CA1119004A - Process for mercury removal - Google Patents

Abstract

ABSTRACT FOR THE DISCLOSURE

A process for the removal of mercury from aqueous media containing mercury and undissolved solids comprises, adding hypochlorite to the aqueous medium, reducing the ionic mercury to elemental mercury, aerating the aqueous medium with a gas to entrain the mercury and separating the mercury from the entraining gas.

Description

~ACKGROUND OF T~E I~VENTION/PRIOR ART
The present invention relates to a process for removal and recovery of mercury from aqueous media which contain mercury and undissolved solids. More particularly, the present invention relates to a process for the removal and recovery of mercury from chloralkali plant effluent.
Efforts are being made to prevent the loss of mercury to the environment. Some of the mercury lost to the waterways originates in the form of plant effluent, such as the liquid employed in scrubbing stack gases or effluent from chloralkali plants employing electrolytic cells with mercury cathodes.
One of the s~.hemes proposed to eliminate these losses of mercury in aqueous media comprises reducing ionic mercury contained therein to elemental mercury, aerating the medium with an inert gas which entrains the mercury, followed by recovering the mercury from the gas by scrubbing it with a chIorinated brine solution. Such a procedure is not entirely satisfactory as it generally fails to remove effectively the mercur~ which is present in the medium. The elemental mercury present in the medium prior to the addition of the reductant, tends to coalesce into larger particles which are not very susceptible to vaporization. Even the agitation of the aqueous medium by the aerating gas and the consequent fragmentation of the agglomerated mercury fails to adequately improve the eva-poration rate of the mercury. In addition, typical ef-fluents contain undissolved solids on which the mercury, both eIemental and ionic tends to adsorb or with which it might somehow associate itself, further impeding its vapori-zation and entrainment by the aerating gas.

A laboratory procedure for quantitative assay of mercury comprises boiling the sample with excess strong oxi-

- 2 -1119()04 " ` dant such as aqua regia for complete digestion of the ele-mental mercury, followed by reduction of the ionic mercury and aeration of the medium for mercury detection in the gas phase.
It is the object of the invention to provide an improved process for the removal and recovery of mercury contained in aqueous media.

