Determination of Mercury Chemical Speciation in the Presence of Low Molecular Mass Thiols and Its Importance for Mercury Methylation

Home , Mercury sulfide

Determination of mercury chemical speciation in the presence of low molecular mass thiols and its importance for mercury methylation

Van Liem-Nguyen

Department of Chemistry Doctoral thesis Umeå, 2016

This work is protected by the Swedish Copyright Legislation (Act 1960:729)

ISBN: 978-91-7601-469-1

Cover: boreal wetland by Van Liem-Nguyen

Electronic version available at http://umu.diva-portal.org/

Printed by: The service centre in KBC, Umeå University, Sweden 2016

To my family

To my beloved wife

iii

Abstract Methylmercury (MeHg) is a neurotoxic compound that threatens the well-being of humans and wildlife. It is formed through the methylation of inorganic mercury (HgII) under suboxic/anoxic conditions in soils, sediment and waters. The chemical speciation of HgII, including specific HgII species in aqueous and solid/adsorbed phases, plays a key role in MeHg formation. Chemical forms of HgII which have been reported to be available for uptake in methylating bacteria include neutral HgII–sulfide complexes, HgII complexes with specific low molecular mass (LMM) thiols, and nanoparticulate HgS(s). Accurate determination of the chemical speciation of HgII is thus crucial when elucidating the mechanism of MeHg formation. The concentration of HgII–LMM thiols complexes is predicted to be extremely low (sub fM range). Current analytical methods do not allow direct quantification of HgII complexes due to the very low concentration of these complexes, and therefore determination rely on thermodynamic modeling. Accurate stability constants for HgII–LMM thiols complexes and quantification of LMM thiol ligands in environments are thus required to precisely determine the concentration of such complexes. In this thesis, a novel analytical method was developed based on online pre-concentration coupled with liquid chromatography tandem mass spectrometry to determine the concentration of 16 LMM thiols (Paper I). This method was successful in detecting 8 LMM thiols in boreal wetland porewaters, with mercaptoacetic acid and cysteine being the most abundant. The total concentration of individual detected LMM thiols ranged from sub nM (LOD=0.1 nM) to 77 nM. Moreover, the stability constant (β2) for HgII complexes with 15 LMM thiols were directly determined for the first time by competing ligand exchange experiments combined with liquid chromatography ICPMS analysis (Paper II). Values of log β2 for the reaction Hg2+ + 2LMM-RS- = Hg(LMM-RS)2 ranged from 34.6 for. Based on the determined constants of Hg(LMM-RS)2 complexes and state-of-the-art constants from literature for other HgII complexes, we established comprehensive thermodynamic speciation models for MeHg and HgII in boreal wetlands (Paper III). The speciation of HgII was coupled with the HgII methylation rate constant (km) determined with different enriched Hg isotope tracers (Paper IV). There was a good correlation (R2=0.88) between the km determined by a HgII(aq) tracer added as Hg(NO3)2 with high bioavailability and a tracer where HgII was bond to thiol groups in natural organic matter (HgII-NOM(ads)) and has a lower bioavailability. The HgII(aq) tracer was consistently methylated at 5 times higher rate than the HgII-NOM(ads) tracer. A good correlation was observed between the concentration of biologically produced LMM thiols and km in the boreal wetlands. In a mesocosm study of estuarine sediment-brackish water systems, increased concentration of phytoplankton chlorophyll α due to macro nutrient additions led to an increase in HgII methylation rate of the HgII(aq) but not of the HgII-NOM(ads) tracer or ambient HgII species (Paper V). Furthermore, simulated newly deposited HgII species from atmospheric and terrestrial sources were exhibited significantly higher HgII methylation rates when compared with simulated aged sediment HgII pools. Through the development and adoption of novel analytical methods, this thesis reveals the significance of LMM thiols in Hg biogeochemistry by precise determination of HgII–LMM thiol complexes in natural environmental systems.

Table of Contents

List of publications ...... ix Abbreviation...... xi 1. Introduction ...... 1 1.1. Mercury ...... 1 1.2. Chemical species and chemical speciation of mercury ...... 2 1.3. Methylation of HgII ...... 3 1.4. Aims of the thesis ...... 5 2. Materials and experimental approaches ...... 6 2.1. Study sites...... 6 2.1.1. Boreal wetlands ...... 6 2.1.2. Coastal marine system ...... 7 2.2. Experimental approaches ...... 7 2.2.1. Determination of LMM thiols in natural waters ...... 7 2.2.2. Determination of total thiols in natural waters and soils...... 8 2.2.3. Determination of total Hg and MeHg in natural waters and soils ...... 9 2.2.4. Determination of the methylation and demethylation rate constants .. 9

2.2.5. Determination of stability constant for Hg(LMM-RS)2 complexes ..... 10 2.2.6. Chemical speciation of MeHg and HgII in boreal wetlands ...... 10 2.2.7. Mesocosm experiment ...... 11 2.2.8. Mercury isotope tracers and nutrient preparation ...... 11 3. Results and Discussion ...... 13 3.1. Determination of LMM thiols in boreal wetland porewaters (Paper I) ...... 14 3.1.1. Limits of detection of the method ...... 15 3.1.2. Determination of LMM thiols in boreal wetland porewaters ...... 15

3.2. Determination of stability constants for Hg(LMM-RS)2 complexes (Paper II) ...... 16 3.3. Mercury chemical speciation in the boreal wetlands (Paper III) ...... 18 3.3.1. Geochemical properties at the wetland sites ...... 19 3.3.2. Chemical speciation model for MeHg ...... 19 3.3.3. Chemical speciation model for HgII ...... 21 3.3.4. Formation of HgS(s) ...... 23 vii

3.4. Factors controlling MeHg formation (Paper IV &V) ...... 25 3.4.1. Determination of HgII methylation rate constant ...... 25 3.4.2. Effect of microbial activity on labile HgII tracer added ex situ ...... 26 3.4.3. Effect of microbial activity on net methylation in situ ...... 28 3.4.4. Net methylation and demethylation of different Hg pools ...... 30 3.5. Remarks for the future research ...... 31 3.5.1. Determination of Hg chemical speciation on the methylating bacteria surface proximity ...... 31 3.5.2. Experimental approaches for determination of MeHg, and HgII complexes to LMM thiols in pure bacteria culture ...... 33 4. Conclusions ...... 34 5. Acknowledgements ...... 36 6. References ...... 38

viii

List of publications Publications included in this thesis I. Determination of Sub-Nanomolar Levels of Low Molecular Mass Thiols in Natural Waters by Liquid Chromatography Tandem Mass Spectrometry after Derivatization with p‑(Hydroxymercuri) Benzoate and Online Preconcentration Van Liem-Nguyen, Sylvain Bouchet and Erik Björn Analytical Chemistry, 2015, 87, 1089-1096

II. Stability constants for mercury (II) complexes with low molecular mass thiols as determined by competing ligand exchange and liquid chromatography inductively coupled plasma mass spectrometry Van Liem-Nguyen, Ulf Skyllberg, Kwangho Nam and Erik Björn Manuscript in preparation

III. Thermodynamic modelling of the chemical speciation of mercury and methylmercury under sulfidic conditions in boreal wetland soils Van Liem-Nguyen, Ulf Skyllberg, and Erik Björn Manuscript in preparation

IV. Mercury chemical speciation and biological production of low molecular mass thiols in control of methylmercury formation in boreal wetlands Van Liem-Nguyen, Ulf Skyllberg and Erik Björn Manuscript in preparation

V. Effects of nutrient loading and mercury chemical speciation on the formation and degradation of methylmercury in estuarine sediment Van Liem-Nguyen, Sofi Jonsson, Ulf Skyllberg, Mats B Nilsson, Agneta Andersson, Erik Lundberg and Erik Björn Under review in Environmental Science & Technology

Author contribution: Paper I. The author formulated the scientific approach, led the project planning work, performed all of the experimental work and made major contributions to the data evaluation and writing of the paper. Paper II. The author was largely involved in formulating the scientific research objectives and approaches, led the project planning work, conducted most of the experimental work and the data evaluation and was lead author in writing the manuscript. Paper III. The author was largely involved in formulating the scientific research objectives and approaches and in the project planning work, performed most of the experiments and data processing and made major contributions to writing the manuscript. Paper IV. The author was largely involved in formulating the scientific research objectives and approaches and in the project planning work, performed most of the experiments and data processing and made major contributions to writing the manuscript. Paper V. The author was largely involved in data processing and interpretation and made major contributions to the writing of the manuscript.

Abbreviation

2-MPA 2-Mercaptoproprionic acid AMA Advanced mercury analyzer Aq Aqueous Bio Biological Chl α Chlorophyllα Cys L-Cysteine DOC Dissolved organic carbon DOM Dissolved organic matter EXAFS Extended X-Ray absorption fine structure GC Gas chromatography Glyc Monothioglycerol GSH Glutathione GTN Gästern HDPE High density polyethylene Hg All chemical forms of mercury Hg0 Elemental mercury HgII Inorganic divalent mercury IC Ion chromatography ICP Inductively coupled plasma IQ Intelligence quotient

K Stability constant Ka Acid dissociation constant

Kd Distribution between solid/adsorbed and dissolved phases kd Methylmercury demethylation rate constant km Mercury methylation rate constant KSN Kroksjön LC Liquid chromatography LDN Långedalen LMM Low molecular mass LODs Limits of detection xi

MAC Mercaptoacetic acid MeHg All chemical forms of methylmercury MS Mass spectrometry MS/MS Tandem mass spectrometry NACCys N-acetyl cysteine NOM Natural organic matter

Org-SRED Organic reduced sulfur P.P Photosynthetic primary biomass production rate Pen Penicillamine

PHMB p‑(hydroxymercuri) benzoate PP Polypropylene

R2 Correlation coefficient RSH Thiol RSR Organic monosulfide RSSR Organic disulfides S XANES Sulfur X-ray absorption near edge structure S(-II) Inorganic sulfide S0 Elemental sulfur SD Standard deviation sed Sediment SKM Storkälsmyran SPE Solid phase extraction SRB Sulfate reducing bacteria STEB Sodium tetra ethyl borate SUC Mercaptosuccinic acid t48 Time after 48 h from the beginning of the incubation to Time at the beginning of the incubation US EPA United States Environmental Protection Agency UV Ultra violet wt Water

β2 Combined stability constant, β2= K1×K2

xii

1. Introduction 1.1. Mercury Mercury (Hg) is a highly neurotoxic compound, causing negative effects to humans health and to wildlife. Mercury is a naturally occurring metal in the Earth's crust with an average concentration of approximately 0.05 mg kg-1. While the ore of Hg has been found to contain up to 12-14% of naturally occurring Hg, it exists mainly in the form of cinnabar mineral.1, 2 Mercury is present in the atmosphere at low levels (1-170 ng m-3), mostly as elemental mercury (Hgo).3, 4 It can be released into the atmosphere by natural, e.g. volcanic eruptions, and anthropogenic activities such as incineration of fossil fuels, mining and industrial activities.5-7 Globally, industrialization have significantly increased anthropogenic Hg release by a factor of 3 to 5 and have increased chronic Hg exposure to humans as well as aquatic and terrestrial organisms.5, 8 Despite societal undertakings, Hg pollution is still a major environmental issue globally.9, 10

Exposure to high Hg concentration can damage the central nervous system and in worst case cause death.11, 12 This mostly happens for those who work in mines under poor conditions and limited knowledge of health and safety measures.3 Consumption of fish and rice is the main source of Hg in humans.13-16 The first officially discovered Hg pollution incident was in Minamata city, Japan’s Kyushu Island in May 1956.11 This led to the discovery of the “Minamata disease”; a disease of methylmercury (MeHg) poisoning through the consumption of fish and shellfish.11 In March 1992, 2252 patients were officially diagnosed of Minamata disease and 1043 fatalities was recorded.12

In Sweden, Hg levels in piscivorous fish greatly exceed the levels (0.02 mg Hg kg- 1 fresh weight (EU 2008))17 for environmental monitoring target in most of the hundred thousand lakes.18 The Hg concentration in fish 0.2-0.5 mg Hg kg-1 (wet weight) is considered as safety level for fish consumption according to United Nations Environmental Programme (UNEP).19 About 7 % of women of child- bearing age in the USA are estimated to have consumed food with Hg concentrations that exceeds health guideline.3. A recent study concluded that about 1.8 million children are born every year in Europe with Hg concentration exceeding the limit of 0.58 µg g-1 (in hair), which has negative effects on the IQ 1 and that the cost for the society of Hg exposure is 9000 million € per year.20 In Sweden, around 10% of children are born with Hg concentration above the limit (0.58 µg g-1, in hair), and the cost for Hg exposure is 150 million € per year for the Swedish society.20

Natural events and anthropogenic activities emit mostly inorganic forms of Hg, but the organic form, methylmercury (MeHg), is the dominant in fish and rice.16, 21, 22 Methylmercury is much more toxic than inorganic mercury thus in combination with its persistence and bioaccumulating properties the formation of MeHg in the environment is of major importance.

