497 F. Supp. 661 | D. Del. | 1980
HERCULES INCORPORATED, Plaintiff,
v.
EXXON CORPORATION, Defendant.
United States District Court, D. Delaware.
*662 *663 David F. Anderson of Potter, Anderson & Corroon, Wilmington, Del., for plaintiff; Caspar C. Schneider, Jr., Stanley L. Amberg, William J. Hone, Richard P. Ferrara, and Deborah A. Brandstater of Davis, Hoxie, Faithfull & Hapgood, New York City, of counsel.
Arthur G. Connolly, Jr. of Connolly, Bove & Lodge, Wilmington, Del., for defendant; Thomas F. Reddy, Jr., Berj A. Terzian, Jonathan A. Marshall, and Ronald A. Bleeker of Pennie & Edmonds, New York City, Lee Chasan of Exxon Chemical Co., of counsel.
OPINION
CALEB M. WRIGHT, Senior District Judge.
In this action, Hercules Incorporated ("Hercules") charges Exxon Corporation ("Exxon")[1] with infringing both literally and under the Doctrine of Equivalents, United States Patent 3,211,709 which covers a form of sulfur-vulcanizable ethylene propylene terpolymers ("EPT's") useful as synthetic rubbers.[2] Hercules's interest in this patent arises by virtue of an assignment from Great Britain's Dunlop Rubber Company, Ltd. ("Dunlop") whose interest in turn is derived from an assignment by its three employees, Drs. Stephen Adamek, Edward Allen Dudley and Raymond Woodhams ("Adamek et al.") whom the patent lists as inventors. Exxon defends by charging that some, if not all, of the claims of the Adamek et al. patent are invalid because (1) the patent involves an obvious development; (2) its disclosure is inadequate; and (3) it was procured by fraud. It is the opinion of this Court that some of Adamek et al.'s involved claims are invalid because Adamek et al.'s disclosures are inadequate. The remaining involved claims, although valid, are not infringed by Exxon's products.
THE CLAIMED PRODUCTS
Hercules charges Exxon with infringing Adamek et al. patent claims 1-2, 4-7, 10, 13-14 and 16-18[3] of which claims 1 and 17 are representative:
1. A sulfur-vulcanizable, elastomeric copolymer of at least two straight chain [alpha]-olefins of from 2 to 10 carbon atoms and an ethylenically unsaturated bridged-ring hydrocarbon containing at least two ethylenic double bonds, at least one of said double bonds being in one of the rings of the bridged ring present in said hydrocarbon, said hydrocarbon being present in the copolymer in an amount imparting sulfur-vulcanizability, said hydrocarbon having from 7 to 20 carbon atoms, the total straight chain [alpha]-olefin content of said copolymer being at least 50%.
. . . . .
17. A rubbery copolymer of ethylene consisting of ethylene, at least one alpha-olefin having the structure R-CH = CH2, in which alpha-olefin R is a C1-C8 alkyl *664 radical, and dicyclopentadiene, there being at least about 2.5 to 92.6% ethylene units by weight and at least about 2.5% to about 92.6% of said alpha-olefin units by weight, and, about 2.5 to 50% of dicyclopentadiene units by weight in said copolymer.[4]
Claim 1 (which typifies claims 1-2, 4-7, 10 and 16) and claim 17 (which typifies claims 13-14 and 17-18), thus cover terpolymers containing three hydrocarbons, at least two of which must be alpha-olefins,[5] including especially ethylene[6] and propylene.[7] Adamek et al.'s terpolymers must also contain a bridged-ring polyolefin.[8] The essential difference between claim 1 and claim 17 is that while claim 1 broadly specifies the use of a large group of bridged ring polyolefins, claim 17 narrowly specifies the use of only one dicyclopentadiene ("DCP").[9]
During polymerization, at least one double bond from each olefin becomes a single bond, releasing two electrons-one to each previously doubly bonded carbon atom. These electrons form the single bonds through which the two previously doubly bonded carbon atoms become incorporated within the main chain of the polymer. Ethylene, for instance, polymerizes in this fashion, known as "head-to-tail" or "1,2-" addition:[10]
*665 Bridged-ring polyolefins are added to the polymer in order to make it curable. Curing refers to a treatment by which rubber is stiffened from a putty-like compound into a harder more resilient and useful material:
[M]olecules . . . in a piece of .. unvulcanized, rubber . . . exist in a randomly coiled arrangement. If that article is stretched . . . the randomly coiled molecules extend in the direction of stress. . . . [and] the [main chain of the] molecules tend to align in the direction of the stress. . . . [I]f you immediately release the article, the molecules will then return to their randomly coiled shape in their original position and the molecule will resume its original shape. . . . If, however, you stress an uncured sample of rubber for a long enough time . . . the [main chain of the] molecules will flow past one another [This is known as creep][11] .. And now when you release the stress on that article, it will no longer retract to its original shape. It will have deformed into its new shape.
In contrast, if you can cross-link or cure the polymer, ... and .. stress the molecules by stretching the assembly, again the molecules will tend to stretch out in the direction of the stress. ... Now, however, the [main chain of the] molecules are tied together in a three-dimensional network by the cross-links. They can no longer flow past one another because they're tied together. ... The molecules remain in the stressed condition for as long as you hold it stressed. When you release it, they want to go back to their original randomly coiled arrangement, and that takes the assembly back to its original shape. ...[12]
Curing is successful when it significantly increases a polymer's stiffness or resistance to stretching. Stiffness is measured by hanging weights from a molded ring of polymeric material and measuring the resulting elongation.[13]
Curing is usually accomplished by incorporating cross-links of sulfur. Sulfur curing requires the presence of double bonds, which the curing process converts to single bonds. Hence, the need for using polyolefins which, by definition, contain at least two double bonds. The theory is that a polyolefin will sacrifice one of its double bonds for polymerization. The remaining double bonds will remain, providing sites for sulfur vulcanization.[14]
THE ACCUSED PATENT
During World War II, Japan cut off the Allies' traditional sources of natural rubber.[15] This caused a massive Allied research program to develop a synthetic substitute.[16] For reasons of security, this program was centered among the American rubber companies;[17] as a result, American rubber makers developed their technology far in advance of their foreign counterparts, including Dunlop.[18] In its subsequent efforts to catch up with the American technology, Dunlop established the North American Research Center ("N.A.R. C.") in Toronto, Canada about 1950.[19] About 1954, N.A.R.C. hired Drs. Raymond Woodhams, Stephen Adamek and Edward Allan Dudley,[20] who together set out to make the invention involved in this suit.
Adamek, Dudley and Woodhams began their work during the fall of 1956, when they homopolymerized several olefins. They also copolymerized ethylene and propylene,[21]*666 and soon they began searching for a suitable diene to polymerize with ethylene and propylene.[22] They initially investigated isoprene and butadiene, but when this research proved unfruitful, they considered other dienes including DCP.[23]
The idea of using DCP may have originated with Dudley, who remembered that in June, 1956, Enjay advertised "that [DCP] was available and [that Enjay] would be interested in people finding some use for it. . . . It struck [Dudley] as an interesting possibility as a diene ... [so he] wrote away and asked for . . . samples."[24]
The idea of using bridged ring structures such as DCP may also have originated with Woodhams who attended a September, 1956 meeting of the American Chemical Society in Atlantic City, New Jersey, where he listened to a lecture by three E. I. du Pont de Nemours & Company ("Du Pont") employees discussing the preparation of homopolymers of norbornene.[25] Woodhams thought that norbornene, which closely resembles DCP's bridged-ring,[26] "would be a useful monomer to incorporate in our screening program."[27]
On April 26, 1957, Adamek et al. began preparing and testing EPT's containing numerous third monomer candidates including two bridged-ring dienes-DCP and norbornadiene[28] and a variety of other compounds.[29] Only those experiments with the bridged-ring dienes proved successful.[30]
Adamek et al. then prepared a "rough draft for patent application"[31] which they sent on June 12, 1957 to Dunlop's Patent Department in Birmingham, England.[32] From this draft, Dunlop's Patent Department prepared a patent application and filed it in Great Britain on July 17, 1957.[33] On July 14, 1958, Adamek et al. filed an American application,[34] claiming the benefit of their British filing date under 35 U.S.C. § 119.
The American Patent Examiners initially rejected all of Adamek et al.'s claims.[35] Dunlop attempted to meet these objections,[36] but was unable to do so[37] prior to its November 30, 1961 assignment of the Adamek et al.'s American rights to Hercules.[38] Hercules's Edwin H. Dafter then assumed responsibility for Adamek et al.'s application. Dafter immediately cancelled all claims then on file and submitted a new set.[39] At about the same time, Dafter copied claims from and requested an interference[40] with U.S. Patent 3,000,866, which *667 had previously issued, on September 19, 1961, to Du Pont's Robert Edward Tarney.[41] The Patent Office granted this request[42] and Hercules's interference expert Clinton F. Miller[43] assumed responsibility for its prosecution.[44] On or about July 16, 1964, the interference terminated in favor of Adamek et al.[45] Adamek et al.'s patent eventually issued on October 12, 1965.[46]
THE LITIGATION
During Hercules's prosecution of Adamek et al.'s application, Exxon began making EPT's[47] containing methylene norbornene ("MNB")[48] and ethylidene norbornene ("ENB").[49] Both MNB and ENB contain two double bonds, one of which is located within a bridged ring. Since the broad language of Adamek et al.'s patent claims 1-2, 4-7, 10 and 16 arguably cover EPT's containing these structures, Exxon entered licensing negotiations with Hercules.[50] In the midst of these negotiations, Hercules initiated the present litigation on November 17, 1967.[51] The parties subsequently prepared for trial, generating 18,000 pages of deposition from thirty-nine witnesses.[52] A forty-six day bench trial followed, continuing between December 6, 1978 and July 17, 1979, during which time the parties generated a record of 6,900 pages and nearly 1,800 exhibits.
In deciding this case, the Court is particularly guided by the presumption of validity that normally attaches to issued patents under 35 U.S.C. § 282. This presumption places the burden of persuasion on the party attacking the validity of the patent.[53] If the Patent Office has previously considered the evidence that is cited to establish invalidity in a judicial proceeding, the party attacking the patent must assume a heavy burden and he must make a "clear and cogent" showing in order to prevail.[54] If, on the other hand, the Patent Office did not consider the cited evidence, the presumption is weakened in accordance with a balancing of the pertinence of the newly cited evidence against the pertinence of the evidence actually considered by the Patent Office.[55]
OBVIOUSNESS
The Law
Exxon's first challenge to the validity of the Adamek et al. patent arises out of 35 U.S.C. § 103, which, in relevant part, provides:
A patent may not be obtained . . . if the differences between the subject *668 matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains.
Graham v. John Deere Co.,[56] provides the following guidelines for testing a patent under Section 103:
[T]he scope and content of the prior art are to be determined; differences between the prior art and the claims at issue are to be ascertained; and the level of ordinary skill in the pertinent are resolved. Against this background, the obviousness or nonobviousness of the subject matter is determined. Such secondary considerations as commercial success, long felt but unsolved needs, failure of others, etc., might be utilized to give light to the circumstances surrounding the origin of the subject matter sought to be patented.[57]
Since Section 103 does not define the term "prior art", courts generally look to 35 U.S.C. § 102,[58] which defines prior art for purposes of judging novelty, as opposed to obviousness. In relevant part, Section 102 provides:
A person shall be entitled to a patent unless
(a) the invention was known or used by others in this country, or patented or described in a printed publication in this or a foreign country, before the invention thereof by the applicant for patent, or
(b) the invention was patented or described in a printed publication in this or a foreign country or in public use or on sale in this country, more than one year prior to the date of the application for patent in the United States, or
. . . . .
(e) the invention was described in a patent granted on an application for patent by another filed in the United States before the invention thereof by the applicant for patent, or
. . . . .
(g) before the applicants invention thereof the invention was made in this country by another who had not abandoned, suppressed, or concealed it.
Hercules insists that Section 102(g) is applicable to interferences only, and that in an infringement action, a court cannot consider unpatented inventions.[59] Hercules's contention, however, contradicts recent case law development recognizing that Section 102(g) is not so limited.[60] Although courts in this jurisdiction have not previously decided the issue, Hercules fails to convince this Court that the settled authority from other jurisdictions should not be followed.
Section 102(g) clearly covers actual reductions to practice,[61] which consist of three elements: (1) production of the claimed composition; (2) recognition of its properties; and (3) identification of its specific practical utility.[62] Each of these elements must be established using corroborated proof.[63]
*669 Exxon argues that Section 102(g) also covers American patent applications, which, although abandoned here, issue in a foreign country.[64] The basis for Exxon's argument is unclear. "Invention" might conceivably be equated to a constructive reduction to practice,[65] which is established by filing a patent application.[66] This application, however, must meet the standards of American patent law. Whether or not a foreign application ultimately issues is irrelevant for determining the sufficiency of its American counterpart.[67]
The Facts[68]
Natural rubber consists of long chains of isoprene units[69] which, because they contain double bonds, may be vulcanized with sulfur.[70] During World War II, the Allies polymerized styrene and butadiene,[71] producing a suitable natural rubber replacement,[72] which because it contains double bonds, is also sulfur vulcanizable.[73] Polymer chemists, most of whom held Ph.D.'s in chemistry or chemical engineering,[74] subsequently developed other sulfur vulcanizable rubbers including butyl rubber,[75] which contains isobutylene and isoprene.[76]
The first group of patents that Exxon specifically cites as prior art were issued to Brown,[77] Frederick,[78] and Parrish,[79] and relate to artificial rubbers, all of which were prepared using Friedel-Crafts type catalysts under restrictive conditions.[80] Brown's patent seeks to solve the following problems:
The introduction of unsaturation [i. e. a double bond] even a small amount makes the isobutylene isoprene copolymers and their sulfur vulcanizates [i. e. butyl rubbers] susceptible to oxidation and makes them less flexible at low temperatures. *670 Moreover, it is not possible, ordinarily, to produce sulfur vulcanizates from known isobutylene isoprene copolymers having desirably high tensile strength and modulus, especially in the unreinforced condition.[81]
Brown suggests the copolymerization of 50% to 98% (weight) of an aliphatic isoolefin, e. g., isobutylene, and 1% to 30% (weight) of an acid chloride.[82] Brown's polymer does not contain any carbon-to-carbon double bonds and so it apparently is not sulfur vulcanizable.[83] Brown's patent further specifies that one or more other components may be polymerized with isobutylene and acid chloride in order to vary "the properties of the interpolymers of this invention."[84] The patent then lists more than a dozen optional components including DCP and cyclopentadiene.[85]
Frederick's patent addresses the following problems:
Butyl rubber cures more slowly with sulfur than other useful sulfur vulcanizable rubbers; it does not adhere well to other materials such as natural rubber; when it is mixed with natural rubber and the mixture is vulcanized the resulting vulcanizates are generally less useful and valuable than vulcanizates from either of these rubbery materials alone.[86]
Frederick's patent claims that these problems may be solved by exposing butyl rubber type polymers to a hydrogen halide[87] as an intermediate step between their formation and sulfur vulcanization.[88] The patent thus specifies the copolymerization of between 70% and 99% (weight) of an isoolefin, e. g. isobutylene, and 1% to 30% (weight) of one or possibly two or three polyolefins,[89] each containing between four and eighteen carbon atoms. Possible polyolefins include a very wide variety of hydrocarbons, including cyclopentadiene and DCP which Frederick specifically lists.[90] Formation of the polymer consumes one double bond from each polyolefin, leaving at least one other double bond available as the site for attaching a hydrogen halide.[91]
Parrish's patent addresses the following problems:
Although . . . isoolefin diolefin copolymers [butyl type rubbers] are quite useful in the final vulcanized condition, they possess in both the vulcanized and unvulcanized condition several distinct and serious disadvantages. The modulus of elasticity of their vulcanizates is often undesirably low. In addition, they are soft and weak in the unvulcanized raw state such that "cold flow" occurs making it difficult to handle them as in storage and shipment, yet, when they are processed and compounded in the rubber factory, they are tough, "nervy", difficult to form into a smooth sheet on the mill and not readily admixed with compounding ingredients.[92]
In order to overcome these problems, Parrish copolymerized 85% to 99.5% (weight) of an olefin, preferably isobutylene, 1% to 14.5% (weight) of a diolefin, preferably isoprene, and 0.5% to 5% (weight) of a substituted *671 fulvene.[93] The patent claims that the resulting copolymer was sulfur vulcanizable. The patent also specifies that "minor proportions, preferably" 1% to 30% (weight) of a fourth component might be included in his polymers although it does not indicate the purpose for doing so. Parrish's patent then lists a wide variety of such possible ingredients including cyclopentadiene and DCP.
The Brown, Frederick and Parrish patents disclose inventions that clearly are different from that of Adamek et al. Where Adamek et al. use Ziegler catalysts,[94] Brown, Frederick and Parrish use Friedel-Crafts type catalysts. Exxon makes no showing that the art relating to Friedel-Crafts catalysts renders obvious the use of Ziegler catalysts. Where Adamek et al. require the polymerization of at least three components, Brown, Frederick and Parrish require only two. Moreover, the various components are substantially different. Adamek et al. require that their polymers contain two alpha-olefins and a bridged-ring polyolefin. Contrastingly, Brown requires one isoolefin and one acid chloride, Frederick requires an isoolefin and a diolefin and Parrish requires an isoolefin and a substituted fulvene. The products of Adamek et al. and those of at least Brown and Frederick,[95] also differ in their vulcanization properties. Adamek et al.'s products may be directly vulcanized with sulfur using pendant group vulcanizations sites.[96] In contrast, Brown's products are not sulfur vulcanizable. Frederick's products, though sulfur vulcanizable, contain their vulcanization sites in the polymer's main chain,[97] and they also require intermediate reaction with a hydrogen halide prior to sulfur vulcanization.
