Studiengesellschaft Kohle mbH v. Dart Industries

Full title: STUDIENGESELLSCHAFT KOHLE mbH, Plaintiff, v. DART INDUSTRIES, INC.…

Court: United States District Court, D. Delaware

Date published: Oct 5, 1982

549 F. Supp. 716 (D. Del. 1982)

Opinion

Civ. A. No. 3952.

October 5, 1982.

Howard M. Handelman, of Bayard, Brill Handelman, Wilmington, Del., for plaintiff; Arnold Sprung, and Nathaniel D. Kramer, of Sprung, Horn, Kramer Woods, New York City, of counsel.

Arthur G. Connolly, Jr., of Connolly, Bove Lodge, Wilmington, Del., for defendant; Thomas F. Reddy, Jr., Gerald J. Flintoft, Stanton T. Lawrence, III, and Brian M. Poissant, of Pennie Edmonds, New York City, of counsel.

OPINION

CALEB WRIGHT, Senior District Judge

BACKGROUND

This is a suit alleging infringement of United States Patent No. 3,113,115 ('115), issued to Dr. Karl Ziegler of the Max Planck Institute for Coal Research in 1963. The plaintiff, Studiengesellschaft Kohl mbH, (SGK), is the successor in interest to Professor Ziegler. The defendant Dart Industries is the successor corporation to Rexall Drug Chemical Company. This action was initiated in July, 1970. The life of the patent expired before trial. This Court has jurisdiction under 28 U.S.C. § 1338(a).

Briefly, the '115 patent teaches a system for the polymerization of ethylene and other lower olefins, including propylene. The catalyst system consists of alkyl aluminum halides, especially diethyl aluminum chloride, and titanium halides, especially titanium tetrachloride. The interaction of these components produces a catalyst that has proven singularly active, effective and efficient in producing high grade plastic polymers.

SGK contends that Dart's process for the production of commercial grade polypropylene infringes SGK's patent. Dart defends first, that the patent is invalid because anticipated by and obvious from the prior art; second, that its catalyst system is so distinct from the system disclosed by the SGK '115 patent that it does not infringe; and finally, that the equitable doctrines of laches and estoppel render the '115 unenforceable because SGK unreasonably delayed instituting any infringement action, affirmatively misrepresented its enforcement intentions and, in so doing, materially prejudiced Dart.

The '115 patent has been the subject of previous litigation. The United States Court of Appeals for the Fifth Circuit addressed the issue of the validity of the '115 in its 1973 opinion in Ziegler v. Phillips Petroleum Co., 483 F.2d 858, cert. denied, 414 U.S. 1079, 94 S.Ct. 597, 38 L.Ed.2d 485 (1973). The Fifth Circuit upheld the validity of the patent and found it to be infringed by the catalyst system used by Phillips. While this Court is and has been aware of Phillips, this Opinion reflects independent findings of fact and conclusions of law based on all the evidence in the case presented during February and March of this year.

The Opinion addresses the issues of validity, infringement and equitable doctrines in that order. The sheer magnitude of the evidence presented precludes addressing each issue raised. Failure explicitly to mention an issue or argument in this Opinion indicates only that, after due consideration, the Court found the issue or argument to be without merit or of no consequence to the outcome of the case. This Opinion represents the Court's findings of facts and conclusions of law as required by Fed.R.Civ.P. 52.

I. VALIDITY

Dart Industries argues that U.S. Patent 3,113,115 issued to Karl Ziegler, Heinz Breil, Heinz Martin and Erhard Holzkamp on December 3, 1963, is invalid. Dart contends that the claims and content of the '115 violate 35 U.S.C. §§ 102 and 103, arguing that the '115 was anticipated by and obvious from prior art in the field. In support of these arguments, which proved to be the principal part of its case, Dart offered a wide range of evidence, much of it highly technical. Central to Dart's thesis, however, are two strands of evidence: the German Patent No. 874,215¹ granted to Dr. Max Fischer of Badische Anilin Soda Fabrik (BASF); and the experimental work done at the Du Pont Company which resulted in a number of U.S. patents, particularly those to Dr. Arthur W. Anderson.

1 Defendant's Exhibit 30t [hereinafter DX-30t, where "t" indicates the english translation of a document offered into evidence both in its original non-English and English forms]. This patent is referred to as "the Fischer Patent" or "the Fischer work" throughout the trial transcript and this Opinion.

The Fischer work was first patented in 1943, but was not issued or published by the post-war Federal Republic of Germany until 1953. The patent was not known even to members of the scientific community until the later date. Dr. Fischer's laboratory was destroyed during the latter part of World War II, as were his experimental notes and data. Consequently, there is very little contemporaneous information to aid the Court, or any other reader, in interpreting the Fischer patent. The patent itself is very short — only three and one-half typed pages as translated — and contains only one working example.

The Du Pont work is much better documented and has been effectively supplemented and explained by both deposition and live testimony. The Du Pont research bears only on the question of obviousness under § 103, not anticipation under § 102. Ziegler's '115, entitled to priority dates based on foreign applications, pre-dates the relevant Du Pont patents to Anderson et al. A. Anticipation

Dart's anticipation defense is based on its assertion that the Fischer patent meets the standards of 35 U.S.C. § 102(a) and (b). Since the Fischer patent was awarded before even the earliest of Ziegler's German applications, the question of anticipation reduces to whether Fischer actually describes the same invention as that claimed in the '115. The Court heard an enormous amount of testimony about the Fischer work, most of it post hoc experimentation and interpretation of what is actually described in the Fischer patent. As noted, the destruction of contemporaneous records and the brevity of the patent itself made such an approach almost inevitable.

In evaluating a claim of anticipation, the Court will apply a test of strict construction. See Research Corp. v. NASCO Industries, Inc., 501 F.2d 358 (7th Cir.), cert. denied, 419 U.S. 1096, 95 S.Ct. 689, 42 L.Ed.2d 688 (1974); Alco Std. Corp. v. Tenn. Valley Auth., 516 F.Supp. 431 (W.D.Tenn. 1981); 1 Chisum, Patents § 3.02 p. 3-5 (1982) ("The standard for novelty, that is, for "anticipation," is one of strict identity."). To anticipate the '115, the Fischer patent must disclose "the very invention of the ['115] patent"² or "a device substantially identical to that claimed under the terms of the patent."³ The Third Circuit has further elaborated the test for anticipation:

2 Tokyo Shibaura Elec. Co., Ltd. v. Zenith Radio Corp., 404 F.Supp. 547, 558 (D.Del. 1975), aff'd, 548 F.2d 88 (3d Cir. 1977). See General Battery Corp. v. Gould, Inc., 545 F.Supp. 731 (D.Del. 1982); Grefco v. Kewanee Industs., Inc., 499 F.Supp. 844 (D.Del. 1980), aff'd, 671 F.2d 495 (3d Cir.), cert. denied, 454 U.S. 1086, 102 S.Ct. 644, 70 L.Ed.2d 621 (1981).

3 Grefco, supra, 499 F.Supp. at 850, quoting DeLong Corp. v. Raymond Int'l., Inc., 622 F.2d 1135, 1141 (3d Cir. 1980).

For a prior publication to defeat a patent it must exhibit a substantial representation of the invention in such full, clear and exact terms that one skilled in the art may make, construct and practice the invention without having to depend on either the patent or on his own inventive skills.⁴

4 Universal Athletic Sales Co. v. Amer. Gym, Recreational Athletic Equip. Corp., 546 F.2d 530, 544 (3d Cir. 1976), cert. denied, 430 U.S. 984, 97 S.Ct. 1681, 52 L.Ed.2d 378 (1977), quoting Philips Electronic Pharmacautical Industs. Corp. v. Thermal Electronic Industries, Inc., 450 F.2d 1164, 1169 (3d Cir. 1971). See Johnson Johnson v. W.L. Gore Assocs., Inc., 436 F.Supp. 704, 726 (D.Del. 1977) ("Unless all of the same elements are found in exactly the same situation and united in the same way to perform the identical function in a single prior art reference, there is no anticipation.") ( quoting Walker v. Gen'l. Motors Corp., 362 F.2d 56, 58 (9th Cir. 1966)). See also Alco Std. Corp., supra, 516 F.Supp. at 435.

Anticipation, to be effective, must be both clear and contained in a single reference. The publication argued to anticipate must contain "within its four corners, adequate directions for the practice of the patent claim sought to be invalidated." Congoleum Industries, Inc. v. Armstrong Cork Co., 339 F.Supp. 1036, 1052 (E.D.Pa. 1972), aff'd, 510 F.2d 334 (3d Cir.), cert. denied, 421 U.S. 988, 95 S.Ct. 1991, 44 L.Ed.2d 478 (1975). "Anticipation cannot be shown by combining more than one reference to show the elements of the claimed invention."⁵

5 1 Chisum, Patents § 3.02 p. 3-5 (1982). See Grefco, supra, 499 F.Supp. at 850, ("`it must appear that every material element of the claim in question was disclosed by a single prior art reference.'") ( quoting Mueller Brass Co. v. Reading Industs. Inc., 352 F.Supp. 1357, 1367 (E.D.Pa. 1972), aff'd, 487 F.2d 1395 (3d Cir. 1973)).

To invalidate the '115 on grounds of anticipation by Fischer, Dart must show that the Fischer patent clearly, fully and independently demonstrates the essentials of the '115 claims Dart challenges. To make this determination, the Court must examine and compare the '115 and the Fischer patent.

Briefly, Ziegler's '115 patent claims a polymerization catalyst of the general formula RR'A1X, where R and R' are organic radicals, especially alkyl radicals, and X is a halogen, combined with a heavy metal salt, usually a titanium chloride.⁶ The most widely used organo-aluminum component is diethyl aluminum chloride (DEAC or Et₂ A1C1). The '115 teaches the use of these claimed catalyst components for the polymerization of "lower olefins up to about C₅, and particularly ethylene."⁷ SGK, as successor in interest to Karl Ziegler on the '115, has brought this infringement action against Dart on the basis of Dart's use of DEAC and titanium trichloride (TiCl₃) to polymerize propylene (C₃H₆).⁸ SGK claims that propylene, as well as ethylene, polymerization is covered by the '115.

6 See DX-1. The details of the '115 are more fully explored in the Court's section on infringement. See text accompanying notes 75-83, infra.

7 DX-1, Col. 1, lines 53-55.

8 See note 76 infra, for a diagram of propylene.

The Fischer patent⁹ relates to the production of polyethylene. In particular, the language of the Fischer patent claims novelty in the ease with which the invention permits the production of solid or near-solid polyethylene as opposed to the oils predominantly formed by previous methods.

9 DX-30t.

It is known that one can polymerize ethylene or ethylene-rich gases to form lubricating oil-like products with the aid of aluminum chloride under pressure and at elevated temperature. . . .

It has now been found that the course of the polymerization can be influenced in such a way that the formation of liquid polymerizate can be suppressed in favor of the formation of higher solid substances and that these will be formed in a greater amount if one uses as catalyst aluminum chloride with addition of titanium tetrachloride and, if necessary, with addition of other halides of the Friedel-Crafts catalyst series. These solid substances are formed especially easily, moreover, if one also adds hydrogen chloride-fixing metals, such as iron, zinc, or aluminum in finely divided form, especially aluminum powder.¹⁰

10 DX-30t at p. 1.

The Fischer patent teaches the combination of aluminum powder (Al), aluminum chloride (AlCl₃) and titanium tetrachloride (TiCl₄) to polymerize ethylene. Nowhere in the patent is there any mention of diethyl aluminum chloride or any organo-aluminum molecule. Indeed, none of the substances listed above as catalyst components even contains a carbon source for the formation of the organo-aluminums, essential to the claims of the '115. Thus, on its face the patent does not clearly indicate that the polymerization involves or may involve intermediate products that might include dialkyl aluminum halides.

As strict as the test for anticipation may be, it is not based on whether a trial judge can find within the four corners of the asserted anticipatory reference the full, clear and exact representation of the challenged patent. Thus, while it is to Dart's disadvantage that Fischer does not on its face mention or unambiguously indicate the dialkyl aluminum halide component or reaction product that is the essence of the '115 system, that absence is not conclusive. If, from the text of the Fischer patent, a chemist could "practice the invention [of the '115] without having to depend on the ['115] patent or his inventive skills,"¹¹ then the '115 claims will be anticipated.

11 Universal Athletic Sales, supra, 546 F.2d at 544.

Dart introduced voluminous testimony in its attempt to prove that the Fischer system as disclosed necessarily and inevitably evolves organo-aluminum reaction products that are active in the observed polymerization of ethylene. Dart further argued that skilled chemists of the 1950's would have looked at the Fischer patent and understood that the evolution of organo-aluminum species was a part of the process even though no organo-aluminums are mentioned in the text of the patent. After a careful review of all the evidence, and for the reasons set forth below, the Court finds that Dart has failed to meet the burden of proving that the Fischer patent anticipates Ziegler's '115.

1. Initial Experimental Evidence

In preparation for its invalidity defense, Dart employed Dr. George A. Olah, formerly of Case Western Reserve University and presently on the faculty of the University of Southern California, to run a series of tests aimed at determining whether the Fischer process, as disclosed in the patent, evolves organo-aluminum compounds in the course of polymerizing ethylene.¹² Dr. Olah ran a number of tests reproducing Fischer with experimental variations that he expected would allow him better to understand the character of the Fischer reaction.¹³ Counsel for both parties examined Dr. Olah extensively on the nature of his experiments and the validity of his conclusions. See, e.g., T-2450-2870, 4153-4330. Based on testimony and questions during his first appearance at trial, Dr. Olah left Wilmington, returned to his lab, conducted additional tests, and was recalled for further testimony.

12 The Court's discussion does not, at this point, touch on or evaluate the quality of the polyethylene produced by the Fischer patent or compare that product to the polyethylene produced by the '115 process.

13 Dr. Olah's report summarizing his experimental data and conclusions is DX-142. See T-2471-2488.

In the course of his testing to detect the presence of alkyl aluminums in the supernatant of the Fischer mixture, Dr. Olah used some of the most current analytic technology, including nuclear magnetic resonance (NMR) studies and other non-destructive analytic techniques developed in the last several years. Despite the extent and duration of his effort, Dr. Olah never clearly detected organo-aluminum compounds in any experimental runs that exactly duplicated the proportions of aluminum powder, aluminum chloride, titanium tetrachloride, and hydrocarbon solvent that were enumerated in the working example of the Fischer patent.

The conclusion of Olah's report reflected his belief that, given a choice between ascribing the Fischer results "to some kind of organo-metallics related catalysis, or . . . to a typical Friedel-Crafts polymerization,"¹⁴ an experienced chemist of the time would undoubtedly have opted for the former. The choice is tendentious and does not reflect the inquiry before the Court at this time. Given the absence of any mention of organo-metallics in the Fischer patent, the relevant inquiry is whether an experienced chemist of the time, looking at the patent and the results it reached under the conditions it discloses, would have concluded, without further suggestion, that the mechanism of the reaction not only represented a significant departure from traditional Friedel-Crafts catalysis, but, further, that the actual mechanism must have involved organo-metallics. Olah's conclusion, correct as far as it goes, does not meet the standards for anticipation under 35 U.S.C. § 102.

14 DX-142, p. 10.

In the course of his report, prepared before trial and therefore not reflecting the experiments he conducted during the course of the trial, Olah states that "no ethyl-aluminum compound could be detected in the supernatant heptane layer of the Fischer polymerizations."¹⁵ While Olah offers an explanation for the inability of NMR spectroscopy to detect organo-aluminums which he nonetheless believes to be present, the inability directly to prove their existence, even when seeking to do exactly that, suggests that a chemist reading the Fischer patent in 1953 would not inevitably arrive at the polymerization catalyst taught by Ziegler's '115 patent.

15 DX-142, p. 12. See also T-2821.

2. Evidence from Repeated Experiments

After he had returned from California with additional test results, Olah testified that a carbon-13 NMR spectrum of a particular experimental run indicated to him the presence of "a very small amount of an ethyl aluminum compound."¹⁶ Similarly, Olah concluded that an aluminum-27 spectrum of the same experimental sample indicated "the presence of some organic aluminum compound, but I couldn't make, based on these spectra, any determination of the exact nature."¹⁷ The sample on which these spectra are based reflected the proportions of aluminum powder (1g) and aluminum chloride (10g) of the Fischer example in half the amount of heptane solvent (250ml rather than 500ml of heptane). The experimental run did not use any titanium tetrachloride. The temperature was maintained at 65°-72°C during Olah's repeat experiment. The only indication of temperature in the Fischer example is that "ethylene is pressured in so that the temperature reaches 155°."¹⁸

16 T-4194. See DX-625, 626 (reproductions of the spectra to which Olah referred).

17 T-4195. See DX-627 (aluminum-27 spectrum).

18 DX-30t. The Court notes that while the significance of temperature control and temperature thresholds was sharply contested throughout testimony on Fischer, their significance went to the nature and quantity of the polyethylene produced rather than to whether alkyl aluminum compounds were evolved by Fischer.

