WORLEY, Chief Judge.

This appeal is from the decision of the Board of Appeals which affirmed the examiner's rejection of product claims 2 and 3 in appellant's application for "Arylsulfur Pentafluorides and their Preparation." Two process claims were allowed.

The nature of the subject matter involved is reflected in the appealed claims:

"2. Compounds of the formula

X_nR(SF₅)_m,

wherein R is an aromatic hydrocarbon group, X has a maximum of eighteen carbons and is a member of the group consisting of halogen, hydroxyl, thiol, hydrocarbyl, halohydrocarbyl, oxahydrocarbyl, thiahydrocarbyl, hydrocarbonyl, hydrocarbonyloxy, carboxy and groups hydrolyzable thereto, and sulfo and groups hydrolyzable thereto, m is a whole number of 1-3, inclusive, and n is a cardinal number of from zero to 5, inclusive.

"3. Phenylsulfur pentafluoride."

Appellant states in his specification that "the compounds of the invention are characterized by excellent thermal and chemical stability" and "* * * the sulfur pentafluoride group is resistant to hydrolysis under acidic, basic and neutral conditions." The compounds are said to be useful as fluids for high temperature power transmissions, hydraulic systems or liquid-coupled mechanical drives, and as solvents for highly fluorinated polymer compositions useful in waterproofing cellulose materials.

The reference relied on is: Emeleus et al., J. Chem.Soc. (London), Vol. of 1946, pp. 1126-1131.

That article relates to preparation of alkyl- and aryl- substituted fluorides of sulfur, selenium and tellurium, with the following portions particularly pertinent:

"The fluorides of sulfur, selenium, and tellurium show decided differences from the other halides of these elements. For instance, in all three cases the maximum covalency of 6 is reached in combination with fluorine, but not with the other halogens, even at low temperatures. The hexafluorides are exceptionally stable, whereas the other halides, including the fluorides, easily hydrolyse and dissociate when heated. Indeed, some doubt exists as to the formation of lower fluorides of selenium and tellurium. Considerable interest, therefore, attaches to the preparation and properties of the alkyl- and aryl- substituted fluorides of these elements. Experiments are described below which seem to show that substituted hexafluorides such as RSF₅ are not stable * * *.

* * * * * *

"Experimental.

* * * * * *

"Fluorination of the Substituted Dihalides. — No fluorine derivatives in this series were isolated. Attempts were made to replace chlorine by fluorine in PhSCl, o-NO₂.C₆H₄.SCl, o-NO₂.C₆H₄. SBr, and p-NO₂.C₆H₄.SCl, with silver, mercurous, mercuric, and hydrogen fluoride. All these substances reacted with the metallic fluorides, but the product always consisted mostly of the corresponding disulphide, mixed with tarry matter containing some fluorine * * *.

* * * * * *

"Higher-substituted Fluorides. — No analogues of the compounds SRF₃, SRF₅ and SeRF₅ exist for the other halogens. It was therefore necessary to choose methods of preparation other than by way of the usual metal fluorides or hydrogen fluoride. Experiments were made on the following lines: (1) The action of Grignard reagents upon sulphur tetra- and hexa-fluoride. (2) The action of fluorine, or phenyliodonium fluoride, upon the organic disulphides and diselenides.

"(1) Gaseous sulphur hexafluoride bubbled through an ethereal solution of phenylmagnesium bromide did not react below the b.p. of ether. Liquid `sulphur tetrafluoride' * * * reacted violently with phenylmagnesium bromide in ether at -60°. The products, however, contained no volatile fluorine compound and apparently consisted of PhSBr and bromine * * *.

"(2) The reaction at 0° between fluorine-nitrogen mixtures and chloroform solutions of diphenyl diselenide or o- or p-nitro-phenyl disulphide yielded only hydrogen fluoride and tarry or high-boiling condensation products. * * *

* * * * * *

"Discussion

"From the work described, the following conclusions emerge: (1) The substituted fluorides of sulphur, selenium, and tellurium are much less stable than the other halides. Stability increases from fluorine to iodine. (2) Stability increases in all the halides with the number of organic radicals in the molecule, and most noticeably in the fluorides. (3) Stability increases from sulphur to tellurium.

* * * * * *

"* * * The results of the attempts to fluorinate the phenyl sulphur halides can be explained as follows: The phenyl sulphur fluoride first formed disproportionated to a mixture of the disulphide and a higher fluoride [e.g., p-NO₂.C₆H₄.SF to (p-NO₂.C₆H₄.S)₂ and p-NO₂.C₆H₄. SF₃ or p-NO₂.C₆H₄.SF₅]; the latter then decomposed by intermolecular condensations to hydrogen fluoride and tarry products. * * *"

The examiner rejected claims 2 and 3 over Emeleus, stating that the claimed compounds differ from the reference compound, p-NO₂.C₆H₄.SF₅, in that they contain either no substituent on the phenyl group (as in phenylsulfur pentafluoride), or some substitutent other than a nitro group (as in the Markush grouping set forth in claim 2). He regarded the claimed compounds as "unpatentable analogs" of the nitro compound of Emeleus, since the "nitro group is as close to some of the members of the Markush group as they are to each other." Presumably the statutory basis for the examiner's rejection is 35 U.S.C. § 103.

