T 0307/04 () of 6.10.2006

Summary of Facts and Submissions

I. Mention of the grant of European Patent No. 0 857 194 in the name of Eastman Chemical Company, in respect of European patent application No. 96 936 793.7, filed on 21 October 1996 as international application No. PCT/US96/16854 and claiming priority of US patent application no. 548 162 dated 25 October 1995, published as WO 97/15629 on 1 May 1997, was announced on 28 March 2001 (Bulletin 2001/13) on the basis of 22 claims. Independent claims 1 and 2 read as follows:

"1. A low color polyester blend composition comprising:

(a) 98.0 to 99.95 weight percent of polyester formed from:

(1) a dicarboxylic acid component comprising repeat units consisting of at least 85 mole percent of terephthalic acid, naphthalene dicarboxylic acid and mixtures thereof; and (2) a diol component comprising repeat units from at least 85 mole percent ethylene glycol, based on 100 mole percent dicarboxylic acid and 100 mole percent diol,

(b) 2.0 to 0.05 weight percent of a polyamide; wherein the combined weight percents of (a) and (b) total 100 percent,

characterised in that a major proportion amounting to 80 to 100 wt% of said polyester (a) consists of one or more polyesters obtained by direct condensation of the acid form of the dicarboxylic acid component (a)(l), and a minor proportion amounting to 0 to 20 wt% of the polyester (a) consists of polyester obtained by ester interchange using an ester form of the dicarboxylic acid component (a)(1), thereby to form a polyester blend having less color as measured in b* units than a corresponding polyester blend which is formed from polyester (a) the major proportion of which consists of polyester obtained by ester interchange using the ester form of the dicarboxylic acid component (a)(1) and the minor proportion of which consists of polyester obtained by direct condensation of the acid form of component (a)(1).

2. A low color polyester blend composition comprising:

(a) 80 to 99 weight% of base polyester which comprises:

(1) a dicarboxylic acid component comprising repeat units from at least 85 mole percent of: terephthalic acid, naphthalene dicarboxylic acid and mixtures thereof; and (2) a diol component comprising repeat units from at least 85 mole percent ethylene glycol, based on 100 mole percent dicarboxylic acid and 100 mole percent diol; and

(b) 1 to 20 weight % of a concentrate comprising:

(1) 1 to 99 weight % of a carrier resin comprising a dicarboxylic acid component comprising repeat units from at least 60 mole percent aromatic dicarboxylic acid selected from the group consisting of terephthalic acid, naphthalenedicarboxylic acid and mixtures thereof, and a diol component comprising repeat units from at least 50 mole percent ethylene glycol, based on 100 mole percent dicarboxylic acid and 100 mole percent diol; and

(2) 1 to 99 weight % of a polyamide which displays a melting point below the melting point of said carrier resin,

wherein said base polyester (a) consists of one or more polyesters obtained by direct condensation of the acid form of the dicarboxylic acid component (a)(1), the polyester blend thereby formed having less color as measured in b* units than a corresponding polyester blend formed from a base polyester (a) which consists of polyester obtained by ester interchange using an ester form of the acid component (a)(1)."

Independent claims 3 and 4 each defined a process for forming a low colour polyester blend comprising blending the components (a) and (b) of claims 1 and 2 respectively. Claims 3 and 4 contained the same characterising features as claims 1 and 2 respectively.

Independent claims 5 and 6 defined the use of a polyester (a) defined as in claims 1 and 2 respectively in a polyester blend with a polyamide (claim 5) or a polyamide concentrate (claim 6), as defined in claims 1 and 2 respectively to form a low colour polyester blend. The characterising portion of claims 1 and 2 was retained in claims 5 and 6 respectively.

Dependent claims 7-20 defined preferred embodiments. Claim 21 read as follows:

"21. A composition or process or use as claimed in any preceding claim wherein said polyester blend is formed into an article by extrusion or injection molding".

Claim 22 specified that said article was a bottle.

II. Oppositions against the grant of the patent were filed on 21 December 2001 by Schmalbach-Lubeca AG (OI) and on 27 December 2001 by Arteva Technologies S.à.r.l (OII). Both opponents based their oppositions on the grounds according to Articles 100(a), 100(b) and 100(c) EPC, specifically that the subject matter of the patent was not new contrary to Article 54 EPC, and was not inventive contrary to Article 56 EPC, that the invention was not sufficiently disclosed contrary to Article 83 EPC and that the subject matter of the patent extended beyond the content of the application as filed contrary to Article 123(2) EPC.

III. In a decision announced orally on 5 November 2003 and issued in writing on 30 December 2003, the opposition division revoked the patent. The oral proceedings were attended by the patentee and OII (Arteva Technologies S.à.r.l), OI (Schmalbach-Lubeca AG) having previously indicated in a letter of 25 June 2003 that it would not attend the oral proceedings.

The decision was based on a main and an auxiliary request. The claims according to the main request were those of the patent as granted (claims 1-22).

The auxiliary request consisted of 20 claims.

Independent claims 1-6 had been amended compared to the claims of the granted patent in that instead of being directed to polyester blends they were directed to a low colour blow-moulded bottle (claims 1 and 2), a process for forming a low colour blow-moulded bottle (claims 3 and 4) and the use of a polyester to form a low colour blow moulded bottle (claims 5 and 6).

The characterising part of claims 1 and 2 were correspondingly amended as follows (the differences compared to the granted claims being indicated in bold by the board):

Claim 1:

"…characterised in that a major proportion amounting to 80 to 100 wt% of said polyester (a) consists of one or more polyesters obtained by direct condensation of the acid form of the dicarboxylic acid component (a)(1), and a minor proportion amounting to 0 to 20 wt% of the polyester (a) consists of polyester obtained by ester interchange using an ester form of the dicarboxylic acid component (a)(1) so that the bottle has less sidewall color as measured in b* units than a corresponding bottle formed from a corresponding polyester blend which is formed from polyester (a) the major proportion of which consists of a polyester obtained by ester interchange using the ester form of the dicarboxylic acid component (a)(1) and the minor proportion of which consists of polyester obtained by direct condensation of the acid form of component (a)(1)".

