Application of Jan Ide De Jong, 416 F.2d 1401 (C.C.P.A. 1969)

This is an appeal from the decision of the Patent Office Board of Appeals, adhered to on reconsideration, affirming the rejection of claims 1 and 3 to 8, inclusive, of appellant's application entitled "Processes for Obtaining Substantially Pure 4,4'-Dihydroxy-Diphenyl-Propane 2.2."¹  No claims have been allowed.

The invention relates to an improved process for obtaining substantially pure diphenylol propane (4.4'-dihydroxy-diphenyl-propane 2.2), hereinafter DPP, from the product of the condensation reaction of phenol with acetone in the presence of an acid catalyst. Generally, the DPP thus obtained in the condensation reaction product is in the form of very fine crystals, which causes difficulty in effecting the separation of the solid product from the mother liquid by either filtering or centrifuging. Prior attempts to obtain substantially pure DPP, including the manufacture of DPP in the presence of inert liquid diluents such as liquid hydrocarbons or halogenated hydrocarbons or the recrystallization of DPP out of an organic solvent such as toluene or xylene following heating of the crude condensation product with water in the presence of a base, are said to suffer from economical drawbacks due to requirements of large amounts of acid or solvent as well as from the danger of decomposition of the DPP during recrystallization if the starting product is not acid free.

The present invention allegedly obviates the disadvantages mentioned above and results in a product of greatly improved color and purity. More specifically, this is accomplished by adding either to the crude condensation product or to the condensation reactants suitable amounts of additional water and a water-immiscible organic solvent or diluent, heating the mixture to a temperature below the boiling point thereof to form a wholly liquid system having present an organic phase containing at least 20% by weight of DPP, and cooling the system below the crystallization point thereof so that pure DPP crystallizes from the organic phase. The solubility of DPP in the diluent or solvent preferably should be poor while the diluent should provide a mixture with DPP and water that is fully liquid upon heating to a temperature that lies beneath the boiling point of the mixture.

Claims 1 and 8 are illustrative:

1. In the process for producing diphenylol propane [DPP] by the condensation reaction of phenol with acetone in the presence of an acid catalyst, the steps for obtaining substantially pure diphenylol propane comprising

heating a mixture consisting essentially of the acid containing reaction product, water in an amount between 10% and 1000%, by weight, of the amount of diphenylol propane in said reaction product, and a water immiscible organic solvent in an amount between 10% and 400%, by weight, of said amount of diphenylol propane, said mixture being heated to a temperature between 70°C and 100°C which is below the boiling point thereof, and said organic solvent being selected from the group consisting of toluene, xylene, monochlorobenzene, 1-1-2-2-tetrachloroethane and 1-2-dichloroethane to form, with said diphenylol propane and water in said relative amounts, a system which is wholly liquid at said temperature and which includes at least a first organic phase containing said solvent, at least 20% of diphenylol propane and a minor part of water, and a second aqueous, acid-containing phase;

cooling below the crystallization point of said system, thereby to cause substantially pure diphenylol propane to crystallize therefrom; and

collecting the crystallized pure diphenylol propane.

8. The process as in claim 1;

wherein said organic solvent is added to the condensation reactants so that the condensation reaction occurs in the presence of said organic solvent, and said water is added upon the completion of the condensation reaction.

Claims 3 to 7, inclusive, are dependent on claim 1 and the limitations added by such claims are not in issue, the parties having agreed that consideration of representative claim 1 will be dispositive of claims 1 and 3 to 7.

The appealed claims were rejected as being unpatentable under 35 U.S.C. § 103. The references relied upon are:

Grimme et al. 2,959,622 November 8, 1960 (Grimme) Stanley et al. 557,976 December 14, 1943 (Stanley-Great Britain)

Grimme discolses a process for purifying DPP whereby the condensation reaction product, crude DPP, is washed to remove therefrom the acid condensation agent and then neutralized with a compound having an alkaline reaction. The crude neutralized product is then dissolved while hot in an organic water-immiscible solvent such as ethylene dichloride or monochlorobenzene. After completion of the dissolving process, there forms an aqueous layer and an organic layer containing the DPP and unreacted phenol. After separation, the organic phase is cooled to precipitate out DPP which, after washing, is a practically colorless homogenous crystalline product. The quantity of the solvent used is variable, but at least such a large quantity is to be used as will completely dissolve the DPP.

Stanley discloses that the condensation reaction of phenol with acetone to produce DPP may be advantageously carried out in the presence of a solvent such as dichloroethylene or chlorobenzene.

The examiner rejected claims 1 and 3-7 as unpatentable over Grimme under 35 U.S.C. § 103. While noting that the amount of water-immiscible organic solvent employed by Grimme is 500% or more of the crude DPP to be purified, he reasoned that "it is within the expected skill of the art to determine optimum proportions for the purpose of achieving better results."

Claim 8 was rejected as unpatentable over Grimme in view of Stanley under 35 U.S.C. § 103 since the examiner considered it obvious to add the organic solvent to the reactants prior to the production of DPP.

The board affirmed the rejections, stating as to claims 1 and 3-7:

It is a common, routine expedient in operating industrial processes to use the minimum practical amount of materials. We consider that it would be obvious to those of average skill in the art to determine the practical minimum proportion of organic solvent in the examples of the reference. It appears that such minimum will be materially lower than the figures suggested by the table in column 3 of the reference, but this alone does not, in our opinion, provide a basis for patentability. We do not find that the reference requires an amount of solvent at least equal to the solubility at the boiling point in the table in column 3. We believe that workers in the art would recognize that solubility values obtained with pure two component systems (as in the table) are rarely the same as for the same system with impurities and a third component added, viz., water in the present case. Appellant states that he "has found that, in the presence of water the amount of solvent required to produce a wholly liquid with D.P.P. is far less than the amount of solvent required to dissolve the D.P.P. therein, as disclosed by Grimme et al." * * * We do not dispute this, but it is a result that is inherent in an obvious variation of the working examples of the reference. Note In re Libby, 45 CCPA 944; 1958 CD 324; 733 OG 294; 255 F.2d 412; 118 USPQ 94.

