G5 Remediation and control technologies

Tuesday, 26 July, 2011

TG5-P1 — 11:00-12:00 and 17:30-18:30
Authors: MERRITT, Karen1, GLESSNER, Allison1, HENNING, Miranda1, MAGAR, Victor1
(1) ENVIRON International Corp., kmerritt@environcorp.com

We examine the justification and proposed strategy for site remediation in Jellicoe Cove, Ontario, through thin-layer placement of a sand cap over the area with surface sediment total mercury (HgT) concentrations exceeding 3 mg/kg. The other principal chemical of concern in Jellicoe Cove is polychlorinated biphenyls (PCBs), with maximum surface sediment concentrations of 1.2 mg/kg. Capping will prove effective for reducing the HgT concentration in surface sediment, and is expected to prove effective for limiting biological exposure to HgT, methylmercury (MeHg) and PCBs, as well as limiting on-going transport of HgT and PCBs from Jellicoe Cove. As assessed herein through the application of a design tool for estimating chemical migration through cap layers, the placement of a 15 cm sand cap in Jellicoe Cove can reduce porewater HgT flux by 97% within the cap area footprint. For MeHg, the placement of a 15 cm sand cap in Jellicoe Cove can reduce porewater flux by > 99%. Results for HgT and MeHg—measured in terms of the percent reduction in diffusive flux to the biologically active layer following cap placement—will conservatively satisfy concerns for PCB flux. As determined by application of the design tool, the time to containment breakthrough for a 15 cm cap is > 250 years for HgT and ~250 years for MeHg. The magnitude of the porewater flux and the time to containment breakthrough are sensitive to the choice of distribution coefficients (KD) for HgT and MeHg, as well as the groundwater advection rate. If the current conceptual understanding of site conditions is correct, a 15 cm cap at this site will likely achieve adequate reduction in solid phase and porewater HgT and MeHg concentrations within the cap area footprint, even considering a 10-fold decrease in KD values and a doubling of the groundwater advection rate.

TG5-P2 — 11:00-12:00 and 17:30-18:30
Authors: FOSTIER, Anne1, BEHAINNE, Jhon J.R.2, GOLDSTEIN JR., Leonardo 2, PÉCORA, Arai A.B.2, HORY, Rogério I. 2, MICHELAZZO, Paula A.M.2
(1) Universidade de Campinas - UNICAMP, fostier@iqm.unicamp.br; (2) UNICAMP;

Emissions of particulate mercury (Hgp) originated from combustion of the CE-4500 Brazilian coal previously mixed with dolomite were measured in a bench-scale circulating fluidized bed (CFB) unit. A procedure based on the Ontario Hydro method was adopted to collect flue gas samples in a stack point located downstream a fabric filter operating at temperatures lower than 80°C. Additionally, mercury content was measured in both the bottom ashes of the combustor and in the fly ashes trapped by the particulate control device. The cold vapor atomic fluorescence spectrometry (CVAFS) analytical technique was used to quantify the mercury in samples obtained from eleven runs carried out at specific operational conditions. Results showed that only 1% of the total Hgp emissions were released to the atmosphere together with the flue gas. For all cases analyzed, mercury concentrations in the fly ashes were clearly higher than those in residual particles of the bed. The experiments showed that the combination of CFB combustion and fabric filter is an interesting alternative for controlling mercury in coal combustion, specially when the particulate separator works at relatively low temperature for retaining fly ash with some carbon content. Additionally, for the operational conditions tested it was not observed a significant relationship between Hgp concentrations and both, excess air used in the combustion and Ca/S molar ratio of the fuel mixture.

TG5-P3 — 11:00-12:00 and 17:30-18:30
Authors: AUSTIN, Carrie1, KIRCHNER, James W.2, WHYTE, Dyan C.3, MYERS, Alexandra2
(1) CalEPA, SF Bay Water Board, caustin@waterboards.ca.gov; (2) U.C. Berkeley; (3) CalEPA, SFBay Water Board;