Claim 2:

"…wherein said base polyester (a) consists of one or more polyesters obtained by direct condensation of the acid form of the dicarboxylic acid component (a)(1), so that the bottle has less sidewall color as measured in b* units than a corresponding bottle formed from a corresponding polyester blend formed from a base polyester (a) which consists of polyester obtained by ester interchange using an ester form of the acid component (a)(1)."

Corresponding amendments were made to the characterizing parts of independent claims 3 and 4 and independent claims 5 and 6.

The dependent claims 7-20 corresponded to those of the main request.

The former dependent claims 21 and 22 were deleted.

According to the decision under appeal the independent claim 1 according to the main request did not meet the requirements of Article 83 EPC. The opposition division noted that the feature of comparing the b* values between compositions made by direct esterification and transesterification had been introduced during proceedings before the examining division. The opposition division held that the fact that the claim defined neither the method by which the b* value (yellowness) was determined nor the form of the sample upon which the determination was carried out represented a defect pursuant to Article 83 EPC. In particular, it was held that there were several standard methods for measuring the yellow colour, and it was not correct that any b* test could be applied to determine the relative values of the level of the yellow colour with the same result regardless of the form of the tested specimen. Thus it was held that the patent did not provide sufficient information to enable the skilled person to define whether a polyester blend compositions satisfied the b* requirement of claim 1 and hence that the patent did not disclose the invention in a manner sufficiently clear and complete, contrary to the requirements of Article 83 EPC.

With regard to the auxiliary request in which the claims were directed to blow-moulded bottles, it was held that the requirements of Articles 54, 83 and 123(2) EPC were satisfied. An inventive step was however denied. Accordingly the patent was revoked.

IV. On 4 March 2004, the patentee filed an appeal against this decision, the appeal fee being paid on the same day.

It was requested that the decision under appeal be set aside and the patent maintained on the basis of the set of claims according to the auxiliary request on which the decision of the opposition division had been based. The appellant indicated that it might wish to file further requests according to the circumstances of the appeal proceedings.

V. The statement of grounds of appeal was filed on 30 April 2004.

The request as set out in the notice of appeal was reiterated. An auxiliary request for oral proceedings was made.

Arguments in support of inventive step were presented.

VI. In a rejoinder dated 17 September 2004 the former opponent II (hereinafter "respondent II") notified a change of name to "Invista Technologies S.à.r.l".

Rejection of the appeal was requested.

The respondent stated that objections under Article 123(2) EPC with respect to the claims of the main request were not raised.

Objections pursuant to Articles 54, 56 and 83 EPC were raised.

VII. The board issued, on 26 July 2006, a summons to attend oral proceedings.

(a) In the accompanying communication, the preliminary, provisional view was expressed that certain of the amendments made did not meet the requirements of Article 123(2) EPC.

(i) With respect to claim 1, it was noted that the characterising feature that the bottle had a sidewall colour less than that of a "corresponding bottle" formed from a "corresponding polyester blend" having (non-specified) "major" and "minor" amounts of polyester derived from transesterification and direct esterification respectively had not been in the application as originally filed, but had been introduced during examination proceedings.

The only disclosure in the application as originally filed concerning relative degrees of yellowness (page 4 lines 2-6 and 24-29) related either to prior art blends or constituted a general statement that colour was improved when the polyester was derived from the acid form rather than from the ester form of the acid. Neither of these passages however provided a disclosure of "corresponding polyester blends".

This feature appeared to be a generalisation from certain of the examples. The evidence of the examples was however that compositions derived from polyesters obtained by direct esterification were not in all cases superior in terms of colour to those obtained by transesterification.

(ii) Further objections pursuant to Article 123(2) EPC were raised, inter alia against the definition in claim 1 of a "blow molded bottle", since the application as originally filed referred either generally to "bottle(s)" or more specifically to "injection stretch blow-molded bottles", or "extrusion blow-molded bottles". Compared to these disclosures "blow-molded bottle" constituted an inadmissible intermediate definition.

(b) It was considered that the wording of the claims, notwithstanding the defects in respect of Article 123(2) EPC, did not clearly specify the basis for the comparison upon which the colour was to be determined. In particular, the claims failed to specify which percentage of the constituents formed the "major" and "minor" proportion of the second polyester blend recited in the claims. Further since the evidence of the examples was that the colour of the blends did not yield unambiguous information about the composition thereof and in particular the manner in which they had been obtained this feature could not serve to characterise the bottles.

VIII. Together with a letter of 6 September 2006 the appellant submitted three sets of claims as a main and first and second auxiliary requests, and presented arguments in support of the admissibility of these with respect to Articles 54, 56, 83 and 123(2) EPC. These arguments relied, inter alia on a Table providing a reorganised presentation of the examples such that compositions with identical amounts of polyamide were adjacent to each other.

IX. With a letter dated 7 September 2006 the respondent Ball Packaging Europe Holding GmbH & Co.KG, the legal successor to Schmalbach-Lubeca AG (former Opponent I) stated that it was not interested in pursuing the present appeal proceedings and that the proceedings could thus be terminated.

X. With a letter dated 14 September 2006 the respondent Invista Technologies S.à.r.l. withdrew its opposition.

XI. Oral proceedings took place before the board on 6 October 2006, attended only by the appellant.

At the beginning of the oral proceedings the appellant maintained the main and second auxiliary requests but reversed their order so that the second auxiliary request became the final main request and the previous main request became the final second auxiliary request. The first auxiliary request was withdrawn and an amended set of 17 claims designated as the third auxiliary request submitted (see XII below).

XII. The submissions of the appellant with respect to the admissibility of the claims according to the resulting main, and second and third auxiliary requests pursuant to Article 123(2) EPC in the letter of 6 September 2006 and at the oral proceedings held on 6 October 2006 can be summarised as follows.

(a) Common to all requests were amendments to take account inter alia of the objection relating to the manner in which the bottle was produced (see VII.(a)(ii) above) by introducing the wording:

"injection stretch blow-molded bottle or extrusion blow-molded bottle"

into the preamble of the independent claims. The appellant submitted that this wording was to be found at page 11, lines 26 and 27 of the application as originally filed.

