Circuit Court of Appeals, Second Circuit.
Merrell E. Clark and Harold F. Watson, both of New York City, for appellant.
Watson, Bristol, Johnson & Leavenworth, of New York City (Lawrence Bristol and Charles P. Bauer, both of New York City, and Leon Robbin, of Washington, D. C., of counsel), for appellees.
Before MANTON, L. HAND, and CHASE, Circuit Judges.
L. HAND, Circuit Judge.
This appeal comes up from a decree for the plaintiffs in a suit in equity enjoining the infringement of claims 2, 5, 6, 7 and 8 of patent No. 1,710,073, issued to Samuel Ruben on April 23, 1929; and of claims 4, 5, 6 and 7 of patent No. 1,714,191, issued to the same person on May 21, 1929. The second of these was applied for first, but the disclosure of the other is more complete and we shall confine our attention to it. The invention is for an electric condenser, to be used in a self-starter on motor cars, in radio sets and the like; it is *538 of the dry electrolytic type, and is made as follows. The electrodes consist of thin aluminum plates of large area, wound upon each other into a cylinder with a "spacer" between them. Upon one of these, the anode, an oxide film of molecular thickness is formed which acts as the dielectric; the thinner the film, the more effective the condenser. The name "electrolytic condenser" has come by usage to mean this type. The film is subject to rupture with increases in voltage, and when broken, the condenser is inoperative until the break is restored; the electrolyte is interposed between the plates both to act as a conductor and to restore such brakes. The "spacer," a fabric whose fibre is coated with paraffin or the like to prevent its being burned, holds a paste in its mesh made of glycerin as a carrier and an electrolyte, which is described in the specifications in these words: "The particular film-forming electrolyte employed may be any one of a large number of well-known electrolytes but I have found a mixture of boric acid and sodium borate to be especially satisfactory." (Page 3 of the patent, lines 36-40). Again: "In forming the paste the electrolyte is first prepared preferably by dissolving the film-forming electrolyte materials in hot water and this solution is then thoroughly intermixed with glycerin to form a product of uniform composition." (Page 3, lines 59-64). The advantage of this mixture is that the glycerin being hygroscopic lengthens the life of the electrolyte which would otherwise evaporate, reduces its conductivity when it would otherwise ionize and tend to destroy the plates, and acts itself as "a dielectric constant." Claim two calls for spaced "film-forming" electrodes, and a "film-forming" electrolyte suspended in glycerin in such proportions as to make a paste. Claim five describes the paste as "a conductive material" made of a "supersaturated" electrolyte, suspended "in a medium" which contributes viscosity, hygroscopicity and "a relatively high dielectric constant." Claim six describes the glycerin and electrolyte in terms of function. Claim seven includes the "spacer" and specifies the elements of the electrolyte, boric acid and sodium borate. Claim eight is for the paste, the elements again being described by function. The upshot of all of these is a condenser of the electrolytic type, the conductive medium being composed of free glycerin and any well-known electrolyte mixed into a paste and smeared upon a "spacer." The defendant denies infringement for two reasons; first, it says that its conductive medium is not a paste; and second, that the electrolyte is not suspended in glycerin. As to the first, it is merely a question of degree, and as the case does not turn upon it, we need not discuss it. The word is not strictly one of art; the point where a substance gets viscous enough to become "paste" we shall not stop to determine. As to the second, the electrolyte of the defendant contains no free glycerin at all; it is made by heating about two parts of glycerin with one of ammonium borate, so that they reäct chemically and produce a solution of glyceryl boric acid and ammonium glyceryl borate. This is less pasty than Ruben's and is smeared upon a "spacer" of much finer mesh. How far it is an equivalent of his "paste" is the determining issue in the case.
