Full opinion text
OPINION LATCHUM, Senior District Judge. I. INTRODUCTION This is a patent infringement suit in which the plaintiff, E.I. duPont de Nemours and Company (“DuPont”), has charged the defendants, Phillips Petroleum Company, Phillips Chemical Company, whose name was changed during the pend-ency of this case to Phillips 66 Company, and Phillips Driscopipe, Inc., with the infringement of Claims 1, 2, 5, 10, 12, and 14 of DuPont’s U.S. Patent No. 4,076,698 (“the ’698 patent”). (Docket Item [“D.I.”] 212 at 1.) Since Phillips Chemical Company, now named Phillips 66 Company, and Phillips Driscopipe, Inc., are wholly owned Delaware corporate subsidiaries of Phillips Petroleum Company, all three defendants will be referred to collectively as “Phillips.” (D.I. 212 at 1 and 252.) Phillips’ Answer, Affirmative Defenses, and Counterclaims allege that the ’698 patent is invalid and unenforceable for various reasons and there has been no infringement. (D.1.159.) In addition, Phillips filed a pretrial motion for summary judgment to dismiss DuPont’s suit under the doctrine of collateral estoppel based on a rejection of all the claims of the ’698 patent by a Patent Examiner entered in a reissue-reexamination proceeding in the Patent Office. (D.I. 210.) The Court reserved decision on the summary judgment motion until after trial. The parties stipulated with Court approval that the liability issues of validity, enforceability, and infringement would be bifurcated from the trial of the damage issues. (D.I. 161.) The case was tried on the liability issues to the Court without a jury from July 21, 1986 through August 18, 1986. After carefully considering the sufficiency, weight, and credibility of the testimony of the witnesses, their demeanor on the stand, the documentary evidence admitted at trial, and the post-trial submissions of thé parties, the Court enters the following findings of fact and conclusions of law which are embodied in this opinion as permitted by Rule 52(a), Fed.R.Civ.P. II. THE FACTS A. The Invention In Issue The ’698 patent in suit relates to copolymers of ethylene and higher alpha-ole-fins. The original patent application was filed on March 1, 1956; a continuation-in-part application was filed on January 4, 1957, and the patent issued on February 28, 1978 (PX Í). The invention of the patent is based on DuPont’s discovery that the “impact strength” and the “environmental stress crack resistance” of melt processable ethylene copolymers can be unexpectedly improved by incorporating a higher alpha-olefin comonomer having five or more carbon atoms (Tr. 87-93). The ’698 patent specifically discloses that the alpha-olefin comonomers which are “most outstanding in producing resins with high impact strength and excellent stress crack resistance” are those having “preferably 5 to 18 carbon atoms per molecule” (PX 1, col. 3, 11. 19-23). There are two aspects to the ’698 patent: (1) the composition of the ethylene copolymers, and (2) the superior impact strength and the superior environmental stress crack resistance that can be obtained by using those ethylene copolymers. 1. The Composition of the Ethylene Copolymers Polymers are large molecules made by chemically joining together many small molecules called monomers. Polyethylene is the polymer made by polymerizing ethylene monomer: ETHYLENE — C = C POLYETHYLENE— .. ,-C-C-C-C-CC-C-C- ... (PX 1200; Tr. 36-37) There are two types of polyethylenes. The first, free-radical polyethylene, was developed in the 1930’s (Tr. 1180), and was commercialized after the end of World War II (Tr. 52). Free-radical polyethylene is formed by highly reactive growing molecules containing free, radicals. The polymer molecules formed by this process have both long and short chain branches. The structure can be analogized to a rose bush (Tr. 37-41). The second type of polyethylene is linear polyethylene made by coordination catalyst processes developed in the 1950’s (Tr. 61-62; 1181). Linear polyethylene molecules are long straight chain structures. These molecules do not have the branches that free-radical polyethylene molecules have (Tr. 62). Copolymers are made by polymerizing two monomers (Tr. 66). For example, when ethylene and hexene are copolymerized, an ethylene-hexene copolymer is formed. (PX 1202; Tr. 64-67) Hexene (or hexene-1) is a six carbon “alpha-olefin.” It has a double bond at one end of the molecule. Its structure, and the structures of other typical alpha-olefin co-monomers are as follows: (PX 1203; Tr. 67-68) The compositions described in the ’698 patent are linear copolymers of ethylene and higher alpha-olefin comonomers (PX 1, col. 1, 1. 61 to col. 2,1. 23; Tr. 67-68). The higher alpha-olefin comonomers begin with pentene which has 5 carbon atoms, and include hexene (6 carbon atoms), heptene (7 carbon atoms), octene (8 carbon atoms), and the remaining higher alpha-olefins through octadecene, which has 18 carbon atoms (Tr. 67). DuPont uses octene-1 in its commercial ethylene copolymers and has in some instances used decene (Tr. 1090-91). Phillips uses hexene-1 (PX 1237; Tr. 2275; 3161-62; 3168-69; 3206; 3212-13). Copolymers made with the lower alphaolefins, such as butene (4 carbon atoms) and propylene (3 carbon atoms), are not within the scope of the invention. The ’698 patent specifically discloses that in “producing resins with high impact strength and excellent stress crack resistance ... [p]ropylene is not effective; butene-1 shows only marginal effectiveness in this respect” (PX 1, col. 3, 11. 19-25; Tr. 110-11). The ethylene copolymers of the ’698 patent can be processed in conventional melt processing equipment (Tr. 88). Melt index is a measure of the rate at which the copolymer flows when melted and relates to the melt processability of the material. The patent specifically discloses that the ethylene copolymers of the ’698 patent “have melt indexes in the range of 0.2 to 20” to enable them “to be fabricated by conventional fabricating techniques” (PX 1, col. 4, 11. 12-15). The ethylene-higher alpha-olefin copolymers of the invention are characterized by several parameters so that they can be distinguished by external tests from linear ethylene homopolymers and from linear ethylene copolymers that are rubber-like (Tr. 856, 858-60, 888-89). Those parameters are measurements of comonomer content, density, and percent crystallinity. Comonomer content is a measurement of the presence and amount of comonomer in the ethylene-higher alpha-olefin copolymer which aids in distinguishing the copolymers of the invention from linear ethylene homo-polymers and linear ethylene copolymers that are rubber-like. The ’698 patent discloses that the amount of higher alpha-olefin can vary from a very small amount of comonomer (linear ethylene homopolymer has none) on the order of 1% by weight up to about 20% by weight (above 20% the copolymers become rubber-like) (Tr. 139, 859-60, 888-89). The measurement technique for comonomer content disclosed in the patent is infrared spectroscopy (PX 1, col. 7, 11. 62-64 and Table I; Tr. 363-64, 380-90). Infrared spectroscopy comonomer content measurements as low as 1.4% by weight for a heptene (7 carbon) polymer and 1% by weight for a decene (10 carbon) copolymer (PX 1, Table I [Example 1] and col. 7, 11. 