T 1268/11 () of 20.10.2015

Summary of Facts and Submissions

I. By way of its interlocutory decision, the opposition division found that European Patent No. 1 425 454 as amended according to Auxiliary Request 4 met the requirements of the European Patent Convention (EPC).

II. The patent proprietor filed an appeal against this decision and paid the appeal fee. In its grounds of appeal, the appellant-proprietor requested according to its Main Request in version A, to maintain the patent as granted under consideration of its request to correct the wording in granted claim 7; according to its main request in version B to maintain the patent without any correction, i. e. with the claims as granted; according to its main request in version C to maintain the patent with a set of claims wherein claim 7 is deleted. All requests filed previously before the opposition division were maintained.

III. The opponents I, II, III and IV each filed an appeal and paid the appeal fee. In their grounds of appeal, they requested to set aside the decision of the opposition division and to revoke the patent.

The appellant-opponent OI objected inter alia to the term "smallest diameter" in claim 1 as lacking clarity (Article 84 EPC) as well as to the corresponding feature not being sufficiently disclosed (Article 83 EPC). It objected to the request to amend the description. Additionally, it considered the definition in the patent specification of the "average diameter" such as set out in paragraph [0029] as being somewhat eccentric. Moreover, objections under Article 100(b) EPC concerning the method set out in claim 7 and objections under Article 100(a) EPC with regard to the subject-matter of claims 1 and 7 were raised.

IV. The appellant-opponents OII, OIII and OIV filed grounds of appeal and referred to lack of clarity, to insufficient disclosure of the invention as well as to lack of novelty and lack of inventive step. However, in the course of the proceedings, these appellants withdrew their oppositions.

V. With letter of 27 April 2012, in response to the grounds of appeal of the appellant-opponents, the appellant-proprietor

- maintained the A, B and C versions of the requests and

- submitted auxiliary requests 1 to 4 in three versions (A, B and C),

- submitted auxiliary request 5 in A and B versions,

- indicated that the clarifying amendment regarding the determination of the smallest diameter such as made in Auxiliary Request 1D, which was found by the opposition division to meet the requirements of the EPC, could be made to any of the requests on file, and

- commented on the experiments of the opponents.

With letter of 11 July 2014, the appellant-proprietor commented on the further observations of the appellant-opponents.

VI. In a communication sent on 28 April 2015 and annexed to the summons to oral proceedings, the Board indicated its preliminary view that the requirement of Article 83 EPC appeared not to be met, that Rule 139 EPC was not applicable and referred to objections under Article 84 EPC and Article 123(2) EPC. Additionally, the parties were notified that the discussion during the oral proceedings would be limited to these issues.

VII. Respondent-opponent OV withdrew its opposition with letter of 21 August 2015.

VIII. The appellant/proprietor replied with letter of 18 September 2015 to the communication of the Board and withdrew the "A" and "B" sets of claims. New sets of claim requests were filed, nominated as "C", "D" and "E" sets of requests, each set comprising a main request and auxiliary requests 1 to 7.

IX. Oral proceedings were held before the Board on

20 October 2015.

The appellant-opponent OI requested that the decision under appeal be set aside and that the patent be revoked.

The appellant-proprietor requested that the decision under appeal be set aside and that the patent be maintained in amended form on the basis of the Main Request or Auxiliary Requests 1 to 7 in the versions C and E filed with letter dated 18 September 2015, or on the basis of the Main Request or Auxiliary Requests 1 to 7 in the version G filed during the oral proceedings. An overview regarding Example 1 was submitted during the oral proceedings.

X. Claim 1 of the Main request (C), (corresponding to claim 1 as granted) reads as follows:

"An alumina fiber aggregate comprising alumina short fibers whose average diameter is 4.0 to 10.0 µm and smallest diameter is not less than 3.0 µm."

