Opinion
Appeal No. 81-555.
June 24, 1982.
Herbert B. Keil, Washington, D.C., for appellants.
Joseph F. Nakamura, Sol., and Harris A. Pitlick, Asst. Sol., Washington, D.C., for the Patent and Trademark Office.
Appeal from the United States Patent and Trade Mark Office Board of Appeals.
Before MARKEY, Chief Judge, and RICH, BALDWIN, MILLER and NIES, Judges.
This appeal is from the decision of the United States Patent and Trademark Office (PTO) Board of Appeals (board) sustaining the examiner's rejection of claims 8 and 9 in application serial No. 413,664, filed November 7, 1973, for "Trisazo Dyes," under 35 U.S.C. § 103. We affirm the rejection of claim 9 but reverse the rejection of claim 8.
The Invention
Appellants claim two trisazo dyes. The rejected claims are:
8. The dye of the formula
9. The dye of the formula
The significant difference from the compounds of the references, according to appellants, is the presence of a sulfonamido bridging group (-SO[2]NH) between the phenyl groups of the benzidine component of the dye.
The References
French patent 1,109,435 to Geigy discloses trisazo dyes in which the phenyl rings of the tetrazo component are either directly linked or are joined by any one of the groups -S-, -O-, -CH[2]-, -S-S-, -CH = CH-, -NH-, -NH-CO-, -NH-SO[2]-, -NH-CO-NH-, -SO-SO[2]- [sic, -SO-], -SO[2]- and -CO-. Badische, French patent 1,242,226, discloses green trisazo dyes of the formula (which we have transposed right to left)
wherein A is either phenol or 1,3-dihydroxy-benzene. Where A is the latter, the dye molecule is identical to appellants' claim 9 dye except that the tetrazo component contains a -S- link rather than a -SO[2]NH- link.
German patent 2,110,771 to Huhne et al. (Huhne) discloses trisazo dyes of the formula
where A can be selected from groups including appellants' nitrophenyl radical, and B can be selected from groups including hydroxphenyl (as in the Badische patent) and the diaminophenyl radical of appellants' claim 8 dye. The Huhne example 1 dye has the formula
The Huhne example 2 dye, which differs from appellants' claim 8 dye only in that it has a -CONH- linking group rather than an -SO[2]NH- link, has the formula
Huhne states that his dyes are
* * * outstandingly suitable for dyeing natural or synthetic fibre materials containing hydroxyl groups or containing nitrogen, for example cellulose fibres in all states of processing, especially cotton and regenerated cellulose, and also wool, wool/cellulose union, silk, nylon, leather and paper. Green to black dyeings have good fastness properties, especially good fastness to water and to perspiration following an after-treatment with cationic auxiliaries, are obtained.
The Rejections
Claim 8 was rejected on Huhne in view of Geigy and claim 9 was rejected on Badische in view of Geigy, both under 35 U.S.C. § 103.
The Declarations
Dr. Dietrich Lach (Lach), a staff dye chemist for the assignee of appellants' application, BASF Aktiengesellschaft, made a declaration in response to the above rejections. Stating that Huhne "is considered to represent the closest prior art," Lach compared the Huhne example 2 dye with appellants' claim 8 dye (labeling them dyes 1 and 2 respectively), and the Huhne example 1 dye with appellants' claim 9 dye (labeling them dyes 3 and 4, respectively). Each dye was tested for water-solubility, dyeing of chrome side leather and retanned chrome side leather, and washfastness. The visual results of Lach's tests are included in an exhibit accompanying the declaration, and a table in the declaration indicates that appellants' dyes both had a solubility of 40 grams per liter (g/1), while the Huhne dyes had solubilities of 18 and 10 g/1.
Lach stated:
It is obvious from these Exhibits that Dyes 2 and 4 [appellants' dyes] have surprising advantages in the dyeing of retanned chrome leather and in the washfastness.
The new dyes have further a much better solubility than the prior art Dyes 1 and 3. As to the importance of these properties, I would like to remark that nowadays mostly retanned chrome leather is in demand, because of its better properties, and that the dyes therefor must be effective for this material.
Washfastness is very important for glove and garment leather and is a requirement for these uses. Finally the solubility is extremely important in the dyeing process and 20 g/1 is considered to be the lowest value acceptable for practical purposes.
In response to the examiner's statement that the differences in the dyes noted in the first declaration "are not great enough to tip the balance in favor of patentability," Lach made a supplemental declaration. In it he stated:
It is generally known in the trade that the dyeing of chrome leather in full shades normally does not pose any great problems because the anionic leather dyes have a good affinity for this substrate. However, when the leather has been pretreated with synthetic or vegetable tanning agents, as is the case with retanned chrome leather, the affinity for most dyes is decreased considerably. It is therefore a big and, in many cases, unsolved problem to dye retanned chrome leather in deep shades. It is therefore very surprising and unobvious that Dyes 2 and 4 do dye retanned chrome leather in sufficiently deep shades with the same amount of dye as is used for non-retanned chrome leather. This behavior is quite in contrast to that of Dyes 1 and 3 which show the normal behavior, i.e. they more or less only stain the retanned chrome leather.
