GB2548889A

GB2548889A - Flow device

Info

Publication number: GB2548889A
Application number: GB1605503.0A
Authority: GB
Inventors: Begu Ermira; Horvat Milena; Shlyapnikov Yaroslav; Stergarsek Andrej; Frkal Peter; Kotnik Joze
Original assignee: Institut Jozef Stefan
Current assignee: Institut Jozef Stefan
Priority date: 2016-03-31
Filing date: 2016-03-31
Publication date: 2017-10-04
Also published as: SI25182A

Abstract

A flow device for extracting a dissolved analyte from a liquid sample, comprising (i) a liquid sample inlet 1, (ii) a carrier gas inlet 2, (iii) an equilibration section 5, where flows of liquid sample and carrier gas coming from the inlets are mixed and equilibration of the content of analyte in the two phases takes place, and (iv) a gas-liquid phase separator 6 downstream of the equilibration section, for separating the gaseous phase with the extracted analyte from the liquid phase of the mixture. Mixing of the two phases may take place in a mixing chamber prior to entry into the equilibration section, which may comprise a helical tube. The lower part of the gas-liquid separator may comprise a U-shaped tube 7 branching off to two outlets, one for the treated liquid sample and another for residual carrier gas. The upper part thereof may be connected to a device 9 for measuring the analyte concentration in the carrier gas, whereby its concentration in the liquid sample may be determined via the value of partition coefficient. Preferably the analyte is elemental mercury, the liquid sample is wet flue gas desulfurisation scrubber solution and the carrier gas is nitrogen.

Description

FLOW DEVICE

Technica! Field

The present invention applies to the field of environmental technology and in particular relates to measurements of an analyte such as mercury in solutions such as wet flue gas desulfurisation (WFGD) suspensions where mercury is dissolved as part of industrial exhaust purification procedures.

Background

Various toxins and other harmful materials may exist in a dissolved form in liquids, particularly aqueous solutions. Those solutions may be naturally occurring and contaminated, or may be man-made.

In either case, the presence of unwanted materials dissolved in those solutions can be of environmental concern. It is therefore beneficial to remove such materials from the solutions, and to find ways of measuring the content present.

Many such methods exist. However, there remains a need for useful devices and methods which permit continuous (or semi-continuous) monitoring of the content of such harmful materials and the change of that over time.

The present invention relates to this field, it is described in the context of mercury, a commonly cited toxin, however its wider applicability will be apparent.

Mercury in aqueous systems can be present in different chemical forms. Of these, dissolved elemental Hg(0) (DEM) is of great importance because it can readily be partitioned between air and water. Mercury is of global concern since it enters the environment through natural and anthropogenic sources. There are several Hg species in constant interchange between atmospheric, aquatic and terrestrial environments, forming the global biogeochemicai cycle of Hg. in water systems, the most representative species include dissolved gaseous elemental Hg (DEM), dimethylmercury (DMeHg), inorganic Hg (Hg(il)) and methylmercury (MeHg) while in the atmosphere, Hg is mostly found in its elemental form (Hg(0)), with traces of oxidized Hg(ll) forms and Hg bound to particles.

Due to its toxicity, standards for the emission of Hg into the atmosphere are becoming stricter. Coal burning has been confirmed to be among the most important anthropogenic sources and procedures are needed not only to safely prevent Hg release from industrial installations but also to monitor the efficiency of such procedures.

About 70% of coal burning thermo power plants (TPPs) today are equipped with a wet flue gas desulfurisation technology (WFGD) which was originally developed to reduce SO2 emissions but later adapted for removal of Hg as well. One of the most investigated options for achieving high levels of removal of Hg in the WFGD is based on modified chemistry in the scrubber solution that leads to increased oxidation of Hg(0) to Hg(ll) [see, for example, Stergarsek, A. ei al, Fuel. 107, 183 (2013); Stergarsek, A. et al. Fuel. 89, 3167 (2010); P. Cordoba, Fuel. 144, 274 (2015); and Krzyzynska, R. et al, J. Air Waste Manage. Assoc. 62, 212 (2012)].

Hg(ll) can be captured effectively by taking advantage of its high solubility in water. In order to control oxidation/reduction processes in the VWGD, it is important to control DEM levels in the scrubber solution as an alternative to measuring Hg(0) in the flue gas. DEM concentrations are relatively high in scrubber solutions but rarely reported in the literature, probabiy due to the lack of tools for its speciation in such samples.

In natural waters, DEM is present at much lower levels that usually do not exceed 0.2 ng L"k The presence of DEM in natural waters is related to the exchange of Hg between the atmosphere and water, to the photo-reduction and/or oxidation of Hg in surface waters, to chemical reduction, to biologically and bio-geochemicaiiy mediated reduction processes, and to natural geogenic sources. Most surface waters are supersaturated with Hg(0) relative to the atmosphere. Due to its gaseous nature, Hg(0) escapes, leading to increased Hg levels in the atmosphere. Re-emission from oceans is estimated to be an important source of Hg on the global scale, with around 2000-2950 i y~7 Further, the oxidation of Hg(0} in natural waters serves as a source of Hg(ll), which is reactive and can be converted into MeHg, one of the most toxic Hg forms that is bioaccumulated and biomagnified in food webs. DEM has been reported in surface and deep oceanic waters. DEM in water media is usually measured manually by purging Hg(0) at room temperature from a sample by a gas flow and trapping it in a gold (Au) trap, then determining it either by cold vapour atomic absorption spectrometry (CV AAS) or cold vapour atomic fluorescence spectrometry (CV AFS). The problems when measuring DEM are mostly related to the instability of the species (gas), the low concentrations (mainly in natural waters) and the nature of the sample matrix.

Existing manual methods have the potential to measure DEM accurately but the analysis time is relatively long. Further, due to its discontinuous nature, it is not useful for monitoring rapid changes of concentration of Hg(0) in aqueous media.

