wheel-barrow load being used to a charge, and only in the daytime. On that account the fuel consumption is doubtless too high, varying from 25 to 33 per cent of coke, based on the weight of the ore.

At first the standard lead-smelting slags containing various proportions of iron, silica, and lime were aimed at, but the recoveries of zinc were low, varying from 10 to 50 per cent. Slags so calculated that an excess of one of the slag-forming constituents over the others was used gave much better results. If lime or silica were used in excess, 80 to 95 per cent of the total zinc was obtained, but the slag had a much higher melting point, and more fuel was needed to make it pass from the furnace. Presumably the use of slag with a high melting point permits all the zinc to distill from the ore before the gangue is fluxed; thus, the slag has no opportunity to take up zinc oxide. The use of a high percentage of lime might well be expected chemically to displace zinc from the slag, but the successful use of a high percentage of silica can be explained only on the assumption that the formation temperature of the slag is raised above the boiling point of zinc. In fact, one slag containing 67 per cent SiO, was formed during a series of tests with increasing amounts of silica. This slag was so sticky that the tap hole had to be drilled and the slag removed by "balling" on the end of a cold iron rod thrust through the hole.

The results of some of these tests are tabulated in Table 37. Recoveries were low with the high-iron slags, but good recoveries were obtained with the high-lime, high-silica, or high-alumina slags. Slags which are unusual in lead, copper, or iron blast-furnace work were formed in this small furnace and removed with very little trouble. The small size of the furnace permitted quick adjustment of the air blast when the slag seemed in danger of sticking; also sticky slag could be removed by balling whenever the furnace was freezing. However, the slag need not be unbalanced to the danger point in order to get as much as 85 per cent recovery of the zinc.

TABLE 37.-Results of igneous concentration of oxidized zinc ore in a small blast furnace.

Some of the most difficult zinc ores to handle by present commercial methods are the partly oxidized ores in which only the sulphide of zinc can be concentrated. The results of four tests of oxidized ores containing some zinc sulphide are recorded in Table 38. These indicate that the presence of as much as 10 per cent sulphur does not affect the zinc recovery. The ore used was from the Daly-Judge mine at Park City, Utah, and in addition to zinc sulphide also contained sulphide of lead. No lead collected in the bottom of the furnace, and the final slag contained only traces of that metal; consequently, the lead went into the zinc oxide and the product made would be a "leaded" pigment. A serious drawback to making pigment from such ore by this method is the formation of sulphur compounds, which contaminate the product. On the other hand, in making a zinc concentrate, the presence of lead in the product would be a disadvantage. Hence, for making pigment, a preliminary roasting to drive off the sulphur before igneous concentration might be necessary, whereas if the product were sold as zinc concentrate, the smelter would not pay full value for the zinc on account of the lead present.

In

Some explanation of the figures in Table 38 is necessary. all these tests the zinc assays were obtained by decomposing the slag samples with a mixture of nitric and hydrochloric acids, followed by the use of hydrofluoric acid to break up ferrates of zinc. However, only hydrochloric and nitric acids were used in determining the "insoluble", which is not the "absolute silica;" hence, the slag analyses are largely of comparative value only.

TABLE 38.-Results of igneous concentration of low-grade sulphide ores of lead and zine in a small blast furnace.

Tho experionco of the River Smelting Co., at Florence, Colo., as mentioned previously, has been that in smelting zinc ore in a blast furnace accretions of zinc oxide and "blue powder" tend to form on the walls of the furnace above the smelting zone. Such masses are so hard to bar down while the furnace is running that the practice is to shut down the furnace every ten days or so for removing the accretions. This method is, consequently, rather more expensive than is desirable. The tests at the Salt Lake City station were not sufficiently extended to observe this trouble.

In order to obtain an idea of the composition of the zinc oxide product and its physical characteristics, several methods were used in collecting it. A wire screen placed across the throat of the blast furnace collected material that contained only about 67 per cent zinc and was contaminated with flue dust. A stream of the furnace gases by-passed through a cold pipe dropped part of its load as a 73 per cent zinc product. Gases from the discharge of the blast furnace sucked through an electrical precipitator comprising a single 4-inch iron pipe to which was applied 18,000 volts, could be almost entirely cleared at velocities ranging up to 18 feet per second, although most of the time a rate of not more than 10 feet per second was maintained. The product, a fluffy white material containing 77.6 per cent zinc, had a slight bluish tint, possibly due to the presence of small amounts of carbon or of some unoxidized zinc (blue powder). There seems to be no reason why this product could not be used for pigment.