1.2. Chemical species and chemical speciation of mercury According to IUPAC, chemical species is defined as “specific form of an element defined as to isotopic composition, electronic or oxidation state, and/or complex or molecular structure” and the chemical speciation is defined as “distribution of an element amongst defined chemical species in a system”.23 Therefore, the chemical speciation of Hg with respect to redox is the distribution amongst divalent (HgII), monovalent (HgI) and elemental mercury (Hgo). The chemical speciation with respect to molecular complexes composition of inorganic divalent

II II Hg (Hg ) is the distribution of different Hg complexes, e.g. Hg(SH)2, HgCl2 and

Hg(SR)2, in for instance wetland system. Also the term “geochemical pools” of HgII and MeHg is relevant in this thesis and was defined by Jonsson et al.: “geochemical pools of HgII and MeHg as differing with respect to their i) spatial distribution (e.g. buried in sediment or localized at the sediment surface), and/or with respect to their ii) chemical speciation”.24

In the solid/adsorbed phase HgII is mostly present as cinnabar (HgS(s)) form or bound to thiol (RSH) functional groups of NOM. Inorganic divalent mercury is also present in the aqueous phase as strong complexes with soft Lewis bases (ligands) such as inorganic sulfide, thiols and halides (but not fluoride).25, 26Typically, HgII complexes are low coordinated. Two-coordinated complexes are most common followed by three- and four-coordinated.27-30 The linear two- coordination of HgII is caused by relativistic effects and is a very unusual structure for a metal.31 On the other hand, the strong covalent character of the C– Hg bond is the reason for the high stability of the MeHg molecule and MeHg also 2 has similar characteristics as HgII when it comes to the strong complexation with soft Lewis bases.28, 32, 33

Reports on complete chemical speciation of HgII in natural environments, covering both solid/adsorbed and aqueous phases are quite few due to the lack of analytical methods to directly measure specific relevant HgII species at natural concentrations, and the need to rely on thermodynamic modeling. Most studies focus on the chemical speciation of HgII in either soil34 or aqueous (dissolved) phases.35, 36 One reason for this is a significant uncertainty in the stability constants for HgII complexes formed by ligands such as low molecular mass (LMM) thiols, thiols associated with natural organic matter (NOM-RSH) and inorganic sulfides (S(-II)).25, 37 In this work, comprehensive thermodynamic models were constructed for MeHg and HgII, including both the solid/adsorbed and dissolved phases. The models were evaluated by comparing the model results with actual measurements of concentrations of MeHg, HgII and important ligands (particularly LMM thiols) in soil and porewater of the boreal wetlands. Such an evaluation has not been done in these environments before. Even though LMM thiols are present at low concentrations in environments they are believed to play an important role in transformations of HgII.38-40

1.3. Methylation of HgII The transformation of HgII to MeHg (methylation) is mainly the result of biotic processes under suboxic and anoxic conditions. Sulfate reducing bacteria (SRB) have been identified as important methylators.41, 42 Recently, iron reducing bacteria (IRB),43, 44 methanogens,45 and other types of microbes46 have been reported producing MeHg in natural environments. The methylation rates have been suggested to be limited by the amount of HgII available for uptake by methylating bacteria, which is controlled by inorganic sulfide (S(-II)) concentration, natural organic matter (NOM) functional groups, and pH.47 Due to the strong affinity of HgII to reduced sulfur, the chemical speciation of HgII is mostly driven by the relative proportions of S(-II) and thiol (RSH) groups.25, 47Inorganic divalent mercury forms complexes with RSH groups in a two-

27 coordinated structure, e.g.Hg(SR)2. In addition, different geochemical pools of HgII have also been demonstrated to control MeHg production.48 The presence of 3 mackinawite FeS(s) has been shown to decrease MeHg production in wetland slurries due to reduced availability of HgII for methylation.49, 50

0 II It has been shown that Hg(SH)2 (aq) and Hg complexed by specific low molecular mass (LMM) thiols i.e. Hg(LMM-RS)2 are available for uptake by methylating bacteria.36, 38 A recently developed theory suggests that nanoparticulate HgS(s) is indirectly or directly taken up by methylating bacteria, yet the mechanism for the uptake remains to be identified.51 In natural waters, sediments, and wetland environments concentrations of LMM thiols and S(-II) vary largely. Low molecular mass thiols normally range from sub nM levels (LODs) to hundreds nM [Paper I and Manuscript III & IV], but in some cases they may reach µM52, 53 or even mM54 concentrations. Under suboxic to anoxic conditions S(-II) concentrations may vary from 20 nM (LOD)55 to mM.56, 57 Therefore, concentrations of bioavailable HgII complexes may vary substantially depending on the concentrations and molar ratio of LMM thiols and S(-II). The rate of HgII methylation has been shown about 5–6 times higher in the presence of specific LMM thiols e.g. Cys or MAC, as compared to the presence of similar concentration of S(-II).38 Thus, it is of utmost importance to develop relevant thermodynamic models to predict at which molar ratio between S(-II) and LMM thiols, the uptake by methylaing bacteria is mainly driven by HgII–sulfides or by HgII–LMM thiols complexes. Wetlands have seasonally variable water regimes covering oxic-suboxic-anoxic conditions. These frequent seasonal fluctuations turn wetlands into dynamic redox environments and often enhance MeHg formation. This makes wetlands become critical sources of MeHg release, and very important environments for Hg cycling.58, 59 Geochemical pools of Hg have been assumed to be different with respect to their availability for methylation, bioaccumulation, transportation and mobilization.47, 60. Jonsson et al. reported a higher net MeHg formation of HgII newly deposited to sediment, as compared to the aged HgII sediment pool, by use of a mesocosm approach.48 In this thesis work, a similar observation of differentiated availability of HgII geochemical pools for methylation was extended by investigating the effect of nutrient loading on rates of methylation and demethylation.

1.4. Aims of the thesis The aims of this thesis are:

(1) To determine LMM thiols concentration in natural waters as well as to establish stability constants for different HgII–LMM thiols complexes by developing and apply novel analytical methods (Paper I & II).

(2) To determine the concentration of important MeHg and HgII species in soils and porewaters by establishing comprehensive thermodynamic speciation models for MeHg and HgII (Paper III).

(3) To investigate if the chemical speciation of HgII control MeHg formation in boreal wetlands and estuarine sediment environments and control the effect of nutrient loading on HgII methylation (Paper IV &V).

2. Materials and experimental approaches 2.1. Study sites 2.1.1. Boreal wetlands Boreal wetlands have been reported as sources of MeHg formation and transportation.56, 58 Detailed information on the characteristics of the wetlands selected for the study are found in Paper III and in Tjerngren et al.58 Briefly, four different boreal wetlands were selected and studied: two in southern and two in northern Sweden. The southern sites were Långedalen (LDN) and Gästern (GTN), and Storkälsmyran (SKM) and Kroksjön (KSN) represent the northern sites. The map and location of sampling spots are shown in Figure 1 and Table 1.

Figure 1. Maps showing sites and sampling locations (black filled squares). Black stars indicate locations of small dams, and black solid lines denote streams.

Table 1. Sampling locations and properties of the boreal wetlands (information from Tjerngren et al., 2012).58

Central Location Central wetland Catchment Wetlands wetland area Donating vegetation Group type area (km2) Long (N) Lat (E) (ha) and (%) Carex spp., Sphagnum spp., Northern, nutrient SKM 63°57'52.0" 20°38'47.0" Riparian 0.48 0.020 (4.2) Polytrichum spp. poor peatlands Dystrophic lake- Northern, nutrient KSN 63°56'41.6" 20°38'8.7" 1.0 0.26 (26) Carex spp., Sphagnum spp. peatland complex poor peatlands Scirpus spp., Carex spp., Peatland nutrient LDN 58°19'45.7" 12°29'8.1" Fen/bog 2.0 0.106 (11) Sphagnum spp., broad-leaved gradient grasses, Calluna vulgaris. Mesotrophic lake- Phragmites australis, thypa Southern, nutrient GTN 57°27'18.8" 16°35'0.4" 23 0.58 (2.5) peatland complex latifolia, Sphagnum spp. rich wetlands

2.1.2. Coastal marine system A marine sediment was sampled from Öre river estuary. The estuary is located in the Bothnian Sea (site 63° 33.905′ N, 19° 50.898′ E) at the Swedish east coast and covers an area of 50 km2 with a total water volume of 109 m3 and an average depth of 16.4 m. The sediment has a total Hg concentration of approximately 200 pmol g-1 (d.w) which means it is classified as non-contaminated (background site) sediment.

2.2. Experimental approaches 2.2.1. Determination of LMM thiols in natural waters Thiols can be associated with natural organic matter (NOM) or low molecular mass organic molecules and are present in so-called thiol (RSH) or organic disulfides (RSSR) forms. Previous studies mostly reported LMM thiols concentration as the sum of RSH and RSSR forms after adding a reducing agent to convert RSSR to RSH prior to measurement.61, 62 However, the affinity of the RSH is much stronger for most metals (including HgII) as compared to the RSSR form. It is therefore important to determine the concentration of RSH separately from RSSR. In Paper I, we developed a novel analytical method to determine 16 different LMM thiols using solid phase extraction online pre-concentration hyphenated to liquid chromatography tandem mass spectrometry (SPE/LC- MS/MS). The method allows determination of LMM thiols in natural waters at sub nM levels of limits of detection (LODs).

To increase the sensitivity, selectivity and preservation of LMM thiols, p‑(Hydroxymercuri) Benzoate (PHMB) was used as a probe. The PHMB probe is selective and has strong affinity for the RSH groups. It is however an extremely toxic compound,61 and therefore its concentration should be optimized for the derivatization step. A simplified version of the optimization (Paper I) suggests that the minimal concentration of PHMB for the derivatization could be calculated by the following equation:

[PHMB] = 3×([NOM-RSH(aq)] + [S(-II)])

The pH of the boreal wetland porewaters is relatively low, ranging from 4 to 6.

(-II) Consequently, inorganic sulfide (S ) is mostly present as H2S(aq, g), which was evaporated during sample preparation. When dealing with samples containing high (S-II) concentration at neutral or alkaline pH (pH>7), for instance sediment porewater, an acidification step can be used to remove S(-II).

2.2.2. Determination of total thiols in natural waters and soils Total thiol (LMM thiols and thiol associated with NOM) in waters and soils was determined using Sulfur K-edge X-ray absorption near edge structure spectroscopy (S K-edge XANES). For porewater samples, one liter was filtered through a 0.22 µm membrane filter in a glove box filled with N2 gas. This sample was then frozen at -20°C and freeze-dried. As illustrated by the spectrum in Figure S3, Paper III, one peak of reduced organic sulfur and one peak of oxidized

2- sulfur was obtained. The latter peak was highly dominated by sulfate (SO4 ) in the porewater. The reduced sulfur peak includes the signal of the three organic sulfur forms: thiol (RSH), organic disulfides (RSSR) and organic monosulfide (RSR). The absorption energies of those three groups of compounds are close to each other and partly overlapping, and therefore often summed up as organic

63 2- reduced sulfur (Org-SRED). The concentration of SO4 in porewater was also determined by ion chromatography (IC). Based on the measured concentration of

2- 2- SO4 by IC and the relative contribution of Org-SRED and SO4 to the S XANES spectrum, the absolute concentration of Org-SRED was determined.

Prior to S XANES analyses of soil samples, wet soils were freeze-dried and homogenized. The total sulfur in soil samples was determined by LECO analyzer 8 whereas elemental sulfur (S0), extracted from fresh soils, were determined by liquid chromatography with UV adsorption.64 The obtained S XANES spectra were deconvoluted into absolute concentrations of Org-SRED, sulfoxide, sulfonate and sulfate based on the relative normalized XANES signal of the respective peaks for these S functional groups and the total sulfur concentration in the sample (Figure S3, Paper III). Finally, the concentration of RSH groups was calculated as 30% of Org-SRED, based on previously combined Hg LIII-edge EXAFS and S K-edge XANES determinations of HgII bonding to RSH groups in NOM from similar types of boreal soils and waters.27, 63

2.2.3. Determination of total Hg and MeHg in natural waters and soils Total Hg in waters was analyzed by isotope dilution analysis (IDA) combined with mercury cold vapor generation ICPMS while the analysis of total Hg in soils was performed by AMA 254 (LECO). Total MeHg in waters was analyzed by IDA combined with thermal desorption gas chromatography (GC) ICPMS after derivatizing with sodium tetraethyl borate (STEB).4 Prior to analysis, soil samples were spiked enriched MeHg isotope for isotope dilution analysis (IDA). Following the samples were homogenized and MeHg in soils were extracted using

KBr/CuSO4/H2SO4/CH2Cl2 and determined by GC-ICPMS after derivatizing with STEB.65

2.2.4. Determination of the methylation and demethylation rate constants

II The methylation rate constant of Hg (km) and MeHg demethylation rate constant (kd) in soils and sediments were determined using enriched stable isotope tracers. Both km and kd were described by irreversible pseudo first-order kinetic models.66 Experimentally, approximately 10.00 g (with 2 decimal digits) of wet soil/sediment samples were weighed into two sample sets inside the glove

199 204 200 II box. Dissolved Me HgCl(aq) and Hg(NO3)2(aq) (and/or Hg -NOM(ads) adsorbed tracer), corresponding to 10-30 % of the total MeHg and HgII concentrations in the samples were added. The synthesis of 200HgII-NOM(ads) tracer was described by Jonsson et al.67 based on Skyllberg and Drott.68 One

o sample set was immediately frozen at -20 C, designated to, while the other sample set was incubated in a N2 filled glovebox at room temperature in the dark for 48

9 h, designated t48. The incubation was terminated by placing the t48 sample set into a freezer. The determination of MeHg in the incubation study was carried out with the same procedure for the determination of MeHg in soil/sediment.

