Appellant seems to see some significance in the reference in claims. 24, 25 and 26 to the stock being "reduced to" and "having" a "desired thickness dimension." It is apparent, however, that "desired" does not distinguish from the thickness dimension which Weiss desires to provide with the corresponding steps of his method.

It is further urged by appellant that claim 28 is not met by Weiss because of its recitation of performing the repeat steps in "discrete spaced apart areas to form spaced sections and portions connecting the sections." We do not agree. Repeated rolling steps over adjacent areas in Weiss will result in a plurality of spaced apart raised portions with lower connecting portions therebetween and thereby meet the terms of the claim.

Appellant's arguments depend essentially on the meaning of "adjacent" and raise no points with respect to claims 25 and 26 other than those relied on with regard to base claim 24. Moreover, separate consideration of claims 25 and 26 compels the conclusion that their rejection, like that of the other appealed claims, is free of reversible error. The decision is affirmed.

387 F. 2d 663; 156 USPQ 39

IN RE SYDNEY M. SPATZ AND MARVIN KORAL (No. 7666)

PATENTS

1. APPEALS TO COURT-ISSUES DETERMINED EX PARTE CASES, EVIDENCE-JUDICIAL NOTICE

Article published in scientific journal four years after applicants' filing date is not properly before court where Board had refused to consider article because it was not presented to examiner during prosecution; court will not consider article or take judicial notice of it.

2. PATENTABILITY—EVIDENCE OF DELAY AND FAILURE OF OTHERS TO PRODUCE

INVENTION

Such secondary considerations as long felt but unsolved needs and failure of others are given effect in determining obviousness under 35 USC 103; however, rule does not apply where it is not clear from evidence that problem existed or that skilled workers knowledgeable in the art had failed to solve problem.

3. WORDS AND PHRASES

"Did not consistently" produce polyesters is equivocal, relative expression broad enough to allow production of polyesters as infrequently as one out of 100 times or as frequently as 99 out of 100 times; it does not establish that problem necessarily existed in prior art or the criticality of claimed content alleged in specification.

United States Court of Customs and Patent Appeals, December 14,

[Affirmed.]

1967

Appeal from Patent Office, Serial No. 6,635

Kenneth D. Tremain, Frederick N. Weinfeldt, George B. Campbell, and Michael S. Jarosz for appellants.

Joseph Schimel (Jack E. Armore, of counsel) for the Commissioner of Patents.

[Oral argument October 4, 1967 by Mr. Jarosz and Mr. Armore]

Before WORLEY, Chief Judge, and Judges RICH, SMITH, ALMOND, KIRKPATRICK*

WORLEY, Chief Judge, delivered the opinion of the court:

1

This appeal is from a decision of the Board of Appeals affirming the rejection of claims 1-12 in appellants' application 1 for "Fire-Retardant Composition Derived from Tetrabromophthalic Acid Anhydride,” as "unpatentable" in view of certain prior art.

The subject matter is reflected in claims 1 and 6:

1. A method of preparing polyester resins that have fire-retardant characteristics when cured, which comprises: preparing a mixture of glycol and a component of the group consisting of saturated dicarboxylic acids and anhydrides thereof, said component containing at least 10 mol percent of tetrabromophthalic acid anhydride that has not above about 5 ppm. of ferric iron impurity associated therewith, and heating said mixture to form a polyester therefrom.

6. A polyester resin that has fire-retardant characteristics when cured, which comprises the reaction product of glycol and a component of the group consisting of saturated dicarboxylic acids and anhydrides thereof that contain at least 10 mol percent of tetrabromophthalic acid anhydride that has not above about 5 ppm. of ferric iron impurity associated therewith.

The feature of the claims asserted to be nonobvious is the use of fat least 10 mol percent of tetrabromophthalic acid anhydride (TBPAA) that has not more than 5 parts/million (ppm) ferric iron impurity as the acid component of a fire-retardant polyester resin.

