Electric Smelting & Aluminum Co. v. Pittsburgh Reduction Co.

111 F. 742, 1901 U.S. App. LEXIS 4992

Opinion

HAZED, District Judge.

This is a suit in equity by the Electric Smelting & Aluminum Company against the Pittsburgh Reduction Company to restrain the infringement of two United States letters patent for a process of separating aluminum and for a process of obtaining metals from their ores or compounds by electrolysis. The patents are owned by the complainant under several mesne assignments. The original application for patent was filed by Charles S-Bradley, February 23, 1883; on December 8, 1891, patent No. 464,933 was issued on a division of the original application; and on February 2, 1892, patent No. 468,148 was granted to him. Both patents relate to a process for the reduction of highly refractory and nonconducting metallic ores or compounds in an unfused state by electrólisis; that is, by subjecting the ores or compounds to the action of an electric current, resulting in a fusion of the entire mass of ore, and, while in a state of fusion, by conjoint and simultaneous action of the electric current to separate, disassociate, or decompose the fluid mass, so that the metal contained therein will be gradually deposited at the cathode, then to be drawn from the bottom of the receptacle or basin in which it is contained. Claims 1, 2, and 3 of patent No. 468,148 cover and control the process broadly, while claims 4, 5,, and 6 cover and control the process described as specifically applied: to the separation of aluminum from its ores. Patent No. 464,933 relates to the same process, but confines its claims to a particular structure used and a special mode of practicing the process. The answer sets up various defenses, among them want of novelty anti noninf rin gem eut.

The case is of very great importance. The questions involved are of such magnitude and complexity that a great deal of time has been consumed in becoming familiar with the voluminous testimony, consisting of about 3,000 printed pages, including exhibits. The court has proceeded, however, in the hope that, when a decision is rendered, the propositions involved may have received such careftd consideration as their importance requires. The claim of the complainant and the various defenses interposed are of a difficult and technical nature. The testimony of the expert witnesses, men of a high reputation in chemistry and electro-metallurgy, is of widely divergent character as to the application of important scientific and electrical principles. A knowledge of the complicated and intricate subjects of chemistry and electricity, and the application of their laws, would seem necessary to a proper disposition of this case. The evidence consumes great space, and should make the questions at issue clear of comprehension. The arguments of counsel generously amplify the copious and discursive opinions of the expert witnesses. Everything appertaining to the discovery for which the patents were granted has been clearly set forth, so that in the consideration of the facts and the law applicable thereto the task of the court may be as clear as the subject will admit. Some of the facts are obscured in more or less uncertainty, and are difficult of solution; but it is hoped that a sufficient understanding of the questions involved has been reached for a proper- determination of the controversy.

The process by which aluminum is extracted from its ores has been completely revolutionized. The patent of Hall, used commercially by the defendant, and claimed by complainant to infringe its patents in suit, is No. 400,766, issued April 2, 1889, for a process of reduction of aluminum by electrolysis. This patent unquestionably resulted in the phenomenal reduction in the price of aluminum. In 1886 its price ranged from $5 to $8 per pound, while in 1897 it was from 25 to 30 cents. The entire consumption of aluminum in the United States is supplied by the defendant. From an extremely rare and only occasional employment of this metal in the commercial world, it has become useful in the manufacture of articles for which theretofore brass, copper, and tin only were known to be serviceable. Its price being reduced to that of copper and brass resulted in thé advantageous use of aluminum for .many purposes only made possible by the lowered cost of manufacture. It is not seriously disputed thát this result was due to the discovery of the Hall process and its subsequent operation. Indeed, immediately after the defendant began its use commercially, manufacturers using other processes ceased to work on account of their inability to profitably make aluminum at the price for which the defendant was able to place its product on sale. What brought about this revolution in the art ? Hall’s answer is that he was the first to discover a suitable-bath material, which, when fused, on account of certain properties contained therein, would at a low temperature freely dissolve alumina; that his bath contained .a higher chemical stability than the dissolved alumina itself, so that, by the passage of the electric current, the alumina would be decomposed, and the bath material and its properties left unaffected.

Alumina in the form of an ore or compound is abundant, and therefore cheap. Prior to Hall’s invention, it was difficult to melt or fuse, while cryolite, a natural double fluoride of aluminum, was known to be fusible at a high temperature, and its resistance to electrical conduction was regarded so great as to render its use practically prohibitive. Hall’s idea was that, inasmuch as it was commercially impracticable to fuse alumina, he might discover a method of dissolving it in a fused bath chemically more stable than alumina, and containing substances not easily volatilized nor subject to decomposition by air. The bath mush possess high electrical conductivity and low specific gravity. Moreover, it appeared to him vitally essential that the solvent must be incapable of decomposition by the electric current required to decompose the dissolved alumina; otherwise, the solution would become decomposed by the action of the current, and the alumina remain in the bath unaffected. Therefore the bath material used as a solvent must have a higher electrolytic stability than the alumina. He experimented with flúor spar and the fluorides of sodium, potassium, and magnesium. These fluorides were either insufficiently fusible, or, when in a fused state, failed to satisfactorily operate as solvents for alumina. Cryolite, a double fluoride of aluminum and sodium, when fused, was finally ascertained to meet with highly favorable results. Hall says he “found that cryolite fused easily, and, when fused, dissolved alumina as easily as water dissolved salt, without raising the melting point of cryolite or apparently changing its physical properties.” This discovery was quickly followed by repeated successes. Carbon crucibles were used, in which to dissolve the alumina in a double fluoride, or cryolite, bath, and then to decompose the dissolved alumina by passing an electric current through the solution by means of carbon rods used as electrodes. Defendant contends that through this discovery Hall solved the problem of the practical electrolytic production of aluminum, one which had remained unsolved for half a century; that, in comparison with the prior processes, he made a vast improvement, when he planned to fuse a practically nonclectrolyzable ore, and to electrolyze an uufusible one. The extraction of aluminum from its ores by a method of electrical fusion and decomposition, so that the metal regarded as precious could be commercially utilized, had for many years baffled specialists, who had given untiring attention to the subject. In 1888, following the successful experiments conducted by Hall, this defendant corporation was organized. Fifty pounds of aluminum for a time were produced per day at its Pittsburg manufactory. Soon afterwards its plant was enlarged by the erection of two additional establishments at Niagara Falls, using a total of 6,000 electrical horse power and having an output daily of 9,000 pounds of aluminum.