The km and kd were calculated using the equations below.

[Me204Hg]t – [Me204Hg]t 48 0 (d-1) km = 204 II [ Hg ]addition×2

199 199 -1×(ln[MeHg Hg]t48 – ln[Me Hg]t0) -1 kd = (d ) 2

2.2.5. Determination of stability constant for Hg(LMM-RS)2 complexes

The stability constants f0r 15 Hg(LMM-RS)2 complexes were determined using a competing ligand exchange method combined with liquid chromatography ICPMS. In the first step, iodide (I-) ligand was used as the competing ligand to determine the stability constant for Hg(MAC)2 and Hg(2-MPA)2 due to well- known stability constants for the complexation of HgII with I-.69 In the second step, to enable determination of LMM thiol ligands having different retention time in the LC column, either MAC or 2-MPA was used as a competing ligand when establishing the stability constant for the 13 other Hg(LMM-RS)2 complexes.

2.2.6. Chemical speciation of MeHg and HgII in boreal wetlands Concentrations of MeHg, HgII and important ligands for the determination of MeHg and HgII chemical speciation in soils and porewaters were measured. Those parameters were then used as input to the WinSGW software to calculate the concentration of MeHg and HgII complexes pertaining to solid/adsorbed and dissolved phases. The reactions (1) to (7) were the most significant reactions with strong influence on the chemical speciation of MeHg and HgII in the wetland system.

+ + MeHg + H2S = MeHgSH + H (1) MeHg+ + LMM-RSH = MeHgSR-LMM + H+ (2)

MeHg+ + NOM-RSH = MeHgSR-NOM + H+ (3)

2+ + Hg + 2H2S = Hg(SH)2+ 2H (4)

2+ + Hg + H2S = HgS(s) + 2H (5)

2+ + Hg + 2LMM-RSH = Hg(LMM-RS)2+ 2H (6)

2+ + Hg + 2NOM-RSH = Hg(NOM-RS)2+ 2H (7)

2.2.7. Mesocosm experiment Mesocosm scale experiments are a powerful tool for simulating natural environmental conditions and to control factors which affect, in our case, MeHg formation in marine ecosystems. Our mesocosm experiments were conducted in 6 double-mantled high density polyethylene (HDPE) tanks (5 m × 0.75 m ⌀), located at the Umeå Marine Sciences Centre. Intact sediment cores were manually sampled from the Öre river estuary by divers at the site coordinates 63° 33.905′ N, 19° 50.898′ E and at water depth of 5-7 m using custom-made sampling devices, then inserted into the mesocosm tanks. Solid/adsorbed mercury tracers were injected into the sediments and dissolved mercury tracers were added to the water columns. The tanks were temperature controlled at different heights via an outer glycol layer. The temperature of the water column was controlled in two sections with 15oC in the upper part (above 3.5 m) and 10oC in the lower part (below 3.5 m). The convection of the upper part of the water column was obtained by purging with ~20 mL s-1 air at 3.2 m from the water surface. 150 W metal halogen lamps (Master color CDM-T 150w/942 G12 1CT) was used as light sources with 12/12 hours on/off cycles.

200 II 201 198 The three Hg sediment tracers i.e. Hg -NOMsed, β- HgSsed and Me Hg-

NOMsed were injected to the sediments at 1 cm depth prior to loading of the

204 II brackish water to the tanks. The other two Hg water tracers, Hg wt and

199 Me Hgwt, were then added to the water columns. Nutrients (N, P) were also added to the water columns from moderate to rich nutrient level. Total experimental time was 30 days.

2.2.8. Mercury isotope tracers and nutrient preparation The Hg enriched 196Hg (50%), 198Hg (92.78%), 199Hg (91.95%), 200Hg (96.41%),

201 204 Hg (98.11%) and Hg (98.11%) (as HgO or HgCl2) were purchased from Oak

204 II 199 Ridge National Laboratory (TN, USA). The Hg wt and Me Hgwt tracers were prepared by diluting 204HgII(aq) and MeHg199Hg(aq) in MQ water. The β-

201 HgSsed and isotopically enriched MeHg tracers were synthesized as described

67 70 200 II 198 by Jonsson et al. and Snell et al. The Hg -NOMsed and Me Hg-NOMsed tracers were prepared by adding 200HgII(aq) and Me198Hg(aq) to a homogenized organic soil previously characterized by Skyllberg and Drott.68 The Hg/RSH molar ratio was kept in the range of 1:2 HgII:RSH and 1:1 MeHg:RSH complex

68 - 3- + stoichiometry. Nitrate (NO3 ), phosphate (PO4 ) and ammonium (NH4 ) solutions were prepared from salts of NaNO3, NaH2PO4×H2O and NH4Cl, respectively.

3. Results and Discussion The presence of LMM thiols has, in pure bacteria culture, been demonstrated to enhance methylation rates of HgII. 38, 40 However, the characterization of LMM thiols in natural environments has so far been very limited due to a lack of powerful (high accuracy, low detection limits) analytical methods. Therefore, in Paper I, we focused on the development of a new analytical method for determination of LMM thiols in natural environments. Complexes of HgII with LMM thiols are believed to be present mainly as two-coordinated monodentate

27, 29, 30, complexes i.e. Hg(LMM-RS)2, and to be taken up by methylating bacteria. 38 Many different types of LMM thiols occur in natural environments39, 52, 53 and their effects on HgII methylation rates are different.38 In the literature, not all reported LMM thiols have established stability constants for complex with HgII, and some reported constants also vary substantially. Hence, the aim of Paper II is to fill a gap of knowledge and to establish stability constants for an environmentally important group of Hg(LMM-RS)2 complexes (15 complexes). Paper I and II are essentially analytical chemistry oriented laboratory studies. In a contrast, Papers III, IV and V were more focusing on applying the methods and knowledge generated in the laboratory experiments to study processes in natural environments. The chemical speciation of MeHg and HgII in boreal wetlands was determined by the establishment of comprehensive thermodynamic models for MeHg and HgII in Paper III. The relation between HgII species in porewaters and methylation as well as factors in control of the methylation were addressed in Paper IV. By the use of a mesocosm setup, Paper V investigated the methylation of different HgII sediment pools, and dissolved and solid species of HgII. Further more, the effects of nutrient loading on MeHg formation was also investigated. It demonstrated the roles of HgII availability and methylating bacteria activity for the methylation. The scheme in Figure 2 shows the contribution of each Paper in addressing MeHg formation

Paper I Paper II

LMM thiols Log k Hg(LMM-RS)2

MeHg formation HgII species in boreal wetlands *Solid/absorbed phase of HgII *Effect of nutrient loading Paper III

Relation of HgII species in Paper V porewaters and MeHg formation

Paper IV Figure 2. Schematic aims for each Paper of the thesis work

3.1. Determination of LMM thiols in boreal wetland porewaters (Paper I) Low molecular mass (LMM) thiols are a diverse group of compounds, which may play important roles in aquatic ecosystems even at nM concentrations.52, 53, 71 Methods for determination of LMM thiols are mostly dedicated for biological samples.61, 62, 72 Methods based on liquid chromatography (LC) separation with UV adsorption73, 74 and fluorescence detection75-78 are usually constrained by relatively high limits of detection (LODs) and auto-quenching, respectively. With mass spectrometry, the selectivity and sensitivity for LMM thiols determination are improved. Rao et al.61 and Bakirdere et al.62 reported LODs from 1 to 32 nM for 6 LMM thiols determined by high resolution Orbitrap mass spectrometry. Still, the LODs of those methods are not low enough to cope with natural water matrices. Because of these shortcomings, we developed an analytical method for the determination of LMM thiols in natural waters (can also be applied for biological samples) using SPE online pre-concentration liquid chromatography tandem mass spectrometry (SPE/LC-MS/MS).

3.1.1. Limits of detection of the method By the use of SPE online pre-concentration, the analysis time and running costs were highly reduced when compared to conventional SPE offline pre- concentration method. The LODs of 16 investigated LMM thiols ranged from 0.06 nM to 0.5 nM (Table 3, Paper I). The LOD of each LMM thiol substantially depends on both its molecular mass and hydrophobicity. LMM thiols with higher molecular mass and stronger retention on reversed phased LC column (i.e. more hydrophobic) showed better LODs. This could be explained by the recovery of PHMB-thiol complexes on the HLB oasis SPE online cartridge and the species- specific sensitivity of electrospray ionization MS. The SPE online column covers mixed mode (lipophilic and hydrophilic) interactions but compounds with high hydrophobicity show strong interaction and high recovery on the SPE column. For electrospray ionization MS, a high signal-to-noise-ratio was most likely achieved for PHMB-thiol complexes with high molecular mass. By combining the two characteristics, N-acetyl-penicillamine (the second largest molecular mass, 191 Da, and the most hydrophobic) and mercaptoethanol (the smallest molecular mass, 78 Da) showed the best (0.06 nM) and worst (0.5 nM) LOD, respectively (Table 3, Paper I).

3.1.2. Determination of LMM thiols in boreal wetland porewaters In total eight different LMM thiols were detected in the boreal wetland porewaters, with concentrations in nM range (Table 2). For individual LMM thiol, concentration of separate RSH form ranged from 0.1-77 nM whereas the sum of RSH and RSSR forms gave a range of 0.7-110 nM. Low molecular mass thiols such as Cys, NACCys and GSH are known to be of direct biological origin (biomolecules) while MAC, 2-MPA, Glyc, Pen and SUC have an indirect biological origin, can be formed by addition of sulfide to unsaturated alkyl chain of organic compounds.73, 79-81 Average concentrations of total LMM thiols and reduced LMM thiols in the porewaters of all four studied wetlands were 41 nM and 22 nM, respectively. Concentrations of LMM thiols were highest at the northern lake- wetland site (KSN), and the bog-fen peatland gradient (LDN), with average total LMM thiols concentration of 50 nM. At the northern riparian wetland site (SKM), the average concentration was 40 nM, and 20 nM for the mesotrophic southern

15 wetland (GTN). At the KSN site, Glyc was the most abundant, contributing more than 90% of total LMM thiols. At LDN, SKM and GTN sites, Cys and MAC were the most abundant thiols. A positive relationship could be established between the concentration of LMM thiols of direct biological origin and rates of HgII methylation.

Table 2. The concentration of Σ(disulfides (RSSR) and thiol (RSH) forms), and separate RSH form in parenthesis for LMM thiols in the boreal wetland porewaters.