Bjorksten, Polyesters and Their Application, pub. 1956, Reinhold Publishing Corp., New York, N. Y., page 160.2

* Senior District Judge, Eastern District of Pennsylvania, sitting by designation. 1 Serial No. 6,635, filed February 4, 1960.

The appeal was originally argued Oct. 6, 1966. Reargument was heard Oct. 4, 1967. 2 The examiner cited Bjorksten to show that diallyl phenyl phosphonate and hexachloroendomethylene tetrahydrophthalic acid anhydride, materials recited in some of the claims, are conventional components of flame resistant polyester resins. Appellants do not argue otherwise.

[1] Appellants ask us to take judicial notice of an article published in Chemical and

A detailed discussion of the Lundberg, Phillips and Wismer reference is unnecessary, since appellants admit that the art, as represented by those references, knew that TBPAA "could be used in the preparation for polyesters with fire-retardant properties.”

Mindful of a similar concession below, the board sustained the examiner's rejection of the claims as "unpatentable over" Lundberg alone or considered with Phillips. With respect to the limitation in the claims calling for not more than about 5 ppm of ferric impurity, the board stated:

One skilled in this art would not be expected to employ an impure reactant, but would follow normal chemical practice and use a pure phthalic anhydride in the esterification. Since the references are silent as to any content of ferric ion it cannot be presumed that such ion is necessarily present in the starting material. * * * We therefore agree with the Examiner that a pure or iron-free reactant would normally be employed, thus rendering obvious appellants' contribution.

We also note there to be no evidence in this record that all commercially available tetrabromophthalic anhydride would necessarily contain ferric ion in a proportion above the 5 ppm. which appellants consider to be the operative limit of their contribution. The use of commercially available tetrabromophthalic anhydride having a low ion content is equally obvious in the polyesterifications of the prior art as the high iron content urged by appellants.

The board found it to be apparent from the background references cited by the examiner that purified phthalic acids would normally be employed by one of ordinary skill in the art in various esterification. reactions. O'Neill, for example, discloses the desirability of removing colored impurities in terephthalic acid prior to using it in the manufacture of polyester fibers and films in order to obtain a product having good color. Tate removes certain impurities from various phthalic acids prior to preparing polyesters therefrom in order to improve the speed of the esterification reaction as well as the color, crystallinity and melting point of the ultimate polyester. Hoffman discloses that crystalline tetrachlorophthalic acid anhydride 3 "must be thoroughly purified" before it may be used for industrial purposes, and details a process for the removal of ferric iron and other impurities therefrom to obtain a product with the degree of purity required for use in the plastics and dye industries.

3

While appellants concede that those in the art were aware that TBPAA could be used in the preparation of polyesters with fireretardant properties, they contend that there was a practical industrial

Engineering News about 4 years after appellants' filing date. Appellants made a similar request to the board but the board refused to consider the article because it was not presented to the examiner during prosecution. Under the circumstances, the article is not properly before us and will not be considered. In re Cofer, 53 CCPA 830, 354 F. 2d 664, 148 USPQ 268.

* The record shows that tetrachlorophthalic acid anhydride, like TBPAA, is a material known to be useful in formation of fire-retardant polyesters.

problem which made it difficult to prepare polyesters from commercially available TBPAA. Appellants urge that their contribution resides in the "triple-faceted discovery” that: (1) the cause of the difficulty in processes for preparing polyesters with TBPAA resides in the presence of ferric iron in impure TBPAA; (2) the remedy resides in the use of TBPAA containing not more than 5 ppm ferric impurity in the esterification process; and (3) when TBPAA of such purity is used, fire-retardant properties far in excess of expectation are achieved in the products of the process.

