T 1618/07 24-09-2008

Reasons for the Decision

1. The appeal is admissible.

2. Appellant's requests

2.1 The appellant requested as main request that the decision under appeal be set aside and that the patent be revoked in its entirety.

However, at the first oral proceedings before the opposition division, on 12 March 2003, the opposition division decided to maintain the patent in suit in amended form based on the proprietor's auxiliary request 2 (point III, above). Only the patent proprietor appealed against that decision.

Thus, according to the principle of reformatio in peius, as set out in G 9/92 (OJ EPO 1994, 875), neither the board of appeal nor any of the non-appealing opponents may challenge the maintenance of the patent as amended in accordance with that decision of the opposition division announced orally on 12 March 2003 and issued in writing on 4 April 2003.

Since Basell Polyolefine GmbH did not appeal against that decision announced orally on 12 March 2003, the request of Basell Polyolefine GmbH in the present proceedings to revoke the patent in suit in its entirety is contrary to the principle of reformatio in peius and is therefore inadmissible.

2.2 Consequently, the only acceptable request of Basell Polyolefine GmbH is its auxiliary request to set aside the decision under appeal and to maintain the patent in amended form according to the decision of the opposition division announced orally on 12 March 2003 with claims corresponding to auxiliary request 2. The claims of that auxiliary request 2 correspond to those of the present 4th auxiliary request.

3. Amendments (main request)

The claims of the respondent's main request corresponded to Claims 1 and 2 as granted, except that in Claim 1 the upper limit of the ratio (Mw/Mn) has been amended to 3.0 (the upper limit in Claim 1 as granted was 4.0). A basis for the amendment can be found on page 18, line 13 of the application as filed. Thus, no objections under Articles 123 or 84 EPC arise against the amendment in Claim 1. Nor has the appellant or the other party advanced any objection in this connection.

4. Sufficiency of disclosure

The appellant raised no objection with respect to sufficiency of disclosure. Nor does the board see a reason to challenge the finding of the opposition division in that respect (point V (b), above).

5. Novelty (main request)

5.1 The appellant does not challenge the finding of the opposition division that the subject-matter of the main request is novel over the alleged further prior uses occurring with the sale of Vistalon® MDV-746 to Republic Plastic Co. and the sale of Vistalon® MDV-746-1 to Dexter Corp. (point V (d), above). Nor does the board see a reason to raise an objection in this connection on its own.

However, the appellant challenges the finding of the opposition division that the subject-matter of the main request is novel over Comparative Example 17 of D1, D7 and Comparative Example 5 of D9.

5.2 Comparative Example 17 of D1

D1 relates in general terms to an ethylene/alpha-olefin copolymer composition comprising two ethylene/alpha-olefin copolymers which are different in density, intrinsic viscosity and the number of short chain branching per 1000 carbon atoms.

In Comparative Example 17 of D1 a low density ethylene/butene-1 copolymer of the conventional technique was synthesized, using the catalyst produced in Example 1 of D1 and triethyl aluminium (cocatalyst) and employing the polymerization conditions shown in Table 21. The copolymer had a density of 0.920 g/cm**(3), a melt index (MI) of 0.5 g/10 min and a melt flow ratio MI21.6/MI2.16 of 30 as shown in Table 18.

5.2.1 It is evident from the above that Comparative Example 17 of Dl provides very little information with regard to the properties (a) to (d) of Claim 1 of the main request. Comparative Example 17 discloses the density, namely parameter (a). However, no indication is given of the intrinsic viscosity (b), molecular weight distribution (c), nor MFR10/MFR2 ratio (d). Further, there is no explicit disclosure in Comparative Example 17 as to the relative amounts of ethylene and comonomer. Table 21 merely indicates the polymerisation conditions, eg amount of butene-1 (in kg), partial pressure of ethylene, polymerisation time. For these reasons, without additional data repeating Comparative Example 17 of Dl in reliable fashion, it is not possible to determine whether this polymer falls within the scope of Claim 1 or not.

5.2.2 Nevertheless, the appellant is of the opinion that the relevant parameters (b) to (d) are inherent to the copolymer produced in Comparative Example 17 of D1. The appellant's argumentation in this connection is, however, not convincing for the following reasons:

(a) The appellant provides a discussion of the invention of Dl, indicating that the general information provided in Dl contains information on the density, intrinsic viscosity and molecular weight distribution. Thus, the general information in D1 would provide information with respect to parameters (b) and (c) of the copolymer produced in Comparative Example 17 of D1. However, any discussion of the invention of Dl is irrelevant in the context of Comparative Example 17. The copolymer produced in Comparative Example 17 is acknowledged by the Dl inventors to have properties different from those of the claimed invention. The appellant cannot therefore rely on any combination of the features explicitly disclosed in Comparative Example 17 with the general disclosure of the Dl invention. It is therefore not possible to determine either from Dl itself or from the argumentation of the appellant, what the intrinsic viscosity or molecular weight distribution of the copolymer produced in Comparative Example 17 of Dl would be.

