436 F.2d 1046
Application of Alfred LANDGRAF, Hans Nienburg, Norbert Loesch, Max Appl and Diethard Francke.
Patent Appeal No. 8403.
United States Court of Customs and Patent Appeals.
February 11, 1971.
Herbert B. Keil, Chicago, Ill. (Johnston, Root, O'Keeffe, Keil, Thompson & Shurtleff, Chicago, Ill.), attorney of record, for appellant.
S. Wm. Cochran, Washington, D. C., for the Commissioner of Patents. L. Randall, Bethesda, Md., of counsel.
Before RICH, ALMOND, BALDWIN and LANE, Judges, and NEWMAN, Judge, United States Customs Court, sitting by designation.
The sole issue in this appeal is whether the Patent Office Board of Appeals committed error in affirming the rejection of all the claims in appellants' application1 as being obvious under 35 U.S.C. § 103 in view of the prior art.
The subject matter sought to be patented relates to the well known OXO process of producing aldehydes and alcohols from the catalyzed reaction of carbon monoxide and hydrogen with olefinic hydrocarbons. According to the specification,
[r]eaction of olefins with carbon monoxide and hydrogen to form aldehydes and alcohols having one more carbon atom than the initial olefins proceeds exothermically. In carrying out this reaction continuously on an industrial scale it is therefore necessary to distribute the heat of reaction through the reaction mixture to avoid local temperature peaks which may lead to undesirable side reactions and secondary reactions. It has therefore already been proposed to carry out the reaction in great dilution. For this purpose circulation equipment may be used and the initial olefin, carbon monoxide and hydrogen (and also as a rule the catalyst) are passed into the reaction product already formed so that the said substances are distributed therein and are thus present in low concentration.
* * * * * *
We have now found that oxygen-containing compounds, such as aldehydes and alcohols, may be advantageously prepared continuously by reaction of olefins with carbon monoxide and hydrogen in the presence of cobalt carbonyl compounds at elevated temperature and superatmospheric pressure in the liquid phase in a circulation system, by introducing the liquid reactants at high linear velocity into the circulation system so that circulation of the liquid phase is effected largely by transfer of momentum.
Claim 1 is representative of the claims on appeal:
1. In a process for the production of oxygen-containing compounds selected from the class consisting of aldehydes and alcohols by reaction of olefins in the liquid phase with carbon monoxide and hydrogen in the presence of a cobalt carbonyl compound at elevated temperature and at superatmospheric pressure, the improvement which comprises introducing the liquid reactants at a high linear velocity into a reaction vessel so that the circulation of the liquid phase in said vessel is effected to the extent of 10 to 80% by transfer of momentum.
THE PRIOR ART
Throughout the prosecution of this application before the Patent Office, the appealed claims were consistently rejected as being obvious over the disclosures in two2 prior art references, Tramm et al.3 and a British patent4 referred to in the Patent Office proceeding (and hereinafter) as "F.H.A." (using the first three initials of the assignee, a German Corporation). The Tramm et al. patent discloses the basic OXO synthesis using as catalysts aqueous solutions of cobalt salts such as those recited in the instant claims. The F.H.A. reference is directed to solving the same problem of excessive local heat build-up in the OXO reaction as is the process claimed by appellants. The reference disclosure contains an extensive discussion of prior art methods of solving the problem of local superheating. Listed are such things as the use of high boiling solvents, the introduction of reactants at many places along the reaction zone, employment of very long and narrow reaction chambers, circulating the gas at high speed and recycling the liquid product in various ways. For one reason or another, such methods are stated to be unsatisfactory. The improvement taught in the F.H.A. patent involves creating turbulence in the reaction zone by introducing the gaseous starting materials at the bottom end of the reaction zone "at a high rate per unit time". "In this way", as appellants explain it, "an airlift pump effect is produced which increases the rate of circulation of the liquid reaction mixture." In other words, the gases are bubbled up through the liquid creating a pressure which forces some of the liquid out of the reactor. The recirculation of this liquid adds to the turbulent effect.5
Two further aspects of the F.H.A. disclosure need mentioning. At one point in the reference teaching, after certain comments regarding the determination of optimum quantity of gas throughput, it is stated that "it is unnecessary and even detrimental to use nozzles or jet appliances such as are frequently used to accelerate the circulation of liquids." At the conclusion of the written description the following is stated:
The process of the present invention is distinguished from the known processes which are carried out in the liquid phase and wherein portions of the liquid product are drawn off at the top of the reactor and re-introduced at the bottom thereof without the use of mechanical assistants, by the fact that circulation of the liquid is brought about without special undercooling of the reflux current and without the use of nozzles or jets but simply by a high rate of input of gas, which is so rapid that the temperature remains practically constant throughout the apparatus. This is not possible in the known processes.
IN THE PATENT OFFICE
The instant claims were rejected "under 35 USC 103 as unpatentable over F. H.A. either alone or taken in view of Tramm et al." After stating that the particular catalyst solutions recited in the claims "are conventional as disclosed by F.H.A., and also Tramm et al." the examiner asserted:
The only alleged difference between the instantly claimed process and that of the F.H.A. reference is that in the instantly claimed process circulation of the reactants is accomplished by high velocity input of the liquid feed whereas in the F.H.A. process this circulation is accomplished "by a high rate of input of gas, which is so rapid that the temperature remains practically constant throughout the apparatus". (See page 5, lines 37-41).
