740 F. Supp. 305 | D. Del. | 1990
GENERAL ELECTRIC COMPANY, Plaintiff,
v.
HOECHST CELANESE CORPORATION and Celanese Engineering Resins Inc., Defendants.
United States District Court, D. Delaware.
*306 *307 David A. Anderson, of Potter, Anderson & Corroon, Wilmington, Del. (William F. Kilgannon, William J. Hone, and Stanley L. Amberg, of Davis, Hoxie, Faithfull & Hapgood, New York City, of counsel), for plaintiff.
Howard M. Handelman, of Bayard, Handelman & Murdoch, Wilmington, Del. (John F. Lynch, Michael Macklin, Glenn W. Rhodes, and Melinda L. Patterson, of Arnold, White & Durkee, Houston, Tex., of counsel), for defendants.
OPINION
ROTH, District Judge.
General Electric Company ("GE") filed this action alleging infringement of U.S. Patent No. 3,953,394 ("the GE patent") against Hoechst Celanese Corporation ("HCC") and Celanese Engineering Resins, Inc. ("CER"). HCC and CER have each counterclaimed for a declaration of non-infringement, invalidity and unenforceability of the GE patent.
On November 9, 1988, the Court ruled on two summary judgment motions which had been filed by the defendants, holding that claims 1-37, 39 and 40 were not infringed by HCC, that claims 31-37, 39 and 40 were invalid, and that the remaining assertions could not be resolved on summary judgment. See General Electric Co. v. Hoechst Celanese Corp., 698 F. Supp. 1181 (D.Del.1988) (hereinafter "General Electric I").
Presently before the Court are three new summary judgment motions filed by the defendants regarding (1) invalidity of claims 1-4 and 38 of the GE patent under 35 U.S.C. § 102(b); (2) invalidity of all claims on the grounds of undue breadth, which renders the claims invalid under 35 U.S.C. §§ 102, 103 or § 112; and (3) nonliability on the grounds of laches.
FACTS
The following undisputed facts of record are relevant to all three summary judgment motions.
The GE Patent
GE is the owner of U.S. Patent No. 3,953,394, entitled "Polyester Alloys and Molding Compositions Containing the Same." GE applied for the GE patent on November 15, 1971. The patent was issued in 1976 and reexamined in 1986-1987. GE's original patent described:
thermoplastic, stable, blended compositions comprising in combination
a. from about 1 to about 99 parts by weight of a poly(ethylene terephthalate) resin [referred to as "PET"] and
b. from about 99 to about 1 part by weight of a poly(1,4-butylene terephthalate) resin [referred to as "PBT"] or a copolyester thereof with a minor amount of an aliphatic or aromatic dicarboxylic acid or an aliphatic polyol.
(D.I. 111A at A3) PET and PBT are both polymers, known as polyesters, which contain an ester functional group as part of the repeating unit in the polymer. In PET, the repeating unit is ethylene terephthalate; in PBT, the repeating unit is butylene terephthalate.
*308 PET and PBT can be combined in different ways. Under some conditions a mixture of PET and PBT results in a "copolyester" (also referred to as a "copolymer"). A copolyester of PET and PBT has both ethylene terephthalate units and butylene terephthalate units chemically combined in a single polymer chain. One method of forming a copolyester of PET and PBT is to combine the two copolymers in a melt and allow a reaction ("transesterification") to proceed. As a result of the transesterification reaction, polymer chains having both "ET" and "BT" units are formed. Alternatively, PET and PBT can be combined in a blend or mixture in such a way that a transesterification does not occur. One method of achieving this is to combine PET and PBT in a melt, as is done when forming a copolymer, but to retard or prevent the occurrence of a transesterification by adding a phosphorous compound to the melt at or before mixing. (D.I. 111A at A30 ¶ 8)
The novelty and objective of the compositions claimed by the GE patent is explained in the patent specification as follows:
Because [PET] is known to crystallize only very slowly from the melt and [PBT] is known to crystallize very rapidly from the melt, in view of the above, it would be entirely unexpected to find that blends of these two resins prove to be highly compatible both on the macro and molecular scale. In other words, these two polyester resins, which should be incompatible on the basis of the wide difference in their rates of crystallization, have, in fact, been discovered to form a stable alloy.
. . . . .
It is an object of this invention to improve the moldability of [PET] while improving the properties of [PBT] and to provide compositions having many properties improved over those compositions containing either resin alone.
(Id. at A2, col. 2, lines 17-27, 63-69)
The Reexamination
The GE patent was reexamined in 1986-1987. The Patent Examiner ordered reexamination in light of British Patent No. 1,060,401, entitled "Method of Manufacturing Stable Block Copolyesters and Polyester Mixtures," [hereinafter "Kurashiki"] and Belgian Patent No. 747,243, entitled "Elastic Polyester Fibers." [hereinafter "Fiber"] (D.I. 111A at A35, A41) Kurashiki was published on March 1, 1967 and Fiber was published September 14, 1970.
The invention of Kurashiki relates to "stable compositions containing block copolyesters or polyester mixtures containing at least two polyesters." (D.I. 111A at A35, lines 9-11) The principal object of the Kurashiki invention, as stated in the patent, "is to manufacture stable block copolyesters ... or polyester mixtures which can exist as stable mixtures without causing chemical combination of the two or more polyesters." (Id. at A35 lines 16-26) A further object of the Kurashiki invention is to "provide a method of manufacturing ... shaped articles from block copolyesters or polyester mixtures having improved ... mouldability...." (Id. at A35 lines 27-31) Kurashiki recognizes that "it is difficult to obtain a polyester mixture wherein two or more than two kinds of polyesters are not chemically combined with each other owing to the fact that there occurs the ester interchange reaction when the mixture is heated." (Id. at A35 lines 63-68) Yet Kurashiki discloses that this difficulty can be overcome: "[I]f two or more than two kinds of polyesters are heated in the presence of the A-material [a phosphorous-containing compound which acts as a catalyst inhibitor], a substantially stable block copolyester or a polyester mixture wherein two polyesters are not substantially chemically combined together with each other can be obtained." (Id. at A36 lines 18-26) Example 4 of the Kurashiki patent describes a combination of 78 parts PET by weight to 22 parts PBT. (Id. at A38 lines 35-46)
The British Patent Office issued an abridgment of the Kurashiki patent in 1966-1967. The abridgment discloses mixtures of PET and PBT, and states that "[a] polyester mixture or block copolymer is stabilized by adding ... a phosphorous compound which retards or prevents esterinterchange." *309 (D.I. 216A at A19) Fiber describes a process whereby 50 parts PET and 50 parts PBT are combined and extruded through an annular nozzle at an outlet temperature of 275°C. (D.I. 111A at A59)
On reexamination, GE argued to the Examiner that the GE patent was distinguishable from these prior art references because the references claimed compounds that contained copolymers of PET and PBT, rather than a blend of substantially unreacted PET and PBT as called for in the GE Patent. (Id. at A81) The Examiner did not accept GE's argument. He noted that the terminology of the GE patent was not limited to "blends," and that blends of PET and PBT made under the conditions of GE's original patent inevitably contain small amounts of block copolymer. (D.I. 152A at A76-A77) The Examiner concluded that:
The claim terminology, when reasonably interpreted, encompasses blended compositions which may comprise various components, including blocked [copolymer] material, as well as the two recited linear polymers. The other components may be in major amounts. The composition blended with all components is a blended composition. Hence, a blended composition as encompassed by the claims may include major proportions of block polymeric materials.
The Examiner rejected the GE patent on November 17, 1986, "under 35 U.S.C. 103 as being unpatentable over Kurashiki and Fiber...." (Id. at A75)
Shortly thereafter, representatives of GE met with the Examiner to discuss limiting the GE claims. GE proposed amending its claims from "comprising" PET and PBT, to "consisting essentially of" PET and PBT. (Id. at A86)[1] This change in claim terminology would avoid prior art by eliminating copolymer from the claimed blend. The Examiner requested that "objective evidence and discussion would be necessary to ... show that what is being eliminated [block copolymer] would alter the basic nature of the compositions desired to be claimed." (Id. at A82) Thus, the Examiner requested GE to narrow its claims and to show that the compositions covered by the narrowed claims had basic and novel characteristics different from mixtures which contained the prior art copolymer along with the linear polymers.
GE amended its claims to incorporate the "consisting essentially of" terminology, and conducted an experiment which compared the moldability of a PET/PBT copolymer set out in Kurashiki with that of GE's claimed "blend." GE submitted the following summary of its test results in support of its amended claims:
[S]ample A-C bars were made by injection molding at a temperature of 170°F with a mold cycle of 30 seconds to produce a bar one quarter inch (¼") thick.
In contrast to the uniformly milky, nondistorted bar of sample C, which bar was prepared from the claimed blend of PET and PBT, Samples A [Kurashiki] and B [Kurashiki/GE mix] were badly distorted and non uniform in that the edges of the bar are semi-transparent and milky only in the center.
. . . . .
