WO2008008475A2

WO2008008475A2 - Process for the removal of mercury from gases

Info

Publication number: WO2008008475A2
Application number: PCT/US2007/015965
Authority: WO
Inventors: Peter J. Hurley
Original assignee: Solucorp Industries, Ltd.
Priority date: 2006-07-11
Filing date: 2007-07-11
Publication date: 2008-01-17
Also published as: WO2008008475A3

Abstract

A method is provided for removing elemental mercury from a combustion gas by allowing the combustion gas to contact an aqueous dispersion or slurry (which can be a spray or aerosol) containing an alkaline-earth metal sulfide and a hyperdispersant or other surface- active agent, and preferably a phosphate and a pH buffer.

Description

PROCESS AND REAGENTS FOR THE REMOVAL OF MERCURY FROM

COMBUSTION GASES

CROSS-REFERENCE TO RELATED APPLICATION^)

[0001] This Patent Application claims the priority benefit of U.S. Provisional Application Serial No. 60/830,193, filed July 11, 2006, entitled, "Process and Reagents for the Removal of Mercury from Combustion Gases," the entire content of which is hereby expressly incorporated by reference.

BACKGROUND OF THE INVENTION

[0002] Mercury pollution emitted by the burning of lignite coal has become of increasing concern in the United States in recent years, and resulted in new regulation and restriction of mercury in power station stack emissions. Mercury is present in spent combustion gas in the form of ionized mercury, (Hg¹ and Hg¹¹) and in the form of vaporized elemental mercury (Hg°). The principal problem experience by the various mercury removal technologies is to remove the mercury in both its elemental and ionic forms. Numerous alternative reagents and techniques have been employed to remove the mercury pollution from the spent gas, including: activated carbon, other sorbents, and reagents including various sulfide technologies.

[0003] The relative ease of removal of ionic mercury by aqueous scrubbing has been demonstrated by several technologies. To capitalize of this phenomenon, processes have been developed to treat coal feedstock with oxidizing agents such as calcium chloride prior to combustion. These oxidizing agents act within the combustion process to largely convert the elemental mercury present to its ionic form and thereby facilitate total mercury removal by downstream capture in wet and dry scrubbing systems. However, addition of chloride salts to the coal feedstock may act to increase corrosion of the incinerator and flue system and may also facilitate the formation of halogenated dioxins within the exhaust gases. Also, once captured in the scrubbing system, the ionic mercury compounds may still require treatment to render them insoluble and non-hazardous to human health and the environment. [0004] Molecular Bonding Systems (MBS™) reagents from Solucorp Industries Limited (West Nyack, NY), are known reagents for the successful remediation of a broad spectrum of heavy metal contaminants. Molecular Bonding Systems SITE Report EPA/540/R-97/507. They have known efficacy as fixation agents for heavy metals in both their compound and elemental forms. U.S. Patent Application No. 1 1/1 18,107 (Pub. No. 2005-0244319-Al, the entire contents of which are incorporated by reference herein, describes the use of MBS reagents for removing both ionic and elemental mercury from coal combustion gases. The success of MBS™ reagents in removing mercury from coal combustion gas is also detailed in "Mercury Control Technologies for the Electric Utilities Burning Sub-bituminous Coals ", Final Report, Benson S. et al. Energy & Environmental Research Center, University of North Dakota. June 2005.

[0005] To remove mercury or, indeed, other heavy metals, from a combustion gas, one or more heavy metal fixation agents (sometimes referred to as "integrated fixation system" (IFS) reagents), as sparingly soluble to insoluble solid material, are dispersed in water and typically applied as a mobilized wet slurry or aerosol spray, within a wet or dry scrubber system. When an aqueous slurry of MBS reagents of 30-40μm mean particle size was introduced into a dry scrubbing system, where total (ionic and elemental) mercury content of inbound coal combustion was assayed at 9.0μg/M³ and elemental mercury content assayed at 7.8μg/M³, the reagents removed 83.3% of the ionic mercury component and 12.8% of the elemental component. (NB: All mercury determinations undertaken by Ontario Hydro mercury assay.) [0006] Unlike Mercury (I) and (II), elemental mercury has little affinity for water, its solubility being a mere 0.28μ moles (56 μg) per liter at 25°C. In addition, the equilibrium concentration of mercury in air at 25°C and latm is approximately 400 μg per liter. Thus, the capture rate of Hg° by the reagent(s) will depend upon reaction of the Hg⁰ with the reagent(s) at the surface of the aerosol droplet or water surface and/or its ability to diffuse into the aerosol droplet or water surface. Due to the unfavorable partition between the two environments, the conditions for capture of elemental mercury in the aqueous phase are regarded as suboptimal.