Despite these experimental differences and the diligence of Dr. Olah's investigation, Dr. Bailey, SGK's expert, examined the data and spectra upon which Olah had commented and concluded that "you'd have to have a great imagination to look at that curve and say that there was evidence of a material, an ethyl aluminum compound. . . ." T-4387. It is not at all unusual to find experts in disagreement, but here the data are those developed by Dart's expert to supplement the conclusions one would draw upon examining the text of the Fischer patent. The fact that Olah's support for the inherency of ethyl aluminum chlorides is so weak, even though going beyond the "four corners" of the Fischer patent, leads the Court to conclude that Dart has failed to meet its burden of proving anticipation. The Court finds that the Fischer patent does not exhibit the substantial representation of the invention of the '115 in sufficiently full, clear and exact terms to allow the skilled chemist to recreate the challenged '115 claims without the benefit of any source outside of Fischer.

Thirty years after the fact, experts of extraordinary skill in the field, armed with exceptional technology and, perhaps more important, hindsight, did not conclusively demonstrate that ethyl aluminum chlorides are created by the Fischer reaction. It is unlikely that a skilled chemist who read the Fischer patent would independently surmise and be able to practice the dialkyl aluminum halide and titanium tetrachloride combination of the '115 patent. The Court finds that Fischer does not anticipate the '115.

B. Obviousness

Section 103 denies patentability to inventions that are obvious even though "not identically disclosed or described as set forth in section 102 of this title." 35 U.S.C. § 103. Thus, the Court's inquiry into Dart's defense that the '115 is invalid because obvious will be broader than the previous inquiry on the defense of anticipation. In evaluating a claim of obviousness under § 103, the Court applies the standards set out by the Supreme Court in Graham v. John Deere Co., 383 U.S. 1, 86 S.Ct. 684, 15 L.Ed.2d 545 (1966).

Under § 103, the scope and content of the prior art are to be determined; differences between the prior art and the claims at issue are to be ascertained; and the level of ordinary skill in the pertinent art resolved. Against this background, the obviousness or nonobviousness of the subject matter is determined. Such secondary considerations as commercial success, long felt but unsolved needs, failure of others, etc., might be utilized to give light to the circumstances surrounding the origin of the subject matter sought to be patented. As indicia of obviousness or nonobviousness, these inquiries may have relevancy.

383 U.S. at 17-18, 86 S.Ct. at 693-694. Because Graham provides a certain organizational regularity in inquiries as to obviousness, the Court will, for the most part, structure its analysis around the Graham standards as elaborated and interpreted by more recent rulings.

Before undertaking an analysis based on Graham elements, however, it will be useful to explain the experimental work done at Du Pont and how it relates to or demonstrates prior art. The Du Pont work, examined below in some detail, is not itself prior art relative to the Ziegler '115 patent being challenged here. The Court rejects Dart's arguments to that effect. At best, Du Pont's investigation was contemporaneous with Ziegler's.

Dart, in introducing voluminous testimony on the Du Pont work, sought to establish that Du Pont, working from sources completely independent of Ziegler, developed a polymerization catalyst system¹⁹ functionally similar to that developed by Ziegler. Furthermore, Dart argues that the relative rapidity with which the Du Pont team accomplished its work reinforces the obviousness of the solution to those skilled in the art. Implicit in Dart's arguments are the assumptions that Du Pont's work was clearly independent of Ziegler and that the Du Pont team's research represented the work of people "having ordinary skill in the art to which said subject matter pertains." 35 U.S.C. § 103. The accuracy and import of these assumptions will be examined, where relevant, in the course of the Court's analysis of Dart's arguments for obviousness.

19 Du Pont researchers called their catalysts "coordination catalysts." See T-2154-56.

Dart's obviousness theory, like its argument relating to anticipation, hinges on the Fischer patent as prior art.²⁰ The basic theory of obviousness is that knowledge of the Fischer patent and two articles by Hall Nash written in the 1930s²¹ would lead the chemist of 1952-54 to conclude that solid polymers of ethylene could best be obtained through organo-aluminum-based catalysts. Hall Nash, it is argued, provided information that would cause the skilled chemist to conclude that the interaction of aluminum powder, aluminum chloride, titanium tetrachloride and ethylene would produce as an intermediate and active catalyst component alkyl aluminum compounds that would necessarily include the dialkyl monohalide forms specified in the claims of the '115 patent.

20 DX-30t. See text accompanying notes 9-18, supra.

21 DX-47, 48. The two articles were written in 1937 and 1938 respectively. Throughout the testimony, the two articles were referred to merely as "Hall Nash." No effort was made to distinguish between the two. Consequently, unspecified references to "Hall Nash" within this Opinion will refer to both articles.

1. Natta '987 as Prior Art

Dart argues that Natta's United States Patent No. 3,587,987 ('987) is prior art against the '115 based on the '987's July 27, 1954 Italian priority date. The Court finds that, in arguing the Natta patent, Dart may not assert the Italian priority date of a U.S. patent as a prior art reference. The Court agrees with the interpretation of §§ 119 and 102 in Application of Hilmer (Hilmer I), 359 F.2d 859 (Cust. Pat.App. 1966) and finds that a prior art reference patent is effective only as of its U.S. filing date. See Studiengesellschaft Kohle v. Eastman Kodak Co., 616 F.2d 1315, 1337 (5th Cir. 1980); Application of Hilmer (Hilmer II), 424 F.2d 1108 (Cust. Pat.App. 1970); Eli Lilly Co. v. Brenner, 375 F.2d 599 (D.C. Cir. 1967); 3 Chisum, Patents § 1405[3] pp. 14-56-14-59 (1982).

Dart argues that Sperry Rand Corp. v. Knapp-Monarch Co., 307 F.2d 344 (3d Cir. 1962) represents a different position than that adopted by the Fifth Circuit, D.C. Circuit and Court of Customs and Patent Appeals. After a re-examination of Knapp-Monarch, this Court is persuaded that the position it adopted in Waterman-Bic Pen Corp. v. W.A. Shaeffer Pen Co., 267 F.Supp. 849, 854 (D.Del. 1967) is not inconsistent with the Third Circuit's opinion in Knapp-Monarch. ²² The Court finds the rule reflected by Hilmer I and II to represent the better interpretation of the statute as written and does not read Knapp-Monarch to prohibit it from adopting that position here. Natta's '987, when asserted by Dart as a prior art patent reference, is entitled only to its U.S. filing date of June 8, 1955. See SGK v. Kodak, supra, 616 F.2d at 1338. Consequently, it is not prior art against the '115.

22 The Court notes that its position does conflict with the district court's opinion in Knapp-Monarch. See 193 F.Supp. 756, 757 n. 3 (E.D. Pa. 1960), aff'd, 307 F.2d 344 (3d Cir. 1962). The circuit court's treatment of the Wimberger patent is ambiguous as to whether the patent stands as prior art or whether sources and references in the patent indicate the existence of references prior even to Wimberger's patent.

2. The Prior Art: Hall Nash

The Hall Nash articles describe the authors' observations in a series of ethylene polymerizations in which aluminum powder was added to the typical Friedel-Crafts catalyst, aluminum chloride. Hall Nash observed that the addition of metallic aluminum increased both the rate of the polymerization and the viscosity of the lubricating oils produced.²³ Hall Nash did not produce or distill a solid polyethylene. Their goal was to produce synthetic lubricating oils.

23 DX-48 at p. 478-79.

All the Hall Nash data reflect experiments run at 100-200°C. The single recorded attempt to operate below that range was performed at 15° and produced no detectable reaction.²⁴ In contrast to the later work of Fischer and the Ziegler team, there is no mention of any heavy metal compounds; no titanium chloride component is present in the Hall Nash system.

24 DX-48 at p. 476. See T-3556.

From the perspective of Dart's argument to this Court, the principal discovery of the Hall Nash work is that the combination of aluminum powder, aluminum chloride and gaseous ethylene produced ethyl aluminum chlorides. Hall Nash specifically observed and characterized both diethyl aluminum chloride and monoethyl aluminum dichloride.²⁵ Because the Hall Nash articles preceded Fischer, Dart argues that a skilled chemist of the time would infer from the presence of aluminum powder and aluminum chloride in the Fischer reaction that organo-aluminums, including dialkyl aluminum halides, must be evolved.

25 DX-47, 48. As noted above, by convention and usage, diethyl aluminum chloride has come to be abbreviated in a number of ways, predominantly as DEAC. Hall Nash, however, labelled the diethyl form "aluminum diethyl chloride" and symbolized it Al(C₂H₅)₂Cl. The monoethyl form, referred to in the trial transcript as EtAlCl₂ or as "the dichloro," was labelled by Hall Nash as aluminum ethyl dichloride and empirically written as Al(C₂H₅)Cl₂. See DX-48.

3. Fischer with Hall Nash

The Fischer patent makes only passing reference to the state of the art in 1943. Fischer does not mention Hall Nash or, as pointed out earlier,²⁶ any organo-aluminum. The sole reference to work preceding his own refers to ethylene polymerizations "to form lubricating oil-like products with the aid of aluminum chloride under pressure and at elevated temperature."²⁷ Fischer does not mention aluminum powder as a component previously discovered to have beneficial effects on product viscosity. When the Fischer patent does mention aluminum powder, it implies that the utility of aluminum powder is part of the new discovery. This does not suggest a strong relation between Fischer and Hall Nash.

26 See text accompanying notes 9-11, supra.

27 DX-30t at p. 1.

Moreover, Fischer sets out aluminum chloride, titanium chloride and Friedel-Crafts halides in reference to "the catalyst:"

[Solids] will be formed in a greater amount if one uses as catalyst aluminum chloride with addition of titanium tetrachloride and, if necessary, with addition of other halides of the Friedel-Crafts catalyst series.²⁸

28 DX-30t at p. 1 (emphasis supplied).

The reference to aluminum powder does not occur until the following sentence: "[Solids] are formed especially easily, moreover, if one also adds hydrogen-chloride fixing metals, such as iron, zinc or aluminum, in finely divided form, preferably aluminum powder."²⁹ Together, the two sentences could plausibly be read to imply that the aluminum powder was not part of the active catalyst and that its function was to bind hydrogen chloride that, were it evolved in the reaction, could produce "cracking" of the hydrocarbons.³⁰ Regardless of whether this is what Fischer actually intended, it plausibly suggests that the Hall Nash reaction producing aluminum alkyls may not have been either in the forefront of the mind of Fischer or of a chemist reading the Fischer patent.

29 Id.

30 See T-1707, 2778-81. The parties disagreed on how one properly ought to interpret the description "hydrogen chloride-fixing." Dr. Olah contended that the phrase did not describe the function of the metals in the context of the patent, but rather specified the group of metals useful in the Fischer system according to a characteristic. See, e.g., T-2780-81. He admitted, however, that using the phrase as a description left one with a very broad class of eligible metals. T-2781. Dr. Bailey, by contrast, testified that he interpreted the Fischer patent to indicate that aluminum powder was included to bind HCI and to suppress possible cracking. T-3465-67.

The parties devoted considerable attention to the experimental differences between Hall Nash and Fischer. The principal difference is that Hall Nash recommended a 3:1 ratio of aluminum chloride to aluminum, using 30 grams of AlCl₃ and 10 grams of Al. See DX-48 at p. 483-85. Hall Nash also experimented with higher proportions of aluminum powder, but found that while more aluminum alkyls were produced, the ethylene product tended to be less satisfactory. Id. The Fischer working example teaches a ratio of 30g AlCl₃:10g TiCl₄:1g Al. Thus, the AlCl₃:Al ratio is ten times greater in Fischer than in Hall Nash.³¹

31 The significance of this reduction in proportionate amount of Al will be examined below. The Court's purpose here is merely to note that the difference in the AlCl₃: Al ratio, along with other experimental conditions, indicates that the relationship between Hall Nash and Fischer is not so immediate or apparent as an initial examination might suggest.

Secondary differences between Hall Nash and Fischer, such as differences between parameters of temperature, pressure and the presence or absence of a heptane solvent, are of significantly less consequence in determining the likelihood of whether the two would naturally be read together by a chemist of ordinary skill in 1953.

4. Other Prior Art References

Dart cited a number of other references designed to prove that one or more of the components of the '115 patent were obvious from the prior art.³² The Court has taken these into consideration, but finds none individually significant enough to merit comment here.

32 See T-1330-1438, 1642-72, 1695, 3421-45, 3449-57. See also DX-155, 156, 158, 159, 162, 165, 217, 222, 225, 241, 242, 245, 254, 290. Each of these asserted prior art references diverged even more widely, from the invention in Ziegler than did Fischer and Hall Nash. While many were in fact prior to the '115 and taught polymerization, or at least alkylation of simple hydrocarbons, none approached being sufficient to render the '115 obvious. Even in combination with the references discussed in the text, none made such a significant addition that it warranted specific treatment.

5. The '115 and the Prior Art

The principal difference between the prior art, especially Fischer, and the '115 is the '115's clear and intended use of dialkyl aluminum halides as indispensible catalyst components to polymerize lower olefins, including but not limited to ethylene, to solid, high molecular weight polymers. As discussed above, Fischer never explicitly mentions organometallics. Hall Nash did not produce solid polyethylene. Neither Fischer nor Hall Nash specifically addressed the possibility of polymerizing propylene or other olefins. The '115 explicitly applies to "olefins up to about C₅."³³ The nature of this difference has been explored in the preceding subsections and in the earlier discussion of anticipation. Consequently, the Court will now briefly consider other differences that, within the § 103 analysis dictated by Graham, supra, may bear upon the issue of obviousness.

33 DX-1, Col. 1, lines 54-55. See text accompanying notes 76-78, infra.

a. Physical Characteristics of the Product

The Fischer patent characterizes its polyethylene product as "yellow brown masses . . . which, after purification through extraction, form an almost white, loose powder."³⁴ "The solid polymerizate formed is insoluble in common organic solvents."³⁵ In the Fischer working example, 10 grams of TiCl₄, 30g AlCl₃ and 1g Al plus 165 atmospheres (atm) of ethylene produced 145g of solid product and 194g of oils and other amorphous residues.³⁶

34 DX-30t at p. 2.

35 Id. at p. 3.

36 Id. at p. 3-4.

The '115 initially produces a "paste consisting of the solvent used and a white ethylene polymer,"³⁷ which when worked up by extraction methods comparable to those in Fischer yields "a snow-white polyethylene."³⁸ The working examples of the '115 uniformly indicate that the proportionate yield of solid polymer substantially exceeded that produced in the Fischer working example.³⁹

37 DX-1, Col. 11, lines 68-69.

38 DX-1, Cols. 11-12.

39 Id. at Cols. 11-17. Cf. DX-1 at p. 3-4; text accompanying note 36, supra.

The Fischer patent does not analyze its solid product in terms of crystallinity, molecular weight or linearity. Professor Koenig's analysis of the products of Dr. Olah's attempts to reproduce Fischer in terms of these variables is not particularly persuasive evidence that the powder in Fischer approached the Ziegler products in crystallinity or molecular weight. The parties never agreed on which, if any, of Olah's samples closely approximated Fischer. Since the Fischer work itself is silent on the point, the Court will not undertake a comparison of Fischer and the '115 on these characteristics. The Court notes in passing, however, that the polyethylene commercially available before the '115 system became known was substantially more branched than that claimed in the '115. Earlier samples showed 5-10 methyl groups per 100 carbon atoms whereas the '115 product contains.1-.3 methyls per 100 carbons.⁴⁰ b. Experimental Conditions

40 T-3852-54. See DX-1, Col. 11, lines 1-8. See also DX-1, Col. 10, lines 59-62: "The temperature stability or resistance of the polymers, such as polyethylenes, is greater than that of the known conventional polyethylenes." To the extent that temperature stability reflects crystallinity, the '115 shows increased crystallinity relative to products available at the time.