It is urged that the experimental evidence submitted herewith overwhelmingly establishes that a copolymer of PET and PBT present in an amount such as in samples A and B produces a molded product that is unsatisfactory and thus constitutes an alteration of the basic nature of the claimed compositions.
(D.I. 152A at A87-A88) [hereinafter referred to as "the ABC experiment"] Table I summarizes the ABC experiment.
*310
Table I
ABC Experiment
Barrel Temperature: 500°F
Mold Temperature: 170°F
Cycle Time: 30 sec.
========================================================================
| Sample Description PET Result |
| Content |
| (%) |
|========================================================================|
| A 100% prior art 78 Distorted, rubbery, |
| "Kurashiki" stuck in mold |
|------------------------------------------------------------------------|
| B 90% prior art 75 Distorted, rubbery |
| "Kurashiki" |
| 10% of 50/50 PET/PBT |
| blend |
|------------------------------------------------------------------------|
| C PET/PBT blend 38.5 Uniformly milky, non |
| distorted |
========================================================================
On March 18, 1987, the Examiner confirmed GE's amended claims:
The claims have been amended, narrowed, to exclude any materials which would alter the basic nature of the compositions claimed. It appears from the affidavit of W. F. H. Borman, submitted in support of the amendments, that at least major amounts of block copolymeric material present would alter the basic nature of the blends claimed.
(D.I. 152A at A96)[2]
The German Experiment
In January, 1986, approximately one year before the reexamination of the GE patent, GE was prosecuting the German counter-part to its patent. In connection with the prosecution of the German patent, Dr. Borman performed an experiment ("the German experiment") which compared the moldability of blends, containing different ratios of PET and PBT, in which there was *311 no block copolymer. (D.I. 109A at A109) Dr. Borman related the results of the German experiment to GE counsel in a letter dated January 21, 1986:
[T]he blends containing 25 and 35% PET yielded well crystallized parts that released easily from the mold without sticking or distortion. The composition containing 55% PET occasionally stuck in the mold and showed distortion as molded. The composition containing 75% PET stuck badly in the mold and was severely distorted due to the rubbery nature of the product as it exited the mold.
(D.I. 109A at A109) Table II summarizes the German experiment.
Table II
The German Experiment
Barrel Temperature: 500°F
Mold Temperature: 170°F
Cycle Time: 30 sec.
=========================================================================
| Sample Description PET Result |
| Content |
| (%) |
|=========================================================================|
| A PET/PBT blend 25 OK |
|-------------------------------------------------------------------------|
| B PET/PBT blend 35 OK |
|-------------------------------------------------------------------------|
| C PET/PBT blend 55 Distorted, stuck in mold |
|-------------------------------------------------------------------------|
| D PET/PBT blend 75 Distorted, rubbery, |
| stuck in mold |
=========================================================================
The German experiment shows that under certain conditions, high concentrations of PET in blended compositions produce an unsatisfactory mold product.
Other References
In 1972 GE filed a U.S. patent application entitled "Reinforced Rapidly Crystallizable Copolyester Resin Compositions." (D.I. 152A at A204) (hereinafter "Floryan application") The Floryan application, which GE subsequently abandoned, was concerned with injection molding compositions that contained PET/PBT copolymers as opposed to blends. The Floryan application discloses that PET/PBT copolymers could contain up to 20 mole%[3] of PET without substantially lowering the fast crystallization rate associated with the PBT polymer. (D.I. 152A at A207)
In 1971 a U.S. patent was granted,[4] entitled "Flame Retardant Poly(tetramethylene Terephthalate) Molding Compositions." (D.I. 152A at A157) (hereinafter "Caldwell") Caldwell discloses a PET/PBT copolymer containing 10 mole% PET and 90 mole% PBT.[5] This composition is described as having "[d]esirable molding properties." (D.I. 152A at A161, col 7, lines 33-34) The Caldwell application was filed February 24, 1970, more than one year before the GE application.
Both Floryan and Caldwell show that under certain conditions, PET/PBT copolyesters exhibit the same molding characteristics as PET/PBT "blends."
*312 DISCUSSION
I. Summary Judgment Standard
Summary judgment is appropriate under Fed.R.Civ.P. 56(c) "if the pleadings, depositions, answers to interrogatories, and admissions on file, together with the affidavits, if any, show that there is no genuine issue as to any material fact...." The Court reviewed the requirements for summary judgment In General Electric I:
Summary judgment should be granted when there are no genuine issues of material fact and the movant is entitled to judgment as a matter of law. Moeller v. Ionetics, Inc., 794 F.2d 653, 656 (Fed.Cir. 1986) (citing Fed.R.Civ.P. 56). The burden of proving that there are no genuine issues of material fact rests on the moving party. Celotex Corp. v. Catrett, 477 U.S. 317, 323, 106 S. Ct. 2548, 2553, 91 L. Ed. 2d 265 (1986). Summary judgment may be granted if there is "evidence that is merely colorable ... or [that] is not significantly probative." Anderson v. Liberty Lobby, Inc., 477 U.S. 242, 249-50, 106 S. Ct. 2505, 2510-11, 91 L. Ed. 2d 202 (1986) (citations omitted). "Factual disputes that are irrelevant or unnecessary will not be counted." Id. at 248, 106 S. Ct. at 2510. Furthermore, reliance on mere assertions in the pleadings is not permissible. Id. There must be enough evidence to enable a reasonable jury to find for the nonmoving party on the issue for which summary judgment is sought. Id. at 249, 106 S. Ct. at 2510.
... Finally, it is well settled law that evidence should be viewed in a light most favorable to the nonmoving party. Adickes v. S.H. Cress & Co., 398 U.S. 144, 157, 90 S. Ct. 1598, 1608, 26 L. Ed. 2d 142 (1970)....
Summary judgment is appropriate in patent cases. Avia Group International, Inc. v. L.A. Gear California, Inc., 853 F.2d 1557, 1561 (Fed.Cir.1988) (quoting Brenner v. United States, 773 F.2d 306, 307 (Fed.Cir.1985)). Anticipation is a finding of fact, whereas obviousness and claim interpretation are questions of law. Titanium Metals Corp. of America v. Banner, 778 F.2d 775, 780, 782 (Fed.Cir.1985). Yet claim interpretation has factual underpinnings. In General Electric I, we held that the determination of "the basic and novel characteristics" of the GE patent is part of determining the scope of the claims. 698 F. Supp. at 1187. This was a question of fact which we were unable to resolve at that juncture. However, for the purpose of the present motions, brought by HCC and CER, we will adopt GE's characterization of the "basic and novel characteristics" of its invention. Those characteristics will be discussed more fully below. With these standards in mind we will proceed to apply the evidence of record to defendants' summary judgment motions.
II. Invalidity Under 35 U.S.C. § 102(b)
A. Legal Standard
Defendants claim that claims 1-4 and 38 of the GE patent are invalid under 35 U.S.C. § 102(b). That section provides:
A person shall be entitled to a patent unless
(b) the invention was patented or described in a printed publication in this or a foreign country or in public use or on sale in this country, more than one year prior to the date of the application for patent in the United States....
Defendants assert that the GE claims are described in Kurashiki, the Kurashiki Abridgment, and Fiber. GE filed its U.S. patent application on November 15, 1971. Kurashiki was published on March 1, 1967, the Kurashiki abridgment was publicly available prior to 1968, and Fiber was published on September 14, 1970. None of these references were before the Examiner during GE's original application, but both Kurashiki and Fiber were considered by the Examiner during the reexamination.
GE's patent is presumed valid. 35 U.S.C. § 282. Defendants have the burden of proving invalidity. Defendants must show by clear and convincing evidence that the GE patent is anticipated under section 102. Tyler Refrigeration v. Kysor Industrial Corp., 777 F.2d 687, 690 (Fed.Cir. 1985). The Examiner's determination that *313 the GE patent is valid is evidence of the claims' validity, but is not binding on the Court. See Fromson v. Advance Offset Plate, Inc., 755 F.2d 1549, 1555 (Fed.Cir. 1985). As evidence of validity, the Examiner's decision increases defendants' burden on summary judgment. See Hughes Aircraft Co. v. United States, 717 F.2d 1351, 1359 (Fed.Cir.1983). Yet defendants' burden can be met. See, e.g., Surface Technology, Inc. v. United States International Trade Commission, 801 F.2d 1336, 1340 (Fed.Cir.1986) (Claims in a patent may be found invalid even though no prior art more pertinent than that considered by the PTO is presented.); Tyler Refrigeration, 777 F.2d at 690 ("The trial court did not err in finding the patent claims invalid as anticipated by a prior art reference even though the PTO examiner had considered that reference during prosecution of the patent.").
