T 0257/89 () of 25.10.1991

Summary of Facts and Submissions

I. European patent No. 0 056 703 was granted on 11 March 1987 on the basis of European patent application No. 82 300 150.8 filed on 12 January 1982.

II. Three oppositions were filed against this patent on the grounds of lack of novelty and inventive step (Article 100(a) EPC), insufficiency of disclosure (Article 100(b) EPC) and inadmissible extension of subject-matter (Article 100(c) EPC).

Of the documents cited in support of the oppositions, the following ones are of particular relevance:

(D1) CH-A-500 060 (corresponding to US-A-4 037 011),

(D2) US-A-4 058 581,

(D3) US-A-3 765 998,

(D4) US-A-3 993 726,

(D5) US-A-3 154 908,

(D6) McMahon, P.E. and Maximovich, M., International Conference on Composite Materials - III Proceedings, Vol. II, 1980, pages 1662 to 1673,

(D7) US-A-3 920 879 (corresponding to DE-B-2 253 048),

(D8) US-A-3 022 210 and

(D9) US-A-3 367 814.

III. In a decision dated 27 February 1989, the Opposition Division revoked the patent on the grounds that the product according to Claim 1 as granted was not novel over document D2, and the processes according to the independent Claims 8 and 11 as granted did not involve an inventive step in the light of documents D1, D2 and D5.

IV. The Appellant (Proprietor of the patent) filed a Notice of Appeal against this decision on 10 April 1989. The appeal fee was paid on the same date. The Statement of Grounds was received on 5 July 1989.

V. Oral proceedings were held on 25 October 1991. Respondent III did not attend as previously announced by letter dated 8 October 1991. The Appellant presented a modified set of Claims 1 to 13 together with a correspondingly revised description.

The independent Claims 1, 4, 8 and 10 are worded as follows:

"1. A thermoformable, fibre-reinforced structure comprising a thermoplastic polymer and at least 30% by volume of parallel, aligned reinforcing filaments, for example carbon filaments of 7 micron diameter or glass filaments up to 24 micron diameter, characterised in that the filaments have been wetted by molten thermoplastic polymer from a melt which has a viscosity at low shear rates of less than 100 Ns/m2 in a melt pultrusion process so as to give a continuous structure having a longitudinal flexural modulus determined by ASTM D790-80 of at least 70% of the theoretically attainable flexural modulus.

4. The use of a thermoformable fibre reinforced structure according to any one of Claims 1 to 3 to make a product in which the molecular weight of the thermoplastic polymer has been increased after the filaments have been wetted by the thermoplastic polymer.

8. A process of producing a fibre-reinforced structure according to Claim 3 comprising drawing a plurality of continuous filaments through a melt of a thermoplastic polymer having a melt viscosity at low shear rates of less than 30 Ns/m2 to wet the filaments with molten polymer, the filaments being aligned along the direction of draw and being drawn under tension over at least one spreader surface situated in the molten polymer.

10. A process of producing a fibre-reinforced structure according to Claim 1 comprising tensioning and aligning a plurality of continuous filaments to provide a band of contiguous filaments urging the band against a heated spreader surface so as to form a nip between the band and the spreader surface, as the band is pulled over the spreader surface maintaining a feed of a thermoplastic polymer at the nip, the temperature of the spreader surface being sufficiently high to give a polymer melt of viscosity at low shear rates of less than 100 Ns/m2 capable of wetting the continuous filaments as they are urged against the spreader surface."

VI. The arguments brought forward by the Respondents against the present set of claims can be summarised as follows:

As to formal aspects, Respondent I argued that the feature of Claim 1 "from a melt which has a viscosity at low shear rates of less than 100 Ns/m2" was a process feature which described the state of the melt but not the product itself. This feature was, therefore, not appropriate to distinguish the product of Claim 1 from the prior art. Respondent II took the view that the feature "viscosity ... of less than 100 Ns/m2" had no basis in the application as originally filed and thus contravened Article 123(2) EPC. In addition, he pointed out that the viscosity depended on temperature.

As far as novelty was concerned, the Respondents, whilst admitting that the processes according to Claims 8 and 10 were novel, argued that the product of Claim 1 was anticipated by document D6. It was true that this document did not disclose the feature concerning the viscosity of less than 100 Ns/m2, but this feature could be concluded from it.