Claim 1 of the Main request (E) reads as follows:

"An alumina fiber aggregate comprising alumina short fibers whose average diameter is 4.0 to 10.0 µm and smallest diameter is not less than 3.0 µm, wherein the smallest diameter is determined as follows:

(i) an earpickful of fibers are nipped off from an alumina fiber mat by forceps;

(ii) the fibers nipped off in (i) are placed, in such a manner as will not overlap each other as much as possible, on a carbon-made conductive tape pasted to an observation specimen holder of a scanning electron microscope (SEM),

(iii) a platinum-palladium film is deposited to a thickness of 1 to 3 nm on the specimen surface to provide conductivity to the surfaces of the individual fibers of (ii);

(iv) the deposited specimen for analysis is placed in a measuring chamber of a JSM-6320F SEM manufactured by Nippon Electron Co., Ltd., and observed at an acceleration voltage of 15 kV, work distance of 15mm, and magnification range of from x1000 to x3000, and then the observed image is photographed;

(v) the fiber diameter of 100 randomly collected fibers is measured to the unit of 0.1 mm by a slide caliper or a straightedge from the SEM photograph obtained in (iv); and

(vi) the diameter whose frequency distribution is not less than 1% is determined to be the smallest fiber diameter."

Claim 1 of the Main Request (G) differs from claim 1 of the Main Request (E) in step (vi) which reads:

"(vi) wherein the measurement is repeated four times and the diameter whose frequency distribution is not less than 1% on average is determined to be the smallest fiber diameter."

In claim 1 of auxiliary requests 1 of all the versions, the average diameter range is limited to 4.0 to 8.0 µm.

Claim 1 of auxiliary requests 2 of all the versions includes further the feature concerning the composition of the alumina short fibers being Al2O3 : SiO2 = 99:1 to 65:35 by weight.

Claim 1 of auxiliary requests 3 of all the versions includes further the feature concerning the preparation of the alumina short fibers by a spinning method wherein spinning is carried out by a blowing method.

Claim 1 of auxiliary requests 4 of all the versions includes further the feature concerning the length of the fibers being 10 to 500 mm.

Claim 1 of auxiliary requests 5 of all the versions is limited to an average diameter range of 5.0 to 8.0 µm

and includes further the feature concerning the composition of the alumina short fibers being Al2O3 : SiO2 = 99:1 to 65:35 by weight.

Claim 1 of auxiliary requests 6 of all the versions is limited to an average diameter range of 5.0 to 8.0 µm, includes further the feature concerning the composition of the alumina short fibers being Al2O3 : SiO2 = 99:1 to 65:35 by weight and specifies that the preparation of the alumina short fibers by a spinning method wherein spinning is carried out by a blowing method.

Claim 1 of auxiliary requests 7 of all the versions is limited to an average diameter range of 5.0 to 8.0 µm, includes further the feature concerning the length of the fibers being 10 to 500 mm and specifies that the preparation of the alumina short fibers by a spinning method wherein spinning is carried out by a blowing method.

XI. The appellant-proprietor essentially argued as follows:

The main request (C) should be admitted. Claim 1 corresponded to claim 1 as granted. The method disclosed in paragraph [0029] had to be read into the claim. For consistency, paragraph [0029] should be corrected to read "the frequency distribution is not less than 1% on the average" rather than the incorrectly translated "greater than 1%". A signed declaration by a certified translator attested that this wording was present in the original Japanese text in PCT/JP02/07383. When taking this correction into account, the protection conferred by the claim became narrower. Hence, the requirement of Article 123(3) EPC was met. With regard to sufficiency of disclosure, the established principles should be applied such as to interpret the claim with regard to the description and such as to read the claim with a mind willing to understand.

The invention underlay the basic aim to provide an aggregate being essentially free of fibers whose diameter were less than 3 µm. Hence, purity of the aggregate was the concept.

The skilled person understood that the number of fibers in the aggregate having a diameter that is smaller than (and equal to) the reported "smallest fiber diameter" had to be determined. Thus, the skilled person would turn to the description to investigate which method to use for determination. The only method in the description was to be found in paragraph [0029] and led directly to exactly this method as being applicable. Moreover, the fact that this method was the only technically meaningful way to construe the feature was also evidenced by the opponents, which had had no problem in applying it. An ambiguity - if at all present - existed only at the edges of the range. However, the skilled person inherently knew when a distribution was abnormal such that doubt was cast on the reliability of the obtained data. Repetition in case of doubt was standard practice when evaluating data. In such case, the repetition of the measurement two to four times was usually sufficient to remove any doubt concerning the frequency distribution of the fibre diameter. There was no reason to fixate the precise number of repetitions, since the measurement should simply be repeated until the skilled person was satisfied that an accurate value had been obtained. Hence, the issue was the precision and accuracy of the measurement rather than the absence of relevant information. Concerning the basis for the repetition, the skilled person would repeat the whole measurement. Therefore, the patent in suit was sufficiently disclosed.

