product. Such material needs to be calcined at a temperature high enough to drive off the sulphur dioxide. As regards the preparation of a zinc oxide product suitable for distillation of zinc, the presence of zinc sulphate and sulphur dioxide would probably not be objectionable. Hence, as a method of concentrating the zinc in a mixed sulphide ore, igneous concentration is practicable.

At Florence, Colo., the River Smelting Co. has recently installed a blast furnace for direct treatment of mixed sulphide ores. A strong blast is used and a shallow bed; thus the furnace operates with a hot top. Moxt of the pyrite of the ore oxidizes and the iron oxide enters the slag. Some of the iron and all of the copper go into a matte, which also collects the precious metals. The lead and most of the zinc are distilled as fume and are caught in the bag houses. Some of the zinc goes into the matte and the slag, and to date the recoveries of zinc in the fume have been lower than is desirable. Also, fume deposits as crusts on the walls of the furnace, necessitating a shutdown every 10 days or two weeks for cleaning the walls.

At the Tooele smelter in Utah an attempt to smelt complex sulphides and then "blow" them in a copper converter met the same difficulty, accretions of zinc oxide forming on the walls of the con

verter.

ELECTROSTATIC SEPARATION.

Electrostatic separation, as previously mentioned, is one of the processes used for separating mixtures of sulphides of the same specific gravity that are not amenable to gravity concentration. Zinc ore mixtures of pyrite and sphalerite in coarse crystals are often treated by this process. Such ore needs only drying before treatment on the electrostatic separator. The principle of the electrostatic separators need not be explained here, as there is ample literature on the subject. From 1905 to date there have usually been several papers published every year on this process in such publications as Mineral Industry, Engineering and Mining Journal, Metallurgical and Chemical Engineering, and the Transactions of the American Electrochemical Society. The nonconductive mineral particles pass through the separator inert, but the particles of the conductive minerals acquire an electrical charge and are repelled from one electrode. Sphalerite is a poor conductor or a nonconductor, and pyrite is a good conductor. Galena is likewise a good conductor and accompanies the pyrite. Fine dust can not be treated on such a machine, its use being limited to the treatment of "sands."

As ores vary widely in their electrical properties, the only way of finding whether a given ore of zinc sulphide and other sulphides can be separated by this method is to test the ore in a commercial machine. With some ores electrostatic separation produces concentrates of commercial grade without further treatment, but as a rule none of the

electrostatic processes will make a clean separation of the average complex sulphide ore found in the intermountain States. The fact that pyritic ore does not have to be roasted before electrostatic separation is important, as the sulphur can be saved, if desired. If the magnetic separation process were used, some of the sulphur in the ore would be burned off and escape in the smelter gases, forming a nuisance. Another advantage of the electrostatic method is that mixtures of galena, pyrite, chalcopyrite, and sphalerite can be separated into a product containing the first three minerals mentioned and a second product containing the sphalerite. These two products can both be marketed to good advantage. The magnetic separation process, however, makes a product containing the iron and the copper and another product containing the lead and the zinc; and this latter mixture can not be marketed to as good advantage as if the lead and zinc minerals were separated.

MAGNETIC ROASTING AND SEPARATION.

The method of separating zinc and iron sulphides by the use of a magnetic separator is an old one. There are many ores, in districts like the Wisconsin zinc district, which are mixtures of the two sulphides that can be easily separated by this method, and in these places the practice has become standard. In the intermountain States, however, there are many sulphide ores that do not respond to this treatment. Determining the character of such ores by their appearance is difficult, and the best way of finding whether a given sample of ore is amenable to magnetic separation is to test it.

EXPERIMENTS AT SALT LAKE CITY STATION.

The Zinc Concentrating Co., in cooperation with the University of Utah and the Bureau of Mines, set up one of its Campbell magnetic separators and a small Etherington-Singer roaster at the Salt Lake City station in order to test mixed sulphide ores from various western districts.

DESCRIPTION OF ROASTER AND MAGNETIC SEPARATOR.

The Etherington-Singer roaster is a revolving cylindrical kiln, tapering slightly toward the flue end and being fired with oil or gas. The small laboratory roaster used for the tests was an iron cylinder 7 feet long and about 1 foot in diameter. Heating was by means of two large oil blow-torch flames, the roaster generally requiring about one hour's heating to attain the temperature necessary for a quick roast under these conditions. The roaster was so inclined that when it was making four revolutions per minute the ore passed through it in 7 minutes. The ore, just before discharging, showered through the flame from the torches. It was found that under these conditions enough of the sulphur in the pyrite was quickly eliminated and the product,

because of the nonoxidizing character of the flame, had not been oxidized too much. Of course, the object in magnetic roasting is to raise the ore to such a temperature that the pyrite breaks up into FeS or Fe, S, +1, and S. Pyrite is not magnetic, but the lower sulphides of iron are ferromagnetic, and are capable of being lifted by a magnet of low intensity. Sphalerite is nonmagnetic; hence the magnetic separator removes the iron and leaves the sphalerite, with any other nonmagnetic minerals present.

