Opinion
Civ. A. No. 4347.
December 23, 1975.
Vincent A. Theisen and John G. Mulford of Theisen, Lank Mulford, Wilmington, Del., for plaintiff; David E. Varner of Cushman, Darby Cushman, Washington, D.C. of counsel.
Arthur G. Connolly, Sr., and Paul E. Crawford of Connolly, Bove Lodge, Wilmington, Del., for defendants; Carl F. Peters, Elizabethon, Tenn., of counsel.
OPINION
This is an action for infringement of claims 1, 2, and 9 of United States Letter Patent No. 3,412,062. The patent (hereinafter referred to as the '062 patent) was issued November 19, 1968, pursuant to an application filed April 19, 1965 by William Johnson, Leslie Nathan Phillips, and William Watt. National Research and Development Corporation ("NRDC") is the assignee of the patent and a British corporation. Defendant Great Lakes Research Corporation is a wholly owned subsidiary of Defendant Great Lakes Carbon Corporation (hereinafter referred to interchangeably as "Great Lakes"). Both defendants are Delaware corporations and residents of this District.
The scope of the trial upon which this opinion is based encompassed infringement of "any of the claims of the patent in suit." (Docket No. 137, ¶ (a).) At final argument, plaintiff formally abandoned allegations of infringement based upon all but these three claims.
At the time of filing priority was claimed under 35 U.S.C. § 119 as to an earlier filed British patent application dated April 24, 1964.
Jurisdiction is founded on 28 U.S.C. § 1338(a). Venue is properly placed in this District under 28 U.S.C. § 1391 and 1400(b).
Defendants, in their answer, denied infringement, raised a series of affirmative defenses challenging the scope and validity of the patent and its claims, and counterclaimed for appropriate declaratory relief. The case proceeded through discovery and two unsuccessful motions for summary judgment. Thereafter, the Court granted defendants' motion for a separate trial pursuant to Fed.R.Civ.P. 42(b) limited to the issues of infringement and alleged invalidity of the patent in suit for failure to comply with the provisions of 35 U.S.C. § 112. A trial without jury was held; this Opinion constitutes the Court's findings of fact and conclusions of law with respect to the issues presented and tried in accordance with Rule 52, Fed.R.Civ.P.
Docket No. 111.
The patent in suit is entitled "Production of Carbon Fibres and Compositions Containing Said Fibres." According to the patent specification, carbon fibres are members of a class of non-metallic fibres which have application as a high strength and stiffening element in composite materials. The inventions claimed in the patent include a process for producing carbon fibres, the fibres produced thereby, and certain composite material comprising the fibres as elements thereof.
The continuous production process employed by defendant, which is alleged to infringe plaintiff's patent, consists of three primary steps: (1) The beginning raw material fiber (hereinafter referred to as a precursor fiber) is taken from a tow box and passed through a steam bath and stretched for purposes of reducing the denier of the fiber, (2) thereafter the stretched fiber is passed through a forced draft heated air furnace (hereinafter the furnace is referred to as an oxidizer) while under tension, and finally (3) the oxidized fiber is subjected to further heat treatment in a non-oxidizing atmosphere (hereinafter referred to as carbonization) at temperatures significantly above that employed in the oxidizer, after which the final product, now a carbon fiber, is taken up on rolls which are then sold to defendants' customers.
Denier is the weight in grams of 9,000 meters of a given fiber. H. Bennett, Concise Chemical and Technical Dictionary 282 (2d ed. 1962); T. 386.
While on the facts of this case there is close interaction between infringement, indefiniteness, under 35 U.S.C. § 112, and claim interpretation, infringement and invalidity will be treated separately.
Many other issues have been the subject of both the pleadings and discovery. At defendants' request, the Court granted a separate trial on the following issues: "(1) Does the defendants' process for the manufacture of carbon fibres infringe any of the claims of the patent in suit? (2) Is the patent in suit invalid for failure to comply with the provisions of 35 U.S.C. § 112?" (Docket 137, Pre-Trial Order ("PTO").
I INFRINGEMENT Claim 1
The only independent claim of the '062 patent is Claim 1, which sets forth "a method of making carbon fibers" comprising several steps. Claim 1, in full, reads as follows:
"1. A method of making carbon fibers having a Young's modulus parallel to the fiber axis of not less than 16 x 106 pounds per square inch comprising the steps of oxidizing an organic polymer fiber by simultaneously heating the fiber in an oxidizing atmosphere at a temperature of from about 200° C. to 250° C. for a time sufficient to permit substantially complete permeation of oxygen throughout the core of the fiber while the fiber is held under longitudinal tension, said tension being sufficient at least to limit shrinkage of the fibers during heating to not more than about 12% of the length of the fiber, and carbonizing the fiber by heating the oxidized fiber in a non-oxidizing atmosphere to a temperature of up to about at least 1000° C."
In its consideration of the alleged infringement of Claim 1, the Court proceeds in accordance with guidelines established by the Supreme Court:
"In determining whether an accused device or composition infringes a valid patent, resort must be had in the first instance to the words of the claim. If accused matter falls clearly within the claim, infringement is made out and that is the end of it." Graver Tank Mfg. Co., Inc. v. Linde Air Products Co., 339 U.S. 605, 607, 70 S.Ct. 854, 855, 94 L.Ed. 1097 (1950).
Furthermore, a finding of infringement requires that each essential element recited in the claim be present in the accused device.
Claim 1 begins:
"A method of making carbon fibers having a Young's modulus parallel to the fiber axis of not less than 16 x 106 pounds per square inch. . . ."
Young's modulus, designated "E" in scientific literature, is a constant for a given material. It is the ratio of stress to corresponding strain for a material when it is behaving elastically. As such it is a measure of the material's stiffness. See, 14 McGraw-Hill Encyclopedia of Science and Technology 690 (1971).
The symbol "106" represents million; 16 x 106 is 16 million.
The Court finds this element of the Claim reads on the Great Lakes process.
See, e. g., Admitted Fact No. 2, PTO-Docket No. 137, p. 3; Testimony of Great Lakes Research Corporation Vice President and General Manager, Trial Transcript page (hereinafter designed "T.") 1137. Supportive citations paginated to the record are intended to be illustrative and are not to be considered as the only evidence upon which the Court relies.
The Claim continued:
". . . comprising the steps of oxidizing an organic polymer fiber. . . ."
The Great Lakes process oxidizes Orlon Type 42 or 43, a type of organic polymer fiber known as polyacrylonitrile (PAN). This element of the Claim reads on the Great Lake process.
Admitted Facts Nos. 3 and 5 — (PTO), Docket No. 135, pp. 3 and 4.
The Claim continues:
". . . by simultaneously heating the fiber in an oxidizing atmosphere at a temperature of from about 200° to 250° C.. . . ."
"250° C." means 250 degrees on the Centigrade scale.
