be increased to a satisfactory degree by double roasting of the material after finer grinding, followed by brine leaching. As much as 92.7 per cent of the lead was recovered in this way, nor does it seem possible to get good extractions without such treatment.

TABLE 25.-Results of shaft roasting of material from Wilbert tailings dump.

[Tests by C. L. Larson.]

Assay of original material: Lead, 5.51 per cent; silver, 0.75 ounce per ton.

The anomalous behavior of a number of the roasts led to the belief that the lead chloride was being converted to some other form not soluble in water. In every test, more lead was soluble in brine than in hot water, although in many of the tests it seemed that the 39094°-18-Bull. 157- 5

percentage of insoluble lead was larger than it should be. This might be explained on the assumption that either silicates of lead were forming or some oxychloride of lead. Some of the quick hightemperature roasts that were not in contact with air as long as some of the others gave higher extractions, so it was thought that possibly the exclusion of air during cooling of the calcine might prevent the formation of such compounds. Covering the furnace during cooling partly excluded the air, and it seems that the extraction was higher. Quenching in water also seemed to considerably improve the process. Due to variations in roasting, it was not possible to be sure that these methods of handling the hot calcines were doing what they seemed to do. Therefore, in test 21 part of the calcine was quenched in water as soon as the roasting was completed, and the rest was covered to exclude air. The quenched calcine gave an extraction of 49 per cent in hot water, whereas the calcine that was cooled slowly with partial exclusion of air only gave 27 per cent extraction. This seemed to prove that the lead chloride was being converted into some insoluble complex compound during cooling.

In any event, double roasting of such low-grade material would not be practicable, so the effect of adding sulphide was next taken up. The results of tests 27 and 29 show that with sulphide alone less lead sulphate is formed than when both pyrite and sodium chloride are used. Whether the pyrite helps in the chloridizing or the sodium chloride helps in the sulphating of the lead is impossible to say, as any lead chloride formed would react with the sodium sulphate as soon as leaching was begun. Hence brine is needed for the leaching, as it will dissolve both lead chloride and lead sulphate. In tests 33 to 39 more salt was used and better extractions were obtained. This would make it appear that the presence of pyrite helps chloridizing. Evidently about 8 per cent of sodium chloride and the same of pyrite are necessary to get higher than an 80 per cent extraction of the lead in this material.

An analysis of some of the tailing showed that about 80 per cent of the silver in the original material was also chloridized and extracted.

The action of the lead in the roasts where no pyrite was added was still puzzling. There are several known oxychlorides of lead that are insoluble, and one of them is prepared by heating lead chloride. Some chemically pure lead chloride was heated in a porcelain crucible to a low red heat and first melted down. At red heat it began to give off fumes of chlorine and probably some lead chloride volatilized. A yellow residue resulted, which, on being boiled with water, turned white and was mostly insoluble. This seems to indicate the cause

o Muir, M. M. P., and Morley, II. F., Watts' dictionary of chemistry, vol. 3, 1912, p. 126.

of the trouble. In the highly chloridizing atmosphere afforded by the reaction of the oxides of sulphur from burning pyrite on the sodium chloride, the lead chloride formed could not decompose thermally, whereas in the presence of sodium chloride alone the thermal decomposition of lead chloride gradually took place, especially in a stream of air.

CONCLUSIONS FROM RESULTS OF TESTS.

The conclusions to be drawn from these experiments are as follows: 1. Argentiferous lead carbonate ores in which the lead and silver are not already in soluble forms can be chloridized in a shaft roaster of the type used. The experience of two mills operating for considerable lengths of time has shown that these ores will yield their silver, and experiments at the Bureau of Mines station show the conditions necessary for keeping the lead in soluble form.

2. The addition of both salt and sulphide to the charge is necessary in roasting in order that the lead may be converted to a form soluble in a neutral brine. Lead carbonate and the compounds into which it is converted in roasting are soluble in acid brine, so that the roast need not be controlled with the idea of converting the lead into forms soluble in water or in neutral brines. The requisite conditions for properly chloridizing the silver are the addition of both salt and sulphides to the roast in order to get good chloridizing, and these conditions will not interfere with efficient extraction of the lead.

3. Some lead is volatilized at the temperatures used in roasting, but if the temperature is not allowed to rise above 700° C. the amount volatilized will be small, and the use of an electrical precipitator, as shown in later experiments, will collect all of the material that volatilizes.

VOLATILIZATION OF LEAD FROM OXIDIZED ORES.