In order to prove that Adamek et al.'s product was anticipated by prior art, Exxon first relies heavily upon the fact that Brown and Parrish specify that their polymers might possibly contain cyclopentadiene.[98] The Court concludes, however, that since cyclopentadiene is not a bridged-ring compound, its polymerization is not sufficiently relevant to the subject matter of this litigation to constitute prior art. Although the fact that Woodhams believed that DCP and cyclopentadiene were equivalent dienes might indicate a relevant connection,[99] subsequent research proved this belief wrong.[100] Another possible connection with this litigation might be based upon cyclopentadiene's "spontaneous" formation of DCP at room temperature.[101] While Brown and Parrish report the possible use of cyclopentadiene, then, a polymer chemist might read their specifications and realize that cyclopentadiene dimerizes so quickly that, in effect, they are using DCP. When used in chemical parlance, however, the word "spontaneous", does not suggest the notion of immediacy normally associated with the word; rather, it usually means merely that a reaction will proceed without external interference, such as that provided by a catalyst.[102] In fact, the reaction by *672 which DCP is formed from cyclopentadiene is so slow that its impact is unimportant. Exxon's Dr. Maurice Morton testified that the dimerization reaction occurs only "over a long period of time".[103]
Exxon also relies upon the fact that Brown and Parrish specify the possible incorporation of DCP in their polymers.[104] The Court concludes, however, that this fact has little to do with this litigation. Brown and Parrish only mention DCP as an optional ingredient without specifying any particular purpose for its inclusion. A polymer chemist reading their specifications would thus not be significantly influenced to incorporate DCP as a polymeric constituent necessary for introducing sulfur vulcanization sites to an EPT. In sum, the inventions of Adamek et al. and those of Brown, Frederick and Parrish are very different, and the Brown, Frederick and Parrish patents do not render obvious Adamek et al.'s invention.
The second group of inventions claimed by Exxon as prior art relate to artificial rubbers prepared with Ziegler type catalysts[105] under flexible conditions.[106] This group includes prior art patents to Goodrich-Gulf Chemicals, Inc.'s S. E. Horne, Jr., et al.,[107] Montecatini, S.p.A.'s Guilio Natta et al.[108] and Du Pont's William F. Gresham, et al.[109] The initial pages of the Horne et al. patent specify homopolymers of conjugated diolefins, including isoprene and cyclopentadiene. The Horne et al. patent later discloses copolymers, trimers, etc. prepared from mixtures of "a large proportion" of "one, two, [or] three or more" conjugated polyolefins and one or more olefins.[110] This language arguably covers polymers of cyclopentadiene, ethylene and propylene. The patent does not, however, mention or exemplify this specific combination, although it does exemplify copolymers of isoprene and cyclopentadiene.[111]
The Horne et al. patent does not render that of Adamek et al. obvious. Although Exxon emphasizes Horne et al.'s polymerization of cyclopentadiene,[112] this fact is not sufficiently relevant.[113] Further, Horne et *673 al. use a conjugated structure for locating sulfur vulcanization sites within their polymer, while Adamek et al. use bridged ring structures which may, but need not, be conjugated. These developments are very different. Conjugated dienes permit 1,4-addition rather than 1,2-addition.[114] In 1,4-addition, all four carbon atoms comprising the conjugated structure[115]
1 2 3 4 CH = CH CH = CH enter the main chain of the polymer and a double bond forms between carbon atoms two and three.[116] If sulfur vulcanization does not consume this unsaturation, it remains available for reaction with other chemicals, especially ozone, which, in reacting with doubly bonded carbons, may split the previously doubly bonded carbon atoms apart from each other. This causes the polymer's main chain to break into many smaller pieces and thus destroy the integrity of the vulcanized rubber network, with the result that the rubber becomes useless.[117] In contrast, Adamek et al. use a bridged-ring structure with the result that the residual unsaturation becomes more favorably located pendant to the polymer's main chain.[118] Ozone attack does not occur in the main chain of the polymer and the integrity of the structural network is preserved. The result is a more durable rubber.[119] Horne et al.'s patent does not disclose the use of any bridged-ring structure and Exxon fails to adduce reliable evidence that its use was rendered obvious by Horne et al.'s disclosures.
Natta et al.'s patent discloses the polymerization of one or more monoolefins with between 7% and 88% (weight) of a diolefin that yields an elastomeric and sulfur vulcanizable product.[120] Possible monoolefins include compounds of the "general formula CH2=CHR, in which R" represents a wide variety of straight chain hydrocarbon radicals.[121] Although this language arguably is broad enough to cover ethylene or propylene polymers, Natta et al.'s patent does not mention or exemplify their use. Possible diolefins, including butadiene, 1,3-pentadiene and 1,5-hexadiene must contain a terminal double bond.[122] This requirement excludes the use of many bridged ring dienes, such as DCP which do not contain any terminal double bonds. Indeed, Natta et al. fail to mention or exemplify the use of any bridge ring dienes.
Gresham et al.'s patent describes the polymerization of at least one alpha olefin, including specifically ethylene or propylene,[123] with a straight chain diolefin, including 1,4-hexadiene "in which the double bonds are . . . [non-conjugated], and in which at least one of the double bonds is terminally located,"[124] and the other is preferably internal. Gresham et al. exemplify polymers containing equal molar *674 amounts of ethylene, propylene,[125] and a diene, obtaining "particularly useful .. products with rubber-like properties of the kind usually desired."[126] The patent does not, however, disclose the use of DCP, which is not a straight chain diolefin.
The patents of Adamek et al., Natta et al. and Gresham et al. cover different inventions. The three research teams all attempted to find a suitable polyolefin having at least two double bonds of differing reactivities. One double bond had to react under the relatively mild conditions used in Ziegler catalysis, while the other had to resist reacting until exposed to the much more severe conditions used in sulfur vulcanization.[127] Where Adamek et al. utilize a bridged-ring structure for this purpose, Natta et al. and Gresham et al. employ straight chain dienes having terminal unsaturation.
Each of these structures works for an entirely different reason. The structures of Natta et al. and Gresham et al. work because a terminal double bond is more reactive than an internal one.[128] Bridged-ring hydrocarbons, on the other hand, work because they are strained due to the drawing together of many carbon atoms in unnaturally close proximity. In order to relieve some of this strain, a bridged-ring diolefin will, even under the relatively mild conditions of Ziegler catalysis, change one bridged-ring double bond into a single bond. If the polymerization conditions are properly adjusted, the second double bond will not react until it is exposed to the more rigorous conditions of sulfur vulcanization.[129]
Since Exxon fails to adduce reliable evidence that the use of straight chain dienes would necessarily imply the use of bridged-ring dienes, the Court concludes that the use of each was a significant, though perhaps narrow, development in the apparently well developed rubber technology field. This conclusion seems especially justified since the Adamek et al. patent issued despite the fact that Hercules cited the Patent Examiners[130] to one of the articles[131] in which Natta et al. describe their discovery.
The third group of patents Exxon cites as prior art includes three new polymers made with Ziegler catalysts. Du Pont's Arthur W. Anderson et al. patented two of these polymers, the first being a homopolymer of *675 norbornene,[132] and the other a copolymer of ethylene and preferably less than fifty mole percent of norbornene.[133] Du Pont's Gelu Steoff Stamatoff patented the third cited polymer, a copolymer of ethylene and a bridged-ring monoolefin,[134] including, among others, dihydrodo dicyclopentadiene.[135] All of these polymers are saturated, and therefore are not sulfur vulcanizable.[136] This third group of patents also includes a patent to Stamatoff covering a modified Ziegler catalyst suitable for polymerizing ethylene, "over a wide range of conditions".[137] Example III of the patent describes the copolymerization of ethylene and 27% (weight) of DCP. Stamatoff's specification does not indicate whether this product contains residual unsaturation or whether it is sulfur vulcanizable.[138]
Exxon's third group of patents involves inventions that are very different from Adamek et al.'s invention and do not render it obvious. Where Adamek et al. use at least two alpha olefins, Anderson et al. and Stamatoff use only one. Where Adamek et al. produce materials containing three or more components, the products of Anderson et al. and Stamatoff are all homopolymers or copolymers. Where Adamek et al.'s products are sulfur vulcanizable, those of Anderson et al. and Stamatoff are not, or at least as in the case of the product of Example III of Stamatoff's patent, they are not revealed to be.
Exxon claims that the Anderson et al. and Stamatoff patents demonstrate that bridged-ring dienes might be polymerized using Ziegler catalysts.[139] Exxon fails to show, however, that this information implies the necessary fact that the second double bond in a bridged-ring hydrocarbon would not affect polymerization, and would remain available as a sulfur vulcanization site. Du Pont's Dr. Edward K. Gladding, thus testified that he would be "really speculating" to say that knowledge that norbornene could be polymerized would indicate that bridged-ring hydrocarbons might be used to prepare sulfur vulcanizable EPT's. "[T]he influence of some other structure on that norbornene double bond, to my knowledge was not known. . . . I wouldn't be able to predict that it would or would not influence the polymerization reaction or deactivate the catalyst or what have you."[140] Indeed, Exxon's own witness, Dr. Maurice Morton, confirmed the difficulty of making such a priori conclusions in the synthetic rubber field. "There is still very little understood, and there was even less understood then. . . . It became a highly empirical science. I should say highly empirical art. It's more of an art than a science."[141]
The fourth group of allegedly prior art patents upon which Exxon relies consists of a series of foreign patents issuing to Du Pont's Stamatoff[142] and carrying a November *676 5, 1956 priority date, based upon an American application. These foreign patents disclose the copolymerization of ethylene and 0.1% to 50% (mole) of a bridged-ring diene,[143] including DCP or MNB, both of which are exemplified by the patents. The resulting polymers may be sulfur vulcanized.[144]
Although Stamatoff's patents may involve an invention that is closely akin to that of Adamek et al., Stamatoff's patents do not qualify as prior art. Sections 102(a) and 102(b), covering inventions that are patented abroad, specify that such inventions may not be considered as prior art until the date on which they are described in a printed publication. Neither party contends that Stamatoff's invention was so described prior to the time that Adamek et al. completed their invention and filed their British application on July 17, 1957. Section 102(e), which makes American patents prior art, is also inapplicable because Stamatoff abandoned his American patent application before it could issue in the United States.
Exxon finally relies upon Section 102(g), which defines completed inventions as prior art. Exxon's reliance is misplaced, however, because of a lack of proof that Stamatoff completed his discovery prior to Adamek et al.'s invention. In attempting to show this fact, Exxon alludes to Stamatoff's abandoned American application. Since Exxon fails to produce a copy of this application, however, the Court is without knowledge regarding the information it might contain. Further, abandoned patent applications are insufficient to prove prior invention.[145] Indeed, the fact that the applicants abandoned their application may evidence a belief in, if not an admission of, the inadequacy of their claim.
Exxon next produces what the Court understands to be undated and unwitnessed entries from some unspecified laboratory notebook, possibly belonging to Stamatoff.[146] Exxon also cites a July 13, 1956 Memorandum, from Stamatoff to Gresham, stating:
Dicyclopentadiene can be copolymerized with ethylene with [Ziegler] catalysts in a variety of manners and a number of interesting copolymers have been obtained with stiffness varying from slightly stiffer than polyethylene to as stiff or stiffer than polystyrene. This work, of course, is far from complete but, at this time, we are bringing to your attention some pertinent observations which might be of interest *677 to those concerned with the ethylene copolymers.
. . . . .
5. Physical Properties
We have just started this phase of the problem and have only a few physical measurements which we present in Table II. From these data we concluded that these copolymers are attractive and deserve consideration.
. . . . .
Batch . . . . Film Characteristics
NB. p.
5234 - 158 tough & stiff
159 clear, touch & stiff
160 stiff, brittle & clear
161 stiff & brittle
162 tough, clear & flexible
163 brittle & very stiff.
. . . .
Physical Tests of Copolymers of Ethylene/dicyclopentadiene
. . . Tensile Impact Flex Modulus × 103
Batch Mol. % DCP 25°C. 25°C. 50°C.
5234 - 157 5.05 179 236 116
158 5.09 113 262 160
159 5.15 28.6 293 171
160 9.98 25.7 343 263
162 5.05 70 376 148[147]
163 9.98 16 396 330
Since Exxon fails to produce any witness to substantiate the nature and authenticity of these documents, the Court believes that it should not receive and consider them. Assuming arguendo that the Court consider the documents, they do not show that Stamatoff reduced his claimed copolymers to practice. In particular, the submitted notebook pages fail to show that Stamatoff ever identified a specific purpose for which his polymers might be used. Because the July 13, 1956 memorandum includes complex scientific data, and because the Court does not have the benefit of expert testimony regarding this memo's meaning, the Court cannot conclude that the memo provides enough specific information to demonstrate that Stamatoff's polymer was useful. Indeed, the opposite conclusion seems justified in light of Stamatoff's statement that he had "just started" the physical testing phase of the development program and the fact that nearly four months passed before he filed his American patent application which was eventually abandoned.
Finally, Exxon seeks to rely upon testimony apparently adduced as an after-thought during the cross-examination of Hercules's rebuttal witness, C. Marshall Dann. Prior to becoming the Patent Commissioner, Dann served as a patent attorney for Du Pont from 1945 to 1974, and became Chief Patent Counsel in 1969.[148] Dann testified that insofar as he was aware, Du Pont's "common practice" was to avoid including in "any patent application any example which had not been actually done in the laboratory." Dann further testified that "[T]he way [Stamatoff's foreign patent examples] are written, they certainly sound *678 to me like actual working examples."[149] Although Dann's testimony might prove helpful to corroborate the testimony of witnesses familiar with the specifics of Stamatoff's work, his work at Du Pont was so remotely related to Stamatoff's patent application that his testimony, considered by itself, is insufficient to supply the proof that is otherwise missing from Exxon's case.
The determination that the discovery of EPT's containing bridged-ring hydrocarbons was unobvious, is supported by what Graham v. John Deere Co., supra, characterized as secondary considerations, including the commercial success enjoyed by EPT's containing bridged-ring dienes. Beginning in 1966, when American rubber makers produced about 12,500 metric tons of EPT's containing a bridged-ring diene, they steadily increased production, manufacturing 75,000 metric tons in 1977. Experts expect 1982 production to total more than 100,000 metric tons.[150] That Adamek et al. were the first scientists to conceive inventions to capture this large demand, indicates the unobviousness of their development.
Another secondary consideration, the failure of other parties in the field, including Du Pont, Exxon and Hercules, to discover many other competitive EPT's, further supports the conclusion that Adamek et al.'s discovery was unobvious. Du Pont centered its efforts in a team of research chemists[151] who were given first class laboratory and library facilities.[152] Beginning as early as 1955,[153] they investigated the utility of alternative cross-linking methods not requiring residual unsaturation,[154] but this research was so unsuccessful that it served only to confirm the value of the long used sulfur vulcanization technique.[155] Du Pont's researchers next searched for dienes suitable for incorporating sulfur vulcanization sites within ethylene-propylene polymers.[156] They soon discovered the workability of straight or branched chain diolefins, especially 1,4-hexadiene, and in January, 1956, Gresham applied for his patent.[157] Gresham's diolefins, however, could then be made only by using Grignard-style synthesis, which, though suitable for the laboratory work, is generally uneconomical for large scale commercial operations.[158] Moreover, "Addition of even small amounts of unconjugated diolefins of the type represented by 1,4-hexadiene . . . almost completely inhibit[ed] ethylene, propylene copolymerization at atmospheric pressure."[159]
Du Pont researchers next discovered, during 1959, a process to synthesize 1,4-hexadiene *679 cheaply,[160] and they also discovered that 1,4-hexadiene would not deactivate the catalyst, as long as a low reaction temperature (between 0 to -10°C.) was maintained.[161] Use of 1,4-hexadiene remained problematical,[162] however: the lowered temperature reduced conversions,[163] necessitating costly recovery and recycling of unused monomers.[164]
During the autumn of 1959, four years after Du Pont began its intense research program, Tarney finally conceived using DCP.[165] Initial testing proved so "highly successful"[166] that Du Pont viewed EPT's containing DCP as its leading commercial candidate.[167] Shortly thereafter, Du Pont drafted and executed a patent application which it rushed by company car from Wilmington, Delaware to the Patent Office.[168] Only when Du Pont subsequently learned of Adamek et al.'s outstanding application[169] did it abandon the use of DCP in favor of 1,4-hexadiene.
Hercules began its research as early as April 30, 1956,[170] assigning three or four predominantly Ph.D. level polymer chemists to find a suitable diene.[171] Within the first five months, Hercules tried a dozen dienes, but, as one researcher reported, "No outstanding polymer has been found."[172] Research continued without success.[173] Indeed, by 1960, Hercules's scientists performed as many as 140 experiments.[174] These experiments revealed only the possible utility of straight or branched chain dienes.[175] Hercules did not conceive trying any bridged-ring termonomers like DCP until after 1961 when it learned of Adamek et al.'s initial success. Thereafter, it immediately abandoned its other work and concentrated upon the use of bridged-ring termonomers.[176]
Exxon, beginning as early as September, 1956,[177] also worked to identify suitable dienes. Exxon's Dr. Henry Markowski and his associates screened a number of possible dienes, including cyclopentadiene[178] which he and his colleagues prepared by splitting DCP.[179] Unlike Adamek et al., however, Exxon's workers did not conceive using unsplit DCP. Consequently, they never experienced the success enjoyed by Adamek et al.[180] Indeed, Exxon scientists almost abandoned their project for making sulfur vulcanizable *680 EPT's.[181] After learning of Du Pont's success with 1,4-hexadiene[182] and Dunlop's success with DCP[183] however, Exxon's scientists renewed their interest in sulfur vulcanizable EPT's.[184] In September, 1961, after five years of research Exxon finally prepared a sulfur vulcanizable EPT containing DCP, its "first successful diolefin".[185]
In addition to Dunlop, at least three other major chemical companies, Du Pont, Hercules and Exxon, sought suitable dienes for terpolymerizing with ethylene and propylene. Their efforts were intense, yet, Du Pont's researchers worked for more than four years before realizing the suitability of bridged-ring hydrocarbons, and researchers at Hercules and Exxon learned of this fact only through Dunlop's disclosures. That so many companies conducted parallel research without making the same discovery as Adamek et al., further demonstrates the unobviousness of their invention.
Another factor to consider is what Hercules calls the industry tribute to the patent in suit. Indeed, a number of companies have taken licenses, paying Hercules substantial sums.[186] Exxon counters that the licensing payments are not tribute to the patent but that, instead, they are an inexpensive means for avoiding costly litigation.[187] Exxon points out that Hercules's original licensing agreements offered 50% to 70% royalty rebates should Adamek et al.'s patent be invalidated.[188] As this trial neared, Hercules renegotiated these license agreements, granting paid up licenses in return for a release from its possible duty to rebate previously collected royalties.[189]
The Court concludes that the evidence regarding licensing is not helpful because the parties did not present any evidence regarding normal licensing arrangements. Consequently, the Court is without sufficient knowledge to judge whether the licensing agreements and their subsequent amendment favor Hercules or its licensees.