(0007] However, Hg° has a higher affinity for certain organic solvents systems, its solubility being approximately 10-12 μ moles per liter in aromatic solvents, 7-8.5 μ moles per liter in aliphatic solvents, and 5-7 μ moles per liter in ethers. "Thermodynamics of the Solution of Mercury Metal", J. Phys. Chem., Spencer N. & VoigtA., VoI 72 (1968) 464-470. and "The Solubility of Mercury and Some Sparingly Soluble Mercury Salts in Water and Aqueous Electrolyte Solutions " H. Lawrence Clever, Susan A. Johnson, and M. Elizabeth Derrick, J. Phys. Chem. (1985), Vol. 14, Issue 3, 631-680. Mercury's "lipophilicity" is evidenced by its uptake in the flesh of large fish, notably tuna and salmon. The present invention draws upon this property to address the problem of elemental mercury in flue and other combustion gases.

SUMMARY OF THE INVENTION

[0008] The present invention provides an improved air pollution control process in which a heavy metal fixation reagent is preferably prepared as an aqueous dispersion or emulsion of a sorbant, (e.g. activated carbon) and/or a fixation reagent, (e.g. an alkali or alkaline earth sulfide), which is mixed or finely milled with a surface modification agent or phase-transfer catalyst, so that the dielectric character of the slurry or spray is rendered more apolar or organic solvent-like, thereby improving the diffusion and capture of elemental mercury in the water-based slurry or injected-spray reagent from the gas phase. Preferably, a phosphate and a pH buffer are also present.

BRIEF DESCRIPTION OF THE DRAWINGS

[0009] These and other features of the invention will become better understood when considered in conjunction with the following Detailed Description and accompanying drawings, wherein:

[0010] FIG. 1 is a graph of scrubber vessel hydrodynamics for various rates of feed of mercury-removal agents according to the present invention;

[0011] FIG.2 is a graph of scrubber vessel hydrodynamics for various rates of feed of mercury-removal agents according to the present invention;

[0012] FIGS. 3A-C are scanning electron microscope images of solid residues obtained from a mercury scrubber according to the present invention;

[0013] FIG. 4 is a scanning electron microscope image of ash obtained from the residue of a scrubber according to the present invention;

[0014] FIG. 5 is a graph depicting mercury removal vs. IFS reagent concentration at pH

4, according to the present invention;

[0015] FIG. 6 is a graph of mercury removal vs. IFS reagent at concentration pH 6 according to the present invention, and;

[0016] FIG. 7 is a graph of mercury concentration vs. time for a test run in a scrubber according to the present invention.

DETAILED DESCRIPTION

[0017] In one embodiment of the invention, an improved fixation agent slurry for use in a wet or "dry" scrubber is prepared from an alkaline earth metal sulfide, a phosphate (such as trisuperphosphate or calcium phosphate), and a pH buffer, e.g., an alkaline earth metal carbonate and/or hydroxide, and a surface-modifying agent or phase transfer catalyst. Optionally, the slurry also includes a coagulating agent and/or an anti-foaming agent, the latter to reduce foaming when used in a wet scrubbing system.

[0018] A preferred fixation agent for the remediation of mercury (and other heavy metals) is technical MBS 2.1™, a 3:2:1 (wt:wt) mixture of calcium carbonate, calcium sulfide, and trisuperphosphate, available from Solucorp Industries Ltd. (West Nyack, NY). In addition to the CaS, CaCO₃, and TSP, MBS 2.1™ can also contain CaO, mixed calcium phosphates, magnesium adducts, calcium silicates, silicon carbide grit, and trace amounts of iron and other metals. Unstabilized without a surface-modifying agent, MBS 2.1™ has a tendency to agglomerate or recrystalize from solution.