62-64) are reported. When the ’698 patent application was filed in the mid-1950’s, the margin of error for infrared comonomer content measurements was much greater than the degree of accuracy that can be obtained today by much later developed techniques. For example, with a hexene copolymer, the margin of error was at the very least ± 0.6%. Thus, a nominal infrared measurement of 1% by weight for hexene comonomer would have included copolymers with hexene comonomer contents ranging from about 0.4% to about 1.6% by weight (Tr. 3616). Density and degree of crystallinity are also indications of the amount of comonomer, aiding to distinguish the copolymers of the invention from linear ethylene homopolymers and linear ethylene copolymers that are rubber-like. As comonomer is incorporated in the linear polyethylene chain, both the density and degree of crystallinity decrease from the values measured with linear polyethylene homopolymer alone (Tr. 132). The density data that are included in the ’698 patent disclose that small amounts of comonomer can be used. Specifically, the patent discloses that with the materials the DuPont researchers were using, for linear polyethylene homopolymers the densities ranged from 0.945 to 0.960, and that as comonomer was added, the density decreased. Similarly, the patent discloses that the degree of crystallinity likewise decreased as comonomer was added (Tr. 132). 2. Superior Impact Strength And Environmental Stress Crack Resistance Obtained With Higher Alpha-Olefin Copolymers The invention of the ’698 patent was not the discovery that ethylene could be copolymerized with higher alpha-olefins, nor was it the discovery of a new comonomer content range, a new density range, or a new X-ray crystallinity range (Tr. 135, 607-OS). The invention was the discovery that melt processable copolymers of ethylene and higher alpha-olefins having five or more carbon atoms unexpectedly can achieve impact strength and environmental stress crack resistance properties that are superior to those obtained with free-radical polyethylene, with linear polyethylene and with comparable copolymers of ethylene (that is, with similar densities and melt index) and the lower alpha-olefins propylene (3 carbons) and butene (4 carbons) (Tr. 86-93). Impact strength is a measure of a polymer’s ability to sustain a sudden mechanical blow. Impact strength tests include the Izod impact test in which a sample is struck by a weight that falls in a pendulum fashion and the Elmendorf tear strength test in which a film sample is torn by a weighted pendulum which rapidly tears the film (T|r. 88-90; PX 601, which is the 1955 ASTM D 689-44 for Elmendorf Tear Strength). Environmental stress crack resistance is a measure of the length of time that a polymer sample can be subjected to a stress load without cracking. The stress load can be imposed by bending the sample. The environment can be heated air or liquids which cause stress cracking such as detergents. One of the standard environmental stress crack resistance tests is the Bell ESCR test in which samples are notched and bent and placed in the stress cracking agent Igepal (Tr. 90-92, 932-36; PX 135; PX 1256 at 389-93). Another stress test is the hoop stress test for plastic pipe (Tr. 92-93). The ’698 patent discloses that superior impact and environmental stress crack resistance properties can be obtained in ethylene copolymers with alpha-olefins containing five or more carbons up to eighteen carbons. Based on prior experience on the effect of branching in free radical polyethylene, this result was totally unexpected and unpredictable prior to DuPont’s discovery in 1955 (Tr. 99-103). When an alpha-olefin comonomer molecule is incorporated in an ethylene copolymer, a side chain or branch is formed (Tr. 66). This can be seen graphically in the structure diagram of an ethylene-hexene copolymer {supra at 1349). In the mid-1950’s, it was known on the basis of prior work with free-radical polyethylene that decreasing the number of branch points would cause an impairment of both stress crack resistance and impact toughness (PX 1, col. 1,11. 51-60; Tr. 94-96). In other words, to improve the properties of stress crack resistance and impact toughness, one would increase the number of branch points. On the basis of this experience, one skilled in the art would have expected lower alphaolefins to be more effective than higher alpha-olefins in improving these properties. When lower and higher alpha-olefins are separately copolymerized with ethylene in equal amounts by weight, the lower alpha-olefin comonomers being of lower molecular weight will give more branch points in the copolymer (Tr. 102). Dr. John Beasley explained this by comparing copolymers made with butene, which has four carbons, and octene, which has eight. At equal weights, twice as many branch points will be obtained using butene than will be obtained with octene. Thus, to one skilled in the art in the 1950’s, better stress crack resistance and impact strength would have been expected with the butene copolymer (Tr. 101-09). Unexpectedly, DuPont discovered the opposite to be true. The higher alpha-olefin copolymers achieved these superior properties (Tr. 101, 103, 109). Based on this discovery, DuPont filed the applications for the ’698 patent in 1956 and 1957 from which the ’698 patent issued. The specifications of the ’698 patent clearly disclose the existence of unexpected superior properties of the invention of the ’698 patent and they read in pertinent part: a. A copolymer which can be extruded into film ... surprisingly, with an impact toughness higher than any film forming polyethylene resin ever tested. b. A copolymer which can be extruded into wire coatings ... which, surprisingly, have stress crack resistance at least equal to any other wire coating polyethylene resin ever tested. c. A copolymer which can be extruded into pipe ... which, surprisingly, has a longer time to failure when subjected to internal pressure both at room temperature and at elevated temperatures than does any other pipe forming polyethylene resin ever tested. d. A copolymer which can be injection molded into various articles ... which, surprisingly have good impact toughness and stress crack resistance at least equal to any injection moldable resin ever tested (PX 1, column 2, lines 24-47). * * * * * * Those which are most outstanding in producing resins with high impact strength and excellent stress crack resistance while at the same time being relatively high in crystallinity contain more than 4 carbon atoms and preferably 5 to 18 carbon atoms per molecule. Propylene is not effective; butene-1 shows only marginal effectiveness in this respect. (PX 1, column 3, lines 19-25.) Phillips argues that it is improper to ascertain the invention of a patent from the specifications outside of the patent claims. But this is not entirely so. The Supreme Court held in United States v. Adams, 383 U.S. 39, 40, 86 S.Ct. 708, 709, 15 L.Ed.2d 572 (1966), that “it is fundamental that claims are to be construed in the light of the specifications and both are to be read with a view to ascertaining the invention.” Other courts have likewise held that limitations that are essential to an invention are to be read into the claims in ascertaining the invention. Roberts Dairy Co. v. United States, 530 F.2d 1342, 1352-53, 208 Ct.Cl. 830 (1976); Stanley Works v. McKinney Mfg. Co., 520 F.Supp. 1101, 1110 (D.Del.1981). In this case the superior impact strength and environmental stress crack resistance which are essential elements of the invention are distinctly disclosed in the specifications as the