REPOSITORY > RESULTS

Doctoral dissertation

Determination of dissolved elemental mercury in surface and industrial waters

Author(s): Ermira Begu (Author), Milena Horvat (Supervisor)

Thesis defense date: 22.04.2016

Organization: MPŠ - Mednarodna podiplomska šola Jožefa Stefana

PID: 20.500.12556/ReVIS-15053

Views: 2 | Downloads: 0

Abstract

Mercury (Hg) is a toxic metal that occurs naturally in the Earth's crust principally as the ore cinnabar, HgS, but centuries of human activities such as mining and fossil fuel burning have been mobilizing increasing amounts in the atmosphere, oceans, and terrestrial systems. It is emitted to the atmosphere from a variety of natural and anthropogenic sources, after which it redistributes globally via a complex combination of chemical, physical, and biological processes throughout the atmosphere and the rest of the environment. Due to its volatility and long residence time, it can travel long distances and is therefore considered a global pollutant. The recently signed global treaty “Minamata Convention” acknowledged the negative impacts of mercury on humans and ecosystems at the global scale and it primarily aims at the reduction of mercury emissions, particularly from current anthropogenic sources. Among these, coal burning power plants represent the greatest source, followed by other industrial sectors using high temperature processes. In recent years clean technologies have been developed, which significantly reduced the emissions of Hg(0) into the atmosphere. In the coal burning sector wet desulphurisation equipment (WFGD) with forced oxidation has shown to be very effective. Removal efficiency is commonly monitored by Hg measurements in the emitted flue gas, but these measurements are costly and time consuming. These methodologies are still not fully optimized; therefore, comparability of the results is questionable.
In the WFGD systems elemental Hg is oxidized and trapped in the slurry solution; however, a small fraction of dissolved elemental mercury (DEM) leaves the WFGD and is emitted to the atmosphere through the stack gas. It has been suggested that by accurate and continuous measurements of DEM in the slurry solution it is possible to follow the oxidation/reduction processes in the WFGD slurry and estimate the quantity of Hg emitted to the flue gas. In order to simplify the protocol for checking the emissions of Hg(0) through the flue gas, we proposed the determination of DEM in the scrubber solution as a measure of Hg removal efficiency.
In natural waters Hg exists in different forms such as inorganic mercury (Hg(II)) compounds, dissolved elemental mercury (DEM), monomethylmercury compounds (MeHg) and dimethylmercury (DMeHg). These forms can be present in dissolved forms and/or adsorbed on suspended particles. DEM and DMeHg are gaseous forms at room temperature and are present as dissolved gaseous Hg (DGM) forms. While DMeHg is rarely found in natural waters, DEM can represent a significant fraction of total Hg in surface waters. The evasion of DEM as Hg(0) from surface waters represents one of the most important sources of elemental Hg in the global atmosphere. Hg(II) compounds are frequently named reactive mercury (RHg), which is an operationally defined fraction of mercury determined after reduction of Hg(II) in aqueous solution by stannous chloride. The methods for DEM determination in surface waters as well as RHg have not been standardized as yet.
The main objective of the present PhD was to develop and validate a simplified semi-continuous method which enables DEM and RHg measurements in both natural waters and in the scrubber solution of the WFGD.
Analytical methods used for determining DEM are conventionally based on the removal of Hg(0) by purging, pre-concentration on gold and detection by either cold vapour atomic absorption (CV AAS) or atomic fluorescence spectrophotometry (CV AFS). At present there is no agreed protocol for the measurement of DEM in aqueous samples. A new method has been developed in this work which is based on continuous stripping of DEM by mercury-free nitrogen in a flow injection mode and detection by CV AAS. The partitioning of Hg(0) between aqueous and gas phases is largely dependent on the composition of the former. Moreover, calibration using the standard addition method was not possible due to the reactivity of DEM and RHg introduced from calibration solutions. Calibration is therefore done by reference measurements using a manual method for DEM involving quantitative removal and CV AFS detection. DEM was then determined in the water sample by applying the partitioning factor. The optimized method is precise, sensitive and linear over a wide concentration range. It has provided comparable results with the manual method when applied on board a research vessel in the Mediterranean Sea (20 to 50 pg L⁻¹) and during a pilot laboratory-scale experiment on industrial aqueous media from wet flue gas desulphurisation equipment (2 to 300 ng L⁻¹).
By adding an additional line in the system which enables a flow of tin chloride (used as a reducing agent) to be introduced continuously into the system, the method was successfully adapted and applied also for RHg measurements in the scrubber solution. The existing manual method was used as a reference method along with the semi-continuous one. Good comparability of the data was obtained regardless of the sample matrix. We were able to track changes in Hg chemistry depending on different physical/chemical factors of the natural and industrial water media (scrubber solutions). Such measurements have not been conducted yet in scrubber solutions of WFGD due to large difficulties related to the sample matrix. The intention of this PhD, however, was not to discuss the parameters affecting the chemistry of Hg in aqueous media but to demonstrate the ability of the analytical method to accurately track the changes. In natural waters this is important in order to better estimate Hg(0) emissions from water surfaces into the atmosphere and to determine the availability of Hg(II) to be methylated. In the scrubber solutions, the accurate determination of DEM and RHg can be used as a tool to improve the efficiency of cleaning technologies.
The PhD also addressed some aspects of the analytical protocol that contribute considerably to the uncertainty of the results. These include (i) stability of DEM and RHg in aqueous samples, (ii) passivation of gold traps used for pre-concentration of Hg(0) prior to detection on CV AAS, and (iii) calibration using gaseous Hg at saturated conditions and Hg(II) in aqueous media.
(i) As both species, DEM and RHg, are unstable, their stability in relation to storing conditions including temperature and exposure to light and the addition of storage agents was studied. The tested water matrices included freshwater and seawater samples and scrubber solutions. The results showed good performance for freshwater samples and a maximum of three hours was found to be not critical for DEM measurements when stored in glass bottles in dark conditions. The use of HNO₃ and AuCl₃ as agents for preserving the samples demonstrated the oxidation of Hg(0) into Hg(II) and the formed Hg(II) was stable for 2–3 days. Preservation of RHg and DEM is therefore not possible.
(ii) The most widely used material for mercury pre-concentration is gold-coated silica oxide, which tends to be passivated over an extended period of time. This results in irreproducible and lower signals during measurements. It was found that this is due to inhomogeneity of the coating of the surface and reactivity of the SiO₂ substrate at higher temperatures. A new substrate, alumina, was used and coated with a nanolayer of Au and was tested with very promising results; however, longer-term testing is needed to ascertain the resistance towards passivation. The obtained results were very promising, but further work is needed. As humidity was found to affect the performance of gold traps and the measurement system, various drying procedures were tested, among which soda lime traps showed the best results.
(iii) Recent detailed studies have shown that Hg(0) measurements are questionable due to systematic biases in calibration. Several empirical equations are available to describe the vapour pressure of mercury at a given temperature, which is the basis for calibrating Hg instruments. The agreement between these equations differs by 5% or more. In order to remove the dependence of mercury vapour measurement on these empirical equations, and to provide stability, comparability and coherence for mercury vapour measurements, the calibration in the present work was also done using an Hg(II) calibration solution certified for Hg content. The results obtained by the two methods provided comparable results; therefore, instrument calibration was mainly done with Hg(0) vapour.
Opomba: Popravil sem razmike, podvojene dele besedila, očitne napake pri pretvorbi iz PDF in zapise kemijskih formul ter enot. V izvirniku manjka številčna vrednost pri navedbi deleža DEM v površinskih vodah, zato je nisem ugibal.

Attachments

Cite this work