The breadth of the claims was next apparently (though not explicitly) attacked, the examiner pointing out that the claims "do not exclude" effecting the circulation of the reactants by introducing the synthesis gas at a high rate while at the same time introducing the liquid reactants at a high linear velocity. It was also emphasized that "the velocity of the input liquid feed in the instantly claimed process may be lower than the velocity of the liquid circulating in the reaction vessel" (emphasis quoted) and the conclusion drawn that
Thus example 2 in F.H.A. which discloses continuously introducing a liquid olefin and also introducing the synthesis gas at "a high rate of input" is deemed to be indistinguishable from the instantly claimed process.
The Board of Appeals affirmed the rejection
because we are convinced that appellants' process which involves the introduction of OXO liquid reactants into a reaction vessel at a high linear velocity "so that the circulation of the liquid phase in said vessel is effected to the extent of 10 to 80% by transfer of momentum," would have been obvious to a person of merely ordinary skill in the art from the F.H.A. patent whether considering with the Tramm et al. patent or not.
Citing its own criticisms of the claim language (e. g., that "the relative term `high' would provide dubious basis for distinction," and that the specific linear velocities mentioned in some of the claims cannot be accorded significance because they are not associated with any parameters to be measured against), the board then shored up its conclusion by interpreting the disclosure in the F.H.A. reference as indicating that it was known in the prior art to obtain a circulation effect by introducing liquid into the reaction chamber and to use nozzles or jets for such liquid introduction.
In response to the examiner's reluctance to allow the claims, appellants had submitted an affidavit by one of their number, Dr. Nienburg, which attempted to set out a comparison showing the efficiency of the present process per unit time as compared with the F.H.A. process. The comparative experiments recited therein indicate that the process employing a high linear velocity in introducing the reactants converted 95.1 percent of the olefin with a space-time yield of 24.8 kilograms of product per liter of reaction space per day. In the experiment using the F.H.A. process (with a low linear velocity), 86.9 percent of the olefin was converted and the space-time yield measured 16.5. Certain reaction conditions employed in the two experiments were not the same. The affiant mentioned this, explaining that the experiments had been carried out at different times, but asserting that, in his opinion, the experiments do show that the use of a higher linear velocity of the starting materials will produce a high conversion and a higher space time yield. This affidavit was held unpersuasive by the examiner and the board, not only because it was considered not to be a good comparison but also because it was "insufficient to establish criticality of the claimed high velocity of liquid feed."
Upon our consideration of the record and the briefs and arguments of the parties, we are satisfied that the administrative decision below was correct and should be affirmed. The situation here is not unlike that involved in the recently decided In re Muchmore, 433 F.2d 824, 58 CCPA (1970). In that case, as here, we had claims before us to a process which was, superficially at least, very similar to that already disclosed in the prior art. Patentability was asserted on the basis of a single distinction, the limitation to which in the claims was predicated on language which was not as precise and definite as it could have been. In Muchmore, too, the Patent Office tribunals criticized the claim language but made no rejection under 35 U.S.C. § 112. And, in Muchmore, the applicant submitted an affidavit attempting to demonstrate improved results over the process of the prior art. We held that while the affidavit indicated that some of the processes within the claim might be said to yield unexpectedly superior results over that of the reference, the totality of processes covered by the claim, as a class, did not, and, accordingly, affirmed the rejection.
In the case before us, we are satisfied that the disclosure of the F.H.A. patent supports a prima facie conclusion that the process of the claims on appeal is obvious (indeed, appellants, by the comments in their specification and arguments both here and below, appear to concede that such is the case). With regard to appellants' attempt to overcome this prima facie showing, we note first of all that we have been shown nothing which would justify our disagreeing with the administrative finding that the results therein demonstrated did not establish the "criticality" by which appellants hoped to prove unobviousness. Under these circumstances we would have no choice but to accept the conclusion made below that the affidavit would not convince one of ordinary skill in the art of the unobviousness of the claimed process.
More importantly, however, even if we were to assume that the results which are indicated in the affidavit would have been unexpected, we nevertheless find the limited evidence presented therein insufficient to convince us of the probability that comparable results would be attained with the totality of processes covered by the appealed claims. Since such would be necessary to support a holding that the initial conclusion of obviousness was rebutted, the decision of the Board of Appeals is affirmed.
Serial No. 416,027, filed December 4, 1964, for "Continuous Production of Oxygen-Containing Compounds."
Two other references were cited but were not relied on by the examiner on appeal as they were felt to be only cumulative
U. S. Patent 2,802,843, issued August 13, 1957
Specification No. 920,417, made public on March 6, 1963
It is of some interest to note at this point that the F.H.A. process was discussed at some length in appellants' specification. In their brief here appellants assert that "[t]he present process was designed to overcome the defects found in the reference method."