surprising aspect of the copolymers involved and they must be read with the claims in ascertaining the invention. III. COLLATERAL ESTOPPEL DEFENSE As noted above, just prior to the commencement of trial in this Court, Phillips moved for summary judgment to dismiss DuPont’s action under the doctrine of collateral estoppel. (D.I. 210.) Consideration of that motion was deferred until after trial. Phillips’ motion must now be decided. Phillips points out that the Patent Office conducted a merged reissue and reexamination proceeding of DuPont’s ’698 patent over an extended period of time. (D.I. 211A at A3-9.) In a final office action, Edward J. Smith, Patent Examiner, on May 12, 1986, rejected all the claims of the ’698 patent. (Id. at 102-119.) On June 11, 1986, DuPont appealed the Examiner’s final office action to the Board of Patent Appeals and Interferences. (Id. at 133-34.) On August 6, 1986, the Assistant Commissioner of Patents entered an Order-staying all further proceedings in the DuPont reissue/reexamination pending in the Patent Office. (D.I. 257, App. C.) Phillips strenuously argues that the Examiner’s Final Office Action of May 12, 1986, rejecting all the claims of the ’698 patent, is entitled to preclusive effect against DuPont in this Court which it contends is a collateral proceeding. The Court finds Phillips’ contention to be without legal merit and therefore will deny Phillips’ summary judgment motion for the following reasons. First, Phillips’ argument is premised on the assumption that a patent examiner's rejection of the ’698 patent claim is a final decision of an administrative agency. The Court concludes that this assumption is incorrect. A patent examiner’s rejection of claims in a reissue/reexamination proceeding is not a final decision by the Patent Office. Any final Patent Office decision in the reissue/reexamination matter is a question that would have to be determined by the Board of Patent Appeals and Interferences. MPEP §§ 1442.02 and 2286. (D.I. 257, App. C.) But since the reissue/reexamination proceedings in the Patent Office were stayed (id., App. C), the Board of Patent Appeals and Interferences has never considered the Examiner’s rejection of the claims, has never made a final determination, and never will, because the MPEP provides that the decision of this Court on validity and enforceability of the ’698 patent will be considered controlling in the Patent Office. See MPEP § 2286 (id., App. C). Phillips has cited no precedent, and the Court has found none on its own, which holds that an Examiner’s final office action in a reissue/reexamination proceeding is a final decision of the Patent Office which should be given collateral estoppel effect by this Court. Second, Phillips’ effort to bring an Examiner’s final office action under the Supreme Court’s decision on collateral estoppel in Blonder-Tongue v. University Founda tion, 402 U.S. 313, 91 S.Ct. 1434, 28 L.Ed.2d 788 (1971), is equally unavailing. Phillips argues that in Blonder-Tongue, the Supreme Court has established that the doctrine of collateral estoppel prevents a patentee from relitigating the validity of claims of a declared invalid patent in a “prior proceeding.” But Phillips ignores the fact that the only “prior proceeding” covered by the collateral estoppel rule in Blonder-Tongue was a “trial” in which there has been a “judicial resolution of the same issue.” In Blonder-Tongue, there had been a prior decision of patent invalidity by a United States District Court after trial. Thus, the Supreme Court held in that situation, that collateral estoppel could be pleaded and given effect in subsequent litigation if the patent owner had been given a full and fair opportunity to pursue his claim in the prior trial “procedurally, substantively and evidentially.” Blonder-Tongue, 402 U.S. at 333, 91 S.Ct. at 1445. Apart from the fact that the Examiner’s rejection was not a final decision of the Patent Office, it would be absurd to hold that an Examiner’s final office action before a decision was reached by the Board of Patent Appeals and Interferences in the stayed reissue/reexamination proceedings was preclusive in this Court. The proceedings to the point reached before the Examiner afforded no evidentiary hearing, no live testimony in order to determine credibility, no right of cross-examination, and no application of the rules of evidence. Certainly the absence of these features before the Examiner cannot be said to fulfill the Blonder-Tongue requirements of a full and fair opportunity “procedurally, substantively and evidentially.” In contrast to the mere paper record considered by the Examiner, this Court conducted a full trial on the issues of infringement, validity, and enforceability of the ’698 patent between July 21, 1986 to August 18, 1986. The parties called a total of 29 fact and expert witnesses to testify and introduced 999 exhibits into evidence. The trial transcript covers more than 1900 pages of direct testimony and over 1400 pages of cross-examination. (D.I. 554 A-U.) Trial was not merely limited to a comparison of the ’698 patent and prior art references. Extensive and detailed evidence was also received on DuPont’s unexpected and surprising discovery that superior impact strength and stress crack resistance could be obtained with ethylene-higher alpha-olefin copolymers, on Phillips’ failure to make that discovery when its researchers also worked in the same field in the mid-1950’s, on Phillips’ later recognition of DuPont’s discovery and the commercial success of that invention. The Court, therefore, holds that the Examiner’s final office action rejecting all claims of the ’698 patent in the now terminated reissue/reexamination proceedings is not entitled to a preclusive effect in this Court based on the doctrine of collateral estoppel, and Phillips’ summary judgment motion will be denied. IV. VALIDITY A. Burden of Proof The parties disagree as to which has the burden of proof on the issue of patent validity. Phillips contends that its “burden under [35 U.S.C.] § 282 is met simply by showing prima facie invalidity based on the final office action in the reissue proceeding” (D.I. 259 at 33), or as it otherwise argued, DuPont has the burden of overcoming the deference that this Court must give to the Examiner’s rejection of the claims of the ’698 patent in the reissue/reexamination proceedings. The Court finds Phillips’ argument to be without merit. 