TABLE 4.-Results of electric iron smelting with furnace at Trollhättan. a 1,209,825 kg. represented Kiiruna "A" ore containing 69.61 per cent of Fe. The table shows that step by step the results have been improved, the quantity of iron per horsepower-year increased, whereas the electrode consumption and time for repairs have been reduced, these three items in conjunction with the saving in charcoal being the decisive factors as regards electric iron smelting. The figures show that during the last few months of the experimental work as much as 3.20 to 3.10 tons of iron were obtained per horsepower-year, whereas before the alterations were made the highest average figure was 2.86 tons. The highest expected yield was 3 tons per horsepower-year, which was therefore exceeded. It may also be mentioned that during single periods of several weeks when especially suitable ores were used, the highest average figures above mentioned were materially exceeded. It will thus be seen that as regards efficiency the results of the furnace surpassed expectations. The electrode consumption shows a remarkable decrease. The consumption of 13.8 kilograms during the first period of working was finally reduced to about 3 kilograms per ton of pig iron. The cost and time required for repairs were items that could not be estimated beforehand. Experience has shown that both the cost and the time required are less than could have been expected. Utilization of the gas, when practicable, should also be taken into account. At Hagfors the use of gas for firing the open-hearth furnaces is estimated to reduce the cost of the pig iron about 65 cents per ton. DURABILITY OF THE ROOF OF THE CRUCIBLE. In perfecting the electric furnace one of the most serious difficulties that had to be overcome was maintaining the brickwork of the crucible. In the operation of the furnace at Trollhättan from August 4, 1911, to June 21, 1912, the following repairs were made: October 26, 1911, the arch over the crucible was partly repaired; the February 13, 1912, the arch was entirely rebuilt; April 27, 1912, arch was partly repaired. At Hagfors the arch was rebuilt after working four and one-half months. COMMERCIAL FURNACES OF THE SWEDISH TYPE. As a result of the successful working of the furnace at Trollhättan the following furnaces have been built: Furnaces built subsequent to the Trollhättan furnace. However, the furnaces at Hardanger, after having been operated for about nine months, were closed down, the reason given being that suitable raw materials were not available in Norway. In Sweden charcoal is used as a reducing agent, whereas in Norway only coke is available for that purpose. During the experimental work at Trollhättan, coke was tried as a reducing agent, but was found unsatisfactory, and the failure at Hardanger, where coke only was used, confirms the conclusions reached by the engineers at Trollhättan regarding the unsatisfactory results with the use of coke as a reducing agent. The bus-bar connections and the arrangements of electrodes around the crucible in the Swedish type of electric shaft furnace FIGURE 13.-Bus-bar connections and arrangement of electrodes in Swedish type of electric shaft furnace. are shown in figure 13. The Uddeholm Co., at Hagfors, Sweden, is adding a third furnace to its electric furnace installation at Hagfors, and is constructing three similar furnaces at its Nykroppa works. It will probably substitute electric smelting altogether for its blast furnaces. The Stora Kopparbergs Bergslags is also preparing to introduce electric smelting on a large scale, and according to reports, other iron works in Norway and Sweden will follow its example. The cost of construction of the plant at Trollhättan is given below: Cost of construction of electric iron smelting plant at Trollhättan, Sweden. Preparation of site: Excavations, foundations, and fencing; narrow-gage tram lines, 3 turntables, and 5 trucks; 1 5-ton and 1 40-ton weigh bridge; Buildings: Furnace house with 2 windlasses and inclined tracks $14, 735. 00 Office, laboratory, and storeroom_ General storehouse__. Furnaces: 1 furnace for 2,500 horsepower (ironwork, castings, masonry, etc.)- 6, 032. 51 13, 117. 12 13, 781. 70 $10, 727. 42 24, 728:62 Electric equipment_ Bars and cables to the electrodes__ Exhaust fan for furnace gases- Pump and motor (for water)__ Water tank and pipes.__. 3, 381.93 826.94 1, 452. 64 942. 19 33, 502. 52 1, 010. 80 Crusher Electric lighting and motors: Transformers and wiring; 4 motors for exhaust fan, crusher, ore and coal elevators, and conveyors; direct-current transformers_. Laboratory equipment--- Instruments (self-recording) for measuring temperature, pressure, 5, 157. 24 889.88 1, 894. 25 1, 212. 30 675.00 5,637.86 7,500.00 92, 935. 