Opinion
Patent Appeal No. 6202.
May 15, 1956.
John T. Kelton, New York City (Paul A. Rose, Washington, D.C., David A. Woodcock, Charles E. Howson, New York City, of counsel), for appellants.
J.M. Castle, Jr., and A. Newton Huff, Wilmington, Del., for appellee.
Before JOHNSON, Acting Chief Judge, and WORLEY and JACKSON (retired), Judges.
This is an appeal from the decision of the Board of Patent Interferences of the United States Patent Office awarding priority of invention of the subject matter to the senior party Bullitt, appellee here. The interfering subject matter is defined in the single count as follows:
A fluorinated acyl peroxide having the formula
O O | | F(CF2)n C-O-O-C(CF2)n-F wherein n is an integer of at least 2.
Bullitt is a patentee, but since the application of the junior party, Young and Stoops, was copending with that on which Bullitt's patent was granted, Young and Stoops have the burden of proving their case by a preponderance of the evidence.
Bullitt did not take testimony and is therefore restricted to his filing date, March 4, 1950, for conception and constructive reduction to practice. Young and Stoops took testimony purporting to show the preparation and successful use of bis (heptafluorobutyryl) peroxide in November 1949, and the preparation of an additional amount of the same compound in January or February 1950. It is conceded in Bullitt's brief that such compound satisfies the requirements of the interference count. The board held that the work described in the testimony was insufficient to establish that the compound prepared was actually bis (heptafluorobutyryl) peroxide, and that the junior party had therefore failed to establish an actual reduction to practice. The correctness of that holding is the sole issue to be determined here.
The testimony shows that during 1949 and 1950 Young and Stoops were employed as research chemists by the Carbide and Carbon Chemicals Corporation, which, on January 1, 1950, became a division of Union Carbide and Carbon Corporation, at Charleston, West Virginia. One of the problems with which they were concerned at that time was in finding a catalyst suitable for use in commercial preparation of polychlorotrifluoroethylene, also called fluorothene, which was prepared by a polymerization process. Prior to September 1949 a certain amount of success had been had with the use of bis (chloroacetyl) peroxides as catalysts, but it was not considered sufficiently attractive for commercial use.
It appears those peroxides were prepared from the corresponding chloroacetic acid by first converting the acid to the acid chloride by reacting it with benzotrichloride, usually in the presence of zinc chloride as a catalyst, and then dissolving the acid chloride in an organic solvent known as a "Freon" and adding that solution to an aqueous solution of sodium peroxide, with vigorous agitation, at a temperature of about zero degrees Centigrade. This resulted in an exothermic reaction, after which an aqueous layer and a Freon layer were formed. The bis (chloroacetyl) peroxide resulting from the reaction was contained in the Freon layer which was drawn off and stored at a low temperature until it was to be used. The preparation of the acid chloride was normally carried out by one Byron Stewart, and that of the peroxide by a Marvin P. Wright, both of whom worked under the supervision of, and in consultation with, Young.
The testimony of Stoops and Young, corroborated by that of Stewart and Wright, indicates that by about September 1949 it had been decided to try certain fluorine compounds as polymerization catalysts instead of the chlorine compounds previously used. Accordingly, two pounds of trifluoroacetic acid were obtained from the Minnesota Mining and Manufacturing Company.
Stewart testified that, at the request of Young, he reacted some of that acid with benzotrichloride in the presence of zinc chloride. The report of that work, dated September 27, 1949, and identified by Stewart, was offered in evidence and is headed "Preparation of Trifluoroacetyl Chloride." Stewart testified that that chloride was collected and turned over to Wright.
The testimony of Wright is to the effect that he received the trifluoroacetyl chloride from Stewart and reacted it with sodium peroxide in an aqueous solution in the manner previously followed in connection with similar chlorine compounds, except that sodium chloride was added so the reaction temperature might be kept lower because of the low boiling point of the chloride.
The record of that work shows that it was done on October 12, 1949. Wright stated that the resulting peroxide was immediately turned over to Richard T. Wood. That is corroborated by Wood, who testified that some of the material was used in a polymerization on the same day and that a second run was made on October 17, 1949. Apparently no complete analysis of the peroxide was made.
It is not contended that the fluorine compound above discussed satisfied the requirements of the interference count, but that the work in connection with it is pertinent as leading up to the alleged production and use of bis (heptafluorobutyryl) peroxide, and as showing the laboratory procedures followed by Stoops and Young and their associates in the preparation of compounds generally similar to that here in issue.
As the result of information that heptafluorobutyric acid was obtainable from the Minnesota Mining and Manufacturing Company, Stoops and Young decided to try that compound and ordered one pound of it, which was received on November 10, 1949. It was turned over to Stewart and Young with instructions to prepare the acid chloride from it in the usual way. Stewart accordingly proceeded to react the heptafluorobutyric acid with benzotrichloride in the presence of a zinc chloride catalyst in the same general manner which had been employed in preparing the acid chlorides above referred to. The resultant product, which Stewart described in his testimony as heptafluorobutyryl chloride, and which is so referred to in his contemporary work report, was turned over to Wright. The latter, on November 17, 1949, added it to a Freon solution and then reacted it with an aqueous solution of sodium peroxide, the mixture being agitated vigorously. That resulted in an exothermic reaction, after which the Freon layer, containing the peroxide, was drawn off in accordance with the usual procedure.
