|European Case Law Identifier:||ECLI:EP:BA:1986:T001785.19860606|
|Date of decision:||06 June 1986|
|Case number:||T 0017/85|
|IPC class:||C09C 1/02|
|Language of proceedings:||DE|
|Download and more information:||
|Title of application:||-|
|Headnote:||If the preferred numerical range in a citation in part anticipates a range claimed in an application, the said claimed range cannot be regarded as novel at least in cases where the values in the examples given in the citation lie just outside the claimed range and teach the skilled person that it is possible to use the whole of this range.|
|Relevant legal provisions:||
Selection from a numerical range
Limitation of a numerical range
Amendment of claim
Summary of Facts and Submissions
I. European patent application No. 80 106 420.5 (publication No. 27 997) filed on 21 October 1980 with a German priority of 29 October 1979 was refused by the Examining Division in a decision taken on 6 July 1984. This decision was based on two claims, the first of which read as follows: "Mineral filler, particularly calcium carbonate, characterised by the ratio % by weight of particles < 1µm = R = 4 to 10." % by weight of particles < 0.2 µm An alternative request concerned a further set of claims differing from that filed with the main request in the following disclaimer at the end of Claim 1: "except for 9.4 and 9.7".
II. The decision laid down that the subject-matter of Claim 1 under the main request was not new in view of the products C, D and E with (calculable) R values of 9.7, 9.4 and 10 respectively, which were mentioned in citation FR-A- 2 418 263; but although the subject-matter of Claim 1 under the alternative request was new in so far as a selection was involved, it had nonetheless been suggested by the citation. The same applied to the preferred R value according to Claim 2 under the main and alternative requests.
III. The appellants filed an appeal against this decision on 13 August 1984, paying the prescribed fee at the same time, and submitted grounds along the following lines on 9 November 1984. The citation said nothing whatever about an R value, nor would a skilled person be prompted without a knowledge of the invention to formulate an R value. The citation did not disclose the dimensioning rule according to the invention, but only a number of values which accorded "fortuitously" with that rule. The subject- matter of the application was therefore new. It was not obvious, moreover, since although the citation referred to certain adverse effects of an excessively high content of particles <0.2 µm it did not suggest the teaching that this content be reduced in favour of the proportion ranging from 0.2 to 1 µm; the skilled person, after all, was aware that the process of grinding to obtain products with a low proportion of particles <0.2 µm automatically also produces lower proportions of particles <1 µm, i.e. R values outside the claimed ratio. In this connection the appellants also drew attention to results of tests which had already been submitted to the department of first instance and which purportedly confirmed the unexpectedly advantageous properties of the claimed products.
IV. Following two communications from the Board expressing doubts about the novelty of the claims under the original and alternative requests, which were refused by the department of first instance, and about the limited claims filed subsequently, the appellants continue to defend the present limited claims under the main request; an alternative request is for the granting of a patent further limited by the inclusion of another feature in Claim 1.
V. The claims under the main request are worded as follows: "1. Mineral filler, particularly natural calcium carbonate, for paints and paper coating compositions, characterised in that the ratio is % by weight of particles < 1µm = R = 4 - 8. % by weight of particles < 0.2 µm
2. Filler according to Claim 1 characterised in that R is approximately 5." Under the alternative request, Claim 1 is to be worded as follows (with Claim 2 unchanged and an obvious typing error corrected): "Mineral filler, particularly natural calcium carbonate, for paints and paper coating compositions, characterised in that the ratio is % by weight of particles < 1 µm = R = 4 - 8 % by weight of particles < 0.2 µm and that at least 93% by weight of the particles are smaller than 2 µm." Finally, the appellants have filed for the first time a separate version of the claims for Austria in which Claim 1 is worded as follows: ....