The Fischer patent teaches the use of "elevated temperatures, preferably at 130 to 180°C and elevated pressures, preferably at 30 to 80 atmospheres or above."⁴¹ The Fischer example operates at temperatures that reach 155° and 58 atmospheres.⁴² By contrast, the '115 operates efficiently across a wide range of temperatures and pressures, including experimentally mild conditions. Polymerization will occur at atmospheric pressure and at room temperature or below.⁴³ These characteristics sharply distinguish the '115 system from the free radical polymerization methods generally in use up to that time. Free radical polymerization typically requires extremes of both temperature and pressure and is consequently very expensive for commercial processes.⁴⁴ The mild conditions under which the '115 system operates enormously increased the commercial practicability of olefin polymerization.

41 DX-30t at p. 21.

42 Id. at p. 3.

43 DX-1, Col. 9, lines 34-37, 40-42, 46-48: "It is an outstanding advantage of the invention that one may operate at ordinary atmospheric pressure with excellent results."; lines 60-61: "The temperature of the contacting is not critical and the same may be effected at room temperature or below."

44 See text accompanying notes 88-89, infra; T-31, 3852-54.

6. Findings As To Obviousness

The determination of what constituted ordinary skill in the relevant art at the time of the Ziegler '115 invention will not be separately made by the Court in this Opinion. The issue was not specifically addressed at trial. The determination of ordinary skill is so closely intertwined with the Court's overall findings on the obviousness defense that it will be made in the context of that analysis and those findings. Two aspects of Dart's obviousness defense depend importantly on issues of ordinary and extraordinary skill and will be dealt with below.

a. Ziegler's Statements on Fischer

In January, 1954, Ziegler wrote to von Kreisler, his German patent counsel, regarding an addition to the German patent applications he was filing. These applications later became the basis for Ziegler's '115 and '332 patents in the United States. In the letter, Ziegler included the following passages:

I would like to draw to your attention another thing as follows. We have become aware of a patent application by the BASF. The subject matter of this application is undoubtedly a forerunner of our new processes. Only the inventors did not recognize what they had in their hands. . . . The true mechanism of the process according to the BASF patent is thus the primary formation of a small amount of ethyl aluminum sesquichloride . . . and these genuine aluminum-organic compounds lead then in combination with titanium tetrachloride to the solid polyethylenes as also obtained by us. Naturally, the inventors did not see this connection. Furthermore, they have everything else than a smooth reaction. The excess in aluminum chloride . . . and finally, quite different temperatures are employed for the reaction, than are required by us. All told, our new processes thus have nothing to do at all with this BASF application.

With the now almost unbelievable widening of the entire field which even a short time ago could not be foreseen, I fear that quite a dangerous situation has arisen. There are indeed here and there in the literature hints which can easily lead a clever man, if he combines them with our publications in the aluminum-organic field, to this new field.⁴⁵

45 DX-31t pp. 1-2.

Dart argues that the statements in the quoted passages indicate that Ziegler himself knew that Fischer's patent (the BASF application) and Hall Nash made Ziegler's own polymerization invention obvious. The Court cannot agree.

First, despite his examination of the Fischer patent and his recitation of its features similar to his own work, Ziegler nonetheless concluded that his processes were wholly different than those disclosed by Fischer. While Ziegler saw and admitted similarities, most of which he read into the Fischer patent based on his own knowledge, he concluded that his system was different.

Second, Ziegler's theorizing about the evolution in Fischer of ethyl aluminum compounds did not represent the application of "ordinary skill in the art of which said subject matter pertains." 35 U.S.C. § 103. Ziegler was probably the world's leading authority on organo-aluminums. See T-2324. Even in the quoted passage, he refers to his previous work in the area. His '115 and '332 patents and the fact that he was awarded the Nobel Prize in 1963⁴⁶ further support the Court's findings in this respect.

46 See note 84, infra, and accompanying text.

Given his background in organo-metallic, especially organo-aluminum, chemistry, the Court finds that Ziegler would read organo-aluminums into the Fischer patent more quickly than would the ordinarily skilled polymer chemist of the day. Ziegler's assessment of the inevitability of combining Fischer with "other references", presumably Hall Nash, actually represented hindsight on Ziegler's part. Ziegler and his team had been over that ground, whether aware of Fischer or not, and already performed the work that supported the applications of which he wrote. In reading Fischer, Ziegler was the captive of his own work just as subsequent scientists speculating about the actual mechanism of Fischer are captive to the chemistry that has intervened, especially the work of Ziegler and Natta.⁴⁷

47 The Court's finding on this point is supported by the testimony of Dr. Martin, T-3862. The Court's finding that subsequent experts who have speculated that organo-aluminums inhere in the Fischer mechanism are influenced by the subsequent chemistry is supported by Dr. Olah's acknowledgment that by 1958, when BASF reperformed the Fischer experiments ( See DX-614), Ziegler's work was already well and widely known and could conceivably have influenced BASF's choice of experimental conditions. T-4239-4244. For the work of various post-Ziegler commentators, see DX-149 (Dr. Mark); DX-593 (Dr. Hopff); DX-150, 151 (Dr. Tornqvist); DX-154 (Dr. Kennedy); DX-153 (Dr. Boor). None of these publications was prepared before 1958. See also text accompanying notes 48-49, infra.

The Court notes that Dr. Tornqvist's most recent publication on the subject of Ziegler and Fischer strongly concludes that "[n]othing could be more wrong" than the "suggestion that Ziegler's discovery was in some way influenced or followed as a logical consequence of Fischer's disclosures."⁴⁸ Tornqvist goes on to state that while he believes that alkyl aluminums are formed in Fischer, "the composition of the catalyst used in the example of the Fischer patent is such that no useful Ziegler catalyst can possibly be formed. . . ."⁴⁹

48 PX-57 at p. G62.

49 Id. (citations omitted).

The effect of "hindsight" on the Court's analysis of obviousness will be further discussed below.

Dart argues that the Breil doctoral thesis,⁵⁰ in which Breil examined the Fischer patent in light of Breil's own experiments, further indicates that Ziegler, as Breil's teacher and thesis adviser, and Breil, as a co-inventor of the patent, understood Fischer to read on their own work with aluminum alkyls. Dart relies principally on the following passage from Breil's thesis which refers to the Fischer patent: "It seems to me natural that a catalyst of the type of my catalysts was formed in tests of this type."⁵¹

50 DX-25t.

51 DX-25t at p. 18.

The Court finds that Breil's statement, as was the case with Ziegler's, does not represent the conclusions of a chemist of ordinary skill in the art at the time. While, as a graduate student, Breil is not to be held up as someone extraordinarily accomplished or beyond the normal range of ability, his close association with Ziegler and intimate familiarity with the work that led to the '332 and '115 patents did make him especially likely to find alkyl aluminums in the Fischer work or other patents or research he encountered. Moreover, Martin testified that, at the time Breil was writing, his conclusions were purely speculative. The SGK research team had not performed any experimental analysis of Fischer at that time. T-3849.

In addition to the Ziegler letter and Breil thesis, Ziegler commented upon the Fischer patent in the course of his defense to the German Opposition proceedings brought by BASF and the Nullity proceedings subsequently brought by Ziegler. While Ziegler clearly had an interest in resolving these proceedings in his favor, his September 25, 1957 analysis of Fischer⁵² is worth noting for the distinctions it draws between Fischer's and Ziegler's work.

52 PX-67.

The opponents . . . would like to consider it today as completely selfunderstood that the expert while reading the Max-Fischer-Patent had to arrive immediately at the known combination: Hall Nash produced organic aluminum compounds from aluminum, aluminum chloride and ethylene and it was, therefore, selfunderstood and obvious to prepare the polymerization catalysts by mixing such aluminum compounds with titanium tetrachloride. Precisely the opposite conclusion is correct. [O]ne cannot avoid the data that also other metals such as iron and zinc can be used . . . instead of aluminum. Max Fischer . . . [prefers aluminum but] clearly states: It also works with other metals," and this is decisive. No metal-organic compounds are known from iron. . . . a very special inventive effort was required to recognize that contrary to these data of the Fischer-patent and by Hall Nash the organic aluminum compounds were extremely important for the development of the polymerization catalysts.⁵³

53 Id. at 6, 7.

In support of this interpretation, Dr. Martin testified that, if reacted with TiCl₄ and AlCl₃, iron would not produce organic compounds. T-3871-873. While Dr. Olah did not directly concur in this analysis, he did indicate that iron would be considerably less active than aluminum or zinc in producing organo-metallics. T-2778-2780.

To the extent that other scientists of the time regarded the formation of organo-iron compounds as unlikely or impossible, the Fischer reference to iron as a possible substitute for aluminum⁵⁴ tends to teach away from the recognition of organo-metallics as the active intermediate product of the Fischer process. Prior art that teaches away from the methods actually used is relevant to indicate nonobviousness. To the extent that Fischer teaches away from organo-metallics, their recognition and use in the '115 more clearly represents inventive and patentable work. See General Battery Corp. v. Gould, 545 F.Supp. 731 (D.Del. 1982); W.L. Gore Assocs., Inc. v. Carlisle Corp., 529 F.2d 614 (3d Cir. 1976).

54 DX-30t at p. 1.

Overall, the Court finds that statements by Ziegler and Breil about the inherency of organo-aluminums in the Fischer mixture are unpersuasive evidence of obviousness. Ziegler and all the members of his team were far more likely than most polymer chemists of the day to read organo-aluminums into the Fischer patent.

b. The Du Pont Work

United States Patent No. 3,050,471⁵⁵ claims a catalyst system for the production of solid ethylene polymers and co-polymers. The patent was issued to Arthur W. Anderson, et al. of Du Pont and, along with other patents issued to Anderson and his Du Pont research team, reflects the experimental work on olefin polymerization done at Du Pont during the mid-1950s. Dart argues that the work done by Anderson and other Du Pont chemists during that period starting from Fischer, Hall Nash and other prior art references demonstrates that Ziegler's invention was obvious within the meaning of 35 U.S.C. § 103.

55 DX-288 [hereinafter frequently "the '471" or Anderson's '471"].

The Anderson '471 patent teaches reacting "a titanium trihalide with an organo-metallic compound containing at least one hydrocarbon radical bonded to a metal. . . ."⁵⁶ The patent specifically claims titanium trichloride (TiCl₃) and alkyl metal halides "sufficient to lower the valence state of the titanium, at least in part, to below three."⁵⁷ The patent specifically discloses diethyl aluminum chloride (DEAC) as an effective reducing agent although DEAC is not a component of any of the catalysts described in the claims of the '471.⁵⁸ Thus, the Du Pont system is quite similar to the '115 which is entitled to an earlier filing date and, hence, to priority.

56 Id. at Col. 9, lines 8-10.

57 Id. at lines 11-13.

58 DX-288 at Col. 3, lines 23-24.

In early February of 1954, Dr. Merckling at Du Pont performed experiments along the lines of the example in Fischer. T-2111, DX-56. By mid-February, Merckling and others, including Truett, had done enough work with the Fischer system to persuade Anderson and other senior chemists that the work was significant and that Fischer had not grasped that significance because he perceived his system to be merely an advancement of Friedel-Crafts catalysis. Because the Du Pont chemists had been able to produce polyethylene of high molecular weight, a feat beyond known Friedel-Crafts methods, they concluded that the Fischer work involved something novel. T-2113-2118.

Based on the experiments, a literature review, and his own knowledge and creativity, Anderson intuited⁵⁹ that the active component in Fischer was a reduced valence species of titanium. T-2120, 2127, 2227-29, 2270-74. Consequently, the exploratory group, comprised of a number of Du Pont chemists under Anderson's direction, undertook to investigate various methods of reducing titanium tetrachloride. In the course of this research, the Du Pont team investigated a number of reducing agents. One technique was direct reduction with finely divided metals. A great deal of early attention was devoted to the reduction of TiCl₄ and TiCl₃ through alkylation with a Grignard reagent.⁶⁰ Grignards produced more high molecular weight polymer and less oil than catalysts produced by aluminum-reduced titanium.

59 T-2280.

60 A Grignard reagent is a compound of the type RMgX, in which R is an organic radical, e.g., an alkyl or aryl radical, and X is a halogen. Ethyl magnesium bromide (C₂H₅)MgBr was among those tested by the Du Pont researchers. See DX-203, T-2431-34, 2438-39.

There is no evidence of any alkyl aluminum compound being used until sometime in March or April of 1954.⁶¹ Anderson testified that in March, the alkyl aluminum being used as alkyl aluminum iodide. An April 27, 1954 report⁶² to members of the chemical department mentioned di-butyl aluminum iodide (Bu₂AlI) and butyl aluminum di-iodide (BuAlI₂) in combination with TiCl₄, but mentioned no ethyl aluminums or alkyl aluminum chlorides. The report stated that:

61 T-2308.

62 DX-73. See also PX-50 (April 21 report from Robinson to Gresham).

So far the best halides are titanium, vanadium and zirconium and the best second component either magnesium or the Grignard reagent. . . . At the moment there is no clear view of the mechanism of this catalysis. Polychem's thoughts . . . favor organometallic intermediates somewhat reminiscent of Ziegler's polymerization with aluminum alkyls.⁶³

63 Id. at PP00648. A June research report by Robinson, a senior chemist at Du Pont, also indicated that "the possibility of organo-metallic intermediates suggested a possible relationship to Ziegler's work." DX-74 at p. 4.

Anderson testified that through June of 1954, the team, which had grown in size and increased the intensity of its work, was primarily concentrating on Grignards. T-2323.

In early August, additional actors became involved in directing the Du Pont research effort toward turning out patents. By that time, Du Pont had information that Ziegler had developed a low pressure polymerization process.⁶⁴ Gresham, Anderson's superior at Du Pont, began to press for further activity. On August 3, Anderson wrote a report to Gresham outlining the difference between Ziegler's and Anderson's patent applications and the Du Pont strategy to find areas undominated by Ziegler.

64 T-2323, PX-52.

[D]u Pont has claimed alkylating agents broadly with each transition element, whereas Ziegler has claimed each alkylating agent class separately with all the transition elements. This makes it easier for us to find areas undominated by Ziegler. . . .

However, it is important to remember that finding an undominated new alkylating agent only permits us to use it without infringing his patents.⁶⁵

65 DX-55 at p. 1.

It was not until August and the addition of Hyson to the exploratory group that alkyl aluminums were actively investigated and used to produce solid polymers. See T-3116-3125. Hyson understood his task to be to find a catalyst system more efficient than those produced by Du Pont to that date. T-3131. In the abstract to an experiment run during this period, Hyson wrote: "The titanium tetrachloride/lithium aluminum R₄ catalyst system is the most versatile of any yet developed for the production of linear polyethylene." T-3124. The statement was based on comparing this TiCl₄/organo-aluminum with the work done by others prior to that date. T-3124-25.

After an extensive review of the Du Pont work, including the deposition testimony of those not available to testify in person, the Court finds that the efforts of the exploratory group do not demonstrate that Fischer and Hall Nash or other prior art render Ziegler's '115 obvious. Du Pont chemists read the Fischer patent early. Despite a major effort, involving many experienced scientists with high priority access to the resources of Du Pont's experimental station,⁶⁶ the Du Pont group explored a number of alternatives before investigating alkyl aluminums. If, as Dart argues, ethyl aluminum chlorides are "obviously" the mechanism of the Fischer patent, one would expect Du Pont to have looked into them more immediately. Du Pont's failure to do so takes on additional significance in light of the recognition by Hyson and, subsequently, the polymer industry at large, that alkyl aluminum halides were particularly effective.

66 See T-2316, 2327; PX-52.

Du Pont's work represented a major effort by fine scientists, headed by at least one, Anderson, who possessed more than ordinary skill in the relevant art. Yet, despite Anderson's exceptional insight in attributing the activity in Fischer to reduced valence titanium species, the team under his direction did not arrive at the organo-aluminum component essential to the Ziegler '115 until after an intensive and sustained effort. Fischer and Hall Nash may have been the starting point for the Du Pont work, but the path that Du Pont's skilled and highly motivated scientists took cannot be said to demonstrate that Ziegler's work was obvious from the prior art. The ordinarily skilled polymer chemist of 1953-54 would not have read Fischer plus Hall Nash and concluded that alkyl aluminums inhered in the Fischer reaction or that they were responsible for the activation of the catalyst by reducing TiCl₄ to a lower valence. The information available in the prior art required the application of significant additional inventiveness before it resulted in the catalyst system that Dart seeks to challenge.

c. Secondary Considerations

In addition to the basic inquiries it mandates, Graham, supra, 383 U.S. at 17-18, 86 S.Ct. at 693-694, counsels investigation of a number of secondary factors relevant to obviousness. Because analysis of these factors merely reinforces the Court's conclusion, based on the primary factors already examined, that the '115 is not obvious from the prior art, they will be treated rather briefly.

