Opinion
No. 4070.
December 23, 1929. Rehearing Denied January 30, 1930.
Appeal from the District Court of the United States for the Western District of Pennsylvania; Frederic P. Schoonmaker, Judge.
Patent infringement suit by Alan Kissock against the Duquesne Steel Foundry Company and another. From a decree dismissing the bill, plaintiff appeals. Affirmed.
Christy Christy, of Pittsburgh, Pa. (Drury W. Cooper and John D. Morgan, both of New York City, and Harvey Lechner, of Philadelphia, Pa., of counsel), for appellant.
Warfield Watson, of New York City (Lawrence Bristol, of New York City, and Paul N. Critchlow, of Pittsburgh, Pa., of counsel), for appellees.
Before BUFFINGTON, WOOLLEY, and DAVIS, Circuit Judges.
The plaintiff appeals from a decree of the District Court dismissing his bill for infringement of Reissue Patent No. 16,396, granted to him on July 27, 1926 for a process of making alloy steel. Direct infringement of all claims was charged against Duquesne Steel Foundry Company and contributory infringement against Molybdenum Corporation of America. The court, passing by some of the defenses, found the reissue patent void because for an invention other than that disclosed by the original patent, or, if technically valid, still void because anticipated and for want of novelty in view of the art.
The invention disclosed by the specification (with the critical words italicized by us to emphasize its elements and bring out the main points in controversy) is as follows:
"In the manufacture of steel, alloyed particularly with molybdenum, the usual practice is to add or introduce into the fused steel the molybdenum in the elemental form or in the form of a high grade alloy of molybdenum, usually with iron, known as the ferro state, or ferro-molybdenum. This involves as a preliminary step, in order to obtain the molybdenum in the elemental or ferro state, the reduction of the ores or salts of molybdenum to the metallic form, this being usually done in an electric or other suitable furnace.
"The oxides of molybdenum combine with the oxides of certain other elements, more especially with certain of the alkali metals and the alkaline earth metals, of which calcium oxide may be regarded as an example, to form a salt of the general type known as molybdate of calcium or other such element. In the case of these particular elements just enumerated these salts would be of the general form CaOMoO3.
"At the present time, I prefer to employ the salts of calcium, molybdenum being the constituent element of the salt which reduces to produce the alloying metal for the steel. These salts of calcium can be produced from the ores of molybdenum without the use of furnacing, and often these salts are necessarily produced in the ore-reduction process. That is, the ferro state may be obtained or produced by furnace reduction of their calcium salts which have in turn been produced from the ores.
" The present invention obviates the preliminary or preparatory reduction of the ores or salts of molybdenum to the metallic state, and provides for introducing the alloying molybdenum directly from the salt thereof into the steel."
Kissock next states the thing which, he says, distinguishes his invention from prior art practices, as follows:
"The procedure under my present method is essentially that instead of reducing the ores of molybdenum to the metallic state in an outside furnace, and then introducing such alloying element into the molten steel, the calcium salt of molybdenum is added directly into the steel melting furnace."
He then describes what happens in the furnace:
"The carbon content of the steel and its bath, which carbon is actually a previous alloy of the steel, reduces the calcium salt of the molybdenum, the so reduced molybdenum alloying directly with the steel. (Note by the court. The word "reduce" used in this connection means the separation of the molybdenum in metallic form from the calcium and oxygen with which it is combined.) If desired, other carbon may be added to, or substituted for, the carbon content of the steel and its bath. The calcium oxide which is a constituent of the salt fixes the molybdenum and prevents its volatilization, thus avoiding loss, and presents the molybdenum capable of alloying directly with the steel in the melting furnace. Some suitable reducing agent other than carbon, such as silicon, which also may be a previous alloy of the steel, may be employed in the melting furnace. There are thereby eliminated the outside reduction of the ores or salts of molybdenum, with the attendant time, cost and labor, and large furnace losses of this costly element are likewise avoided."
The claims cover the invention in varying degrees of breadth. The first speaks of calcium molybdate as a constituent of the steel alloy reduced by carbon as a reducing agent; the second is the same as the first without naming the reducing agent; the third mentions a salt of molybdenum without limiting it to calcium molybdate and without naming the reducing agent; and the fourth calls for a calcium salt of molybdenum with carbon or silicon as the reducing agent.
To show novelty, the plaintiff (appellant) states in his brief that the prior practice of making an alloy of steel and molybdenum was that described in the specification. To show utility, the plaintiff asserting that his invention is of great practical value and has revolutionized the industry, points to the testimony of a very respectable number of metallurgists and steel makers who, first doubting its practicability, found it entirely operative in experiments and highly useful in actual practice.