What does Bradley claim? His invention consists in the regulation of the heat generated by current energy to perform three functions. He utilizes the electricity developed between electrodes placed in a mass of refractory ores,—nonconductors in an unfused state. He maintains the necessary current energy communicated from the initial point throughout the entire mass, thus securing fusion and simultaneous electrolysis, and progressively holds a mass of ore in a continuously fused state, resulting in its reduction. Thus the pure aluminum will be deposited at the bottom of the cavity or basin, whence it may be withdrawn. To produce this result, the electric energy must be applied according to the method specified by the patent, and stated in complainant’s brief to be:

■“(1) Melt and maintain melted the material in the initial line of flow. (2i To apply the current in such a manner tliat the melted material will spread its heat, so as to propagate sufficient heat to liquify the surrounding material. (8) Decompose or separate the atoms of the fused mass, made a conductor of electricity by the diffusion of the heat generated.”

The main object of the invention, the patentee says, “is to dispense with the external application of heat on the ore in order to keep it fused.” This he proposes to accomplish by the employment of “an electric current of greater strength or intensity than would be required to produce electrolytic decomposition alone.” The decomposition or disassociation of refractory ores by electrolysis is not claimed to be new in the art. The utilization of a sufficient quantity of heat from a center of fusion throughout a mass .of nonconducting and refractory material, of which cryolite is a .‘sample, by means of an electric current concentrated upon a por>tion of such mass, is claimed to have been unknown, and that Bradley is a pioneer in the art.

Complainant claims that its process is not limited to any particular class of ores. The patent covers the extraction of metals 'from their compounds, and is not restrictive in its scope. Therefore the Hall process is the same as Bradley’s, although more economically performed, because Hall uses a well-known flux or solvent for the alumina, which facilitates fusion and consequent electrolysis. I am quite satisfied that it was not a well-known flux for this purpose. The utility of electricity as a potent agent for the transmission or diffusion of heat was well known for fusing refractory ores. Various anticipatory patents will be hereafter discussed. The mere application of this agent to decompose alumina, after ir has passed through an intensely heated cryolite bath, does not necessarily fuse the alumina. Hall-does not endeavor to do this. Instead of using a fusing process, he creates a bath by which alumina is dissolved. This is done by charging it into the fused bath, resulting in the process of solution. This distinction, claimed to exist between fusion and solution in the chemical and metallurgical arts, has been the basis for considerable expert testimony, ■extremely contradictory in its nature. It will not serve any good -purpose, with reference to the various definitions of fusing, fluxing, melting, solution, and dissolution, to in detail or at length make any comment upon these definitions.

This very same question, which is regarded with so much importance, was decided in the circuit court for the Northern district -of Ohio in the case of Pittsburgh Reduction Co. v. Cowles Electric Smelting & Aluminum Co. (predecessor of complainant) 55 Fed. 301. That was a suit brought by the defendant in this case against the Cowles Electric Smelting & Aluminum Company, a corporation .then in existence and engaged in the manufacture of aluminum .alloys, to'restrain infringement of the Hall process on the part of the Cowles Company. It was contended that the use of cryolite as a flux for alumina was well known in the art prior to the date of the Hall invention; that fluxing and dissolving were synonymous, terms, and therefore the Hall process was old and his patent void. The suit was tried before two eminent jurists, Judges Taft and Ricks. The learned and exhaustive opinion given by the court in that case is persuasive of the finding there made on the questions before it. The learned court, after an examination of all the patents claimed in anticipation, and after full consideration of the terms “fluxing,” “fusion,” and “solution,” held that the Hall process consisted simply in solution of the alumina and subsequent electrolysis. The process of De Ville was said to be a process of electrolysis, followed by “electro-chemical solution of the alumina.’7 Judge Taft was of the opinion that by Hall’s process the alumina dissolved and became the electrolyte. The court had before it substantially all the evidence and exhibits that are in this case. The questions which arise here with reference to fluxes and solution were exhaustively argued. Many of the expert witnesses in that case are witnesses here, notably Mr. Cowles and Drs. Chandler' and Morton. Indeed,, the chief contention was that the use of cryolite- as a flux for alumina was old in the art, and that “fluxing” and “dissolving” were synonymous terms. Hall maintained, and the conclusion of the court was that the novelt}' of the process consisted in that the alumina dissolved in a fused double fluoride of sodium and aluminum as freely as sugar dissolves in water; that, until he discovered that alumina would thus freely dissolve in this bath, it had never been known, and that no one had ever before succeeded in disrupting the constituent elements of alumina by an electric current; and therefore the patent was a pioneer patent, entitling it to liberal construction. The court quoted with approval, at page 309, the testimony of Prof. Chandler, who also gave testimony for defendant in this case:

“The alumina dissolves in the fused hath of cryolite, as sugar does ik water, quietly, without any of the usual evidences of chemical action, such as production of heat, or the reduction of fusibility, or thickening of the liquid. This seems to show that the alumina in solution remains chemically unchanged.”