[0019] Broadly, two classes of surface-modifying agents can be identified for use in the practice of the invention: surfactants and dispersants. Within these two classes are a number of sub-genuses, including linear surfactants or detergents (having a polar region and a lipophilic region, which can be aromatic, aliphatic, or mixed aliphatic/aromatic), cyclic analogs, and so-called "hyperdispersants" having multiple polar functionalities. A number of particular surface-modifying materials can be identified, including water soluble linear and cyclic ethers based on polyethylene glycol (PEG), modified with one or more phosphate or other polar functional groups; copolymers of such compounds; etc. Polymeric hyperdispersants, such as Solsperse 4000 and Solplus D540 (described below), are two specific examples.

[0020] The surfactant is ideally of a type that can mimic a lipid-like coating on the surface of water, such as a linear alkyl phosphate or linear alkyl-alkoxy phosphate, Solsperse 4000 and Solplus D540 being two examples. The IFS reagent, being calcium-rich, has a tendency to suppress the surface-active properties of such materials through formation of calcium surfactant salts. Thus, these phosphate-type surfactants can have only a limited lifespan in the presence of the reagent. The loss of surfactant quality can be detected by a loss of slip or slipperiness when rubbed between the fingers, and by an increase in surface tension over time. Additionally, as these surfactants have an affinity for the reagent surface, as well as the water-air interface surface, they should be added at a sufficiently high level to populate both the solid reagent-water interface surface and the water-air interface surface. Thus, a preferred reagent system is a freshly mixed preparation of a 20w/w milled, aqueous dispersion of IFS solid reagent (e.g., MBS 2.1TM, a 2:3:1 w/w/w blend of technical calcium sulfide, trisuper phosphate, and calcium carbonate) blended lOOpbw to ≥ lpbw with surfactant.

[0021] In a preferred embodiment, the surfactant is non-foaming. Foaming arises when bubbles form at the air-liquid interface when a solution of the surfactant is aerated, and, on cessation of aeration, the bubbles at the interface survive for more than about 30 seconds. Agents whose bubbles do not survive for more than 30 seconds are defined as non-foaming. [0022] According to one embodiment of the invention, MBS 2.1™ is used in combination with one or more surface-modifying agents, such as Solsperse 40000, a polyethylene glycol polymer containing pendant phosphate groups, neutralized with diethanol amine (DEA) or other alkylated amines, and Solplus D540, a polyethylene glycol non-ionic surfactant. Both materials are available from Noveon Inc, Cleveland OH 41141. Other dispersants are described in published U.S. application, Publication No. US 2004/0258608 Al (incorporated by reference herein). Combinations of surface-active modifying agents (hyperdispersants, sufactants, phase-transfer catalysts, etc.) can also be used. [0023] Use of the present invention as a means of sorbing elemental mercury from a flue gas or other combustion gas is facilitated by keeping the particle size of each dispersed particle of MBS reagent or other fixation agent below 30 microns, more preferably less than 5 microns; most preferably as low as 1 micron or smaller. Particles having a size less than 30 microns can be used in a scrubber that incorporates 50 micron jets in the slurry atomizer nozzles. The reduced particle size also increases the efficiency of the overall process by maximizing the surface area available for interaction with mercury.

[0024] Ln one embodiment, the fixation agent is prepared by blending 20 parts by weight (pbw) MBS 2.1™, 1 pbw Solsperse 40000 or Solplus D540 and 79 pbw water. The resulting slurry is fine milled by passing three times through a Dyno mill bead mill containing 2mm diameter zirconium beads, resulting in a dispersion particle size of less than 1 micron mean diameter. Although settlement does occur, the formulation may be easily re-suspended by moderate agitation or stirring. The resulting dispersion is homogenised and introduced into a wet or dry scrubbing system, as described, for example in U.S. Patent Publication No. 2005/0244319, pages 2-3. The coagulating influence of the reagents is indicated by their influence on the settlement of the dispersion. A control dispersion of 20pbw MBS and 80pbw water settled 15% in 3 days. Replacing 1 pbw water with Solsperse 40000 or Solplus D540 caused the dispersion to settle by 70% and 85%, respectively, in the same period. [0025] Although not bound by theory, it is believed that surface modifying materials such as Solsperse 40000 or Solplus D540 form micelles with the MBS reagent particles in which the polar regions of the surfactant molecules are inwardly directed to the center of the micelle, and the lipid (organic) regions of the surfactant molecules point outward, away from the MBS reagent. Thus, the resulting emulsified MBS reagent particles are effectively rendered apolar in character, and can interact with lipophilic elemental mercury particles in the combustion gas. The polarity of the micelles in this model apparently is the reverse of that typically achieved by a dispersing aid or polymeric hyperdispersant. [0026] In another embodiment of the invention, activated carbon or other sorbent is used in combination with a surface-modifying agent to remove elemental mercury from a flue or other combustion gas. An aerosol of carbon particles in water is formed with the aid of a surface-modifying agent, such as a polymeric hyperdispersant described above. The resulting particles are allowed to contact a mercury-containing combustion gas in, e.g., a dry scrubber. Although not bound by theory, it is believed that, as the particles dry (shed water molecules), the graphite-like sheet structures of the activated carbon come into contact with, and sorb, vaporized mercury atoms.

