This appeal is from the decision of the Patent and Trademark Office (PTO) Board of Appeals (board) affirming the rejection under 35 U.S.C. § 103 of claims 1 and 8 in application serial No. 114,034, filed February 9, 1971, entitled “Hydroxyaryldialkyl Sulfonium Halides.” We affirm.
The Invention
The invention involves a specific class of hydroxyaryl asymmetric dialkyl sulfonium halide compounds, which are disclosed as having unusual germicidal activity, especially against gram positive bacteria. Claim 1 is illustrative:
1. Hydroxyaryldialkyl sulfonium halides having the structural formula
wherein R is a normal alkyl group having 10 to 20 carbon atoms, wherein R' is hydrogen or methyl, and wherein X is bromine or chlorine, provided that when R' is hydrogen, the hydroxyl group is in the ortho or para position with respect to the sulfur atom, and provided that when R' is methyl, the hydroxyl group is in the para position with respect to the sulfur atom.
Claim 8 differs from claim 1 in that it limits the class of compounds of claim 1 to those in which R is a normal alkyl group having 10-18 carbon atoms.
The References
The references relied upon by the board are:
MacGregor 3,133,971 May 19, 1964
Cisney 3,259,660 July 5, 1966
MacGregor discloses a class of hydroxyaryl sulfonium chloride compounds of the following formula:
wherein R and R' are represented by an alkyl, aryl, aralkyl, or alkaryl radical, and substituted derivatives thereof, and a process for forming these chlorides. The process comprises reacting an organic thioether, a phenol, and chlorine in the following manner:
*749MacGregor specifically discloses that the organic thioether reactant (which forms the R — S — R' portion of the prior art class of compounds) is “preferably” one of “those containing up to 20 carbon atoms and having at least one methylene group attached to the sulfur atom,” and illustrates the preferred alkyls with, inter alia, methyl and dodecyl. Usefulness of the compounds as biological toxicants is shown.
Cisney, assigned to the same assignee as MacGregor, discloses a class of hydroxyaryl sulfonium chloride compounds made by a process improved over that of MacGregor and having the same uses shown by MacGregor. It has substantially the same disclosure as MacGregor with respect to the described class of sulfonium chlorides, the only difference being that Cisney defines the R and R' radicals as selected from alkyl of 1 — 20 carbon atoms, 2-chloroethyl, aryl, arylalkyl of 7 — 8 carbon atoms, and alkylaryl of 7-10 carbon atoms. This is a narrower group than that defined by MacGregor. The improved process (which reduces formation of by-products) reacts a thioether with chlorine to form a first reaction product and then reacts this product with a phenol to produce the sulfonium chloride, in the following manner:
Rebuttal Evidence
Appellants have relied on data in their specification and on an affidavit of Lamberti to show unobviousness. The specification states that the compounds in the claimed class contain unexpectedly good germicidal activity, especially against gram positive bacteria. It sets forth examples and comparative data which allegedly show that the minimum effective concentration for germicidal activity of the claimed compounds was unexpectedly lower than for the other compounds tested.
The affidavit of Lamberti allegedly shows the inoperability of the process of producing specific hydroxyaryl dioctyl sulfonium chlorides (Example 9 of MacGregor and Example 5 of Cisney). The affidavit, in pertinent part, states:
3. I have attempted to duplicate certain of the examples described in the above-identified applications of Cisney and MacGregor and to that end I directed persons working under my direction to prepare dioctyl sulfonium compounds according to Example 5 of Cisney and Example 9 of MacGregor. Five separate attempts to produce the dioctyl compound according to the referenced patents were made by three different chemists, and each such attempt was unsuccessful.
6. Based upon the five attempts by three different chemists acting under my supervision, I have concluded that the Cisney and MacGregor references do not teach the preparation of the dioctyl compounds which the patents purport to disclose.
The affidavit includes an exhibit showing the unsuccessful attempts to prepare the compounds according to Example 9 of MacGregor, but it does not contain the facts upon which the stated inoperability of Example 5 of Cisney was based.
Proceedings Below
The examiner rejected the claims, inter alia, under 35 U.S.C. 103 as obvious in view of either MacGregor or Cisney.
The board affirmed only the 35 U.S.C. § 103 rejection, holding that it would have been obvious to use an asymmetric dialkyl thioether in the disclosed process of MacGregor or Cisney to produce a hydroxyphenyl dialkyl sulfonium halide having an asymmetric dialkyl grouping. The board noted that “[e]ach of MacGregor and Cisney teach [sic] hydroxy phenyl dialkylsulfonium halides useful as biological toxicants which necessarily includes bactericides.”
The board also held that the rebuttal evidence did not overcome the prima facie *750case of obviousness established by the examiner. With respect to the data in appellants’ specification, it found that the evidence “does not establish that appellants’ compounds viewed as a whole have unexpected bactericidal properties as compared with the lower homologous compound [sic].” As to the Lamberti affidavit, it noted that appellants did not show the inoperability of a method for producing a sulfonium chloride having an asymmetric dialkyl group, as claimed. It also noted that the conclusion regarding the Cisney process lacked probative value since the facts upon which it was based had not been shown.