On the presumption of validity arising from the grant of the patent, implicit in which is the element of novelty, and on the evidence of utility indicated by its commercial use, we should, if that were all the case, have no trouble in finding the subject-matter a true invention. But unless the process be both new and useful, it is not invention. Its public reception, no matter how favorable, will not take the place of novelty, if lacking, for such reception figures only when invention is in doubt. Duer v. Corbin Cabinet Lock Co., 149 U.S. 216, 223, 224, 13 S. Ct. 850, 37 L. Ed. 707; McClain v. Ortmayer, 141 U.S. 419, 429, 12 S. Ct. 76, 35 L. Ed. 800; Package Machinery Co. v. Johnson (C.C.A.) 246 F. 598, 602. Therefore we shall not consider the commercial use of the patented process or place it in the scale unless a doubt as to invention should arise.
On the issue of patentability the defendants challenge the novelty of the process. They say that the alleged invention had been disclosed in patents and described in printed publications in this and other countries and had been in public use in this country for more than two years before Kissock conceived and patented it; and that, anyway, its subject-matter was within the common knowledge of those skilled in the art to which the patent relates.
With the issue thus drawn we turn to the history of the invention and then to the art.
The event in its history, next to the last, was the grant of Letters Patent No. 1,300,279 to Kissock on April 15, 1919 for a process of making alloy steel. Two of its claims and the entire specification were identical with those of the reissue patent in suit except for the inclusion of more than one chemical element to be treated and the necessary grammatical differences in expression. Instead of dealing with salts of molybdenum alone as in the reissue, the first or original patent dealt with the alloying of steel by the introduction specifically of salts of the alloying elements of tungsten, chromium, vanadium and molybdenum, and generally of salts of other members of the fifth and sixth groups of the periodic system. Kissock thought, honestly no doubt, that he had invented a process applicable alike to all these elements. And so did the Patent Office for it granted him a patent covering all of them. On that patent Kissock brought suit in the same court against the Duquesne Steel Foundry Company, defendant in this suit. The Molybdenum Corporation of America, as here, intervened and defended. When the defendants filed their answer showing prior patents, publications and practices covering the same process in its relation to tungsten, chromium and vanadium, yet not extending, as he thought, to molybdenum, Kissock, realizing that his patent would not stand in view of the prior art thus disclosed, surrendered it and applied for a reissue under section 4916 of the Revised Statutes (35 USCA § 64) limiting his invention to the process of making alloy steel with the alloying element of molybdenum alone. The reissue was granted (the last event in the history of the invention) and on it Kissock brought this suit, consenting to a decree dismissing the first suit.
Coming to the art, the defendants, aside from their claim that the reissue for a process limited to the single alloying element of molybdenum is invalid because anticipated in respect to that specific element, contend that the practice of Kissock's process in respect to tungsten, chromium and vanadium as disclosed by prior patents, publications and uses and as disclosed also by his own original patent, and admitted by him on surrendering it, constitutes art prior to the process of the restricted reissue. This of course the plaintiff vigorously contests, raising an issue which should be met at the threshold.
Ordinarily a court would hesitate either to sustain or strike down a process patent concerning a chemical element merely because that element is, in characteristics and behavior, similar to other elements embraced within a recognized chemical group or family, General Electric Co. v. De Forest Radio Co. (C.C.A.) 28 F.2d 641, 647, just as a court would hesitate to acquit or convict a person on a family trait. But in this case the situation is different, not because tungsten, chromium, vanadium and molybdenum are recognized members of the fifth and sixth groups of the periodic system, but because Kissock in his original patent grouped them as being alike in their characteristics and behavior when subjected to his process. In other words Kissock by his original patent disclosed a process which he said applied alike to all these four elements and, when practiced, would produce like results chemically, operatively and financially. When he included them in his original patent he made them a part of the art of his process, even if they had not been there before, and when he surrendered his original patent he left three of them in the art still connected with the process he had patented in respect to them and had abandoned. Therefore we are constrained to look upon tungsten, chromium and vanadium with an eye to what had been done with them in the art, and what Kissock himself admits had been done by himself and others, for if Kissock's process, which under his original patent applied to all of them, had already been practiced with respect to three of them — tungsten, chromium and vanadium — they constitute an art which, if not precisely prior art, is certainly a kindred art that, in the circumstances, cannot be overlooked in determining whether Kissock's reissue process was novel and involved invention in view of that art.