BRIEF DESCRIPTION OF THE INVENTION
The present invention provides a process for the removal of mercury from an aqueous medium containing ~ercury and undissolved solids~ comprising the steps of treating said medium with hypochlorite in a molar amount which comprises at least 20 times the molar amount of said mercury in said medium, adding a reducing agent to said hypochlorite treated medium to convert substantially all ionic mercury in said medium to elemental mercury, passing an entraining gas through said medium to entrain said elemental mercury, and separating said vaporized mercury from said gas.
The single figure of the drawings is a schematic representation of the sequence of steps of an embodiment of the present process integrated into a chlor-alkali operation.
DETAILED DESCRIPTION OF THE INVENTION
A necessary step of the process of the present in-vention comprises adding hypochlorite to the mercury-contain-ing aqueous medium to convert any non-ionic mercury in the medium to ionic mercury and to dissociate any mercury from the undissolved solids with which it may somehow be as-sociated. This may be carried out at ambient temperature.
The hypochlorite added is preferably sodium hypochlorite, but calcium hypochlorite or hypochlorites of other al~ali or alkali earth metals may be used. The hypochlorite ion may alternatively be produced in situ by the reaction of chlorine gas and the respective hydroxide in the aqueous medium. The quantity of hypochlorite added to the medium must be sufficient to ensure the complete oxidation of the elemental mercury that may be present. Additionally some hypochlorite will be needed to dissociate the undissolved solids from the associated mercury. Such an amount is nor-mally at least 20 times the estimated molar concentration of mercury (elemental and ionic) in the aqueous medium.
Combining all the aqueous effluent from a chloralkali plant would normally result in a mercury level about 10 ppm (which corresponds to a ~oncentration of 0.5 x 10`4 moles/
litre), hence about 100 ppm (i.e. 13.4 x 10 4 moles/litre) of hypochlorite would be used. For an adequate oxidation of the non-ionic mercury to take place, a certain residence time is normally required. At ambient temperature, at least ~ hour is normally required, with time periods of about 3-6 hours being preferred.
The oxidized mercury, which is believed to be dis-persed throughout the aqueous medium in predominantly ionic form, is then reduced to e-emental mercury by the addition of reducing agent (such as are taught in Canadian Patent 993,174 issued July 20, 1976, to Allied Chemical, assignee of J. Guptill et al) which may be sodium boro-hydride, sodium sulfite, stannous chloride dihydrate, hy-droxylamine hydrochloride, sodium thiosulfate, triethano-lamine, hydrazine, hydroquinone, ascorbic acid, sucrose or D-glucose; or a combination of the preceding.
The preferred reducing chemical (in a non-acidic environment1 is sodium borohydride, the dosage of which is normally at least about 10 times the stoichiometric pro-portion i.e in the combined effluent of a chloralkali plant, of the type mentioned earlier, a minimum of about 5 ppm tWhich corresponds to 1.7 x 10 3 moles/litre) of sodium borohydride would be required. Since, however, the reduc-tant is first consumed by the excess hypochlorite from the preceding step, the actual dosage is much greater and may range upto about 500 ppm..
In order to prevent much coalescence of the mer-cury, the passage of an entraining gas through the medium is commenced immediately after the reduction step. This gas may be nitrogen, oxygen, or air, (air partially freed of carbon dioxide may be used if the aqueous medium contains substantial quantities of caustic soda or caustic potash).
The gas is preferably at a temperature above the ambient and may be thus obtained by passing it through a heater, heat exchange unit or by introducing steam into the gas.
The aerating gas is preferably at a temperature greater than 30C.
The mercury thus entrained may be separated from the gas by scrubbing the gas with chlorine containing brine, which is preferably acidified or by dry absorption on a bed of absorber such as activated carbon or the like. The former technique, namely, using the chlorine containing brine is more appropriate for chIoralkali plants where chIorine and brine are both readily available and the "spent" scrubbing liquid may be combined with the brine solution which is retur-ned to the electrolytic cells where the mercury in the brine solution is plated out on the mercury cathode.
One of the advantages accruing from this process is the removal of mercury from the sludge tor other aqueous media~ which would otherwise have been lost to the environ-ment or carted away and stored indefinitely. In the case of a chloralXali system, the various mercury-containing ef~luents may be combined and treated simultaneously, as this technique is effective in the presence of undissolved solids, and the gas stream bearing mercury may be combined with the other stack gases for simultaneous removal of the mercury. Furthermore, the scrubbing liquid may be combined with the rest of the brine solution and recycled to the electrolytic cells for a restoration of the mercury to the cathode, whence it was lost.
The preferred embodiment of the present invention illustrated schematically in the single figure shows the present in~ention integrated with a chloralkali operation, 100, in the diagram. The combined effluent from 100 passes through the oxidation stage 102, followed by reduction 104, following which hot gas is passed through the treated ef-fluent in the aeration stage 106. The mercury-containing gases issuing from the aeration stage 106 are contacted with the scrubbing liquid in contactor 110, while the liquid effluent from 106 proceeds to the solid separation stage 108. The separated solids from 108 may be discharged along with a portion of the treated effluent, which the rest of the effluent may be recycled to the chloalkali system 100, possibly for reuse. A portion of the scrubbing liquid in the contactor 110 may be bled off and returned to the elec-trolysis section of the chloralkali plant 100 ~or the re-moval of mercury contained therein.
As illustrated in the drawing, the various efflu-ents from the chloralkali plant 100 are combined and pro-ceed to the oxidation stage 102 where hypochIorite is added to the combined effluent. The residence time of the efflu-ent in this stage (102) is preferably about 3-6 hours to permit the oxidation to take place. The next step is the reduction, (104) the length of which will depend on the re-ductant used. Immediately following the reduction, theeffluent is aerated ~106) with a hot gas. Preferably, the aeration is carried out in a turbulent gas-liquid contactor, which enhances the transfer of mercury from the liquid to 11~9004 the gas phase. The hot gas i5 preferably air which has been heated eg. with low pressure steam. After aeration, the residual liquid is filtered or decanted, (108) the solids discharged in the form of a filter cake or thick sludge;
while a portion of the liquid is recycled to the chloralkali system (100) and the remainder is bled off and discharged.
The hot gas containing mercury may be combined with other stack gases which contain mercury, and contacted with acid-ified brine containing chlorine to strip off a portion of its mercury content in a contactor (110) which may be a packed tower or fluidized bed or a turbulent contact absor-ber. Some of the scrubbed gas may be vented while a portion may be recirculated through the scrubber and another protion may be added to the incoming hot gas for aeration of the effluent. A portion of the scrubbing liquid containing mercury is bled off and returned to the electrolytic cells where the mercury is plated out on the cathode. Fresh - scrubbing liquid is added as make-up to the remainder of the scrubbing liquid containing mercury and this mlxture circulated through the scrubber.
The examples below are intended to provide an illustration of specific embodiments of the present inven-tion, and should not be construed as limiting it in any way.
EXAMRLE I
Three 200 ml. samples of brine sludge from a chlor-alkali plant, having a solids content of 12.7% and a mercury content which had been previously measured, were treated for comparison~ according to a prior art technique and in accord-ance with the present invention. The sample of run 1 wastreated according to a prior art technique which comprised adding a reducing agent (sodium borohydride in this case) to the sample, followed by aerating the sample. Runs 2 and 3 11~9004 were carried out in accordance with the process of the pre-sent invention which additiona ly comprised adding sodium hypochlorite to the sample followed by a retention time of about 8-12 hours, adding sodium borohydride to the sample and aerating the sample after the addition of the reducing agent.
The concentration of the mercury in the treated sludge was measured at the end of the aeration; and the strip-ping efficiency of this procedure, defined as the percent decrease of the mercury content in the sludge due this treat-ment, was obtained.
The experimental conditions are summarized in Table I belo~.
TABLE I