How effective is site cleanup? Simple before-and-after comparisons of contaminant concentrations or loads can be misleading measures of remediation effectiveness, due to factors external to the remediation, including weather. We present a new analytical method, before-and-after contaminant rating curves, to correct for variations in external driving forces, and thus to clarify remediation effectiveness. We demonstrate our method using monitoring data from the Gambonini mercury mine in the California Coast Range. Measured mercury loads in a stream draining the mine site were a factor of ~1000 lower five years after remediation, compared to measurements made prior to remediation. However, the post-remediation year also had much lower rainfall, and thus would be expected to give lower mercury loads because most of the mercury comes from erosion of the waste pile. Our calculations show that the variation in rainfall would account for a factor of ~60-80 decrease in mercury loads (corresponding to an apparent 98-99% removal efficiency), even without remediation. Our results also show, that (a) concentrations of mercury in sediment went down (less mine waste in stream sediment), (b) concentrations of sediment at a given stream flow went down (less sediment mobilized due to greater slope stability), and therefore (c) for a given stream flow, a much smaller load of mercury left the site. By comparing pre- and post-remediation contaminant rating curves, we show that the mine remediation can be shown to have reduced mercury loads by a factor of 10-20 (90-95%) on an all-else-equal basis.

TG5-P4 — 11:00-12:00 and 17:30-18:30
Authors: TSAI, Cheng-Yen1, HSI, Hsing-Cheng2, FAN, Kuo-Shuh1
(1) National Kaohsiung First University of Science and Technology, u9315918@nkfust.edu.tw; (2) National Taipei University of Technology;

Mercury (Hg) emissions from natural and anthropogenic sources have tempted substantial attention due to its toxicity and bioaccumulative effects via the food chain. Research pertaining to enhance the adsorption and catalyst oxidation of Hg0 has drawn increasing attention in recent years. Titanium dioxide (TiO2) photocatalyst is known to effectively decompose various pollutants in gaseous and aqueous phases. In this study, nitrogen-doped TiO2 (i.e., TiO2-xNx) nanoparticles were synthesized using thermal plasma as a heating source via evaporation condensation for removing elemental mercury (Hg0) under UV and visible light irradiation. Titanium metal of 99.8% purity was used as the Ti source. Desired plasma power level was maintained by controlling the flow rate of the plasma gases. The mixture of nitrogen, argon and helium was used as plasma gas. The flow rate of Ar + He was controlled at 2 L min-1; N2 was added to control N2/(Ar + He) = 0, 5, 7 and 10% by volume. The formed TiO2-xNx photocatalyst nanoparticles were characterized with TEM, XRD and UV-Vis. Hg removal effectiveness of the TiO2-xNx nanoparticles was subsequently evaluated at a low Hg0 concentration (ppb) level.

Sample characterization results showed that the formed TiO2-xNx nanoparticles have a size within 10 – 40 nm. The formed TiO2 nanopowders were a mixture of anatase and rutile. For the 0%, 5% and 7% N-doped TiO2 nanoparticles, the absorption edge shifts towards the visible light regions depending on the N2 ratio in the plasma gas. The new absorption band can be attributed to the contribution from N atoms doped into TiO2 nanoparticles. Hg0 removal by the TiO2-xNx nanoparticles was significantly enhanced by increasing the O2 concentration. However, the presence of water vapor in the test gas stream reduced Hg0 capture, especially when light irradiation was applied. It was suspected that the reduction in Hg0 capture was mainly attributed to the competitive adsorption of H2O on the active sites of TiO2-xNx with Hg0 and transformed Hg2+.

TG5-P5 — 11:00-12:00 and 17:30-18:30
Authors: IE, Iau-Ren1, CHEN, Wei-Chin1, YUAN, Chung-Shin1, HUNG, Chung-Hsuang2, LIN, Yuan-Chung3
(1)Institute of Environmental Engineering, National Sun Yat-Sen University, peter-34@yahoo.com.tw; (2) Department of Safety, Health and Environmental Engineering, National Kaohsiung First University of Science and Technology; (3) Institute of Environmental Engineering, National Sun Yat-Sen University.

