Appeal from the District Court of the United States for the District of Connecticut.
Before L. HAND, AUGUSTUS N. HAND, and CHASE, Circuit Judges.
This is a suit in equity upon a patent for a process of plating metals with chromium. Infringement being established, the issues turn upon the validity of the patent, and that in turn upon whether it disclosed an invention. The judge held that it did, and gave a decree upon all the claims in suit; it is not necessary to consider the verbal variations between them, except, as appears hereafter, in regard to claim sixteen. Before describing the disclosure it is necessary to say something about the art at large. All electroplating involves the immersion, in a bath of proper solution, of the two poles -- the anode and the cathode -- of an electric circuit. The current passes through the solution to complete the circuit, and as a result the metal in the solution is deposited upon the cathode -- plates it. Thus the object to be plated must be the cathode. All this was a commonplace in other kinds of electroplating, silver, copper, nickel and the like; but the difficulties in regard to chromium were many, the most troublesome of which, at least in 1925, was the composition of the solution or bath; and it is about this that the suit principally turns. Chromic acid in solution had been found to be the most available form of the metal more than twenty years before Fink's application was filed at the very end of the year 1925.It is soluble in water and in anhydrous form consists of a molecule of chromium and oxygen in the relation of one atom of the first, to three of the second (Cr. O3); the current breaks down the molecule and sets free the chromium near the cathode. The art had added other substances to the solution, as for example sulphuric acid, but after 1920 had come chiefly to rely upon chromic sulphate, which is made up of chromium and a sulphate radical in the form indicated by the chemical symbol Cr2 (S O4)3, two atoms of chromium to three molecules of the composite radical. It is not clear whether the art had recognized that the function of this substance was catalytic, that is, to assisting the chemical reactions at which it was present, but not itself to enter into combination. However, Fink stood upon some ground already gained; he was not the first person who succeeded in electroplating chromium; nor did he claim to be. He did claim to be the discoverer of "a practical and commercially available process"; the question is whether he was justified.
The disclosure is unusually clear; it prescribes the usual detail of electroplating with "a chromium-carrying electrolytic solution, in the presence of a catalyst." The solution is of chromic acid of from one hundred and fifty grams to a litre to saturation, an entirely adequate description. The catalyst must contain an acid radical, stable in the bath, among which are suggested acids and salts having a sulphate, fluoride, phospate, or borate radical. Apparently the art has in practice only used the first, and to it many of the claims are confined. The catalyst is to be calculated from all sources, both the radical in the chromic acid -- which comes in as an impurity -- and in the substance added as catalyst, properly speaking. It is never safe to assume that commercial chromic acid is pure; the safest way is to test the solution of chromic acid and correct by adding or subtracting substances which will supplement or decrease the catalytic agent already there. Sometimes instable radicals are in the chromic acid; they will disappear with use, and thus, though initially calculated correctly, they must be replaced by stable radicals so as always to keep the proper ratio. That ratio is between four-tenths of one per cent. and two per cent. with an optimum of one per cent., which is what the defendant and the art generally has come to use with entire success. It has displaced generally, if not altogether, earlier processes of chromium plating.
The disclosure also contained a description of the creation of a hydrogen film about the cathode, designed to protect the chromium cation; that is, the chromium atom positively charged which seeks the cathode. As this is highly sensitive to oxygen, it will readily oxidize unless at the moment of its separation out of the chromic acid it is surrounded by a protecting hydrogen film. On the other hand the film must not be too thick to prevent access of the ion to the cathode. The proper thickness of the film is secured by regulation of the current density which varies with the temperature. As all this plays no part in the case, it will not be necessary to consider it further.
Having obtained his claims on April 20, 1926, Fink apparently became fearful that he might have occupied more of the field than was properly his; and about two years later (June 20, 1928), he filed a disclaimer by which he narrowed all the claims in suit but 3, 16 and 18, adding as an element that the radical must be "regulated * * * in maintaining the efficiency of the bath." With the defendant we read this as meaning that the claims will cover nothing but a process in which the bath is watched as the work goes on, and the proper proportion of the "radical" is always maintained. There is no dispute that the defendant does just this, and perhaps it is necessary to any economic use of the process. At any rate the result of the disclaimer is that an electroplater might prepare a bath after the recipe of the specifications and electroplate the cathode without infringing the claims to which the disclaimer applied; he would infringe only in case he in addition kept an eye on the radicals in the solution.
