Robeson appeals from the board's affirmance of the rejection of claims 2 to 31, all the claims of his application1 for a patent on a chemical process.

The invention relates to the production of trimethylolpropane, hereafter TMP, and its recovery from the reacted mixture. In that process formaldehyde is reacted with butyraldehyde in the presence of sodium hydroxide. Sodium formate is produced as a by-product.

The claims may conveniently be divided into three groups. Claim 8, representative of claims drawn to a process for the production of TMP, reads:

"8. Process for the production of trimethylolpropane which comprises continuously reacting, in an aqueous medium, a mixture of butyraldehyde, formaldehyde and sodium hydroxide, there being about 6 to 10 moles of formaldehyde per mole of butyraldehyde, to produce trimethylolpropane and sodium formate, continuously reducing the pH of the reacted mixture, continuously removing unreacted formaldehyde by distillation from the mixture and separating the trimethylolpropane from the sodium formate."

Claim 2, representative of claims drawn to the recovery of TMP by solvent extraction from a reaction mixture containing sodium formate, reads:

"2. Process for the separation of trimethylolpropane from an aqueous solution containing trimethylolpropane and sodium formate, which comprises extracting said solution with ethyl acetate to obtain a first liquid phase comprising a water solution of sodium formate and a second liquid phase comprising an ethyl acetate solution of trimethylolpropane, physically separating said phases to free said second liquid phase from said first liquid phase, and recovering the trimethylolpropane from said second liquid phase."

Claim 25, illustrative of claims drawn to the washing of the solvent extract with water to remove residual sodium formate, reads:

"25. Process for the separation of trimethylolpropane from an aqueous solution containing trimethylolpropane and sodium formate, which comprises extracting said solution with a liquid, substantially water-immiscible, solvent, for trimethylolpropane to obtain first liquid phase comprising a water solution of sodium formate and a second liquid phase comprising a solution of trimethylolpropane in said solvent, said solvent having a solubility of less than 0.5% in said water solution of sodium formate, physically separating said phases to free said second liquid phase from said first liquid phase, washing said second liquid phase with water to remove sodium formate therefrom while retaining the trimethylolpropane in solution in said solvent, and recovering the trimethylolpropane from the washed second liquid phase."

  Walker et al.     2,135,063       November 1, 1938
  Poitras et al.    2,420,496       May 13, 1947
  Elgin             2,479,041       August 16, 1949
  Robeson           2,790,837       April 30, 1957
                    (Filed June 8, 1954)
  Gottesman et al.  2,806,889       September 17, 1957
                    (Filed May 20, 1954)
  Gottesman         2,806,890       September 17, 1957
                    (Filed May 20, 1954)
  De Lorenzo        2,806,892       September 17, 1957
                    (Filed June 2, 1954)
  Albert (France)   1,081,691       June 9, 1954

Certain of the claims are rejected on the ground of double patenting, others as being obvious in view of the prior art. The rejections will be treated separately and the references will be discussed in connection with each ground of rejection.

Claims 7, 8, 9, 13, 14, 15 and 30, drawn to the reaction of butyraldehyde, formaldehyde, and sodium hydroxide, are rejected as being obvious over the claims of Robeson, appellant's own patent,2 either alone or in view of De Lorenzo and Poitras et al.

"1. Process for the production of trimethylolethane, which comprises continuously reacting, in an aqueous medium, a mixture of propionaldehyde, 5 to 15 moles of formaldehyde per mole of propionaldehyde and sodium hydroxide, to produce trimethylolethane and sodium formate, continuously reducing the pH of the reacted mixture, then continuously removing unreacted formaldehyde by distillation of the mixture under superatmospheric pressure, and thereafter isolating crystals comprising trimethylolethane from the mixture."

The only material difference between the claimed process, as illustrated by claim 8, and that of Robeson's patent claims is the substitution of butyraldehyde for propionaldehyde to obtain trimethylolpropane (TMP) instead of trimethylolethane (TME). No alteration in the type of reaction appears to be involved since the reaction claimed in the patent takes place at the functional groups and the present process merely involves lengthening the side chain by a CH₂ group.

Claims 7 and 8 additionally recite the extraction of TMP with ethyl acetate. For that reason the examiner also relied on De Lorenzo and Poitras et al. De Lorenzo discloses the extraction of trimethylolethane with ethyl acetate, and Poitras et al. disclose that TMP and TME have similar solubility characteristics. The claims of the Robeson patent recite extraction of TME with a lower aliphatic alcohol.