This study investigated the adsorption of vapor-phase mercury chloride (HgCl2) by a composite sulfurized activated carbon (AC) impregnated by vapor-phase elemental sulfur (S0) and aqueous-phase sodium sulfide (Na2S). The composite sulfurized AC was prepared to enhance its adsorptive capacity of HgCl2. Experiments on the adsorptive capacity of vapor-phase HgCl2 onto the composite sulfurized AC were conducted with a thermogravimetric analyzer (TGA). The operating parameters investigated in this study included the types of sulfurized AC, the adsorption temperature, and the influent HgCl2 concentration. Moreover, the surface characteristics and chemical properties of AC before and after impregnation were also compared. Experimental results indicated that the innovative composite sulfurized AC impregnation process could greatly increase the sulfur content of AC, which is nearly the summation of the sulfur content of AC obtained from the impregnation of both vapor-phase S0 and aqueous-phase Na2S. Further investigation on the sequence of sulfur-impregnation processes showed that AC impregnated by aqueous-phase Na2S and followed by vapor-phase S0 had much higher HgCl2 adsorptive capacity than that impregnated in the reverse sequence. Comparing the HgCl2 absorptive capacity of various sulfurized AC showed that the composite sulfurized AC had higher HgCl2 adsorptive capability, and its HgCl2 adsorptive capacity (6,720 µg-HgCl2/g-C) was higher than the summation of the adsorptive capacity of AC impregnated with aqueous-phase Na2S (3,223 µg-HgCl2/g-C) and vapor-phase S0 (1,781 µg-HgCl2/g-C). The HgCl2 adsorptive capacity of the innovative composite sulfurized AC was approximately four times higher than previous sulfurized AC.

TG5-P7 — 11:00-12:00 and 17:30-18:30
Author: LUO, Jinjing1
(1)Xiamen University, luojj27@xmu.edu.cn

This study evaluated the mercury capture ability of a novel modified activated carbon. Mercury adsorption tests were performed under pure N2 atmosphere and the presence of flue gas component(s) (CO2, O2, SO2, NO, HCl, NO2 and water vapor). Results show that the modified activated carbon sample has a relatively large mercury adsorption capacity compared to those of the commercial activated carbons (DARCO FGD and FLUPAC). NO2, NO and O2 can strongly increase the mercury adsorption capacity of the modified sample. On the contrary, SO2 and water can significantly inhibit mercury adsorption by the modified activated carbon. In addition, CO2 and HCl do not play important roles in determining the mercury adsorption capacity of the modified sample. Adsorption tests under the flue gas mixture revealed that the modified activated carbon is a promising sorbent for capturing mercury vapor in flue gas.

TG5-P8 — 11:00-12:00 and 17:30-18:30
Authors: ECONOMY, James1, YAO, Yaxuan1, VELPARI, Vedagiri2
(1) University of Illinois, jeconomy@uiuc.edu; (2) PPG.

Mercury is considered one of the most harmful elements to human beings due to its health impacts, persistence, bioaccumulation and volatility. Every year, around 85 tons of mercury is generated from burning of coals in USA. Of which, only 40% is captured, and 60% is emitted into the atmosphere. The present study involved preparation of chemically treated activated carbon fibers (ACFs) for mercury removal from power plant effluents. The incorporation of surface functional groups on ACFs, such as bromine, chlorine and sulfur species, can strongly enhance mercury adsorption performance. Thus, with one of the brominated ACFs, mercury uptake capacity was approximately 103 mg/g C, which was around 15 times compared to the original ACF. Another effective system was based on a 6~7 wt% sulfur treated ACF with a mercury uptake of 11~15 mg/g C. Besides specific chemistry, pore properties also play an important role on mercury adsorption. Generally, samples with 35 to 45 Å in size with an equal volume of micropores and mesopores appear to show good mercury adsorption performance. The micropores appear to be the active sites for mercury adsorption while the mesopores acted as transport routes.

In a typical situation, activated carbon powder (ACP) is injected into the power plant effluents, adsorbs mercury, and then is captured along with fly ash using a FF (Fabric Filter) or an ESP (Electrostatic Precipitator). Although ACP provides superior contact efficiency, a problem arises in collecting and handling. Compared with ACP, ACF shows greatly improved contact efficiency and the potential for continuous use. Additionally, such fibers can be cost competitive to ACP and exhibit excellent wear resistance. By coating the glass fabric with the preferred coating discussed previously, FF will show a combined function of fly ash removal of the glass fabric along with mercury removal by the carbonaceous material.

In this presentation, preparation and evaluation of the above indicated Hg adsorbents is discussed.

TG5-P9 — 11:00-12:00 and 17:30-18:30
Authors: GILMOUR, Cynthia1, RIEDEL, Gerhardt F.1, RIEDEL, Georgia S.1, CHU, Paul2, GOODRICH-MAHONEY, John2
(1) Smithsonian Environmental Research Center, gilmourc@si.edu; (2) EPRI;

Flue-gas desulfurization (FGD) is used to sequester SOx from stack gases of coal-fired power plants. Stack gases pass through a calcium carbonate spray chamber, where SOx is trapped as solid CaSO4 (gypsum). FGD can also capture flue gas Hg (as well as Se) emissions from power plants. Mercury, Se and other elements accumulate in FGD process waters, presenting a wastewater treatment challenge to plant operators.