The plaintiffs assert that Ruben was the first to make available a dry electrolytic condenser. The art had devised wet electrolytic condensers twenty years before; for example, Zimmerman, 1,074,231; and Peek, 1,008,860. In such wet condensers the electrodes had to be kept at some distance apart to avoid the ionization of the plates, and in both the disclosures just mentioned the electrolyte contained glycerin. Zimmerman plainly recommends its use; the plaintiffs' interpretation to the contrary we cannot accept. The first mention of glycerin is as follows, (page 1, lines 26-29): "The presence of glycerin has also been found advantageous in reducing corrosion of the electrode"; the second, (page 2, lines 18-21): "The presence of glycerin and sugars * * * has the effect of reducing corrosion at the electrodes and increasing the efficiency of the cell." Taken together there cannot be any doubt. What follows is not so plain, but it seems to mean that just as glycerin excites ortho-boric acid, so it serves to activate the acid radicals which reform the film when broken. The amount of glycerin to be added is not given, but its use and purpose are very plainly prescribed. Peek used ten per cent. of glycerin, in order to insure an acid reäction for the electrolyte; the purpose of this he does not state and perhaps the discovery was merely empirical; but none the less he prescribed it. Dry non-electrolytic condensers, which the art calls "electrostatic," appeared earlier. Wurts disclosed a cylindrical one in 1892, the dielectric being a fibrous "spacer" impregnated *539 with oil; Bradley disclosed another in 1897, in which the dielectric was stearate of lead. Russenberger in 1909 disclosed a dry electrolytic storage battery without any glycerin. A storage battery is apparently to be distinguished from a true condenser, though it may be used as a condenser under certain conditions, but we think it unnecessary to decide whether this patent disclosed a condenser or not. Indeed, the defendant's case is further advanced if it did not, because in that case there was no dry electrolytic condenser until 1925, when the art broke out in several, substantially at the same time. In September, 1925, Spelian filed an application for a storage battery, which consisted of a stack of iron plates separated by filter paper impregnated with sodium hydroxide to which thirty per cent. by volume of glycerin was added to reduce evaporation. This was not a condenser of the film-forming type and was in any case only intended for very low voltages; but the use of glycerin as a preservative of the conductive medium is significant in so closely related a device. That it is closely related is amply shown by the very patents in suit, which declare that a condenser with lead plates is one form of the invention, though useful only for low voltages. ("First" patent, pages 3, lines 65-82; "second" patent, page 1, lines 74-85). The plaintiffs attempt to explain these passages as intended merely to contrast the patents in suit with storage batteries; but it is clear that this was not their purpose but that Ruben regarded such cells as variants of his condenser. In May of 1926 Nodon, a French electrical engineer, read a paper before the French Academy of Sciences describing what he called a "colloidal condenser." This was made up of two sheets of aluminum, separated by cotton gauze, whose interstices were filled with a "thick paste of colloidal sesquioxide of iron and of glycerin." When placed in an alternating current, "this device fulfills the same functions of an electrostatic condenser of very great capacity." Whether the colloidal sesquioxide of iron, or glycerin, or both, were an electrolyte comparable with the electrolyte of Ruben are matters which we reserve for the moment. On October 1, 1926, Bush, a distinguished electrical engineer, filed an application for a dry electrolytic condenser in which the electrodes, which might be of "film-forming" aluminum, were separated by a "spacer" asbestos, blotting paper or the like to keep the electrodes apart and to absorb the film-forming liquid, preferably potassium acetate. Bush had no glycerin and his potassium acetate was a liquid; but the result was none the less a dry electrolytic condenser, though the low resistance of his electrolyte would probably decompose the plates rather soon. Ruben filed his application for the "second patent" in December, 1926.
