Commonwealth Scientific and Industrial Research Organisation v Cytec Industries Inc - [2021] APO 27

IP AUSTRALIA

AUSTRALIAN PATENT OFFICE

Commonwealth Scientific and Industrial Research Organisation v Cytec Industries Inc. [2021] APO 27

Patent Application: 2014370386

Title:Polyacrylonitrile (PAN) polymers with low polydispersity index (PDI) and carbon fibers made therefrom

Patent Applicant: Cytec Industries Inc.

Opponent:Commonwealth Scientific and Industrial Research Organisation

Delegate:Les McCaffery

Decision Date: 1 July 2021

Hearing Date: 24 March 2021, in Canberra by Videoconference

Catchwords: PATENTS – opposition to grant of a patent under section 59 – clarity – clear enough and complete enough disclosure – support – inventive step – amendments and priority – opposition fails on all grounds – parties give 2 weeks to provide written submissions on costs.

Representation: Counsel for the applicant: Mr Ben Fitzpatrick
Patent attorney for the applicant: Dr Toby Thompson (Griffith Hack)
Counsel for the opponent: Mr Ian Horak
Patent attorney for the opponent: Dr Ramon Tozer, Dr Dario Buso (Davies Collison Cave), Dr Hishani Prabaharan (CSIRO)

IP AUSTRALIA
AUSTRALIAN PATENT OFFICE

Patent Application: 2014370386

Title:Polyacrylonitrile (PAN) polymers with low polydispersity index (PDI) and carbon fibers made therefrom

Patent Applicant: Cytec Industries Inc.
Date of Decision: 1 July 2021
DECISION
The opposition under section 59 is unsuccessful on all grounds.
Parties give 2 weeks to provide written submissions on costs.

REASONS FOR DECISION

Background

Australian Patent Application 2014370386 in the name of Cytec Industries Inc. (the applicant) was filed on 21 November 2014 and claims an earliest priority date of 23 December 2013 from US Application 61/919,843. I note that the priority date of the claims has been challenged, but this relates to an amendment that was made to the application during prosecution and not specifically to priority derived from the priority document.
The application was accepted on 27 June 2018. A notice of opposition was filed by Davies Collison Cave on 19 October 2018. A request to amend the notice of opposition and for the opposition to proceed in the name of Commonwealth Scientific and Industrial Research Organisation (the opponent) was allowed on 9 March 2021.
A statement of grounds and particulars was filed on 18 January 2019. This was amended on 9 April 2019 and again on 24 October 2019.
Evidence in support was completed on and consists of declarations by:
Dr Graeme Moad (Moad 1), dated 17 April 2019 together with exhibits GM-1 to GM-15.
Dr Richard Alexander Evans (Evans 1), dated 16 April 2019, together with exhibits RE-1 to RE-3.
Mr Christopher Henry Such (Such 1), dated 16 April 2019, together with exhibits CS-1 to CS-13.
Dr Derek Buckmaster (Buckmaster 1), dated 17 April 2019, together with exhibits DB-1 to DB-8.
Evidence in answer was completed on and consists of declarations by:
Dr Sreekumar Thaliyil Veedu (Veedu), dated 24 July 2019, together with exhibits SV-1 and SV-2.
Dr David James Wilson (Wilson), dated 23 July 2019, together with exhibits DJW-1 to DJW-4.
Evidence in reply was completed on and consists of declarations by:
Dr Moad (Moad 2), dated 24 September 2019 together with exhibits GM-16 to GM-21.
Dr Evans (Evans 2), dated 24 September 2019, together with exhibits RE-4 and RE-5.
Mr Such (Such 2), dated 24 September 2019, together with exhibits CS-14 to CS-24.
Dr Buckmaster (Buckmaster 2), dated 24 September 2019, together with exhibits DB-9 and DB-10.
The amended statement of grounds and particulars set out the grounds of opposition as manner of manufacture, novelty, inventive step, usefulness and all ground under section 40(2) and (3). Inventive step, section 40(2)(a) and section 40(3) were pressed at the hearing.
The hearing was held on 24 March 2021 by videoconference.
Amendments brought about by the Intellectual Property Laws Amendment (Raising the Bar) Act 2012 apply to this application. This includes section 60(3A), which provides that the Commissioner may refuse an application if satisfied on the balance of probabilities that a ground of opposition exists. The opponent carries the onus of proof.

The specification

10. The specification provides a useful background to the invention. I have attempted to summarise this material to provide some context for the issues in dispute.

11. The technical field of the invention is said to relate to the “synthesis of polyacrylonitrile (PAN) polymers and methods of forming carbon fibres from PAN copolymers”.[1]

[1] Specification at page 1, lines 4 to 5.

12. PAN copolymers are said to be useful for the production of carbon fibres, which because of properties such as high specific strength and stiffness, high chemical resistance and low thermal expansion, have been widely used in aerospace, sports, automobiles, wind energy and other energy saving areas.[2]

[2] Specification at page 1, lines 7 to 10.

13. PAN copolymers have traditionally been made by free radical polymerisation[3] techniques. Polymerisation uses a catalyst or initiator to produce free radical species which then react with monomers to create free monomer radicals. These react with other monomers to propagate the molecular chain, resulting in polymer radicals. The propagation reactions can be terminated in several ways. Combination termination occurs when two free radical chains combine to form a dead chain. Disproportionation termination can occur when the free radical end of one radical chain attacks a hydrogen on the second to last carbon of a second radical chain, resulting in two dead chains.

[3] I have generally used the spelling polymerisation throughout the decision except where quoting or citing material that uses the alternative spelling polymerization.

14. A third termination method involves the use of chain transfer agents. The polymer radical reacts with the chain transfer agent, thereby terminating the propagation reaction and forming a new radical from the chain transfer agent. The new radical then starts a new chain propagation. Chain transfer agents thereby act to reduce the length of the polymer chains. If the rate of termination is greater than the rate of propagation, then shorter polymer chains are obtained.

15. If the chains are not terminated before all the monomer is consumed, then propagation can continue if more monomer is added. This polymerisation is referred to as a living polymerisation. In an ideal living polymerisation, all chains are initiated at the beginning of the reaction and grow at a similar rate. There is no irreversible chain transfer or termination. If initiation is rapid compared to propagation, the molecular weight distribution is very narrow, and the chains can be extended by the addition of more monomer.

16. However, in a radical polymerisation all chains cannot be simultaneously active. One approach to address this problem has been to use a Reversible Addition-Fragmentation Chain Transfer (RAFT) agent to control the propagation and its rate by forming a dormant stage. The present invention makes use of RAFT agents.

17. By reversibly deactivating or activating the propagation, a rapid equilibrium between active and dormant chains can be achieved to control the chain growth at a similar rate and achieve a narrow molecular weight distribution. This is called “controlled/living radical polymerisation”.

18. Because of the different termination reactions that can occur, the molecular chains obtained by free radical polymerisation have different lengths or molecular weights – that is, they have a distribution. This distribution can be defined by its polydispersity index (PDI). The PDI can be calculated in the following way:

PDI = Mw/Mn

Wherein Mw is the weight average molecular weight, and
Mn is the number average molecular weight.

19. The specification states that an alternative means of expressing PDI can be calculated as:

PDI = Mz/Mw

Wherein Mz is the average molecular weight, and
Mw is the weight average molecular weight.

20. Mw, Mn and Mz are measured by gel permeation chromatography. A high PDI indicates the polymer has a large molecular weight distribution. This means that the polymer comprises molecular chains that vary greatly in length. A high PDI is said to affect the processability of the polymer into fibres by spinning, as well as the fibre properties, since small molecular weight polymers act as a kind of molecular defect that impacts on the mechanical properties of the polymer.

21. The specification states that conventional radical polymerisation does not provide adequate control over PAN copolymerisation and large molecular weight distributions occur. The invention is said to provide a method for synthesising PAN with a narrow molecular weight distribution and a method for producing carbon fibres from such polymer. The polymer is said to be synthesised by controlled/living radical polymerisation using a “special” RAFT agent.

22. Furthermore, carbon fibres produced from these polymers are said to exhibit uniform cross-section, and low micro- and molecular defects since the low PDI results in low molecular and micro defects during carbon fibre manufacturing processes. In the present invention the PDI (Mw/Mn) is said to be “targeted to around 2 or less, preferably PDI (Mw/Mn) of 1.2 to 1.9 (or an alternative PDI (Mz/Mw) of 1.2 to 1.7)”. The molecular weight of the polymers obtained by the present process are said to be within the range of 60 to 500, preferably 90 to 250 and most preferably 115 to 180 kg/mol.[4]

[4] Throughout the decision the molecular weight may be expressed in terms of kg/mol or g/mol. For example, 50 kg/mol would be the equivalent of 50,000 g/mol.

23. The invention specifically relates to a method for synthesizing a PAN copolymer with a narrow molecular weight distribution, the method comprising:

a)Combining acrylonitrile (AN) monomer with a solvent, at least one co-monomer, and a thiocarbonylthio compound to form a solution;
b)Heating the solution to a temperature within the range of 40℃ – 85℃; and
c)Adding an initiator to the heated solution to affect polymerisation reaction,

wherein polymerisation is affected by controlled/living radical polymerisation, in which the thiocarbonylthio compound functions as a Reversible Addition/Fragmentation Chain Transfer (RAFT) agent,

wherein the thiocarbonylthio compound is selected from the following structures:

wherein the PAN copolymer has a polydispersibility index (PDI) of about 2 or less and a molecular weight within the range of 60 kg/mol to 500 kg/mol.