The Electric Power Research Institute is working with power plants to develop and test technologies for FGD wastewater treatment. One potential treatment approach is anaerobic biological reduction systems.

We undertook a study to investigate the primary mechanismsof Se and Hg removal in compost-based anaerobic columns Our overall goals are 1) to examine the forms in which Hg and Se are sequestered in bed materials, 2) to understand how the chemistry of the treatment water impacts Hg and Se removal.

We used anaerobic sequential extractions to help characterize the chemical forms of Hg and Se in bed materials from a column test system at a power plant. We found that about 75% of both Hg and Se were associated with organic matter, defined as material extractable in KOH. A small fraction in the residual was soluble in HCl, which represents amorphous FeS and carbonates. For Se, a small, but measurable fraction was CS2 extractable (elemental Se). The remaining roughly 25% of both elements remained in the residual, pyritic phase.

Flow-through laboratory columns packed with spent mushroom compost were used to begin to characterize the chemical conditions and microbial process that support Hg and Se sequestration. Mercury retention by the columns began as soon as the columns became anoxic; however sulfide production was not required for Hg retention, nor did it enhance retention. Columns removed >99% of Hg and reduced Hg in effluent to ~15 ng/L. Se retention ranged from 80-95%, poor compared to Hg, but the data strongly suggest that sulfate reduction and/or sulfide production is important from the effective removal of Se, as the decline in Se is more pronounced in the portions of the columns where sulfide is present. Thus, measures to promote sulfate reduction (and sulfide production) may be productive for Se removal.

TG5-P10 — 11:00-12:00 and 17:30-18:30
Authors: SALIM, Vera M M1, RESENDE, Neuman S.1
(1) Federal University of Rio de Janeiro UFRJ-Coppe, vera@peq.coppe.ufrj.br

Hg species immobilization by metal doped calcium phosphate

Decontamination of natural gas and water polluted by mercury is still a challenge to the environmental science regarding safe storage and post-removal treatment of used sorbent.

Mercury has been detected at low trace levels in natural gas worldwide and must be removed below detectable levels of 0.01 µg/Nm3 to avoid corrosion and several additional failures in natural gas plant.

The contaminated material resulting from such processes needs to obey more and more restrictive laws and regulations for safe disposal.

Concerning this request we studied the mercury immobilization in Cu doped calcium phosphates materials used in decontamination process on gas system.

Sorption experiments for mercury removal from contaminated gas (9 ng Hg0/mL N2) were carried out in fixed bed reactor at room temperature and atmospheric pressure with on line cold-vapor Hg0 analyzer - Zeeman atomic absorption spectrometer (RA-915+, Lumex). The mercury concentration of spent adsorbent (24 mg Hg/gadsorbent) was measured by Lumex equipment RA-915+ with RP-91C attachment for solid samples analysis.

Stabilization of metallic mercury (Hg0) in a CuSx doped synthetic calcium apatite has been evaluated by lixiviation tests and thermal treatment.

Exhausted samples were suspended in water, with pH 5 and room temperature, from 24h to 495h. Hg0 from extraction solution was measured by cold vapor atomic absorption method using FIMS-400 – Perkin-Elmer. The Hg released (accumulated mass) from multiple extractions after 495h was 1.18x10-4 mg Hg/50 mL that corresponds to 0.52%relative to initial Hg mass presents in adsorbent.

The samples were also submitted to thermal treatment in the same sorption system using N2 flow (30 ml/min) and heating, first at constant rate (10°C/min), up to 60°C, and then, kept at this maximum temperature for 10 h. The released mercury was measured on line by Zeeman atomic absorption spectrometer (RA-915+, Lumex) equipment. The results show that no mercury was detected on gas effluent.

From the stabilization results and solid material characterization we were able to point out that the immobilization phenomenon maintains Hg0 ions firmly on CuSx held into Ca10(PO4)6OH2 structure.

TG5-P11 — 11:00-12:00 and 17:30-18:30
Author: SNIDER, Graydon 1
(1)McGill University, graydon.snider@mcgill.ca

Coal burning and waste incineration are major sources of atmospheric gaseous mercury. We are exploring means to remove mercury gas directly from these emissions.