DESCRIPTION
The Law
Exxon's next set of challenges are based on 35 U.S.C. § 112. This section, which traces its history to the first patent statute,[190] now provides in relevant part:
The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor of carrying out his invention.
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
O'Reilly v. Morse[191] is a fountainhead case in which the disclosure requirement *681 was applied. Samuel F. B. Morse claimed infringement of his telegraph patent which included at least one very broad claim:
Eighth. I do not propose to limit myself to the specific machinery, or parts of machinery, described in the foregoing specifications and claims; the essence of my invention being the use of the motive power of the electric or galvanic current, which I call electro-magnetism, however developed, for making or printing intelligible characters, letters, or signs, at any distances, being a new application of that power, of which I claim to be the first inventor or discoverer.[192]
O'Reilly challenged this claim as covering a "principle or effect, and not a machine, manufacturer, or composition of matter."[193] The Supreme Court agreed, declaring Morse's eighth claim illegal and void:
Whoever discovers that a certain useful result will be produced, in any art, machine, manufacture, or composition of matter, by the use of certain means, is entitled to a patent for it; provided he specifies the means he uses in a manner so full and exact, that any one skilled in the science to which it appertains, can, by using the means he specified, without any addition to, or subtraction from them, produce precisely the result he describes. . . .
Indeed, if the eighth claim of the patentee can be maintained, there was no necessity for any specification, further than to say that he had discovered that, by using the motive power of electro-magnetism, he could print intelligible characters at any distance. We presume it will be admitted on all hands, that no patent could have issued on such a specification.[194]
Since Morse, courts have evolved a number of specific description rules that Exxon charges were violated by Adamek et al.'s patent specification. The first such rule requires an applicant to draft claims that closely correspond to the language of his patent specification. In In re Ruschig,[195] an applicant sought to amend his original application by adding a new claim 13, specifically naming the compound for which he sought patent coverage. The applicant claimed that this amendment corresponded to the language of his original specification, and that it was therefore permitted under 35 U.S.C. § 132.[196] The original specification, however, did not include any language remotely resembling that of claim 13. Rather, the original specification described the compounds using a formula that included several chemical constituent lists, from each of which one constituent was to be selected. This formula covered "something like half a million possible compounds,"[197] including the compound of claim 13 which could be made only by selecting the correct constituent from each list. The Examiner rejected the applicant's claim that his specification and his later-filed claim 13 were equivalent[198] and the Court of Customs and Patent Appeals ("C.C.P.A.") upheld this rejection:
It is an old custom in the woods to mark trails by making blaze marks on the trees. It is no help in finding a trail or in finding one's way through the woods where the trails have disappeared-or have not yet been made, which is more like the case here-to be confronted simply by a large number of unmarked trees. Appellants are pointing to trees. We are looking for blaze marks which single out particular trees. We see none.
. . . . .
Working backward from a knowledge of [the resulting compound], that is, by hindsight, it is all very clear what route one would travel through the forest of the *682 specification to arrive at it. But looking at the problem, as we must, from the standpoint of one with no foreknowledge of the specific compound, it is our considered opinion that the board was correct in saying:
Not having been specifically named or mentioned in any manner, one is left to selection from the myriads of possibilities encompassed by the broad disclosure, with no guide indicating or directing that this particular selection should be made rather than any of the many others which could also be made.[199]
The second description rule requires that a specification provide enough information to justify the utility of the claimed compounds. In In re Cavallito,[200] the applicants sought to patent "several hundred thousand possible compounds, of which only thirty [were] specifically identified in appellant's application."[201] The C.C. P.A. noted:
[W]here the applicant seeks to obtain a monopoly in exchange for his disclosure of a group of compounds there should be a disclosure which gives reasonable assurance that all, or substantially all of them are useful. . . . An applicant is not entitled to a claim for a large group of compounds merely on the basis of a showing that a selected few are useful and a general suggestion of a similar utility in the others.[202]
In setting forth this rule, the C.C.P.A. apparently recognized one exception: only a very few compounds need be discussed when it is shown that they and the large group of compounds they represent share a key structural feature from which a common utility derives. Arguments for applying this exception are most persuasive when the application specifies the basic formula necessary to import useful properties or when it is shown that all of the specific compounds have equal potency indicating that their utility derives solely from the basic structural formula common to all of them.[203]
The third description rule prevents an applicant's claim from covering inoperative as well as operative subject matter. In Graver Tank & Mfg. Co. v. Linde Air Products Co.,[204] the patentees sought to enforce a part claiming a new welding composition, comprising either "metallic silicate and calcium fluoride" (Claim 24) or chiefly "silicates" (Claim 26). The trial court refused to enforce these claims because "The evidence is clear and convincing that many silicates, even many metallic silicates, are inoperative as major constituents in a welding composition having for its objectives those stated in the patent."[205] The Supreme Court upheld the trial court's decision:
[Claims] must be sufficient to "particularly point out and distinctly claim" an identifiable invention or discovery. . . . While the cases more often have dealt with efforts to resort to specifications to expand claims, it is clear that the latter fail equally to perform their function as a measure of the grant when they overclaim the invention. When they do so to the point of invalidity and are free from ambiguity which might justify resort to the specifications, we agree with the District Court that they are not to be saved because the latter are less inclusive.[206]
The Facts
Exxon clearly and convincingly establishes that the rule of In re Ruschig is an obstacle to a finding that Adamek et al.'s *683 specification supports claims 1-2, 4-7, 10 and 16.[207] The parallel between the facts of this instant case and those of In re Ruschig is striking. Like Ruschig's specification, that of Adamek et al. broadly specifies a wide variety of possible chemical constituents:
The preferred [termonomer] compounds are the endocyclic[208] hydrocarbons containing seven to ten carbon atoms and two double bonds, especially those containing a bridge of one or two methylene groups. Substituted endocyclic hydrocarbons such as the alkyl, aryl and cycloalkyl derivatives can also be used.
Examples of suitable bridged-ring hydrocarbons are:
(a) Unsaturated derivatives of bicyclo(2,2,1) heptane, including
bicyclo(2,2,1) hepta-2-ene (norbornene);
ticyclo(2,2,1) hepta-2.5-diene; bicyclo pentadiene [DCP], tricyclopentadiene and tetracyclopentadiene.
(b) Unsaturated derivatives of bicyclo(2,2,2) octane, including
bicyclo(2,2,2) octa-2-ene and bicyclo(2,2,2)
octa-2.5-diene.
(c) Unsaturated derivatives of bicyclo(3,2,1) octane.
(d) Unsaturated derivatives of bicyclo(3,3,1) nonane.
(e) Unsaturated derivatives of bicyclo(3,2,2) nonane.[209]
Indeed, Adamek et al.'s language is so broad that Dunlop's Woodhams testified that it encompasses an infinite number of compounds.[210] Hercules's Dr. Charles Price testified that the number is very large, perhaps a million or more,[211] a conclusion in which Exxon's two experts Dr. John Stille[212] and Mr. Irving Marcus[213] each joined.
As in In re Ruschig, Adamek et al. now attempt to amend their application to cover a more narrow class of chemical constituents to wit, EPT's containing "a bridged-ring hydrocarbon containing at least two ethylenic double bonds, at least one of said double bonds being present in one of the rings of the bridged ring present in said hydrocarbon."[214] As in Ruschig, *684 however, the specification does not contain any language remotely resembling that of these amended claims 1-2, 4-7, 10 and 16. Accordingly, the Court invalidates these claims.
The breadth of claims 1-2, 4-7, 10 and 16 must be contrasted with the narrowness of claims 13-14 and 17-18. The latter group of claims name DCP as the third monomer. Since Adamek et al.'s specification explicitly specifies and exemplifies DCP's use, the rule of In re Ruschig does not prevent the enforcement of these claims.
These rulings are particularly appropriate in light of what appears to be a calculated effort on behalf of Adamek et al. to claim the benefit of far more knowledge than they actually generated.
Indeed, Adamek et al. knew very little about their invention and the third monomers that might work when they filed their July, 1957 patent application.[215] Thus, Adamek et al.'s draft application, which contained thirty-one examples,[216] exemplifies the use of only four termonomers, including three dienes: DCP, norbornadiene, and methyl cyclopentadiene dimer,[217] and one monoolefin: norbornene.[218] The application specifically mentions only four other possible bridged-ring termonomers: tricyclopentadiene,[219] tetracyclopentadiene,[220] bicyclo(2,2,2) *685 octa-2-ene[221] and bicyclo(2,2,2) octa-2,5-diene.[222]
Hercules, which assumed the lead in commercializing Adamek et al.'s inventions, narrowly focused its attention on the use of DCP until 1963 when its use began to present problems. On December 10, 1963, Keim thus noted:
The preparation of a sulfur vulcanized EPT is complicated by the fact that the two double bonds in dienes such as DCP are both reactive with Ziegler-type catalysts. The more reactive double bond in DCP is in the bridged six-member ring; it is about 30 times as reactive as the double bond in the five-member ring. Therefore, it will be used almost exclusively to incorporate DCP into the growing polymer chain. Occasionally [however], the DCP will add through the double bond in the five-member ring. When this happens, the reactive bond in the bridged ring will react rapidly with another growing chain to form a long chain branch or cross-link.[223]
The result was excessive cross-linking resulting in reactor fouling. Keim suggested designing a better diene[224] and stressed the need for its development.[225]
By August 13, 1964, Hercules learned of additional problems with using DCP:
Our EPT appears to be less stable in the presence of heat and oxygen than does [Gresham et al.'s product]. . . . Treads of our EPT appear to be more subject to groove cracking during road tests than does [Gresham's product]. These two phenomena are not known with certainty to be inter-dependent; however, heat aging of our EPT results in an increase in cross-linking and therefore a higher modulus-higher modulus can be expected to result in a tendency to crack.[226]
Eight days later, Hercules's B. F. Brown summarized what was rapidly becoming a gloomy picture:
We have uncovered a deficiency of our EPT, as we know it today, relating to instability and to tire tread groove cracking. "Instability" refers to oxidation as well as to cross-linking or gelation of the raw gum in the absence of added curatives. These problems . . . are real and serious. Unless resolved, our entire position in EPT may be in jeopardy.[227]
In order to "salvage [its] position, Hercules sought to uncover an EPT which would cure rapidly, would not groove-crack and yet [would] possess the other physical properties necessary to a good rubber." To this end, Keim and Christman screened twenty-two dienes between October, 1964 and June, 1965, running "some two hundred polymerizations."[228]
As Hercules was realizing the limitations of DCP and searching for a suitable substitute, it began to learn of the prior success of Du Pont and Exxon with EPT's containing MNB.[229] On November 22, 1963, Dafter thus reported to Miller:
Dr. [G. I.] Keim [one of Hercules's researchers] called me this morning to advise that our latest information on Du *686 Pont's EPT indicates that it is made in accordance with [its MNB Patent]. . . This patent issued June 11, 1963, on an application filed March 29, 1960. Its claims are very much like the Tarney patent except that the third monomer is defined as [MNB] instead of dicyclopentadiene.
Dr. Keim also informed me that the Du Pont EPT cures at a lower temperature than Hercules' EPT and that this could be a material advantage to the trade.
Dr. Keim's prime question was whether or not our Adamek et al. application could be construed to cover the use of [MNB] as the third monomer in the manufacture of EPT. I told him that I think there is reasonable basis for interpreting one or more of the present claims of the application as covering the terpolymers of [Du Pont's MNB Patent]. At the present time there is nothing we can do except bear in mind the information given by Dr. Keim and to make every effort to obtain as broad claims as possible when the Adamek et al. application [which was then in interference with Tarney's patent] returns to Ex parte prosecution.
(Dr. Keim also informed me that Enjay's [now Exxon's] new EPT apparently does not contain dicyclopentadiene but may contain instead a substituted norbornene similar to that used by Du Pont.)[230]
Three days later, another Hercules memorandum reported: "Enjay EPT-an unexpectedly good polymer based on preliminary sample-appears easier to cure than our current one. May become our most serious competition."[231]
Thereafter, Hercules restructured its application introducing more specific claims, just broad enough to cover EPT's containing MNB.[232] After several submissions and interviews,[233] Adamek et al.'s patent including claims 1-2, 4-7, 10 and 16 issued.
In sum, Adamek et al., who knew little about the scope of their invention, submitted a broadly worded specification covering myriads of chemical agents worthy of further testing. When Du Pont and Exxon completed that testing, Adamek et al. carefully reworded their claims so as to cover only the most worthy agents. Insofar as Adamek et al. seek to benefit from disclosures they did not make and research they did not initiate, the Court must void their claims.
Hercules insists, however, that Adamek et al.'s original application enabled a skilled artisan to reason that the polymerization double bond "must be in a bridged-ring" and that the "structure containing the [vulcanizable] double bond [which is] capable of wide variation" might be located outside of any ring whatsoever. Hence, Hercules argues, Adamek et al.'s specification revealed the use of MNB.[234] Although, in hindsight, it might seem obvious that MNB and its homologues work for the same reason as DCP, the test is whether Adamek et al.'s application prospectively discloses the workability of MNB and its homologues. Since Adamek et al. failed to mention or exemplify the use of any termonomers which, like MNB, contain one double bond in a bridged ring and another outside of any ring whatsoever, the Court cannot say that the choice, viewed prospectively, is as obvious as Hercules would make it seem. Indeed, Hercules failed, at least until after learning of the success of Du Pont and Exxon, to discover MNB's suitability.
Moreover, Hercules's argument erroneously focuses on the individual attributes of each double bond rather than their ability to work in tandem. First double bonds are useful only if they undergo Ziegler catalysis under conditions that do not affect the remaining double bonds. Second, third and fourth double bonds are useful only if they do not interfere with the polymerization reaction, but remain available for sulfur *687 vulcanization. Adamek et al.'s application fails to indicate, because Adamek et al. did not know, that MNB's double bonds would satisfy these requirements. In fact, scientists at Du Pont, Exxon and eventually Hercules were able to show the operability of MNB only after undertaking extensive research programs.
Exxon also presents clear and convincing evidence that Adamek et al.'s claims 1-2, 4-7, 10 and 16 should be voided for another independent reason-the rule of In re Cavallito-that an application demonstrate the utility of the claimed compounds. Indeed, Adamek et al.'s claims 1-2, 4-7, 10 and 16 present an even more compelling case for applying this rule. Cavallito sought to patent several hundred thousand compounds, thirty of which he identified and exemplified. Adamek et al. claim EPT's containing an unspecified but very large number of possible third components, yet their original application lists only eight possible termonomers and exemplifies only four.
In re Cavallito, of course, recognizes that an application covering many compounds may occasionally mention or exemplify only a few of their number. Hercules seeks to apply this exception, arguing that the broad class of compounds it claims all feature the same basic structure: a bridged-ring containing one or more double bonds.[235] Hercules ignores, however, the requirements for applying this exception. Adamek et al.'s specification fails to identify the key utility giving feature of their invention, be it a hydrocarbon, a hydrocarbon ring, a bridged-ring hydrocarbon or a bridged-ring hydrocarbon containing two double bonds, at least one of which is in the bridged-ring, etc. Moreover, their specification fails to disclose compounds having nearly equal utilities. Indeed, EPT's containing either norbornene[236] or dimers of methyl cyclopentadiene,[237] are far inferior to those containing DCP.
Claims 1-2, 4-7, 10 and 16 must again be contrasted with claims 13-14 and 17-18 which name DCP as the third monomer. Adamek et al.'s specification clearly exemplifies DCP's use, and indicates its volcanizability. The Court, therefore, finds that the rule of In re Cavallito does not invalidate these claims.
Exxon poses other, less worthy, objections to the quality of Adamek et al.'s utility disclosures. Exxon's first charge relies primarily upon its witness, Dr. Maurice Morton, who claimed that the single property differentiating rubbers from other materials was the ability of the former to retract almost completely following stretching.[238] Morton claims that since Adamek et al. did not include any numerical data regarding retraction, their specification failed to justify the claimed utility of even their exemplified products. Morton supported his claim by producing a portion of a 1962 American Society for Testing Materials (A.S.T.M.) definition:
RUBBER. As elastomeric material that can be or already is modified [usually through sulfur vulcanization] to a state exhibiting . . . quick and nearly complete recovery from an extending force, [to wit:]
(1) It is capable of being stretched 100 per cent, and
(2) After being stretched 100 per cent, held for 5 min. and then released, it is capable of retracting to within 10 per cent of its original length [90% recovery] within 5 min. after release. . . .[239]
Morton's testimony counts for very little because it is premised upon an extreme definition of a rubber. A.S.T.M. did not attempt to define rubber until September, 1960, when it proposed the standard about which Morton testified. This relatively strict standard achieved only tentative status,[240] a fact that Morton did not at first reveal. During 1962, A.S.T.M. revised this *688 standard to require only a 50% recovery rate rather than 90%, as initially proposed.[241] This latter definition is more in line with that normally associated with the term "rubber", which in actuality implies no more than the ability to stretch and retract.[242] These properties may be tested without reporting quantitative data. "[S]imply stretch the material and watch the return. This is [invariably] done in [our] factory as a good rough and ready test of elastomeric properties."[243]
Morton's testimony also improperly discounts as meaningless[244] such terms as "elastic", "fair recovery" and "some recovery", appearing on Adamek et al.'s data sheets and in their patent specification. According to Adamek et al., these terms are a sufficient indication of utility. This position seems well supported by many rubber technology patents wherein such terms are commonly used without including the underlying numerical data.[245] Indeed, Morton willingly accepted such terms when used by others besides Adamek et al. as sufficiently indicative of utility as a rubber.[246] "I had assumed that the [other] authors were knowledgeable in the matter and would not call an elastomer a substance that, let us say, did not recover well and in fact retained its original extension." The Court sees no reason to presume that Adamek et al. were any less competent than any of the other scientists appearing at trial or holding patents in this field. Since the other scientists accepted such terms as indicating an adequate level of retraction, so, too, will the Court. The Court finds, therefore, that Adamek et al.'s application adequately discloses the rubbery nature of the products that are otherwise disclosed.