35 U.S.C. § 282 provides, in pertinent part: A patent shall be presumed valid____ The burden of establishing invalidity of a patent or any claim thereof shall rest on the party asserting such invalidity. The party asserting invalidity must prove invalidity with facts supported by clear and convincing evidence. Loctite Corp. v. Ultraseal, Ltd., 781 F.2d 861, 872 (Fed.Cir.1985). That burden of persuasion is “permanently on the party asserting invalidity,” Richdel, Inc. v. Sunspool Corp., 714 F.2d 1573, 1579 (Fed.Cir.1983), is “constant and never changes,” American Hoist & Derrick Co. v. Sowa & Sons, 725 F.2d 1350, 1361 (Fed.Cir.1984), and is “never annihilated, destroyed, or even weakened, regardless of what facts are of record,” ACS Hosp. Systems Inc. v. Montefiore Hosp., 732 F.2d 1572, 1574-5 (Fed.Cir.1984); Stratoflex, Inc. v. Aeroquip Corp., 713 F.2d 1530 (Fed.Cir.1983). These holdings of the Federal Circuit are overwhelming and the presumption of validity created by § 282 is applicable to this case in this Court. The Federal Circuit has also noted in Fromson v. Advance Offset Plate, Inc., 755 F.2d 1549, 1555 (Fed.Cir.1985): The Examiner’s decision, on an original or reissue application, is never binding on a court. It is, however, evidence the court must consider in determining whether the party asserting invalidity has met its statutory burden by clear and convincing evidence. In short, Phillips bears the burden of persuading this Court that the ’698 patent is invalid by clear and convincing evidence. The fact that the reissue/reexamination examiner rejected the claim is only one piece of the total evidence presented at trial which this Court must consider in determining whether Phillips has met its burden of proving invalidity. That piece of evidence, however, is not binding upon this Court and it does not weaken the presumption of validity or reduce Phillips’ burden of proving invalidity by clear and convincing evidence. B. Phillips’ Anticipation Defenses Phillips contends that the claims of the ’698 patent are invalid as anticipated under 35 U.S.C. § 102 by the prior work of Phillips and by other prior art. 1. Phillips’ Work In The Mid-1950’s DuPont concedes that in 1954-55, Donald Witt and Gerald Leatherman, researchers for Phillips, made ethylene copolymers using both lower alpha-olefins (propylene and butene) and higher alpha-olefins (pentene and hexene) (Tr. 1279) and that they were disclosed in a patent application which they filed in August 1956. (PX 88.) Phillips contends that this earlier work of Phillips’ researchers, which was not abandoned, suppressed, or concealed, invalidates the claims of the '698 patent under 35 U.S.C. § 102(g). The Court is unable to agree because Phillips’ researchers, unlike DuPont, failed to discover that superior impact strength or environmental stress crack resistance can be achieved with higher alpha-olefin copolymers. Indeed, the Izod impact strength of the butene copolymers made by Witt and Leatherman was superior to their pentene and hexene copolymers. (PX 88, Table VI; Tr. 168-73; 2059; 1472). These disclosures are just the opposite of DuPont’s discovery. Phillips also argued that by the fall of 1955, “Phillips’ researchers also determined” that the Witt and Leatherman pentene and hexene copolymers “exhibited excellent toughness” (D.I. 259 at 13), citing to DX 858 and the testimony (Tr. 1935-38) of Professor Bryce Maxwell, one of Phillips’ experts. Phillips’ contention is based on stress-strain multiplication analysis which Professor Maxwell admitted on cross-examination cannot be used to determine impact strength. (Tr. 2123.) Phillips has also throughout its post-trial briefs created the impression that its researchers concluded in 1955 that the Witt and Leatherman pentene and hexene copolymers had environmental stress crack resistance superior to that of lower alpha-olefin propylene and butene copolymers. The evidence at trial demonstrated that Phillips’ researchers never arrived at that conclusion or made that discovery. It is true that the Witt and Leatherman application contains specific stress crack data for propylene and butene copolymers, but no such data is given for pentene and hexene copolymers. (PX 88, Table VI.) The application contains no hints whatsoever of the possibility that pentene and hexene copolymers could be superior to propylene and butene copolymers in stress crack resistance. Indeed, Mr. Leatherman admitted on cross-examination that he and Mr. Witt did not discover that higher alpha-olefin copolymers had impact strength or stress crack resistance superior to that of the lower alpha-olefin (propylene and butene) copolymers. (Tr. 2541-44.) Their contemporaneous actions in 1954 and 1955 confirm that Witt and Leatherman prepared detailed research reports summarizing their work on propylene and butene copolymers, but they did not prepare any reports on copolymers made with higher alpha-olefins pentene or hexene. (Tr. 2547-48; 2570.) Also after making the pentene and hexene copolymers, they abandoned them and returned to using propylene and butene copolymers. (Tr. 2536; 2569; DX 204, 208A, 215, 227, 230, 234.) Phillips belatedly produced on March 5, 1986, a few months before trial, the notebook of John N. Scott, a Phillips researcher. (DX 206.) The stress crack data contained in that notebook show that butene and pentene copolymers achieved comparable stress crack resistance, while the hexene copolymers had not reached that level. (Tr. 188.) Mr. Scott admitted that his re-suits showed no difference between butene, pentene or hexene (Tr. 2755) and Mr. J. Paul Hogan, a principal witness for Phillips, testified that there was no way of telling which would have the highest stress crack resistance. (Tr. 2385-86.) But even if the Scott data could be said, as Phillips contends, to show any improvement of stress crack resistance, that data was suppressed and concealed for more than 30 years and reliance upon that data, as uncertain as it is, cannot be relied upon to support Phillips’ prior invention defense. In short, Phillips has not proved by clear and convincing evidence that the pentene and hexene copolymers made by Witt and Leatherman were superior to propylene and butene copolymers in impact strength or stress crack resistance. The Court further finds that Phillips has not borne its burden of proving that Witt and Leather-man’s pentene and hexene copolymers inherently had stress crack resistance superi- or to that of the lower alpha-olefins propylene and butene. However, even if Phillips had established at trial that the Witt and Leatherman pentene or hexene copolymers inherently had such superior impact strength or stress crack resistance, its prior invention defense would still be unavailing because it is clear that these superior properties were neither recognized nor appreciated by Phillips prior to DuPont’s discovery of those superior properties in the higher alpha-olefin copolymers. To establish prior invention, there must be evidence that the alleged prior inventors appreciated at the time of their work all the elements of the invention. Standard Oil Company v. Montedison, 494 F.Supp. 370 (D.Del.1980), aff'd, Standard Oil Company v. Montedison, 664 F.2d 356 (3d Cir.1981); see also Kimberly-Clark Corp. v. Johnson & Johnson, 745 F.2d 1437, 1444 (Fed.Cir.1984). Thus, Phillips not only failed to prove that the Witt and Leatherman pentene and hexene copolymers had superior impact strength and stress crack resistance over propylene and butene, but it also failed to prove by contemporaneous evidence that it recognized that these superior properties could be obtained with higher alpha-olefins. As noted above, the only stress crack resistance data for the Witt and Leatherman pentene and hexene copolymers appear in one document in Mr. John Scott’s notebook number 6606 (DX 206) which Phillips did not produce, although called for, until March 5, 1986. (PX 1364.) Phillips advances the argument that the production of this notebook negates DuPont’s contention of abandonment, suppression or concealment. The Court finds otherwise. In the first place, the Scott data is not conclusive that the pentene and hexene copolymers of Witt and Leatherman were superior to the lower alpha-olefins. But even if it did make that showing, disclosing this data after 30 years is not evidence negating the argument of abandonment, suppression or concealment. DuPont’s contention that Phillips abandoned, suppressed or concealed any data