24. The specification goes on to provide a general description of suitable solvents, co-monomers and initiators, as well as preferred ratios of reactants. The low-PDI polymers are said to have good spinning ability. The spun polymers can be converted to carbon fibres by standard techniques. The specification provides several examples to illustrate the invention. Example 1 relates to the synthesis of PAN copolymers, Example 2 to the preparation of white fibres and Example 3 to the preparation of carbon fibres.

25. The specification ends with fifteen claims. Claims 1 to 11 define methods of preparing PAN copolymers corresponding to the method described above. Claims 12 to 15 define methods of making carbon fibres using the polymers made by the methods of Claims 1 to 11. The claims are provided in full in the Attachment. I have undertaken a detailed construction of claim 1 below.

The person skilled in the art (PSA)

26. The person skilled in the art is the hypothetical person to whom the specification is addressed.[5] This determination plays a central role in determining the validity of the patent:

“He is the person to whom the patent is addressed and who must construe it. He is the person whose knowledge will determine whether a patent is novel. He is the person who will judge whether a patent is obvious...”[6]
[5] General Tire & Rubber Co v Firestone Tyre & Rubber Co Ltd (1971) 1A IPR 121 at 134.
[6] Root Quality Pty Ltd v Root Control Technologies Pty Ltd [2000] FCA 980; 49 IPR 225 at [70].

27. The PSA is assumed to be a skilled but unimaginative and non-inventive worker in the field of the invention. One formulation describes them as “those likely to have a practical interest in the subject matter of [the] invention.”^[7]

[7] Ibid at [70]-[72].

28. In the present case, there appears to be a fundamental difference in the characterisation of the relevant PSA by the parties. This appears to stem from the different characterisation of the problem to be solved.

29. The applicant submitted that the field of the invention lies in methods for synthesising polyacrylonitrile (PAN) polymer suitable for use in the production of carbon fibres and methods for forming carbon fibres from such PAN copolymers. They considered that the relevant addressee would be experienced in the manufacture of advanced fibre and carbon fibre, as well as the polymer synthesis processes required to produce fibre. The opponent did not characterise the PSA in their submissions, but their submissions in relation to the CGK focus heavily on RAFT polymerisation.

30. I acknowledge that the invention relates to the use of RAFT polymerisation to provide PAN copolymers. However, the specification is aimed at addressing shortcomings in existing free radical polymerisation techniques for the preparation of PAN copolymers specifically for use in making carbon fibres. This is a broader consideration than the one suggested by the opponent. I therefore consider that the PSA would comprise a team of polymer scientists with a broad knowledge of free radical polymerisation techniques used in the preparation of PAN copolymers, as well as their use in the preparation of carbon fibres.

31. Evidence was provided by the following witnesses:

Dr Moad is a senior polymer scientist at CSIRO. Dr Moad stated that he has expertise in radical polymerisation, and, in particular, RAFT polymerisation.[8] Dr Moad has been an author on over 200 scientific papers and an inventor on 34 patent families. He provided a list of key publications which he had co-authored, which combined had been cited in about 14000 articles as of the priority date of the present application.[9] Dr Moad stated that he is currently co-editing a book on RAFT polymerisation.[10]
Dr Evans is a senior principal research scientist at CSIRO. Dr Evans stated that his principal research interest is in the development of highly functional materials using living radical polymerisation methods, and that RAFT polymerisation is a “recurring theme” of his research projects.[11]
Mr Such retired as a Chief Scientist at Dulux in 2017. He stated that he specialised in the design and operational management of plants for the industrial production of polymers, including the production of polymers by RAFT polymerisation.[12] Mr Such stated that he had pioneered the industrial implementation of RAFT-mediated processes for commercial production of polymers.[13] His CV listed several journal articles and patents relating to RAFT polymerization.[14]
Dr Buckmaster is the Director of Carbon Nexus, which he stated is the first open-access carbon fibre/composite research facility in the world.[15] Dr Buckmaster is responsible for developing and implementing strategies to maximise industry participation in the open access Carbon Nexus research facilities.[16] He also stated that between 2007 and 2014 he was employed in various General Manager and Technical Director positions where he acquired extensive experience in the synthesis, manufacture and commercialisation of carbon fibre and carbon fibre-related technologies.[17]
Dr Veedu is the Director of Technorbital Advanced Materials Pvt Ltd, a company that manufactures carbon-grade PAN copolymer and PAN-based hollow fibre membrane.[18] He also consults on advanced fibre and carbon fibre technologies.[19] Dr Veedu is co-author or inventor on several journal articles and patents relating to polyacrylonitrile polymers.[20]
Dr Wilson is a laboratory manager and Technology Leader for Synthetic Polymers at the Aubervilliers Research and Innovation Centre of Novecare.[21] His CV states that he has worked on RAFT/MADIX[22] polymerisation techniques and products for the last 14 years.[23]
[8] Moad 1 at [1].
[9] Exhibit GM-3.
[10] Moad 1 at [21].
[11] Evans 1 at [5].
[12] Such 1 at [1].
[13] Such 1 at [7].
[14] Exhibit CS-2.
[15] Buckmaster 1 at [1].
[16] Buckmaster 1 at [8].
[17] Buckmaster 1 at [7].
[18] Veedu at [1].
[19] Veedu at [11].
[20] Exhibit SV-1.
[21] Wilson at [1].
[22] MADIX is a polymerisation process which is similar in mechanism to RAFT. See Wilson at [23].
[23] Exhibit DJW-1.

32. Consistent with the problem the opponent formulated, they submitted that their experts were well qualified to give evidence regarding polymerisation procedures known in the art for reducing the PDI of polymers, including RAFT.[24] They argued that the evidence of their experts should be preferred over the evidence from Dr Veedu and Dr Wilson because:

Dr Veedu had knowledge of RAFT polymerisation but had never used the process. The specification is directed to a person with extensive knowledge of RAFT polymerisation, and that needs to be taken into account in the weight given Dr Veedu’s evidence.[25]
Dr Wilson’s evidence was supported by only a single document published before the priority date,[26] and that Dr Wilson’s evidence attempted to paint an inaccurate picture of the general applicability of RAFT polymerisation.[27]
[24] Opponent’s written submissions at [51].
[25] Opponent’s written submissions at [53].
[26] Opponent’s submissions at [56].
[27] Opponent’s submissions at [57].

33. The applicant accepted that all the experts are qualified in the field of the invention. However, they highlighted aspects of the evidence given by the Opponent’s witnesses that affect the weight that should be given to their evidence:

The identity of the true opponent was only disclosed shortly before the hearing, and that two of the witnesses are employees of the opponent while a third is the director of a joint venture set up by the opponent. None of the witnesses indicated in their evidence that they knew the identity of the true opponent or whether they had connections or an association with the true opponent.[28]
Dr Moad has significant experience in RAFT polymerisation. His views as to how RAFT polymerisation might be used and modified therefore does not reflect the approach that the notional skilled person would have taken as a matter of routine at the priority date.[29]
[28] Applicant’s submissions at [88].
[29] Applicant’s submission at [90].

34. Contrary to the opponent’s submissions I do not consider it necessary that the person have extensive experience of RAFT polymerisation. As noted previously, the PSA is a team of polymer scientists with a broad knowledge of free radical polymerisation techniques, as well as PAN copolymers and their use in the preparation of carbon fibres. I consider that each of the experts has relevant knowledge or experience in that regard.

35. However, I share the concerns of the applicant in respect to the evidence provided by Dr Moad. The opponent has made a number of submissions based on information they allege is common general knowledge information where Dr Moad is a co-author or co-inventor of the supporting documents. Dr Moad has demonstrated a high level of inventive capacity and is clearly at the forefront of RAFT polymerisation. I do not consider that Dr Moad is truly representative of a polymer scientist of ordinary skill in the art.

36. But it does not follow that I should exclude Dr Moad’s evidence on the basis that his experience is too extensive. The key consideration will be, as it is with all of the witnesses, whether the evidence goes to the issue of common general knowledge and what the PSA in the relevant art would do, as opposed to what the witness themselves knew and would have done in a particular instance.

37. I also acknowledge the applicant’s concerns that the identity of the true opponent was not disclosed earlier in the process, particularly where there could be questions of potential conflicts of interest. But this approach to an opposition is allowed, and it does not necessarily mean that the evidence is tainted. As recognised by the applicant in their submission, the witnesses were provided with guidance in the form of the Federal Court’s Expert Evidence Practice Note. They were therefore aware of their role in assisting the determination by providing relevant and unbiased evidence, rather than acting as an advocate for any party.

38. Nevertheless, in the present case declarants appear at times to go beyond the role of providing evidence on technical issues, and to draw inferences from the evidence, suggesting, for example, that witnesses are “misleading” in their evidence, that their evidence is based on “hearsay”, and “perhaps deliberately trying to discredit the implementation and function of RAFT agents”. Witnesses will understandably disagree with the evidence of other witnesses, and on occasions will express those differences in strong terms. This can also be no more than a poor choice of language or statements made in response to leading questions. However, such statements can indicate a lack of objectivity, or suggest that the witness is acting as an advocate for the party. If there are differences in the evidence of witnesses it is the role of the hearing officer to reconcile those differences, and to weigh the evidence before them in view of all the circumstances. It is not the place of an expert to draw inferences in their evidence.

