Mercury pollution is a problem in estuarine waters in Europe, USA, and elsewhere, since the element is toxic in both the elemental and in combined forms. Many treatment methods for aqueous solutions of mercury rely on the precipitation of insoluble compounds, such as the sulphide. However, this may only transfer the problem from one source to another, since breakdown of these compounds can sometimes occur during storage at the disposal site. A method for recovering the mercury as the metal would, therefore, be advantageous, since recycling could then be achieved.

At present many types of mercury-containing wastes are disposed of in landfills. This route, obviously, does not recover the mercury, and there is also the danger that mercury may be leached from the system and give rise to contamination of ground or surface water.

Mercury may be recovered by a conventional distillation process, which involves heating the waste material in a retort, followed by condensing the metal from the vapour. However, the possibility then arises of mercury escaping into the atmosphere. To prevent this occurrence carbon fibre traps are situated at the exists of the retort, but this merely produces another waste to be treated.

The UK government has recently awarded a contract to a waste services organisation to recycle or"safely dispose"of up to 106 fluorescent tubes a year. The distillation method will be used. Suggested costs for this process at present would be of the order of £0.40 per tube. The disadvantages of the distillation method may be summarised as: the high cost (because large quantities of material have to be heated, especially if the mercury is initially present at low concentrations); the possibility of the escape of mercury vapour to the atmosphere; and the traps used to prevent emissions have to be treated themselves.

The present invention addresses the above disadvantages since the processing steps are carried out at relatively low temperature, thereby avoiding heating costs and emissions of mercury vapour.

An electrochemical cell for the removal of metals such as copper, lead, silver, tellurium, platinum, palladium and nickel, from dilute solutions of the metal is described in PCT/GB94/01929 (WO 95/07375).

Accordingly, in a first aspect, the present invention provides a process for the recovery of mercury from solid waste materials, such as, for example, sludges from brine electrolysis plants and residues from mercury vapour lamps, which process comprises: (i) contacting the waste material with a leaching agent to leach mercury from said material; (ii) passing the mercury-containing leachate from step (i) or a solution thereof into an electrochemical cell having an anode and a

cathode, and passing a direct current between the anode and cathode to thereby deposit liquid mercury on the surface of the cathode; and (iii) collecting the liquid mercury deposited on the cathode of the electrochemical cell from step (ii).

The leaching agent preferably comprises an oxidising agent and/or an acid. Suitable oxidising agents include salts of chloric acid containing hypochlorite, chlorite and/or perchlorate ions, preferably alkali metal salts. Advantageously, the oxidising agent comprises a sodium salt of chloric acid, preferably a sodium hypochlorite solution. Such a leaching agent may be used where the waste material comprises brine sludges. A concentration in the range of from 10 to 60 g Cl2/l has been found to be particularly effective, preferably from 30 to 50 g Cl2/l. If an acid is used to leach mercury from the solid waste, then it may comprise, for example, hydrochloric acid or nitric acid. An acid leaching agent is particularly effective where the waste material comprises residues from mercury vapour lamps, for example fluorescent lamps. The acid concentration is preferably in the range of from 1 to 10 %, more preferably from 3 to 7%, still more preferably approximately 5%.

The temperature at which the leaching step is carried out will generally be in the range of from 20 to 100°C. The leaching temperature will depend inter alia on the material from which the leaching reactor is formed. For a reactor formed from polypropylene, the inventors have found that a temperature in the range of from 20 to 50°C may advantageously be used, more preferably approximately 50°C.

The process according to the present invention may also enable regeneration of hypochlorite at the cell anode.

The mercury-containing leachate or solution thereof is preferably acidic since this has been found to prevent deposition of mercury as the oxide. If the leachate is not intrinsically acidic after the leaching step then it may be acidified in a conventional manner by, for example, adding an acid such as hydrochloric acid. The mercury-containing leachate or solution thereof preferably has a pH: 5, more preferably in the range of from 1 to 3, still more preferably from 1 to 2.

In a preferred aspect of the process, electrolysis in step (ii) is carried out using an electrochemical cell having a cathode formed of a porous carbon material. Electrolysis will generally be carried out at room temperature under ambient conditions.