In order to anticipate a later claim, a single prior source must contain all of the essential limitations of the claim. See, e.g., Hybritech, Inc. v. Monoclonal Antibodies, Inc., 802 F.2d 1367, 1379 (Fed. Cir.1986) ("for prior art to anticipate under § 102 it has to meet every element of the claimed invention"), cert. denied, 480 U.S. 947, 107 S. Ct. 1606, 94 L. Ed. 2d 792 (1987). A later claim is anticipated by an earlier claim if the earlier claim would literally infringe the later claim if later in time. See Lewmar Marine, Inc. v. Barient, Inc., 827 F.2d 744, 747 (Fed.Cir.1987), cert. denied, 484 U.S. 1007, 108 S. Ct. 702, 98 L. Ed. 2d 653 (1988). However, "[a]nticipation is not avoided where the prior achievement was deliberate or a necessary consequence of what was intended, even though the achiever did not fully appreciate the uses, purposes, or properties of the product or process." 1 D. Chisum Patents § 3.03[1] at 3-27 (1990). This concept is known as "inherency." See generally H. Schwartz, Patent Law and Practice 46 (1988) (discussing judicially created doctrine of inherency). Finally, even if a prior publication discloses the claimed invention, "it will not suffice as prior art if it was not enabling." In re Donohue, 766 F.2d 531, 533 (Fed.Cir.1985). Therefore defendants must "show that each element of the claim in issue is found in a prior patent or publication, either expressly or under principles of inherency," Tyler Refrigeration, 777 F.2d at 689; see Verdegaal Brothers, Inc. v. Union Oil Co. of California, 814 F.2d 628, 631-32 (Fed.Cir.), cert. denied, 484 U.S. 827, 108 S. Ct. 95, 98 L. Ed. 2d 56 (1987), and that "one of ordinary skill in the art could have combined the publication's description of the invention with his own knowledge to make the claimed invention." In re Donohue, 766 F.2d at 533.
B. Claim Interpretation
A question of claim anticipation necessarily involves interpretation of the "claim." As noted above, claim interpretation is a question of law, Titanium Metals, 778 F.2d at 780, and anticipation is a question of fact. Id.; Ralston Purina Co. v. Far-Mar-Co, Inc., 772 F.2d 1570, 1574 (Fed.Cir.1985). The Court must interpret GE's claims, and separately examine each of the prior art references to conclude if any one of them anticipates.
GE's amended claims 1-4 and 38 call for a stable, homogeneous, blended composition of PET and PBT, containing no significant amount of copolymer. The Court must assume for present purposes that the basic and novel characteristics of GE's invention are "stable homogeneity and uniform crystallinity." General Electric I, 698 F.Supp. at 1186; see also D.I. 172 at 59 (GE describing the basic and novel characteristics as "the ability of the claimed compatible blend to impart uniformity of crystallinity."). Moldability is an indicium of these basic and novel characteristics. Under the GE patent the claimed PET/PBT blends can be injection molded to a uniformly crystalline test bar in from 10 to 90 seconds at from 75° to 200°F. (D.I. 111A at A8 col. 13)
GE's amended claims can be categorized into three groups based on the allowable ratios of PET to PBT. One group of claims covers compositions ranging from 1:99 to 99:1 PET to PBT. (claims 1, 2, 5-9, 12-17, 20, 21, 24-30 and 38; D.I. 152A at A11-A14). A second group of claims narrows the allowable variation in PET/PBT *314 ratio to from 10:90 to 90:10 PET/PBT. (claims 3, 10, 18, 22; D.I. 152A at A12-A13). The third group of claims narrows the allowable ratio to from 20:80 to 80:20 PET/PBT. (claims 4, 11, 19, 23; D.I. 152A at A12-A13).
1. Kurashiki
The question presented by Kurashiki is whether it discloses PET/PBT "blends" within the ranges claimed in the GE patent. Defendants argue that Kurashiki discloses every element of the claimed compositions. GE responds that Kurashiki discloses mixtures of PET/PBT copolymer, not the "stable blended composition" of the GE claims.
There is no doubt that GE's patent claims a "stable blended composition." The parties agree that this composition is a substantially unreacted mixture of PET and PBT. The parties disagree on the interpretation of "mixture" as that word is used in Kurashiki, and its relationship to the "stable blended composition" claimed by GE. The following table compares the GE claims with the Kurashiki claims.
GE Kurashiki
(Claim 1 (specification and claim 4)
A thermoplastic The principal object of the invention is to
manufacture stable block copolyesters which
do not change melting point ... when ...
heated ... or polyester mixtures.... (D.I.
111A at A35)
Stable blended composition Stable composition containing ... a polyester
[unreacted] mixture and 0.02% to 4.0% (based on the
weight of the polyesters) of a phosphorous
compound which retards or prevents ester
interchange.
Consisting essentially of Containing (D.I. 111A at A39)
1 to 99 parts PET At least 2 polyesters wherein a crystalline
99 to 1 parts PBT polyester [PET] constitutes more than 30
mole% of the total polyester. (A PET/PBT
copolymer containing 78 parts PET and 22
parts PBT is disclosed in example 4) (D.I.
111A at A34)
GE states that the Kurashiki "mixture" is not a "blend of PET and PBT ... that is compatible on the molecular level." GE argues that its claimed blends are "`stable' against physical separation into two amorphous phases when the blend was cooled from the melt and formed into an article that was `rigid'...." GE contends that the Kurashiki "mixtures" do not disclose this "stable blended composition." (Declaration of Dr. Arthur W. Anderson, D.I. 168A at A2, A5-A11). In other words, GE claims that Kurashiki does not disclose how to make melted products that are compatible blends which can be molded successfully. (D.I. 168A at A10)
Defendants counter that Kurashiki does disclose "stable blended compositions." The Kurashiki specification claims "stable compositions containing block copolyesters or polyester mixtures containing at least two polyesters and from 0.02% to 0.4% ... of a phosphorous-containing compound ... which retards or prevents ester exchange." (D.I. 111A at A35 lines 10-15) (emphasis added) The next paragraph states that "[t]he principle object of the invention is to manufacture ... polyester mixtures which can exist as stable mixtures without causing chemical combination of the two or more polyesters." (Id. lines 16-26) And the following paragraph provides: "A further *315 object of the invention is to provide a method of manufacturing ... shaped articles from ... polyester mixtures having improved ... mouldability...." (Id. lines 27-31)
We find that these passages from the Kurashiki patent, when given their ordinary meaning, disclose the same homogeneous, stable, uniformly crystalline mixtures of PET and PBT that are disclosed by GE's patent. Therefore, it is crucial whether Kurashiki's disclosure placed GE's invention in the hands of the public so as to enable one of ordinary skill in the art to make stable blended compositions of PET and PBT. Kurashiki teaches that the addition of a phosphorous-containing compound to a mixture of polyesters "retards or prevents ester exchange." (D.I. 111A at A35) (emphasis added) However, the parties do not agree whether this phosphorous compound can completely eliminate transesterification. Defendants' expert, Roger S. Porter, states that "[t]he phosphorous compounds described in Kurashiki are, as shown by the Kurashiki data and as admitted by GE, effective in preventing the chemical reaction of PET and PBT to form a copolymer." (Porter Declaration, D.I. 111A at A32 ¶ 18) GE's expert, Dr. Arthur W. Anderson states that "[t]he [Kurashiki] melting-point data show [phosphorous compounds] do not prevent chemical reaction at any of the heating times stated in Kurashiki." (Anderson Declaration, D.I. 168A at A8 ¶ 28) The experts also disagree on whether Kurashiki would have enabled, in 1971, a person of ordinary skill in the art to achieve an unreacted blend of PET and PBT. Compare D.I. 168A at A10 ¶ 39 ("In 1971 a person of ordinary skill in the art would not have found in Kurashiki a disclosure of how to melt together two polyesters to achieve, as a product, any type of unreacted mixture of those two polyesters, with or without a phosphorous compound."), with D.I. 111A at A31 ¶ 17) ("The Kurashiki reference ... unequivocally discloses and enables polyester mixtures wherein the two polyester components are not chemically combined.").
On their face these conflicting statements appear to create a triable issue of fact. But a careful reading reveals that this is not so. First, the GE expert's statement that phosphorous compounds do not prevent chemical reaction at any of the "heating times stated in Kurashiki" refers to the heating times exemplified by the Kurashiki data, not the heating times which will produce blends. Neither party disputes that the specified Kurashiki examples disclose copolymer mixtures, not "blends." Therefore, the above statements are not material to the creation of blends. More importantly, however, GE's denial of enablement flies in the face of its earlier representations to the Examiner:
prior knowledge, such as, incorporating certain phosphorous type compounds in the mix of PET and PBT to preclude copolymer formation prior to subjecting same to a melt cycle were well within the skill of the art at the time [of the GE patent] application.
(D.I. 111A at A84) (citing Kurashiki and other references)
Use of certain phosphorous compounds to preclude copolymerization was known prior to [the GE patent].
(D.I. 111A at A84 n.*) (citing Kurashiki and other references)
[A]ddition of the well-known phosphorous compounds to the mix of PET and PBT prior to the thermal treatment, ... is part of the art prior to [the GE patent] and utilizable by those skilled in the art to preclude copolymerization in the polyester field.
