Elsevier

Analytica Chimica Acta

Volume 663, Issue 2, 24 March 2010, Pages 127-138
Analytica Chimica Acta

Review
Methods for the determination and speciation of mercury in natural waters—A review

https://doi.org/10.1016/j.aca.2010.01.048Get rights and content

Abstract

This review summarises current knowledge on Hg species and their distribution in the hydrosphere and gives typical concentration ranges in open ocean, coastal and estuarine waters, as well as in rivers, lakes, rain and ground waters. The importance of reliable methods for the determination of Hg species in natural waters and the analytical challenges associated with them are discussed. Approaches for sample collection and storage, pre-concentration, separation, and detection are critically compared. The review covers well established methods for total mercury determination and identifies new approaches that offer advantages such as ease of use and reduced risk of contamination. Pre-concentration and separation techniques for Hg speciation are divided into chromatographic and non-chromatographic methods. Derivatisation methods and the coupling of pre-concentration and/or separation methods to suitable detection techniques are also discussed. Techniques for sample pre-treatment, pre-concentration, separation, and quantification of Hg species, together with examples of total Hg determination and Hg speciation analysis in different natural (non-spiked) waters are summarised in tables, with a focus on applications from the last decade.

Introduction

Mercury is one of the most toxic elements impacting on human and ecosystem health and therefore is one of the most studied environmental pollutants. All mercury species are toxic, with organic mercury compounds generally being more toxic than inorganic species. The United States Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA) lists mercury and its compounds in third place on the “Priority List of Hazardous Substances” and the European Water Framework Directive (2000/60/EG) classifies mercury as one of 30 “precarious dangerous pollutants”.

A wide range of mercury species exists within our environment and the chemical form of mercury controls its bioavailability, transport, persistence and impact on the human body, with methylmercury being the most toxic species. Because of its high bioaccumulation, mercury concentrations escalate up the food chain and, e.g. predatory fish can have up to 106 times higher mercury concentrations than the ambient water and up to 95% of this mercury can be in the form of methylmercury. This was the cause of the well documented events in Minamata Bay in Japan in 1953, where over 2000 people, the majority of whom died, were affected by mercury poisoning due to the consumption of methylmercury contaminated fish [1]. Another poisoning incident due to organic mercury occurred in Iraq in 1972 where grain, treated with organomercury fungicides, was consumed [1].

The World Health Organisation (WHO) recommends a maximum intake of methylmercury of 1.6 μg kg−1 per week [2] and the United States Environmental Protection Agency (EPA) and National Research Council (NRC) [3] developed a reference dose of 0.1 μg kg−1 body weight per day for adults. Organomercury compounds were banned from agricultural use in the 1970s in Europe and elsewhere [4]. Nowadays it is well known that any mercury released into the environment undergoes biogeochemical transformation processes and can be converted into the most toxic methylmercury form. Therefore, industrial countries have made great efforts to replace any mercury in products (e.g. amalgam fillings, thermometers, switches) and industrial processes (catalyst in e.g. acetaldehyde production, amalgam in chlorine-alkali electrolysis) by other substances or processes within recent decades. In 2006 the European Union (EU) banned the use of mercury in thermometers and other measuring instruments and from 2011 export of mercury from the EU will be prohibited in order to further reduce its release. Nonetheless, more than 2000 t are emitted annually from global anthropogenic sources [5], with about two-thirds originating from coal and fuel combustion and increasing emissions from local sources in some regions. In addition, significant natural emission sources, e.g. volcanoes, contribute ca. 2500 t per year to the total mercury release [6]. The biogeochemical transformations and high mobility of mercury species make mercury a global pollutant that can be transported thousands of kilometers through the atmosphere. Therefore, an understanding of mercury species transformations and accurate monitoring of mercury and its species in the environment are essential for reliable risk assessment.