The electrochemical cell used in step (ii) advantageously comprises a proton-conducting membrane disposed between the cathode and the anode to thereby separate the cell into a cathode compartment and an anode compartment. Proton-conducting membranes are commercially available and one sold under the trade name Nafion (Dow Chemicals) has been used. In this manner, the solution of the mercury-containing leachate may be circulated through the cathode compartment and an anolyte may be circulated through the anode compartment. The anolyte may comprise, for example, a solution of nitric acid, a solution of sulphuric acid, a solution of sodium nitrate or a solution of sodium sulphate. A suitable concentration for the anolyte is preferably in the range of from 0.2

to 0.8 M, more preferably from 0.3 to 0.7 M, still more preferably approximately 0.5 M.

A particularly preferred cell for use in the process according to the present invention is described below with reference to the second aspect of the invention.

The process preferably further includes the step of removing solid residues from the mercury-containing leachate prior to step (ii). Such residues removed from the leachate may be washed, preferably with water, one or more times to remove traces of mercury.

In this case, electrolysis in step (ii) is preferably carried out on a dilute solution of the mercury- containing leachate and the washings resulting from washing the solid residues.

Depending on the concentration of mercury in the electrochemical cell effluent after step (ii), an additional step of extracting mercury from the cell effluent may be carried out to reduce the concentration further. This may be achieved in a conventional manner, for example by means of solvent extraction. A solvent comprising a mixture of a tertiary amine and kerosene may advantageously be used.

Once the process according to the present invention has been carried out, then the resulting mercury-containing leachate and/or the solid residues and/or cell effluent may be disposed of. Prior to disposal, it is preferably to test the mercury- containing leachate and/or the solid residues and/or the cell effluent to ascertain the mercury concentration.

The process according to the present invention may be used to recover mercury from any solid waste, although the solid waste will generally comprise sludges from brine electrolysis plants, which operate the conventional process in mercury cells; and/or residues from mercury vapour lamps. Other possible sources of waste include dental waste (for example mercury-containing amalgams) and waste from the electronics industry, for example electrical relays.

In a second aspect, the present invention provides an electrochemical cell suitable for the removal of metals which are liquid at atmospheric temperature, such as mercury, from a dilute solution of the metal, the cell comprising: (a) a tubular cathode comprising a porous carbon material; (b) a current feeder for the cathode; (c) a tubular anode spaced from the cathode; (d) a current feeder for the anode; (e) an outer casing enclosing the anode and cathode; (f) the outer casing being provided with an inlet through which, in use, a dilute solution of the metal is introduced into the cell and an outlet through which, in use, a dilute solution of the metal exits the cell; (g) collecting means disposed below the cathode for collecting liquid metal deposited on the cathode during electrolysis of a dilute solution of the metal;

(h) a second outlet through which, in use, liquid metal collected in the collecting means exits the cell; the arrangement being such that, during electrolysis, the dilute solution from which the metal is to be removed is introduced into the cell by means of the inlet and flows through the porous carbon cathode and out from the cell by means of the outlet, whereby liquid metal is deposited on the cathode and is collected in the collecting means therebelow and subsequently exits the cell by means of the second outlet.

Metals which are liquid at atmospheric temperature include mercury and gallium.

It will be appreciated that the dilute solution of the metal will generally be in the form of metal ions in solution.

The electrochemical cell according to the second aspect of the present invention may advantageously be used in a process for recovering mercury from solid waste materials as herein described.

The cell will generally have radial symmetry with the tubular anode surrounding the tubular cathode, being spaced therefrom and being substantially concentric therewith.

The cathode of the cell preferably comprises a porous carbon fibre material. Using a cathode comprising carbon or graphite fibres, typically 5 to 15 microns diameter, results in an electrode having a very high surface area to volume ratio. Carbon fibres can be obtained in several forms e. g. papers, veils,

yarn, tow, chopped or milled fibres, needled, non- woven mat and as felts. These fibres can, therefore, be made up into a variety of forms e. g. flat felts or cylinders. Many of these carbon fibres have relatively high electrical conductivities which can be optimised depending on the heat treatment applied during the production process. Typically a single filament can have a resistivity of 3.1 x 10-3 to 22.6 x 10-3 ohm-cam mechanism.

