Precipitate Gold Mercury with Electricity

Precipitate Gold Mercury with Electricity

The treatment of slimes has of late years been, the chief difficulty in connection with cyanide work, in which much time and money has been spent in endeavouring to find a practical solution of this problem.

The processes at present in use are:

  1. Filter pressing, as practised in Western Australia.
  2. Decantation, as practised in South Africa.

Filter pressing after agitation affords a ready and efficient means of separating the gold-bearing solution from the slimes and admits a perfect washing, thus ensuring a high extraction ; but the very high working costs forbids its application except on high grade material, the chief items being the necessary large amount of labour, compressed air, and general upkeep, none of which can be materially-reduced.

The decantation method has the advantage of low working cost; but on the other hand the low extraction obtained, due to imperfect washing, the length of time occupied in the treatment, and the enormous bulk of solutions to be handled, are serious disadvantages.

In either process it is absolutely necessary to obtain a perfectly clean solution before starting the gold precipitation by zinc, which is a source of much trouble. Also the high first cost of installing either plant is such that they can only be adopted on mines with very large working capital.

The advantages to be gained by electrical precipitation on mercury are:

  1. The extraction and subsequent precipitation in the solution without the necessity of filtration.
  2. A clear solution is not necessary for precipitation.
  3. Precipitation takes place equally well from strong or very dilute solutions of cyanide.
  4. The presence of copper salts in solution has no injurious effects.
  5. The gold is recovered as amalgam which only requires retorting and melting.
  6. In the treatment of slimes previous amalgamation is not necessary, and any gold too coarse for solution is instantly amalgamated by contact.

The essential features about a mercury cathode are:

  1. That the surface of the mercury shall be as large as possible.
  2. That the surface shall be kept in a perfect condition by some simple means.
  3. That a minimum quantity of mercury be used.
  4. That the right current density be used, i.e., the amperes per surface area of cathode.

Palatan-Clerici process

The Palatan-Clerici process seems to have been the first to have used a mercury cathode with any success.

The apparatus consists of a circular vat of wood having amalgamated copper plate on the bottom forming the cathode. Sufficient mercury is added to keep the plates coated with mercury.

The agitator comprises the anode and consists of six or eight radial arms supported by a central shaft, the arms of which are a few inches above the cathode and consist of flat iron mounted on wood, having wooden pegs to keep the slime from settling on the cathode. The anode is connected by a suitable arrangement to the positive pole of the dynamo and the cathode to the negative pole.

A discharge valve for the pulp is placed a few inches above the mercury, also a valve in the bottom to draw off the mercury when cleaning up. This process was in use at the De Lamar Mill in Idaho.

Riecken Process

The Riecken Process involves the use of an iron vat with vertical ends, inclined sides and rounded bottom (Plate I.) Its dimensions are 13 feet long, 8 feet wide and 11 feet deep; holding 17 tons of ore. A horizontal shaft, to which is attached paddles, is passed through the ends of the vat by means, of stuffing boxes, and is driven at about twelve revolutions per minute. The sides and bottom of the vat, which form the cathode, are lined with amalgamated copper plate, and a constant stream of mercury is kept flowing over the plates to keep them in good condition; the mercury being drawn off the bottom of the vat and elevated to the top by means of a jet of compressed air. The anodes are iron bars three inches by one inch, hung from an iron girder parallel to the side and the bottom of-the vat, and about 18 inches distant. A discharge valve for the pulp is placed four inches from the bottom, so that the mercury remains in the vat when discharging the pulp. The mercury pipe is brought out of the lower end of the vat, which has a fall of two inches in the bottom, and is connected to the elevator which delivers into the basin feeding the mercury distributing pipes.

From 400 lb. to 500 lb. of mercury are required for each vat.

Mumford’s Electro Cylinder

Mumford’s Electro Cylinder—The cylinder (Plate II.) is made of steel plates, is copper lined, supported on rollers and driven by a spur wheel at five revolutions per minute. The ends are of wood from which the anodes, being two inch round iron, are supported, as also are wooden bars to assist the agitation.


Wooden ends are used to ensure a good insulation from the cylinder which forms the cathode. A 6 inch valve is placed in the cylinder end 3 inches from the bottom, having an elbow pointing downwards which is used for discharging the pulp, the mercury remaining in the cylinder. For filling, the position of the cylinder is reversed so that the valve is at the top and the pulp can be led in from an overhead pipe. There is also a spring valve in the end of the cylinder, near the outer edge, which at its highest point hits a tappet and allows any gas formed to escape.