89 CHEMISTRY OF THE REDUCTION OF IRON ORES IN THE ELECTRIC FURNACE. As is well known, the reduction of iron ores is effected by heating the ore and the reducing agent to such a temperature that a reaction. takes place between the oxygen and the reducing agent, which usually is some form of carbon. For the most part, iron ores at the present time are reduced in the blast furnace. In order to get a clear idea of the difference between a blast furnace and an electric-reduction furnace, let us first briefly note how the work of reduction and subsequent melting of the reduced iron and fluxes is brought about in the blast furnace. The charge is composed of ore, oxide of iron plus gangue materials, fluxes for combining with the gangue materials of the ore in such proportions determined by analysis and calculation as will make a fusible slag, and the fuel, which is needed to provide heat and also acts as a reducing agent. This fuel is generally coke or charcoal and contains practically no volatile matter. In the tuyère zone the larger part of the fixed carbon passes to CO, but whether it first passes through the CO2 state and is then reduced to CO by incandescent carbon is still a debated subject. At any rate CO is the final product, and in its production a high enough temperature is produced to melt down the previously reduced iron and to form a fluid slag from the fluxes and gangue materials present. At this point the difference between the blast furnace and an electric-reduction furnace should be noted, for in the electric furnace the heat necessary for melting the charge is furnished by the electric current. In the blast furnace the quantity of fuel necessary to produce the smelting temperature, and not the amount necessary for the reduction of the oxides, determines the amount of fuel that is used. In the electric furnace the amount of electric energy necessary for producing the smelting temperature is used in place of coke or charcoal; and as only one-third or less of the coke that is used in blast-furnace work is needed in the electric furnace for the reduction of the iron oxides, it is possible to produce in the electric furnace three times as much iron with 1 ton of coke or charcoal as in a blast furnace. Such being the case, the amount of electrical energy and carbon that are necessary to produce 1 ton of pig iron in an electric furnace may now be considered. C This problem has been carefully worked out by Yngstrom, Richards, and others, and hence it is necessary in this connection to give only the essential details in such a calculation. 2 In determining the amount of electrical energy necessary to produce 1 ton of pig iron from a given ore the amount of heat absorbed by the following processes must be calculated: The reduction of the iron oxides to iron; the reduction of the SiO, to Si; the melting and superheating of n kilograms of iron; the melting and superheating of n kilograms of slag; the heating of a kilograms of CO, plus CO to whatever temperature it is determined that they should escape from the shaft. a Belden, A. W., Foundry-cupola gases and temperatures: Bull. 54, Bureau of Mines, 1913, pp. 14-19. Yngstrom, Lars, Electric production of iron from iron ore at Domnarfvet, Sweden: Engineer (London), Feb. 25, 1910, p. 206. c Richards, J. W., Metallurgical calculations, pt. 2, 1907, p. 403. From the number of calories absorbed in this manner may be deducted the heat that would be developed by the combustion of the x kilograms of carbon needed for the reduction. The calculation of the amount of carbon needed for the reduction of the oxide is not an easy matter, for, as has been pointed out by Richards, in calcuiating the amount of carbon necessary to reduce the oxide of iron there is no way of knowing what proportion of carbon will form CO and what proportion CO,. The amount of each that will be formed. depends upon the temperature, CO being almost the only product that is formed at a high temperature, whereas CO2 is formed in increasing quantities as the temperature decreases, the reactions being represented as follows: From these equations we note that if the gaseous product of the reduction is all CO, not more than one-third as much carbon is required for the reduction as is required in the case of blast-furnace practice. Yngstrom in his calculations assumes that when carbon combines with oxygen in the electric furnace a mixture is formed that contains 30 per cent (by volume) of CO2, and on this basis he calculates the carbon necessary for the reduction of the iron oxides present and the subsequent development of heat by the formation of CO and CO, by the following equation: 14Fe2O3+30C = 12CO2+18CO+28Fe In his computations he uses the following theoretical values: To reduce 1 kg. of Fe from Fe3O4 there is required 1,650 calories. 1 kg. of pig iron requries for melting and superheating 280 calories. As a result of his calculations it is seen that for the production of ton of pig iron containing 3 per cent of carbon, 1 per cent of silicon, 96 per cent of iron, and traces of manganese, phosphorus, and sulphur, made from a charge containing 60 per cent iron in the form of FeО and with the escaping gases containing 30 per cent of CO2, 248 kg. of carbon, or 292 kg. of coke, and 1,460 kilowatt-hours would be required, which would correspond to 4.4 tons of pig iron per horsepower year of 365 days. 4 The production of this amount of iron per kilowatt-year has not been attained in actual practice, for, according to the latest reports from Trollhättan, only 3.94 tons is produced per kilowatt-year. At this point it may be interesting to compare the number of heat |