1i: Comparable measurement results for mercury analysis and speciation

Within the “Traceability for mercury measurements” project (MeTra; European Metrology Research Programme) a mercury vapour generator is being developed to establish traceability of mercury vapour measurement results, based upon a gravimetric approach, for ambient air levels as well as higher concentrations.

Current measurement capabilities are maintained at levels of 0,25 – 350 μg Hg/m3, whereas the aim of the novel gravimetric primary standard is to realize metrological traceability for the range 5 ng Hg/m3 - 60 μg Hg/m3, with a target measurement uncertainty of 1 %. This to cover key requirements for ambient air monitoring (1 - 2 ng Hg/m3), health-based exposure standards (50 ng Hg/m3), concentrations relevant to stationary source emissions (upwards of 1 μg Hg/m3) and the minimum alveolar concentration value (20 μg Hg/m3).

The development and characterization of the primary vapour generator will be presented. Furthermore the results will be presented of two comparisons held to demonstrate the robustness and comparability of the novel primary standard. One comparison was performed against the Dumarey vapour pressure equation, while another comparison was performed against current calibration facilities present at national metrology institutes.

The newly developed mercury vapour generator will contribute to comparable measurement results of mercury vapour at ambient and background air levels, and also to higher safety standards and cost reductions in e.g. the LNG field, where aluminium main cryogenic heat exchangers are used which are particular prone to corrosion caused by mercury.

A standard addition system was developed that utilizes the Gaseous Elemental Mercury (GEM) output from a Tekran(r) 2537X permeation source and ports it to the entry glassware of a speciated mercury system. Valve timing and control is performed using National Instruments LabVIEW(r) software. At our Atmospheric Mercury Network (AMNet) site in Beltsville, MD standard addition spikes are made into the inlet sample flow at points upstream and downstream of the inlet impactor frit during both GEM sampling and desorption periods. This allows ambient and zero-air matrix effects to be investigated, and provides a measure of any losses of GEM in the complete sampling train. From these spikes an effective response factor is derived and compared to that obtained using calibrations through the internal flow path of the 2537X analyzer. Applying interpolated effective response factors to each five-minute integrated ambient GEM sample reduces the within-hour standard deviation by correcting matrix-effect trap bias, and also results in an increase in hourly-averaged GEM measurements. Changes in standard addition instrument response are correlated with meteorology and trace gas measurements.

The assessement of Mercury impact on ecosystems and human health is of major concern and European Directives are waiting for a real traceability procedure on its measurement as well as the reduction of the associated uncertainties. Reliability and comparability of mercury measurements are essential to assess concentrations and trends of this highly toxic element. The European Project JRP ENV51 MeTra Traceability for Mercury Measurements aims to meet these needs in the different environmental compartments.

Recently, the exploration of mercury isotopes allows the discrimination of Hg sources for a better understanding of its biogeochemical cycle in the environment. Many factors control Hg isotopes fractionation in the environment such as biological or photoreduction processes leading to a specific isotopic fingerprint in the final product. To ensure the quality and validate these measurements, it is necessary to understand and control all the fractionation processes which can occur during each steps of the analytical chain (sampling, storage, sample prep) but also during the sample introduction (CVG/GC) and the analysis on the MC-ICP-MS.

To evaluate of all these potential sources of fractionation, we have controled in detail the whole analytical chain for Hg factionation in environmental samples. At first, we have studied the storage condition of the sample by measuring several NIST 3133 solution over a long time period. The accuracy and precision of different sample preparation (digestion/derivatization) and different sample introduction (several CVG, GC) were assessed by comparing HotBlock, Microwave and High Pressure Asher digestion as well as derivatization step which is critical for compound specific isotopic analysis which can lead to inaccuracy on total mercury budgets. Further, depending on digestion methods used or reactives introduced, matrix effects can occur during the Hg isotopes analysis. We have therefore investigated several compounds added to standard solution such as oxydant (H2O2, BrCl) or organic matter in order to understand the potential fractionation induced during measurements. Finally, mass discrimination in the MC-ICP-MS can play a major role on isotopic fractionation and mass bias correction by Tl internal standard, which can be also mass discriminated, can lead to an underestimation of the analytical uncertainty. This implies that different mass bias correction methods were explored by comparing sample-standard bracketing and internal standard correction method. We will present and discuss all the potential biais that may be introduced during Hg isotopic fractionation measurements in environmental samples all along the analytical chain.

Field techniques for measuring Hg in contaminated soils have always been an appealing prospect. In situ field measurements have the potential to provide significant savings in sampling, shipping, and laboratory costs. More importantly, the real-time reporting of field instruments can greatly increase the efficiency of monitoring efforts by quickly locating hotspots, making real-time decisions to develop more targeted sampling strategies, and efficiently delineating contaminated areas in the field.

Handheld X-ray fluorescence spectrometers (XRF) provide a rugged and portable style that is ideal for in situ monitoring. Many researchers have successfully used XRF for field screening of Pb, As, and other metals at contaminated sites, however, previous attempts at using handheld XRF for Hg analysis have been disappointing. Devices have typically provided insufficient sensitivity or poor accuracy and precision.

