Simple methodology for methylmercury and inorganic mercury determinations by high-performance liquid chromatography–cold vapour atomic fluorescence spectrometry
Introduction
Mercury in natural environments may be present in several inorganic and organic forms, being methylmercury (MeHg) the dominant toxic occurring organomercury specie in environmental matrices. In order to assess fluxes, bioavailability and toxicity of the different mercury forms, the chemical species present in the environmental compartments should be determined with a high degree of analytical confidence.
Several analytical methods for separating mercury species based either on gas chromatography (GC) or high-performance liquid chromatography (HPLC) coupled with element-specific detection for mercury have been developed in the last few years. Coupled with GC, a large variety of mercury detection and quantification methods have been used: electron capture detection (ECD) [1]; microwave induced plasma (MIP) [2], [3], [4]; atomic fluorescence spectrometry (AFS) [5], [6], [7]; atomic absorption spectrometry (AAS) [8], [9]; inductively coupled plasma-mass spectrometry (ICP-MS) [10], [11]; and glow discharge atomic emission spectrometry (GD-AES) [12]. Coupled with HPLC, the detection and quantification methods for mercury speciation have included: ICP-MS [13]; photometry [14]; cold vapour atomic absorption spectrometry (CV-AAS) [15]; cold vapour atomic fluorescence spectrometry (CV-AFS) [16], [17]; atomic emission spectrometry (AES) [18], [19]; and atmospheric pressure ionisation mass spectrometry (API-MS) [20]. Although GC has been the most widely used separation technique in mercury speciation studies, it is necessary to perform column passivation by repeated mercury salt injections [21] and derivatisation of mercury compounds to form volatile species [22]. HPLC allows an easier sample treatment, elimination of the derivatisation step, separation of mercury compounds at laboratory temperature, injection of large volumes of sample, easy performance and automatisation. One disadvantage of HPLC systems is the highest detection limits (DLs), usually ranging between 20 and 1000 pg [15], [16], [23], [24].
The most commonly used detection techniques for organomercury compounds determination by HPLC are the CV-AAS and the CV-AFS. The mobile phase is mixed with a reducing agent, e.g. tin(II) chloride (SnCl2) in an alkaline solution [16], [25], [26] or sodium tetrahydroborate (NaBH4) [15], [17], [27], [28]. The main drawback in the use of NaBH4 in the reduction step is the generation of hydrogen simultaneously with the production of elemental mercury. Hydrogen, besides requiring careful safety precautions, quenches the mercury fluorescence signal, thus acting as an interference. The reduced mercury vapour formed in the reduction step is then separated in a gas–liquid separator cell and stripped by an inert gas directly into the flow cell of the detector. Usually, when SnCl2 is used for mercury reduction, organomercury compounds should be transformed into inorganic mercury prior to reduction. Several oxidising solutions, e.g. potassium persulfate [16], [25], [29], a mixture of potassium bromate and bromide [30], [31], as well as, UV radiation [15] have been used for organomercury compounds oxidation. When NaBH4 is used as reducing agent, the subsequent oxidation step seems to be unnecessary since this compound is efficient in decomposing inorganic and organic mercury to elemental form [32].
A simple HPLC-based methodology, with SnCl2 reduction in acidic medium, for the determination of methylmercury and inorganic mercury, with a DL of around 10 pg without any pre-concentration step, is proposed in this work. The composition of the mobile phase and the type of oxidation (chemical oxidation versus UV irradiation) were selected, and the flow rates of the mobile phase and of the reducing agent, length of reduction reaction coil, flow rates of carrier and shield gas were optimised by a modified simplex algorithm. The high performance of the methodology has been confirmed by its application to the analysis of a certified reference material (CRM) sediment reference material.
Section snippets
Assembling of instrumentation
A schematic view of the HPLC–CV-AFS system used in this work is presented in Fig. 1, including a reversed-phase HPLC system for the separation of different mercury species and a CV-AFS for element-specific detection.
The chromatographic system consisted of a HPLC pump (Knauer, Berlin, Germany; 10 cm3), a six-port injection valve (Type Rheodyne; six-port) (Rheodyne, California, USA) equipped with a 200 μl PEEK loop and a reversed-phase analytical column packed with Nucleosil ODS (RPC18, 25 cm×4.6 mm,
Wet oxidation as oxidation procedure
The wet oxidation was performed at three temperatures (30, 50 and 65°C) and in four coil lengths (3.4, 10.0, 13.4 and 19.0 m). The formation of bromine, which is an oxidising agent, was easily observed by the yellow colour of the solution. The wet oxidation recovery, measured as the ratio between MeHg+ and Hg(II) peaks heights, ranged from 37 to nearly 74% when temperature varied from 30 to 65°C, with a coil length of 3.4 m. In fact, with this coil, the methylmercury signal increased almost
Conclusion
A low DL methodology (10±2 pg) for separation and determination of MeHg+ and Hg(II) has been accomplished by HPLC-UV–CV-AFS. The UV oxidation of organomercury compounds discards the addition of any chemicals to the sample and can be performed at laboratory temperature in smaller coil lengths when comparing to the wet oxidation method. The tin(II) chloride reduction of mercury avoids the production of hydrogen that causes quenching of mercury fluorescence. The water vapour trap based on H2SO4/CaCl
Acknowledgements
The authors are pleased to acknowledge the help of Dr. Millward (University of Plymouth) in suggesting different water removal systems and to PSA in helping us with the AFS detector. This work was partially supported by grants provided by PRAXIS (BD/11153/97) and Xunta of Galicia to two of the authors (E.C.D. Ramalhosa and S. Rı́o Segade).
References (48)
- et al.
Anal. Chim. Acta
(1993) - et al.
Anal. Chim. Acta
(1998) - et al.
J. Chromatogr. A
(1994) - et al.
J. Chromatogr.
(1990) - et al.
Microchem. J.
(1997) Chemosphere
(1999)- et al.
Talanta
(1996) - et al.
J. Chromatogr. A
(1993) - et al.
J. Chromatogr. A
(1994) - et al.
Anal. Chim. Acta
(1990)