Plaintiff has appealed from the judgment (unpublished) of the United States District Court in favor of defendants that Claim 1 of U.S. Patent No. 3,268,636 ("636 patent")1 has not been infringed; that said claim is invalid over the prior art under 35 U.S.C. § 102(e) (anticipation) and under 35 U.S.C. § 103 (obviousness); and that the 636 patent is invalid in its entirety for failure to disclose the best mode contemplated to carry out the invention as required by 35 U.S.C. § 112, first paragraph.2 We affirm on the best mode issue, so it is unnecessary to reach the sections 102(e) and 103 issues. The infringement issue thus becomes moot.

Appellant is the assignee of the inventor, Richard G. Angell, a scientist and one of its employees. Angell's invention is a process3 for molding foamed thermoplastic products that are useful as substitutes for wood. These have a high density shell, an integral low density cellular core, and a high strength to weight ratio. (The best mode issue involves a valve and an extruder used in apparatus to perform the process.)

1. Process for molding foamed thermoplastic articles which comprises the steps of

(a) melting a mixture of a blowing agent and a foamable thermoplastic material in an extruder at a temperature above the foaming temperature of said blowing agent and at a pressure above the foaming pressure thereof;

(b) extruding the resulting molten mixture into an expanding accumulation zone while maintaining said mixture therein in the molten state and at a pressure above the foaming pressure thereof;

(c) establishing communication between said accumulation zone and a mold maintained at a pressure no greater than the foaming pressure of said molten mixture;

(d) rapidly forcing said molten mixture from said accumulation zone into said mold whereby the pressure differential between said accumulation zone and said mold causes said mixture to rapidly expand in said mold.

Referring to the drawings, Angell's specification explains the process as follows:

FIG. 2.

NOTE: OPINION CONTAINS TABLE OR OTHER DATA THAT IS NOT VIEWABLE

As shown in FIGURE 2, the extruder charges molten polymer 26, through a valve 21, and into an accumulator 20. The molten polymer 26, must be mixed with a blowing agent prior to the charging of the polymer into the mold.

. . . .

While the various types of blowing agents which can be employed to produce a foamed product can be mixed with the plastic material at various stages in the process, it is advantageous to add liquid or gaseous agents directly to the polymeric material, in the extruder, while the material is in a molten state, in order to obtain a uniform dispersal, of the agent within the molten plastic without employing additional mixing apparatus. Similarly, a decomposable chemical blowing agent is advantageously pre-mixed with the polymer prior to the charging of the polymer into the extruder.

. . . .

Alternatively, the accumulator or the barrel of the extruder, can if desired, contain means, not shown, for the direct addition thereto of a liquid or gaseous blowing agent.

. . . .

The accumulator 20, as shown in FIGURE 2, includes a piston 22, which divides the accumulator into two chambers and which resists the filling of the accumulator because of gas under pressure in the chamber 24 of the accumulator 20. The gas pressure within the chamber 24 is at least equal to the pressure necessary to maintain the mixture of thermoplastic material and gas in an unfoamed state. . . .

Since the charging of the accumulator with molten polymer 26, by the extruder, is opposed by the piston 22, the filling of the accumulator cannot start until the extruder pressure exceeds the back-pressure of the piston 22. As the piston 22 moves in the accumulator 20, the gas volume in the chamber 24 is decreased and the gas pressure is increased. The extruder pressure must continuously exceed the increasing piston pressure during the filling operation.

The amount of material charged into the accumulator chamber 26 is a function of the increase in pressure in the chamber 24, and therefore, a pressure gauge which communicates with chamber 24 may be directly calibrated in terms of amount of material charged to the accumulator.

When the desired amount of thermoplastic material has been charged into accumulator 20, the position of valve 21 is changed to the position as shown in FIGURE 3, whereby communication is suddenly established between the chamber 26 of the accumulator 20, and the cavity of the mold 28.

The back-pressure on the piston 22, in combination with the gas contained within the thermoplastic material serves to rapidly fill the mold. . . .

In April of 1963, prior to Angell's filing his application which matured into the involved patent, appellant commenced "scale-up" operations, preliminary to commercial production, in a pilot plant using Angell's process. Testimony by Theodore Loeser, who was directly responsible for the pilot plant and was Angell's supervisor, shows that the valve used was not the one disclosed in the 636 patent, but one which "did not leave in the discharge passages of the machine any slug of semifoamed or unfoamed plastic that would be injected into the next part to be made"; and that the valve so used "was a better valve than is shown in the 636 patent."

