T 0022/82 (Bis-epoxy ethers) of 22.6.1982

Summary of Facts and Submissions

I. European Patent Application No. 79 102 431.8, filed on 13 July 1979 and published on 5 March 1980 under publication No. 0 008 358, claiming the priority of the prior application in the Federal Republic of Germany of 19 July 1978, was refused by decision of the Examining Division of the European Patent Office dated 19 June 1981. The decision was based on the original Claims 1 to 4.

II. The stated grounds for the refusal were that the compounds and processes claimed contained no unexpected properties or advantages and were obvious to the person skilled in the art. It was to be expected that the compounds according to Claim 1, based on the starting products known from DE-C-2 105 014 (hereafter referred to as document 4), could have been prepared by the process described in Houben-Weyl, Methoden der Organischen Chemie, 4. Auflage, Band VI, Teil 3, 1965, Thieme-Verlag, Stuttgart, page 413 (hereafter document 1). In such a reaction, as embodied in Claim 2, there was no reason to fear disturbance by simultaneously occurring polymerisation reactions, since according to DE-A-1 768 649 (hereafter document 3) and document 4 similar reactions with vicinal dicarbonyl compounds had shown no problem in this respect. The process according to Claim 3 and likewise to be judged obvious, as the direct cyclization of the diepoxides was by no means unexpected in view of Houben-Weyl, loc. cit, page 197 (hereafter simply referred to as document 2) and document 3. Claim 4 merely contained a combination of the process steps according to Claims 2 and 3 which, in view of the problem to be solved, viz. the preparation of dihydro furans, was obvious.

III. On 19 August 1981 the applicant lodged an appeal against the decision dated 19 June 1981 and on 20 October 1981 submitted a Statement of Grounds, the substance of which was as follows: The nature of a substance's availability can only exceptionally affect the question of whether a substance is to be judged inventive. In the present case the claimed diepoxides, given the close adjoining of the functions compared with the known monofunctional compounds, are already fundamentally original because of their property of potentially enriching chemistry. The bifunctional diepoxides are not simply a modification of the monofunctional compounds, but provide fundamentally different possibilities for synthesis. It is, furthermore, extremely remarkable that the preparation of the claimed diepoxides according to Claim 2 succeeded at all, for according to document 4 dichloro diketones react in aqueous alkali to give compounds having the energetically-favoured five-membered ring structure. Claims 3 and 4 satisfy the patentability criteria by the mere fact that new chemical ground is broken with the diepoxides without which these processes are impossible, and as a result a novel method for preparing a desired end product is made available. Any improved effects such as yield, purity, etc., are therefore no longer important. Furthermore, the Examining Division has studiously ignored the fact that the yield from the preparation of the 2,5 dimethyl-dihydro furanone compounds according to the combination in Examples 2 and 9 was 45%, considerably better than the 37% according to document 4. Additionally, when citing document 2, it has failed to take account of the fact that there the two radicals R and methoxyl are located on the same carbon atom of the oxirane, which is not the case in the present process. The appellant requested that the decision refusing the application be set aside and that a patent be granted on the basis of the Claims 1 to 4 with the amendments requested in the submissions dated 11 December 1980 and 20 October 1981. The presently proposed claims are as follows: "1 Bis-epoxy-diylkoxy alkanes of the general formula I

(Formula I)

where R1 to R3 are alkyls having 1 to 4 C atoms.

2. Process for the preparation of the novel bis-epoxy-dialkoxy alkanes of the general formula I according to Claim 1, characterised in that alpha, alpha'-dihalogen-1,2-diketones of the general formula II

(Formula II)

where R1 and R2 have the meanings indicated in Claim 1 and X is C1 or Br, are reacted with an alkali hydroxide or an alkali alkoxide of the formula R3-O-Me, where Me is K, Na or Li, and an alkanol of the formula R3-OH, at temperatures of from - 20° to + 100° C, preferably from O° to 25° C.

3. Use of the novel bis-epoxy-dialkoxy alkanes of the general fomula I according to Claim 1 for the preparation of flavour substances of the general formula III

(Formula III)

where R1 and R2 have the meanings indicated in Claim 1, by heating with strong mineral acids or strong organic acids at from 60° to 120°C.