Samples Σ RSSR and RSH, and (RSH) concentration (nM)

MAC Glyc Cys Pen SUC NACCys GSH 2-MPA Total SKM-1 19 (14) 7.6 (2.7) 0.59 27 (17) SKM-2 29 (23) 4.8 (2.0) 34 (25) SKM-3 40 (15) 4.5 (2.2) 5.7 (1.5) 3.2 54 (19) SKM-4 35 (8.9) 5.1 (1.8) 2.0 1.0 4.6 47 (11) SKM-5 3.1 (1.2) 3.1 (1.2) SKM-6 38 (20) 12 (4.3) 0.55 7.8 (3.9) 2.5 (1.2) 0.86 (0.24) 62 (30) SKM-7 22 (14) 3.5 (3.2) 26 (17) SKM-8 29 (18) 9.2 (4.6) 0.59 (0.22) 39 (23) KSN-1 5.3 (2.3) 50 (30) 2.9 (1.1) 0.8 (0.3) 56 (33) KSN-2 7.1 (3.1) 74 (45) 3.2 (1.0) 1.0 (0.4) 85 (50) KSN-3 47 (31) 2.5 49 (31) KSN-4 28 (10) 1.5 29 (10) KSN-5 22 (16) 22 (16) KSN-6 6.0 (2.4) 21 (12) 2.8 (1.2) 0.62 (0.25) 0.73 (0.38) 30 (16) KSN-7 26 (20) 3.4 0.86 (0.50) 30 (21) KSN-8 9.0 (3.2) 110 (77) 2.8 1.2 (0.87) 120 (84) LDN-1 28 (16) 9.1 (3.1) 0.78 11 (2.8) 0.70 (0.17) 11 (2.2) 61 (24) LDN-2 62 (32) 8.0 (2.4) 1.6 (0.21) 15 (2.8) 15 (2.1) 100 (40) LDN-3 51 (38) 6.3 (2.6) 8.7 (2.3) 1.2 (0.47) 2.8 70 (46) LDN-4 38 (30) 6.6 (2.6) 1.0 (0.63) 0.88 47 (34) LDN-5 12 (9.2) 2.1 (1.5) 0.51 15 (11) LDN-6 16 (13) 2.9 2 (1.5) 21 (15) LDN-7 14 (8.5) 2.5 (2) 1.5 (1.5) 18 (12) LDN-8 10 (6.9) 3.2 13 (6.9) LDN-9 6.8 (5.0) 2.8 0.42 0.6 11 (5.0) LDN-10 6.5 (2.1) 1.5 0.52 (0.35) 8.5 (2.5) GTN-1 10 (3.0) 2.3 (0.92) 0.71 (0.45) 2.8 (1.2) 13 (4.3) GTN-2 9.2 (5.4) 3.1 (1.2) 0.71 (0.58) 13 (7.2) GTN-3 15 (3.2) 5.2 (1.6) 0.73 1.8 (0.80) 3.2 25 (5.6) GTN-4 15 (10) 3.3 (0.92) 0.64 (0.43) 19 (11) GTN-5 20 (12) 2.5 (1.1) 0.84 (0.25) 23 (13)

3.2. Determination of stability constants for Hg(LMM- RS)2 complexes (Paper II) It is a challenge to determine stability constants for the very stable HgII–LMM thiol complexes. The wide range of stability constants reported for these complexes before 1980 was extensively discussed at by Casas and Jones.82 By applying an direct determination of very low concentration of free Hg2+ ions, they reported the log β2 for Hg(Cys)2, Hg(Pen)2 and Hg(MAC)2 to be in the range of 16

40.0 to 45.0.82, 83 In later works, addition of a competing ligands such as iodide (I-) or bromide (Br-) ion or the lipophilic thiol dithizone shifted the HgII speciation towards a separable, measurable and identifiable entity.30, 84-86 Even with this approach, there are still uncertainties in stability constants reported, as

30 86 exemplified by log β2 for Hg(Cys)2 complex, ranging between 38.2 and 43.5 . Therefore, there is still a substantial uncertainty remaining before a consensus on stability constants for HgII complexes with LMM thiols can be reached. It has been established that Hg(LMM-RS)2, Hg(LMM-RS)3, and Hg(LMM-RS)4 complexes may all form depending on pH and the HgII/RSH molar ratio.29, 30 At a pH lower than 6, Hg(LMM-RS)2 is the most predominant form, while the proportions of Hg(LMM-RS)3 and Hg(LMM-RS)4 increase at higher pH as demonstrated for the complexation of HgII with Cys, Pen and NACCys.29, 30, 87, 88 For larger LMM thiols like GSH, the formation of higher coordination numbers

II than two is less common. At pH of 7.4, Hg(GSH)2 contributes 95% of all Hg complexes.29 This could be explained by steric hindrance effects of large ligand molecules, limiting the coordination number of the complex. In line with this, spectroscopic methods such as EXAFS and NMR indicate a high dominance of

II the Hg(NOM-RS)2 structure in the complexation of Hg with NOM associated RSH (NOM-RSH) in soil and natural water at pH below 7.27, 30

The concentration of Hg(LMM-RS)2 complexes in the environment is estimated to be extremely low, normally below fM range, in porewaters.89 Current analytical methodologies do not allow direct detection of such complexes. Therefore, the determination of Hg(LMM-RS)2 still needs to rely on thermodynamic modeling and correct stability constants for Hg(LMM-RS)2 complexes are obviously crucial. In this study, we determined stability constants for 15 different Hg(LMM-

RS)2 complexes using a novel analytical method based on competing ligand exchange (CLE), liquid chromatography separation and ICPMS detection. Determined stability constants for the 15 investigated complexes are reported in Table 3. When writing the reaction without considering the protonation of RSH group (reaction 1a) the log K of Hg(LMM-RS)2 range from the lowest value of

19.6 to the highest value of 21.0 corresponding to Hg(Cyst)2 and Hg(2-MPA)2, respectively. For the reaction 1b (Table 3), where the pKa value of the RSH groups is included in the reaction (log β2= log K + 2pKa), the log β2 showed a 17 wider range from the lowest value of 34.6 for Hg(CysGly)2 to highest value of 42.1 for Hg(3-MPA)2. The pKa value of the RSH group is important for the determination of β2, they were thus calculated by the Atomic Charge model (quantum mechanics model) reported by Ugur et al.90 as shown in Paper II.

Table 3. Stability constants for Hg(LMM-RS)2 complexes reported in agreement with reaction 1a and 1b and pKa values for RSH groups (reaction 2).Complexes are sorted from highest to lowest log K for reaction (1a) at an ion strength of zero (I=0) M and pH=3.0.

Complexes Stability constant (± SD)

log K (1a) log β 2 (1b) pKa (2)

Hg(2-MPA)2 21.0 41.5 (±0.1) 10.27

Hg(NACPen)2 20.9 40.1 (±0.2) 9.59

Hg(NACCys)2 20.6 40.2 (±0.2) 9.79

Hg(SUC)2 20.6 41.7 (±0.2) 10.55

Hg(3-MPA)2 20.6 42.1 (±0.2) 10.76

Hg(MAC)2 20.6 40.9 (±0.2) 10.16

Hg(Glyc)2 20.6 39.4 (±0.2) 9.42

Hg(GluCys)2 20.5 40.3 (±0.3) 9.92

Hg(ETH)2 20.5 40.3 (±0.2) 9.89

Hg(Pen)2 20.3 36.9 (±0.2) 8.30

Hg(GSH)2 20.2 38.8 (±0.2) 9.28

Hg(Cys)2 20.1 37.5 (±0.2) 8.64

Hg(CysGly)2 19.9 34.6 (±0.3) 7.34

Hg(HCys)2 19.7 39.4 (±0.2) 9.87

Hg(Cyst)2 19.6 40.3 (±0.2) 10.33

2+ + (1a) Hg + 2RSH = Hg(SR)2 + 2H 2+ - (1b) Hg + 2RS = Hg(SR)2 - + (2) RSH = RS + H

3.3. Mercury chemical speciation in the boreal wetlands (Paper III) The biogeochemistry of Hg in the environment is complex. A multitude of parameters can affect the distribution of Hg species, control their solubility and subsequent availability for methylation, demethylation and bioaccumulation.48, 63, 68, 89, 91 18

In the 1990s Morel et al.92 gave an overview of the chemical speciation of Hg in aquatic and terrestrial environments but they did not include the role of LMM thiols and NOM-RSH. Later studies revealed the important role of NOM and elucidated that Hg has a strong bond with NOM via complexation with RSH groups.27, 63, 93-95 Hitherto, there are very few studies including LMM thiols in Hg chemical speciation due to lack of stability constants for HgII–LMM thiol complexes and relevant analytical methods for determination of LMM thiols concentration in natural environments. Although LMM thiols are present at a low concentration of sub nM to 200 nM39 (in some cases can up to µM52, 53 and mM54 ranges) in natural environments, they are however believed to play an important role in transformation of Hg.38-40, 54

3.3.1. Geochemical properties at the wetland sites The boreal wetland porewaters examined in this study showed a typically low pH, ranging from 4.17 to 6.06 with an average of 5.01. Due to the low pH of the

(-II) porewaters, H2S(aq, g) was the dominant species of inorganic sulfide (S ), with a concentration range from below limit of detection (LOD = 0.3 µM) to 51 µM. The average concentration of HgII in soils was similar for the three wetlands SKM, KSN and GTN, about 470 nmol kg-1, and almost double at the fourth site: LDN, with an average of 900 nmol kg-1. The average MeHg concentrations were highest at LDN 90 nmol kg-1 , followed by SKM 47 nmol kg-1, KSN 19 nmol kg-1, and GTN 10 nmol kg-1. The NOM-RSH(aq) concentration in the porewaters ranged from 0.84 µM to 9.1 µM with an average of 3.8 µM. The sum of concentrations of the 8 detected LMM thiols (MAC, Cys, Pen, NACCys, GSH, Glyc, 2-MPA and SUC) ranged between 1.2 nM to 84 nM with an average of 22 nM. In the solid/adsorbed phase of soils, the concentration of NOM-RSH(ads) was about 800 times higher than NOM-RSH(aq) in the aqueous phase. Elemental sulfur (S0) was detected in all soil samples, with an increasing average concentration in the order KSN (12 µmol kg-1)

3.3.2. Chemical speciation model for MeHg The input for MeHg chemical speciation model included pH, concentrations of

- II 0 Cl , Hg , MeHg, NOM-RSH and LMM thiols, S and H2S(aq) in soil and

19 porewater compartments. The fit of MeHg model was tested by comparing measured and modeled concentrations of MeHg in porewaters. In the literature, stability constants for the formation of MeHgSR-NOM complexes were reported to vary between 15.6 and 17.5 and pKa of the RSH group to vary between 8.5 and 10.18.32, 33, 63 Therefore, during fitting of our model, the log K of MeHgSR-NOM complex and the pKa of RSH group were allowed to vary between 16.0 and 17.5 and between 8.5 and 10.0, respectively. The best fit between measured and modeled concentration of MeHg in the dissolved phase was achieved with a log K of MeHgSR-NOM of 17.5 and a pKa of 9.0 for the RSH group as shown in Figure 3. The linear relationship between modeled and measured data had a slope of 0.77 (ideally it should be 1.0) and a coefficient-of-determination (R2) of 0.75 (n=31), which can be considered an excellent fit considering uncertainties in measurements and lack of true equilibrium in natural systems.

Due to high abundance and strong affinity for RSH groups, more than 99% of MeHg was predicted to bind to thiol associated with NOM i.e. MeHgSR- NOM(ads) in the solid/adsorbed phase, with a log concentration of -8.5±0.5 M. In porewater, the average concentration of MeHgSR-NOM(aq) was ten times higher than the average concentration of MeHgSH0(aq), with log concentration of -11.4±0.4 M and -12.3±0.9 M, respectively. Due to low concentration of LMM thiols as compared to NOM-RSH(aq), the concentration of MeHgSR-LMM(aq) was about a factor of 100 lower than MeHgSR-NOM(aq), with a log concentration of -13.9±0.7 M (Figure S5, Paper III).

Log modeled MeHg (M) -9.5 -12.5 -11.5 -10.5 -9.5

High H2S(aq): 20-60 µM

Medium H2S(aq): 1-20 µM

) M

Low H S(aq): <1 µM ( 2

-10.5

d e

y = 0.77x -2.8 r u

R² = 0.75 s a

-11.5 e

o L

1:1 -12.5

Figure 3. Relationship between modeled and measured solubility of MeHg. The log K for the formation of MeHgSH0(aq) was 14.6 and MeHgSR-NOM was 17.5. The pKa values of

H2S(aq) and RSH group was 7.0 and 9.0, respectively.

3.3.3. Chemical speciation model for HgII Model fits for HgII solubility were evaluated by comparing modeled and observed

II -1 distribution of Hg between solid/adsorbed and aqueous phases (log Kd, L kg ). Inorganic divalent mercury, in contrast to MeHg, forms a solid phase i.e. HgS(s). The formation of HgS(s) has strong influence on the chemical speciation of HgII,

II its solubility and availability for Hg transformation processes. The log Kd is thus considered an appropriate parameter to evaluate the HgII model fit.

