gosar, Urban sediment contamination in a former Hg mining district, Idrija, Slovenia, Environ. Geochem. Health, 2014, 36, 427–439 CrossRef PubMed.
  55. M. Gosar and J. Čar, Vpliv žgalnic živosrebrove rude iz 16. in 17. stoletja na razširjenost živega srebra v okolici Idrije [influence of mercury ore roasting sites from 16th and 17th century on the mercury dispersion in surroundings of Idrija], Geologija, 2006, 49, 91–101 CrossRef.
  56. T. Teršič, M. Gosar and H. Biester, Distribution and speciation of mercury in soil in the area of an ancient mercury ore roasting site, Frbejžene trate (Idrija area, Slovenia), J. Geochem. Explor., 2011, 110, 136–145 CrossRef.
  57. T. Teršič, M. Gosar and H. Biester, Environmental impact of ancient small-scale mercury ore processing at Pšenk on soil (Idrija area, Slovenia), Appl. Geochem., 2011, 26, 1867–1876 CrossRef.
  58. L. D. Hylander and M. Meili, The rise and fall of mercury: converting a resource to refuse after 500 years of mining and pollution, Crit. Rev. Environ. Sci. Technol., 2005, 35, 1–36 CrossRef CAS.
  59. D. G. Streets, M. K. Devane, Z. Lu, T. C. Bond, E. M. Sunderland and D. J. Jacob, All-time releases of mercury to the atmosphere from human activities, Environ. Sci. Technol., 2011, 45, 10485–10491 CrossRef CAS PubMed.
  60. J. G. Wiederhold, R. S. Smith, H. Siebner, A. D. Jew, G. E. Brown Jr, B. Bourdon and R. Kretzschmar, Mercury isotope signatures as tracers for Hg cycling at the New Idria Hg mine, Environ. Sci. Technol., 2013, 47, 6137–6145 CrossRef CAS PubMed.
  61. R. Sun, D. G. Streets, H. M. Horowitz, H. M. Amos, G. Liu, V. Perrot, J.-P. Toutain, H. Hintelmann, E. M. Sunderland and J. E. Sonke, Historical (1850–2010) mercury stable isotope inventory from anthropogenic sources to the atmosphere, Elementa: Science of the Anthropocene, 2016, 4, 91 Search PubMed.
  62. C. S. Kim, J. J. Rytuba and G. E. Brown, Geological and anthropogenic factors influencing mercury speciation in mine wastes: an EXAFS spectroscopy study, Appl. Geochem., 2004, 19, 379–393 CrossRef CAS.
  63. J. D. Demers, J. D. Blum and D. R. Zak, Mercury isotopes in a forested ecosystem: Implications for air-surface exchange dynamics and the global mercury cycle, Global Biogeochem. Cycles, 2013, 27, 222–238 CrossRef CAS.
  64. J. G. Wiederhold, C. J. Cramer, K. Daniel, I. Infante, B. Bourdon and R. Kretzschmar, Equilibrium mercury isotope fractionation between dissolved Hg(II) species and thiol-bound Hg, Environ. Sci. Technol., 2010, 44, 4191–4197 CrossRef CAS PubMed.
  65. R. Das, W. Landing, M. Bizimis, L. Odom and J. Caffrey, Mass Independent Fractionation of Mercury Isotopes as Source Tracers in Sediments, Procedia Earth Planet. Sci., 2015, 13, 151–157 CrossRef CAS.
  66. M. Schnitzer, Humic Substances: Chemistry and Reactions. in Developments in Soil Science, Elsevier, 1978, vol. 8, ch. 1, pp. 1–64 Search PubMed.
  67. T. Robson, Weathering of sulfide ores in model soils, potentially toxic element release and bioavailability, PhD thesis, Plymouth University, 2013.
  68. B. A. Bergquist and J. D. Blum, Mass-Dependent and -Independent Fractionation of Hg Isotopes by Photoreduction in Aquatic Systems, Science, 2007, 318, 417–420 CrossRef CAS PubMed.
  69. N. Estrade, J. Carignan, J. E. Sonke and O. F. X. Donard, Mercury isotope fractionation during liquid–vapor evaporation experiments, Geochim. Cosmochim. Acta, 2009, 73, 2693–2711 CrossRef CAS.
  70. S. Covelli, R. Piani, A. Acquavita, S. Predonzani and J. Faganeli, Transport and dispersion of particulate Hg associated with a river plume in coastal Northern Adriatic environments, Mar. Pollut. Bull., 2007, 55, 436–450 CrossRef CAS PubMed.
  71. A. Malej, P. Mozetič, V. Malač and V. Turk, Response of summer phytoplankton to episodic meteorological events (Gulf of Trieste, Adriatic Sea), Mar. Ecol., 1997, 18, 273–288 CrossRef.
  72. T. Kanduč, D. Kocman and N. Ogrinc, Hydrogeochemical and stable isotope characteristics of the river Idrijca (Slovenia), the boundary watershed between the Adriatic and Black seas, Aquat. Geochem., 2008, 14, 239–262 CrossRef.

Footnote

Electronic supplementary information (ESI) available. See DOI: 10.1039/c7em00443e

This journal is © The Royal Society of Chemistry 2018