1. Commercial Success

There can be no doubt that the '115 was an enormous commercial success. By radically changing both the conditions under which commercial polymers could be made and the physical characteristics of the polymer product, Ziegler catalysts revolutionized the plastics industry.⁶⁷ Other than Dart, the principal producers of olefin polymers throughout the United States and much of the rest of the world have taken licenses under the '115⁶⁸ or been found to infringe the '115. See Ziegler v. Phillips Petroleum Co., 483 F.2d 858 (5th Cir.), cert. denied, 414 U.S. 1079, 94 S.Ct. 597, 38 L.Ed.2d 485 (1973). By contrast, Dart was the only company known to have taken a license under the Anderson '471 patent.⁶⁹ Evidence of commercial success unequivocally indicates that the scientists and business people of the industry overwhelmingly concluded that Ziegler's work represented an important commercial breakthrough.

67 See text accompanying notes 83-87, infra.

68 See T-3839-40; see also notes 137, 147, infra.

69 See T-836-38, 2342; see also note 131, infra.

The Court finds the widespread and immediate industry recognition to be strong evidence of the nonobviousness of the '115. As was stated in another case in which an obviousness defense was rejected: "All of the indicators of unobviousness approved by the courts are present: . . . the revolutionary commercial success; the enormous savings gained by using the invention; the fact that every significant tire manufacturer except Firestone has taken a license. . . ." General Tire Rubber Co. v. Firestone Tire Rubber Co., 349 F.Supp. 345, 360 (N.D. Ohio 1972).

2. Unsolved Needs

The evidence that many major chemical companies were working to find a more economical way to produce high quality polyolefins suggests that the industry felt the lack of a catalyst system such as Ziegler's. The extent and intensity of the Du Pont work supports the conclusion that olefin polymerization was being actively studied by some of the finest chemists in the world. Similarly, the fact that Ziegler's work won him the 1963 Nobel prize for chemistry indicates the intellectual as well as commercial importance of his work.

The fact that a solution to the problem was being actively and widely pursued, but was actually solved by Ziegler and the SGK team further supports the Court's finding of nonobviousness.

The fact that the solution to the problem solved. . . . had baffled workers of extraordinary skill in the catalytic art for many years, but, when revealed, was immediately and extensively adopted, establishes the existence of patentable invention worthy of protection by the courts under the patent laws.

Mobil Oil Corp. v. W.R. Grace Co., 367 F.Supp. 207, 235 (D.Conn. 1973).

3. Hindsight

Finally, the Court finds that the superficial persuasiveness of Dart's arguments and assembled exhibits is substantially undercut by keeping clearly in view the difference between the hypothetical inventor of 1953 working in his shop with all the prior art references, and the current expert witness looking now at what was available then. The current scientist brings with him or her the "baggage" of thirty years of astonishing advances in polymer chemistry. It is a far different inquiry that he or she undertakes, and what is undoubtedly chemically obvious now was not "obvious" under the patents laws to the 1953 inventor to whom the prior art was readily available.

In arguing its case, Dart has assembled a substantial number of references that seem to read on one or more elements of the '115. Skillful arguments create a tempting path along which the hypothetical inventor is asserted unavoidably to have travelled. But there is hindsight involved; not only in discerning the path, but in choosing the starting points. The fundamental fact remains that Ziegler and SGK developed the revolutionary system that others actively sought. That the solution now seems elegantly simply does not vitiate the importance of the fact that Ziegler discovered it first. "After the fact, Firestone has assembled dozens of references from which they argue obviousness, yet they cannot explain why the industry did not think to use this `obvious' invention. . . ." General Tire Rubber, supra, 349 F.Supp. at 360. Courts must resist the significant temptation to include the art of the invention being challenged among the data or knowledge imputed to the inventor of the time.

While the defendant now argues in retrospect and with the benefit of hindsight, that an analysis of the prior art indicates that the puzzle might have been put together, had the right parts only been placed in the proper position, the problem was not one of putting together a specific chemical substance.

Mobil Oil Corp., supra, 367 F.Supp. at 235. Thus, it has been recognized to be "absolutely essential to guard against piecemeal reconstruction of the [challenged] invention by using the [challenged] patent as a blueprint." Tate Engineering Inc. v. U.S., 477 F.2d 1336, 1345 (Ct.Cl. 1973).

The Court finds that when the entire range of evidence is scrutinized without the deceptive clarity of hindsight, the Ziegler '115 patent is demonstrated to be both nonobvious and not anticipated.

Even when all the references are viewed in conjunction with all the testimony taken at trial, the evidence falls short of "clear and convincing" proof that the method disclosed by the patent would have been obvious to a person with ordinary skill in the relevant art at the time.

Trio Process Corp. v. L. Goldstein's Son's Inc., 461 F.2d 66, 72 (3d Cir.), cert. denied, 409 U.S. 997, 93 S.Ct. 319, 34 L.Ed.2d 262 (1972).

C. Prosecution Before The Patent Office

Dart argues that the manner in which SGK prosecuted the '115 before the Patent Office manifested sufficient bad faith that the '115 should be held invalid or unenforceable.⁷⁰ 1. Failure to Cite Fischer

70 The Court notes that, although mentioned in the pre-trial stipulation of February 17, 1982, there was no testimony on this subject at trial and that it was first developed in the defendant's post-trial briefs.

The Court finds no fraud in SGK's failure to cite Fischer to the Patent Office Examiner. Dart's argument to this effect is rejected.

2. Inventorship

Dart argues that the inventors named on the face of the '115 — Ziegler, Martin, Breil and Holzkamp — do not accurately reflect the inventorship of the patent. Dart contends that the inventorship determinations for both the '332⁷¹ and the '115 were manipulated to allow Ziegler's claims for polypropylene and DEAC under the '115 to share the early foreign filing dates more properly attributable to the applications associated only with the '332. The Court rejects these arguments and finds in them no reason to declare the '115 invalid or unenforceable.

71 United States Patent No. 3,257,332 issued to the same inventors on June 21, 1966 and awarded a foreign filing date of November 17, 1953. This patent, originally part of this action, was dropped from the suit at the pre-trial conference of February 5, 1982.

The issue was raised before the Patent Office in a different proceeding and resolved in favor of SGK. The Patent Office opinion stated:

Finally, in view of [exhibits] which show the acceptance by the German Patent Office of the naming of the four inventors, Ziegler, Martin, Breil and Holzkamp . . . we hold that said application complies with the requirement of 35 U.S.C. § 119 that the privilege afforded thereby be claimed by the inventors designated in the United States application.⁷²

72 DX-250 at p. 18.

Moreover, the Patent Office accepted the change in inventorship list and held "that Ziegler has complied with 35 U.S.C. § 116 and Rule [of Practice] 45(c). Furthermore, we can find no basis in the record for concluding that there was any deceptive intention by any of the inventors in originally omitting the names of Breil and Holzkamp when the involved Ziegler application was filed."⁷³

73 DX-250 at p. 22-23.

In addition to this previous decision by the Patent Office, the Court finds, as is discussed below in the context of pioneer status for the '115, that each of the named inventors played a role in the essentially unitary program of research that led to both the '332 and the '115.⁷⁴ The Court finds adequate factual bases for listing all four men as inventors of the '115. In the light of the available evidence, the Court finds that Dart has failed to prove its essentially technical inventorship defense by clear and convincing evidence. See Eli Lilly Co. v. Premo Pharmaceutical Labs, 630 F.2d 120 (3d Cir. 1980).

74 See text accompanying notes 83-85, infra.

The Court holds that the '115 patent is valid.

II. INFRINGEMENT

Dart argues that its catalyst system does not infringe the '115 patent. In order to evaluate these claims, the Court must establish the scope of the claims of the '115. See Lockheed Aircraft Corp. v. U.S., 553 F.2d 69 (Ct.Cl. 1977); Tate Engineering, Inc. v. U.S., 477 F.2d 1336 (Ct.Cl. 1973). See also Kodak, supra, 616 F.2d at 1323-24. The first place to turn is to the patent of itself.

A. Scope of the '115 1. Claims

The initially stated object of the '115 is "a new catalyst useable, inter alia, for the polymerization of ethylenically unsaturated hydrocarbon products."⁷⁵ Propylene (C₃H₆), diagrammed in the margin,⁷⁶ is undeniably an ethylenically unsaturated hydrocarbon or olefin. The olefinic bond is the "double bond" attaching one of the terminal carbons to the middle carbon atom. Propylene, as diagrammed, is an alpha olefin ( -olefin).

75 DX-1, Col. 1, lines 46-48.

76

The fourth-mentioned objective of the patent is the invention of "a polymerization catalyst for lower olefins up to about C₅ and particularly ethylene."⁷⁷ Propylene, C₃ H₆, is an olefin included within the range of "up to about C₅." See Phillips, supra, 483 F.2d at 876. Indeed, it is only one methyl group larger than ethylene,⁷⁸ which is specifically mentioned as a monomer to be polymerized by the patented catalysts. Propylene is also disclosed as one of the monomers to which '115 catalysts will be applied.

77 DX-1, Col. 1, lines 53-55.

78 See note 76, supra.

Claim 1 of the '115 states:

1. Polymerization catalyst essentially consisting of an aluminum compound having the general formula RR'AIX, in which R is a member selected from the group consisting of hydrogen, alkyl radicals and aryl radicals, R' is a member selected from the group consisting of hydrogen, alkyl radicals and aryl radicals, and in which X is a member selected from the group consisting of hydrogen, halogen atoms, alkoxy radicals, aryloxy radicals, secondary amino radicals of the formula

in which R" and R"' are hydrocarbon radicals, secondary acid amide radicals of the formula

in which R" and R"' are as given above, mercapto, radicals, and radicals of carboxylic acids of the formula

in which R" is as given above, with a heavy metal compound selected from the group consisting of salts, freshly precipitated oxides and hydroxides of metals of groups IV — B, V — B, and VI — B of the periodic system, including thorium and uranium, metals of group VIII of the periodic system and manganese.

Dart's DEAC falls squarely within the class of aluminum compounds RR'AIX in which R and R' are both alkyl radicals.⁷⁹ Specifically, in DEAC the univalent alkyl radicals are ethyl groups, C₂H₅. Dart's diethyl aluminum compound satisfies the general formula of '115 where X is a halogen atom,⁸⁰ in this case chlorine.

79 For the purposes of this Opinion, it is adequate to describe an alkyl radical or an alkyl group as a hydrocarbon of the general formula C_nH_{2n + 1}. Here the alkyl is ethyl, C₂H₅, which could be formed by dropping one hydrogen from a saturated ethane molecule.

80 Halogens are the elements of Group VIIA of the Periodic Table. Of this group, flourine and chlorine are the most reactive.

The heavy metal component of a '115 catalyst, as described in Claim 1, clearly includes TiCl₄. Titanium is a metal of Periodic Group IVB. Combined with chlorine, it forms a heavy metal "salt" within the definition of the patent.⁸¹ TiCl₃ falls within the same general category of "heavy metal salts" as defined under the terms of the patent, as does TiCl₄. See Phillips, supra, 483 F.2d at 878. Thus, a DEAC + TiCl ₃ combination falls squarely within the description of Claim 1 of the '115.

81 The patent states that, in relation to the description of heavy metal salts, the term is employed "in its broadest sense." DX-1, Col. 2, lines 27-29.

DEAC and various titanium chlorides are more narrowly specified in Claim 4:

Catalyst according to claim 1, in which said aluminum compound is a dialkyl aluminum monohalide and in which said heavy metal compound is a compound of a metal from group IV-B of the periodic system.⁸²

82 DX-1, Col. 20, lines 44-47.

As discussed above, DEAC is a dialkyl aluminum monohalide and TiCl₄ and TiCl₃ are heavy metal compounds comprised of a group VI-B heavy metal, titanium. This is true of various species of TiCl₃. Regardless of whether the TiCl₃ is produced by reducing TiCl₄ with aluminum powder, with DEAC, with aluminum triethyl (ATE or Et₃ A1) or with hydrogen, the TiCl₃ produced is still a salt of titanium and chlorine. Similarly, changes in the orientation of the titanium and chlorine atoms within the TiCl₃ molecule, described as -TiCl₃, -TiCl₃, -TiCl₃, and -TiCl₃,⁸³ do not violate the description of claim 4. These forms are all titanium salts.

83 See DX-398.

2. Development

The '115 embodies work done by Professor Karl Ziegler and his team of chemists during the early 1950s. It differs from his '332 patent in that it covers dialkyl aluminum compounds as opposed to the aluminum trialkyls. The two patents represent revolutionary discoveries that came out of a unitary research effort by a single team of scientists. Ziegler was widely regarded as a prominent authority on organo-aluminum compounds and their application. T-2324. In 1963, Ziegler was awarded the Nobel Prize in chemistry for his work in developing organo-metallic catalysts for polymerizing olefins.⁸⁴

84 Ziegler shared the Nobel Prize with Guilio Natta. See PX-4.

The catalysts that Ziegler and his team developed revolutionized the production of high molecular weight polymers of simple hydrocarbons. These catalysts are crucial in making production of these plastics, especially propylene, commercially practicable. Before Ziegler's work, there was doubt about whether propylene could be polymerized at all. See Standard Oil Company (Indiana) v. Montedison, S.p.A., 494 F.Supp. 370, 374 (D.Del. 1980), aff'd, 664 F.2d 356 (3d Cir. 1981), cert. denied, ___ U.S. ___, 102 S.Ct. 1769, 72 L.Ed.2d 174 (1982).⁸⁵ Through application of Ziegler's discovery, Natta at Montecatini, Martin at the Max Planck Institute, and, subsequently, many others were able to produce crystalline polypropylene on a commercial scale. Indeed, the plastics produced by improvements on the basic organo-metallic catalyst specified in the '115 have become ubiquitous, particularly in the United States.

85 See also DX-390 at 857-858.

In professional circles, the catalysts themselves have come to be known as "Ziegler catalysts."⁸⁶ Stauffer, the company that supplies Dart with its preformed TiCl₃ component, explains the use of its product in the following manner:

86 See DX-398 at p. 447 ("Since the development of complex catalysts based on transition metal compounds and alkylmetals by Ziegler in 1953. . . ."). See also DX-390 at 858: "the classical Ziegler organometallic catalytic systems"; at 859: "the original catalysts proposed by Ziegler"; at 860: "the typical Ziegler catalysts." The Court also notes the entry in Hawley, Condensed Chemical Dictionary (10th ed. 1981) under "polymer, stereospecific" which mentions Ziegler-type catalysts. Indeed, there are separate entries for "Ziegler catalysts" and "Ziegler process."

All grades of Stauffer TiCl₃ are normally used in combination with an aluminum alkyl to prepare catalysts which produce stereo-specific polyolefins. The catalyst combination is known as the "Ziegler-Natta" system. . . .⁸⁷

3. Pioneer Status

87 DX-392; PX-9A, Stauffer brochure, at 2. The Court also notes the following quotation from a Stauffer brochure not in evidence here, but which was cited by the Fifth Circuit in Phillips, supra, 483 F.2d at 878:

Titanium Trichloride is an important co-catalyst used with Aluminum Alkyls, in the polymerization of olefins. Titanium Trichloride and an Aluminum Alkyl is the Ziegler-Natta catalyst system.

The Supreme Court, in Westinghouse v. Boyden Power Brake Co., 170 U.S. 537, 18 S.Ct. 707, 42 L.Ed. 1136 (1898) established a workable definition of a pioneer patent:

To what liberality of construction these claims are entitled depends to a certain extent upon the character of the invention, and whether it is what is termed in ordinary parlance a "pioneer". This word, although used somewhat loosely, is commonly understood to denote a patent covering a function never before performed, a wholly novel device, or one of such novelty and importance as to mark a distinct step in the progress of the art as distinguished from a mere improvement or perfection of what had gone before.⁸⁸

88 170 U.S. at 561-62, 18 S.Ct. at 718. See Corning Glass Works v. Anchor Hocking Glass Corp., 374 F.2d 473, 476 (3d Cir. 1967) ( quoting Westinghouse, supra). See also 4 Chisum Patents § 18.04 pp. 18-32.3-18-54 (1982).