(D.I. 111A at A86)
The conditions of time and temperature and/or the addition of agents that tend to preclude copolymerization of PET and PBT were well known at the time of the filing of the [GE patent] application. Thus, one skilled in the art had the tools available to produce a blend or a copolymer.
(D.I. 111A at A85) (emphasis added) These admissions by GE constitute clear and convincing evidence of enablement.
The fact that Kurashiki never actually produced any unreacted mixtures of PET and PBT does not prevent Kurashiki *316 from anticipating GE. "It is not ... necessary that an invention disclosed in a publication shall have actually been made in order to satisfy the enablement requirement." In re Donohue, 766 F.2d 531, 533 (Fed.Cir.1985). Moreover, there is no evidence that Kurashiki's authors attempted to make the disclosed unreacted mixture. "[T]he fact that the author of a publication did not attempt to make his disclosed invention does not indicate one way or the other whether the publication would have been enabling." In re Donohue, 766 F.2d at 533.
GE also argues that Kurashiki's disclosure of an unreacted PET/PBT mixture is speculative. See W.L. Gore & Associates, Inc. v. Garlock, Inc., 721 F.2d 1540, 1554 (Fed.Cir.1983), cert. denied, 469 U.S. 851, 105 S. Ct. 172, 83 L. Ed. 2d 107 (1984). Gore held that "anticipation of inventions set forth in product claims cannot be predicated on mere conjecture respecting the characteristics of products that might result from the practice of processes disclosed in references." Id. Gore is distinguishable from this case because the prior references in Gore did not disclose the patented product. Here, GE's claimed "stable blended composition" is specifically disclosed by Kurashiki's "polyester mixtures which can exist as stable mixtures without causing chemical combination of the two or more polyesters." (D.I. 111A at A35)
The Examiner's statement allowing GE's amended claims indicates that he was persuaded that GE's claims were not obvious because GE's blends are not exemplified in Kurashiki:
The prior art supplied and utilized in the rejection of record are directed almost entirely to tye [sic] synthesis of block copolymeric materials. Throughout some of the material references blends are alluded to. Nowhere, however, in these references are any blends clearly set forth, exemplified or actually produced in any examples.
(D.I. 111A at A106) However, the fact that Kurashiki does not set out a specific example of an unreacted PET/PBT blend does not prevent anticipation under section 102(b) in view of the PET/PBT copolymer, set out in example 4. the described object of the manufacture of stable polyester mixtures; and GE's acknowledgment of the ability of one skilled in the art to produce a blend and to know the "conditions of time, temperature, and/or the addition of agents" to preclude copolymerization. (D.I. 111A at A86) See In re Donohue, 766 F.2d at 533, Kurashiki discloses both the product and the process necessary to make GE's claimed "blends." Kurashiki need not actually exemplify GE's composition to anticipate under section 102(b).
This case is similar in many respects to Titanium Metals Corp. v. Banner, 778 F.2d 775 (Fed.Cir.1985). Titanium Metals involved a patent claim for an alloy comprised of titanium combined with specified amounts of nickel and molybdenum. A prior Russian technical article disclosed a similar alloy, but did not discuss any of the unique corrosion-resistant properties of the alloy discovered by the patent applicant. The court was surprised that the trial court had failed to address the issue of novelty: "The patent law imposes certain fundamental conditions for patentability, paramount among them being the condition that what is sought to be patented, as determined by the claims, be new." Titanium Metals, 778 F.2d at 780. Since the applicant's claims only covered the alloy itself, the court held that "it is immaterial, on the issue of their novelty, what inherent properties the alloys have or whether these applicants discovered certain inherent properties." Titanium Metals, 778 F.2d at 782. The court reversed the district court's decision authorizing the issuance of a patent. Id. at 783.
In the final analysis GE's argument in favor of its patent is premised on the discovery of unexpected properties inherent in an unreacted mixture of PET and PBT. As GE urged during the reexamination: "That ... a solidified melt of PET and PBT would yield a homogeneous product having excellent physical properties was not known and unexpected. It is submitted that this is the `right stuff' of patentability." (D.I. 111A at A86) However, *317 inherent properties of known compositions are not patentable. Titanium Metals, 778 F.2d at 782; Friction Division Products, Inc. v. E.I. DuPont de Nemours & Co., 658 F. Supp. 998, 1010 (D.Del.1987), aff'd, 883 F.2d 1027, reported in full, 12 U.S.P.Q.2d 1575 (Fed.Cir.1989).
Kurashiki discloses unreacted mixtures of PET and PBT. Kurashiki example 4 discloses a copolymer mixture of 78 parts PET to 22 parts PBT. This mixture is within the ranges stated in GE claims 1-4 and 38. The GE claims encompass and, if allowed, would enable GE to exclude others from making, using, or selling unreacted mixtures of PET and PBT as they can be perceived in Kurashiki. The Kurashiki patent, the prosecution history of the GE patent, GE's admissions, and the expert affidavits constitute clear and convincing evidence of invalidity. GE claims 1-4 and 38 are invalid as anticipated by Kurashiki under 35 U.S.C. § 102(b).
Alternatively, even if Kurashiki does not enable one of ordinary skill in the art to produce unreacted mixtures of PET and PBT, Kurashiki's disclosure of how to achieve such mixtures is at least as great as the GE patent's disclosure. If Kurashiki lacks enabling disclosure, then the GE patent is also nonenabling. See Constant v. Advanced Micro Devices, Inc., 848 F.2d 1560, 1569 (Fed.Cir.), cert. denied, 488 U.S. 892, 109 S. Ct. 228, 102 L. Ed. 2d 218 (1988). In Constant, the court considered whether a patented computer chip was anticipated by a specification sheet for a similar, rival chip. The patent owner argued that the specification sheet did not enable one of ordinary skill in the art to use the rival chip, because it did not teach how to program the chip. The court responded that "[t]he disclosure in [the specification sheet] is at least at the same level of technical detail as the disclosure in the ... patent. If disclosure of a computer program is essential for an anticipating reference, then the disclosure in the ... patent would fail to satisfy the enablement requirement of 35 U.S.C. § 112, First ¶."[6]Constant, 848 F.2d at 1569. Therefore, if Kurashiki is nonenabling, so too is the GE patent.
2. Kurashiki Abridgment
The British Patent Office issued an abridgment of the Kurashiki patent in 1966-1967. The abridgment discloses mixtures of PET and PBT, and states that "[a] polyester mixture or block copolymer is stabilized by adding ... a phosphorous compound which retards or prevents ester-interchange." (D.I. 216A at A19) The Abridgment states that the mixture "may then be heated, e.g. for molding ... without randomization." Id. Defendants rely on the same arguments noted above to conclude that the abridgment anticipates the GE claims.
GE responds that "a person of ordinary skill in the art would have not understood the Kurashiki abridgment to contain a definite disclosure of any materials that are unreacted mixtures. That person would have understood `mixture' to refer to unheated starting materials." D.I. 168 at 24 (citing Anderson Decl., 168A at A11-A12). This argument has merit because the abridgment is reasonably open to two interpretations. It does not unambiguously disclose stable blended compositions of PET and PBT. In ruling on the present motion, the Court must view the abridgment in the light most favorable to GE. Therefore, considered in that light, the abridgment does not anticipate the GE claims. A finder of fact might of course come to a different conclusion.
3. Fiber
The Fiber reference was published September 14, 1970. GE's patent application was filed November 15, 1971. Thus the Fiber reference was published more than one year before GE's patent application.
Example 8 of Fiber describes a process whereby 50 parts PET and 50 parts PBT are combined and extruded through an annular nozzle at an outlet temperature of *318 275°C. (D.I. 111A at A59) This ratio of PET and PBT is covered by GE claims 1-4 and 38.
GE argues that examples 8 and 9 of Fiber disclose copolymers, not unreacted mixtures of PET and PBT. (Anderson Decl., D.I. 168A at A13) GE cites example 9 of the Fiber patent to support this proposition:
In following the procedure described in Example 8, copolymers are prepared with the mixtures and the physical properties indicated in Table II.
* * * * * *
As is seen from Examples 8 and 9 and Table II, copolymers of [PET] and [PBT], ... result likewise, just as the homopolymers themselves, in the desired properties of elasticity.
(D.I. 111A at A60) (emphasis added) Defendants do not deny that Fiber is concerned entirely with copolymers. Defendants argue that GE's "blend" claims are inherently disclosed by the Fiber reference because Fiber's disclosure of PET/PBT copolymers placed GE's claimed "blend" of PET and PBT in the hands of the public. In other words, one skilled in the art could have read Fiber's description of a 50/50 PET/PBT copolymer and combined that description with his own knowledge that phosphorous compounds prevent copolymerization to make GE's claimed "blend."[7]See In re Donohue, 766 F.2d at 533. This anticipation by inherency argument requires the court to look to a second reference for the knowledge of how to make an unreacted PET/PBT mixture. See Id.; 2 P. Rosenberg, Patent Law Fundamentals § 15.06[1] at 15-94 (1989 rev.); H. Schwartz, Patent Law and Practice 46 (1988). In this case the enabling reference is Kurashiki. In fact, GE cited Kurashiki to the Examiner to argue that "[u]se of certain phosphorous compounds to preclude copolymerization was known prior to Fiber...." (D.I. 111A at A84 n.*) Yet defendants' reliance on Kurashiki as evidence that Fiber anticipates the GE patent renders defendants' Fiber argument a watered down version of their Kurashiki argument: If the GE patent is anticipated by Fiber, it surely is anticipated by Kurashiki. Therefore this Court's attention is best focused on Kurashiki.
