T 0799/95 18-06-1998
Process for dimensionally stable polyester yarn
Akzo Nobel Fraser AG
Hoechst Aktiengesellschaft Zentrale Patentabteilung
Extension of protection conferred (no) - scope of protection not extended by the use of a different unit of measurement
Sufficiency of disclosure (no) - undue burden
I. European patent application No. 90 902 093.5, corresponding to PCT/US89/05081 (WO 90/07592) in the name of AlliedSignal Inc. (formerly Allied-Signal Inc.), which had been filed on 17 November 1989, claiming priority from a US application filed on 3. January 1989, resulted in the grant of European patent No. 452 405 on 16 September 1992 on the basis of 9. claims, independent Claim 1 reading as follows:
"1. A process for the production of a dimensionally stable drawn polyethylene terephthalate multifilament yarn having filaments of at least 2.5 denier per filament (2.8 dtexpf) comprising the steps of:
a) extruding a polyethylene terephthalate polymer melt through a spinnerette having a plurality of extrusion orifices to form filaments;
b) advancing the extruded multifilament yarn first through a delay zone then through a quenching zone to solidify the filaments in a controlled manner;
c) withdrawing the solidified multifilament yarn from the quenching zone at a desired spinning speed V (km/min); whereby the steps a) through c) are performed under conditions to form a partially-oriented multifilament yarn having an undrawn birefringence (Delta nu) of at least 0.020 and wherein Delta nu = RfV2.0IV2.4 where IV is the intrinsic viscosity of the undrawn yarn and is at least 0.80 and Rf = RrRe is at least 9.0 x 10-3, Rr being a constant (0
d) hot drawing the partially-oriented multifilament yarn."
Claims 2 to 9 were dependent on Claim 1.
II. Notices of Opposition were filed by
- Akzo Faser AG, now Akzo Nobel Faser AG (Opponent I) on 9 June 1993 and by
- Hoechst AG (Opponent II) on 15 June 1993
requesting revocation of the patent in its entirety, on the grounds of Article 100(a) and (b) EPC.
III. By its decision announced orally on 6 April 1995 and issued in writing on 25 July 1995 the Opposition Division revoked the patent.
(i) The appealed decision held that the claimed subject-matter was not barred from protection by Article 52(2)(a) EPC.
(ii) That decision furthermore held that the patent in suit met the requirements of Article 83 EPC, because (1) the temperature of measurement of the intrinsic viscosity of 25 C was within the implicit disclosure of the application, (2) in view of the information contained in the description the meaning of the parameter Rr was not such as to inhibit the invention to be carried out by a person skilled in the art, and (3) the examples in the patent specification could be repeated.
(iii) However, in the Opposition Division's opinion, Claim 1 of the Main Request did not relate to novel subject-matter because the claimed functional relationship between the undrawn birefringence nu, the spinning speed V, the intrinsic viscosity IV, the spinnerette capillary diameter D, the polymer flow rate Q and the quenching conditions was known from both documents D2: EP-A-169 415 and D5: EP-A-423 213.
(iv) Although novel because of the enhanced lower limit of the parameter Rf, the subject-matter of Claim 1 of the Auxiliary Request was not considered inventive over D2, because the modification of this factor was merely the result of routine experiments based on trial and error.
IV. Notice of Appeal against the above decision was filed by the Patentee (Appellant) on 25 September 1995. The fee for the appeal was paid on the same day and the Statement of Grounds of Appeal was submitted on 1. December 1995.
Further written submissions of the Appellant dated from 18. May 1998 and 5 June 1998.
Written submissions of the Respondent I (Opponent I) dated from 13 October 1995 and 10 June 1996; of the Respondent II (Opponent II) from 20 June 1996 and 14. May 1998.
V. In the course of the oral proceedings held on 18 June 1998 the Appellant filed a new Main and a new Auxiliary Request, both comprising a set of six claims each consisting of an independent Claim 1 and five claims dependent thereupon.