39. On balance, I do not consider it necessary to entirely discount the evidence of any of the witnesses. But under the circumstances outlined above, I am led to approach the evidence with some caution.

Construction

40. The parties differed on some key aspects of the process defined by the claims. Some of these are dealt with in more detail under the section 40 grounds. The majority of these differences lie in Claim 1, so I will focus on this claim in my construction analysis.

41. Claim 1 defines a method for synthesizing a polyacrylonitrile (PAN) polymer using specific RAFT agents under specific conditions. I understand the reference to a PAN polymer as being one in which acrylonitrile is the main component in the copolymer. This is also consistent with the references in Claim 1 to “acrylonitrile (AN) monomer and “at least one co-monomer”.[30] I note also that there are several errors in the claims, but I do not consider that these impact on the construction of the claims. I have dealt with this issue in more detail later in the decision under the grounds of Clarity.

[30] Applicant’s submissions at [217] indicate that the main component is referred to as the monomer, and the minor component is referred to as the comonomer.

42. The language used in the preamble of the claim is that the method comprises the three steps. The language of the claim is open-ended – that is, in addition to the steps defined there may be other features used in the process. But I do not understand this to mean that this can encompass any and all possible variations to the process. The language used to define the individual steps is closed such that it links them as consecutive steps: step (a) forms a solution; step (b) involves heating the solution; and step (c) involves adding an initiator to the heated solution. In these consecutive steps the definite article links the solution which is heated in step (b) to the solution which was formed in step (a). Similarly, “the heated solution” in step (c) is the solution which was heated in step (b). The language in the three steps is also closed inasmuch as it does not provide for additional components. For example, the definition of the combination of components in step (a) does not provide for additional reactants other than further comonomers.

43. I therefore consider the claim is limited to a process wherein the three steps are carried out consecutively. However, the open-ended language of the claim in relation to the method as a whole would include processes where in addition to the three consecutive steps, there may be other steps. For example, there may be pre-heating steps, purification steps, or even (potentially), the addition of further reactants following step (c).

44. This analysis goes towards some of the opponent’s submissions. For example, the opponent submitted that the claim did not explicitly exclude addition of initiator in step (a).[31] Dr Moad also suggested that “heated” is not the same as “hot”,[32] meaning that the process would involve heating the solution as defined in step (b), then an intermediate step of allowing the solution to cool, then adding the initiator to the (previously heated) cooled solution and then reheating the solution with the initiator from room temperature.

[31] Such 1 at [76], Moad 2 at [155].

[32] Moad 1 at [128] and Moad 2 at [131].

45. Based on my interpretation the claim does not include such steps, and the aim of the argument appears to be to provide support for the conclusions made in relation to obviousness. The evidence given by the declarants in this regard is, as noted by the applicant, inconsistent with the established principles of construction.[33] As Middleton J stated:

“It is well settled that the Court should, from the outset, approach the task of patent construction with a generous measure of common sense. The Court must place itself in the position of a person skilled in the relevant art, being the subject matter of the patent. From this perspective, the patent is to be read as a whole, in the context of the specification and in light of the prevailing common general knowledge and state of the relevant art at the priority date.”[34]
[33] Applicant submissions at [56].
[34] Eli Lilly and Co Ltd v Apotex Pty Ltd [2013] FCA 214; 100 IPR 451 at 482.

46. Notably the construction given to the language in a claim is determined in light of the CGK and the understanding of the PSA. Once determined, the construction given a term should be used consistently and not varied in view of the prior art documents under consideration in order to make out a case of invalidity. In this respect, I prefer the evidence of Dr Veedu and Dr Wilson, which is consistent with my analysis above.[35]

[35] Wilson at [79], Veedu at [128].

47. The claim includes a requirement that the “polymerization is affected (sic) by controlled/living radical polymerisation in which the thiocarbonylthio compound functions as a Reversible Addition/Fragmentation Chain Transfer (RAFT) agent”. The conditions used in the process, and particularly the use of thiocarbonylthio RAFT agents would suggest that the definition that the mechanism by which polymerisation occurs is superfluous.[36] While this could be no more than a “belt and braces” approach to defining the process, it appears to me that the claim requires that the conditions used in the process be such that a controlled/living radical polymerisation occurs rather than any other possible type of polymerisation.

[36] Moad 1 at [132].

48. Perhaps the most contentious issue in the construction relates to the use of the term “affect” in step (c). The opponent submitted that the ordinary meaning of the term “affect” is “to have an effect on/to make a difference to”,[37] rather than to trigger or start as submitted by the applicant.[38] They considered this indicates that the polymerisation is already underway when the initiator is added to the mixture in step (c). This was used in support of the opponent’s other submissions in relation to initiator being added earlier in the process at step (a). I have already determined that the claim does not provide for the addition of initiator at an earlier stage of the process, so this argument is moot.

[37] Moad 1 at [130], Such 1 at [76].

[38] Wilson [81], Veedu [37].

49. In any case, the issue appears to arise because the claim uses the term “affect” (to have an effect on/to make a difference to), rather than “effect” (which when used as a verb can mean to bring about, such as for example to effect a change). The use of “affect” in this way may be archaic, but I consider that, in the context of the specification as a whole, the reader would readily understand that the term is being used synonymously with the term “effect”.

50. I therefore understand step (c) to define that the initiator is added to the heated solution, thereby starting the polymerisation reaction.

51. There are a couple of other construction issues that were also raised under section 40 – the meaning of the terms “PDI” and “molecular weight”. These are discussed later, but my understanding is that “PDI” refers the Mw/Mn measure of dispersity, and that “molecular weight” refers to the weight average molecular weight.

52. At this point I will also address the evidence of Dr Evans since it aims to establish that the addition of initiator after heating is, in effect, an inessential feature of the claimed invention.

53. Dr Evans was “briefed” on experimental tests done by Dr Melissa Skidmore and Ms Lisa O’Brien, who are employees of the opponent,[39] and provided evidence on what he considered the results showed and how those compared with the present application. Based on his analysis of the results he was provided, he concluded that the same product would be obtained irrespective of whether the initiator was added before heating or after heating.[40] He also provided comments on the use of certain initiators under certain conditions, but I will deal with those submissions later in the decision.

[39] Evans 1 at [7].

[40] Evans 1 at [33].

54. The applicant questioned the probative value of the experimental evidence. They referred to Rule 34.50 of the Federal Court Rules, which include certain requirements that had not been met here.[41] I will not repeat the requirements here, but note they include the other party being informed of the particulars of the experiments and being permitted to attend the conduct of the experiments. The applicant recognised that there is no formal requirement in Patent Office proceedings to comply with these requirements but submitted that it highlighted the dangers in accepting this evidence.

[41] Applicant’s submissions at [153].

55. As noted by the applicant, there is no formal requirement in Patent Office proceedings to comply with the Federal Court Rules. However, I share their concerns in relation to the evidence. Dr Evans did not carry out the experiments and his evidence is only in relation to the material he has been provided. He does not appear to have first-hand knowledge, for example, of the questions posed in developing the experiments, how the conditions were determined (for example, ratio of reagents), nor whether further results were obtained that were not provided to Dr Evans. The experiments do not include comparative experiments where the conditions of the present application were employed under the same conditions or provide relevant details of other conditions such as the heating rate that could potentially change the course of the reactions.[42] Given these factors and their potential impacts, I do not consider that this material can be relied upon to establish that the step of adding the initiator following heating is inessential to the invention.

[42] Wilson at [112].

56. I will adhere to this construction of the claim in my decision.

Inventive step

57. Section 7 of the Act sets out that an invention is taken to involve an inventive step unless it would have been obvious to the person skilled in the art in the light of the common general knowledge, either considered alone or together with the prior art.

58. The test for whether an invention is obvious is whether it would have been a matter of routine to proceed to the claimed invention.[43] In Alphapharm,[44] the High Court accepted the approach taken by Graham J in Olin Mathieson, where he used a reformulated Cripps question.[45] Notably in the case of a combination, the question is whether the combination as a whole is obvious, not whether each individual integer in the combination is obvious.[46]

[43] Wellcome Foundation Ltd v VR Laboratories (Aust) Pty Ltd [1981] HCA 12; 148 CLR 262 at 286.

[44] Aktiebolaget Hassle v Alphapharm Pty Ltd [2002] HCA 59 at [53]; (2002) 212 CLR 411 at [53].

[45] Olin Mathieson Chemical Corporation v Biorex Laboratories Ltd [1970] RPC 157 at [187].

[46] Aktiebolaget Hassle v Alphapharm Pty Ltd, ibid at [65].

59. Matters that do not constitute common general knowledge cannot form part of the problem addressed by the specification. As noted by the Full Bench of the Federal Court in AstraZeneca v Apotex:[47]

“...whether a claim of a patent is invalid for lack of inventive step is to be determined by comparing the invention, so far as claimed, against the common general knowledge and any 7(3) information. The question is then whether the invention would have been obvious to the hypothetical person skilled in the art in light of that knowledge considered separately from, or together with, the s 7(3) information. So understood, it is apparent that the relevant provisions of the Act do not expressly or impliedly contemplate that the body of knowledge against which the question whether or not an invention, so far as claimed, involves an inventive step is to be determined may be enlarged by reference to the inventor’s (or patent applicant’s) description in the complete specification of the invention including, in particular, any problem that the invention is explicitly or implicitly directed at solving.
If the problem addressed by a patent specification is itself common general knowledge, or if knowledge of the problem is s 7(3) information, then such knowledge or information will be attributed to the hypothetical person skilled in the art for the purpose of assessing obviousness. But if the problem cannot be attributed to the hypothetical person skilled in the art in either of these ways then it is not permissible to attribute a knowledge of the problem on the basis of the inventor’s ‘starting point’ such as might be gleaned from a reading of the complete specification as a whole.”
[47] AstraZeneca AB v Apotex Pty Ltd (2014) 107 IPR 177; [2014] FCAFC 99 at [202]- [203].

