Metallurgy is the art of extracting metals from their ores and bringing them into that state of purity which is necessary for their industrial application. Electro-metallurgy is that branch of the metallurgic art in which the agency of electricity is employed. We would then define electro-metallurgy as the art of extracting metals from their ores or of refining them, on a commercial scale, by the agency of the electric current.
Before going into the further classification of this subject, let us inquire into its history. We can hardly realize, in this age of electrical wonders, that it is less than a century since Volta discovered current electricity. Messrs. Nicholson and Carlisle first made known, in 1800, the chemical powers of an electric current, that it would decompose water and certain saline solutions. Kissinger and Berzelius, in 1S03, and Davy, in 1807, enlarged upon this subject, the latter especially achieving renown by decomposing the fixed alkalies by the electric current and first isolating the alkaline metals. Faraday was the first to determine accurately the laws governing the electric deposition of metals from solution, a phenomenon to which he gave the term of Electrolysis. It was thus known, early in this century, that the electric current would, if properly applied, deposit metals from solutions of their salts in water.
In 1836, De la Rue, working with Daniell's recently-devised constant-current battery, discovered that when a copper plate was electrically coated with a sheet of metallic copper, and the sheet stripped from the plate, every scratch in the plate had its counterpart in the sheet which was deposited on it. This discovery gave rise to the very useful art of galvano-plasty, by which fac-simile impressions are so easily obtained, and which has its widest application in the modern methods of electrotyping.
In 1838, Messrs. Elkington and Barratt obtained patents for processes of electrically depositing gold, silver, platinum and zinc upon articles, to serve as a protective plating. These were the first practical electric-plating processes, which have expanded to such a wonderful degree at the present day.
However, there are many metals which cannot be electrically deposited from aqueous solution, and Professor Bunsen, in 1853, devised an extremely ingenious method of treating such cases. He was experimenting on the electrolytic production of magnesium, and instead of a solution in water he simply fused anhydrous magnesium chloride by heat, in a crucible, and used the molten salt as the liquid bath or electrolyte. This device was successful, and opened a new field for investigation of electric action. By using a similar method, in 1855, H. Saint-Claire Deville succeeded in first producing a bar or stick of aluminium.
As far back as 1847, Maximilian, Duke of Leuchtenberg, proved that when impure copper containing precious metals is used as an anode in a copper sulphate solution, the copper deposited on the cathode is of exceptional purity, while the precious metals are left undissolved in a concentrated form ready for further treatment. He foresaw that a day might come when this discovery would be of great importance.
And now, we may well ask, what obstacle prevented the inauguration of electro-metallurgic processes? The electric current deposits many metals from aqueous solution in a very pure state; metals not yielding to this method can be obtained by electrolysis of a molten bath of their salts, and an excellent process for refining copper and extracting its gold and silver is worked out; yet the real art of electro-metallurgy, as I have defined its meaning, was non-existent. The cause is not hard to find. Until the introduction of Wilde's magneto-electric machine, in 1865, all electrolytic operations were conducted with the current from batteries, and it needs no explanation to see that the application of this electric process to the extraction of metals from their ores, or refining them, was a commercial impossibility. The introduction of Wilde's machine may be taken as the starting-point of all our commercial electro-metallurgic success, for it furnished large electric currents at a cost many times less than the battery, and rendered financially possible several methods of electrolysis. Electro-metallurgy, as a practiced art, dates from 1865.
It will be readily seen that 1 exclude from the meaning or scope of the term electro-metallurgy the processes of electrotyping, electroplating, and all electric processes which are not metallurgic, in the true sense of that word. I confine the term simply to the extraction of metals from their ores and their refining on a commercial scale. Many of the so-called treatises on electro-metallurgy are really treatises on electro-plating, etc., dismissing the metallurgic side of the question in probably one or two short chapters. The distinction which I have made is a real one, and has been fully appreciated by Dr. Gore, who, in the preface to his recently published "Electric Separation of Metals," really a work on electro-metallurgy, says: "This volume is written to supply a want. No book entirely devoted to the electrolytic separation and refining of metals exists at present (1890) in any language; those hitherto written on the subject of electro-metallurgy are more or less devoted to electro-plating, the molding or copying of works of art, etc., by electro-deposition."
