General Foods Corp. v. Studiengesellschaft Kohle mbH, 765 F. Supp. 121 (S.D.N.Y. 1991)

US District Court for the Southern District of New York - 765 F. Supp. 121 (S.D.N.Y. 1991)
May 30, 1991

765 F. Supp. 121 (1991)


No. 88 Civ. 8343 (GLG).

United States District Court, S.D. New York.

May 30, 1991.

Kenyon & Kenyon by Paul H. Heller, James Galbraith, Richard L. DeLucia, Richard S. Gresalfi, Peggy A. Climenson, Frederick H. Rein, New York City, for plaintiff.

Sprung Horn Kramer & Woods by Arnold Sprung, Nathaniel D. Kramer, Tarrytown, N.Y., for defendant.


GOETTEL, District Judge.


1. This is an action by General Foods Corporation ("GF") for a declaratory judgment that five patents United States Patents 4,260,639; 3,806,619; 3,969,196; *122 4,247,570; and 4,348,422 owned by its licensor Studiengesellschaft Kohle mbH ("SGK"), are invalid and unenforceable, and do not cover a coffee decaffeination process employed by GF at a facility in Houston, Texas, known as "AMCO."

2. Although each of the five patents is licensed, SGK asserts that only claims 1 and 4 of United States Patent 4,260,639 (the "`639 patent") (PTX-1) cover GF's operation at the AMCO plant. It does not contend that royalties are due under any of the other four patents subject to this declaratory judgment action. Thus, the remaining patents are no longer in suit, and the court need only determine whether claims 1 and 4 of the '639 patent are invalid, unenforceable, or do not cover the AMCO process.

3. GF is a Delaware corporation having a place of business in White Plains, New York. GF owns Maxwell House Coffee Company, which manufactures decaffeinated coffee at the AMCO plant for use in brands such as Sanka, Maxim, Brim, Yuban, and Maxwell House.

4. GF also owns Kaffee HAG, or Kaffee Handels Aktiengesellschaft ("HAG"), a leading coffee processor and research company in coffee-related technology located in Bremen, Germany. HAG was acquired by GF in 1979, but was originally founded in 1907 by Ludwig Roselius. In the early 1900's, Roselius and other HAG scientists invented the first processes for the decaffeination of coffee by employing organic solvents capable of dissolving caffeine from moistened coffee beans.

5. Defendant is a western German company, having a place of business in Mulheim/Ruhr, Germany. It acts as trustee for the Max-Planck-Institute for Coal Research, a not-for-profit research institute in Mulheim/Ruhr, Germany. SGK is in the business of procuring and licensing patents for a non-profit German research organization called the Max-Planck-Institut fur Kohlenforschung ("MPI"), which has facilities in Mulheim/Ruhr.

6. Pursuant to a February 1978 license agreement with SGK, GF now holds a non-exclusive license under the '639 patent (it was originally an exclusive license), as well as under the four other SGK patents initially in suit. GF, however, contends that the decaffeination process used in its newly built AMCO facility does not come within claims 1 or 4 of the '639 patent, and hence that royalty payments are not due. Nevertheless, it has continued to pay royalties under protest during the pendency of this suit in order to avoid cancellation of its license.

7. GF further contends that the claims in suit are invalid under the judicially created doctrine of obviousness-type double patenting, and that they are obvious under 35 U.S.C. § 103 and non-enabled by the patent's specification under 35 U.S.C. § 112. GF also claims that during the prosecution of the '639 patent, the conduct of SGK, the named inventor, Dr. Kurt Zosel ("Zosel"), and SGK's agents and attorneys mandates a finding that the patent claims in suit are unenforceable due to inequitable conduct before the United States Patent and Trademark Office ("Patent Office"). SGK denies GF's contentions. Plaintiff has requested repayment of certain royalties. Defendant has counterclaimed for repayment of its attorneys' fees under the terms of the license.

8. On April 5, 1991, the court issued an Order of Trial:

[GF] has moved to bifurcate the issue of double patenting. The application is granted. However, the issue will be decided directly following the presentation of the evidence. If the decision does not dispose of the action, trial will then immediately continue on the remaining issues.

Because the court concludes that the '639 patent claims in suit are invalid for double patenting, trial of the remaining issues is not required.


9. The '639 patent relates to and is entitled "Process for the Decaffeination of Coffee." The patent issued on April 7, 1981 from an application (Ser. No. 364,190) filed May 25, 1973 (the "1973 application"), which is a continuation-in-part ("C-I-P") of *123 the original parent application (Ser. No. 110,428), filed January 28, 1971 (the "1971 application") in the name of Dr. Kurt Zosel, who is now deceased. The '639 patent is assigned to SGK.

10. Both the 1971 and 1973 applications leading to the '639 patent claim priority from German and Austrian patent applications filed in February 1970.