Whether the reissue is for an invention other than that disclosed by the original patent and is for that reason invalid, as held by the learned trial court, is a matter which, in view of our conclusions on other points, we shall not decide. The substantial question is whether the reissue, assuming it technically valid, is for an invention in view of the art.
To prove that it is invention the plaintiff, first conceding that the alloying elements of tungsten, chromium, vanadium and molybdenum chemically belong to the same family and have some properties in common and in certain respects behave alike, distinguishes between the first three elements on the one hand and molybdenum on the other, particularly in their different degrees of volatility in the furnace. He maintains that oxides of molybdenum are highly volatile while oxides of tungsten, with or without calcium oxide, are not volatile at all; that, somewhere between these extremes, chromium, vanadium, and titanium, other members of the family group, have strong affinities for oxygen and are volatile in varying degrees. However, drawing the distinction between molybdenum and the other elements as to their points of volatility, he maintains that by reason thereof they presented different problems as alloying elements and that he solved the problem in respect to molybdenum by using calcium oxide which, a constituent of the salt, " fixes the metallic alloy element and prevents its volatilization, thus avoiding loss," etc. This, he says, is the essence of his invention. Yet he said the same thing in the same words in his original patent with reference to fixation of tungsten, chromium and vanadium against volatilization. There he made no distinction between the characteristics and furnace behavior of tungsten, chromium, venadium and molybdenum but recognized their full equivalency in the process. Had molybdenum presented a different problem the subject-matter of the reissue might be invention, but as Kissock himself, in the original patent, said in effect that the one process solved the problems of all four elements, we have been unable to find any difference in the problems or in the means to solve them. While Kissock asserts and vigorously stresses the theory of fixing molybdenum against volatilization by combining calcium oxide with it, and, in his brief, points to that as the essence of the invention in the reissue patent, we are constrained to accept his same assertion in respect to the other elements in his original patent and believe that the same fixation against the same undesirable action occurs by combining calcium oxide with tungsten, chromium and vanadium, producing combinations long known as calcium tungstate, calcium chromate, calcium vanadate and calcium molybdate. Thus we are back to the original patent for a process to do the same thing with all four elements, which raises again the question whether there is invention in picking out one element of that group and limiting the process to it.
Kissock has conceded that calcinated salts of tungsten, chromium, vanadium and other elements of the same chemical group, except molybdenum, had before his invention been used directly in the furnace as alloying elements in place of ferro alloys in the manufacture of steel alloys, thus avoiding, in respect to them, the very intermediate operation to bring the elements to a pure or ferro state which he claims is the economic feature of his invention in respect to molybdenum. Or, stated differently, as Kissock has admitted that all he invented in respect to molybdenum was old in the patent and practical arts in respect to tungsten, chromium and vanadium, we shall not review those arts but, accepting his concession, address the remainder of this discussion to the use of calcium molybdate, the single subject of the reissue patent, as an element in the process of alloying steel.
Back in 1911, H. Mennicke published in German a text-book in which among other things he disclosed processes having great similarity to those for the preparation of tungsten, including among them processes for the preparation of ferro molybdenum. Of the formulæ he gave, there was one for magnesium molybdate, magnesium being an alkaline-earth metal within the terms of the reissue patent, and another for the production of ferro-silicon, silicon also being an earth metal within the reissue patent, in the presence of lime (calcium), thus:
"(c) Reduction of MgMoO4 (magnesium molybdate) by carbon in the presence of iron oxide as well as kaolin.
"(d) Reduction of MoO2 of 53 per cent. of ferro-silicon in the presence of lime, as CaMoO4 in the electric furnace. The ferro-silicon is first melted or produced primarily in the electric furnace."
Kissock does not limit his process to a particular kind of furnace. It includes open hearth, basic or acid, and electric. The lime of which Mennicke speaks is the calcium of the patent and the resultant is a molybdenum salt or calcium molybdate. Ferro-silicon is, like carbon, a reducing agent permitted by the patent. Kissock resists whatever force this publication may have by saying that it contains no disclosure of the use of calcium oxide or calcium in chemical combination with an oxide of molybdenum to "fix" the latter against volatilization in the furnace, which is true, yet in composition and operation it may be similar to, or close to, Kissock's process without his theory.