(b) Main request

Claim 1 of the main request read as follows, the differences compared to the characterizing part of the main request submitted together with the statement of grounds of appeal, corresponding to the auxiliary request underlying the decision under appeal (sections III, IV, V above), being indicated in bold by the board:

"1. A low color injection stretch blow-molded bottle or extrusion blow-molded bottle formed from a polyester blend composition comprising:

(a) 98.0 to 99.95 weight percent of polyester formed from:

(1) a dicarboxylic acid component comprising repeat units consisting of at least 85 mole percent of terephthalic acid, naphthalenedicarboxylic acid and mixtures thereof; and

(2) a diol component comprising repeat units from at least 85 mole percent ethylene glycol, based on 100 mole percent dicarboxylic acid and 100 mole percent diol,

(b) 2.0 to 0.05 weight percent of a polyamide; wherein the combined weight percents of (a) and (b) total 100 percent,

characterised in that a major proportion amounting to 80 to 100 wt% of said polyester (a) consists of one or more polyesters obtained by direct condensation of the acid form of the dicarboxylic acid component (a)(1), and a minor proportion amounting to 0 to 20 wt% of the polyester (a) consists of polyester obtained by ester interchange using an ester form of the dicarboxylic acid component (a)(1), so that the bottle has less sidewall color as measured in b* units than a corresponding bottle formed from a corresponding polyester blend in which the amount of acid form of the dicarboxylic acid component (a)(1) in the polyester has been replaced with the ester form of the dicarboxylic acid component and the polyester is thereby obtained by ester interchange."

Corresponding amendments were made to claims 3 and 5.

Claims 2, 4 and 6 of the main request were identical to claims 2, 4 and 6 as submitted with the statement of grounds of appeal with the exception of being amended in the same manner as claims 1, 3 and 5 in respect of the specification of the manner in which the bottles were moulded.

The remaining claims 7-20 are not of importance for this decision and will not be considered in further detail.

In respect of the admissibility of the amendments made to claim 1, the appellant submitted as follows:

(i) The basis for the feature "so that the bottle has less sidewall colour", was to be found in Table 1, page 16 and page 15 lines 31 and 32 of the application as originally filed.

(ii) It was acknowledged that the comparative feature relating to "corresponding" bottles and "corresponding" polyester blends, was not to be found explicitly in the application as originally filed. This feature meant that only the change specified was made, i.e. for the purposes of the comparison the 80-100 wt% of polyester prepared by direct esterification was replaced in its entirety by one prepared by transesterification and that all other features of the composition remained the same. This was derivable from claim 18 and page 11, lines 15-20 of the application as originally filed. It was submitted to be clear from the disclosure at page 4 line 3-6 of the application as originally filed that when the polyester was obtained by direct esterification from the free acid form better colour was obtained in the bottles than when it was obtained by transesterification from the ester form. The description clearly disclosed a direct comparison between these two alternatives - there was no reference to any other compositions.

(iii) The data of Table 1 of the patent in suit allowed the skilled person to compare the sidewall colour of bottles according to the invention with that of bottles wherein the acid form of the dicarboxylic acid component in the polyester had been replaced with the ester form such that the polyester had been obtained by ester interchange rather than by direct esterification. From these data the skilled person would have been able immediately and unambiguously to derive the subject matter of the amendment to claims 1, 3 and 5 from the application as originally filed. In presenting this argument the appellant relied not on Table 1 in the form in which it was present in the application as originally filed or in the granted patent, but on a subset of the data from said Table 1 presented in a modified - reordered - form in which the examples to which the appellant referred were in direct juxtaposition.

(iv) With regard to the question of whether an improvement in colour was obtained in all cases (cf. VII.(a)(i) final section), the appellant submitted that the correct comparison was between blends which were identical in all respects except for the source of the polyester. In making the comparison it was necessary to take into account also the polyamide portion such that blends with the same amount of polyamide were compared. This was derivable from the bottom of page 4 of the application.

The Table provided specific examples of the comparison indicated at the bottom of page 4 and demonstrated than in a direct comparison consistently better colour was obtained when direct esterification was used. The appellant emphasised however that it was not being argued that there was a consistent improvement in the level of acetaldehyde generated.

(c) Second auxiliary request

Claim 1 of the second auxiliary request read as follows, the differences compared to the main request (section XII.(b) above) being indicated in bold:

"1. A low color injection stretch blow-molded bottle or extrusion blow-molded bottle formed from a polyester blend composition comprising:

(a) 98.0 to 99.95 weight percent of polyester formed from:

(1) a dicarboxylic acid component comprising repeat units consisting of at least 85 mole percent of terephthalic acid, naphthalenedicarboxylic acid and mixtures thereof; and

(2) a diol component comprising repeat units from at least 85 mole percent ethylene glycol, based on 100 mole percent dicarboxylic acid and 100 mole percent diol,

(b) 2.0 to 0.05 weight percent of a polyamide; wherein the combined weight percents of (a) and (b) total 100 percent,

characterised in that a major proportion amounting to 80 to 100 wt% of said polyester (a) consists of one or more polyesters obtained by direct condensation of the acid form of the dicarboxylic acid component (a)(1), and a minor proportion amounting to 0 to 20 wt% of the polyester (a) consists of polyester obtained by ester interchange using an ester form of the dicarboxylic acid component (a)(1), so that the bottle has less sidewall color as measured in b* units than a corresponding bottle formed from a corresponding polyester blend which is formed from polyester (a) wherein the major proportion amounting to 80 to 100 wt% consists of polyester obtained by ester interchange using the ester form of the dicarboxylic acid component (a)(1) and the minor proportion amounting to 0 to 20 wt% consists of polyester obtained by direct condensation of the acid form of component (a)(1)."

Corresponding amendments were made to the wording of claims 3 and 5.

Claims 2, 4, 6 and 7-20 of the second auxiliary request were identical to the correspondingly numbered claims of the main request.

(i) The appellant submitted that this request differed from the main request in the final part of the characterising portion in that the major and minor proportions in the comparative composition were quantified. The comparison was between two compositions in which the proportions of polyesters obtained by the two routes was "symmetrical" such that in the comparative composition the proportions of polyester derived by direct esterification and transesterification respectively were reversed. Thus, for example, an inventive bottle prepared from a blend of 90 wt% directly esterified polyester and 10 wt% transesterified polyester would be compared with one prepared from a blend wherein the proportions were 10 wt% and 90 wt% respectively. It was emphasised that the claim was not to be interpreted as defining a comparison between two compositions the constitution of which could vary independently of each other within the ranges defined by corresponding parts of the claim.