A division of the subject of electro-metallurgy might be made into the science and the art; that is, the theoretic principles of electro-deposition on which the art is based, and the art itself, of practically applying those principles. The theoretical principles underlying the art are simply those of electrolysis, common to the whole subject of electro-deposition; their practical application to metallurgic operations constitutes the art of electro-metallurgy. The principles were mostly well known prior to 1865, but their commercial application dates from that time.
Electro-metallurgy falls naturally into two divisions:
I. Extraction of metals from their ores by electricity.
II. Refining of metals by electricity.
The latter division was the first to be put into practical operation. By its nature it must be an adjunct to some other metallurgic operation for reducing the ore to metal, and constitutes only a subsidiary part of some ordinary metallurgic process. We will, therefore, consider this latter division first, in order to clear the ground for a discussion of processes of the first division, the true, independent electro-metallurgic processes.
REFINING OF METALS BY ELECTRICITY.
In 1865, immediately on the introduction of Wilde's electromagnetic machines, Mr. Elkington of Birmingham, England, started a plant for refining copper which has been in practical operation ever since. It has been already explained that the possibility of this method had been proven many years before, so that Mr. Elkington's enterprise consisted essentially in starting on a large commercial scale what had been done on a small scale, with the battery, almost twenty years before. The plant was commercially successful, and was the father of the many large copper-refining plants now scattered through Europe and America.
The rationale of the electric copper-refining is as follows: The metallurgy of copper has always been a rather complicated affair. By one or two smeltings the ore can readily be reduced to an impure copper, but the heaviest part of the work has still to be done in refining this to pure copper. Especially is the question made difficult when the impure copper contains silver, which is frequently the case. In this event, the only practical way to get out the precious metal was to dissolve up the entire mass of copper in acid and separate the silver chemically, by precipitation. It was at this point that the electric method of refining stepped in. It took the impure copper, produced by ordinary dry smelting from the ores, and converted this at one operation into the very purest copper; meanwhile, at the same time, extracting all the precious metals. It is thus seen that a very wide field was open to this art of refining, and that the financial side of the question was materially assisted by the high price commanded by the superior quality of copper produced. The operation of refining may be briefly described as follows: The impure copper is cast into plates about 18 inches square and ½ inch thick, with lugs projecting from the corners at one end. These are connected with the positive pole of the electric generator, and hung at intervals of four to six inches in a trough filled with solution of sulphate of copper. Between these are hung thin sheets of pure copper of similar shape, connected with the other pole of the dynamo, on which the pure copper is deposited. To, ensure success, close attention has to be given to the concentration of the bath, its temperature, and that it has free circulation. When working properly, only pure copper will be transferred from the anodes to the thin sheet cathodes. The impurities in the copper behave as follows : The iron is dissolved, goes into solution as sulphate and accumulates in the bath, not being deposited with the copper. When the bath contains a certain amount of iron, it can be purified by being run out, concentrated and crystallized, the iron sulphate crystallizing out first. Bismuth, tin and arsenic also pass into solution, but need not be deposited with the copper if the manager attends carefully to the various details. Gold, silver, platimim, cuprous oxide and cvpric sulphide, with most of the bismuth and some tin and arsenic, remain undissolved and fall as mud to the bottom of the bath. This residue, therefore, contains all the precious metals present, in a very concentrated form suitable for further treatment by ordinary cupellation methods. The deposited copper ought to be very nearly chemically pure.
The electric refining of copper has developed into an immense business. There are in operation twelve works in Germany, one in Italy, five in France, six in England, and six in the United States. Their annual production is many thousand tons, being a considerable proportion of the entire production of pure copper.