(b) As regards the MFR10/MFR2 ratio, ie parameter (d) of Claim 1, the appellant provided already in the notice of opposition a calculation. Via that calculation, the appellant determines from values disclosed for Comparative Example 17, namely MI and MI21.6/MI2.16, the MFR10/MFR2 ratio which is not disclosed in D1. This calculation relies on the use of the Ostwaldt-de Waele equation which includes a parameter m relating to the polydispersity index PI. The appellant indicates that the value of m (0.45) has been selected for a polydispersity in the relevant range and calculates a value of 14 for the MFR10/MFR2 ratio for Comparative Example 17 of D1. However, as pointed out by the respondent, since the polydispersity of Comparative Example 17 of Dl is not known, it is not clear whether the chosen value of m is correct or not.

The respondent's objection is supported by the Tsutsui declaration D10. As is apparent from the table provided by Mr Tsutsui on page 6 of Dl0, the MFR10/MFR2 ratio and the molecular weight distribution are intrinsically linked. A precise calculation of the MFR10/MFR2 ratio of a particular polymer therefore cannot be carried out without knowing the molecular weight distribution. Since Dl discloses no value for the molecular weight distribution, it is clear that the appellant´s calculation for MFR10/MFR2 cannot be relied on.

Furthermore, as pointed out by the respondent, the appellant's calculation does not appear to correspond with experimental results for such polymers. The table appearing on page 6 of Dl0 provides the expected relationship between molecular weight distribution and melt flow properties of polymers produced using conventional Ziegler catalysts. In accordance with this table, a polymer having a melt flow ratio MI21.6/MI2.16 of 30, such as that described in Comparative Example 17 of D1 and also produced using a conventional Ziegler catalyst, would be expected to have a molecular weight distribution of about 3.5 and a MFR10/MFR2 ratio of about 7.9. Thus, based on expected results, Comparative Example 17 of D1 would even fall outside the scope of Claim 1 of the main request.

5.2.3 In conclusion, the appellant´s arguments go no further than those previously submitted before the opposition division. The opposition division´s conclusion that Comparative Example 17 has not been demonstrated to fall within the scope of Claim 1 of the main request therefore still stands.

5.3 D7

5.3.1 The article D7 by W. Kaminski et al. relates to the use of a catalyst composition comprising a group IVB transition metal complex and methylalumoxane (MAO) for the production of various types of polymers. The introduction of the article discloses that titanocene, zirconocene and hafnocene, for example biscyclopentadienyl- and bisindenyl transition metal complexes, form together with alumoxane highly active Ziegler-Natta catalysts. Further, it is stated that MAO is a key substance responsible for the high activity of the catalyst composition. Table 2 in D7 lists ethylene homopolymerization and ethylene copolymerization with a zirconocene/alumoxane catalyst. The last paragraph on page 291 states that "The analogous titanium and hafnium compounds form active catalysts too". The copolymerisation of 4-methlypentene-1 and ethylene with two different zirconocenes, namely Cp2ZrCl2 and Et(Ind)2ZrCl2, and methylalumoxane as cocatalyst under various conditions is shown in Table 4 of D7. As regards the properties of the final copolymer, Table 4 discloses only the mol% of 4-methlypentene-1 in the copolymer and the Mn of the copolymer.

5.3.2 Firstly, it is evident from the above that D7, and Table 4 in particular, provides no information with regard to a copolymer having the properties (a) to (d) of Claim 1 of the main request. Secondly, the appellant's assertion (contended at the opposition stage) that D7 discloses the same catalyst and the same starting monomers as required by Claim 2 of the main request and therefore the same polymers as claimed in Claim 1 of the main request is not supported by the disclosure of D7. There is no disclosure in D7 of a process wherein a hafnium containing catalyst is used to produce an ethylene copolymer with properties (a) to (d). Indeed the only reference to the use of hafnium appears on page 291 in the final paragraph, a reference which does neither directly relate to ethylene copolymerisation nor to hafnium complexes having a multidentate ligand as required in Claim 2 of the main request.

5.3.3 In the present appeal, the appellant basically relied on a repetition of the copolymerisation of ethylene and 4-methlypentene-1 with Et(Ind)2ZrCl2 and methyl alumoxane (MAO) according to the 6th condition in Table 4 of D7 to attack the novelty of the subject-matter of Claim 1 of the main request. This repetition is reported in D8.

