404 F.2d 616 - In re Broadley — 404 F.2d 616

¶156 CCPA

¶2Jay P. Friedenson, Morristown, N. J., for appellant.

¶3Joseph Schimmel, Washington, D. C. (Raymond E. Martin, Washington, D. C., of counsel) for the Commissioner of Patents.

¶4Before WORLEY, Chief Judge, and RICH, SMITH, ALMOND, and BALDWIN, Judges.

¶5WORLEY, Chief Judge.

¶6Broadley alleges reversible error in the decision of the Board of Appeals affirming the examiner’s rejection of claims 1-6 1 as unpatentable over Lewis 2 in view of Midgley3 under 35 U.S.C. § 103.

¶7The invention relates to a high capacity,4 low boiling, azeotropic refrigerant composition consisting of difluoro-me thane (CH2F2; boiling point of -51.7° C.) and monochloropentafluoroe-thane (C2F5C1; boiling point of -38.7° C.), as well as to a method of refrigerating using that composition. It appears from the record that azeotropic compositions — ones in which the vapor composition is the same as the boiling liquid composition with which it is in equilibrium — -are advantageous refrigerants compared to non-azeotropic multicomponent mixtures since fractionation or separation of the components of the composition, with attendant increase in condenser pressures and overall loss of efficiency, does not occur during the refrigeration cycle. Appellant’s azeotropic compositions consisting of 30-33 mol % C2F5C1 exhibit a constant, minimum boiling point of -57.3° C., whereas mixtures consisting of 13-65% C2F5C1 have boiling points differing only slightly — about 0.7° C. or less — from that of the azeo-trope. According to appellant, his compositions boil at temperatures lower than the boiling point of either pure component, and thus exhibit a negative deviation from Raoult’s law, an empirical rule to the effect that boiling points and vapor pressures of two-component mixtures vary with concentration between the val*617ues for the pure components. It appears that the high capacity of appellant’s compositions relative to that of either component is due to the fact that the compositions are azeotropic and boil at a lower temperature than either component alone,

¶8Olaim lis representative-Vw/IaxIII JL lo X cUl cocil bw L1V v •

¶9, ... 1. A low boiling composition consisting of a mixture of difluorometh-ane and monochloropentafluoroe-thane, in which mixture the mol percent of monochloropentafluoroethane is substantially m the range of 13-65.

¶10The examiner noted that Lewis discloses an essentially azeotropic refrigerant composition consisting of 20-40 mol percent of 1, l-difluoroeíftcme and 60-80 mol percent of the heretofore men-tinned C2F„C1. The actual azeotrope, formed at 69 mol percent C2F5C1, has a minimum boiling point of -41.3" C„ and thus exhibits a negative deviation from Raoult’s Law. Observing that Lewis does not disclose the use of CH2F2 in refrigerant compositions or as a component of an azeotropic mixture, the examiner turned to Midgley, who discloses that CH2F2, C2F5C1 or CH3CHF2, among many other fluorohydrocarbons, each may be used individually as refrigerants 5 and also shows the boiling point of each of those compounds in his figures 1 and 2. In the examiner’s view, it “would be obvious to one having ordinary skill in the art to substitute … [CH2F2]. for the … [CH3CH2] of Lewis with the expectation of obtaining suitable results” in view of the fact ill /n tt -n • i i i • t • that CH2F2 is a known low boiling* refrigerant and ig the adjacent lower homo. jQg of CH¡¡CHF2- Quite evidently, the examiner felt that the homologous relationship between CH2f2 and CH3CHF2 wouW be suffident in and of itself to enabk one of ordinary skill to predict that CH2F2 and C2F5C1, like the known CH3CHF2 and C2F5C1, would form a useful azeotropic refrigerant composition. The board agreed.

¶11Appellant contends here, as below, that notbia? “ P"" ar* ^ reco"d gests that CH2F2 and C2.F?C1 would form an azeotropic composition possessing hl*h refrigeration capacity.^ Indeed, apPellant£ urges that there is no known basis, for Pr^ietmg the formation of an ^eotrope between two given fluorohydrocarbons. In support of that argument, he points to the Kvalnes patent6 cited by him in his brief before the board, as well as to certain proprietary data appearing in the same brief.7 It *618is appellant’s position that he has established that the mere fact of homology between CH2F2 and CH3CHF2 would be of no material assistance to one in the art in determining or predicting a priori the proclivity of CH2F2 to form an azeotrope with C2F5C1, while the examiner has offered no evidence to support the bare assertion that homology would be of assistance in that regard.

¶12The board acknowledged that appellant “documents the unpredictable nature of determining azeotropic pairs of compounds” in his brief before it, thus apparently accepting the comparisons made by appellant and the statements of Kvalnes as factual and accurate.8 Like the examiner, however, the board found it could “give no weight to appellant’s argument that the determination of azeo-tropic mixtures is empirical in nature.” The only reason each assigned was that “the claims are not limited to azeo-tropic mixtures.”

