investigating what had been done at the station in treating lead carbonate ores containing some silver and gold. Most of the roasting tests had been in blast roasters, but Wigton used an assayer's muffle. By mixing the ore with common salt and heating to a relatively high temperature, he was able to volatilize not only the lead but also the gold and silver. He attributed this result to the more highly oxidizing atmosphere of the muffle. Several somewhat similar experiments were made by the bureau after being apprised of his success. Chloridizing roasting has been applied to the treatment of silver ores for many years. The process consists in converting the silver in a nonsmelting ore into silver chloride, dissolving the chloride with a desirable lixiviant, and precipitating and recovering the metal from solution. Recently this practice has been extended to the ores of copper and gold. The work at the Salt Lake station has shown the possibility of treating lead ores in this way. It is well known that this chloridizing roast, when done for leaching only, entails volatilization losses. The temperature is maintained at 700° C. and sometimes rises higher, causing large losses of lead or copper, and comparatively large losses of silver by the old method. The temperature must be kept high in order to effect good chloridization. Therefore, in order to prevent the losses mentioned above, complete volatilization and precipitation by electrical precipitator of the metals is suggested. The complete volatilization process consists in mixing the ore with common salt in the usual way, heating to sufficiently high temperature to drive off the metallic chlorides, precipitating the chlorides, calcining these chlorides, and converting them into the metals by refining.

This method is by no means a new one. It was repeatedly proposed in the earlier days of western metallurgy. A large number of patents have been taken out within the past 25 years, in which several different methods of volatilization are covered. The processes proposed in these patents involve the use of the alkaline or alkaline-earth chlorides, and most of them of sulphur or iron pyrites, it being claimed that the presence of sulphur is necessary for chloridization. The Pohlé-Croasdale process was tested commercially, a testing plant of fairly large capacity being operated in 1903 by a metal refining company, of Denver, Colo. The fumes were collected in cooling towers, filtered, and refined. The failure was due chiefly to the inefficient method of catching the volatilized metals, also the tonnage treated seemed to be extremely small for the size of the furnace used.

• Chanute, Arthur, U. S. patent 501559, Jan. 9, 1892; Besemfelder, E. R., U. S. patent 543284, Dec. 27, 1893; McNight, Robert, U. S. patent 693982, Dec. 5, 1900; Pohlé, E. C., and Croasdale, Stuart, U. S. patent 741712, Jan. 23, 1900; Clawson, S. I., U. S. patent 1192037, Oct. 28, 1912; Forland, T. R., U. S. patent 1078779, Mar. 27, 1913; Clawson, S. I., U. S. patent 1169530, May 24, 1911; Biggs, W. H., English patent 11031, July 29, 1915.

DESCRIPTION OF EXPERIMENTS.

Tests were made at the Salt Lake City station by C. E. Williams, of various oxidized and semioxidized ores of lead and copper that contained gold and silver, and some contained zinc. The roasts were made in a gas-fired assay muffle, small roasting dishes containing 100-gram charges of ore mixed with sodium chloride and other reagents being used. The proper proportions of sodium chloride and calcium chloride, the effect of oxidizing and reducing atmospheres, and the desirability of using iron pyrites, coal, or manganese dioxide, were determined. In all of these tests the fumes were allowed to escape. The results of the tests are shown in Tables 28 to 35, following.

TABLE 28. Data on tests of material from Nevada United mine, Ely, Nev.

[Analysis of heading: Pb, 5.58 per cent; Ag, 3.20 ounces per ton; Fe, 37.6 per cent; CaO, 1 per cent; Al2O3, 2.2 per cent; S, 1.15 per cent; insoluble, 11.6 per cent. Size: Passed 20-mesh sieve. Charge: 100 grams ore, 10 grams water.]

TABLE 29. Data on tests of material from Dry Valley mill dump, Pioche, Nev.

[Analysis of heading: Pb, 7.12 per cent; Cu, 0.25 per cent; Ag, 10.04 ounces per ton; Au, 0.10 ounce per ton; Fe, 5.36 per cent; CaO, 0.71 per cent; Al2O2, 3.80 per cent; S, 1.08 per cent; insoluble, 78 per cent. Size, 90 per cent through 80-mesh. Charge: 100 grams ore, 5 grams water; variable chlorides.]