This was derivable from claim 18 as originally filed.

(ii) It was further submitted that the person skilled in the art would be able to derive immediately and unambiguously the subject matter from the specification as filed. From original claim 18 and examples 9-14 the skilled person would understand that from 80 to 100% of the polyester in the bottles of the invention could be obtained by direct condensation of the acid form of the dicarboxylic acid component. The above cited passage at page 4 lines 3-6 of the originally filed description was also referred to. Thus the skilled person would immediately and unambiguously conclude that a bottle made from a polymer blend using 80 to 100% of the polyester obtained by direct esterification would have a lower colour than a corresponding bottle wherein 80-100% of the polyester had been obtained by transesterification. Further upon examining Table 1 the skilled person would be able to compare the sidewall colour of bottles of the invention with corresponding bottles wherein, instead of the acid form the ester form had been used in the preparation of the polyester of the polymer blend.

(iii) As in the case of the main request (cf. XII.(b)(iv) above) the need to compare bottles made out of blends having a similar content of polyamide was emphasised.

(d) Third auxiliary request

Following the discussion at the oral proceedings of the main and second auxiliary requests, the Appellant submitted a further set of 17 claims as the third auxiliary request. Claim 1 of this request read as follows, the differences in the characterizing part as compared to claim 2 of the main request (sections III, IV and XII.(b) above) being indicated in bold by the board:

"1. A low color injection stretch blow-molded bottle or extrusion blow-molded bottle formed from a polyester blend composition comprising:

(a) 80 to 99 weight% of base polyester which comprises:

(1) a dicarboxylic acid component comprising repeat units from at least 85 mole percent of: terephthalic acid, naphthalenedicarboxylic acid and mixtures thereof; and

(2) a diol component comprising repeat units from at least 85 mole percent ethylene glycol, based on 100 mole percent dicarboxylic acid and 100 mole percent diol; and

(b) 1 to 20 weight % of a concentrate comprising:

(1) 1 to 99 weight % of a carrier resin comprising a dicarboxylic acid component comprising repeat units from at least 60 mole percent aromatic dicarboxylic acid selected from the group consisting of terephthalic acid, naphthalenedicarboxylic acid and mixtures thereof, and a diol component comprising repeat units from at least 50 mole percent ethylene glycol, based on 100 mole percent dicarboxylic acid and 100 mole percent diol; and

(2) 1 to 99 weight % of a polyamide which displays a melting point below the melting point of said carrier resin,

wherein said base polyester (a) consists of one or more polyesters obtained by direct condensation of the acid form of the dicarboxylic acid component (a)(1), so that the bottle has less sidewall color as measured in b* units than a corresponding bottle formed from a corresponding polyester blend in which said acid component (a)(1) is instead derived from an ester form thereof".

Former claims 4 and 6, correspondingly amended, became claims 2 and 3. The remaining claims were renumbered 4 to 17 and the dependencies amended where necessary, but were otherwise identical to claims 7 to 20 of the main request (section XII.(b) above).

(i) The appellant submitted that the request had been filed in response to new objections that had been raised in respect of the embodiment according to the second auxiliary request, i.e. that in which - as submitted by the appellant - the proportions of polyester derived from the direct esterification route and the transesterification route were reversed.

The amendments constituted a change in wording but not in the sense and hence there was no disadvantage for the opponents.

(ii) With respect to the term "derived" and an observation of the board that this could encompass the steps of starting from the ester form of the acid, deesterifying and using the so obtained free acid in the polyesterification, the appellant submitted that this was to be understood as meaning that the polyester was produced by transesterification.

(iii) From page 4 of the application as originally filed it was apparent that the correct comparison was with polyesters produced by the transesterification route and that the thrust of the invention was that the base polyester be derived from the free acid form. The examples supported the advantage of employing an acid derived polyester as the base resin. Only preferably was the polyester employed for the concentrate also obtained by direct esterification. The data of the examples showed that the improvement resulted from the nature of the base resin. The data showed that in the case that a polyester derived from transesterification was employed for the concentrate, then employing a direct esterification polyester as the base led to an improvement in both colour and AA content (examples 3 and 12). When the base polyester was derived from direct esterification then changing the concentrate resin from a transesterification polyester to a direct esterification polyester resulted in an improvement in acetaldehyde emission, but a poorer colour, i.e. a higher b* value (examples 9 and 12). The results were consistent in that there was in all cases an improvement in colour when a direct esterification polyester rather than a transesterification polyester was employed as the base resin.

(iv) The polyester employed for the concentrate was not taken into account when making the comparison. The examples demonstrated that an improvement was obtained regardless of how the resin of the concentrate was produced.

However, in the case that additionally the carrier resin in the polyamide concentrate was also derived from the acid form there was a further improvement in respect of the AA content. The fact that according to one pair of examples the colour was worse in the case of the "inventive" composition did not constitute a discrepancy.

(e) With respect to the further observation of the board (see section VII.b above) that certain inventive bottles exhibited worse colour than some of the comparative bottles it was submitted that the data in Table 1 provided support for the feature of superior colour. A further reorganised version of this Table was provided, incorporating in this case all of the data. This was in order to "allow a better comparison of the examples", namely to facilitate comparison between bottles wherein the only difference in the corresponding bottle was whether the polyesters of the base resin were derived from the acid or the ester form of the dicarboxylic acid component. The need to carry out comparisons between compositions matched for polyamide content was reiterated (cf. XII.(b)(iv) above).

XIII. The appellant requested that the decision under appeal be set aside and that the patent be maintained on the basis of the main request (claims 1-20) filed at oral proceedings (formerly auxiliary request 2 filed with letter dated 6 September 2006), or in the alternative, on the basis of auxiliary request 2 (claims 1 to 20) filed at oral proceedings (formerly main request filed with letter dated 6 September 2006), or auxiliary request 3 (claims 1 to 17) filed at oral proceedings.