The examiner was of the opinion that, in view of the similar solubility characteristics of TME and TMP, extraction of one with ethyl acetate would suggest extraction of the other in the same way. Appellant submitted an affidavit to show that ethyl acetate was an unexpectedly superior solvent for the extraction of TMP. Although we shall comment on that affidavit further in considering the remaining claims, we note that with respect to the claims rejected on the ground of double patenting, appellant has not argued or even asserted in his brief that the claimed process is unobvious in view of that claimed in the Robeson patent. Indeed, at oral argument, counsel for appellant expressly conceded obviousness of the process herein over that already claimed in the earlier patent.

Appellant takes the position that, notwithstanding the conceded obviousness of the process now being claimed from that already patented, the two processes are distinct and the execution of his terminal disclaimer, of record, obviates all objections to issuance of a second patent claiming subject matter distinct from, but obvious in view of, the claims of the first patent.

Whether a terminal disclaimer can overcome the objections to double patenting has been the subject of much discussion. The second paragraph of 35 U.S.C. § 253,3 providing for a terminal disclaimer, was first enacted in 1952 and has no known antecedent in the earlier patent statutes. The legislative history and the Reviser's Notes shed little light on exactly why Congress enacted that particular provision.4

Where the claims of a second application are substantially the same as those of the first patent, they are barred under 35 U.S.C. § 101. In re Ockert, 245 F.2d 467, 44 CCPA 1024. Miller v. Eagle Mfg. Co., 151 U.S. 186, 14 S. Ct. 310, 38 L. Ed. 121 (1894).

Where, as here, the claimed subject matter is an obvious modification of what has already been claimed, a second patent is contrary to one of the fundamental principles underlying the patent system, namely, that when the right to exclude granted by a patent expires at the end of the patent term, the public shall be free to use the invention5 as well as obvious modifications thereof or obvious improvements thereon.6 Thus, to grant a second patent for an obvious variation deprives the public of those rights.7 If, however, the second patent expires simultaneously with the first, the right to fully utilize the patented discovery at the expiration date remains unimpaired. Thus the terminal disclaimer here precludes any extension of monopoly.

We do not find, as the Government contends, In re Siu, 222 F.2d 267, 42 CCPA 864, to be contrary.

"that the manipulative steps of Ladisch, including the characteristics of the gas stream and its relationship to the stream of molten liquid, substantially correspond to those of the Siu application. The difference between the two groups of claims is that appellant's claims are directed toward the specific use of the method with molten glass, while those of the patent to Ladisch are directed toward the method generically, and toward its use with resinous materials specifically." The court observed that "* * * the claims in issue [Siu's] represent a species of Ladisch's invention." Speaking of 35 U.S.C. § 253 and the provision for a terminal disclaimer, this court said

"However, while the disclaimer has the effect of permitting less close scrutiny of the distinctions between claims issuing to an applicant in separate patents, it was not, and could not have been the legislative intent to permit indiscriminate issuance of numerous patents directed to mere colorable variations of the same idea. Nothing in the statute or its legislative history suggests abandonment of the settled rule of Underwood v. Gerber, 149 U.S. 224, 13 S. Ct. 854, 37 L. Ed. 710, forbidding more than one patent for what is obviously only one invention, whether or not the grants expire on the same day." (Emphasis supplied.)

Thus Siu continues the basic rule of only one patent for one invention, and where claims are directed to "mere colorable variations of the same idea" they are directed to the identical invention.8 It seems to us, however, that here the production of TMP, although an obvious variation of the invention defined in the claims of Robeson, is more than a "mere colorable variation" thereof.

As noted in Siu, extension of monopoly is not the only objection to double patenting.9 Others include possible harassment by multiple assignees, inconvenience to the Patent Office, and the possibility that one might avoid the effect of file wrapper estoppel by filing a second application.10 We do not minimize those possibilities, but we must decide this case on the facts before us. We are not here confronted by a situation where any abuse of the terminal disclaimer is suggested. We conclude that on the facts here, the only real objection to granting appellant's application is an extension of the monopoly. The terminal disclaimer, which Congress has expressly provided, removes any danger of such result, thus we are obliged to reverse the rejection of claims 7, 8, 9, 13 to 15 and 30.