Ultra-violet light irradiated over titanium dioxide will readily oxidize mercury to HgO(s). The surface-based reaction follows a Langmuir-Hinshelwood mechanism. In a previous study the rate constant k and Langmuir adsorption constant K were compared with literature values.

Current work focuses on the effects of mercury oxidation and desorption with concomitant coal flue gases (H2O, CO, NO2, and SO2) at ambient and elevated (T > 100°C) temperatures. We have found high concentrations of SO2 and H2O show no apparent effect on reaction rates. High levels of NO2 oxidize mercury to Hg(NO3)2 by an uncertain mechanism. Above 100°C, CO will reduce HgO back to elemental mercury. Details of the experimental conditions, and how they apply to a coal flue gases, will be discussed.

TG5-P12 — 11:00-12:00 and 17:30-18:30
Authors: BOLLEN, Anne1, BIESTER, Harald2
(1) Institute of Earth Sciences, University of Heidelberg, anne.bollen@geow.uni-heidelberg.de; (2) Institute of Environmental Geology, Technical University of Braunschweig.

A sometimes considered approach in soil remediation techniques is the use of natural heavy metal complexing agents such as amino acids to form soluble organic complexes for soil leaching. L-cysteine is the essential amino acid with the highest affinity for soft metal ions and is considered as a potential leaching agent to remove Hg from contaminated soils. Mercury (Hg) mobilisation from soils by L-cysteine has rarely been investigated and little is known about the stability of Hg-L-Cystein complexes. The objective of this study was to evaluate the applicability of L-cysteine for mobilisation of Hg from contaminated soils containing different Hg binding forms such as Hg adsorbed to mineral surfaces, Hg bound to soil organic matter, and Hg sulphide (HgS). Using laboratory-scale batch and column experiments, soils were subjected to extraction using L-cysteine solutions with S:Hg-molar ratios of 1, 2, 10, 20, 100 and 200. In 24 h-batch experiments, the addition of L-cysteine led to an increase of Hg in the leachates of 42% for soils with Hg bound to mineral surfaces. In additional column experiments the maximum Hg removal rate was 75%. In comparison, leaching with water could only mobilise 1% of inorganically bound Hg, proving the high mobilisation potential of L-cysteine. For soils with organically bound Hg or HgS, only 1-5% of Hg could be mobilised. Thus, the extraction of Hg from soils with L-cysteine is highly dependent on soil composition and the Hg binding form in the soil. Hg speciation analyses of leachates indicate that Hg-L-cysteine-complexes are mainly easy reducible and labile complexes. Speciation analysis in soil samples using a Hg thermo desorption method revealed that, besides the formation of Hg-L-cysteine-complexes, reduction to elemental mercury Hg(0) takes place at low S:Hg-ratios (1 to 10), assumingly by microbial activity. At higher S:Hg-ratios of 10 and 100 precipitation of stable Hg-S complexes could be observed. L-cysteine shows a high mobilisation potential for inorganically bound Hg, but Hg species transformation processes and the labile character of Hg-L-cysteine complexes are limitations for considering L-cysteine leaching as a remediation strategy.

TG5-P13 — 11:00-12:00 and 17:30-18:30
Author: TURNER, Ralph1
(1) Azimuth Consulting Group, rrtgeo@direct.ca

Extensive remediation of soil and groundwater at a large chlor-alkali plant site in British Columbia was completed in 2003. Onsite and intertidal groundwater, as well as surface water and selected biota within the groundwater discharge zone, have subsequently been monitored intensively to demonstrate compliance with established criteria and to track recovery of these environmental media. In spite of extensive excavation, treatment and/or removal of contaminated soil, as well as a period of exhaustive extraction and treatment of contaminated groundwater, subsequent monitoring failed to demonstrate acceptable continued recovery three years after remediation was completed. The initial groundwater extraction program was implemented in July 2001and halted in September 2002 after 50,000 m3 of contaminated groundwater had been extracted and treated. At this time mercury concentrations in boundary wells had decreased below 1µg/L, an interior plume had virtually disappeared and most soil remediation was nearing completion. However, within a few months mercury in boundary wells had returned to former levels and by 2006 mercury in foreshore (intertidal) groundwater was exceeding established “trigger” levels. A new program of groundwater extraction and treatment was initiated in 2007 focused on maximizing mercury capture instead of complete plume capture. Monitoring results since 2006 have demonstrated both effectiveness of the renewed groundwater extraction as well as a “superimposed” longer term trend of decreasing mercury concentrations in intertidal groundwater, surface water and biota. This work illustrates both the complexity of tracking mercury transport in intertidal groundwater and the challenges of assuring acceptable restoration of Hg-contaminated sites, few of which are monitored sufficiently following remediation.