Schwarzenbach and Widmer reported stability constants (log K) for the

0 96 formation of Hg(SH)2 (aq) and HgS(s) of 38.6 and 37.6, respectively. Recently, Drott et al. determined those stability constants and reported a log K of 36.8 for the formation of crystalline metacinnabar (β-HgS(s)) and a log K of 39.1 for the

0 37 formation of the aqueous complex (Hg(SH)2 (aq)). The best fit (merit-of-

0 fit=0.041) was obtained with log K of 38.6 for the formation of Hg(SH)2 (aq) and 37.3 for the formation of HgS(s) as shown in Figure 4. Hence, the best fitted log K value for the formation of HgS(s) was 0.3 units below the constant reported by

Schwarzenbach and Widmer96 and 0.5 units above the constant reported by Drott

37 0 et al. The constant for the formation of Hg(SH)2 (aq) was in agreement with the value reported by Schwarzenbach and Widmer.96 With these constants, the average log Kd for observation (4.04±0.35) and prediction by the model (3.97±0.68) were in good agreement. The influence of log K for the formation of

Hg(NOM-RS)2 was also illustrated by varying the log K value from 38.0 to 42.0. The results showed that the log K of 40.0 (and pKa = 9.0) gave a better fit than either lower or higher stability of Hg(NOM-RS)2 complex. This is in excellent agreement with the stability constant of Hg(NOM-RS)2 previously reported by Skyllberg.25

Although the average modeled and observed log Kd are in a good agreement, the model showing the best merit-of-fit (Figure 4) predicted a larger variability in

II Hg solubility, largely driven by the H2S(aq) concentration, than what was observed by measurements. One possible explanation for this could be that colloidal Hg is passing through the filter membrane (0.22 µm) used to filter the porewater samples. It has been shown that nanoparticulate HgS(s) is preserved in the presence of LMM thiols and NOM-RSH(aq) by stabilizing clusters of HgS and thus inhibit aggregation and crystallization.97-99

6 1:1

)

1 -

g 5

(

g 4

d 3

s a

e Low H2S(aq): <1 µM

M 2 Medium H2S(aq): 1-20 µM

High H2S(aq): 20-60 µM 1 1 2 3 4 5 6 II -1 Modeled Hg log Kd (L kg )

Figure 4. Observed and modeled distribution of HgII between solid/adsorbed and aqueous phases (log Kd, L kg-1). The log K values for the formation of Hg(SH)20(aq),

HgS(s) and Hg(NOM-RS)2 are 38.6, 37.3 and 40.0, respectively.

3.3.4. Formation of HgS(s) HgS(s) is formed when the ion activity product [H+]/([Hg2+]×[HS-]) exceeds the value of the stability constant for the reaction Hg2+ +HS-= HgS(s) + H+. Hence, free concentrations of the Hg2+ ion and inorganic sulfide together with pH control the formation of HgS(s). Essentially, the formation of HgS(s) is controlled by two processes: (1) the competition between dissolved aqueous HgII–sulfides complexes and HgS(s) at high S(-II) concentration, and (2) the competition between the complexation of Hg2+ with NOM-RSH(ads+aq) and HgS(s) at low S(- II) concentration. As illustrated in Figure 5, more than 10 mM of S(-II) is required

0 for Hg(SH)2 (aq) to outcompete and inhibit the formation of HgS(s). In the type of wetland soils investigated in this study, S(-II) concentrations never reach 10 mM and the formation of HgS(s) is thus never controlled by dissolved HgII–sulfides species. When it comes to the competition with NOM-RSH, three scenarios are illustrated in Figure 5 based on the measured concentrations of NOM-RSH(ads) in the wetland soils. The first scenario covers the lowest measured concentration

(0.26 mmol L-1), the second covers an intermediate (the average of all concentrations = 2.6 mmol L-1) and the third comprises the highest observed concentration (7.1 mmol L-1). At the lowest concentration of NOM-RSH(ads), the formation of HgS(s) is initiated at 6 nM and equal amounts of HgS(s) and

(-II) Hg(NOM-RS)2(ads) are reached at 10 nM of S . At the average NOM-RSH(ads) concentration the threshold for HgS(s) formation is 0.5 µM and equal amounts of

(-II) HgS(s) and Hg(NOM-RS)2(ads) are reached at 0.9 µM S . At the highest concentration of NOM-RSH(ads) at least 4 µM of S(-II) is required to initiate the formation of HgS(s) and equal amounts of HgS(s) and Hg(NOM-RS)2(ads) are reached at approximately 7 µM of S(-II). This means that depending on the NOM content in wetland soils, the threshold for HgS(s) formation may vary between 6 nM and 4 µM of S(-II).

-6.5

) 1

HgS(s) (a) -

HgS(s) (b) o

m (

-7.0 n

HgS(s) (c) o

a r Hg(NOM-RS) (ads) (a) t

2 n e

-7.5 c Hg(NOM-RS) (ads) (b) n

2 o

s e Hg(NOM-RS) (ads) (c) i

2 c

e p

-8.0 s

0 I

Hg(SH)2 (aq) (a, b, c) I

o L -8.5 -9 -8 -7 -6 -5 -4 -3 -2 -1 (-II) Log S concentration (M)

Figure 5. Concentration of HgII species in solid/adsorbed phase per liter of porewater, as a function of S(-II) concentrations in boreal wetlands. The HgII concentration and pH value were kept constant at the average measured concentrations of 82 nmol L-1 and 5.0, respectively. The NOM-RSH(ads) concentration was varied from the low to high measured concentration in the wetland soils: (a) low (0.26 mmol L-1), (b) intermediate (2.6 mmol L-

1), and (c) high (7.1 mmol L-1). The intersections between HgS(s) and Hg(NOM-RS)2(ads) demonstrate equal amounts of those species. Lines representing HgS(s) begin at the threshold point for the initiation of HgS(s) precipitation. The log K values for the formation of Hg(SH)20(aq), HgS(s) and Hg(NOM-RS)2(ads) are 38.6, 37.3 and 40.0, respectively.

3.4. Factors controlling MeHg formation (Paper IV &V) 3.4.1. Determination of HgII methylation rate constant Dissolved HgII (HgII(aq)) enriched isotope are typically used as tracers for the

II 36, 56, 60, 66 determination of Hg methylation rate constant (km). In the environment, HgII is predominantly associated with NOM in the solid/adsorbed phase and/or occurs as HgS(s). These forms of HgII are less available for the methylation as compared to dissolved HgII(aq) tracers.48, 67 The methylation of added dissolved HgII(aq) tracers is somewhat overestimating the methylation capacity of the environments. In this study, two HgII tracers: 198HgII(aq) and

200 II 198 II Hg -NOM(ads) were used to determine the km. The Hg (aq) tracer was prepared from 198HgII-nitrate in water while 200HgII-NOM(ads) was prepared by equilibrating 200HgII-nitrate with natural organic matter from an organic soil.68 The 200HgII tracer was demonstrated to be bound to RSH groups of NOM as

II Hg(NOM-RS)2(ads) complex due to the strong affinity of Hg for RSH functional groups.26, 27

2 Figure 6 shows an excellent correlation (R = 0.88) for the km determined by the two tracers, with approximately 5 times higher methylation rate of 198HgII(aq) as compared to 200HgII-NOM(ads). This result is in an agreement with results from previous studies of an estuary sediment.48, 67 A strong positive correlation between 198HgII(aq) and 200HgII-NOM(ads) tracers indicates that the 200HgII- NOM(ads) tracer is possibly useful for the determination of ambient HgII methylation in natural environments in which Hg(NOM-RS)2 complexes dominate. It also shows that the 198HgII(aq) tracer gives results which can be used to compare relative methylation rates for different sites/samples/conditions even if the absolute rates are overestimated in comparison to methylation rates of ambient HgII in the system.

Figure 6. Correlation between the methylation rate constants (km) determined for dissolved labile 198HgII(aq) and adsorbed 200HgII-NOM(ads) tracers.

3.4.2. Effect of microbial activity on labile HgII tracer added ex situ As mentioned in part 3.1.2, microorganisms excrete biological (Bio) LMM thiols.61, 62, 100 If they are formed as a consequence of metabolism, concentrations of Bio-LMM thiols can be expected to correlate positively with microbial activity. If so, Bio-LMM thiols may be used as proxy for potential MeHg formation within a wetland site. In the boreal wetlands (Paper IV), we observed a positively significant correlation between total concentrations of Bio-LMM thiols and km with R2 coefficient greater than 0.72 for samples taken at the wetlands LDN, GTN and SKM sites (Figure 7). The fourth site: KSN had a very low concentration of direct Bio-LMM thiols, but showed high concentrations of Glyc which was most likely originating from both direct and indirect biological processes.101 With an assumption that Glyc belongs to the group of Bio-LMM thiols, total LMM thiol concentration at KSN also showed a fairly strong positive correlation with km (Figure 7, R2= 0.41). Within the sampling site, the methylation rate in the sample with presumably highest microbial activity is 8, 10, 30 and 70 times higher as

26 compared to the lowest one in SKM, KSN, GTN and LDN sites, respectively (Figure 7).

0.06 SKM 0.08 KSN

0.05 R² = 0.72

0.06 R² = 0.41

) 1

- 0.04

(

m K 0.03 0.04

0.02 0.02 0.01

0 0 0 5 10 15 20 0 50 100 150 Total Bio-LMM thiols (nM) Total LMM thiols (nM)

0.1 0.07 LDN GTN 0.06 R² = 0.72 R² = 0.80 0.08

0.05 )

1 0.06 -

d 0.04

(

m K 0.03 0.04 0.02 0.02 0.01

0 0 0 10 20 30 40 0 5 10 15 Total Bio-LMM thiols (nM) Total Bio-LMM thiols (nM)

Figure 7. Relationship between the methylation rate constant (km) of the 198HgII(aq) tracer and LMM thiols concentration in the four boreal wetlands.

In the estuarine mesocosm-scale model ecosystem (Paper V), the microbial activity (or biomass) of the pelagic system was increased following nutrient loading to the water column, as reflected by the increase in chlorophyll α (Chl α) and the rate of photosynthetic primary biomass production (P.P) as shown in Figure 8a. The sediment samples were collected at different time points to determine km. Following the increase of microbial activity, km also increased and reached a maximum at the condition with highest microbial activity in Figure 8b.

The km value was approximately 4 times higher at the highest microbial activity conditions102 as compared to the moderate conditions103. A significant correlation

2 between km and Chl α concentration (R =0.96) was observed. After day 22 of the 27 experiment Chl α and P.P decreased, despite the maintained nutrient loading. This may be explained by the growth rate of predatory zooplankton which commonly increases subsequent to phytoplankton growth, generating a predator prey‒dynamics.104

The effect of microbial activity on km in the wetlands (Paper IV) and in the estuarine sediment (Paper V) point towards the conclusion that bacterial activity limits the methylation rate of HgII species being a high bioavailability.

Figure 8. (a) Increase of chlorophyll α (Chl α) and the rate of photosynthetic primary biomass production (P.P) in response to increased nutrient (N, P) addition. (b)

Methylation rate constant (km) of 199HgII(aq) in the sediment samples as determined at different time during the experiment.

3.4.3. Effect of microbial activity on net methylation in situ The MeHg/HgII molar ratio is frequently used to investigate the net MeHg formation in natural environments.48, 56, 60 In contrast to a significant correlation between km and Bio-LMM thiols concentration in the SKM, LDN and GTN sites (Figure 7), the MeHg/HgII molar ratios showed a weak correlation (non- significant, R2=0.18) to the Bio-LMM thiols concentration as shown in Figure 9. Considering the fact that HgII is predominantly associated with thiol groups in

NOM in the solid/adsorbed phase and as an solid phase: HgS(s),25, 26 it is much less available for methylation when compared to dissolved labile HgII(aq) tracers.48, 67

Figure 9. Correlation between total biological (Bio) LMM thiol concentrations in porewaters at the SKM, LDN and GTN wetland sites and MeHg/HgII molar ratios in the wetland soils.

199 II In the studied estuarine sediment (Paper V), km of the dissolved labile Hg (aq) tracer increased 4 times at the maximum nutrient conditions. In contrast to this, the increase in MeHg/HgII molar ratio of the 200HgII-NOM(ads) tracer and ambient HgII was not significant following increased nutrient addition (Figure 10). The different response in HgII methylation of the 200HgII-NOM(ads) tracer and ambient HgII, as compared with the HgII(aq) tracer, is explained by the different availability of HgII from these three sources for methylation. The 200HgII-NOM(ads) tracer was pre-equilibrated as an organic soil slurry in degassed MQ water for one week prior to the experiment and200HgII is expected to be strongly bind to RSH groups in the NOM, reducing its availability for methylation. Ambient HgII is also expected to predominantly present as solid/adsorbed species i.e. Hg(NOM-RS)2(ads) and HgS(s).

The results of MeHg formation ex situ and in situ supported the conclusion that MeHg formation in the environment is driven by two factors: (1) the activity of methylating bacteria, and (2) the availability of HgII for methylation. It however remains to be clarified whether there is a breaking point at which one of the two factors take over the control in natural environments.

Nutrient 200HgII-NOM(ads) 0.04

Ambient Hg 150

i 120

t 0.03

)

g o

90 µ

(

I t

I 0.02

/ t

g 60

N e

M 0.01 30

0.00 0 0 5 10 15 20 25 30 35 Time (d)

Figure 10. Responding of MeHg/HgII molar ratios in the sediment according to nutrient loading for 200HgII-NOM(ads) and ambient Hg.

3.4.4. Net methylation and demethylation of different Hg pools Different Hg tracers were added to the sediment at 1 cm depth below the surface

201 200 II 198 (β- HgSsed, Hg -NOMsed, Me Hg-NOMsed) and to the water column

204 II 199 201 200 II 198 ( Hg wt and Me Hgwt). The β- HgSsed, Hg -NOMsed and Me Hg-NOMsed tracers represent pools of HgS(s) and organically complexed HgII and MeHg in

204 II 199 the sediment, whereas Hg wt and Me Hgwt represent the sum of atmospheric deposition and terrestrial run off for HgII and MeHg. The average net methylation as reflected by the MeHg/HgII molar ratio, was significantly different for the different HgII tracers (ANOVA, p<0.05). The highest value was obtained

204 II 200 II for the Hg wt, 0.041 ± 0.013, intermediate for Hg -NOMsed, 0.020 ± 0.004

201 and lowest for β- HgSsed tracer, 0.004 ± 0.006 (Figure 2a, Paper V). These 30 results indicate a high availability for methylation of HgII deposited in the water phase.