In Corning Glass Works v. Anchor Hocking Glass Corp., 374 F.2d 473 (3d Cir. 1967), the Third Circuit applied the Westinghouse standard to a situation in some ways analogous to this one.

In Corning Glass, the circuit court affirmed this Court's finding that the development of a crystallized glass marketed under the name Pyroceram constituted a pioneer patent. Before the Pyroceram patent, crystallized glass, with distinctive thermal properties, had "accounted for only a fraction of the total world glass production." Id. 374 F.2d at 476. Citing its adaptability to "a wide variety of commercial applications," the circuit court found that, in the case of "such a pioneer patent, liberality becomes the keynote of construction requiring the court to give the patentee a wide breadth of protection in construing the patent claims and specifications." Id. at 476.

Alpha olefins were polymerized before Ziegler's inventions. The Ziegler team's work, however, did revolutionize a process of great commercial import that had, up to that point, essentially relied upon free radical polymerization methods. Free radical polymerization requires extremes of temperature and pressure, conditions costly to maintain on a commercial scale. See T-31. Ziegler's novel catalysts are effective across a wide but significantly lower range of pressures and temperatures. Ziegler's invention is a basic one, marking a distinctive step in the progress of the art. See Westinghouse, supra, 170 U.S. at 562, 18 S.Ct. 718.

Dart has argued that even if the Ziegler work taken as a whole represents a pioneering breakthrough for the art, the '115, standing alone, is not a pioneer patent because it departs only minimally from the claims and specifications of Ziegler's '332 patent. The '332, although issued later than the '115, is argued to be based on German patent applications submitted a few months earlier than the priority claim for the '115.

The Court acknowledges the difference between deciding that an individual was a pioneer inventor and deciding that a particular patent represents a pioneering addition to the art. See Moore v. U.S., 211 U.S.P.Q. 800, 808 (Ct.Cl. 1981). After an extensive review of the record in this case, however, the Court finds itself in agreement with the Fifth Circuit in Phillips, supra, and finds that the '115 is a pioneer patent. The '115 itself represents a significant step in the progress of the art.

The Court considers it significant that the work done by Ziegler's team at the Institute in 1953-54 was continuous and, for all intents and purposes, a unitary inquiry. The work was done primarily by a single, relatively small team in a sustained effort. The development of the catalyst system was brilliant, novel and of the utmost important to the commercial world of polymers. Under the circumstances of this case, and the substance of the testimony before it, the Court cannot accept an argument that "pioneering" status must be attributed exclusively to the '332 or '115 patent. The Court finds the unity of research that led to these two Ziegler patents is not fatal to the '115's status as a patent entitled to a broad range of equivalents.⁸⁹

89 See T-12-60 for testimony regarding the events surrounding the discoveries patented in the '115 and '332. See, generally, 4 Chisum, Patents, § 18.04[2] pp. 18-36-18-42.

On cross-examination, Dart's counsel asked Dr. Martin if the novelty of the '115 vis-a-vis the '332 was "substituting a chloride for the ethyl." T-148. Martin agreed. Dart's counsel, drawing on a large sketch pad, DX-565, then inquired whether that difference — between DEAC (or Et₂AlCl or AlR₂Cl) and ATE (AlEt₃ or AlR₃) — was a pioneer invention. Martin stated emphatically that it was and that an organo-metallic chemist would easily recognize that fact. Martin then volunteered to explain why he thought the substitution of a chlorine atom for the third ethyl group on ATE was revolutionary. Dart's counsel, however, declined to hear Martin, and the point was left for plaintiff's counsel to elicit on redirect. T-148.

Although he never precisely revived the question on redirect, plaintiff's counsel did eventually ask Martin about the difference in reaction product and polymer product when TiCl₄ was reduced by DEAC as opposed to ATE. T-459-461. Basically, the difference lies in the capacity of ATE to reduce TiCl₄ to a valence below 3 whereas DEAC, a weaker reducing agent, can reduce TiCl₄ no further than to a valence of 3. Martin testified that reduction of TiCl₄ by ATE produced both black and brown colored precipitates. The black dominated and represented a divalent titanium halide, probably TiCl₂. The brown precipitate was the same -TiCl₃ produced when using DEAC.⁹⁰

90 See note 96, infra.

The significance of the difference between a mixture of divalent and trivalent titanium chlorides produced by ATE reduction and unmixed -TiCl₃ (brown trivalent) produced by DEAC is reflected in the crystallinity (isotacticity) of the polymer produced. While the exact mechanism of this stereoselectivity is still incompletely understood,⁹¹ it is established that DEAC is more effective than ATE in producing polypropylene of high crystalline content or, in Natta's lexicon, polypropylene of high isotacticity.⁹² It is this highly isotactic polypropylene that is of principal commercial value worldwide and which Dart produces at its Odessa, Texas plant.

91 See Pino Mülhaupt "Stereospecific Polymerization of Propylene: An Outlook 25 Years After Its Discovery" 19 Angew. Chem. ( Int.Ed. Engl.) 857 (1980), DX-390.

92 See DX-390, supra, note 91. See also, Tornqvist, "Stereospecific Polymerization of Olefins." 155(2) Annals of the New York Academy of Sciences 447 (1969) (use of dialkylaluminium halide, e.g., AlEt₂I, as opposed to trialkylaluminum yields a more stereospecific catalyst), DX-398.

Whether in his first offer to demonstrate why the '115 was a "pioneer" invention vis-a-vis the '332, Martin would have recited only the facts elicited in the later redirect or whether he would have argued that there was some further revolutionary aspect in Ziegler's substituting DEAC for ATE, is not established by the record. Regardless, in the absence of persuasive evidence by the defendant that this was not the major step that Martin claims for it, the Court finds that the distinctions between the '115 and '332, especially as they relate to the polymerization of propylene, are significant.

Dart's representations that Ziegler's January 7, 1954 letter to von Kreisler in which Ziegler asks that von Kreisler act swiftly to file several additional German applications, one or more of which became bases for the U.S. '115, are ultimately inconclusive. While the Ziegler letter,⁹³ does, at one point, state that, "[f]rom the point of view of patent law the newly won chemical knowledge should be of rather insignificant importance," it also makes claims for the significant novelty of the new developments:

93 See DX-31t.

We have now found that it is not necessary at all to start from aluminum trialkyls.

In this connection, however, we have only one experiment so far and the same relates to the combination diethyl aluminum chloride + titanium tetrachloride. . . . With [this] discovery. . . . there exists as against the complex of our previously newly found reactions a perfectly new discovery and novelty. . . . For the chemist it is clearly evident from this that the now newly discovered catalytic processes have nothing at all to do with our earlier reactions.⁹⁴

94 DX-31t at p. 1 (emphasis supplied).

Ziegler's estimation of the import of his new discoveries in terms of patent law seems to argue against construing the '115 as a pioneer. Ziegler, however, was not a lawyer and, in any case, used "should be" in a way that qualified his assertion. Moreover, he was predicting the legal importance of the discoveries in terms of German rather than United States patent law. Taken in the context of the entire letter, Ziegler's suggestions about legal novelty go more to his hope for a speedy filing and granting of these German applications than to any consideration relevant to how this Court should read his U.S. patents. In light of the longer of the passages quoted above, the Court finds Ziegler's letter to be inconclusive of whether the '115 is a pioneer patent.

The system of patents and patent law is designed both to reward past inventors and to encourage new invention. Courts that too zealously protect the former class may thwart the law's goal of promoting the latter. Thus, the balance of these values implicit in the doctrine of equivalents dictates that trial courts carefully weigh decisions regarding pioneer patent status. Yet, patent law has as its further goal the protection and advancement of the public good. That goal is accomplished and encouraged by requiring that inventors, in exchange for their temporary legal monopoly, publically disclose the fruits of their labors. These are the elements of the contract between patentee and public. See Catanzaro v. Masco Corp., 423 F.Supp. 415 (D.Del. 1976), aff'd sub. nom., Catanzaro v. Int'l Tel. Tel., 575 F.2d 1085 (3d Cir.), cert. denied, 439 U.S. 989, 99 S.Ct. 586, 58 L.Ed.2d 662 (1978). Furthermore, the public's right is to early disclosure in order to maximize the invention's availability before it is superseded or rendered obsolete.

A policy in favor of prompt disclosure is particularly appropriate in the case of pioneer patents, where publication of a novel invention is all the more likely to stimulate others. "[B]asic inventions have led to basic patents, which amounted to real incentives, not only to invention and its disclosure, but its prompt, early disclosure." Application of Hogan, 559 F.2d 595, 606 (C.C.P.A. 1977).

If courts were to sever every aspect of a duly patented invention developed later than the earliest application upon which that patent is based, and refuse protection to those later aspects, no matter how independently important or developmentally integrated with the disclosures of the first application, they could well discourage inventors from making the prompt, early disclosure the patent law seeks to encourage. In the case before the Court, Ziegler's early filing of applications, whether here or in Germany, is consistent with the policies and goals of the patent laws. Where, as with the '115 and '332 patents, concurrently developed yet independent novel inventions, by a single inventor or group of inventors, share a partial reliance upon a single revolutionary technology, the Court may permit more than one to partake of the protections accorded "pioneer" status.

In this case, without having the '332 patent before it and without deciding the status of that patent, the Court finds the '115 patent to be a pioneering advance in the field of olefin polymerization. It is "a great basic invention which has brought substantial benefit to the public at large." Corning Glass, supra, 374 F.2d at 479. Consequently, the patent will be read liberally with a broad range of equivalents. The Court's broad construction of the '115's claims will be limited by what is reasonable under the circumstances and will not expand the claims so far as to risk rendering the patent invalid. Tate Engineering, supra, 477 F.2d at 1341.

B. Dart's Catalyst

Dart's catalyst consists of diethyl aluminum chloride (DEAC or Et₂AlCl) and titanium trichloride (TiCl₃). There is little contention about the fundamental identity of the chemical character of the DEAC disclosed in the '115 and that used by Dart. It is the parties' divergent positions regarding the titanium component of the Dart catalyst that have generated the infringement claims before the Court.

Dart has attempted to distinguish its catalyst from the Ziegler catalyst by emphasizing numerous differences between the titanium tetrachloride (TiCl₄) explicitly mentioned in the '115 and the TiCl₃ Dart employs. Dart argues, in essence, that the cumulative effect of these quantitative and qualitative distinctions takes its titanium component outside the claims of the '115.

Even Dart, however, does not read the '115 to include only TiCl₄.⁹⁵ Throughout most of the testimony of Dart's chemical experts — Drs. Grubbs, Kelley and Olah — comparisons were based on an assumption that the '115 processes start out with tetravelent titanium chloride which is reduced to a trivalent species of titanium chloride. That reduction reaction, as summarized by Dr. Robert Grubbs, T-1845-1858, is approximately diagrammed in the margin.⁹⁶ The TiCl₃ produced by this reaction is of a brownish color and either crystalline or near-crystalline in form. TiCl₄ is a clear liquid. Although this species of TiCl₃ was variously referred to as "brown TiCl₃," " -TiCl₃," "beta-TiCl₃," or even just "beta," the Court will attempt uniformly to refer to it as -TiCl₃.

95 And if, contrary to the weight of its experts' testimony, Dart does intend to read the '115 as entirely limited to TiCl₄ the Court finds that the alkyl aluminums enumerated in the '115 are effective in reducing TiCl₄ to TiCl₃, see note 96, infra, and that the '115 contemplated reduction of titanium valences. See DX-1, Col. 7, lines 14-17.

96 Where DEAC =

The TiCl₃ species generated by the reduction of TiCl₄ by DEAC is the beta form, -TiCl₃. The active catalyst is produced by the interaction of DEAC and -TiCl₃. Because the DEAC is instrumental both in reducing the TiCl₄ and in forming the active catalyst, the initial molar ratio of DEAC: TiCl₄ should exceed 1:1 and ought probably to approach 3:1. (T-1856). See also DX-6t at 4, Example 1: "20 grams of diethyl aluminum chloride are carefully mixed with 1 gram of titanium tetrachloride, this solution thereby taking on a brown color and depositing a precipitate." Thus, in Ziegler's January, 1954, application he disclosed the reduction of TiCl₄ to -TiCl₃. DX-7t also refers to the "effect of the aluminum-organic or generally organo-metallic compounds upon the metal halides is to be interpreted, in the first place, as a reduction." DX-7t at p. 4. See also DX-1, Col. 7, lines 14-17.

Dart attempts to distinguish its titanium halide catalyst component first by noting that it starts with a crystalline species of TiCl₃ rather than with TiCl₄. Dart purchases its "preformed" TiCl₃ from Stauffer Chemical Company which produces four grades of TiCl₃.⁹⁷ Dart uses what Stauffer calls its "TiCl₃ A.A.". The A.A. designation indicates that the TiCl₃ has been reduced by aluminum⁹⁸ and that it has been "activated" by dry ball milling. The particular TiCl₃ A.A. that Dart uses is actually a co-crystal of titanium trichloride and aluminum trichloride.⁹⁹ This co-crystal is then ball milled. As procured by Dart, the TiCl₃ A.A. is a crystalline powder of violet color.¹⁰⁰

97 See DX-392; PX-9A (hereinafter sometimes referred to as the "Stauffer brochure").

98 Stauffer also markets TiCl₃H. that is reduced by hydrogen. See DX-392 at 2. See also Kodak, supra, 616 F.2d at 1322 (referring to H — -TiCl₃). Hydrogen-reduced TiCl₃ is not at issue in this case.

99 The formula for the co-crystal is variously represented as: (TiCl₃)₃ AlCl₃, TiCl₃•.33 AlCl₃ or simply TiCl₃•AlCl₃. The first two forms simply indicate the 3:1 ratio of TiCl₃ to AlCl₃. Unless otherwise noted, TiCl₃•AlCl₃ shall indicate a 3:1 ratio. Other ratios, e.g., TiCl₃•.2 AlCl₃ will be specifically noted.

100 Sometimes also referred to as purple-TiCl₃. See DX-392 at 3.

Regardless of what treatment they undergo before contact with DEAC or propylene, all species of TiCl₃ are reduced from TiCl₄ as the basic starting material.¹⁰¹ Thus, the violet TiCl₃ A.A. used by Dart is directly analogous to the -TiCl₃ produced by the interaction of DEAC and TiCl₄. Both are trivalent forms reduced from The tetravalent TiCl₄. The question is whether the differences between the two forms or between the activities of the catalysts they produce take the Dart system beyond the scope of the '115.

101 T-250.

Dart's effort to differentiate between its titanium catalyst component and the claims of the '115 depends critically on arguments going to the fundamentality of the following factors: a preformed violet crystalline starting component, dry ball milling, co-crystallization, in situ formation of the catalyst, and the use of liquid propylene as both monomer source and diluent. Dart argues that these factors collectively result in a catalyst that is far more active and efficient than anything contemplated by Ziegler and his co-workers.

1. Preformed TiCl ₃

This item has largely been addressed above. The Court finds that the functional similarity, setting aside momentarily the possible differences in activity and efficiency, between Dart's "preformed crystalline violet TiCl₃" and the -TiCl₃ produced by DEAC reduction of TiCl₄ is sufficient to bring Dart's violet TiCl₃ within a proper reading of the claims of the '115. The Fifth Circuit in Phillips, supra, 483 F.2d at 878, reached the same conclusion about the inherency of TiCl₃ in the language of the claims of the '115. Dart's use of "violet, crystalline TiCl₃" may take its titanium starting component outside of the literal language of the '115, but it does not take it beyond the reasonable equivalents of the claims of the patent. The '115 explicitly covers all species of TiCl₃ and the Court finds that Dart's violet crystalline TiCl₃ is within the scope of the '115's properly interpreted claims.¹⁰² 2. Ball Milling

102 Without relying on it as a basis for this aspect of its infringement decision, the Court notes that -TiCl₃ can be converted into one of the crystalline ( -, — or -TiCl₃) forms under the proper conditions. See T-1607-18; 1896; DX-398. See also DX-287, Tornqvist's '510 Patent, Col. 2, lines 40-43 and text accompanying note 103, infra.

Dart argues that the fact that its TiCl₃•AlCl₃ complex is dry ball milled before use in the catalytic process distinguishes it from the catalyst components claimed or contemplated by the '115. Dart also argues that dry milling so greatly increases the activity of the catalyst that it constitutes an invention beyond the scope of the '115.