For these reasons, semi-continuous/continuous methods have been developed that enable real-time measurements to be made on natural surface waters [see, for example, Krzyzynska, R, et ai, J. Air Waste Manage. Assoc. 62, 212 (2012); O’Driscoll, N. J. et al, Sci. Total Environ. 304, 285 (2003); Amyot, M. et al, Anal. Chim, Acta. 447, 153 (2001); Lindberg, S. E. et al. Biogeochemistry. 48, 237 (2000); Andersson, M. E. et al. Anal. Bioanal. Chem. 391, 2277 (2008); Stergarsek, A. et al. Fuel. 87, 3504 (2008)].

Removal of gaseous mercury from gases by passing the latter through scrubber or absorber solutions has been discussed often in the prior art. Treatment of slurries in WFGD systems is also discussed in the direction of lowering the redox potential of the slurries or removal of mercury via application of UV light and / or an oxidation step, or a precipitation step. CA2422103A1 describes adsorption of mercury in aqueous solutions and subsequent conversion of DEM to the gaseous form for subsequent analysis. However this is achieved through reduction reactions. Furthermore, in this document fractionation is followed by measurement of both fractions.

The state of the art includes numerous methods for measuring the concentration of Hg in liquid phase of aqueous, for example WFGD, systems, performance of which is complex and time-consuming. There is a need for less complex, less time consuming, but still reliable procedures enabling continuous monitoring of the concentrations of Hg in aqueous, such as WFGD, systems.

General Notes A number of patents and publications are cited herein in order to more fully describe and disclose the invention and the state of the art to which the invention pertains. Each of these references is incorporated herein by reference in its entirety into the present disclosure, to the same extent as if each individual reference was specifically and individually indicated to be incorporated by reference.

Throughout this specification, including the claims which follow, unless the context requires otherwise, the word “comprise” and variations such as “comprises” and “comprising,” will be understood to imply the inclusion of a stated integer or step or group of integers or steps but not the exclusion of any other integer or step or group of integers or steps.

It must be noted that, as used in the specification and the appended claims, the singular forms “a,” “an,” and “the” include plural referents unless the context dearly dictates otherwise. Thus, for example, reference to “a surface layer formation step” includes combinations of two or more such surface layer formation steps, and the like.

Ranges may be expressed herein as from “about” one particular value, and/or to “about” another particular value. Wtien such a range Is expressed, another embodiment includes from the one particular value and/or to the other particular value. Similarly, when values are expressed as approximations, by the use of the antecedent “about,” it will be understood that the particular value forms another embodiment.

This disclosure includes information that may be useful in understanding the present invention. It is not an admission that any of the Information provided herein is prior art or relevant to the presently claimed invention, or that any publication specifically or implicitly referenced is prior art.

Summary of the Invention

The present invention relates to a flow apparatus or flow device through which a sample to be analysed can flow.

In a first aspect, the present invention provides a flow device for extracting a dissolved analyte from a liquid sample, the device comprising: a liquid sample inlet, for receiving a sample flow at a sample flow rate; a carrier gas inlet for receiving a carrier gas flow at a carrier gas flow rate; an equilibration section, in fluid communication with the liquid sample inlet and the carrier gas inlet such that a sample flow from the liquid sample inlet and a carrier gas flow from the carrier gas inlet flow into the equilibration section, for equilibration of the content of the analyte in the liquid sample and In the gas phase; and a gas-liquid phase separator, in fluid communication with an outlet from the equilibration section such that a flow from the outlet of the equilibration section flows into the phase separator, for separating the analyte in the gas phase from the liquid mixture output from the equilibration section. in use, a sample is pumped or otherwise introduced into the liquid sample inlet. That sample contains an analyte of interest, or, at least, is suspected of having such content. Also flowing into the device is a carrier gas. These two flows enter an inlet of an equilibration section. In that section, the joined flow permits partitioning or equilibration of the content of the dissolved analyte between the liquid phase and the gas phase. A residence time within the equilibration section can be controlled by design or modification of the equilibration section and the flow rate of the sample and carrier gas, for example.

The mixed, equilibrated sample and carrier gas then flow out of an outlet from the equilibration section to the gas-liquid phase separator, through an inlet of the gas-liquid phase separator. That separates off the gas phase containing a portion of the analyte from the sample. That separated gas phase can be analysed to find the content of the analyte.

By equilibrating or partitioning the dissolved analyte between the gas and liquid phases in the equilibration section and detecting the analyte in the resultant gas phase, the total content of analyte in the sample can be derived based on the partitioning factor or partition coefficient of the iiquid-gas biphasic system used.

The analyte is therefore generally one which can be present in both dissolved {liquid phase) and free gas phase forms in equilibrium. That is, it may be one which is partitioned or capable of being partitioned between the liquid phase of the sample and the gas phase of the carrier gas.

In certain embodiments of the invention, the dissolved analyte is elemental mercury, that is, DEM, Hg(0). in certain embodiments, the carrier gas inlet and the liquid sample inlet are in fluid communication such that a sample flow from the liquid sample inlet and a carrier gas flow from the carrier gas inlet join to form a single flow before flowing into the equilibration section. This allows equilibration in the equilibration section to proceed efficiently.

For example, the device may further comprise a mixing chamber in fluid communication with the liquid sample inlet and the carrier gas inlet, such that the sample flow and carrier gas flow both enter the mixing chamber, where they join and mix; the mixing chamber also being in fluid communication with the equilibration section such that the mixed liquid sample and carrier gas flow can flow into the equilibration section.

In such an embodiment, a single flow enters the inlet of the equilibration section. Use of the mixing chamber, to mix the flows of sample and carrier gas, may again increase the efficiency of equilibration.

The equilibration section may have various shapes and sizes. For example, it may comprise a helical tube to increase residence time and mixing. The residence time within the equilibration section may be, for example, 2 to 40 minutes. The residence time may be, for example, 5 to 20 minutes, for example 7 to 15 minutes, for example 8 to 13 minutes.