Although he testified that, as of the period June-July, 1963, he believed the valve shown in the 636 patent "was the best for practicing my process," Angell acknowledged that prior to June 1, 1963, he had recognized that material being carried into the mold from the previous shot was "undesirable"; and that the pilot plant "eliminates this problem." In a report dated June 1, 1963, entitled "STRUCTURAL FOAM MOLDING," Angell described the pilot plant unit as including: "A unique valve system connected to the valve that is self purging so that no material is carried into the mold from the previous shot." Angell's notebook page 96, dated December 13, 1962, subject "Melt Foam Process Improved Apparatus," states the object of experiment 6413-26 as follows:

To design a piece of equipment that will hold material and force it into the mold and then purge out any remaining material.

The idea is to eliminate the solid pieces of material that result from foam collapsing in the valve of the present equipment. This can be accomplished by running a rod through the entire apparatus . . . .

Angell admitted that the valve shown in the drawing on said page 96 is the same as the one used in the pilot plant and essentially the same as one shown in his patent application No. 364,382, filed May 4, 1964. This application was subsequently abandoned in favor of Angell's application No. 556,103, filed June 8, 1966, as a continuation in part of three applications (including No. 291,898 (636 patent) and No. 364,382). No. 556,103 matured into U.S. Patent No. 3,436,446 ("446 patent"). Angell's testimony shows that the valve in the above-referred-to notebook drawing is "essentially the same" as the one shown in a drawing in the 446 patent. Loeser readily identified the valve originally used in the pilot plant as the same as the one shown in said drawing in the 446 patent.

The following statement appears in Angell's 446 patent:

Perhaps the most serious problems met with injection molding of foamed articles is (sic) the formation of a sprue on the uniformly porous structure and the formation in the equipment of an unfoamed slug which enters the mold on the next succeeding molding cycle creating an undesirable unfoamed, solid region in what should otherwise be a completely foamed article. This unfoamed slug is generally formed in the valve which connects the injection zone and the mold and/or in the conduits leading to and from the valve.

Angell testified that, because of "problems" with his new valve, "we did not have a satisfactory (pilot plant) operation in the spring of 1963." He stated that there were "problems" with the valve rod "sticking" and "severe problems" with "leakage"; that the reliability of the valve was "poor"; and that there were "many failures and periods of inoperability."4 Loeser testified that the pilot plant was operating "adequately to allow us to operate the Angell process; but we had a lot of difficulties with it, really in the form of mechanical problems . . . ." However, the only contemporaneous record evidence cited by appellant of "problems" prior to the time Angell executed his 636 patent application on or about June 19, 1963, is a Union Carbide work order dated April 22, 1963, requesting heat treatment work on the valve rod and showing an amount of $22.50.

With respect to the pilot plant operation, the following comments by Angell appear in his notebook:

The equipment works fine and boxes can be molded easily. (April 19, 1963)

Good boxes are being made. (April 22, 1963, and May 7, 1963)

Good bowling pins are being made on the pilot plant equipment. (May 15, 1963)

The equipment is working well. (May 21, 1963)

A melt temperature of 330o F makes good boxes with very few voids. (June 6, 1963)

In reducing his invention to practice, Angell used a single-stage screw extruder. However, the extruder installed in appellant's pilot plant was a two-stage screw extruder manufactured by the Egan Company. This, according to Angell's deposition, "was made special to order with extra expense." Angell stated that it "cost more than several thousand dollars," and that it was made according to a design "we recommended." However, he denied that this was done because he thought it would do a better job in practicing his process than the single-stage screw extruder used in his first experiments. He said that he did not think that the Egan extruder "necessarily" did a better job than the single-stage screw extruder and that "(t)here is more than one specific design that will satisfy the requirement" to practice the process. Further, he left the impression that the only "job" an extruder must perform in his process is to "mix and melt the (thermoplastic material and blowing agent) sufficiently."

Angell's testimony on cross-examination shows that not just any screw extruder, "regardless of the particular design of its flights and its dimensional properties," would satisfactorily work in his process; and he acknowledged that from his experience he had found that "some screw extruders do a better job of mixing than others." Angell agreed that "the purpose of this two-stage (Egan) screw with its pressure maximum in the middle is to inhibit the backflow of gas," and he stated that the Egan screw was ordered "so we could inject gas but we did encounter problems of gas leaking back through that screw several times." Referring to the Egan extruder, Angell deposed that he did not believe a standard screw extruder had "this kind of fly vent"; that "(w)ith this construction we were able to melt a plastic material . . . before we injected gas at the injection port"; and that he considered this to be a "desirable achievement." He admitted that there is no information that the reader of the patent specification is given regarding the physical properties or design of the extruder to be used in his process.