Exxon next attacks Adamek et al.'s reported data as inconsistent with the laboratory work on which it is supposedly based.[247] Adamek et al.'s draft application includes a table showing that some of their products were so poor that recovery could not be tested and that other products showed no better than fair to poor recovery.[248] Adamek et al.'s patent application, Exxon claims, ignores some of these unfavorable results, and, in at least one example, reports data where, in fact, none was gathered. Exxon does not argue, however, that Adamek et al.'s utility claims are inaccurate or overblown, and the Court has been unable to understand how the lone fact that Adamek et al. might have neglected to report some of the results affects the adequacy of the utility disclosures.[249] Exxon also faults the application for claiming the use of inoperative compounds in violation of the rule of Graver Tank & Mfg. Co., that prevents an applicant from claiming operative as well as inoperative compounds. Exxon argues, without opposition, that the involved claims cover EPT's containing vinyl norbornene ("VNB")[250] or methyl cyclopentadiene dimer. Exxon charges that such polymers are inoperative and that the Adamek et al. claims are, therefore, overly broad.
The validity of this charge turns upon the meaning of the term "operative" as applied *689 to the field of rubbermaking. The Patent Office apparently dealt with this question, defining "operativeness" as relating to "the matter of success in being able to [prepare[251] and] cure the elastomeric" products.[252] EPT's containing VNB meet this definition.[253] So too, do EPT's containing methyl cyclopentadiene dimer. In fact, Hercules reported:
Methylcyclopentadiene dimer (MCPD) appeared to be as reactive as endo-DCP in terpolymerizations. Mileage was slightly poorer and the level of unsaturation must be determined by titration rather than by infrared analysis. The properties of the vulcanizates were inferior to those made with the DCP.[254]
Exxon thus failed to show that Adamek et al.'s claims covered inoperable polymers.
Exxon also charges Adamek et al. with violating the doctrine of Muncie Gear Works, Inc. v. Outboard Marine & Mfg. Co.,[255] that a patentee cannot rely upon the filing date of its original application if that application "in no way suggested the combination" that a later application "asserted as his invention".[256]
The instant case, however, presents a slightly different situation than that involved in Muncie Gear. Outboard Marine, the patentee there, submitted an application that did not include any language covering the allegedly infringing matter. Contrastingly, Adamek et al.'s original specification may have claimed the allegedly infringing matter, but only in such broad terms that it is impossible to know the metes and bounds of the claim. Although Exxon cites a number of cases applying the Muncie Gear doctrine,[257] the Court has discovered only one case, Kahn v. Dynamics Corp. of America,[258] holding that Muncie Gear prohibits the use of an overly broad, as opposed to a silent, application for supporting amendatory claims. Even this holding has been criticized as unnecessary in light of Section 112's description requirement.[259] The Court, however, need not resolve these difficult disputes surrounding the scope of the Muncie Gear doctrine at the present time, since the doctrine at most provides an additional reason for invalidating Adamek et al.'s previously invalidated claims 1-2, 4-7, 10 and 16.
PATENT LAW FRAUD
Exxon contends that Adamek et al.'s patent is invalid and unenforceable because of numerous allegedly fraudulent events. The Court finds that Exxon failed to prove any of its contentions.
The Law
Patent law fraud traditionally consists of two elements, each of which must be established by clear and convincing evidence. The first element is misrepresentation, which may consist of either an intentional lie or a deceitful or reckless suppression of relevant information. The second element is materiality. A misrepresentation is material when its existence might have influenced a patent examiner.[260]
The Facts[261]
Exxon first charges that Hercules's Dafter should have cited the Examiners to *690 the patents within Exxon's second groups of prior art patents,[262] particularly that of Gresham et al. which Exxon concedes to be the most relevant element of prior art,[263] but that, through either deceit or recklessness, he failed to do so, and that his failure materially affected the Examiners' decision to grant the Adamek et al. patent. As evidence of Dafter's deceit, Exxon cites only the following exchange between its counsel and Dafter:
Q. Well, is there anything about the references that has changed so that you would not at that time think, as you do today, that Gresham is the closest?
A. It is hard for me to resurrect the state of mind in 1960 to '65 to know that anything has changed or not. A lot of things have happened over the years in my mind.
Q. I know there have, this lawsuit for example.
A. Yes.
Q. An examination of the Adamek patent in inter parties [sic] proceedings. That has happened. Does that help change your mind?
A. I have learned quite a bit from this lawsuit, Mr. Reedy, quite a bit.
Q. Is this just one of the things that you learned the pertinency of Gresham?
A. Not the pertinency of Gresham, but the fact that what I should have done, in the light of this lawsuit, is a separate question. I have learned a lot.
Q. It will only get worse instead of better if I keep going.
A. Perhaps we wouldn't have had this lawsuit if I had cited Gresham, I don't know.[264]
Exxon apparently interprets this testimony as an admission that Dafter practiced deceit in refusing to cite the patents of Gresham et al. and others and that but for his deceit this lawsuit would have been unnecessary. The Court rejects this interpretation. The interpretation flatly contradicts Dafter's other testimony that his refusal to cite the patents of Gresham et al. and others was motivated only by his belief in their irrelevance.[265] A far more likely interpretation, which comports with Dafter's other testimony, is that he now wishes he had cited the Examiners to the patents of Gresham et al. and others so that the presumption of validity would have operated more strongly in favor of Hercules, whose position regarding obviousness would have been strengthened, possibly enabling enforcement of its patent rights without resorting to this litigation. Another likely interpretation is that had Dafter been able to foresee the future trends that made patent law fraud a commonly asserted claim, he would have cited the patents of Gresham et al. and others, whether or not he thought they were relevant in order to avoid the risk of being charged with fraud.
The Court does not find that Dafter's decision to avoid citing the patents of Gresham et al. and others was reckless under the circumstances of this case. Failure to report scientific data often constitutes fraud because the Patent Office does not have any independent facilities for gathering its own information.[266] In contrast, the Patent Office has adequate capacity to search out possibly relevant prior art references. Indeed, the Patent Office maintains a library of prior art references as well as its own indexing and, before issuing a patent, Examiners devote considerable time to studying this index and *691 the art it accesses.[267] Courts are therefore reluctant to find fraud in an attorney's failure to cite prior art references.[268]
In United States v. Standard Electric Time Company, the United States, claiming fraud, sought to cancel a previously issued patent. The basis of its claim was that while the applicant knew of four prior art masters' degree theses, he decided against reporting them because of their irrelevance to the Patent Office. Judge Wyzanski ruled in the applicant's favor:[269]. 155 F.Supp. 949 (D.Mass.1957).
Of course, a putative inventor must disclose any printed publication which he either knows or believes describes the very invention claimed. United States v. American Bell Telephone Co., 128 U.S. 315, 355-356, 9 S.Ct. 90, 92-93, 32 L.Ed. 450. More than this, if he knows of a printed publication which plainly describes his claimed invention, or comes so close thereto that every reasonable man would say the invention claimed was not original but had been anticipated, then regardless of his personal view that he is the original inventor, he will not be excused for his failure to disclose his knowledge. But the applicant has no duty to cite every publication of which he knows, or which he has used, merely because the publication is one likely to be referred to by a vigilant examiner in the Patent Office, or by a rival in an interference or other proceeding. It is not the object of the quoted statute or rule to supply all available evidence to the Patent Office, or to force the applicant to set up what he regards in good faith as straw men which he reasonably and in good faith believes he can knock down.[270]
*692 In attempting to show that Dafter violated this rule, Exxon stresses the import of a September 18, 1961 memo from Dr. Eleanor R. Bartholomew to Hercules's Chief Patent Counsel E. G. Peterson. This memo notes that in view of "the Keim-Voris letter, it should be possible to argue that the Dunlop terpolymer has improved properties, at least insofar as ozone resistance is concerned, over [Gresham et al.'s] terpolymer."[271] The "Keim-Voris letter"[272] fails, however, to provide any valid basis for distinguishing Gresham et al.'s invention from that of Adamek et al. Indeed, its only passage regarding ozone resistance notes, "Since the unsaturation is presumably in the side chain rather than the backbone, ozone or other reagents should not cause excessive degradation." Gresham et al.'s polymer also features the location of residual unsaturation within a side chain.[273] Exxon apparently believes Bartholomew wanted to distinguish between the inventions of Gresham et al. and Adamek et al., but that she was unable to do so because the two patents involved indistinguishable inventions. Exxon's argument ignores Bartholomew's purpose, which was to critique the Adamek et al. application in preparation for licensing negotiations[274]-not to defend it. Exxon's argument also ignores another paragraph of the Bartholomew memo which presents a well-founded distinction[275] between all possible prior art and Adamek et al.'s invention:
[I]t should be possible to argue that the bicyclo dienes, such as dicyclopentadiene, are in no way equivalent to the aliphatic dienes and that the teaching of the two parents cannot be combined. In [the Gresham et al. patent] it is emphasized that the aliphatic diene [the termonomer] must contain at least one terminal double. ... Since the double bonds of dicyclopentadiene are both internal double bonds, it can be argued that there would be nothing to teach that such a cyclic diene could be substituted for the aliphatic diene of [the Gresham et al. patent].[276]
Therefore, as early as 1962 Bartholomew articulated the key and significant distinction between Gresham et al.'s work and that of Adamek et al. Hercules believed in this distinction as evidenced by its eventual purchase of Adamek et al.'s American patent rights for $140,000 plus two percent of net sales,[277] and its subsequent investment of considerable time and effort toward developing these rights. Based on this evidence, the Court is unable to conclude that every reasonable person would agree that Adamek et al.'s invention was rendered obvious by Gresham et al.'s or other concededly less relevant patents. Indeed, the Court finds that Exxon fails to adduce even a preponderance of evidence on this fact, much less enough to convince every reasonable person.[278]
Moreover, Dafter's failure to cite the Gresham et al. patent did not affect the Examiners' determinations regarding the Adamek et al. patent since the Examiners were aware of Gresham et al.'s patent or equivalent information. Hercules's attorneys cited the Examiners to an article by Natta, disclosing information very similar to that contained within the patents of Gresham et al. and others.[279] More importantly, the Examiners searched Class 260/Subclass 80.5, containing Gresham et al.'s patent,[280] during July, 1962.[281] It *693 stretches credulity to believe that Examiners failed to uncover the Gresham et al. patent during this search.[282] Exxon insists that the Examiners soon forgot its possible relevance, enabling Adamek et al.'s 1964-65 claim alteration. The problem with this argument is that, as Adamek et al.'s patent neared issuance, Hercules successively narrowed its claims.[283] If the Gresham et al. patent was irrelevant to the earlier broader claims, it was certainly irrelevant to the later narrower ones.[284]
Furthermore, the Patent Examiners working on the Adamek et al. patent application often cited the Gresham et al. patent Indeed, one Examiner L. P. Quast began working on the Adamek et al. application on March 28, 1960, at the same time that she was completing her work on the Gresham et al. application[285] which issued three weeks later.[286] That Quast remembered Gresham et al.'s patent[287] is shown by an October, 1960 decision of hers denying Du Pont's claims for trimers of ethylene, another alpha-olefin and a non-conjugated, straight chain diene.[288] The basis of her decision was that these claims were "fully met by Gresham et al., which discloses terpolymers of ethylene, propylene and non-conjugated dienes...."[289] Quast was succeeded during 1964 by W. H. Hoover and J. A. Seidlich, along with their supervisor J. L. Schofer.[290] That these Examiners knew about Gresham et al.'s patent is demonstrated by their numerous citations of it. Hoover and Schofer first cited Gresham et al.'s patent in October, 1963, when they denied Shell Development Company a patent covering a "process for the terpolymerization of mono-olefins."[291] On December 8, 1964, Schofer denied as unpatentable *694 over Gresham et al. an application to Hercules claiming "sulfur vulcanizable copolymers[292] Six weeks later, Schofer, Hoover and Seidlich rejected an application submitted by Shell Development Company, covering "sulfur vulcanizable olefin copolymers and process for their preparation ... as unpatentable over Gresham, et al.... Gresham [et al.] is seen to anticipate the essential limitations of the instant claims. In particular, Gresham [et al.] teaches the incorporation of straight and branched chain dienes into an alpha-olefin polymer."[293] On March 9, 1965, Schofer and Seidlich used Gresham et al.'s patent to reject another Natta application covering copolymers of diolefins and olefins and method of production.[294] In late March, Schofer and Hoover rejected, in light of Gresham et al., another patent submitted by Du Pont, claiming "a new catalyst for the polymerization of hydrocarbon monomers."[295] On May 13, 1965, Hoover and Schofer cited Gresham et al.'s patent in rejecting a filing by Du Pont Gladding and Ro, for a "chemical process."[296] The next day, Schofer and Hoover again cited Gresham et al. in rejecting another Du Pont application claiming a polymerization "process".[297] Finally, in August 1965, Schofer and Hoover denied Natta et al.'s claims for a patent covering olefinic copolymers and a process for preparing them[298] as unpatentable over Gresham et al. "The primary references teach interpolymers of at least one alpha-olefin and at least one diolefin. ... Gresham et al. also teaches the advantages of adding dienes to monoolefin polymers."
Although the evidence is not entirely conclusive, it indicates that the Examiners rejected Gresham et al.'s patent as prior art because it is irrelevant to the subject matter involved by Adamek et al.'s application. Thus, they failed to cite Gresham et al.'s patent in considering other applications involving subject matter closely akin to that covered by Adamek et al.'s patent. Examiner Quast, in reviewing Tarney's patent involving subject matter so similar to Adamek et al.'s that the Patent Office placed the two in interference with each other,[299] searched Class 260/Subclass 80.5 on January 21, 1961.[300] Although Gresham et al.'s patent was in this file, she apparently judged that it was irrelevant to a proper consideration of Tarney's patent, and did not cite it. Quast also worked on Du Pont's patents, claiming sulfur vulcanizable EPT's containing MNB.[301] Her work caused her to search Class 260/Subclass 80.5, during August, 1960.[302] Once again, though, Quast judged that Gresham et al.'s patent was irrelevant and did not cite it.
Exxon counters that many of the above cited cases did not concern the same subject matter as Adamek et al. or that they represented citations of Gresham et al. that could not have been used against Adamek et al.[303] Many of Exxon's proposed distinctions, however, are weak, and Exxon does not attempt to distinguish Hoover and Schofer's citation against Gladding and Ro. Assuming arguendo the validity of Exxon's distinctions, an important truth still emerges. At about the time the Examiners were working upon Adamek et al.'s application, *695 they were also deciding whether other applications were allowable. Their decision making process led them to Gresham et al.'s patent. They read it, decided it was relevant and cited it. Indeed, they occasionally cited passages from Gresham et al.'s patent very near those cited by Exxon as anticipating the Adamek et al. patent. Thus, the Examiners knew about and understood the meaning of Gresham et al.'s patent and the fact that Dafter refused to cite it is simply irrelevant.
Exxon's next fraud charge involves a series of experiments by Woodhams and Christman. Exxon claims that Christman and Woodhams fraudulently suppressed some of their results and that Hercules was thus able to argue that [VNB] terpolymers were operative when, in fact, Exxon claims, they were not.
The Christman and Woodhams experiments began with a February 2, 1965 letter from Miller instructing Christman to "duplicate Example II(a) of the Adamek et al.
Subsequently, on March 1, 1965, Christman prepared an affidavit, which he submitted to the Patent Office, stating:
I was asked by counsel to duplicate a representative working example of the Adamek et al. application....
. . . . .
I chose ... to use [VNB] as the unsaturated bridged-ring hydrocarbon. ...
. . . . .
The product of this experiment was a colorless gum. It was rubbery in character and had the properties of extensibility and retractability. It could also be described as an elastomer.
. . . . .
... application exactly, substituting [VNB] for the DCP.[304] ... The product should be compounded and cured using the exact formulation and procedure set forth at pages 6 and 7 of the application." Christman prepared the sample as instructed and shipped it to N.A.R.C., writing Woodhams, "Our understanding is that you will vulcanize this polymer using the recipe described in the patent application, press ring molds and determine the extensibility of the sample."[305] On February 18, 1965, Woodhams cabled Hercules "TEST RESULTS COMPLETE ON EPT SAMPLE CURING TEST POSITIVE BUT PROPERTIES INDIFFERENT WILL CALL FRIDAY AS PROMISED."[306] That same day, he prepared a letter to Hercules, reporting the following results:
Extension in Centimeters
Sample without Sample with
Load (1 min.) Curatives Curatives
500 grams 8.0 1.6
1000 grams broke 3.5
1500 grams broke 7.5
2000 grams broke broke
Dr. Woodhams [cured the product, tested it] and reported back to me the following results. [Christman then quoted from the above cited portion of Woodhams's letter tabulating his results.]
. . . . .
In view of the above results, the copolymer which I have prepared from ethylene, propylene, and [VNB] in accordance with working Example II of the subject application is one which possessed a sufficient degree of unsaturation to enable it to be vulcanized with sulfur. In other words, it was sulfur-vulcanizable or sulfur-curable.[307]
Woodhams simultaneously prepared and submitted an affidavit, stating:
On February 16, 1965, I received in the mail a sample of a material in form of a *696 colorless gum from Dr. Donald L. Christman of Hercules Powder Company, Wilmington, Delaware, U. S. A.; ... I was asked to determine whether the sample was sulfur-vulcanizable using the identical test procedure set forth in the application at pages 6 and 7; I did this and reported the results as follows to Dr. Christman.... [Woodhams then quoted from the above cited portion of his February 18th letter, tabulating his results] ... It is my opinion that the gum which was prepared by Dr. Christman and submitted to me for compounding, vulcanizing and testing was sulfur-vulcanizable; the test results clearly so indicate.[308]
Exxon charges that Woodhams and Christman fraudulently suppressed the following information that Woodhams first reported in his February 18th letter:
You will note that we could not employ the 3000 gram load because the sample did not cure up to give a high tensile strength. In our experience vinyl substituents are not very suitable for vulcanization purposes and so the poor results might be explained on this basis. An allyl substituent or any unsaturated substituent containing easily accessible alpha-methylenic hydrogen atoms would be preferred.