The cathode is preferably provided on a porous tubular support. The porous support is preferably fabricated from a non-conducting substance such as porous polyethylene, an open mesh structure or an appropriate filter cloth supported on the open structure so that the flow regime required can be obtained. The support may also be a conducting material in which case it can also act as the cathode current feeder. The cathode advantageously comprises carbon felt which is wrapped around a porous support with at least one complete winding around the support The combination of the support and the carbon cathode in intimate contact acts to control the flow distribution of the electrolyte through the cathode.

A pressure drop is inevitably created when the electrolyte passes through the porous support and cathode. In order that the pressure drops and hence the flow is regulated, the relative pore sizes of the cathode and porous support may be adjusted in different embodiments of the invention. For example, when a cathode of open structure (i. e. large pore size) is employed, a porous support with relatively small pores is preferably used. In the case of a relatively dense cathode, however, a more open support material of larger pore is sufficient. The same principle applies whether the electrolyte is flowing

from the cathode outer side or the support side. The resultant even flow distribution achieved provides a steady flux of metal ions to all parts of the cathode and therefore sustains a maximum level of current for metal deposition.

Additional or alternative regulation of flow may be achieved by overwinding the tubular porous support with string before applying the cathode. By varying the tension to the string (described as a string filter) the pressure drop across the cathode can be adjusted as desired.

Current is supplied to the cathode by means of one or more current feeders which may be supported on a cathode tubular porous support. A suitable current feeder comprises two rods made of a low electrical resistance metal, such as stainless steel, diametrically opposed to each other and in intimate contact with the porous support (if present) and the cathode. The rods preferably extend outside upper and lower end plate assemblies which seal the ends of the cell. The rods provide the means for making the electrical connection to the cathode. Because the electrical resistance of the carbon electrode is so much greater than that of a metal such as steel, the current feeders, in order to carry current to the entire effective carbon electrode surface, preferably extend substantially along (or through) the entire length of the cathode. If a porous tubular support for the cathode is provided, then the current feeder is preferably supported on the support and, again, preferably extends substantially along the entire length of the cathode.

As a result of the metal feeders running along the entire length of the cathode, a very even

distribution of electrode potential is normally obtained. Accordingly, there is very little variation in the rate of metal deposition along the cathode length and circumference and a very high percentage of the surface area is active. The current efficiency for metal electro-deposition therefore can be maximised. The result is that there are no discrete areas operating at relatively high cathode potentials and this minimises the formation of excessive quantities of hydrogen. No discrete areas are operating at relatively low cathode potentials below the threshold required for metal deposition.

Because mercury is deposited on the cathode in liquid form it flows down the cathode under the influence of gravity to be collected in a receptacle positioned beneath the cathode. The cathode rarely, if ever, needs to be changed and this is an important advantage.

In certain circumstances it is desirable to improve the distribution of electrode potential by including more than two current feeders along the length of the cathode or by incorporating a mesh or a spiral current feeder.

Where the cathode is carbon felt, to ensure that electrical contact is maintained with the stainless steel rods, the carbon felt is pulled down onto the current feeder by means of, for example cable ties or similar means of fastener.

Alternatively, a metal strip can be employed as the current feeder. Such a strip can be either in the form of separate lengths fed from a common point or in the form of a spiral wound along the length of the carbon felt or as a mesh which produces an even

current distribution throughout the cathode. Whether the current feeder is in the form of a strip, rod, spiral or mesh it extends along the entire length of the electrode so as to uniformly distribute the current to the entire effective carbon surface.

It is preferable that in the case of large electrodeposition cells the electrical resistance of the current feeders be as. low as possible. This is to ensure that resistive losses which result in heating of the current feeders are minimised. Similarly, it is desirable for the current feeder selected to be corrosion resistant in the electrolyte in use.

The cathode preferably also extends into the metal collecting means such that, in use, the cathode contacts the liquid metal collected in said means.

This is advantageous because it prevents re- dissolution of the metal.

The base of the cell preferably comprises a plate having a recessed portion therein which serves to collect liquid metal deposited on the cathode during electrolysis.

The second outlet is advantageously provided with a valve to control the flow of liquid metal exiting the cell.