In a report dated February 4, 1964, by one of Angell's associates, it is stated:

In the first half of 1963, as the next step in the structural foam program, a pilot plant was designed and installed. This consisted of an extruder, accumulator and mold handling equipment. This scale-up was based on previous experience, including use of a chemical blowing agent to obtain foaming. However, the extruder was additionally designed to permit direct injection of gas through the barrel into the plastic melt by having a hole drilled through the barrel approximately half way down its length and corresponding to a decompression zone in a 2-stage screw. The extruder was a 3 1/2 Egan, with a 24:1 length to diameter ratio.

The specially designed screw is shown in Figure . . .. The injection port was located at the 14th flight. This location was chosen to be far enough forward to permit the plastic to have melted sufficiently to provide a melt seal against gas working back through the hopper, and yet far enough back to leave enough screw length left to accomplish mixing of the gas and plastic. (Emphasis supplied.)

Additionally, Angell's supervisor, Loeser, testified as follows:

Q. Now, is it also a fact that in the pilot plant that was in operation by April or May of 1963 that a screw extruder was employed? A. Yes, that's true.

Q. In fact, that screw extruder had been engineered at your laboratory in Boundbrook and made on a special order by the Egan Company? A. Yes, I believe that's correct.

Q. That special extruder was constructed in such a way, was it not, that it would prevent blowing gas from backing up and escaping through the material hopper at the input? A. Yes, I believe that was the case.

. . . .

Q. But in fact the reason that this engineering work was done and the two-stage Egan machine with a dam in it was ordered is to overcome this problem of gas leakage, wasn't it? A. Yes, I guess that's correct.

Appellees in their brief state that between 1966 and time of trial in 1975 appellant had signed up fifty-four licensees, each of which had paid for a "know-how package" (in addition to non-exclusive rights under the 636 and 446 patents) containing extensive written specifications, including the "secret" extruder design of appellant's pilot plant. In their reply brief, appellants have not traversed the statement, which has substantial support in the evidence of record.

The district court concluded that, as a matter of law, appellant's 636 patent is invalid "for failure of the inventor (Angell) to disclose in full, clear, concise and exact terms the best mode contemplated for carrying out the invention." Appellant takes exception to the court's use of the phrase "disclose in full, clear, concise and exact terms" rather than the words "set forth," which appear in the best mode portion of paragraph 1 of 35 U.S.C. § 112. Although the statute does not prescribe the standard according to which the best mode is to be "set forth," it would seem that it should be fully, clearly, and precisely disclosed if the public is to receive the benefits intended to result from the grant of the patent monopoly. Flick-Reedy Corp. v. Hydro-Line Manufacturing Co., 351 F.2d 546, 551, 146 U.S.P.Q. (BNA) 694, 697 (CA 7 1965), cert. denied, 383 U.S. 958, 86 S. Ct. 1222, 16 L. Ed. 2d 301, 148 U.S.P.Q. (BNA) 771 (1966). Nevertheless, any error possibly arising from the court's use of the disputed phrase would be harmless, since it is clear that Angell's 636 patent application did not set forth in any manner what the district court determined to be the best mode contemplated by Angell.

The primary question is whether, bearing in mind the presumption of validity of the Angell patent (35 U.S.C. § 282), either the district court's finding that the valve used in appellant's pilot plant was the best mode contemplated by Angell for carrying out his invention at the time he filed his 636 patent application, or the district court's similar finding with respect to the extruder used in appellant's pilot plant, is supported by substantial evidence and is not clearly erroneous. Life Time Doors, Inc. v. Walled Lake Door Co., 505 F.2d 1165, 184 U.S.P.Q. (BNA) 1 (CA 6 1974).