11. Asserted claims 1 and 4 of the '639 patent describe a process in which moist carbon dioxide ("CO2") gas at supercritical conditions (i.e., high temperatures and pressures) is contacted with raw coffee to provide a "substantially decaffeinated" coffee. Specifically, claim 1 covers:

A process for the decaffeination of raw coffee which comprises contacting the raw coffee with water-moist carbon dioxide above its critical temperature and critical pressure to effect removal of caffeine therefrom and recovering a substantially decaffeinated coffee, the amount of water in the carbon dioxide being sufficient to effectuate said removal of the caffeine from the coffee.

Claim 4 covers a decaffeination time of 5 to 30 hours:

A process as claimed in claim 1 in which the contact with the moist carbon dioxide is effected for a period of from 5 to 30 hours.

12. A "substantially decaffeinated" coffee is one from which a large degree of caffeine has been removed. According to present industry standards, decaffeinated coffees retain only 3% of their caffeine.

13. The '639 patent did not issue and the patent term did not commence in the United States until more than 11 years after its German and Austrian counter-parts were filed, and 8 years after SGK rejected, in 1973, patent claims which the Patent Office had allowed.

14. SGK filed some 22 foreign counter-parts to the '639 patent in 1971. Every foreign patent which issued from those counterpart applications has expired except the '639 U.S. patent, which does not expire until 1998.

A. Zosel's Decaffeination Work in Connection with Green Coffee

15. In about 1969, Dr. Zosel of MPI conducted decaffeination experiments with supercritical fluids. Zosel's first experiments involved attempts to decaffeinate roast coffee with moist supercritical carbon dioxide. The result was a product that yielded a terrible tasting beverage. The coffee was unsuitable for its intended purpose. Dr. Zosel then carried out experiments, which were successful, involving a process for the decaffeination of raw, undigested, green coffee using moist, supercritical CO2 as the solvent.

16. Raw coffee has a greenish color prior to roasting. The coffee oil containing the coffee aroma and flavor is only produced when the raw coffee is roasted.

17. Based on Zosel's work, in February 1970, SGK had filed a patent application in Germany and, in 1971, filed the first U.S. counterpart application which led to the '639 patent in suit. In his patent application, Zosel reported that caffeine could be removed from raw coffee by the application of moist supercritical CO2. This process led to a decaffeinated coffee bean which could be roasted and brewed into a palatable coffee beverage. The application noted that if the process were practiced on roasted beans, the aroma and flavor would be lost.

18. Under German law, which requires that inventors receive a percentage of gross royalties accrued by assignees of their licensed patents, Zosel would profit from any patent issuing from the application. Zosel received (and his estate receives) 20% of the royalties generated from the '639 patent in suit.

19. Zosel described his moist, supercritical CO2 decaffeination process as applicable only to raw coffee beans. Consequently, the 1971 U.S. application for the '639 patent states that in the supercritical CO2 process, "[t]he caffeine must be extracted from the raw beans, since extraction [of the caffeine] of the roasted beans cannot *124 be carried out without loss of aroma." The patent claims were limited to a decaffeination process for raw or green coffee with moist supercritical CO2.

20. Every claim of every corresponding foreign counterpart patent that issued, as well as the patent in suit, is explicitly limited to the decaffeination of raw (or green) coffee.

B. HAG's Developments

21. With a view towards licensing, Zosel first disclosed his decaffeination system to HAG in June 1970 and later in the form of his Austrian patent, which describes and claims a decaffeination system as depicted in Figure 1 of the later issued '639 patent in suit.

22. As illustrated in Figure 1 of the original U.S. application, caffeine-laden supercritical CO2 is passed through an apparatus called a heat exchanger, which causes the gas to leave the supercritical state and become a liquid. The caffeineladen liquid CO2 is then passed into a carbon containing vessel for removal of the caffeine from the liquid CO2.

23. In June 1971, Dr. Otto Vitzthum of HAG saw an actual decaffeination system at MPI in Mulheim. It was not the system depicted in the Austrian patent, but a further embodiment which had three high pressure vessels called autoclaves: caffeine was extracted from green coffee in the first autoclave by moist supercritical CO2; the caffeine was then removed from the moist supercritical CO2 by dissolving the caffeine in water in a second autoclave; periodically, the caffeine-laden water was sent through a third autoclave that contained carbon, which removed the caffeine from the water.

24. The three autoclave system Dr. Vitzthum saw in Mulheim was an improvement over the system disclosed in the Austrian patent, and depicted in Figure 1 of the later issued '639 patent, because the system did not require the CO2 decaffeination solvent to leave the supercritical state in order to remove caffeine from the solvent before the solvent was recycled.