Although the defendants' expert testified that Kissock's reissue patent is nothing more nor less than Mennicke's disclosure of a process for the direct reduction of calcium molybdate or magnesium molybdate by carbon or silicon at furnace temperatures and is anticipated by Mennicke, we are not inclined to follow him. Nor are we inclined to follow the plaintiff in his contention that Mennicke did not disclose calcium molybdate "nor any molybdenum compound in making molybdenum alloy steel." What Mennicke gave the art with respect to molybdenum was "some data of the comparison to the analogous conditions of tungsten" and, showing how magnesium and calcium molybdate can be made, told the art that "as an addition to steel, both metals (tungsten and molybdenum) produce similar properties." This, if not the starting point, was a contribution to the process which later the art, or Kissock, invented and practiced. Its value to the art, while limited perhaps, cannot be ignored in appraising the conception which Kissock says was his alone.
In 1914 the United States issued patent No. 1,119,643 to Byramji D. Saklatwalla for "process of producing alloy steels." The process of this patent is so like the process of the patent in suit that we shall quote the relevant parts of the specification and, to shorten discussion, italicize critical words and bracket comparisons with Kissock's specification. The patent states:
"This invention relates to the production of alloy steels containing one or more of the following alloying elements, namely, nickel, covalt, chromium, manganese, tungsten, vanadium, molybdenum, titanium, uranium, and the like." (Members of the fifth and sixth groups of the periodic system referred to by Kissock in his original patent.)
"One object of my present invention is to provide a method of producing alloy steels, directly from raw materials containing such alloying elements, without the necessity of reducing such raw materials to the metallic state, either in the form of ferro-alloy or pure metal before mixing such alloying elements with the steel." (The main object of Kissock's invention.)
"A further object is to provide a method whereby alloy steels could be produced with a higher degree of efficiency, reducing the usual loss of the alloying element." (Another object of Kissock's invention.)
"My present invention obviates these defects inasmuch as it reduces the alloying element from raw materials directly in the body of the finished steel thus avoiding intermediate losses." (Kissock says that his invention obviates the same things.)
"Further I use a reducing agent, namely, silicon," (Kissock says, some suitable reducing agent other than carbon, such as silicon, may be employed) which is always used as a deoxidizer in the manufacture of the alloy steels just before the addition of the ferro-alloy or pure metal.
"My improved process is carried out substantially as follows: The raw material containing the alloying elements, generally in the form of an oxid, and consisting of an ore or product of a previous metallurgical operation, after being pulverized, is mixed with ground silicon metal or ferro silicon, or other silicon bearing material and pressed into briquettes. For example, in the manufacture" of steel with any one of these alloying elements, the oxid, "obtained by roasting the matte, is mixed (in a named proportion) with ferro silicon (of named fineness and weight). To this mixture is added half the weight of the ferro silicon used in a mixture consisting of equal parts of lime (calcium oxide of Kissock's process) fluorspar and soda ash for the purpose of fluxing the silicic acid formed during the reduction."
"Briquettes obtained in the above described method are introduced into a molten charge of steel either in an open hearth furnace, crucible, or ladle." (That is what Kissock does with his granular molybdate.)
This is Kissock's practice of obtaining a salt of molybdenum and charging it directly into the furnace except that, instead of using calcium oxide alone, it calls for fluorspar and soda ash in addition. The plaintiff meets this patent not by denying that Saklatwalla disclosed the introduction of a salt or oxide of molybdenum directly into the furnace, thus avoiding the intermediate step, but by saying that Saklatwalla did not disclose the "fixed" oxide or salt of the Kissock patent to avoid volatilization in the molten bath but that he used it for another purpose, that is, to flux the silicic acid formed during the reduction. True, the patentee did say the function of lime, fluorspar and soda ash is to flux the silicic acid, but if he had the same ingredients that Kissock had and used them in the same way, then, even without a conscious purpose, he patented the same means of fixing the alloy against volatilization. In other words, it may be that, wholly without a declared purpose, Saklatwalla disclosed Kissock's process. It is the process, not the theory of its operation, that is patentable.
Although the two processes seem, in elements, functions and objects, to be almost identical, we hesitate, in view of the subtleties of chemistry, to hold that Kissock is anticipated by Saklatwalla in the strict sense of the law stated in Skelly Oil Co. v. Universal Oil Products Co. (C.C.A.) 31 F.2d 427, but we have no doubt that the step from Saklatwalla to Kissock, if a step it were, and the step away from prior like processes, having to do with the related alloying elements to which Kissock first linked his process, did not involve invention.
Holding the reissue patent invalid for want of invention in view of the art, the decree is affirmed.
BUFFINGTON, Circuit Judge, dissents.