A cock, which can be locked, is placed in the periphery of the cylinder to draw off the mercury when cleaning up. Mercury is added until it forms a bath at the bottom, about half an inch in depth. When working, the cylinder is filled within a few inches of the top and is revolved at five revolutions per minute about its horizontal axis. A current of about half an ampere per square foot of cathode surface is passed through the pulp. This causes the soluble gold to be precipitated on the amalgamated copper lining, which is kept in the highest possible state of efficiency by every portion of it passing under the bath of mercury at every revolution of the cylinder ; any gold too coarse for solution being immediately amalgamated by contact.

In cleaning up it will be found that the mercury contains most of the amalgam, so that it can be drawn off through the mercury cock, and squeezed in the ordinary way to obtain the amalgam. If any amalgam remains on the copper lining, the squeezed mercury can be returned and agitated with sand. Steam can also be added to soften the amalgam if necessary. This will cause all the adhering amalgam to dissolve in the mercury, which can then be drawn off and squeezed.

A cylinder 20 feet long and 5 feet in diameter will hold 16 tons of pulp of equal weights of solution and ore, i.e., 8 tons of dry ore and 8 tons solution, and requires about 300 lbs, of mercury when running, and 150 amperes.

With a slime assaying about 6 dwt. of gold, a complete precipitation would take about 8 hours.

With a slime assaying from 1 to 2 oz. it would take from 12 to 16 hours. This cylinder would therefore have a capacity of about 24 tons per day of 6 dwt. ore, or a capacity of 12 to 16 tons of 1 to 2 oz. ore. The process can be used on any slimes whatever, and has the advantage over all other slime processes of being able to catch any coarse gold which remains undissolved. When using this process, the tailings resulting from battery treatment are passed over classifiers in order to separate the sand from the slimes. After thickening the slimes to a suitable consistency with spitzkasten, the resulting pulp is run into circular agitation vats, just enough cyanide being used to get the gold in solution; in cases of oxidized ore .015% KCN is generally sufficient, which represents about 1/3 lb. per ton ; in many cases less cyanide can be used.

The pulp can then be run into the electro cylinder in charges. After the necessary time it is discharged into a settler to settle any mercury it may contain, and thence to the dam or race.

The sands can be re-ground or treated by percolation.

A 50 ton plant per day for treating low grade slimes would therefore require three 18 ton circular agitation vats; two electro cylinders, 20 feet by 5 feet diameter; one dynamo, 400 amperes at 3 volts; one mercury settler, and some means of elevating the slimes if the country is flat. This will show a very large saving in first cost compared with either a decantation or filter pressing plant.

With the Kalgoorlie ores the process can be used on the roasting plants, or on wet crushing bromo cyanide plants, to replace filter pressing precipitation by zinc and previous amalgamation. The arrangement in a roasting plant on sulpho-telluride ore is as follows:—After crushing with Krupp ball mills or Griffin mills, and roasting in a suitable furnace, the roasted ore would be mixed with water in order to separate


the sands from the slimes, the sands being ground in a flint mill or other suitable slimming machine till the whole is reduced to a slime. The slime is then thickened to a consistency of equal weights of water and ore by means of spitzkasten, and run into a circular agitation vat, in which cyanide is added till the solution titrates .03% to .05% KCy, which is in most cases sufficient to get the gold into solution. After six hours agitation the bulk of the gold will be in solution. The pulp can then be run into the electro cylinder, when the solution of gold will continue ; but it is advisable to dissolve the bulk of the gold before running into the electro cylinder. After the necessary time in the electro cylinder, the pulp can be run through a settler into the dam and the solution recovered.

Current Density

Current Density, i.e., the amperage per surface area to be used, is a most important point, and one on which opinions seem to vary. If too high a current density is used the gold does not amalgamate, being deposited as a black powder, which is easily rubbed off-the cathode, and re-dissolves in the presence of cyanide, though not rapidly. My experiments were performed on Kalgoorlie sulpho-telluride ore, previously roasted and slimed, the gold being already in a state of solution, cyanide of .03% to .06% being used. The water, contained 10 to 20% of solids in solution, chiefly sodium chloride, being the ordinary mine water. I used an experimental cylinder 14 inches in diameter and 26 inches long, made of copper plate, the inside of which was amalgamated. The ends were of wood with a two inch shaft through the centre, The whole was mounted on bearings and driven at five revolutions per minute. The shaft also acted as the anode, and was connected to the positive pole of a 3 volt dynamo; 30 lbs. of mercury being added to the cylinder. With this apparatus I found that good results were obtained up to .8 amperes per square foot of cathode, all the gold deposited being amalgamated; but over that current density the results were uncertain, a portion only of the gold being amalgamated and the rest washed off into the pulp. At over one ampere per square foot, gold was always thrown down, which did not amalgamate. At two amperes per square foot, none of the gold deposited was amalgamated, but all was washed off into the pulp which, by adding stronger cyanide, about one per cent., and agitating for eight hours, was redissolved.