Main Request (E) should be admitted. Claim 1 included the method steps for determining the smallest diameter. Only trivial features of the determination method were omitted. A repetition in case of doubt applied usually for any determination method and, if necessary, the skilled person would do so anyway, hence this feature was trivial and could be omitted. Also the number of repetitions required was within the knowledge of the skilled person and was unnecessary to define. The omission of the term "on the average" did not make a difference with regard to the determination of the smallest diameter since also the opponents had been capable of defining in their test results an averaged smallest diameter.

Main Request (G) should be admitted. Claim 1 referred to four repetitive measurements for determining the smallest diameter. Hence, it was possible to determine an average of the frequency distribution of the smallest fiber diameter. The requirement to repeat the determination four times clearly gave a reliable and reproducible result for the frequency distribution of the smallest diameter. By including such a requirement, the scope of the claim was limited further. It was clear that five measurements had to be done. It was not necessary to define whether the whole procedure, such as set out in paragraph [0029], had to be repeated or whether only the measurement with regard to the scanning electron microscope had to be repeated since in any case an average could be established.

XII. The arguments of the appellant/opponent may be summarised as follows:

Concerning the Main Request (C), it should not be admitted. Post-grant errors could no longer be corrected in a patent specification. When requiring a correction of a translation at this stage, this amounted to an amendment and had to meet the requirements of Articles 84 EPC, 123(2) EPC and 123(3) EPC. The request was late filed and the provisions of Article 13 Rules of Procedure of the Boards of Appeal (RPBA) had to be considered.

Additionally, the test method was not included in claim 1 and thus any method could be applied. However, without a reliable and reproducible test method, the patent in suit did not enable the skilled person to obtain the claimed fiber aggregate (Article 83 EPC). There was no instruction of how to define the smallest diameter and there was only an obscure determination method in the specification. For obtaining a monopoly such as a patent, the disclosure of all relevant information in a clear and reproducible manner was necessary. No such information was present concerning how to obtain the small diameter fibres nor how to determine the smallest diameter.

Claim 1 of none of the requests submitted with the statement of grounds of appeal included the test method despite just such a claim having been allowed by the opposition division. For this reason alone, the subsequently filed "E" and "G" sets of requests - which respectively included a reference to the test method and the steps of the test method itself - should not be admitted at this late stage (Article 13(1) RPBA).

Concerning the Main request (E), the determination method set out in paragraph [0029] was included in claim 1 but various steps were omitted which led to a violation of the requirement of Article 123(2) EPC:

By omission of the repetition steps, a new teaching was introduced in that a repetition of the steps was no longer claimed, which according to the description was necessary in case of doubt on the value of the lower threshold fiber diameter.

By omission of the term "on the average" in relation to the frequency distribution of the smallest fiber diameter, it was not even clear whether a different definition of the fiber diameter was introduced. There was no reason why an incomplete test scenario should be claimed.

Concerning the Main request (G), step (vi) of claim 1 was not clear (Article 84 EPC). The fact that the measurement should be "repeated four times" and the expression regarding the diameter "whose frequency distribution is not less than 1% on average is determined to be the smallest fiber diameter" lacked clarity. It was not clear whether the repetition of the measurement concerned the whole procedure or only the measurement in the measuring chamber of the SEM. Moreover, it was not clear how to determine the average. Additionally, according to the determination of fiber diameter distribution method in paragraph [0029], particularly step (7), it was unclear when a case of doubt would exist, necessitating a repetition of the measurement 2 to 4 times.

Reasons for the Decision

1. Main Request in the version C - Admittance

1.1 According to Article 13(1) of the Rules of Procedure of the Boards of Appeal (RPBA), it lies within the discretion of the Board to admit any amendment to a party's case after it has filed its grounds of appeal or reply. Although the request was filed in reply to the communication sent as an annex to the summons to oral proceedings, the current claim set differs from the claim set as granted only in that claim 7 - against which the main objections in the opposition proceedings were raised - has been deleted.