The Great Lakes plant has two manufacturing lines. In the "pilot" line, the fiber is heated in an oxidizer having a uniform interior temperature of 235° C. The Claim element reads on the pilot line process. In the "production" line, the oxidizer has four successive interior zones of increasing temperatures, the last two of which operate at temperatures from 250° C. to 265° C. or occasionally 275° C. The above element of the Claim reads on the production line process only if the phrase "from about 200° C. to 250° C." can be interpreted to include temperatures as high as 275° C. Plaintiff's expert witness testified that considering the kinetics of the physical and chemical reactions involved, the claimed range of temperature would be interpreted as approximate at the higher level. Defendants, in the post trial briefing and argument, have not seriously contended otherwise.
Admitted Facts No. 2 (PTO), Docket No. 137, p. 3.
Admitted Facts Nos. 5 and 6.
T. 520-521.
The temperature for the atmosphere in the oxidizer sets forth the operative range with minor departures not critical. In light of the evidence, I find the element as written reads on the temperatures found in the oxidizer of the Great Lakes "production" line.
Cf., Ex Parte Johnson, et al., 82 U.S.P.Q. 452 (Patent Office Bd. App. 1949).
Claim 1 continues:
". . . for a time sufficient to permit substantially complete permeation of oxygen throughout the core of the fiber. . . ."
Starting with the phrase "throughout the core of the fiber", the term "core" was defined by plaintiff's skilled in the art experts as:
". . . central region of the fiber, the region around the axis of the fiber, if it happens to be a circular fiber, but. . . ."
* * * * * *
". . . 10 percent of the cross-sectional area." (Morley T. pp. 349-50).
". . . oxygen has spread throughout the fiber, even including the central region of the fiber." (Uhlmann T., p. 623).
"I would consider the core as the complete fiber, everything except the outer surface." (Winer T., p. 181).
Plaintiff urges all of the foregoing definitions of "core" can be synthesized, are clear, and have the meaning attributed thereto by common everyday usage; i. e., the central region or portion of a given object. Plaintiff would have the Court find that "throughout the core of the fiber" means simply throughout the entire cross sectional area of the fiber.
T. 380-81, 625, 677-78, 1355, 1360-63.
Defendants, without benefit of expert testimony by one skilled in the art, would have the Court find that the term "core" refers to a phenomenon peculiar to certain types of PAN fiber, with special emphasis on Courtelle, and, in no event applicable to its precursor PAN Fiber, Orlon. In brief, Courtelle heated in air for two hours at 200° C. develops zones, the outer zone being oxidized, while the inner core is unoxidized PAN. In contrast, when heated in air, defendants' precursor fiber does not develop a zone structure and consequently no "core".
Significantly, the word core was added to the claim at the insistence and suggestion of the patent examiner. Upon direct examination, plaintiff's witness, Brenner, a former commissioner of patents, a patent expert but not skilled in the art of carbon fibers, testified: "Q. What does the word `core' mean to you in this context? A. Well, this goes back again to the Examiner's initial statement. You remember when we looked into the specification and, I guess it was Page 5, it said that if the oxidizing step is too short in duration the fibers are left with a soft core, and upon subsequent high temperature heat treatments, holes are formed in the remaining fibers. So this was in my view the purpose of the amendment." (Brenner T., p. 44).
DX-U Photograph Series 3.
DX-N(1) pp. 7-8; DX-N(2) p. 131 and Fig. 10.
T. 1099-1100, 1181-82.
In primary support of its theory of the meaning of "core", defendants, over the objection of plaintiff, relied upon the English non-party inventors' handwritten notebook of recorded experiments and observations to conclusively establish the derivation of "core" as found in the patent specification and Claim 1. It is concluded that the inventor's notebook and testimony based thereon is inadmissible.
Plaintiff objected to the use of the inventors' notes (DX-M) contending they constitute inadmissible hearsay. Defendants urged their admissibility relying upon the business records exception to the hearsay rule and a theory of adoptive admission. The business records exception argument fails because no showing has been made that it was the regular course of plaintiff's business to make notes within a reasonable time after noted events occurred and that such notes were made in that regular course, as required by the Federal Business Records Act, 28 U.S.C. § 1732(a), and Fed.R.Evidence 803(6). Hagans v. Ellerman Bucknail Steamship Co. Ltd., 318 F.2d 563, 575-76 (3rd Cir. 1963). Moreover, it is not at all clear that documents such as DX-M come within the scope of the term "business records". See, Alpert v. Slatin, 305 F.2d 891, 49 C.C.P.A. 1343 (1962) (§ 1732 not applicable to progress reports of scientific research and tests). Finally, basic elements of the Federal Rules of Evidence 803(6) exception for "Records of regularly conducted activity" are lacking in that there has been no showing by a "custodian or other qualified witness" that the notes were either made or kept in the regular practice of inventors' professional activity or that they were made "at or near the time . . ." of the event or occurrence. Contrary to defendants' apparent contention, the parties' stipulations as to DX-M's authenticity and the fact that the documents are dated do not constitute a proper foundation under any of the tests discussed above.
Defendants' second theory for admitting DX-M asserts that plaintiff adoptively admitted parts of DX-M in its Answers to Interrogatories (DX-O). In essence, defendants are arguing that plaintiff's reference to portions of DX-M in his answers to certain interrogatories constitute in effect an "express" adoption of "another's statement as his own", McCormick, Evidence § 269 (2d ed. 1972), or "(A) his own statement, in either his individual or a representative capacity or (B) a statement of which he has manifested his adoption or belief in its truth." Fed. Rules Evid. 801(d)(2)(A), (B). There is some support for the proposition that answers to interrogatories constitute admissions. See, Ohio Valley Electric Corp. v. General Electric Co., 244 F. Supp. 914, 954 (S.D.N.Y. 1965). However, it is extremely tenuous to argue that any reference by plaintiff to the contents of a DX-M page in the course of responding to an interrogatory constitutes an admission of the page's entire contents for all purposes. In this context, the language of each interrogatory and the wording of the corresponding reply becomes significant since the range of admissible material in DX-M is theoretically circumscribed by the meaning of the questions and responses. For instance, defendants' interrogatory 79 and its subparts (in DX-O), worded in a fashion typical of defendants' other interrogatories, reads: "79. Define the phrase `substantially complete permeation of oxygen throughout the core of the fiber' as that phrase is used in the claims of the '062 patent. a) State the factual and documentary basis for such definition, if any, in work done by or on behalf of the alleged inventors of the '062 patent, and identify all documents which refer, reflect or relate to such work by the '062 inventors. b) State where the quoted phrase is described or defined in the '062 patent specification, if at all and state in detail plaintiff's contentions as to how the specification described or defines this phrase." See, DX-O interrogatories 24, 26-27, 81(a), 84(a), 94(a), 96, 100, 101, 104(a), 109, and answers and supplemental answers thereto.
This Court is not convinced that plaintiff's admission that a particular page in DX-M is a "factual and documentary basis" or that it may "refer, reflect or relate" to work done by the '062 inventors constitutes more than a mere designation of material in response to a directive specifically demanding such referencing. Furthermore, the alternative wording of many of the questions (i. e. "refer, reflect, or relate to") naturally creates ambiguity in the interpretation of the corresponding response since it remains unclear whether any DX-M material designated was referring, or relating, or, perhaps, reflecting. Semantically, only the word "reflect" imparts any adoptive connotations and, given the general wording of the questions and responses, a positive finding of adoption and consequent admission is not possible.