relating to the stress crack resistance of Witt and Leatherman’s pentene and hexene copolymers is a valid one. There was nothing in the talks of Dr. William Reynolds, of Phillips Research Division, presented at various conferences or in papers presented at American Chemical Society meetings which disclosed the superior stress crack resistance of pentene, hexene, or other higher alpha-olefins. (Tr. 2690-93.) Likewise, the reference in the Witt and Leatherman patent application (PX 88) relied upon by Phillips to negate abandonment, suppression and concealment proves nothing with regard to stress crack resistance of copolymers with higher alpha-ole-fins. For example, the reference in the Witt and Leatherman application to the “high degree of flexibility” (PX 88 at L000007) continues on to disclose that the “copolymers have very high environmental stress crack resistance value greater than 420 hours.” This, however, was the stress crack resistance obtained by Witt and Leatherman with propylene and butene co-polymers. (PX 88 at Table VI, L000016.) Phillips’ reliance on references in the Witt and Leatherman application to copolymers as being useful for pipe, tubing or film because of their “stress cracking properties” (PX 88 at L000010) does not anticipate DuPont’s discovery. This is so because Mr. Leatherman testified at trial that such general references to stress cracking in the application were based on stress crack data for propylene and butene copolymers. (Tr. 2545-46.) Furthermore, Mr. Leatherman testified at his deposition that he did not recall any pipes, tubing or film being made with the copolymers he and Mr. Witt prepared. (PX 1402 at 72-73.) It is also significant that the actions Phillips took in regard to ethylene alpha-olefin copolymers in the latter part of the 1950’s and early 1960 were consistent with Phillips’ failure to discover, or even to recognize, that superior properties can be achieved with higher alpha-olefin copolymers. The first ethylene alpha-olefin copolymers Phillips introduced commercially in 1958 were made with butene. (Tr. 2268.) Almost ten years passed before Phillips in 1967 eventually introduced a higher alpha-olefin hexene copolymer. (Tr. 2275; D.I. 137 at 15; DX 2580.) In the early 1960’s, Phillips continued to rely on butene and considered using propylene. When a project was undertaken in 1960 to develop a “copolymer which will compete with high pressure polyethylene in film applications,” Mr. Hogan listed the comonomers to be studied as follows: “I. Comonomers A. 1-Butene B. Propylene C. 1-Butene-Propylene Mixtures.” (PX 121; Tr. 159.) Phillips did not introduce its first ethylene higher alpha-olefin copolymer to the market until 1967, five years after DuPont introduced its higher alpha-olefin copolymers to the market and more than ten years after DuPont filed for the ’698 patent. That Phillips’ copolymer was made with hexene. (Tr. 2275; D.I. 137 at 15.) The manner in which Phillips entered the higher alpha-olefin copolymer business supports the significance of the invention of the ’698 patent. The evidence also strongly supports the inference that Phillips initiated its program to switch from butene to hexene only after it learned of DuPont’s discovery that the higher alpha-olefin copolymers have superior properties. On June 4, 1963, the DuPont Canadian patent (the counterpart to the ’698 patent in suit) issued. (PX 4.) Phillips was using four carbon butene as a comonomer at that time. (Tr. 2268-75; D.I. 137 at 14-15.) The DuPont Canadian patent (PX 4) specifically disclosed that superior impact toughness and environmental stress crack resistance can be achieved by using higher alpha-olefins above butene (i.e., pentene and up). (Tr. 205-06.) On July 9, 1963, a month after the issuance of the Canadian patent, Phillips held a Marlex Task Force Meeting. (PX 125.) The environmental stress crack resistance of Phillips’ ethylene-butene copolymers was considered and the Task Force reported that “ESC improvements resulting from incorporation of larger chain ole-fins are being studied for PF bottle grade resins.” (PX 125 at 4.) On July 12, 1963, three days after the Task Force meeting, Phillips made an ethylene-hexene copolymer which was then immediately tested for environmental stress crack resistance. (PX 136 at L95078; Tr. 2425.) The stress crack resistance time of greater than 1000 hours for this sample, which Phillips obtained sometime after August 7, 1963 (Tr. 2425), was the earliest Phillips data Mr. Hogan could point to in support of the statement in an August 1964 memorandum (DX 405; PX 126) that “we have found that ethylene/hexene-1 PF co-polymers in the processable melt index range possess higher ESCR than ethylene/butene copolymers.” (Tr. 2420-24.) In January 1964, Phillips issued a Research Division Progress Report entitled “Improved Methods for Preparation of Marlex Polymers,” and according to a later report, dated May 24, 1965, the research studies discussed in that January 1964 report “showed that improved ESCR values could be obtained by utilizing hexene-1 instead of butene-1 as the process comonomer.” (PX 34 at L08134-35; Tr. 2447-48.) Although Phillips “searched and searched” for the January 1964 report (Tr. 2450), it was never found and produced. Mr. Hogan, who prepared a portion of the missing January 1964 report and who received a copy (Tr. 2451), had no recollection whether the missing report referred to DuPont’s Canadian patent (the counterpart to the ’698 patent in suit). (Tr. 2451-52.) Because of the timing, a strong inference persists that Phillips turned to a closer examination of copolymers with higher alpha-olefins only after the superior properties were disclosed in 1963 in DuPont’s Canadian patent. Phillips’ work with hexene and other higher alpha-olefins continued after the January 1964 report. In November 1965, Mr. Hogan and Mr. A.G. Kitchen issued a report which summarized work done with several higher alpha-olefins after June 1963 when DuPont’s Canadian patent issued. (PX 83; Tr. 2452-54.) Hogan and Kitchen stated at the outset of the November 1965 report that: Bench reactor studies have shown that the environmental stress crack resistance (ESCR) of processable PF polymers is greatly increased by the use of alpha olefins other than butene-1 as comonomers. Hexene-1, which introduces butyl branches to the polymer chain, was the preferred comonomer. (PX 83 at 1, L04404.) In the “Results And Discussion” section of the report, Messrs. Hogan and Kitchen stated, “it is quite apparent that hexene-1 incorporation caused a large increase in ESCR, compared to butene-1 incorporation.” (PX 83 at 3, L04407.) This result is graphically expressed in Figure 2 of that report which shows the significantly better stress crack resistance Phillips obtained with hexene compared to butene. (PX 83, Figure 2, reproduced as chart PX 1215.) Mr. Kitchen at his deposition testified that he was surprised by the large increase in stress crack resistance achieved with hexene in place of butene and he also explained that similar results were obtained with octene, decene and dodecene. (D.I. 95 at 47.) Based on all this evidence the Court concludes that Phillips has not proved by clear and convincing evidence that the research efforts or patent application of Witt and Leatherman in the mid-1950’s anticipated DuPont’s invention of the ’698 patent so as to invalidate that patent under 35 U.S.C. § 102(g). 