For the two MeHg tracers, the net demethylation was significantly lower for the

198 199 Me Hg-NOMsed compared to Me Hgwt, as manifested by a higher average

II 198 MeHg/Hg molar ratio of 0.34 ± 0.07 for Me Hg-NOMsed compared to 0.14 ±

199 0.04 for Me Hgwt (Figure 2b, Paper V). This could be explained by longer and

199 more intense light exposed to the Me Hgwt (in water column) compared to the

198 105, 106 Me Hg-NOMsed (at 1 cm sediment depth). Although the demethylation

199 198 rate of Me Hgwt was higher than Me Hg-NOMsed, it has been shown that the

199 198 48 bioaccumulation factor of Me Hgwt is much higher than Me Hg-NOMsed.

To summarize, HgII in aquatic ecosystem originated from atmosphere deposition and terrestrial run off shows a higher availability for net MeHg formation and bioaccumulation48 than the sediment pools of HgII and MeHg. This finding strongly points at processes and activities which contribute to an overall release of Hg into aquatic environments, such as e.g. fossil fuel burning, gold mining, forest clear cut, and agriculture harvesting could contribute to an overall release of Hg into aquatic environments, thus having deteriorating effects on the aquatic ecosystem food web.6, 107-110

3.5. Remarks for the future research 3.5.1. Determination of Hg chemical speciation on the methylating bacteria surface proximity Despite the extensive efforts dedicated towards the elucidation of the mechanism of the MeHg formation over the past decade, factors and processes that control MeHg in soils and waters are still not fully understood. Both passive35 and active38 uptake mechanism have been suggested for the uptake of available

0 Hg(SH)2 (aq), and specific Hg(LMM-RS)2 species by methylating bacteria. Recently, nanoparticulate HgS(s) has also been reported available for uptake of methylating bacteria but the mechanism has not been fully described.51 The boreal wetlands study (Paper IV) showed that neither concentrations of

0 Hg(SH)2 (aq) nor any specific Hg(LMM-RS)2(aq) complexes in bulk wetland

II porewaters were significantly correlated to MeHg formation (km and MeHg/Hg molar ratio). It is noteworthy that bacteria (including methylating bacteria) 31 excrete LMM thiols and/or sulfide (sulfate reducing bacteria (SRB)).41, 42, 111 The chemical speciation of HgII on the surrounding bacterial surface is thereby likely very much different from bulk solution in terms of LMM thiols and/or sulfide concentration. In a recent study on biofilm, Leclerc et al. indeed reported up to 3 orders of magnitude greater concentration of extracellular LMM thiols inside biofilms as compared to the bulk water column.100 The aqueous environment surrounding methylating bacteria can be expected to control the HgII uptake by changing the chemical speciation of HgII as illustrated in Figure 11.

It is however challenging to isolate and quantify concentrations of the chemical species at this small scale and low concentrations. The definition for the bacterial surface vicinity is not given, this is however expected to be correlated with bacterial dimension. Techniques such as Raman spectroscopy, Fourier transform Infrared Spectroscopy (FTIR), transmission electron microscopy (TEM), and synchrotron are suggested for exploring the issues.

Figure 11. Illustration for HgII chemical speciation at the near surface environment of HgII methylating bacterium. The figure illustrates for two different types of methylating bacteria that are iron reducing bacteria (IRB, green zone) and sulfate reducing bacteria (SRB, yellow zone). Methylating bacteria can produce both LMM thiols and inorganic sulfide. Based on the properties of both bacteria strains and in order to simplify the illustration, LMM thiols (Cys as representative) and (S(-II)) are assumed to be excreted by IRB and SRB, respectively.

3.5.2. Experimental approaches for determination of MeHg, and HgII complexes to LMM thiols in pure bacteria culture In order to improve understanding on the mechanism of MeHg formation, the actual chemical forms of HgII and MeHgII must be experimentally determined to elucidate the HgII uptake and the MeHg secretion processes. We are therefore developing robust novel analytical method for determination of such complexes based on liquid chromatography (LC) and mass spectrometry (MS) detection.

For the determination of MeHg-LMM thiol complexes, preliminary results show that 8 different MeHg-LMM thiol complexes were detected in pure Geobacter sulfurreducens pca culture experiment by amendment of 100 nM of HgII(aq) into the growing medium with the bacteria optical density of 0.09-0.1 at 660 nm wave length. The LODs for 14 investigated MeHg-LMM thiol complexes range from pM to low nM concentration. This indicates that the method can be applied to study the transformation of HgII in bacteria culture with amended HgII concentration relevant to natural environments as well as the bioaccumulation of different MeHg-LMM thiol complexes in phytoplankton.

For the determination of HgII-LMM thiol complexes, 15 different complexes have been successfully separated and detected by LC coupled to ICPMS. Detection limits in the sub nM range is expected for the optimized method.

4. Conclusions This thesis developed new techniques for the determination of LMM thiols in the natural environments and stability constants of relatively large number of Hg(LMM-RS)2 complexes (15 complexes). By using the same methodology, the relative binding strength of different LMM thiols to HgII are directly compared. We contributes some important aspects to improve the understanding on MeHg formation in the natural environment.

 LMM thiols are frequently detected in boreal wetlands and their concentration. In the boreal wetlands 8 different LMM thiols (Cys, MAC, GSH, NACCys, SUC, GluCys, 2-MPA, Glyc) were detected out of which Cys and MAC were the most abundant with the concentration of individual LMM thiols ranging from 0.1 nM (LOD) to 130 nM. The developed analytical method presented (as presented in Paper I) can be applied to various type of natural waters such as freshwater, wetland porewater, marine and lake sediment porewaters.

 Thermodynamically stable for complexes between HgII and LMM thiols

This thesis also reports more precise stability constant for Hg(LMM-RS)2 complexes based on direct measurement of complexes. The stability constant for HgII complexation with LMM thiols as expressed by the reaction Hg2+ + LMM-

- RS = Hg(LMM-RS)2 ranges from 34.6 for Hg(CysGly)2 to 42.1 for Hg(3-MPA)2 for 15 investigated LMM-RSH compounds. The Hg(LMM-RS)2 complexes can therefore be classified as thermodynamically very stable. The stability constants for 15 frequently reported LMM thiols complexed by HgII are presented in Paper II.

 Concentration of MeHg species in boreal wetlands The research reports MeHgSR-NOM(ads) as the most abundant chemical form of MeHg with an average concentration of 5 nmol kg-1. In the aqueous phase, the concentration of MeHgSR-NOM(aq) dominates with an average of 10 pM, followed by MeHgSH0(aq) of 3 pM and MeHgSR-LMM(aq) of 0.03 pM.

 Concentration of HgII species in boreal wetlands In the solid/adsorbed phase, the average amounts of HgS(s) and Hg(NOM-

-1 RS)2(ads) are 60 and 30 nmol kg , respectively. In the aqueous phase, the

0 average concentration of Hg(SH)2 (aq) is the dominant with 16 pM. Average

- concentrations of HgSnSH (aq)and Hg(NOM-RS)2(aq) are 2 pM and 0.1 pM, respectively. Due to the very low concentration of LMM thiols in bulk porewaters, the average log concentration of Hg(LMM-RS)2(aq) complexes are determined to be as low as -19 M.

 Conditions under which HgS(s) form in boreal wetlands The formation of HgS(s) depends on the concentrations of HgII, NOM-RSH(ads) and S(-II). At the average concentration of HgII and NOM-RSH(ads) in the boreal wetlands in this study of 82 nmol L-1and 2.65 mmol L-1, respectively, HgS(s) begins to form at an S(-II) concentration of 0.5 µM. When S(-II) concentration

0 exceeds 30mM, the Hg(SH)2 (aq) complex constrains the formation of HgS(s) but such high S(-II) concentration is not present in most boreal wetland porewaters.

 Factors controlling MeHg formation MeHg formation is driven by two factors: (1) the activity of methylating bacteria, and (2) the availability of HgII for methylation. It remains to be elucidate whether there is a breaking point at which one of the two factors take over the control in natural environments. The microbial activity (including methylating bacteria) is provoked by nutrient loading, seasonal fluctuation and flooding via agricultural activities and climate change. The availability of HgII is controlled by the distribution of HgII between dissolved and solid/adsorbed phases and depends heavily on the S(-II) concentration in porewater and the RSH(aq)/RSH(ads) molar ratio.

5. Acknowledgements I would take this opportunity to express my greatly indebted to my supervisor Erik Björn. Thank you very much for offering me the opportunity to work in the interesting projects. I am greatly appreciated for your nice and patient supervision, I have learnt many things from your supervision. Thank you for your “wine lessons” and “group activities”, that was unforgettable. It was remarkable and my honor to be with you in the international mercury conferences. Thousand thanks for joining the wetlands sampling campaign with me, for your cooking and supporting, it was a tough time but we knew we leant somethings from that. Ulf Skyllberg, my co-supervisor, I would also take this opportunity to show my greatly grateful to you. “Miljon tacks” to you for being super nice supervisor, for your thoughtful helps and explanations, I have learnt a lot from you. Thank you for joining the mercury conference in Jeju and greatly hiking trip there.

I would deeply grateful our Vietnamese Big Father, Lars Lundmark, thank you for open the door for many Vietnamese students to go to Umeå for studying in chemistry. I am greatly indebted your incredible supports for the wetlands sampling campaign, without you we could not make it. Rest in peace, Big Father.

Sofi Jonsson, I would give many thanks to you for inspiriting me to the “mercury world”, thank you for joining me to the conferences and fantastic drinking company. I am deeply thankful to Peter Hedlund for your immense help with WinSGW. Many thanks to Jerker Fick, Richard Lindberg and Marcus Östman for your help with LC-MSMS. Lars Lambertsson, Yu Song (Rain), Tao Jiang (Tower), Wei Zhu, Gbotemi Adediran, Andreas Drott and Aleksandra Skrobonja, my colleagues, thank you very much for your huge supports in the Lab and scientific discussions.

I would give special thank to Phuoc Dinh for your valuable advices and being a best friend. Many thanks to Hoa Vo, Chau Phan, Minh Anh Nguyen for being best friends and your supports. I am grateful to Solomon Tesfalidet for your

36 encouragements, and your wife for her fantastic handmade coffee. Thanks to Helen Genberg, Rose-Marie Kronberg, Erik Steinvall for nice fika and drinking company. I would like to thank Ngoc Tung Pham for operating our badminton group, and Tinh Hoang for your ardor helps. Thanks to Tung Nguyen, Tri Tran, Trung Le, Cuong Pham, Hung Tran, Khanh Nguyen, Minh Tuan Nguyen, and the other Vietnamese friends for your supports and being nice company.

To my family. Con Cám ơn bố mẹ đã sinh ra con, nuôi dưỡng con, cho con cơ hội được học tập và trưởng thành. Con tự hào và hạnh phúc được làm con của bố mẹ, chúc bố mẹ luôn mạnh khỏe, hạnh phúc. Cám ơn vì là chị của em, chị Hiền, cám ơn anh Tỉnh vì trở thành thành viên trong gia đình. Chúc Tiến Thành và Gia Phúc chăm ngoan, học giỏi và trưởng thành.

To my family in law. Con rất vui vì trở thành một thành viên trong gia đình. Con cám ơn những lời khuyên của bố mẹ. Chúc bố mẹ luôn mạnh khỏe, vui vẻ, hạnh phúc.

Finally, I am hugely indebted to my beloved wife, Na Pham. Thank you very much for coming together with me, for your cooking and everything. Wishing you soon complete your tasks and dreams.