OPINION
Both Cisney and MacGregor disclose a class of compounds which embraces the claimed class when the “X” of the claimed class is chlorine. Moreover, both references disclose that the compounds are useful as biological toxicants, which term is generic to germicidal agents. Finally, the statement in both MacGregor and Cisney that the reactant which forms the R-S-R' portion of the compound has “at least one methylene group attached to the sulfur atom” suggests a symmetric dialkyl moieties, since the phrase “at least one methylene group attached to the sulfur atom” implies that the other group attached to the sulfur atom need not be a methylene (e. g., it can be a methyl) group.
Appellants argue that neither of the prior art references suggests the use of asymmetric dialkyl moieties, because the disclosures of both references are broad, asymmetric dialkyl compounds are neither expressly disclosed nor exemplified, and the preferred embodiment is a symmetric dialkyl sulfonium chloride. However, as noted previously, the disclosure in the prior art of “at least one methylene group attached to the sulfur atom” would suggest the asymmetric aspect of the claimed invention. The fact that neither of the references expressly discloses asymmetric dialkyl moieties is not controlling; the question under 35 U.S.C. § 103 is not merely what the references expressly teach, but what they would have suggested to one of ordinary skill in the art at the time the invention was made. In re Simon, 461 F.2d 1387, 59 CCPA 1140 (1972). Finally, the fact that a specific symmetric dialkyl is taught to be preferred is not controlling, since all disclosures of the prior art, including unpreferred embodiments, must be considered. In re Mills, 470 F.2d 649, 176 USPQ 196 (Oust. & Pat.App.1972).
Appellants further argue that the references do not suggest the use of the claimed compounds as germicidal agents against gram positive bacteria. Although conceding that it is possible for the biological toxicants disclosed by the references to include germicidal agents, they take exception to the board’s statement that the disclosure of biological toxicants necessarily includes these agents. Appellants are correct in saying that these prior art disclosures do not necessarily include germicides. However, this does not preclude the disclosures from making the claimed compounds obvious. As this court has previously pointed out, obviousness does not require absolute predictability. In re Rinehart, 531 F.2d 1048, 189 USPQ 143 (Cust. & Pat.App. 1976); In re Clinton, 527 F.2d 1226, 188 USPQ 365 (Oust. & Pat.App.1976).
In criticizing the board’s holding that the rebuttal evidence did not overcome the prima facie case of obviousness established by the examiner, appellants point to the comparative data in their specification (considered by the board) to show that the claimed compounds have an unexpectedly good germicidal activity against gram positive bacteria. However, for such a showing, the claimed compounds must be compared with more closely related compounds of the prior art. In re Mageli, 470 F.2d 1380, 176 USPQ 305 (Oust. & Pat.App.1973). As indicated by the Solicitor, such prior art compounds would be drawn from three groups:
(1) symmetrical dialkyl sulfonium chlorides having normal alkyl groups with 1 to 20 carbon atoms, (2) asymmetrical dialkyl sulfonium chlorides having normal alkyl groups with 1 to 9 carbon atoms, and *751(3) asymmetrical dialkyl sulfonium chlorides having normal alkyl groups of 2 to 20 carbon atoms.
The comparative data in appellants’ specification do not show lack of germicidal activity of any compound drawn from these groups. Thus, Example 8, (4-hydroxyphe-nyl)-2,4-dichlorobenzylmethyl sulfonium chloride, Example 9f, (2-methyl-4-hydroxy-phenyl)-2,4-dichlorobenzylmethyl sulfonium chloride, and Example 11, (2-methyl-4-hy-droxyphenyl) benzylmethyl sulfonium chloride, all comprise an arylalkyl sulfonium chloride — not a dialkyl sulfonium chloride.1
Appellants argue that the Lamberti affidavit (also considered by the board) adequately rebuts “any inference or presumption that it would be prima facie obvious ‘to employ an asymmetric dialkyl thioether in either of the references to produce the corresponding hydroxyphenyl dialkylsulfonium halide’.” However, the affidavit is not probative with respect to employment of an asymmetric dialkyl thioether since Lamberti employed a symmetric dialkyl thioether (dioctyl sulfide).2 Moreover, the affidavit contains no factual evidence from which the inoperability of the Cisney example can be established. See In re Lindner, 457 F.2d 506, 59 CCPA 920 (1972).
In view of the foregoing, and considering the claimed invention as a whole vis-á-vis the evidence produced by the PTO, which shows both a close structural similarity between the prior art compounds and the claimed compounds and a close similarity between the disclosed uses, we hold that appellants’ invention would have been obvious to one of ordinary skill in the pertinent art.
Accordingly, the decision of the board is affirmed.
AFFIRMED.
. The difference in properties of arylalkyl, as opposed to dialkyl, groupings on the sulfur atom is noted by appellants where, in arguing against the board’s treatment of the Lamberti affidavit, they state:
The introduction of a phenyl or an aryl compound as a substitutive moiety onto the sulfur atom in question would, in all likelihood, change the properties of the class of compounds being claimed. To go from an alkyl to an aryl moiety certainly changes the complexion of what one skilled in the art would ordinarily expect in terms of the properties exhibited thereby. This is even suggested by the data of Table 1 with respect to Examples 8, 9f and 11 wherein methyl-benzyl substitutions are involved.
. At best, appellants have merely shown that it is possible to follow the process in one example each of MacGregor and Cisney without success. And even then, there is no showing that one of ordinary skill in the art, making adaptations within the skill of the art, could not have successfully carried out each process. See In re Michalek, 162 F.2d 229, 34 CCPA 1124 (1947).