Claims 2, 6, 10, 11, 16, 21, 22, 24 and 2911 have been rejected as being obvious over the Gottesman or De Lorenzo patents taken in view of the Poitras et al. patent. The aforesaid rejected claims relate to the extraction of TMP from the reacted aqueous mixture with ethyl acetate. Gottesman discloses the production of TME by the analogous condensation of formaldehyde with propionaldehyde in the presence of sodium hydroxide. The reacted mixture is then evaporated to dryness and TME is extracted with ethyl acetate from the reacted mixture containing the by-product sodium formate. De Lorenzo teaches the same chemical reaction as Gottesman and discloses the continuous extraction of TME with ethyl acetate. Poitras et al. is relied on to show that TME and TMP have similar solubility characteristics. The same type of condensation reaction, namely, an aliphatic aldehyde, formaldehyde, and sodium hydroxide are reacted to produce a mixture of polymethylolalkanes. The patent discloses that an alkanol will preferentially extract either TME or TMP from the reacted mixture containing sodium formate as a by-product.

The examiner held that it would be obvious to run the Gottesman or De Lorenzo extraction processes with a homologous compound, that is, TMP for TME, particularly in view of their similar solubility characteristics as shown by Poitras. The board agreed.

Appellant contends that De Lorenzo leads away from the use of ethyl acetate as an extractant for TMP because it states that ethyl acetate is a poor solvent for TME. We note that the whole sentence referred to states:

"* * * While ethyl acetate is a poor solvent for trimethylolethane, it has been found that ethyl acetate is a poorer solvent for the impurities and particularly for those complex impurities which lower the melting point of the trimethylolethane so that with ethyl acetate a high degree of purification can be effected at a high rate of recovery."

De Lorenzo also states that TME may be "recovered in a relatively pure condition" and "at a relatively high yield" by extraction with ethyl acetate. The extraction is a liquid-liquid one similar to that being claimed.

Appellant asserts that his affidavit establishes the unexpected superiority of extraction with ethyl acetate in lieu of alkanols. That argument ignores the basis of the rejection. The instant claims are not rejected on the ground that it would be obvious to substitute ethyl acetate for an alkanol but that it would be obvious to substitute TMP for TME in an analogous reference process which employs ethyl acetate as the extractant. Thus, as to the obviousness of the extraction steps, it matters little that ethyl acetate is superior to alcohols as extractants. In our view the substitution of TMP for TME in the extraction process of De Lorenzo would be obvious to one of ordinary skill in the art.

There remains for our consideration the rejection of claims 25, 26, 27, 28 and 31, drawn to the washing of extracted TMP with water.

The rejection is based on the disclosure of Elgin, Walker et al. or Gottesman et al. in view of the Albert patent or over the Albert patent in view of the Walker patent.

Elgin discloses the extraction of polyols with a water-immiscible solvent to separate them from inorganic salts, and re-extraction of the polyol with water.

Walker et al. discloses the extraction of dimethylolpropane with a water-immiscible solvent and subsequent distillation of dimethylolpropane from the solvent.

Gottesman et al. extract TME with a water-immiscible solvent to separate it from sodium formate. The TME is re-extracted from the solvent with water from which it is precipitated in the form of crystals. The crystals may then be washed with water.

Albert relates to the preferential solubility of TMP in higher aliphatic alcohols. It teaches the separation of TMP from sodium formate by the addition of the alcohol and subsequent dehydration. Since the salt is not soluble in the alcohol it precipitates. The TMP may then be distilled.

Although none of the references discloses water washing, the references do teach the separation and recovery of TMP from sodium formate by solvent extraction and distillation.

Appellant contends that since TMP is infinitely soluble in water, a chemist would be led away from using it to wash away remaining impurities in the ethyl acetate extract. However, as the record indicates, solvent extraction is an old technique and its basis is the preferential solubility of a solute in one solvent over others. The fact that TMP is extracted from an aqueous medium (40-60% water) by ethyl acetate indicates that as between ethyl acetate and water, TMP is preferentially soluble in the former. Therefore, a chemist would not be led away from using water to wash away the residual salts.

"* * * such step of washing the organic phase is deemed within the skill of the art since obviously the sodium formate is more soluble in water than in the non-aqueous solvent and if it is desired that additional sodium formate be removed from the non-aqueous extract, then additional water would effect such a removal. * * *"

We find no basis in this record for holding that statement in error.

The rejection of claims 7 to 9, 13 to 15 and 30 is reversed.

The rejection of claims 2, 6, 10, 11, 16, 21, 22, 24 to 29 and 31 is affirmed.

Modified.