6. References

1. USEPA Agency for Toxic Subtances & Disease Registry, Mercury in your environment. 2015. 2. UNEP Sources, Emissions, Releases and Environmental Transport. UNEP Chemicals Branch, Geneva, Switzerland. 2013. 3. USEPA EPA Mercury Study Report to Congress; Office of Air Quality and Standards and Office of Research and Developement Washington (DC), 1997. 4. Bjorn, E.; Larsson, T.; Lambertsson, L.; Skyllberg, U.; Frech, W., Recent advances in mercury speciation analysis with focus on spectrometric methods and enriched stable isotope applications. Ambio 2007, 36, (6), 443-451. 5. Selin, N. E., Global Biogeochemical Cycling of Mercury: A Review. Annual Review of Environment and Resources 2009, 34, 43-63. 6. Malm, O.; Pfeiffer, W. C.; Souza, C. M.; Reuther, R., Mercury pollution due to gold mining in the Madeira River basin, Brazil. Ambio 1990, 19, (1), 11-15. 7. Energy, U. S. D. o. Mercury Emission Control R&D. http://www.fossil.energy.gov/programs/powersystems/pollutioncontrols/overview_merc urycontrols.html (2 november 2011), 8. Lamborg, C. H.; Hammerschmidt, C. R.; Bowman, K. L.; Swarr, G. J.; Munson, K. M.; Ohnemus, D. C.; Lam, P. J.; Heimbürger, L.-E.; Rijkenberg, M. J.; Saito, M. A., A global ocean inventory of anthropogenic mercury based on water column measurements. Nature 2014, 512, (7512), 65-68. 9. UNEP Chemical Branch, D. The Global Atmosperic Mercury Assessment: Sources, Emissions and Transport; Geneva, 2008. 10. Pacyna, E. G.; Pacyna, J. M.; Sundseth, K.; Munthe, J.; Kindbom, K.; Wilson, S.; Steenhuisen, F.; Maxson, P., Global emission of mercury to the atmosphere from anthropogenic sources in 2005 and projections to 2020. Atmos. Environ. 2010, 44, (20), 2487-2499. 11. Kurland, T.; Faro, S. N.; Siedler, H., Minamata disease. The outbreak of a neurologic disorder in Minamata, Japan, and its relationship to the ingestion of seafood contaminated by mercuric compounds. World neurology 1960, 1, (5), 370-95. 12. Harada, M., Minamata disease: methylmercury poisoning in Japan caused by environmental pollution. Crit. Rev. Toxicol. 1995, 25, (1), 1-24. 13. Agency, U. S. E. P., National Listing of Fish Advisories" (EPA Fact Sheet EPA-823-F-09-007). 2008. 14. Jiang, G. B.; Shi, J. B.; Feng, X. B., Mercury pollution in China. Environmental Science & Technology 2006, 40, (12), 3672-3678. 15. Hakanson, L.; Nilsson, A.; Andersson, T., MERCURY IN FISH IN SWEDISH LAKES. Environ. Pollut. 1988, 49, (2), 145-162. 16. Zhang, H.; Feng, X.; Larssen, T.; Qiu, G.; Vogt, R. D., In inland China, rice, rather than fish, is the major pathway for methylmercury exposure. Environ. Health Perspect. 2010, 118, (9), 1183.

17. Union, E., Directive 2008/105/EC of the European parliament and of the council. Official journal of the European Union 2008. 18. Lindqvist, O.; Johansson, K.; Aastrup, M.; Andersson, A.; Bringmark, L.; Hovsenius, G.; Hakanson, L.; Iverfeldt, A.; Meili, M.; Timm, B., MERCURY IN THE SWEDISH ENVIRONMENT - RECENT RESEARCH ON CAUSES, CONSEQUENCES AND CORRECTIVE METHODS. Water Air and Soil Pollution 1991, 55, (1-2), R11-&. 19. Chemicals, U., Global mercury assessment. United Nations Environmental Programme. Chemicals. Inter-Organization Programme for the Sound Managements of Chemicals. Geneva, Switzerland, 258pp 2002. 20. Bellanger, M.; Pichery, C.; Aerts, D.; Berglund, M.; Castaño, A.; Cejchanová, M.; Crettaz, P.; Davidson, F.; Esteban, M.; Fischer, M. E., Economic benefits of methylmercury exposure control in Europe: monetary value of neurotoxicity prevention. Environ Health 2013, 12, (3). 21. Clarkson, T. W.; Magos, L., The toxicology of mercury and its chemical compounds. Crit. Rev. Toxicol. 2006, 36, (8), 609-662. 22. Qiu, G.; Feng, X.; Li, P.; Wang, S.; Li, G.; Shang, L.; Fu, X., Methylmercury accumulation in rice (Oryza sativa L.) grown at abandoned mercury mines in Guizhou, China. J. Agric. Food. Chem. 2008, 56, (7), 2465-2468. 23. Templeton, D. M.; Ariese, F.; Cornelis, R.; Danielsson, L. G.; Muntau, H.; Van Leeuwen, H. P.; Lobinski, R., Guidelines for terms related to chemical speciation and fractionation of elements. Definitions, structural aspects, and methodological approaches (IUPAC Recommendations 2000). Pure Appl. Chem. 2000, 72, (8), 1453-1470. 24. Jonsson, S., Unraveling the importance of solid and adsorbed phase mercury speciation for methylmercury formation, evasion and bioaccumulation. 2013. 25. Skyllberg, U., Competition among thiols and inorganic sulfides and polysulfides for Hg and MeHg in wetland soils and sediments under suboxic conditions: Illumination of controversies and implications for MeHg net production. Journal of Geophysical Research- Biogeosciences 2008, 113, G00C03. 26. Skyllberg, U., Chemical speciation of mercury in soil and sediment. Environmental chemistry and toxicology of mercury 2012, 219-258. 27. Skyllberg, U.; Bloom, P. R.; Qian, J.; Lin, C. M.; Bleam, W. F., Complexation of mercury(II) in soil organic matter: EXAFS evidence for linear two-coordination with reduced sulfur groups. Environmental Science & Technology 2006, 40, (13), 4174-4180. 28. Dyrssen, D.; Wedborg, M., The sulphur-mercury (II) system in natural waters. Water Air & Soil Pollution 1991, 56, (1), 507-519. 29. Mah, V.; Jalilehvand, F., Glutathione complex formation with mercury (II) in aqueous solution at physiological pH. Chem. Res. Toxicol. 2010, 23, (11), 1815-1823. 30. Koszegi-Szalai, H.; Paal, T., Equilibrium studies of mercury (II) complexes with penicillamine. Talanta 1999, 48, (2), 393-402. 31. Tossell, J.; Vaughan, D., Relationships between valence orbital binding energies and crystal structures in compounds of copper, silver, gold, zinc, cadmium, and mercury. Inorg. Chem. 1981, 20, (10), 3333-3340.

32. Reid, R. S.; Rabenstein, D. L., Nuclear magnetic resonance studies of the solution chemistry of metal complexes. XVII. Formation constants for the complexation of methylmercury by sulfhydryl-containing amino acids and related molecules. Can. J. Chem. 1981, 59, (10), 1505-1514. 33. Karlsson, T.; Skyllberg, U., Bonding of ppb levels of methyl mercury to reduced sulfur groups in soil organic matter. Environmental Science & Technology 2003, 37, (21), 4912- 4918. 34. Skyllberg, U.; Xia, K.; Bloom, P. R.; Nater, E. A.; Bleam, W. F., Binding of mercury (II) to reduced sulfur in soil organic matter along upland-peat soil transects. Journal of environmental quality 2000, 29, (3), 855-865. 35. Benoit, J. M.; Gilmour, C. C.; Mason, R. P.; Heyes, A., Sulfide controls on mercury speciation and bioavailability to methylating bacteria in sediment pore waters. Environmental Science & Technology 1999, 33, (6), 951-957. 36. Drott, A.; Lambertsson, L.; Bjorn, E.; Skyllberg, U., Importance of dissolved neutral mercury sulfides for methyl mercury production in contaminated sediments. Environmental Science & Technology 2007, 41, (7), 2270-2276. 37. Drott, A.; Björn, E.; Bouchet, S.; Skyllberg, U., Refining Thermodynamic Constants for Mercury(II)-Sulfides in Equilibrium with Metacinnabar at Sub-Micromolar Aqueous Sulfide Concentrations. Environmental Science & Technology 2013, 47, (9), 4197-4203. 38. Schaefer, J. K.; Rocks, S. S.; Zheng, W.; Liang, L. Y.; Gu, B. H.; Morel, F. M. M., Active transport, substrate specificity, and methylation of Hg(II) in anaerobic bacteria. Proceedings of the National Academy of Sciences of the United States of America 2011, 108, (21), 8714-8719. 39. Liem-Nguyen, V.; Bouchet, S.; Björn, E., Determination of Sub-Nanomolar Levels of Low Molecular Mass Thiols in Natural Waters by Liquid Chromatography Tandem Mass Spectrometry after Derivatization with p-(Hydroxymercuri) Benzoate and Online Preconcentration. Anal. Chem. 2014, 87, (2), 1089-1096. 40. Schaefer, J. K.; Morel, F. M. M., High methylation rates of mercury bound to cysteine by Geobacter sulfurreducens. Nature Geoscience 2009, 2, (2), 123-126. 41. Gilmour, C. C.; Henry, E. A.; Mitchell, R., Sulfate Stimulation of Mercury Methylation in Fresh-Water Sediments. Environmental Science & Technology 1992, 26, (11), 2281-2287. 42. Compeau, G. C.; Bartha, R., SULFATE-REDUCING BACTERIA - PRINCIPAL METHYLATORS OF MERCURY IN ANOXIC ESTUARINE SEDIMENT. Applied and Environmental Microbiology 1985, 50, (2), 498-502. 43. Kerin, E. J.; Gilmour, C. C.; Roden, E.; Suzuki, M. T.; Coates, J. D.; Mason, R. P., Mercury methylation by dissimilatory iron-reducing bacteria. Applied and Environmental Microbiology 2006, 72, (12), 7919-7921. 44. Fleming, E. J.; Mack, E. E.; Green, P. G.; Nelson, D. C., Mercury methylation from unexpected sources: Molybdate-inhibited freshwater sediments and an iron-reducing bacterium. Applied and Environmental Microbiology 2006, 72, (1), 457-464. 45. Hamelin, S.; Amyot, M.; Barkay, T.; Wang, Y.; Planas, D., Methanogens: principal methylators of mercury in lake periphyton. Environmental science & technology 2011, 45, (18), 7693-7700.

46. Gilmour, C. C.; Podar, M.; Bullock, A. L.; Graham, A. M.; Brown, S. D.; Somenahally, A. C.; Johs, A.; Hurt Jr, R. A.; Bailey, K. L.; Elias, D. A., Mercury methylation by novel microorganisms from new environments. Environmental science & technology 2013, 47, (20), 11810-11820. 47. Benoit, J. M.; Gilmour, C. C.; Heyes, A.; Mason, R. P.; Miller, C. L., Geochemical and biological controls over methylmercury production and degradation in aquatic ecosystems. 2003; p 262-297. 48. Jonsson, S.; Skyllberg, U.; Nilsson, M. B.; Lundberg, E.; Andersson, A.; Björn, E., Differentiated availability of geochemical mercury pools controls methylmercury levels in estuarine sediment and biota. Nature communications 2014, 5. 49. Mehrotra, A. S.; Sedlak, D. L., Decrease in net mercury methylation rates following iron amendment to anoxic wetland sediment slurries. Environmental science & technology 2005, 39, (8), 2564-2570. 50. Jeong, H. Y.; Klaue, B.; Blum, J. D.; Hayes, K. F., Sorption of mercuric ion by synthetic manocrystalline mackinawite (FeS). Environmental Science & Technology 2007, 41, (22), 7699-7705. 51. Hsu-Kim, H.; Kucharzyk, K. H.; Zhang, T.; Deshusses, M. A., Mechanisms regulating mercury bioavailability for methylating microorganisms in the aquatic environment: a critical review. Environmental science & technology 2013, 47, (6), 2441-2456. 52. Zhang, J.; Wang, F.; House, J. D.; Page, B., Thiols in wetland interstitial waters and their role in mercury and methylmercury speciation. Limnology and oceanography 2004, 49, (6), 2276-2286. 53. Dryden, C. L.; Gordon, A. S.; Donat, J. R., Seasonal survey of copper-complexing ligands and thiol compounds in a heavily utilized, urban estuary: Elizabeth River, Virginia. Mar. Chem. 2007, 103, (3), 276-288. 54. MacCrehan, W.; Shea, D.; Vairavamurthy, M.; Schoonen, M.; Eglinton, T.; Luther III, G.; Manowitz, B. In Temporal relationship of thiols to inorganic sulfur compounds in anoxic Chesapeake Bay sediment porewater, ACS Symp. Ser., 1995; Washington, DC: American Chemical Society,[1974]-: 1995; pp 294-310. 55. Cline, J. D., SPECTROPHOTOMETRIC DETERMINATION OF HYDROGEN SULFIDE IN NATURAL WATERS. Limnology and Oceanography 1969, 14, (3), 454-&. 56. Tjerngren, I.; Karlsson, T.; Bjorn, E.; Skyllberg, U., Potential Hg methylation and MeHg demethylation rates related to the nutrient status of different boreal wetlands. Biogeochemistry 2012, 108, (1-3), 335-350. 57. Zheng, Y.; Anderson, R. F.; van Geen, A.; Kuwabara, J., Authigenic molybdenum formation in marine sediments: a link to pore water sulfide in the Santa Barbara Basin. Geochim. Cosmochim. Acta 2000, 64, (24), 4165-4178. 58. Tjerngren, I.; Meili, M.; Björn, E.; Skyllberg, U., Eight Boreal Wetlands as Sources and Sinks for Methyl Mercury in Relation to Soil Acidity, C/N Ratio, and Small-Scale Flooding. Environmental Science & Technology 2012, 46, (15), 8052-8060. 59. St. Louis, V. L.; Rudd, J. W.; Kelly, C. A.; Beaty, K. G.; Bloom, N. S.; Flett, R. J., Importance of wetlands as sources of methyl mercury to boreal forest ecosystems. Canadian Journal of fisheries and aquatic sciences 1994, 51, (5), 1065-1076.