Exp. Run Number 1 2 3 Total solids (%)12.712.7 12.7 HypochIorite Dosage tas ppm C12) 0 800 1600 Sodium Boro-hydride dosage tppm) ` 250 250 300 Temp. of Strip~ing gas ~ C1 57 54 57 Stripping Time tmin.) 20 16 22 Total mercury 30 concentration in untreated sludge tppm)15.98 16.32 16.12 Residual mercury con-centration in treated sludge tppm~ 8.02 6.66 5.77 Stripping 40 Efficiency t%) 49.8 59.2 64.2 ~119004 EXAMPLE II
This example illustrates the pilot plant scale application of the present invention in the treatment of chloralkali plant effluents. The process comprised adding sodium hypochlorite to the mercury containing liquid, follow-ed by a retention time of about 8-12 hours; then adding sodium borohydride to the liquid before contacting it with the entraining gas.
The mercury-containing aqueous medium was contact-ed with the estraining gas in counter current, in a turbulentcontactor of the type taught in Canadian Patent No.740,853 (issued August 16, 1966, to Dominion Tar and Chemical Co., assignee of H.R. Douglas) and known by the tradename "Tur-bulent Contact Absorber" tT.C.A.). A 3-stage T.C.A. with a 6 inch internal diameter, through which air heated by steam to about 55-70C, flowing at a rate of 2366-2602 l./m was used for aeration of the sample and entrainment of the mercury. The liquid circulation rate through the contractor was 18-22 l./m. The stripping efficiency in this case is obtained from the percent differences of the mercury content between the untreated aqueous medium and the bleed-off stream from the line which recycles the aerated aqueous medium for additional passes through the contactor.
The aqueous medium employed in experimental runs 1 and 2 was taken from the cell sewer of a chIoralkali plant and contained the water employed in washing the floor of the cellroom and has an undissolved solids content of 0.007~.
These experiments demonstrate the difference in mercury removal between a prior art technique and the present tech-nique when substantially all the mercury present in theaqueous medium is ionic and elemental mercury is substan-tially absent. The addition of 100 ppm. of sodium hypochl-orite to the aqueous medium before the reduction step, in experiment 2 resulted in an increase of 10% in mercury re-moval over the previous run. The slightly higher amount of sodium borohydride employed in the second experiment is due to its partial consumption by the excess hypochlorite added in the previous step.
The samples for experiments 3 and 4 were taken from a chloralkali plant perimeter sewer; the contents of which originated in spills and run-off water. The total mercury in the sample for experiment 3 was 0.9 ppm. while the sample for experiment 4 contained 5 ppm. The higher mercury content in the latter sample was a result of the addition of ionic mercury in the form of mercuric chloride solution to the perimeter sewer sample. The lack of signi-ficant difference in mercury removal between experimental runs 3 and 4 suggests that increases in the ionic mercury content of the sample does not adversely affect the removal of mercury.
Experiment 5 deals with the removal of mercury from a sample of brine sludge from a chloralkali plant, using the process of the present invention.
The experimental conditions for the various runs are summarized in Table II below.