The layer thus drawn off was analyzed for peroxide content. It is not necessary to consider the manner in which that was done but it is clear that while the tests showed some peroxide to be present in substantial amounts, they were not of such a nature, in themselves, to establish what the particular peroxide was. The peroxide was not separated from the Freon solution nor identified by a complete analysis. It is assumed by Wright and all the others associated with the work that the peroxide produced was bis (heptafluorobutyryl) peroxide, and it is so identified in the work reports. It was tested as a polymerization catalyst on November 17, 1949, with satisfactory results.
The testimony as to further work by Stoops and Young in 1950 is similar to that discussed above and need not be considered in detail. If the work done in 1949 was insufficient to effect a reduction to practice, it is clear that the work done in 1950 was similarly insufficient.
The testimony offered by the junior party is full, clear, and amply supported by contemporary documents, and there is no reason to doubt that the facts are substantially as above outlined. Bullitt does not attack the testimony as to its veracity, but contends that it fails to show the required degree of certainty that the compound made and tested in 1949 was actually bis (heptafluorobutyrul) peroxide. That contention was sustained by the Board of Patent Interferences.
Bullitt urges that the evidence presented fails to establish the identity of the starting material used in the November 1949 work as being heptafluorobutyric acid. Admittedly that material was not analyzed by Young and Stoops after it was received, but reliance was placed primarily on the labeling of the bottle in which it was contained. As to Bullitt's contention, the board said:
"Applying the foregoing principles, we agree with Young et al. that it was not necessary to verify the identity of the one pound sample of starting material received from Minnesota Mining. Even though the product came from a pilot plant operation, it was nevertheless shipped in the course of business by a reputable manufacturer and there seems little likelihood that the bottle did not contain the chemical as labeled. The possibility of error must be recognized, but it is too remote to require independent testing for identification."
Under the circumstances here, we think that conclusion sound and proper. It is, of course, always possible that the actual composition of a commercial product may differ from that indicated by its label, but it is certainly not the usual practice for purchasers to analyze such products before using them. The fact that the heptafluorobutyric acid here involved was a comparatively new product and came from a pilot plant does not necessarily alter that situation. We think it reasonable to assume the operators of such a plant would take care to see that their product was what it purported to be. The testimony that the acid purchased in November 1949 had a "goaty" odor, suggests that it was correctly labeled, although it is by no means conclusive, since the record indicates that other butyric compounds also have such an odor.
Under the circumstances of this case, and in view of the fact that there is nothing to indicate that the acid in question was not properly labeled, we are of the opinion the starting material used by Young and Stoops in 1949 has been satisfactorily shown to have been heptafluorobutyric acid.
Notwithstanding its holding that the starting product of Young and Stoops was heptafluorobutyric acid, the board held the evidence was insufficient to identify the final product as bis (heptafluorobutyryl) peroxide. In so holding, the board noted that the material in question was not isolated from the solution at any time prior to Bullitt's filing date.
Since the interference count does not call for the product claimed in pure or substantially pure form, the failure to isolate it from solution would not necessarily be fatal to a claim of reduction to practice. In that connection it is pointed out by Young and Stoops that in the Bullitt patent the only example which describes the production of a compound coming within the interference count does not show or suggest its recovery from solution in an organic solvent; and that the only disclosed analysis of the product relates to analysis of that solution for peroxide content.
The board cited the following decisions as indicating the degree of certainty with which a product must be identified when its composition is material to the decision in an interference: Farrington v. Mikeska, 155 F.2d 412, 33 C.C.P.A., Patents 1073; Brooker v. Riester, 161 F.2d 745, 34 C.C.P.A., Patents, 1088; Searle v. Glarum, 179 F.2d 974, 37 C.C.P.A., Patents 896; Kvalnes v. Wright, 183 F.2d 193, 37 C.C.P.A., Patents, 1147; and Guinot v. Hull, 204 F.2d 281, 40 C.C.P.A., Patents, 982.
Those cases have been discussed extensively in the board's decision and in the briefs of the parties. They relate to the rule that a complete chemical analysis is not always necessary to identify a product, but that various other factors, including the materials used, the procedures followed, and the skill and experience of the experimenters, are to be considered.
The question generally is whether, when all the circumstances are considered together, there is a reasonable certainty as to the identity of the product. Proof beyond a reasonable doubt is not necessary where, as here, the applications of the parties were copending.