The only other metal to which electric refining has been applied on a commercial scale is lead. Metallic lead can be refined with much more ease, by ordinary furnace methods, than impure copper, yet it is a difficult matter to extract from it the precious metals. These are usually removed by the ancient method of cupellation, or by de-silverizing by zinc (Parke's process). Dr. Keith of New York devised, in 1878, an electric method of refining argentiferous lead, whereby the silver was extracted and a very pure lead obtained. The process was similar in most respects to the refining of copper, the lead anodes being, however, enclosed in thin muslin bags, which allowed the solution to pass through them but retained all insoluble residue. The solution consisted of acetate of soda in which sulphate of lead was dissolved. During the operation the iron and zinc present go into solution, but are not deposited with the lead. Antimony, arsenic, copper, silver and gold remain in the residue, which is treated in a similar manner to the residue from copper-refining. The baths were kept at about 100° F. This process was operated for some time on a large scale at Rome, N. Y,, but it was not sufficiently economical to compete with later improvements in other methods of de-silverizing bullion, and has been abandoned.
It is thus seen that the copper-refining is the only kind of electric refining in practical operation, and it is rendered possible by the difficulties of the methods of refining by the ordinary furnace processes. The chief items of expense in a refining plant are the large number of depositing vats needed for even a small-sized works, and interest on the large stock of metal locked up in the anodes and being in course of deposition. In one of the largest plants in the United States as much as 350 tons of copper are in course of treatment at one time, while the plant covers several acres. The relative slowness of deposition by electrolytic action is the chief difficulty with which all electric processes have to contend.
EXTRACTION OF METALS FROM THEIR ORES BY ELECTRICITY.
Metallurgically, there may be distinguished three distinct methods of applying the electric current to the extraction of metals from their compounds or ores:
I. Electro-deposition from aqueous solution.
II. Electro-deposition from a fused electrolyte.
III. Electro-thermal reduction.
I.
This heading includes a great number of electro-metallurgic processes. The method has been principally applied to the metallurgy of copper, silver, gold and zinc, and has developed on two distinct lines.
1st. Preparation of a solution of the metal and electrolysis of this by means of insoluble anodes, or anodes of a metal other than that being deposited.
2d. The use of anodes made of the metallic compound or ore, the solution being regenerated by the acid set free attacking these anodes and dissolving out the metal.
Operations of the first class are particularly applicable to the isolation of copper or zinc, which are easily brought into solution. Copper exists as sulphate in many mine-waters, which need only to be concentrated by evaporation to be ready for treatment. Many of the ores of copper and zinc can be treated so as to convert the metal into soluble sulphate. Thus, if copper pyrites is carefully burnt, most of the copper will form sulphate and can be washed out of the residue. Oxide or carbonate ores can easily be brought into solution by treatment with sulphuric acid. When a solution of copper sulphate thus formed is electrolysed, using sheet-iron anodes and sheet-copper cathodes, copper is deposited on the latter, while the anodes are dissolved and ferrous sulphate goes into solution. When all the copper has been deposited the solution can be evaporated to dryness, and the sulphate of iron regained and used over in the roasting operation, converting copper oxides into soluble sulphate. This method of electrolysis is often performed without the aid of an outside current, the copper and iron electrodes being simply connected by wires outside of the bath, the electricity generated by this galvanic couple being sufficient to electrolyse the solution and deposit the copper.
Zinc ores can be treated in a very similar manner. Letrange's process consists in taking zinc sulphide (blende), roasting it so as to convert as much as possible into sulphate, and leaching the product. Some zinc oxide will be formed, which, with unchanged zinc sulphide, will remain undissolved in the residue. The sulphate solution is electrolysed, using thin plates of zinc for cathodes and lead plates for anodes. The lead being insoluble in sulphuric acid is unattacked by the solution, which therefore gradually becomes more acid as the zinc is removed and the free sulphuric acid accumulates. When the zinc has been removed to a certain extent, the acid solution is run out and passed over the residues left from the leaching operation. The acid extracts the rest of the zinc from these, the solution being at the same time replenished with zinc and its acidity taken away. The extraction of zinc from the ore is thus practically complete, a result far from being reached by the ordinary zinc processes. Letrange's process has been worked in France, and the whole question of its applicability seems to be that of cost of metal, the process being industrially quite a success.
