Home >> Encyclopedia-britannica-volume-8-part-1-edward-extract >> Einsiedeln to Elephant Head >> Electrometallurgy

Electrometallurgy

Loading


ELECTROMETALLURGY. As different methods of pro ducing electric currents in increasing magnitude were discovered, one of the earliest applications made, in every case, was the study of their effects on chemical change both by electrolytic action and, on account of the high temperatures attainable, by their electro thermal effects. Thus Davy in 181o, using a voltaic pile of i,000 plates, isolated the alkali metals and aluminium by electrolysis, and conducted experiments on the fusion of iron wire. Pepys, in 1815, by means of an electrically heated iron rod demonstrated the cementation of iron by absorption of carbon. The experiment was conducted by bending a piece of soft iron, cutting a longi tudinal groove at the bend, and filling with diamond dust. The wire was then mounted between wider metal poles, and, after covering with talc to protect from oxidation, was heated by the passage of an electric current when the diamond was absorbed and the iron converted into steel.

Despretz in 1849 carried out experiments in which high tem peratures produced electrically were utilized. The earlier work forms the subject of a paper entitled "The Fusion and Volatili zation of some Refractory Bodies : Notes on some experiments carried out with the triple acid of the voltaic pile, the sun, and the blow-pipe" (Comptes Rendus, 1849, xxviii. 755). In later ex perifnents an apparatus is described which consists of a tube of sugar-charcoal about 4 in. wide and 2 in. long closed by two char coal plugs which, with its contents, was raised to a high tempera ture by the passage of an electric current.

H. Wilde, in 1886, by means of a current from one of the earliest types of magneto-electric machines, was able to melt a bar of platinum 6 mm. thick and 2 ft. long. William von Siemens, in 1878, designed several forms of arc furnaces which contained all the important features of modern types. One, referred to in many textbooks, has a carbon electrode adjusted to the distance neces sary to maintain an arc with the desired current by a magnetic solenoid operated by the main current passing through a sur rounding coil. The crucible is insulated by surrounding it with carbon packed in a wider container. In place of a carbon electrode an adjustable water-cooled iron electrode, was also employed, as being more effective. One of the earliest electric furnace processes to be brought into commercial operation was that of the Cowles Brothers which was installed at Milton, Staffordshire, in 1886 (Industries, 1888, vol. v., 237). The process consisted of the reduction of alumina by carbon in the presence of iron or other metals, resulting in the formation of aluminium alloys. The fur naces were rectangular in shape, constructed of fireclay, and pro vided at either end with an inclined cast-iron pipe, through which the electrode was introduced. Heating was brought about by the formation of an arc between the carbon electrodes.

Each electrode consisted of a bundle of from seven to nine carbons (each 21 in. diam.) held together by a cylindrical head of metal which was connected to the flexible cables leading in the current. The charge consisted of a mixture of corundum metal and charcoal. A current of 5,000 amp. at 6o volts was passed and a period of one and a half hours was required for the reduction of each charge. The fur nace was covered with a cast-iron case provided with a hole in the centre, through which the gases generated during the process escaped. The reduced alum inium and alloying metal were volatilized, and the ascending vapours became con densed in the upper and cooler layers of charcoal. Combination occurred here and the liquid alloy flowed to the base of the furnace and was run off through a tap hole. The daily output at Milton amounted to about 20 cwt. of ferro aluminium or aluminium bronze contain ing 15 to 17% aluminium. The current was obtained from a Crompton dynamo of 30o kw. capacity, and its capacity con stituted at that time a record in electric generators.

In 1887 Heroult, in France, and Hall in America, devised an electrolytic process for the production of aluminium and the Cowles process suspended operation.

Moissan, commencing in 1892, carried out a series of experi mental researches with the high temperatures obtained by means of an electric arc furnace. Metals such as chromium tungsten, molybdenum, uranium and titanium were thus obtained in a fused state for the first time. The method adopted, as illustrated in fig. I, consisted in placing a powerful arc in a cavity of minimum size in a limestone block and at a certain distance above the substance to be heated. In this way, actual contact with the carbon vapour from the arc is avoided and at the same time the thermal action of the current is separated from any electrolytic effect. The cur rent generally used by Moissan was about 450 amp. at 6o volts.