60. The opponent submitted that the problem the specification seeks to address is the inability to produce PAN with low PDI, the solution being the use of RAFT as an alternative to conventional radical polymerisation.[48] They framed the Cripp’s question in the following way:

“would the person skilled in the art, in all the circumstances, directly be led as a matter of course to try the method of claim 1 to synthesize PAN in the expectation that it might well produce the desired result (being PAN having a PDI of less than about 2).”
[48] Opponent’s written submissions at [22].

61. The applicant submitted that the problem lies in providing a PAN copolymer suitable for use in producing carbon fibre and having a PDI of less than 2 and an average molecular weight of 60-500 kg/mol.[49] Notably the intended use of the PAN copolymer for producing carbon fibre and the molecular weight of the PAN copolymer is included in their formulation of the problem.

[49] Applicant’s written submissions at [118].

62. The PAN copolymers prepared by the method of the invention are purported to have properties that address issues with existing methods of preparing PAN copolymers which useful in the manufacture of carbon fibres with particularly advantageous mechanical properties. Admittedly Claims 1 to 11 define the method of producing the PAN copolymer without any limitation as to their intended use. But this is not unusual in patent claims covering a product which is intended to be used for a specific purpose (or its method of preparation).

63. It follows, in my view, that the provision of such products (or their precursors) constitutes part of the problem to be solved. In addition to the PDI, the molecular weight of the PAN copolymer contributes to making it suitable for use in carbon fibres. I would therefore include this additional feature in the Cripp’s question posed by the opponent – namely the desired result is a PAN having a PDI of less than about 2 and a molecular weight of 60 to 500 kg/mol which is suitable for the preparation of carbon fibres.

64. The opponent made submissions on inventive step in view of common general knowledge alone, as well as common general knowledge taken together with each of the following documents:

WO98/01478 (D5)
Xiao-hui Liu et al., “Well defined higher-molecular-weight polyacrylonitrile via RAFT technique in the presence of disulfide compounds as a source of chain transfer agent”, European Polymer Journal 44 (2008) 1200-1208 (D32)
KR20120115029 (based on an Espacenet machine translation provided by the opponent in evidence) (D26)
US2010/0003515 (D20)
WO2008063886 (D31)

65. I will consider each of the submissions in relation to CGK first, and then each of the documents in turn.

Inventive step in light of the CGK

66. The gist of the opponent’s submissions is that RAFT polymerisation was a preferred polymerisation technique for lowering PDI,[50] and that the conditions used in the opposed application were routine in the art.[51] The PSA would be led directly to try RAFT polymerisation in the expectation that it might well produce the desired polymer. The arguments were framed around it being obvious to use RAFT for the preparation of a PAN copolymer having a low PDI, and having initially arrived at the use of RAFT, the specific reagents and conditions were routine.[52]

[50] Opponent’s submissions at [115] to [126].

[51] Opponent’s submissions at [127] to [137].

[52] Opponent’s submissions at [127].

67. I provided a summary of the specification above which covered some background to the invention, none of which appears to be in dispute. The opponent made additional submissions concerning the CGK, including:

The use of RAFT agents in radical polymerisation was routine by the earliest priority date.[53]
RAFT polymerisation was known for its versatility and ability to produce polymers with low PDI.[54]
It was routine practice to adjust various reaction parameters as a means of controlling the molecular weight of polymers. This included the choice of RAFT agent, the reagents and quantities used, and the use of different conditions (batch, semi-continuous etc.).[55]
RAFT produced polymers having a molecular weight between 60 kg/mol and 500 kg/mol.[56]
[53] Opponent’s submissions at [75].
[54] Opponent’s submissions at [76].
[55] Opponent’s submissions at [92] to [96].
[56] Opponent’s submissions at [78].

68. Perhaps the key point of contention between the parties lies in the opponent’s submissions that it was CGK that RAFT polymerisation could be used for the production of PAN copolymers, and it was also CGK that RAFT could be used to produce PAN with a molecular weight above 60,000 g/mol and PDI below 2.[57]

[57] Opponent’s submissions at [97] to [100].

69. The applicant submitted that the opponent’s evidence and submissions in relation to the CGK are general propositions concerning RAFT technology without proper regard to the fact that the claimed method involves a number of specific conditions to produce PAN copolymer suitable for carbon fibre. The evidence was not specifically directed to the preparation of PAN copolymers, nor to the specific molecular weight and PDI suited to the preparation of carbon fibre. [58]

[58] Applicant submissions at [141].

70. Turning to the evidence on the CGK, it was routine practice to perform radical polymerisation reactions using one or more means to achieve improved control the molecular weight characteristics of the resulting polymer.[59] This included the use of irreversible chain transfer agents which act by substitution, and irreversible addition/fragmentation transfer agents.[60] Another means of controlling molecular weight involves reversible-deactivation radical polymerization (RDRP). RAFT, which was developed in the mid-1990s, is a type of RDRP process.[61] Others include atom transfer radical polymerization (ATRP) and nitroxide-mediated radical polymerisation (NMP).[62] RDRPs provide for lower PDI compared to conventional free radical polymerisations. [63]

[59] Moad 1 at [18].

[60] Moad 1 at [20].

[61] Moad 1 at [21].

[62] Moad 1 at [23].

[63] Moad 1 at [30].

71. NMP was considered restrictive in terms of the range of applicable monomers (usually polystyrene and certain acrylates), and its use would have been “conditional to the fine tuning of operational conditions”. ATRP is usually performed using organometallic complexes that inevitably lead to metal contamination and the poor solubility of the reagents limits the range of monomers and solvents that can be used.[64] Dr Such further stated that polymerisation procedures such as ATRP and NMP would have limited applicability since they require very strict reaction conditions to actually work, particularly in an industrial setting.[65]

[64] Moad 1 at [33].

[65] Such 1 at [19].

72. Dr Moad considered that in terms of lowering the PDI of a polymer, RAFT would have been considered as one of the more versatile RDRP techniques in terms of its applicability and effectiveness.[66] He suggested that RAFT agents were “certainly in everyone’s ‘toolkit’ to use in radical polymerisation”. This seems a fair statement, noting that this does not mean that it was the only technique available to polymer scientists.

[66] Moad 1 at [34].

73. There is no apparent dispute between the parties on these general matters. However, this evidence is in relation to radical polymerisation reactions in general and not specifically to PAN.

74. To this end, Dr Moad considered that the principles for RAFT polymerisation applied to PAN copolymers as they did to any other polymer.[67] He stated that in choosing a suitable means for RDRP he would have routinely turned to the use of RAFT because of its versatility and efficiency, “together with it being known at the time that RAFT polymerisation was applicable to the polymerisation of acrylonitrile”.[68]

[67] Moad 1 at [72].

[68] Moad 1 at [75].

75. He also stated that it was commonly accepted that it was possible to use RAFT for producing polymers having a molecular weight higher than 60 kg/mol in combination with a PDI of less than 2. Such characteristics could also be achieved using conventional radical polymerisation.[69] He went on to state that it would have been common general knowledge that obtaining polymers having these properties “was rather trivial”.[70] He referred to several documents that he said disclosed PAN copolymers having the defined characteristics which were made via RAFT polymerisation.[71] The opponent submitted, based on this statement, that the documents represented common general knowledge at the time.[72]

[69] Moad 1 at [32].

[70] Moad 2 at [37].

[71] Moad 2 at [42] to [46].

[72] Opponent’s submissions at [100].

76. I do not find this submission persuasive. The statements made by Dr Moad go to the issue of whether PAN copolymers could be obtained by RAFT polymerisation, and not whether or not this was CGK. While the documents referred to may have been public knowledge, no evidence was adduced that would support the submission that they represented common general knowledge in the field.

77. Dr Moad also referred to D5, which could arguably represent common general knowledge. This patent document was one of the first disclosures of RAFT polymerisations and has been cited numerous times in other papers.[73] This is certainly a strong indication that the document has been widely read and accepted in the art, and the RAFT agents described appear to have found widespread use. However, D5 is a broad disclosure of a wide range of different polymer types and conditions, and the evidence does not indicate the context in which D5 has been cited in those documents. I consider this a relevant consideration in whether the application of RAFT to the synthesis of PAN copolymers was CGK, and the extent to which D5 shows it to be so.

[73] Moad 2 at [46].