25. Prior to learning of Dr. Zosel's invention involving the use of supercritical carbon dioxide to decaffeinate coffee, it had never occurred to Dr. Vitzthum, or to anyone else at HAG, to use moist supercritical carbon dioxide to decaffeinate raw coffee.

26. While many solvents are capable of extracting caffeine from coffee, only a relatively few have been recognized as effective solvents for producting palatable decaffeinated coffee.

27. Dr. Vitzthum and others at HAG, after learning of Dr. Zosel's invention involving the decaffeination of coffee with supercritical carbon dioxide, devised the first supercritical decaffeination process for roast coffee. It included the following steps:

(a) coffee oil, including its aroma, is extracted from roast coffee utilizing dry, supercritical carbon dioxide;
(b) caffeine is then extracted from roast coffee utilizing moist, supercritical carbon dioxide;
(c) the previously separated coffee oil containing the aroma and flavor components is then re-added to provide a palatable instant coffee product.

28. HAG filed an application for this multistep process in Germany in April 1971, and to resolve SGK's allegation that it was built upon a private disclosure of Dr. Zosel's invention of the decaffeination of raw coffee with moist supercritical carbon dioxide, in April 1973 it assigned the application to SGK, retaining 30% of the royalties that might be obtained thereon.

1. Lipka's Work

29. From August to October 1971, HAG scientist Bernd Lipka and his assistants, Messrs. Breden and Steinwedel, periodically visited Mulheim to evaluate for potential commercial application the system Dr. Vitzthum saw in Mulheim.

30. Lipka found that Dr. Zosel's system required at least 16 hours to decaffeinate coffee, a clearly unacceptable level for commercial application. HAG's process required only 9 to 10 hours. Lipka believed the system to be insufficient in some other *125 respects as well and discussed various possible improvements with Dr. Zosel.

31. In October 1971, Lipka discovered that he could eliminate both the intermediate water autoclave and the associated need to remove caffeine from the water by directly contacting the caffeine-laden supercritical CO2 with activated carbon which could effect caffeine removal directly from the supercritical gas.

32. HAG investigated the patentability of Lipka's system, but decided not to pursue the matter because it was embraced by Dr. Zosel's Austrian patent on the basic idea of using moist supercritical CO2 to decaffeinate raw coffee.

33. Lipka took steps to construct his improved system. He requested that Dr. Zosel ship the equipment needed to build it, i.e., only two autoclaves, and Zosel complied. Lipka intended to use the equipment requested from MPI to operate a pilot plant in Bremen, Germany, but stopped when German authorities found that the equipment had not received official safety approval.

34. Lipka's efficient two pressure vessels system was an improvement over the system disclosed in the originally filed U.S. application leading to the '639 patent (Figure 1) and the later developed three autoclave water system.

35. Lipka's system (Figure 2) is an improvement over the system depicted in Figure 1 because it saves the time and energy consuming processing step of entering and exiting the supercritical state, and it eliminates costly heat exchangers.

36. The pilot plant which incorporated Lipka's system was ultimately successfully commercialized in 1978, and could decaffeinate in 8 to 9 hours. Dr. Zosel was fully advised of Lipka's new decaffeination system in 1971 when it was designed and, later, in early 1973, Lipka showed him his pilot plant.

37. In the 1973 C-I-P application for the '639 patent, SGK added a second illustration of a decaffeination system (Figure 2), a written description, and patent claims (now claim 13), all of which were directed to Lipka's development. Figure 2 does not appear in any foreign counterpart to the '639 patent.

38. The '639 patent includes claim 13, which was added as part of the 1973 application. Claim 13 requires:

the moist carbon dioxide in the supercritical state containing caffeine is contacted with the active carbon and following said contacting the moist carbon dioxide in supercritical state is recycled to the contacting of coffee with carbon dioxide.

This claim does not cover the system of Figure 1, nor does it cover Dr. Zosel's three autoclave system.

39. In his submission to the Patent Office along with the 1973 application, Dr. Zosel represented that he was the inventor of this improvement.

40. Shortly after SGK learned that Vitzthum et al. had successfully developed a process for the decaffeination of roast coffee with supercritical carbon dioxide, SGK abandoned its 1971 application and filed the 1973 application. SGK presented claims in the 1973 application covering roast coffee decaffeination.

41. Claim 1 of the 1973 application as filed embraced the Vitzthum et al. roast coffee process.

42. The statement that appears in the 1971 application that "[t]he caffeine must be extracted from the raw bean, since extraction from the roasted bean cannot be carried out without loss of aroma" is not a description of a process for the decaffeination of roast coffee.

43. To read the statement as a description of a process for the decaffeination of roast coffee would contradict the entire thrust of the application as is apparent from other statements in the application, such as "almost only caffeine is exclusively removed" (not aroma constituents) and from the repeated and exclusive reference to raw coffee.