This type of furnace, which is still suitable for many laboratory experiments, consists of two slabs of lime carefully cut and posed. The lower slab has a long groove in which the electrodes rest, and in the middle is a small cavity which serves to hold a small carbon crucible containing the substance to be heated. The electrodes are easily rendered movable by means of two able supports, or better, by using two sliders which rest on a plate. The freedom of motion of the electrodes enables an ment of the length of the arc to be conveniently made. The arc at first is less than i cm. long, but towards the end of the experi ment the length usually increases to from 2 to 2.5 centimetres. If the furnace is filled with a good conducting metallic vapour (e.g., aluminium), the electrodes may be 5 to 6 cm. apart. The length of the arc will thus be regulated so that an approximately constant resistance can be maintained.

Electric Smelting of Iron Ore and Production of Steel.— The principle of the Moissan furnace was applied by Stassano in Turin in 1903 for the smelting of iron ore and production of steel.

A type of furnace arranged to rotate about an inclined axis which was finally adopted is illustrated in fig. 2, which shows a vertical section through two electrodes. Three equidistantly-spaced elec trodes are employed in the larger units in order to utilize three phase current. The electrodes are surrounded by water-cooled jackets. Gases from the reaction pass out through a vent at the head of the en closure and the furnace is provided with a charging hopper and tapping spout not shown in the diagram. The furnace was originally employed for the smelting of magnetite and haematite whereby malle able iron was produced directly, and later for the production of steel from cast-iron and turnings.

Large-scale experiments on the electric smelting of iron ore and the production of ferro alloys were undertaken at Sault Ste. Marie, Canada, by a commission appointed by the Canadian Government in 1903. The furnace consisted, as seen in fig. 3, of an iron casing, bolted to a base plate of cast-iron. Rods of iron were then cast into the plate to secure a good contact with the carbon paste rammed into the lower part of the fur nace and lining the bottom and sides of the crucible. One end of the electrode was planed to fit into a steel shoe and was held tight by means of wedges. A pipe was put in the electrode holder, through which a current of air was circu lated on to the holder. The power applied was 200 kw. single phase alternating current at a mean value of 5,000 amp. at 40 volts. By the use of charcoal as reducing agent it was shown that magnetic ores could be smelted efficiently in this form of electric furnace, though the ore is too high in sulphur to be treated by the usual blast-furnace process. It was also shown that a ferro-nickel pig could be produced practically free from sulphur and of fine quality from roasted nickeliferous pyrrhotite.

For the production of steel the main advantages which have been gained by the employment of electric furnace operation is that a product can be obtained which is equal to the best grade of crucible steel at a considerably lower cost and on a much larger scale than is possible with the crucible process. The advantage is mainly due to the higher temperature of operation which is pos sible in the electric furnace and this gives more latitude in the composition of the slags which are employed for the removal of sulphur and phosphorus. Access of nitrogen to the metal which has a deleterious action, can moreover be more completely ex cluded by the electric furnace, and manganese and other metals employed in the purification can be added with less loss through oxidation.

The three main types of elec tric steel furnaces now in use are ( i) the direct arc type, ( 2) the radiation arc type and (3 ) the induction furnace.

Direct Arc Furnace.—The main example of this type is the Heroult furnace, the principle of which is illustrated in the diagram in fig. 4.

This furnace is operated by forming arcs between the surface of the charge and carbon electrodes which are suspended vertically above the furnace and connected to the poles of an alternating current supply.

Formerly only single-phase current was used together with two electrodes joined in series, but the larger furnaces now in use are operated with three-phase current and contain three electrodes. In all cases when the furnace charge is molten, the current, as shown in fig. 4 passes in the form of an arc from the electrode to the surface of the slag, through the slag to the metal, and hori zontally through the surface layers of the metal to the adjacent electrode where a second arc is formed. A zone of very high tem perature is thus produced in the slag in the neighbourhood of the electrodes, and rapidly brings about the chemical reactions in volved in the purification of the iron and its conversion into steel.

These furnaces are all fitted with a tilting mechanism and a spout so as to enable the fused slag or metal to be poured.

A variation in the design of the Heroult furnace is shown in the Snyder furnace, which as originally designed for single-phase cur rent is illustrated in fig. 5.