1.2 The amendment to delete claim 7 was made as a reaction to the objections and comments made during the opposition proceedings and the written appeal proceedings. Hence, the amendment does not change the arguments put forward with regard to claim 1.

1.3 Current claim 1 always formed the basis of the opposition and appeal proceedings. Thus, the Board exercised its discretion under Article 13(1) RPBA and admitted the request into the proceedings.

2. Main Request in the version C - Sufficiency

2.1 Claim 1 refers to an alumina fiber aggregate comprising alumina short fibers which are characterised by their average diameter and their smallest diameter. No method for establishing the average diameter or the smallest diameter of the fibers is defined in claim 1.

Accordingly, the scope of the claim allows the skilled person to apply any appropriate method. The question to be answered was whether the skilled person would be able to ascertain that he had achieved the aggregate of claim 1.

2.2 Sufficiency of the disclosure must be assessed on the basis of the specification as a whole. There is no doubt that the disclosure is aimed at the skilled person who may use his common general knowledge to supplement the information contained in the specification. Nevertheless, the information contained in the specification must be sufficient to allow the invention to be performed over the whole scope of the claim and without undue burden.

2.3 Accordingly, when trying to identify the claimed product, the skilled person would read the specification. The appellant-proprietor's view that the skilled person would apply the method disclosed in the specification and thus be capable to obtain the claimed product was thus investigated.

2.4 In paragraph [0029] the patent specification discloses a measuring method concerning the determination of the fiber diameter distribution. Paragraph [0026] states that the frequency distribution "means the values determined according to the fiber diameter distribution measuring method described later". This reference thus leads directly to the method disclosed in paragraph [0029]. Additionally, there are two examples disclosed (paragraphs [0030] to [0035]) of providing alumina fiber aggregates. Table 1 (paragraph [0041]) indicates for both examples - by determining the frequency distribution of the fiber diameters of 100 fibers in example 1 and 97 fibres in example 2 - the distribution of the fiber diameters in steps of 0.5 µm and specifies an average diameter and a smallest diameter. The average fiber diameter appears to be calculated in Table 1 on the basis of the actually determined values which are actually, however, not provided since only "bins" in steps of 0.5 µm are provided. The smallest fiber diameter appears to indicate the actually measured smallest fiber diameter. Hence, in the examples and in corresponding Table 1, the average value and the smallest value for the fiber diameter are not established such as set out in the method of paragraph [0029].

2.5 The appellant-proprietor argued that the skilled person would simply follow the method. However, the definition of the average fiber diameter by way of the calculation and the method steps referred to in paragraph [0029] are not clear. Nonetheless, the Board agrees with the position of the appellant-proprietor that the skilled person would read the calculation with its background knowledge and would be capable of calculating an average fiber diameter when following the method steps. Since also the appellant-opponent(s) provided test results indicating an average value without the appellant-proprietor contesting their correctness, it can only be concluded that any average value - independent on its calculation method - is to be considered within the scope of claim 1.

2.6 It remained to be scrutinized whether the determination of the "smallest" (or "minimum") fiber diameter is sufficiently disclosed. When disregarding the information which could be taken from Table 1 that the smallest diameter is the one which actually should be determined, the information given for the method had to be scrutinized.

2.7 The method disclosed in paragraph [0029] defines in step (7) that

"in case where a doubt is produced on the lower threshold fiber diameter in view of the tendency of frequency distribution of fiber diameter, measurement is repeated usually 2 to 4 times and the diameter whose frequency distribution is greater than 1% on the average is reckoned as the minimum fiber diameter."

2.8 Accordingly, it has to be taken into account that the repetition(s) of the measurement is(are) disclosed as being dependent on "doubt" as to whether the tendency of frequency distribution of the fibre diameter is considered correct. No indication of how to define a case of doubt is given. The wording "in case where a doubt is produced" would need an applicable definition to ensure a non-arbitrary method step.