The Campbell magnetic separator comprises a shaking plate so inclined that the ore travels down it in a thin sheet; above this plate are suspended three electromagnets. Beneath the face of each magnet passes a canvas belt, so that any particle attracted to the magnet is arrested by the belt, which carries the particle over the edge of the shaking table and beyond the field of the magnet, when the particle drops from the belt. The three magnets used are of increasing intensity, so that the separator makes three products which are decreasingly magnetic, and a nonmagnetic product. The distance of the faces of the magnet poles from the table is adjustable, and rheostats control the amperage flowing through each magnet, so that a stronger or weaker magnetic field is obtainable.

TESTS OF ORES.

Very few of the western ores not now being worked that were tested had sufficiently coarse crystallization to permit magnetic separation of the iron sulphide from the zinc sulphide. Tests were made of ore from the Utah-Apex mine, Bingham, Utah; zinc middling from the Broadwater mill at Park City, Utah; zinc middling from the Consolidated Nevada-Utah mill at Pioche, Nev.; ore from the Greenwood lease, Pioche, Nev.; ore from the Carlyle mine, Duncan, Ariz.; ore from the Middle Golconda mine, and ore from the C. O. D. mine, near Chloride, Ariz., and others.

As stated above, only a few of the ores responded satisfactorily in the tests. The essential data on the more important tests follows.

ZINC-IRON MIDDLING FROM UTAH-APEX MINE, BINGHAM, UTAH.

The Utah-Apex mine at Bingham, like most of the other Bingham lead mines, has not saved the zinc in its ore, much of the ore being shipped crude to the lead smelters. This ore has the following analysis:

Composition of ore from Utah-Apex mine, Bingham, Utah.

The lead mineral in this ore can be separated by mechanical concentration, but the zinc and iron minerals, having the same specific gravity, are obtained as a zinc-iron middling and the high iron makes this middling unsalable as a zinc product. As the ratio of the iron content to the silica content is such that the ore is self-fluxing, it is sought by the smelters and often contracts can be obtained which ignore the high zinc content. Hence very little concentration of this grade of ore has been practiced. By making a middling on Wilfley tables, a product containing zinc sulphide, iron sulphide, and quartz was obtained, and on tabling the gangue a product of the following analysis was obtained:

Such a product could not be marketed to advantage as zinc concentrate; so this middling was subjected to magnetic roasting and separation. In the treatment on the magnetic separator, the intent was to remove most of the iron, the magnets being set for a low intensity, and then re-treat the zinc concentrates with a more intense field. The results of these tests are given in Table 7 following: TABLE 7.-Results of magnetic separation of Utah-Apex zinc-iron middling.

ORES FROM THE CONSOLIDATED NEVADA-UTAH AND THE GREENWOOD MINES, PIOCHE,

NEV.

The ore from the Consolidated Nevada-Utah mine and that from the Greenwood lease are representative of two types of ore found in the Pioche, Nev., district. The former represents the deeper lying ore in the vein deposits of that district, and the latter the complex, partly oxidized sulphides nearer the surface. This latter type of ore is much more complex and difficult to separate than the first. It can be seen from Tables 8 and 9 that marketable zinc concentrate could be prepared from the Consolidated ore, whereas

a good separation of the zinc and lead minerals from the iron sulphide in the Greenwood ore is practically impossible. High-iron zinc ore or zinc-lead ore is sold for the manufacture of zinc oxide or leaded zinc oxide, although the price paid for the ore is not equivalent to that paid for the separate lead and zinc concentrates that might be made. The companies buying such ore usually specify 35 per cent combined zinc and lead content as the lower limit of the material that they will accept for making zinc oxide. The Greenwood ore can be prepared for such a market by double treatment with magnetic separators, as can be seen from Table 9, but the loss of zinc in the iron concentrate is too great.

TABLE 8.-Results of magnetic separation of ore from Consolidated Nevada- Utah mine.

arated, makingNo. 1 "iron".

No. 2 "iron".

3.0 2.14 2.3 45.9 5.3 7.8 1.5 5.8 1.5 4.88 2.7 23.0 1.3 21.8 2.0 14.4 Zinc concentrate. 24.0 5.23 45.8 6.5 6.0 44.6 67.1 12.2

TABLE 9.-Results of magnetic separation of ore from Greenwood lease.