Reasons for the Decision

1. The appeal is admissible.

2. The patent in suit; the application as originally filed

2.1 The patent in suit relates to low colour polyester blend compositions.

The blend according to independent claim 1 of the patent as granted comprises 98.0 to 99.95 wt% of polyester and 2.0 to 0.05 wt% of a polyamide as set out in section I above.

This claim specifies that a major proportion (defined as 80 to 100 wt%) of the polyester consists of polyester(s) obtained by direct condensation, a minor proportion (defined as 0-20 wt%) being obtained by ester interchange using an ester form of the dicarboxylic acid component (transesterification). The claim further specifies that the colour of the blend is less than that of a blend wherein the major proportion of the polyester is obtained by ester interchange and the minor proportion is obtained by direct condensation. Independent claim 2 - the wording of which is also reported in section I above - defines a second aspect wherein the polyamide is added in the form of a concentrate in a polyester carrier resin. The blend composition according to claim 2 comprises 80-99 wt% of polyester- designated the "base" polyester and 1 to 20 wt% of a concentrate consisting of 1 to 99 wt% of a resin designated the "carrier" and 1 to 99 wt% of a polyamide.

Claim 2 specifies that the base polyester has repeat units from at least 85 mole % of terephthalic acid, naphthalene dicarboxylic acid and mixtures thereof, i.e. has a major proportion of polyester obtained by direct esterification and further specifies that the polyester blend has less colour than one formed from a base polyester which consists of a polyester obtained by ester interchange. Thus the subject matter claimed in the patent in suit as granted falls into two parts or aspects: claim 1 on the one hand which defines a blend composition having a polyester component (a) and a polyamide component (b), the colour comparison being made with a composition wherein all the polyester present in the composition is taken into consideration and claim 2 on the other hand in which only the nature of the polyester forming the "base" resin is taken into account in making the colour comparison. In claim 1 furthermore, the nature of the comparison requires an "inversion" or switch (cf. XII.(c)(i)) in that the polyester blend comprising a major proportion (80 to 100 wt%) of polyester obtained by direct condensation and a minor proportion (0 to 20 wt%) of polyester obtained by transesterification is to have less colour measured in b* units than the corresponding polyester blend having a major proportion of polyester obtained by transesterification and a minor proportion of polyester obtained by direct esterification.

In claim 2 in contrast, there is no "inversion" involved in the comparison since the "base" polyester having a major proportion of polyester obtained by direct esterification is compared with a corresponding base polyester blend which consists of polyester obtained by transesterification. In other words, the latter comparison involves replacing the base polyester resin having a major proportion thereof obtained by direct esterification and a minor proportion obtained by transesterification by a base polyester obtained to the extent of 100% by transesterification. The polyester present in the concentrate - which can constitute up to ca 20% of the total composition i.e. the "carrier" resin - is moreover not specified in this definition and thus forms no part of the colour comparison.

Claims 3 and 4 relate to a process for forming a low colour polyester blend composition having the features of claims 1 and 2 respectively, while claims 5 and 6 relate to the use of a polyester blend composition defined according to claims 1 and 2 respectively.

The originally filed application

2.2 The originally filed claims

Claim 1 as originally filed read as follows:

"1. Polyester compositions having improved flavor retaining properties and low color, comprising:

(a) 98.0 to 99.95 weight percent of a polyester which comprises

(1) a dicarboxylic acid component comprising repeat units from at least 85 mole percent aromatic dicarboxylic acid selected from the group consisting of terephthalic acid, naphthalenedicarboxylic acid and mixtures thereof, wherein said acid is derived from terephthalic acid or naphthalenedicarboxylic acid respectively; and

(2) a diol component comprising repeat units from at least 85 mole percent ethylene glycol, based on 100 mole percent dicarboxylic acid and 100 mole percent diol; and

(b) 2.0 to 0.05 weight percent of a polyamide; wherein the combined weight percents of (a) and (b) total 100 percent."

Originally filed claim 18, dependent from claim 1 specified that "up to 20 weight % of said polyester is derived from an ester form of said terephthalic acid or naphthalenedicarboxylic acid".

Claim 20 as originally filed related to a process for forming a blend comprising blending a base polyester together with a concentrate comprising a polyester carrier resin and a polyamide in the proportions specified in granted claim 2.

None of the claims of the application as originally filed contained any comparative definition of the colour, let alone the definitions referred to above in relation to the granted claims 1 and 2.

2.3 The originally filed description

According to page 4 of the application as originally filed, (corresponding to paragraphs [0007]-[0009] of the patent in suit) the polyester/polyamide blends have surprising low AA (acetaldehyde) and low colour (page 4, lines 2 and 3). In particular, it is stated with respect to the nature of the polyester in general, i.e. the aspect according to granted claim 1:

- "By using the acid form of the acid component of the polyester instead of the ester form, the resulting polymer blends display lower AA and color than the previously produced blends" (original application page 4 lines 3-6, patent paragraph [0007], 2nd sentence).

- "When the acid component of the polyester is derived from the acid form of the acid component the amount of acetaldehyde (AA) generated is less that (sic) which is typically generated when the acid component is derived from the ester form and the color is much better" (original application page 4 lines 24-29, patent paragraph [0009], first sentence).

The influence of the manner in which the polyester is obtained is further discussed at page 13, lines 6-12 of the original application (corresponding to paragraph [0035] of the granted patent) where it is stated with respect in particular to the addition of the polyamide:

"It has been surprisingly found that by using the acid form of the acid component of the polyester and adding a polyamide, the resulting polyesters display surprisingly low AA content and good color. This was particularly surprising as the addition of polyamide is known to deleteriously effect the color of the resultant polyesters".

With respect specifically to the aspect according to which a concentrate is employed, i.e. that of originally filed claim 20 and granted claim 2, it is taught in the original description that:

- "The desired AA and color properties are achieved when the acid based polyesters are used as the base polymer" (page 4 lines 10 and 11, paragraph [0007], 4th sentence).

- "It should also be understood that the base resin may contain small amounts of the ester form of the acid component, so long as the total amount of the ester form of the polyester/polyamide blend does not exceed 20 weight %, and preferably not more than 10 weight%" (original application, page 11 lines 15-20, paragraph [0030], 2nd part of the granted patent).