Mercury emissions from both anthropogenic and natural sources are mainly in the form of elemental mercury (Hg0) which constitutes >99% of the total Hg in the atmosphere [4]. However (bio)geochemical transformations can lead to Hg species in the oxidation states +I and +II. Most inorganic HgI compounds are sparingly soluble in water and are therefore found mainly in soils and sediments. In contrast, inorganic HgII forms complexes or compounds with inorganic and/or organic species (e.g. [HgClx]2−x; [HgII-DOC]; HgS), dependent on the local chemical environment. Their residence time in the air is very short (minutes) and they are rapidly removed by deposition processes due to their high surface activity and water solubility [7]. More than 90% of the mercury loading of surface waters results from atmospheric deposition. Inorganic HgII undergoes biomethylation, resulting in the formation of monomethylmercury (MMeHg; CH3Hg+) and dimethylmercury (DMeHg; (CH3)2Hg). These reactions are reversible with demethylation facilitated by microorganisms and/or photolytic decomposition. All of these species are highly mobile as shown by the biogeochemical mercury cycle in Fig. 1[8]. In natural waters there are three main forms of mercury, namely elemental mercury (Hg0), inorganic HgII (Hg2+ and its complexes), and organic mercury (MMeHg and its complexes and DMeHg) [9]. With an aqueous solubility of 0.08 mg L−1[10] (at 25 °C) Hg0 is a ubiquitous component of natural waters [11]. In the oceans Hg0 is found at all depths and is usually supersaturated, especially in surface waters, and can be nearly 50% of the total Hg in the mixed layer [12]. In the Mediterranean Sea high Hg0 concentrations are also found in bottom waters, probably due to intense tectonic activity [13]. In freshwaters it constitutes 10–30% of total mercury [14], [15]. Inorganic HgII and MMeHg in freshwaters are complexed by dissolved humic matter [16], whereas in sea water mainly chloro-complexes are formed. Dimethylmercury has been found in the deep ocean [17] but it has not been detected in the mixed layer or in peripheral seas where loss by evaporation can occur and decomposition via photolysis and thermal instability are likely [11], [13]. In freshwaters DMeHg is normally not detected [9]. The proportion of organic mercury in marine waters is typically <5% of the total mercury concentration [18], although the Mediterranean Sea shows a “mercury anomaly” with an organic mercury content of up to 30% [19], whereas in freshwater systems the organic mercury fraction is typically 30% of the total mercury pool [20]. Average total mercury concentrations and percentages of mercury species/fractions dissolved in different natural waters are listed in Table 1.

In addition to the monitoring of total mercury concentrations in the environment, speciation analysis provides very useful additional information. In accordance to the official definition of the International Union of Pure and Applied Chemistry (IUPAC), a ‘chemical species’ is a specific form of an element defined as to isotopic composition, electronic or oxidation state, and/or complex or molecular structure. Speciation analysis represents the analytical activity of identifying and/or measuring the quantities of one or more individual chemical species in a sample[31]. Therefore, speciation analysis always implies the determination of very low concentrations of minor species and in the case of mercury speciation in waters it means that concentrations in a range of ng to pg L−1 have to be handled. Highly sensitive detection techniques are available for mercury (e.g. cold vapour atomic fluorescence spectrometry, CVAFS; inductively coupled plasma-mass spectrometry, ICP-MS) but speciation analysis also requires effective and selective pre-concentration procedures. In addition to the demand for high sensitivity, the preservation and integrity of the sample and the Hg species of interest during sampling, storage and pre-treatment are crucial. This requires less aggressive procedures than those used in total Hg analysis.

Depending on the objective, different strategies for speciation analysis are necessary. In order to elucidate Hg transformation and transport processes the determination of all Hg species in a sample is desirable. Chromatographic techniques (gas chromatography, GC); high performance liquid chromatography, HPLC) coupled to element specific detectors are able to separate Hg species but in practice a combination of methods is usually applied to accurately determine all of the different Hg species in a natural water sample. These include non-chromatographic methods based on the different chemical and/or physical behaviour of the Hg species.

When investigating the adverse effects of toxic Hg species on living organisms it could be sufficient to determine the most toxic species, i.e. methylmercury. In other applications the determination of a particular Hg fraction could be appropriate, e.g. dissolved gaseous Hg (DGHg; Hg0 and DMeHg) or “reactive Hg”, which is defined as inorganic HgII that can easily be reduced to Hg0 by addition of a reductant. A schematic diagram showing commonly applied techniques for sample pre-treatment, pre-concentration, separation and quantification in waters is given in Fig. 2. For sample pre-treatment several of the listed techniques are often combined whereas for pre-concentration, separation and quantification a particular technique is usually chosen.

This review discusses the drivers for the determination and speciation of mercury in natural waters and critically compares approaches for sample collection and storage, pre-concentration, separation and detection. Approaches used and figures of merit for selected literature are presented in tabular form. The review focuses on methods published since 2000 that have been applied to real water samples. All abbreviations used in this review are listed in alphabetical order in Table 2.

Section snippets

Sample collection and storage

Clean sample collection and suitable pre-treatment procedures are as important in ultra trace Hg analysis as selective pre-concentration and highly sensitive detection methods. Due to the extremely low concentrations in natural waters, even minor contamination of the reagents used, storage containers or other tools have a deleterious effect. Beside the risk of contamination, the high volatility of some Hg species (Hg0 and DMeHg) and the strong affinity of elemental and inorganic Hg species for

Total dissolved mercury determination

The most common technique used to monitor total mercury in natural waters is cold vapour atomic fluorescence spectrometry (CVAFS). As a pre-treatment step, natural water samples are digested to transform all mercury species to “reducible” mercury (free inorganic HgII ions) by addition of a strong oxidant, e.g. bromine chloride, permanganate, or UV light. After pre-reduction of excess oxidant, HgII is reduced to elemental Hg0 by, e.g. tin(II) chloride or sodium borohydride, which is then purged

Future trends and recommendations

Developments in analytical techniques have made the quantification and separation of Hg species in the aquatic environment possible at the pg L−1 level. Nevertheless, the determination of Hg species in natural waters is still a challenge for analytical scientists due to the high risk of contamination, analyte losses, and inter-species conversion during sampling, storage, and pre-treatment. The ubiquity of Hg makes sample pre-treatment procedures a key issue for its sensitive and accurate

Acknowledgement

The authors are very grateful to the Bavarian research foundation (Bayerische Forschungsstiftung) for financial support.