Moreover, even if plaintiffs responsive designations of DX-M material were to constitute admissions, their scope would be narrowly limited by the wording of their respective interrogatories. Thus, a concession that the pages designated in the answer to interrogatory 96, for example, represented admissions as to the detailed description of "the method of testing, analysis, measurement used" in heating Orlon fibers, would say nothing as to whether the plaintiff was thereby manifesting, for example, his belief in the truth of a particular use by the inventors of the term "core" found elsewhere on the referenced page. The response to interrogatory 96, indicating that "[a]ll information is contained in document Nos. 07409-07419 . . ." points to a repository of solicited information — a description — but does not independently embrace the truth of the designated documents' total contents.
Finally, assuming no financial constraints on defendants, they could have utilized the procedures available under Fed.R.Civ.P. 28(b) to take the depositions in England of the non-resident non-party inventors and subsequently introduce the inventors' deposition under Fed.R.Civ.P. 32(a)(3). Alternatively during such depositions, the defendants could have developed a record sufficient to indicate that the inventor's notebook satisfied the requirements of the business records exception to the hearsay rule of 28 U.S.C. § 1732(a). However, the record presently before the Court indicates that such efforts were apparently not undertaken. Fed. Rules of Evidence 803(6).
Secondary support for defendants' theory is the use of "soft core" in the specification of the '062 patent, combined with reference to the same phenomenon by the inventors in subsequent publications describing earlier work wherein they state Courtelle was used, and creation of a distinct core in Courtelle when run on defendants' process.
"In cases where the preliminary low temperature oxidizing step forms part of the process, if this step is of too short duration the fibres are left with a soft core and upon subsequent high temperature heat treatment holes are formed in the resulting fibres." (PX-1, Col. 2, lines 6-10).
DX-N(1) pp. 7, 8, 12 and Fig. 10; DX-N(2) p. 131.
DX-N(1) p. 7.
DX-U Photograph Series 3.
I, as fact-finder, have serious reservations about the correctness of the definition of core as asserted by plaintiff. One of plaintiff's expert witnesses, Donald R. Uhlmann, an M.I.T. professor with impeccable credentials, while adhering to plaintiff's definition of the word "core", conceded that use of the word in Claim 1, neither contributed (T-1389) nor taught anything to one skilled in the art (T-1388). On the other hand, being unable to consider the inventor's notebook and there being nothing in the patent specification, claims, or in evidence which specifically limit "core" as used in the claim, or "soft core" as used in the specification to the Courtelle phenomenon, I am left with a record which is devoid of any direct reference or testimony contradicting the testimony of plaintiff's experts. A trier of fact may reject testimony of a witness where he is not convinced of its merit. However, failure to accept the uncontradicted testimony of plaintiff's three skilled in the art expert witnesses, because of insight gained from unavoidable examination of a document subsequently held inadmissible, would be without the bounds of reason. See, NBO Industries Treadway Companies, Inc., et al. v. Brunswick Corporation, et al., 523 F.2d 262 (3rd Cir. 1975). This is especially true where the defendant must overcome a statutory presumption in favor of the validity of issued patents. 35 U.S.C. § 282; Trio Process Corp. v. L. Goldstein's Sons, Inc., 461 F.2d 66, 70 (3d Cir. 1972); Tilotson Mfg. Co. v. Textron, Inc., Homelite, 337 F.2d 833, 841 (6th Cir. 1964); Weisbrod v. S. Gluck Co., Inc., 185 F. Supp. 238, 240 (E.D.Pa. 1960). The burden of proof is a heavy one requiring clear and convincing evidence. Trio Process Corp. v. L. Goldstein's Sons, Inc., 461 F.2d 66, 70 (3d Cir. 1972); Schmidinger v. Welsh, 383 F.2d 455, 462, n. 10 (3d Cir. 1967); Struthers Scientific International Corp. v. General Foods Corp., 314 F. Supp. 313, 316 (D.Del. 1970). Defendants have not met their burden. Accordingly, I find for the limited purpose of defining the word "core" as used in Claim 1, that "core" means "cross-sectional" area.
See n. 20, supra.
Reopening of the record was considered and ultimately determined to be not feasible because a directive addressed to the scope of the problems would have constituted an unwarranted intrusion into the adversary process. See n. 20, supra, and n. 127, infra.
PX-1, Col. 5, line 31.
Id., Col. 2, line 8.
See discussion infra pp. 7-8 and T. 380-81.
The term "permeation of oxygen" refers to the transport of oxygen by the mechanism of diffusion and perhaps other physico-chemical mechanisms. As used in the Claim, it refers to the penetration of oxygen from the oxidizing atmosphere inward from the surface of the fiber toward its innermost center. Joining the two terms, "permeation of oxygen throughout the core of the fiber" means the penetration of oxygen as described above throughout the cross-section of the fiber. Permeation is distinct from oxidation. The oxygen transport being described by this portion of the claim has both a physical and chemical aspeet. At the temperatures specified in the claim, some chemical reaction of oxygen with the fiber (oxidation) undoubtedly accompanies the permeation. The more time spent in the oxidizing atmosphere at a given temperature the more oxidation will occur until a point of saturation is reached. Likewise, a longer period spent under the conditions prescribed in the patent will yield more permeation of oxygen. But permeation as a general principle of science can be a strictly physical phenomenon involving no chemical reaction. In the context of the patent claim not all of the permeated oxygen will necessarily be chemically combined.
T. 227, 248-49.
T. 271, 380, 508-15.
T. 723-24.
T. 515.
T. 516-17.
T. 253-54, 518, 717-18.
PX-1, Col. 2, lines 20-25.
T. 352.
T. 250, 253, 350-51, 353, 723-24.
"Complete permeation of oxygen", therefore, does not mean complete oxidation; that is, the chemical combination of the fibrous material with the maximum amount of oxygen theoretically possible for a given temperature. Nor does the weight of the evidence support an interpretation that "complete permeation of oxygen" means the transport or diffusion inward of the maximum amount of oxygen which the fiber can absorb, i. e., complete saturation of the fiber with oxygen. Rather, the evidence demonstrates that to those skilled in the art "complete permeation of oxygen throughout the core of the fiber" means permeation of some oxygen throughout the entire cross-sectional area of the fiber. As such, its meaning varies little if at all from the same phrase absent the term "complete."
T. 518, 621-22, 717-18.
To be sure, there is evidence in this case which suggests not only that this is the proper meaning to ascribe to the phrase but also that it is one that was employed in the past by various agents of plaintiff. See, e. g., Defendants' Exhibit (hereinafter DX) J at 3, DX-Q at 250, DX-K. This evidence is weak, however. The Court views it as neither convincing as to this term's meaning to those skilled in the art nor as compelling a finding of indefiniteness by virtue of its conflict with the interpretation supported by the greater weight of the evidence.