It is not clear why the art should have become especially interested in dry electrolytic condensers about ten years ago; but it was about then that four such condensers appeared within two years; Nodon, Bush, Ruben and Edelman; and five, if we accept Slepian's storage battery. There was an active interest in such condensers at the time, and since all the elements which Ruben chose were to be found in the others, even though not combined as he combined them, the inference that his particular combination was inspired by unusual talent is weaker than if the efforts had been spread over a long period. The most important of the references is Nodon and on it the defendant particularly relies. There can be no dispute that the construction of it was like the defendant's condenser except for the supposed electrolyte, colloidal sesquioxide of iron. It had aluminum electrodes, a gauze "spacer," and glycerin apparently to suspend the electrolyte. The plaintiffs deny that the electrodes carried a film, but it seems to us perfectly apparent that they did. We cannot see any other possible meaning to the declaration that when the voltage gets too high "the dielectric is perforated and the discharge passes through the colloid." Nor can we otherwise understand the reference to "a dielectric whose thickness is 107 centimeter, that is to say, of molecular order." In a paper read at St. Louis twenty-two years before about electrolytic valves, Nodon used much of the same language as in this paper. He then spoke of such valves as "electrolytic condensers," and it was abundantly plain that the dielectric was a film upon the anode; indeed there was no other unless the glycerin was to act as such. That Nodon expected ruptures in this dielectric and provided for their repair does not indeed appear in the 1904 paper, but in 1926 after mentioning possible perforation of the dielectric, he went on to say that "the condenser recovers its original properties as soon as the rate of charge has again become normal"; and again that "the `break down' or internal discharge by no means puts it out of service." *540 Finally, one result of his condenser was "to obviate any danger of destruction of the apparatus by internal discharges."
The plaintiffs insist that he was not trying to make a condenser at all, though he repeatedly called it such, and spoke of its "very great capacity," as we have said. He contrasted it with an electrolytic valve though it had the "valve effect"; he knew the difference and he clearly supposed that this was a condenser. Was he right? The plaintiffs again say no, because with an alternating current no "leakage" was discovered. If that meant that on an alternating current no current passed, the criticism would be just, and a milliammeter, which Nodon suggested, apparently does not show "leakage" properly speaking. But the word, "leakage," would not have been used if Nodon had meant current, and it is perfectly clear that he did not mean it, because he speaks of the "perfect preservation" of the charging current. Next the plaintiffs say that Nodon's statement that on direct current there is an "appreciable leakage," shows that it was not a condenser. But there is some "leakage" through any condenser on direct current, as the plaintiffs' expert, Waterman, agreed, though it is "negligible"; and there is no reason to suppose that Nodon meant anything more. In further criticism Waterman believed that the disclosure could not be for a condenser because glycerin and sesquioxide of iron would not make a proper conductive medium or a good electrolyte. He thought that Nodon must have been deceived by impurities in the materials which he used. In support of these conclusions Barsky, a chemist, made tests of the conductivity of glycerin, sesquioxide of iron, and of both in a mixture. The sesquioxide helped the glycerin a little, but so little that the mixture would not have served for a conductive medium. Barsky did not say that he used the colloidal form; perhaps he did, but we cannot tell, nor whether colloids would have been more successful. He did say that sesquioxide of iron was not an electrolyte, but then he also said that glycerin was not, which is true only of the chemically pure substance, because commercial glycerin being mixed with water is a conductor and can be used as an electrolyte, as Ruben himself said in a later patent. (No. 1,920,151, page 1, line 72; page 2, line 54). Moreover, Grosvenor, for the defendant, was equivocal about sesquioxide of iron as an electrolyte, and though in still another of Ruben's patents (No. 1,912,223), he spoke of magnetic oxide of iron as an electrolyte, the proportions of iron and oxygen in that are different from those in sesquioxide, and it is unsafe to generalize in such matters. It is therefore open to doubt theoretically whether there was any electrolyte in Nodon except the glycerin.
Nevertheless Nodon appears to have been a scientist of standing, working as he did at the Sorbonne and the College of France and invited to address scientists in this country. It cannot well have been true that he was not competent to know whether he was getting the condenser which he certainly thought he was. It is a poor reply to his definite assertion to speculate about the properties of the substances he used; and while Barsky's tests were more persuasive than speculation, there was a far better way to prove that the supposed condenser was no condenser. That was by following out his disclosure and testing the result. That was exactly what the defendant did do; it made a condenser after Nodon's model, tested it and the plaintiffs either tested it, or made and tested a condenser of their own. That might have told us something, but the judge excluded the testimony because the defendant reserved it for surrebuttal. All we know is that after testing it Grosvenor remained of the opinion that Nodon had made a condenser. It does not seem to us, taking all the evidence together, that the plaintiffs' attack has been successful; we conclude that Nodon did make a dry "electrolytic condenser." We have already mentioned Bush, and within a few weeks after Ruben's date Edelman disclosed another dry "electrolytic condenser"; electrolytic plates, an absorbent holder for the electrolyte, an electrolyte in solid form, and although this, like Bush's, did not prescribe glycerin, Edelman used glycerin in practice without any cue from Ruben. Thus stood the art in 1927.