In case the claimed product was considered to be novel, it would at least involve no inventive step. Two approaches could be followed. Starting from document D2 as the closest state of the art and trying to overcome the disadvantage of voids present in the solution impregnated products of this document, it would be obvious to make use of the teaching of documents D7 or D3. Indeed, these latter documents disclosed the improvement of impregnation and some mechanical properties and the reduction of void contents by using polymers of extremely low molecular weight and thus extremely low viscosity for melt impregnation of fibrous reinforcements. On the other hand, when starting from document D6 as the nearest prior art, the disadvantage of insufficient wetting, assuming that there was any, would be avoided again by an obvious combination with document D7 or D3, thus leading to the claimed product. In addition, documents D1, D8 and D9 would give a hint in the direction of solving the problem underlying the patent in suit.

VII. In response, the Appellant submitted that Claim 1 was admissible under Article 123(2) EPC and that its subject- matter was novel and involved an inventive step. Irrespective of which document was taken as the closest state of the art, the combination of documents as suggested by the Respondents was neither obvious nor would it necessarily lead to the subject-matter of Claim 1.

VIII. The Appellant requests that the decision under appeal be set aside and the patent be maintained on the basis of Claims 1 to 13 and description pages 2, 4 and 9 as submitted at the oral proceedings in combination with description pages 3, 5 to 8 and 10 to 23 as granted. The marginal note "As Claim 1" on page 2 of the description means that the text "a thermoformable, ... at least 70%" on page 2, lines 43 to 47 is to be replaced by the text of Claim 1. Similarly, the note "As Claim 10" on page 4 means that the text "According to Claim 11 ... spreader surface." on page 4, lines 55 to 60 is to be replaced by "According to Claim 10 there is provided" followed by the text of Claim 10.

The Respondents request that the appeal be dismissed.

Reasons for the Decision

1. The appeal is admissible.

2. Amendments

The amended claims meet the requirement of Article 123(2) EPC. In particular, the feature concerning the viscosity at low shear rates of less than 100 Ns/m2 can, in the view of the Board, unequivocally be deduced from the description as originally filed. Indeed, it follows from page 9, line 33 to page 10, line 2 of this description that the claimed structure can be obtained even though the thermoplastic polymer used has a melt viscosity significantly in excess of 30 Ns/m2. On the other hand, the Appellant has shown in Examples 4, 31 and 44 of the original description that conventional grades of polymers, because of their high viscosities, lead to poor wetting and low flexural moduli and are not conveniently processed by pultrusion. Such conventional thermoplastic polymers are stated as having melt viscosities of 100 Ns/m2 or more (cf. page 15, lines 24 and 25 and Examples 4, 31 and 44). Thus, it is directly and unambiguously derivable from the original disclosure that the thermoplastic polymers used to achieve the claimed product have melt viscosities of less than 100 Ns/m2.

The term "at low shear rates" in the claims has its basis on page 3, lines 12 and 13 of the original description.

Furthermore, the features now incorporated in the claims represent a clear limitation of the scope of protection in comparison with the granted claims. The amended claims, therefore, do not contravene Article 123(3) EPC either.

3. Interpretation of the claims

The feature of Claim 1 that the filaments have been wetted by molten thermoplastic polymer in a melt pultrusion process is to be construed as meaning that the achieved structure is substantially free of solvents and that the product of this melt pultrusion process thereby differs from a product obtained by a solution pultrusion process.

Based on the Appellant's statement made during the oral proceedings that the crosshead extrusion process as mentioned in documents D6, page 1666, and D4, being a high pressure method, is also distinct from the melt pultrusion process used in the patent in suit, wherein a spreader surface is mandatory, the claims are to be interpreted in this way.

Unlike Respondent I, the Board is of the opinion that the feature of Claim 1 "wetted by molten thermoplastic polymer from a melt which has a viscosity at low shear rates of less than 100 Ns/m2", i.e. a low molecular weight, characterises the product itself insofar as it defines the structural feature that the polymer matrix of the product has a low molecular weight.

4. Sufficiency

This matter was no longer pursued in the appeal. It is understood that the two methods claimed in Claims 8 and 10 necessarily provide the products according to Claim 1. The former, of course, is confined to the preferred set with the melt viscosity of less than 30 Ns/m2, while the latter produces products within the full range having viscosities of less than 100 Ns/m2. Although the process claims do not show the critical characteristic of Claim 1, i.e. the "at least 70% of the theoretically attainable modulus", the Board was given to understand by the Appellant that this necessarily follows from the processes claimed. There is no reason to assume that the disclosure is insufficient to enable the skilled person to carry out the inventions claimed, or that the process claims are broader than the product claims.