After the decision in that case a motion was made for a rehearing before Judge Taft, and the meaning and effect of metallurgical fusion and solution again discussed. Pittsburgh Reduction Co. v. Cowles Electric Smelting & Aluminum Co. (C. C.) 64 Fed. 125. It is true, in that case the Bradley patents were not relied on. The Cowles patents, however, substantially claimed to be the same, were’ before the court. On the motion for rehearing it was again argued that the use of cryolite as a flux is the same as its use as a dissolvent for alumina. The De Ville patent, claimed in anticipation, was again reviewed and distinguished. The court rested its decision on the degree of fusibility of alumina in cryolite. This De Ville’s description of his process shows to a very small per cent., and it does not appear that his process will render fusible a very large per cent.,, which Hall discovered and which his process performs. The court said:

“It is utterly immaterial, therefore, whether the cryolite he called a ‘flux’ or a ‘dissolvent’ of the alumina, because in either case the extent to which De Yille’s experiments showed that it could dissolve or flux alumina was very small.”

The court further said:

“Neither the Bradley nor the 'Cowles patents contain the slightest suggestion that cryolite should be used as a flux for the refractory alumina, and no evidence is sought to be introduced that any one ever actually used cryolite in furnaces for that purpose.”

From the foregoing extracts, and a careful review of the evidence and authorities cited by counsel in the present case, I am constrained to hold that Hall’s invention was not a mere alteration in the Bradley process, or the substitution of one ingredient for another, which was well known at the time. The subsequently conceived process of Hall is substantially different from the former of Bradley, and the addition of the alumina in the cryolite flux tended to perform a different function, not covered by the Bradley process. It tended towards the solution of an ore which, if operated upon by the Bradley process, construing it broadly, and not limiting it to any particular class of ores, would have necessitated its fusion with the contents of the bath, and would, therefore, have required an extremely high electrical voltage to perform the office of disassociating the aluminum.

Complainant’s counsel maintain with great force and ability that, inasmuch as the Hall bath is a compound one, consisting of fluorides of aluminum, of sodium, and of calcium, and oxide of aluminum, and the electric current passes through every substance in the bath, therefore the Hall bath is fused in the same manner specifically claimed by the Bradley patents. Nevertheless, Hall’s heat is quite differently applied and utilized than Bradley ever contemplated. Hall maintains that he reduces his current to a minimum and obtains his heat from the anodes, the combustion of the anodes, and the lining of the crucible or pot; that these developments of heat result from the use of the electric current for decomposition. He adds pulverized carbon to his bath and throws on additional alumina. After a short space of time, it is intermingled with the bath material. This method also utilized a certain amount of heat by preventing its escape by radiation. In each pot or crucible are fixed 40 carbon rods, made of compressed carbon, each weighing eight pounds. When these rods are immersed in the bath, they are gradually consumed, and as ( the ends are consumed they are pushed down, to keep them nearly ■ in contact with the lining of the crucible or pot. The oxygen, as it is released from its compounds, combines with the carbon, by which carbonic acid gas is formed, resulting in the production of heat. This heat maintains the fluidity of the bath, and suffices to keep the bath liquid, without any excess of current. Hall claims that to use an excessive current, as described by the Bradley patent, would be unnecessary, and would involve considerable expenditure of money for the production of its current energy. In the Bane-Fox British patent of 1878 it was demonstrated that electric currents could be employed for external heating as a substitute for fires involving combustion'; the heat being produced by resistance to the current in or at the surface of the containing vessel. The greater part of defendant’s heat being obtained through the carbon anodes and the crucib!e lining would suggest a similarity with the Fane-Pox patent. The heat generated is not permitted to be of such intensity as to fuse the bath, and also decompose it, but only of such low intensity as to decompose only the dissolved alumina, and not the bath itself. The alumina is heated to the temperature of the bath before it becomes a part of it, and is then easily dissolved, without participating in any fluxing action. It is quite true that a chemical reaction may be said to take place, producing from substances practically nonelectrolytic a new substance, the electrolyte, which, when acted upon by the amperage, deposits aluminum at the cathode, and that the initial step of the process to create a chemical change of substances to form the electrolyte cannot take place without heat. Hall’s claims do not say what heat shall be employed to fuse the bath. The process described by his patent speaks of dissolving alumina in a fused bath, composed of fluorides of aluminum and a metal more electro-positive than aluminum, and then passing an electric current through the fused mass. With reference to obtaining the required temperature to maintain the bath the patent says:

“Tills crucible is placed in a suitable furnace, B, and subjected to a sufficient heat to fuse the materials placed therein; such materials fusing at approximately the same temperature as common salt.”

From this it is deduced that Hall never intended, certainly that he did not conceive, the plan of internal heating by current energy, and that its present use for bath fusion and electrolysis is an infringement of Bradley’s process. The latter describes a method for maintaining both fusion and electrolysis by continuous and simultaneous action of the electrical current applied to his bath. The specifications declare:

“The main object of my invention, therefore, is to dispense with the external application of heat to the ore in order to keep it fused. In order to accomplish this object, I employ an electric current of greater strength or intensity than what would be required to produce the electrolytic decomposition alone, and I maintain the ore or other substance in a state of fusion by the heat developed by the passage of the current through the molted mass, so that by my invention the electric current is employed to perform two distinct functions; one of these being to keep the ore melted by having a portion of its electrical energy converted into heat by the electrical resistance offered by the fused ore, and the other being to effect the desired electrolytic decomposition, by which means the heat, being produced in the ore itself, is concentrated at exactly the point where it is required to keep the ore in a state of fusion.”

The question presented is not free from difficulty. Counsel for complainant argue with great force that the heat development had first to be conceived, and the advantage of utilizing the Bradley plan of heating as a substitute for intense external heat had first to be realized. Was the conception of the application of electrical heat, in the art as it existed at the date of the application for the Bradley patent, such a discovery as to entitle the patentee to the protection of the patent laws? It is urged that the claims of the Bradley patent must be broadly construed with reference to its limitation to the use of í'ii electric arc for initial fusion, in the light of the decision rendered in Aluminum Co. v. Lowrey, 24 C. C. A. 616, 79 Fed. 331. The court in that case did not pass upon the validity of Bradley’s claims, but it did have before it what had been previously done by Sir Humphrey Davy, Siemens, and others. The court expressly said that the record in that case was not made up for the purpose of enabling it to decide how far the Bradley invention advanced the prior art. It also used the following language:

“But the researches of Sir Humphrey Davy, Siemens, and other scientific men had already made a considerable advance in develoifing this new art.”