Examples and Test Results

[0027] To demonstrate that IFS dispersion reagents applied in wet scrubber systems can capture elemental mercury from simulated combustion gases, and to evaluate the impact of phase transfer agents on mercury transfer and capture by IFS reagents, a number of IFS reagents were tested.

Example 1 ("IFS-2-CM [0028] A 20% w/w dispersion of MBS 2.1™ in water was prepared in a ball mill (200kg MBS 2.1™; 800 kg water). Starting with an average particle size of about 75 microns, the MBS particles were reduced in size so that about 99.5% had a particle size of less than 30 microns, and greater than 95% had a particle size of less than 5 microns. Many particles were 1 micron in size or smaller.

Example 2 ("IFS-2-K")

[0029] A 20% w/w dispersion of MBS 2.1™ in water was prepared as in Example 1 , but with Solplus D540 added as a surface-modifying agent (20 pbw MBS 2.1™, 1 pbw Solplus D540, 79 pbw). Solplus D540 has a "graduated" hydrophilicty; as a result of its phosphate- PEG-aliphatic chain structure.

Example 3 ("IFS-2-2D")

[0030] A 20% w/w dispersion of MBS 2.1™ in water was prepared as in Example 2, but with Solsperse 4000 added as a surface-modifying agent in place of Solplus D540.

[0031] To test the effectiveness of mercury sorbants according to the invention, a synthetic combustion gas was prepared by blending air, nitrogen, carbon dioxide, nitrous oxide, and sulfur dioxide, to the approximate component composition shown in Table 1 , using a nitrogen carrier gas. The combustion gas was then spiked with mercury, and the resulting gas was heated and fed into a mercury scrubber vessel.

Table 1

[0032] At the start of each test run, prior to introduction of MBS™ reagent into the scrubber, the baseline outbound gas level was established (Table 2). Other than removal of sulfur dioxide, the levels of nitrogen oxides and elemental mercury remained unchanged in the control.

Table 2 Typical Inbound Gas Analysis

The scrubber vessel was connected separately to a source of saturated calcium hydroxide solution and to a source of an aqueous calcium sulfate suspension. Outlet gases from the scrubber vessel was directed to spectrophotometers for continuous monitoring of SO₂ and NO_x, and Hg.

[0033] The studies undertaken examined the effects of IFS reagents at two pH's: 3.8 ± 0.2, typical of what might be expected under extreme conditions of excess gases experienced by the scrubber liquor at the spray head, and 5.8 ± 0.2, typical of the whole scrubber performance.

[0034] Assessment of the impact on reagent and limewater feed on eventual reagent concentration in the scrubber vessel and the time required to achieve steady state concentration of IFS reagent at various feed rates are illustrated below, in Figures 1 and 2. Overall, the hydrodynamic models indicate that the scrubber vessel will achieve a steady state concentration within 45 minutes and 30 for low and high pH regions, respectively. The theoretical steady states predicated by calculation were observed in practice during the tests, within the parameters of the periodic varying lime feed.

[0035] The test procedure involved establishing stable baseline conditions of response for pH, mercury, sulfur dioxide, and nitrogen oxides for a gypsum (calcium sulfate) suspension in the scrubber vessel, and feeding the gypsum suspension whilst simultaneously feeding in mercury-contaminated gases. The feed of gypsum was ceased and then the reagent under test, being continually stirred, was charged into the scrubber vessel at a constant rate. The subsequent impact on mercury, sulfur dioxide, and nitrogen oxides was monitored against time, whilst maintaining broadly stable pH conditions by periodically adjusting the lime water (calcium hydroxide) introduced into the scrubber.