T. 271, 380, 508-15.
T. 624-25, 723.
The term "substantially" in this element of the claim is one commonly used in patents to prevent the avoidance of literal infringement by minor changes which do not themselves cause a loss of the benefit of the invention. 1 A. W. Deller, Patent Claims § 305 (2d ed. 1971). Indeed, there is authority for the proposition that its presence should always be implied in every claim, even when not introduced. Musher Foundation, Inc. v. Alba Trading Co., Inc., 150 F.2d 885, 889 (2d Cir. 1945). The Court finds the term "substantially" in Claim 1 fulfills the function described above and carried with it a meaning commensurate to that function.
T. 1387.
PAN fiber has an initial chemical composition consisting of about three percent, by weight, oxygen. Being a part of the chemical makeup of the fiber itself, this oxygen is distributed throughout the cross-section. Thus, for the claim to prescribe heating for a time sufficient to permit substantially complete permeation of oxygen throughout the core of the fiber means for a time sufficient to permit a measurable increase in oxygen content at each location throughout the cross-section of the fiber over oxygen content prior to heat treatment in the oxidizer. The Court finds that any measurable increase in oxygen content, however small, if it occurs through-out the fiber's cross-sectional area, would be included in the meaning of the phrase, hence, the scope of the claim.
T. 721; DX-N(1) at 9.
T. 271, 718, 723-24.
T. 718-19. In thus defining the meaning of this claim element, the Court is sensitive to the distinction between a change in oxygen content at each location throughout the cross-section of a fiber and a change in total oxygen content of a fiber. Total oxygen content could increase, but if the increase occurred only near the fiber surface and not throughout its cross-section, the condition set forth in the claim would not be satisfied. This phenomenon is confirmed by experiments reported by the inventors in which fibers heated for too short a period resulted in a substandard product even though a considerable increase in total oxygen content had occurred. (DX-N(1) at 7-9). Compare also the minimum time calculated by the inventors required to achieve complete permeation of oxygen throughout a particular fiber with a plot showing considerable increase in total oxygen content of the fiber long before the time of complete permeation (DX-N(1) at 8 and Fig. 12). To the same effect is the inventors' comment upon experiments performed by Shindo, an early pioneer in the field of carbon fibers (DX-N(1) at 12; DX-F(1)).
The ion probe mass spectrometer, an instrument not urged by plaintiff as having afforded a test for substantially complete permeation at the time of the '062 application, was nonetheless employed by plaintiff to demonstrate that this element of the claim reads on the Great Lakes process. This device has the capability of measuring oxygen content in a fiber at various points throughout its cross-section. A careful review of all the evidence convinces the Court that, although not widely available at the time of the patent application, the ion probe mass spectrometer is a device capable today of making accurate qualitative comparisons between oxygen content throughout the cross-section of PAN fiber before and after oxidative treatment. further find that the method employed by plaintiff's expert in the conduct of the laboratory measurements of defendants' fibers was reliable and provided adequate safeguards to protect the integrity of the data obtained.
T. 462-66.
T. 697-98, 700, 706-07, 1315-17, 1320-23, 1372-73; PX-31.
T. 700, 707, 1315-17, 1321-22; DX-Z. But see, T. 851-53, 1320-21; DX-KK at 513-15, 527-28, DX-JJ at 529, 535-36.
Examined with the ion probe mass spectrometer were samples of Great Lakes fiber from both manufacturing lines. Samples were taken immediately prior to entry into the oxidizer and upon exit. After testing, scanning electron microscopy was used to correlate data obtained from the ion probe mass spectrometer with physical features of the various samples. The Court finds the data conclusively show that the oxygen content throughout the cross-section of Great Lakes fiber is generally greater in samples taken after oxidation than in samples taken prior to oxidation. Further testing by neutron activation analysis tends to confirm this conclusion by reliably showing a substantial increase in total oxygen content between the orlon precursor fiber and the oxidized samples. The results from neutron activation analysis closely parallel findings made by independent laboratory measurements sponsored by the defendants in 1971.
T. 468-69.
PX-28.
T. 494-504; PX-29.
DX-J. See also T. 1125.
These tests demonstrate significant amounts of additional oxygen penetrate throughout the entire cross-section of defendants' fibers during their oxidation step. As such, defendants' process achieves substantially complete permeation as defined by plaintiff's experts. This conclusion is confirmed by the admission of an officer of Great Lakes Research Corporation.
T. 494. Attempts by defendants to show otherwise from the data obtained fail to raise a serious challenge to this conclusion. See, e. g., T. 1120-27, 1133-34, 1392.
This conclusion is not based on any test for "substantially complete permeation of oxygen", as defined, such as presence or absence of holes in fibers carbonized from the Great Lakes process. Indeed, save for one photomicrograph taken of defendants' fiber found in their sales brochure, there is no evidence in the case on this point (PX-20(a)). See n. 118 and 127. Instead, the meaning of the phrase itself has been used as the main basis upon which this finding is made. The situation would be quite different had it been demonstrated that measurable increases in oxygen content throughout the fiber was inoperative in eliminating holes and/or shortening the required carbonization step, and/or improving final mechanical properties. See n. 127. Further, had DX-M been admissible and defendants' Courtelle phenomenon theory ultimately adopted, the result also would have been different even if the definition of core were unchanged if defendants had been able to demonstrate the Courtelle "inner zone" had a "trace increase" of oxygen, whether or not the same was chemically bound.
T. 1125, 1139.
The Court finds this element of Claim 1 reads upon the Great Lakes Process.
T. 520, 1350.
Claim 1 continues:
". . . while the fiber is held under longitundinal tension, such tension being sufficient at least to limit shrinkage of the fibers during heating to not more than about 12% of the length of the fiber,. . . ."
When PAN fibers are heated, as in defendants' oxidizer, there is a strong tendency for shrinkage to occur, sometimes times by as much as thirty percent of fiber length. This is undesirable and may be counteracted by restraining the fibers during heat treatment. By doing so, measurable tension develops. The Court finds, by virtue of defendants' admissions, that this element of the claim reads on the Great Lakes process.
T. 1310-13.
T. 1312-13.
Admitted Fact No. 5; T. 1158.
Claim 1 concludes:
". . . and carbonizing the fiber by heating the oxidized fiber in a non-oxidizing atmosphere to a temperature of up to about at least 1000° C."
Carbonization is the final step required to achieve carbon fibers under the patent. By heating the oxidized precursor fiber to high temperatures, the nitrogen, hydrogen, and oxygen that were chemical constituents of the precursor, as well as the newly penetrated oxygen from the oxidation step, are driven off, leaving a relatively pure carbon fiber arranged in a structural format which produces the desired mechanical properties. The process must occur in a non-oxidizing atmosphere; otherwise at the prescribed temperatures the fiber would literally burn up.
T. 459.
T. 255-56, 456.
Defendants' process provides for a carbonization step following oxidation. They allege, however, that because their "primary carbonizer" unit has an atmosphere containing 0.7 percent air in their pilot unit and 4.0 percent air in the production unit, that their process does not carbonize in a "non-oxidizing atmosphere" and therefore does not infringe.
T. 458-61; Admitted Fact Nos. 7 and 8.
T. 457-58, 1089; PX-26.
Air is an oxidizing atmosphere containing about 20 percent oxygen. The oxygen content, therefore, amounts to about 0.14 percent in the pilot line and 0.8 percent in the production line. After the "primary carbonizer", which reaches a temperature no higher than 600° C., the fibers in each line of defendants' process pass through a "graphitizer" which further performs the function of carbonization at a temperature from 1100° C. to 2700° C. Each "graphitizer" operates in an atmosphere containing no oxygen whatsoever.