[2] We find no reversible error in the manner in which the examiner and board employed the primary and background references as evidence of the obviousness of the claimed subject matter. While such secondary considerations as "long felt but unsolved needs, [and] failure of others" are to be given effect in determining obviousness or nonobviousness under § 103, it is not clear from the evidence in the record here that the industrial problem of which appellants speak in fact existed, or that skilled workers knowledgeable in the art had failed to find any solution to the problem. See In re Allen, 51 CCPA 809, 813-14, 324 F.2d 993, 997, 139 USPQ 492, 495. Indeed, the one tangible, extraneous piece of evidence appellants cite in their specification in an effort to show that a problem did exist is in marked contrast with the later disclosures of Phillips and Wismer, who speak matter-of-factly of the use of TBPAA in forming polyesters as if no problem existed at all.

5

[3] In further support of their arguments, appellants point to Example II (particularly Table 1) and Example V of their application. We do not think those examples support appellants' assertions of nonobviousness of the process and composition claims. With respect to the professed criticality of the use of TBPAA containing the claimed ferric iron content to facilitate an allegedly difficult esterification reaction, it is true that Example V of the specification states that "No polyester could be produced" when a sample of "commercial" grade TBPAA, containing 34 ppm ferric impurity, was employed in the esterification process, while the use of "purified" TBPAA containing 1.2 to 5 ppm of ferric iron "consistently produced polyesters." However, the specification also states that "technical" grade TBPAA, said in Example III to be prepared by "the conventional method wherein phthalic acid anhydride is brominated in the presence of

See Graham v. John Deere, 383 U.S. 1 at 17, 18, 35 and 36.

The specification says:

✦✦✦ it has been stated that attempts to incorporate desirable flame-retardant properties into high polymer polyesters by use of chloro- and bromo-derivatives of phthalic anhydride for use in surface coatings have been either unsuccessful or economically unattractive (Kirk-Othmer, Encyclopedia of Chemical Technology, Volume 10, page 595-Interscience Encyclopedia, Inc., 1953).

oleum" and to contain 8 ppm ferric impurity, "did not consistently produce polyesters" when used in the esterification process. (Emphasis supplied). The record contains no data to further elucidate what the expression "did not consistently" means in the present context. It seems apparent that the phrase is broad enough to allow production of polyesters from "technical" grade TBPAA containing 8 ppm ferric impurity as infrequently as 1 out of a 100 times, or, for that matter, as frequently as 99 out of a 100 times. We do not think the equivocal, relative language employed by appellants establishes that a problem necessarily existed with respect to the use of the "technical" grade TBPAA of the prior art or the criticality of the claimed ferric iron content they allege here.

Insofar as the efficacy of Example II of the application to show nonobviousness of the claimed polyester composition is concerned, it should be noted that appellants have only shown that resins produced from TBPAA containing less than 5 ppm ferric impurity have superior fire-retardant properties vis-a-vis resins produced from other phthalic acids, subject matter not relied on by the Patent Office in the rejection. There is no comparison of the fire-retardant properties of the claimed resin with like properties of a resin produced from TBPAA containing slightly more than 5 ppm ferric impurity.

While we appreciate appellants' arguments, we are satisfied the board committed no reversible error in sustaining the rejection. The decision is affirmed.

ALMOND, Judge, concurring.

The question before us is whether, at the time the invention was made, it was or was not obvious to use TBPAA having less than 5 ppm ferric iron in a polyesterification reaction. It seems clear that if TBPAA having the required degree of purity was in fact readily commercially available at the time the invention mas made, it would have been quite obvious to use it in such reactions, particularly in view of the suggestion of Hoffman that ferric iron is desirably removed from the very similar compound tetrachlorophthalic acid anhydride before its use in plastics production.

On the other hand, if all TBPAA which was commercially available at the time the invention was made had a high iron content and hence was unusable in polyesterification reactions, it would appear that appellants have made an advance in the art by recognizing a solution to an existing problem. Such a solution can be given effect in determining the obviousness or nonobviousness under section 103.

Our problem, therefore, seems to revolve about the availability or nonavailability of TBPAA of low ferric iron content at the time the