The district court found that Angell's conception involved using a valve to establish communication between the accumulator and the mold; that, as first devised, his valve permitted the formulation of an unfoamed slug of solid plastic which, on each subsequent molding operation or cycle, was injected into the mold along with the new charge of foaming molten plastic; and that the slug presumably had some effect on the structural integrity of the foamed product and was thus thought to be unsatisfactory. The district court further found that by the time Angell filed his patent application, he had devised an improved valve which cured the problem of the unwanted slug; and that he did not disclose this valve in his application even though he conceived of the improved valve as an integral part of the best mode of practicing his process.5

We hold that these findings are supported by substantial evidence and are not clearly erroneous. Absent overriding economic or other considerations, not apparent here, it would be expected that what Angell contemplated as the best valve for practicing his process would have been installed in appellant's pilot plant. Although Angell and Loeser testified that there were "problems" with the valve used in the pilot plant, we note that there is contemporaneous record evidence of only one such instance prior to the date that Angell signed his application, and this had to do with a rod used in the valve. The fact that there were "problems" with a component part (which could be repaired or replaced) of what Angell described as a "unique valve system . . . that is self purging so that no material is carried into the mold from the previous shot," and what Loeser stated "was a better valve than is shown in the 636 patent," would be unlikely to alter Angell's contemplation of the valve installed in the pilot plant. This is underscored by the seriousness of the slug problem (described in the 446 patent) which the 636 patent valve occasioned and which the pilot plant valve eliminated.

Although the parties have concentrated considerable conflicting evidence and argument over the "problems" issue, we do not consider such evidence as it relates to "problems" that occurred after July 1, 1963, to be helpful in determining the state of mind of the inventor on July 1, 1963. On the other hand, evidence showing that the valve installed in the pilot plant was carried forward into a patent application filed May 4, 1964, and ultimately into the 446 patent tends to support the finding that Angell contemplated this valve as the best mode for practicing his process at the time he filed his 636 patent application.

The district court found that Angell's 636 patent application does not disclose with precision the kind of extruder to be used in his process, that the extruder is illustrated with a blank box, and that the best mode contemplated by Angell to practice his invention (which he did not disclose) requires a particular kind of screw extruder having peculiar characteristics, i.e., having flights of a certain shape, length, and number to mix and melt the plastic and blowing agent in a proper manner for the product in mind.

We hold that these findings are supported by substantial evidence and are not clearly erroneous. Although Angell testified that not just any extruder would satisfactorily work in his process, use of a blank box, without more, would not necessarily defeat his patent. It could be argued that one of ordinary skill in the plastics molding art would be able, without undue experimentation, to select an extruder that would work. Cf. In re Gunn, 537 F.2d 1123, 190 U.S.P.Q. (BNA) 402 (CCPA 1976). This would satisfy the enablement requirement of 35 U.S.C. § 112, first paragraph. However, the question here involves the best mode requirement, and Angell's own testimony shows that some extruders do a better job than others. Clearly, this case does not come within what has been termed an "exception" to the best mode requirement where the " best mode" is, in fact, the basic teaching and actual invention claimed. Ziegler v. Phillips Petroleum Co., 483 F.2d 858, 872, 177 U.S.P.Q. (BNA) 481, 489 (CA 5), cert. denied, 414 U.S. 1079, 94 S. Ct. 597, 38 L. Ed. 2d 485, 180 U.S.P.Q. (BNA) I (1973).

As in the case of Angell's "unique valve system," absent overriding economic or other considerations, not apparent here, it would be expected that what was contemplated by Angell as the best extruder for practicing his process would have been installed in appellant's pilot plant. That it was specially designed and "cost more than several thousand dollars" is a basis for inferring that Angell did not regard any extruder that would work as the best mode for practicing his process at the time he filed his 636 patent application. The facts of this case readily distinguish it from Benger Laboratories Ltd. v. R. K. Laros Co., 209 F. Supp. 639, 135 U.S.P.Q. (BNA) 11 (E.D.Pa.1962), aff'd 317 F.2d 455, 137 U.S.P.Q. (BNA) 693 (CA 3), cert. denied, 375 U.S. 833, 84 S. Ct. 69, 11 L. Ed. 2d 64, 139 U.S.P.Q. (BNA) 566 (1963), where there was an active in-house dispute over which of two methods was "better" at the time the patent application was filed, a dispute which was not settled until nearly two years later. Finally, appellant's argument that the Egan extruder was not "crucial" to practicing Angell's process is not responsive to the best mode question.

Appellant argues that "(t)here is no evidence that Angell was aware of any unique function performed by the Egan two-stage screw extruder which could not be performed by other screw extruders on the market." However, this argument is refuted by Angell's testimony, earlier described, that the Egan extruder was ordered "so we could inject gas," and that he considered this a "desirable achievement."