However, the above example would nevertheless be satisfactory for purposes of illustration.[309]
Although Woodhams admittedly believed that the extent of cure was poor,[310] Exxon fails to present any evidence demonstrating that the Examiners were sufficiently interested in the level as opposed to the fact of curing to justify the conclusion that Hercules intentionally or recklessly failed to report Woodhams's observations or that such observations might have affected the Examiners. Indeed, Woodhams testified that his observations were entirely irrelevant. VNB terpolymers cured and, given time, better cures could be effectuated:
The vinyl substituent of the type that would be attached to the norbornene nucleus in this monomer has a different level of reactivity than, say, arylic type unsaturation, so that the rate of vulcanization or the extent of vulcanization would be different with these curatives when a vinyl substituent was used. And in most cases the curing system has to be modified in order to take the difference in reactivity into account. In this case, however, we were required to use exactly the same curative concentrations and curing conditions as employed in the original patent. And since we had such a small sample, we could not carry out experiments to optimize the cure.[311]
Exxon insists, however, that in addition to sulfur vulcanization, the Examiners were also concerned that EPT's containing VNB might become excessively cross-linked during polymerization and that Woodhams's observations somehow spoke to this issue.[312] Exxon ignores the fact that the Examiners did not raise the possibility of excessive cross-linking until at least as late as April 27, 1965, nearly two months after Christman and Woodhams submitted their affidavits.[313] Moreover, excessive cross-linking usually results in the production of overly stiff polymers.[314] Yet, the withheld information indicated, if anything, that VNB terpolymers were not stiff enough even to support a 3000 gram load. To the extent that excessive cross-linking was a relevant consideration, Hercules had no incentive to withhold the subject information, and the Court sees no reason for believing that Adamek et al.'s prosecution was materially advanced thereby.
*697 Exxon seeks to bolster this fraud charge by referring to reproduction runs undertaken during trial by one of its employees Dr. Edward Kresge. Exxon claims that these reproduction runs "demonstrate the [unsatisfactory] nature of the polymeric materials obtained by Christman and tested by Woodhams in 1965".[315]
The Court can afford little, if any, weight to Kresge's alleged reproduction experiments because Hercules identified important differences between the Christman-Woodhams and Kresge experiments. Christman reported the use of a closed reactor,[316] i. e., one from which unused reactants could not escape. Kresge, however, used an "open end" reactor,[317] i. e. one from which unused reactants could escape. Different types of reactors can, in turn, produce differently constituted products.[318] Thus, while Christman's products contained only about one and three quarters (by weight) times as much ethylene as propylene,[319] Kresge's products contained about twice[320] as much ethylene as propylene.[321] Moreover, where the viscosity of Christman's products was about 1.5, the viscosity of Kresge's products was about 4.3.[322] This indicated that Kresge "made a higher weight polymer than Dr. Christman produced."[323] Furthermore, Christman's polymers in their uncured state, could be stretched to several times their original length. Yet, Kresge's polymers were so stiff that even in their uncured state, they hardly extended at all. Upon curing, Christman's polymers became a great deal stiffer; Kresge's showed little or no change. In short, Kresge's products were very different from those of Christman. There is, therefore, little, if any, reason to consider Kresge's products.
Exxon next charges fraud against Hercules for failing to report the results of another subsequent test. Christman and Woodhams began this test on March 5, 1965, shortly after submitting their affidavits. On March 10, 1965, Christman shipped the second polymer to Woodhams[324] who reported to Miller:
The test results on the elastomer sample ... are now complete and are summarized in the following table. The sample was treated in exactly the prescribed manner.
Elongation (cm.)
Load (grams) Uncured Cured
500 3.5 2
1000 13 4.5
1500 23.5 9.4
2000 broke broke
It is evident that the cured specimen shows evidence of vulcanization or cross-linking.[325]
Exxon claims that Hercules hoped that the test would yield more favorable results than the first, but that when it did not, Hercules practiced fraud by failing to reveal *698 its results to the Patent Office.[326] Exxon's claims suffer several weaknesses, however. Exxon fails to adduce any evidence concerning Hercules's expectations for its second test. Moreover, Exxon fails to prove that the second test yielded poorer results than the first. Indeed, Hercules proves that it was impossible to tell which experiment was more successful.[327] Furthermore, Exxon fails to show that the Examiners were interested in anything but the existence of a sulfur cure, a fact that the second test results served to confirm.[328] Exxon thus fails to prove both that Hercules acted either intentionally or recklessly, in failing to reveal Christman and Woodhams's second test and that the decision in any way affected the Examiners.
Exxon next levels a series of fraud allegations involving Hercules's April, 1965 submission of an affidavit by Keim reporting the results of another set of tests. The genesis of this affidavit was a March 19, 1965 Hercules strategy session at which:
E. G. Peterson reported on the most recent meeting (March 18, 1965) with the Patent Office re the "Adamek Case". . . . The supervising examiner has stated that he will allow the broad claims (e. g. alkenyl norbornenes) if we can show that a terpolymer prepared and cured as disclosed in the application, using an alkenyl norbornene, yields a rubber as good as a terpolymer using DCP prepared and cured in the same way. . . . Clint Miller has already met with Dr. Spurlin to get the necessary work started.[329]
Five days thereafter, Spurlin wrote Keim instructing him to:
Run two . . . duplications of Example VIII substituting in each case a different substituted 2-norbornene having an exocyclic double bond in the substituent. . . . Each of the copolymers should be compounded, cured, and tested to determine its curing characteristics using the extensibility test of the Adamek et al. application.[330]
Spurlin did not suggest any particular termonomer[331] stating only that the particular substituted 2-norbornene employed should be chosen from those presently available in our laboratories due to the time limitations involved.[332] Within this limitation, Keim perceived that he was free to chose any alkenyl norbornenes that he wished.[333] On March 30, Keim responded that he planned to use VNB in one of his experiments.[334] Keim later changed his experimental plan, however, testing MNB rather than VNB.[335] Exxon first claims that this substitution was fraudulent because Hercules hoped to obtain better results with MNB than with VNB.
Exxon's argument, premised as it is upon the idea that the mere act of selecting promising experiments can be fraudulent, goes far beyond available precedent which requires no more than the full disclosure of relevant experimental results. Further, Exxon fails to present any direct evidence of fraud and its circumstantial evidence depends upon a number of incorrect assumptions and suppositions. Exxon, thus, claims that Christman and Woodhams's VNB experiments were unsuccessful, which is incorrect.[336]
Exxon further claims that Hercules failed to deny or explain Keim's substitution. Hercules, however, offers a reasonable and benign explanation: after initially choosing VNB, Keim later discovered that VNB was not readily available to him, and that MNB *699 was.[337] Exxon next claims that Miller told Keim not to use VNB, but it fails to adduce even a shred of evidence that Miller ever communicated with Keim regarding his choice of possible third monomers. Furthermore, Hercules's personnel had little or no basis for believing that EPT's containing MNB would be any better than their VNB counterparts. On the one hand, Hercules prepared several EPT's containing MNB that were poor. As early as December 7, 1964, Keim thus reported "[MNB] caused gelling in nearly every instance that it used indicating that cross-linking had occurred."[338] Indeed, work between October, 1964 and June, 1965, led Keim and Spurlin to reject the use of "[MNB which] gave conflicting results, sometimes cross-linking during the polymerization, curing more rapidly than DCP-EPT but with the vulcanizate having only fair physical properties."[339] On the other hand, VNB yielded promising results.[340] In short, Exxon fails to present clear and convincing evidence that Keim's substitution of MNB for VNB was either misleading or material.
Exxon next claims that Keim's use of Adamek et al.'s Example VIII, rather than Example II which Christman and Woodhams used was fraudulent because "Example VIII presented the greatest opportunity for success."[341] The Court, once again, expresses doubt that the mere act of choosing promising experiments can, by itself, be either misleading or material. Assuming arguendo that it can be, Exxon fails to produce a shred of evidence confirming its suspicions. Exxon focuses particularly on Keim's use of a 25° C. reaction temperature rather than the -20° C. temperature specified by Example II. Keim explained, however, that he chose a reaction temperature of 25° C. only because this was the temperature at which his laboratory was maintained. Keim felt this temperature was justified because "there are at least two examples in the patent where 20 degrees was used. The range of temperatures is minus 60 to plus 50, I believe, in the various examples."[342]
Exxon's next fraud claim involves minor variations between the procedures Keim actually followed and those specified by Adamek et al.'s Example VIII. Example VIII specifies the addition of 10 mls. (milliliters) of aluminum tri-iso-butyl (a catalyst component) to a reactor along with 300 mls. of petroleum ether (a solvent). 1 ml. of vanadium tetrachloride (the other catalyst component) is then dissolved in 100 mls. of ether and the combination is filtered to remove the undissolved vanadium tetrachloride. The filtrate, ethylene, propylene and 10 g. (grams) of DCP, are then added continuously to the reactor during the next hour. The resulting product is washed with plain methanol.[343] Keim generally followed this procedure except: (1) he separated the undissolved vanadium tetrachloride by carefully decanting the solvent and the dissolved vanadium tetrachloride; (2) after the reaction was complete, he precipitated the product using acidified methanol, filtered the product and washed it twice more with acidified methanol; and (3) Keim treated the resulting product with phenyl beta naphthylanine,[344] an antioxidant.
Keim did not mislead the Patent Office regarding these changes. Indeed, with one minor exception,[345] Keim's affidavit reported the exact procedures that he followed. *700 Assuming, however, that Keim's treatment was somehow misleading, it was immaterial. The Examiners would have reached the same conclusions regarding Keim's products because the properties of these products were not affected by the changes. Keim thus testified, without opposition, that careful decanting was equivalent to filtration,[346] and that he only undertook alternate procedures 2 and 3 in order to recover[347] and protect his products,[348] not to alter them.
Exxon next claims fraud[349] in Keim's reporting about only one, rather than the three, terpolymerizations he actually performed.[350] Exxon fails, however, to explain how this failure would have been either misleading or material.
Exxon's next fraud charge is that a Hercules May 24, 1965 affidavit by Spurlin was misleading in that it stated, "[I]t is a fact that [VNB] is operable. . . . A useful rubber has been produced from [VNB] in accordance with Adamek et al. and under conditions which make it attractive for commercial use,"[351] although Spurlin knew that EPT's containing VNB were subject to excessive cross-linking.[352] It is possible that the first EPT's containing VNB showed excessive cross-linking. As early as January, 1963, a Dr. Junghanns thus reported to Keim and Spurlin that VNB "terpolymerized but also cross-linked."[353] Junghanns simultaneously reported, however, that cross-linking could be avoided simply by dissolving the reaction catalyst in tetrachloroethylene,[354] and shortly thereafter, Hercules discovered that cross-linking could be avoided by using pure grades of VNB.[355] The result was that by May, 1965 scientists were regularly polymerizing useful EPT's containing VNB. Among these scientists were Keim and Christman who polymerized, cured and tested vinyl norbornene terpolymers, producing results similar to those obtained with such commercial grade diene termonomers as DCP and MNB.[356] EPT's containing VNB cured well.[357]
Moreover, workers at a number of other companies, including Farbwerke Hoechst, a German chemical maker, as well as Exxon[358] and Du Pont[359] prepared useful EPT's containing VNB. Hoechst reported its success in a February 19, 1965 report,[360] a partial English translation of which was circulated at Hercules.[361] The translation notes, among other favorable data, that terpolymerization with VNB produced less crosslinking than those with MNB and DCP.[362] Hoechst considered EPT's containing VNB merited further examination.[363]
Far from being fraudulent, Spurlin's May, 1965 affidavit therefore truthfully reported well-known scientific facts.
Exxon's next fraud charge is that Hercules misrepresented that Adamek et al. were pioneers in the broad EPT field, when, in fact, they were anticipated by Horne et al., *701 Natta et al. and Gresham, et al. As support for this charge, Exxon submits several closely edited excerpts from the Adamek et al. patent file. Adamek et al.'s statements in context, however, only claim the discovery of EPT's containing bridged-ring hydrocarbons. Exxon fails to show that this claim is either misleading or material.
Exxon first excerpts[364] Dafter's statement that:
Applicants' major contribution to the art is submitted to be a new class of rubbery copolymers of at least three different monomers.
. . . . .
The object of the present invention, therefore, is to provide a copolymer which has the essential attributes of elastomeric copolymers of ethylene and propylene but which contains some residual unsaturation and is sulfur vulcanizable.[365]
Although the cited material might be interpreted as claiming the invention of all EPT's, it does so only because Exxon excised a sentence that clearly indicated that Adamek et al.'s claims extended only to EPT's containing "an unsaturated bridged-ring hydrocarbon, e. g., dicyclopentadiene."[366]
Exxon also excerpts[367] Spurlin's affidavit statement that "[t]he problem faced by Adamek et al. was to find compounds that would not only provide the . . . structure known to provide sulfur-curability, but would provide also the capacity to copolymerize successfully with the particular [alpha]-olefins that make good elastomers."[368] Exxon cannot rely on this statement as claiming the discovery of all EPT's since Spurlin's next two paragraphs clearly indicate that Hercules's claim was limited to terpolymers containing bridged-ring dienes:
This discovery that the bridged-ring unsaturated hydrocarbons had such outstanding utility for this purpose was totally unexpected to those in the art who had not tried them. . . .
The recognizable new concept was that the successful materials are those hydrocarbons having a bridged-ring nucleus, with or without substituents, providing a double bond for copolymerization and a double bond for sulfur-vulcanization. (footnote omitted).[369]
Exxon next cites[370] Hercules's claim that Adamek et al.'s contribution was "a basic invention".[371] Hercules did not claim, however, that its basic invention covered all EPT's, but rather that "Adamek et al. have made a basic invention with the common characteristic of the termonomer being the bridged-ring hydrocarbon nucleus and the requirement that the hydrocarbon contains at least two double bonds."[372]
Exxon also excerpts Hercules's argument that "Eminent polymer specialists had tackled this problem [of imparting sulfur curability to elastomeric polymers] and had failed to do what applicants did . . . and note also that the eminent research establish [Du Pont] . . . did not get into applicants' field until after applicants' invention. . . ."[373] Exxon, however, ignores the immediately succeeding paragraph of Hercules's Patent Office submission, which states that Adamek et al.'s "basic contribution was the selection, as a termonomer, of bridged-ring hydrocarbons."[374]
Exxon next charges fraud in Hercules's May 24, 1965 argument that "[t]he .. others had attempted with no real success to . . . incorporat[e] a termonomer, *702 ... that would not only provide sulfur curability, but also would give successful, unimpaired copolymerization under the desired catalyst conditions."[375] Exxon claims that this argument is misleading in that others, including particularly Gresham et al. successfully polymerized EPT's. The next sentence of Hercules's Patent Office submission, however, clearly specifies the criteria it used for judging the success of an EPT. "This last means inter alia avoidance of gelling, attainment of desired molecular weight, avoidance of catalyst poisoning, and avoidance of excess monomeric material requiring costly recovery and recycling. ..."[376] Exxon failed to show that any prior inventions satisfied these criteria. Indeed, Hercules showed that Gresham et al.'s polymerizations required excess monomeric material in violation of Hercules's final criteria.[377]
Exxon charges fraud in the following statement from one of Spurlin's affidavits:
14. There was a lot of information extant in 1957 when the Adamek et al. application was filed on the requirements for sulfur-vulcanization....
15. In regard to the general theoretical aspects of vulcanization, it was well known by 1950 at the latest that the only requirement for sulfur-vulcanization was the presence of the [allylic] structure that is, a carbon-carbon double bond with a hydrogen on an adjacent carbon. The mechanism for the reaction of sulfur with an olefin (including the reactions of vulcanization) was first put forward by Farmer and co-workers.... These investigators studied the reactions of such compounds as cyclohexene and isobutylene with sulfur. Earlier workers had studied the reaction of trimethyl ethylene with sulfur. These studies were extended, prior to 1957, to numerous other unsaturated compounds by Bateman, Moore, and coworkers at NRPRA. This general field has been reviewed many times, e. g., Florey, Principles of Polymer Chemistry, pp. 454 et seq. (published 1953), Walling, Free Radicals in Solution, pp. 335 et seq. (published 1957). (footnotes omitted).[378]
Exxon claims that this statement was false, first because "some dienes which do not have the allylic structure do provide vulcanizable materials." (emphasis in original), (footnote omitted).[379] Exxon also faults Spurlin's statement because he failed to "disclose that in 1950, and in fact as late as 1960, there were conflicting theories on the subject [of vulcanization]." (emphasis in original).[380] Exxon fails to show, however, how these facts render Spurlin's statement misleading. After all, Spurlin's statement did not exclude, either directly or by implication, the possible existence of other useful structures besides the one on which he focused, or other reliable theories besides those of Farmer. Moreover, Exxon failed to show how Spurlin's failure to mention other available structures or theories might have affected the Examiners's decision.
Exxon also claims that Spurlin's statement was inaccurate because "some dienes which do have the `required structure' do not impart sulfur-vulcanizability."[381] (footnote omitted) (emphasis in original). As support for its claim, Exxon cites three laboratory reports covering EPT research at Hercules.[382] Although these reports may evidence some difficulty in preparing EPT's containing the allylic structure, they do not evidence any failure of the allylic structure to impart sulfur vulcanizability. Exxon, thus, fails to produce any other evidence showing that Spurlin's statement was misleading.
Exxon's next fraud charge concerns a May 12, 1965 interview at which:
*703 The Examiners were shown [by Hercules's personnel] samples of copolymer made with ... [VNB], vis., samples of .. (a) the copolymer solution, a somewhat sirupy clear liquid; (b) the uncured copolymer as recovered from solution, an elastomeric unshaped piece; (b) the "oil extended" copolymer, uncured, a jade green disc; (d) the same after compounding, a thin black rectangular piece; and (e) the same after the sulfur cure, also a thin black rectangular piece.[383]
Although there exists no record of the specific discussion regarding these samples, Exxon claims that Hercules probably used them for showing the operability of VNB terpolymers. Exxon claims that their use in this manner was fraudulent because the VNB terpolymer sample "was one of the terpolymers prepared by Hercules Research during the Christman-Keim screening program [and] was considered unsuitable for further investigation...." (emphasis in original).[384] This argument confuses operability and commercial success. Although Hercules might have wished to postpone marketing its VNB terpolymers, they were operable[385] and Hercules did not mislead the Examiners, if indeed it made this representation.
Exxon also claims that the VNB terpolymer sample "was definitely not made `in accordance with the Adamek et al. disclosure.'" (emphasis in original).[386] Exxon then lists a number of alleged differences between the procedures exemplified in Adamek et al.'s application and the procedures followed by Keim and Christman.[387] Exxon's list, however, is irrelevant. Although Keim and Christman may not have followed Adamek et al.'s examples, they did follow the procedures specified by Adamek et al.'s application. Where Adamek et al.'s specification discloses that "[o]ne method of preparing the copolymers of this invention consists in reacting the monomers together in a common solvent in the presence of a surface-active catalyst",[388] Keim and Christman used a Sutherland reactor[389] for contacting reacting monomers in the presence of surface active catalysts.[390] Where Adamek et al.'s specification discloses the use of a Ziegler catalyst,[391] Keim and Christman used a Ziegler catalyst.[392] Where Adamek et al.'s specification discloses that "[t]he molecular weight of the copolymer and its distribution may be modified and regulated by varying ... temperature."[393] Keim and Christman used a reaction temperature of 34°C.[394] Where Adamek et al.'s specification discloses that "[i]t has been found that the copolymers of this invention can be vulcanized with sulfur,"[395] Keim and Christman vulcanized their product with sulfur.[396] Thus, Keim and Christman accurately reported that their experimental procedures were in accordance with Adamek et al.'s application, and they did not commit fraud in so doing.
