1e: Advances in analytical methods for environmental mercury speciation

Analysis of mercury often requires not only measurement of the total mercury content in the sample but also speciation studies of its inorganic (Hg0,Hg+1, Hg+2) and organic compounds (methylmercury). Many years experience of mercury measurements in industry (coal combustion, flue gases) and environmental samples (tissues, sediments, aerosoles, PM) performed by the authors has resulted in several modifications in instrumentation and methodology. The following innovationas will be presented with the examples of applications.

1.Portable system for continuous mercury speciation in flue gas and process gases.Effective reduction of mercury emission requires the application of speciation systems for emission control. In order to operate in flue gas the Nippon Instrument Corporation EMP-2 WLE-8 set was additionally equipped with the fly ash filter, steel probe, transfer line and the tee connector. The portable system was successfully tested in the laboratory and during the industrial tests in Poland and Czech Republic.
2. Automatic system for methylmercury determiniation. In order to facilitate the determination of methylmercury (MeHg) in biota samples with the headspace Tenax trap gas chromatography atomic fluorescence method, an automatic system was designed, constructed and tested. The system consists of Tekran 2500 detector, Gas Chromatograph (Hewlett Packard 5890), a pyrolysis unit, a Tenax trap with a heater, a flow unit, an Electronic Control Unit and a PC with Peak Simple software. A ten-stage mercury determination procedure was elaborated and optimized.
3. Portable Hg+2 ultrasonic calibrator for the control and calibration of mercury speciation systems. The calibrator generates a stable stream of HgCl2 (RSD=2.8% for Hg =28 ug/m3).
4. Innovative tests of non-carbon particulate sorbents in large scale demo installation in coal-fired power plant.
5. Innovations in methodology of performing data collection for the total particulate mercury (TPM) in PM10 and PM2.5 in many urban sites region of Southern Poland and Krakow agglomeration.

Acknowledgment: This work was partially financed by the AGH University grant 11.11.210.244.

The most commonly employed technique for the determination of mercuryusing Atomic Fluorescence Spectrometry in a variety of matrices involves cold vapour generation. Usually, thisisachieved chemically by oxidising all organic mercury species to Hg2+, followed by a reduction to Hg0with SnCl2or NaBH4. However, this method usesa plethora ofchemicals which is costly andprone to error and instrumental issues due to the complex wet chemistry necessary.Another pathway is photochemical vapour generation (photo-CVG), whichhas been used inthe past inconjunction with CV-AFS as an alternative method of cold vapour generation. Previously, photo-CVG has been applied to total mercury measurements using formic acid. It has been shown that formic acid rapidly decomposes when subjectedto UV radiation, giving rise to species which easily reduce all mercuryspeciesto Hg0usingshort irradiation intervals. This methodpromises amuch simplerapproach, whichuses fewer and more environmentally friendly chemicals to give results which are comparableor even more sensitive thanthe cold vapour approach.However, the future of this technique lies in the ability to successfully apply this to mercury speciation. Thus, new LC-UV-CV techniques based on the method arebeing sought.One successful route used amethod whereby a 70/30 MeOH/H2O solution containing mercapto-ethanol as a mobile phase for reverse phase C18 separation,withon-line addition of formic acid: water samples spiked with Hg2+, MeHg+and EtHg+producedgood quality chromatogramsusing this approach. Inthis work, we apply this methodto a wide range of matrices such as wastewater, food, fish, and clinical samples.

A growing body of evidence is showing that conventional, KCl denuder-based measurement systems for atmospheric oxidized mercury are biased low, and that the extent of bias depends on environmental conditions. To make matters worse, many of the oxidized mercury trends that have been observed may, at least in part, be the result of measurement bias. For example, very low ozone concentrations during Arctic mercury depletion events likely decrease oxidized mercury measurement bias, artificially enhancing the contrast between depleted plumes and background air. Similarly, dry, relatively pristine conditions in the upper atmosphere can be expected to decrease bias, clouding our picture of the vertical distribution of oxidized mercury. We must develop new measurement and validation techniques or we will never be able to gain a clear understanding of atmospheric oxidized mercury concentrations and cycling.