T. 457, 1179.
T. 1180.
Admitted Fact No. 7; T. 457-61.
Id.
The Court adopts plaintiff's definition of "non-oxidizing atmosphere" as being "a surrounding medium or envelope, usually fluid, which has no tendency to oxidize a substance it surrounds." An officer of one of the defendants testified that the purpose of the small amounts of air added to the otherwise inert atmosphere of the "primary carbonizer" is not to oxidize the fiber. Rather, defendants have found that it improves the final product. There is a tendency for some of the material being vaporized from the fiber during carbonization to recondense on its surface in deposits which result in an inferior product. The air oxidizes these vaporized materials, apparently preventing the undesirable deposits from occurring.
DX-O, plaintiff's answer to interrogatory 94(v).
T. 1168-69.
T. 1090.
Plaintiff's expert witness testified that the phrase "non-oxidizing atmosphere" would encompass atmospheres containing trace amounts of oxygen. The trace amount of oxygen is not for the purpose of further oxidizing the fibers themselves nor does it have that effect. The Court finds that this phrase of the claim element reads on the atmosphere in defendants' "primary carbonizer". Subsequently subjecting the fiber to the graphitizer adds further support to the Court's conclusion that defendants' process includes a step of carbonization in a non-oxidizing atmosphere.
T. 256. But see T. 1090.
The duration of defendants' carbonization step is not at issue. Docket No. 156 at 43, 114.
Defendants next contend that the phrase "up to about at least 1000° C." is vague and indefinite. Specifically, they allege the claim fails to make clear whether 1000° C. is the minimum or maximum temperature prescribed. The specification at one point refers to a "carbonizing temperature of at least 1000° C." but is otherwise unenlightening.
PX-1 at Col. 2, line 18. See also "up to about at least 1000° C." at Col. 1, lines 19-20, "carbonizing to about 1000° C." at Col. 2, lines 31-32, "raising the temperature of the furnace to 1000° C." at Col. 2, lines 51-52, "carbonized . . . to 1000° C." at Col. 4, lines 32-33.
Syntactically, "at least 1000° C." means a minimum temperature of 1000° C. "Up to . . . at least 1000° C." emphasizes that carbonization is often, if not always accomplished by a progressive rise in the temperature to which the fibers are subjected. As such "1000° C." is thus read as the minimum value of an upper range of temperatures which would culminate the carbonization step. Addition of the term "about" is not uncommon to claim language and generally is meant to infer "of some tolerance." In this context, "about . . . 1000° C." would suggest the minimum value for the upper range of temperatures need not be precisely 1000° C. but could itself be a range of temperatures on either side of 1000° C. In sum, the phrase appears to mean that fibers should be heated to a temperature of about 1000° C. or somewhat higher. This reading is in close accord with that ascribed to the claim by both legal and technical expert witnesses and is adopted by the Court. I find this phrase of the claim element reads on defendants' process. Therefore, both phrases constituting this claim read on defendants' process.
PX-1 at Col. 2, lines 51-52; Admitted Facts No. 7; PTO-Docket No. 137, pp. 4-5.
See Ex Parte King, 82 U.S.P.Q. 450, 451 (Patent Office Bd.App. 1948); In re Ayers, 69 U.S.P.Q. 109, 112 (C.C.P.A. 1946). See generally 1 A. W. Deller, Patent Claims § 309 (2d ed. 1971).
Admitted Facts Nos. 7 and 8. PTO-Docket No. 137, pp. 4-5.
The only remaining infringement issue is "file wrapper estoppel". Application of the conventional file wrapper estoppel doctrine is inappropriate because plaintiff did not rely upon the doctrine of equivalents, nor can it be said Claim 1 was drawn more narrowly to avoid rejection on the basis of prior art. Trio Process Corporation v. L. Goldstein's Sons, Inc., 461 F.2d 66 (3rd Cir. 1972). In answer, defendants urge adoption of a doctrine known as "file wrapper estoppel by admission". This concept is explained in Duplan Corp. v. Deering Milliken, Inc., 181 U.S.P.Q. 621 (D.S.C. 1974). Basically stated, after a claim has been rejected, if the applicant clarifies the meaning of its claim without rewriting the claim, it will be bound by the limitations expressed in the clarification.
The term "core" was added to distinguish over the prior art. PX-16 at 49-50. But the distinction drawn by this amendment went not to the duration of the inventor's oxidation process compared to the prior art as defendants would have the Court find. Rather, it went to the physical distinction between penetration throughout the cross-section compared to mere surface blackening.
Defendants urge that plaintiff's claim was clarified by reason of the following language in the file wrapper to the extent that the patent examiner could reasonably conclude that prior art was distinguished by plaintiff in terms of the duration of the oxidation step:
"Applicant has found that the duration of the oxidation step to achieve permeation of the case is generally quite long, for example, one day. It is considered quite clear, therefore, that Tsunoda does not inferentially teach what applicant specifically requires in that Tsunoda teaches an oxidation time of preferably less than one hour and in one instance mentions three hours."
PX-16 at 49-50.
Defendants' contention with respect to the effect of the above quoted representation of the patent examiner is possibly meritorious. Nonetheless, this case is inappropriate for application of the "file wrapper estoppel by admission" doctrine even if that doctrine were to be imported into the law of this district. Like classic file wrapper estoppel, the doctrine of file wrapper estoppel by admission has only been applied where the patent owner asserts a charge of direct infringement under the doctrine of equivalents. Plaintiff, in the matter sub judice, has expressly and carefully declined to rely upon or assert the doctrine of equivalents.
The Court expressly declines to rule on whether the doctrine of file wrapper estoppel by admission does or should constitute part of the case law of this district.
Duplan Corp. v. Deering Milliken, Inc., supra, at 631 and 636; Marston v. J. C. Penney Co., 324 F. Supp. 889 (E.D.Va. 1971), aff'd 469 F.2d 694 (4th Cir. 1972); Welch v. General Motors Corp., 330 F. Supp. 80 at 83-84 (E.D.Va. 1970).
Accordingly, I find Claim 1 in its entirety reads on the Great Lakes process. It is held on the record as limited in this opinion that the Great Lakes process infringes on Claim 1 of the '062 patent.
Claims 2 and 9
Claim 2 reads:
"A method according to claim 1 wherein said organic polymer is polyacrylonitrile."
A claim written in dependent form such as this one, must be construed to include all the limitations of the claim incorporated by reference. 35 U.S.C. § 112. The only new element is a specifically named organic polymer. On the basis of Admitted Fact No. 3, the Court concludes Claim 2 reads on the Great Lakes process.
Claim 9 reads:
"Carbon fiber made by the process of claim 1 having a Young's modulus parallel to the fiber axis of not less than 16 X 106 pounds per square inch."
On the basis of admissions by defendants, the Court finds Claim 9 reads on the final products of the Great Lakes process.
Admitted Fact No. 2; T. 1137; PTO-Docket No. 137, p. 3.