The main features of the furnace are a hinged roof which en ables the roof, electrodes and all overhead gear to be removed by a pivoted mechanism. This de- • vice facilitates charging and en ables mechanical charging to be employed. With the single-phase type, the current enters by the cables shown at the base which lead to the central carbon elec trode at the top and after form ing an arc between the electrode and the surface of the bath, the current leaves the furnace through a water-cooled metal contact at the base. A number of advantages are obtained by the use of single-phase units chiefly due to the central distribution of the current. Single-phase units are, however, limited to sizes of a capacity of about one ton. For larger furnaces a system has consequently been adopted which makes use of three-phase current, two phases being con nected to two top or movable carbon electrodes, while the third is connected to the metal electrode through the hearth.

Radiation Arc Furnace.

The chief example of this type of furnace is that of Stassano, which is described above. A modifica tion of this is in the Rennerfeld arc furnace. In this system the electrodes are arranged so that the arc is deflected downwards away from the roof and on to the charge in the furnace. As shown in fig. 6 the two side electrodes are arranged slightly inclined to the horizontal, while the third electrode is suspended vertically in the centre of the furnace in a position equidistant from the ends of the side electrodes. By means of the so-called Scott-connections the electrodes are joined to a three-phase current supply in such a way that the middle or vertical electrode forms the common re turn for the two phases, and thus carries about 40% more current than the side electrodes. The electromagnetic action of the current arranged in this way causes the flame of the arc to strike downwards and spread over the surface of the thinned slag. The furnace is supported on trunnions and arranged to tilt.

The usual size of a Rennerfeld furnace is three tons, though 12 ton and larger units have been designed. The current supply for a three-ton furnace is usually amp. X Ioo volts X 2 = 75o k.v.a.

Induction Furnaces.---In

the usual type of induction furnace the crucible holding the metal un der treatment is in the form of a ring or loop and constitutes the single-turn secondary of a transformer. The iron core or yoke of the transformer passes through and interlinks the primary coil with the single-turn secondary and thus provides the nec essary close coupling between the two. The first example of an induction furnace was designed by Ferranti (Eng. Pat. 700 of 1887), while the first practical type to be brought into industrial use was constructed by Colby in America in 1890. This con tained a double-yoked magnetic core and a primary winding which in an early type was arranged outside the metal bath as shown in fig. 7, but in a later development the primary winding was placed in the centre of the metal ring, as in the Kjellin furnace. A unit of 13o k.v.a. capacity holding about 190 lb. of steel was installed in Philadelphia about 1906 for the manu facture of high-grade steels. The primary consisted of 28 turns of copper tube cooled by water circulation and arranged to utilize S40 amp. at 24o volts.

The main advantages obtained by induction furnaces are the absence of contamination of the metal by contact with carbon electrodes ; better exclusion of furnace gases from the metal; convenience in introduction of current in that a high potential circuit can be supplied without the use of transformers or copper cables of large cross-section; fluctuations of the current do not occur as in other types, and a much steadier load is therefore offered to the power supply. Tappings are obtained which are very large com pared with the contents of a crucible, and the steel produced is stated to be of better quality than that given in crucibles from the same raw material. The alter nating current, through its electromag netic action, causes an efficient circulation of the metal. One disadvantage, how ever, in the original type of furnace, containing a ring of metal of uniform section, is that the temperature could not be taken to the same high degree as in the direct arc type, and the powerful reducing action which is accompanied by the formation of calcium carbide is thus not secured. On account of this lower temperature the furnace could not be applied for carrying out any extensive refining, but rather a melting which required the use of pure materials in order to yield a high grade of steel. However, in more recent types of the induction furnace attempts are made to overcome this limitation by arranging a narrowing of the channel of a portion of the circuit, or by other means causing an increased resistance and higher temperature to be obtained locally. A good circulation of the metal is in all cases ensured in the induction type of furnace through the operation of electromagnetic forces.

The Kjellin furnace was introduced in 1899, and as seen in figs.

8 and 9 contained either a single or double yoke core with one primary winding in the centre. The Kjellin furnace was erected at Gysinge in Sweden and applied to the production of steel in 1900. Larger units of this furnace were later installed at the Roch ling steel works at Volklingen, Essen, Germany. Furnaces of 75o kw. capacity were operated with single phase current of five periods at 4,500 to 4,900 volts. The circular crucible of this furnace, which had a capacity of 85 tons, was built up from masonry and lined with a suitable basic material as used in the Bessemer converters or Siemens Martin furnace. During operation the annular crucible is roofed over by cover ing with segmental iron plates. The fur nace is arranged to tilt and is provided with a pouring spout for emptying. The Kjellin furnace has proved to be very suitable for making the highest class of steel from pure raw materials, and has been able to compete with the crucible process for this purpose, even when the power is generated from coal.

