In Mexico and elsewhere there are large bodies of valuable silver ore containing manganese which present great, and hitherto insuperable, obstacles to a satisfactory extraction by cyanide. The proportion of the silver that is soluble in cyanide varies greatly and may be as low as 5% only. The mere presence of manganese in the ore is not invariably accompanied by refractoriness of the silver because the writer has known of an instance where ores of apparently similar composition and character and coming from adjacent mines behaved quite differently, one yielding only 50% of its silver and the other 90% under the same conditions. As a general rule, however, when oxides of manganese are present in a silver ore trouble may confidently be expected.

The manganese usually occurs as the dioxide but in what way, if at all, it is associated with the silver has never been determined. There are three possible theories on the subject. The first is that the refractory silver occurs as a silver-manganese mineral. Although no such mineral is described in the text books, there are certain facts which would seem to point to its existence:

(a) The silver is insoluble in nitric acid, in connection with which it is to be noted that MnO2 is also more or less insoluble in nitric acid.

(b) If the oxides of manganese be removed by some solvent, such as hydrochloric or sulphurous acid, the silver is at once rendered soluble in cyanide solution.

On the other hand there are circumstances which seem to point rather to the idea that the silver may be free from the manganese but occurring in some unknown insoluble form, the result of oxidation. In favor of this theory may be mentioned the fact that:

1. in the case of manganese ores it commonly happens that the ore in the upper levels is more refractory than that in the lower, though containing the same percentage of manganese.
2. An oxidizing roast applied to any silver ore usually renders the bulk of the silver insoluble in cyanide.
3. A preliminary reducing treatment given to a refractory manganese ore, as for instance agitation with an alkaline-hydrosulphide or a roast with sulphur or crude oil in a reducing atmosphere, will increase the percentage of the refractory silver subsequently soluble in cyanide.
4. If a sample which has been through such a reducing treatment be subsequently subjected to an oxidizing effect, as for instance blowing air through the Wet pulp, there is a tendency to throw the silver back into the insoluble form. It may be admitted as against this theory, that a simple silver oxide or salt insoluble both in nitric acid and potassium cyanide has hitherto been unknown.

The third hypothesis is that the silver is combined with some mineral other than manganese dioxide, but that the mineral combination never occurs except in company with the manganese though the manganese may occasionally occur unaccompanied by that particular silver mineral. None of these theories, however, seems to conform to all the known facts.

Following up the indications of the benefit of preliminary reducing treatment for these ores, a process has been patented in which the ground ore is roasted in an atmosphere of producer gas previous to cyanidation by which means the inventor claims to be able to obtain extractions up to 90% and over. It would seem, however, that as in the case of the chloridizing roast the ore would need to have an unusually high silver content to make this process possible.

Where the grade of the ore is sufficiently high to make the process commercially profitable a good extraction can usually be obtained by a chloridizing roast followed by cyanidation. The feasibility of this method will depend largely on the cost of fuel and salt at the mill as well as the assay value of the ore. There will also be a loss entailed by volatilization of a part of the gold if gold be present and also to a less extent of the silver. This could no doubt be corrected to a large extent by installing a Cottrell precipitator for treating the fume from the furnace though at the expense of increasing both installation and running costs. It would also be necessary to mill the ore dry, a method which is usually to be avoided whenever possible. Flotation has been tried for dealing with such ores with the result of producing a small quantity of high-grade concentrate, but a very low percentage recovery from the ore; moreover, generally speaking, the silver recovered by flotation is just that part of the silver content that is amenable to cyanide while the refractory silver remains in the tailing.

Probably the most hopeful chance of finding a solution of this problem lies in the process first suggested by the writer early in 1909 and mentioned in the Mining and Scientific Press of December 4th in that year. It consists in treating the ground ore in a water pulp with SO2 gas, or its solution in water by which the manganese is dissolved and washed out leaving the silver free to be acted on by cyanide. Extractions up to 95% of the silver contents have been made on refractory ores by this process and while the extra expense is considerable it will in most cases be a good deal lower than that involved in roasting processes, and might under favorable conditions be found of payable application to ore of average grade. At the present time (1919) experiments are being conducted with very promising results along these lines in Pachuca using the concentrates from the sulphide ore to furnish sulphur gas for the treatment of the refractory ore. Walter Neal appears to prefer the use of sulphuric acid and ferrous sulphate as a substitute for the sulphurous acid stating that he found the extraction due to these reagents somewhat superior to those from the SO2 but the method would seem a good deal more expensive unless the solutions could be regenerated electrolytically. The writer has made some tests along these lines, but without very encouraging results.

For those who wish to experiment with this process it may be said that the most satisfactory method for laboratory use consists in the addition of varying amounts of a solution of SO2 gas of known strength which may be prepared by boiling strips of metallic copper in concentrated sulphuric acid, passing the gas through a washing bottle and thence into distilled water until saturated. The method of determining the SO2 content of a solution is as follows:

Take from 50 to 200 cc of N/10 iodine solution according to the strength of the SO2 solution and add to it from a pipette 5 to 10 cc of the SO2 solution. Should the brown color of the iodine be dispelled or nearly so, it will be necessary to make a new test using more iodine since the iodine must be in excess. Then titrate the excess iodine with N/10 Na2S2O3. The original amount of N/10 iodine less the quantity shown by titration gives the iodine consumed by the SO2.

1 cc N/10 liter consumed = 0.0032 gr. SO2