Claim 1 of the Main Request reads as follows:
"1. A process for the production of a dimensionally stable drawn polyethylene terephthalate, multifilament yarn having filaments of at least 2.5 denier per filament (2.8 dtexpf) comprising the steps of:
a) extruding a polyethylene terephthalate polymer melt through a spinnerette having a plurality of extrusion orifices to form filaments;
b) advancing the extruded, multifilament yarn first through a delay zone then through a quenching zone to solidify the filaments in a controlled manner;
c) withdrawing the resulting solidified-partially-oriented multifilament yarn from the quenching zone at a desired spinning speed V (km/min); and
d) hot drawing the partially-oriented multifilament yarn
characterised in that the conditions in steps a) to c) are selected to form a partially-oriented multifilament yarn having an undrawn birefringence (Delta nu) of at least 0.020. and wherein nu = RfV2.0IV2.4, where IV is the intrinsic viscosity of the undrawn yarn and is at least 0.80. and Rf = RrRe is at least 15 x 10-3, Rr being a constant (0
Claim 1 of the Auxiliary Request differs from the same claim of the Main Request only by the introduction of an upper limit of the intrinsic viscosity of the undrawn yarn of 0.95.
VI. The written and oral arguments of the Appellant may be summarized as follows:
(i) Claims 1 of both requests did not contravene Article 123(2) and (3) EPC, because the change of the measurement unit for the diameter "D" from "cm" (version as granted) to "inches" was based on the disclosure as originally filed, and because no new subject-matter was brought into the scope of the patent by this amendment;
(ii) the patent specification complied with the requirement of Article 100(b) EPC, because the mathematical equation in Claim 1, in particular the parameter Rr therein, was sufficiently supported by the description to enable the skilled person to carry out the claimed invention within its full claimed scope; one skilled in the art was aware of the possible conditions of treatment of the extruded filaments in the delay and quench zones, which determine Rr, and it needed only a few routine experiments to find out the values of the other variables of the mathematical equation, particularly the spinnerette diameter D and the polymer flow rate Q (which define the parameter Re, the actual "core" of the invention) in order to arrive at a satisfactory value of birefringence Delta nu.
The many arguments of the Appellant with respect to the issues of novelty and inventive step need not be referred to here, because, irrespective of the possible outcome of these issues, the Board upheld the conclusion of the first instance to revoke the patent albeit on the different ground of non-compliance with the requirement of Article 100(b) EPC.
VII. The arguments of the Respondents may be summarized as follows:
(i) the change in Claims 1 of both requests of the unit "cm" to "inches" contravened Article 123(3) EPC, because it caused the parameter Re to become smaller and thus extending to values below the lower limit of 10.5 x 10-2 in granted Claim 1; since it was not apparent from granted Claim 1 that the unit "cm" was "wrong", Rule 88 EPC could also not be invoked to justify this change;
(ii) the parameter Rr was not a constant, but - as even admitted in the patent specification - was itself a function of several variables, whose interdependency was unknown; moreover the condition 0 < Rr < 1 was of no technical significance because the mathematical equation in Claim 1 was not valid within this broad range; since the patent specification did not comprise sufficient information to reduce the parameter Rr into practice, the patent in suit did not comply with the requirement laid down in Article 100(b) EPC;
(iii) concerning the issues of novelty and inventive step, for the reason given in the last paragraph of the preceding Section there is no need to refer to the respective arguments of the Respondents; the same applies to the issue of Article 52(2)(a) EPC.
VIII. The Appellant requested that the decision under appeal be set aside and the European patent No. 452 405 be maintained on the basis of the Main Request or the Auxiliary Request both submitted during oral proceedings.
The Respondents requested dismissal of the appeal.
1. The appeal is admissible.
2. Article 123(2) and (3) EPC
2.1. The process to be carried out according to Claim 1 of both requests requires i.a. that the following conditions be met:
that the undrawn birefringence nu satisfies the equation nu = RfV2.0IV2.4, where Rf = RrRe is at least 15 x 10-3, Rr being a constant (0
As opposed to the above amended version, the granted Claim 1 required that the spinnerette capillary diameter D be measured in "cm", not in "inches" (see point I).
2.2. According to the application as filed the spinnerette capillary diameter D is to be determined in inches (Claim 8; sentence bridging pages 8 and 9; Examples 1 to 4: page 9, lines 25 to 27; page 10, lines 15 to 17 and 33 to 36; page 11, lines 6 to 8) and there is no reference to a measurement in "cm".