Throughout the experiments described in the foregoing pages it was noticed that some lead always volatilized in roasting, especially when a charge of shallow depth was used. Presumably the volatilized lead was lead chloride, as the boiling point of lead chloride is placed at 861° C. Before any attempt to utilize this fact had been made by the bureau, it was learned that N. C. Christensen had been utilizing this idea in a proposed new process that he was developing for a mining company at Pioche, Nev. Subsequently a number of tests were made by Christensen at the Bureau of Mines station, with the assistance of M. J. Udy.

a Landolt, H. H., Landolt-Börnstein physikalische-chemische Tabellen, 1905, 3rd ed., p. 270.

TESTS WITH CHRISTENSEN ROASTER.

Christensen used a laboratory roaster he had developed for testing ore in the Knight-Christensen mill, at Silver City, that had burned. As previously stated, the commercial Christensen roaster for chloridizing ores is a down-draft roaster with a circular hearth of annular shape and a suction box beneath for creating the draft. A shallow bed of ore is used (4 to 8 inches) and conditions for volatilization of lead chloride are much better, as there was no zone of cold ore to condense volatilized material.

The roaster brought by him to the bureau laboratory is shown in figure 11. This roaster was constructed from a piece of 8-inch pipe with a cap on the lower end. The charge was supported on an iron-wire screen. Air was sucked down through the charge with a Root cycloidal blower, the gases being discharged from the blower

FIGURE 11.-Christensen down-draft laboratory roaster.

on the pressure side. To regulate the draft through the charge a valve in a T joint in the suction pipe was set to cause the proper suction. The charge was ignited by means of an electric igniter made of resistance wire. This igniter consisted of a piece of perforated sheet asbestos on the underside of which was wound a spiral helix of No. 22 B. & S. gage nichrome wire, connected to a 110-volt lighting circuit. The furnace was charged with material that had been crushed and prepared in the same way as for the Holt-Dern laboratory roaster. The igniter was placed over the charge and the current turned on. After the ore on top showed that it was getting hot, by a slight crackling or glowing, the blower was started and immediately the charge would ignite. As soon as the surface of the charge was glowing the current was turned off and the igniter removed. The combustion zone would travel down. through the charge in 5 to 20 minutes, and the fire would then die out unless the charge consisted principally of sulphide minerals,

when the whole mass would ignite and burn for as much as an hour. It was found best to support the screen in a smaller piece of pipe that would fit into the pipe shown in the drawing. This piece could be conveniently lifted out with the roasted charge of ore.

The experiments by Christensen with his roaster in the bureau laboratory are described in the following paragraphs. The basic theory was that at higher temperatures than those necessary for chloridizing of silver the lead chloride formed is volatilized. The temperatures used should be near to or above the boiling point of lead chloride, 861°C. Too high a temperature will decompose any silver chloride formed in the roasting process. In roasting a carbonate ore mixed with powdered coal and salt it is a peculiar fact that the only metal to be volatilized as chloride in important amounts at these temperatures and in the length of time allowed for roasting in this furnace is lead. Only a very small amount of silver, gold, or copper volatilizes, and hardly any zinc chloride is formed. In order to get a good extraction of the lead by volatilization, the temperature must be raised so high that all of the silver will not be soluble in saturated brine.

TESTS OF MATERIAL FROM BULLIONVILLE TAILING DUMP.

The first material tested was that from the Bullionville tailing dump near Pioche, Nev. This is finely ground oxidized material containing lead carbonate and some gold and silver. The sample tested had the following content of valuable metals: Lead, 9.7 per cent; copper, 0.32 per cent; silver 10.9 ounces per ton; and gold, 0.135 ounce per ton. The material in this dump had been treated three times previous to being left in its present condition, having been pan-amalgamated twice and cyanided once.

The results of the tests are shown in Table 26. In some of the tests pyritic sulphide was added to the charge to assist chloridizing, and in others limited proportions of sulphide were used and ground coal was added to provide sufficient fuel. In other tests only coal was used in order to determine whether it was possible to chloridize and volatilize the lead without using sulphides. In these tests the stronger the suction the higher was the temperature with a given quantity of fuel.

With increasing temperature more and more of the lead was volatilized. However, in order to leach gold and silver from the tailing with a brine solution, the optimum conditions for their extraction are such that 30 per cent of the lead is not volatilized. The calcines were leached with four parts of brine to one of original material, hence those charges in the tests that show low volatilization of the lead contained too much lead to be taken into solution.