5. Novelty

The structure according to Claim 1 is novel over the prior art documents mentioned during the proceedings. In particular, documents D1, D3, D4, D5, D7, D8 and D9 do not disclose continuous products having a longitudinal flexural modulus of at least 70% of the theoretically attainable flexural modulus. Document D2 concerns a solution pultrusion process and, therefore, a product which is, in contrast to the claimed structure, not substantially free of solvents.

Document D6 (cf. page 1666 in combination with Table 14) discloses thermoplastic/carbon fibre composites whose flexural moduli meet the requirement specified in Claim 1. However, this document is, as admitted by the Respondents, silent as to the melt viscosity of the molten thermoplastic polymers applied for wetting the filaments. One of the Respondents alleged during the oral proceedings that polybutylene terephthalate having a viscosity of less than 100 Ns/m2 was actually applied. Since no evidence was produced, the Board has no reason to doubt that the polymers nylon and polybutylene terephthalate used for coating are of conventional grades, i.e. have melt viscosities in excess of 100 Ns/m2, and that, for this reason, the polymers of the matrix of the product have high molecular weights. At least, the disclosure of document D6 is silent about this feature, and there is no compelling reason at all to assume that this was unequivocally implied in view of the above.

In addition, the flexural moduli mentioned on page 1666 and in Table 14 were measured using specimens formed by hot pressing of continuous tapes which were apparently cut to the appropriate dimensions, stacked to provide the desired thickness and then pressed (cf. page 1664, fifth paragraph of document D6). Thus, the flexural moduli measured were those of a non-continuous rather than a continuous structure as mentioned in Claim 1.

The novelty of the processes according to the independent Claims 8 and 10 was not disputed by the parties, and follows anyway automatically from the novelty of the products in the present case.

6. Closest state of the art

As acknowledged by the Respondents in one of their approaches, document D6 represents the state of the art which is closest to the subject-matter of Claim 1. According to page 1666, last two paragraphs of this document, continuous tapes were made by a crosshead extrusion coating method. After hot pressing the tapes to form composite laminate test specimens, which pressing step is, as mentioned in paragraph 5 above, apparently non-continuous, flexural moduli were measured. The data shown in Table 14 are 127 GN/m2 for a carbon fibre/nylon composite and 142 GN/m2 for a carbon fibre/polybutylene terephthalate composite. These figures are within the range specified in Claim 1, i.e. "at least 70% of the theoretically attainable flexural modulus".

Nevertheless, the disclosure itself is somewhat problematic for the skilled person in the absence of any proper examples, and needs to be supplemented on the basis of general knowledge to obtain testable embodiments.

7. Problem and solution

In the light of the closest prior art mentioned above, the objective technical problem relevant to the product according to the patent in suit can be formulated. It is to be seen in the provision of further thermoformable, fibre-reinforced structures which as continuous structures exhibit a flexural modulus of at least 70%, and preferably significantly more, of the theoretically attainable flexural modulus.

The problem is solved in accordance with Claim 1 by the use of molten thermoplastic polymer having a viscosity at low shear rates of less than 100 Ns/m2 in a melt pultrusion process so as to give a continuous structure comprising a matrix of low molecular weight thermoplastic polymer and having the wanted flexural modulus.

In view of the numerous examples mentioned in the patent specification which also show that higher than 90% of theory can be attained, the Board is satisfied that the problem defined above is effectively solved.

8. Inventive step of Claim 1

8.1. The primary question is whether or not it was obvious to a person skilled in the art and faced with the problem indicated above to modify the structure as known from document D6 so as to arrive at the structure according to Claim 1. In this respect, the following is observed:

8.2. The only documents mentioned in the proceedings which concern the wetting of filaments by molten thermoplastic polymers having viscosities of less than 100 Ns/m2 are documents D3, D7, D8 and D9. It is true that, as stated by one of the Respondents, document D1 teaches that the amount of polymer picked up by a strand or roving can be controlled by the viscosity of the thermoplastic melt. However, this document does not suggest that the viscosity should be as low as less than 100 Ns/m2.

8.3. Hence, the question as to whether the structure of Claim 1 involves an inventive step is reduced to the question whether any of the documents D3, D7, D8 and D9 would have given the skilled person a hint in the direction that thermoplastic polymers having melt viscosities of less than 100 Ns/m2, i.e. polymers of extremely low molecular weight, could have been used instead of conventional high molecular weight thermoplastic polymers as applied in document D6 without any detrimental effect on the high value of flexural modulus.