References (125)

  • M. Gochfeld

    Ecotoxicol. Environ. Saf.

    (2003)
  • J. Kotnik et al.

    Mar. Chem.

    (2007)
  • N.J. O’Driscoll et al.

    Sci. Total Environ.

    (2003)
  • R.P. Mason et al.

    Deep-Sea Res. Part II: Top. Stud. Oceanogr.

    (1999)
  • R.P. Mason et al.

    Mar. Chem.

    (1998)
  • F.J.G. Laurier et al.

    Mar. Chem.

    (2004)
  • R.P. Mason et al.

    Deep-Sea Res. Part I

    (1993)
  • D. Cossa et al.

    Mar. Chem.

    (2004)
  • C.H. Lamborg et al.

    Mar. Chem.

    (2008)
  • J.L. Barringer et al.

    Sci. Total Environ.

    (2005)
  • M. Leermakers et al.

    Trends Anal. Chem.

    (2005)
  • J.L. Parker et al.

    Sci. Total Environ.

    (2005)
  • Y. Cai et al.

    J. Chromatogr. A

    (1995)
  • R.B. Geerdink et al.

    J. Chromatogr. A

    (2007)
  • L.R. Bravo-Sanchez et al.

    Spectrochim. Acta B: Atom. Spectrosc.

    (2004)
  • S. Díez et al.

    Talanta

    (2008)
  • K. Leopold et al.

    Trends Anal. Chem.

    (2009)
  • K. Leopold et al.

    Talanta

    (2008)
  • O. Wurl et al.

    Talanta

    (2000)
  • S. Gil et al.

    Spectrochim. Acta Part B

    (2007)
  • T. Stoichev et al.

    Talanta

    (2004)
  • J. Sanz et al.

    Anal. Chim. Acta

    (2003)
  • A. Castillo et al.

    Anal. Chim. Acta

    (2006)
  • J.L. Gómez-Ariza et al.

    J. Chromatogr. A

    (2004)
  • J. Margetinova et al.

    Anal. Chim. Acta

    (2008)
  • C.R. Hammerschmidt et al.

    Geochim. Cosmochim. Acta

    (2006)
  • J. Faganeli et al.

    Sci. Total Environ.

    (2003)
  • M.V. Balarama Krishna et al.

    Talanta

    (2005)
  • B.C. Mondal et al.

    Anal. Chim. Acta

    (2003)
  • I. Dakova et al.

    Talanta

    (2009)
  • D. Cossa et al.

    Mar. Chem.

    (2004)
  • R. Puk et al.

    Anal. Chim. Acta

    (1994)
  • P.J. Craig et al.

    Chemosphere

    (1999)
  • M. Horvat et al.

    Anal. Chim. Acta

    (1993)
  • M. Horvat et al.

    Anal. Chim. Acta

    (1993)
  • T. De Smaele et al.

    J. Chromatogr. A

    (1998)
  • E. Bulska et al.

    Anal. Chim. Acta

    (1991)
  • C.M. Tseng et al.

    Chemosphere

    (1999)
  • Food & Agriculture Organization (FAO) and World Health Organization (WHO), Joint FAO/WHO, Expert Committee on Food...
  • NRC

    Toxicological Effects of Methylmercury

    (2000)
  • P. Jitaru et al.

    J. Phys. IV France

    (2004)
  • E.G. Pacynaa et al.

    Atmos. Environ.

    (2006)
  • J.O. Nriagu

    Nature

    (1989)
  • O.R. Bullock 259 (2000)...
  • E.D. Stein et al.

    Crit. Rev. Environ. Sci. Technol.

    (1996)
  • S.M. Ullrich et al.

    Crit. Rev. Environ. Sci. Technol.

    (2001)
  • Agency for Toxic Substances and Disease Registry (ATSDR)

    Toxicologic Profile for Mercury

    (2000)
  • W.F. Fitzgerald et al.

    Chem. Rev.

    (2007)
  • R.P. Mason et al.

    Water Air Soil Pollut.

    (1995)
  • M. Amyot et al.

    Can. J. Fish. Aquat. Sci.

    (2000)
  • Cited by (447)

    View all citing articles on Scopus
    View full text