Ball milling is not a mysterious or novel procedure. Basically, it is a process by which the physical characteristics of a substance are altered by grinding it in a rotating mill vessel in the presence of steel balls, or, sometimes, flint pebbles. The rotation of the mill imparts kinetic energy to the steel balls which in turn transfer it to the substance to be ground through repeated collisions. The usual result of the process is to break the substance into smaller particles. If the substance is suspended in a diluent or solvent during grinding, the process is "wet" milling; if there is no diluent, the milling is referred to as "dry."

United States Patent No. 3,032,510 ('510) issued to Dr. Erik Tornqvist of Esso (now Exxon) in 1962, teaches dry milling of the titanium components of olefin catalysts.¹⁰³ Tornqvist's '510 patent teaches milling the TiCl₃•AlCl₃ co-crystal with steel balls in a moisture-free inert atmosphere. The claimed result is a substantial increase in both the activity level, measured in terms of units of polymer produced per unit of milled catalyst component, and in the efficiency, measured in terms of units of polymer product insoluble in boiling heptane per unit of catalyst component, of the catalysts produced. It is this heptane insoluble fraction that is the crystalline portion of the polymer, e.g., isotactic polypropylene.

103 DX-287 [hereinafter sometimes referred to as "the '510" or "the Tornqvist patent"].

The '115 patent, in several of its examples,¹⁰⁴ teaches wet ball milling. It does not, however, mention dry milling, nor does it apply its wet milling techniques to titanium compounds. That titanium compounds were not milled is of little consequence since the patent teaches several groups of heavy metal compounds, only one of which is the class of titanium halides. Similarly, wet and dry ball milling serve sufficiently similar functions, in general, that the '115's failure explicitly to mention dry milling is not, of itself, fatal to SGK's claim of infringement.

104 See DX-1, Examples 4, 6, 7 and 13.

Qualitatively, the Tornqvist patent under which Stauffer presumably ball mills its TiCl₃ A. to produce TiCl₃ A.A. is merely an improvement on the basic catalyst system developed by Ziegler.¹⁰⁵ That the improvement constitutes a validly patentable invention does not vitiate a basic patent. "Improvement of the plaintiff's invention . . . while possibly affording a basis for a separate patent on the improvements, does not constitute a defense of noninfringement." Bendix Corp. v. U.S., 199 U.S.P.Q. 203, 222 (Ct.Cl. 1979). See Marvel Specialty Co. v. Bell Hosiery Hills, Inc., 330 F.2d 164 (4th Cir.), cert. denied, 379 U.S. 899, 85 S.Ct. 187, 13 L.Ed.2d 175 (1964);¹⁰⁶ 4 Chisum, Patents § 16.02[1] p. 16-7, 16-7 n. 10 (1982).

105 See note 110, infra, and accompanying text.

106 "[T]he making of patentable improvements does not avoid infringement." 330 F.2d at 167 n. 3. The Court notes that the defendant in 330 F.2d 164 is Bell Hosiery Mills and that the caption, as published, is probably a typographical error.

Since, standing alone, the fact of dry milling does not take Dart's catalyst outside the scope of the '115, Dart must rely on its argument that the quantitative differences between dry milled catalyst components and '115 components are of sufficient magnitude to go beyond the valid claims and equivalents of the '115. In evaluating this argument, the Court's task is to determine whether the quantitative differences are so substantial as to indicate that the Dart catalyst is not "performing the same work, in substantially the same manner to achieve substantially the same result as the claimed device." Lockheed, supra, 553 F.2d at 80.

Analyzing ball milling as a discrete factor, independent of the effects upon activity and efficiency claimed for co-crystallization, is quite difficult. The documentary evidence relied upon for quantitative comparisons, especially the work of Dr. Tornqvist of Exxon, and of unidentified scientists at Stauffer, are not primarily designed to isolate the effect of each incremental improvement in the technology of olefin polymerization.

The Court heard extensive testimony comparing the activity and efficiency levels of various combinations of catalyst components, milled and unmilled. While much of the evidence is contradictory, a number of themes emerge that make useful starting points in an analytic framework. Chief among these were: (1) the nature of increases in or tradeoffs between activity level and efficiency; and (2) whether dry milling produced results that could not be explained in terms of changes in the surface area of the milled substance. As previously noted, the replacement of ATE by DEAC in polymerization catalysis increases the crystallinity of the resulting polymer.¹⁰⁷ The increased stereospecificity of DEAC-reduced TiCl₃, however, is achieved at the price of a reduction in catalyst activity.¹⁰⁸ Thus, while DEAC is more efficient than ATE in producing isotactic polypropylene — it produced a higher percentage of heptane insoluble material — it is less active.¹⁰⁹ According to Tornqvist, the existence of this apparent trade off "caused us to start looking for improved solid catalyst components which would make it possible to prepare solid, crystalline polypropylene of attractive properties at a moderate cost."¹¹⁰ Dry ball milling and co-crystallization were the two principal improvements developed by Tornqvist and his co-workers at Exxon. Dart, by buying Stauffer TiCl₃ A.A., has adopted both improvements.

107 See text accompanying note 92, supra. See also DX-398.

108 T-1495; DX-398.

109 See DX-398, p. 447.

110 Id. The Court notes that Tornqvist's goal from the very outset was to improve upon the existing catalyst technology. In light of the opening sentence of his article: "Since the development of complex catalysts based on transition metal compounds and alkyl-metals by Ziegler in 1953. . . ." — there is little doubt that Tornqvist considered Ziegler's technology to be the essence of what he was seeking to improve. DX-398 at 447.

Milling is primarily directed at increasing the activity of catalyst preparations. Data presented in Tornqvist's article¹¹¹ and his '510 patent¹¹² indicate substantial activity increases among preparations that are dry milled over a period of several days. Isolating, for the moment, the effect of milling from the related effects of co-crystallization, the data and testimony suggest that dry milling increased catalyst activity by approximately a factor of 10 compared to unmilled TiCl₃ preparations.¹¹³ This increase in activity is accomplished with little or no change in stereoselectivity, at least insofar as it is measured by the percentage of heptane insolubles produced. Ball milling, whether wet or dry, does not appreciably affect the efficiency of the catalysts produced.¹¹⁴

111 DX-398.

112 DX-287.

113 T-1898. See also DX-392; PX-9A at p. 9, Table b. While the only comparison here is that between hydrogen-reduced, TiCl₃H., and the hydrogen-reduced, activated (ball milled) species of TiCl₃•AlCl₃ H.A., this is the comparison that was relied upon by Dart's expert witness.

114 DX-392, at p. 9, Table b. Figures for heptane insolubles for TiCl₃ H. and TiCl₃ H.A., when used with DEAC show an insignificant increase, from 80-85% to 88-91%. Comparison of TiCl₃ A. and TiCl₃ A.A., when used with ATE, show identical ranges of heptane insoluble percentages — 70-75%. Comparison of entries of TiCl₃ H. and TiCl₃ H.A. used with ATE also indicate that milling affects activity with no appreciable influence on efficiency. T-1896-1902, 1455-60.
The Court notes an entry in DX-287, the Tornqvist '510 patent, Table II, Cols. 5-6, lines 55-75 labelled "catalyst efficiency g/g." This entry, however, seems to measure grams of total polymer, atactic as well as isotactic, per gram of catalyst. Thus, the apparent relation there between milling and "efficiency" is a relationship between catalyst concentration and grams of total polymer. As such, it is closer to being a measure of activity as used in this Opinion.

SGK's expert, Dr. William Bailey, testified that there was no difference in the effect of wet and dry milling. T-3402. Bailey suggested that both processes produce the same result because they subject the milled substance to nearly identical forces, but in different magnitudes. Bailey's position was that wet milling could produce the same results as dry milling if a substance were wet milled for a longer period of time or in such a way that the energy of the steel balls were increased sufficiently to compensate for the amount of energy absorbed by the suspension medium. T-3402-3403. Bailey further testified that intense wet milling might increase the temperature of the system in the mill vessel sufficiently to convert -TiCl₃ to -TiCl₃ and, thereafter, to the -TiCl₃ claimed by Dart's experts to represent the most active crystalline configuration of the catalyst component.¹¹⁵

115 See T-3404. The Court notes that DX-398, Fig. 7 does not directly compare activity levels of the — and — forms. It does, however, compare -TiCl₃ to the — and — forms. There is evidence that any TiCl₃ must go through either the — or -form before it can be converted to -TiCl₃. T-1616. Fig. 7 comparisons are also for co-crystallized TiCl₃•AlCl₃.

SGK's witnesses testified that the entire increase in activity through ball milling was attributable to changes in the particle size and, hence, the surface area of the milled titanium compound. SGK argues that data presented or relied upon by Dart's experts are inconclusive because those data do not yield information on particle size. SGK contends that, -TiCl₃ and the violet- -TiCl₃ would prove comparably active if particle size were the same. Other than citing to the statements in the Stauffer brochure¹¹⁶ which note a correlation between particle size and activity, and appealing to one's natural inclination to relate a grinding process with an increase in surface area, SGK introduced no evidence supporting its proposition that the entire covariance between activity level and grinding could be explained by changes in surface area.

116 DX-392 at 4.

In contrast, Dart's experts relied upon data and conclusions set forth in the Tornqvist '510 patent indicating that "the large increase in catalyst activity obtained with the dry milled preformed metal halides is due to changes in the crystal structure of the metal halides rather than a mere increase in surface area."¹¹⁷ Table II of the '510 patent presents data indicating that the correlation between surface area and activity is incomplete. In a situation in which the sole variable was the duration of the milling, TiCl₃•.2AlCl₃ removed after nine days had a surface area calculated to be 16.5 m².g and yielded 169 grams of polymer. TiCl₃•.2AlCl₃ removed after twenty-one days was measured at 29^m2/g, yet yielded only 111 grams of polymer.

117 DX-287, Col. 3, lines 15-19.

The rest of the table, however, confuses the situation. Measurements at days 0, 2 and 6 show a non-linear but direct relation among time, change in surface area and change in activity. Measured surface area declined in samples from days 9 and 13. Activity, measured as yield, continued to increase between days 6 and 9, but declined between days 6 and 13. Between days 13 and 21, surface area increased and yield continued to drop. The relevant figures from Table II are set out in the margin.¹¹⁸

118 DX-287, Col. 5-6, Table VIII.

Time, days 0 2 6 9 13 21

Surface Area m²/g 2.5 5.6 35 16.5 14.5 29.0
Yield, g 18.6 47.5 138.0 169.0 156.0 111.0

While these data suggest that increases in activity associated with dry milling cannot be fully explained as a function of changes in surface area, no alternative pattern emerges to suggest a different relationship. Indeed, Tornqvist's analysis of Table II acknowledges that "the optimum milling time for a particular catalyst and for the particular milling equipment used can easily be determined by routine experimentation."¹¹⁹ A suggestion by the inventor that optima would have to be empirically determined and that relevant variables include not only the particular catalyst components, but even the equipment used, tends to reduce the explanatory power of the data.

119 DX-287, Col. 7, lines 6-9.

Table II reflects polymerizations with a co-crystal of titanium trichloride and aluminum trichloride of different ratio than that used by Dart.¹²⁰ Table II involved ATE rather than DEAC. More importantly, there is only one strong data point suggesting that surface area increased while activity (yield) declined. There are equally strong or stronger data for different intervals tending to show parallel increases in surface area and activity. On the whole, Tornqvist's data, undoubtedly accurate for the experiments they reflect, are a less than satisfactory basis for a general conclusion that dry ball milling affects catalyst activity in a way substantially independent of changes in the surface area of the milled particles.

120 Dart's TiCl₃•AlCl₃ ratio is 3:1, the experiment in Table II used a 5:1 ratio in its co-crystal (TiCl₃•.2AlCl₃).

Testimony by Martin and Bailey, although unsupported by documentary evidence, suggested that the incomplete correlation between measured surface area and activity reflected an inadequacy of measuring technique. In Tornqvist's work, the surface area was measured by the absorption of nitrogen at liquid nitrogen temperatures.

Bailey testified that since the nitrogen atom was smaller than the catalyst complex, one could observe increases in surface area measurable by nitrogen absorption but not available as an active catalyst site. T-3780-3783. This is an interesting theory, but does not explain the anomaly in Tornqvist's data. Bailey's hypothesis would explain an increase in nitrogen-measured surface area accompanied by a decrease in activity or no change. Tornqvist's data are troublesome because they suggest that activity increases exceed surface area changes, not that they lag behind. Bailey elaborated that measured surface area could decrease even though surface area increased, but he neither postulated a mechanism for that reaction nor explained how it would work in light of his preceding explanation.¹²¹

121 T-3784. In his testimony, Bailey mentions parallel decreases in measured surface area and in particle size. Decreases in particle size, of course, will typically increase total surface area.

Martin testified that SGK experimenters had also noticed and investigated the discontinuity in the surface area/activity relationship. Martin concluded that it was inappropriate to measure surface area with nitrogen at temperatures approximately 200°C. below the temperatures at which olefin polymerization occurs. He testified that SGK experiments measuring surface area with an organo-aluminum at approximately room temperature indicated a consistent relation between particle size and activity in the systems under examination. T-3836-3838. Martin did not produce the data or tests to which he referred. In the absence of these data, Dart's cross examination was limited to Martin's recollection of the tests and Dart's counsel had no opportunity to cross examine him on methodology or specific outcomes. Consequently, the Court is reluctant to credit fully the Martin testimony, elicited as it was on redirect examination directly responsive to Dart's evidence based on Tornqvist.

The sum of the scientific evidence before the Court indicates that a dry ball-milled crystalline starting component reliably produces a substantially higher activity level than is encountered with unmilled components. Within the limits of its ability to unravel the actual chemical reaction, however, the Court finds that activity increases result from increasing the number of active sites available for attachment of propylene molecules. It is not clear whether dry milling techniques accomplish this merely by increasing the surface area of the milled catalyst component or whether there is some additional factor involved.

Regardless, the Court finds that the evidence before it does not permit the conclusion that increased exposure of catalytically active sites represents a qualitatively different interaction between catalyst and monomer. The evidence indicates that the active sites of dry milled catalyst components are functionally equivalent to those that are produced by the unmilled or wet milled catalysts of the '115 patent. Consequently, the Court finds that the milled catalyst components used to produce crystalline polypropylene through the use of a titanium halide (TiCl₃) and diethyl aluminum chloride (DEAC) perform substantially the same work in substantially the same manner and to the same result as Ziegler's non-milled components. Given the fundamentality of the '115, the admittedly substantial increase in activity attainable through dry milling is not enough to take Dart's process outside of the proper range of equivalents when the evidence supports a difference that is quantitative, but not qualitative.

3. Co-Crystallization

The Dart process uses a TiCl₃ component that is both dry milled and a co-crystal of titanium chloride and aluminum chloride. In co-crystallized forms, aluminum atoms occupy some of the sites that, in a pure titanium chloride crystal, would be occupied by titanium atoms.¹²² At least in part because of the appreciable difference in the relative sizes of titanium and aluminum atoms, incorporation of AlCl₃ in the lattice of the TiCl₃ crystal changes its physical characteristics. These changes result in a catalyst both more active than noncocrystallized forms, and apparently, capable of producing a higher percentage of heptane insoluble polypropylene. Co-crystallization, in combination with dry milling, increases both activity and efficiency.

122 See T-1546-47, 1866; see also DX-398 at 450-452.

Throughout the trial, evidence relating to co-crystallization was presented in conjunction with evidence regarding ball milling. Their impacts upon activity and efficiency are not wholly independent of one another. Thus, observed differences between the '115 and the co-crystallized TiCl₃ A.A. of Dart's process will not necessarily be completely distinct from differences observed above. It may sometimes be less than accurate, where quantification is possible at all, to add the magnitude of differences observed in this subsection to differences observed in the previous subsection.

Co-crystallized TiCl₃•AlCl₃ is usually produced by reducing TiCl₄ with metallic aluminum in the presence of AlCl₃ in a heated environment.¹²³ Aluminum powder is a stronger reducing agent than DEAC. The resulting co-crystal is one in which, according to Dr. Grubbs, the incorporation of the AlCl₃ changes the spatial relationship of the titanium atoms in the complex. T-1547. Varying the ratio of AlCl₃ to TiCl₃ in the complex affects both the activity of the catalyst and the percentage of heptane insolubles produced.¹²⁴ Maximum activity is achieved with a TiCl₃ to AlCl₃ molar ratio of 3:1.¹²⁵ This is the TiCl₃•.33AlCl₃ that Dart acquires by purchasing Stauffer TiCl₃ A.A.