The Lowrey suit was brought by Lowrey’s executors to clear the title to the Bradley patents, which Cowles claimed to own by virtue of an assignment of May 18, 1885, assigning to Cowles inventions of Bradley which interfered with the patents of Cowles. The complainant claimed title under a subsequent assignment, and contended that the Bradley claims related to a process essentially different from that of the Cowles patents, and therefore that the assignment by Bradley to the defendant, by reason of a restrictive limitation in the assignment, did not include the Bradley patent. The determination of the court in that case with reference to the construction placed upon the claim of the patent in suit can have no bearing upon the questions before me. This suit is not between the same parties or their privies in interest, and the subject of the action is widely different. Moreover, in the view that I have taken, it would serve no useful purpose to determine the question whether the patents of the complainant are limited to initial fusion by an arc. The proof on the part of the complainant—and the patent so indicates—is that the electrode and the carbon slab upon which the ore may be placed shall first form a contact. This contact would offer a resistance* resulting in the production of intense heat at the point of contact. The electrode'is withdrawn when the heat is generated by reason of the contact between the electrode and carbon slab. This permits a heat-producing action, and aids the development of a larger bath or body of fused material. Complainant’s experts described this as an incandescent operation, as distinguished from that of an arc. The proofs show that, when a current of sufficient voltage is applied to metallic compounds, it will fuse, maintain fusion, and decompose them without the aid of external fire. Defendants contend that this was well known to be old in the art.

Prior to 1884 Messrs. Cowles were engaged in the reduction of metals, aluminum alloys. By their process the refractory and nonconducting material was mixed with carbon or other similar conducting, although highly resisting, material. At this time'the diffusion of electrical heat throughout the mass of material in the furnace, invented by Cowles, was justly regarded as a remarkable process, marking an era in practical metallurgy. Bradley’s application for a patent was then on file. The general adoption of the practice of internally heating the baths by the current, from the failure of external heating unquestionably, is due directly to Messrs. Cowles. Doubtless great credit is due to their enterprise and progressive commercial foresight.. Nevertheless it was known that Sir Humphrey Davy, in a lecture delivered before the Royal Society of London in 1807, and published in the Philosophic Transactions for 1808, used this language;

“1 tried several experiments in the electrization of potash rendered fluid by heat, with the hopes of bom?;' able to collect the combustible matter, but without, success, ¡md I only attained my object by employing electricity as a common agent for fusion and decomposition.”

This was a mere experiment, possibly, but it was an account of a process ihat was complete and capable of practical operation. Bradley admitted, in the proceedings at the patent office, in his second brief on appeal to the commissioner, that:

“Sir Humphrey Davy’s reduction of potassium and sodium, as described by him, on án inlinitesimal scale, was the process described and claimed by Bradley.”

I cannot agree with the argument that Sir Humphrey Davy’s process revealed nothing to the world. His laboratory operations with reference to a process for reduction have since been carried out oil a larger scale. They have been proven practical by successive improvements, and bis method has been tested by actual use. Castner’s patent, No. 452,030, granted May 12, 1891, is cited as an improvement in the process described by Davy. Siemens invented a mode of initial fusion, using the arc, which made Davy’s process applicable to other fusible electrolytes than those upon which,he experimented.

It is vigorously contended by complainant that Davy’s experiment did not demonstrate that electric heating could be practically applied internally by conduction to surrounding nonconducting and refractory substances; that his experiment consisted of its operation on minute quantities; and that soda and potassa, Davy’s material, does not justify a similar operation with materials of different character. Dr Morton, expert for complainant, says, in speaking of the Davy experiment, after declaring that a battery analogous to that of Davy ought to be used to ascertain the conditions which existed in his case:

“I Pave no doubt, that Davy’s experiment may be successfully carried out, even with a battery, as in the experiment that I made, or with a dynamo machine and adjustable resistance properly manipulated.”

Dr. Chandler met with fair success in his trial experiments for use in this case by his substantial repetition of Davy’s process. In 1881 the Compagnie Generate Beige, in its French patent, spoke of the use of the. electric arc for heating substances. Nothing is said of reduction, but it is suggestive of the use of electrical heat for fusion of metals. Prof. Howe in 1882 published a paper in the regular animal volume of Transactions of the Society of Alining Engineers, and copies thereof were distributed as early as May, 1882. In this paper Prof. 1 [owe speaks of fusing his material before decomposing it by electrolysis. Bunsen fused chloride of magnesium, and electrolyzed it with the aid of electrodes, and demonstrated the necessity of a definite voltage. The United {states patent of Ball & Guest, No. 236,478, described the application of an electric current for carbonizing arcs of paper or other materials while confined in a case of nonconducting material. Siemens’ British patent, No. 2,110 of 1879, to which I have already referred, describes the use of the. electric arc for the fusion of metals. It makes no reference to decomposition, but it is dearly stated by the patent that the electric current is ap plied to produce intense heat for the fusion of electrolyzable substances. Two carbon rods, fitted to slide towards each other horizontally within water-cased tubes, are employed. The substances to be fused are introduced in the crucible, and the carbon rods are advanced sufficiently near each other to form an arc within the crucible. Dr. Morton says that with modern dynamos, generating large currents, Siemens’ apparatus, which was useful' only on a small scale, could be enlarged, so as to fuse large masses of material requiring high temperature, provided the Siemens method of cooling his electrode with water is dispensed with. Bell in 1861 had electrolyzed cryolite, and Gauduin in 1869 decomposed aluminum salts, by the electric current in crucibles by means -of external heat. As the aluminum was extracted, fresh material was added. This is also done by complainant’s process, and by that of the defendant, when necessity requires. Other patents are in evidence from which it appears that the decomposition of metallic compounds was old in the art. In every case external heat was used. No electrolytic heat generated in the crucible was utilized for continuous heating. That seems not to have been practical until the discovery by Hall of a suitable bath, which would, when fused, dissolve alumina. I think that these patents quite clearly show the state of the art with reference to electrolysis and internal heating of ores or compounds. They disclose the state of the art to be such that what Bradley did was to adopt a well-known process for heating to the special purpose of fusing,—a carrying forward or new application of what others had already done or attempted in reducing refractory metals by similar means. His claims must therefore be narrowly construed. Manufacturing Co. v. Cary, 147 U. S. 623, 13 Sup. Ct. 472, 37 L. Ed. 307.