78. Dr Moad referred specifically to the results provided in Table 33, and stated that:

“[D5] (which would have been commonly known by polymer scientists in the field of RAFT polymerisation at the Earliest Priority Date) shows a RAFT synthesis route that can readily achieve synthesis of PAN copolymer having molecular weight above 60,000 g/mol and PDI well below 2. When considering the reaction times of the test samples shown in Table 33, it is clear that for polymerisation times higher than 18 hours one could readily expect to produce PAN copolymer having molecular weight above 60,000 g/mol and PDI well below 2…”

79. Even if I were to accept that D5 represents common general knowledge, including the specific information referred to by Dr Moad above, I do not find this evidence useful in the determination of the common general knowledge in relation to PAN copolymers. In particular, the example referred to relates to the preparation of a styrene-acrylonitrile copolymer and not a PAN copolymer. The predominant component is styrene (about 62%), and acrylonitrile is the minor component (about 38%). The conditions used in the example appear to be specific to styrene monomer (for example the manner in which the polymerisation is initiated). I therefore consider that D5 does not support the opponent’s submissions that it was CGK that RAFT could be used for the preparation of PAN copolymers.

80. The opponent also relied on statements by Dr Such in reference to exhibit CS-3.[74] Dr Such stated that at the time “there existed established and well documented guidelines one would have routinely consulted for guidance on the choice of specific agents for given monomers”. He provided a two-page extract from the Handbook of RAFT Polymerization which provides a table outlining guidelines for the selection of RAFT agents for various polymerisations (Figure 6.2). The Figure is provided below and indicates the suitability of various substituted thiocarbonyl RAFT derivatives for polymerisation of various classes of monomers. Acrylonitrile is abbreviated as “AN”.

[74] Such 1 at [21].

81. The inclusion of this information in a reference such as a handbook is indicative that it constitutes common general knowledge, though, as cautioned by the applicant, evidence of its general acceptance and assimilation by persons skilled in the art is generally required to confirm that this is the case.[75] A later Sigma-Aldrich publication (GM4), provides the essentially the same information and would, on the face of it, appear to support a proposition that the information had been widely assimilated in the art. However, GM4 states that:

“Selection of the RAFT agent (ZC(=S)SR) for the monomers and reaction conditions is crucial for the success of a RAFT polymerization experiment. However, this should not be a daunting task. The effectiveness of RAFT agents is determined by the substituents R and Z and guidelines for selection have been proposed…” [emphasis added].
[75] Applicant’s submissions at [108] to [112].

82. The phrase “has been proposed” suggests to me that these guidelines had yet to be considered widely accepted in the art. This is probably a moot point since I consider that these guidelines do not actually provide clear evidence that it was common general knowledge that RAFT polymerisation could be used to prepare PAN copolymers.

83. The table shown in these documents is a guideline for identifying RAFT agents that may be useful for polymerisation of a particular class of monomer. It is not a disclosure of the preparation of a PAN copolymer using RAFT, and therefore does not support the opponent’s submissions that RAFT was CGK for the preparation of PAN copolymers. Indeed, the documents state that even with selection of a RAFT agent from the table, successful polymerisation requires careful selection of conditions. The evidence also suggests that there are other RAFT agents available in the art that are not shown in the figure.

84. While the table refers to acrylonitrile as a class of monomer, there is nothing to indicate that the product of the polymerisation is a PAN polymer rather than any other acrylonitrile-containing copolymer. The guidelines appear to be based on reported results, but the evidence provides no information on that point, including whether these were based on PAN copolymers, how many reports were used to generate the guideline, and whether reports used the full range of RAFT agents indicated as being suitable for acrylonitrile. It may be that the reports used to generate the table included the preparation of PAN polymers. But, even if that were the case, this alone would not be sufficient to elevate those particular disclosures to the level of common general knowledge.

85. On balance I do not consider that the evidence establishes that it was common general knowledge that RAFT could be used to prepare PAN copolymers, and particularly those having the characteristics defined in the present claims. To the contrary, I consider the evidence establishes that it was not CGK.

86. Of the declarants, Dr Veedu appears to have the most direct experience of the manufacture of PAN copolymers. He stated that a number of techniques can be used to make PAN copolymers, but the most commonly used methods in industry are solution-based free radical polymerisations. This is consistent with the evidence of Dr Buckmaster, who stated that PAN copolymer for use in preparing carbon fibre was conveniently synthesised by solution or suspension polymerisation.[76]

[76] Buckmaster 1 at [13].

87. Dr Veedu was aware of RAFT polymerisation techniques, but he had not used RAFT techniques in his work and was not aware of RAFT being used in the manufacture of PAN by any commercial supplier prior to 2013.[77] He stated that a balance of molecular weight and polydispersity are important factors for the production of high-quality carbon fibre.[78] Dr Veedu noted that the molecular weight of the PAN copolymer could be adversely affected by chain transfer reactions of the monomer with other reagents such as the solvent and, as RAFT agents are a form of chain transfer agent, they could act to reduce molecular weight.[79]

[77] Veedu at [27].

[78] Veedu at [24].

[79] Veedu at [28].

88. Similarly, Dr Wilson noted that acrylonitrile is a “very particular monomer” as it is more susceptible to irreversible chain transfer reactions than other monomers. This in turn leads to more branching and chain growth arresting events and a higher PDI than would be expected with other monomers.[80] Dr Wilson stated that it was a challenge to achieve a polymer with a high molecular weight and a low PDI using RAFT techniques,[81] since it was considered that RAFT gives greater control over PDI but usually results in the production of lower molecular weight polymers than obtained through conventional free radical polymerisation.[82]

[80] Wilson at [41].

[81] Wilson at [43].

[82] Wilson at [44].

89. Dr Moad disagreed with the suggestion that RAFT polymerisation necessarily lowers the molecular mass from that obtained in a conventional polymerisation and that, if desired, conditions could be selected to provide a targeted molecular weight.[83] He referred to Table 30 in D5, in which acrylate polymers (not PAN copolymers) were prepared which had similar molecular weight of their conventional counterparts, and D32 in which samples 5 and 6 had about the same molecular weight as the control sample (additionally using 2-2.5 times the quantity of monomer). He suggested that Dr Veedu and Dr Wilson seemed “to be misleading as they attempt to convey a sense of dysfunction of RAFT polymerisation by selectively presenting only a narrow commentary on the operation of RAFT polymerisation”.

[83] Moad 2 at [11].

90. Consistent with Dr Moad’s submission that RAFT does not necessarily lower molecular weight in RAFT polymerisations, I note that in addition to the examples cited where there was no decrease in molecular weight compared to controls, there are also uncited examples where there is a significant drop in molecular weight (D5, tables 22, 32 and 33; D32, samples 2 and 4 which differ from the control only in the addition of the RAFT agent). But contrary to the inferences Dr Moad drew, I do not consider that Dr Wilson and Dr Veedu are attempting to misrepresent the benefits that RAFT polymerisation can provide or that they have been selective in their evidence. I understand their evidence to relate to their knowledge of the art in relation to PAN copolymers and RAFT at the time.

91. To this end there appears to have been a technical prejudice against the use of RAFT for the polymerisation of PAN copolymers, and particularly for the preparation of PAN copolymers with a suitable balance of PDI and molecular weight for carbon fibre production. This technical prejudice appears to have support in the contemporaneous document provided in evidence as DJW-4 (the CSIRO paper).[84]

[84] Jasjeet Kaur et al., Producing high-quality precursor polymer and fibers to achieve theoretical strength in carbon fibers: A review, J. Appl. Polym. Sci., 2016, DOI: 10.1002/APP.43963.

92. The parties differed significantly as to the conclusions the paper drew concerning the state of the art in PAN copolymerisation. I have taken these conflicting opinions into account in attempting to undertake an objective analysis of the paper and resolve this conflict.

93. The CSIRO paper was published in 2016, about 2 years after the priority date of the present application and reviews the production of high-quality carbon fibre from PAN copolymers. The paper purports to highlight “opportunities to enhance the quality of polyacrylonitrile-based precursor fibres”. The paper notes that there is scope to improve the average tensile strength of carbon fibres based on PAN fibres since they have attained about 70% of their theoretic modulus, but only 10% of their theoretical strength. It suggests that precursor fibre should have high molecular weight (>200 kDa), but lower molecular weight fibres are used for practical purposes since the higher molecular weight polymer is difficult to dissolve due to its high PDI.

94. The paper states that the polymerisation conditions used strongly influence the molecular weight, PDI and molecular defects occurring in the produced polymers. Solution, aqueous suspension, and solvent-water suspension methods have been used for production of PAN. Conventionally as the molecular weight of the polymer increases, so too does its PDI. Obtaining a high molecular weight polymer with low PDI is therefore a challenge.

95. The paper goes on to say that in more recent times considerable attention has focused on using living or controlled radical polymerisation such as RAFT and ATRP. RAFT is said to be one means of achieving better control over molecular weight and PDI, with the work of Dr Moad having “opened up space for using RAFT polymerization for the manufacture of the PAN-based polymers for carbon fiber production, as recently shown by Cai”.[85] The Cai patent document dates from around the same time as the present application. I understand the CSIRO paper to suggest this work represented a significant development in the art. It was not published at the priority date of the present application.[86]

[85] Cai is a patent application made by CSIRO which describes preparation of PAN having a high molecular weight and low PDI. The PAN polymers are made using RAFT techniques.

[86] This patent document was not published at the priority date of the present application and was not entered into evidence. While the parties made verbal representations at the hearing, I have considered the document and do not consider the information provided in the document is crucial to the present determination beyond the mention it receives in the CSIRO paper.

96. The paper then states there are very few reports on the RAFT polymerisation of acrylonitrile because of its high reactivity. The paper sets out some desirable outcomes such as control over molecular weight, PDI and conversion yields (>70%). Lack of high molecular weight fractions due to low PDI can apparently present certain challenges, such as low structural viscosity.