2. Phillips’ Other Anticipation References Phillips has cited a handful of other references in support of its anticipation defense, viz: Vandenberg U.S. Patent 3,058,963 (DX 100); Vandenberg U.S. Patent 3,015,690 (DX 101); Belgian Patent 533,362 (DX 107A); Nowlin et al. U.S. Patent, 3,219,649 (DX 77); Brown U.S. Patent 2,728,752 (DX 14); Hogan et al. U.S. Patents 2,846,425 (DX 44) and 2,825,721 (DX 45). Curiously, Phillips argues that this Court should combine and cull the teachings of its references to build an anticipation. This the Court refuses to do because it is contrary to well established law. The Federal Circuit has held time and again that anticipation is a defense that is established only when a party challenging validity proves that a single prior art reference discloses each and every element of the claimed invention. E.g., Great Northern Corp. v. Davis Core & Pad Co., Inc., 782 F.2d 159, 165 (Fed.Cir.1986); Structural Rubber Products v. Park Rubber, 749 F.2d 707, 715 (Fed.Cir.1984); Studiengesellschaft Kohle v. Dart Industries, 726 F.2d 724, 727 (Fed.Cir.1984); W.L. Gore & Associates, Inc. v. Garlock, Inc., 721 F.2d 1540, 1541 (Fed.Cir.1983). As a matter of fact, none of these other references teach the achievement of superior impact strength or stress crack resistance with ethylene higher alpha-olefin copolymers, which was the discovery and invention of DuPont’s ’698 patent. But even when these other references are considered in combination as suggested by Phillips, they do not support Phillips’ anticipation defense. Vandenberg ’963 and ’690 Patents and Belgian Patent The Vandenberg ’963 patent is directed to a coordination catalyst system with which “[a]ny ethylenically unsaturated hydrocarbon or mixtures thereof may be polymerized.” (DX 100, col. 4, 11. 45-46.) Vandenberg ’963 goes on to state: Exemplary of the ethylenically unsaturated hydrocarbons which may be homopolymerized or mixtures of which may be copolymerized are the linear 1-olefins such as ethylene, propylene, butene-1, hexene-1, heptene-1, octene-1, octadecene-1, dodecene-1, etc., and branched chain 1-olefins and other olefins such as isobutylene, cis-butene, diisobutylene, tert-butylethylene, 4- and 5-methylhep-tenes-1, tetramethylethylene, and substituted derivatives thereof such as styrene, a-methylstyrene, vinylcyclohexane, diolefins such as hexadiene-1,4, 6-methylhep-tadiene-1,5 and conjugated diolefins such as butadiene, isoprene, pentadiene-1,3, cyclic olefins such as cyclopentadiene, cyclohexene, 4-vinyl-cyclohexene-l, terpenes such as B-pinene, etc. (DX 100, col. 4, 11. 63 to col. 5, 1. 1.) This paragraph merely offers, as Professor Charles Price, one of Phillips’ experts, concluded, a “[l]ot of possibilities.” (Tr. 1540.) The Vandenberg ’963 patent contains sixty-four examples, principally of the homo-polymerization of ethylene or propylene, which confirm that Vandenberg’s process results in extremely high molecular weight polymers. In almost every example, Vandenberg’s reported value for reduced specific viscosity shows that the resulting polymer had a zero melt index and was not melt processable. (Tr. 1532.) The only example in Vandenberg which discloses a copolymer made from ethylene and an alpha-olefin is Example 53, in Table V at columns 11 and 12. (Tr. 1306, 1532.) This ethylene-octene-1 copolymer made in a batch polymerization contained 5.8% octene and had a reduced specific viscosity of 9.0. (DX 100, Table V.) This copolymer was of very high molecular weight, on the order of 1,000,000, and would be a zero melt index or no flow copolymer. (Tr. 1532; Tr. 3495-96.) It was not melt processable in the mid-1950’s. (Tr. 87-88.) Vandenberg’s copolymer was not tested for impact strength or stress crack resistance. (Tr. 3474-75.) In short, Vandenberg’s ’963 patent does not disclose a melt processable ethylene-octene-1 copolymer or any other melt processable higher alpha-olefin copolymers. Vandenberg’s ’963 patent also contains no hint of the superior impact strength or stress crack resistance of such copolymers. Mr. Edwin J. Vandenberg, the inventor, admitted at trial that he had not discovered that ethylene higher alpha-olefin copolymers would have superior impact strength and stress crack resistance over butene copolymers at the same melt index and density (Tr. 3497-98; 3501), nor did he predict this discovery in 1955 when he made his ethylene-octene copolymers. (Id.) Indeed, Mr. Vandenberg, a highly skilled polymer chemist (Tr. 3464-65), admitted that even today he had no explanation why the stress crack resistance of the higher alpha-olefin copolymers is superior to the propylene and butene copolymers. (Tr. 3500-00A.) Furthermore, Phillips’ attempt to rely on the Vandenberg ’690 patent (DX 101) as an anticipation, either alone or combined with the Vandenberg ’963 patent, highlights the deficiencies of the ’963 patent. The Vandenberg ’690 patent discloses a process for polymerizing olefins using hydrogen to control the molecular weight and refers to the application for the Vandenberg ’963 patent as describing a process which can be improved by such use of hydrogen. (DX 101, col. 1, 11. 48-55.) The Vandenberg ’690 patent has no specific examples of ethylene alpha-olefin copolymers. Phillips’ argument can be reduced to the following two propositions that (1) one skilled in this art would use hydrogen as taught by the Vandenberg ’690 patent to reduce the molecular weight of the ethylene-octene copolymer (Example 53) of the Vandenberg ’963 patent to a melt processable range, and (2), if Example 53 were modified with such use of hydrogen, that copolymer produced inherently would have impact strength or stress crack resistance superior to that of the lower alpha-olefin propylene and butene copolymers. Phillips, however, did not prove its second proposition of inherency at trial any more than it did with the Witt and Leather-man prior invention defense. Furthermore, the first proposition is incorrect factually. Mr. Vandenberg was highly skilled in this art and was the inventor of both the ’690 and ’963 patents. (Tr. 3462-65.) He knew that hydrogen could be used to reduce the molecular weight of polymers before he made the high molecular weight, no flow ethylene-octene copolymer (Example 53) of his ’963 patent. (Tr. 3493-94.) Yet, even Mr. Vandenberg never thereafter used hydrogen to reduce the molecular weight of that ethylene-octene copolymer. He did something else. After Mr. Vandenberg had conducted his Example 53 experiment (Run 13) using 10 grams of octene to 1.9 grams of ethylene (Tr. 3468-69; PX 11 at 1), he reported that “it is probable that some useful copolymers may be developed” and that “it may be possible to obtain sufficient copolymerization by operating at the very high monomer ratios that can be obtained by using the more reluctant monomer (as isobutylene, octene-1, etc.) as solvent for the polymerization.” (PX 930 at 4; Tr. 3479-80.) Thereafter, Vandenberg attempted two runs using octene to ethylene ratios of about 25 to 1, or about 5 times the ratio of Example 53. Insignificant amounts of product were obtained. (Tr. 3481-86; PX 11 at 6, 10.) Mr. Vandenberg then made another attempt to make a copolymer using a 25 to 1 ratio of octene to ethylene (PX 12, Run 8) and he obtained a polymer containing 31% octene and having a reduced specific viscosity of 10.0, even higher than his Example 53 copolymer. (Tr. 3487-90.) From this Vandenberg concluded (PX 13 at 2) that “copolymerizations go