TG5-P14 — 11:00-12:00 and 17:30-18:30
Authors: HE, Feng 1, LIANG, Liyuan1, MILLER, Carrie1
(1)Oak Ridge National Laboratory, hef2@ornl.gov

Despite decades of efforts to reduce mercury (Hg) contamination, low levels of mercury continue being discharged into the East Fork Poplar Creek (EFPC) from the Y-12 National Security Complex (NSC), Oak Ridge, Tennessee. This study has been carried out to address how to remove waterborne mercury at concentrations ranging from <1 to ~40 microgram/L (µg/L). To protect the ecosystem of EFPC, significant mercury removal has to be achieved before the discharge at the headwater of EFPC. However, the high flow rate (1,300 gpm) presents a great challenge for mercury removal using traditional treatment technologies. Two alternative methods were evaluated. One was using SnCl2 to reduce Hg(II) in the enclosed storm drain network to dissolved gaseous mercury (DGM) prior to discharge at the headwater. Field tests conducted in 2009 using ascorbic acid as dechlorinating agent achieved an overall conversion rate of Hg(II) to Hg(0) of 92% at a molar ratios of Sn:Hg of 25 (Southworth et al., 2010, ORNL/TM-2009). Engineering calculation suggested that it is feasible to construct an air-stripping system to remove DGM from 1300 gpm flow at the headwater. A pilot test is being designed and implemented by Y-12 NSC to address key questions such as Hg-removal efficiency, and the fate of tin in the environment. The other method was to remove mercury at source areas. Extensive mercury data collected in the past decades show that the mercury flux at the EFPC headwater is mainly from a small number of relatively short sections of storm drains. Laboratory testing has been conducted to evaluate the efficiency of mercury removal by 8 different sorbents, including thiol-based resins, amine-coating Forager Sponge (Dynaphore), self-assembled monolayers on mesoporous supports (SAMMS) (Chen et al., 1999, Sep. Sci. Technol. 34, 1121), and a layered metal sulfides sorbent (KMS-1) (Manos and Kanatzidis, 2009, Chem. Eur. J. 15, 4779). The micro-scale KMS-1 particles were first incorporated into the calcium-alginate beads to generate an engineering form that is suitable for water treatment. The experimental results show that the thiol-functionalized resins, SAMMS, and KMS-1 have faster mercury sorption rates and potentially higher Hg sorption capacity than GAC. Therefore, the use of these materials can reduce the size of sorption vessel and the frequency of sorbent replacement, as well as minimize the generation of secondary waste.

TG5-P15 — 11:00-12:00 and 17:30-18:30
Author: PETERSON, Mark1
(1) ORNL, petersonmj@ornl.gov

A conceptual model of mercury flux from an industrial facility to a stream in Oak Ridge, Tennessee was developed to assist in environmental management decisions and in mitigating the impacts of mercury on the surrounding environment. The task of developing a comprehensive conceptual model for the facility was challenging because of widespread distribution of mercury sources and the complex or not well understood mercury transport pathways. To develop the current model, a multi-organizational team reviewed existing conceptual models from a variety of sources, consolidated historical data and source information, gathered input from staff members with extensive site knowledge, and used recent mercury flux data from a variety of sampling programs. The conceptual model indicates that the nature and extent of mercury concentration and contaminant flux at the facility has significantly changed in the ten years since conceptual models were used for previous remedial action decisions. A new water treatment system has effectively reduced mercury inputs to the creek and is removing substantially greater quantities of mercury from groundwater than was expected. Flux from one large outfall at the creek’s headwaters appears to be a greater percentage of the overall flux leaving the site than previous years, albeit year to year variation in flux is large, and the many small sources of mercury identified in the model may also be important if the goal is to reach very low mercury levels in stream water and fish. One of the most important results of the conceptual model exercise was the identification of data gaps and potential research needs. Research necessary to develop a more robust conceptual model for the facility and creek include evaluation of mercury speciation within the facility, using new tools to detect and quantify elemental mercury; evaluation of the importance of mercury concentration versus flux on downstream media and bioavailability; evaluation of the importance of base flow versus storm flow flux; and understanding of the connections between concentrations of inorganic mercury precursors and methylmercury concentration, bioavailability, and bioaccumulation.