In response to this need, we are developing a suite of next-generation instrumentation for measurement and calibration of oxidized mercury. These include (1) an automated calibrator that can deliver stable concentrations of oxidized mercury compounds directly to the inlet of mercury instrumentation, verifying performance under ambient conditions; (2) A high-time resolution dual channel instrument to measure elemental and oxidized mercury, and (3) a GC/MS-based system to identify individual oxidized mercury compounds. We will discuss the development of and results obtained from these instruments. While they are not perfect, these instruments represent a vast improvement over conventional methods, especially because they incorporate routine verification of measurements in real-world conditions.

The assessment of the health and environmental effects of mercury requires the precise and accurate determination of mercury species. Numerous techniques can be used for this purpose, but the low concentrations of mercury species usually found in environmental samples make necessary to include additional steps of preconcentration and/or cleaning in the sample preparation for mercury speciation analysis. Novel methodologies based on the use of new nanomaterials have been currently developed for trace metal analysis. A particularly promising combination is the application of magnetic nanoparticles coated with carbon nanotubes for solid phase extraction. It provides a fast and simple separation of analytes using an external magnet and avoids the time-consuming traditional on-column solid phase extraction procedures. However, this interesting approach did not have been applied up to now for the extraction of mercury species.

Therefore, in this work we have studied the potential of Fe3O4 magnetic nanoparticles coated with different types of carbon nanotubes (single, double and multi-walled) as sorbent material for the determination of mercury species. In all cases, mercury speciation analysis were carried out using gas chromatography coupled to atomic fluorescence detection after derivatization by ethylation. A careful optimization of the conditions affecting both sorption and desorption steps (i.e. type of carbon nanotube, amount of composite, time of sorption or desorption steps, desorbing medium) have been performed aiming to achieve an effective and selective preconcentration and subsequent analysis of monomethylmercury. Single-walled carbon nanotubes magnetic nanoparticles exhibited higher adsorption capacity compared to both double-walled and multi-walled carbon nanotubes. The reuse capability of these nanomaterials during several adsorption/desorption cycles has also been investigated. In addition, the potential interferences and/or interactions between mercury species in the sorption and desorption processes have been studied regarding different mercury species concentration ratios. The results demonstrated that the developed methodology enables not only the preconcentration of monomethylmercury but also the simultaneous clean-up of inorganic mercury.

Acknowledgements: The authors thank the Spanish Ministry of Economy and Competitiveness (CTQ2016-78793-P) for supporting this work.

Coal combustion is the largest anthropenic emission source of mercury in the world, but it is still unclear of the existing form of oxidized gaseous mercury in its flue gas. This study aims to explore the species of mercury in the flue gas of coal-fired power plants by HPLC-AFS. Gaseous mercury in the flue gas of a sedimentation furnace was sampled on nylon, PTFE and quartz membranes. Mercury adsorbed on various membranes was extracted by chromatographic grade acetonitrile and the sampling extrating solutions ( ES1) was used to analyze the mercury species. Six mercury reference substance (Hg0, HgCl2, Hg2Cl2, HgO, HgS and HgSO4) were dissoved into acetonitrile solutions, which served as the reference substance solution, respectively. It was found that mercury capture efficiency of nylon, PTFE and quartz membranes were 72%, 35% and 28%, respectively. The HPLC-AFS result showed all of the three sampling extrating solutions (ES1) have three characteristic peaks at 211-218s (first), 257-268s (second) and 290-291s (third), and the three peak areas follow the order: first (1.1-18.2%) < second (19.5-29.5%) < third (52.4-78.0%). The characteristic peaks of six reference solutions appeared at 253-267s, belonging to the second mercury type. This result indicates that the six mercury references are not main existing form of mercury in flue gas, which is different from the traditional view that the HgCl2 is the main existing form of oxidized mercury in flue gas. The HPLC-MS/MS will be used to further identify the mercury species in flue gas in next stage.