97. Several documents are cited that use RAFT polymerisation to prepare PAN polymers – most of these are said to suffer disadvantages: while low PDI is achieved, molecular weights are also low; high molecular weight and low PDI are achieved but for reasonable conversion yields longer reaction times were required. The paper suggests that a topical research question is to see if such polymers can be spun into fibres and thermally converted to produce carbon fibres with high average tensile strength.

98. The present application is cited as providing carbon fibres with uniform cross-section and low microdefects, as well as the average tensile and tensile modulus mentioned above. In the context in which these figures are cited, I consider the paper to suggest that these are considered an advantageous feature of the fibres produced by the present invention.

99. The paper also mentions the use of ATRP for the preparation of PAN fibres with a molecular weight of 120,000 g/mol but suggests the PDI of about 2 was too high.

100. On balance I consider the CSIRO paper indicates that the manufacture of PAN polymers having high molecular weight and low PDI and that were suitable for the production of high-quality carbon fibres had presented challenges. This is consistent with the evidence of Dr Veedu and Dr Wilson. Focus had shifted to the use of RAFT polymerisation as a means of better controlling molecular weight and PDI. The present work and that of other researchers around 2014 appear to have represented significant advances in the achieving those aims, but there remains scope for improvement to achieve fibres that are suitable for high quality carbon fibres.

101. Moreover, the reactivity of acrylonitrile appears to have deterred research on the use of RAFT with acrylonitrile, and as a consequence there are very few reports of its use. Dr Moad argued that the absence of reports on the RAFT polymerisation of acrylonitrile was due to issues relating to the handling and toxicity in a laboratory setting rather than any thought that acrylonitrile might not be amenable to control using RAFT polymerisation.[87]

[87] Moad 2 at [60].

102. I find this argument unconvincing. Dr Moad’s line of argument suggests that there would be a general technical prejudice against the use polyacrylonitrile in general, rather than in RAFT alone. I see no evidence of a general reluctance to use acrylonitrile in any type of polymerisation, and the CSIRO paper clearly states that it is the result of the reactivity of acrylonitrile. Similar comments are made in D32 and D33.

103. On balance I prefer the evidence of Dr Wilson and Dr Veedu on this point. It was not common general knowledge that RAFT was the technique of choice for the preparation of PAN copolymers. To the contrary, polymer scientists working in the field of PAN copolymers may have been aware of living polymerisation techniques, including RAFT, but there was a technical prejudice that RAFT presented challenges in achieving the appropriate balance of PDI and molecular weight. At the time (and indeed later), achieving a suitable balance for use in the preparation of carbon fibres was considered a considerable advance.

104. On that basis I do not consider that the skilled person would, as a matter of routine, been led to try the defined invention with a reasonable expectation of success in view of the common general knowledge alone.

Inventive step in light of CGK combined with section 7(3) documents

105. The opponent relied on the following documents under this ground:

WO98/01478 (D5)
Xiao-hui Liu et al., “Well defined higher-molecular-weight polyacrylonitrile via RAFT technique in the presence of disulfide compounds as a source of chain transfer agent”, European Polymer Journal 44 (2008) 1200-1208 (D32)
KR20120115029 (based on an Espacenet machine translation provided by the opponent in evidence) (D26)

US2010/0003515 (D20)
WO2008063886 (D31)

106. I will consider each of these in turn.

WO98/01478 (D5)

107. D5, of which Dr Moad is a co-inventor, discloses polymerisation using thiocarbonylthio RAFT agents. The opponent submitted that “D5 presents extensive experimental detail supporting the efficacy of RAFT in achieving polymerisation of a plethora of polymers having PDI consistently below 2”. They relied on the evidence of Dr Such who, based on Example 67 of D5 and the general teachings in the page 24, lines 5 to 17,[88] arrived at a process that is said to replicate the present invention.[89] The applicant disagreed with this submission and submitted that the suggested process drastically changes the reaction conditions from those in Example 67, and that the person skilled in the art would have no understanding, let alone an expectation, as to what type of polymer might be obtained from such a vastly different process.[90]

[88] Such 1 at [42] to [43].

[89] Opponent’s submissions at [172] to [174].

[90] Applicant submissions at [215].

108. The applicant also noted that Mr Such uses the language of the opposed application (that is, “comonomer” rather than specifically referring to styrene, which is the second monomer in Example 67). This was despite indicating that he was not provided with the opposed application until a later time.[91] I have nothing before me that would lead me to determine that Dr Such’s evidence was tainted by hindsight and therefore can give little weight to this submission.

[91] Applicant submissions at [217].

109. However, I consider that the questions posed to Dr Such (and as a consequence the problem to be solved) appear to be aimed primarily at upscaling a particular reaction to a large scale, rather than the choice of a particular combination of features for preparation of PAN copolymer having the defined properties. Dr Such was asked how he would have performed copolymerisation of acrylonitrile with narrow PDI,[92] and specifically how, as a matter of routine, he would have adapted the procedure in Example 67 to be performed at a large scale.[93] As noted by the applicant there are a number of differences between the process proposed by Dr Such and the process described in Example 67:

·     Example 67 is a styrene-acrylonitrile copolymer in which acrylonitrile is the minor component. Dr Such states that his process is for the preparation of PAN copolymers. These would appear to require entirely different monomer compositions, and specifically an excess of acrylonitrile.
·     Example 67 does not mention an initiator and the reaction appears to be self-initiated by heating. Dr Such does not explain this change in any detail but justifies addition of an initiator in his statement that commercial monomers contain inhibitors that “would require some adjustment including the possible introduction of some thermal initiator to assist in decomposing the inhibitor”.[94] This suggests that the initiator is added for the purpose of removing an inhibitor rather than for the purpose of the polymerisation.
·     The reaction mixture was heated at 100 degrees in Example 67 (presumably for self-initiation of the styrene copolymer), while Dr Such’s process heats the reaction mixture at 65-70 degrees.
·     Example 67 uses bulk conditions in that no solvent is used. Dr Such stated that polymerisation of acrylonitrile would be expected to be exothermic, and dissipation of heat would become a critical design feature as the production scale is increased.[95] In bulk polymerisations where no solvent is used, as polymer forms, the viscosity of the bulk mass increases and makes heat transfer through the walls of the reactor more difficult. Therefore, addition of a processing solvent would be necessary.
·     The process described in Example 67 is a batch polymerisation. The process described by Dr Such uses a semi-continuous feed process. He stated that based on the teachings of D5, he would have adapted the procedure of Example 67 to be performed as a semi-continuous feed solution polymerisation.
[92] Such 1 at [38].
[93] Such 1 at [41].
[94] Such 1 at [42].
[95] Such at [41].

110. On balance, I consider the conditions proposed by Dr Such represent a significant departure from those used in Example 67. Dr Such provides no real detail as to how he would have selected the particular RAFT agents, temperatures and the like to adapt the specific polymerisation given in Example 67 to the preparation of PAN copolymers having the properties defined in the present claims. Turning to the last of the differences above, there is mention of feed-type processes in D5, but there is also detailed discussion of a range of other process types (“emulsion, solution or suspension in either a batch, semi-batch, continuous or feed mode”).[96] Most of the examples use batch conditions, with several using emulsion processes. It is therefore not readily apparent why the PSA, based on the disclosure in D5, would be led to use a semi-continuous mode over any other approach.

[96] D5, page 23, line 28 to 30.

111. In response to evidence from Dr Wilson and Dr Veedu questioning the motivation to select such conditions, Dr Such referred to exhibit CS-17, which states that removal of this heat was easy in small reactors but became a problem in larger ones. He referenced a passage which noted that “heat removal is easier in semicontinuous operations where monomer and/or initiator is added gradually”.[97] This would suggest semi-continuous feed modes would be the natural choice for upscaling a reaction, but I note at the same place in CS-17 there is a discussion of alternative reactor arrangements using internal cooling baffles or bundles of cooling coils. The document goes on to state that one advantage that batch polymerisation has over continuous polymerisation is that the polymerisation time may be kept exactly uniform for the entire batch, resulting in a product of narrow specifications, such as molecular weight distribution (PDI).[98] This would seem to me to teach that other considerations could come into play when determining the feed mode. In any case, as noted by the applicant, while the process arrived at by Dr Such is a semi-continuous feed process, it differs from the process defined by the present claims.

[97] Such 2 at [18].

[98] Exhibit CS-17 at page 8.

112. The present application claims a very specific process in terms of the specific RAFT agents used, and the manner in which the polymerisation is carried out in order to obtain the desired PDI and molecular weight. In my view the use of the processes described in D5 would require a de novo determination of the reaction conditions that would be employed for the preparation of PAN copolymers. D5 generally teaches that their RAFT process can be applied to any monomer or monomer combinations that are susceptible to free-radical polymerisation. A large number of monomers or comonomers are listed, including acrylonitrile, and there is general guidance for the choice of RAFT (pages 21 and 22). But there is no specific mention of PAN copolymers, and there appears to be a large degree of variability in the effectiveness in controlling PDI and molecular weight depending on the reaction conditions and monomers.

113. In my opinion there is nothing that would directly lead the skilled person, in view of this broad disclosure, to the specific combination of reagents and process steps for the preparation of PAN copolymers with the defined molecular weights and PDI as set out in the present claims. The claims are therefore inventive in view of D5.