It is also known that dyes having low water-solubility are in general more suitable for the dyeing of retanned chrome leather than dyes with good solubility. It is therefore all the more surprising that the much more readily soluble Dyes 2 and 4 give the deeper dyeings.
Finally, the differences in the washfastness between the new and the known dyes are extreme; in fact, the differences are as big as that between usefulness and uselessness. Here again, this behavior is unexpected in view of the solubilities. Taking things overall, the new Dyes 2 and 4 show an unforeseeable combination of advantageous properties which are very different from those of Dyes 1 and 3 and are very surprising and unobvious, particularly if one concedes a structural similarity.
The Board
With respect to claim 8, the board stated that the sole difference between the claimed dye and the example 2 dye of Huhne is the presence of a sulfonamido (-SO[2]NH-) instead of a carbonamido (-CONH-) linking group. Since Geigy teaches the interchangeability of those groups, the board concluded that the claim 8 dye was prima facie obvious. Turning to the declarations of Lach, they stated:
While the minor differences in results demonstrated in the Declarations are not expressly spelled out in the references, it would be somewhat unexpected not to experience some differences in degree at least when a structural change is made to a basic molecular configuration. We therefore think it appropriate to insist upon a more dramatic difference in results where, as here, the showings are limited as to the number of properties tested.
The board also sustained the rejection of claim 9. Because Lach compared the claim 9 dye with the Huhne example 1 dye, the board stated that his "Declaration is of little probative value * * * since it does not provide a comparison of the claimed compound with the closest prior art compound disclosed in Badische."
Arguments On Appeal
Appellants state that the CCPA has often held that an affidavit of a skilled worker in the field must be given its fair weight and that neither the examiner nor the board should substitute their own speculations for the factual knowledge of those skilled in the art. "The fact that the dyes of claims 8 and 9 impart full and deep shades to retanned chrome leather with the same amount of dyestuff as is used for non-retanned chrome leather was surprising to Dr. Lach. This property of Appellants' dyes was in sharp contrast to the prior art dyes which only stained retanned chrome leather." Further, appellants argue, the result was surprising because dyes with lower solubility usually dye retanned chrome leather better than those with high solubility. Appellants thus conclude that it was error for the board to substitute its judgment for that of an expert whose qualifications and findings are unchallenged. The board's statement that the differences in the properties of the dyes are "minor" is, appellants state, unsupported by the record.
Appellants also assert that the claim 9 dye was compared with the closest prior art. Dr. Lach did not consider the Badische dye to be the closest prior art; he stated that the closest dyes were the Huhne example 1 and 2 dyes. Additionally, appellants argue that a sulfonamido group is closer in characteristics and properties to the carbonamido group of Huhne than to the thio group disclosed in Badische. Further:
The dye of Example 1 of the Huhne patent also differs from the dye of claim 9 in that the terminal benzene of Huhne is substituted by a 4-hydroxy group rather than a 2,4-hydroxy group. This difference, however, is not significent [sic] for purposes of comparison. The Badische patent teaches that the terminal benzene may be substituted by either 4-hydroxy or 2,4-hydroxy and yet have the same properties. Thus, a proper comparison could be made using either of the hydroxy substituted benzenes as the terminal unit. Since the mono- or di-hydroxy substituted benzene behave similarly and since the sulfonamido, the important feature of the invention, and the carbonamido bridging groups are more closely related, the dye of Example 1 of the Huhne patent was considered to be the closest prior art.
Finally, appellants argue that the starting material needed to produce the Badische compound is carcinogenic and was not available to appellants and their laboratories. Thus, even if the Badische dye and the Huhne example 1 dye are considered to be "equally close," there were sufficient practical considerations for choosing the latter.
The PTO argues that a prima facie case of obviousness, the existence of which has not been contested by appellants, stands unrebutted. As to claim 8, the solicitor abandoned at oral argument his arguments relating to washfastness and water solubility comparisons. What remains is his argument regarding Lach's leather dyeing comparisons. The PTO argues only that "the holdings of both the examiner and the board are supported by the results shown in Exhibit 1" (the dyed swatches of leather).