The main disadvantages of the Kjellin furnace are the incon venient shape and narrow width of the annular crucible which pre vents the hearth from being readily accessible and easily surveyed, and the specially low frequency of the alternating current required in the larger furnaces, which involves the construction of special expensive generators. The above disadvantages have for the most part been overcome in the modified type of induction furnace devised by Rochling and Rodenhauser, and this has now generally replaced the earlier type of Kjellin furnace.

Rochling-Rodenhauser Furnace.

The Rochling-Roden hauser furnace was first brought into operation in Germany in 1906 for the final refining of fluid steel which had first been treated in the Bessemer converter. The chief features which dis tinguish it from the Colby-Kjellin type is that the crucible, in stead of being in the form of a plain ring, is arranged in the shape of a figure 8, with a wide central space. Heating is effected by a combination of two systems. Firstly, by induction currents, whereby the furnace acts like two combined ordinary induction furnaces, and secondly, by an auxiliary circuit in which by means of a separate winding around the primary coil an induced current is applied to the metal bath through electrodes consisting of plates of soft cast-steel covered by grids of refractory oxides which be come conducting by electrolytic action at high temperature. A diagram of the furnace as designed for single-phase current is shown in fig. 10 in horizontal plan. Like the ordinary induction furnace, this furnace is essentially a transformer with a single primary winding A around both cores H of the transformer. The secondaries are two in number, one is the molten bath C in the form of a figure 8, the channel D between the two cores being comparatively broad and the other consists of extra windings B surrounding the primary coils and conducting currents to the me tallic plates E which are inserted in the furnace walls, and are raised to a high temperature by the passage of the current to the metal bath. During operation the molten charge is kept in good circulation by the electromagnetic influence of the current. The whole furnace is built as a tilting furnace by mounting on rollers, and in general design is similar to the Siemens-Martin open-hearth furnace.

The disadvantage attending the generation of single-phase cur rent has been overcome by a three-phase induction furnace, which in units of from 3 to 15 tons capacity is operated with a fre quency of 5o periods, and thus makes possible the use of standard three-phase generators.

The hearth shown in the I.5-ton furnaces is i ft. 7 in. wide, and 4 ft. 9 in. long. The three transformer cores are surrounded by heating channels. At the places in which two such channels enter into the main hearth, special electrode plates are arranged with great care in the general form of rectangular black slots built in the furnace walls. These electrodes are embedded in the furnace wall and separated from the fused charge by a refractory wall which becomes conducting when heated. Each of the three transformer cores is provided with a primary winding, and above each primary winding a secondary winding is also arranged. While one end of the latter is connected to a bus-bar, the other ends of the three windings are connected to the three electrodes.

At the Rochling steel works in Germany, the procedure most generally adopted is a "triplex" treatment, in which the steel is first "blown" in a convertor, further purified in a basic open hearth, and then subjected to a final refining treatment in the electric furnace. At present the largest application of electric steel furnaces is in combination with the older fuel-heated fur naces to give the final refining to molten steel which has undergone a preliminary treatment in the convertor or open hearth furnace.

Electrolytic Processes with Fused Electrolytes.—The procedure in all in stances in the production of metals from fused components consists in utilizing the property these compounds possess of be coming conductors at high temperatures. Electrolytic dissociation is thus brought about as in aqueous solutions at ordinary temperatures. By the application of a suitable potential, the cation is then sepa rated in the metallic condition on the surface of the cathode. During the elec trolysis, the necessary high temperature may be maintained through heat generated by the passage of the current.

Aluminium.

In all methods which have so far received prac tical application, the preparation of aluminium is brought about by dissolving alumina in the halogen compounds of the alkalis, alkaline earths or of aluminium, and separating the aluminium by electrolysis.

The raw materials employed consist almost exclusively of bauxite, which is a hydrate of alumina of the formula and cryolite which has the composition An addi tion is generally made to the bath of an excess of aluminium fluoride, calcium fluoride, sodium chloride or other compounds in order to reduce the melting point of the electrolyte and diminish its density so as to facilitate the sinking of the aluminium to the bottom of the bath.