The complainant’s argument is that by the Bradlej’’ patents an electrolytic current is employed sufficiently powerful to do two things, to wit, to electrolyze and to develop sufficient heat to keep the ore fused. This may be done, as the patent points out, in the following way:

“For the purpose of perfectly managing and controlling my process, I have my electric generator or source of current so arranged that the strength of the electrolytic current may he properly regulated, and the mass of ore thereby kept at proper temperature. The most convenient way to accomplish this is to raise or lower the electro-motive force of the generator by any of the well-known methods employed,—for example, in incandescent electric lighting. * * * By these means I am able to dispense with the necessity of keeping the ore in a fused state by the application of heat from without through the walls of the refractory vessel, and to concentrate the hoat required for this purpose just where it is needed between the two electrodes.”

True, Bradley explicitly states that by his invention external heating is dispensed with. To do that, as already stated, simültaneous and progressive fusion and electrolysis are relied .upon; but, in the light of the heat which is incidentally generated by the Hall method to perform this work in his process and the prior state of the art, a construction cannot be placed upon the Bradley patent sufficiently broad to permit his claim with regard to simultaneous fusion and electrolysis to inure to his benefit as against the claims of Hall. This construction, however, requires that defendant’s operation of its process clearly appear. Its apparatus consists of a crucible, pot, or tank, about six feet long and from three to four feet wide, wherein the double fluoride of aluminum and some other more eleciro-posicive metal, such as sodium and calcium, are fused. In this fused bath alumina is dissolved and decomposed. The receptacle in which the process takes place is lined with carbon, and, with the aluminum it may contain, constitutes the cathode. The anodes are carbon cylinders, adjustably suspended by copper rods from an overhanging frame, which is connected with the positive side of the dynamo. The anodes are immersed in the bath, and the current passes to the lining of the pot, accomplishing in its passage the electrolytic decomposition of the dissolved mass. At the beginning the flat ends of the carbon anodes are pressed into close contact with the flat bottom lining of the pot, the current is then turned on, and, owing to the resistance offered at the contacts of the anodes and cathode, great heat is developed, rapidly heating up the lower ends of the anodes and the lining of the pot at and near the contacts. The anodes are pressed down from time to time, as the carbon is destroyed by combustion. By defendant’s method, molten bath material is ladled into the apparatus from one going out of'use or requiring repairs. Other raw mal erial is added to the crucible. Alumina is added and quickly dissolved. The anodes are raised a few inches, and decomposition takes place. This operation continues for months, until the apparatus gives out or repairs are required. Prom the evidence of Jeff, complainant’s witness, it appears that the hath sometimes becomes thick when running under a voltage of less than 5 ; that the carbon is then raised to supply more voltage, a volt meter being connected with each apparatus. Efforts are made to keep 5J4 volts in each crucible, and about 45 are in operation at the saíne time. Is the current regulated, and the mass of ore thereby kept at the proper temperature, for the purpose of managing and controlling the process, as claimed by the Bradley patent? It is well understood by those familiar with the art that an electric current, applied to a conducting clement, generates heat proportionate with the watts of electrical energy used to overcome the electrical resistance. Joule’s law, so called, determines the law of electrical heating; in the following words:

•‘The hoot which is evolved by the proper voltaic action of any voltaic current is proportional to the Ripiare of intensity of the current multiplied, by the resistance to conduction which it experiences.”

Prof. Thurston, of the fibley College of Mechanical Engineering, Cornell University, called as an expert witness for defendant, says:

‘•.Toule's law, as above expressed, holds good, whatever the nature of the substances producing the resistance, and the heat evolved in the passage of the electric current is precisely ¡lie same in amount; the same current being passed and the Maine resistance being met, whether the pesbding material he a solid, or liquid, or ;> solid in a fused ututo. The .quantity of heat is a function of current and resistance alone, and is absolutely independent of the nature of the substance traversed by the current.”

To electrolyze a fused bath, a practical voltage is required to overcome the resistance of the carbon electrodes. This voltage is distinguished from theoretical voltage by Prof. Chandler. He says:

“Extra electrical energy is always required in every electrolytic process to overcome conductive resistance of tlie hath and to go on with reasonable rapidity. It is essentially the case in fused baths where a voltage considerable above the theoretical voltage is always required to produce satisfactory results. * * * It is necessary to use a sufficiently high voltage to cause electrolytic action to proceed with rapidity, and experience has shown that from five to six volts are necessary with the solution of the defendant’s process. It is immaterial whether defendant’s process is carried on in crucibles or pots heated by the external heat of the fire or by the electrical heat of conduction.”

