T 1122/02 24-08-2005

Reasons for the Decision

1. The appeal is admissible.

2. Admissibility of documents

2.1 In addition to documents D1 to D4 mentioned in the Notice of Opposition, the Opponent cited three further documents after the nine month opposition period (Article 99(1) and Rule 55(c) EPC), namely with its letters dated 13 July 2000 and 16 July 2002. Moreover, in the oral proceedings before the Opposition Division, yet another document was referred to by the Opponent, which had a relevant filing date more than two years after the filing date of the patent in suit and more than three years after the claimed priority date and, therefore, did not comply with the definition of "state of the art" in Article 54(2) EPC.

According to the minutes of those oral proceedings (item 5) and the decision under appeal (item I.10), the Opposition Division, after consideration of the relevance of the prior art documents, did not admit them into the proceedings under Article 114(2) EPC, with the sole exception of D7, which was considered as representing the closest state of the art.

The Board has no reason to take a different position with respect to the admissibility of these late-filed documents. Moreover, the Respondent gave its consent to take D7 into account, and the Appellant argued that D7 was more relevant than other available prior art.

2.2 With regard to the further experimental data in documents E2 and E3 as filed with the Statement of Grounds of Appeal (section VII, above, paragraphs 2 to 5), the Board has come to the following conclusions:

2.2.1 These data are a direct reply to the reasons (Nos. 3.4 and 3.5) in the decision under appeal, that the Opponents had not provided evidence to support their allegations and, consequently, had not shown that the claimed invention was made obvious by the cited prior art, and the data were filed at the earliest possible occasion (the Statement of Grounds of Appeal). Hence, they neither delay the proceedings, nor constitute an abuse of procedure.

2.2.2 The experimental report contains Examples 1 to 13, all relating to gas phase polymerisations of ethylene, carried out in different marginal conditions. They contain some examples which prima facie show, that the catalyst activity, measured in the steady state, in the course of the polymerisation in a given example, before and after changes of the molar HHC/Ti ratio within the limits of Claim 1, was affected by these changes. This was found valid for Examples 4, 9, 10, 11 and 12 of E3 as displayed in the tables of E2.

2.2.3 Therefore, the Board decided that the above examples of E2/E3 were relevant to the case and informed the parties during the oral proceedings, that they were admitted into the proceedings (section VII, above, paragraph 6).

3. Article 123(2) and (3) EPC

No objections have been raised in these appeal proceedings against the claims on file in respect of these requirements. Nor has the Board any reason to do so.

4. Problem and solution

4.1 The patent in suit concerns a process wherein ethylene is polymerised in the presence of hydrogen and of a halogen-containing compound with the aid of a Ziegler- Natta type catalyst comprising (i) a catalyst component containing Ti, Mg and halogen atoms and (ii) an organometallic compound of groups I to III of the periodic table as the cocatalyst (Claim 1). The process aims at the substantial reduction of the formation of ethane during the polymerisation (column 1, lines 26 to 31).

4.2 Such a process is also known from D7, which refers to the same problem of increased formation of paraffins such as ethane, propane etc. by hydrogenation of the monomer in olefin polymerisation in the presence of large quantities of hydrogen, necessary for controlling the molecular weight of the resulting polyolefin, by means of a high activity catalyst (D7: page 3, lines 22 to 33). The catalyst component (A) of D7 undisputedly also comprises Mg, Ti and halogen, and it is used together with (B) an organometallic cocatalyst of metals of groups Ia, IIa, IIb, IIIb or IVb of the periodic table and (C) a halogen compound selected from HHC compounds, halogens, interhalogenous compounds and halides of Al, Sn, Pb, P, Sb and S, preferably HHC (Claims 1, 8 and 9; page 8, lines 53 to 56).

The amount of component (A) used is preferably equivalent to 0.001 to 2.5 mmol Ti or higher. Components (B) and (C) are used in amounts of 0.02 to 50 mmol and 10/-5 to 500 mmol, respectively. The basis for each of these amounts is given in terms of "mmol per 1 l of solvent or per 1 l of reactor" (page 9, lines 1 to 5).

On the basis of these data, the Opponent had calculated a general range of the molar ratio halogen-containing compound to titanium in the catalyst used to range from 4·10/-6 to 5·10/5.

