T 1958/22 (Etex/Gypsum) of 2.4.2025

Summary of Facts and Submissions

I. The appeal of the opponent is directed against the decision of the opposition division to reject the opposition against European patent No. 3 312 339, claim 1 thereof reading:

"1. A gypsum board comprising a gypsum core with at least one side covered by a fibrous mat comprising at least one ply of a non-woven fabric and a binder composition, wherein:

- said binder composition represents from 10 to 40 wt% of the total weight of the mat; and

- said binder composition comprises a copolymer comprising a co-monomer unit of a vinyl ester of an alpha branched aliphatic monocarboxylic acid, said copolymer being present in an amount from 25 to 100 wt% of the binder composition weight."

II. In its statement of grounds of appeal, the appellant requested that the decision under appeal be set aside and the patent be revoked in its entirety, arguing that the claims as granted were not inventive over D10 (EP 2230075 A1) combined with D24 (EP 1980540 A1); D17 (WO 2010/026065 A1) combined with common general knowledge or with D3 (WO 94/12549), D6 (EP 0731207 A1), D7 (US 5,763,022), D8 (US 6,174,568 B1), D10 or D34 (WO 2013/113459); D18 (US 2006/0068186 A1) combined with common general knowledge (as shown in D1) and/or with D7, D8, D12 (WO 00/22016), D17 or D24; or starting from D19 or D23 (US 2004/0209074 A1). It also requested that the case be remitted to the opposition division if further auxiliary requests were to be discussed.

III. In its reply, the patent proprietor (respondent) requested that the appeal be dismissed and the decision to maintain the patent as granted be upheld, arguing that the grounds of appeal merely reiterated the arguments brought forward during the first-instance proceedings, which should not be admitted under Article 12(3) RPBA. As an auxiliary request, it sought to maintain the patent in an amended form based on the claims of one of the auxiliary requests 1 to 3 submitted with the reply. It also requested that the case be remitted to the first instance if the board intended to reject any of these requests. Additionally, it requested that the reports D28 and D29 (filed and not admitted during first instance proceedings) be admitted into the appeal proceedings if assessing inventive step required evidence of the effect of the binder amount in the mat.

IV. In a submission dated 14 September 2023, the appellant argued that auxiliary requests 1 and 2 did not overcome the inventive step objections, auxiliary request 2 extended beyond the content of the application as filed and auxiliary request 3 was late filed and should not be admitted into the appeal proceedings.

V. In its preliminary opinion, the board concluded that none of the requests on file appeared to meet the requirement of inventive step when starting from D17 as the closest prior art.

VI. In response to this opinion, the patent proprietor filed auxiliary requests 1 to 5 with a submission dated 31 October 2024, with auxiliary requests 1 and 2 being new and auxiliary requests 3 to 5 corresponding to the previously submitted auxiliary requests 1 to 3.

Claim 1 of auxiliary request 3 corresponds to that of the main request with the following additional feature:

" ... wherein said comonomer unit of a vinyl ester of alpha branched aliphatic monocarboxylic acid is present in said copolymer in an amount of 20 to 70 wt.% ... ".

Claim 1 of auxiliary request 4 corresponds to that of the main request with the following additional feature:

" ... wherein said comonomer unit of a vinyl ester of alpha branched aliphatic monocarboxylic acid is present in said copolymer in an amount of 40 to 70 wt.% ... ".

Claim 1 of auxiliary request 5 corresponds to that of the main request with the following amendment (highlighted by the Board): "... - said binder composition represents from [deleted: 10] 15 to [deleted: 40] 35 wt% of the total weight of the mat ... ".

VII. In a submission dated 15 January 2025, the opponent requested that the newly filed auxiliary requests 1 and 2 not be admitted into the appeal proceedings under Article 13(2) RPBA.

VIII. At the oral proceedings, which took place on 2 April 2025, the parties confirmed that the present decision should be based on the following requests:

The appellant requested that the decision of the opposition division be set aside and the patent be revoked in its entirety.

The respondent requested that the appeal be dismissed and that the decision to maintain the patent as granted be confirmed (main request) or, in the alternative, that the patent be maintained in amended form on the basis of one of auxiliary requests 1 to 5, as filed with letter dated 31 October 2024.

Reasons for the Decision

1. Admittance of opponent's objections under Article 12(3) RPBA

1.1 The patent proprietor argued that opponent's appeal failed to meet the requirements of Article 12(3) RPBA, and requested that its objections and arguments which merely reiterated those raised at first instance - rather than directly addressing the reasoning of the decision under appeal - be disregarded.