The value of the igneous concentration method for concentrating low-grade oxidized ores of zinc may be estimated from the following two representative contracts, which are commonly possible in the marketing of oxidized zinc ores in Utah. With these two contracts as a basis the possible profits from a plant using the process have been calculated.

Oxidized zinc ores are sold at prices calculated on base prices for 35 per cent zinc ore delivered at the smelter. The base prices for 35 per cent zinc ore are as follows: $17 a ton when spelter is 5 cents per pound, $29 when spelter is 8 cents, and $50 when spelter is 15 cents. The base price changes $3 per ton for every change of 1 cent per pound in the price of spelter between 8 cents and 15 cents, and $4 for every change of 1 cent in price between 5 cents and 8 cents. For ore containing less than 35 per cent zinc, a penalty of $2.50 per unit is charged for each 1 per cent deficiency in zinc as far as 30 per cent, and $2 per unit for each deficiency of 1 per cent below 30 per cent zinc content. A credit of $1 per unit is allowed for each 1 per cent. of zinc the ore contains in excess of 35 per cent. The freight rate to the "gas belt" where the zinc ores are smelted is $7.50 per ton of ore.

From these figures the net values of oxidized zinc ores in Salt Lake City have been calculated and plotted as shown in figure 11. It can be seen that with normal 5-cent spelter the lowest grade of ore which can be shipped from Salt Lake City to the "gas belt" is about 31.5 per cent zinc ore. With spelter at 8 cents the lowest grade possible is 26.2 per cent zinc ore. Contracts vary somewhat, so that the minimum zinc content for different contracts may vary 2 or 3 per cent zinc in either direction from the figures given.

For the sale of zinc oxide concentrate the scale of ore values given might be used in calculating the value of the product. The net value in Salt Lake City of such concentrate for different percentages of zinc and for different market prices of spelter has been calculated and plotted on the solid-line curves in figure 12. As

the credit for each extra unit of zinc above 35 per cent zinc is only $1, even when spelter is selling at 10 cents, the contract price for the zinc oxide when spelter is selling at 7 cents or better is too low.

A better offer is one made tentatively by a zinc-smelting company which said it was willing to pay for a ton of zinc oxide product containing 75 per cent zinc a base price of 48.75 per cent of the price of a ton of spelter on the day of sale. Thus an 80 per cent zinc product would bring 80/75 of the base price and a 70 per cent zinc product, 70/75 of the base price. The value under these terms of zinc oxide concentrate for varying conditions is also plotted in figure 12 for comparison with the rates paid under the ore contract mentioned. A comparison of the sale prices of zinc oxide under these two contracts shows that with 5-cent spelter the ore contract offers the best price, but with spelter at 7 cents or more the oxide contract offers the best price. In fact, the difference between the two contracts is so marked

that careful arrangements would have to be made in regard to the sale of the zinc oxide before commercial production by the process proposed could be undertaken. The zinc smelting company making the offer mentioned wished to have a trial carload of the oxide product before it would close the contract, and thus might bid lower for the next car.

If the oxide product were acceptable as pigment, as it should be, it would probably bring a higher price than is shown by either of the two methods used, but this consideration was eliminated in making the calculations, in order to show only the minimum possible

returns.

The cost of a plant for making zinc oxide product would, of course, be dependent on the daily tonnage of ore proposed to be treated. On the supposition that anyone interested in this type of plant would first be concerned in building a small unit in order that every doubt

ZINC IN "OXIDE" CONCENTRATE, PER CENT. FIGURE 12.-Curves showing change in value of zinc oxide concentrate with varying zinc content. SolidIne curves are based on "ore" contract; dotted-line curves on "zinc oxide" contract. Figure near each curve shows price in cents per pound for spelter based on "35 per cent zinc" ore.

ful point might be tested out, the necessary data for estimating the cost of a plant capable of treating 25 tons of ore per 24 hours is presented. The two most important figures that need to be known in order to design such a plant are the volume of air used and the rate of cooling of the hot gases. The figures chosen represent the average amount of air commonly used in lead blast furnaces when they are run with hot tops, and the average rate of cooling of gases in iron flues at a number of smelters. Thus the assumption is made that 200 cubic feet of air per minute is needed per square foot of furnace area measured at the tuyères, and that each square foot of

86198°-19-Bull. 168- -8