~119004 TABLE II
Exp, Run Number 1 2 3 4 5 Total solids (%) 0.007 0.007 0.028 0.0280.48 Sample Volume Treated 40 40 60 40 40 (litres3 10 Hypoch lorite dosage (as ppm C12) 100 700 850 780 Sodium Borohy-dride dosage (ppm) 35 S0 200 250 250 20 Stripping Time (min.) 50 50 40 50 50 Temp. of Stripping Gas tC) 55-62 5~-65 57-63 58-70 . 57-63 Initial Mercury Concen-30 tration in un-treated aqueous medium tppm) 4 4 0.9 5 0.65 Residual Mercury Concen-tration 40 in treated aqueous medium tppm~ 1.4 1.0 0.~54 0.1 0.026 Stripping Efficiency t%) 65 75 94 98 96 1~19U04 EXAMPLE III
This example illustrates the application of the present invention to a process fGr the continuous removal - of mercury from aqueous media, characterized in that the residence time required for substantial mercury removal is greatly decreased as compared with the previous example.
The sample of experimental run number 1 comprised brine sludge diluted to a consistency (about 0.25% solids) similar to that of the combined effluents from a chlor-alkali plant. This dilution was carried out with a mercuricchloride solution in order to prevent the mercury concen-tration in the sample from falling below detectable levels because of this dilution. The diluted sludge sample was first treated with a sodium-hypochlorite solution and held for a period of 3-6 hours. Following this period of oxi-dation, sodium borohydride was added continuously to the oxidized sample before it entered the feed pump of the T.C.A.
The reduced mercury in the sample was continuously stripped by means of an air stream, heated by steam in such a way that a gas temperature of about 60C was maintained. The stripping efficiency of this system was given by the percent decrease in mercury concentration and was obtained from the mercury concentrations in the feed stream, prior to its dilution by the recycle and in the bleed off stream.
The sample for experimental run 2 originated as caustic filter backwash sludge and was treated in a manner essentially identical to the previous sample.
The sample for experimental run 3 was a mixture of brine sludge, perimeter sewer sludge, caustic backwash sludge, bleed off from the acid sewer, cell sewer and spent brine of a chlor-alkali plant in quantities to duplicate the essential characteristics (e.g. consistency, chemical composition, etc.) of the combined effluent from a chIor-alkali plant. The procedure employed was essentially thesame as in the previous experimental runs, with the only difference being the reducing agent, which in this case was a mixture of sodium sulfite and sodium borohydride.
The experimental conditions are summarized in Table III below:
TABLE III

Experimental Run Number 1 2 3 Solids Content (%)0.25 0.25 0.30 Sodium Hypochlorite Added tAs ppm C12) 850 745 1200 Sodium Sulfite dosage (ppm) 0 0 2800 Sodium Borohydride dosage (ppm) 300 260 97 Liquid feed Rate to T.C.A.(litres) t min~. ) 2.1 2.0 3.79 Circulation Rate in T.C.A. (litres) ( min. ) 19 17 18.93 T&mp of Stripping Gas ( C) 60-62 57-58 64.5 Stripping Time (min.)* 1-2 1-2 1-2 Initial Mercury Conc.(ppm) 4.72 1.74 35 Residual Mercury Conc.(ppm) 0.15 0.115 2.4 Stripping Efficiency96.8 93.4 93 * Average residence time of the liquid in the tower, based on residence times in each stage.
Modifications to the above will be evident to those skilled in the art, without departing from the spirit of the invention as defined in appended claims.

Claims (7)

1. A process for the removal and recovery of mercury from an aqueous medium containing solids having mercury associated therewith, substantially all of said mercury be-ing in the form of inorganic ionic mercury, comprising the steps of: treating said medium with hypochlorite in a molar amount comprising at least 20 times the molar amount of said mercury in said medium thereby to dissociate said ionic mercury from said solids, adding a reducing agent to said hypochlorite treated medium to ensure conversion of sub-stantially all said ionic mercury in said medium to elemental mercury, passing an entraining gas through said medium to entrain said elemental mercury, separating said entrained mercury from said gas thereby to recover said mercury.

2. A process as described in claim 1 wherein said aqueous medium comprises aqueous effluent from a chloralkali plant having electrolysis cells with mercury cathodes.

3. A process as described in claim 1 wherein said reducing agent contains at least one member of the group comprising sodium borohydride and sodium sulfite.

4. A process as described in claim 1 wherein said entraining gas is air at a temperature of at least 30°C.

5. A process as described in claim 2 wherein said entrained mercury is separated from said gas by stripping said mercury from said gas with chloride-containing brine.

6. A process as described in claim 5 wherein at least a portion of said chlorine-containing brine containing said stripped mercury is returned to said electroysis cells in said chlor-alkali plant.

7. A process as described in claim 1 wherein said entrained mercury is separated from said gas by passing it through an absorber.