II INVALIDITY
Defendants vigorously contend that one portion of one element of Claim 1 is fatally indefinite under 35 U.S.C. § 112 because of: (a) failure to include variables which would effect the length of heat treatment in the oxidizer; (b) undue breadth; (c) failure to disclose best mode for carrying out the invention; (d) failure to distinguish over prior art; and (e) there being no known test in existence at the time of filing the application for the patent to determine whether one skilled in the art following the process of Claim 1 had achieved substantially complete permeation. Each contention will be separately treated.A. Failure to Include Variables
The record in this case makes clear that the duration of heating required to achieve "substantially complete permeation of oxygen throughout the core of the fiber" is dependent upon a number of factors, among which are temperature, oxidizing atmosphere, the kind of fiber used, fiber denier, and the fiber's cross-sectional area and configuration. Defendants assert the claim and specifications of the '062 patent are fatally indefinite for failing to list these variables.
T. 218-21, 247-48, 333, 338, 377, 521-22, 619, 1148.
A claim is to be read and interpreted in light of the specification and will not be considered indefinite if it reasonably apprises those skilled in the art both of the utilization and scope of the invention. Georgia-Pacific Corp. v. United States Plywood Corp., 258 F.2d 124 (2d Cir. 1958).
The specification itself refers at one point to the effect which changing the denier will have on duration of the oxidation step. The evidence, moreover, is ample that the existence of such variables and their effect on the duration required to achieve permeation is knowledge that would have been possessed by anyone skilled in the art at the time of filing of the application. Under 35 U.S.C. § 112, a specification need not teach that which is known and obvious to those skilled in the art. In re Sureau, Kremer, and Dupre, 153 U.S.P.Q. 66 (C.C.P.A. 1967). The patent does not fail for indefiniteness by not referring explicitly to such factors.
PX-1, Col. 2, lines 21-25.
T. 231, 248, 333, 335-37, 341-42, 344, 386, 521-23, 525-27; PX-6; PX-7.
B. Undue Breadth
There is precedent for the proposition that a patent claim so broad as to include elements which are inoperative or ineffective for the purpose of the invention is invalid under 35 U.S.C. § 112. Having defined "substantially complete permeation" to encompass any measurable increase in oxygen content, the issue here would be whether there is evidence to demonstrate that small but measurable increases in oxygen do not result in either the desired mechanical properties or the elimination of deleterious holes in the center of subsequently carbonized fibers. The work of the experimenter Shindo could be interpreted as demonstrating that improved mechanical properties result from even small increases in oxygen. Defendants did not introduce conclusive evidence on the matter, which was an issue in this trial. Defendants have not established that the terms, as defined in Claim 1, are so broad as to be fatal by reason of being of undue breath.
See, e. g., Corona Cord Tire Co. v. Dovan Chemical Corp., 276 U.S. 358, 385, 48 S.Ct. 380, 72 L.Ed. 610 (1928); Hercules Powder Co. v. Rohm Haas Co., 66 F. Supp. 899, 902 (D.Del. 1946). Teleflex, Inc. v. American Chain Cable Co., 273 F. Supp. 573, 579-80 (S.D.N.Y. 1967). See, generally, 1 A. W. Deller, Patent Claims § 192 (2d ed. 1971).
PX-1, Col. 2, lines 6-10.
DX-F(1) at 5, 9, 26. But see DX-N(1) at 12.
Evidence of a sort does exist in the record. One purpose of achieving substantially complete permeation of oxygen, according to expert testimony, is to "stabilize" the fiber for subsequent high temperature carbonization (T. 371, 517, 722). The speed with which the carbonization step can be performed increases with increased "stabilization" (T. 1149-52; PX-12 at 13). The Vice President of Great Lakes Research Corporation testified that after a very short period in defendants' oxidizer orlon fiber is "oxidizing essentially uniformly from the center to the outside" (T. 1181-82). It may be that at this early stage, the fiber is already "stabilized" to some degree so that carbonization could proceed at a somewhat faster rate than would otherwise be possible (T. 1149-52). Apparently the present state of the art is such that no one knows for sure (T. 718-19, 722-23). The Court is not unaware there is a possibility that very early in the Great Lakes process there is permeation, as defined throughout the cross-section of the fiber, without chemical combination with consequent possible loss of stabilization. Following trial, this shortcoming was mentioned to counsel. Docket No. 155, and see n. 127.
Defendants' alleged inoperativeness in their Answer as part of their First Affirmative Defense (Docket No. 5, ¶ 9(e)). Trial was held on all allegations of invalidity arising from 35 U.S.C. § 112.
"A patent claim shall be presumed valid. . . . The burden of establishing invalidity of a patent or any claim thereof shall rest on the party asserting it." 35 U.S.C.A. § 282 (Supp. 1955-1968).
C. Best Mode for Carrying Out the Invention
Defendants assert that the inventors failed to set forth in the '062 patent the best mode contemplated for carrying out their invention. 35 U.S.C. § 112. The best mode requirement is fixed as of the time an inventor executes his patent application; its purpose is to restrain inventors from applying for patents while concealing from the public preferred embodiments which they have in fact conceived. In re Glass, 181 U.S.P.Q. 31, 35 (C.C.P.A. 1974). Defendants' position is that while the only times referred to in the patent as sufficient for substantially complete permeation are 24 hours and 50 hours, the inventors had themselves conducted experiments which established that for some smaller (1 1/2 denier) fibers a period of from four to eight hours was sufficient.Plaintiff points to the absence of any evidence in the record which shows the applicants knew of this information at the time of filing in April, 1965.
PX-1, Col. 1, lines 8-9.
Further, the specification suggests permeation throughout the fiber will vary with the diameter of the fiber. Moreover, the specification also notes that the smallest diameter carbon fibers will have the highest specific strength. Furthermore, while shorter oxidation times may be generally desirable, it does not follow that the shorter duration associated with 1 1/2 denier fiber is the preferred embodiment for manufacture of carbon fibers. Defendants' process itself uses a precursor of larger denier than 1 1/2, even after it is subjected to a stretching process prior to oxidation. It is the customer who determines what size precursor is used, not the duration of the corresponding oxidation step required.
PX-1, Col. 2, lines 20-22.
PX-1, Col. 2, lines 20-26.
T. 1149-50.
T. 1093-94.
T. 1094.
The Court finds the applicants did not fail to set forth the best mode for carrying out their invention, as required by 35 U.S.C. § 112.
D. Prior Art
Defendants also challenge the validity of the patent for failure of its claims to distinguish over prior art. This grounds for attack is normally associated with provisions of the patent statute other than Section 112, i. e., 35 U.S.C. § 102, 103. The Supreme Court, however, has elucidated the connection between prior art and § 112 in two major opinions. Thus, to satisfy § 112, a claim must "clearly distinguish what is claimed from what went before in the art . . . ." United Carbon Co. v. Binney Smith Co., 317 U.S. 228, 236, 63 S.Ct. 165, 170, 87 L.Ed. 232 (1942); accord, General Electric Co. v. Wabash Appliance Corp., 304 U.S. 364, 368-73, 58 S.Ct. 899, 82 L.Ed. 1402 (1938); Pennsylvania Crusher Co. v. Bethlehem Steel Co., 193 F.2d 445, 448 (3d Cir. 1951).
