T 0657/02 21-12-2005

1. The appeal complies with Articles 106 and 108 EPC and Rule 64 EPC and is therefore admissible.

2. Late-filed documents D6b and D9

2.1 The appellant requested that documents D6b and D9, filed after expiry of the time limit set by Article 99(1) EPC and not considered by the opposition division , be admitted into the appeal proceedings for consideration.

2.2 Since, however D6b (publication date 1998) is not prior art within the meaning of Article 54 EPC with respect to the patent in suit (priority date 27 April 1995) and the appellant was not in a position to show why D9 was more relevant than the documents already in the proceedings, in particular more relevant than D8, the board decided not to admit D6b or D9 into the proceedings for consideration (Article 114(2) EPC).

3. Amendments (main request)

According to the decision under appeal, the amendment in Claim 1 of the main request ("having at least three monosaccharide units") meets the requirements of Article 123 EPC (point IV (a), above). The board sees no reason to challenge this finding. Nor was any objection in this respect raised by the appellant.

4. Sufficiency of disclosure (main request)

4.1 Claim 6 of the main request is directed to a process for producing a fructan derivative according to any one of Claims 1-5, wherein fructan is reacted with a nitrile, an amide or an ester, and the addition product is optionally hydrolysed, reduced, alkylated and/or oxidised (points I and III, above).

4.2 According to the appellant, Claim 6 does not define all the necessary steps to arrive at all variants claimed in Claim 1. In particular, it is not possible to obtain with the process steps listed in Claim 6 the amidoxime derivative referred to in Claim 1 (ie when R**(3) represents -C(=NX)-NR**(4)R5).

4.2.1 However, the process of Claim 6 is not limited to the process steps listed therein. The wording "… wherein fructan is reacted with …" in Claim 6 merely introduces the essential step(s) in terms of the invention but does not exclude further reaction steps. In other words, the term "wherein" is not to be interpreted in the sense of "consisting of" but rather in the sense of "comprising" where it is not in contradiction to have further reaction steps.

4.2.2 Furthermore, paragraph [0019] in the patent specification teaches that the cyanoethyl fructan (ie the reaction product of fructan with a nitrile of the formula R**(1H)C=CR**(2)-CN) can be converted to an amidoxime by reaction with hydroxylamine. Thus, the patent specification clearly teaches how to obtain the amidoxime variant referred to in Claim 1.

4.2.3 Even if one interpreted the term "wherein" in more narrow way, ie that the process consists of the reaction steps listed in Claim 6 (as suggested by the appellant), the missing reference to the possible reaction with hydroxylamine would merely be an incompatibility between Claim 6 and Claim 1. Thus, the fact that Claim 6 is not quite in line with Claim 1 could at most give rise to an objection under Article 84 EPC (lack of clarity). However, Article 102(3) EPC does not allow objections to be based upon Article 84 EPC, if such objections do not arise out of the amendments made in the course of the opposition or opposition appeal proceedings, respectively (eg T 301/87 (OJ EPO 1990, 335, point 3.8 of the reasons) or T 381/02 of 26 August 2004 (not published in the OJ EPO, points 2.3.2 to 2.3.5 of the reasons)).

4.3 In view of the above, the appellant's objection under Article 83 EPC must fail.

5. Novelty (main request)

The appellant did not challenge the finding of the opposition division that the claimed subject-matter was novel with respect to the cited prior art. Nor does the board see any reason to raise an objection in this respect.

6. Problem and solution

6.1 Claim 1 of the main request is directed in general terms to fructan derivatives which contain three-carbon side-chains having functional groups on the side chain, eg carboxyethyl groups. The fructan derivatives have useful properties as surfactants, emulsifiers and as agents inhibiting the crystal growth of calcium salts, and have low viscosity (page 2, lines 10 to 12 of the patent specification).