Turning to claim 9 and the "closest prior art" argument, the solicitor argues that the only difference between the claimed dye and the Badische dye is a sulfonamido bridge used in place of a sulfur bridge, while there are two differences between the claimed dye and the Huhne example 1 dye, a "sulfonamido bridge is used in place of a carbonamido bridge * * * and a dihydroxy phenyl moiety is used in place of a hydroxy phenyl moiety * * *." Further, Geigy teaches the equivalence of sulfonamido, carbonamido and sulfur bridges and "appellants have presented no evidence, nor offered any reasons, why one skilled in the art would not accept this teaching on its face." The solicitor also argues that the teaching in Badische that hydroxy and dihydroxy phenyl moities are equivalent is limited to the sulfur-bridge containing dyes claimed therein. "Should the Court determine that the Example 1 dye of Huhne does indeed qualify as the closest prior art to the claimed dye, it is submitted that the deficiencies of the Lach declarations pointed out in the discussion of appealed claim 8 above apply equally well herein and are hereby incorporated by reference."
The PTO also argues that the declarations are "deficient with regard to both claims 8 and 9 on appeal because they rely on properties which have no basis in the specification."
From the declarations, it is clear that it is in conjunction with retanned chrome leather that appellants consider the claimed dyes to be particularly beneficial and unobvious as compared to prior art dyes. One skilled in the art, however, reading the specification is not even given a hint that the claimed dyes are especially useful for retanned chrome leather as distinguished from other leathers.
That is, the PTO states, knowledge of the undisclosed properties would not "inherently flow" from what was disclosed in the specification, citing and quoting from In re Davies, 475 F.2d 667, 177 U.S.P.Q. 381 (C.C.P.A. 1973).
OPINION
We agree with the solicitor that the cited references establish prima facie obviousness. One of ordinary skill would have had reason to expect, given the close structural similarity of the Huhne and Badische compounds and the teachings of Geigy, that use of a sulfonamido bridge in the benzidine component of the claim 8 and 9 compounds would have resulted in dyes possessing the same or only slightly different properties from prior art dyes.
Although appellants stated at the end of their reply brief that, "the existence of a reasonable case of prima facie obviousness is resolutely disputed," they stated in their main brief only that "citation of the Huhne and Geigy patents by the Examiner * * * made it necessary for Appellants to show that their dyes had unobvious properties." That a prima facie case of obviousness had been established was never seriously questioned by appellants during prosecution or on appeal.
The Lach declarations, however, state that "the differences in the washfastness between the new and the known dyes are extreme; in fact, the differences are as big as that between usefulness and uselessness." Those declarations also state that appellants' dyes have a much higher water-solubility than the prior art dyes and that such a characteristic is desirable as such dyes will, as corroborated by Lach's experiments, better color chrome side leather. Because dyes with lower water-solubility are generally more suitable for dyeing retanned chrome side leather, Lach was surprised that the claimed compounds also dyed retanned chrome side leather better than the prior art dyes; the result was not what, as an expert, he would have expected.
Although perception of color may, in essence, be a "subjective" determination, we believe that an expert's evaluation in this field is entitled to more weight than that of a layman. In re Neave, 54 C.C.P.A. 999, 1007, 370 F.2d 961, 968, 152 U.S.P.Q. 274, 279-80 (1967). Therefore, because the qualifications of Lach and the test procedures which he employed are unchallenged, the board's holding that "a more dramatic difference in results" is required constitutes reversible error, the board having erroneously substituted its judgment for that of an established expert in the art. Lach's unequivocal analysis of washfastness, water-solubility, and color cannot be ignored in favor of the board's visual inspection and characterization of the evidence. In re Neave, supra.
Accordingly, weighing all the facts, we hold that the record establishes that the claim 8 dye would have been unobvious because it possesses overall, unexpectedly superior dye properties compared with the prior art. Cf. In re Crounse, 53 C.C.P.A. 1390, 1393, 363 F.2d 881, 884, 150 U.S.P.Q. 554, 557 (1966).
With respect to the claim 9 dye, however, we are persuaded that the prima facie case stands unrebutted because appellants did not compare it with the closest prior art.
Lach stated that Huhne "is considered to represent the closest prior art in the present case and the dyes used for the comparisons were chosen from this reference." Thus, Lach compared the claim 9 dye with the Huhne example 1 dye.
The rejection of claim 9, however, was made on Badische, which discloses a dye which is identical except for a thio (-S-) linking group rather than appellants' sulfonamido (-SO[2]NH-) link. Geigy teaches that those linking groups are interchangeable as used in trisazo dyes. Because this placed the burden on appellants to demonstrate that they are not equivalent, that is, that the substitution of the sulfonamido link led to an unexpectedly superior dye, it was not sufficient that appellants compared the claim 9 dye to one which had two differences, a carbonamido linkage and a mono-hydroxy terminal phenyl group. Such comparison would not demonstrate the criticality of the linkage change. Accordingly, we hold that the rejection of claim 9 was justified.
The assertion by the solicitor that the Lach declarations are deficient because they allegedly rely upon properties which have no basis in the specification is one of three new arguments raised for the first time on appeal. Accordingly, it will not be considered. The other two were withdrawn by the solicitor at oral argument.
The decision of the board is reversed with respect to claim 8 and affirmed as to claim 9.
MODIFIED.