poorly, in general, with this system.” No further work was done by Mr. Vandenberg or others at Hercules, Inc., on copolymers of ethylene with an alpha-olefin with five or more carbon atoms. (Tr. 3491-93.) Finally, the Belgian ’362 patent (DX 107A), which Phillips relies on in combination with Vandenberg ’963 to argue that claims 1,10 and 12 are anticipated (D.I. 259 at 53-54), only discloses ethylene homopolymers. (Tr. 1536-37.) The Vandenberg ’963 patent refers to this Belgian patent only for its description of “a new process of polymerizing ethylene to a high molecular weight polyethylene” (DX 100, col. 1, 11.14-17), not for uses or properties for the polymers of Vandenberg ’963. (Tr. 1537-38.) The Court finds nothing in either Vandenberg patent which discloses that the Vandenberg copolymers would have the properties and uses of the Ziegler homopolymers (the Belgian patent ’362) or that Vandenberg copolymers would meet the numerical values for tear strength in Claim 1 and for hoop stress in Claim 12 of the ’698 patent. The Court concludes that Phillips has not sustained its burden by clear and convincing evidence that the invention of the ’698 patent was anticipated by the two Vandenberg patents or the Belgian patent. The Nowlin et al. ’649 Patent The Nowlin et al. '649 patent (DX 77) also discloses a single ethylene higher alpha-olefin copolymer (ethylene-hexene) that is not melt processable. Dr. Charles Price, a Phillips’ expert witness, admitted that fact and also testified that because this ethylene-hexene copolymer would not be melt processable, it could not be used to form bottles or extruded into pipe as stated in col. 9, lines 46-51 of the Nowlin et al. patent. (Tr. 1551-52.) Example II of Nowlin is the only example of an ethylene higher alpha-olefin copolymer (ethylene-hexene). Ethylene homo-polymers are disclosed in Examples I and III of Nowlin et al. The Example II hexene copolymer, like the Example I no flow, zero melt index ethylene homopolymer, was difficult to grind in a blender. (DX 77, col. 6, 11. 50-74 and col. 7, 11. 68-70.) Although not stated in Nowlin et al., the melt index of the Example II ethylene-hexene copolymer was 0.057. (Tr. 1545-46; PX 1304; PX 1302 at L07685, Run 6868-14.) Nowlin et al., however, knew how to reduce the molecular weight of a polymer. The Example III ethylene homopolymer they made had a melt index of about 6.0. Yet, there was also a severe decrease in falling ball impact strength. (DX 77, col. 8, 11. 57-65.) Phillips again makes the argument based on two propositions that (1) one skilled in this art would reduce the molecular weight of the Example II ethylene-hexene copolymer of Nowlin et al. to the melt processable range, and (2), if Example II were modified to so reduce molecular weight, the copolymer inherently would have impact strength or stress crack resistance superior to that of the lower alpha-olefin propylene and butene copolymers. Phillips, however, failed to prove at trial the second inherency proposition. The facts are also contrary to the premise in its first proposition that one skilled in the art would reduce molecular weight. Dr. Nowlin and his co-inventor Lyons, like Vandenberg, were skilled in this art and knew how to reduce the molecular weight of a polymer. Yet after obtaining the ethylene-hexene copolymer of Example II of their patent, they did not rerun that experiment to reduce the molecular weight of that copolymer. Instead, they suggested doing what Yandenberg did — add more comonomer: In Run 6694-11 and -14 [the 6694 should be 6868] efforts were made to build-in a plasticizer in the form of 1-hexene. Preliminary evaluations listed in Table XI show there to be no marked difference in polyethylene and these co-polymers which also are very stiff and insoluble in hydrocarbons. Larger 1-hexene/ethylene ratios should be investigated in the polymerizations to increase the plasticity and other properties. (PX 1302 at L07672.) Nowlin et al., like Vandenberg, were skilled polymer researchers, yet they all missed the discovery made by DuPont that is the subject of the ’698 patent. The Brown ’752 Patent Standard Oil’s Brown ’752 patent (DX 14) differs from Phillips’ other anticipation references in that it relates to free-radical polymerization, as opposed to coordination polymerization. (Tr. 1778.) Brown specifically relates to the use of tertiary butyl ethylene as a “polymerization modifier” in the free-radical initiated polymerization of ethylene. (DX 14, col. 1, 11. 51-58; col. 2, 11. 12-21.) Brown also discloses that it is not clear what the tertiary butyl ethylene “polymerization modifier” does. Brown explains that: Whether the tertiary butyl ethylene acts by increasing the solubility of polyethylene in ethylene during polymerization, or by modifying the polymerization process, in somewhat the same manner as do higher temperatures or higher pressures, to, for example, increase branching and decrease polymer density, or whether some other effect, such as chain termination, is predominant, is not determined. (DX 14, col. 2, 11. 19-26.) Tertiary butyl ethylene, preferably present in amounts between about 5.0% and 20% by weight based on the ethylene, modifies the polymerization so that more extensible, less dense polymers are produced. The larger amounts of tertiary butyl ethylene within the given range apparently copolymerize with the ethylene. (DX 14, col. 1, 11. 56-61.) Phillips relies on the indication of apparent copolymerization because tertiary butyl ethylene is a higher alpha-olefin. Regardless of what it discloses about tertiary butyl ethylene, Brown does not anticipate the ’698 patent. Dr. George A. Mortimer, who Phillips produced at trial to testify on Brown, conceded that there is absolutely no disclosure in the Brown patent that the Brown process resulted in ethylene copolymers with improved impact strength or stress crack resistance. (Tr. 1777, 1789-93.) Dr. Beasley also testified to the same effect. (Tr. 3649, 3665.) Dr. Mortimer’s “toughness and flexibility” testimony (D.I. 257 at 56-57) does not overcome this fundamental deficiency in the Brown disclosure. Once again there is a complete failure of trial proof by Phillips that any ethylene tertiary butyl ethylene copolymer that might be produced in the examples of Brown would inherently possess impact strength or stress crack resistance superior to the lower alpha-olefin propylene and butene copolymers. Phillips’ reliance on Dr. Mortimer’s testimony that Brown’s apparent copolymers would have densities less than .9300 (D.I. 257 at 57) overlooks the fact that Dr. Mortimer’s anticipation opinions were also based on density data for polyethylene prepared without tertiary butyl ethylene which he attempted to read into the Brown disclosure from Franta patent 2,586,322. (DX 38; Tr. 1758-60, 1786-88.) Dr. Mortimer conceded, however, that Brown says nothing about incorporating data from Franta. (Tr. 1760.) The trial evidence confirmed that there was a sound basis for Brown’s uncertainty as to what the tertiary butyl ethylene actually did. Brown disclosed only how much tertiary butyl ethylene was included in the reaction mixture at the start of each run and said nothing about how much tertiary butyl ethylene was incorporated into any copolymers that might have been formed. (DX 14, col. 3, 11. 24-60; Tr. 1785-86.) The evidence confirms Brown’s disclosure that “some other effect, such as chain termination” may have been “predominant.” (DX 14, col. 2, 11. 