At page 12, lines 11-13 of the original application corresponding to paragraph [0033] of the patent, it is stated that the compositions of the invention display better colour than those disclosed in a specified US patent (US 5 258 233). This patent is among those referred to in the passage bridging pages 1 and 2 of the application as originally filed, corresponding to paragraph [0003] of the patent where it is further stated that the AA and/or colour levels of the products (of the cited patents) "are still undesirable for certain applications".

2.4 The examples.

The examples of the originally filed application and the granted patent demonstrate bottles formed from polyester compositions wherein the polyamide is added in the form of a concentrate in a polyester "carrier", i.e. corresponding to the embodiment of originally filed claim 20 and granted claim 2. The base resin is blended with the concentrate in a ratio of 50:1 meaning that the final compositions contain ca 98% of the base polyester and ca 2% of the concentrate polyester. The sidewall colour of the bottles is measured.

The examples are summarised in Table 1 of the application and patent, which is reproduced below:

FORMULA/TABLE/GRAPHIC

Inspection of the Table reveals that examples are provided wherein:

- Both the base and concentrate are produced by transesterification (examples 3-5)

- Both the base and concentrate are produced by direct esterification (examples 9-11)

- The base is produced by transesterification and the concentrate by direct esterification (examples 6-8)

- The base is produced by direct esterification and the concentrate produced by transesterification (examples 12-14).

3. Main request - Article 123(2) EPC.

The wording of claim 1 of the main request is reported in paragraph XII.(b) above.

This claim does not specify the manner in which the polyamide is introduced into the polyester, i.e. does not require that this be added in the form of a concentrate.

3.1 The subject matter "low color injection stretch blow-molded or extrusion blow-molded bottle formed from a polyester blend" is disclosed in the application as originally filed at page 4 lines 3-6 (see paragraph 2.3 above) and at page 11 lines 26 and 27 (relating to the manner in which the bottles are formed).

3.2 The constitution of the components (a) and (b) of the polyester blend is to be found in claim 1 as originally filed, recited in section 2.2 above.

3.3 The feature relating to the specified "major" proportion of 80-100 wt% and "minor" proportion of 0-20 wt% of the polyesters present is disclosed in claim 18 as originally filed as explained in paragraph 2.2 above.

3.4 The final feature of claim 1 of the main request, reported in section XII.(b), above defines a comparison of the sidewall colour, in terms of b* units with a "corresponding" bottle formed from a "corresponding" polyester blend in which "the amount of acid form of the dicarboxylic acid component (a)(1) in the polyester has been replaced with the ester form of the dicarboxylic acid component and the polyester is thereby obtained by ester interchange".

3.4.1 As clearly acknowledged by the appellant at the oral proceedings (XII.(b)(ii) above), there is no explicit mention in the application as originally filed of any comparison between "corresponding" bottles prepared from "corresponding" polyester blends.

3.4.2 It is therefore necessary to examine whether, despite the lack of an explicit, literal basis for this comparison, said comparison is nevertheless derivable implicitly from the disclosure of the application as originally filed.

The question of implicit disclosure in the context of the admissibility of amendments pursuant to Article 123(2) EPC was considered in decision T 823/96 (28 January 1997, not published in the OJ EPO). In paragraph 4.5 of the reasons it was observed that "implicit disclosure" should not be construed as meaning matter that does not belong to the content of the technical information provided by a document but may be rendered obvious on the basis of that content (emphasis of decision cited). Rather the board in that case considered that the term "implicit disclosure" related solely to matter which is not explicitly mentioned but is a clear and unambiguous consequence of what is explicitly mentioned.

3.4.3 The comparison required by the characterizing feature of claim 1 of the main request is between

- a bottle formed from a polyester/polyamide blend in which 80 to 100 wt% of the polyester is formed by direct esterification employing the acid form of the acid component and 0 to 20 wt% of which is formed from a polyester obtained by ester interchange (transesterification)

and

- a bottle formed from a polyester blend in which the polyester is entirely obtained by trans ester ification.

According to the characterising feature of claim 1, in the case that the polyester component of the inventive composition is formed entirely by direct esterification, then for the purposes of the comparison this will be replaced entirely by polyester formed by transesterification. Alternatively, in the case where the polyester component of the inventive composition is prepared from a mixture of polyester formed by direct esterification and polyester formed by transesterification then that portion of the polyester formed by direct esterification will be replaced by one formed from transesterification, while the portion of polyester prepared by transesterification will be retained, meaning that in this case as well the comparative composition is one wherein the polyester is in its entirety formed by transesterification.

3.4.4 Considering the disclosure of the description of the application as originally filed, the passage at page 4, lines 3-6, recited in section 2.3 above, constitutes a comparative assessment of the colour of blends. However this comparison is with non-specified "previously produced blends" and does not identify the technical feature(s) by which the blends of the invention differ from said "previously produced blends". Further, this passage does not relate to the sidewalls of extrusion or injection stretch blow-moulded bottles. Thus the comparison disclosed at page 4 lines 3-6 of the originally filed application is of a more general nature than the comparison defined according to claim 1 of the main request.

The disclosure at page 4 lines 24-29 also quoted in paragraph 2.3 above appears to disclose - in general terms - an improvement in properties, specifically AA content and colour when employing the polyester obtained by direct esterification rather than that obtained by transesterification. This passage discloses that there is a reduction in AA content as compared to that which is "typically generated" when the acid component is "derived" from the ester form.

Neither the nature of the "typical" compositions "derived" from the ester form of the acid component on which this statement is based nor the precise nature of the difference between the compositions which are to be compared is however elucidated. This passage therefore does not reflect the - quasi mathematical - relationship involving complete replacement of one component by another to which the claim is directed. In particular, this passage does not state that the improvement is observed with respect to polyester blends or compositions wherein the polyester present in the composition which had been obtained by direct esterification is replaced entirely with polyester obtained by transesterification, as required by the characterising feature of claim 1.

The appellant also referred to claim 18 and page 11 lines 15 to 20 as providing a basis for this comparison (see XII.(b)(ii) above).