6.2 D11 reviews methods for introducing carboxyl groups into sucrose and inulin, the latter being the preferred fructan in the patent in suit. These carbohydrate based polycarboxylates have a wide range of potential applications, eg as sequestering agents for Ca/Mg in detergent formulations, as dispersing agents or as metal ion carrier (last paragraph of point 1). In particular, D11 describes the carboxymethylation of inulin (point 4.1) and gives results of the crystallization inhibition properties of carboxymethyl inulin (CMI) whereby it is stated that CMI does not have a great impact on the viscosity of the solution (last paragraph of point 4.3). Since CMI is structurally closely related to the claimed subject-matter (instead of the three-carbon side-chains CMI contains two-carbon side-chains), and D11 discloses technical effects, purpose and intended use most similar to the claimed subject-matter, D11 is considered to represent the closest prior art.

6.3 The respondent has shown by comparative tests disclosed in A2a (excerpts from the dissertation of one of the inventors of the patent in suit) that carboxyethyl inulin (CEI) according to Claim 1 has better calcium carbonate precipitation inhibition properties than the closest prior art compound, ie CMI. As can be seen from Table 5 of A2a, CEI has a much longer induction period, tind, than a CMI of comparable molecular weight. tind is, according to the respondent, the key parameter for measuring the precipitation inhibition ability of CMI and CEI, respectively.

6.3.1 Although the CEI in Table 5 of A2a has a degree of substitution (ds) of 0.65 and the CMI has a ds of 1.05, these tests represent a valuable comparison between CEI and CMI. As explained by the respondent, a higher ds is associated with a better performance in precipitation inhibition. In this context, the respondent referred to Table 2 in A2b (other parts of the dissertation of one of the inventors of the patent in suit) which confirms this relationship between ds and performance in precipitation inhibition. Thus, a CEI with a ds of 1.05 would show, under the same conditions, an even better performance in precipitation inhibition than the CEI of Table 5 of D2a (ds = 0.65).

6.3.2 The appellant questioned the reliability of the experiments taken from Table 5 in A2a in view of a discordance with experiments elsewhere in the same dissertation, ie Table 2 in A2b. Thus, Table 5 reports different data on tind and the apparent equilibrium concentration Ce than Table 2 for apparently one and the same CMI.

However, as explained by the author of the dissertation (declaration dated 18 February 2003), the data contained in Table 5 of the dissertation are reliable for the purpose of internal comparison, since these results follow from experiments that were carried out on the same day using the same reagents - apart from the particular inhibitor - under the same ambient conditions. On the other hand, results obtained in different tables are not automatically comparable, since they follow from experiments carried out on different days, and the qualities of the equipment and reagents and the ambient conditions were not necessarily exactly identical. Since, however, the outcome of the experiments depends on a variety of conditions (such as temperature, quality of water roughness of the glass surface and of the surface of the magnetic stirrer), the outcome of the experiments performed on different occasions may be different. The results described in Table 2 are obtained from an experiment done on another day than the experiments described in Table 5 and for the reasons given above need not be fully comparable with each other.

6.3.3 With the declaration on file and in the absence of any evidence from the appellant that the results in Table 5 are wrong, the board has no choice but to accept these results as a valid comparison which establish the superiority of the claimed subject-matter over the closest prior art with respect to precipitation inhibition properties.

6.4 Therefore, the objective technical problem to be solved by the patent in suit has to be seen in the provision of fructan derivatives having improved calcium carbonate precipitation inhibition properties.

In view of the comparative tests in Table 5 of A2a, the board is satisfied that the above identified objective technical problem is solved by the features identified in Claim 1 of the main request.

7. Inventive step (main request)

7.1 It remains to be decided whether the proposed solution, ie the introduction of a three-carbon side chain into a fructan, is obvious from the available prior art.

7.2 In D11 itself there is no suggestion as to how the calcium carbonate precipitation inhibition property of CMI could be further improved, let a lone a hint to the introduction of a three-carbon side-chain.