The use of metallic compounds for anodes affords a direct method of extracting metal from its ore in a minimum number of operations. These kind of electric processes were evolved from the copperrefining processes by a natural transition. In the latter, the impure copper used as anodes is dissolved by the acid set free by electrolysis, and thus the solution is regenerated. Marchesi, of Genoa, had the idea that since cuprous sulphide is attacked by free acid, that the impure copper might be replaced by copper matte from an earlier stage of the ordinary smelting processes, and thus one or more of the smelting operations be rendered unnecessary. He found the operation somewhat more difficult than with impure copper, yet he succeeded in making it practicable and it is now used on a large scale. The copper matte, sometimes obtained by only a single smelting operation direct from the ore, is cast into slabs, which are used in a copper sulphate solution exactly as if they were impure copper. The reactions are similar to those in refining impure copper; the precious metals particularly being thus very easily separated in the residues or mud.
In Luckow's zinc process, a bath is made of solution of zinc sulphate; the cathode is a thin sheet of pure zinc, and the anode is a mixture of zinc ore and coke, finely ground and well mixed together and held in an open-work case of wood. As zinc is removed from solution by the electrolytic action, the free acid attacks the anode, dissolving out the zinc ore. The carbon is placed in the anode to conduct the electricity; for, while impure copper and even copper matte conduct electricity, the zinc ore is practically a non-conductor of the current.
II.
About 1854, Bunsen made a new departure in electrolytic methods by subjecting a fused salt to the action of the current. He placed anhydrous magnesium chloride in a crucible, melted it at a gentle heat, and then dipped into it two electrodes of dense carbon, such as comes from gas-retorts. Magnesium was obtained at one electrode and chlorine gas at the other. Soon after, Deville electrolysed in a similar manner the anhydrous double chloride of aluminium and sodium. In such a bath, the current decomposes only the aluminium chloride, producing aluminium and chlorine; the sodium chloride being a more fixed compound is not decomposed if the current is properly regulated. In this way Deville made the first masses of aluminium which had ever been produced. He tried hard to perfect the process. He operated on a large scale, and tried to effect the regeneration of the bath and stop the evolution of chlorine at the anode by making the latter of a mixture of carbon and alumina, made by mixing the latter with pitch, moulding into shape and coking at a high heat. The electrolysis went on easily at 500° to 600° C, but the greatest difficulty met with was the disintegration of the electrodes, particularly the anode. A fundamental difficulty in the way of commercial success lay in the use of the battery to generate the current, an obstacle only overcome by the introduction of dynamo-machines many years later.
As early as 1879 it was proposed to produce aluminium by a method similar to Deville's, yet using dynamo-currents. In 1883, Dr. Richard Gratzel of Bremen obtained patents for a similar process, which was operated for about four years by the "Aluminium und Magnesium Fabrik" at Hemelingen, and many thousand kilos of aluminium made. Dr. Kleiner's process for producing aluminium consists in fusing the mineral cryolite (a double fluoride of aluminium and sodium) between two carbon electrodes which touch each other, producing a large electric arc. When the bath is well fused the electrodes are drawn apart, and the fused mineral is electrolyzed by the current into aluminium and fluorine (the sodium fluoride remaining unattacked if the current is properly regulated), while the bath is maintained in fusion by the heat generated by the passage of the current. Mr. Hall, whose process is being operated by the Pittsburgh Reduction Company, takes a bath of fused cryolite and stirs into it alumina until it is saturated. On passing an electric current through this bath, by carbon electrodes, the alumina, which is as it were dissolved in the cryolite, is the only compound attacked by the current, because it is the weakest of the three present, and thus the cryolite solvent remains untouched. When the alumina is all decomposed, the bath is regenerated by simply stirring in some more, and thus the operation is continuous. In practice only one electrode dips into the bath, the positive one, while the carbon lining of the iron pot holding the bath is made the negative. The bath is kept fluid by the heat generated by the current, which can be regulated by the distance between the positive carbons and the bottom of the pot. A plant of 500 horse-power is now manufacturing about six tons of aluminium a month by this process.
We cannot take the space even to name all the different devices used in electrolysing fused aluminium salts; one hundred pages would no more than suffice to describe them all.