The polymerisation is carried out in liquid phase or in gas phase (D7: page 9, lines 20/21), which, according to the Appellant, is confirmed by the reference to "per 1 l of the reactor", thus, clearly referring to gas phase polymerisation as opposed to "per 1 l of solvent" concerning the liquid phase polymerisation (section VII, above, paragraph 7).

Whilst the claims are silent in this respect, all the examples in D7 refer to liquid phase polymerisations. Most of them describe the polymerisation of ethylene in the presence of an HHC compound. The reduction of ethane formation in these examples is significantly better than in the respective comparative examples and those using other halogen-containing compounds (see D7: Tables 1 to 4). According to E1, the molar HHC/Ti ratios in the examples of D7 ranged, as calculated by the Appellant and not disputed by the Respondent, from 3.2 to 13.4, and the activities of the catalysts in the examples of D7 ranged from 400 to 1600 g polyethylene/ mmol Ti·h·0.1 MPa (see eg Examples 9 and 11).

4.3 The decision under appeal (Nos. 2.4 and 2.5 of the reasons) held that the subject-matter of the patent in suit did not constitute a "selection invention", because, besides the general statement in the description, no mention was made in D7 of a gas phase polymerisation and the document was clearly directed to a liquid phase polymerisation. Moreover, D7 did not disclose the molar HHC/Ti range of from 0.01 to 1.8, and a selection of the halogenated compounds had to be made from different groups of compounds, which meant that the selected range of molar HHC/Ti ratios was not the only difference between the two processes.

Furthermore, in the decision under appeal, the technical problem to be solved vis-à-vis D7 was seen, on the basis of column 2, lines 7 to 13, of the patent in suit, in the reduction of ethane formation without substantial variation of the average activity of the catalyst (section III, above). In its letter of 20 August 2003 (page 6, item 3.1), the Respondent agreed to this formulation of the technical problem.

4.3.1 In view of the arguments of the Appellant that the subject-matter of the claims under consideration constitutes a selection from the process of D7 in the sense of T 198/84 and T 279/89, above; section IV, last paragraph), the Board has come to a conclusion different from the decision under appeal, for the following reasons:

Both D7 and the patent in suit relate, in the first place, to an olefin polymerisation process, wherein the hydrogenation of olefin monomer to the corresponding saturated hydrocarbon, in particular of ethylene to ethane, is to be reduced (D7: page 4, lines 19 to 21 in conjunction with page 3, lines 29 to 33; the patent in suit: column 1, lines 10 to 15, 26 to 31 and 38 to 51).

As conceded by the Respondent, the catalyst system as defined in Claim 1 fulfils all the requirements of the definition of the catalytic system in Claim 1 of D7. Thus, both catalytic systems in D7 and the patent in suit undisputedly comprise atoms of titanium, magnesium and halogen, so that, in this respect, the claimed subject-matter is not distinct from the process of D7.

In the description of D7, mention is also made of the polymerisation of olefin in either liquid or gas phase, as does the patent in suit in its granted version (column 8, lines 43/44; as regards D7, see section VII, above, paragraph 7; and section 4.2, above).

Moreover and as already pointed out in section 4.2, above, HHC compounds are the preferred choice of the halogen-containing compounds in D7. In this respect, even the decision under appeal referred to a selection, however, without referring to this preference of HHC compounds in D7 (No. 2.4 of the reasons, item (3)).

Since D7 is, however, silent about a range of the molar HHC/Ti ratio, the Appellant had calculated specific values of the molar HHC/Ti ratio for those examples of D7, which comprised the use of a HHC, and, on the basis of the ranges of concentrations/amounts of its components (A), (B) and (C) (D7: page 9, lines 1 to 13), a general range for this feature extending over about eleven magnitudes including the range in Claim 1 of the patent in suit (cf. the above sections IV, VII (paragraph 8) and 4.2).

Hence, in the Board's view, the claimed subject-matter relates to particular elaborations of the process of D7. This finding gives rise to the question of whether the selection made has been purposive rather than arbitrary.