[0036] Sulfur dioxide scrubbers generally operate at mean pH between 5.5 and 6.0, and achieve greater than 95% removal of inbound sulfur dioxide. However, in regions within the scrubber, specifically near the region of gas inlet and high aeration near the spray jets, localized acidity can occur, leading to a pH in these regions approaching 3.5. Thus, the control reagent was evaluated at two pH ranges to represent the chemistry of these regions, pH 3.8 and pH 5.8, with the carbonate-based standard MBS 2.1™ chosen as the basis for a control reagent, "IFS-2C" (Example 1) prepared as described above. Additionally, to facilitate ease of pH control, maintain pH stability, and counter pH drift that can occur through sulfur dioxide absorption, lime water (dilute soluble hydrated calcium hydroxide) was added rather than calcium carbonate (as occurs in industrial scrubbers), such that impact on pH was immediate and no lag in response caused by alkali dissolution would be experienced as would be the case if calcium carbonate slurry were solely relied on. [0037] Testing reagent IFS-2C at both high and low pH's indicated good levels of reaction with elemental mercury. From 5-55% Hg⁰ removal from a carrier gas mixture containing 86 μg/M³ elemental mercury was observed, the magnitude of effect depending upon reagent concentration and pH. (See Tables 3 and 4 below).

[0038] Under the acidic conditions of the low pH region, there was a noticeable odor of hydrogen sulfide in the outlet gas, roughly estimated at 5-10 mg/M³. In contract, at the higher pH, reagent feed concentrations of less than 2g/Liter gave rise to a barely detectable odor. Given the results, it might be reasoned that, on scale up, there may be some generation of free hydrogen sulfide in the low pH regions of a scrubber. However, from the comparative absorption of sulfur dioxide and hydrogen sulfide it can also be reasoned that any hydrogen sulfide generated should be reabsorbed in the higher pH regions of the scrubber along with the accompanying sulfur dioxide. Therefore, any evolution of detectable odor of hydrogen sulfide should only be apparent if the scrubber failed to meet its overall sulfur dioxide removal performance.

[0039] Spent scrubber liquors retained a slight odor of hydrogen sulfide. On scale-up, a discharge agent, such as weak peroxide, can be utilized to minimize or eliminate this odor. Alternatively they can be simply treated with Hydros (sodium hypochlorite). [0040] Solid residues were isolated by filtration on a Whatman No.44 paper, yielding a firm press-cake and clear mother liquor. The solids were slightly grey colored, arising from 'ash' contamination within the IFS reagent. Calcium sulphite hydrate is unstable in air giving rise to an odor of sulfur dioxide. The isolated residues have no significant odor, of either sulfide or sulfur dioxide. Furthermore, calcium sulphite is known to be soluble in acidic liquors and solutions of sulfur dioxide. The principal reason for industrial concern over slow oxidation of sulphite to sulfate is the possibility of precipitation of calcium sulphite-sulfate, which is of finer, and poorer, crystal form, which in turn can cause presscake de-watering problems. SEM analysis indicates that the residue is predominantly monoclinic crystals of 10-200 micron particle size (Figure 3a-c). Quantitative analysis indicates that the isolated residues were predominantly gypsum hydrate with some apatite. Industrially, gypsum is isolated from scrubber slurry with a typical particle size range of 5-150 micron. Thus, we can say that the IFS-laboratory process will generate a comparable product. [0041] There is a common misconception within the power industry that trace metals are 'necessary' to catalyze the oxidation of calcium sulphite to sulfate. Manganese and iron have been shown to accelerate the oxidation, but they are not essential for the process and this catalysis cannot replace the benefit of providing excess oxygen through correct design of liquid gas mixing within the scrubber system. Sulphite, provided with excess oxygen will 'rapidly' revert to sulfate, without the need for catalysis. However, IFS does carry iron as a known contaminant arising from residual ash contamination. The ash iron component was identified by SEM as an iron magnesium aluminosilicate hydrate, (dark phase, Figure 4). This 'ash' will have similar chemical and physical properties as the fly ash, which routinely contaminates gypsum in the industrial scrubbing process and is widely attributed to be a catalyst favoring sulphite to sulfate oxidation. As an additional note, IFS reagents generate a significant grey-green color on aging, due to the interaction of its active components with trace metal (principally iron) contaminants freed during processing. This color, arising from 5 week old reagent, significantly tinted the scrubber solid residues when used at concentrations above Ig / Liter in the vessel. However, it was noted that this color discharged on exposure to excess sulfur dioxide. Color arising from fresh reagent should have negligible impact on residue color of industrially produced gypsum. [0042] Utilizing estimated scrubber vessel concentration figures, derived using data from a vessel hydrodynamic model, it was possible to plot the influence on elemental mercury removal against reagent concentration in the vessel. The graphs (Figure 5 and 6) demonstrated a plateau effect, where increasing concentration of reagent in the scrubber vessel beyond a certain level gives ever decreasing benefit in terms of mercury removal. The occurrence of this effect clearly indicates a limitation on transfer of elemental mercury from the gas to the aqueous phase, within the system constraints of the laboratory model. [0043] Whilst overall practical concentrations of IFS-2C addition - within the laboratory model - are 0.5-1 dry weight gram /liter, these may represent only a starting point guide for scale-up evaluation. These levels can to be re-assessed in the light of the material dynamics of the industrial scale process.