As noted above (section 2.3) the passage at page 11 relates not to the aspect of claim 1 but to the different aspect set out in claim 2 in which a base resin and a concentrate are defined. The two aspects are however presented completely independently from one another, with the information from the "base/concentrate" aspect of claim 2 thus not being directly applicable to the "simple" compositional aspect of claim 1. Consequently, the cited disclosure at page 11 cannot provide a basis for the subject matter of claim 1 of the main request.

Whilst originally filed claim 18 (see section 2.2) is dependent on originally filed claim 1 and does specify that up to 20 weight% of the polyester be derived from an ester form of the acid it does not, however, define any comparative assessment of the properties of the blends or bottles prepared therefrom and accordingly also does not provide a basis for the comparison defined in claim 1 of the main request.

3.4.5 Regarding the examples, it is first noted that although the methodology employed, in which the polyamide is added as a concentrate in a polyester resin is defined in claim 2, the resulting compositions, implicitly disclosed by the examples, are nevertheless relevant to the subject matter of claim 1 insofar as the proportions of components exemplified in the examples are such as to yield compositions within the ranges permitted by this claim.

Examination of the examples of the application as originally filed, as summarised in Table 1, reproduced above reveals that the improvement required by the comparison defined in claim 1 is not in fact obtained in all cases where the polyester present is obtained by direct esterification rather than by transesterification is (see discussion in section 2.4 above).

In particular it is noted that:

- The bottle of Example 11, obtained from a polyester blend wherein both the base and the concentrate are obtained by direct esterification exhibits worse sidewall colour (higher b* value) than examples 3, 4, 6 and 7 in which the base resin is obtained by the transesterification route and the concentrate is derived either by transesterification (examples 3, 4) or by direct esterification (examples 6, 7).

- The bottle of example 14, obtained from a polyester blend wherein the base is obtained by the direct esterification route and the concentrate is obtained by transesterification exhibits poorer sidewall colour than bottles in which both polyesters are obtained by transesterification (examples 3 and 4) and also poorer colour than bottles in which the base resin is obtained by transesterification and the concentrate is obtained by direct esterification (examples 6 and 7).

The appellant submitted that in making the comparison it was necessary to consider blends which were identical in all respects except for the nature of the polyester. In particular, it was necessary to take account of the polyamide component and make comparisons between compositions with the same content of polyamide.

Thus, according to this submission, and as set out in the restructured Table presented in the submission of 6 September 2006 (see section VIII above) the correct comparisons to be made are between (comparative) examples 3, 4, 5, 6, 7 and 8 with, respectively, inventive examples 12, 13, 14, 9, 10 and 11.

According to the submission of the appellant, this was apparent from the passage at page 4, lines 24-29 cited in section 2.3 above (see paragraph XII.(b)(iv) above). While this submission, introducing a further condition for the comparison, is entirely consistent with one possible interpretation of the results reported in the Table, it is neither explicitly nor even implicitly derivable from the aforementioned passages at pages 1, 4 or 11 of the original description. In particular the passage on page 4, lines 24-29 referred to by the appellant makes no mention that the improvement in colour and acetaldehyde content is contingent on any factor other than the manner in which the polyester had been obtained. It is not derivable from these passages that this improvement is subject to the condition that the content of polyamide be identical in the compositions compared. While it would indeed be consistent with these passages to apply the interpretation preferred by the appellant, it is equally consistent to interpret these passages as indicating that the source of the polyester is the dominating factor that - independently of any other variation including the amount of polyamide present - results in improved colour. This second interpretation is supported by the passage at page 13, cited in section 2.3, above which teaches that use of the directly esterified polyester leads to improved colour despite the presence of the polyamide - which normally deleteriously affects the colour. There is no indication that this finding is subject to any further condition, for example that the compositions on which this comparative assessment is based be matched for the amount and/or type of polyamide.

It is also the case that this highly specific comparison requiring that the content of polyamide in the "corresponding" compositions be identical to that of the composition defined as being the subject matter of the claim is not in fact specified in claim 1 of the main request.

3.4.6 The claim therefore defines a comparison that is consistent with but more specific than that set out in the most general part of the description in that it specifies the proportions of polyester derived by direct esterification and transesterification in the compositions to be compared, in particular requiring that the comparative composition be one in which the entire quantity of polyester is derived from transesterification. However, the comparison in the claim is also more general than that which, according to the submission of the appellant, is derivable from the Table and examples in the application as originally filed, due to the fact that the condition that the content of polyamide be identical in the compositions compared is absent. Although the comparison imputed is fully consistent with a possible interpretation of the data in the Table in combination with the more general aspects of the description, it is inconsistent when taking into account instead a more specific part of the description, namely the passage at page 13, lines 6-12 cited in section 2.3 above.

Hence the claim defines a condition which is intermediate between the very general statements of the description and a highly specific and precise comparison based on one possible interpretation of the data presented in the Table but is not unambiguously supported - even implicitly - by either.

3.5 Therefore claim 1 of the main request does not meet the requirements of Article 123(2) EPC and the request must be refused for this reason.

4. Second auxiliary request - Article 123(2) EPC

4.1 Claim 1 of the second auxiliary request differs from claim 1 of the main request in that the characterising portion requires that the sidewall colour, measured in b* units of a bottle prepared from a polyester blend formed from 80 to 100 wt % of polyester prepared by direct esterification and 0 to 20 wt% of a polyester prepared by transesterification be less than that of a bottle prepared from a polyester blend formed from 80 to 100 wt% of a polyester prepared by transesterification and 0 to 20 wt% of a polyester prepared by direct esterification.

Thus the comparison required according to the second auxiliary request corresponds to the situation in claim 1 of the patent in suit as granted for the "corresponding composition" but in which the "major" and "minor" proportions in the comparative polyester are now quantified. In other words, there is some kind of interdependency or inversion between the "major" and "minor" proportions of what is being compared. This is in contrast to the definition in claim 1 of the main request in which the comparative composition is always formed entirely from polyester obtained by transesterification.

4.2 To the extent that the description of the application as originally filed discloses any comparison (page 4 lines 24-29 cited in section 2.3, above) it is evident that the comparison is between polyesters prepared entirely by direct esterification and those prepared entirely by transesterification.