It will be well understood by those skilled in the art that it is also highly desirable that the anode is stable under the specific conditions employed in each electrolyte composition. For example, with alkaline electrolytes it is preferable to use a nickel anode or some other suitable corrosion resistant material which is stable in alkaline conditions such as stainless steel or mild steel. A noble metal

coated titanium electrode may also be used in alkaline electrolytes. With acid electrolytes it is preferable to use a suitably corrosion resistant material stable in acidic conditions, for example, a noble metal coated titanium anode or Ebonex, a titanium sub-oxide (this is a proprietary electrode material which has a high conductivity and excellent corrosion resistance) or, under certain conditions, lead dioxide on titanium can be used. The conditions which enable a lead dioxide on titanium anode to be used are those where the applied current density is very low and there are no organic materials present to complex with the lead which would remove lead dioxide from the surface of the titanium. The titanium substrate for the noble metal coated titanium electrode may be mesh or plate.

The cell is also provided with one or more current feeders for the anode which may also extend outside the end plate assemblies. Rods may be connected to the anode (S) and provide a means of supplying current to the anode assembly. Current feeders for the anode may comprise, for example, titanium rods, preferably spot welded to the anode.

Alternatively, titanium bolts which extend through the outer casing of the cell and make contact with the anode can be used. In such a case the head of the bolt may be located within the cell and in contact with the active surface of the anode.

It is also preferred that the outer casing provides a support for the tubular anode which extends for substantially the full length of the tube commensurate with the length of the cathode or cathodes. The material for the casing can be chosen from U-PVC, C-PVC, ABS, polypropylene or other suitable non-porous materials. The choice will depend to some extent on the electrolyte being used and

hence the chemical and temperature resistance required. The outer casing is preferably tubular with separate, removable top and bottom end plates.

In one embodiment of this invention the cell may include two anodes, one within a tubular support but spaced therefrom and the second surrounding the cathode as described above. Since, in operation the metal may deposit preferably in the section of the cathode nearest the anode, by operating with two anodes it is possible to load the cathode more rapidly and uniformly throughout the volume of the cathode.

This embodiment is particularly useful when low conducting electrolytes are employed. For such an embodiment it is preferable if the second anode inside the tubular support is in the form of a mesh. This acts to minimise the restriction in flow and therefore the pressure drop due to the second anode.

In another embodiment the cell of the invention may include a microporous separator, for example a polymer mesh tube, with a high open area possessing small apertures (<20 microns) located in the space between and separating the anode and the cathode. In some circumstances oxygen may be produced at the anode while hydrogen is released at the cathode. For safety reasons it is desirable to substantially prevent the mixing of the oxygen gas produced at the anode and the hydrogen gas produced at the cathode. The microporous separator serves to minimise the mixing of the hydrogen and oxygen and is hence a safety feature.

Support for the separator may be provided by means of perforated discs or a cage assembly supported off the porous support. The tubular separator is thus concentric with the cathode and spaced off from the cathode. The ends of the separator may be closed so

encouraging the hydrogen rich solution stream to exit with the depleted metal stream via the top solution outlet of the cell. The microporous tube also acts to contain any metal that is loosely adherent to the cathode. The oxygen gas evolved from the anode may exit by a channel appropriately machined in a top plate of the cell so that the oxygen enters in the channel. A bleed pipe which extends into said channel may then be provided in the top plate. This allows gas to be bled off with electrolyte via a transmission tube at a rate commensurate with its rate of formation.

In yet another embodiment of the invention a tubular ion-exchange membrane may be located between the anode and cathode so that two separate electrolyte compartments can be realised. This structure enables two different electrolyte streams to be used in the two compartments. This is desirable when, for instance, metal is to be removed from a chloride containing electrolyte. In a cell without an ion-exchange membrane chlorine will be evolved at a noble metal anode. This is obviously undesirable from the safety point of view and also, as the chlorine concentration builds up in the solution the electrodeposit may redissolve. The ion-exchange membrane ensures that chloride ions do not enter the anolyte compartment to any great extent. It enables two different electrolytes to be employed as the anolyte and the catholyte. Only oxygen is then produced at the anode in this case. Seals may be provided in top and bottom end plates of the cell to ensure that electrolyte mixing between the anolyte and catholyte compartments cannot occur.

During operation of the electrochemical cell, a source of current is connected to the cathode and the

anode and mercury is deposited as liquid droplets on the porous carbon cathode. As the reaction proceeds, the droplets coalesce and run down the carbon cathode, under the action of gravity, and are collected in the collecting means therebelow.