4. Process for the preparation of flavour substances of the general formula III

(Formula III) where R1 and R2 are alkyl of 1 to 4 C atoms, characterised in that

(a) alpha, alpha'-dihalogen-1,2-diketones of the general formula II (Formula II) where R1 and R2 have the meanings indicated in Claim 1 and X is C1 or Br, are reacted with an alkali hydroxide or an alkali alkoxide of the formula R3-O-Me, where Me is K, Na or Li, and an alkanol of the formula R3-OH, at from - 20° to + 100°C, and

(b) the novel bis-epoxy-dialkoxy alkanes obtained of the general formula

(Formula I)

where R1 to R3 are alkyls having 1 to 4 C atoms, are heated with strong mineral acids or strong organic acids at from 60° to 120°C."

Reasons for the Decision

1. The appeal complies with Article 106 to 108 and Rules 64 and 78(3) EPC and is, therefore, admissible.

2. With the exception of the partial deletion of the indications of preferred embodiments, the present claims are the same as in the original version, in which the obvious errors of transcription in formula II of Claims 2 and 4 have been corrected in accordance with Rule 88 EPC (cf. in this connection Examples 1 to 6, 8 and 11 to 14).

3. As is clear from the preamble to the description, the object of the invention has been to find novel methods and novel starting compounds by means of which the desired flavour substance 2,5-dimethyl-4-hydroxy-2,3 dihydrofuran-3-one (hereinafter abbreviated DHF) can be prepared more economically and technologically more simply than hitherto (cf. the description, page 3, lines 29-33).

4. This problem is solved by making available the relevant diepoxides (better described as bis-epoxy ethers) (Claim 1), which are more closely designated by formulae and to be regarded as so-called intermediates, a complete two-stage process for the preparation of the flavour substance DHF (Claim 4), and the two stages of that process (Claims 2 and 3).

5. According to the finding of the Examining Division, which is not to be faulted, such a technical teaching is novel. However, in assessing inventive step the Examining Division directed its attention only at individual facets of the invention without seeing the intimate connection - both technical and in patent terms - between intermediate and end product.

6. The Board is on the other hand unable to concur with the appellant's view that an intermediate is patentable merely because it potentially enriches chemistry. In citing the differences between monofunctional compounds on the one hand and bifunctional compounds with close interaction of the two functions on the other, the appellant's approach is to concentrate on the structural originality of the new compounds vis-à-vis those comprised in the state of the art. In so doing he overlooks the fact that structural difference in a compound has no intrinsic value or significance for the assessment of inventive step as long as it does not manifest itself in a valuable property in the widest sense, an effect or an increase in the potency of an effect.

7. In the present case, in assessing the patentability of the new intermediates according to Claim 1 leading to DHF, the Board's first consideration is the following: Where, because of an existing need, an applicant sets himself the task of developing an economically and technologically advantageous new complete chemical process for the preparation of known and desired end products, the solution to such a problem may be inventive if - despite the choice of the same starting materials as for the nearest comparable process in the state of the art - the advantageous result actually achieved is surprising (in this case quantitatively). If that result (in this case higher overall yield) is achieved via new intermediates made available in the course of the complete process, the "process effect" will also support the inventive step for the intermediates themselves, without which the advantageous complete process is not conceivable.

8. These conditions are here satisfied. In document 4, in addition to the proposed solution proper as disclosed therein, five other methods for preparing DHF, each having disadvantages, are described. Moreover, in DE-A-2 359 891, cited in the European search report, a multi-stage process for preparing the same compound is described, the last stage of which yields only 30% DHF (Example 13). Furthermore, the preamble to the description of the present application describes three other processes for preparing DHF, all having disadvantages (cf. preamble page 2, paragraphs 3 and 4, and page 3, paragraph 3). The existence of at least 10 known processes for preparing DHF, some covered by patent applications, demonstrates the major interest shown in this compound by those skilled in the art and the efforts to discover an economical way of preparing it. The need for an improved process of preparation is clearly evident.