The phase separator may comprise an upper part through which the analyte in the gas phase flows or through which it is extracted and a lower part through which the remaining sample and carrier gas output from the equilibration section flows or is extracted. For example, the upper part and lower part of the phase separator may each comprise a tube, the two tubes diverging from a point proximal the outlet of the equilibration section.

The lower part of the phase separator may terminate in an outlet for the sample and an outlet for the carrier gas. The lower part of the phase separator may comprise a U-shaped tube section.

The upper part of the phase separator may comprise an outlet for analyte vapours, or a gas phase including the analyte, and may be suitable for connection to a device for measuring the analyte concentration. In certain embodiments, such a measurement or detection device is present. That is, In some embodiments the upper part of the phase separator is connected to a device for measuring the analyte concentration.

The present invention therefore also relates to an apparatus for analysing or determining the content of a dissolved analyte in a sample, comprising a device according the present invention and a detector for the analyte connected to the analyte vapour or gas phase outlet of the phase separator.

The detector, or device for measuring the analyte concentration may be an atomic fluorescence spectrometer or an atomic absorption spectrometer.

Some or all of the device may be made from or internally coated with borosiiicate glass or polytetrafiuoroethylene. This prevents the analyte, such as mercury Hg{0), from depositing or adhering within the device which would of course affect the readings obtained.

The present invention also relates to a method for extracting a dissolved anaiyte from a liquid sample, the method using a device according to the present invention, the method comprising the steps of: passing the sample into the liquid sample inlet at the sample flow rate; passing a carrier gas into the carrier gas inlet at a carrier gas flow rate; flowing the sample and the carrier gas through the equilibration section, to equilibrate the content of the analyte in the sample and in the gas phase; and extracting any analyte in the gas phase from the phase separator. in aspects of the present invention, the sample may be aqueous. For example, the sample may be a wet flue gas desulferisation solution; a wet flue gas desulferisaiion scrubber solution; or is a sample collected from a wet flue gas desulferisation absorber tank.

So, for example, the measured Hg(0) or elemental mercury dissolved in the liquid phase (which can be found through knowledge of the amount in the gas phase and the partitioning factor or partition coefficient) is an indicator of how much mercury will eventually pass to the stack gas. Therefore the present invention provides a simple on line system which can replace complicated Hg measurements in the stack gas and significantly reduce the costs of measurements in the future. The present method is a simple predictor for the quantity of mercury to be released to the atmosphere.

In the present invention, the sample flow rate may be less than or equal to 15 mLmin'·. For example it may be 5 to 15 mLmin·''.

The sample and the carrier gas may be mixed before flowing through the equilibration section.

The carrier gas may be one capable of purging dissolved eiementai mercury from the sample. For example it may be nitrogen (Na). The carrier gas flow rate may be 65 to 80 mLmin··'.

The method of the present invention may comprise the further steps of determining the content of analyte in the gas phase extracted from the phase separator; and using a known partition coefficient to calculate the content of analyte in the sample.

The present invention also relates to use of a device according to the present invention in a method for extracting dissolved elemental mercury from a liquid sample.

The present invention also relates to use of a device according to the present invention in a method for determining the content of dissolved elemental mercury in a liquid sample.

Brief Description of the Figures

The invention will now be described, without limitation, in the following discussion of embodiments of the invention and with reference to the accompanying Figures, in which:

Figure 1 shows a schematic representation of a device according to the present invention.

Figure 2 shows the correlation between calibrations using Hg(0) vapours and vapours derived from an Hg(ll) standard solution.

Figure 3 shows an optimization of the flow rate of a nitrogen (N2) carrier gas.

Figure 4 shows the difference in equilibration time needed for distilled water and for a WTGD scrubber solution.

Figure 5 shows the correlation between a manual method and the method according to the invention for determining DEM concentration in distilled water, over a range of tested concentrations.

Figure 6 shows the correlation between a manual method and the method according to the invention for determining DEM concentration in WTGD solutions with three different DEM concentrations.

Figure 7 shows the influence of particle density on the effectiveness of stripping Hg(0) during mixing in the reaction coil.

Figure 8 shows the correlation between a manual method and the method according to the invention for determining DEM concentration in a series of solutions, prepared from a stock Hg(0) solution, over a range of tested concentrations.

Figure 9 shows the correlation between a manual method and the method according to the invention for determining DEM concentration in solutions exposed to different environmental parameters (redox potential, temperature, addition of oxidants).

Figure 10 shows the correlation between a manual method and the method according to the invention for determining DEM concentration in solutions collected from a WFGD absorber tank.

Figure 11 shows the correlation between a manual method and the method according to the invention for determining DEM concentration in ‘real-time’, that is, over a course of several hours, with the capability of tracking changes in levels of reduction and oxidation of Hg in a WFGD scrubber solution.

Figure 12 shows the application of the method according to the invention, of determination of DEM concentration in surface seawater samples. Figure 12A shows that the semi-continuous method was able to detect diurnal changes in DEM concentration. Figure 12B shows the correlation between the manual method and the method according to the invention.

The Device

The device of the present invention is illustrated, in one embodiment, in Figure 1. The illustrated parts may be installed on, for example a base such as a wooden plate. A sample enters the device through the sample inlet 1. It may be flown into the device by injection or, for example, by means such as a pump (for example a peristaltic pump), not illustrated. In the present invention such introduction of the sample can be effectively continuous or semi-continuous, allowing for ‘real-time’ mercury content analysis.

Similarly, in this embodiment, a carrier gas such as nitrogen (N2), which is preferably inert and/or Hg-free, and which may act to purge elemental mercury from a sample, is also flown into the device, through the carrier gas inlet 2.

The two inlets are connected by pipes and meet at mixing chamber 3, where they form a mixed flow. The mixed flow then travels through a further tube 4, which may be of, for example, glass, to an equilibration section 5 of the device (sometimes called the “reaction coil” herein). in the equilibration section 5, elemental mercury dissolved in the sample equilibrates between the liquid phase (In this embodiment, remaining in the sample) and the gas phase (in this embodiment, being flown with the carrier gas).