1.2 The board is not persuaded that any of said arguments or objections should be disregarded solely on the grounds that they were allegedly repeated from the first-instance proceedings. While an appeal may be inadmissible if it relies exclusively on the same case presented at first instance, this does not preclude a party from referring to arguments previously made. On the contrary, some degree of repetition is often inevitable, as the purpose of the appeal is typically to put forward, and eventually further develop the original arguments in order to challenge the contested decision. Accordingly, reiterating arguments from the first-instance proceedings is not only permissible, but often necessary to present a complete case pursuant to Article 12(3) RPBA.

1.3 The board therefore concludes that the arguments and objections presented by the opponent in the statement of grounds of appeal meet the requirements of Article 12(3) RPBA and are not disregarded.

2. Main request - Inventive step

The opposition ground under Article 100(a) EPC in combination with Article 56 EPC prejudices the maintenance of the patent as granted for the following reasons:

2.1 Closest prior art

2.1.1 The appellant cited documents D10, D17, D18, D19 and D23 as possible starting points for the inventive step argumentation. The present decision will focus on the objections starting from D17 as the closest prior art, as this is considered as the most promising springboard to arrive at the claimed subject-matter.

2.1.2 As a matter of fact, D17 discloses the use of binders in architectural elements, including gypsum boards (see page 1, lines 7-9), and specifically addresses the issue of weather resistance, such as resistance to mould and moisture. The document proposes a construction in which a gypsum board is sandwiched between fibreglass scrims or facers, with the top scrim being coated using specific formulations (see page 1, lines 17-21). More particularly, D17 describes (see page 12, lines 17-21) the use of a copolymer dispersion formulated into a coating that is applied to the gypsum board. In certain embodiments, the fibreglass scrim is coated with this copolymer-based formulation and then placed onto a wet gypsum board to form a glass-gypsum composite.

The proposed binder (see page 3, line 21 - page 4, line 8) includes a copolymer with a co-monomer derived from a vinyl ester of neodecanoic acid such as VeoVa 10®, preferably in an amount of 30-50 phm (parts per hundred monomers). Moreover, according to two specific embodiments (see formulations 8 and 9 on page 16, lines 5-7), the binder includes a copolymer with 40 wt.% of the co-monomer from VeoVa 10®, which falls within the scope of the copolymer defined in claim 1 at issue.

2.1.3 The opposition division concluded that D17 could not be considered as the closest prior art, because it did not address the problem of improving the bonding of the facer to the gypsum core, but was mostly concerned with the use of the gypsum board in wet or humid areas. This document was therefore more remote from the subject-matter of claim 1 than D10, so that the latter should be taken as the closest prior art.

2.1.4 The patent proprietor argued that D17 made no reference to the bonding properties of the coating composition, and that the proposed formulations were externally applied in the form of a coating and not impregnated as in the opposed patent. Moreover, there was no indication in D17 that the coating composition was in contact with the gypsum board, and the reference to the 'binding' function was actually intended to indicate that one of the functions of the copolymer was to retain the other components of the formulation. There was thus no reason to conclude that the copolymers in D17 played any role in binding the fibres of the mat and/or the mat to the gypsum board. In this respect, since the described attachment of the scrims to the board was carried out by bringing the scrims onto the wet gypsum board, it was clear that the binding resulted from an embedding of the scrims into the gypsum material and not from the addition of the coating formulation. The skilled person would therefore not realistically regard document D17 as the closest prior art.

2.1.5 The board disagrees therewith for the following reasons:

According to established case law of the Boards of Appeal, a prior art disclosure should only be discarded as a starting point for the inventive step argumentation where it is so far removed from the technical context of the invention that a skilled person would only consider it with the benefit of hindsight.

This is clearly not the case for D17, which does not only belong to the same technical field as the alleged invention, but also discloses the majority of its features and pursues closely related technical objectives. Specifically, the main purpose of D17 is to provide a coating or binding composition that enhances the weather resistance of fibreglass-gypsum composite boards intended for use in construction panels. This objective is nearly identical to that of the opposed patent which, as stated in paragraphs [0004] and [0005], aims to provide gypsum board composites for constructions panels with an improved mechanical resistance while maintaining desirable properties such as weather and moisture resistance. Thus, the main distinction between the patent and D17 lies not in the underlying purpose but only in the emphasis, since D17 primarily focuses on weather resistance, whereas the patent places greater emphasis on the mechanical strength of the composite.

