United States Mitis Co. v. Midvale Steel Co.

135 F. 103 | U.S. Circuit Court for the District of Eastern Pennsylvania | 1904

ARCHBAL,D, District Judge.¹

The process for making wrought iron and steel castings which is the subject of this suit was invented by Carl Gustave Wittenstrom, a Swede. It was patented in Great Britain, France, and Belgium, July 8, 1885, and in the United States December 29 of the same year, and, while the patent has now expired, it had not at the time suit was brought, and the complainants, who are the owners, are entitled to damages if it was infringed while it was in force. Chinnock v. Paterson, 112 Fed. 531, 50 C. C. A. 384. Infringement is denied, and the validity of the patent vigorously assailed, and these are the questions, therefore, which are presented for determination. The patent was considered and sustained by Judge Acheson in a suit against the Carnegie Steel Co. (C. C.) 89 Fed. 343, and it would be sufficient, so far as the one issue is concerned, to rest upon the opinion which he expressed; but, as it is contended that the matter is somewhat differently presented here, an independent examination of it in the light of the present argument will be undertaken.

The process described in the patent is a simple one, consisting merely in the addition of a small piece of aluminum to the iron or steel after it has been fully melted, and just as it is about to be poured into the mold. The object sought to be attained is the making of a solid and more perfect casting, free from superficial blisters and internal cavities known as “blow holes.” The great difficulty in securing this without a deterioration of the product was the problem to which the mind of the inventor was addressed, and his invention consisted in successfully solving it. High-water mark in steel casting up to that time is admittedly represented by the so-called “Terre Noire” process, patented in France in May, 1876, and in the United States in April, 1879. It consisted in the organized use of manganese and silicon, with specific reference to the casting, as distinct from the refining art; but, much as it added to existing metallurgical knowledge, it had its recognized limitations. While the resulting product was undoubtedly more solid and homogeneous, its quality was affected, being made more brittle or red short. The process was only serviceable where the steel was hard,' and the preservation of its intrinsic character was not important; and it was entirely inadequate, therefore, for the production of castings of wrought iron or mild steel, of which there were none at the time Wittenstrom brought forward his invention. Referring to this state of the art in his specifications, the inventor says:

“It Is well known that one of the great difficulties in making castings from steel is to get a product which is solid, sound, homogeneous, or free from blisters or cavities. Lately the manufacture has been much improved by adding to the metal fero-manganese and other compounds containing carbon, silicon, and manganese; but, although all these admixtures make the product somewhat more solid, they deteriorate the quality in other respects, as the product gets harder and more brittle or red short.”

Proceeding thereupon to disclose the discovery which was the basis of his invention:

“I have found,” he says, “that eastings of wrought iron or mild steel may be obtained solid without changing the intrinsic quality of the metal by the addition of the metal aluminium either alone or in the shape of an alloy, such addition to be made after the iron or steel has been melted, and preferably just before the pouring is commenced.”

With regard to the quantity to be used he states:

“I have found that the use of a minute quantity, never exceeding one per cent, by weight—preferably from one-fifth to one-tenth of one per cent, by weight—of metallic aluminium added to the molten iron has the desired influence, and even a very much smaller percentage has an appreciable influence.” “By this, my new invention,” as he declares, “I have succeeded in making perfect castings from the softest wrought iron, which castings in every respect retain their ductility and nature of wrought iron, though their tensile strength is greatly enhanced.”

In thus making possible that which was not so before, there can be little question that the inventor rendered a material service to, and made an important advance upon, the existing art.

An effort is made, however, to limit the invention by the suggestion that all that was in mind was the production of an aluminum alloy, use being made of the well-known principle that the alloy of two metals has a lower melting point than that of either of them, the fluidity necessary to make a proper casting being thereby obtained without .injurious superheating. This contention is based upon that part of the specifications where it is said:

“The melting point of aluminium is about 800 Fahrenheit, and the effect of such addition [reférring to the aluminium] is to lower the melting point of the misture, and thereby render it more fluid (as it at once becomes superheated) so that the gases in the metal pass away easily, the metal runs freely into the mold, and a more perfect product is obtained.”