Developments of new techniques for determination of monomethylmercury are primarily focused on high tech equipment, including mass spectrometry. However, methods based on mass spectrometry and other currently available instrumental techniques for speciation are based on principles that require clean laboratory conditions, expensive gases and a lot of experience. For most routine laboratories such conditions are unsuitable which results in the fact that speciation analysis are only done in expert laboratories. Moreover, recently adopted Minamata convention calls for simplified and easy-to-use methods that allow speciation of mercury world-wide.

Simple extraction method that can be used to measure monomethylmercury (MMHg) in biological samples has been developed. The method is based on the simultaneous extraction of Hg compounds into toluene as their dithizonates. Hg(II) and MMHg dithozonates are then separated on silica gel column and selectively collected into separate test tubes followed by digestion with mixture of acids (HNO3/HClO4 and H2SO4) and measured by CV AAS. The average repeatability and reproducibility of the method are 1.5% and 4.3%, respectively. Expanded relative standard uncertainty according to Nordtest is 12.8% using coverage factor k=2.

The method separates inorganic and organic mercury quantitatively in biological samples. Separation is very good also in samples where the percentage of MMHg is low. The results obtained were verified by the analysis of Certified Reference Materials and comparison of the results obtained by alternative methods based on ethylation and CV AFS detection.

Compared to other methods this method is simple and inexpensive from the point of view that the instrumentation which is used for total mercury measurements can be used to measure organomercury compounds. The detailed analytical methods as well as the results of MMHg in biological samples obtained using this method will be presented.

Mercury is one relevant element for many researcher due the potential impact to the enviromental as well as to humans. It recent years, many authors have determined the total concentrations and different species present in soils, atmosphere, water, biota mining and petroleum activities¹, particularly in petroleum business. Hg can be present in natural gas, crude oil and its products. The most dominant species are Hg⁰ and Hg particulate, most likely associated to HgS^2-4. Recently, some researchers reported that particulate HgS could be the most abundant species in stored petroleum hydrocarbons⁵. Additionally, when crude oil is extracted, the water used in the oil extraction and production consequently may contain Hg species, as Hg particulate. Researchers are now interested in determining if Hg nanoparticles are present in these waters. In an attempt to characterize these Hg nanoparticles, single particle inductively coupled plasma mass spectrometry (sp-ICP-MS) has been employed. Sp-ICP-MS allows simultaneous determination of particle number, concentration and size, as well as measuring the dissolved metal concentration.

References

[1] Sakamoto, M., Murata, K., Kakita, A., Sasaki, M., A review of mercury toxicity with special reference to methyl mercury. In Environmental Chemistry and Toxicology of Mercury, Liu, G., Cai, Y., O'Driscoll, N., Ed. John Wiley & Sons, Inc.: New Jersey, 2012; pp 501-516.

[2] Wilhelm, S. M.; Bloom, N., Mercury in petroleum. Fuel Processing Technology 2000, 63 (1), 1-27.

[3] Bloom, N. S., Analysis and stability of mercury speciation in petroleum hydrocarbons. Fresenius' Journal of Analytical Chemistry 2000, 366 (5), 438-443.

[4] B. Bouyssiere, F. B., L. Savary, R. Lobinski, Speciation analysis for mercury in gas condensates by capillary gas chromatography with inductively coupled plasma mass spectrometry detection Journal of Chromatography A 2002, 976, 431 - 439.