In this respect, the reference in D17 to the dual function of the formulation for 'coating or binding' is interpreted as a clear and explicit indication that the copolymer dispersions described therein serve both as a coating for enhancing weather/moisture resistance and as a binder for increasing the mechanical stability of the composite. This dual function is explicitly supported by the passage on page 12, lines 3-4 of D17, that states that 'The emulsion polymer dispersion can be used as a binder in any coating system, particularly where water resistance is beneficial'. It follows that the board does not share the proprietor's interpretation that the function of 'binding' would only concern the retention of other components, such as pigments, within the coating composition. Such an interpretation would not only conflict with a straightforward and technically reasonable reading of D17, but would also be inconsistent with the fact that the copolymer is the sole essential component of the formulations disclosed therein (see page 3, lines 4-7 and claim 1), which would raise the question as to why the copolymer would be specifically included for the purpose of retaining components that are entirely optional.

Furthermore, the board considers the passage on page 12, lines 17-21 of D17, which describes the application of the copolymer when the scrims are affixed to the gypsum board to form the composite, as further evidence that the coating or binding formulations play a substantive role in securing the scrims to the gypsum core. The fact that the scrims are applied to wet gypsum does not undermine this conclusion; rather, it indicates that the scrims are embedded within the gypsum matrix, without implying that such an embedding is the sole means of attachment. In this context, given that the coating or binding formulation is applied to both the gypsum board and the scrim, it is clear that it contributes to the overall bonding within the fibreglass-gypsum composite.

Finally, the board does not agree with the assertion that D17 would be more remote than D10 in terms of shared technical features with the claimed invention. In any case, this argument is irrelevant when assessing whether D17 can be considered as a suitable starting point. The distinguishing features relative to D17 are namely different from those relative to D10, and as such, any conclusion of non-obviousness based on D10 cannot be automatically extrapolated to a problem-solution analysis starting from D17. Consequently, the only way to conclude that the invention meets the requirement of inventive step would involve providing an independent assessment starting from each of these documents.

There is thus no reason to disregard D17 as the starting point for the inventive step argumentation.

2.1.6 Concerning the differentiating features with respect to D17, the appellant argued that for a person skilled in the art of construction materials it would be clear that the concept of 'scrim' in D17 encompassed non-woven materials. Moreover, within the specific scope of this document, the skilled person would readily understand that the fibreglass scrims were non-woven, so this feature could not be seen as a differentiating feature.

2.1.7 The board disagrees, as the terms in a claim should be construed using the broadest possible interpretation within the relevant technical context. In other words, even where a specific alternative is regarded as more preferred or normal within the technical field, this knowledge cannot be used to narrow down the interpretation of a clearly broader concept. In this respect, it is clear that the concept of 'fibreglass scrim' is not restricted to woven or non-woven glass fibres, but encompasses both alternatives. Document D17 does therefore not anticipate the use of non-woven scrims.

2.1.8 The proprietor argued that according to table 1 of D17, the copolymer only amounted to 7 wt% of the binder composition, so that the subject-matter of claim 1 also differed from D17 in that the copolymer was present in an amount of 25 to 100 wt% of the binder composition.

2.1.9 The board disagrees that the range of 25 to 100 wt% would represent a further distinguishing feature over D17 because the formulations disclosed in Table 1 of D17 do not necessarily correspond to a 'binder composition' as defined in claim 1 at issue, since both the patent as a whole and claim 1 in particular describe this concept in rather vague terms. Specifically, claim 1 only requires the presence of a 'binder composition' which comprises the inventive copolymer in an amount of 25 to 100 wt%. Since the copolymer present in the dispersion disclosed in Table 1 of D17 falls within the scope of claim 1 and there is no other ingredient, it can be concluded that this dispersion falls as such within the scope of a 'binder composition' as defined in claim 1, i.e. in this dispersion the copolymer will be in an amount of around 100 wt% or slightly lower if other secondary components (e.g. dispersants) are present. In any case, it is directly and unambiguously clear that the amount of copolymer in that dispersion will fall within the broad range of 25 to 100 wt%.