44. The 1971 application, and every one of the twenty-two foreign counterparts corresponding to the patent in suit describe and claim only a process for decaffeinated raw or green coffee.

*126 45. SGK also added the following text to the specification of the 1973 application:

[I]n situations when the coffee produced is reconstituted with aromatics by addition thereof after treatment, as for example in the manufacture of instant coffee, caffeine may be extracted from the roasted bean.

This step appears in the Vitzthum et al. patent.

46. The re-addition of aroma is necessary in order to prepare a palatable decaffeinated beverage from roast coffee. Dr. Zosel never did any work in this area. SGK concedes that Dr. Vitzthum et al., not Dr. Zosel, is the inventor of the roast coffee decaffeination process described above.

A. The '639 Patent

47. Shortly after the April 1973 agreement between HAG and SGK, supra, was negotiated, the Patent Office issued a Notice of Allowance finding the claims of the 1971 application for the decaffeination patent in condition for allowance. These claims included all claims that were originally filed; that is, the Patent Office gave SGK every claim it originally asked for. The Examiner had not required any amendments to limit their scope. SGK, however, defaulted in payment of the $112.00 issue fee and in June of 1973 abandoned the application.

48. Instead, SGK refiled the 1971 '639 application as a C-I-P, and amended the specification and claims to cover the work of Dr. Vitzthum et al. and Lipka. As filed, SGK's new claims embraced Lipka's decaffeination system and, by asserting claims directed broadly to "coffee," Dr. Vitzthum's roast coffee process.

49. In the C-I-P, submitted in 1973, Dr. Zosel swore that he was the original and sole inventor of the improvements described and claimed.

50. In November 1973, the improvement patent for caffeine recovery, '619, was approved. It issued in April 1974.

51. There followed a lengthy delay with respect to the '639 patent, which did not issue until April 7, 1981. Some of the additional delays were due to administrative procedures and some to the actions of the applicant. However, the decision not to accept the original patent application in June of 1973 made it certain that the improvement patent would issue first.

52. Defendant could have taken a patent on the allowed claims in parent application 110,428, filed a continuing application with the broader claims, and filed a terminal disclaimer for the broader claims, so that the same would expire with the expiration of the patent issuing on the parent application.

53. One purpose of filing the continuing application was to obtain claims which would cover Dr. Vitzthum's process involving the use of roasted coffee, of which defendant had become cognizant.

54. Another purpose of filing the continuing application was to disclose and claim (in patent claim 13) the embodiment of Figure 2 of the patent, which Lipka at HAG claims was his invention.

55. The broader claims sought in SGK's continuing application covered the decaffeination of roast as well as raw coffee with moist, supercritical carbon dioxide to obtain a palatable coffee beverage.

56. The HAG multistep process was known to the Examiner who oversaw the prosecution of the continuing application, and who cited it as anticipating the claims during the course of such prosecution. Applicant's response was that his invention was prior to HAG's.

57. Claims 1 and 4 of the '639 decaffeination patent are supported by the disclosure of parent application 110,428, filed January 28, 1971. They are, therefore, entitled to an effective filing date of January 28, 1971. The application which matured into the '619 patent was later filed, on May 3, 1972, so that the application itself is not prior art.

58. It was obvious to one of ordinary skill in the art that supercritical carbon dioxide would substantially decaffeinate *127 coffee, as otherwise it would not be claimed in claim 1 of the '619 patent, since the claim was directed to a commercially attractive process and the process would not be commercially attractive without the recovery of substantially decaffeinated coffee, the value of the recovered caffeine being far less than the cost of the coffee beans used.

59. During the prosecution of the 1973 application, the Examiner rejected the claims brought forward from the first application on the basis, inter alia, of double patenting over U.S. Patent 3,806,619, which issued April 23, 1974. He rejected the claims covering roast coffee on the basis of HAG's patent on roast coffee decaffeination (U.S. Patent 3,843,824) and Zosel's British patent, which had issued on an application corresponding to the U.S. parent application. SGK argued that the cited references were not available as prior art because the claims covering roast coffee were supported by the original 1971 application, which pre-dated the references.

60. SGK's arguments that the broad claims were supported by the original application were unreasonable and insupportable.

61. When the Examiner found that this patent presented a double patenting problem, SGK's patent attorney indicated that he would seek his client's permission to file a terminal disclaimer. However, one was not filed.

62. In 1980, the Board of Patent Appeals rejected all of Zosel's claims embracing the decaffeination of roast coffee. The claim to the improvements claimed by Lipka was allowed and appears in the '639 patent as claim 13 and Figure 2. The '639 patent finally issued in April 1981, limited, as the original application had been, to the decaffeination of green coffee.