A flow rate of about 2 to about 200 litres/minute may be used with the apparatus of the invention.

Preferably however, the flow rate will be about 15 to about 150 litres/minute.

For all the embodiments described above, it is envisaged that at least two cells can be arranged in series or in parallel in the flow path of the solution. Alternatively, a plurality of cathodes may be arranged in series or in parallel within a single unitary anode and housing. Either way it is preferable if, when in use, the electrolyte solution passes in the first cathode or cell from inside the cell through the cathode towards the anode, and passes through the second cathode in the opposite direction away from the anode.

Additional benefits can be gained by reversing the direction of the electrolyte flow in this way.

In certain operational schemes, the concentration of the metal ions in solution will fall significantly while passing through the cell. This is particularly true when single pass operation is underway in cells containing two or more carbon cathodes in series.

Under these conditions metal is deposited to a greater extent on the first cathode where the solution concentration is relatively high and to a lesser extent on the second and subsequent cathodes. Flow reversal under these circumstances achieves a more even electrodeposited metal distribution between the cathodes by reversing the direction of the soluble

metal concentration gradient within the cell. The result is increased operational efficiency and a greater metal loading capacity.

The benefit of flow reversal can also be obtained within a single cell having just one cathode by blocking the tubular support at various points so that the electrolyte solution is forced in the opposite direction and travels up through the cathode in a zig-zag fashion.

The electrochemical cell may be operated as a method for final treatment of metal-bearing effluent prior to discharge of a clean waste stream with metal concentrations below the local effluent content limits.

In a third aspect, the present invention provides a method for removing metals which are liquid at atmospheric temperature, such as mercury, from a dilute solution of the metal, which method comprises: (A) passing a dilute solution of the metal into a cell as herein described; (B) passing a direct current between the anode and cathode to deposit liquid metal on the surface of the carbon cathode whereby said metal is removed from said solution and is then collected in the collecting means ; and (C) withdrawing liquid metal collected in the collecting means by means of the second outlet.

The dilute solution of the metal is preferably acidic since this has been found to prevent deposition of the metal as the oxide. If the solution is not

intrinsically acidic, then it may be acidified in a conventional manner as herein described. The dilute solution of the metal preferably has a pH: 5, more preferably in the range of from 1 to 3, still more preferably from 1 to 2.

It will be appreciated that the dilute solution of the metal will generally be in the form of metal ions in solution.

A proton-conducting membrane as herein described is advantageously disposed between the cathode and the anode to thereby separate the cell into a cathode compartment and an anode compartment. In this manner, the solution of the metal may be circulated through the cathode compartment and an anolyte may be circulated through the anode compartment. The anolyte may comprise, for example, a solution of nitric acid, sulphuric acid, sodium nitrate or sodium sulphate. A suitable concentration for the anolyte is preferably in the range of from 0.2 to 0.8 M, more preferably from 0.3 to 0.7 M, still more preferably approximately 0.5 M.

In carrying out the method of the invention flow rates of the dilute solution of from about 2 to about 200 litres/minute may be used although typically the flow rate used will be between about 15 and about 150 litres/minute. The current density which may be employed is preferably between about 100 and about 300 A/m2. However, in certain applications, for example when operating with electrolytes of very high acidity or alkalinity a current density in excess of 300 A/m2 may be needed to deposit metal on the cathode and prevent its redissolution in the electrolyte. In such applications the current density employed may be between about 300 A/m2 and 8C0 A/m2.

The invention will now be described, by way of example, with reference to the accompanying drawings in which: Figure 1 is a schematic illustration of an electrochemical cell according to a preferred embodiment of the present invention; and Figure 2 is a plot of mercury concentration (ppm) vs charge passed for Example 5.

The electrodeposition cell shown in Figure 1 is a tubular design, with radial symmetry for most of the components. The design in most respects is the same or similar to the cell described in the applicant's earlier patent application PCT/GB94/01929 (WO 95/07375). The outer cell body (1) may be constructed from polyvinyl chloride, or other suitable material.