The same conclusion applies even when the term 'binder composition' in claim 1 is interpreted in the context of the patent specification. In particular, it is apparent in view of paragraphs [0177]-[0180] and Table 1 that the 25 to 100 wt% range defined in claim 1 refers to the concentration of the copolymer according to the invention within a dispersion that may also include other copolymers or ingredients. Therefore, the purpose of the 25 to 100 wt% requirement is simply to ensure that the dispersion either consists solely of the claimed copolymer or, if other copolymers are present, that the allegedly inventive copolymer constitutes at least 25 wt% of the copolymer mixture. This condition is clearly met in D17, as the only essential copolymer present in the dispersion (or the only copolymer used in the composition of Table 1) falls within the scope of claim 1 at issue. Therefore, also from this perspective, it is the copolymer dispersion disclosed in Table 1 of D17, rather than the whole coating composition, which corresponds to the binder compositions described in Table 1 of the patent. From this perspective, the coating composition as a whole in D17 - including the copolymer dispersion, dispersant, pigment, and rheology modifier - is analogous to the slurries described in Table 2 of the patent.

2.1.10 In light of the foregoing considerations, the board concludes that D17 constitutes a suitable starting point, and that the subject-matter of claim 1 differs therefrom in that (i) the fibreglass scrim is non-woven, and (ii) the binder is present in an amount of 10 to 40 wt%.

2.2 Problem solved by the invention

2.2.1 According to the examples in the patent - particularly the results shown in Tables 5 and 7 - the application of a binding composition in accordance with the invention (tests F7 to F11) to gypsum boards results in enhanced mechanical properties under both dry and humid conditions, as well as improved water resistance, when compared to composites (F1 to F6) using other binding compositions, in particular formulations containing fluorocarbons as water-repellent (see par. [0211]). Notably, the binder concentration in both the inventive and the comparative examples is maintained within a narrow range of 22 to 25 wt%.

2.2.2 To further support the argument that the amount of binder of 10 to 40 wt% in claim 1 at issue had not been arbitrarily selected but was aimed at achieving a specific technical effect, the patent proprietor submitted documents D28 and D29, which show that adding an amount of approximately 50 wt% of the binder composition (i.e. outside the claimed range) results in poorer peeling forces when compared to the addition of around 30 wt% of binder (within the claimed range).

2.2.3 According to the proprietor, these results demonstrated that if too much or too little binder was used the adhesion or the hydrophobicity of the mat would be negatively affected. Consequently, the range of 10 to 40 wt% of binder composition was linked to an improved adhesion of the non-woven scrim to the gypsum board. The problem solved by the invention was therefore to improve the mechanical properties of the gypsum composite while maintaining other desirables properties.

2.2.4 The Board notes that, although the tests presented in the patent are primarily aimed at demonstrating the effects of binder compositions known from D17, they are nonetheless considered sufficient to show that the defined binder range of 10 to 40 wt% is not arbitrarily selected, but rather represents suitable amounts to achieve the desired properties - namely, mechanical strength under dry and humid conditions, as well as weather resistance. Accordingly, if the data in documents D28 and D29 are intended to support this effect, they are considered redundant and unnecessary.

On the other hand, the board has concluded that D28 and D29 cannot be relied upon to substantiate the argument that the binder range of 10 to 40 wt% would achieve an unexpected or surprising improvement when compared to gypsum composites with binder amounts falling outside the claimed range. Firstly, it cannot be derived that such an effect is encompassed by the teachings of the original application, as required by decision G 2/21 (point II of the Headnote), given that in the original documents the range of 10 to 40 wt% is only presented as a suitable option - i.e. as an appropriate balanced amount to achieve good adhesion and weather resistance properties. Furthermore, there are no comparative examples to substantiate such a surprising improvement, since even if D28 and D29 were taken into account, these documents do not include comparative tests involving binder concentrations below 10 wt%, which is particularly relevant considering that the closest prior art D17 discloses binder composition levels within or slightly below the lower end of the claimed range (see point 2.3.2 below). The board therefore maintains, as stated in its preliminary opinion, that the content of D28 and D29 is not critical for determining the problem solved by the invention. Consequently, there is no need to decide on the admittance of these documents into the appeal proceedings.

2.2.5 In view of the above considerations, the board sees no basis to conclude that the gypsum composites according to the invention would necessarily achieve improved mechanical strength and/or weather resistance when compared to those in D17. However, the board does also not agree that the defined range of 10 to 40 wt% simply provides an arbitrary alternative.

2.2.6 All in all, the board concludes that the problem solved by the alleged invention is to provide an alternative gypsum board composite which is mechanically stable in dry and in humid conditions while maintaining other desirable properties such as weather resistance.