123 See T-2887-88, 1547-48. Tornqvist suggests that intense grinding alone may be sufficient to co-crystallize -TiCl₃ and AlCl₃. DX-398 at 460.

124 See DX-397 at p. 11, Table k; DX-398 at 457-48.

125 Id.

While the exact function of the AlCl₃ is not entirely clear, it has been theorized to increase the dispersion of the catalyst in the liquid monomer or in the diluent into which the monomer is introduced.¹²⁶ Were AlCl₃ to be removed from the co-crystal, its removal would not substantially inhibit the catalytic capacity of the crystal.¹²⁷ Removal of the AlCl₃ in such a way as to leave intact the existing spatial arrangement of the titanium atoms would not destroy the gains in either activity or stereospecificity that are achieved through co-crystallization.

126 DX-298 U.S. Patent 3,128,252 to Tornqvist [sometimes hereinafter "the '252"] at Col. 4, lines 62-63.

127 T-203-06, 1547-49. The testimony on this point leaves unclear to what extent AlCl₃ could practicably be removed from the co-crystal without destroying the crystal.

In determining whether co-crystallized TiCl₃•AlCl₃ falls within the range of equivalents of the '115, the fundamental issue is the role co-crystallization plays in catalysis. As a compulsory by-product of DEAC reduction of TiCl₄,¹²⁸ AlCl₃ is present under the conditions of the '115. Absent high temperatures or intense milling, however, it will not co-crystallize with TiCl₃. When it does co-crystallize, it can apparently be removed without proportionate damage to the catalytic activity of the system. These factors comport with the view that the function of AlCl₃ in the crystal is to promote more efficient and regular exposure of catalytically active sites to the propylene monomer. While this role is not chemically insignificant, the evidence presented to the Court suggests that the active sites so exposed are fundamentally identical to those of Ziegler catalyst systems. The role of AlCl₃ in the crystal is to orient the titanium atoms and to provide a particular electrochemical environment at the edges of the co-crystal lattice.¹²⁹ Co-crystallization results in the exposure of more active sites to the propylene monomer; it seems also to orient the titanium atoms in a way that facilitates construction in an isotactic polymer. Nonetheless, the chemistry at the active site is still fundamentally Ziegler catalysis.

128 See note 96, supra. See also T-212.

129 DX-398 at 463-65.

The Court finds that co-crystallization of TiCl₃ with AlCl₃ in the starting catalyst component falls within that broad range of equivalents to which the '115 is entitled. Tornqvist's work on both ball milling and co-crystallization is important and represents a valuable addition to knowledge of and control over the chemistry of olefin polymerization. Yet the Tornqvist work in both areas improves upon the basic Ziegler system. The text of Tornqvist's patents and of his article acknowledge that his novel inventions improve upon the basic polymerization chemistry that the Court has found to fall within the claims of the '115.¹³⁰ Beyond using TiCl₃ A.A. upon which Stauffer has performed the improvements developed by Tornqvist, Dart adds nothing in terms either of dry milling or co-crystallization.

130 See DX-289, the '252 patent, at Col. 1, lines 14-17: "More particularly [the patent] relates to improved modified catalysts comprising an at least partially reduced compound of a reducible heavy metal . . . activated by an organometal compound. . . ."; Id. at lines 52-53: "The main object of this invention is to modify and improve the above mentioned type of catalysts. . . ." See also note 92, supra; note 110, supra; DX-398; DX-291 at Col. 1.

Though not dispositive of its findings in either area — ball milling or co-crystallization — the Court finds it significant that only Dart took a license under the Anderson '471¹³¹ and that Exxon took a license under the '115. First, urging that the quantitative improvements in activity and efficiency achieved by incorporating the Tornqvist discoveries into its process takes it out of the scope of the '115, ignores the fact that Dart, and Dart alone, took a license under the Anderson '471 patent.¹³² Dart seeks to distinguish its process from that of the '115 by quantitative comparisons between the two. Yet a comparison between Dart's commercial process and the '471, the patent Dart, by taking a license, implicitly conceded it would infringe, reveals a divergence greater than that between Dart's process and the '115 example.

131 See T-836-38, 2342. Valles testified that he was not aware either of any other company that took a '471 license or of Du Pont ever asserting the '471 against anyone. Anderson could not name any '471 licensees other than Dart, although he believed there had been more than one licensee early on.

132 See DX-288 U.S. Patent No. 3,050,471 (August 21, 1962). Dart's license agreement with Du Pont under the '471 is reproduced at DX-424.

In testimony comparing the activity and efficiency of the '115 to Dart's process, Dart's witnesses compared working example 1 of Ziegler's August, 1954 German Patent Application¹³³ with Dart's commercial process.¹³⁴ Dart's expert calculated that in terms of polymer per gram of catalyst, the Dart system outproduced Example 1 by a factor of ten — 700 grams of polypropylene per gram of TiCl₃ for Dart to 71 grams of polypropylene per gram of TiCl₄ for the Ziegler example.¹³⁵ If the comparison is extended to include the example from the '471 that Dart's expert testified to be the closest to Dart's commercial operation,¹³⁶ one finds that the '471 catalyst demonstrates an even lower degree of polymer production. Dr. Grubbs testified that Example VII of the '471 would yield 35 grams of polypropylene per gram of TiCl₃. T-1936. That is approximately half of the figure arrived at for Ziegler's Example 1.

133 DX-7t.

134 For its present purposes, the Court will accept as reasonable the highly questionable proposition that a bench experiment can be validly compared to a commercial process. In the Court's analysis the burden of that potential fallacy is symmetric.

135 T-1556-72. For purposes of this comparison, the Court will not go beyond the calculation of relative activities. In his further calculations relating to Ziegler's example, Dart's expert postulated an efficiency rate (percent heptane insolubles) based on the literature because it could not be calculated from the example itself.

136 T-1931-32. Dr. Grubbs chose Example VII (Col. 4-5) of the '471 as the example closest to the Dart catalyst.

Dart contends that quantitative comparison alone sets its system outside the scope of the '115 and yet the basic patent under which Dart elected to take a license and pay royalties compares even less favorably, at least with regard to activity, than the Ziegler working example Dart chose to hold up to comparison. Viewed in this context, the claims of non-infringement by virtue of purely quantitative improvements lose a substantial amount of their superficial persuasiveness.

Second, and similarly, the debate over the magnitude of the difference between the '115 system and Dart's system relying on Tornqvist's improvements, dry milling and co-crystallization, is put into a more concrete perspective in light of the testimony that Exxon, under whose auspices Tornqvist developed his improvements, is a licensee under the '115 for the production of polypropylene. Exxon's process uses a TiCl₃ A.A. as a catalyst component.¹³⁷

137 See T-3839-40. See also note 147, infra, listing other U.S. '115 licensees for polypropylene production.

The Court finds that Dart's co-crystallized dry ball milled catalyst system is within the scope of the '115. Dart's use of -TiCl₃ A.A. from Stauffer does not, standing alone, save it from infringement of the '115. Despite the differences between the systems, Dart's catalyst system as explained performs substantially the same work in substantially the same way as the mechanisms of the '115 patent.

4. Other Claimed Distinctions

Dart claims that other comparisons between its system and the specifications of the '115 indicate non-infringement. These factors are subordinate to the distinctions discussed in the subsections above. They have also been litigated before in Phillips, supra, and declared not to take the Phillips catalyst out of the scope of the '115. Consequently, their treatment here will be somewhat abbreviated.

a. "Essentially Consisting Of"

Dart argues that inclusion of AlCl₃ in its titanium component means that its catalyst does not "essentially consist of"¹³⁸ a heavy metal salt and an alkyl aluminum as specified by the '115. The Court's previous discussion of co-crystallization should demonstrate that the essential feature, the sine qua non, of the catalyst systems being examined is the interaction of the titanium halide and organo-aluminum compounds. They are the essence of the catalysts.¹³⁹ Dart attempts to interpret the disputed passage as meaning "consisting exclusively of." The evidence does not bear such a reading. See Phillips, supra, 483 F.2d at 878.

138 DX-1, claim 1.

139 See DX-398 at pp. 447, 462.

Titanium and organo-aluminums are the "essence" of both Dart's catalyst and the '115. The addition of AlCl₃ in a co-crystal with TiCl₃ does not change the essence of the invention.

b. IN SITU Formation

Dart attempts to distinguish its process on the grounds that its catalyst is formed in situ, in the presence of the monomer. Dart contends, as Phillips did before it, "that the '115 discloses only a catalyst that is mixed and reacted beforehand, out of the presence of the monomer." Phillips, supra, 483 F.2d at 879.

Dart's argument ignores the broad range of equivalents to which the '115 is entitled. While the '115 teaches preformation of the catalyst components as the preferred mode, it is not limited to that preferred embodiment. See Continental Paper Bag Co. v. Eastern Paper Bag Co., 210 U.S. 405, 28 S.Ct. 748, 52 L.Ed. 1122 (1908); Phillips, supra; Sterner Lighting, Inc. v. Allied Electrical Supply, Inc., 431 F.2d 539 (5th Cir. 1970), cert. denied, 401 U.S. 909, 91 S.Ct. 869, 37 L.Ed.2d 807 (1971). The Court finds that the SGK patent neither requires preformation nor excludes in situ formation and that the Dart catalyst falls within the equivalents of the '115.¹⁴⁰ c. Liquid Propylene as Suspension Medium

140 The Court also notes that, in one sense, the Dart catalyst is pre-formed in that, before it is contacted with proplyene, TiCl₄ is reduced by aluminum powder to produce the -TiCl₃ that Dart buys from Stauffer. This pre-contact reduction can be loosely analogized to the reduction of TiCl₄ by DEAC before contacting with propylene as taught by the '115. See note 96, supra.

Dart's commercial process uses liquid propylene as both monomer source and suspension medium. The liquid propylene that comes into contact with the active site of the catalyst is polymerized. The unreacted liquid propylene suspends the polymer particles, holding them in solution. T-2945-46. By contrast, the '115 prefers suspension media of aromatic hydrocarbons, but also mentions aliphatic hydrocarbons¹⁴¹ which would include propylene. Thus, propylene as a diluent, while not explicitly mentioned, is contemplated by the '115.

141 See DX-1, Col. 10, lines 32-39.

5. Summary

Despite the additions and improvements of which Dart, largely through Stauffer, has availed itself, the evidence demonstrates that the fundamental polymerization chemistry of the Dart process is that of the '115. The evidence establishes that the '115 contemplates and is capable of the production of isotactic polypropylene through the interaction of titanium tetrachloride, diethyl aluminum chloride and monomeric propylene. The valence of the titanium halide is reduced to 3 and, in further interaction with DEAC, an active catalyst is formed. Dart's process appropriates the essence of the '115 system. It uses a trivalent titanium chloride and diethyl aluminum chloride to construct a catalyst of sufficient activity and stereospecificity to produce isotactic polypropylene. Thus, the Dart catalyst performs essentially the same function, the polymerization of propylene, in essentially the same way, catalysis at an active site created by the interaction of trivalent titanium chloride and diethyl aluminum chloride, to achieve substantially the same result, the production of commercially useful, e.g., isotactic, polypropylene. See Lockheed, supra, 553 F.2d at 80; see also Phillips, supra, 483 F.2d at 879.

After a careful review of all the factual and theoretical evidence presented, the Court finds that Dart's process infringes the '115 patent.

III. LACHES AND ESTOPPEL

Laches and estoppel are equitable doctrines enforced in patent cases because the federal patent statutes prescribe no specific period of limitations for a patentee's initiation of an infringement suit. As a consequence of their basis in equity, the appropriateness of the defenses will be determined in each individual case under its particular facts and circumstances. Wayne-Gossard Corp. v. Sondra, Inc., 434 F.Supp. 1340, 1361 (E.D.Pa. 1977), aff'd, 579 F.2d 41 (3d Cir. 1978). Furthermore, the decision regarding the applicability of either laches or estoppel rests predominantly with the trial judge in the exercise of his or her sound discretion. See Studiengesells-chaft Kohle (SGK) v. Eastman Kodak Co., 616 F.2d 1315 (5th Cir.), cert. denied, 449 U.S. 1014, 101 S.Ct. 573, 66 L.Ed.2d 473 (1980); Advanced Hydraulics, Inc. v. Otis Elevator Co., 525 F.2d 477 (7th Cir.), cert. denied, 423 U.S. 869, 96 S.Ct. 132, 46 L.Ed.2d 99 (1975); Baker Mfg. Co. v. Whitewater Mfg. Co., 430 F.2d 1008 (7th Cir. 1970), cert. denied, 401 U.S. 956, 91 S.Ct. 978, 28 L.Ed.2d 240 (1971); Johnson Johnson v. W.L. Gore Assoc., Inc., 436 F.Supp. 704 (D.Del. 1977).

A. Laches

To make out a defense of laches, a defendant must establish: (1) that the delay in bringing suit was unreasonable or inexcusable; and (2) that the defendant has been prejudiced by the delay. Advanced Hydraulics, supra; Jenn-Air Corp. v. Penn Ventilator Co., 464 F.2d 48 (3d Cir. 1972). The burden of so demonstrating rests with the defendant asserting the defense. Simply to establish that there was a delay is not adequate.

In determining what constitutes unreasonable or inexcusable delay, courts have frequently turned to 35 U.S.C. § 286, which restricts the period for which a party may collect damages to the six years prior to the commencement of suit. While the six year period of § 286 provides a useful frame of reference, the Third Circuit, in contrast with some of the other circuits,¹⁴² has not unequivocally taken the further step of shifting from defendant to plaintiff the burden of showing reasonable or excusable delay or lack of prejudice if the suit was not commenced within six years of the date by which plaintiff knew or should have known that defendant's activity infringed. Minnesota Mining and Mfg. Co. v. Berwick Industries, Inc., 532 F.2d 330 (3d Cir. 1976); Johnson Johnson, supra, 436 F.Supp. at 733, 733 n. 56. Recent Third Circuit cases in areas other than patent law, however, clearly indicate that the burden does shift to plaintiff if he "sleeps on his rights for a period of time greater than the applicable statute of limitations." University of Pittsburgh v. Champion Products, Inc., 686 F.2d 1040 (3d Cir. 1982); see Gruca v. U.S. Steel Corp., 495 F.2d 1252, 1258-59 (3d Cir. 1974). The Court finds it unnecessary to resolve this issue since it finds that the period during which SGK was chargeable with knowledge of Dart's probably infringement and yet took no action was substantially less than six years. In this case, the burden of persuasion remains with Dart.

142 See, e.g., SGK v. Eastman Kodak, 616 F.2d 1315, 1325-30 (5th Cir. 1980); TWM Mfg. Co. v. Dura Corp., 592 F.2d 346 (6th Cir. 1979); Continental Coatings Corp. v. Metco, Inc., 464 F.2d 1375 (7th Cir. 1972).

An essential component of an unreasonable delay is the idea that plaintiff has "slept on his rights" to the detriment of the defendant. Consequently, plaintiff must be aware of an infringer's activities before any period of delay starts to run. The standard of notice applied in this circuit is that of Johnston v. Standard Mining, 148 U.S. 360, 370, 13 S.Ct. 585, 589, 37 L.Ed. 480 (1893):

[W]here the question of laches is in issue, the plaintiff is chargeable with such knowledge as he might have obtained by inquiry, provided the facts already known by him were such as to put upon a man of ordinary intelligence the duty of inquiry.

See Minnesota Mining Mfg. Co. v. Berwick Industries, Inc., 373 F.Supp. 851, 863 (M.D.Pa. 1974), aff'd, 532 F.2d 330 (3d Cir. 1976).

Dart argues that the period of delay should be measured as running from May, 1964, when it broke off licensing negotiations with Montecatini that included Montecatini's repeated offer of a license¹⁴³ under Ziegler's '115, until July, 1970, when SGK filed this infringement suit. Dart asserts a delay of six years and two months.

143 Under a pooling arrangement between Ziegler and Montecatini, Montecatini was the authorized licensing agent for the pool in the United States until 1969. See T-608, 613; DX-408, 409, 445.

SGK contends that it had no reason to believe that Dart was infringing Ziegler's '115 patent until it was so informed by its licensee, Hercules, in 1969.¹⁴⁴ SGK argues that until that time it had neither actual notice nor a duty to inquire. It further contends that it immediately embarked on a path of offer, negotiation and litigation to enforce its rights under '115.