This method of electrolysis has also been applied to the isolation of sodium. Jablochoff devised apparatus for decomposing sodium chloride (common salt), which consisted of a large pot in which the salt was fused, with arrangements to feed the bath as it was used up. Dipping into the salt were two electrodes of carbon, encased in tubes which also dipped under the surface of the bath. The products of electrolysis in this case were both vapors, the sodium vapor being led into a condenser, while the chlorine gas from the positive carbon was led into chambers where it was utilized for making bleaching powder.
III.
The electro-thermal processes are primarily dependent on the utilization of the enormous temperature of the electric arc, by interrupting a powerful current, by this agency bringing about chemical reactions which would not take place at temperatures attainable by any other means.
As far back as 1853, John Henry Johnson applied for a patent in England for "smelting iron and other ores" by electricity. He states that the metallic ores are to be ground, mixed with charcoal, and dropped between the poles of large electrodes, across which a voltaic arc is established. The ore thus treated separates into molten metal and slag, which are run out of the reduction chamber into an exterior vessel, where they may separate. In 1873 Werderman claimed the process of crushing the metallic ore, mixing with carbonaceous matter, heating to redness, and then raising the temperature to the point necessary for reduction by passing an electric current, led into the mass by terminal electrodes of carbon or other refractory conductor of electricity. Many advantages are thus gained by reduction in an enclosed space, where the atmosphere is perfectly reducing and the temperature almost unlimited. Such apparatus have been very appropriately called "electric furnaces." It will readily be recognized that such operations are expensive, and could not apply profitably to the production of the common metals. They have been used almost exclusively for reducing the most refractory ores.
Messrs. A. & E. H. Cowles, of Cleveland, Ohio, were the first to apply the electric furnace to the reduction of aluminium compounds on a commercial scale. Their type of furnace consists of a horizontal fire-brick-lined cavity, in which the mixture for reduction is placed, and through the ends of which pass two large carbon electrodes. The charge is carbon, alumina and a metal, usually granulated copper or iron; and the furnace is covered with a fire-clay slab. On passing the current from a 300 horse-power dynamo machine, and gradually drawing the electrodes apart, an interrupted arc of several feet in length is produced, and, at the temperature obtained, alumina melts, copper vaporizes, carbon crystallizes, and alumina is reduced by carbon. The product is an aluminium alloy. If the alloying metal is left out, no quantity of pure aluminium can be obtained, since it partly vaporizes and obstinately sticks in thin sheets to the lumps of carbon, refusing to run together. Heroult's furnace for reducing alumina works on the same principle, but is arranged differently. A large iron case is filled with carbon, a cavity hollowed out on top, and a large carbon electrode hung so as to dip into this cavity. On placing copper in the hole and lowering the carbon, the iron case being connected with the negative pole of the dynamo, the arc formed between the carbon rod and the copper soon melts the latter. Then alumina is thrown in, which is also liquefied by the arc. The operation then proceeds as if it were the simple electrolysis of a fused bath, the copper being the negative electrode and the alumina the electrolyte. Aluminium being set free, the copper absorbs it and forms aluminium bronze.
Several other forms of electric furnaces for reduction have been devised. In one, the two electrodes are made of a mixture of the ore and carbon, and when the arc is passed between their points the reduced material falls into a crucible beneath. In another, the carbon electrodes are made hollow, and the material to be reduced fed through the rods into the arc, where it is reduced. Some of these forms may yet be made serviceable, but the Cowles and Heroult furnaces are the only ones which have so far been successfully operated on a commercial scale.
CALCULATIONS.
Having briefly reviewed the various kinds of electro-metallurgical processes, we will note, by means of a few illustrations, the method of calculating the amount of power required to decompose compounds by electrolysis, and thus obtain means of estimating the percentage of useful effect in any process for which we have the necessary details.
An electric current has two factors—quantity and tension; the former measures its absolute amount, the latter its power of overcoming resistance. The unit of quantity is an ampere (measured on an ampere meter), the unit of tension is a volt (measured by a volt meter). Whether the affinities of a chemical compound will be overcome by a given current will depend on whether the current is of sufficient tension; when a current is of the required tension, the amount of chemical action performed will be proportional solely to the quantity of the current. The dynamic energy of an electric current is proportional to the product of its quantity by its tension; i. e. a current of one ampere at a tension of one volt has a definite mechanical value, and if this force is exerted in one second, the unit is called a Watt. This unit is at the foundation of all our subsequent calculations, and its absolute value is of first importance. The mean of the best experimental determinations make one Watt equal to 0.00024 calories of heat or to 0.1 kilogrammeter of work, and therefore nearly 1-750th of a horse-power. (French measures.)