Prior art is "`the existing state of knowledge in a particular art at the time an invention is made. It includes the issued patents . . . , publications, and all other knowledge deemed to be common thereto such as trade skills, trade practices, and the like.' A. Smith, Patent Law, Cases, Comments and Materials 2 (1964)." Trio Process Corp. v. L. Goldstein's Sons, Inc., 461 F.2d 66, 69 n. 3 (3d Cir. 1972).
A United States patent issued to the Japanese inventor Tsunoda is cited as prior art over which the '062 patent fails to distinguish. In Example 1, found in the Tsunoda specification, the inventor describes heating fiber in air for three hours "in order to blacken it completely." Defendants allege to have replicated this example in their photomicrographic studies and found the fibers so treated to be discolored evenly throughout their cross-sections, thus proving Tsunoda achieved substantially complete permeation.
PX-10a, filed April 5, 1963, issued November 15, 1966.
PX-10a, Col. 2, lines 5-9. Plaintiff asserted an error in translation of this portion of the Tsunoda patent and proffered into evidence an alleged correct translation of Example 1. Defendants object to its admission on grounds of relevancy and hearsay. The evidence in its present form will not be admitted as it is hearsay. It is also irrelevant as to all issues under 35 U.S.C. § 112 and as to any argument going to file wrapper estoppel. Decision as to its relevancy on issues emanating from 35 U.S.C. § 102 and 103 is reserved.
T. 743-50, 778-79, 932; DX-U Photograph Series 1 and 2.
The Court is not, at this time, convinced from the evidence presented that the treatment described for defendants' Samples 1 and 2, as depicted in DX-U, is an accurate replication of Tsunoda's Example 1. Further, the Court is not convinced on the present record that the technique of photomicrography by itself is an accurate test for substantially complete permeation. For these reasons, the Court finds the '062 patent adequately distinguishes itself from Tsunoda.
See n. 127 and docket items referenced therein.
Neither this finding nor any other remark, finding or conclusion made by the Court in connection with distinction over prior art for definiteness purposes should be interpreted as prejudging or otherwise relating to any issues within the purview of 35 U.S.C. § 102 and 103. In particular, the Court is making no judgment on novelty or obviousness, neither of which were issues tried to the Court.
In a related issue, Great Lakes asserts its process follows the prior art of Tsunoda and not the patent in suit, therefore qualifying for a finding of non-infringement according to a long line of cases. See, e. g., Scott Paper Co. v. Marcalus Co., 326 U.S. 249, 66 S.Ct. 101, 90 L.Ed. 47 (1945). But mere similarity of oxidizing times, and temperatures is not enough, where, as here, Claim 1 of the '062 patent reads on defendants' process, but not on Tsunoda's.
Compare, T. 1098 with PX-10a, Col. 2, line 10.
Compare, Admitted Fact No. 6 with PX-10a.
Defendants also cite the prior art of Shindo as an example of the achievement of similar mechanical properties in fibers subjected to a similar process. As to the Shindo patent, it appears to teach halogenation followed by carbonization. Of the two most relevant examples in the Shindo specification, Example 1 does not refer to the application of tension and Example 2 refers to tension in the carbonization process only. However, in a published article by Shindo, he states: "filaments . . . were . . . prepared that they might be straight." Fuller development, if ever, of the meaning of this language and its relationship to tension will necessarily have to await trial on the issues under 35 U.S.C. § 102 and 103.
DX-E, filed December 17, 1963, issued February 11, 1969; DX-F(1).
DX-E.
DX-E, Col. 2, line 69 to Col. 4, line 12; T. 888-91.
DX-F(1).
A British patent cited by defendant is likewise distinguishable in several ways from the patent in suit.
DX-G, filed December 16, 1959, published April 18, 1962. The attention here is on treatment or cellulosic material as distinguished from PAN fibers. See T. 856, 885. No explicit mention is made of the application of tension to avoid shrinkage. See T. 885. But see DX-G at 2, lines 4-11, at 4, line 1 through line 17. No explicit mention is made of an oxidation step.
In sum, the Court has been shown no prior art for limited application under § 112 which combines together all the elements of the claim to achieve the stated results of the '062 patent. Whether the combination, per se, of individual elements found in the prior art to replicate the '062 patent was obvious or lacked novelty at the time of filing of the application is an issue going to Section 102 or 103 of the statute, not Section 112.
The Court finds the '062 patent does not fail to clearly state a distinction from the prior art for Section 112 purposes.
E. Test for Substantially Complete Permeation
Defendants argue that there exists no known test or technique for determining whether or not a given process has achieved substantially complete permeation of oxygen throughout the core of the fiber; moreover, they allege that no such test existed at the time of filing the application for the patent in suit (T. 1097). To satisfy the statute, there must have been a test available at the time of the filing of the patent application which could have been employed by a person skilled in the art to which the patent applies. See, Raybestos-Manhattan, Inc. v. Texon, Inc., 268 F.2d 839, 842 (1st Cir. 1959). See also, Allen-Bradley Co. v. Erie Resistor Corp., 104 F.2d 150, 152 (3d Cir. 1939). Furthermore, while some preliminary tests may be required to apprise those skilled in the art of how to practice or avoid practicing the claimed process, a requirement for costly independent experimentation amounting to a separate research effort would not meet the statutory standard. See, Corning Glass Works v. Anchor Hocking Glass Corp., 374 F.2d 473, 478-79 (3d Cir.), cert. denied, 389 U.S. 826, 88 S.Ct. 65, 19 L.Ed.2d 80 (1967); Merck Co. v. Chase Chemical Co., 273 F. Supp. 68, 75-76 (D.N.J. 1967). Testing which is necessary not for determining what the patent discloses but merely to adopt a process to particular materials will not invalidate the patent. Minerals Separations, Ltd. v. Hyde, 242 U.S. 261, 271, 37 S.Ct. 82, 61 L.Ed. 286 (1916); Lever Bros. Co. v. Proctor Gamble Mfg. Co., 139 F.2d 633, 638-39 (4th Cir. 1943); Gray Company v. Spee-Flo Mfg. Corp., 361 F.2d 489, 493 (5th Cir. 1966).
The patent specification itself discloses that if the duration of the oxidative heating step is too short "fibres are left with a soft core and upon subsequent high temperature heat treatment holes are found in the resulting fibres." According to a publication of the inventors, the holes are not observable in carbonized fibers that have been oxidized for a sufficiently long period. The holes are observable by taking longitudinal and transverse cross-sections of fiber samples and inspecting these under a microscope. Evidence was introduced that others have made use of the presence or absence of holes as a test for substantially complete permeation.
PX-1, supra, note 43.
DX-N(1) at 7.
Id. at 8, and Fig. 10.
T. 177-83, 212-14. However, the precursor fiber used was manufactured by Courtaulds, Ltd., the same manufacturer of "Cortel", the precursor fiber defendants urge was used by the inventors of the '062 patent. See DX-N(1) at 7.