Appellant also points to Angell's testimony that there were gas leakage problems with the Egan extruder. At the same time, appellant notes that the first run utilizing gas instead of fluid or chemical blowing agents with the Egan extruder did not occur until June 25, 1963, "six days after Angell executed the application on June 19, 1963." As in the case of Angell's improved valve, evidence relating to such "problems" appears to have little relevance to the state of mind of the inventor on July 1, 1963. The subsequent use of the design of the specially ordered Egan extruder in appellant's "know-how package" does, on the other hand, lend support to the district court's finding that this extruder was the best mode contemplated by Angell for practicing his invention.

Appellant argues that the best mode requirement is satisfied "when the inventor acts in good faith and discloses what he believes the best mode to be." It cites the district court's statement in Benger Laboratories Ltd. v. R. K. Laros Co., supra at 644, 135 U.S.P.Q. (BNA) at 15, that:

A patentee must disclose the best method known to him to carry out the invention. Even if there is a better method, his failure to disclose it will not invalidate his patent if he does not know of it or if he does not appreciate that it is the best method. It is enough that he act in good faith in his patent disclosure.

Appellant then points to the following finding of fact by the court below:

29) Notwithstanding the failure to disclose the best mode, the Court finds that plaintiff acted in good faith, albeit somewhat guarded; that it was not guilty of any fraud or other unconscionable or inequitable conduct; and that therefore, there are no facts making this an exceptional case, as that term is used in 35 U.S.C., § 285.6

The matter of attorney fees is not before us on this appeal, so the above finding and the court's later statement that "the Court declines to exercise discretion to make such an award" need not be reviewed.

We would point out, however, that there is no statutory basis for evaluating the failure to comply with the best mode requirement according to the same standard as that used in determining that a case is "exceptional" for the purpose of awarding attorney fees under 35 U.S.C. § 285. Thus, attorney fees may be denied even though there has been a lack of candor on the part of a patentee in his dealings with the Patent and Trademark Office. Indiana General Corp. v. Krystinel Corp., 421 F.2d 1023, 164 U.S.P.Q. (BNA) 321 (CA 2), cert. denied, 398 U.S. 928, 90 S. Ct. 1820, 26 L. Ed. 2d 91, 165 U.S.P.Q. (BNA) 609 (1970).

Appellant also argues that there must be evidence of deliberate withholding7 of something critical pertaining to the best mode. It cites Flick-Reedy Corp. v. Hydro-Line Manufacturing Co., supra. There the Seventh Circuit held that the district court's findings concerning the inventor's withholding of specific information on a "special tool" needed for an essential element in the claims were not clearly erroneous and that the court's conclusion of invalidity because of failure to comply with the best mode requirement was supported by substantial evidence. The district court had concluded that the involved patent was invalid and unenforceable because of the patentee's conduct in maintaining the tool a trade secret. Appellant also cites In re Gay, 309 F.2d 769, 772, 50 CCPA 725, 731, 135 U.S.P.Q. (BNA) 311, 315 (1962), for the following statement:

Manifestly, the sole purpose of this latter requirement (best mode) is to restrain inventors from applying for patents while at the same time concealing from the public preferred embodiments of their inventions which they have in fact conceived.

See also In re Bosy, 360 F.2d 972, 976, 53 CCPA 1231, 1236, 149 U.S.P.Q. (BNA) 789, 792 (1966).

We have no quarrel with the results in the above-cited cases. However, we are not persuaded that the failure to set forth the best mode contemplated by an inventor of carrying out his invention must rise to the level of active concealment or grossly inequitable conduct in order to warrant invalidation of a patent. Thus, in Dale Electronics, Inc. v. R.C.L. Electronics, Inc., 488 F.2d 382, 389, 180 U.S.P.Q. (BNA) 225, 230 (CA 1 1973), the court said:

We do not accept the proposition that where an inventor, as here, clearly knows a specific material that will make possible the successful reproduction of the effects claimed by his patent, but does not disclose it, speaking instead in terms of broad categories, he may nevertheless be considered as having described the best mode contemplated by him. . . . Unintentional obtuseness or obfuscation might be a reason not to penalize someone; we do not see it as a reason for granting a seventeen year monopoly.

See Frantz Manufacturing Co. v. Phenix Manufacturing Co., 457 F.2d 314, 325, 173 U.S.P.Q. (BNA) 266, 274 (CA 7 1972); Columbia Broadcasting System v. Sylvania Electric Products, Inc., 415 F.2d 719, 725, 162 U.S.P.Q. (BNA) 577, 580 (CA 1 1969).8

In view of all the foregoing, the judgment of the district court that appellant's 636 patent is invalid for failure to disclose the best mode as required by 35 U.S.C. § 112, first paragraph, is affirmed.