To summarize, Kurashiki anticipates GE claims 1-4 and 38 under 35 U.S.C. § 102(b). Kurashiki was published in 1967, more than one year before GE's patent application. Kurashiki discloses every material claimed feature of GE's claims; i.e., stable unreacted polyester mixtures within the specified limits. Although the Kurashiki examples themselves contain no unreacted mixtures of PET and PBT, Kurashiki teaches that transesterification can be inhibited by use of a phosphorous containing compound. GE has admitted that Kurashiki enabled one skilled in the art to produce a copolymer or a blend. Example 4 of Kurashiki specifically discloses a 78/22 PET/PBT copolymer mixture. This mixture is within the ranges described GE claims 1-4 and 38.
III. Undue Breadth Under 35 U.S.C. §§ 102 and 103, or § 112
Defendants' October 1989 summary judgment motion is based on "overbreadth." Defendants assert that all of the GE claims are overbroad on a number of grounds. First, defendants contend that the GE claims encompass prior art that was supposedly distinguished by the claims so that they are overbroad under the novelty requirement of 35 U.S.C. § 102 and the nonobviousness requirement of 35 U.S.C. § 103. See generally 2 D. Chisum Patents § 7.03[7] at 7-72 (1990) ("if an inventor claims a group encompassing A, B, and C, and A is old while B and C are new and *319 nonobvious, the claim will be rejected"). Second, defendants argue that the GE claims are overbroad under paragraph 2 of 35 U.S.C. § 112, because they are broader than the subject matter which GE regards as its invention. See generally id. (claims are invalid when they "encompass matter that the inventor does not in fact regard as part of his invention").[8]
A. Overbreadth Under §§ 102 and 103
1. Section 102.
Defendants' "overbreadth" argument under section 102 is basically the same as their argument under section 102(b). The underlying theory is that broad claims are more likely to be found invalid. See Andco Environmental Processes, Inc. v. Niagara Environmental Associates, Inc., 220 U.S.P.Q. 468, 475 (W.D. N.Y.1983). For one reason, they are more likely to read on prior art. A claim covering multiple chemical compositions, such as the GE claims, is anticipated if any one composition is disclosed by prior art. See Titanium Metals, 778 F.2d at 782. Defendants cite Kurashiki, Fiber, and Caldwell as prior art references that anticipate GE's claims.
The GE patent claims substantially unreacted blends of PET and PBT. This conclusion is necessitated by the amendments GE made to its claims during the reexamination. Of the cited references, only Kurashiki discloses such "blends" of PET and PBT, yet this fact is disputed and defendants allow for purposes of their "overbreadth" motion that Kurashiki does not disclose such blends. (D.I. 152 at 2 n. 1) Fiber and Caldwell are concerned with copolymers, not with GE's claimed "blends."
Defendants assert that it is irrelevant that the prior art fails to disclose blends because GE's blends do not behave differently than the prior art copolymers. Defendants assert that the "basic and novel characteristic" of GE's claims is uniformity of crystallinity, irrespective of the presence of a "blend." In other words, defendants' interpretation of the claims differs from GE's in that GE links the characteristics, "stable homogeneity" and "uniformity of crystallinity," with the requirement that the composition be a "blend," whereas defendants do not. Defendants extract their interpretation of GE's claims from the reexamination proceeding.
The Examiner rejected GE's original claims because the original claims encompassed mixtures that contained prior art block copolymer. GE proposed to limit the claim language to "consisting essentially of." The Examiner apparently was not satisfied with this amendment. He requested "objective evidence ... to show that what is being eliminated [block copolymer] would alter the basic nature of the compositions desired to be claimed." (D.I. 152A at A82) (emphasis added) GE responded by submitting the ABC experiment. This experiment did not prove that blends behave differently than copolymer mixtures across the entire range of GE's claims, or even that the blends behave differently than copolymer at a single ratio of PET to PBT. Nevertheless the Examiner was satisfied. Defendants now argue that GE failed to establish that its blends behave differently than copolymers. This may be so. However, in connection with the present "overbreadth" motion, defendants' argument that the composition need not be a blend is flawed. In order to avoid prior art, GE had to limit its claims to blends, and show that the blends behaved differently (had different basic and novel characteristics) than the prior art. These were two separate requirements. The Court addressed GE's change in claim terminology on the earlier summary judgment motion:
*320 "consisting essentially of" restricts the inclusion of unspecified ingredients to those that do not "materially affect the basic and novel characteristics" of the claim.
General Electric I, 698 F.Supp. at 1185. It does not follow that GE's claimed compositions can contain copolymer as long as they behave the same as a blend. GE's compositions must be a blend and must behave differently than copolymer. For purposes of this motion none of the cited references disclose "blends." Therefore the GE patent is not "overbroad" under section 102.
2. Obviousness under section 103.
The second facet of defendants' "overbreadth" argument incorporates the nonobviousness requirement of section 103. The Examiner's original rejection of the GE patent was made under section 103 in light of Kurashiki and Fiber. (D.I. 152A at A75) Defendants now contend that the amended claims are still obvious. Section 103 states in pertinent part:
A patent may not be obtained ... if the differences between the subject matter sought to be patented and the prior art are such that the subject matter as a whole would have been obvious at the time the invention was made to a person having ordinary skill in the art to which said subject matter pertains.[9]
The Court must make the following factual determinations to decide whether a claim is obvious under section 103:
(1) What does the prior art disclose;
(2) what is claimed, and the differences between what is claimed and the prior art;
(3) the level of ordinary skill in the pertinent art; and
(4) whether secondary evidence suggests that the invention was not obvious.
Graham v. John Deere Co., 383 U.S. 1, 17-18, 86 S. Ct. 684, 694, 15 L. Ed. 2d 545 (1966); Metrodonic Inc. v. Intermedics, Inc., 799 F.2d 734, 738 (Fed.Cir.1986), cert. denied, 479 U.S. 1033, 107 S. Ct. 882, 93 L. Ed. 2d 836 (1987). In making these findings under section 103 the Court may consider a broader spectrum of references than under section 102:
In contrast to the novelty determination's strict identity requirement, that is, anticipatory prior art must disclose each element of the claim either expressly or inherently, the nonobvious [sic] determination does not require strict identity.
H. Schwartz, Patent Law and Practice 53 (1988); see Metrodonic, 799 F.2d at 738 (reviewing multiple references in obviousness determination). Consequently, all prior art (i.e., Kurashiki, Kurashiki Abridgment, Fiber, and Caldwell) is relevant to the obviousness inquiry.
a. The Prior Art
As noted above, Kurashiki discloses stable blended mixtures of PET and PBT within the ranges stated in the GE claims. Defendants concede for the sake of this argument that these are copolymers. Fiber discloses a 50/50 PET/PBT copolymer which is within the allowable ranges of the GE claims. Caldwell discloses a PET/PBT copolymer containing 10 mole% PET and 90 mole% PBT, which has "[d]esirable molding properties." (D.I. 152A at 161) The Caldwell mixture is within the ranges permitted by the broadest group of GE claims,[10] but is not within the ranges allowed under the narrower groups of claims.[11]
b. Differences Between the GE Patent and the Prior Art
For purposes of this motion defendants argue that even if the prior art does not disclose GE's claimed "blends," still GE's claimed "blends" and the prior art copolymer are indistinguishable. This is due to the fact that the GE patent claims broad ranges of PET/PBT blends and the compositions *321 at either extreme behave the same whether copolymer is present or not.
The record shows that mixtures containing the lowest concentrations of PET permitted by GE's claims exhibit the same basic and novel characteristics whether the mixture is a copolymer or a blend. The copolymer mixtures in the Floryan application (20 mole% PET) and Caldwell (10 mole% PET) both mold well.[12] The question is raised whether obviousness follows when a patented composition does not exhibit the characteristics said to distinguish it from the prior art throughout the claimed ranges.
A somewhat analogous question was addressed in In re Dill, 604 F.2d 1356, 1361 (C.C.P.A.1979), and In re Tiffin, 448 F.2d 791, 792 (C.C.P.A.1971). Both of these cases held that evidence presented to overcome a prima facie case of obviousness had to be commensurate with the scope of the claims in support of which the evidence is offered. Here, the ABC experiment was probably inadequate to support the broad scope of GE's claims in light of the prior art that suggested low PET copolymers could mold well (Caldwell; Kurashiki; Fiber).[13] However, Dill and Tiffin are distinguishable from this case because both cases involved appeals from a patent rejection.[14] This case does not involve a rejection: the Examiner allowed GE's claims on reexamination. Although obviousness is a question of law, Titanium Metals, 778 F.2d at 780, GE's patent is presumed valid and defendants must prove invalidity by clear and convincing evidence. Moreover, GE amended its claims on reexamination. Neither Dill nor Tiffin involved amended claims.