60. Hammerschmidt, C. R.; Fitzgerald, W. F., Geochemical controls on the production and distribution of methylmercury in near-shore marine sediments. Environmental Science & Technology 2004, 38, (5), 1487-1495. 61. Rao, Y.; Xiang, B.; Bramanti, E.; D’Ulivo, A.; Mester, Z., Determination of thiols in yeast by HPLC coupled with LTQ-orbitrap mass spectrometry after derivatization with p- (Hydroxymercuri) benzoate. J. Agric. Food. Chem. 2010, 58, (3), 1462-1468. 62. Bakirdere, S.; Bramanti, E.; D’ulivo, A.; Ataman, O. Y.; Mester, Z., Speciation and determination of thiols in biological samples using high performance liquid chromatography–inductively coupled plasma-mass spectrometry and high performance liquid chromatography–Orbitrap MS. Anal. Chim. Acta 2010, 680, (1), 41-47. 63. Qian, J.; Skyllberg, U.; Frech, W.; Bleam, W. F.; Bloom, P. R.; Petit, P. E., Bonding of methyl mercury to reduced sulfur groups in soil and stream organic matter as determined by X- ray absorption spectroscopy and binding affinity studies. Geochim. Cosmochim. Acta 2002, 66, (22), 3873-3885. 64. Burton, E. D.; Bush, R. T.; Johnston, S. G.; Sullivan, L. A.; Keene, A. F., Sulfur biogeochemical cycling and novel Fe–S mineralization pathways in a tidally re-flooded wetland. Geochim. Cosmochim. Acta 2011, 75, (12), 3434-3451. 65. Lambertsson, L.; Lundberg, E.; Nilsson, M.; Frech, W., Applications of enriched stable isotope tracers in combination with isotope dilution GC-ICP-MS to study mercury species transformation in sea sediments during in situ ethylation and determination. J. Anal. At. Spectrom. 2001, 16, (11), 1296-1301. 66. Hintelmann, H.; Keppel-Jones, K.; Evans, R. D., Constants of mercury methylation and demethylation rates in sediments and comparison of tracer and ambient mercury availability. Environ. Toxicol. Chem. 2000, 19, (9), 2204-2211. 67. Jonsson, S.; Skyllberg, U.; Nilsson, M. B.; Westlund, P.-O.; Shchukarev, A.; Lundberg, E.; Björn, E., Mercury Methylation Rates for Geochemically Relevant HgII Species in Sediments. Environmental Science & Technology 2012, 46, (21), 11653-11659. 68. Skyllberg, U.; Drott, A., Competition between Disordered Iron Sulfide and Natural Organic Matter Associated Thiols for Mercury(II)-An EXAFS Study. Environmental Science & Technology 2010, 44, (4), 1254-1259. 69. Martell, A.; Smith, R.; Motekaitis, R., National Institute of Standard and Technology, NIST. Critically selected stability constants of metal complexes. PC-based Database, Gaithersburg, MD 2004, 20899. 70. Snell, J. P.; Stewart, I. I.; Sturgeon, R. E.; Frech, W., Species specific isotope dilution calibration for determination of mercury species by gas chromatography coupled to inductively coupled plasma- or furnace atomisation plasma ionisation-mass spectrometry. J. Anal. At. Spectrom. 2000, 15, (12), 1540-1545. 71. Hu, H.; Mylon, S. E.; Benoit, G., Distribution of the thiols glutathione and 3‐mercaptopropionic acid in Connecticut lakes. Limnology and oceanography 2006, 51, (6), 2763-2774. 72. Seiwert, B.; Karst, U., Simultaneous LC/MS/MS determination of thiols and disulfides in urine samples based on differential labeling with ferrocene-based maleimides. Anal. Chem. 2007, 79, (18), 7131-7138.

73. Vairavamurthy, A.; Mopper, K., Field method for determination of traces of thiols in natural waters. Anal. Chim. Acta 1990, 236, 363-370. 74. Kuśmierek, K.; Chwatko, G.; Głowacki, R.; Kubalczyk, P.; Bald, E., Ultraviolet derivatization of low-molecular-mass thiols for high performance liquid chromatography and capillary electrophoresis analysis. J. Chromatogr. B 2011, 879, (17), 1290-1307. 75. Toyo’oka, T., Recent advances in separation and detection methods for thiol compounds in biological samples. J. Chromatogr. B 2009, 877, (28), 3318-3330. 76. Tang, D.; Wen, L.-S.; Santschi, P. H., Analysis of biogenic thiols in natural water samples by high-performance liquid chromatographic separation and fluorescence detection with ammonium 7-fluorobenzo-2-oxa-1, 3-diazole-4-sulfonate (SBD-F). Anal. Chim. Acta 2000, 408, (1), 299-307. 77. McMenamin, M. E.; Himmelfarb, J.; Nolin, T. D., Simultaneous analysis of multiple aminothiols in human plasma by high performance liquid chromatography with fluorescence detection. J. Chromatogr. B 2009, 877, (28), 3274-3281. 78. Isokawa, M.; Funatsu, T.; Tsunoda, M., Fast and simultaneous analysis of biothiols by high- performance liquid chromatography with fluorescence detection under hydrophilic interaction chromatography conditions. Analyst 2013, 138, (13), 3802-3808. 79. Dickinson, D. A.; Forman, H. J., Glutathione in defense and signaling. Ann. N.Y. Acad. Sci. 2002, 973, (1), 488-504. 80. Gulati, P.; Klöhn, P. C.; Krug, H.; Göttlicher, M.; Markova, B.; Böhmer, F. D.; Herrlich, P., Redox regulation in mammalian signal transduction. IUBMB life 2001, 52, (1), 25-28. 81. Moriarty-Craige, S. E.; Jones, D. P., Extracellular thiols and thiol/disulfide redox in metabolism. Annu. Rev. Nutr. 2004, 24, 481-509. 82. Casas, J.; Jones, M. M., Mercury (II) complexes with sulfhydryl containing chelating agents: stability constant inconsistencies and their resolution. J. Inorg. Nucl. Chem. 1980, 42, (1), 99-102. 83. Basinger, M. A.; Casas, J.; Jones, M. M.; Weaver, A. D.; Weinstein, N. H., Structural requirements for Hg (II) antidotes. J. Inorg. Nucl. Chem. 1981, 43, (6), 1419-1425. 84. Starý, J.; Kratzer, K., Radiometric determination of stability constants of mercury species complexes with L-cysteine. J. Radioanal. Nucl. Chem. 1988, 126, (1), 69-75. 85. Cardiano, P.; Cucinotta, D.; Foti, C.; Giuffre, O.; Sammartano, S., Potentiometric, Calorimetric, and 1H NMR Investigation on Hg2+-Mercaptocarboxylate Interaction in Aqueous Solution. Journal of Chemical & Engineering Data 2011, 56, (5), 1995-2004. 86. Cardiano, P.; Falcone, G.; Foti, C.; Sammartano, S., Sequestration of Hg2+ by some biologically important thiols. Journal of Chemical & Engineering Data 2011, 56, (12), 4741- 4750. 87. Jalilehvand, F.; Leung, B. O.; Izadifard, M.; Damian, E., Mercury (II) cysteine complexes in alkaline aqueous solution. Inorg. Chem. 2006, 45, (1), 66-73. 88. Leung, B. O.; Jalilehvand, F.; Mah, V., Mercury (II) penicillamine complex formation in alkaline aqueous solution. Dalton Transactions 2007, (41), 4666-4674. 89. Liu, G.; Cai, Y.; O'Driscoll, N., Environmental chemistry and toxicology of mercury. John Wiley & Sons: 2011. 43

90. Ugur, I.; Marion, A.; Parant, S.; Jensen, J. H.; Monard, G., Rationalization of the p K a Values of Alcohols and Thiols Using Atomic Charge Descriptors and Its Application to the Prediction of Amino Acid p K a’s. Journal of chemical information and modeling 2014, 54, (8), 2200-2213. 91. Lambertsson, L.; Nilsson, M., Organic material: The primary control on mercury methylation and ambient methyl mercury concentrations in estuarine sediments. Environmental Science & Technology 2006, 40, (6), 1822-1829. 92. Morel, F. M. M.; Kraepiel, A. M. L.; Amyot, M., The chemical cycle and bioaccumulation of mercury. Annual Review of Ecology and Systematics 1998, 29, 543-566. 93. Xia, K.; Skyllberg, U.; Bleam, W.; Bloom, P.; Nater, E.; Helmke, P., X-ray absorption spectroscopic evidence for the complexation of Hg (II) by reduced sulfur in soil humic substances. Environmental science & technology 1999, 33, (2), 257-261. 94. Hesterberg, D.; Chou, J. W.; Hutchison, K. J.; Sayers, D. E., Bonding of Hg (II) to reduced organic sulfur in humic acid as affected by S/Hg ratio. Environmental science & technology 2001, 35, (13), 2741-2745. 95. Yoon, S.-J.; Diener, L. M.; Bloom, P. R.; Nater, E. A.; Bleam, W. F., X-ray absorption studies of CH 3 Hg+-binding sites in humic substances. Geochim. Cosmochim. Acta 2005, 69, (5), 1111-1121. 96. Schwarzenbach, G.; Widmer, M., Die löslichkeit von metallsulfiden I. schwarzes quecksilbersulfid. Helv. Chim. Acta 1963, 46, (7), 2613-2628. 97. Deonarine, A.; Hsu-Kim, H., Precipitation of mercuric sulfide nanoparticles in NOM- containing water: Implications for the natural environment. Environmental science & technology 2009, 43, (7), 2368-2373. 98. Slowey, A. J., Rate of formation and dissolution of mercury sulfide nanoparticles: The dual role of natural organic matter. Geochim. Cosmochim. Acta 2010, 74, (16), 4693-4708. 99. Pham, A. L.-T.; Morris, A.; Zhang, T.; Ticknor, J.; Levard, C.; Hsu-Kim, H., Precipitation of nanoscale mercuric sulfides in the presence of natural organic matter: Structural properties, aggregation, and biotransformation. Geochim. Cosmochim. Acta 2014, 133, 204-215. 100. Leclerc, M.; Planas, D.; Amyot, M., Relationship between extracellular low-molecular- weight thiols and mercury species in natural lake periphytic biofilms. Environmental science & technology 2015, 49, (13), 7709-7716. 101. Shea, D.; MacCrehan, W. A., Determination of hydrophilic thiols in sediment porewater using ion-pair liquid chromatography coupled to electrochemical detection. Anal. Chem. 1988, 60, (14), 1449-1454. 102. Andersson, A.; Hajdu, S.; Haecky, P.; Kuparinen, J.; Wikner, J., Succession and growth limitation of phytoplankton in the Gulf of Bothnia (Baltic Sea). Mar. Biol. 1996, 126, (4), 791-801. 103. Andersson, A.; Haecky, P.; Hagström, Å., Effect of temperature and light on the growth of micro-nano-and pico-plankton: impact on algal succession. Mar. Biol. 1994, 120, (4), 511- 520. 104. Elmgren, R., Trophic dynamics in the enclosed, brackish Baltic Sea. Rapp Pv Réun. Cons. Int. Explor. Mer 1984, 183, 152-169. 44

105. Lehnherr, I.; St Louis, V. L.; Hintelmann, H.; Kirk, J. L., Methylation of inorganic mercury in polar marine waters. Nature geoscience 2011, 4, (5), 298-302. 106. Hammerschmidt, C. R.; Fitzgerald, W. F., Photodecomposition of methylmercury in an arctic Alaskan lake. Environmental science & technology 2006, 40, (4), 1212-1216. 107. Rytuba, J. J., Mercury from mineral deposits and potential environmental impact. Environ. Geol. 2003, 43, (3), 326-338. 108. Malm, O., Gold mining as a source of mercury exposure in the Brazilian Amazon. Environ. Res. 1998, 77, (2), 73-78. 109. Porvari, P.; Verta, M.; Munthe, J.; Haapanen, M., Forestry practices increase mercury and methyl mercury output from boreal forest catchments. Environmental Science & Technology 2003, 37, (11), 2389-2393. 110. Kronberg, R.-M., The boreal journey of methyl mercury. 2014. 111. Mangal, V.; Guéguen, C., Examining concentrations and molecular weights of thiols in microorganism cultures and in Churchill River (Manitoba) using a fluorescent-labeling method coupled to asymmetrical flow field-flow fractionation. Analytical and bioanalytical chemistry 2015, 407, (15), 4305-4313.