In this study, we demonstrated that handheld XRF can be successfully used in the field to measure Hg in soils with accuracy, precision, and sensitivity. Rather than using manufacturer configurations and calibrations, this study developed a Hg-specific calibration set and used Hg-specific configurations to optimize sensitivity and increase the accuracy of Hg detection. This study measured 239 bank soil samples in the field at the South River remediation site in Waynesboro, Virginia. After field analysis using the handheld XRF, samples were shipped to a laboratory for confirmatory Hg analysis using EPA Method 7471A.

The handheld XRF was able to achieve a method detection limit of 7.4 mg/kg Hg with a 60-second analysis time, sufficient for comparison to most risk-based soil screening levels. Accuracy of the XRF method was determined by analysis of spiked Hg samples and by comparison of field results to laboratory Method 7471A results for the 239 collected samples. Percent recovery for spiked samples averaged 102%, and the regression of spiked Hg levels to XRF reported values achieved an R2 value of 0.999. XRF and Method 7471A results were highly correlated with an R2 value of 0.934. Precision of the method was determined by the analysis of field triplicates. The median CV (coefficient of variation) for XRF field triplicates was 18%, compared to a median CV of 19% for field duplicates measured using Method 7471A. These findings demonstrate that the XRF method, if calibrated and optimized properly for Hg, can be used to quickly and reliably analyze the Hg content of soils in the field with sufficient accuracy, precision, and sensitivity.

The launch of the United Nations Environment Programme Global Legally Binding Treaty on Mercury, will lead to the establishment/strengthening of Mercury monitoring efforts of the countries in different environmental compartments, in order to assess environmental mercury contamination, as well as to control the efficiency of the control measures undertaken. Therefore, Mercury monitoring will become an integral part of all marine monitoring programmes implemented at national or/and regional level around the world. Because of the expected proliferation of the generated Mercury monitoring data, and taking into consideration the difficulties related to the accurate analysis of Mercury and its species in marine samples, the use of recommended analytical methods by the laboratories involved in this effort, as well as the strengthening of the quality assurance of their data, is a necessity for the effective use of the generated data for environmental assessments and decision making. Therefore, many developing countries will need assistance to build their capacity for accurate and representative mercury analysis in the marine environment.

To assist countries in the strengthening of their Mercury data quality assurance, the Marine Environmental Studies Laboratory of IAEA is developing fit-for-purpose recommended analytical methods for Mercury and Methyl Mercury in marine sediment and biota, organises Inter-Laboratory Comparison (ILC) exercises and produces relevant Certified Reference Materials to be used by laboratories involved in mercury monitoring in the marine environment. Recently MESL has produced one new CRM for Mercury and Methyl Mercury in marine sediment and two CRMs in biota samples (clam and scallop). The Recommended Methods for Mercury and Methyl Mercury, as well as the relevant CRMs in marine sediment and biota, will be outlined in the presentation.

Gaseous elemental mercury (Hg0) and oxidized mercury (HgCl2) reference standards are integral to the use of mercury continuous emissions monitoring systems (Hg CEMS) for regulatory compliance emissions monitoring. EPA and NIST have collaborated to establish the necessary procedures for establishing the required NIST traceability of commercially-provided Hg0 and HgCl2 reference generators. However, a quantitative disparity of approximately 7-10% has been observed between commercial Hg0 and HgCl2 reference gases which currently limits the use of (HgCl2) reference gas standards. The source of this disparity is the focus of considerable debate. Moreover, the actual disparity has not been accurately quantified. This presentation will highlight the approach and results of a joint NIST/EPA study to accurately quantify the true concentrations of Hg0 and HgCl2 reference gases produced from high quality, NIST-traceable, commercial Hg0 and HgCl2 generators. The results of this study will not only quantify the disparity, but will help provide an understanding of the source of this disparity, if confirmed.

Mercury porewater measurements are an important indicator of mobility and availability of mercury in natural systems. Mercury associated with the solid phase is typically not as available for uptake or methylation as mercury present in porewater in either dissolved or suspended phase. Techniques for evaluating porewater mercury concentrations include direct withdrawal of porewater, collection of minimally disturbed sediment cores for processing via porewater displacement or centrifugation, as well as passive sampling techniques including dialysis membrane sampling devices (peepers), and flux based diffusive gradient in thin film samplers (DGTs). Direct withdrawal is difficult in fine-grained low permeability media and often difficult to ensure that only porewater is collected. Porewater expression from cores can be effective but care must be taken to avoid disturbance of the core. Centrifugation is often used but resuspend colloidal and suspended particles that can then scavenge dissolved mercury during filtration. All of these methods suffer from sensitivity to the filtration approach (e.g. sensitivity to filtration size).

As compared to previous techniques, passive samplers have advantage of greatly reducing the disturbance of the physical and chemical conditions present at the site. Dialysis membrane passive samplers typically require collection of relatively large volumes potentially limiting vertical resolution and lengthening sampling time. Diffusive Gradient in Thin-Film (DGT) samplers require relatively short exposure times and exhibit low detection limits by concentrating mercury or methyl mercury on a sorbing resin. A concern with both methods is the question of the relationship of mercury speciation and the measurement obtained by the sampler. In our research, we have investigated the mercury species and complexes most likely to be measured by DGTs and used a combination of microbial and macrobenthic assays to show that these species and complexes are the most likely to be biologically relevant.

This paper compares these various approaches to the measurement of mercury in porewater. The relative strengths and weaknesses of each approach and their biological relevance will be identified. The specific advantages of DGTs for the assessment of mobile and bioavailable mercury will be emphasized and the rationale behind expanding use of DGTs identified.