The object in thus seeking to narrow the scope of the invention is to deprive it of other beneficial results, particularly the deoxidizing of the metal, which now, according to the better opinion, is the real effect of the addition of the aluminum, and for which purpose, in the defendants’ ingot practice at least, it is admittedly employed. It was not a theory, however, that was patented, but a process; and, even if a theory was advanced to explain it which was erroneous, the inventor is not precluded from laying claim to all the benefits legitimately to be derived therefrom. Eames v. Andrews, 122 U. S. 40, 7 Sup. Ct. 1073, 30 L. Ed. 1064; Thomson Meter Co. v. National Meter Co., 65 Fed. 427, 12 C. C. A. 671. But the truth is, no mistake was made. While the invention no, doubt fundamentally consists in increasing the fluidity of the molten metal, the inventor unquestionably had in mind the deoxidizing effect of the aluminum as an essential part of it, and did not stop short with the idea of simply creating an alloy. That idea was thrown up to him in the course through the Patent Office, as disclosed by the file wrapper, and distinctly repudiated.

“This invention,” he there says, “does not relate to the production of a new alloy, or a new method of producing iron or steel. * * * My invention makes use for the production of castings from wrought iron or steel of one quality of metallic aluminium which I have discovered, * * * that of making the molten iron or steel highly fluid, * * * and this quality is of the utmost importance when making castings, as the metal will run freely into the finest parts of the mold.” “The quality possessed by aluminium of rendering wrought iron fluid enough for castings without changing it to steel X regard as my discovery.”

The reducing of the melting point on the other hand is referred to as a distinct, if not subordinate, feature, as follows:

“Another important feature in my production of castings in wrought iron is that by the trifling addition of aluminium the melting point is lowered; that is, whereas the wrought iron alone would solidify by a very slight decrease of temperature, it will, after the addition of aluminium, stand a considerable cooling before solidifying. This is of great importance, as it gives time for casting the wrought iron in a practical way into a series of molds.”

The rejection of the idea of' an alloy is further shown in the patent itself, where the inventor disclaims the purpose of obtaining any product made up of aluminum and iron, such as “Wootz steel,” or any product in fact in which the aluminum and iron or steel are fused together. In that connection he explains that, while “a superheated state of the metal is essential for the practical performance of casting into several molds,” yet “by adding the aluminium to the wrought iron or steel, and fusing them together, a superheating would result in injury, as the metal would become red short, or take up gases, whereas by first melting the iron or steel, and then adding the aluminium before pouring, the ‘super-heating’ (if it may be so called) produced by a sudden lowering of the melting point does not injure the metal.” That the inventor understood that the addition of aluminum when the metal was in a molten state had the effect of deoxidizing it, or freeing it from gases, is also disclosed by the file wrapper. “Furthermore,” as it is there said in one of the communications, “the sudden creation of a greater fluidity just before pouring allows all gases to pass easily away; whereas, if the aluminium is melted together with the iron, some gases are always taken up during the melting, and the product becomes unsound.” This is repeated in the specifications in the very part relied upon by the defendants, where, speaking of the lowering of the melting point, and the rendering of the mixture more fluid, it is pointed out, that as the result “the gases in the metal pass away easily, the metal runs freely into the mold, and a more perfect product is obtained.” But it is not necessary to prolong this part of the discussion further. Without intending to withdraw the force of anything that has been said, even if the deoxidizing effect of the process was more dimly disclosed than it is, a certain liberality of treatment is to be accorded it, sufficient to protect this feature, the inventor having claimed, as he has, as the particular desideratum and aim of the process, the increased fluidity of the metal, which is now recognized as resulting from the removal of the oxygen through its affinity with the aluminum by which the mass is made less wild. And although he may have had in mind the production of a momentary or theoretical alloy, as a means for reducing the melting point—which is certainly as far as it can be carried in view of what he declares to the contrary—this is not the exclusive idea, and the invention is not to-be confined to it.