Exxon's next fraud charge involves a June 15, 1965, telephone conversation in which an examiner called Miller and informed him "that [the examiners] had decided *704 to allow [Adamek et al.'s] application."[397] Miller failed to file a record of this conversation. Exxon apparently believes that Miller violated Patent Office Rule 133(b), which provides, "In every instance where reconsideration is requested in view of an interview with an examiner, a complete written statement of the reasons presented at the interview as warranting favorable action must be filed by the application."[398] Miller, however, did not request reconsideration of the Examiners's announced decision which, after all, was favorable to Hercules.[399] Thus, he was not required to record his interview. Assuming the contrary arguendo, Exxon failed to show that Miller's actions were either misleading or material, rising to the level of fraud as opposed to the level of a mere violation of Patent Office rules.
Exxon finally charges fraud in the destruction of documents relevant to this litigation, to wit, the work diaries of Hercules's patent counsel. Exxon claims that it requested these work diaries in July, 1972, but that Dafter, who was charged with fulfilling Exxon's document requests,[400] failed to supply them and permitted their destruction during August, 1972.[401]
Dafter's failure to supply the work diaries of Hercules's patent counsel was neither intentional nor reckless. Exxon asked for them only as part of an extensive document request, including the following interrogatory among more than forty others:
State whether or not plaintiff has produced or identified all documents in the possession, custody or control of or known to either plaintiff or Dunlop which relate or pertain to the conception, reduction to practice, testing, development or evaluation of the subject matter of the Adamek et al. patent No. 3,211,709.
. . . . .
To the extent that the answer to this interrogatory is other than in the affirmative:
. . . . .
(c) Identify all documents in the possession, custody or control of or known to plaintiff or Dunlop which relate or pertain to the above matters.[402]
For purposes of this request, Exxon defined the term "document" as including,
without limitation, the following items, whether printed, or recorded, or reproduced by any other mechanical process, or written or produced by hand, and whether or not claimed to be privileged against discovery on any ground, namely; agreements; communications, including intracompany communications; correspondence; telegrams, memoranda, summaries or records of telephone conversations; summaries or records of personal conversations or interviews; diaries; forecasts; statistical statements; graphs; laboratory and engineering reports and notebooks; charts; plans; drawings; minutes or records of statements of policy; lists of persons attending meetings or conferences; reports and/or summaries of interviews; reports and/or summaries of investigations; opinions of counsel; records, reports or summaries of negotiations; brochures; pamphlets; advertisements; circulars; trade letters; press releases; drafts of any document; original or preliminary notes; batch sheets or pilot and manufacturing plants; and marginal comments appearing on any document.[403]*705 It is understandable that Dafter would fail to produce some documents from this extensive list. Exxon failed to produce any evidence that Dafter's oversight was either intentional or reckless.
The subsequent destruction of the requested diaries is innocently explained. Maynard Turk, Hercules's General Counsel, recounted that when he joined Hercules's Patent Staff in 1970, all patent attorneys maintained diaries. "[T]hey really were used as a control measure. They were something that was required by the then head of the Department, [Mr. Peterson][404] and he used them, I think, to sort of determine what was going on in the Department. They were something that was read for his benefit."[405] Hercules then retained these diaries for fifteen years. When, in August, 1972, Turk became Hercules's Chief Patent Counsel,[406] he stopped requiring patent attorneys to keep work diaries:[407]
I had not been in the department a very long period of time. I was sort of an outsider coming in really. I was following someone who had been head of the department for oh, approximately thirty years, and that's a pretty hard act to follow in a lot of ways. And I wanted to establish good rapport as quickly as I could. And I wanted to establish it in those areas where something could be done. And keeping diaries was not a very popular item. And so one of the first things I did was decide that we would not maintain diaries any more.
Diaries that were more than two years old were to be "shredded and gotten rid of in the normal way any kind of materials that were no longer useful were gotten rid of".[408] Turk "never gave ... a thought" to the possibility that the diaries might contain material relevant to any litigation.[409]
Moreover, Dafter's failure to produce the work diaries of Hercules's patent counsel, and their subsequent destruction are immaterial because the undisclosed diaries, which were only time records, did not contain any substantive information which would have aided Exxon's case. Indeed, this is precisely the reason that Dafter decided against producing them.[410]
HERCULES DOCTRINE OF EQUIVALENTS ARGUMENT.
In sum, this Court finds that Claims 1-2, 4-7, 10 and 16, are invalid, void and unenforceable, and rejects Hercules's claim that terpolymers containing MNB and its homologues infringe these claims. Claims 13-14 and 17-18 and 19, however, are valid. Hercules contends that these claims covering terpolymers containing DCP may be stretched to cover EPT's containing MNB or one of its homologues under the doctrine of equivalents.[411] This doctrine permits a patentee to cover a composition outside his claims providing that composition "performs substantially the same function in substantially the same way to *706 obtain the same result"[412] as its validly claimed invention.
The doctrine of equivalents is, however, inapplicable under the facts of the instant case. DCP and MNB and its homologues do not work in substantially the same way to obtain the same result. Although these compounds all contain two double bonds, one of which is located in a bridged-ring, DCP's second double bond is located within a ring, while the second double bond of MNB is located in a straight chain. This single structural difference leads to a host of differences between EPT's containing DCP and those containing MNB or ENB. In particular, DCP's second double bond is a more vulnerable target of Ziegler catalysis as compared with the second double bond of either MNB or ENB.[413] Consequently, EPT's containing DCP gel more easily, an undesirable result. Moreover, DCP terpolymers are more difficult to cure than EPT's containing MNB or its homologues.[414] This is a problem because "an uneconomically large proportion of dicyclopentadiene nomoner units must ... be present in the copolymer or an undesirably long cure time must ... be used."[415] On the other hand, DCP terpolymers are favored because they resist ozone attack better than EPT's containing MNB or its homologues.[416]
Hercules argues that "[t]he differences which the rubber industry had found between dicyclopentadiene terpolymers on the one hand, and methylene or ethylidene norbornene terpolymers on the other, are matters of degree, not kind."[417] The doctrine of equivalents, however, is only applicable where the results are the same. They cannot vary by degree. Though plaintiff may belittle degree differences, they can be extremely important. Thus, rubbermakers use MNB terpolymers where cost considerations are paramount[418] and DCP terpolymers where ozone resistance is more important.[419]
That the doctrine of equivalents should be found inapplicable seems particularly appropriate in light of Triax Co. v. Hartman Fabricators, Inc.,[420] where it is said, "[S]ome inventions by their very nature deserve slightly more extensive legal protection than others .... [I]f the patent is a pioneering patent, the patentee is allowed a wide range of equivalents, but if the patent is a narrow patent or the art is crowded the patentee is only allowed a correspondingly narrower range." Because their invention, though significant, constitutes a narrow contribution to a crowded art, Adamek et al. may not extend their control of EPT's containing MNB or its homologues which are well beyond the limits of their valid claims.
The foregoing Opinion constitutes this Court's Findings of Fact and Conclusions of Law as required by Federal Rule of Civil Procedure 52. Any issues not specifically addressed in this Opinion have nevertheless been considered by this Court and found to be without merit.
NOTES
[1] Hercules initially sued the Enjay Chemical Company. Pleading Paper 1. On March 29, 1973, Exxon substituted itself for its predecessor Enjay. Pleading Paper 207, ¶ 5.
[2] Plaintiff's Exhibit (hereinafter cited as "PTX-") 6, Col. 1.
[3] Plaintiff's Brief, pp. 21-24.
[4] PTX-6, Cols. 7-8 and 10.
[5] An alpha-olefin is a hydrocarbon containing exactly one carbon-to-carbon double bond which is located in the terminal position between the first and second carbon atoms in the structure's main chain.
[6] Fig. 4.
[7] Diagrams similar to those of ethylene and propylene can become very complicated when larger molecules are drawn. For this reason, chemists often omit depicting the single bonds. Instead, they draw the symbols of singly bonded atoms close to one another. When more than one hydrogen is bonded to a carbon atom, they include only one "H" and indicate with a subscripted numeral the number of hydrogens that are actually present. See Weininger, Contemporary Organic Chemistry 91 (1972). Ethylene and propylene may thus be depicted:
1 2 1 2 3 CH2 = CH2 CH2 = CH2 CH3 ethylene propylene | | | | | 1 2 | 1 2 | 1 2 | -|- CH2 CH2 | CH2 CH2 | CH2 CH2 -|- | | | | | | | |[8] A bridged ring polyolefin is a hydrocarbon, containing more than one double bond, in which the carbon atoms are arranged in a ring, or closed chain structure, and two non-adjacent carbon atoms in the ring are joined by a bridge. The parties disagree over the meaning of the word "bridged". Hercules argues that the bridge must contain one or more carbon atoms. PTX-33. Exxon argues that a bridge may consist of a bond alone. Defendant's Exhibit (hereinafter cited as "DX") 355, p. 32. The Court sees no reason for preferring the definition of Exxon, which party carries the burden of proof. Accordingly, it adopts Hercules's definition. The Court notes this disagreement, however, as further evidence of the inexact nature of Adamek et al.'s specification. See text accompanying notes 207-234, infra.
[9] 7, Fig. 12A. Chemists picture cyclic compounds by drawing only the carbon-to-carbon bonds. Chemists assume that carbon atoms occupy each of the figure's vertices. They also assume that each carbon atom is bonded to exactly as many hydrogens as are needed to occupy it in four bonds. See Weininger, Contemporary Organic Chemistry 116 (1972).
[10] Id., pp. 218-219.
[11] Hackh's Chemical Dictionary 231 (3d Ed. 1944).
[12] Tr., pp. 77-79.
[13] Id., pp. 407-411 and 6366.
[14] PTX-7, Fig. 7.
[15] Pleading Paper 119, pp. 23-29.
[16] Id., p. 48.
[17] Id., pp. 32, 41 and 48.
[18] Id., p. 50.
[19] Id., pp. 74-76.
[20] Tr., pp. 306-07 and 331-32.
[21] PTX-69; Tr., pp. 359-60.
[22] Tr., p. 376.
[23] PTX-79.
[24] Pleading Paper 78, pp. 74-75. Dudley did not make a record of his idea, however, because "we simply did not know that we should do this." Id., p. 83.
[25] Fig. 12A.
[26] Compare notes 9 and 25.
[27] Tr., pp. 354-55.
[28] Fig. 12B.
[29] Among these compounds were "1-hexyne, 4-vinyl cyclo-hexenal-1, 2-methyl-penta-2, 4-diene, 1, 3-pentadiene, 1, 8-nonadiyne, and phenyl acetylene." PTX-84.
[30] Id.
[31] PTX-92.
[32] PTX-91.
[33] PTX-113, p. 83.
[34] Id., p. 1.
[35] Id., pp. 29-30.
[36] PTX-113, pp. 31-37, 40-44 and 47-53.
[37] Id., pp. 38-39 and 45-46.
[38] PTX-476.
[39] PTX-113, pp. 57-63.
[40] Id., pp. 65-68.
[41] PTX-528; DX-678.
[42] PTX-113, p. 100.
[43] Pleading Paper 229, p. 53.
[44] PTX-113, p. 106.
[45] Id., p. 105.
[46] PTX-6.
[47] Pleading Paper 207, ¶ ¶ 6-7.
[48] Fig. 12.
[49] ">49.
Id.
[50] Defendant's Brief, p. 8.
[51] Pleading Paper 1.
[52] Defendant's Brief, p. 6.
[53] Although the vast majority of cases discuss the relationship between the presumption of validity and the question of obviousness, the presumption appears to extend as well to other questions including, for instance, the operability of an invention, and the permissibility of amending a previously filed specification. See e. g. Remington Cash Register Co. v. National Cash Register Co., 6 F.2d 585, 618 (D.Conn. 1925); R. Ederer, On Operability as an Aspect of Patent Law, 42 J.P.O.S. 398, 410 (June, 1960); Chisum, Patents § 11.04[6].
[54] Chicago Rawhide Mfg. Co. v. Crane Packing Co., 523 F.2d 452, 457-58 (7th Cir. 1975).
[55] Aluminum Co. of America v. Amerola Products Corp., 552 F.2d 1020, 1025 (3d Cir. 1977).
[56] 383 U.S. 1, 86 S.Ct. 684, 15 L.Ed.2d 545 (1966).
[57] Id., pp. 17-18, 86 S.Ct. p. 694.
[58] Chisum, Patents § 5.03[3] (1980).
[59] Plaintiff's Brief, pp. 64-79.
[60] Del Mar Engineering Laboratories v. United States, 524 F.2d 1178, 1184 (Ct.Cl.1975); Dunlop Holdings Limited v. Ram Golf Corp., 524 F.2d 33 (7th Cir. 1975); Allen v. Brady, 508 F.2d 64 (7th Cir. 1974); International Glass Co. v. United States, 408 F.2d 395, 187 Ct.Cl. 376 (1969); Amerline Corp. v. Cosmo Plastics Co., 407 F.2d 666 (7th Cir. 1969); Dunlop Company, Ltd. v. Kelsey-Hayes Co., 364 F.Supp. 1094, 1099 (E.D.Mich.1972).
[61] See e. g. Del Mar Engineering Laboratories v. United States, 524 F.2d 1178, 1184 (Ct.Cl. 1975); Dunlop Holdings Limited v. Ram Golf Corp., 524 F.2d 33 (7th Cir. 1975); International Glass Co. v. United States, 408 F.2d 395, 187 Ct.Cl. 376 (1969).
[62] Standard Oil Co. v. Montedison, S.p.A., 494 F.Supp. 370 (D.Del.1980); Chisum, Patents § 10.06 (1980).
[63] Mann v. Werner, 347 F.2d 636, 640, 52 CCPA 1578 (1965).
[64] Defendant's Brief, pp. 92-96.
[65] Whether "invention" covers constructive reductions to practice appears to be an open question. Possibly, Congress, by using the term "invention" as opposed to "reduction to practice" intended to exclude from Section 102(g)'s coverage constructive reductions to practice which are more particularly identified by Section 102(e). That Section 102(g) requires this construction, however, is unclear and there appear to be few good reasons for differentiating between constructive and actual practice reductions. The Court need not now resolve this difficult problem, however, because Exxon failed to adduce sufficient evidence showing a prior reduction to practice either actual or constructive.
[66] Chisum, Patents § 10.05 (1980).
[67] Thus, In Re Hafner, 410 F.2d 1403, 56 CCPA 1424 (1969), deciding that a German patent application was insufficient to support an American priority date, did not mention whether the German application had issued. See also Yasuko Kawai v. Metlesics, 480 F.2d 880 (C.C. P.A.1973).
[68] This factual examination is, by necessity, extensive. The Patent Examiners recorded their consideration of only a very limited number of patents that Exxon now claims as prior art. Indeed, they failed to record their consideration of American patents issuing to Brown, Parrish and Gresham et al. and Belgium patents issuing to Horne et al. and Natta et al. and a series of foreign patents issuing to Stamatoff. Each of these patents is discussed below, along with other patents cited by the Examiners.
[69] PTX-7, Fig. 2.
[70] Tr., p. 73.
[71] PTX-7, Fig. 8.
[72] Id., Fig. 9; Tr., p. 83.
[73] Tr., p. 85.
[74] Id., pp. 907, 1256 and 2682-83; Pleading Paper 152, pp. 6-7.
[75] Tr., p. 86.
[76] PTX-7, Figs. 10A and 10B.
[77] U.S. Patent 2,671,074; PTX-38; DX-662. It issued on March 2, 1954 upon an application dated November 24, 1950.
[78] U.S. Patent 2,809,372; PTX-39; DX-371. It issued on October 8, 1957 upon an application dated January 13, 1953.
[79] U.S. Patent 2,628,955; DX-777. It issued on February 17, 1953 upon an application dated June 8, 1950.
[80] Brown, Frederick and Parrish's reactions all require low reaction temperatures "usually below room temperature (i. e. 25°C.) and ordinarily between -10°C. and -160°C., and more preferably between -75°C. and -125°C." PTX-38, Cols. 3-4. Compare PTX-39, Col. 3; DX-777, Col. 4.
[81] PTX-38, Col. 1; DX-662, Col. 1.
[82] Acid chlorides may be depicted as follows: (where R signifies a hydrocarbon group):
PTX-38, Col. 3.
[83] PTX-38, Col. 7.
[84] PTX-38, Col. 3.
[85] of Chemistry and Physics C-276 (48th Ed. 1967).
[86] PTX-38, Col. 1.
[87] Hydrogen halides are compounds having the formula HX wherein H is hydrogen and X is a halogen, i. e. fluorine, chlorine, bromine or iodine. Hackh's Chemical Dictionary 394 and 420-21 (3d Ed. 1944).
[88] PTX-39, Col. 1.
[89] Id., Col. 2.
[90] PTX-39, Cols. 2-3.
[91] Id., Col. 4.
[92] DX-777, Col. 1.
[93] E. g.
wherein R1 and R2 are hydrocarbon radicals. DX-777, Col. 3.
[94] PTX-6, Col. 2.
[95] The parties failed to present evidence relating to the vulcanization properties of Parrish's products except to indicate that they were sulfur vulcanizable.
[96] PTX-6, Col. 3. See text accompanying notes 117-119, infra.
[97] PTX-7A, Fig. 16.
[98] Defendant's Brief, pp. 96-101.
[99] PTX-111, p. 1.
[100] Tr., p. 514.
[101] DX-900; Tr., pp. 4856-58.
[102] Hackh's Chemical Dictionary 631 (4th Ed. 1969); Chemical and Technical Dictionary 867 (2d Ed. 1962); Hackh's Chemical Dictionary 799 (3d Ed. 1944); A. Garrett, W. Lippincott, F. Verhoek, Chemistry 433 (1968). When asked if the DCP forming reaction was "spontaneous", Exxon's Morton thus replied: "Yes. As a matter of fact, it is spontaneous in the chemical sense. It is not catalyzed." Tr., p. 4102.