Consequently, the amendment to "inches" does not introduce subject-matter which extends beyond the content of the application as filed (Article 123(2) EPC).
2.3. The issue whether this amendment contravenes Article 123(3) EPC depends on the influence it has on the size of the parameter Re, which must not be lower than the numerical value of 10.5 x 10-2 when D is measured in "cm", as required by Claim 1 of the patent as granted.
The Respondents argued that the conversion of "cm" into "inches" led to a reduction of the numerical value of D (and thus Re), because 1 cm corresponded to only 0.39 inch. However, this reasoning does not take proper account of the specific situation here.
Rather, in the Board's judgment, the amendment causes the value of Re to go up, because a spinnerette capillary diameter of a certain numerical value in inches, is actually 2.54 times bigger than when the same numerical value is expressed in cm. Thus, since Re is directly proportional to D0.5, Re [D in inches] is, in real terms, 1.6 times (square root of 2.54) bigger than Re [D in cm] which has the effect to increase by the same factor the lower limit of Rf and nu. As pointed out by the Appellant during the oral proceedings, none of these parameters can now have values which would fall in substance outside the ranges arising from the previous definition of D, which means that the change to the measurement unit "inches" does not extend the protection conferred.
2.4. Apart from the change of the measurement unit of the spinnerette capillary diameter, Claim 1 of the Main Request differs from its granted version only by some formal rearrangements and by the insertion of the feature that "D is at least 0.027 inches (0.069 cm)", which was disclosed in Claim 8 of the application as filed (= granted Claim 7).
Claim 1 of the Auxiliary Request additionally comprises the upper limit of 0.95 for the intrinsic viscosity, which was disclosed on page 5, lines 35 to 37 of the application as filed (page 4, lines 1 to 3 of the granted patent).
These amendments to Claims 1 do not extent the scope of the patent as granted.
Claims 2 of both requests are based on Claim 5 of the application as filed (granted Claim 4), Claims 3 to 6 of both requests correspond to Claims 6, 7, 9 and 10 of the application as filed (granted Claims 5, 6, 8 and 9).
2.5. Thus, both requests comply with the requirements of Article 123(2) and (3) EPC.
3. Sufficiency of disclosure (Article 100 (b) EPC)
3.1. According to the Enlarged Board's decision G 10/91 (OJ EPO 1993, 420) only those grounds for opposition already cited at the opposition stage can be considered on appeal. New grounds can be introduced only with the Proprietor's consent.
3.1.1. In the present case, objections under Article 100(b) EPC concerning the nature of the parameter Rr have been raised in the opposition briefs of both Respondents, were dealt with in the decision under appeal (point C.1 of the Reasons) and were brought up again during the appeal by Respondent I (Section II of the submission of 10. June 1996).
3.1.2. The further objection under Article 100(b) EPC raised by the Opponents during the first instance opposition proceedings, namely the one concerning the measurement temperature of the intrinsic viscosity, was decided by the Opposition Division in the Proprietor's favour; that finding was not objected to by the Respondents during these appeal proceedings and also the Board concurs with that conclusion.
3.1.3. Considering the situation summarized in point 3.1.1 supra and contrary to the Appellant's conflicting assertion, the Board is therefore empowered to examine and decide the issue of the sufficiency of the disclosure of the patent specification with respect to the meaning of the parameter Rr without the Appellant's (Proprietor's) consent.
3.2. Article 100(b) EPC states that an opposition may be filed on the ground that the "European patent does not disclose the invention in a manner sufficiently clear and complete for it to be carried out by a person skilled in the art".
It is well established case law that a disclosure is regarded as sufficient only if it allows the invention to be reproduced without undue burden (cf. eg. T 226/85 OJ EPO 1988, 336; T 14/83 OJ EPO 1984, 105; T 32/85 of 5. June 1986, not published in the OJ EPO).
As set out below, this is not the case here.