But it is urged that, even assuming the patent to cover all that is claimed for it, -the deoxidizing effect of aluminum, the same as that of manganese and silicon, was well understood, and its use as a reagent for removing the gases from iron involved nothing patentable or novel. Its price, it is said, was prohibitive prior to the date of the invention, and it has come to be used since then simply because this fell and made.it possible. Nothing—in this view— was contributed by Wittenstrom to the metallurgical art, or at least nothing in practical results; the theoretical interest at first aroused in his so-called discovery disappearing when it was found out how little he had done. But the fact is that, whatever had been the case with steel, wrought-iron castings had not been made up to the time of the invention. And if Mr. Ostberg, at whose works in Sweden the experiments were made which brought about the discovery of the process, is to be believed, the metallurgical world was astonished at the result, most of the papers reproducing his Pittsburg address, and the president of the British association at the next annual meeting making a part of it his own. It will hardly do in the face of this to say that nothing was contributed by the inventor to the art in which it stands, the high testimony to the contrary given at the time being safer to be trusted than expert opinions, after long familiarity with it, controversially expressed. Nor can the extended use into which it has gone since, which equally disproves the assertion, be entirely attributed to the cheapening of the price of aluminum, however much it may have been aided thereby. Furthermore, whatever knowledge of the deoxydizing property of aluminum there may have been, its utility in producing iron and steel castings without a deterioration of the product certainly was not understood before, nor yet the time and manner of applying it, which were also involved. One had to know not merely that aluminum was a deoxidizer like manganese and silicon, but that, in marked contrast with both, it could be made to disappear in the process, and leave no bad effect. This it was left to Wittenstrom to discover and give to the world, and it does not do to belittle his achievement after the fact. The process which he evolved may be a simple one—merely casting a bit of aluminum into the molten mass at the moment of pouring—but it is not to be judged by its simplicity, but by its effect (Carnegie Steel Co. v. Cambria Iron Co., 185 U. S. 403, 429, 22 Sup. Ct. 698, 46 L. Ed. 968), and of this we can hardly doubt.

But, again, it is said, the Marbeau and Abel patents disclose the use of aluminum in iron compounds in the same minute quantity (the fraction of a per cent.), at the same point in the process (the moment of pouring) for the same purpose (the removal of the last vestiges of oxygen), and with the same object in view (a more solid and homogeneous product); and that it involved ,no invention to extend this process bodily (as was practically done) to iron and steel. The Marbeau patents, to which reference is thus made, consist of a series or group of patents, embodying the discoveries of the eminent French metallurgist Henry Marbeau, taken out in France in 1883 in the name of the “Société Anonyme Dite Fonderie de Nickel et Métaux Blanc,” and carried down to January, 1885, by sundry certificates of addition, with intermediate partial counterparts in England and the United States. There were also two independent English patents of October, 1885, and March, 1888, respectively, and a still later one, taken out in the United States by Marbeau himself in 1890. Although much is made of these patents by the defendants as establishing a prior analogous use, I am far from convinced that they are to be given any such effect. The subject to which they are addressed is not the casting of wrought iron or steel, but the manufacture of ductile and malleable nickel and cobalt, and the compounds, known from the admixture of iron, as fero-nickel and fero-cobalt. The use of a small quantity of aluminum at the moment of pouring is, indeed, suggested; and the asserted purpose is to profit by the properties thereby contributed, in order to produce a better molded casting when the resultant product should come to be so used; although the facilitating of a homogeneous alloy of the metals is also brought forward. But, in addition to the fact that the process is thus made to relate to the manufacturing, if not to the alloying, in contradistinction to the casting art, which is metallurgically important, the subject is always treated from the nickel, and not the iron, standpoint, which is recognized by the inventor as controlling, éven where, as in the formula given in the British patents of 1884 and 1885, the relative proportion of the two metals is reversed, the nickel being reduced to 25 per cent, and the iron raised to 75. For proof of this we have his own words in the United States patent of 1890, where, distinguishing the process embodied in the latter from those which had preceded it, he says:

“Heretofore I have obtained fero-nickels and steel nickels with a high percentage of nickel, varying from ninety-nine per cent, to twenty-five per cent, of nickel, possessing the peculiar properties of the latter metal, such as brilliancy, incapability of oxidation, etc., and which are capable of being substituted for nickel for many purposes. * * - * In pursuing experiments in this direction I have now succeeded, by reducing the percentage of nickel below twenty-five per cent., in producing a series of alloys, which, although belonging to the same [general] class as those formerly described and patented, are possessed, on account of their constitutive elements and mode of production, of new properties, and constitute a distinct class of alloys, forming a new manufacture.”