As before stated, assuming that a current is of sufficient tension, the chemical work which it will do depends solely on its quantity. Some unit of chemical work per unit of electrical quantity would seem to be needed here, and this is given in the determination that when an electric current is decomposing water, each ampere passing sets free 0.000010352 gramme of hydrogen. The amount of oxygen liberated at the same time is necessarily eight times as great, and we can therefore pass directly to the law that the amounts of different elements liberated by a current of given quantity are proportional to their chemical equivalents. The amount of any element set free by one ampere is its electro-chemical equivalent, and is obtained by multiplying the electro-chemical equivalent of hydrogen by the chemical equivalent weight of the element.
The question of the tension necessary to decompose a compound follows immediately the statements of the two preceding paragraphs. We know from thermal data that to liberate 0.00001035 gramme of hydrogen from water requires an expenditure of energy represented by 0.00001035 X 34.162 = 0.000358 calories. But a current of one ampere at a tension of one volt is mechanically equivalent to only 0.00024 calories, and therefore the work being done in decomposing the water absolutely requires that the strength of the current shall be at least 0.000358/0.00024 = 1.49 volts. This is the absolute minimum electro-motive force which will operate the decomposition of water. The force of this reasoning may appear clearer if we were to assume, for argument's sake, that a current with a tension of 1 volt could decompose water. If so, every ampere passing represents one Watt of energy, or 0.00024 calories ; but it sets free 0.00001035 gramme of hydrogen, which if burnt back to water would set free 0.000358 calories. We have therefore created energy, being able to get one and a half times as much energy from the product as were expended. Of course we consider this an impossibility, and see at once that the one ampere must be propelled by a tension of at least i^ volts in order that its mechanical energy may be equal to the work which we know that it does. The tension practically required will always be greater than this calculated minimum, for the reason that the transfer resistance (the resistance which the current meets in passing from the electrodes into the electrolyte) and the conduction resistance (that met by the current in passing through the electrolyte) have to be overcome. These resistances do not result in the accomplishment of any chemical work, but cause a proportional part of the energy of the current to be converted into heat, which warms up the electrolyte. These latter resistances will vary principally with the temperature of the bath (as far as it affects the conductivity of the electrolyte) and the distance of the electrodes apart ; the transfer resistance is apt to be abnormally increased by the electrodes becoming coated over with a layer of non-conducting gas or liquid, a phenomenon called polarisation, and which we have time only to mention. Among all these variable resistances, that required for decomposition is the only one which is constant, and even it is not absolutely so, if critically examined, but decreases slightly with an increase in temperature of the bath.
A careful application of the principles just reviewed will enable us to discuss any of the problems presented in electro-metallurgy. In order for electrolysis to take place at all, it is necessary that the electrolyte be in the fluid state and that it be, when fluid, a conductor of electricity. These conditions being filled, and proper electrodes put in place, then the current passing between the electrodes must be of a certain minimum tension to accomplish decomposition. When the anode is soluble, and is gradually dissolved by the bath, the chemical heat of its solution may be set against the chemical work which the current does in decomposition, thus lessening the decomposition resistance. For instance, when, as in refining copper, metal is dissolved from the anode, the action at the anode is just the reverse of the decomposition taking place in the electrolyte, one off sets the other, and the only resistances to be overcome by the current are those of transference, conduction and polarisation. When, as in producing aluminium from cryolite, a metallic compound is broken up at the anode, such as alumina, its heat of formation will be the measure of the decomposition resistance, lessened, if the alumina is mixed with carbon, by the heat of union of the oxygen with carbon. We will conclude these calculations by analysing an example of each of the four kinds of electro-metallurgic processes, viz. refining, and the three divisions of electro-metallurgic processes proper.