2.3 Obviousness of the solution

2.3.1 The patent proprietor argued that none of the cited documents rendered the solution obvious, because they were not related to gypsum boards, did not clearly anticipate the proposed solution and/or did not teach that the proposed solution would lead to an improved bonding between a non-woven scrim and the gypsum board.

According to the definitions in dictionaries and Wikipedia, the concept 'scrim' was typically associated with woven materials. Although D26 suggested that scrims could also be non-woven, the drawings actually indicated that the concept of 'non-woven' in this document was associated with highly oriented fibres, i.e. intersecting at regular 90° angles. There was thus no incentive to choose a non-woven scrim when starting from D17 as the closest prior art.

The cited prior art documents would also not lead the skilled person to select an amount of 10 to 40 wt% of the binder composition. The coating or binding compositions in the exemplary embodiments of D17 (see Table 2 on page 16) were added in amounts ranging from 74 to 90 wt% of the total weight of the mat, which was significantly above the range of 10 to 40 wt% defined in claim 1 at issue.

The amount of binder was also not rendered obvious by any one of D3, D6, D7, D8, D10 or any other cited prior art document. In particular, the tensile strengths achieved in D3 (see Table 1 on page 17) were significantly lower than those achieved with the claimed subject-matter, so this document did not provide a clear incentive to take into account its teachings; D10 referred to a latex (i.e. an emulsion), and there was no information as to how much copolymer was used; D6 did not even mention gypsum boards, so the skilled person would not take its teachings into account; and D7 and D8 did also not relate to gypsum boards, nor did they discuss any adhesion properties.

2.3.2 The board has concluded that the proposed solution lacks an inventive step in light of the teachings of D17 combined with common general knowledge, because first of all, the term 'scrim' encompasses both woven and non-woven fibre mats, and it is undisputed that the use of both types of mats in gypsum boards is well known in the field of construction (as acknowledged by the patent itself (see par. [0002])). Furthermore, there is no evidence on file demonstrating that the use of a non-woven structure would produce any unexpected or surprising technical effect in the context of the claimed invention. From this standpoint alone, the choice of one known alternative over the other would not involve an inventive step, as a skilled person starting from D17 would necessarily have to choose between a woven and a non-woven scrim, and making such selection based on the well-known properties of each option - i.e. rather than for achieving an unexpected effect - constitutes an obvious choice among known alternatives.

Additionally, the board maintains its view that D26 reinforces the lack of inventiveness of this feature, as it indicates a trend toward using non-woven scrims over woven ones, also in construction-related applications such as 'roofing' and 'air ducts'. The proprietor's argument that the non-woven scrims in D26 comprise highly oriented fibres is unpersuasive, because regardless of the schematic illustrations therein, the text of D26 explicitly states that 'non-woven processes can place yarns at various angles and can lay down multiple layers of yarns with various orientations', which explicitly highlights the well-known fact that the fibres in non-woven scrims can be randomly oriented.

The board also finds no inventive contribution in the provision of the binder composition in an amount of 10 to 40 wt%. As discussed above, the 74 to 90 wt% range derivable from Table 2 in D17 concerns the coating compositions (equivalent to the slurries in Table 2 of the patent) and not the binder composition (corresponding to the copolymer dispersion). Since the copolymer dispersion represents approximately 7 wt% of such coating composition, in D17 the actual amount of the inventive copolymer (i.e. the binder composition) in the mat is from about 5 to 6 wt% (i.e. 7 wt% of 74 to 90 wt%). Even though this would still fall slightly below the 10 to 40 wt% range recited in claim 1, neither the patent nor the subsequently filed experimental data in D28 or D29 demonstrate any technical effect associated with higher binder concentrations. Since, as discussed above, the functions of the copolymer dispersion in D17 are the same as in the alleged invention (namely to provide mechanical strength and weather resistance), the board has concluded that arriving at the defined range would be a matter of obvious trial and error optimisation. The selection of amounts of binder falling within the range of 10 to 40 wt% is therefore obvious in view of D17 combined with common general knowledge.