4.3.2 The Appellant further disputed that the decision under appeal referred to the relevant technical problem to be solved by the claimed process with regard to D7 (cf. sections IV and 4.3, above). In support of this view, it argued that the patent in suit contained only examples to batch polymerisations, so that the patent in suit did not demonstrate that both of those aspects of the problem as defined in the decision under appeal, ie the reduction of ethane formation without substantial variation of the average catalyst activity, were solved within the whole range of the claims. Rather, the experiments in E2/E3 would show that, in continuous gas phase polymerisations of ethylene, the variation of the molar HHC/Ti ratio also affected the average catalyst activity, contrary to the Respondent's assertions in this respect (patent in suit: column 2, lines 11 to 13: "Surprisingly within these ranges no substantial variation of the average activity of the catalyst is observed".). Hence, the selection was, in the Appellant's view, arbitrary, because the selected range had only the same capabilities and properties as the whole range (section IV, above, Statement of Grounds of Appeal, page 5, first complete paragraph).

4.3.3 Although the Appellant's argument that the patent in suit did not provide data concerning the continuous polymerisation of ethylene is not valid in view of its Example 1 describing such a process, the Board nevertheless takes the view that the patent in suit does not contain sufficient information which would convincingly confirm the assertion in the specification, quoted above, and would invalidate the results in E2/E3.

Thus, Example 1 of the patent in suit does not provide any information as to the activity of the catalyst, let alone the influence of the molar HHC/Ti ratio thereon.

Nor can Examples 2 to 4 convincingly provide such information. These latter examples relate to discontinuous polymerisations of ethylene, carried out in the absence of a HHC (Example 2), in the presence of 0.6 mmol of CHCl3 (Example 3) and in the presence of 1 mmol of CH3-CCl3 (Example 4), respectively. Hence, Examples 3 and 4 do not fulfil the requirements for convincing evidence with regard to the above assertion concerning a constant catalyst activity either, because they differ in both the nature of the HHC used and the amounts thereof. Despite the fact that they were carried out in liquid phase, this finding is also supported by the results of Examples 1 and 3 to 7 of D7, which show that the different compounds affect the ethane formation to a different degree. Namely, in Examples 1 and 6, wherein not only identical amounts of the respective HHC compounds were used, but wherein also the molar HHC/Ti ratios were the same, different degrees of reduction of the ethane formation were found: the use of 1,2-dichloroethane in Example 1 gave 0.005% ethane formation (page 12, lines 8/9), that of 1,1-dichloroethane in Example 6, however, 0.013% (page 13, Table 1).

Consequently, the examples in the patent in suit do not allow any meaningful conclusion to be drawn with regard to the influence of different molar HHC/Ti ratios. They only show that, by adding such compounds, the ethane formation can, in principle, be reduced, as suggested by D7.

In fact, nowhere in the patent in suit has an assertion been made that specific molar HHC/Ti ratios would result in an improvement in respect the degree of reduction of the ethane formation. Rather, the patent in suit refers to the importance of the specific range of molar HHC/Ti ratios only in the context of the alleged absence of any variation of the average catalyst activity (column 2, lines 11 to 13).

4.3.4 The Appellant has disputed, however, on the basis of its additional experiments filed in E2/E3, that the claimed effect of constant catalyst activity would be achieved over the whole range of the claims.

The data in E2/E3 as admitted into these proceedings (sections 2.2.2 and 2.2.3, above) demonstrate, indeed, that, in a given example, the further addition or reduction of the molar HHC/Ti ratio, within the limits of the range defined in Claim 1, in the course of a continuous gas phase polymerisation of ethylene resulted in a variation of the average catalyst activity, as demonstrated by measurements in steady state before and after the changes of the HHC feed (E3: page 1, first paragraph).

Thus, in Example 4, the decrease of the molar CHCl3/Ti ratio in one step from 1.5 to 0.5 resulted in a reduction of the average catalyst activity from 205 to 165 g PE/mmol Ti·h·0.1 MPa, when the steady state was again reached after approximately 6 h of transition (E3: page 8, lines 2 to 16).

In Example 9, an increase of the molar CHCl3/Ti ratio in one step from 0.04 to 0.1 (ie in the region of low contents) resulted in a slight increase of the catalyst activity from 90 to 92 g PE/mmol Ti·h·0.1 MPa after 8.5 h of transition (page 18, lines 6 to 13).