Refining.—As before remarked, in refining, the decomposition resistance becomes nil, and the current has only to overcome the transfer resistances, etc. Since these latter are small, a very small electric current will refine a large weight of copper, if the baths are placed in series, the quantity deposited in each bath being proportional to the number of amperes of current. The amount of anode surface in each bath must be regulated according to the quantity of the current. It is found that the purest copper deposits, and in best condition for further handling, when about 5 ounces are deposited per square foot per 24 hours (1 ½ kilos per square meter). So, while the conduction resistance in each bath would be lessened, and the number of baths which could be used in a series increased by enlarging the anode surface, yet the total anode surface per bath will be regulated by the above principle.
At Elkington's works at Pembrey, near Swansea, an engine of 65 indicated horse-power ran a dynamo giving a current of 350 amperes at no volts, equal to (350 x 110)/750 = 51 1/8 electrical horse-power. (Efficiency of dynamo, 80 per cent.) This current was sent through a series of 200 vats, each with an anode surface of 44 square feet, with electrodes about two inches apart. The output was 4000 pounds in 24 hours. Looking into these figures, we see that a current of 350 amperes should deposit in 200 vats the following quantity of copper per second equal to 22.887 grammes per second, or 1980 kilos, equal to 4355 pounds per day. There was therefore an efficiency in this regard of 92 per cent. The number of volts absorbed by each bath was 110/200 = 55, and the density of the current was 350/44 = 8 amperes per square foot of anode surface. If a greater density had been used, in order to produce more copper with a given anode surface, the quality of the deposited copper would have suffered. The amount of copper deposited by this current was about 7 ounces per square foot of anode per day, an amount rather above the average.
Electric deposition from aqueous solution.—The case of deposition with insoluble anodes may be illustrated by an experiment made with Letrange's zinc process. With five vats in series, a current of 75 amperes at 13.05 volts, continued 4 ½ hours, deposited 1.475 kilogrammes of zinc. Let us first investigate the efficiency of the deposition. The chemical equivalent of zinc is 32.5, so that 0.00001035 X 32.5 X 75 = o.025263gramme should have been deposited in each vat per second, or 32.2 kilos in the 5 vats in 44 hours. The efficiency is therefore but 4.6 per cent. The reason for this very small return is to be found in considering the voltage required and used. The separation of the electro-chemical equivalent of zinc from zinc sulphate represents a thermal value of 0.000566 calories, and the voltage required to decompose zinc sulphate will therefore be this quantity divided by 0.00024 calories, or 2.359 volts. But it requires only 1.49 volts to decompose water, therefore we see why only 46 per cent of the current isolated zinc,—the rest was used up in decomposing the water of the bath into its elements. We see that 13.05/5 = 2.61 volts were actually used to each vat. If 6 vats had been used, the voltage for each would have been 2.17, and no zinc would have been deposited at all, but all the current wasted in decomposing water. If less than 5 vats had been used with this current, a larger proportion of deposited zinc would have been secured in each bath than 4.6 per cent of what the current might deposit, that is, more zinc and less hydrogen would have been produced in each bath, but the gain in this respect would not have made up in the sum-total for the dropping off in the number of baths. It is thus seen that the decomposition of a salt in solution, with insoluble anodes, is a very uneconomical proceeding if the decomposition resistance is large enough to involve the decomposition of the water.
If, on the other hand, a soluble anode is used, the decomposition resistance may be greatly decreased, as has been before explained. For instance, copper sulphate requires 1.25 volts for its decomposition; but if an iron anode is used, the solution of the iron sets up an auxiliary current of 2.01 volts. Therefore the iron helps the decomposition to such an extent that outside help is unnecessary, for about 0.76 volt more than is required for decomposition is furnished, enough to overcome all the other resistances of conduction, etc. We therefore see why, if the iron and copper cathode are simply connected by a wire outside the bath, the use of external currents is unnecessary. If an outside current were used in such a case it would, after supplying losses by conduction resistances, simply increase the voltage above 2.01, and thus begin to deposit iron with the copper.