2.9 The suggestion that in case where a doubt is produced, the measurement "is repeated usually 2 to 4 times" is not specific and hence the number of the repetition depends on the arbitrary selection of whether a doubt exists, and whether a doubt continues to exist and hence, of which number of repetition(s) shall apply. Implicitly, the measurement could also be repeated until a desired result is obtained since the method is not clearly limited to such "usual" repetition of two to four times. Moreover, the resultant value for the smallest fiber diameter depends on further arbitrary choices such as the order in which the samples are determined. This effect was demonstrated by the appellant-opponent OI in the test settings provided with letter of 8 October 2015. The data show that when examining 4 groups of 100 fibres, the test result depends on the order in which the samples are measured.

2.10 Moreover, the method defines the smallest (minimum) fiber diameter in relation to a frequency distribution and in relation to an average without specifying how to obtain this average. When applying a single measurement (no case of doubt, no repetition of the measurement), an average cannot be calculated since there is only one value produced. When applying multiple measurements, the test procedure could be read such as to require averaging across multiple samples or possibly not since the test method is not clear on this point.

2.11 When considering multiple measurement (2 to 4 times) as applicable, there is further no clarity on which sample the measurement shall be based. There is no method step present which indicates whether the repetition in case of doubt requires re-measuring the diameters of the 100 fibers which were already examined, selecting another 100 fibres from the same earpickful and measuring them or taking a fresh earpickful of fibres and restarting the complete measurement method. Hence, the result of the repetition of the measurement depends additionally on the arbitrary selection of the sample.

2.12 Accordingly, a reliably reproducible value of smallest fiber diameter cannot be obtained. In such event it is not a question of an undue burden but it is the mere lack of a consistent and reproducible instruction in the method which renders the skilled person incapable to supplement the lacking information by his common general knowledge. Thus the specification neither alone nor in combination with the relevant common general knowledge provides a fully self-sufficient technical concept as to how the smallest diameter of the fiber is to be determined. Accordingly, the skilled person is unable to be sure that he has achieved the aggregate of claim 1 because he cannot verify it by a reproducible and reliable method confirming the smallest diameter being less than 3 µm. Accordingly, the skilled person would not be able to identify reliably and reproducibly the aggregate of claim 1.

2.13 Hence, claim 1 of the Main Request in the version C fails to meet the requirement of Article 100(b)/83 EPC. This conclusion applies irrespective of whether the requested amendment/correction of translation in step (7) of paragraph [0029] of the specification regarding the term "greater than 1% on the average" to be replaced by "not less than 1% on the average" would be allowable. Thus there was no need to take a decision on the requested amendment/correction.

3. Auxiliary requests 1 to 7 in the version C

3.1 The reasoning given for the Main Request above also applies to claim 1 of the Auxiliary Requests 1 to 7 in the version C, which thus consequently do also not meet the requirement of Article 83 EPC.

4. Main request in the version E - Admittance

4.1 The requests in the version E were filed in reply to the communication of the Board sent as an annex to the summons to oral proceedings. As set out already under point 1.1 above, according to Article 13(1) RPBA, it lies within the discretion of the Board to admit any amendment to a party's case after it has filed its grounds of appeal or reply.

4.2 Claim 1 was amended to include additionally some method steps. The added method steps were based upon the method disclosed in paragraph [0029]. The therein disclosed method steps (1) to (3) were added unaltered. The therein disclosed method steps (4) and (5) were adapted to the claim language yet maintaining all relevant elements of the method step. The method step (6) as well as the equation for calculating the average diameter were not inserted and the final method step (7) was only partly inserted into claim 1.

4.3 Hence, parts of method step (7) were omitted. These parts concern the prescription of what to do in case of doubt, (which was to repeat the measurement "usually 2 to 4 times"), and the determination of the (minimum or) smallest fiber diameter by determining a frequency distribution and the diameter whose frequency distribution is greater than 1% "on the average".

4.4 The deletion of the wording "on the average" alters the skilled person's understanding of the test conditions for establishing the parameter of the ("minimum" or) "smallest" fiber diameter so that he would be confronted with information which was not directly and unambiguously derivable from that previously presented by the application. Although the wording in step (7) is ambiguous to a certain extent, it was acknowledged during the oral proceedings that the wording should be understood such that it was the fiber diameter whose frequency distribution "on the average" is greater than 1% which is reckoned as the minimum (smallest) fiber diameter. Hence, an average was necessary to be established. Accordingly, the proposed amendment is prima facie not allowable under the provisions of Article 123(2) EPC.