A tubular anode (14) surrounds a tubular proton- conducting membrane (8), which, in turn, surrounds a tubular cathode (3). The anode, cathode and separator are spaced apart and are generally concentric with each other. The solution to be treated enters the cell through inlet (2), and passes outwards through the porous carbon cathode (3). The cathode (3) is supplied with a metallic current feeder (4). The solution exits the cell via outlet (5). As shown, anode and cathode compartments of the cell, (6) and (7) respect vely, are separated by the proton- conducting membrane (8). Anolyte solution then passes through the anode compartment (6), entering through the port (9,, and exiting through the port (10). This arrangement becomes necessary when species present in the catholy e would give rise to deleterious reactions at the anode (14). In other circumstances the membrane (6, could be eliminated, and the cell run with one compartment only. The cell as shown has a

base plate (11) having an annular recessed section or groove (12), which serves to collect liquid mercury which has been deposited on the cathode (3). An outlet with a tap (13) can be used to run off the mercury.

During operation of the electrochemical cell, a source of current is connected to the cathode (3) and the anode (14), and the mercury is deposited as liquid droplets on the porous carbon cathode (3). As the reaction proceeds, the droplets coalesce, and run down the carbon fibres, under the action of gravity, and into the grooved part (12) of the cell base (11).

As will be understood by those skilled in the art, the cell in use may be integrated into plant by means of conventional pumps and tanks. Metal removal can be achieved by either recirculating of the electrolyte through the cell at a variety of flow rates and current densities; or in a single pass through the cell from holding tanks to discharge pipe.

Individual cell modules are suited to flexible operation in combination, i. e. employing fluid flow in series and/or in parallel.

The bulk of the current is carried by a supporting electrolyte such as an acid, alkali or a neutral salt. When very low conductivity electrolytes are used the supporting electrolyte is added to the metal bearing waste and reduces the cell voltage to a suitable level.

In some instances it may be necessary to raise the temperature of the electrolyte or to add buffering agents to enhance deposition of the metal. Control of the pH in the feed tank of a recirculating electrolyte

is important to avoid precipitation of the metal.

The solutions which would be most suitable for treatment using the cell as described above fall broadly into two categories: those where there is a need to comply with local consent limits for discharge of metal bearing effluents (these are often toxic and environmentally damaging metals); and those where there is an intrinsic metal value which would cause a financial loss if the metal was not recovered. Often these solutions have already undergone some conventional chemical treatment such as precipitation or ion exchange but there remains a metal content which, for the reasons listed above, needs to be treated.

The present invention will now be described further by way of example.

Leaching Waste materials containing mercury in the metallic form may be leached with, for example, hydrochloric acid, or with sodium hypochlorite solution. The former may be preferred, but the latter should be used in cases where the bulk waste would be attacked by acid.

Example 1 The leaching conditions for two types of waste material are summarised below in Table 1. The leaching was performed in a polypropylene reactor, with an immersed heater, a temperature controller, a pH controller, a stirrer, and a condenser. The metal parts of the lamps were removed, and the crushed glass and powder were leached.

Table 1

Brine sludges Fluorescent lamps Leaching agent Sodium Hydrochloric hypochlorite acid Concentration 30-50 g Cl2/l 5% Liquid/Solid 3/1 7/1 ratio Liquid and solid 13.6 Kg of dry 30 lamps (6.6Kg) quantities used sludge 40 1 of 50 1 of leaching leaching solution solution Temperature 50°C 50°C Reaction time 4h 4h Agitation Mechanical Mechanical Flow of air The efficiency of mercury abstraction from the wastes was high, as shown below in Table 2, even though the initial levels of mercury are low.

Washing and Filtering The lamp residues are readily filtered from the leachate, both glass and powder are easily separated.

The residues from the above leach were washed with two 25 1 batches of water.

The sludge residues are quite fine, and are preferably separated from the leach solution with a filter press. They must then be washed thoroughly to remove traces of mercury. This may be done in two stages, using 20 1 of water in each.