From the experiment of Dr. Morton, made on April 26, 1899, and from other measurements of counter electro-motive force made by Mr. Davis, Prof. Geyer, and Prof. Rogers, it is argued that the defendant uses an electric current of sufficient strength and density to decompose cryolite, and that it uses a greater voltage than would be required for electrolysis alone. Defendant’s experts do not agree with this contention. I am constrained to defer to the judgment of defendant’s expert witnesses, in view of the vast amount of corroborating evidence found in the case, showing that Hall, in experimenting with his process, never contemplated the use of an electric current of greater voltage than to produce decomposition. He was primarily of the opinion that the heat produced by the electrical energy necessarily used to decompose the metal in the crucible would be entirely sufficient for both fusion and decomposition. Indeed, in a letter written to his sister, as early as August, 1886, he said:

“In some respects tbis invention is going to be better than I anticipated. Thus, the resistance of the liquid is exceedingly low. Also, it is evident from the experiments that the waste heat of electricity, which must be used anyway, will be nearly, if not quite, enough to keep the solvent melted.”

The proofs show that Flail’s anticipations in that regard proved true. The current strength sufficient for electrolysis, as applied b3r the Flail process, is sufficient to fuse and decompose. The development of heat is an adjunct to the process’, or, as said on the argument, it is incidental thereto. This fact, of course, aids in minimizing the voltage and amperage required to keep the bath fused and to perform electrolysis. Obviousfy, this renders it unnecessary that Hall should use an excessive current for fusion of the ore. By Bradley’s method, it is distinctly stated that the current must be sufficiently powerful to both keep the ore in a proper state of fusion and to perform the required electrolytic work. It appears that in securing this result by his process a double strength of current is required. Cryolite is the ore cited by' Bradley as an illustration in his patent, and doubtless was in his mind when he conceived his invention. Its fusion requires, as has been said, a current of greater strength and intensity than would be required to produce decomposition alone. In the case of Pittsburgh Reduction Co. v. Cowles Electric Smelting & Aluminum Co., supra, the defendants sought to distinguish their process then employed b)' denying infringement of the Hall patent, for the reason that they used the electric current internally applied to fuse their bath material. The court said:

“This defense is based on a narrow and highly impossible construction of the Hall patent.”

The application of that decision to the case at bar would certainly compel the conclusion that, if the only alleged infringement by Hall is the application of internal heat, then in reality no infringement exists. The theoretical voltage required by Hall to decompose alumina is 2.8 volts. The working voltage required is from 6 to 7 volts. Defendant’s contention that this amount of voltage would be required to electrolyze the contents of the pot, even though the mass had been fused by external heat or by the electrical heat of conduction, is disputed. The Heroult process is cited as an instance where the theoretical voltage which is required for this same purpose, using external heat to keep the hath fused, is 3 volts. Inasmuch as the Heroult process is substantially that of the defendant, complainant argues that the defendant, in its use of a current of 6 volts, infringes this feature of the patent, because of its use of an electric current of greater strength or intensity than would be required for decomposition alone, and that a regulation of the current is necessarily so arranged as to keep the ore at proper temperature. A practical voltage, as heretofore defined, is required; but the heat required to maintain fusion is obtained by the heat radiation, and from such sources as are incidental to the use of the process, and not from any independent process of electric heating. The record does not bear out this claim of the complainant with- reference to anticipation. The Heroult process fused the alumina by internal electric heat, melting it above a base metal, so that it would alloy with aluminum without the presence of cryolite. These alloys contain 10 or 15 per cent, of aluminum only. This process was used in Rurope in 1887 and 1888. Subsequent to the Hall process, it was put in use in the United States. His application for United States patent covering the pure aluminum process was in interference with the application of Hall, and was determined in Hall’s favor.

It is quite true that there must be a regulation of the current utilized by the Hall process; but manifestly, from what has already-been said, that must be found due to the regulation of the application of electricity for the decomposition of alumina. After the current has passed through the electrolyte, an increase of voltage will not Increase the production of metal. That is affected only by increasing the amperes. No statement appears in the Bradley patent as to the number of volts employed either for reduction performed with internally or externally heated crucibles. Complainant’s witness Rogers testified that, if Bradley used twice the voltage necessary to effect decomposition, he would not be employing, according to Prof. Chandler’s figures of the theoretical voltage, more than 6 to 10 volts, and that his voltage would simply be determined by the amount necessary to decompose and maintain fusion. Obviously, it is essential by tlie Bradley process to first ascertain the amount of voltage required for electrical decomposition, and then increase the electric current to one about twice as strong as would be employed to perform a given amount of electrolytic work in the ordinary way in externally heated crucibles. By so doing the patentee is enabled to keep the ore fused without the application of any external heat whatever. It is quite plain that this is not Hall’s process, who, as we have seen, uses only such heat as is incidental to every operation of electrolysis, which it is admitted is not covered by the patent and is not patentable. Considering this feature, together with the application of alumina to a cryolite bath, it appears to me that the Hall process is a new mode or operation by means of which a new result is obtained. I do not think, for reasons hereinafter stated, that the Bradley patent, claiming broadly the disassociation of aluminum from its ores or compounds, or of other like highly refractory metallic compounds, can be construed to cover the claims of Hall. It is quite true that when a patentee describes an invention, and then claims it as described, he is protected by law, not only in the precise form which he has described, but all other forms which embody his invention. Winans v. Denmead, 15 How. 329, 14 L. Ed. 717. But here in the Hall process there was a different employment of the ingredients -which go to make up the process, and an entirely new mode of operation. The means employed and the method of operation produced a new or better result. It may be said to be equally true “that, when a person has invented some mode of carrying into effect a law of natural science or a rule of practice, it is the application of that law or rule which constitutes the peculiar feature of the invention.” Sewall v. Jones, 91 U. S. 184, 23 L. Ed. 277. But Hall’s process was not applied in the same way. By his method of solution' the work was performed differently.