Ruben's disclosure appears upon this record to have been used only once; by the Grigsby-Grunow-Hunds Co. in the latter part of 1927. They made about seventy-five thousand condensers, though even these did not have an electrolyte made of sodium bicarbonate; but except for them there is no evidence of any use whatever of the invention, conceived in even its broadest terms; that is, as a "spacer" holding glycerin as a carrier for any "well-known electrolyte." The plaintiffs have *541 indeed received a large income from licenses, but that only means that the licensees have preferred to make their peace rather than fight, and little can be judged from it. Such acquiescence has often been taken as evidence of invention upon preliminary injunction, but the force of the inference is very variable. The reasoning is that if licensees consent to pay tribute, they must believe the disclosure to be patentable, and that this is evidence that it was beyond commonplace contrivance. But when a manufacturer takes a license, it is the resultant of two opposed motives, its cost and the dangers of infringement; and the magnitude of neither factor can be ascertained without knowing that of the other. If the patentee is wise he will make his terms low rather than invite a contest, and every reduction lessens the motive of the licensee to resist and the inference that he thinks successful resistance impossible. Walker on Patents, § 669. This is especially true when the license, as here, covers a number of patents, the threat from any one of which might be enough to induce payment. Upton v. Wayland (C. C.) 36 F. 691; Ertel Co. v. Stahl (C.C.A.) 65 F. 519; Felt & Tarrant Mfg. Co. v. Mechanical Accountant Co. (C.C.) 129 F. 386.
There being no evidence that Ruben really contributed to the advancement of the art, we see no reason to construe his patent broadly. Literally the defendant does not infringe even though the substance it uses is a "paste." It is certainly not an electrolyte suspended in free glycerin; and the glycerin in combination has not been proved to do all that the free glycerin does in Ruben's disclosure. Certainly it does not provide hygroscopicity, for the defendant's condenser is sealed. Whether it provides a "dielectric constant" is not clear; Waterman merely said that he "understood" so, but whether this was his independent judgment or from someone else, we do not know. We cannot find anything more in the record about the action of the glyceryl element in the defendant's electrolyte. Moreover, it might act like free glycerin, and still not answer as an infringement. Assuming that the patent is good at all, as to which we express no opinion, it should be confined to the claims as they read. There was nothing in the mere conception of a dry "electrolytic condenser" which required any invention. Given dry "electrostatic condensers" and wet "electrolytic," it was natural to wish to combine the compactness of the first with the capacities of the second. So much had certainly been anticipated. Ruben claimed nothing for his electrolyte; his invention rests altogether upon glycerin as a carrier. Zimmerman and Slepian had shown its use to limit conductivity, and Nodon had used it in a dry condenser, whatever his reasons. Inventors are charged with knowledge of what the art has done, and all the elements had been used before in this, or closely allied, fields. We do not of course forget that it is always the combination which counts, and that no patents, or almost none, are made from new elements. But all new combinations are not patentable combinations. Especially in chemical and electrical experiments happy solutions may be reached by testing out variants reached merely by permutations of old elements. Possibly the patent law should protect such industrial achievements, but it does not; the work of the well-equipped laboratory which by trial and error checks off a series of formulas, may be the proper path of industrial advance, as the Germans found it to be before the Great War; but it does not demand what we call "invention." Something more personal to the inventor, something which better measures his imaginative powers, is required. The setting here is a familiar one; at about the same time several inventors began to supply the same need; they naturally varied in their answers and one may be better than the others. But there is little antecedent reason for saying that any of such spontaneous outcroppings required unusual abilities, or that a successful device was not sure to be soon found. That is quite a different picture from that of a need long existing, with inconclusive answers spread throughout its duration, finally capped by success. The claims at bar may perhaps be valid, if confined to what they say, but they can go no further; they cannot cover the defendant's condenser.
Decree reversed; bill dismissed for noninfringement.