The equilibration section can have a size and shape determined by the desired residence time for equilibration. For example, if it is determined that a long residence time is wanted a longer equiiibration section can be used.

In this embodiment, the equilibration section is in the form of a spiral or helical glass tube, although of course other materials and shapes are contemplated. Such a shape allows for a longer residence time in a given size of device. It may be positioned substantially vertically in normal use, for example. In some embodiments which use a spiral or helical tube as the equilibration section, various lengths are suitable. For example, a length of 50 mm to 500 mm, for example 70 mm to 250 mm, for example 150 to 200 mm or even 180 mm may be used.

The equilibrated mixed flow leaves the end of the equilibration section 5, in this case an outlet at the top of the helical tube, to a gas-liquid phase separator 6. Such separators are understood in this technical field. The illustrated example comprises an introduction section connected to the outlet of the equiiibration section 5. in this embodiment the introduction section is bulbous. That introduction section has an outlet at its lower portion, which is here in the form of a U-shaped tube 7. The liquid sample and excess carrier gas flows along this section to respective outlets from the device itself. The U-shaped portion may have a diameter of, for example 5 to 15 mm, to encourage suitable flow.

At the top of the introduction section of the phase separator is a further outlet pipe or tube 8, through which the gas phase containing Hg(0), elemental mercury, is flown.

This outlet pipe may have a diameter larger than that of the U-shaped portion, for example a diameter of 10 to 25 mm, for example 15 to 20 mm, to encourage preferable flow of the gas phase mercury through this outlet. Hg(0) vapors accumulate in the upper portion of the phase separator. There they can be extracted or flown out through the outlet 8.

From there the extracted gas phase elemental mercury can be, for example, detected or analysed by a further device. Illustrated is a detector 9. That might be, for example, a CVAAS (cold vapor atomic absorption spectrometry) or CVAFS (cold vapor atomic fluorescence spectroscopy) device.

The detector 9 may be connected to the outlet 8 by, for example, a tube of Teflon, for example (that is, polyteirafluoroethylene), of a suitable length. An example apparatus might use a length of, for example, 50 to 100 cm.

Preparation and Use of the Device

Since the present semi-continuous method operates on the basis of partitioning of the analyte, such as Hg(0), between the equilibrated liquid and gaseous phases, the concentration of the analyte in the gaseous phase is proportional to the amount of analyte dissolved in the aqueous phase.

This means that a “partitioning factor” or partition coefficient for the analyte, such as Hg(0), between the carrier gas and liquid matrix of the sample can be calculated to allow concentrations of the analyte measured in the gas phase to be correlated to a total amount of DEM in that sample.

As the partitioning factor or partition coefficient is affected by the nature/composition of the sample solution, that is, for example, the content of an aqueous sample solution, it can be determined in advance.

For example, in order to define the partitioning factor of Hg(0) between aqueous and gaseous phases, reference measurements can be made using a method which directly determines the Hg(0) content of a sample. For example, a manual method that removes DEM quantitatively. The ratio between Hg(0) in the gaseous phase (semi-continuous method) and the DEM concentration in the aqueous phase (manual method), can be calculated for a given ‘matrix’, that is, sample solution. Thai partitioning factor or partition coefficient can then be used repeatedly with that sample solution. The partitioning factor is used to calculate the concentration of DEM in aqueous samples in the semi-continuous method.

This permits the present semi-continuous method to give considerable speed, convenience and responsiveness advantages as compared to previously described manual methods.

Specific examples of these procedures, and other investigations performed by the present inventors, will now be explained.

Preparation of chemicals, sampling, and calibration.

Hg(0) stock solution: DEM standard solutions were prepared by putting a few grams of liquid Hg into a 5 L glass bottle (Duran, acid pre-cleaned) filled with MilliQ water. The bottle was protected from direct light exposure to prevent possible photo-oxidation.

The concentration of DEM in this solution at room temperature after about 24 hours was 46 pg L”^ and it was stable for about 2 months at room temperature. The concentration of the stock solution was determined each time before its use by the reference manual method,

Hg(ll) standard solution: A standard solution of Hg(ll) with concentration 1 ng mL-^ was prepared in 5% HNO3 by appropriate dilution by weight of the certified Hg standard solution NIST SRM 3133 (9.954 mg g-’' ± 0.053 mg g-'). The working standard was used for calibration purposes.

Stannous chloride solution: A 3% (by weight) solution of SnCb x 2H2O (MERCK, Emsure, in 250 g dark bottles) was prepared in 1% fuming HCI (MERCK, Em sure, 37%, in 2.5 L dark bottles). The solution was prepared fresh every day and purged for at least one hour with Hg free N2 before use to remove any Hg present.

Nitrogen: Standard laboratory grade (grade 4.5) has been further purified by removal of mercury using a gold-coated sand trap.

Scrubber solution from WFGD: Scrubber solutions were taken from the lignite-fired thermal power plant (TPP) in Sostanj, Slovenia and were used for validation purposes.

The ΤΡΡ uses the wet limestone process with forced oxidation. Previous work on this TPP has reported very high levels of removal of Hg in the WFGD.

Sampling. 100 L of scrubber solution was collected in a plastic container during the WFGD operation. The solution was characterised by a brownish colour with a solid particle density of 88 g L-·'. The concentration of DEM immediately after sampling was 79 ng L-'' in the unfiltered sample and 5 ng L-·· in the filtered (0.45 pm) sample. The total Hg (THg) concentrations of unfiltered and filtered samples of scrubber solution, determined by commercial methods, were 200 pg L-' and 3 pg L-·* respectively. The high concentration of THg in particles, 2.24 pg g-·, demonstrated the high adsorption efficiency of the latter for Hg, The particles represent the finer fraction of gypsum produced in the WFGD during operation as a result of SO2 capture. The samples were used to validate the present semi-continuous method before applying it to the WFGD pilot plant for DEM determination.