[103] Tr., p. 4102.
[104] Defendant's Brief, pp. 96-101.
[105] Along the way toward discovering this form of catalysts, Ziegler discovered that what he called "polymerization activators" were useful "for polymerizing ethylene and ethylene mixtures with other unsaturated hydrocarbons". PTX-51; DX-425. Ziegler patented this discovery in the United States on January 11, 1955, based upon a German filing dated four and one-half years previously. Although defendant initially relied on this patent application as prior art, its briefs do not mention it and the Court treats it less fully than the other patents, which defendant's briefs cite and exhaustively discuss. Ziegler's 1955 patent and that of Adamek et al. clearly involve different inventions. Ziegler's 1955 patent does not mention ethylene propylene diene terpolymers, nor does it allude to sulfur vulcanization. Tr., p. 1478. Ziegler's activators, moreover, produce low molecular weight oligomers, i. e. polymers comprising only a few monomer units. Hackh's Chemical Dictionary, 473 (4th Ed. 1969), rather than the high molecular weight polymers involved in this litigation. Tr., pp. 1477-78.
[106] "Neither the temperature nor the pressure at which the monomeric hydrocarbon is contacted with the catalyst is critical, and it is possible to use temperatures as low as -100°C. and as high as + 100°C., or to adopt pressures of one atmosphere above or below atmospheric pressure." PTX-45, p. 14. See also PTX-40, p. 3; PTX-47, p. 2.
[107] Belgian Patent No. 543,292; PTX-45; DX-616. It issued on June 2, 1956, based upon an application filed six months previously.
[108] Belgian Patent 546,150; PTX-40; DX-613. It issued on September 17, 1956, based upon an application filed six months previously.
[109] U.S. Patent No. 2,933,480; PTX-46; DX-4. It issued on April 19, 1960, based on both a continuation-in-part application dated June 17, 1959, and the original application, PTX-47, dated January 25, 1956.
[110] PTX-45, p. 6.
[111] Id., pp. 54 and 56.
[112] Defendant's Brief, p. 81.
[113] See text accompanying notes 99-101, supra.
[114] See text accompanying note 10, supra.
[115] Hackh's Chemical Dictionary 221-22 (3d Ed. 1944).
[116] PTX-7A, Fig. 23.
[117] Tr., pp. 4002-03; Pleading Paper 230, pp. 65-67.
[118] PTX-7, Fig. 14.
[119] Tr., pp. 4003-04; Pleading Paper 230, pp. 66-67.
[120] PTX-40, pp. 8-9. Natta publicized his success widely. In March, 1957, he lectured at an International Synthetic Rubber Symposium, publishing this lecture as an article three months later. G. Natta, New Synthetic Elastomers, Rubbers & Plastics Age (June 1957); PTX-29; DX-376.
[121] PTX-40, p. 2.
[122] PTX-40, p. 4. Some of these dienes are conjugated and insofar as they behave like conjugated dienes, their use occasions the same difficulties as those employed by Horne et al. Later patents to Natta et al. indicate, however, that these dienes probably behave, under the reaction conditions they employ, as if they were non-conjugated. Consequently, the Court groups and discusses Natta et al.'s work with that of Gresham et al. who use non-conjugated dienes. PTX-43, p. 7; PTX-44, Col. 7; Tr., pp. 1438-39 and 1441. Exxon does not rely upon Natta et al.'s later patents as prior art. Tr., pp. 1441-42.
[123] PTX-47, p. 1.
[124] Id., p. 2.
[125] PTX-47, pp. 3-4.
[126] Id., p. 4.
[127] Tr., p. 4007. Hercules argues that "Double [b]onds of [d]iffering [r]eactivity [is] [n]ot the [k]ey. . . . Adamek's norbornadiene is a successful terpolymer even though it had two identical double bonds." Plaintiff's Brief, pp. 87-88. The problem with this argument is that it focuses on norbornadiene as it enters the reactor when its double bonds are identically located within a bridged ring structure containing two double bonds. See note 28, supra. Inside the reactor, however, norbornadiene polymerizes using one of its two double bonds. The effect of this reaction is to change the ring structure so that the remaining double bond is located within a bridged-ring structure containing only one, rather than two, double bonds. This apparently changes the remaining double bond so that it is preserved as a sulfur vulcanization site.
[128] This results from the attractive forces between a double bond and the carbon atoms immediately adjacent to the doubly bonded carbons. When the double bond is internal, there can be as many as four such adjacent carbon atoms, but when a double bond is terminal, this number reduces to two. Consequently, an internal double bond is characterized by greater coherence and stability than a terminal one and, thus, more resistance to any reactions that might alter the status quo. Weininger, Contemporary Organic Chemistry, 174 (1972).
Gresham et al. realized yet another less desirable solution involving the use of dienes containing two terminal double bonds which, though chemically similar, are differentiated by the law of averages governing polymerization reactions. Because these double bonds are terminal, they react more readily and they "tend to be used up in the copolymerization with the monoolefins. . . ." A few of these dienes, however, buck the trend and use only one double bond for polymerization, leaving the other available for sulfur vulcanization. A suitable number of vulcanization sites are incorporated by using "large initial proportions of diolefins." PTX-46, Col. 2.
[129] Tr., p. 6568.
[130] PTX-113, p. 244.
[131] See note 120, supra.
[132] U.S. Patent 2,721,189; DX-617. It issued on October 18, 1955, based upon a patent application dated August 30, 1954.
[133] U.S. Patent 2,799,668; PTX-48; DX-424. It issued on July 16, 1957, based upon an application dated September 19, 1956.
[134] U.S. Patent 2,883,372; PTX-49; DX-423. It issued on April 21, 1959, based upon an application dated March 14, 1957.
[135] Col. 1.
[136] Tr., pp. 1462-65.
[137] U.S. Patent 2,939,845; DX-885. It issued on June 7, 1960, based upon an application dated September 13, 1957.
[138] Tr., pp. 6455-56.
[139] Defendant's Brief pp. 83-96.
[140] Pleading Paper 244, p. 72. See also Pleading Paper 230, pp. 126-27.
[141] Tr., pp. 4012-13.
[142] Australian Patent No. 228,056 (DX-668) issued on May 2, 1960 based upon an application dated November 4, 1957. British Patent No. 863,373 (PTX-50; DX-373), issued on March 22, 1961 based upon an application dated October 11, 1957. Canadian Patent No. 607,108 (DX-946), issued on October 18, 1960 based upon an application dated October 19, 1957. German Patent No. 1,102,411 (DX-945, 945A), issued on March 16, 1961, based upon an application dated November 4, 1957. All of the disclosures are essentially identical.
[143] PTX-50; DX-373, p. 1.
[144] Id., p. 2.
[145] Carboline Co. v. Mobil Oil Corp., 301 F.Supp. 141, 146-47 (N.D.Ill.1969). See also Monarch Marking System Co. v. Dennison Mfg. Co., 92 F.2d 90 (6th Cir. 1937).
[146] DX-959. These entries are typified by their first page, providing:
NB 5626, p. 33.
Copolymerization of the dimer of methylcyclopentadiene with ethylene
Added to a 1 liter flask .50 grams of the dimer of methylcyclopentadiene (freshly distilled from LiA1H4 at [illegible] and 500 cc cyclohexane. The mixture was stirred while passing ethylene to [illegible] the air from the flask and also to [illegible] saturated the mixture with ethylene added .01 mole A1 (i-C4H9)3 and stirred for a few minutes and then .005 mole TiCl4 stirred while passing ethylene
Time Temp. Remarks
3:05 30°C.
3:10 54°
3:20 58
3:45 53 Added .005 mole Al (i-C4H9)3
4:00 53°C.
The polymer was washed with methanol [illegible]
Wt. = 20.3 g.
[illegible] = 7.13
Film Very tough and fairly stiff. By infra-red analyses the copolymer contains 24.4% by weight of the dimer.
DX 959, p. 1.
[147] DX 959, pp. 1-3.
[148] Tr., pp. 1999-2007.
[149] Tr., pp. 6790-91.
[150] PTX-501, p. 300.5405I; PTX-502. The cited materials do not directly supply the above estimates. Instead, they list the total ethylene-propylene rubber production. The Court arrived at the above estimates by first noting that about 10 percent of the total ethylene-propylene rubber production is saturated product containing only ethylene and propylene, PTX-501, p. 300.5405H, and that Du Pont, which controls about forty percent market, uses a straight chain, rather than a bridged-ring diene. PTX-499, pp. 33-34; PTX-500, p. 38. Thus, only about half of the ethylene-propylene rubbers made in America contain a bridged-ring diene. The parties failed to offer any evidence about how much of these products contain DCP vs. MNB or ENB.
[151] Tr., p. 1256.
[152] Id.
[153] See e. g. PTX-302.
[154] Among the techniques they investigated were chlorosulfonation, Tr., pp. 1262-3, PTX-312, peroxide curing, Tr., pp. 1355-56, and beta-radiation curing, PTX-310.
[155] Tr., pp. 1356-57.
[156] Among the candidates they investigated were butadiene, PTX-306, PTX-309, isoprene, PTX-301, and divinyl benzene, PTX-302, PTX-309.
[157] See text accompanying note 109, supra.
[158] Tr., pp. 1291-94; PTX-313, p. 2.
[159] PTX-321, p. 3. During 1958, Du Pont's researchers demonstrated that this inhibition could be overcome by using dienes in which the two double bonds were widely separated. PTX-320. The fact that such dienes could be made only by using the costly Grignard-style synthesis, however, precluded their utilization.
[160] PTX-573.
[161] PTX-560, p. 4; PTX-562, p. 3.
[162] PTX-573.
[163] PTX-561, p. 7; PTX-562, p. 8.
[164] DX-461, Col. 1; DX-407, Col. 1. See also PTX-52.
[165] PTX-333; Tr., p. 1359.
[166] PTX-334, p. 7; See also PTX-333.
[167] PTX-569.
[168] PTX-570.
[169] PTX-339, p. 3.
[170] PTX-127, p. 2; Tr., pp. 917-18.
[171] Tr., p. 907.
[172] Among these dienes were butadiene, isoprene, 2-3-dimethyl butadiene, allo-ocimene, dipentene, myrcene and vinyl cyclohexene. PTX-511, pp. 10-12.
[173] PTX-149, p. 3. In July, 1957, Hercules's Dr. Peter VanWyck thus reported. "We have under test or have completed, sulfur vulcanization of 87 samples of unsaturated co- and terpolymers received from Central Research over the past month. To date, none of these materials offer any appreciable promise."
[174] By then Hercules had tested cyclopentadiene, Tr., pp. 991-92, butadiene, isoprene, biallyl, cyclohexadiene, vinylcyclohexadiene, divinylbenzene, allene, dimethylbutadiene, PTX-158, p. 11, 1,5-hexadiene, trans-1,3-pentadiene, PTX-166, p. 2, allo-ocimene and 11-methyldidecadiene-1, 10, PTX-167, p. 2.
[175] Tr., p. 1019; PTX-160, p. H34026; PTX-167, p. 2.
[176] Tr., pp. 1027-28.
[177] Tr., pp. 1194-96; PTX-280.
[178] Tr., p. 1200; PTX-233, p. 4. Other dienes included isoprene, vinyl cyclohexene, 1,5-hexadiene, allyl chloride, 1-hexene, Tr., p. 1200, and hexyne-1, Tr. pp. 1208-09.
[179] Tr., p. 1217; PTX-284, p. EO3004.
[180] PTX-233, p. 4.
[181] PTX-230, p. 58.
[182] DX-814; Tr., pp. 2819-20.
[183] PTX-278.
[184] DX-814; See also PTX-244.
[185] PTX-255, p. 1.
[186] Hercules claims that Sun Oil paid $50,000.00 down payment toward a license, Tr., pp. 1707-10, and Polymer Corporation of Canada, Tr., pp. 1715-17, and Dutch State Mines of the Netherlands, Tr., pp. 1739-41, each paid $10,000.00. Because these companies later decided not to enter the American EPT market, they were not required to pay any additional royalties. On the other hand, Uniroyal, Inc., B. F. Goodrich Company and Copolymer Rubber & Chemical Corporation, which later commercialized EPT's, have paid Hercules substantial sums for their licenses. Indeed, Uniroyal has paid $1,967,574, Tr., p. 1727, Goodrich has paid about $725,000, Tr., pp. 1738-39, and Copolymer has paid $1,055,656, Tr., pp. 1750-51.
[187] Defendant's Brief, pp. 110-124.
[188] PTX-384; PTX-366; PTX-367.
[189] DX-775; PTX-390; PTX-368; Pleading Paper 236, pp. 78-80.
[190] See 1 Stat. 110; 1 Stat. 321-22; 3 Stat. 119; 16 Stat. 201.
[191] 56 U.S. 62, 14 L.Ed. 601 (1854).
[192] Id., p. 85.
[193] Id., p. 99.
[194] Id., pp. 118-119.
[195] 154 U.S.P.Q. 118 (C.C.P.A.1967).
[196] See Chisum, Patents §§ 11.04 and 11.05 (1980).
[197] 154 U.S.P.Q. 121 (C.C.P.A.1967).
[198] Id., p. 120.
[199] 154 U.S.P.Q. pp. 122-23.
[200] 282 F.2d 357, 48 C.C.P.A. 711 (1960).
[201] Id., pp. 361-62.
[202] 282 F.2d 361, 48 C.C.P.A. 711 (1960).
[203] Id., p. 362.
[204] 336 U.S. 271, 69 S.Ct. 535, 93 L.Ed. 672 (1949).
[205] Graver Tank & Mfg. Co. v. Linde Air Products Co., 75 U.S.P.Q. 231, 237 (N.D.Ind.1947).
[206] Graver Tank & Mfg. Co. v. Linde Air Products Co., 336 U.S. 271, 277, 69 S.Ct. 535, 538, 93 L.Ed. 672 (1949).
[207] Possibly Exxon need adduce no more than a preponderance of evidence on this issue since the examiners initially considered very little evidence regarding the adequacy of Adamek et al.'s description. There is, thus, no record informing the Examiners, among other facts, that Adamek et al.'s specification was so broadly worded as to cover a million or more possible third monomers and that Hercules first used MNB only after Du Pont and Exxon discovered its utility and Du Pont submitted a patent application claiming its usage. There is no need, however, to examine how the Examiners' apparent lack of knowledge affects the presumption of validity normally attaching to issued patents because Exxon's case meets the clear and convincing standard.
[208] The prefix "endo" is ambiguous. On the one hand, it can indicate the presence of a hydrocarbon bridge. Hackh's Chemical Dictionary 305 (3d Ed. 1944). Hercules thus argues that endocyclic implies the use of bridged-ring compounds including MNB. Plaintiff's Brief, pp. 17-20 and 123-25. On the other hand, "endo" can be used to indicate the presence of a substituent at an inner position. PTX-35. Exxon thus argues that endocyclic connotes a compound in which all double bonds are contained within a ring. Defendant's Brief, p. 53. This definition excludes MNB which contains an "exocyclic" double bond. The Court sees no reason to prefer Hercules's definition any less than that of Exxon which party carries the burden of persuasion on this issue. Although the Court therefore accepts Hercules's definition, it notes this dispute as indicative of the lack of precision which characterizes Adamek et al.'s specification.
[209] PTX-113, pp. 3-4. The quoted language is from Dunlop's American application. Slight variations appear in Adamek et al.'s British application. Compare Id., pp. 85-86.
[210] Tr., p. 694. See also Pleading Paper 82K, pp. 1876-77.
[211] Tr., pp. 1644-45.
[212] Id., p. 4776.
[213] Id., p. 5780.
[214] See text accompanying note 4, supra.
[215] This lack of knowledge is explained by the short time-twelve weeks-that passed between Adamek et al.'s earliest successful polymerizations and the submission of their initial British application. The reason for this "screaming rush", Pleading Paper 78, p. 11, was that Dunlop's Edward Murphy intended soon to visit Montecatini's Natta. "[I]f there had been any possible disclosure on [Natta's] part which happened to be similar to the work we were doing, had we not-had we taken out the patent after that meeting, we might have been accused of using some of the information we had received from Natta; hence, it was important to file this patent to make it quite clear that it was in no way connected to the visit to Natta." Pleading Paper 119D, p. 536.
[216] The draft application actually sub-titled thirty-two examples, but it failed to include any examples under the last sub-title. PTX-92.
[217] symbol CH3 with a line drawn to the center of a ring is used to indicate that a -CH3 is attached to one undetermined carbon atom in the ring structure. PTX-7, Fig. 12B.
[218] Adamek et al. apparently specified the use of norbornene only because they did not know the underlying theory of their invention. As late as June 12, 1957, Woodhams thus noted: "It is an important feature of our work that the `bridged ring' compounds do not polymerize in the normal [1,2] manner. [See text accompanying note 10, supra.]. According to Du Pont workers who have done a thorough study on norbornene . . . the polymerization proceeds by a ring opening", DX-529, which may be depicted:
PTX 67. This mechanism would leave double bonds available as sites for sulfur vulcanization. Scientists later learned that Woodhams was wrong and that norbornene polymerizes in the normal fashion, consuming its one double bond in polymerization. The language specifying the use of norbornene was then struck from Adamek et al.'s application. PTX 113, pp. 249-251. Dudley testified that he had reservations about including at least the norbornene examples: "I was told [however] that this was a provisional application and that we had a one-year period to make additions or remove errors. On that basis, I could hardly argue strenuously against its inclusion, so I simply did not." Pleading Paper 78, p. 114.
[219] ">219.
DX-895.
[220] ">220.
DX-895.
[221] p. 3.
[222] p. 3.
[223] DX-303, p. 1.
[224] Id., p. 2.
[225] Id., p. 3.
[226] DX-68.
[227] DX-70, p. 1.
[228] DX-83, p. 2.
[229] Du Pont's work was spurred by the early realization of DCP's limitations. See DX-761. EPT's containing MNB, on the other hand, produced favorable results leading Du Pont to prepare and submit on March 29, 1960 patent applications claiming their invention. DX-407; DX-461. These applications issued on June 11, 1963 to Edward K. Gladding. Exxon's experience was similar. After learning of DCP's limitations, DX-818, Exxon began experimenting with MNB: "That diene gave a polymer that was superior and closer to the [commercially marketable Gresham] polymer than anything we had seen up to date", Tr., pp. 2838-39, and in January, 1962, one of Exxon's researchers reported: "METHYLENE NORBORNENE TERPOLYMERS CONTAINING AS LITTLE AS 2 WEIGHT PERCENT DIOLEFIN GIVE GOOD VULCANIZATE PROPERTIES." DX-821, p. 4.