In Example 10, the molar CHCl3/Ti ratio was increased in one step from 0.1 to 0.2, resulting in an increase of the catalyst activity from 92 to 111 g PE/ mmol Ti·h·0.1 MPa after the transition which took approximately 4.5 h (page 19, lines 3 to 10).

In Example 12, the molar CHCl3/Ti ratio was increased in three equal steps from 0.3 to 0.6. The initial catalyst activity prior to any change of the molar CHCl3/Ti ratio was 143 g PE/mmol Ti·h·0.1 MPa, the final activity at a molar CHCl3/Ti ratio of 0.6 was found to be 180 g PE/ mmol Ti·h·0.1 MPa (page 20, last three lines, page 21, last three lines and page 22, lines 1/2). In addition to these data, the initial prepolymer productivity and the prepolymer productivity (in terms of kg of polyethylene per kg of prepolymer) for all subsequent steps is provided.

Whilst conceding that the catalytic system in the experimental report complied with the definition in Claim 1 of the patent in suit, the Respondent argued that it would not be apparent in E2/E3 that the other reaction conditions had been maintained constant, that, during the transition, further variations of the activity had occurred, and that the selection of a HHC from the various types of compounds in D7 had been made by hindsight. Therefore, no reliable and meaningful conclusions from these experiments could be drawn. However, the Board cannot concur with these arguments, because, in each of these examples, explicit reference is made to the process conditions in a table contained in the report and no inconsistencies have been found between the values in the table referred to and those values relating to process conditions specified in the description of each example, and the use of HHC was preferred in D7 (sections 4.2 and 4.3.1, above).

Moreover, the values of the molar HHC/Ti ratios of 0.04, 0.1, 0.2, 0.3, 0.5, 0.6 and 1.5 in these experiments are, in the Board's view, representative of the claimed range of from 0.01 to 1.8, and variations of activity during a transition, ie in unstable conditions, cannot invalidate comparisons of the measurements carried out in steady state before and after the transition. Therefore, the Board cannot accept the argument that the examples in E2/E3 would only relate to a "snapshot" which is not representative for a continuous polymerisation process according to the claims of the patent in suit.

In any case, the reported results in the admitted experiments of E2/E3 per se have not been disputed by the Respondent. Neither within the time limit of four months set in the communication of 20 February 2003 with which the Statement of Grounds of Appeal and E1 to E3 had been communicated, nor in the meantime, has the Respondent filed any counter-experiments to refute these experimental data.

4.3.5 Consequently, the Board has come to the conclusion that these experiments of the Appellant, in contrast to the statement in column 2, lines 11 to 13 of the patent in suit, convincingly show that the catalyst activity is affected by the molar HHC/Ti ratio in the reaction mixture of a process for the gas phase polymerisation of ethylene within the definitions in Claim 1.

4.4 In view of these facts and findings, the Board is not satisfied that the technical problem as formulated in the decision under appeal (sections III and 4.3, above) is indeed solved by the process claimed throughout the full range of Claim 1.

Consequently, the technical problem to be solved with regard to D7 has to be formulated in a less ambitious way, because the assertion that no substantial variation of the average catalyst activity would occur has been disproved. Consequently, the relevant problem can only be directed to an alternative gas phase polymerisation process for ethylene in which the ethane formation (by hydrogenation of ethylene) is reduced.

5. Novelty

5.1 The decision under appeal in detail dealt with each of D1 to D4 and D7 and came to the conclusion that none of these documents anticipated the subject-matter of the Main Request. As regards D1 to D4, these findings have not been challenged by the Appellant. Nor has the Board any reason to take a different view in this respect.

5.2 As pointed out above (section 4.3.1, above), the claimed process relates to a selection from the disclosure of D7, contrary to the finding in the decision under appeal.

5.3 According to established jurisdiction, certain requirements have to be met so that novelty of a selection can be acknowledged.