Table 2 Waste Hg in initial Dissolved Hg in the Remaining Hg in the Ext@ sludge leachate after the solid after the Eff@ leaching process leaching process Sludge 1 24.6 mg/Kg 5.5 mg/l 4.5 mg/kg Sludge 2 18.8 mg/Kg 5.4 mg/l 1.5 mg/Kg Fluorescent 28 mg/Kg 3.8 mg/l 0.8 mg/Kg lamps

Electrolysis An electrochemical cell according to the present invention as described herein has been operated as indicated in the examples which follow to recover mercury from solutions. It was found that the best results were obtained if the pH of the solution to be treated was in the acidic region. This prevents the mercury depositing as the oxide, which is a solid and therefore cannot run off the electrode. The cell is capable of reducing the mercury levels in solution to 0.1 parts per million (ppm) or less. Electrolysis is preferably carried out using a porous carbon cathode cell as herein described. The cell is preferably divided by a proton conducting membrane to avoid the production of chlorine at the anode. The solution of the metal is circulated through the cathode compartment and is preferably acidified to prevent the deposition of mercury as the oxide. An anolyte is circulated through the anode compartment and preferably comprises a solution of nitric acid, sulphuric acid or sodium nitrate. The metal forms as droplets on the cathode, from whence it may be collected. Examples of the electrolysis of typical solutions are provided below.

Example 2 The conditions of electrolysis were: Area of felt electrode (nominal): 240 cm2 Cell current: 5 A Catholyte: Leachate from brine sludge, pH 1 Catholyte volume: 40 1 Anolyte: 0.5 M H2SO4 (3 1) Flow rates: Catholyte 75 1/h Anolyte 50 1/hr

The mercury concentrations in the catholyte, as a function of time are given in Table 3 below: Table 3 Time (hours) Hg (ppm) 0 5.50 1 3.30 2 2. 3 1.50 4 0.98 5 0.75 6 0.66 7 0.54 8 0.40 9 0.32 10 0.25 12 0.20 14 0.18 16 0.14 200.10 Example 3 In this experiment mercury was removed from a solution with a high initial concentration. The cell used is illustrated schematically in Figure 1 and comprises a membrane separating the cathode and anode. The anolyte and catholyte solutions were pumped through the cell compartments is closed loops including flowmeters and holding tanks.

Area of felt electrode (nominal): 150cm2 Cell current: 2.25 A-4.5 A (after 15 min) Catholyte: 0.1M NaCl, adjusted to pH2 (18800 ppm Hg) Catholyte volume: 3 1 Anolyte: 0.5M NaNO3 (3 1) Flow rates: Catholyte 150 1/hr Anolyte 100 1/hr Samples were withdrawn from the catholyte and analysed for mercury and the results are given in Table 4 below.

Table 4 Elapsed Cell Cell Catholyte Hg (ppm) time voltage current pH (min) (V) (A) 0 2. 5 2. 25 1.83 18800 30 3. 25 4. 5 1.85 17220 60 3. 06 4. 5 1.61 14620 90 2. 96 4. 5 1.6 11690 120 2. 88 4. 5 1.12 8600 150 2. 88 4. 5 2.13 6400 180 2. 83 4. 5 1.61 3620 240 3. 9 4. 5 2.01 0.6 During the first 3 hours of the experiment, the current efficiency was 98%, and over the whole experiment it was 89%.

Example 4 In this example the operating conditions were the same as those in Example 3, except that the initial concentration of mercury in solution was approximately 1,000 ppm. Samples were withdrawn from the catholyte and analysed for mercury and the results are given in Table 5 below.

Table 5 Elapsed Cell Cell pH Mercury time current voltage [ppm] (mins) (A) (V) 0 2. 25 4. 14 1. 97 940 3 2. 25 3. 54 2. 08 808 6 2. 25 3. 25 1. 96 690 9 2. 25 3. 26 185 552 12 2. 25 3. 34 1. 96 414 15 2. 25 3. 36 2. 14 282 18 2. 25 3. 49 1. 96 177 21 2. 25 3. 71 1. 92 82 24 2. 25 3. 95 1. 89 40 The current efficiency in the mercury concentration range down to 200 ppm was 90%, and overall it was 80%.

Example 5 Using the same conditions as in Example 3, a cell current of 2.25 A, and an initial mercury level of approximately 1,000 ppm, the effect of pH was investigated. The results are shown in Figure 2 and indicate that electrodeposition is best carried out

within the pH range of from 1 to 3. At a higher pH the current efficiency declines, especially at low mercury concentrations.

Example 6 A solution obtained by leaching a mercury waste material was electrolysed. The initial mercury concentration was about 5. ppm.