After observing that halogen is a member of the Markush group set forth by appellant, the board affirmed the examiner's rejection of claim 2 on Emeleus. It noted "the analogy between a nitro group and a chloro group as a substituent has long been recognized," citing In re Taub, 125 F.2d 719, 29 CCPA 893. In response to appellant's argument that his compounds are stable, whereas the reference compounds are described as being unstable, the board stated:

"* * * In view of the great variety of compounds that are described by appellant as being very stable it seems highly unlikely that only the nitro variant would be unstable even if it were pure. If in fact there is a difference in stability it can with reason only be attributed to differences in the mode of preparation resulting either in differing kinds and/or degrees of impurity * * *."

On the basis of its own study of Emeleus, the board also regarded the reference to anticipate claim 3, hence claim 2, under 35 U.S.C. § 102(a), reasoning:

"* * * the p-nitro phenylsulfur pentafluoride compound mentioned * * * is merely exemplary of results obtained by the experiments described * * * in the paragraph entitled, `Fluorination of the Substituted Dihalides.' * * * [That paragraph] does not mention any sulfur pentafluorides but states that attempted fluorination of phenylsulfur chloride and p-nitrophenyl sulfur chloride with metallic fluorides yielded corresponding disulfides and a tarry matter and hydrogen fluoride. * * * [The "Discussion" portion of Emeleus] has a more detailed explanation in some respects and states that it is a `higher fluoride' which decomposes to yield the hydrogen fluoride and tarry products and that this higher fluoride could be a phenyl sulfur pentafluoride. Appellant has not denied that such a pentafluoride is obtained, and, this being the case, we must assume that the experiments yielded phenyl sulfur pentafluoride as well as the p-nitro compound. The reference clearly teaches one skilled in the art the possible existence of phenyl sulfur pentafluoride just as well as it teaches that of the nitro compound.

* * * * * *

"* * * We have made it clear that the compound of claim 3 is explicitly taught by Emeleus et al. and we, accordingly, hold that claims 2 and 3 are unpatentable over the reference for this reason also. Their allowance would be contrary to 35 U.S.C. § 102(a) * * *."

With that background information, we must determine the correctness of the rejections of claims 2 and 3 under 102(a) and claim 2 under 103. From our evaluation of the record we are inclined to conclude that the board erred in both instances.

REJECTION UNDER 35 U.S.C. § 102(a)

We begin by agreeing with the board that Emeleus "teaches one skilled in the art the possible existence of phenyl sulfur pentafluoride just as well as it teaches that of the nitro compound." Emeleus says "all these substances," referring to PhSCI and p-NO₂.C₆H₄.SCl, reacted with the designated fluorinating agents. It seems to us that a person of ordinary skill in the art could fairly expect those compounds to follow a similar hypothetical reaction path. Thus, under "Discussion" Emeleus states:

"* * * The results of the attempts to fluorinate the phenyl sulfur halides can be explained as follows: The phenyl sulfur fluoride first formed disproportionated to a mixture of the disulphide and a higher fluoride [e.g., p-NO₂.C₆H₄.SF to (p-NO₂.C₆H₄.S)₂ and p-NO₂.C₆H₄.SF₃ or p-NO₂.C₆H₄.SF₅]; the latter then decomposed * * *."

If we regard Emeleus as suggesting that the corresponding unsubstituted compound, C₆H₅SCl (PhSCI), undergoes a similar reaction, the disclosure in effect says:

"* * * [C₆H₅.SF to (C₆H₅.S)₂ and C₆H₅.SF₃ or C₆H₅.SF₅]; the latter then decomposed * * *"

(All emphasis supplied)

However, we reach an opposite conclusion than the board that phenyl sulfur pentafluoride disclosed in that manner is an enabling disclosure of the claimed invention within our understanding of 35 U.S.C. § 102(a). We agree with appellant that the board erred in concluding that either p-NO₂.C₆H₄.SF₅ or C₆H₅. SF₅ is "explicitly taught" by Emeleus. We find that teaching to be less than unequivocal, noting that he states that p-NO₂.C₆H₄.SF₃ or p-NO₂.C₆H₄.SF₅ is believed to be formed. Indeed, the board's language in its initial opinion indicates its uncertainty on that point as well as on the part of Emeleus. The board stated:

"* * * [Emeleus] states that it is a `higher fluoride' which decomposes * * * and that this higher fluoride could be a phenyl sulfur pentafluoride. * * * The reference clearly teaches one skilled in the art the possible existence of phenyl sulfur pentafluoride * * *." (Emphasis supplied)

In its decision on reconsideration, the board said:

"Such words as `could' and `possible' were used by us merely because the authors of the reference were not certain that the named compounds were actually obtained * * *. We are not aware of any decision which requires that the author of a reference must be certain that a compound is obtained for that reference to be a valid one."