9. In this connection, the process according to document 4 must be considered to be the nearest state of the art, as its starting and end products are the same as those in the process according to Claim 4 of the present application. The only difference is the method, which in the known alkali cyclization process comprises a single stage but in the process claimed two stages, the first involving an epoxidation reaction in the presence of alkali and alcohol to form the bis-epoxy ethers as intermediates and the second the latter's cleavage and re-cyclization to form the dihydrofuran ring. Looking at the optimum DHF yield in the two processes, this is 37.6% in the case of the known process cited, according to the Examining Division's calculations, which are not disputed, compared with 45% for the claimed process (Example 2: bis-epoxy ether preparation 100%; Example 9: DHF preparation 45%). The overall yield rate from a multi-stage process normally diminishes as the number of steps in the process increases. Consequently, the above-mentioned result obtained in accordance with the invention, whereby the yield from the known single-stage process for preparing DHF, starting from dihalogen diketone, can be significantly increased by means of the new two-stage method using bis-epoxy ethers as intermediates, could not be foreseen by the person skilled in the art. Such a person who, as here, had set himself the task of improving on the yield of the known process of preparing DHF using the relevant dihalogen diketone and who knew of the processes according to documents 1, 2 and 3, which do not admit of any forecast as to yield, would not therefore have arrived at the solution claimed unless there had been an inventive step. This leads to the patentability of the complete process according to Claim 4, insofar as it relates to the preparation of DHF, as well as of the intermediates according to Claim 1 in this way made available for the first time, without which the advantageous complete process is inconceivable in the sense of following from the teaching of causality.

10. However, an "effect" such as that shown in point 9 cannot be inferred for the intermediates according to Claim 1, where R1 and R2 are alkyl radicals having 1 to 4 carbon atoms but not simultaneously methyl, and which therefore are suited only for the preparation of the higher homologues of DHF, because, firstly, the yields in the two-stage process according to the application are lower (Examples 8 + 10: 27: 4%; Examples 11 + 15: 33.6%: Examples 16 + 12:31.2%; Examples 13 + 17: 16.2%; Examples 18 + 14: 33.6%) and, secondly, document 4 merely discloses a preparation process for DHF without explaining how the higher homologues of DHF are prepared, so that a comparison is not possible. On the other hand, DE-A-2 359 891, mentioned above, names the group of organoleptically effective 4-hydroxy-dihydrofuran-3-ones, which are substituted in 2,5 position by straight or branched alkyl groups having 1-4 carbon atoms (cf. Claim 1 in conjunction with page 4, formula III and the last paragraph) and whose preparation is also claimed in the application (Claims 3 and 4). It furthermore indicates their three-stage preparation (acylacetic acid ester condensation, thermal decomposition of the tertiary butyl ester, oxidation with hydrogen peroxide, cf. page 3, lines 2-16 and page 5, penultimate line, to page 6, line 3). However, an exact calculation of the yield is not possible for the higher DHF homologues, because general yield data are not given for the first reaction stage (although for stages 2 and 3 they are: yield 100% and 20-40% respectively). It is likewise not possible to calculate the yield for the complete process from Examples 14-16.

11. Nevertheless, the novel intermediates referred to under point 10 are also patentable, since their being made available opens the way to a new process for preparing known and desired end products, itself involving an inventive step. The Board is here guided by the principle that the surprising properties which derive from a chemical substance and manifest themselves in the form of a particular effectiveness in a technical field of application are to be rated no higher when assessing that substances's inventive step, than those revealed when it undergoes further chemical processing (equivalence of the traditional product effect and the process effect). If, in this connection, the state of the art nearest to the desired 4-hydroxy-dihydrofuran-3-ones is considered, it is noticeable that apart from isolated suggestions relating to the oxidation of dihydrofuran-3-ones is considered, it is noticeable that apart from isolated suggestions relating to the oxidation of dihydrofuran-3-one-4-carboxylic acids (DE-A-2 359 891) and the cyclization of 2,5-dihalogen-hexane-3,4-dione (document 4) or vic- diketo diester (cf. description according to the application, page 3, paragraph 3), the preparation method used more frequently by those skilled in the art is that involving cyclization of hexane-2,5-diol-3,4,dione (cf. the preamble to the description of the present application, page 2, paragraphs 3 and 4, and page 3, paragraph 2). It might therefore have been regarded as obvious for the applicant likewise to have selected this compound as an intermediate.