TG5-P16 — 11:00-12:00 and 17:30-18:30
Authors: YEE, Donald1, MCKEE, Lester J.1, LEWIS, Aurana2
(1)San Francisco Estuary Institute, donald@sfei.org; (2) Brooks Rand.

Many controls being considered for treatment of stormwater rely at least in part on removal of pollutants adsorbed to particulate fractions. A study was conducted to evaluate the fractionation of mercury for various settling times in San Francisco Bay Area storm drain runoff and sediments. Typically, around 10% of mercury in stormwater runoff is removed with the largest size fraction (~2 minutes settling). Longer settling times (~20 minutes) removed an additional 10-20% of the mercury in stormwater runoff, but the majority of mercury was not removed, likely remaining in a finer suspended phase, as other evaluations of stormwater runoff in the region using mechanical filtering show a very small percentage in the dissolved (<0.45 um) phase. In contrast, 50-80% of the mercury in sediments from areas with visible sedimentation was removed in the most rapidly settled fraction. An additional 10-40% of mercury was in a finer fraction settling within 20 minutes, with a small percentage (<10%) remaining suspended (>1 um) or in a dissolved or colloidal phase (<2% in a <1 um fraction). These results suggest that storm drain systems behave at times as sediment sorting systems, preferentially retaining larger faster settling sediments, but also trapping finer fractions. Measures for stormwater management targeting solids removal from relatively quickly settled fractions will therefore be useful for reducing mercury loads from urban drainages to receiving water bodies.

TG5-P17 — 11:00-12:00 and 17:30-18:30
Authors: THOMPSON, Jay M1, LUTHY, Richard G.2
(1)Stanford University, jayt@stanford.edu; (2) Stanford Univ..

The effect of activated carbon (AC) on mercury and polycyclic aromatic hydrocarbon (PAH) bioavailability in contaminated sediments was tested. In addition to the expected reductions in PAH availability (not discussed here) addition of 5% activated carbon (by dry weight) reduced total mercury in the pore water phase by 75% in laboratory testing. This work is ongoing. Uptake testing on the marine worm Neanthes arenaceodentata is currently underway to assess the effectiveness of activated carbon in reducing mercury uptake to a sediment ingesting organism.

We hypothesize that the sorption of mercury to the activated carbon phase is due to the sorption of dissolved organic mater (DOM) mercury complexes. To test this hypothesis, we are currently conducting fractionalization procedures based on complex molecular weight, hydrophobicity, and effective diffusion coefficient on pore water from both AC amended and unamended sediment. The identification of the species that are sorbed to the AC phase is critical to understanding the mechanism of the observed reductions in pore water Hg and the eventual creation of models to predict the effectiveness of AC amendment.

TG5-P18 — 11:00-12:00 and 17:30-18:30
Authors: SOUZA, Carlos G.1, OLIVEIRA, Debora M.1, SOBRAL, Luis G.1, BRAGA, Paulo F.A. 1, DA SILVA, Tiago T.2
(1) Centre for Mineral Technology CETEM/MCT; (2) Braga.

The unsuitable discharge of exhausted fluorescent bulbs in the environment has been causing a severe mercury contamination, bearing in mind the huge number of those lamps being produced worldwide. Once broken, elemental mercury vapour is released reaching different environment compartment and bio-metabolized and further transformed into organo-mercurial species and absorbed by bio-indicators causing a real harm to human being. This technical contribution aims at defining a physical-chemical process for transforming the mercury droplets into mercury sulphide while reacting with elemental sulphur. After cutting the lamps ends, the mercury-bearing white stuff, covering the inner surface of the lamp (Chlor-fluorapatite - Ca10(PO4)6(OH,F,Cl)2), is collected and mechanically suspended in water. Before mixing the elemental sulphur powder, few drops of a surfactant is added so as to change its surface into a hydrophilic one for better reacting with the elemental mercury producing mercury sulphide (HgS). Such sulphide is further removed by using a micro-flotation column, and the experimental conditions for this to occur are being defined

Tuesday, 26 July, 2011