[0044] Comparative assessment of a different mercury remediation agent, NaHS at 2 g/Liter, equating to at least 6- fold the comparative strength per equivalent dry weight of the IFS reagents, indicated at best a 4% removal of elemental mercury at the higher pH. As anticipated, NaHS, unstabilized, though more chemically available, is more susceptible than IFS to oxidation and acid displacement loss from the system.

Table 3 Outbound Gas Analysis at Equilibrium, at ~ pH4 (3.8 ± 0.2) and 55°C

Table 4 Outbound Gas Analysis at Equilibrium, at ~ pH6 (5.8 ± 0.2) and 5S°C

# NB: 2.0g /Kg of NaHS contained the comparable sulfide content to 12.5 g/Kg IFS Reagent on dry weight bases

^♦ NB: Inbound HgO baseline corrected for direct comparison. [0045] IFS-2K was an ethylene oxide surface-modified version of IFS-2C. Evaluation of its mercury removal performance indicates that it performs exactly the same as IFS-2C within the scope of experimental and preparation reproducibility. Furthermore, addition of further portions of the 2K additive, Solsperse 4000, to a concentration of 1 part per 1,000, gave no benefit in performance and displayed undesired foaming, within the limited headspace of the laboratory model. Such foaming would be less likely to cause a problem within an industrial scrubber.

[0046] IFS-2D was an alkoxy-alkyl surface modified version of IFS-2C. Evaluation of its mercury removal performance indicates that underperformed against IFS-2C. However, addition of further portions of the 2D additive, Solplus D540, to a concentration of 1 part per 1 ,000, gave a doubling in performance but displayed undesired foaming within the limited headspace of the laboratory model, as previously experienced with IFS-2K. (See left portion Figure 7). After a period of 20-30 minutes, the enhanced performance of the surfactant began to fail and eventually total disappear. Subsequent examination of the scrubber liquors indicated a loss of slip, showing the surfactant to have been chemically degraded or sorbed to the surface of the gypsum and/or reagent, and become expelled from the system. Subsequent discussions with the manufacturer indicates that the latter is the more likely scenario, as the surfactant was deemed to be stable to all components and conditions encountered within the model, with the exception of exposure to calcium hydroxide. Solplus D540 will not be adversely influenced by calcium carbonate, but exposure to high levels of soluble calcium, via the use of hydroxide may have caused some debilitation of surface active properties of the surface active synergist.

[0047] Further study of IFS-2D indicates that the additive D540 gives rise to a sharply increased efficacy of the IFS reagent. However, the negative aspect is that it is unstable to storage, losing 50% of its efficacy within 5 weeks of preparation.

[0048] Given the above, IFS-2C at a concentration of 1 g/Liter was re-evaluated, co- mixing Solplus D540 as a synergist at 1 part per 1000. The result was a 100% improvement in performance of the now fully active reagent, achieving a better than 50 % removal of elemental mercury (See right portion Figure 7). However, the accompanying problem of foaming within the low head space model restricted further evaluation. Subsequent addition of a conventional siloxane antifoam, Fomex 1488 (Tego Chemie AG), at a rate of 30mg/liter, totally suppressed foaming but gave rise to a slow deterioration of observed enhanced performance.