It is clear that the procedure followed with heptafluorobutyric acid in 1949, by Young and Stoops and their associates, paralleled closely similar procedures carried out previously by them in producing bis (chloroacetyl) peroxides from chloroacetic acids. In view of the close relationship between fluorine and chlorine there would be good reason to suppose that corresponding compounds of the two would behave in a similar manner, and the experimenters evidently proceeded on that theory.
Young and Stoops argue that the procedures in question are standard, and cite a number of prior patents and publications allegedly disclosing them. Bullitt has pointed out various differences between the disclosures of the latter and the procedures followed by Young and Stoops. We are of the opinion, however, that those differences are of such a minor nature they fail to show that those skilled in the prior art, as of 1949, were not familiar with processes similar to those employed by Young and Stoops for converting halogenated carboxylic acids to the corresponding acid-chlorides and for converting such chlorides to peroxides of the kind here involved. Naturally such procedures varied somewhat in accordance with the specific materials used.
With respect to that matter, the board said:
"It is true the reactions carried out by these workers [Stewart and Wright] were known and understood but they did not enjoy the routine usages such as the procedures employed in Kvalnes v. Wright."
We agree with the board that the reactions in question were known and understood and in view of that fact it does not appear to be particularly material whether they enjoyed routine usage. Once it is known and understood what the result of a reaction is, no further information of any consequence is normally gained by routine repetition.
Moreover, it is to be noted that the testimony clearly shows that Young, Stoops, Stewart, and Wright were quite familiar with the general type of procedure which was followed in the alleged production of bis (heptafluorobutyryl) peroxide in 1949. Thus Stewart testified that when Young asked him to prepare the acid chloride of heptafluorobutyric acid, "He [Young] only suggested that we prepare it by our usual procedure. There was no other stipulation at that time," while Wright's report of his work, headed "Preparation of Bis (Heptafluorobutyryl) Peroxide," refers to "Usual batch equipment." It thus appears that the work under consideration was regarded by those who carried it out as being of a routine nature, and that they had no doubts as to what was to be done or as to what results were to be expected.
It is our opinion, upon full consideration of the entire record, that the background of the work done by Young and Stoops and their associates in November 1949 was such as to raise a very strong presumption that that work would result in the production of bis (heptafluorobutyryl) peroxide. That presumption is strengthened by the facts that the acid chloride produced by the first step in the process carried out had a narrow boiling point, indicating a reasonably pure product, as distinguished from a mixture of a number of different products, and that the reaction of the acid chloride with sodium peroxide was an exothermic one.
Moreover, analysis showed that the final Freon solution contained peroxide, although that analysis did not prove that it was bis (heptafluorobutyryl) peroxide; and the solubility of the peroxide in the Freon indicated that it was organic. Additionally, the fact that the final product was effective as a polymerization catalyst, for which purpose compounds generally similar to bis (heptafluorobutyrul) peroxide had been found suitable, provides further evidence as to the identity of that product.
It is also noted that there seems to have been no circumstance to suggest that the product in question was anything other than bis (heptafluorobutyryl) peroxide, and no definite suggestion has been made as to what else it could have been. It is quite true, as Bullitt points out, that he does not have the burden of showing what the product actually was, but the lack of any specific alternative suggestion is a factor to be considered in evaluating identification of the product.
It is clear from the record that no one associated with the work done in November 1949 had any doubt that the resultant product was bis (heptafluorobutyryl) peroxide. It is identified in that manner in the reports of its preparation and of its use as a catalyst, as well as in a progress report authored by Stoops, Young, Stewart, and Wright, dated November 25, 1949, which describes its production and proposed use.
While the board seems to discount the value of the opinions of Stewart and Wright, the record shows that, although not college graduates, they had a good basic background in chemistry, together with substantial practical experience in the particular field involved in this case. Moreover, Young and Stoops, who supervised the work of Stewart and Wright, are graduate chemists of long and extensive experience.
It is to be noted that the reports referred to above were made long before any issue of prior inventorship had arisen. There is no reason to suppose they represent anything other than the honest opinions of the persons who made them. Under such circumstances, the uniform conclusion of four skilled workers, familiar with the development, that bis (heptafluorobutyryl) peroxide had been produced is entitled to great weight.
As pointed out in the decisions cited above, it is not necessary that the product relied on by a party in cases such as the present one be identified with absolute certainty, or that a complete analysis should invariably be made. It is sufficient if the evidence, taken as a whole, establishes, with reasonable certainty, the identity of the product. We are of the opinion the evidence in the present case is sufficient to prove, with the required degree of certainty, that bis (heptafluorobutyryl) peroxide was successfully produced by Stoops and Young in November 1949. Since the record shows a successful use of the compound as a catalyst during that month, the junior party is entitled to a date of reduction to practice as of November 1949 and, of course, a date of conception at least as early as that. Those dates being prior to Bullitt's filing date of March 4, 1950, to which he is restricted, Young and Stoops must prevail.
The decision of the Board of Patent Interferences is reversed.
Reversed.
JACKSON, Judge, retired, recalled to participate.
O'CONNELL and COLE, JJ., because of illness, did not participate in the hearing or decision of this case.