Manufacturing Cyanide
Sodium cyanide is a white, deliquescent, crystalline material easily soluble in water. The basic sources are alkalis or alkaline earths, atmospheric nitrogen, and carbon.
In the United States it is derived (1) from sodamide which is produced from sodium and ammonia. The sodamide is heated with charcoal, and the resultant soda cyanamid is then heated with an excess of charcoal, resulting in the formation of sodium cyanide. This is the Castner-Roessler process, which yields 96 to 98 per cent material.
In Europe, 90 to 92 per cent cyanide is manufactured (2) from destructive distillation of beet-sugar refuse, forming hydrogen cyanide. This is absorbed in caustic soda solution, from which the cyanide is obtained by evaporation.
Aero-brand cyanide is manufactured by fusing calcium cyanamid and salt in a continuous electric furnace, from which the product is tapped at regular intervals into a sump outside the furnace. The conversion is an equilibrium reaction, and practically all of the nitrogen is in cyanide form. By proper cooling of the melt, the equilibrium is “frozen” at the high- temperature equilibrium point. Samples of this cyanide, kept in sealed containers for several years, showed no change in cyanide content.
Types of Cyanide and Cyanide Consumption
Of the two types of cyanide used in the cyanidation of precious-metal ores—sodium cyanide and Aero-brand cyanide—sodium cyanide is generally sold in the form of 5-lb. blocks or calves and is packed in drums holding 200 lb. net.
Aero-brand cyanide is calcium cyanide containing 48 to 50 per cent of pure NaCN equivalent, the other half consisting chiefly of common salt and lime. It is manufactured in the form of black flakes and packed for shipment in zinc-coated iron drums. Its dark color is due to a small amount of graphitic carbon derived from the principal raw material cyanamid.
This brand of cyanide dissolves readily in water, leaving only a slight undissolved residue consisting chiefly of lime and graphitic carbon. This insoluble residue has no effect whatever on the dissolution of gold and silver and will not precipitate precious metals already in solution. It is, therefore, not necessary to remove it.
The recommended procedure for introducing cyanide into a cyanide-mill circuit is to provide a tank of sufficient size so that enough cyanide can be dissolved at one time to furnish the mill requirements for at least one shift. In the case of Aero-brand cyanide, a 10 per cent solution is recommended, which will then contain 5 per cent equivalent pure NaCN. It is advisable to aerate the solution with finely divided air (atomized) to eliminate the small amount of soluble sulphides that it contains. The addition of ¼ lb. lead acetate or pulverized litharge will speed up the desulphurizing operation. The tank for dissolving the cyanide should be provided with an agitating mechanism.
The strength of solution used in the cyanidation of precious-metal ores will vary with the type of ore and the precious-metal content. Cyanide plants treating gold ores, with little or no silver, rarely use a solution containing over 0.05 per cent NaCN equivalent. In the cyanidation of silver ores a stronger solution is necessary, and it is common practice to maintain a strength of 0.20 to 0.30 per cent NaCN. Mechanical and chemical losses of cyanide increase in direct proportion to the strength of solution; consequently, the cyanide content of solutions should be kept at the lowest strength consistent with maximum extraction.
The chemical consumption of cyanide, for a given ore, depends on the cyanide-consuming constituents present in the ore, the period of treatment and the strength of solutions. The mechanical loss of cyanide depends on the type of treatment employed. For gold ores the total cyanide consumption will average around 1½ lb. NaCN per ton ore. In the case of silver ores the consumption is generally much greater, probably averaging over 2 lb. High-grade silver ores and concentrates may require as much as 10 lb. NaCN equivalent per ton.
(The foregoing notes were prepared by S. J. Swainson of the American Cyanamid Company, New York, which manufactures Aero cyanide. Reference may be made here to four papers on cyanides and cyanidation presented to the Electrochemical Society in September, 1931: “Present Status and Uses of Cyanamid Process Cyanide” by G. H. Buchanan; “Cyanides in the Metallurgy of Gold and Silver,” by E. M. Hamilton; “Cyanides in Metallurgy,” by M. R. Thompson and “Physical and Mechanical Aspects of the Cyanide Process,” by A. W. Allen.)
Putting Gold into Solution with Cyanide
“The Physics of Gold-Solution,” by H. A. White, in Jour. C.M. and M.S.S.A., July, 1934, is a highly technical dissertation, wholly confined to the surface reaction between metal and cyanide solution. White’s experiments were based on the hypothesis that the rate of solution of gold in cyanide solutions is mainly dependent upon the presence of oxygen. The chemical reaction between gold and cyanide may be expressed as
4Au + 8KCN + O2 + 2H2O = 4KAu(CN)2 + 4KOH
According to this equation, 1 milligram gold requires 0.0406 milligram oxygen, and this corresponds with 5.80 cc solution at 7 milligrams per liter. At an oxygen concentration of 8 milligrams per liter the corresponding KCN strength is 0.01302 per cent or 0.00980 per cent NaCN. It is found, however, that gold dissolves at the maximum rate if the solution contains
0. 027 per cent KCN, equal to 0.020 per cent NaCN, and if it is saturated with oxygen. Tins difference is due to the slower diffusion rate of the cyanide and to incomplete ionization and hydrolysis for which an allowance of 10 per cent must be made.