[0049] The excess Solplus D540 (an alkoxyl-alkyl co-block polymeric surfactant) in the system was projected to give an apolar quality to the aqueous surfaces exposed to the elemental mercury carrying gas bubbles. The fact that addition of this synergist enhances mercury capture indicates support for the hypothesis that capture of elemental mercury, which is 'apolar' in nature, will be enhanced by making the aqueous phase surface more apolar in quality. Furthermore, additional support for this hypothesis was gained when we partially disrupted that modification to the system by addition of an immiscible siloxane surface active agent (the antifoamer), resulting in the positive synergist effect being suppressed. Interestingly, co-mixing all components; BFS-2C, D540 and antifoam, prior to commencement of the experiment, gave no additional beneficial effect on mercury removal over and above that of IFS-2C used solely, indicating micelle formation to be a critical influence on elemental mercury transfer to the aqueous phase.

[0050] NB: Solplus 40000 and Solplus D540 are not considered to have foaming problems in industrial use as their foams are weak and generally disperse on cessation of aeration. They are sold as surface modification agents and not as general surfactants for increased slip and foaming applications. In an alternate embodiment, believed to be useful with a scrubber system that has adequate tolerance to foaming, lower cost Brij 700 alkoxyl- alkyl co-block polymeric surfactant is employed.

[0051] Give the performances of the test reagents IFS-2C, 2K and 2D above, the invention is supportive of a commercially viable mercury removal agent in both wet and dry systems. Based on the evidence of the plateau effect, it is believed that the reagent IFS-2C will achieve a higher percentage mercury removal if tested at elemental mercury concentrations close to those experienced industrially. Also, if surface area of the reagent can be enhanced, through the more effective milling achievable by a production mill, coupled with use of the reagent within 1-2 weeks of preparation, then it is expected that mercury removal of 50-90% is possible for elemental mercury levels of 20-40 μg/M³ within the laboratory model carrier gas.

Claims

WHAT IS CLAIMED:

1. A method for controlling air pollution, comprising: allowing a combustion gas to contact an aqueous dispersion or slurry of an alkaline-earth metal sulfide, wherein the dispersion or slurry includes a hyperdispersant that facilitates interaction between elemental mercury in the combustion gas and the alkaline- earth metal sulfide.

2. A method as recited in claim 1 , wherein the hyperdispersant comprises a polymeric hyperdispersant

3. A method as recited in claim 1 , wherein the hyperdispersant comprises a water soluble linear or cyclic ether based on polyethylene glycol (PEG), modified with one or more phosphate or other polar functional groups, or a copolymer thereof.

4. A method as recited in claim 1 , wherein the polymeric hyerdispersant comprises a copolymer of a linear or cyclic ether based on polyethylene glycol.

5. A method as recited in claim 1, wherein the hyperdispersant comprises a long chain organic phosphate.

6. A method as recited in claim 1, wherein the hyperdispersant is a copolymer that contains a phosphate group, an alkoxy block, and an aliphatic block.

7. A method as recited in claim 6, wherein the alkoxy block comprises a polyethylene glycol.

8. A method as recited in claim 1, wherein the hyperdispersant comprises a polyethylene glycol.

9. A method as recited in any one of claims 1 -8, wherein the dispersion or slurry further contains a phosphate and a pH buffer.

10. A method as recited in claim 9, wherein the phosphate comprises trisuperphosphate.

11. A method as recited in claim 9, wherein the pH buffer comprises calcium hydroxide.

12. A method for controlling air pollution, comprising: allowing a combustion gas containing elemental mercury to contact an aqueous dispersion or slurry of an alkaline-earth metal sulfide, a phosphate, a pH buffer, and a surface-active agent that facilitates interaction between elemental mercury in the combustion gas and the alkaline-earth metal sulfide, to thereby remove elemental mercury from the combustion gas.

13. A method for controlling air pollution, comprising: allowing a combustion gas to contact an aqueous dispersion or slurry of a sorbent stabilized by a surface-modifying agent that facilitates interaction between elemental mercury in the combustion gas and the sorbent.

14. A method as recited in claim 6, wherein the sorbent comprises activated carbon.

15. A method as recited in any one of claims 1-14, wherein the dispersion or slurry is provided as a spray or aerosol.