Calibration. For calibration purposes, either Tekran 2505 or the calibration unit from the UT 3000 detector were used. Aliquots of Hg(0) vapour ranging from 5 to 25 pL, corresponding to 25-140 pg of Hg, w'ere transferred into the double amalgamation system with a gas-tight syringe through a Teflon septum. In the case when the calibration unit from the UT was used, the range of concentrations was wider due to the higher temperature of the Hg vapour (about 2°C below room temperature). When samples with higher concentrations were measured, a larger amount of Hg(O) was taken for calibration (100-600 pg).

Blanks. Blanks of the manual method were determined according to the descriptions provided in the US ERA 1631 method, where bubbler blanks were continuously controlled. The minimum requirement for the bubbler blanks for the WFGD samples ranged was 100 pg/L, Much stricter requirements for the bubbler blank for natural waters were needed (below 5 pg/L),

Manual method (for calibration)

The system used for manual determination of DEM is composed of two main parts, the bubbling system, by which DEM is stripped from the sample matrix, and the double amalgamation system, in which DEM is quantified (CV AFS). In the purging step, 100 mL of sample was transferred to a glass bubbler. The purging gas (Hg-free N?) entered the glass bubbler (500 mL Pyrex glass) through a glass frit sealed at the bottom of the bubbler ‘cup’. The sample was purged for 10 minutes under an N2 flow of about 800 ml min-T This guaranteed the most effective recovery of DEM from the water samples in a short period of time. Repeated purging of the sample resulted in undetectable quantities of DEM, which was a proof of quantitative removal of DEM. Volatile mercury species (primarily DEM) were amalgamated onto a sampling gold-coated silica trap consisting of a 120 cm long quartz tube. In the middle of the quartz tube, 0.5 g of Au-coated silica Is packed. A soda lime trap was installed after the bubbling system to prevent moisture from reaching the Au trap. In the measurement step, the Au sampling trap was transferred to the double amalgamation system of the CV AFS analyser (Tekran 2500).

Semi-continuous method for DEM measurement

Semi-continuous DEM measurements were performed by an independent system, presented schematically in Figure 1. It is based on the flow injection system in which equilibrium is reached between aqueous and gaseous phases, demonstrated by stable readings from UT3000 during the operation of the system.

The liquid sample is introduced Into sample inlet 1 continuously by a peristaltic pump (Masterflex US precision peristaltic pump Modular Drive w/, Remote I/O; 600 rpm; and Masterflex standard pump head for US 16 tubing, PC housing/SS rotor and a Teflon tube about 20 cm long with an external diameter of 3 mm. Hg-free inert carrying gas {N2) is introduced into the system through an N2 inlet 2 at flow rate ranging from 75 to 80 mL min T In mixing chamber 3 liquid sample and Hg-free inert gas are mixed and further passed through glass tube 4 into the spiral glass tube 5 where Hg in liquid phase is equilibrated with the gas phase Hg. in the outlet of the equilibration spiral, a gas-liquid phase separator 6 is installed, it consists of two parts: a U shape tube (diameter of 9 mm), which is constantly kept filled with sample, and which ends with the excess sample and gas outlet 7, and the upper part 8 (headspace, with a diameter of 17 mm) where the Hg(0) vapours are collected and further measured by CV AAS or CV AFS technique 9. Hg(0) vapour stripped from the sample is carried to the detector by a Teflon tube (about 80 cm long, external diameter of 3 mm). In order to ensure the optimal flow for the UT 3000 detector (0.3 L mln~^), a make-up gas (Hg-free air) may be used, Hg(0) released from the liquid sample is, for example, collected onto a detector-inbuilt gold trap. For higher DEM concentrations, the gold trap collection time is set at 3 minutes, which is the minimum required time for the UT3000 instrument. For lower concentrations (seawater) longer amalgamation time may be needed, therefore the measurement times were set at 6 minutes. Quantification of gaseous Hg(0) was performed by CV AAS (Mercury Instruments UT3000) detector.

As Hg tends to accumulate on various surfaces all tubes and parts that are in contact either with liquid sample or gas are made of borosilicate glass or Teflon. Both materials do not accumulate Hg on their surfaces which was proven during several research on Hg cycling.

The light absorbance values from the UT 3000 detector were recorded. The concentration of Hg(0) in the gaseous phase correlates proportionally with the concentration of DEM in the water sample. However, as the partitioning of DEM between gaseous and liquid phases is affected by the composition of the aqueous solution, the partitioning factor to obtain DEM needs to be determined for each sample matrix. This is done by the use of a manual method, which is used as a reference method as it measures total concentration of DEM in liquid samples.

System blanks

System blanks for WFGD samples were regularly controlled by purging scrubber solution produced synthetically with Hg content below LOD. For natural waters bubbler blanks were double checked by distilled water and samples with DEM concentration below 5 pg/L,

Optimisation of the semi-continuous method

The main difference between the manual and semi-continuous modes is that DEM in the semi-continuous mode is not removed quantitatively from the aqueous solution but partitioned between gas and liquid phases. The proportion of Hg in the gas phase depends on several factors, such as the gas and liquid flow rates and the composition of the liquid phase. The present system can be used to perform DEM measurements in WFGD systems with relatively high concentrations of DEM (from a few to 30 ng Lr') in the scrubber solution. For this reason, the concentrations of Hg(0) used during the present exemplary optimisation (validation) steps were in the range of 1 to 100 ng bv

Sample and gas flow rates

The design of the system allows sample flow rates of, for example, up to 15 mL min-'. With samples with higher densities, such as WFGD scrubber solution, the optimal flow rate may be, for example, 8 mL min-''. Against this the gas flow rate can also be adjusted. The optimal flow of N2 was about 75 mL min-'' (Figure 3).

Equilibration time

Elemental Hg is a gas that is slightly soluble in water with reported values varying from 58.8 pg L·' to 63.9 pg Lr' at room temperature. Since the semi-continuous method of the present invention operates on the basis of partitioning of Hg(0) between the equilibrated liquid and gaseous phases, the concentration of Hg(0) in the gaseous phase is proportional to Hg(0) dissolved in the aqueous phase. In the semi-continuous method, the gaseous Hg(0) can be removed through the gas-iiquid separator and measured by the CV AAS detector. A stable signal may be obtained after about three cycles (for example, about 9 min) of continues sample flow for distlNed water and four cycles (for example, about 12 min) for a WFGD scrubber solution (Figure 4). This indicates that the composition of the aqueous phase affects the time needed for equilibrium to be achieved.