5.3.1 Thus, according to each of T 198/84, T 279/89 (both above) and T 265/84 of 17 December 1986 (not published in OJ EPO), such an acknowledgement is only possible when each of the following criteria is fulfilled:

"(i) the selected sub-range should be narrow;

(ii) the selected sub-range should be sufficiently far removed from the known range illustrated by means of examples; and

(iii) the selected area should not provide an arbitrary specimen from the prior art, i.e. not a mere embodiment of the prior description, but another invention (purposive selection)." (see eg in No. 4.1 of the reasons in T 279/89, above, or in T 275/96 of 10 December 1998, not published in OJ EPO, No 4.2.2 of the reasons).

5.3.2 In T 265/84 (above), Number 5 of the reasons reads as follows: "When the question of novelty is considered, it is relevant that most of the claimed brazed varieties fall within the broadest scope of the disclosure in citation (1). The latter document describes all the essential features of the claimed invention in general, without disclosing any example, however, which would be embraced by the claim. Neither does (1) mention the range of limitations in the claim, not considering, of course, those for the optional ingredients in this respect, since these cannot change the novelty situation based on the essential components. In addition, phosphorus is excluded from the subject- matter claimed, whilst the cited art includes its use as a metalloid within a range of alternatives. Such distinctions would not necessarily impart novelty to the selected ranges in the absence of any technical distinction in properties (cf. "Thiochloroformate/BASF" T 198/84, OJ 7/1986, 209, in particular paragraph 7, page 214 on "purposive selection"). This condition for novelty in an essential sense, contrasted to mere formal delimitations, is satisfied in the present case since the brazed articles manifest different, e.g. improved, properties from those in (1). ...".

5.3.3 In Decision T 198/84 (above), No. 7 of the reasons, the Board stated: "... It would be delimited only in respect of the wording of the definition of the invention, but not in respect of its content, if the selection were arbitrary, i.e. if the selected range only had the same properties and capabilities as the whole range, so that what had been selected was only an arbitrary specimen from the prior art. ...".

5.3.4 And in the case of T 766/99 of 26 February 2002 (not published in OJ EPO; No.5.2 of the reasons), a new effect hitherto unknown was neither mentioned in the patent application in suit nor asserted by the appellant/applicant, nor found to be connected to the partial range selected from the broad range of the state of the art, so that a purposive selection could not be acknowledged.

5.4 As shown in sections 4.1 and 4.3.1 to 4.3.5, above, the only feature of Claim 1 which in the present case could serve as a distinction from D7, ie the specific range of the molar HHC/Ti ratio, does not support the assertion of the patent in suit that a variation of the average catalyst activity would be avoided (cf. sections 4.3.4 and 4.4, above). Nor is the reduction of the ethane formation connected to this range, and a purposive selection cannot be acknowledged in this respect, either (see, in particular, section 4.3.3, above).

Rather, the definition of the particular molar HHC/Ti ratio of Claim 1 has to be considered an arbitrary selection which cannot impart novelty to the claimed subject-matter, because of the absence of any technical distinction in the relevant features of and the relevant effects achieved by the processes of D7 and the patent in suit. Therefore, one of the essential conditions for novelty by selection, requiring that a distinguishing feature can be identified, which is clearly related to an additional effect and does not, thus, constitute merely a formal delimitation from the state of the art, is not satisfied in the present case (cf. section 5.3.2, above).

In other words, the range of the molar HHC/Ti ratio as defined in Claim 1 cannot be construed to constitute a purposive selection by which the claimed subject-matter could be distinguished from that of D7, because, by this feature, the claimed subject-matter would be delimited from D7 only in respect of the wording of the definition of the invention, but not in respect of its content (cf. section 5.3.3, above).

5.5 Consequently, criterion (iii) (section 5.3.1, above) required to be fulfilled for the acknowledgement of novelty, is not met.

Since each of the three criteria addressed in that section has to be complied with concomitantly for the acknowledgement of a purposive selection and, on this basis, also for the acknowledgement of novelty, there is no need to further consider the other two criteria (i) broadness of the selection and (ii) remoteness from the examples in the prior art.

5.6 For the reasons given above, novelty of the subject- matter of Claim 1 cannot be acknowledged.

6. Under these circumstances, the consideration of the other issue raised by the Appellant, that of inventive step, is not necessary either, because it cannot result in a different outcome of the decision.

7. Since the above set of claims represents the sole request of the Respondent on file, it follows that the patent in suit cannot be maintained.