B. The '619 Patent

63. In 1972, while the 1971 application for the '639 patent was pending, SGK filed another patent application (Ser. No. 249,809, ultimately issued as U.S. Patent 3,806,619) for the recovery of caffeine as a by-product (the "improvement application"). The improvement application, like the earlier filed 1971 '639 application, required the decaffeination of raw coffee using moist, supercritical CO2, but included an extra step the removal of caffeine from the moist, supercritical CO2 so as to recover pure caffeine:

A process for obtaining caffein[e] from green coffee which comprises:
a. contacting moist carbon dioxide in supercritical state with the coffee in a caffein[e] absorption zone for absorption of caffein[e] by moist carbon dioxide [followed by caffeine recovery steps b-i].

64. SGK has repeatedly referred to the '619 patent as a "decaffeination patent" and concedes it is directed to an improvement of the decaffeination process described and claimed in the '639 patent. A fair reading of the '619 patent clearly reveals it is based on a process for the decaffeination of coffee.

65. The inclusion of caffeine recovery in a decaffeination process such as the '639 process is useful because caffeine can be sold as a byproduct. Many companies, such as General Foods, have done so since the 1940's.

66. It makes no economic sense, however, to conceive or implement a process to recover caffeine from coffee beans (regardless of their quality or lack thereof) as a primary product, and discard the resultant decaffeinated coffee.

67. In April 1974, while the C-I-P was still pending, the 1972 improvement application issued as U.S. Pat. 3,806,619. Thus, in 1974 the 17-year period of patent protection began for the '619 patent, fully disclosing in its claims the treatment of raw coffee with moist, supercritical CO2 to remove caffeine and produce decaffeinated coffee.

68. There cannot be double patenting here unless the plaintiff establishes, inter alia, that claims 1 and 4 of the '639 patent are obvious from the claims of the '619 patent, as interpreted by the specification in view of the prior art.

69. Claim 1(a) of the '619 patent anticipates, or at least renders obvious, the inventions of claims 1 and 4 of the '639 patent because every step of claims 1 and 4 *128 of the '639 patent is set forth in claim 1(a) of the '619 patent. Accordingly, the processes of claims 1 and 4 of the '639 patent would have been obvious from the invention set forth in claim 1(a) of the '619 patent.

70. Although the '639 patent was repeatedly rejected by the Patent Office for double patenting, SGK never argued that claim 1 of the patent was unobvious over the claims of the '619 patent.

71. Patent protection for the invention claimed in the '619 patent expired in April 1991. The subsequently filed C-I-P application issued as the '639 patent some seven years after the improvement patent was issued and expires in 1998. It also protects the subject matter of the '619 improvement patent, in that even though the '619 patent has expired, no one can use its claimed invention, at least in any practical sense, without infringing the '639 patent. Thus, the '639 patent effectively extends the term of the '619 patent from 17 to 24 years.

72. The delay in obtaining claims 1 and 4 of the '639 patent, so that it issued after the '619 improvement patent, is solely attributable to SGK.


1. Patents are entitled to statutory protection for seventeen years. 35 U.S.C. § 154 (1988). Thereafter, the public is entitled to use the invention and the claims disclosed therein.

2. Patents and their individual claims are entitled to a presumption of validity. 35 U.S.C. § 282 (1988); see Custom Accessories, Inc. v. Jeffrey-Allan Indus., 807 F.2d 955, 961 (Fed.Cir.1986). The party challenging the patent's validity bears the burden of establishing by clear and convincing evidence that the patent is invalid, Ralston Purina Co. v. Far-Mar-Co, Inc., 772 F.2d 1570, 1573-74 (Fed.Cir. 1985), and this burden is particularly heavy when the claim is based on double patenting. Carman Indus. v. Wahl, 724 F.2d 932, 940 (Fed.Cir.1983). The burden of going forward, however, shifts to the patent holder to adduce evidence of nonobviousness once the challenger establishes a prima facie case of invalidity. Ashland Oil, Inc. v. Delta Resins & Refractories, Inc., 776 F.2d 281, 291-92 (Fed.Cir.1985), cert. denied, 475 U.S. 1017, 106 S. Ct. 1201, 89 L. Ed. 2d 315 (1986).

3. Double patenting bars, of which there are two types, attempt to prevent patent holders from extending the term of their patents for more than the statutorily permissible seventeen years.

4. The first such bar, known as "same invention" double patenting, prevents an applicant from receiving a patent on the same subject matter as a previously issued patent. In re Longi, 759 F.2d 887, 892 (Fed.Cir.1985). This doctrine directly stems from statutory language permitting issuance of patents only for "new and useful" inventions. 35 U.S.C. § 101 (1988). The parties agree, however, that this doctrine is inapplicable in the case at bar.