Area of felt electrode (nominal): 150 cm2.

Cell current: 3.8-3.3 A Catholyte: Leachate containing chloride, pH 1 Catholyte volume: 3 1 Anolyte: 0.5M H2SO4 (3 1) Samples were withdrawn from the catholyte and analysed for mercury and the results are given in Table 6 below.

Table 6 Time Volts Current pH Hg (ppm) (min) (A) 0 4. 0 3. 8 0. 98 5.40 10 4. 1 3. 3 0. 91 1.40 20 4. 1 3. 4 0. 88 0.45 30 4. 1 3. 3 0. 85 0.04 45 4. 1 3. 3 0. 80 0.04 60 4. 1 3. 3 0. 77 0.04 Hence, the cell can be used to obtain low final mercury concentrations.

Example 7 This experiment was carried out to determine how much mercury needed to be deposited on the carbon felt cathode before the drops ran down into the collection recess. A concentrated solution of mercury chloride was continually dosed into the catholyte. Under this regime, the mercury level in the catholyte tends to attain a steady-state value. The experimental conditions were: Area of felt electrode (nominal): 150 cm2 Cell current: 2.25 A Catholyte: 0.1M NaCl, adjusted to pH2 Catholyte volume: 2.5 1 Anolyte: 0.5M H2SO4 (3 1) Flow rates: Catholyte 150 1/hr Anolyte 100 1/hr Dosing solution: 40 g/1 mercury chloride in 0.1N NaCl The results are given in Table 7 below.

Table 7

Time (hr) Volts pH Hg (ppm) Comments 0 3. 88 1. 89- 1 3. 42 1. 70 52 2 2. 91 1. 95 36 3 3. 24 1. 95.37 5.3.00 2. 10 35 6 3. 05 1. 94 32 6 hr 55 min 3. 32 2. 00 83 Hg dosing stopped 7 hr 10 min 3. 49 1. 90 0.61 Hg dosing continued 8 3. 29 1. 07 66 9 3. 35 1. 81 80 10 3. 39 1. 92 62 11 3. 47 2. 07 47 12 3. 29 1. 98 50 13 3. 32 1. 75 76 13 hr 50 3. 15 1. 85 70 Hg dosing stopped min 14 hr 5 min 3. 55 1. 93 1.4 A small amount of mercury appeared in the bottom of the cell after 6 hours 55 minutes, drops could then be drained after this time. When the dosing was stopped the mercury concentration dropped rapidly to low levels. Over the whole time period of the experiment the amount of mercury deposited was 80 g, calculated from the dosing rate. Droplets run off from the cathode before this amount is deposited.

Extraction of Mercury from the Electrolysis Effluent The effluent from the electrolysis cell, as described above in relation to Example 2, contained about 0.1 ppm mercury. This level needs to be reduced to below the limits set by legislation before the effluent can be discharged. This may be achieved by solvent extraction using a mixture of a tertiary amine in kerosene solvent (1: 3 v/v). The extraction takes place in two stages in a conventional reactor. Following this procedure the mercury concentration in the aqueous phase was determined to be below 0.005 ppm. After neutralisation, the effluent, which contains mainly sodium chloride, may be discharged. The organic phase may be regenerated with sodium chloride solution, and the mercury containing aqueous phase can be recirculated to the electrolysis cell.

Testing of Solid Residues The residues left after leaching and washing the starting wastes have been subjected to standard tests to ascertain whether they are suitable for discarding.

Several tests have been used, including: Toxicity Characteristic Leaching Procedure (TCLP): EPA Method 1311; Availability Test: Dutch norm NEN 7341 ; Leaching Test: Nordest Method NT ENVIR 003; and Leaching test: German norm DIN 38414-S4.

All of these tests are based on leaching of the solids under defined conditions of solid/liquid ratio, temperature, pH, time, leaching reagent, and agitation.

In all cases the mercury concentrations in the resulting leachates were below 0.005 ppm. These levels are less than those required by legislation of the various countries and organisations involved in

establishing the tests.

The electrochemical cell and process according to the present invention for the treatment of mercury containing wastes reduces their mercury contents, and enables these materials to be disposed of safely.

Furthermore, the mercury is recovered in usable form.