3.3. Steps a) to d) of Claim 1 describe in general terms a process for the production of PET yarns by melt extrusion (step a), solidification (step b), filament take-up (step c) and hot drawing (step d). The manner in which steps a), b) and c) are to be carried out is defined by the requirement that the undrawn yarn should have a birefringence (Delta nu) of at least 0.020, wherein nu = RfV2.0IV2.4, where Rf = RrRe is at least 15 x 10-3, Rr being a constant (0
3.4. When carrying out this process with a PET having a certain IV of at least 0.80 it is, thus, necessary to select the spinning speed V, the spinnerette capillary diameter D, the polymer flow rate Q (which latter two parameters determine Re) and the parameter Rr.
While the meaning and the manner of manipulating the parameters V, D and Q is relatively straightforward, the parameter Rr is far from self-explanatory in that it is indeed a function of several variables.
3.5. According to page 4, lines 54 to 58 of the patent specification "Rr has a value which is related to the retention in orientation after thermally induced polymer relaxation within the constraint that 0
Rr is thus related to step b) of Claim 1, i.e. to the treatment of the extruded filaments after having left the spinnerette, finally yielding a solidified undrawn yarn. According to page 4, lines 23 to 29 of the patent specification the spun filaments are passed into a delay zone, which preferably is a quiescent zone or a heated sleeve of desired length, preferably 1 to 40 inches, maintained at a desired temperature of preferably 100 to 450 C. From the delay zone the yarn passes into a quenching zone, preferably a radial inflow quench, e.g. having an elongated chimney of conventional length, for example 1 to 40 inches.
According to all the four examples in the patent specification the delay zone is a heated sleeve of a certain length (Example 1: 1 inch; Example 3: 2 inches) and operated at a certain temperature (Examples 2 and 3: 220 to 300 C). The quench zone is defined in Examples 1 and 2 as a "radial quench stack", but no information is given as to the technical features of its construction and operation.
3.6. From the disclosure of the patent in suit the skilled person confronted with the necessity to determine the parameter Rr within the scope it is defined in Claim 1, is therefore made aware of the following:
(i) Rr is a constant (0
(ii) Rr increases with increasing temperature of the extruded filament,
(iii) Rr depends on the kind of delay zone, preferably a heated sleeve,
(iv) if the delay zone is a heated sleeve, then Rr decreases
(iv-1) with increasing sleeve length (preferably 1 to 40. inches) and
(iv-2) with increasing sleeve temperature (preferably 100 to 450 C),
(v) Rr depends on the severity of the quench.
The above feature (i) does not convey any teaching other than that Rr has a value which is peculiar to the respective concrete conditions of extrusion (including the spinnerette diameter D and the extrusion rate Q) and solidification (including the delay and quenching zones) and is thus different for different extrusion and/or solidification conditions. Thus, the term "constant", not appearing in the application as filed and regarded as not appropriate by the very Applicant in his letter of 5 March 1992, is a misconception.
3.7. The patent in suit does not disclose any precise information concerning the interdependence of the procedural steps referred to in points 3.6 (ii) to (iv) supra, and in particular the patent specification is completely silent on the concrete construction and operation conditions according to which the quenching (point 3.6 (v) supra) should be carried out in order to satisfy the condition nu = Rr.Re.V2.0.IV2.4.
3.8. The lack of information in the patent in suit with regard to the conditions of quenching (i.a. length of quench zone, temperature, velocity and flow pattern of cooling air) is, however, of particular importance for the determination of the parameter Rr, because it is well known that these conditions are of utmost importance for the orientation (which is reflected in the value of birefringence Delta nu) of filaments prepared according to the POY (partially oriented multifilament yarn) technology used according to the patent in suit.
This fact is emphasized as follows on page 28, right column, 2nd paragraph of document
D9: Riehl, Chemiefasern/Textilindustrie, Jan 1985, pages 27 to 31 (cited by Respondent I, then Opponent I, in his Notice of Opposition):
"The decisive area for the formation of the filaments is the plastic zone below the spinnerette, namely the zone where the filament changes from the liquid into the solid state. Since during solidification the filament is also under stress, this is the zone where all the processes of orientation and hence strength, elongation and shrinkage, but also the uniformity of the filament material itself, which expresses itself e.g. in the Uster value, are determined in all their data" (translation by the Board).