For completeness, it is further noted that it is not even clear whether the claimed subject-matter implies that the amount of inventive copolymer in claim 1 is greater than in D17. Claim 1 defines the binder composition as comprising 10 to 40 wt% of the total mat weight, but permits the inventive copolymer to constitute as little as 25 wt% of that binder composition. This means that in some embodiments falling within the scope of the claim, the amount of inventive copolymer can be as low as 2.5 wt% of the total weight of the mat (i.e. 25 wt% of 10 wt%), which is actually lower than the 5 to 6 wt% copolymer by the total weight of the mat disclosed in D17. The only difference in such cases would be the inclusion of higher amounts of other, unspecified components in the binder composition, such as other copolymers, dispersants, and/or fillers (which may account for up to 75 wt% of the binder composition). Since there is no indication as to the effects of including higher amounts of other (undetermined) components, the claimed range of 10 to 40 wt%, also from this perspective, does not involve an inventive contribution over D17.

2.4 The subject-matter of claim 1 is therefore obvious in view of D17 combined with common general knowledge, so that the request does not meet the requirements of Article 56 EPC.

3. Auxiliary requests 1 and 2 - Admittance

3.1 These requests were filed after notification of the communication under Article 15(1) RPBA, so their admittance is governed by Article 13(2) RPBA. The requests should therefore not be admitted unless there are exceptional circumstances justified by cogent reasons.

3.2 The proprietor argued that it had been surprised by the conclusion that D17 should be regarded as the closest prior art rather than D10. This had led to a significant change in the subject-matter of the proceedings.

3.3 The board does not consider that formulating the inventive step argumentation on the basis of a different closest prior art amounts to an exceptional circumstance - particularly in light of the fact that D17 was already proposed as a possible starting point for the inventive step analysis in the notice of opposition. The board has therefore reached a different conclusion from that of the decision under appeal, specifically by determining that D17 can be regarded as a suitable starting point for the assessment of inventive step. While this divergence has naturally led to a shift in the focus of the proceedings, it does not constitute an unforeseen development that would qualify as an exceptional circumstance within the meaning of the RPBA, as parties must always be prepared for the possibility that the board may depart from the reasoning in the decision under appeal.

3.4 In the absence of exceptional circumstances that could justify the filing of amended claims at this late stage of the proceedings, the board decided that auxiliary requests 1 and 2 were not admitted under Article 13(2) RPBA.

4. No reason to remit the case

4.1 Both parties requested to remit the case if any one of the auxiliary requests needed to be discussed (appellant) or was considered not to be allowable (respondent).

4.2 The board sees however no reason to remit the case to the first instance for discussing the auxiliary requests, since the inventive step argumentation against the alleged invention in the claims at issue is based on the same evidence and arguments as for the main request. Moreover, the board notes that the need to discuss requests which were not previously addressed by the first instance is rarely considered to be a special reason which could justify the remittal of the case under Article 11 RPBA.

5. Auxiliary requests 3 to 5 - Inventive step

5.1 Claim 1 of auxiliary request 3 corresponds to that of the main request with the following additional feature:

" ... wherein said comonomer unit of a vinyl ester of alpha branched aliphatic monocarboxylic acid is present in said copolymer in an amount of 20 to 70 wt.% ... ".

5.2 Document D17 discloses (see page 4, lines 5 to 8) that the amount of vinyl ester monomer in the copolymer is 0 to 100 phm. The most preferred range is however 30 to 50 phm and in some exemplary embodiments (see formulations 7 and 8 on page 16, lines 6-7), the vinyl ester monomer is present in an amount of 40 wt.%. The additional range defined in claim 1 at issue is therefore anticipated in D17, so the arguments and conclusions presented for the main request also apply to this request. The requirements of Article 56 EPC are thus not met.

5.3 Claim 1 of auxiliary request 4 corresponds to that of the main request with the following additional feature:

" ... wherein said comonomer unit of a vinyl ester of alpha branched aliphatic monocarboxylic acid is present in said copolymer in an amount of 40 to 70 wt.% ... ".

5.4 In view of the arguments presented in point 5.2 above, this request does also not meet the requirements of Article 56 EPC in view of D17 combined with common general knowledge.

5.5 Claim 1 of auxiliary request 5 corresponds to that of the main request with the following amendment (highlighted by the board): "... - said binder composition represents from [deleted: 10] 15 to [deleted: 40] 35 wt% of the total weight of the mat ...".

5.6 Since the narrower range for the amount of binder has not been associated with any special technical effect, the same arguments and conclusions presented for the main request apply to this request, which does therefore not meet the requirements of inventive step.

6. Since none of the requests submitted by the patent proprietor meets the requirements of the EPC, the patent shall be revoked.

Order

For these reasons it is decided that:

1. The decision under appeal is set aside.

2. The patent is revoked.