Much of the defendant's argument depends upon an illegitimate inference from the disclaimer; that is, that it conclusively concedes all the disclosure to have been in prior art except the regulation of the bath after it was formed. From this with much plausibility it continues that there was no invention in regulating any electroplating bath once it had been made; this was the commonest practice in the art generally. This we need not dispute; it would indeed seem a very plain thing, when success depends upon the proper proportions of the ingredients and the bath is used repeatedly, to take periodic samples of the solution and correct any variations which occurred. We do not however agree that a disclaimer has the effect asserted. It does indeed presuppose that the patentee fears he has claimed more than he is entitled to. In the case at bar Fink said that "he had reason to believe that * * * the specifications and claims * * * are too broad, including that of which said Fink was not the first inventor," though the statute (section 65, title 35, U.S.C. [35 USCA § 65]), speaks only of "claims," of disclaiming "such parts of the thing patented as he shall not choose to claim." It has never been held that the office of a disclaimer is more than to narrow the claim by the added element. The language in Dunbar v. Meyers, 94 U.S. 187, 193, 194, 24 L. Ed. 34, is indeed somewhat equivocal; part of it means that the claims alone are limited; another part seems to intimate that what is disclaimed is finally conceded to be in the prior art. But this part was in no sense necessary to the decision, and is not, we think, to be taken literally. In the only cases we have found in which the point was decided, it was held that a disclaimer had no such effect. Permutit Co. v. Wadham, 13 F.2d 454, 457 (C. C.A. 6);*fn1 Manhattan Co. v. Helios Co. (C. C.) 135 F. 785, 801, 802.
It seems to us that these decisions are right; the patentee has indeed abandoned a part of his original monopoly, and this because he fears that he has gone too far; his fears may not be justified. There is no reason to impute to his caution a declaration about what the prior art actually contains. And even if there were, why should it be more than an admission? It was not a means of procuring his patent; that had already issued; it stood or fell upon the facts as they were. His reasons for voluntarily yielding a part of what he got, cannot stop his mouth, unless it is unjust to speak. But there is no injustice so long as his monopoly is held straitly to the narrowed claims. It is never for instance a good defense to a claim that it might have been broader; the patentee need not at his peril claim all that he might; he does not concede that each element is necessary to avoid anticipation. He may prove an invention broader than he claimed; the unnecessary elements introduced need not be themselves an invention. Else it would be possible to invalidate any claim by showing that the patentee had yielded too much to the Examiner. The function of a disclaimer ends with the retrocession introduced; the patentee is still free to show that the claim covers only a part of what it might. In the case at bar the regulation of the bath was not the invention; it was a trivial part of the claims. What Fink really did was to single out the acid radical as the catalyst, disregarding the substance which happened to contain it. As we shall show, nobody had thought of this before, and for that reason nobody had found a dependable process. As is generally the case in chemical processes, the important thing was to eliminate the inessentials; until this is done practice must be hit or miss.
We may start with the disclosure of Carveth & Curry which appeared in 1905, more than twenty years before Fink filed his application, and almost as long before he claims to have discovered. The date alone is significant, because the extensive exploitation of chromium electroplating followed the appearance of Fink's patent, and it has been exploited very extensively indeed. We have no reason to suppose that it would not have been as welcome in 1905 as in 1926. Carveth & Curry were concededly competent chemists, as were those who followed them. Thus we have at the outset a history which is impressive; something had stood in the way of success until after Fink's invention was known. True, it might not be what he added, but there is some antecedent reason to attribute the change to him. Kirsch Mfg. Co. v. Gould Mersereau Co., 6 F.2d 793, 794 (C. C.A. 2).
These investigators used as one of their solutions about 14% of chromic acid to 1% of sulphuric acid, (H2 S O4); their results were not satisfactory; the deposit flaked off. As we now know the proportion of the radical was in any case far too high; but they had not discovered that the relevant proportion should be between the chromic acid and the radical itself; sulphuric acid contains it, but their paper did not suggest that it was the radical which counted. The proportion between sulphuric acid and chromic acid, even if it be the right one, is not a proportion between chromic acid and S O4. It is true that Carveth & Curry were aware of the difficulty of getting pure chromic acid and of the fact that the commercial products contained enough sulphuric acid and other impurities to be important in the result. But they produced no practicable process; they did not even profess to have done so; they were frank to say that other investigators must press their experiments further.