However, there is considerable doubt as to whether or not D8 is a true reproduction of the 6th condition in Table 4 of D7.

As pointed out by the respondent, the cocatalyst MAO is a compound which has been used in catalytic polymerisation for some time, but its precise chemical structure is still not known. It is however well known in the art that the methods of production of MAO have varied over time and, due to this, the catalytic performance of MAO has improved considerably. The MAO used in D8 is said to be a commercially available sample from Almabale (sic: Albemarle!) Company. Presumably, this is MAO which was commercially available in 1997 rather than that available at the time (1987 or earlier) that the experiments were carried out for D7. The skilled person cannot deduce from D7 the type of MAO which should be used to produce the stated results. Indeed, the respondent has shown by the submission of test reports D19 to D23 that variation of the nature of MAO used in the polymerisation does have a significant effect on the result. Each of D19 to D23 relates to an experiment repeating the work described in D7 but using modern MAOs. Except in D23 the various MAOs were all commercially available and were used either in approximately the amount as suggested in D7 or in reduced amounts to take account of their increased activity compared with materials from 1987 when the work reported in D7 was presumably conducted. Apart from demonstrating the effect of using different MAOs, it can be seen from these reports that each experiment successfully produced polymer but none produced a polymer meeting the MFR10/MFR2 ratio of present claim 1.

Given the significant variations in the properties of the MAO component, it cannot be established that D8 uses the same type of MAO as D7. Consequently, the board agrees with the opposition division´s conclusion of paragraph 6.3 of the reasons for the decision that (i) it cannot be assumed that D8 is a true repetition of the 6th condition in Table 4 of D7 and (ii) the disclosure in Table 4 of D7 is in principle not sufficiently clear and complete for a skilled person to be able to reproduce the catalyst compositions and polymers described.

5.3.4 The appellant contests the opposition division´s conclusion of paragraph 6.3 of the reasons for the decision and argues that there is sufficient information in D7 to enable a skilled person to repeat the polymerisations disclosed in Table 4 of D7, relying on the mere existence of D8 in this regard. This argument, however, does not follow. Simply because D8 exists and is put forward as a repeat of D7 does not in itself show that D8 does provide the inevitable result of repeating D7.

The Appellant is correct that D8 provides a number of details as to how the test was carried out. What is significant, however, is that many of these details are not found in D7 itself. In such a situation, it must be established (a) that the parameters and conditions used in D8 are those which the skilled person would inevitably have used on reading D7, or where this is not possible (b) that variation in any such parameter or condition would have no effect on the end result. Neither of these issues is addressed in any way by the appellant, and in particular not the issue relating to the type of MAO which was used in D7. As concluded by the opposition division, therefore, given the lack of information in D7, it cannot be established whether D8 is a true repetition of D7 or not.

5.3.5 In the final paragraph of the discussion of D7, the appellant alleges that the lack of any mention of the nature of MAO used in D7 is unimportant. This is because the claims of the patent itself do not define the nature of the cocatalyst to be used. Again, this argument does not follow. For D8 to represent the inevitable result of repeating D7, it must be shown that D8 either uses precisely the same conditions and parameters as D7, or that any variation in such conditions and parameters would have no effect on the end result. The manner in which the proprietor has defined its claims is not relevant to this question. The appellant´s final argument in relation to D7 is therefore simply not correct.

5.3.6 In view of the above, the claimed subject-matter is novel over the disclosure of D7 and in particular over the 6th condition in Table 4 of D7.

5.4 Comparative Example 5 of D9

5.4.1 D9 relates to a gas fluidised bed process for the production of ethylene copolymers using conventional Ziegler polymerisation techniques. The polymers produced in all examples except in Comparative Example 5 have a molecular weight distribution Mw/Mn of at least 4.0. Comparative Example 5 of D1 discloses an ethylene copolymer having a density of 0.918 g/cm**(3), a Mw/Mn ratio of 3.0 and a comonomer content of 9.9% by weight (3% butene-1 and 6.9% 4-methyl-1-pentene) (Table 1 on page 25).

5.4.2 The respondent has already argued during the first appeal that the data set out in Comparative Example 5 of D9 are not reliable and the skilled person would consider that Comparative Example 5 contains an error. The ethylene copolymer of Comparative Example 5 is produced by gas phase polymerisation using conventional Ziegler polymerisation techniques and would not be expected to have a Mw/Mn ratio as narrow as the reported "3.0". As demonstrated in the declaration D28, the copolymer of Comparative Example 5 is an anomalous result which does not follow the usual pattern for such conventional polymers, eg polymers as produced in the other examples of D9.