Xiao-hui Liu et al. (D32)

114. D32 discloses RAFT polymerization of acrylonitrile using the RAFT agents CPDB and CPDPA, which are synthesised in situ by the reaction of the corresponding disulfides BTBDS and bis(thiophenylacetoyl) disulfide (BTPADS) with AIBN. CPDB and CPDPA fall within the scope of Formula (II) as defined in the present claims. CPDPA, BTBDS and BTPADS have the following structures:

115. D32 discloses that in the author’s previous work CPDB was shown to be a successful chain transfer agent for the controlled synthesis of PAN via RAFT, producing controlled/living radical polymerisation including controlled molecular weight, linear increase of “molecular weight up to 32800 with an increase in monomer conversion”, and PDI less than 1.3. Dr Moad clarified that this molecular weight value was based on the number-average molecular weight (Mn) and that this would correspond to a “much higher” Mw.[99] He did not provide the actual value, but I understand this to be about 42640 (that is, Mn multiplied by the PDI value based on the formula PDI = Mw/Mn). This is indeed higher than the cited value but is still significantly lower than the values defined in the present claims. Dr Moad also referred to another example in the previous paper that used the same conditions but significantly longer reaction time to obtain a Mw of about 50,000 g/mol and a PDI of 1.35, which he considered was “not far” from 60,000 g/mol.[100]

[99] Moad 2 at [92].

[100] ibid

116. The opponent acknowledged that D32 differed from the claimed invention in two respects: firstly, BTBDS and BTPADS are not strictly RAFT agents as claimed and, secondly, in D32 the reactants are mixed and then heated rather than the initiator being added following heating to the temperature range defined in the present claims. On the first point, they submitted that D32 uses BTBDS and BTPADS to generate CPDB and CPDPA, which are within the scope of the claims. On the second point the opponent submitted that, if moving to a larger scale, then a semi-continuous procedure would be adopted as a matter of routine.[101]

[101] Moad 1 at [96].

117. This submission appears to be based on the premise that the disulfides are “equivalent” to CPDB and CPDPA, and therefore the disclosure of the disulfides is a disclosure of CPDB and CPDPA. Following that line of reasoning, the only difference is choice of conditions to upscale the process to a large scale.

118. If this is indeed the opponent’s argument, then I do not find it persuasive. The reagents may be alternatives, and even equivalent for practical purposes, but they are not the same. A reaction using BTBDS and BTPADS in place of the RAFT agents defined in the present claims simply would not fall within the scope of the claims, even if it provided a similar product. Moreover, the use of the disulfides appears to require conditions specific to those reagents. For example, the reactions appear to use a higher ratio of initiator in reactions,[102] presumably to provide the RAFT agent in the first place. I also understand that Scheme 1, which shows the proposed mechanism for the in-situ generation of the RAFT agents, indicates that the disulfides follow reaction pathways other than the standard RAFT reactions. As a consequence, I am not satisfied on the evidence before me that the disulfides can be considered “equivalent” to the dithioesters to the extent that a disclosure of one could be a disclosure of the other, even if they do generate such agents in situ.

[102] Table 1 in D32 uses a minimum 1:1 ratio of disulfide to initiator for reaction of BTBDS. GM-20 uses a ratio of at least 3:1 for the corresponding reaction of CPDB.

119. Notwithstanding my conclusion in this regard, I note that Dr Moad and Dr Such both stated that in view of D32 they would be led to use CPDB and CPDBA in place of BTBDS and BTPADS in upscaling the reactions disclosed in D32.[103] Dr Moad stated that the disulfide reagents were, from a practical point of view, equivalent to adding CPDB and CPDPA, and for all practical purposes, the polymerisation is based on CPDB and CPDPA acting as RAFT agents.[104] He considered that the authors substituted these with the standard RAFT agents for “pure convenience”.[105] He suggested that either CPDB or CPDBA could be used in place of the disulfides.[106] In his evidence in support, Dr Such provided little explanation of why he would have made this replacement, and therefore this evidence is of little assistance. In evidence in reply, he used similar wording to Dr Moad in suggesting the use of the reagents were “convenient”.[107] I consider Dr Such’s evidence adds little additional value and I can limit my consideration to Dr Moad’s evidence.

[103] Moad 1 at [97], Such 1 at [59].

[104] Moad 1 at [94].

[105] Moad 2 at [94].

[106] Moad 2 at [97].

[107] Such 2 at [54].

120. As an initial point, I do not consider that there is any evidence to establish that this would be the case for CPDB. D32 references the author’s earlier paper,[108] which uses CPDB and varies a number of parameters with the aim of optimising the polymerisation reaction. These include the reaction temperature, reaction time, monomer concentrations, addition of comonomers, and the molar ratios of the monomer, RAFT agent and initiator. None of the changes increased the molecular weight to the range defined by the present claims. BTBDS provides no better results than those described in the earlier paper. I consider that there is no evidence to support the submission that the skilled person would be led to use that reagent and any of the conditions disclosed in D32 in the expectation that they would obtain the polymers defined in the present claims.

[108] Exhibit GM-20.

121. The remaining question is whether the teaching in D32 in relation to BTPADS would motivate the skilled person to use CPDPA using the claimed invention to obtain a PAN copolymer having the particular molecular weight and PDI defined in the present claims.

122. Dr Wilson considered that D32 taught, that in order to obtain PAN with a higher molecular weight, BTPADS should be used in place of CPDPA and that they are not alternatives for one another.[109] He noted the statements in D32 that the most commonly used dithioester RAFT agents are difficult to prepare and purify, and the aim was to find a RAFT agent with those properties that is suitable for preparing well-defined PAN polymers with higher molecular weight and lower PDI.[110]

[109] Wilson at [64] to [67].

[110] Wilson at [66].

123. He considered that the processes shown in D32 only show good control at relatively low molecular weights (Figures 4 and 5). He noted that increases in the ratio of monomer to disulfide and initiator did not result in increased molecular weight (Table 1), which suggested to him that the polymerisation was not well controlled and did not give good control of molecular weight.[111] As a consequence he considered the reagents were not truly equivalent to the CPDB and CPDPA and referred to the process as a degraded RAFT type process. He considered this would discourage the use of a true RAFT process to achieve improvements in molecular weight and PDI.[112]

[111] Wilson at [63].

[112] Wilson at [69].

124. Dr Moad disagreed with Dr Wilson in his reference to the process as a degraded RAFT process and referred to Figures 9 and 10, which he considered show linear evolution of the molecular weight with conversions up to 90000 g/mol. He considered that the article merely taught that the disulfide agents do not convert efficiently into the corresponding RAFT agents. As a result, very small amounts of the agent are formed which are nevertheless effective in promoting a well-controlled and efficient polymerisation. He pointed to the data in Figures 4, 5, 9 and 10 in support of this assertion.[113] He went on to say that based on the inefficient in situ conversion of the disulfides to the RAFT agents, he would be motivated to use the dithioester RAFT agents directly and at low concentration.

[113] Moad 2 at [98].

125. On balance, I am not satisfied that the skilled person would, in view of the teaching provided by D32, be led to replace BTPADS for CPDPA in the expectation that it would provide a polymer with the required molecular weight and PDI. As noted above, based on the evidence before me, I do not consider the reagents are so equivalent that the PSA would view then as a one for one substitution without changes in the conditions. The only motivation for doing so appears to be based on the inefficient conversion of the disulfide in situ to the RAFT agent. In this regard, D32 identifies a non-zero intercept (approximately 10000 in each case) in Figures 9 and 10. The authors explain this in terms of the presence of some hybrid conventional free radical polymerisation, which is said to be caused by less available diothioester agents formed in situ from the reaction of the disulfide with AIBN.

126. This observation appears to suggest that the polymerisation using BTPADS does not follow standard RAFT processes and would be more in line with the evidence provided by Dr Wilson. Rather than teaching towards the use of a reduced amount of the standard RAFT agent, I consider this would suggest that polymerisation using BTPADS is not directly analogous to that using CPDPA and would teach away from an approach of replacing BTPADS with CPDPA.

127. I therefore consider that the claims are inventive in view of D32.

KR20120115029 (D26)

128. D26 describes the synthesis of PAN copolymers by conventional free radical processes. The document specifically relates to the preparation of acrylonitrile copolymers having ethylene monomeric units that are a,a- disubstituted by carboxylate- and nitrile groups. The examples use dihexylammonium 2-cyanoacrylate to introduce this monomer. The process examples involve heating DMSO to 60 degrees with stirring, adding the monomers and then, after stirring for a “certain time”, adding AIBN. The mixture was held at 60 degrees for 16 hours.

129. The polymers produced have molecular weights between 62,600 and 99,900. PDI ranges from 2.09 to 3.19. Comparative examples with a PAN homopolymer and an acrylonitrile-itaconic acid copolymer are also provided which have molecular weight and 110200 and 137600 respectively and PDIs of 2.04 and 1.81 respectively.

130. The opponent submitted that the order of heating and addition of reagents in D26 corresponds to that of Claim 1.[114] Notably D26 does not use a RAFT agent in the polymerisation, nor is there any suggestion of their use. There is another difference in that the monomers are added to the solvent after it has been brought to the required temperature, but I do not consider this a significant difference and nothing turns on the issue.

[114] Opponent’s submissions at [200].