4.3 In this connection however, the general disclosure on page 4 has been found to define a comparison with which claim 1 of the main request is consistent (section 3.4.6 above). Clearly the same passage cannot at the same time provide support for the fundamentally different kind of comparison that is set out in claim 1 of the second auxiliary request.

4.4 Although certain groups of examples might be regarded as showing such a "symmetrical" or even interdependent comparison (examples 3-5 and 12-14) these correspond only to the case of the "major" and "minor" proportions being 100% and 0% respectively. There is no disclosure in relation to the examples that the comparisons should generally be made between compositions having "symmetrical" or even interdependent amounts of polyester obtained via the two routes. Thus the examples do not support the contention of the appellant that necessarily the comparison will be as set out in claim 1 of the second auxiliary request.

Therefore the arguments advanced in support of the admissibility of claim 1 of the second auxiliary request proposed by the appellant are not supported by the original disclosure.

4.5 Accordingly the subject matter of independent claim 1 of the second auxiliary request does not meet the requirements of Article 123(2) EPC.

5. Third auxiliary request

5.1 The third auxiliary request, the wording of which is recited in section XII.(d), above, was introduced during the oral proceedings before the board. It is therefore necessary in a first step to decide whether this request is to be admitted to the procedure in particular taking account of the fact that the other party - the opponent - was not present at the oral proceedings.

5.1.1 Compared to the main request, the third auxiliary request is restricted by deletion of the aspect according to independent claims 1, 3 and 5. Thus the independent claims of this request relate solely to the aspect according to claim 2 of the main request whereby the structure of the composition is such that the polyamide is added in the form of a concentrate in a polyester, i.e. the second aspect (see section 2.1 above). The comparison defined in the claim is with respect to a composition in which the acid component of the base polyester (a)(1) is replaced by one derived from an ester form thereof, i.e. obtained by transesterification and the nature of the comparison is that given in claim 2 of the main request (see sections III, IV, and XII.b above). In particular the nature of the polyester employed for the concentrate is not specified in respect of this comparison; only the nature of the base polyester is defined.

The terms of the comparison, namely:

"…so that the bottle has less sidewall color as measured in b* units than a corresponding bottle formed from a corresponding polyester blend in which said acid component (a)(1) is instead derived from an ester form thereof"

are however amended (as indicated in bold) compared to claim 2 of the main request in order to align it more closely with the disclosure of page 4 (see sections XII.(d)(iii) and 2.3 above).

5.1.2 The appellant submitted at the oral proceedings that the third auxiliary request had been introduced in response to the objections raised in relation to the subject matter of the second auxiliary request (see XII.(d)(i)).

5.1.3 Regarding the absence of the opponent at the oral proceedings, it is noted that a party who elects not to attend oral proceedings cannot be taken by surprise by the fact that a patentee presents at oral proceedings amendments in order to address objections already raised. On the contrary, an absent party must reasonably expect that a patentee would try to overcome all objections (paragraph 7 of the reasons of T 133/92, 18 October 1994, not published in the OJ EPO, with reference to G 4/92, OJ EPO 1994, 149).

5.1.4 The third auxiliary request was accordingly admitted to the procedure.

5.2 Regarding the requirements of Article 123(2) EPC, the features relating to the aspect of adding the polyamide in the form of a concentrate in a polyester is disclosed in claim 20 and at page 10 line 31 to page 11 line 20 of the application as originally filed.

5.2.1 The characterising feature defining a comparison between corresponding bottles is, as noted in respect of the main request (section 3.4.1 above) not explicitly mentioned in the application as originally filed.

5.2.2 Considering the specific difference with respect to the characterising feature in the main request, namely that the comparison relates to bottles differing in the nature of the base resin, the concentrate being excluded from the comparison, the passages at page 4 lines 3-6 and 24-29 of the application as originally filed refer in general terms to an improvement when the acid component of the polyester composition is derived from the acid form of the acid component. At page 10, lines 32-34 it is disclosed that the concentrate carrier resin may be derived either from the acid or ester form of the acid component of the acid component of the polyester, the acid form being preferred. At page 11 lines 13-20, it is stated that the base resin may contain small amounts of the ester form of the acid component as long as the total amount of the ester form of the polyester/polyamide blend does not exceed 20 wt%, preferably not exceed 10 wt%.

These passages do not disclose, or even imply a comparison based exclusively on variations in the base resin as required by claim 1 of the third auxiliary request. Rather they indicate that the overriding factor is the total amount of transesterification derived polyester in the blend, with no differentiation as to the origin of this (base or concentrate).

5.2.3 With regard to the disclosure of the examples, it is noted that the subject matter of claim 1 of the third auxiliary request, in particular the comparison therein defined, as in the case of the main and second auxiliary request, is fully consistent with a possible reading of the data of Table 1. This requires that comparison is made between examples 9-11 and 6-8 for the case where the polyester in the concentrate is derived by direct esterification and examples 12-14 and 3-5 for the case that the concentrate is derived from transesterification.

However there is no indication either in the context of the examples, or the more general description that this is the comparison to be made, nor is this comparison derivable even implicitly from the application as originally filed.

5.2.4 As noted in the case of the claims of the main and second auxiliary requests (see above), the subject matter of the third auxiliary request represents a position intermediate between the disclosure of the description and one possible interpretation of the examples respectively. The interpretation invoked in the case of the third auxiliary request relies, as noted in section 5.2.3 above on one of two possible comparisons, namely between the examples 9-11 and 6-8 in one case and 3-5 and 12-14 in the other case. It is however conspicuous to the board that this set of comparisons requires a different constellation of the examples than that underlying the comparisons invoked by the appellant in the case of the main request (see section 3.4.5 above).

As noted in respect of the main request, the description provides only a basis for a more general comparison, while the data in Table 1 could be interpreted as indicating that further constraints are to be imposed when making a comparison, namely in respect of the polyamide content which constraints are however not explicitly set out in the application as originally filed or, indeed, defined in claim 1 of the third auxiliary request (see also sections 3.4.4 and 3.4.5 above).

5.3 Accordingly claim 1 of the third auxiliary request does not meet the requirements of Article 123(2) and the request must be refused for this reason.

ORDER

For these reasons it is decided that:

The appeal is dismissed.