TABLE 30.-Data on tests of material from Horn Silver dump, Frisco, Utah.

[Analysis of heading: Ag, 6.70 ounces per ton; Au, 0.02 ounce per ton; Pb, 7.4 per cent; Zn, 6.3 per cent; Cu, 0.25 per cent; Fe, 4.4 per cent; insoluble, 53.6 per cent; CaO, 2.2 per cent; Al2O3, 8.5 percent; S.6.13 percent. Salt mixture used, equal parts of NaCl and CaCl. Theoretical quantity of salt mixture required for Pb, 3.1 per cent; for Zn, 11.4 per cent.]

TABLE 31.-Data on tests of material from Bingham Mines Co., Bingham, Utah.

[Analysis of heading: Ag, 8.06 ounces per ton; Au, 0.01 ounce per ton; Pb, 6.85 per cent; Cu, 0.65 per cent; Zn, 4.53 per cent; Fe, 8.85 per cent; insoluble, 56.3 per cent; CaO, 7.3 per cent; Al2O3, 6.2 per cent; S, 10.8 per cent. Salt mixture used consisted of equal parts of NaCl and CaCl. Theoretical quantity of salt mixture required for Pb, 3.8 per cent; for Zn, 8.1 per cent.]

TABLE 32.-Data on tests of material from Yellow Pine mine, Goodsprings, Nev. [Analysis of heading: Ag, 3.9 ounces per ton: Pb, 9.32 per cent; Zn, 29.25 per cent; Fe 2.7 per cent; insoluble, 14.8 per cent; CaO, 6.8 per cent; MgO, 5.5 per cent; S, 0.31 per cent; size, through 10-mesh.]

TABLE 33. Data on tests of material from Trapper mill dump, Melrose, Mont. [Analysis of heading: Pb, 2.58 per cent; Cu, 0.64 per cent; Zn, 11.32 per cent; Ag, 8.7 ounces per ton, Au, 0.04 ounce per ton; Fe, 7.04 per cent; insoluble, 54 per cent; CaO, 5.5 per cent; Al2O3, trace; S, 0.30 per cent; size, through 20-mesh.]

TABLE 34. Data on tests of material from Hayden mill feed, Clarkdale, Ariz. [Analysis of heading: Ag, 7.10 ounces per ton; Au, 0.15 ounce per ton: Pb, 1.10 per cent; Cu, 0.45 per cent; re, 1.42 per cent; insoluble, 66.2 per cent; CaO, 0.66 per cent; S, 1.2 per cent; size, through 20-mesh.]

TABLE 35. Data on tests of material from Sells mine, Alta, Utah.

[Analysis of heading: Ag, 16.35 ounces per ton; Au, 0.04 ounce per ton; Pb, 10.15 per cent; Zn, 6.04 per cent; Fe, 1.98 per cent, insoluble, 6.62 per cent; CaO, 15.75 per cent; MgO, 23.9 per cent; S, trace; size, through 20-mesh.]

Later tests were made in a small oil-fired mechanical roaster, and in these the volatilized fumes were passed through a single-pipe electrical precipitator and collected. The fume was melted and reduced by heating with additions of calcium oxide and coal dust, a slag of CaCl, being obtained.

DISCUSSION OF CONDITIONS OF VOLATILIZATION.

USE OF CALCIUM CHLORIDE.

As calcium chloride may be recovered in the refining process, it would be desirable to make use of that compound as the chloridizing agent. The equations for the reactions are as follows:

PbCO3+CaCl2+SiO2=PbCl2+CaSiO3+CO2.

(In oxidizing atmosphere.)

PbCl2+CaO+C=Pb+CaCl2+CO.

(In reducing atmosphere.)

From these equations it is seen that as much CaCl, is formed as is needed for chloridizing the metal. The amount of sodium chloride necessary for proper volatilization was found to be one and one-half