In support of their motions, defendants have submitted evidence to show that the presence of copolymer does not affect moldability as much as the amount of PET affects it. For example the German experiment shows that under certain conditions moldability of PET/PBT blends deteriorates as the PET content increases above 40%. The Floryan application and Caldwell both show that low PET copolyesters exhibit desirable moldability. Even the lab notebook of Lawrence R. Wallace, a GE employee who conducted experiments in preparation for this litigation shows that 75/25 PET/PBT blends result in "soft" test bars that "hang up" on the mold under the conditions stated in column 13 of the GE patent.[15] (D.I. 172A at A390-A391)
However, the Court must evaluate this evidence in light of defendants' substantial evidentiary burden. GE does not dispute that its claims, read in light of its specification, may contain a substantial number of inoperative combinations. GE does maintain that nearly all of its claimed blends may successfully be molded at some point within the parameters stated in column 13 of the GE patent. Defendants have not presented clear and convincing evidence to the contrary. Indeed, defendants' own Costanza application describes a PET/PBT blend which is comprised of up to 98% by weight of a component which comprises at *322 least 85 mole% PET. (D.I. 172A at A222) The molding conditions called for in the Costanza application are similar to the GE patent. (D.I. 172A at A228) Theoretically, the Costanza blend could contain up to 98% by weight PET. GE's expert Dr. Hepp has opined that a successful mold containing 95% PET could be made under the conditions stated in the GE patent. (D.I. 172A at A349)
Even though the evidence submitted by GE in support of its claim of nonobviousness was less than compelling, the Examiner was satisfied that it distinguished GE's compositions from the prior art. GE's patent is presumed valid. Defendants have not established their position to the contrary. For the sake of this motion, we will consider that GE's blends are distinguishable from the prior art. A finder of fact may determine otherwise.
c. The Level of Ordinary Skill in the Pertinent Art
Kurashiki, Fiber, and Caldwell all disclose PET/PBT mixtures within the ranges allowed by the GE patent. It was within the skill of the art in 1971 to mix PET and PBT together and to mold the resulting mixture. It is irrelevant for purposes of the "overbreadth" motion whether a method of preventing transesterification was known at the time of the GE patent application because defendants concede for this argument that the prior art references all disclose block copolymer. Therefore, to the extent that this motion involves mixing and molding copolymers of PET and PBT, both parties agree that this was within the level of ordinary skill in the art. GE makes the further argument that in 1971, persons skilled in the art understood that a "compatible blend" of two different polymers was rare. (Anderson Declaration, D.I. 172A at A3 ¶ 8) One of the inventors of defendants' Costanza application has testified that the crystallinity behavior of the PET/PBT blend was "significant and unexpected." (D.I. 172A at A258-A259) This evidence is substantially unrebutted. However, this evidence of the expectations of those skilled in the art is relevant not to the issue of level of skill in the art but instead to the objective evidence of nonobviousness, to which we will now turn.
d. Objective Indicia of Nonobviousness
The objective evidence argues in favor of nonobviousness. Commercial success, copying, and unexpected results are all evidence of nonobviousness. Akzo N.V. v. United States Int'l Trade Comm'n, 808 F.2d 1471, 1480 (Fed.Cir. 1986), cert. denied, 482 U.S. 909, 107 S. Ct. 2490, 96 L. Ed. 2d 382 (1987). GE sales of PET/PBT blends have grown from 1,000,000 pounds per year to 12,000,000 pounds per year since 1980. (D.I. 172A at A166-A181) Defendants' sales of allegedly infringing products grew from less than 1,000,000 pounds in 1984 to over 6,000,000 pounds in 1987. (D.I. 120A tab m) This evidence of commercial success argues strongly that GE's claimed compositions were not obvious. Even if defendants had made a prima facie case of obviousness, which they have not, the evidence of commercial success in this case requires the Court to deny defendants' motion under section 103. See Simmons Fastener Corp. v. Illinois Tool Works, Inc., 739 F.2d 1573, 1576 (Fed.Cir.1984), cert. denied, 471 U.S. 1065, 105 S. Ct. 2138, 85 L. Ed. 2d 496 (1985).
B. Overbreadth Under § 112 ¶ 2
Paragraph 2 of 35 U.S.C. § 112 provides:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
A motion under the second paragraph of section 112 is directed to the wording of the claims. The claims themselves must particularly point out and distinctly claim the applicant's invention. Defendants argue that the Court should determine definiteness by looking solely to the claims and ignoring the specification. The law is not so rigid:
The amount of detail required to be included in claims depends on the particular invention and the prior art, and is not to be viewed in the abstract but in conjunction with whether the specification is *323 in compliance with the first paragraph of section 112: "If the claims, read in the light of the specifications, reasonably appraise those skilled in the art both of the utilization and scope of the invention, and if the language is as precise as the subject matter permits, the courts can demand no more." Georgia Pacific Corp. v. United States Plywood Corp., 258 F.2d 124, 136, 118 U.S.P.Q. (BNA) 122, 132 (2d Cir.), cert. denied, 358 U.S. 884, 79 S. Ct. 124, 3 L. Ed. 2d 112, 119 U.S.P.Q. (BNA) 501 (1958).
Shatterproof Glass Corp. v. Libbey-Owens Ford Co., 758 F.2d 613, 624 (Fed.Cir.), cert. dismissed, 474 U.S. 976, 106 S. Ct. 340, 88 L. Ed. 2d 326 (1985). "Compliance with the second paragraph of section 112 is generally a question of law...." Id. Accordingly, we will evaluate GE's claims in light of the prior art and the disclosure contained in the specification.
The applicants of the GE patent (Fox and Wambach) regard their invention to be "a miscible (i.e. compatible) blend of PBT and PET over the range of 1-99%" that exhibits uniformity of crystallinity and stable homogeneity. (D.I. 172 at 61) Defendants argue that GE's claims are indefinite because they contain many possible combinations of PET and PBT that do not exhibit "stable homogeneity and uniform crystallinity," and urge that claims which include a substantial measure of inoperative combinations are properly rejected under section 112. See In re Corkill, 771 F.2d 1496, 1501 (Fed.Cir.1985). In In re Corkill it was established that many of the compositions covered by the claims were truly inoperative they were unsatisfactory for the purpose stated in the patent. In this case defendants have not shown that any of GE's combinations are inoperative under the thermal history parameters taught by the specification and known by those skilled in the art.[16] Defendants' own Costanza application claims PET/PBT "blends" through nearly the entire range claimed by GE. Accordingly, there is no failure of compliance under section 112 ¶ 2 and defendants' argument of invalidity on that ground will be rejected. Shatterproof Glass, 758 F.2d at 624; Application of Cook, 439 F.2d 730, 734-35 (C.C.P.A.1971).
IV. Laches
A. Factual Background
Defendants claim that GE knew of defendants' alleged infringement as early as September 25, 1978, but did not file this suit until August 24, 1987, almost nine years later. During this time defendants' sales of infringing product rose from less than $150,000 to nearly $8,000,000 per year. The following facts appear from the record.
The GE patent was issued on April 27, 1976. On September 21, 1978, A.D. Wambach, an inventor of the GE patent, received a sample of Celanese plastic labelled "J207." He took the sample to Joel Lundgren, an analytical chemist at GE for analysis. Lundgren concluded on September 25 that the sample contained 22.1% PET, 47.3% PBT, and 30.57% glass fiber, and Wambach informed GE patent counsel William Mufatti of an "investigation of infringement by Celanese" the same day. (D.I. 120A tabs B & C; D.I. 120 at 3; GE withheld document list, document no. 205, D.I. 120A tab d at A20) The next day Lundgren wrote to Mufatti concerning an "investigation of infringement by Celanese." (Id., withheld document no. 198 at A19)
GE's lawyers and Celanese's lawyers were in contact almost immediately. On January 30, 1979, Kenneth A. Genoni, patent counsel for Celanese, wrote to Mufatti concerning the GE patent. This letter states Celanese's position that claims 1-40 of the GE patent are unpatentable in view of prior art. (D.I. 120A at A22) The prior art included Kurashiki and Fiber.