Limit of detection (LOD)

The LOD was expressed as three times the standard deviation of the sample with the lowest DEM concentration. Values of LOD for the semi-continuous method ranged from 4 to 14 pg Lr^ while for the manual method were about 11 pg L-L It should be noted, however, that values of LOD from both methods may be affected by, for example the cleanliness of the laboratory air and glassware.

Estimation of uncertainty

The uncertainty of the manual method was estimated in accordance with GUM instructions. A mathematical model is applied to link the output signal from the detector and the DEM concentration, as in Equation 1;

Equation 1 where Cngio) is the calculated concentration of DEM in pg L-T hsampie is the peak height of the sample, hbiank the peak height of the bubbler blank, hbackground the peak height of the background signal of the detector and hstandard the peak height of the standard (Hg(0) vapour).

The expanded uncertainty for an approximate level of confidence of 95% ranged from about 10% to 15%. The total combined uncertainty of the method was calculated by considering the uncertainties of the parameters included in Equation 2. The uncertainty as a function of the volume of Hg(0) vapour injected, was calculated by combining the uncertainties due to the calibration of the syringe, repeatability of the gas tight syringe and the temperature of the calibration unit. In addition, the repeatability of the method was taken into account (4% expressed as RSD). The main components contributing significantly to the overall uncertainty for DEM measurements in water samples by the manual method were the repeatability of the method (about 30% of the total uncertainty) and the calibration with Hg(0) vapour (about 8% of total uncertainty).

The concentration of DEM in water samples in the semUcontinuous method was calculated by considering the partitioning factor which is measured by making paraiiei sample measurements in the manual method. As a result the uncertainty from the manual method is included in the overall uncertainty of the semi-continuous method. Additionally the uncertainty originating from the calibration of UT 3000 detector with Hg(0) vapour was considered. The overall uncertainty of the method ranged from 18 to 22%. The greatest uncertainty contribution to the overall uncertainty of the method was attributed to the use of the manual method (60%); a lower, but significant, contribution was attributed to the calibration of the UT 3000 detector (5-8%).

Calibration

Most of the measurements for Hg(0) are traceable to empirical relationships that describe the vapour pressure of Hg at any given temperature. Consequently, the result of a measurement will depend on which equation is chosen to represent the vapour pressure and its accuracy.

There are several empirical equations and the agreement between the results calculated from them differs by 5% or more. The most used equation to convert the volume of Hg vapours into quantity of Hg mass is the Dumarey equation (Equation 2):

Equation 2 where y°Hg is the mass concentration of mercury vapour in the taken volume (in ng mL-h, T is the temperature of the saturated vapour inside the calibration unit device (in K), A (-8,1344), B (3240,9 K) and D (3216522 K ng mL·^) are constants, δ is the deviation from reality of the theoretical saturated vapour mass concentration of mercury in the calibration unit. It has recently been proven that the data produced by Equation 1 are 5,8% to 7% lower than those presented by other authors.

Due to the inconsistencies of the Dumarey equation, the calibrations using Hg(0) vapours and those derived from the aqueous Hg(i!) standard solution (NIST SRM 3133) were compared. The latter were reduced by SnCi2 and the resulting Hg(0) was measured by CV AFS, The calibration curves obtained are compared in Figure 2,

Uncertainties of measurements have been calculated based on the Guide to the Expression of Uncertainty in Measurement, Other reports have also been considered. The biggest source of uncertainty in the case of Hg(0) calibration was the repeatability of the gas-tight syringe. The contribution to the overall uncertainty arising from the calibration unit was not significant.

The biggest source of uncertainty in the budget, when the calibration was done using Hg(ll) standard solution (9.954 mg g-^ ± 0.053 mg g-·), was the weighing of the stock solution when the working standard needed to be prepared. The uncertainty of calibration with Hg(0) vapours ranged from 2 to 10 % while for Hg(M) standard was about 14 %.

The results obtained are dearly linear and comparable over the tested concentration range, therefore it may be preferable in the present invention to use calibration units based on the Dumarey equation.

Calibration of the semi-continuous method

The CV AAS and CV AFS detectors were calibrated by the gas caiibration units and the measurement systems by DEM in aqueous solutions. in order to define the partitioning factor of Hg(0) between aqueous and gaseous phases, additional reference measurements were made by the manual method that removes DEM quantitatively. The ratio between Hg(0) in the gaseous phase (semicontinuous method) and the DEM concentration in the aqueous phase (manual method), was calculated for each water matrix. The partitioning factor was then used to calculate the concentration of DEM in aqueous samples in the semi-continuous method.

Using the method according to the invention in connection with atomic absorption spectrophotometry (UT 3000 mercury anaiyser).

Semi-continuous measurements of DEM were made using a UT 3000 mercury anaiyser CV with an AAS detector that performs a single amalgamation step analysis. Hg(0) is stripped from the water sample by mercury free N2 and is pumped onto the Au trap where it amalgamates with Au. it is then desorbed by heating and finally detected in the spectrophotometer cell. The minimum time for the whole measurement cycle requires 3 minutes. For samples with low DEM concentrations, longer amalgamation times (up to 0.5 hours) can be applied.

Calibration of the instruments: Two saturated gaseous Hg sources (calibration units) were used, the calibration unit of the UT 3000 detector and a Tekran 2505. The calibration unit consists of a specially designed Hg vessel surrounded by an aluminium jacket cooled by a thermoelectric cooler. The mercury chamber contains liquid Hg of high purity and the temperature is ambient and measured with a high precision temperature sensor. Based on the temperature the concentration of saturated vapours in the vessel is automatically calculated and displayed in the UT 3000 monitor.