It is necessary to submit the alumina employed to a careful purification and to prepare it in a physical condition which enables solution to take place rapidly in the molten bath. During the elec trolysis, aluminium is liberated at the cathode and oxygen at the anode which reacts with the carbon to form carbon monoxide. At the same time small amounts of fluorine are sometimes liberated, the decomposition voltage of the fluoride being only slightly above that of alumina. It is necessary to adjust any loss of fluorine from the bath by the addition of aluminium fluoride. Alumina is re placed as the electrolysis proceeds and the composition of the bath remains constant. This method, based on the power of electricity to split metal salts into their components had been discovered many years before it was possible to apply it commercially. Early attempts to electrolize aqueous solutions had failed because the aluminium produced at once recombined with water. Robert Bun sen, 1854, passed a current through a molten mixture of aluminium chloride and common salt, splitting the former into chlorine, which escaped, and aluminium which he collected as metallic beads.

Heroult Process.—In this process electrolysis was at first at tempted by the use of alumina alone as the charge. In this case the high temperature necessary for the reaction resulted, when using a carbon cathode, in the formation of aluminium carbide. Cathodes of copper and other metals were accordingly applied and aluminium alloys prepared. Later, by the use of cryolite as a flux or solvent, the separation of aluminium metal was found pos sible when using a cathode either of iron or carbon. The anode, in this apparatus, consisted of a bundle of carbon poles fastened together by copper bands so as to form a block i m. Ione and 0.2 sq.m. section. In the more modern type of furnace, such as has been installed at Neuhausen, the anode consists of a bundle of rectangular plates of carbon which are cemented together by means of a mixture of carbon with tar and molasses or glucose solution. The cathode consists of a carbon block contained in an enclosing metal case. The roof of the furnace is covered with graphite plates k, furnished with openings for the admission of the anode in the centre and the charge at C. The tem perature of the bath is from 800° to I,000° and the most favourable compo sition a mixture of cryolite with 1 o% alumina. The aluminium collects on the bed of the cell from whence it can be re moved by tapping. The specific gravity of the molten metal is 2.54 while that of the fused bath is Sodium.—The industrial preparation of sodium was developed by Castner, through investigations begun in 1888, on the electrolysis of fused sodium hydrox ide. The apparatus designed by Castner consists, as shown in fig. II, of a cast iron case A of about 14 in. diam., the upper part being 24 in. high and provided at the base with an extension tube B of 3 in. diam. and 32 in. long. The cathode H consists of a copper bar secured at the base of the surrounding case by a wooden plug, in the space above which fused sodium hydrate is filled and allowed to solidify. The electrolyte in the space above can be heated by means of the gas-ring burner G.

Directly above the cathode is suspended an iron cylinder, closed by the lid N which is provided with an opening for the escape of hydrogen, while from the lower part, a wire gauze cylinder extends so as to form a diaphragm between the elec trodes and prevent the sodium after separation from diffusing to the anode. The anode E was originally made of an alloy of nickel and silver which was found resistant against corrosion.

The sodium liberated at the cathode H together with the hydrogen rises into the cylinder N, and is periodically removed by means of a perforated ladle which allows its separation from the fused electrolyte to take place.

Calcium.—In a process in operation at the Bitterfeld electro chemical works in Germany calcium is prepared from the fused chloride by means of an electrical process in which a cathode, which is kept cold, is, at its base so arranged as to make contact with the surface of an electrolyte of fused calcium chloride. On passing the current, calcium is separated in globular form, adheres to the cathode and solidifies. The electrode and adhering calcium are slowly raised from the bath so that a cylindrical rod of calcium is then gradually built up, and by means of the thin coating of chloride which adheres to it, is protected against any oxidation.

BIBLIOGRAPHY.-Rodenhauser, Schoenawa, and von Baur, Electric Bibliography.-Rodenhauser, Schoenawa, and von Baur, Electric Furnaces in the Iron and Steel Industry (1920) ; C. C. Gow, The Electro-Metallurgy of Steel (1921) ; J. N. Pring, The Electric Furnace (London, 1921).

furnace, current, metal, electrodes, steel, carbon and aluminium