19-26.) Viscosity values listed in the Table in column 3 of Brown show that as the percent tertiary butyl ethylene in the reaction mixture increases the molecular weight of the resultant polymers decreases substantially. These data show that tertiary butyl ethylene acted as a “telogen” or “chain transfer agent” which lowered molecular weight. (Tr. 57-60; 3655a.) Dr. Mortimer, who has stated that the “chain transfer reaction is more important in ethylene polymerization than it is in the polymerization of almost all other monomers,” agreed. (Tr. 1772, 1784.) Standard Oil’s internal report describing the work underlying the Brown patent (PX 628; Tr. 3650-51) confirms that it was not at all clear that any tertiary butyl ethylene had copolymerized. The report states that, in spite of considerable effort, Standard Oil’s researchers could not confirm by infrared examination that any tertiary butyl ethylene had actually copolymerized. (PX 628 at A00023; Tr. 3653-54.) The report states (PX 628 at A00023): Infrared examination of the structure of the polymers prepared in the presence of the higher concentrations of t-butylethylene has, as previously noted in the case of lower concentrations, failed to confirm the presence of any t-butyl groups. Other methods have also failed to verify the general belief that fc-butylethylene participated in the formation of the high polymers but the small decrease observed in polymer density with increasing amounts of t-butylethylene employed (Figure 3) is a strong indication that some copolymerization did occur. There is, however, one property of t-butylethylene about which no uncertainty exists, namely, that it acts as a mild chain terminator. Work reported later in connection with nonolefinic modifiers suggests that this property of t-butylethylene may have been an important factor in its modifying action. The Standard Oil report also shows that even though Brown’s polymers made in the presence of tertiary butyl ethylene had improved extensibility or elongation when compared to polymers made under the same conditions in the absence of tertiary butyl ethylene, the same level of elongation was attained by the Standard Oil researchers using other nonolefinic “modifiers” that would not copolymerize. (PX 628 at A00008,111; Tr. 3655, 3665.) There was no evidence that the observed change in extensibility with tertiary butyl ethylene was due to copolymerization. (Tr. 3665.) The Hogan ’721 And ’425 Patents The Hogan ’721 and '425 patents are the least pertinent of Phillips’ anticipation references. Although Dr. Price and Mr. Hogan testified about these patents at trial, Phillips refers to none of that testimony in its briefs. Indeed, Dr. Price did not attempt to rely on the Hogan patents as disclosing copolymers within the scope of the claims of the ’698 patent. Hogan '721 and '425 relate to processes for making ethylene polymers and copolymers. Phillips implies that these references disclose ethylene higher alpha-olefin copolymers by asserting that “each reference discloses that ethylene and other olefins up to octene may be polymerized by the process and that copolymers of these olefins can likewise be prepared. (D.I. 257 at 58.) The portions of the references Phillips cites relate only to homopolymerization of ole-fins including higher alpha-olefins (DX 45, col. 1, 11. 52-55; DX 700A at 2, 11. 4-6; DX 44, col. 3, 11. 3-8), copolymerization of ethylene and lower alpha-olefins (propylene and butene), and copolymerization of ole-fins other than ethylene (DX 45, col. 9, 1. 55 to col. 10, 1. 21; DX 700A at 14, 11. 1-28; DX 44, col. 3, 11. 8-10). Dr. Price conceded that there are no examples of ethylene higher alpha-olefin co-polymers in either of the Hogan patents (Tr. 1671, 1672, 1674), and there is nothing in either patent which teaches that ethylene higher alpha-olefin copolymers have superior impact strength and stress crack resistance. (Tr. 1675.) The only specific examples of copolymers of ethylene shown in Hogan ’721 which are prior art are copolymers of ethylene and propylene. (DX 45, Examples XXI-XXIV.) The specific examples of Hogan ’425 disclose only the homo-polymerization of ethylene. The only disclosure directed to copolymers is a general statement concerning the possibility of preparing ethylene-propylene copolymers. (DX 44, col. 3, 11. 8-10; Tr. 2341-42.) The Court finds and concludes that Phillips has not met its burden of proving by clear and convincing evidence that the ’698 patent is invalid based on its anticipation defense under 35 U.S.C. § 102 by any of the references cited. C. Phillips’ Obviousness Defense Phillips next contends that the claims of the ’698 patent were obvious and therefore the patent is invalid under 35 U.S.C. § 103. The test for obviousness under § 103 is whether the invention as a whole would have been obvious to one of ordinary skill in the art at the time the invention was made. The Supreme court in Graham v. John Deere Co., 383 U.S. 1, 17-18, 86 S.Ct. 684, 693-694, 15 L.Ed.2d 545 (1966), held that the resolution of obviousness must be made after a consideration of (1) the scope and content of the prior art, (2) differences between prior art and the claims at issue, (3) the level of ordinary skill in the pertinent art, and (4) objective evidence of secondary considerations such as commercial success, long felt but unsolved need, failure of others, etc. The Federal Circuit repeatedly has held that “secondary considerations” or “objective indicia of nonobviousness” must always be considered before the determination of obviousness is made. Interconnect Planning Corp. v. Feil, 774 F.2d 1132, 1144 (Fed.Cir.1985); Simmons Fastener Corp. v. Illinois Tool Works, 739 F.2d 1573, 1574-75 (Fed.Cir.1984); Jones v. Hardy, 727 F.2d 1524, 1530-31 (Fed.Cir.1984). Where a party challenging validity relies on a combination of prior art references to establish obviousness under 35 U.S.C. § 103, that party must show that those references contain some teaching which suggests their use in combination. Ashland Oil, Inc. v. Delta Resins & Refractories, 776 F.2d 281, 293 (Fed.Cir.1985); Hybritech Inc. v. Monoclonal Antibodies, Inc., 802 F.2d 1367, 1381 (Fed.Cir.1986). There must be some reason for the combination other than hindsight gleaned from the disclosure of the patent in suit. Interconnect Planning Corp. v. Feil, 774 F.2d 1132, 1143 (Fed.Cir.1985); Hodosh v. Block Drug Co., Inc., 786 F.2d 1136, 1143 n. 5 (Fed.Cir.1986); Studiengesellschaft Kohle mbH v. Dart Industries, 549 F.Supp. 716, 736-37 (D.Del.1982), aff'd, 726 F.2d 724 (Fed.Cir.1984). Phillips argues, on the basis of Kimberly-Clark v. Johnson & Johnson, 745 F.2d 1437 (Fed.Cir.1984), that all work which qualifies as Section 102(g) work constitutes “prior art” in a Section 103 obviousness analysis. (D.I. 259 at 62.) The Kimberly-Clark holding does not reach that far. In Kimberly-Clark, the work by Champaigne, which the Court found to constitute Section 103 prior art, differs from the Phillips work in an important respect: the patentee was aware of the Champaigne work, both being employees of Kimberly-Clark (745 F.2d at 1444-45), while the DuPont researchers here were unaware of the Phillips work when the invention of the ’698 patent was discovered and made. The court in Kimberly-Clark further observed that an invention which is unknown to both the applicant and the art at the time the applicant makes his invention would constitute secret art {id. at 1445-