When the cyanide concentration is lower than the optimum, its diffusion rate will be the determining factor in a saturated oxygen solution, but a greater concentration can only hinder rate of oxygen diffusion as well as reduce its solubility. On the other hand, the presence of excess oxygen, either by means of increased pressure or in the presence of oxidizing agents which attack the cyanide slowly enough, will raise the optimum cyanide concentration and the maximum rate of gold solution. If less than saturated with oxygen, as is frequently the case in working solutions, the cyanide strength could be correspondingly reduced, and, in any case, the rate of attack on the gold diminished.
It is known that the rate of gold solution is increased by contact with zinc, iron, or carbon, and this may be attributed in the last resort to the extension of surface to which the oxygen may diffuse and likewise involves an increase in the optimum cyanide concentration.
Four methods of treatment were employed in the experiments—slime treatment by agitation, sand treatment by percolation, gold plates hung in still solution, gold plates hung in moving solution. The optimum cyanide strength was used, and the oxygen concentration was 7 milligrams per liter. These tests, with the physics involved, are given in detail.
FACTORS IN DISSOLUTION OF GOLD AND SILVER
A careful study of the factors that influence the rate of dissolution of gold and silver in dilute cyanide solutions was undertaken by George Barsky, S. J. Swainson, and Norman Hedley and published in Trans. 112, A.I.M.E1934.
Cyanide Concentration
The first of the series of experiments had to do with the effect of cyanide concentration on the rate of dissolution of gold and silver. In plant practice, the solution strength for gold approximates 0.05 per cent NaCN, or 1 lb. cyanide per ton solution. Stronger solutions do not seem to hasten the dissolution or improve the extraction, and as the chemical and mechanical loss of cyanide is much higher with strong solutions, obviously it is desirable to hold the solution at the minimum strength consistent with good extraction. The experiments covered the use of pure gold foil, solutions containing up to 0.50 per cent NaCN and a pH of 9 + but without alkali added. The maximum rate of dissolution of pure gold was reached at 0.05 per cent NaCN, corresponding to concentrations used in modern plants. The solubility of oxygen is practically unaffected by the concentration of cyanide.
A similar set of experiments was made on pure silver foil, in 0.01 to 0.50 per cent NaCN. The finding was a maximum rate of dissolution in 0.10 per cent NaCN.
A third lot of experiments was run on gold-silver alloys containing 79.8 per cent silver and 57.5 per cent gold. Sodium cyanide solutions of 0.10 per cent were used. Later, these were assayed and were found to contain gold and silver proportional to the composition of the alloys.
Alkalinity Variations
A study was made of the effect of varying alkalinity on the rate of dissolution of gold in cyanide solutions. All tests were in cyanide solution of 0.10 per cent strength, and varying amounts of lime water or sodium hydroxide were added. The rate of dissolution was greatly reduced at high pH values or high concentration of OH ions. As the curves plotted from the results were so different, further experiments were run to reveal this unexpected action of lime. It was found that lime had no appreciable influence on the solubility of oxygen in the cyanide solutions used, so calcium sulphate and calcium chloride were added. The former had a slight retarding effect on dissolution of the gold, and the other calcium compound had a more pronounced effect, but as it was determined that the reduction in rate of dissolution of gold caused by the addition of lime is due neither to lower solubility of oxygen nor to the presence of calcium ions, apparently both calcium and hydroxyl ions must be present to produce the full effect, as yet unexplained.
DISSOLUTION OF GOLD AS A CORROSION PROCESS
In a recent paper of considerable interest, “The Dissolution of Gold in Cyanide Solution” Trans. Electrochem. Soc., April, 1947, the author, P. F. Thompson, states:
Corrosion research shows that the dissolution of a metal is an anodic process associated with the necessary cathodic action and is therefore called electrochemical; it differs from electrorefining and plating only in that it depends on the electric energy generated within the corrosion cell or local couple itself, and not obtained from without as in these technical operations.
Referring further to the surface agents necessary to maintain the required electromotive potential, it is remarked that:
These cathodic agents must necessarily be oxidizing substances but limited in this respect in the case of cyanide, since the cyanide ion may itself be oxidized by most oxidizers to free cyanogen or further to cyanate. Fortunately nature has provided such a reagent in the form of dissolved oxygen, which, though of high oxidizing potential, is restricted in its action by the fact of its slight solubility in water under the partial pressure of 0.2 atmosphere. This limitation harmonized its action with the need to use weak solution of cyanide and also with the relatively minute amount of gold to be extracted.
The author also describes a number of interesting experiments carried out to determine the form of dissolution of gold; the effect of aeration on the time-potential curve; the effect of certain films and of lead, silver, and copper ions on the dissolution of gold leaf. The electrochemical interpretation of the results obtained lead to conclusions that agree well with laboratory and mill experience.