5. The second bar, which is the focus of this part of the litigation, "prevent[s] the extension of the term of a patent, even when an express statutory basis for the rejection is missing, by prohibiting the issuance of the claims in a second patent not patentably distinct from the claims of the first patent." Longi, 759 F.2d at 892. This judicially created remedy is known as "obviousness-type" double patenting. The public policy rationale behind this doctrine suggests that once the term of a patent expires, the public should be permitted to utilize the claimed invention, as well as variations that would have been obvious to one skilled in the art at the time the initial invention was made, "`taking into account the skill of the art and prior art other than the invention claimed in the issued patent.'" Id. at 892-93 (quoting In re Zickendraht, 319 F.2d 225, 232, 50 CCPA 1529 (1963) (Rich, J., concurring)).

6. The general rule regarding obviousness-type double patenting is that the "claimed invention in the application for the second patent [must not be] obvious from the subject matter of the claims in the first patent." Id. at 893; see also Hartness Int'l, Inc. v. Simplimatic Eng'g Co., 819 F.2d 1100, 1109 (Fed.Cir.1987) (discussing *129 Longi test). This test, therefore, focuses on the pending application (or second patent) and whether the claims of that application (or second patent) are obvious to one skilled in the art from the claims of the previously issued patent to that same applicant.

7. At the summary judgment stage, defendant suggested (and it continues to do so) that the test actually was whether the claims of the second patent were obvious from the claims of the first patent and vice-versa. Thus, such an approach adds the additional requirement of finding that the claims of the first patent are obvious to one skilled in the art from the claims of the second patent. While it is true that there is some authority for such an approach, see Carman Indus. v. Wahl, 724 F.2d 932, 940 (Fed.Cir.1983), this test focuses on situations where the patents at issue involve different statutory classifications. For example, in the Carman Industries case, one patent was for the design of an invention while the other related to its utility. In the case at bar, where two process patents are at issue, there is no authority for making the vice-versa test the general rule. Cf. Longi, 759 F.2d at 893 n. 5 (suggesting that Carman Industries test might properly be limited to cases involving patents in different statutory classifications).

8. At trial, defendant has modified the focus of its argument in light of our ruling on the vice-versa test at the summary judgment stage. Defendant now argues that in situations such as the case at bar, where patents do not issue in the order in which they are filed, the question is whether the second-filed, first-issued patent is obvious from the first-filed, second-issued patent. See In re Borah, 354 F.2d 1009, 1018, 53 CCPA 800 (1966).

9. While we agree that Borah announces such a test in the context of a first-filed generic application issuing after a second-filed species or improvement application, a situation clearly contemplated by defendant in the processing of its applications, there is one crucial factor upon which Borah turned that is not present in the matter before us. Specifically, the court noted that rejections by the Patent Office and Board of Patent Appeals led to the delayed resolution of the first-filed application. Id. 354 F.2d at 1015-16. Defendant suggests that this is not a relevant part of the inquiry, but we disagree. As stated by the Third Circuit Court of Appeals in Pierce v. Allen B. Du Mont Laboratories, 297 F.2d 323, 328-29 (3d Cir. 1961), cert. denied, 371 U.S. 814, 83 S. Ct. 24, 9 L. Ed. 2d 55 (1962), while administrative delay may justify the existence of both patents, "a considered election to postpone ... acquisition of the broader [patent]" should not be tolerated. Simply, even if the Borah rule requires an inquiry contrary to that of the general rule, Borah will not apply if intentional delay practiced by the applicant leads to the issuance of the patents in the opposite order from which they were filed. See 3 D. Chisum, Patents § 9.03, at 9-34 (1991) (Borah will not apply if inexcusable delay by applicant causes reverse issuance of patents).

10. To summarize, the usual inquiry regarding obviousness-type double patenting is whether the claims of the second patent are obvious from the claims of the first. However, if the second patent to issue was filed first, the test becomes whether the claims of the first-issued patent are obvious from those of the second-issued patent. An exception to this latter rule exists when the delay in issuing the first-filed patent is caused by the applicant, be it with sinister or innocent motives, as opposed to mere administrative delay. If this exception applies, the first test governs, and it is this first test that controls the issues at hand.

11. The parties agree that the decaffeination patent is entitled to at least a January 1971 filing date, notwithstanding the subsequent filing of the C-I-P application that ultimately matured into the '639 decaffeination patent at issue. This agreement is based on the fact that the claims of the decaffeination patent issuing in 1981 are precisely those filed in the initial, pre-C-I-P, application. In fact, these claims could have been accepted in 1973. We note, however, that the Patent Examiner, in rejecting *130 certain claims of the C-I-P application, concluded that defendant was not entitled to this 1971 filing date, but rather, was entitled to a filing date of May 1973, the date the C-I-P application was filed. Although the Board of Patent Appeals reversed the examiner on the issue of double patenting, it did not address the question of filing dates. Since the parties have agreed on this issue, we will accept defendant's entitlement to a January 1971 filing date for its decaffeination patent.