7.3 However, also a combination of the closest prior art D11 (or D10) with other documents, in particular with D2 and/or D8, does not make the claimed subject-matter obvious for the following reasons:

7.3.1 D2 teaches the carboxyalkylation of polysaccharides. Although among various polysaccharides, inulin is mentioned (eg Claim 4), and although among carboxy alkylations, carboxyethylation and cyanoethylation are mentioned (eg Claim 3), the only real teaching of D2 is on carboxymethylation of starch, in particular starch derivatives of a certain chain length. There is neither a generic disclosure in D2 about the cyanoethylation or carboxyethylation of inulin, as suggested by the appellant, nor a teaching or suggestion of any useful property of any inulin derivative, let alone of a carboxyethylated or cyanoethylated inulin. The mere fact that D2 discloses the theoretical possibility of a carboxyethylation and cyanoethylation of inulin is not sufficient, as alleged by the appellant, that it was obvious to modify the CMI of the closest prior art in this way. Moreover, the skilled person, trying to solve the objective technical problem, would have no motivation to apply this theoretical teaching of D2 to the CMI of D11, because D2 provides no hint that carboxyethylation and cyanoethylation would improve the calcium carbonate precipitation inhibition properties of the compounds. Consequently, a combination of D11 with D2 would be based on hindsight.

7.3.2 As regards the other documents referred to by the appellant, namely D8 and D10, they cannot make the claimed invention obvious, either.

D8 describes the cyanoethylation of starch. As mentioned under "3 Results and Discussion", the objective of D8 was to optimise the reaction conditions and to identify the chemical structures of the reaction products. The objective of D8 was not to obtain any industrially useful product, nor does D8 give any results allowing concluding that cyanoethyl starch or a further derivative thereof could have interesting properties. The respondent has never disputed that the cyanoethylation on polysaccharides was known. However, D8 does not provide any hint to the skilled person, trying to solve the objective technical problem, that the cyanoethylation would improve the calcium carbonate precipitation inhibition properties of the products. Thus, the skilled person had no reason to combine D8 with D2.

The teaching of D10 is largely the same and not more pertinent than that of D11. The objective appears to be the same, although no experimental results on the properties of the CMI are mentioned yet. It contains a statement of dicarboxy inulin being a better complexing agent that dicarboxy starch, but this is associated with the particular structure resulting from 2,3-oxidation, which is not relevant in the patent in suit.

7.3.3 Consequently, the claimed subject-matter is not obvious from the documents relied upon by the appellant.

7.4 In a second line of argumentation, the appellant considered known polysaccharide derivatives, such as carboxymethyl cellulose, carboxyethyl cellulose and carboxyethyl starch, which are mentioned in paragraph [0002] of the patent specification, as the closest prior art. Since D10 and D11 disclosed useful properties of carboxymethyl substituted inulin, and, furthermore, it would be common knowledge that cellulose and starch are less soluble than inulin and that this difference in solubility persisted in the corresponding derivatives, the appellant's conclusion was that the substitution of starch or cellulose for inulin to form certain derivatives would be obvious.

However, this line of argument is flawed from the very start because there is no common knowledge that certain physical properties such as solubility of certain polysaccharides persist into the respective derivatives. For example, it is known that starch, a poorly soluble and often viscous polysaccharide, on substitution with carboxymethyl groups becomes more soluble and less viscous in solution, whereas dextran, a readily soluble polysaccharide, increases in viscosity and thus becomes less soluble after substitution with carboxymethyl groups. Moreover, Example 1 of the patent in suit shows that derivatisation of inulin with cyanoethyl groups results in decreasing solubility (up to insolubility), and that thus the solubility of inulin does not persist in these derivatives. Hence, there is no incentive based on common knowledge to substitute cellulose and starch by inulin when developing useful derivatives.

7.5 In summary, the solution to the stated problem does not arise in an obvious manner from the state of the art. Consequently, the subject-matter of Claim 1 of the main request, and by the same token, the subject-matter of Claims 2-11 involves an inventive step.

8. Because the respondent succeeded on the main request, there was no need to consider its auxiliary requests.