5.4.3 But even if the data reported in Comparative Example 5 were to be accepted at face value, the fact remains that D9 does not describe the intrinsic viscosity or flow characteristics of the copolymer produced in Comparative Example 5, ie properties (b) and (d) of present Claim 1. The appellant argues that since Comparative Example 5 is similar to Example 1 of D9, this polymer would be expected to have the same intrinsic viscosity and MFR10/MFR2 ratio as was determined for Example 1. Example 1 of D9 was repeated in D10 and it was found that the copolymer had an intrinsic viscosity of 1.75 dl/g and a MFR10/MFR2 ratio of 8.3.

However, the appellant's assumption that the copolymer of Comparative Example 5 of D9 is similar to the polymer produced in Example 1 of D9 is not correct. As pointed out above, Comparative Example 5 is an anomalous result within the disclosure of D9 and differs significantly from Example 1 in terms of its molecular weight distribution and flow parameter n (ratio of the melt index (MI21.6) to the melt index (MI2.16)). The copolymer of Comparative Example 5 has a molecular weight distribution of 3.0 and a flow parameter n of 33, whereas the copolymer of Example 1 has a molecular weight distribution of 4.0 and a flow parameter n of 25. As is apparent from D10, the MFR10/MFR2 parameter is dependent on the ratio n and also on the molecular weight distribution. It is therefore not correct to conclude that the polymers described in Example 1 and Comparative Example 5 of D9 must have the same value for the MFR10/MFR2 ratio. Rather, given the difference in flow parameter n and molecular weight distribution, the skilled person would expect these two polymers to have very different values for MFR10/MFR2.

5.4.4 Hence no reliable evidence has been provided as to the values of the intrinsic viscosity and the MFR10/MFR2 ratio for the copolymer produced in Comparative Example 5 of D9. The appellant has not discharged its burden of proof in this respect. Consequently, the claimed subject-matter is novel over Comparative Example 5 of D9.

5.5 In summary, the subject-matter claimed in the main request is novel over the prior art cited by the appellant, in particular Comparative Example 17 of D1, D7 and Comparative Example 5 of D9.

6. Inventive step (main request)

6.1 The claimed subject-matter relates to ethylene copolymers which are excellent in flowability in spite of their rather narrow molecular weight distribution.

The appellant suggests that either D9 or D11 might be used as the closest prior art. Both documents describe ethylene copolymers produced using conventional catalysts such as Ziegler catalysts. Consequently, they equally qualify as the closest prior art.

6.2 Following the amendment to Claim 1 to the preferred range of 1.2 to 3.0 for the molecular weight distribution, the appellant alleges that the new technical problem to be solved is less exacting. On the contrary, the technical problem to be solved by the subject-matter of amended Claim 1 is in fact more exacting. Amended Claim 1 requires a rather narrow molecular weight distribution.

6.3 Accordingly, the objective technical problem over the closest prior art has to be seen in the provision of ethylene copolymers having an improved range of properties, namely a rather low molecular weight distribution in combination with excellent flow properties.

As demonstrated by the examples in the patent in suit, the above stated objective technical problem is solved by the provision of the polymers as defined in Claim 1 and prepared in accordance with the process of Claim 2 of the main request using a specific hafnium catalyst. Thus, the board is satisfied that the objective technical problem is solved.

6.4 D1, D9 and D11 relate to conventional ethylene copolymers made by conventional Ziegler catalysts. None of the prior art documents teaches the skilled person how the conventional polymers would have to be modified in order to achieve the beneficial combination of narrow molecular weight and excellent flow properties, let alone that this beneficial combination might be associated with the use of a specific catalyst. In fact, as is apparent from D10, when using the conventional catalysts described in all of D1, D9 and D11, polymers having a combination of narrow molecular weight distribution and good flow properties cannot be accessed. Moreover, it is apparent from D10 that on lowering the molecular weight distribution the skilled person would have correspondingly reduced the melt flow rate ratio. There is simply no teaching in D1, D9 or D11 which would enable the skilled person to maintain narrow molecular weight distribution whilst simultaneously increasing melt flow rate ratios. Thus, from whichever document the skilled person would start as the closest prior art (D9 or D11), the claimed subject-matter is not obvious from the cited prior art.

6.5 As regards the appellant's argumentation that the claimed subject-matter would be obvious either from D9 (in particular from Comparative Example 5) or from D11 in combination with D9 or D1 it appears that this argumentation is based on an ex post facto analysis in the knowledge of the patent in suit. The prior art does not provide any hint how the melt flow properties of ethylene copolymers prepared by conventional Ziegler catalysts could be further improved whilst simultaneously reducing the molecular weight distribution.

6.6 In summary, the subject-matter as claimed in the main request is based on an inventive step.

7. Since the respondent's main request is allowable, any discussion of its auxiliary requests is superfluous.