144 PX-64.

The '115 issued in December of 1963. Dart's Odessa, Texas, plant went into operation in February 1964. By December, 1964, Ziegler had negotiated a license agreement with Shell Oil that extended to crystalline polypropylene under '115.¹⁴⁵ Novamont, Montecatini's United States subsidiary, took a '115 polypropylene license in December, 1967.¹⁴⁶ SGK has persuasively argued that, given the breadth, value and rapid wide acceptance of his '115 and '332 patents,¹⁴⁷ Ziegler rationally relied upon his licensees to police possible infringements. Given the fact that Dart was producing approximately twenty-seven million pounds of polypropylene per year after 1964, it seems unlikely that its activities could go unnoticed by the entire range of Ziegler licensees. Consequently, SGK's contention that Ziegler did not know until 1969 that Dart was producing polypropylene¹⁴⁸ strains the Court's credulity.

145 PX-21.

146 PX-11.

147 The two agreements cited above do not exhaust the group of Ziegler licensees in the United States. Other companies licensed for polypropylene production under the '115 included: Hercules, Atlantic Richfield (ARCO), Esso (Std. N.J., now Exxon), American (AMOCO), Diamond Shamrock and Phillips Petroleum.

148 Plaintiff's Opening Brief at 63.

It is plausible, however, that Ziegler was not aware of the specifics of the Dart process or of the catalyst it used, and so might not have been able to determine, in the face of Dart's representations that it did not require a '115 license, whether Dart infringed his '115 catalyst patent. The Court finds that knowledge merely that Dart was producing polypropylene would not initiate a period in which a reasonable catalyst patentholder's failure to prosecute for infringement could justifiably be labeled "delay."

1. Publically Available Information a. Ash analysis

Dart has suggested that SGK could have obtained samples of Dart's polypropylene and, upon discovering traces of aluminum and titanium, based an infringement suit on the combination of that evidence and widely publicized information about the volume of Dart's commercial production of polypropylene. Dart further argues that Ziegler actually brought suit against Avisun in Norway and Switzerland based on as little or perhaps even less knowledge of Avisun's catalyst as ash tests might have shown about Dart's catalyst.

While Dr. Martin admitted that the European suits against Avisun were initiated with very little knowledge of the particulars of Avisun's catalyst, T-4025-28, he also stated that the nature of the patent under which Ziegler and Montecatini were suing, at least in Switzerland, was different than the '115. The Swiss patent apparently protected the catalyst only in combination with protection of the product; there was no separate patent for the catalyst. T-4027. Thus the grounds for bringing suit in Europe could have been rather different than they would be under the United States '115.

Moreover, Martin testified that, upon being notified by Hercules of Dart's possible infringement, SGK actually performed ash analysis tests on samples of Dart polypropylene. The results were inconclusive. They showed trace amounts of a number of metals. While aluminum and titanium were among the trace elements detected, they were not present in sufficient concentrations relative to other metallic elements to allow any conclusion about the nature of Dart's catalyst. T-3805-06.¹⁴⁹ The Court finds that the mere availability of Dart polypropylene to SGK sometime before 1969 or 1970 does not rise to the level of notice that would have raised in Ziegler or SGK a suspicion of infringement. Moreover, Dart did not prove the date from which Ziegler should have been held aware of the content of Dart's product.

149 Martin's further testimony on the analysis of ash product at T-4103-05 suggests that SGK's inability to detect catalyst components may have resulted from improvements between 1954 and 1970 in commercial processes for de-ashing. The testimony does not indicate when, if ever, before 1970 SGK might have been able to take a sample of Dart polypropylene and detect such disproportionate trace concentrations of aluminum and titanium to permit an inference regarding Dart's catalyst.

b. Erchak Patent

Dart also argues that a conscientious patentee would have been aware of a number of United States patents assigned to Dart after February, 1964. Of these, only U.S. Patent No. 3,225,041 issued to Erchak¹⁵⁰ is problematic. The Erchak patent issued in December, 1965. It deals with industrial-scale processes for polymerizing -olefins. Its major departure from the prior art in the field is to teach the possibility of using lower alkanes ( e.g., propane or butane) as inert diluents. Earlier processes relied primarily on heptane or hexane as diluents in the catalysis reaction.

150 DX-441 (hereinafter "'041" or "Erchak patent").

The other major teaching of the '041 patent is its system design which, allows a smoothly continuous output of the desired polymer product. Basically, the patent represents an improvement in efficiency and, hence a cost savings. It does not teach or directly relate to a particular catalyst system, or the chemistry of olefin polymerization.

In the course of describing its teachings and claims, however, the Erchak patent repeatedly mentions a "preferred" catalyst. Moreover, it explicitly mentions titanium tetrachloride (TiCl₄), aluminum trialkyl (ATA), diethyl aluminum monochloride (DEAC) and titanium trichloride (TiCl₃). Column 6, lines 7-9 of the Erchak patent state that "[t]he preferred catalyst is a titanium trichloride activated with diethyl aluminum monochloride or with an aluminum trialkyl and diethyl aluminum monochloride." This statement, if it had come to the attention of Ziegler or his associates at the Institute, should have put them on notice that assignee Dart "preferred" a catalyst that infringed Ziegler's interpretation of his '115 patent. In addition to the patent's description of a preferred catalyst, it explicitly refers to "catalysts heretofore described by Ziegler, Natta and others in the prior art. . . ."¹⁵¹

151 DX-441, Col. 5, lines 70-71.

SGK's response to the existence of this patent is that it never came to the attention of personnel at the Institute and that, even if it had, the patent is not so specific as to lead a reader to suspect actual practices under it that would infringe. The '041 may be a "paper patent."

Martin testified that while the people at the Institute generally followed the progress of patents in the area of olefin polymerization, they concentrated on patents and literature relating to polymerization chemistry. Martin described the Erchak patent as one that improves the technology of polymerization and so as one that would not be examined by scientists at the Institute. T-4042.

By early 1966, when the Erchak patent could reasonably have been expected to have become available to the Institute's scientists, Montecatini had already commenced suit against Dart for infringement of that Natta polypropylene patent. In light of that fact, the Court finds that the '041 patent to Erchak, with its mention of Ziegler and of the components not only of Dart's preferred catalyst, but also the TiCl₄ so basic to Ziegler's research and patents, raised in Ziegler and the Institute the duty to inquire further into Dart's activities relating to polypropylene. Until early 1966, Ziegler did not have sufficient notice to require further inquiry or for the Court now to impute to him "such knowledge as he might have obtained upon inquire." Johnston, supra, 148 U.S. at 370, 13 S.Ct. at 589.

2. Phillips Suit

Ziegler brought suit against Phillips in September, 1967. The original complaint, however, did not include counts for infringement of the '115 but, rather, was restricted to Ziegler's '332 patent. Ziegler amended his complaint in August, 1969, to include allegations that Phillips infringed the '115.

From the outset, Dart officials watched the Phillips litigation with particular interest. Fred S. Valles, patent counsel for Dart, testified that he became aware of the litigation shortly after it was commenced in 1967. T-626. Valles also testified that he had noted that the Phillips case was the first opportunity Ziegler had had to litigate the '115 and had chosen not to assert it against Phillips. T-627. Valles was aware of when the Phillips complaint was amended to allege infringement of the '115. T-627-28. Shortly after the '115 counts were added to the complaint against Phillips, Ziegler wrote directly to Dart offering a license under the '115.¹⁵²

152 DX-445t; letter from von Kreisler to Dart Industries (Sept. 20, 1969) (offering licensing negotiations between Ziegler and Dart).

Based on the testimony of Mr. Valles, the Court finds that Dart had actual notice that Ziegler considered a catalyst system consisting, at least in part of DEAC and TiCl₃ to be covered by his '115 and that he intended to enforce his claims by litigation, Dart, which must have been aware of the composition of its own polypropylene catalyst, could not but interpret the September 20, 1969, letter to indicate that it too might be subject to suit.

Patentees are not required to pursue simultaneously every party suspected of infringing a particular patent. Such a requirement would unduly burden both patentholders and courts. Consequently, courts have recognized the "other litigation" excuse to an infringer's laches defense. See, e.g., Kodak, supra; American Home Products Corp. v. Lockwood Mfg. Co., 483 F.2d 1120 (6th Cir. 1973), cert. denied, 414 U.S. 1158, 94 S.Ct. 917, 39 L.Ed.2d 110 (1974); Jenn-Air, supra; Clair v. Kastar, Inc., 148 F.2d 644 (2d Cir.), cert. denied, 326 U.S. 762, 66 S.Ct. 143, 90 L.Ed. 459 (1945).

While continuing to recognize the other litigation excuse, a number of courts have recently focused on the degree of notice a patentholder must provide to infringers he chooses not to sue immediately. Some circuits now require a patentholder directly to inform the suspected infringer of the patentholder's position on infringement and make clear that he intends to bring a subsequent action or, at least, reserves the right to do so.¹⁵³

153 See Kodak, supra, 616 F.2d at 1327-28 ("[W]here the plaintiff is engaged in other litigation involving the patent, to escape a defense of laches, he must at least inform the potential infringer of his intent to pursue his rights under the patent."); Advanced Hydraulics, supra, at 481, 525 F.2d ([N]otice of `other litigation' must be given to all known parties who are thought to be infringers; otherwise manifest injustice would result."); American Home Products, supra, 483 F.2d at 1123 (duty to "at least assert to the other infringers its intention to bring a subsequent action at the termination of the presently pending action.").

The Third Circuit has not required that other infringers' notice come directly from the litigating patentholder. Instead, this circuit requires simply that the infringer have actual notice of the other litigation.¹⁵⁴ Notice of intent to enforce patent claims eliminates the possibility of innocent reliance upon silence, non-prosecution or delay as between the patentowner and the infringer now asserting laches. The source of that notice is irrelevant to the character that may be imputed to the infringer's post-notice actions.¹⁵⁵ Material changes in position subsequent to notice of other litigation, from whatever source, should not, then, be asserted before a court sitting in equity as part of an infringer's laches defense. Once on notice, the infringer accepts the risk of his subsequent decisions.

154 See Jenn-Air, supra; Berwick Industries, supra, 373 F.Supp. at 867: ("In those cases applying the `other litigation' principle as an excuse for delay in bringing an infringement action, the plaintiff had either given defendant prior notice of the patents, [citations omitted] or the defendant was actually aware of the other litigation involving the patent.").

155 In fact, recognizing as "prejudicial" all detrimental changes in position prior to notification directly from a patentowner would unfairly reward such actions whenever actual knowledge preceded receipt of the patentowner's notice.

The Court finds that as of late August, 1969, Dart has actual notice that the '115 had become part of the Phillips suit and was in a position to know that it too might be sued. Von Kreisler's September, 1969, letter was adequate, combined with Dart's knowledge, to meet any fair standard of notice to Dart about Ziegler's intent to pursue his '115 rights. See T-3205-06. This suit was initiated nearly a year before the Northern District of Texas decided the Phillips case, which was subsequently appealed, so there can be no allegation of laches after that period.

SGK is entitled to assert the "other litigation" excuse against Dart's laches defense. Consequently, the Court finds that a period of approximately three years and eight months actually elapsed between the time that Ziegler or his agents knew or should have suspected Dart's infringement of the '115, and the time that Dart was notified of Ziegler's intention to enforce his claims. That period began in January, 1966, when the Erchak patent could reasonably have been expected to have come to the attention of an Institute employee in Mulheim. It ended with Dart's recognition that the Phillips complaint had been amended, in August, 1969, to include '115 claims.

3. Prejudice to Dart a. Unavailability of Witnesses

Dart argues that the August, 1973, deaths of Ziegler and von Kreisler have prejudiced its ability to prepare its case. However, both men were alive for three years after this suit was commenced. Moreover, Professor Ziegler was exhaustively deposed in Montecatini v. Dart, Consol.Civ.Action No. 3343. That deposition testimony has been read into the record here.¹⁵⁶ Dart has not adequately demonstrated, under the facts of this case, that the absence of any witness is traceable to the three year delay the Court finds may be attributable to Ziegler or SGK. Dart argues that in Kodak, supra, the district court found and the circuit court affirmed that absence of key witnesses was prejudicial to Kodak. That case, however, was distinguishable in that the suit was not filed until 1974, after Ziegler and von Kreisler had already died. b. Expenditures on Plant Expansion

156 DX-508.

A cursory examination of Dart's trial exhibits relating to expansion during the period between January, 1966, and August, 1969, indicates that the company did, in fact, commit significant resources to the expansion of the Odessa polypropylene plant.¹⁵⁷ Dart argues that the commitment of those resources is sufficient proof of prejudice under the laches doctrine.

157 See DX-459-466. Review indicates that Dart authorized the expenditure of over $5½ million between January, 1966, and August, 1969.

SGK contends, beyond arguing that its delay was reasonable and excusable, that Dart would have undertaken the same course of expansion regardless of the assertion or nonassertion of Ziegler of '115 claims. SGK's argument hinges on two lines of evidence. First, the observed fact that a substantial part of the investment Dart claims to prove prejudice was made during the pendency of Montecatini's suit against Dart for infringement of the Natta '300 patent on crystalline polypropylene. That suit was commenced in June, 1965, and was settled in February, 1975. Thus, Dart proceeded with expansion plans in the face of a risk that it would be held to be infringing Montecatini's '300 patent on the product that the Odessa plant was turning out.

Second, SGK argues that the deposition testimony of Ralph M. Knight, then president and chief executive officer of Dart, indicates that even if Ziegler had sued Dart in the mid-1960s, Dart would probably not have changed its pattern of expansion. Knight has stated on the record both that Dart approved expansion during the time it was being sued by Montecatini, T-3207, and that "inasmuch as we were in the business, we were committed to the business, I probably, although I couldn't be absolutely certainly [sic], I probably would have recommended the same course of action, the expansion," T-3209, even if Ziegler had brought suit.

The Court finds that while this evidence is not conclusive of the absence of prejudice, it casts significant doubt on the characterization Dart now seeks to give its acts. Dart's actions in the face of the Montecatini suit and its evaluation of what it would probably have done even if Ziegler had sued, undercut Dart's attempted demonstration of prejudice. The evidence recited above casts doubt on whether Dart did in fact rely on Ziegler's or SGK's silence in making its expansion decision. If Dart would have behaved in exactly or substantially the same way regardless of notice from or suit by Ziegler, it can hardly have been prejudiced by the absence of such information. The evidence does not establish that Dart would actually have pursued the same course of expansion regardless of Ziegler's action, but, in light of the burden on Dart to prove both elements of its laches defense, Knight's testimony diminishes Dart's best evidence of prejudice.

4. Finding

The Court finds that Dart's defense of laches fails. There was no delay of six or six and one-half years. There were only three years and eight months in which Ziegler knew or should have known that Dart infringed the '115 and did not initiate an infringement action. The delay in suing or directly notifying Dart during that relatively short period was not unreasonable or inexcusable. While Dart did expand its polypropylene plant during the relevant period, the actual prejudice it suffered thereby was not clearly proven.

In sum, this Court, evaluating an equitable defense, finds that Dart has failed to prove "that which is decisive for the chancellor's intervention, namely, whether plaintiff has inexcusably slept on his rights so as to make a decree against the defendant unfair." Holmberg v. Armbrecht, 327 U.S. 392, 396, 66 S.Ct. 582, 584, 90 L.Ed. 743 (1946), quoted in Jenn-Air, supra, 464 F.2d at 50.

B. Estoppel

A defendant asserting estoppel in a patent case must prove not only unreasonable or inexcusable delay, but, further, misleading conduct by the patentowner upon which the infringer relied. See Kodak, supra, 616 F.2d at 1330; Berwick, supra, 373 F.Supp. at 869. Dart has failed to demonstrate that Ziegler or SGK engaged in any conduct, including silence, which would lead a reasonable observer in Dart's position to conclude that Ziegler's claims under the '115 had been abandoned or that Dart's business would be unmolested. "For silence to work an estoppel, some evidence must exist to justify an inference that the silence was sufficiently misleading to amount to `bad faith'." TWM Mfg. Co. v. Dura Corp., 592 F.2d at 350 (footnote omitted). Dart has already been found not to have established delay supportive of a laches defense. It has also failed to adduce evidence that would persuade this Court that Ziegler's or SGK's silence constituted bad faith.

IV. CONCLUSION

Based on its review of all the evidence and for the reasons stated above, the Court finds United States Patent No. 3,113,115 to be valid, infringed and enforceable. The Court orders the parties to file proposed orders in accordance with this Opinion by October 18, 1982.