VI. The appellants also argue as follows in support of the main request: (i) In connection with the view they had already expressed previously to the effect that the citation did not disclose the dimensioning rule according to the invention but only a number of values which accorded "fortuitously" with it, they refer to two German decisions: BGH (Federal Court of Justice) "Etikettiermaschine", GRUR 1981, 812, and BPatG (Federal Patent Court) 13W (pat) 87/82 of 20 February 1984. (ii) The decision "Thiochloroformates/HÖCHST" (OJ 7/1985, p. 209), which was cited by the Board, did not apply to the subject-matter of the application since the latter, unlike the case referred to, was not concerned with a selection invention. A selection had to relate to the previously known teaching itself and not to the result thereof. (iii) Finally, the parallel German application had also led to the granting of a patent taking account of the same state of the art (DE-A- 2 808 425 cited therein corresponded to FR-A- 2 418 263 cited in the present case).
Reasons for the Decision
1.The appeal complies with Articles 106 to 108 and Rule 64 EPC and is therefore admissible.
2.First of all, the admissibility of the newly filed claims must be examined.
2.1 The new Claim 1 under the main and alternative requests is limited to a range of 4 to 8 for the claimed R value. The original documents specified a preferred range of 4 to 10 (e.g. the original Claim 2); and the mention of 5.0, 4.53 and 5.33 as examples of values for natural calcium carbonates (page 4, lines 34-36) - which play a prominent role in the whole description as preferred examples of mineral fillers without any distinction being drawn vis-à-vis other substances with regard to the sizes of the particles (see, for instance, page 1, lines 14-15; page 2, lines 20-27; page 3, lines 28-30) - is followed by the statement that "very much higher values ... , for example in excess of 8", are possible (page 5, lines 1-2). Both the minimum value for the originally disclosed preferred range beginning with R = 4 and also the maximum value ("8") were therefore already included in the original documents, the words "also .... over 8" presupposing a sub-range ending at 8. There are no formal reasons why these two values lying within the originally disclosed preferred range could not be combined in order to avoid a clash with the state of the art (cf. Decision T 02/81, "Methylene-bis(phenyl isocyanate)/ MOBAY, OJ 10/1982, 394 et seq.; and T 53/82, "Furnace soots", dated 28 January 1982, paragraph joining pages 4 and 5, unpublished).
2.2 The filler in the new Claim 1 under the main and alternative requests has also been further specified by the addition of the words "for paints and paper coating compounds". The original documents contain numerous references to such a use for the claimed fillers; in particular, "paper coating compositions and paints" are mentioned expressly on page 9, lines 16 - 17. The addition referred to is therefore unexceptionable.
2.3 The new Claim 1 under the alternative request also contains the feature that "at least 93% by weight of the particles are smaller than 0.2 µm". (As may be inferred from the corresponding document DE-C- 2 943 652 this is intended to mean "smaller than 2 µm". For one thing, an R value >4 rules out by definition a larger than 25% proportion of particles <0.2 µm; also, this conclusion is indicated by Tables I and III in the Description.) This is formally unexceptionable since this feature is embodied in all the specifically described products according to the invention (page 5, Table I, products 3 and 4; page 8, Table III, products 3 - 5) with percentages ranging from 93 to 99.5% and was thus originally disclosed.
2.4 Nor can any objection of a formal nature be made to the new Claim 2 under the main and alternative requests since an R value "near to 5" has already been disclosed as preferred on page 5, line 11, of the original documents.
2.5 Leaving aside the separate set of claims for Austria, the new text of the claims under the main and alternative requests is thus formally admissible. ....
3.Both the search report and the department of first instance referred to the citation FR-A- 2 418 263 as the only relevant state of the art. The citation takes as its starting point a state of the art in which it appeared desirable, particularly with a view to the required gloss for paper coating compositions, that the largest possible proportion of the particles of the filler should have the smallest possible diameter (page 1, line 32, to page 2, line 31; in particular, page 2 lines 8 - 11). But it also teaches that excessively small particles of filler are undesirable when the aim is further to improve the gloss of paper coating compositions or paints (page 2, lines 36 - 40). Specifically, it proposes fillers whose particles are as far as possible no smaller than 0.2 µm (page 3, lines 6 - 8). From the practical point of view a maximum of 15% of such "extremely fine" particles is regarded as admissible.