12. The parties also agree that the claims of the caffeine recovery patent are not obvious from the claims of the decaffeination patent. Thus, if the rule announced in Borah were to apply, defendant would prevail on the question of obviousness type double-patenting since the second-filed patent is not obvious from the first-filed patent.

13. As noted, however, we consider the Borah test to be limited precisely to the facts before that court. Thus, in a situation where the applicant does all that is required and administrative delay in the Patent Office causes the reverse issuance of the two patents, the applicant should not be penalized. Administrative delay in such situations often occurs because the broad generic patent does not enjoy as speedy a prosecution as a narrow improvement patent. In the case at bar, however, there is no question but that defendant could have accepted its decaffeination patent in 1973, before the improvement patent was issued. Instead, SGK decided to abandon the allowed claims and file a C-I-P application, which it turns out was less than half a year before the Patent Office notified it of the approval of the caffeine recovery patent. Thus, administrative delay was not a factor in the issuance of the decaffeination patent after the 1974 issuance of the caffeine recovery patent. Rather, it was a deliberate decision by defendant and its counsel that led to this result. As recognized by the court in Pierce, this:

is not a case where administrative action prevented the inventor from having his more comprehensive application patented first, but rather where the choice has been the inventor's, dictated by his own judgment of possible economic advantage. We think these circumstances are not such as to justify an exception to the general rule and policy against double patenting.

Pierce, 297 F.2d at 329.

14. We note that between 1974, when the '619 patent issued, and 1981, when the decaffeination patent ultimately issued, there was both administrative delay and delay caused by the applicant. This is of no consequence, however, since our inquiry must focus only on the reasons why the decaffeination patent did not issue pre-1974, not what happened after 1974.

15. As we previously stated, the reasons why defendant did not accept the patent on the original application are immaterial. Even if defendant's conduct was entirely within the law and within standard patent law practice, the fact remains that it was a decision by defendant that led to this litigation. We note, however, that it is clear that defendant filed the C-I-P application in order to obtain priority on the decaffeination process involving roasted beans over the patent of HAG, as invented by Dr. Vitzthum et al. (In addition, defendant was interested in claiming the improvements embodied in Figure 2.) Without such priority, defendant would not have been able to obtain a patent on the process for roasted beans since it would have been anticipated by the HAG patent. Thus, a C-I-P application allegedly was defendant's only recourse. The problem with this argument, however, is that there was no support for defendant's claims that its 1971 application covered both raw and roasted beans. In fact, the application specifically eschews such an interpretation since it emphatically states that the process cannot be conducted with roasted beans without loss of aroma. Again, while we do not challenge defendant's right to file the C-I-P application, this was the factor that led to the issuance of the decaffeination patent after the issuance of the caffeine recovery patent. Defendant made its bed, and now must lie in it.

16. To avoid the problems it now faces, defendant could simply have filed a terminal *131 disclaimer. Such a filing would have permitted SGK to retain both the '619 and '639 patents, except they both would have expired at the same time (1991 in this case). 35 U.S.C. § 253 (1988). This means of avoiding double patenting rejections was well established in the early 1970's, not-withstanding the contrary conclusion of defendant's expert. In fact, in the case of In re Ziegler, 443 F.2d 1211, 1215, 58 CCPA 1296 (1971), a case in which the patent applicant was represented by the same counsel representing defendant in the case at bar, the court intimated that the filing of a terminal disclaimer was a means of avoiding a double patenting rejection on grounds of extension of monopoly. Moreover, even if the patent bar did not generally utilize such disclaimers at the time when the C-I-P application was filed, as defendant's expert suggests, defendant could have filed such a document at any time up to and including 1981, when the decaffeination patent issued. Although defendant was certainly not required to file such a disclaimer, it was an available means of avoiding having the entire patent disallowed since it would have insured the validity of the patent at least until 1991, when the caffeine recovery patent expired.

17. This brings us to the crucial inquiry in this case, which is whether certain claims of the decaffeination patent are obvious from those of the caffeine recovery patent. Claim 1(a) if the recovery patent states that it is "[a] process for obtaining caffeine from coffee which comprises contacting moist carbon dioxide in supercritical state with the coffee in a caffein[e] absorption zone for absorption of caffein[e] by moist carbon dioxide." The remaining claims of this patent are not relevant to our inquiry since they discuss the further steps of extracting the caffeine from the caffeine/carbon dioxide mixture. Turning now to the allegedly obvious claims of the decaffeination patent, claim 1 states that it is:

A process for the decaffeination of raw coffee which comprises contacting the raw coffee with water-moist carbon dioxide above its critical temperature and critical pressure to effect removal of caffeine therefrom and recovering a substantially decaffeinated coffee, the amount of water in the carbon dioxide being sufficient to effectuate said removal of the caffeine from the coffee.