The uncertainty which pervades the teaching of the reference leads us to conclude that Emeleus cannot properly be held an enabling disclosure.

A further reason why the board erred in concluding C₆H₅.SF₅ is "explicitly taught," is that the Emeleus disclosure is not sufficient to place the public in possession of the invention. Emeleus has expressly stated that at least p-NO₂.C₆H₄. SF₃ or p-NO₂.C₆H₄.SF₅, both postulated transient intermediates, decompose to form tarry products. In the "Experimental" portion of the reference, Emeleus asserts an inability to isolate any fluorine derivatives which might have resulted from the reaction of the substituted dihalides with the designated fluorinating agents. Further on, under "Higher-substituted Fluorides," Emeleus indicates a complete lack of success in preparation of pentafluorides by (1) the reaction of Grignard reagents upon sulfur hexafluoride (2) the action of fluorine upon organic disulphides. It is clear to us from our study of Emeleus that it was beyond the author's skill as a chemist to alter the reaction conditions or reactants in order to make a stable pentafluoride. No other evidence appears in the record before us which shows that a person of ordinary skill in the art, upon reading Emeleus, would be able to combine the teachings of that article with his own knowledge of this art to obtain possession of appellant's stable arylsulfur pentafluorides.

Most of the cases cited by the solicitor in support of the Patent Office position that Emeleus discloses the claimed invention within the meaning of 35 U.S.C. § 102(a) were considered by us in In re LeGrice, 301 F.2d 929, 942-943, 49 CCPA 1124, 1142-1144. As noted there, explicit or implicit in all those cases is the concept of a certain degree of knowledge possessed by one skilled in the arts involved, to the end that such knowledge, taken with the disclosure of the printed publications, was sufficient to place the disclosed invention in the possession of the public. We think In re Doyle, 327 F.2d 513, 51 CCPA 993, and In re Fried, 329 F.2d 323, 51 CCPA 1118, cited by the solicitor in addition to those considered in LeGrice, are of the same nature and are not controlling. Cf. In re Brown, 329 F.2d 1006, 51 CCPA 1254; E.I. DuPont deNemours Co. v. Ladd, 117 U.S.App.D.C. 246, 328 F.2d 547. We have no such explicit or implicit knowledge shown in the record here. The Emeleus article is not so particular and definite that, without undue experimentation, one versed in the art to which it pertains could gain possession of the claimed subject matter. Under the circumstances we feel obliged to reverse the rejection of claims 2 and 3 under 35 U.S.C. § 102(a).

REJECTION UNDER 35 U.S.C. § 103

In finding the compounds of claim 2, particularly the chloro derivative, unpatentable over the p-nitro phenyl sulfur pentafluoride "disclosed" by Emeleus, the board relied on, erroneously, we think, In re Taub as authority for reaching the conclusion that the two substituents are analogous.

In Taub, this court sustained the conclusion of the examiner that nitro and chloro substituents "are considered to be equivalents in a molecule of the size claimed, in the absence of a showing or unexpected properties in the chloro product." (Emphasis supplied) The compounds in Taub were quaternary ammonium salts in which four hydrocarbon groups, one a chloro-benzyl radical, were attached to the nitrogen atom. The prior art disclosed the same compound except a nitro-benzyl group was attached to the nitrogen atom. Both compounds were useful as fungicides, and it was not shown the compounds differed in effectiveness.

Here, however, Emeleus states that the postulated nitro derivative is unstable and decomposes. Appellant, on the other hand, discloses that arylsulfur pentafluorides having specific substituents are stable chemically and thermally. The prior art neither suggested that arylsulfur pentafluorides would be stable, nor was it aware of that desirable property. Indeed, Emeleus would lead one skilled in the art to the opposite conclusion when it states "Experiments * * * seem to show that substituted hexafluorides such as RSF₅ are not stable."

The board speculated that the instability of Emeleus' compounds can only be due to impurities. As appellant points out, Emeleus does not give that reason as the source of instability. The whole tenor of Emeleus leads one of ordinary skill in the art to believe instability is due to the nature of the compounds rather than outside factors. It appears to us that only by hindsight could the board conclude "it seems highly unlikely that only the nitro variant would be unstable."

Based on our evaluation of the record and understanding of the chemistry involved, we are inclined to agree with appellant that the board erred in its interpretation of, and rejection on, the Emeleus reference. Should there be doubt on that score we feel obliged to resolve it in favor of appellant.

The decision is reversed.

Reversed.