The question of the utility of the Bradley process must now be considered.- Bradley conceived his invention February 23, 1883. The patents were not issued until about eight years afterwards. The delays were due to adverse decisions of the patent office, and amendments made to the original application from time to time to meet those decisions; Hall’s patent was granted April 2, 1889, and went into immediate operation. The complainant’s proofs show that, although the Bradley process had never been in actual operation, the fundamental invention of Bradley was substantially used by Cowles Bros., in 1890 and previously, in the manufacture of aluminum alloys. Previous to that time it was substantially used for a short time at Boonton, N. J. This plant was started primarily to show the practicability of the Heroult process. The witness Stetson gives testimony to the effect that this process was operated at Neuheusen under the Heroult patent from 1886 to 1890. Under his direction pure aluminum and aluminum bronze were made. It was claimed to be used by Cowles Bros, in a suit for infringement in Pittsburgh Reduction Co. v. Cowles Electric Smelting & Aluminum Co., supra; it being established in that case that the defendant used Hall’s solution process. An injunction was granted, which is still in force. Inasmuch as Bradley’s patents were not before the court, it can hardly be claimed that the utility- of the process as specifically claimed was established. Although possessing facilities for their operation and practical control of both Bradley patents since the determination of the Eowrey suit in 1897, nothing has been done by complainant to demonstrate their commercial utility. I am not unmindful that the granting of a patent is prima facie evidence of its utility, and that the burden of proof of want of utility is on the defendant. If there is atjy-doubt, it should be resolved in favor of the'patentee. I find no evidence at all of the - practical use of the Bradley process. There was no abandonment of it, yet it is still important to consider its practical operation for the purpose of construing its scope. The witness Yates gave testimony in the suit brought for infringement of Pittsburgh Reduction Co. v. Cowles Electric Smelting & Aluminum Co., supra. He said the Cowles syndicate of England used the same process that the Cowles Company used at Lockport for the manufacture of aluminum bronze; that an experimental run of pure cryolite was made in one of their furnaces, but that the “current was not properly proportioned to the bath and no alumina was added; hence no aluminum was produced.” By the Gauduin bath experiment, made June 15, 1899, it was undertaken to demonstrate that aluminum was readily made by the Bradley process. It had a bath of cryolite, salt, and aluminum fluoride. It is claimed by the Gauduin process that the low temparature and conditions contributing to generate heat and dissolution observable in the Hall process existed, and that this patent is in anticipation oí Hall. Gauduin, in carrying 011 liis process, added aluminum fluoride. The experiment, however, seems not to have been carried out as specified by Gauduin. Aluminum fluoride was added in the beginning to the cryolite. This resulted in electrolyzing and decomposing a solution of aluminum fluoride in cryolile, instead of-fusing and electrolyzing cryolite. .It is thus quite doubtful whether the Bradley process has ever demonstrated its practical utility. The reasons expressed by the supreme court in the case of Westinghouse v. Power-Brake Co., 170 U. S. 537, 562, 18 Sup. Ct. 707, 718, 42 L. Ed. 1136, 1145, aptly apply:

“In view of the fact that the invention in this ease was never put into snccossful operation, and was to a liinito.fi extent anticipated by the Boyd on patent of 1883, it is perhaps an unwarrantable extension of the term to speak of it as a pioneer, although the principle involved, subsequently, and through improvements upon this invention, became one of great value to the public.”

In Deering v. Harvester Works, 155 U. S. 285, 15 Sup. Ct. 118, 39 L. Ed. 153, a case where, in view of the prior art, the invention was held to be of doubtful utility, and never went into practical use, and where the machines manufactured were extensively sold throughout the country for about eight years before any adverse right was asserted, the court narrowly construed the patent in suit. The doctrine of this last case would seem to be applicable to the case at bar.

Counsel for complainant argue that Bradley cannot be limited to the use of a single, kind of ore—for example, cryolite—in the opera-lion of his process; that, inasmuch as he was a pioneer in the discovery of simultaneous and progressive fusion of refractory ores, his patent must be broadly construed. Patent No. 464,933 states:

"My invention * * * is specially designed for the extraction of metals from their ores or compounds, and their reduction to the metallic slate,— for example, the extraction of aluminum from one of its ores, say, cryolite.”

Patent No. 468,148 states:

“My invention relates to the process of effecting by electric current the separation or disassociation of aluminum from its ores' or compounds, or the decomposition in a similar manner of other like highly refractory metallic compounds, (if which aluminum may be considered a type, and which have boon classed together by reason of the great difficulty in their reduction.”

The patents, read in their entirety, broadly claim to cover their operation by simultaneous fusion and electrolysis upon all ores or compounds refractory in their nature, and it is claimed that the compounds under discussion are no more, fully within the scope of the generic patent than any other metal capable of electrolysis. I am unable to view the patent in this light. The process of Hall was a new-mode of accomplishing the same object which Bradley had in mind, and the discovery that alumina would readily dissolve in a cryolite bath, rendering it easy of decomposition, must be considered a new element, and one which not only rendered his process commercially superior, but tended to produce a result in a different way. Although, as has been stated, the Bradley process has never been put to practical use, it does not appear that it was incapable of performing what is claimed for it. No statement is found in the specifications of Bradley’s process advising a person skilled in the state of the art at the time of the filing of the Bradley application that alumina might be dissolved and decomposed as set forth by Hall. An inventor must set forth the nature and the extent of his discovery, so that a person skilled in the art as understood at the time of the discovery may be able to successfully apply the process. Pitts v. Wemple, Fed. Cas. No. 11,194. In New York Paper-Bag Mach. & Mfg. Co. v. Hollingsworth & Whitney Co., 5 C. C. A. 490, 56 Fed. 224, the court used the following language: .

“An analysis of this patent in tlie light of the prior art, taken in connection with the circumstances that no practical commercial machine containing this invention has ever been put into use, forbids us from holding that this is a pioneer patent, which marks a great advance in the art or lies at the basis of a new art.”