It is true that it is somewhat arbitrary to say that, when the percentage of nickel is 25 or under, the alloy may be compared to steel and iron, and when above that not; nor do we have to accept his say-so for it. But that is aside from the argument. The contention is that the device of Wittenstrom is merely the case of an analogous use, so prefigured by that which had preceded it, and particularly by the Marbeau patents, that any one of ordinary metallurgical skill would have had no. difficulty in extending the process which is there disclosed for the manufacture of nickel and cobalt and their compounds to the making of castings of wrought iron and steel, the same deoxidizing property of aluminum being employed in the same way and at the same time. But here we have the most significant proof—whatever may be now asserted to the contrary, when we have the full light of subsequent advance thrown back upon it—that one who was deep in the subject did not grasp the possibility of this until the very last; Wittenstrom in the meantime having developed and perfected it. Marbeau may have been advancing along the same lines as Wittenstrom, but this does not deprive the latter of the merit of his invention, nor establish that his discovery was so little beyond that which was well understood as to be practically obvious. It is proved by Marbeau also that the exact influence of the common re-agent employed by both was not a matter of ordinary knowledge. Speaking of its action in his British patent of 1885, he says:

“And, lastly, the aluminium prevents a change in the molecular condition of the iron; that is to say, the latter does not lose its original fibrous condition and pass into a crystalline state in consequence of its fusion with the nickel.”"

This is far from grasping the deoxidizing effect of this element such as is now recognized, and it is emphasized by the fact that in the same connection he does attribute to manganese and fero-cyanide of potassium the expulsion of the oxide as fast as formed from between the molecules, equivalent, as he says, to a constant molecular scouring. Undoubtedly nickel and cobalt, with their alloys, are to be classified chemically, and to a certain extent, it may be, metallurgically, in the same group with iron, including the differentiated form of it, difficult to define, which is known as steel. But if there are resemblances between these metals, there •are also decided differences, which would seem to preclude the possibility of similar treatment, relating as they do to the behavior of each with carbon and oxygen, and particularly with certain common deoxidizers. Metallurgical processes, moreover, often depend on delicate and obscure reactions, which cannot be forecast with certainty, and with regard to the exact character of which there is not always complete agreement in the end. The present state of the art has been brought about by prolonged and painstaking experimentation, and it is to this, rather than to analogy, that resort must be had to determine what will be efficacious in any given case. To the oft-repeated argument, therefore, that any one with ordinary skill could have accomplished all that was done by the ■inventor by the mere application of existing knowledge, it can with even more than the usual confidence be said in the case before us that no one up to that time had done so, in proof that his discovery was a real achievement. If the present process .was an obvious one when Marbeau brought forward those which he discovered, why, pray, did he stop short at nickel and cobalt, and not go on to steel and iron, which were quite as important metallurgically? He did ■so, as we have seen, in the end, in the course of his progressive work; but that he did not, at the outstart, nor until Wittenstrom had independently produced the process which we have here, goes far to prove its entire novelty and inherent invention.

The Abel British patent of 1884 is also relied on, but, if those of Marbeau do not suggest the process in suit, it can hardly be claimed that this does either. It also relates to the manufacture of nickel and cobalt (without mention of their iron compounds), and a small quantity of aluminum is employed at the termination of the process to remove the oxygen, by which means, as it is said, compact and malleable castings of the best quality can be obtained. But substantially the same answer is to be made as to the patents to Marbeau. The analogies are not so close between nickel and cobalt on the one hand and iron on the other that the metallurgical treatment found efficacious for the one can be extended to the other without investigation, leaving room for inventive discovery such as we have here.