The state of efficiency of the mercury surface has a good deal to do with the current density. The better the surface of mercury, the higher current density can be used, without forming gold that will not amalgamate. Thus, when running with .6 amperes per square foot, all the gold deposited was amalgamated; but, on ceasing to revolve the barrel for one hour, gold was deposited which would not amalgamate. It is therefore absolutely necessary with this form of apparatus to keep the cylinder revolving all the time. If it has to be stopped for more than five minutes, the current must be stopped also.

I find that the rate of precipitation does not vary in proportion to the amperage per square foot, in dealing with solutions low in gold. In practice on a large scale little difference could be noticed between the results obtained when using .5 amperes per square foot and .8 amperes, the time being practically the same. Therefore I consider .5 amperes the most suitable current density to use, as there is a fair margin before throwing gold down which will not amalgamate.

T. K. ROSE says “With .01 amperes per square foot or less, gold is precipitated and amalgamated simultaneously, the plates kept in good order during 24 hours use ; but with currents of greater density part of the gold is deposited as a non-adherent black powder, easily rubbed off.” I conclude that the reason for these low results was the difficulty of keeping the mercury surface in an efficient condition.

The Riecken process recommends .8 amps, per square foot. In the Keith process .06 amps, is mentioned as suitable.

The inventors of the Palatan Clerici state “It is known that when electrolysing a metal, and especially gold, from a solution upon a solid cathode, the precipitation of the metal can be effected with an electric current of a very small fraction of amperes to the square foot of acting surface of cathode; but we have found that when a mercury cathode is employed, a current of one ampere and a half, or at least one ampere per square foot of the mean surface between the anode and cathode (that is to say half the sum of the surfaces of the anode and cathode) should be employed to ensure thorough and rapid precipitation of the gold.”

The rate of precipitation is inversely proportional to the area of cathode surface, other conditions remaining constant. That is, if the number of square feet of cathode surface per cubic foot of pulp under treatment is doubled, the time required for precipitation is halved. The following experiments were performed in the experimental cylinder described above, having eight square feet of cathode surface, of which seven were under the influence of the current.


From the above experiments it will be seen that the gold in a rich solution is precipitated much quicker than that in a poor one. Also from general experience it appears to take longer to deposit the last few dwts. from a rich solution under treatment, than to deposit an equivalent amount from a fresh solution. In many cases this is caused by the gold, continuing to dissolve from the pulp, and thereby prolonging the finishing point.

In a cylinder 14 inches in diameter there are three square feet of cathode surface for every cubic foot of capacity, and it takes four hours to obtain a precipitation from 12 dwt. solutions.

In a cylinder four feet in diameter there is one square foot of cathode per cubic foot of pulp, and consequently it will take 12 hours to obtain a precipitation from the same solution.

In a cylinder five feet in diameter there is .8 square feet of cathode per cubic foot of capacity, and it therefore takes 15 hours to do the same work.

In an electro vat, as in use at the South Kalgurli Gold Mine, working on the same basis, there is .4 square feet of cathode per cubic foot of capacity, which would require thirty hours to obtain a good precipitation on the same solution, and in practice it is found that 24 to 30 hours is required to insure a good precipitation, although it is sometimes obtained in slightly less time.

From the above it will be seen that where power is expensive a four foot cylinder is the most suitable size, such a cylinder 20 feet long would hold five tons of ore at a charge, equal weights of ore and solution being used. The voltage used does not appear to have any effect on the precipitation of the gold. It varies as the resistance of the cylinder, which depends, firstly, on the distance of the anode from the cathode; secondly, on the density of the electrolyte.

Working with water containing from four to 20 per cent, solids in solution, three to five volt dynamos are found to give the best results with the electrodes about 18 inches distant.

Where salt water is not procurable, fresh water can be used with a small amount of salt added to lessen the resistance, and the distance between the electrodes can be reduced, thereby reducing the voltage and the necessary horsepower.

In conclusion I think it will be seen that the electro cylinder shows decided advantages over the other methods, especially in the large cathode surface and therefore the quick precipitation obtained, causing a large saving in power and time.

The simple means of keeping the plates in an efficient condition, which allows little chance of failure in this most essential point, and the absence of fast machinery and numerous working parts, together with the perfect safety of the amalgam during the operation, which can only be withdrawn by one cock, which is kept locked, are also advantages.