4.By contrast the problem of the invention is to optimise the gloss substantially without the disadvantages of excessive viscosity (page 3, line 5 in conjunction with lines 10 - 11 of the original documents) or uneconomical manufacturing processes (page 5, line
5.To solve this problem the invention indicates in Claim 1 under the main request the dimensioning rule according to which R lies within the range 4 to 8.
6.Tables II and IV on pages 6 and 8 respectively of the original documents provide convincing evidence that an increase in the R value improves the gloss (comparison of products 3 and 4 and 3 - 5 with products 1 and 2). Moreover test results filed on 8 March 1983 demonstrate that increasing the R value above 8 produces only a modest further improvement in the gloss but at the same time has the disadvantage of increasing viscosity (comparison of products VP 3 and VP 4, last two lines of the table of results). It is also clear that where R is appreciably below 4 there is also not only a sharp reduction in the gloss but also a substantial increase in viscosity (product VP 1, line 1 of the table). It is therefore possible to accept that the problem is indeed solved by the claimed dimensioning rule.
7.With regard to the novelty of Claim 1 under the main request, the only citation is unquestionably concerned with mineral fillers (particularly natural calcium carbonate) for paints and paper coating compositions. Accordingly the sole matter of interest is whether the skilled person was also able to derive from the same document an R value (as defined in the claim in the present application) between 4 and 8.
7.1 The citation - whose objective is also to improve the gloss - teaches both that the proportion of particles with a diameter <0.2 µm should be kept as low as possible (page 3, lines 3 - 5) and also that the proportion of particles with a diameter <1 µm should be maximised (page 4, lines 33 - 36). Taking both together the only interpretation at which the skilled person could arrive was that a maximum proportion should come within the range between 0.2 and 1 µm or in other words that what mattered was the ratio symbolised in the application by the letter R. The fact that neither this symbol itself nor the mathematical form of the quotient are used in the citation is irrelevant. What matters is the teaching that these two particle size ranges are given a mathematical relation, however that relation may be formulated.
7.2 The preferred ranges for the specified particle sizes are precisely defined in Claims 2 and 5 of the citation (3 - 12% <0.2 µm, 80 - 95% < 1 µm). If the extreme values of these two preferred ranges are set in relation to one another mathematically, the result is a ratio of 80 to 95 = 6.67 to 31.6 123 which corresponds in substance to the R value. The preferred range thus disclosed extends into the claimed range of R = 4 - 8 and accordingly takes away its novelty.
7.3 This disclosure is supplemented by specific embodiments. Products B, C, D and E adduced as examples in the citation with calculable R values of 13, 9.7 9.4 and 10, with a good gloss and containing less than 15% of particles <0.2 µm are contrasted with products A, F and G with more than 15% of particles <0.2 µm, calculable R values of 3.1, 3.6 and 2.6 and a gloss which, although unsatisfactory, is adequate for practical purposes. Even if all these values lie outside the range claimed in the present application, they still do not limit the preferred range in the citation's broader Claims 2 and 5; on the contrary, they demonstrate to the skilled person in a representative way how the gloss of the filler compound can be varied by adjusting the size of the particles and that the preferred range can even be abandoned by forfeiting a degree of gloss. As indicated by the above list of R values, the examples within and outside the preferred range - particularly those with R values of 3.6 and 9.4 - are so close to one another as to give the skilled person a direct pointer to the range R = 4 - 8 claimed in the present case. Accordingly, the skilled person was also bound to infer from the disclosure in the citation the range lying between the R values of 3.6 and 9.4, of which examples are given. This intermediate range is thus also without novelty.
7.4 Sub-paragraphs 7.2 and 7.3 may be summarised as follows. If the preferred numerical range in a citation in part anticipates a range claimed in an application, the said claimed range cannot be regarded as novel at least in cases where the values in the examples given in the citation lie just outside the claimed range and teach the skilled person that it is possible to use the whole of this range.