Claim 4 further states that it is "[a] process as claimed in claim 1 in which the contact with the moist carbon dioxide is effected for a period of from 5 to 30 hours." While plaintiff argues that both claims one and four of the decaffeination patent are obvious from claim 1(a) of the recovery patent, the focus during the trial and in the parties' written submissions has been upon the obviousness of claim 1 only. Since a finding that claim 1 is obvious would render the decaffeination patent ineffectual, we also will focus on that claim, rather than claim 4.

18. In resolving the question of obviousness, the Federal Circuit has stated that the inquiry to be conducted is similar to that required for determinations under 35 U.S.C. § 103 (1988), which is a separate statutory requirement of nonobviousness. See Longi, 759 F.2d at 892 n. 4. However, in this case the only relevant inquiry is whether the claims of the first patent "disclose[] to one of ordinary skill in the art" the claims of the second patent.

19. In the case at bar, Dr. Adler's uncontroverted testimony establishes that one skilled in the art at the time of the invention of the caffeine recovery patent would have known that claim 1(a) involves the decaffeination of coffee. Claim 1(a) discloses that the process requires contacting supercritical moist carbon dioxide with coffee "for absorption of caffein[e]." Claim 1 of the decaffeination patent, in turn, states that moist supercritical carbon dioxide must contact the coffee "to effect removal of caffeine therefrom." There is no doubt that for these purposes, absorption of caffeine is synonymous with removal of caffeine.

20. The principal argument raised by defendant is that claim 1 of the decaffeination patent goes on to state that the removal of caffeine permits the recovery of "a substantially decaffeinated coffee." Claim 1(a), defendant argues, can be practiced *132 without the recovery of any coffee, not to mention a substantially decaffeinated coffee. For example, defendant argues that one might decaffeinate a coffee bean only partially to obtain a fifty percent (as opposed to the standard ninety-seven percent) decaffeinated coffee. Moreover, defendant contends that one might decaffeinate a Robusta bean to upgrade it to an Arabica bean, which generally contains less caffeine. Therefore, defendant claims, since not every use of the recovery patent will violate the decaffeination patent, there is no threat that its patent coverage will extend beyond seventeen years.

21. If patents stand in a genus/species or generic/improvement relationship, it is said that the genus or generic patent dominates the species or improvement patent. That is, every practice of the improvement patent necessarily entails utilization of the generic patent as well. In the case at bar, the patents do not precisely fit within such categorizations since one can theoretically practice the improvement patent without practicing the generic patent. However, the cases involving such relationships remain apposite since the prosecution history of the recovery patent makes it clear that it was meant to be an improvement over the decaffeination patent. In addition, while claim 1(a) of the recovery patent does not address obtaining a substantially decaffeinated coffee, there is no doubt that the patent is directed towards an economically feasible process. In this regard, Dr. Adler's testimony must be credited. He testified that in light of the relatively low content of caffeine in coffee beans, as well as the expense of obtaining coffee beans in the first place, it would be preposterous to suggest that one would perform the recovery patent solely to recover caffeine, without then utilizing the decaffeinated beans to prepare a palatable coffee.

22. The more pertinent question, however, is whether claim 1(a) of the recovery patent makes decaffeination as defined in claim 1 of the decaffeination patent obvious. Clearly it does. One skilled in the art would be aware that obtaining a substantially decaffeinated coffee was an obvious possibility for one conducting the recovery patent.

23. Defendant obtained patent protection for its improvement patent for seventeen years. Now that the patent has expired, the public should be permitted to practice the claims covered by that patent. Notwithstanding the remote possibilities proffered by defendant, virtually every practicable use of the now-expired caffeine recovery patent entails use of the decaffeination patent, and the recovery of a substantially decaffeinated coffee. Thus, one practicing the caffeine recovery patent today would run afoul of the decaffeination patent in most instances. Such a result is precisely what the double patenting bar attempts to avoid.

24. For all the foregoing reasons, we find that claim 1 of the decaffeination patent is obvious from claim 1(a) of the caffeine recovery patent. Consequently, plaintiff is entitled to a judgment declaring claim 1 of the decaffeination patent invalid on grounds of obviousness-type double patenting.

25. Plaintiff concedes it has no rights to recover royalties paid to defendant prior to the institution of this litigation. However, plaintiff does argue that since it continued to make royalty payments under protest after this action was commenced it is entitled to recover these funds. A hearing on this issue will be necessary. The parties will submit papers in accordance with the court's general rules governing motions.