¶13Strictly speaking, of course, the examiner and board are correct in observing that none of appellant’s claims is directed to a true azeotropic composition consisting of 30-33% C2F5C1 and 67-70% CH2F2. But it appears to us that appellant also is correct in stating that all his claims are directed to compositions which are, for all intents and purposes, essentially azeotropic in nature and which possess the desirable features, e. g. negligible fractionation, of azeo-tropic compositions.9 As appellant points out, his specification states:

… It has also been found that mixtures consisting of difluoromethane and monoehloropentafluoroethane substantially in the range of 13-65% monoehloropentafluoroethane have boiling points differing only slightly from the azeotropic composition. Compositions within this range will exhibit only negligible fractionation on boiling and represent useful refrigerants. …

* -X* -X* …

… The 13-65% chloropenta-fluoroethane compositions indicated *619each has a boiling temperature variation of less than 0.75° C., from the minimum boiling minus 57.3° C. azeo-trope composition. As compositions containing 13-65% C2F5CI all boil between about -56.6° C and -57.3° C., representing a maximum boiling temperature variation within 0.75° C., no significant fractionation occurs on distillation of any compositions within this range, and all such compositions are suited for use as refrigerants. ******

The system formed by herein mixtures of difluoromethane and monochloropentafluoroethane provides a broad range of mixture compositions which boil at substantially constant temperature and give a constant boiling composition. The azeotropic and substantially azeotropic mixtures differ from refrigerant mixtures proposed in the past which offer only a relatively narrow range of useful compositions. … [Emphasis supplied.]

¶14In view of those statements, it seems to us that the examiner and board were unduly critical of appellant’s claimed subject matter in unnecessarily requiring appellant to limit his claims to true azeo-tropic compositions consisting of 30-33% C2F5CI before giving consideration to his argument.

¶15Turning to the arguments presented, we think that whatever reasons the examiner and board may have had — but not expressed or documented — for their belief or presumption that the homologous relationship between CH2F2 and CH3CHF2 would be of assistance to those in the art in predicting formation of an azeotrope between CH2F2 and C2F5C1 have been adequately rebutted by appellant. The only other reason offered by the board in support of its finding of obviousness is that one might well predict that substitution of the lower boiling CH2F2 for the higher boiling CH3CHF2 in the Lewis composition would yield a mixture boiling at a lower temperature, and thus possessing a higher refrigerating capacity, than that of Lewis. However true that may be, it would seem there would be no assurance that such a two-component mixture would have a capacity roughly twice that of Lewis as shown by an affidavit of Atwood, or even be a particularly useful or efficient refrigerant, absent a teaching that a minimum boiling, substantially azeotropic, non-fractionating composition would result. It is that teaching which we think is missing here.

¶16Considering the record in its entirety, we find inadequate evidence to sustain the conclusion below that the claimed subject matter as a whole would be obvious to one of ordinary skill within the purview of § 103.

¶17The decision is reversed.

¶18Reversed.

¶19SMITH, J., participated in the hearing of this case but died before a decision was reached.

¶26. U.S. Patent 3,085,065, issued April 9, 1963.

¶27In the course of discussing formation of azeotropic refrigerant compositions of fluorohydrocarbons, Kvalnes states:

¶28… It is impossible to predict that an azeotrope will form between any two compounds and so it is impossible to pich any two refrigerants and combine them to obtain an azeotrope boiling at some particular point by prediction. If an azeotrope occurs at all, its boiling point is a function of the system and is not under the control of the experimenter. The mere existence of a large number of fluorinated hydrocarbons does not aid in predicting the formation of azeotropes by mixtures thereof. [Emphasis supplied.]

¶29The Kvalnes patent does appear to be of evidentiary value in determining the level of skill of those in the art. See Graham v. John Deere Co. of Kansas City, Mo., 383 U.S. 1, 17, 86 S.Ct. 684, 15 L.Ed.2d 545.

¶30. There, appellant made certain comparisons between pairs of two component fluorocarbon refrigerants, stating:

¶31… The ability to make such a prediction [of azeotrope formation] is not aided by closeness in structure of one compound to another or its closeness in structure to a compound which is known to have formed an azeotrope with another material. By way of ex*618ample the following relationship should be considered:

¶321. CH2F2/CF3Br forms an azeotrope. CH3CHF2/OF3Br does not form an azeotrope.

¶332. CHF3/CF3C1 forms an azeotrope. CH3CF3/CF3C1 does not form an azeo-trope.

¶343. CH3CF3/C2F5C1 forms an azeo-trope. CHF3/C2F5C1 does not form an azeotrope.

¶354. CHF2C1/CC12F2 forms an azeo-trope.

¶36CH3CF2C1/CC12F2 does not form an azeotrope.

¶375. 0HF201/02F501 forms an azeo-trope.

¶38CH3CF2C1/C2F5C1 does not form an azeotrope.

¶39…

¶40It should be noted that the … five relationships are homologous relationships as regards the first noted components, thus demonstrating that the doctrine of homology is of no assistance in providing a basis for the prediction of the formation of an azeotrope. …

¶42. Indeed, that those in the art would regard appellant’s claimed compositions to be essentially azeotropic despite minor deviations from the constant boiling point of the azeotrope is clear, we think, from Lewis, who states:

¶43I have discovered that mixtures of 1,1- difluoroethane and monochloropen-tafluoroethane form an azeotrope boiling at approximately -41.3° C., in which about 69 mole percent is monoehloro-pentafluoroethane. The mixture is a highly useful refrigerant … This mixture has the further advantage that only slight changes in the boiling point occur when the composition is varied over a considerable range on either side of the actual azeotropic composition. Mixtures containing between 60 and 80 mole percent monochloropentafluoroe-thane all boil between about -41.1° and -41.3° O., so that there is little tendency for fractionation to take place with any of these mixtures. 'They are all effective refrigerating agents. [Emphasis supplied.]