GE denies that the samples analyzed in 1978 were commercial products. GE asserts that defendants' products were not commercialized until 1982. GE states that defendants' product "J207," which was analyzed by GE, was never commercialized at *324 all. Sales figures, stipulated to by the parties, show that only 3200 pounds of "J207" were sold in 1981. By contrast, GE's sales of PBT/PET blends between 1979 and 1981 were 5,750,000 pounds. GE asserts that it first analyzed a product of defendants that would eventually be commercialized on November 6, 1979. (D.I. 164A at A132) That product was "J227," and was not commercialized until 1982. (D.I. 164A at A97) However, the same sales reports relied upon by GE show "J227" sales in 1980 and 1981 also. (D.I. 164A at A95-A96)[17] Defendants have submitted sales data showing sales of PET/PBT products totalling $136,242.10 for 1980. (D.I. 120A at A104) GE responds with Celanese accounting and marketing documents which show defendants' market share of PET/PBT to be 0% in 1981, and a commercialization date of 1982. (D.I. 164A at A55; A69-A73; A74-A76) GE analyzed additional samples of Celanese products from 1979 through 1986. (D.I. 120A tab F)
GE requested reexamination on March 11, 1985. Celanese also filed a request for reexamination in September of that year, but did not identify itself. GE told the Examiner that it believed Celanese was the anonymous party requesting reexamination. The Examiner issued a Notice of Intent to issue a Reexamination Certificate in March 1987. (D.I. 164A at A196-A197) The Certificate was issued on June 2, 1987. (D.I. 164A at A198-A200) GE filed this suit on August 24, 1987.
During the period from September 25, 1978, through 1987, sales of Celanex PET/PBT products increased from less than $150,000 in 1980 to more than $7,000,000. (Declaration of Donal McNally, D.I. 120A tab m) Predictably, many key witnesses have forgotten what happened during this time. Dr. Fox, one of the inventors of the GE patent, has died. Dr. Wambach, the other inventor, has left GE. (D.I. 120A at A86) Dr. Jaquiss, research and development manager for GE's Valox Products Section during some of the relevant time, has also died.
B. Was GE's Delay Unreasonable?
Defendants claim that GE knew of the alleged infringement in September, 1978, and that GE's delay from September, 1978, until August, 1987, when GE filed suit, was unreasonable. GE counters that it did not know of any infringing products which were eventually commercialized until November, 1979, less than six years before the reexamination. GE also contends that the relevant "end" date of its delay is the commencement of the reexamination proceeding rather than the commencement of this suit.
Laches is an equitable defense. Its applicability depends on the facts of each case. "In order to sustain a defense of laches in a patent case, a defendant must normally prove an unreasonable and inexcusable delay, together with prejudice or injury resulting from the delay." Bott v. Four Star Corp., 807 F.2d 1567, 1575 (Fed.Cir.1986). However, any delay in filing a patent infringement suit that exceeds six years is presumed unreasonable and injurious to the defendant. See Jamesbury Corp. v. Litton Industrial Products, Inc., 839 F.2d 1544, 1552 (Fed.Cir.), cert. denied, 488 U.S. 828, 109 S. Ct. 80, 102 L. Ed. 2d 57 (1988); Hottel Corp. v. Seaman Corp., 833 F.2d 1570, 1572 (Fed.Cir.1987).[18] When the delay exceeds six years, the burden of rebutting the presumption shifts to the patentee who must prove (1) the existence of an excuse for the delay, and (2) that the excuse is reasonable. Jamesbury, 839 F.2d at 1552. Delay is measured from the time when the defendant first learned, or in the exercise of reasonable diligence, should have known, of the alleged infringing activity. Id.[19] Laches bars recovery of damages *325 for any infringement occurring prior to the filing of the suit. Id. at 1551.
Here, the reexamination proceeding began on March 11, 1985. This was more than six years after GE analyzed the "J207" sample, more than six years after the internal GE communications regarding alleged infringement by Celanese, and more than six years after the meeting and correspondence between Genoni and Mufatti in which Celanese asserted that the GE claims were invalid.
GE claims that its delay "began" on November 6, 1979, when GE analyzed the sample of defendants' "J227" that was eventually commercialized. GE cites no authority in support of this proposition. The undisputed facts of record show that GE knew of Celanese's infringing product in September, 1978. More than six years passed from September 1978 until the Re-examination proceeding in 1985. Nearly eight years passed from September 1978 until this suit was filed in 1987. GE's delay of more than six years was unreasonable. GE has the burden to show a reasonable excuse for the delay.
GE argues that it was excused from delay until the infringing compositions were commercialized in 1982 because before that time it was not monetarily worthwhile to litigate. In Autoclave Engineers, Inc. v. The Duriron Company, 190 U.S.P.Q. 125, 130 (E.D.Pa.1976), the court held that the plaintiff was excused from its delay because defendants' sales presented no business threat to plaintiff until defendant made a sale to a major customer. Autoclave is similar to this case because the Autoclave defendants had been marketing infringing products for a number of years in small quantities before making the major sale. 190 U.S.P.Q. at 127. However in Autoclave, defendants' sales showed no steady trend of growth as there is in this case.[20] In addition, the plaintiff in Autoclave was engaged in merger negotiations with defendants during most of the delay and commenced the infringement action within three weeks of defendants' "big" sale. 190 U.S.P.Q. at 126, 129 (findings of fact 38-39). No such mitigating facts exist here.
Although some of the facts, viewed in the light most favorable to GE, might suggest that it was not economically viable to commence an infringement action before 1979, other undisputed facts are damning to GE. GE's internal communications show that it was concerned about infringement as early as September, 1978. GE was in possession of infringing samples at the same time. Celanese confirmed its interest in development of PET/PBT products in 1978. (D.I. 164A at A106; D.I. 173 at 14) As defendants point out, infringing activity includes acts of making, using, or selling a claimed invention. 35 U.S.C. § 271(a). Giving GE every benefit of the doubt, GE was excused at most until early 1980, when Celanese sales began to gather momentum.
GE also asserts that its delay "ended" with the reexamination proceeding. Technically this is not an accurate statement of the law. The Federal Circuit recently held that "a patentee's involvement in litigation against one or more infringers may excuse delay in enforcing the patent against another infringer...." Jamesbury, 839 F.2d at 1552 (emphasis added). Thus, "other litigation" does not "end" the delay, it merely "excuses" a portion of the delay. Moreover, it is not clear whether a reexamination proceeding constitutes "other enforcement litigation" for purposes of this rule. In Hottel, 833 F.2d at 1573 n. 2, the Federal Circuit explicitly declined to decide whether a reexamination proceeding is "other litigation."
*326 For the same reason that the Federal Circuit in Hottel declined to decide whether a reexamination proceeding is "other litigation," we will also decline to do so. In order to avail itself of the "other litigation" excuse, the patentee must give notice to the alleged infringer of the existence of the other action and of an intent to enforce its rights against the infringer at the conclusion of the other litigation. Jamesbury, 839 F.2d at 1553. In Hottel the court denied plaintiff an excuse for its delay because plaintiff's notice to defendant was deficient. 833 F.2d at 1573. Nothing in the record now before us indicates that GE notified defendants of an intention to enforce its rights at any time. GE sent the following letter to Celanese in August, 1984:
We understand that Celanese has been sampling two PBT blends designated JKX706 and J235.
We would appreciate your reviewing the composition of these products in view of our U.S. patents 4,257,937 and 3,953,394. Incidentally, we plan to request shortly the U.S. Patent Office to re-examine these patents.
(D.I. 164A at A141) This letter does not suggest GE's intention to enforce its patent rights against Celanese, nor has GE pointed to any other communication that did. To the contrary, the record shows that GE closely watched defendants' every move but failed to reveal its full intention at any time.
GE should be excused for the delay that occurred prior to defendants' commercialization of the allegedly infringing product in early 1980. GE should not be excused from delay during the reexamination proceeding.
C. Clean Hands
GE's final argument is that defendants can not claim laches because they plagiarized GE's patent. This is a variation on the equitable clean hands theme. GE relies on Bott, 807 F.2d at 1576, in support of this argument. In Bott, the defendant had knowingly copied the plaintiff's patented device. The defendant attempted to invoke laches due to plaintiff's delay of more than six years. The court held that where the record shows the defendant had access to the patent and contemporaneously produced an infringing product, the burden shifts to defendant to show independent development of the infringing device.