Using the method according to the invention in connection with atomic fluorescence spectrophotometry.

The Tekran 2500 detector is connected to the double amalgamation system. DEM stripped from the solution is collected and amalgamated onto a gold sampling trap. Hg is then released by heating at about 500 °C for 1 minute in a flow of argon to a permanent gold trap, released again (1 minute at 500 "'C) and detected by CV AFS (Tekran 2500). The flow of argon ranges between 50 and 60 mL min-^ The measurement cycle requires about 3 minutes. The fluorescence is recorded as a peak on a chart recorder connected to the Tekran 2500 detector.

Calibration. The Tekran 2505 calibration unit is based on the same principle as the calibration unit of the UT 3000 detector but the temperature of the Hg chamber is thermoelectrically controlled to allow the temperature to be set to 0 to 30°C. Hg concentration in both Hg vapour calibration units was calculated on the basis of the Dumarey equation and provided comparable results.

Matrix effects

The solubility of Hg(0) in the aqueous phase may be affected by the temperature and the composition of the liquid phase. In order to determine the effectiveness of stripping Hg(0) from distilled water, a series of samples with different DEM concentrations in the range of 1.5 to 4 ng were prepared and measured by a manual method to determine the exact concentrations of DEM. The Hg(0) stripped from water samples under the equilibrium conditions of the semi-continuous method w'as also determined. The response over the tested concentration range was shown to be linear {Figure 5). The proportion of Hg(0) stripped from the aqueous phase during mixing in the reaction coil ranged from 43 to 54%, with an average of 48%.

The experiment was repeated with the three FGD solutions with different concentrations of DEM. A linear response was shown (Figure 6).

Effect of particle density.

The effect of particle density on the effectiveness of stripping Hg(0) during mixing in the reaction coil can also be considered. Three samples were prepared with different particle densities by diluting WFGD scrubber solution with a particle density of 88 g LrT The percentage of Hg(0) stripped from the aqueous phase depended significantly on the concentration of the particles, ranging from 42% to 22% for particle concentrations of 17,6 g L·-^ and 88 g b' (Figure 7).

Recent studies suggest the importance of adsorption phenomena of DEM in natural samples, particularly in the presence of particles where Hg(0) can be adsorbed onto the particles. This fraction of Hg(0) is then no longer available for removal from water by purging and may result in underestimation of the total Hg(0) present in natural waters, particularly when water samples are not analysed on-site but several hours after sampling.

Linearity A series of solutions with different DEM concentrations were prepared from the stock Hg(0) solution. The samples were measured by both manual and semi-continuous methods. The absorbance signals obtained from the UT 3000 detector used in the semi-continuous method were correlated with those obtained by a manual method (Figure 8). The correlation was very good and linear over the concentration ranges tested.

Precision: Repeatability and reproducibility

To investigate the precision of the methods under conditions of repeatability, 3-4 spiked distilled water samples were measured during the day by both the manual and the seml-continuous methods and the relative standard deviation (RSD) of the measurements calculated. For the manual method, measurements varied from 1 to 9% (5% average) and, for the semi-continuous method, from 0.5 to 8% (4% average). To evaluate the reproducibility of the methods, measurements were performed on consecutive days and their standard deviation calculated. The sample was freshly prepared on successive days and, following the same measurement procedure, the reproducibility for both methods was found to be between 1 and 7% with an average of about 5.6 %.

Combinations

It is appreciated that certain features of the invention, which are, for clarity, described in the context of separate embodiments, may also be provided in combination in a single embodiment. Conversely, various features of the invention, which are, for brevity, described in the context of a single embodiment, may also be provided separately or in any suitable sub-combination.

The same is true for features described in the context of products and methods.

Any one or more of the aspects of the present invention may be combined with any one or more of the other aspects of the present invention. Similarly, any one or more of the features and optional features of any of the aspects may be applied to any one of the other aspects. Thus, the discussion herein of optional and preferred features may apply to some or a!i of the aspects. Furthermore, optional and preferred features associated with a method or use may also apply to a product and vice versa.

The options, features, preferences and so on mentioned herein apply both independently and in any combination, except where such a combination is expressly prohibited or clearly impermissible.

Examples

Example 1

Performance of the method at the iaboratorv scale pilot plant

The presence of DEM in the WFGD scrubber solution may indicate that Hg(0) from a flue gas is not sufficiently oxidized and/or that Hg(ll) already present in the scrubber solution is reduced to Hg{0). These processes can be altered by controlling the chemistry of Hg in the scrubber solution. The present semi-continuous method can continuously monitor DEM concentrations in the WFGD solution as an indicator of Hg removal.

In order to verify the suitability of the semi-continuous method for the above mentioned purpose, the performance of the method at the laboratory scale pilot plant was investigated. Hg(0) dissolved in the scrubber solution was measured. The initial experiments were carried out with a column filied with distilled water. The water was purged by a mixture of Nz, air and Hg(0) vapour introduced at the bottom of the bubbler glass column through a glass frit. In order to make semi-continuous measurements of DEM, aqueous samples were taken continuously from the upper part of the column through a Teflon tube 1.5 m long with a diameter of 3 mm. The chemistry of Hg(0) in the aqueous phase was altered by changing the physico-chemical properties of the latter (redox potential, temperature, the addition of oxidants, etc.), in consequence, the concentration of DEM changed as shown in Figure 9. The results of the semi-continuous method were compared with the manual measurements and excellent agreement was obtained. Differences between results obtained by the two methods ranged from 2 to 12%. The good correlation is demonstrated by the linearity of the regression line. By choosing a measurement cycle time of 3 min, it was possible to detect changes in Hg chemistry in the bubbler column depending as a function of rapid changes in the physicochemical conditions.