4.5 Thus the Board exercised its discretion under Article 13(1) RPBA not to admit the Main Request in the version E into the proceedings.

5. Auxiliary Requests 1 to 7 in the version E

The reasoning given for claim 1 of the Main Request (E) above also applies to claim 1 of the Auxiliary Requests 1 to 7 (E). Accordingly, the respective claims 1 of the aforementioned requests prima facie fail to meet the requirement of Article 123(2) EPC, thus the Board exercised its discretion under Article 13(1) RPBA not to admit the Auxiliary Requests 1 to 7 in the version E into the proceedings.

6. Main request and Auxiliary Requests 1 to 7 in the version G

6.1 These requests were filed during the oral proceedings. As already set out under point 4.1 above, the provisions of Article 13(1) RPBA apply.

6.2 Article 84 EPC provides that the claims shall define the matter for which protection is sought and that they shall be clear and concise and be supported by the description. The requirement laid down in this article that the claims have to be clear reflects the demand for legal certainty, which is of paramount importance in any system where the rights of the public are affected by the grant of a monopoly.

6.3 Claim 1 is amended to include the following step (vi):

"wherein the measurement is repeated four times and the diameter whose frequency distribution is not less than 1% on average is determined to be the smallest fiber diameter".

6.4 The wording "on average" which had been omitted in claim 1 of the (E)-set of requests has been re-inserted in claim 1 in order to overcome the objection under Article 123(2) EPC such as set out above for claim 1 of the (G)-set of requests.

6.5 Additionally, claim 1 has been limited to one option set out in step (7) of the method, namely to repeat the measurement four times, and the word "determined" is included in claim 1, although in step (7) of paragraph [0029] the word "reckoned" is present.

6.6 The appellant failed to convince the Board of the clarity of the amended features.

6.7 The feature that "the diameter whose frequency distribution is not less than 1% on average is determined to be the smallest fiber diameter" is prima facie not clear. According to the explanation of the appellant-proprietor during the oral proceedings, the smallest fiber diameter is the fiber diameter whose frequency distribution on average is not less than 1 % of the total number of fibers which are determined to have the smallest fiber diameter when determined in specific "bins". It is not even clear whether the foregoing phrase correctly describes the procedure to be followed. Accordingly, the Board is not convinced that, in the present context, it would be immediately obvious to the skilled person how to interpret the term "on average" with regard to the smallest fiber diameter. As explained by the appellant-proprietor, the average should be determined on the basis of the four determination steps. However, the set-up of the "bins" remained ambiguous. No method step includes the set-up of the "bins" in 0.5 µm steps such as argued during the oral proceedings. Accordingly, an average could be manipulated by defining the bins in an arbitrary way.

6.8 Additionally, the limitation of claim 1 to one option set out in step (7) of the method, namely to repeat the measurement four times is not disclosed as applying generally but only in case of doubt. Even when applying such repetitive measurement, there is no definition of whether the repetition of the measurement concerns all the method steps (starting with a new sample) or concerns only part of the method steps such as the photography step or the measurement in the measuring chamber of the SEM. Hence, again it is not clear on which basis the average should be determined.

6.9 Moreover, it is not clear which difference may arise due to the change of the word "reckoned" into "determined". When desiring at such a late stage to insert a different word, there remain doubts as to whether the scope of the claim is thus altered. At least it is not clear what effect such a change would have.

6.10 The view of the appellant/proprietor that irrespective of whether the complete procedure or only part of the procedure was repeated four times, an average could be produced is correct in itself. However, when referring to a threshold parameter in a claim, it represents an essential technical feature. Hence, its determination should be reproducibly repeatable without any doubts about whether the result is correct. When not clearly defining on which basis the repeatability of the result can be obtained - here in that the resultant average can be influenced by various choices to be taken - there is at least prima facie a lack of clarity (Article 84 EPC) in that the method steps are deficient in that they do not give a clear instruction of how to proceed.

6.11 Hence, the Board exercised its discretion under Article 13(1) RPBA not to admit the Main Request and Auxiliary Requests 1 to 7 in the version G into the proceedings, since the respective claims 1 of the aforementioned requests prima facie failed at least to meet the requirement of Article 84 EPC.

Order

For these reasons it is decided that:

1. The decision under appeal is set aside.

2. The patent is revoked.