In Westinghouse v. Power-Brake Co., 170 U. S. 568, 18 Sup. Ct. 707, 42 L. Ed. 1136, the supreme court, at page 569, 170 U. S., page 723, 18 Sup. Ct., and page 1148, 42 L. Ed., said:

“But, after all, even if the patent for a machine be a pioneer, the alleged infringer must have done something more than reach the same result. He must have reached it by substantially the same or similar means, or the rule that the function of a machine cannot be patented is of no practical value. To say that the patentee of the pioneer invention for a new mechanism is entitled to every mechanical device which produces the same result is to hold, in other language, that he is entitled to patent his function.”

The Incandescent Lamp Patent Case, 159 U. S. 465, 16 Sup. Ct. 75, 40 L. Ed. 221, was a case where two forms of electrical illumination had for many years been the subject of experiments. One was the arc, and the other the incandescent, light. A chief defense to that patent was defect upon its face in attempting to monopolize the use of all fibrous and textile materials for the purpose of electric illumination. The claim of the patentee, among other things, was:

“An incandescing conductor for an electric Lamp, of carbonized fibrous or textile material and of an arch or horseshoe shape, substantially as herein-before set forth.”

It was admitted that the lamp described in the Saw3'-er & Man patent was never a commercial success, but the contention was that in the conductor used by Edison he made use of the fibrous or textile material covered by the patent in suit. The court said:

“Instead of confining themselves to carbonized paper, as they might properly have done, and in fact did in tlieir third claim, they made a broad claim for ('very fibrous or textile material, when in fact an examination of over (j,C00 vegetable grow Lbs showed that none of them possessed the peculiar qualities that fitted them for that purpose. Was everybody, then, precluded by (his broad claim from making further investigation?”

American Bell Tel. Co. v. National Tel. Mfg. Co. (C. C.) 109 Fed. 976, is a case where the complainant alleged that the invention must be broadly construed, because the patent in suit covered all “constant-contact telephonic transmitters which it is possible to make.” Judge Brown, speaking of this averment of the complaint, said:

“I believe these claims to cover broadly all -transmitters having variable contact of opposing electrodos. They certainly are broad enough to cover every variable-contact transmitter that has been presented in this case. The concessions made by the complainant are sufficient to prove that these claims are what are termed in Carlton v. Bokee, 17 Wall. 471, 21 L. Ed. 519, ‘ingenious attempts to expand a simple invention of a distinct device into an all-embracing claim, calculated by its wide generalizations and ambiguous language to discourage further invention in the same department of industry, and to cover antecedent inventions.' It is a familiar rule that p generalization or definition that is too broad cannot be made good bv maUm." an arbitrary exception of each case that comes within its terms, bui whir should not have been included.”

And further on in his very exhaustive opinion lie sa\:

“I am of the opinion, therefore, as the claims are statutory requirements, prescribed for the purpose of making ; be patentee define precisely what his invention is, and as ihese claims are ! ■* broad as 1<> include every transmitter of the general class of which Hell von, ¡be liist inventor, and which is exemplified in Ids constant-contact transía ¡Her, that (hey are invalid for excessive breadth, and for a vague generafit.v which makes them a constant-menace to all inventors who may ;;; -U to improve the art of telephony by applying what Bell has taught ¡n, i<> ¡he practicability of operation by the constant and variable-contact method. By claiming ‘variable pressure of electrodes’ as an art or method, the complainant seeks to monopolize an indispensable feature of every ronulant-contact transmitter having varying contact and opposing pleci'roihr., and to suppress all subsequent invention in the same field.”

It seems to me iiui this language and the language of the court in the Incandescent Damp Patent Case are peculiarly appropriate to this case. I am of opinion that the complainant’s claim, with respect to its ojie,ration on ores or compounds, must he narrowly construed, am; iliai it does uot cover the claims of Hall.

No useful ptupose will be served by further consideration of these question;., oi, any of the remaining points or contentions presented by the very /able arguments of counsel. Tlic argument and briefs of complainar/i’s counsel most strenuously insist that Bradley was a pioneer ia -he art, that he was the first to fuse and electrolyze by simnkew-'ms application of electrical energy, and that by his method progressive fusion and electrolysis are made practicable. Doubtless that is so. It is not disputed that Bradley pointed out a way by whi.di the electric current, when regulated as described by the patent, will perform two essential features of his process. This was new in the i'ir¡. although >c-ir Humphrey Davy by his experiments suggested electricity as the common agent for fusion and electrolysis. The concepiion and practicability of utilizing the electric current for these purposes simultaneously was unlike the prior art. Hall, ambitious, vigorous, and intellectually strong, competed, experimented, and produced a process in aid of decomposition of refractory ores not contemplated by the complainant’s patents. The same purpose that Bradley had in view is accomplished by the joint action of certain compounds when subject to certain chemical and electrical action. The admixture of material and the chemical and electrical action applied thereto by Hall’s process are essentially different.

Little attention has been given to the controversy arising by reason of initial fusion by an arc or by incandescence. Nor has it been intended to pass on the validity of the various claims of the Bradley patents. To review all the questions urged upon the attention of the court would extend this opinion beyond useful limits. My purpose has only been to determine the issue of infringement, and give such salient reasons as seemed to me apparent from an examination of the record and arguments presented.

The motion made by defendant for leave to amend its answer, reserved by Judge Coxe until the final hearing, is allowed. Although it has not been considered in the determination of this case, the proposed amendment appears to be material, and may be added to the original answer, as provided by equity rule 60. The motion by complainant to suppress testimony of Romáin C. Cole is granted. This testimony was given in another suit between this defendant and complainant’s predecessor, a corporation. Such privity or mutuality between the parties to this suit is not found as will justify retaining in the case the testimony of Cole, deceased. It is therefore stricken out.

It must therefore be held that the process used by defendant under the patent issued to Hall, No. 400,766, does not infringe patents granted to Bradley, Nos. 464,933 and 468,148. The bill is dismissed-.