7.5 his conclusion accords with an earlier decision of the Board which recognised the possibility under certain conditions of a selection from a numerical range, namely the decision "Thiochloroformates/HÖCHST" (loc. cit.). This laid down that for a sub-range selected from a broader range of numbers to be patentable the selected sub-range must be narrow and sufficiently far removed from the known range illustrated by means of examples. In that case the Board regarded as a narrow selection the singling out of a catalyst concentration in a sub-range between 0.02 and 0.2 from the known overall range of >0 - <100 mol%. At the same time the range - illustrated by means of examples - was removed from the selected range by at least a power of ten. These conditions do not obtain in the present case. The claimed sub-range of R = 4 - 8 is not a narrow section of the known sub- range of 6.67 - 31.6 nor is it sufficiently far removed from the known range cited as an example, which lies between 2.6 and 13. In fact, it falls in the middle of it. The skilled person does not see these examples as separate, isolated embodiments in connection with the preferred range indicated but as representatives of a range extending at least from the lower particle distribution value (R = 2.6) to the upper (R = 13) and thus covering all the values in between, i.e. including the range claimed here.
7.6 The fact that a patent has been granted in respect of the corresponding national German application does nothing to alter this, particularly as the claim allowed in that case was limited by a further feature corresponding to one in the present alternative request. Accordingly, the fact that a patent was granted does not permit the conclusion that the German Federal Court would have conceded novelty to Claim 1 under the present main request.
7.7 The Board's view is not in any way affected by the German court decisions mentioned by the appellants. The Federal Court of Justice's decision "Etikettiermaschine" (loc. cit.) concerns a matter which is not comparable with that under consideration - a mere selection from a numerical range - since that case, unlike the present one, was concerned with a selection from an "undifferentiated range without specific parameters" (second headnote). The Federal Patent Court's decision 13W (pat) 87/82, of which the appellants filed a copy on 7 April 1986, does in fact concern a selection from a numerical range (in the field of alloys) and the conclusion, which the appellant quotes, is that " . . results achieved merely fortuitously by the state of the art and not condensed into technical directions . . . cannot call into question the novelty of the teaching according to the application". In the further appeal to the Federal Court of Justice, however, that Court arrived at a different decision (dated 11 July 1985, "Borhaltige Stähle", GRUR 1986, 163), stating in a headnote: "An 'adjustment rule' applying to the manufacture of alloys is not new if alloys with constituents which are the same in respect of quality and quantity belong to the state of the art and the claimed 'adjustment rule' has been applied - even if inadvertently - in their manufacture." The Federal Court of Justice thereby maintained its case law under which a range cannot be re-patented if sub-ranges forming part of it were already known (GRUR 1982, 610, "Langzeitstabilisierung"). The appellants cannot therefore infer anything to their advantage from the Federal Patent Court decision they cite.
8. The facts set out in paragraph 7 show that Claim 1 under the main request is without novelty and its subject-matter therefore unpatentable. Since a decision can only be taken on a request as a whole, neither Claim 2 nor the separate set of claims for Austria in connection with the main request need to be examined. The appeal relating to the main request must therefore be dismissed.
9. The same does not apply to the alternative request relating to the Contracting States DE, FR, BE, NL, SE, GB and IT. The Board is unable to find in FR-A- 2 418 263 the feature incorporated in the Claim 1 in question, according to which "at least 93% by weight of the particles are smaller than 2 µm". This claim and accordingly also the subordinated Claim 2 are therefore new. 10.However, these claims have not yet been submitted to the department of first instance, which has consequently been unable to take a final decision on their patentability. Nor, moreover, was a separate set of claims for Austria examined by the department of first instance, so that there has as yet been no opportunity for that set to be examined as to patentability. To ensure that the appellants do not lose an opportunity to appeal, the Board therefore considers it appropriate to remit the case to the Examining Division for further substantive examination in the light of the claims now filed under the alternative request.
For these reasons, it is decided that:
1. The contested decision is set aside.
2. The case is remitted to the Examining Division for further examination in the light of the claims filed under the alternative request.
3. On all other points the appeal is dismissed.