8.4. Document D7 is concerned with the use of extremely low molecular weight thermoplastic polymer in a composite reinforced with long glass fibres. The extremely low melt viscosity of such polymers in the range of less than 100 Ns/m2 allows better wetting and thorough impregnation of the glass fibre strands by the molten polymer and results in enhanced impact strength and strength index S2I and lower void contents in the finished product. However, although the document furthermore mentions improved stiffness, a comparison of the results of the Examples 1 and 3 with the corresponding comparative Examples 2 and 4 clearly shows that composites comprising polymers having low molecular weights or, equivalently, low melt viscosities exhibit flexural moduli which are roughly only half as high as those comprising standard, high molecular weight polymers despite the fact that the specimens of the comparative examples contain only short glass fibres. This is a clear warning not to use low molecular weight, i.e. low melt viscosity materials of high flexural modulus composites are sought. Hence, the teaching given in document D7 cannot have provided any incentive to replace the standard, high molecular weight polymer of the structure known from document D6 by a low molecular weight or low melt viscosity polymer without impairing the high flexural modulus of the known structure. It would rather prevent the skilled person from doing so.

8.5. Similarly, document D3 discloses the improvement of some mechanical properties, such as impact and flexural strengths, if low molecular weight thermoplastic polymers are used in long fibre-reinforced structures. However, again the flexural modulus of structures containing low molecular weight polymers (Examples 1 to 5) is lower than that of high molecular weight polymers (Example 6) as exemplified by Table I. When considering this table, it should be borne in mind that the sample of Example 6 contains but 30% by weight of short glass fibres whilst the specimens of Examples 1 to 4 have a content of 40% of long glass fibres and that a specimen similar to that of Example 6 but having 40% by weight of glass content should be expected to have an even higher flexural modulus than that of Example 6. This would render the impairment of the flexural modulus in low molecular weight polymer composites even more pronounced. Thus, the skilled person would be led away from using low molecular weight, i.e. low melt viscosity polymer in long fibre composites because the flexural moduli achieved are worse.

8.6. The teaching of document D8 is in some respects similar to that of document D3. It also discloses an improvement of flexural strength by using low molecular weight resinous material. However, such an increase of flexural strength is, as known from documents D7 and D3 for such structures, not connected with an increase of flexural modulus. Document D8 is silent as to the influence of the molecular weight on the flexural modulus. The same applies to document D9. These documents, therefore, do not suggest the use of low molecular weight polymers if the flexural modulus should not be worsened.

8.7. Furthermore, documents D7, D3, D8 and D9 do neither disclose nor suggest that a flexural modulus of at least 70%, let alone higher than 90%, of the theoretically attainable flexural modulus can be obtained by the use of low molecular weight polymers for wetting the filaments.

8.8. Summing up, the Board comes to the conclusion that it was not obvious for a skilled person to modify the teaching of document D6 by using low melt viscosity polymers so as to arrive at the claimed solution of the existing problem. In view of the above the subject-matter of Claim 1, as a structure, displays properties at a level which is unexpected in the light of documents D3, D7, D8 and D9, and opens the possibility to provide the product in a continuous form. This is an important contribution to the art.

8.9. Accordingly, the argument brought forward by the Respondents that the continuous tapes made in accordance with document D6 would probably have exhibited the high flexural modulus indicated in Table 14 of this document even without the following consolidation step by hot pressing is irrelevant. In view of the problematic character of the disclosure of the document which requires the application of common knowledge, any speculation about deviations from the straightforward have little credibility.

8.10. Starting from document D2 as the closest state of the art, as suggested by the Respondents in a second approach, would not have led to a result different from that indicated in paragraph 8.8. Indeed, the teaching of this document is further away from the subject-matter of Claim 1 than document D6. Thus, in addition to changing the type of polymer from high to low molecular weight polymer, which, as reasoned above, is considered as being inventive, it would also have been necessary to replace the solution pultrusion process disclosed in document D2 by a melt pultrusion process.

9. Independent Claims 4, 8 and 10

Since the products of Claims 1 and 3 are novel and non- obvious there is no need to investigate the obviousness of processes which inevitably result in the manufacture of that product or relate to the use of that product.

Claims 4, 8 and 10 are expressly limited to Claim 1 or 3 and are therefore also patentable.

10. Hence, the patent may be maintained on the basis of independent Claims 1, 4, 8 and 10 and all dependent claims.

ORDER

For these reasons, it is decided that:

1. The decision under appeal is set aside.

2. The case is remitted to the first instance with the order to maintain the patent on the basis of the documents specified in paragraph VIII above.