12. However, the applicant did not use this route but instead the one involving bis-epoxy ethers. Contrary to the view held by the Examining Division, this stage of the process has to be accorded inventive content, since it could not have been suggested by documents 2 or 3. Thus document 2 describes the addition of water, catalysed by traces of acid, to epoxy ethers giving alpha-hydroxy aldehydes or alpha-hydroxy ketones according to the equation while document 3 teaches the likewise acid-catalysed cyclization of hexane-4,5-diol-3,4-dione to DHF. A person skilled in the art, possessing average capabilities, who has set himself the task of searching for a new route to preparing known and desired end products will not for that purpose mentally include those substances whose intermediate formation - here diol-diones - is neither intended nor desired, but is purely hypothetical. Furthermore, the structure of the epoxy ethers known from document 2 (R and methohyl radical on the same carbon atom of the oxirane ring; unsubstituted methylene group in this ring) is so different to those in the present Claim 1 (R1 and OR3 not on the same carbon atom of the oxirane ring; substituted methylene group in this ring; duplication of all these functions) that for this reason as well this document was not suitable as a model for the individual stage here being considered.

13. The lengthy nine-year period elapsing in the much researched area of the flavour substances here under consideration (see under point 8) between the time when hexane-4,5-diol-3,4-dione was first proposed as a suitable intermediate for these flavour substances (in this connection see the two French patent documents mentioned in document 4, column 2, paragraphs 2 and 3, and made available to the public in September 1969) and the time when the present application was filed can be seen as a further indication that document 2 (dating from 1965), in conjunction with the knowledge of the acid-catalysed cyclization of the above-mentioned diol-dione (cf- the reference document cited in document 4), was unable to provide any suggestion for the separate stage as in the application, co-ordinate Claim 3, using the intermediates according to Claim 1.

14. The Examining Division needed no less than three documents (documents 1, 2 and 3) to deny inventive step for the process according to co-ordinate Claim 4. In so doing it split the complete process up into two separate reaction stages, judged each to be obvious and concluded that the combination of the two stages was likewise obvious. This approach does not hold up in review, for it overlooks the fact that when a multi-stage process is assessed for inventive step it must be considered integrally from the perspective of the task set, i.e. the discovery of a new method for preparing the known and desired flavour substances. The determinant question here is not whether it was obvious with knowledge of the starting materials to connect the individual stages of the complete process sequentially to lead to the end product, but whether this process presented itself as obvious when looking back step by step from the target substance to the starting materials.

15. Applying this criterion to an assessment of the process according to Claim 4, the final stage in the new complete two-stage process must be rated as inventive, as already explained in points 11-13. As the mere recognition of the importance of the bis-epoxy ethers as intermediates for the synthesis of the known and desired flavour substances called for an inventive step, the choice of starting materials and of the reaction method for these intermediates, i.e. the first stage of the complete process, can no longer be regarded as obvious either, even if the process in question is not chemically original. Nevertheless, the arguments put forward by the Examining Division for denying inventive step for this first stage, in the light of document 1, should be discussed. This standard chemical work teaches that some alpha-haloketones react with alcoholates in accordance with the following equation: (Formula) Results of original experiments are shown in this context, based without exception on mixed aromatic-aliphatic ketones; in four cases the desired result was obtained and in one case the reaction was unsuccessful. The document cited is hardly likely to have induced a person skilled in the art, in the normal course of his work, to apply this very restricted process to the preparation of purely aliphatic bis-epoxy ethers.

16. Claim 2 relates to the preparation of the intermediates according to Claim 1, which are recognised as patentable. It is not necessary to go into the question of whether the preparation process itself has inventive step, so long as the product of this process is recognised as having inventive step. There appears to be no justification at present for denying the patentability of this process claim. So long as a chemical product is protected by a preceding claim, the process for its preparation by chemical means represents a characterisation of this product in the patent category relating to the process of preparation (process for product). Such an additional claim for a product recognised as patentable may be of importance in opposition or revocation proceedings in the light of Article 123(3) EPC or corresponding national regulations, or in infringement proceedings in the light of Article 69(1), first sentence, EPC.

17. No application has been made for reimbursement of the appeal fee in accordance with Rule 67 EPC, and the facts of the case would not justify such a measure.

ORDER

For these reasons, it is decided that:

1. The decision of the Examining Division of the European Patent Office dated 19 June 1981 is set aside.

2. The case is remitted to the first instance with the order to grant a European patent on the basis of the original documents, account being taken of the amendments requested on 11 December 1980 and 20 October 1981.