[5] Gajdosechova Z., Boskamp¹, M.S. Lopez-Linares F., Feldmann¹ J., Krupp E. M. Energy and Fuels, 2016, 30 (1) 130–137

A novel analytical methodology for reagent-free sampling and extraction of mercury from human urine was developed and optimized. The method is based on selective separation of Hg from fresh urine sample onto active nanogold-coated silica material by highly efficient solid-phase extraction. The adsorbent for mercury extraction was prepared by chemical reduction of Au(III) to Au(0) in presence of silica. Subsequently, immobilization of formed gold nanoparticles is achieved by calcination. Characterization of this material by TEM, REM and TXRF confirmed the successful deposition of gold nanoparticles onto the silica, revealing an average particle diameter of 52 nm. Special in-house designed glass devices allow easy and quick extraction and separation of Hg from urine within only 10 minutes. After thermal desorption of Hg from the extractant, detection is performed by atomic fluorescence spectrometry (AFS). The feasibility and validity of the optimized, reagent-free approach was confirmed by recovery experiments in spiked real urine (recovery rate 96.13 5.34 %). Furthermore, mercury concentrations found in three (non-spiked) urine samples originating from occupationally exposed persons were compared to values obtained by application of the reference methods cold vapor - atomic absorption spectrometry (CVAAS) and CV-AFS. A very good agreement of the found values reveals the validity of the proposed approach. The limit of detection (LOD) was found to be as low as 0.004 ug Hg L-1 and a high reproducibility with a relative standard deviations 4.2 % (n=6) is given. Moreover, storage of the samples for up to one week at an ambient temperature of 30°C reveals no analyte losses or contamination. In addition, investigations revealed that calibration of one exemplary collector provides reliable quantification of all other collectors filled with extractant material coming from the same synthesis batch. Moreover, due to the re-usability of the material the proposed method is cost-efficient. In addition, it enables easy-to-handle on-site extraction of mercury traces from human urine ensuring at the same time reagent-free sample stabilization, providing quick and safe sampling, which can be performed by untrained persons.

Aquatic mercury can pose a significant risk to ecological and human health when it is converted to the more toxic and bioaccumulative methylmercury (MeHg) by anaerobic microorganisms near the sediment-water interface. Benthic animals exposed to MeHg in this zone represent an important link in the process of MeHg bioaccumulation in aquatic food webs. As such, quantitative measurements of the bioavailability of MeHg to benthos are critical for risk assessment of contaminated sites. To date, no passive sampling strategy has achieved wide acceptance for generating such measurements. The diffusive gradient in a thin film (DGT) device, which operates in a kinetic accumulation mode, faces questions about the potential for oversampling, as well as the identity of the MeHg species it samples and their relevance to bioavailability. We are working to develop a novel passive sampling technology to emulate the steady-state (pseudo-equilibrium) mode of MeHg bioaccumulation by benthos. A target range of sampler partitioning was set at log K between 3.0 and 4.5 in consideration of analytical detection limits, the logistics of sampler deployment in the field, and typical sediment-water partitioning constants. A variety of custom polymers with either reduced sulfur chemical functionality or embedded activated carbon (AC) particles was prepared and evaluated in a series of increasingly environmentally realistic experiments. In mildly saline water isotherms, many of the materials showed strong, log-linear partitioning of MeHgCl across a relevant range of concentrations. Partitioning of MeHg complexed with dissolved organic matter was decreased but still in or near our target range (log K = 2.83 to 3.31). In slurries of contaminated soil, polymer-predicted water concentrations (Cw) were within factors of one to five of directly measured water samples. In beaker soil microcosms, accumulation of MeHg by our AC-based polymer successfully reflected temporal variations in Cw as well as reductions due to amendment of soil with AC. Polymer-derived and directly measured Cw agreed equally well at 8, 14, 21, and 28 d of exposure. Across experiments, partitioning by this polymer was notably consistent in the saline water, slurry, and unamended beaker soil (log K = 3.41 to 3.78). An experiment currently being planned will attempt to correlate MeHg accumulation by polymers and by the amphipod L. plumulosus exposed side-by-side in sediment microcosms. The polymer data will be used as input for a bioaccumulation model to evaluate the technologys predictive capability.