The validity of the patent being established, the question of infringement remains. So far, at least, as the defendants’ ingot practice is concerned, this is clear. The molding of ingots is certainly a species of casting, and, use being made of aluminum in the manner and for the purpose described in the patent, the process is appropriated and the patent infringed. This is conceded if it is held to extend to the use of aluminum as a deoxidizing agent, and it was for the purpose of escaping from this that effort was made to have it construed simply as intending to produce greater fluidity by the formation of an alloy. When that was disposed of, therefore, infringement stood confessed. Nor is this overcome by the suggestion, already considered, that the deoxidizing properties of aluminum cannot be monopolized by a patent, being inherent and well known. The process in controversy consists not simply in making use of these properties, but in making use of them at a certain time, in certain quantities, and in a certain way, all of which have been duplicated by the defendants, who therefore unquestionably offend.

It is strenuously denied, however, that such is the case with regard to their commercial castings, in which the aluminum is employed, not, as it is said, to make a more perfect casting, which was not called for, the defendants having long since succeeded in making these castings of the highest grade, and free from blowholes, without it, but for the purpose of preventing leakage about the stopper.of the ladle in passing from one mold to another, for which it had been found effective. The defendants, as it is testified, began to make use of aluminum in 1895, being led into it because it was said to produce good skins or surfaces, and their competitors were doing it; but after a series of intermittent experiments they found there was little, if any, improvement from it, and so gave it up. They discovered, however, as they say, that when it was used they secured a good shut-off, and therefore resumed its use for this distinct and limited purpose. The theory on which this result is explained is that, while aluminum, when added to iron to remove the oxide which is present, makes it more fluid, passing off itself as alumina in the slag; when added after the oxide has been removed—as is the case as it is claimed at the time the defendants apply it—it remains in the metal, and makes it thick or viscous. On the other hand, it is maintained that the molten mass, when “wild,” before the aluminum has been added, has a tendency to cool and get in the way of the stopper, impeding a close fit; whereas by the addition of the aluminum, which admittedly makes it more fluid, it is enabled to retain its heat, and so avoid this difficulty. The weakness of the defendants’ theory is that it is based on the entire removal of the oxide before the aluminum is added, when experience teaches that it is almost impossible to effect this, the molten iron taking up oxygen from the very air to which it is exposed. It may be relatively free, but not absolutely so, nor able to be kept so. It is somewhat remarkable, moreover, that aluminum is employed by the defendants in their ingot practice in order to remove the oxide and quiet the metal, so as to obtain a more homogeneous and solid product, and yet, when the delicate matter of making commercial castings is in hand, in which the greatest fluidity is required in order that the metal may flow freely into every configuration and corner, aluminum is added, according to the defendants, for the deliberate purpose of thickening it, and that for the comparatively insignificant object of saving a little escaping metal. But I am relieved from the necessity of passing on the question of motives, or the relative value of opposing theories. The inventor is entitled to' every beneficial result legitimately flowing from his invention, and it does not detract from it that it accomplishes more than was expected if-that is true of it. Dealing, as we are, with a metallurgical process, with regard to which there is not a little difference of opinion, if not some obscurity, it would invite evasion and destroy the value of the patent by which it is protected, if, while pursuing its terms, the charge of infringement could be successfully met simply by assigning the attainment of a different object. The defendants, in their commercial, the same as in their ingot, practice, throw into the metal when molten the fraction of a per cent, of aluminum at the moment of casting. This follows the directions of the patent, and constitutes an infringement upon it. That it results in a more perfect product in the one case, the same as it admittedly does in the other, as designed by the inventor, can hardly be doubted. That it does more than this—if such be the case—by stopping the leakage of the ladle, as asserted by the defendants, does not entitle them to appropriate it upon the plea that this is their only object.

It is finally said that the complainants have not complied with section 4900, Rev. St. [U. S. Comp. St. 1901, p. 3388], which requires that articles manufactured under a patent shall be so marked. But, as was held by Judge Acheson in the Carnegie Case (C. C.) 89 Fed. 206, this requirement is inapplicable to the case of a process, nor is the omission a bar to a recovery where the defendant has been notified, and continues to infringe in disregard of it; and that is the case here. The defendants persisted in the use of the process after they were served with the bill, and would be liable for that, if nothing else.

Let a decree be drawn sustaining ’the patent and referring the case to a master to assess the damages.

1 Specially assigned.