3.9. In the absence of any concrete guidance as to the design and operation of the quench zone the skilled person can only resort to trial and error experimentation in order to find out those conditions which will result in Rr values compatible with nu as required in Claim 1. This situation is further aggravated by the very meagre information the skilled person has with respect to the other factors (cf. point 3.6 (ii) to (iv) above) that influence Rr. In consequence thereof the skilled person, being without information as to how to take account of and modify the various interacting factors that determine Rr, is not in a position to proceed towards success with a few purposeful experiments. Rather he has to execute quite an extensive screening program on the basis of which he will than be able to supplement the present disclosure with the information now missing.
3.10. When considering the issue of sufficiency of disclosure it must also be borne in mind that Rr is not the only parameter whose value has to be determined on an experimental basis by trial and error. The mathematical equation for determining the birefringence nu comprises the further variables D, Q and V (if IV is held constant as property of the PET quality to be used), which also have an important influence on this equation. All these parameters have to be coordinated on experimental trial and error basis.
This fact was not contested by the Appellant before the first instance (cf. his submission of 28 January 1994, page 3, last paragraph to page 4, third paragraph). There the Appellant explained how the skilled person should proceed if he wanted to prepare an undrawn yarn having a birefringence nu = 0.16 from a PET having an IV of 1.0. According to this statement (1) the closest worked example in the patent specification should be chosen as a starting point (there Example 2), (2) the take up-speed V should be enhanced, (3) the polymer extrusion temperature should be lowered, (4) the shortest and (5) coldest heated sleeve which gave good yarn mechanical quality (few broken filaments) and (6) the severest quench which did not cause excessive filament turbulence should be used, and finally (7) the spinnerette diameter D and (8) the polymer flow rate Q should be adjusted to make Rf meet the value necessary to achieve the desired birefringence of nu = 0.16. Thus, in a first approximation, eight steps have to be performed on a trial and error basis without even taking into account that some of the parameters are interdependent, with the consequence that further experiments will be required to find out the structure of these interdependencies. The influence of such interdependencies is illustrated by a comparison of the nu, Rf and Re values of Examples 2 and 4 of the patent in suit, which, apart from a reduction of the spinnerette diameter from 0.027 inch to 0.018 inch and of a moderate increase of the spinning speed from 2.95 km/min (cf. Table 2 on page 6 of the submission of the Respondent II dated 14 May 1998) to 3.50 km/min were carried out under identical conditions (particularly identical quench conditions). However, the respective values changed considerably from Example 2 to Example 4 ( nu from 0.082 to 0.088; Rf value from 11 x 10-3 to 9.8 x 10-3 and Re value from 14 x 10-2 to 11 x 10-2, with the consequential change of the "constant" Rr from 7.86 x 10-2 to 8.91 x 10-2).
This highlights very well the amount of trial and error experimentation which is required in order satisfy the mathematical equation in Claim 1 of the patent under appeal and shows that the Appellant's allegation during the oral proceedings cannot be accepted that, in the case of an experimental failure, nothing more was required than to change the spinnerette.
3.11. In view of the foregoing the Board comes to the conclusion that the disclosure of the patent in suit is not sufficient for the skilled person to determine without undue burden the various conditions, particularly those defining the parameter Rr, which in combination establish the mathematical equation nu = Rf.V2.0.IV2.4 = Rr.Re.V2.0.IV2.4 specified in Claim 1 of the Main Request.
3.12. Consequently, the Main Request fails to comply with the requirement of Article 83 EPC, implied by Article 100(b) EPC, that the patent shall disclose the invention in a manner sufficiently clear and complete for it to be carried out by a person skilled in the art.
The fact that according to the worked examples in the patent specification the claimed invention could be carried out under certain, partly undisclosed conditions (cf. point 3.5 supra, last paragraph), cannot substitute this requirement which extends to the full claimed scope.
3.13. Since the Auxiliary Request differs from the Main Request only by an additional upper limit in Claim 1 for the intrinsic viscosity of the undrawn yarn (cf. point V, last paragraph), the difficulties for the reduction into practice of the mathematical equation nu = Rf.V2.0.IV2.4 = Rr.Re.V2.0.IV2.4 remain the same as for the Main Request. Consequently, the Auxiliary Request too fails to comply with the requirement of sufficiency of disclosure reflected in Article 100(b) EPC.
For these reasons it is decided that:
The appeal is dismissed.