Electro-deposition from a fused electrolyte.—Let us take for illustration the electrolysis of a bath of fused common salt, producing sodium. In some experiments described by Mr. Rogers, of Milwaukee, the voltage absorbed by the bath was 12 volts, varying with the temperature of the bath and the distance of the electrodes apart, and with a current of 70 amperes the amount of sodium obtained averaged 39 grammes per hour. This shows a yield of 0.000155 gramme of sodium per ampere per second. But the electrochemical equivalent of sodium, the amount which one ampere should liberate, is 0.000238 gramme (0.00001035 X 23), therefore we see here that about 65 percent of the sodium liberated by the current is practically obtained ; the other 35 per cent is really set free, but is lost by recombination with chlorine in the bath, or oxidation or imperfect condensation. This cannot, however, be the only source of loss in the process, for a current of 70 amperes at 12 volts represents 840 Watts or 0.205 calories per second, while the liberation of 39 grammes of sodium from sodium chloride (heat of formation 4.2474 calories per gramme of sodium) represents only 165.65 calories per hour or 0.046 calories per second. The net proportion of useful effect over all is therefore only 22.5 per cent. The cause of this low return is to be found in the high voltage used. Calculating the minimum electro-motive force necessary to decompose sodium chloride, we have (0.000238 x 4.2474)/0.00024 = 4.2 volts.
Since, then, only 4.2 volts out of the 12 volts absorbed by the bath are used for actual decomposition, the percentage of the power of the current used in this way is 35 per cent. But since only 65 per cent of the work which this does is represented by the sodium actually obtained, we should have a net utilisation over all of 65 per cent of 35 per cent, or 22.75 per cent, which agrees with the result before obtained.
Electro-thermal reduction.—Let us take for discussion some official figures of the Heroult process for producing aluminium alloys. Current, 8000 amperes at 28 volts tension. In 271 hours of actual operation, during which time the crucible cooled several times, the average production of aluminium was 6.8 kilos of aluminium per hour. The electro-chemical equivalent of aluminium is 9, so that a current of 8000 amperes can deposit, electrolytically, 0,000010352 X 9 X 8000 X 60 X 60 = 2,68 kilos of aluminium per hour. If, then, there was actually produced 6.8 kilos per hour, and as much as 10 to 12 kilos are claimed when the furnace is working steadily and up to full efficiency, it is impossible that more than a fraction of the aluminium is produced by electrolytic decomposition. As far as the total energy of the current is concerned it is large enough to account for all the thermal effects produced. A current of 8000 amperes at 28 volts is equal to 224,000 Watts, or 53.79 calories per second, or 193,644 per hour. This, if it could be applied to nothing but the decomposition of alumina, would isolate 193644/7250 = 26.7 kilos of aluminium per hour. But as 6.8 kilos were obtained, we see that about 25 per cent of the energy of the current is absorbed in setting free aluminium, the other 75 per cent being converted into heat. This source of heat, together with that added by the burning of the carbon anodes, keeps the interior of the crucible at a temperature far above any temperature ever reached by any other means, and at that temperature the writer has not the least doubt that the alumina has its oxygen abstracted from it by the chemical action of the carbon. Similar calculations could be made with data from Cowles' furnace, with similar results and conclusions.
In conclusion I would remark that copper, silver, gold, magnesium and aluminium are the principal metals which are at present being commercially treated by electro-metallurgy. But if ever the problem of converting the energy contained in coal directly into electric energy be solved, there are very few of the metals which might not be cheapened by electrolytic methods. If the conversion could be effected with an efficiency of only 50 per cent, it would still be 10 to 15 times as efficient as our present indirect methods of boilers, engines and dynamos; and the possibilities opened out for the art of electro-metallurgy by such a cheapening of cost of the electric current are so extensive that if we stated them they might appear visionary. A comparison might be made with the revolution in the mechanic arts which would be produced by such a discovery. We have electric motors which turn nearly 90 per cent of the mechanical energy of a current into rotatory motion, and if the current supplied to them represented say only 50 per cent of the total energy of the coal, we would get rotary motion with an expenditure of J of a pound of coal per hour per efficient horse-power.
However, taking matters as they stand and being as moderate as we may in our expectations as to cheap electricity, I think it reasonable to conclude that this new art of Electro-Metallurgy, which had its commencement within our lifetime, will become, perhaps, the leading feature of Metallurgy in the Twentieth Century.