[230] DX-302.
[231] DX-67, p. 2.
[232] PTX-113, pp. 158-168.
[233] See text accompanying notes 261-399, infra.
[234] Plaintiff's Brief, pp. 119-122.
[235] Plaintiff's Brief, pp. 119-122.
[236] See note 218, supra.
[237] See text accompanying note 254, infra.
[238] Tr., p. 4195.
[239] DX-672.
[240] PTX-710A.
[241] PTX-710B. The definition remained nearly unchanged until 1969 when it achieved permanent status. PTX-710B-PTX-710M.
[242] Tr., pp. 77-79.
[243] Pleading Paper 77E, pp. 634-35. Du Pont's Gladding supported Adamek's above quoted testimony. See also Pleading Paper 244, p. 208.
[244] Tr., p. 4282.
[245] See e. g. PTX-47; DX-461; DX-407; PTX-713A; PTX-39; PTX-38; PTX-718.
[246] Tr., p. 4435.
[247] Defendant's Brief, pp. 141-43.
[248] PTX-92, p. 13.
[249] While it is, of course, possible that Adamek et al.'s statements, if untrue, might be fraudulent, See text accompanying note 260, infra, Exxon does not make this allegation and the Court sees no reason to investigate this question further.
[250] ">250.
PTX-470.
[251] PTX-113, p. 241.
[252] Id., p. 240.
[253] See text accompanying notes 351-363, infra.
[254] PTX-657, p. 2. See also PTX-571, pp. 15-16.
[255] 315 U.S. 759, 62 S.Ct. 865, 86 L.Ed. 1171 (1942).
[256] Id., 761, 62 S.Ct. 866.
[257] Defendant's Brief, pp. 145-46.
[258] 367 F.Supp. 63, 72 (S.D.N.Y.1973).
[259] Chisum, Patents § 1105[2][d].
[260] Standard Oil Co. v. Montedison, S.p.A. et al., 494 F.Supp. 370 (D.Del.1980).
[261] The Court's treatment here may at first seem unnecessarily lengthy. Patent law fraud, however, besides being a defense to a claim of infringement, may also justify taxing the guilty party for the attorneys' fees of the offended party. 35 U.S.C. § 285. The Court understands that Exxon's fees may amount to $1.5 million or more. It wishes, therefore, to explain its holdings with special care.
[262] The evidence is clear that Dafter knew about these patents. Indeed, Dafter admitted reading and considering the possible relevance of the Horne et al. and Natta et al. patents. Tr., pp. 1817-19. Furthermore, as early as September 13, 1961, Hercules's Chief Patent Counsel, E. G. Peterson, cited Dafter and Dr. Eleanor R. Bartholomew to Gresham et al.'s patent and asked them to consider its possible relevance. DX-372.
[263] Defendant's Brief, pp. 183-185.
[264] Tr., pp. 5623-24.
[265] Tr., pp. 1816-19.
[266] See e. g. Standard Oil Co. v. Montedison, S.p.A. et al., 494 F.Supp. 370 (D.Del.1980); In re Multidistrict Litigation, Involving Frost Patent, 398 F.Supp. 1353 (D.Del.1975).
[267] Tr., pp. 2014-19.
[268] Defendant's brief, pp. 202-207, cites a number of cases which, because they do not involve the duty to disclose known prior art references, are irrelevant. Hazel-Atlas Glass Co. v. Hartford Empire Co., 322 U.S. 238, 64 S.Ct. 997, 88 L.Ed. 1250 (1944), and Kingsland v. Dorsey, 338 U.S. 318, 70 S.Ct. 123, 94 L.Ed. 123 (1949), involve the duty to disclose the authorship of favorable articles that the applicants cited to the examiners. Precision Instrument Mfg. Co. v. Automotive Maintenance Machinery Co., 324 U.S. 806, 65 S.Ct. 993, 89 L.Ed. 1381 (1945), involves the duty to disclose perjury. Keystone Driller Co. v. General Excavation Co., 290 U.S. 240, 54 S.Ct. 146, 78 L.Ed. 293 (1933), and Acme Precision Products, Inc. v. American Alloys Corp., 422 F.2d 1395 (8th Cir. 1970), only delimit the duty to disclose a prior use. True Temper Corp. v. C. F. & I. Steel Corp., 601 F.2d 495 (10th Cir. 1979); Norton v. Curtiss, 433 F.2d 779, 57 C.C.P.A. 1384 (1970); and In re Multidistrict Litigation, Involving Frost Patent, 398 F.Supp. 1353 (D.Del.1975), all involved the duty to disclose scientific data regarding prior art inventions that were cited and discussed before the Patent Office. Armour & Co. v. Swift & Co., 466 F.2d 767 (7th Cir. 1972); United States v. Masonite Corporation, 316 U.S. 265, 62 S.Ct. 1070, 86 L.Ed. 1461 (1942), Mercoid Corp. v. Mid-Continent Investment Co., 320 U.S. 661, 64 S.Ct. 268, 88 L.Ed. 376 (1943), and Minnesota Mining & Mfg. Co. v. Norton Co., 280 F.Supp. 674 (N.D.Ohio 1967), do not decide any patent law fraud issues.
[269] 155 F.Supp. 949 (D.Mass.1957).
[270] 155 F.Supp. 952 (D.Mass.1957). See also Admiral Corporation v. Zenith Radio Corporation, 296 F.2d 708 (10th Cir. 1961). As early as August 13, 1963, the Patent Office sought to tighten this rule, proposing that an applicant be required to report prior art "known to him at the time of the application, and believed by the applicant to be significantly pertinent to the claimed invention." PTX-522. The patent bar opposed this change, Tr., pp. 2093-94, however, and it was rejected in favor of a rule urging voluntary citation. PTX-523. In 1977, the Patent Office finally adopted a rule requiring applicants to file a prior art statement with their application. See 35 C.F.R. §§ 1.56(a) and 1.97-1.99. The new disclosure rule applies to future applicants only, however, the behavior of past applicants must be judged against the standards then prevailing. Although several cases find that fraud is constituted by a past applicant's failure to cite known prior art, their ruling appears to be conditioned upon a finding that such references would have been extremely relevant to the examiner. In Triumph Hosiery Mills, Inc. v. Alamance Industries, Inc., 191 F.Supp. 652, 656 (M.D.N.C.1961), the Court found "There is no doubt in my mind that the [involved] claims would not have been allowed over [a prior art reference] if the patentees had performed their duty of making a true presentation of the state of the art at the time of the demonstration." See also Beckman Instruments, Inc. v. Chemtronics, 439 F.2d 1369, 1378 (5th Cir. 1970); Intermountain Research & Engineering Co. v. Hercules, Inc., 171 U.S.P.Q. 577, 609 and 631 (C.D.Cal.1971); Aerodyne Machinery Corp. v. Slick Industrial Co., 169 U.S.P.Q. 150, 160 (D.Minn.1971); Kahn v. Dynamics Corp. of America, 367 F.Supp. 63, 71 (S.D.N.Y.1973).
[271] DX 54, p. 5.
[272] PTX-474.
[273] Tr., p. 2560.
[274] DX-372.
[275] See text accompanying notes 127-129, supra.
[276] DX-54, p. 5.
[277] Tr., pp. 1811-12; PTX-529; PTX-476.
[278] See text accompanying notes 127-129, supra.
[279] See note 120, supra.
[280] PTX-46. On or about August 4, 1967, Gresham et al.'s patent was reclassified in the context of an overall reorganization and placed within Class 260/Subclass 80.78. PTX-517, pp. 1, 3, 5; PTX-518.
[281] PTX-113, p. 261. Hercules also notes searches occurring during March, 1960 and November, 1964. The March, 1960 search, however, would not have turned up Gresham et al.'s patent since it did not issue until April 19, 1960. PTX-46. Moreover, it is unlikely that Gresham et al.'s patent would have been uncovered in the November, 1964 search because that search was only an update search designed primarily, if not exclusively, for reviewing the new references placed within a class since the previous search. Tr., pp. 2196-97. But see Plaintiff's Reply Brief, p. 35.
[282] Uarco, Inc. v. Moore Business Forms, Inc., 440 F.2d 580, 585 (7th Cir. 1971); Canaan Products, Inc. v. Edward Don & Co., 388 F.2d 540, 544 (7th Cir. 1968); Minnesota Mining & Mfg. Co. v. Berwick Industries, Inc., 393 F.Supp. 1230, 1234-35 (M.D.Pa.1975); Harris-Hub Co. v. Lear Siegler, Inc., 179 U.S.P.Q. 469, 475 (N.D.Ill.1973).
[283] Adamek et al.'s American application initially claimed "A copolymer of an aliphatic mono-olefin and an unsaturated bridged-ring hydrocarbon." PTX-113, p. 21. Compare text accompanying note 4, supra.
[284] Tr., p. 2204. Exxon, however, seeks to rely on a later transcript portion wherein Dann testified:
Q. Now, recognizing the five or six differences that I called to your attention, do you believe in your opinion that the Examiner would have been alerted to cite Gresham [et al.]?
A. I really don't have a firm opinion on that. If it had been cited. it would not have surprised me. The fact that it was not cited didn't surprise me.
Tr., p. 2205. Exxon's question, however, is ambiguous. It may refer to the "five or six differences" between: (1) Gresham et al.'s patent and the 1962 version of Adamek et al.'s patent application; (2) Gresham et al.'s patent and the 1964-65 version of Adamek et al.'s patent application in which form their patent ultimately issued; or (3) the 1962 version of Adamek et al.'s patent application and the 1964-65 version of Adamek et al.'s patent application. Dann's testimony supports Exxon's position only if the third interpretation is correct. Yet, in view of Dann's other testimony, it seems more likely that Dann understood Exxon's question as referring to the "five or six differences" between Gresham et al.'s patent and the two versions of Adamek et al.'s application. The fact that Dann was not surprised by the Examiners' failure to cite Gresham et al.'s patent comports well with the idea that Gresham et al.'s patent was irrelevant, so irrelevant that although the Examiners were aware of it, they did not deem it worthy of citation.
[285] PTX-113, pp. 38-39.
[286] See note 109, supra.
[287] PTX-588.
[288] PTX-586.
[289] PTX-598, p. 2.
[290] PTX-113, p. 157.
[291] PTX-516, Tab. C.
[292] PTX-516, Tab. A.
[293] PTX-516, Tab. B.
[294] Id., Tab. 1.
[295] Id., Tab. D.
[296] Id., Tab. F.
[297] Id., Tab. H.
[298] Id., Tab. G.
[299] See text accompanying notes 40-44, supra.
[300] PTX-548, p. 34; Tr., pp. 2069-71.
[301] See text accompanying notes 229-230, supra.
[302] PTX-637, Search Notes.
[303] Defendant's Brief, p. 200.
[304] DX-365.
[305] DX-366.
[306] DX-603.
[307] PTX-113, pp. 192-96.
[308] PTX-113, pp. 174-75.
[309] PTX-119.
[310] Tr., p. 532.
[311] Tr., p. 533.
[312] Defendant's Brief, pp. 209-212.
[313] PTX-113, pp. 240-41.
[314] Exxon thus claims that as a result of ethylene cross-linking, Kresge's polymers, See text accompanying note 315, infra, were overly stiff. Defendant's Brief, pp. 218-19.
[315] Defendant's Brief, p. 221.
[316] PTX-113, p. 194.
[317] Tr., p. 3672. Kresge claimed that, like his, Christman's reactor must have been open-ended. Kresge based his hypothesis upon a material balance concept. Since matter is conserved, the amount of any material entering the reactor must equal the amount that leaves. Himmelblau, Basic Principles and Calculations in Chemical Engineering, 59-124 (2d Ed. 1967). In at least his first experiment, Christman reported admitting more than twice as much reactant to his reactor as were formed into solid polymer. Kresge assumed that since the unused reactants must have gone somewhere, they left in an exit gas stream about which Christman did not know. Tr., pp. 3257-3259. Price, however, pointed out that rather than leaving Christman's reactor as gases, the unaccounted for reactants might have formed into a liquid oligomer, and remained within the reactor. Tr., pp. 6608-6617.
[318] PTX-761; Tr., pp. 6367-6374.
[319] See PTX-117.
[320] Tr., p. 6893.
[321] Kresge did not analyze the [VNB] content of his polymers, preventing similar comparisons for this component. Tr., pp. 3714-15.
[322] Tr., p. 3287.
[323] Id., p. 3288.
[324] PTX-120.
[325] PTX-121.
[326] Defendant's Brief, pp. 213-216.
[327] Tr., pp. 536-39, 561-63, 6631-33.
[328] Tr., pp. 536 and 6632-33.
[329] DX-79, p. 2.
[330] DX-310.
[331] Pleading Paper 237, pp. 1041-47.
[332] DX-310.
[333] Pleading Paper 248A, pp. 677-78.
[334] DX-312.
[335] PTX-113, p. 217.
[336] See text accompanying notes 351-363, infra.
[337] Plaintiff's Brief, p. 157.
[338] PTX-207, p. 8.
[339] PTX-218, p. 3.
[340] See text accompanying note 357, infra.
[341] Defendant's Brief, pp. 224-25.
[342] Pleading Paper 248A, p. 692.
[343] PTX-113, pp. 11-12.
[344] Id., pp. 216-17.
[345] Keim reported only two, rather than three, acidified methanol washes. Keim's undisputed testimony shows that this exception is unimportant. "[W]e were attempting to take more catalyst residues out of it by giving it a third wash, that is all. This is just attempting to protect the product, that is all. Has nothing to do with the properties of the product." Pleading Paper 248A, p. 805.
[346] Pleading Paper 248A, pp. 738-39.
[347] Id., pp. 745-49.
[348] Id., p. 752.
[349] Defendant's Brief, pp. 226-27.
[350] DX-316, pp. 32, 34 and 55.
[351] PTX-113, p. 237.
[352] Defendant's Brief, pp. 233-38.
[353] PTX-651, p. 4.
[354] Id., p. 4.
[355] PTX-207.
[356] See PTX-216, pp. 9-11.
[357] PTX-216.
[358] PTX-668, p. EO4744.
[359] PTX-338, p. 21. Indeed, Du Pont considered refiling its Gladding application "in order to broaden [its] scope to include the use of [vinyl norbornene] which [was reported to be] of increasing commercial interest." PTX-594, p. 3.
[360] DX-455.
[361] DX-456. Whether Spurlin saw this report prior to executing his affidavit is unclear. Pleading Paper 247B, pp. 1406-07. Still, DX-456 serves to demonstrate the truth of Spurlin's assertions.
[362] DX-456, p. H38232.
[363] Id., p. H38233.
[364] Defendant's Brief, pp. 228-29.
[365] PTX-113, pp. 59-60.
[366] Id., p. 59.
[367] Defendant's Brief, p. 229.
[368] PTX-113, p. 189.
[369] Id., pp. 189-191.
[370] Defendant's Brief, p. 229.
[371] PTX-113, p. 203.
[372] Id., p. 204.
[373] PTX-113, p. 244.
[374] Id., p. 244.
[375] Id., p. 243.
[376] Id., p. 243.
[377] See text accompanying notes 162-164, supra.
[378] PTX-113, pp. 188-189.
[379] Defendant's Brief, pp. 231-232.
[380] Defendant's Brief, p. 232.
[381] Id., pp. 231-32.
[382] DX-83; DX-307 and DX-309.
[383] PTX-113, p. 242.
[384] Defendant's Brief, p. 239.
[385] See text accompanying notes 353-363, supra. Keim and Christman's major objection to EPT's containing VNB was that they, like their DCP counterparts, cured slowly and groove cracked. PTX-216, pp. 2, 4. While other dienes appeared more promising, PTX-218, pp. 2 3, "[t]he possibility always remains that we might go back and pick up [VNB] later on when we had more time." Tr., p. 2349.
[386] Defendant's Brief, p. 239.
[387] Id., pp. 240-241.
[388] PTX-6, Col. 2.
[389] See DX-569, Fig. 1.
[390] DX-974.
[391] Defendant's Brief, pp. 240-41.
[392] Id.
[393] PTX-6, Col. 2.
[394] DX-974.
[395] PTX-6, Col. 3.
[396] DX-975.
[397] Pleading Paper 229, p. 262.
[398] 37 C.F.R. § 1.133(b).
[399] Tr., pp. 1894-95. Exxon, however, cites a June 15, 1965 memo from Dafter to Miller suggesting that "it would be a good idea to acknowledge telephone interview." PTX-541. Dafter testified, without contradiction, that he recommenced recordation, not because he believed that Miller requested the Examiner to reconsider a prior decision, but because he mistakenly thought that Rule 133 was broader than it is. Tr., pp. 1895-95.
[400] Tr., p. 5386.
[401] Defendant's Brief, pp. 242-47.
[402] DX-921D, pp. 6-7.
[403] DX-921D, p. 3.
[404] Tr., p. 5687.
[405] Tr., pp. 5684-85.
[406] Id., p. 5687.
[407] Id., p. 5688. Miller first proposed a two year retention period on November 17, 1970, nearly two years before Exxon's 1972 request. Miller's proposal argued, "[S]o far as I am aware, retention of diaries for long periods serves no useful purpose. It can be argued that they might be useful in proving diligence in contested interference situations. This is correct. However, in my experience in this area we have never employed them for this purpose. Moreover, the patent files themselves can be used for this purpose because they contain a log of attorney time spent in connection with each case. I don't believe the remote possibility of using the diaries for interference purposes warrants their retention." PTX-752.
[408] Tr., p. 5688.
[409] Id., p. 5689.
[410] Tr., p. 5400.
[411] Plaintiff's Brief, pp. 23-29.
[412] Graver Tank & Mfg. Co. v. Linde Air Products Co., 339 U.S. 605, 608, 70 S.Ct. 854, 856, 94 L.Ed. 1097 (1950).
[413] See text accompanying note 223, supra.
[414] DX-70, p. 2.
[415] DX-407, Col. 1. See also DX-70, p. 2.
[416] Tr., pp. 2712-13.
[417] Plaintiff's Brief, p. 28.
[418] Tr., pp. 2714-16.
[419] Id., pp. 2712-13.
[420] 479 F.2d 951, 958 (2d Cir. 1973).