In this case there is no doubt that defendants were aware of the GE patent or that they contemporaneously produced infringing products. However, defendants argue that they independently developed their products in 1972, four years before GE was granted its patent. The record supports defendants' position. GE claims that defendants plagiarized Examples 13 and 14 of the GE patent and GE's Valox 831, in their J207 and J228 compositions. Examples 13 and 14 of the GE patent disclose the following compositions:
Example 13 Example 14
PBT 56% 35%
PET 14% 35%
Glass Fiber 30% 30%
PBT/PET ratio 80:20 50:50
(D.I. 164 at 3)
In 1972, lab notebooks of Celanese scientists show that they extruded products similar to those set forth in Examples 13 and 14 of the GE patent:
Run # 2 Run # 3
PBT 56% 35%
PET 14% 35%
Glass Fiber 30% 30%
PBT/PET ratio 80:20 50:50
(D.I. 173A at A18, A59-A60) Therefore the Celanese composition predated the issuance of GE patent by four years. Indeed, GE has admitted as much: "[T]hree chemists at Celanese independently made the same discovery as Fox and Wambach had made at GE more than one year earlier." (D.I. 168 at 7) (emphasis added) The evidence also reflects that Celanese scientists had independently formulated the prototypes for J207 and J228 by early 1974. This was two years before the GE patent was issued and three years before GE claims that defendants were aware of Valox 831. (D.I. 173A at A81-A82)
Defendants have satisfied their burden to show that they independently arrived at the infringing composition. Celanese independently *327 discovered the allegedly infringing compositions at least by 1974, and perhaps as early as 1972. GE cannot avoid the laches defense by contending that defendants' hands are not clean.[21]
In summation, GE knew of Celanese' infringing activity by September 25, 1978. GE is presumed to be guilty of laches if it waits more than six years to file suit. More than six years passed from the time GE discovered the infringing activity until the reexamination proceeding. More than eight years passed from 1978 until this action was begun. This delay is presumed unreasonable unless GE has a valid excuse. GE was probably excused from commencing this action until Celanese developed a market for its infringing products. This was sometime in 1979 or early 1980 when Celanese began to sell the allegedly infringing product. However this action was not commenced until 1987. The reexamination proceeding, if it is "other litigation," was commenced in late 1985. By that time defendants had developed a $4.2 million annual market for the infringing PET/PBT blends and GE had known of the infringement for seven years. At such a late date Celanese was entitled to know where it stood. Yet GE never informed Celanese of an intention to enforce its patent rights. Such lassitude is inexcusable in the face of Celanese' continuous assertion that the GE patent was invalid, and in light of the fact that GE was well aware of Celanese's growing market. GE will be barred from recovering damages that accrued before this action was filed on August 24, 1987.
CONCLUSION
1. GE claims 1-4 and 38 are invalid as anticipated by Kurashiki under 35 U.S.C. § 102(b). Defendants' September 1989 motion for partial summary judgment will be granted.
2. GE's claims are not "overbroad" under 35 U.S.C. §§ 102, 103 and 112 ¶ 2. Defendants' October 1989 motion for summary judgment will be denied.
3. GE unreasonably delayed filing this suit. Defendants' motion for partial summary judgment on the grounds of laches will be granted. GE is barred from recovery of damages accruing prior to August 24, 1987.
NOTES
[1] In the lexicon of patent law, "comprising" leaves a claim open for the inclusion of unspecified ingredients; "consisting essentially of" restricts the inclusion of unspecified ingredients to those ingredients that do not "materially affect the basic and novel characteristics" of the claim. General Electric I, at 1185.
[2] Examples of GE's amended claims are set out below. Bracketed language was in the original claims and underlined language is in the amended claims. Amended claims 1-4 and 38 describe:
1. A thermoplastic, stable blended composition that is rigid at a temperature of 75°-90° F [comprising] consisting essentially of
a. from about 1 to about 99 parts by weight of a poly(ethylene terephthalate) resin and
b. from about 99 to about 1 part by weight of a poly(1,4-butylene trerphthalate) resin or a copolyester thereof with a minor amount of an aliphatic or aromatic dicarboxylic acid or an aliphatic polyol.
2. A composition as defined in claim 1 wherein component (b) is a poly(1,4-butylene terephthalate) resin.
3. A composition as defined in claim 1 wherein component (a) [comprises] is from about 10 to about 90 parts by weight and component (b) [comprises] is from about 90 to about 10 parts by weight.
4. A composition as defined in claim 1 wherein component (a) [comprises] is from about 20 to about 80 parts by weight and component (b) [comprises] is from about 80 to about 20 parts by weight.
38. A thermoplastic, stable blended composition that is rigid at a temperature of 75°-90° F, said composition consisting essentially of
a. from about 1 to about 99 parts by weight of a poly(ethylene terephthalate) resin and
b. from about 99 to about 1 part by weight of a poly(1,4-butylene terephthalate) resin or a copolyester thereof with a minor amount of an aliphatic or aromatic dicarboxylic acid or an aliphatic polyol.
(D.I. 111A, A11-A14)
[3] 20 mole% is roughly equivalent to 18% by weight. (D.I. 152 at 13)
[4] U.S. Patent No. 3,624,024.
[5] Interestingly, the Caldwell patent specification refers to this copolymer mixture as a "blend." (D.I. 152A at A161 col. 7, line 23)
[6] 35 U.S.C. § 112 provides in relevant part: "The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art ... to make and use the same...."
[7] On the enablement issue the experts' battle over Fiber mirrors the battle over Kurashiki. Compare Porter Declaration, D.I. 111A at A33 ("The Fiber reference unequivocally allows one of ordinary skill in the art to recognize that PET and PBT are among the polyesters that may be made into a polyester mixture wherein the two polyesters are not chemically combined."), with Anderson Declaration, D.I. 168A at A14-A15 ("In 1971 a person of ordinary skill in the art would not have found in Fiber a disclosure of a compatible blend of any pair of polymers and a person of ordinary skill in the art would not have found in Fiber a disclosure of a compatible blend of PBT and PET.").
[8] Defendants concede for purposes of their overbreadth motion that the prior art (Kurashiki, Fiber, etc.) does not disclose unreacted blends of PET and PBT. (D.I. 152 at 2 n. 1) This concession is in direct conflict with defendants' position in their section 102(b) motion that Kurashiki discloses GE's "blends" of PET and PBT. This concession also permits the seemingly inconsistent conclusion that Kurashiki does not invalidate GE's patent on obviousness grounds but does anticipate under section 102(b). If the reader has felt confusion as a result of picking up these inconsistencies in the arguments defendants are making, you may feel reassured to know that they are deliberate.
[9] Section 103 was enacted in 1952 to codify the judicially created condition of patentability known as "invention." See Graham v. John Deere Co., 383 U.S. 1, 14, 86 S. Ct. 684, 692, 15 L. Ed. 2d 545 (1966).
[10] I.e., from 99:1 to 1:99. Claims 1, 2, 5-9, 12-17, 20, 21, 24-30 and 38-40.
[11] From 90:10 to 10:90, and from 80:20 to 20:80. The 10 mole% specified in Caldwell is slightly less than 10% by weight.
[12] Floryan was not published prior to GE's application for the GE patent and is therefore not properly considered in the obviousness inquiry. It is cited as evidence of the behavior of low PET copolymers. Caldwell is prior art. GE concedes that "[t]he Floryan disclosure is fundamentally the same as Caldwell Example 5." (D.I. 172 at 50)
The German Experiment, also not prior art, shows that under certain conditions mixtures containing high concentrations of PET mold poorly, whether they contain copolymer or not.
[13] The ABC experiment compared copolymers containing 78% PET with blends containing 38.5% PET. See supra note 1.
[14] In Dill, the applicant appealed from a decision of the PTO Board of Appeals affirming the rejection of many of his claims. In Tiffin, the PTO petitioned the Court of Customs and Patent Appeals for a rehearing. The Court of Customs and Patent Appeals had earlier found the patent to be valid, reversing a PTO Board of Appeals decision which found invalidity on the grounds of obviousness.
[15] Column 13 describes the parameters under which GE's claimed blend will produce a satisfactory test bar:
One convenient way to determine if a polyester alloy is suitable for use in this invention is to injection mold it with the mold temperature of 75°-200°F. in a standard short cycle, e.g., 10-90 seconds, into workpieces 1/16 to ½ inch thick.
(D.I. 172A at A32)
[16] The parameters under which the GE compositions will mold successfully are set out in column 13 of the GE patent.
[17] 1980 sales of J227 were 250 pounds. 1981 sales of J227 were 16,381 pounds. (D.I. 164A at A95-A96)
[18] The presumption that a delay of more than six years is unreasonable is based on 35 U.S.C. § 286, which bars recovery of damages that occurred more than six years before the complaint is filed. Leinoff v. Louis Milona & Sons, Inc., 726 F.2d 734, 741-42 (Fed.Cir.1984).
[19] The allegedly infringing sample that GE analyzed in 1979 was manufactured by Celanese Corporation, predecessor in interest to defendants. It is undisputed that any delay by GE can be "tacked" from Celanese Corporation to defendants. See Jamesbury, 839 F.2d at 1555.
[20] Defendants' annual sales of PET/PBT product was as follows:
1978 $ 0
1979 6,869
1980 136,242
1981 233,242
1982 474,279
1983 471,756
1984 2,080,469
1985 4,247,758
1986 6,839,224
1987 4,994,000 (through Aug.)
(D.I. 120A tab M)
[21] In reaching this conclusion, we have considered defendants' June 7, 1983 "interoffice memorandum," submitted by GE, in which J.J. Conway, now General Manager and Vice President of HCC, instructs Celanese's Technical Director, "If a copy of a GE product is what is necessary, let's do it." (D.I. 164A at A29) A careful reading of this memorandum reveals that what defendants were attempting to "copy" was not GE's claimed blend, but other ingredients that are not claimed in the GE patent, such as "brominated polycarbonate, extra polycarbonate, .85 i.v. etc." (Id.)