In a continuation of this investigation the distilled water was substituted with scrubber solution prepared artificially. Also the pilot plant was designed to resemble more closely the wet limestone WFGD system. The goal of the further experiment was to follow the oxidation/reduction reactions occurring under various operating conditions of the WFGD system. The main part of the system setup consisted of an absorber tank with a capacity of 150 L in which the scrubber solution is modified to achieve high retention of Hg(0) from the flue gas. A stream of Hg(0) vapour and a gas flow of SO2 were therefore, passed continuously through the scrubber solution whose temperature was rigorously controlled, as were the other parameters (pH, redox potentiai). Forced oxidation was achieved with air.

The sample for semi-continuous DEM measurements was taken from the WFGD absorber tank, using a 2 m Teflon tube (diameter of 3 mm). Frequent manual measurements were conducted along with the semi-continuous ones and the partitioning factor was calculated (Figure 10). The differences in DEM concentrations achieved by the two methods ranged mostly between 3-10% but in some cases exceeded 20%.

The results obtained by a semi-continuous method (Figure 11) demonstrate its capability to track the fast changes on reduction and oxidation of Hg in the scrubber solution arising from the addition of catalysts.

Example 2 DEM measurements in surface marine waters

In addition to samples with high DEM concentrations, the present method was also successfully applied ίο the measurement of DEM in surface seawater samples, with low Hg content. The system was tested using seawater samples. The sample for semi-continuous measurements was taken from a depth of 3.5 m and the method was tested in parallel w'ith the manual method. Calculations involved consideration of the partitioning factor. As expected, during 2 weeks of measurements the partitioning factor for seawater samples did not change. The semi-continuous method was able to detect diurnal changes in DEM concentration (Figure 12A) together with an example of a comparison between the two methods (Figure 12B) that demonstrates good comparability and linearity.

Claims

1. A flow device for exiraciing a dissolved analyte from a liquid sample, the device comprising: a liquid sample inlet, for receiving a sample flow at a sample flow rate; a carrier gas inlet for receiving a carrier gas flow at a carrier gas flow rate; an equilibration section, in fluid communication with the liquid sample inlet and the carrier gas inlet such that a sample flow from the liquid sample inlet and a carrier gas flow from the carrier gas inlet flow into the equilibration section, for equilibration of the content of the analyte in the liquid sample and in the gas phase; and a gas-liquid phase separator, in fluid communication with an outlet from the equilibration section such that a flow from the outlet of the equilibration section flows into the phase separator, for separating the analyte in the gas phase from the liquid mixture output from the equilibration section.

2. A device according to claim 1, wherein the dissolved analyte is dissolved elemental mercury.

3. A device according to claim 1 or claim 2, wherein the carrier gas inlet and the liquid sample inlet are in fluid communication such that a sample flow from the liquid sample inlet and a carrier gas flow from the carrier gas inlet join to form a single flow before flowing into the equilibration section.

4. A device according to claim 3, further comprising a mixing chamber in fluid communication with the liquid sample inlet and the carrier gas inlet, such that the sample flow and carrier gas flow both enter the mixing chamber, where they join and mix; the mixing chamber also being in fluid communication with the equilibration section such that the mixed liquid sample and carrier gas flow can flow into the equilibration section.

5. A device according to any one of the preceding claims, wherein the equilibration section comprises a helical tube.

6. A device according to any one of the preceding claims, wherein the phase separator comprises an upper part through which the analyte in the gas phase flows and a lower part through which the remaining sample and carrier gas output from the equilibration section flows.

7. A device according to claim 6, wherein the upper part and lower part of the phase separator each comprise a tube, the two tubes diverging from a point proximal the outlet of the equilibration section.

8. A device according to claim 6 or claim 7 wherein the lower part of the phase separator terminates in an outlet for the sample and an outlet for the carrier gas.

9. A device accordingly to any one of claims 6 to 8, wherein the lower part of the phase separator comprises a U-shaped tube section.

10. A device according to any one of claims 6 to 9, wherein the upper part of the phase separator comprises an outlet for analyte vapours.

11. A device according to claim 10, wherein the analyte outlet is suitable for connection to a device for measuring the analyte concentration.

12. A device according to claim 11, wherein the upper part of the phase separator is connected to a device for measuring the analyte concentration.

13. A device according to claim 11 or claim 12, wherein the device for measuring the analyte concentration is an atomic fluorescence spectrophotomeier or an atomic absorption spectrophotometer.

14. A device according to any one of the previous claims, wherein some or all of the device is made from or internally coated with borosilicate glass or polytetrafluoroethylene,

15. A method for extracting a dissolved analyte from a liquid sample, the method using a device according to any of the preceding claims, the method comprising the steps of: passing the sample into the liquid sample inlet at the sample flow rate; passing a carrier gas into the carrier gas inlet at a carrier gas flow rate; flowing the sample and the carrier gas through the equilibration section, to equilibrate the content of the analyte in the sample and in the gas phase; and extracting any analyte in the gas phase from the phase separator.

16. A method according to claim 15, wherein the sample is aqueous.

17. A method according to claim 16, wherein the sample is a wet flue gas desulferisation solution; a wet flue gas desulferisation scrubber solution; or is a sample collected from a wet flue gas desulferisation absorber tank.

18. A method according to any one of claims 14 to 16, wherein sample flow rate is less than or equal to 15 mLmin·’.

19. A method according to claim 18, wherein the sample flow rate is 5 to 15 mLmim·.

20. A method according to any one of claims 15 to 19, wherein the sample and the carrier gas are mixed before flowing through the equilibration section.

21. A method according to any one of claims 15 to 20, wherein the carrier gas is one capable of purging dissolved elemental mercury from the sample.

22. A method according to claim 21, wherein the carrier gas is nitrogen (N2).

23. A method according to claim 22, wherein the carrier gas flow rate is 65 to 80 mLmin··'.

24. A method according to any one of claims 15 to 23, comprising the further steps of determining the content of analyte in the gas phase extracted from the phase separator; and using a known partition coefficient to calculate the content of analyte in the sample.

25. Use of a device according to any one of claims 1 ίο 13 in a method for extracting dissolved elemental mercury from a liquid sample.

26. Use of a device according to any one of claims 1 to 13 in a method for determining the content of dissolved elemental mercury In a liquid sample.