We recognize that all the claims require that the aqueous phase remains acidic, whereas the reference neutralizes this phase. We are not convinced by the record that this difference, which can be only a very slight amount of acid in the acqueous phase, assures an unobvious result.

The board also felt that the step which dependent claim 8 added to its parent claim 1 was shown by Stanley.

Appellant's request for reconsideration, based on the argument that he discovered that the melting point of DPP is reduced in his system, was denied thusly:

Since merely reducing the amount of solvent in the reference process would produce the wholly liquid system defined in the claims, all steps and conditions of the claims would be satisfied. It is therefore patentably immaterial whether the DPP dissolved or melted.

Appellant's position that the rejections are founded on reconstruction of the art based on hindsight gained from his disclosure includes the specific argument that his process is based on the

* * * recognition of the fact, apparently not appreciated or understood by Grimme et al., that when diphenylol propane is mixed with water and selected solvents the melting point of diphenylol propane is greatly reduced from the melting point (about 150°C.) of dry diphenylol propane so that a wholly liquid system can be formed from such mixture at a temperature below the boiling point thereof (in the range between 70°C. and 100°C.) even though the amount of solvent in the mixture is far less than that required to dissolve the diphenylol propane * * *.

In view of the foregoing, it is said, appellant's solvents are not selected on the basis of high solubility therein of DPP at the boiling point, as in Grimme, but rather on the basis of their ability to substantially reduce the melting point of DPP in the mixture. In fact, appellant's solvents are selected to have as a characteristic poor solubility of DPP so that only a small quantity of DPP is taken up in the organic phase of the mixture, with the result that the risk of decomposition, absent prior neutralization, during recrystallization is avoided.

It is urged that the claimed process is distinguished from the Grimme process in the following particulars:

1. The heated mixture has therein the acid containing reaction product and there is no prior neutralization of the reaction product as in Grimme et al.;

2. The maximum amount of the diluent or solvent in the mixture is 400% of the amount of diphenylol propane, whereas, in Grimme et al., the amount of the solvent is at least 450% of the amount of diphenylol propane; and

3. At least 20% of the organic, solvent-containing phase of the wholly liquid system is constituted by diphenylol propane, whereas, in Grimme et al., no more than 17-18% of the organic layer can be constituted by diphenylol propane.

Finally, appellant argues that several of the solvents specifically mentioned by Grimme, such as benzene and chloroform, are unsuitable in appellant's process due to the fact that the amounts of such solvents required to form the wholly liquid mixture are in excess of the 400% of the amount of DPP claimed. Appellant admits, however, that the solvents toluene, chlorobenzene and 1-2-dichloroethane (ethylene dichloride), specifically mentioned by Grimme, are included in the Markush group of appealed claim 1.

The solicitor counters by emphasizing the breadth of claim 1 in pointing out that the use of 400% of organic solvent, as permitted by claim 1, will result in dissolving over 66% DPP when monochlorobenzene is used and over 80% DPP when ethylene dichloride is used. Thus, it is urged, claim 1 permits a major portion of the DPP to be dissolved in those instances contrary to appellant's alleged objective rendering the alleged advantage of lesser amounts of solvent irrelevant. Moreover, it is argued, Grimme discloses a process embodiment wherein neutralization occurs subsequent to mixing and prior to heating and claim 1 does not exclude neutralization in that it recites "steps * * * comprising" thus leaving the claim open to other steps, while claim 4 calls for neutralization prior to crystallization as disclosed in the application.

While we are favorably impressed with appellant's arguments pointing out those instances where the amount of solvent required to provide a wholly liquid mixture with DPP in his process is considerably less than the amount of solvent required to dissolve the DPP therein, as disclosed by Grimme, the breadth of claim 1 requires that we focus our inquiry on the question of whether it would be obvious to one having ordinary skill in the art to employ an amount of solvent which is less than that amount "which is sufficient to completely dissolve the diphenylolpropane" as fairly disclosed by Grimme. When ethylene dichloride is employed, these amounts are 400% and 450%, respectively, for appellant's process and the process of the reference.

Considering the record, we find that the Patent Office's allegations regarding a "common, routine expedient," the "expected skill of the art to determine optimum proportions" or the belief "that workers in the art would recognize that solubility values * * * are rarely the same," true as they may be, are nevertheless unsupported by sufficient facts to overcome the strong suggestion implicit in Grimme that the minimum amount of solvent usable is that quantity which dissolves the DPP. Thus, the reference states:

The quantity of the solvent used is variable, but at least such a large quantity is to be used which is sufficient to completely dissolve the diphenylolpropane. The use of an exceeding quantity of the solvent is not injurious.

As to limiting the amount, Grimme mentions only:

In order to limit the quantity of solvent required, the solvent should preferably furthermore have a sufficiently great dissolving power for the diphenylol propane while hot or at the boiling point.

Clearly, Grimme does not teach the approach taken by appellant. In the absence of a factual basis for the Patent Office's allegations that that approach with its inherent results would be obvious, we are constrained to agree with appellant. The view we take of claim 1 is also dispositive of the remaining dependent claims.

The decision of the board is reversed.

Reversed.