For detailed discussion of the physical chemistry of the cyanide process, the reader is referred to the following articles: (1) Barsky, Swainson, and Hedley “Dissolution of Gold and Silver in Cyanide Solution” Trans. 112, A.I.M.E., 660-667; (2) Reynolds “Brief Notes on the Cyanide Metallurgy of Gold” C.M.J. Vol. 65, p. 681, October, 1944; (3) Reynolds “The Physical Chemistry of Cyanidation” Vol. 66, pp. 525-530, August, 1945. Among other conclusions reached by the author of the last papers from a study of the mechanics of precipitations is that the real function of lead in assisting precipitation in the zinc-dust process is that of a catalyst to increase the reaction rate of precipitation and that the formation of a zinc-lead couple is only incidental.
DISSOLUTION OF SILVER
Metallic silver dissolves in cyanide solution according to the same reaction discussed above for the dissolution of gold.
The dissolution of silver sulphide in cyanide solutions, on the other hand, is usually regarded as taking place according to the following reaction:
Ag2S + 4NaCN = 2NaAg(CN)2 + Na2S
In discussing this reaction, Hamilton in Manual of Cyanidation, McGraw-Hill, says:
This, being a reversible reaction, cannot proceed far before reaching equilibrium, unless the product Na2S is removed out of the sphere of action. The latter, however, happens to be very sensitive to oxidation, so that a change rapidly takes place probably in two directions:
- Na2S + NaCN + O + H2O = NaCNS + 2NaOH
- 2Na2S + 2O2 + H2O = Na2S2O3 + 2NaOH
The thiosulphate would tend later to oxidize to sulphate, and perhaps more sulpho- cyanate would also be formed.
The need for supplying excess oxygen is evident, however, and the reactions also explain why thiocyanates (sulphocyanates) are always present in solution when cyaniding a silver sulphide ore.
The use of a lead salt is usually a material aid to the extraction of silver, and while this has generally been explained as due to its reaction with the Na2S present to produce insoluble lead sulphide, Clennell suggests that the lead is rather to be regarded as an aid in the attack on the silver, thus:
Ag2S + PbO + 4NaCN – 2NaAg(CN)2 + PbS + Na2O
Table 91 compiled from U.S. Bureau of Mines Publications shows the principal silver minerals, with their composition and cyanidation and flotation characteristics.
It will be noted that the silver chloride, Horn silver, reacts rather differently in solution from the sulphides and does not require the presence of oxygen for complete solution.
Because of the greater actual weight of metal to be dissolved from commercial silver ores as compared with gold ores, the consumption of cyanide due to dissolution alone is no longer the almost negligible factor it is in the case of most gold ores. Hamilton points out that in cyaniding a 400-oz.-per-ton silver concentrate, as much as 32 lb. cyanide (as KCN) would be consumed in dissolving the silver. In such cases special methods of cyanide regeneration may become necessary.
Cyanidation
Conditioning operations at the Dome plant in Ontario comprise grinding in water to produce a pulp 83 per cent through 200 mesh which is passed over corduroy. Tailing from the corduroy tables, after the addition of lime, is aerated in four 14- by 42-ft. Pachuca tanks. The pulp is then treated with cyanide and agitated in a further series of Pachuca tanks.
At Noranda an aerating step precedes each section of the primary roughers in the flotation plant and also on the pyrite recleaning circuit. The machines employed for this purpose are described as aerating classifiers and serve the double purpose of aerating the pulp and classifying into an overflow product for further processing and an underflow for return to the grinding circuit. They consist of circular tanks about 15 ft. deep and of diameters varying from 9 to 16 ft., having a plurality of radial air inlet pipes, four rubber air-lift pipes, and a slow-moving rake mechanism at the bottom.
The tonnage delivered to each ball mill is measured by two factors, a revolution counter and the weight of a section of the belt load. The cutter used to remove this consists of two parallel plates attached to and rigidly held by a cross bar. The distance between the plates is equal to one-fifth of a revolution of the head pulley. In taking the belt-weight sample, the conveyor is stopped, the cutter placed down on the belt, and the ore between the plates carefully brushed off into a pan and weighed. This is done every hour. The sample is then passed through a Jones sampler until about ½lb. is left. Determination of moisture is made before, the ore has a chance to dry in the air. At the end of each day the composite sample is again cut down and sent to the assay office.
A unique system for sampling various pulps consists of several air-operated cutters, all controlled by one master controller. The master controller is made up of a timing mechanism driven by a Telechron motor, a Geco sampler, a four-way valve, and an air header. The four-way valve is operated by the Geco sampler. The two pressure ports of the four-way valve are connected to a two-partition header. The cylinders operating the cutter are connected to the header by ¼-in. copper tubing. Wedge- shaped cutters are operated by a cylinder and piston, being made to cut the pulp stream every 15 min.
Black also notes the need of a truly representative sample of press heads and tails. The common fault of a drip wire in the whole stream is that the rate of drip is not proportional to the rate of flow. The apparatus devised by Black is illustrated in Fig. 57 and described as follows in E. and M.J., November, 1944.