The next laboratory step was embodied in a paper of Sargent, a chemist of standing, which he read in 1920, fifteen years later. He acknowledged his indebtedness to Carveth & Curry and had worked extensively upon their foundation; his own results were the cue to whatever else was done till Fink completed his work, and to Fink's work as well. It is clear that he supposed it necessary to add chromic sulphate to the solution. His theory was that when the chromic acid used contained traces of sulphuric acid as an impurity, this united with the chromium and formed chromic sulphate. With this and with this alone, except in one unsuccessful case, he experimented; his results were stated in terms of the proportion between it and chromic acid, and the art followed him. Some of his proportions are well within Fink's, though more than Fink's optimum, but he was thinking in other terms. The step may seem easy, having gone so far; but the art did not find it so. Chromic sulphate contained the radical; the sulphuric acid in the chromic acid which combined to form chromic sulphate also contained it; but Sargent did not intimate that the radical alone need be controlled. He had not found the critical element, and his followers never could depend upon his process.
Schwartz worked with Fink upon the experiments which finally took form in the patent in suit. In 1923 Fink read a paper which Schwartz had written. It is important only as showing what he, intimately acquainted with Sargent's work as he was, and competent to push forward its implications, had in three years been able to do. The percentages again are well within Fink's at times about his optimum; and some of the results were good. But still there was no suggestion that it was not necessary to use chromic sulphate. Some of the discussion that followed Fink's reading of the paper throws light upon the uncertainties that still existed. For example Richardson, who had extensively experimented with chromium plating in the Westinghouse Lamp works, spoke of the erratic conduct of Sargent's baths. It was known, he said, that chromic sulphate was the provoking reagent, but at times, even with the provoking reagent, but at times, even with the purest materials, the bath would only start when dust settled into it; at times it was well to heat it; at times to cool it; sulphuric acid would not serve. This last statement and that about the purity of the materials are alone a conclusive demonstration of the failure up to that time to seize upon the essentials; for sulphuric acid is to-day a common form in which to add the catalyst; as an impurity in chromic acid it is a catalyst. No one understood this or could, until it was learned that it was the radical in any form which did the work. Two years later in 1925, with the work of sergeant before them, and their own extensive and at times very successful plating, the Bureau of Standards could say: "Although chromium can undoubtedly be deposited from a solution of this" (Sargent's) "composition, the metal efficiency and the character of the deposit tend to vary erratically. This behavior indicates that some variable has not been controlled." They were right. In the same year Proctor, at one time the president of the American Electro-Platers Society, spoke of chromium plating as the missing link of the industry, and not yet "a commercial electro-deposited metal."
Schwartz and Fink were skilled chemists. The two in conjunction did not find the invention at once. Fink claims to have reached it in the spring of 1924; we need not decide whether that is the right date or a year later, for nothing of consequence intervened, as we shall show. For the moment the important thing is that, being men well versed in the art, they had to grope their way. It was only after repeated experiments that it occurred to Fink that the radical alone was the catalyst, whether in chromic sulphate, in sulphuric acid, or not a sulphate radical at all. The optimum ratio then followed from trial and error. Unless, therefore, the art had already learned as much empirically, his idea, when verified, has every mark of an invention. The need had long existed; competent investigators had tried to fill it; they had hit the target but not the bull's eye; the art accepted and practiced the disclosure with success. In retrospect it now seems inevitable; perhaps it was in time. Chemists were probably bound in the end to learn how to electroplate chromium; it was another species of an art well known. But if this is to be the test, there will be few inventions, or none. The patent law need look only to the last step which overstrode what had so far balked advance. Those decisions which emphasize the implications of existing knowledge [Atlantic Works v. Brady, 107 U.S. 192, 199, 200, 2 S. Ct. 225, 27 L. Ed. 438; Thompson v. Boisselier, 114 U.S. 1, 18, 5 S. Ct. 1042, 29 L. Ed. 76; Western Elec. Co. v. Rochester Tel. Co., ...