Neither plaintiff nor defendants introduced any evidence as to whether defendants' product had holes. Defendants introduced evidence at trial of its own photomicrographic experiments which were said to demonstrate that the test described in the patent specification has no applicability to the orlon precursor fiber which Great Lakes employs in its process. The tests involved taking photomicrographs of transverse cross-sections of Great Lakes fibers after they had passed through the oxidizer but prior to the carbonization step. These photographs were taken to record the presence or absence of a discolored "soft core" in the center of the fibers, as described in the patent specification's test. The technique for observing this "soft core" derives from publications of the inventors.
Plaintiff has attempted to rely on the evidence of a non-credible nature, viz. absence of discernible holes in photomicrographs of Great Lakes' which appear in defendants' sales brochure. See, DX-O, plaintiff's supplemental answer to defendants' interrogatory 69(c); PX-20(a); also see n. 127 infra, and documents cited therein relating to reopening of the record.
PX-1.
DX-N(1) at 8 and Fig. 10; DX-N(2) at 131 and Fig. 12.
Defendants allege that this test describes a phenomenon observable only in the special "Courtelle" precursor fiber allegedly used by the inventors at the time, and that partially oxidized orlon fiber of the type used by Great Lakes exhibits no such property as "soft cores". Defendants buttress their theory by pointing to their photomicrographs which reveal no discolored core in orlon fibers, but a quite distinct core in partially oxidized "Courtelle" samples which were also taken. Defendants deduce from the foregoing that the test is not applicable to their fiber precursor. They then argue, the patent claim cannot be interpreted as reading on the Great Lakes' process. This aspect of defendants' argument is not convincing.
T. 854.
T. 752-800, 854; DX-U Photograph Series 1-2, 5-7.
DX-U Photograph Series 3.
The most appropriate test is that suggested by the patent specification, i. e., observation of finished carbonized fibers for holes, as distinguished from soft cores or a zone structure in fiber heat treated in the oxidizer prior to treatment in the carbonizer.
PX-1, Col. 2, lines 6-10.
Moreover, one ready explanation for the absence of an observable "soft core" in the photographs of orlon fiber after oxidation or at an intermediate stage while in the oxidizer, is that the fiber had already been substantially completely permeated with oxygen throughout its core as taught by the patent in suit. This possibility is more than hypothetical in view of the fact that all five of the photographed orlon samples had been subjected to oxidation for at least three hours, T. 752-800, and that two of the samples were taken at the conclusion of the normal Great Lakes oxidation process, DX-U Photograph Series 5 and 6. Defendants' expert witness further acknowledged as to the photographed sample from the Great Lakes Production line: "[E]ither it is completely permeated with oxygen all over or it never had a core to begin with." T. 765. See also T. 766, 778-79, 932. Finally, if one assumes that orlon has special properties which do promote, during oxidation, an even distribution of oxygen throughout the cross-section as distinguished from a "Courtelle Phenomenon" of oxidation moving radially inward from the surface to the center, one strong inference to be drawn is that the entire cross-section of orlon is equally "soft" or "hard" at any given time; therefore, subsequent carbonization could never produce in orlon the variable shrinkage rates between "hard" and "soft" regions of fiber which are said to account for the presence of holes in the final product. Oxygen having substantially permeated throughout the core of orlon fiber almost from the start of the oxidation process, the possibility of holes in the resulting product had been vitiated. Thus, while defendants would be correct in that there would be no holes, the absence of holes is meaningless because there would be infringement. It is concluded on the record as presented in this case that holes in the carbonized fiber are an adequate test for determining substantially complete permeation as that term has been used and defined by plaintiff for purposes of Claim 1 of the '062 patent.
Testimony was introduced that such samples have been taken in the past, photographs of which fail to reveal any indication of a discolored core. T. 1099-1100, 1181-82.
DX-N(1) at 7.
The issue then becomes not whether there are holes (as a test), but whether, if in using orlon as a precursor fiber, one heated defendants' orlon precursor fiber in an oxidizer until there was a trace increase of oxygen throughout the core of the fiber, would it be sufficiently stabilized to undergo carbonization? Unhappily, the record is silent in terms of hard scientific fact as to when, during the defendants' process, a "trace increase", as defined by plaintiff's experts, occurs. If it occurs prior to sufficient stabilization of carbonization (as might happen if chemical combination of oxygen rather than permeation were the essential operative condition), plaintiff's experts would have had to have been in error in their definition of the term "substantially complete permeation."
These and other problems were made known to the litigants subsequent to trial, and prompted the Court to suggest a reopening of the record while attempting to avoid intrusion into the adversary process. See Docket No. 155. Both parties were willing to reopen to aid their own cause, but otherwise objected or impeded reopening of the record to the point that a court determination to reopen could have conceivably preordained the result. See: n. 20, supra; Docket Nos. 158-164. Cf., Rochez Bros., Inc. v. Rhoades, 527 F.2d 891 (Rochez 3) (3rd Cir. 1975).
The practice of attempting to submit proposed evidentiary exhibits with letters subsequent to trial, briefing and oral argument is not encouraged. Such exhibits, while necessarily read to determine they were evidentially inappropriate for utilization in the decision making process have not been considered in arriving at the conclusions contained in this opinion.
I find three other remaining tests proposed by plaintiff are inadequate: final mechanical properties of the carbon fiber, presence or absence of an exothermic reaction within the heat treated fibers, and use of the electron probe analyses to measure oxygen distribution throughout the fiber cross-section.
The Court does not doubt that a strong correlation exists between the achievement of substantially complete permeation in the oxidation process and higher values of tensile strength and Young's modulus obtained by the subsequently carbonized product (PX-1, Col. 4, lines 35-42; T. 176, 212, 229, 251, 298-300, 306-10, 1281-82). The record is silent, however, on any precise, reliable test by which one could know whether or not his process is infringing by merely referring to the final mechanical properties obtained thereby. Plaintiff argued that given a process employing the elements of Claim 1, substantially complete permeation would have occurred if the final properties were a Young's modulus of at least about 16 X 106 pounds per square inch and a tensile strength of at least 100,000 pounds per square inch, per Claim 1 and Example 7 of the patent. Experiments performed by the inventors themselves belie this easy deduction. Fiber batches oxidized for two hours produced carbon fibers with central holes yet having mechanical properties far above those named by plaintiff's counsel as indicating substantially complete permeation (DX-N(1) at 7, Table 2, and Figs. 4 and 5). See discussion, supra 41.
While the exotherm phenomenon was described in the record, the evidence is insufficient to conclude as a result thereof that the absence of an exothermic reaction always accompanies substantially complete permeation, or vice versa (T. 180, 244-45, 251, 299, 307-08, 517, 1096, 1135, 1149; DX-N(1) at 7).
The evidence is insufficient to conclude that this laboratory tool was developed to the point at which, at the time the '062 patent application was filed, accurate measurements of oxygen could be achieved. See, e. g., T. 112-13; DX-LL at 53, Col. 1, 54, Col. 3, 60, Col. 2. But see DX-LL at 58, Col. 2, DX-Z at 2.
I do not find the '062 patent invalid by reason of its failure to comply with 35 U.S.C. § 112.
Submit Order on Notice within ten days.