131. The opponent submitted that it would be obvious to perform the process of D26 using a RAFT agent with the expectation of reducing the PDI further relative to the values shown in D26.[115] They submitted that if seeking to reduce the PDI, one would not have varied the parameters of the reaction described, such as temperature, reaction rate, kinetics or initiator concentration because there is no teaching of these changes in D26 and these would not lower the PDI but rather would result in a change in the molecular weight.[116]

[115] Opponent’s submissions at [201].

[116] Opponent’s submissions at [206].

132. Contrary to the opponent’s submission, the document teaches a number of alternative conditions that can be employed in the reaction (counter ions on the carboxylate group, initiators, monomer concentration in the solution, polymerisation temperature). The key variation lies in the comonomer, which is said to be suitable for the replacement of existing comonomers (methyl acrylate, methyl methacrylate and itaconic acid) and provide benefits for the production of carbon fibre.[117]

[117] Applicant’s submissions at [249].

133. The opponent’s submissions on D26 appear to be based on the premise that a RAFT agent can be added to any conventional free radical polymerisation with minimal change required to implement the different process and minimal impact on molecular weight. However, as discussed previously, it was known in the art that a balance of molecular weight and PDI was a challenge in the preparation of PAN polymers. There is no teaching or suggestion in D26 that would lead the skilled person, with any expectation of success, to consider the use of RAFT polymerisation in preference to any other chain transfer agent (or specific RAFT agent), changes to the existing processes provided in D26, or indeed to use any other means of balancing the PDI and molecular weight.

134. On balance I consider that the claims are inventive in view of D26.

US2010/0003515 (D20)

135. D20 discloses PAN polymers which are suitable for the manufacture of carbon fibres and have reduced fibre breakage and can be spun at increased spinning speed.

136. The PAN polymers prepared in D20 are effectively bimodal – that is, the polymer is a mixture of two polymers having different molecular weights and PDIs. In the preferred embodiment the first polymer has a Mw of between 1 and 1.5 million g/mol, and the second component has a Mw of 150,000 – 1 million g/mol. As a bimodal system, each of the components has its own Mw and distribution, which results in a molecular weight distribution which is higher than the component polymers taken alone.[118]

[118] Veedu at [46].

137. There is no disclosure of a PAN polymer having the defined molecular weight and PDI of the present claims. The opponent submitted that any contention that PDI was not a consideration in D20 was unfounded as it teaches that a PDI of 10 or less is preferable. They argued that the document taught that a PDI above 2 could still achieve carbon fibres having similar properties to those of the present application.[119]

[119] Opponent’s submissions at [212].

138. They also submitted that the process of example 15 provided a polymer with Mw of 460 kg/mol and PDI of 2.6. They argued that the process would be easily transferable into large scale manufacture and that it would have been obvious to substitute the octyl mercaptan chain transfer agent with a RAFT agent with the expectation that it would reduce the molecular weight.[120]

[120] Opponent’s submissions at [209].

139. The applicant noted that Example 15 describes a 2-stage process. The first stage provides a high molecular weight polymer but in low concentration due to the use of small amounts of initiator and chain transfer agent. The second stage involves addition of further reagents and further heating. This gives a polymer of lower molecular weight with greater conversion.[121]

[121] Veedu at [47].

140. On balance I do not find the opponent’s submissions persuasive. D20 does not teach or suggest a polymer having the defined properties, but instead is aimed at solving the problem of achieving better processability of PAN fibres through a different approach. There is nothing in D20 that would lead the skilled person to the specific conditions and products defined by the present claims.

WO2008063886 (D31)

141. D31 discloses methods of making carbon fibres. The opponent asserted that this document is relevant to Claims 12 to 15, but the obviousness of these claims is wholly reliant on obviousness of Claims 1 to 11. As I have found these claims to be inventive, it follows that the claims are inventive in view of D31.

Conclusion on inventive step

142. The claims are inventive in view of the common general knowledge alone, and in view of the cited documents when considered in light of the common general knowledge.

Section 40(2)(a) - Clear and complete disclosure

143. Subsection 40(2)(a) of the Act requires that a complete specification disclose the invention in a manner that is clear enough and complete enough for the invention to be performed by a person skilled in the relevant art.

144. In CSR[122] the delegate adopted a three-step test for determining whether the specification provided a clear enough and complete enough disclosure of the claimed invention as follows:

What is the scope of the invention as claimed?
What does the specification disclose to the skilled person?
Does the specification provide an enabling disclosure of all the things that fall within the scope of the claims?
[122] CSR Building Products Limited v United States Gypsum Company [2015] APO 72.

145. An expanded approach was taken in Evolva.[123] The third consideration of enablement set out in CSR was assessed according to the following criteria:

Is it plausible that the invention can be worked across the full scope of the invention?
Can the invention be performed across the full scope of the claims without undue burden?
[123] Evolva SA [2017] APO 57

146. In general, the extent of the disclosure necessary to make the patent sufficient will depend upon the nature of the invention, the scope of the claims and the art in which the invention is made.

147. The opponent made submissions in respect of the following issues:

There is a lack of guidance on working combinations of features.
The procedure of Example 1 could not be repeated without significant trial and error.
There is no, or conflicting, guidance on the selection of initiator.
The diameter of the fibres is not given.

148. I will deal with the last of these summarily. Dr Buckmaster stated that the specification does not provide the diameter of the carbon fibres. He stated that the tensile strength and modulus of the carbon fibres depend strongly on the diameter of the tested fibre.[124] He considered that the definition of these properties per the ASTM D4018 test bear little weight unless the diameter is also given. He referred to various texts (DB-9 and DB-10),[125] but provided no specific evidence in relation to the requirements of ASTM D4018. I have read the standard and my understanding is that it does not require a measure of the fibre diameter. The defined properties appear to be derived from the density and mass per unit length measurements. No evidence was adduced that would lead me to determine otherwise. Even if that is not the case, there is no evidence to suggest that the PSA could not make something with the desired properties and the specification, on its face, indicates they can. This ground has not been made out.

[124] Buckmaster 1 at [59].

[125] Buckmaster 2 at [27] and [28].

149. In arguing that the specification does not provide sufficient guidance as to working combinations of features, the opponent referred to the evidence of Dr Such who stated that “performing the claimed procedure does not inherently guarantee that the resulting PAN polymer would have a molecular weight within the claimed range.”[126]

[126] Such 1 at [81].

150. Similarly, Dr Moad stated that he would be unable to repeat the procedure of Example 1 without significant trial and error due to a lack of details on a number of parameters (amount of solvent, concentration of monomer and comonomers, concentration of initiator, pre-heating temperature of step c) and the temperature at which the co-monomers and the RAFT agent are added).[127]

[127] Moad 1 at [121] to [124].

151. The applicant argued that the person skilled in the art would have no difficulty reproducing the method described in the specification, including Example 1. In each of the examples, the ratio of monomers is provided. The RAFT agents are given as a mole percent based on the total amount of monomers and the monomer concentration can be calculated from the final polymer concentration assuming a quantitative conversion. The amount of initiator and polymerisation temperature are variables that could be determined based on the specification, and, while it may be necessary to experiment in order to get results close to those provided in the examples, this would not be difficult to do.[128] The applicant commented that in seeking to replicate the invention the opponent had no difficulty in working the invention. They considered this a clear indication that the specification provides sufficient information to perform the invention.[129]

[128] Applicant’s submissions at [362].

[129] Applicant’s submissions at [363].

152. I find these submissions persuasive. As noted in Evolva, the specification must be clear enough and complete enough for the invention to be performed by a person skilled in the relevant art. The legislation sets a minimum requirement, and it is long-established law that the skilled reader will draw upon their common general knowledge and skills to work the invention. This was reflected in Lord Hoffmann statement in Kirin-Amgen:

“The skilled person is taken to be trying to make the invention work. If the skilled person would quickly realise that one method would work and another would fail, the specification is not insufficient because the claim is expressed in terms broad enough to include both methods.”[130]
[130] Kirin-Amgen Inc v Hoechst Marion Roussel [2005] All ER 667; [2004] UKHL 46; [2005] RPC 9 at [128].

153. Notably, the person skilled in the art for the consideration of sufficiency is the same as for addressing inventive step, but they have possession of the invention and they are trying to carry out the invention and achieve success. This contrasts with inventive step where the skilled person is searching for a solution in ignorance of the invention. I agree with the applicant’s submissions that the specification, while not exhaustive in the conditions under which the examples are performed, provides sufficient information that they could be performed with a degree of trial and error, but not so extensive to require inventiveness.

154. The other issue raised by the opponent related to the use of conditions under which the initiator would practically not work. They argued that the person skilled in the art would be required to test each known initiator to exclude those that do not work. I do not consider this submission to be consistent with the legal principle that the skilled person is approaching the specification with the intention of making the invention work. To this end, the specification provides a range of initiators that can be used, so the skilled person would not be approaching the task from point of view of having no guidance. The conditions under which a particular initiator would be used appears to be well-known in the art and would act as a starting point in any routine trial and experimentation. Moreover, the example specifically uses AIBN, so there is a clear starting point as to the choice of initiator.

155. The additional concern in this regard is that the claims included initiators that would not give a practically useful polymerisation under the claimed conditions. The first is AIBN, which has a T_1/2 of 65 degrees, and would not be practically useful at temperatures at the lower end of the claimed range. I do not see this as a significant issue since the person skilled in the art would select conditions that were suitable for the particular initiator.

156. In the case of TBPO, which has a T_1/2 of 123 degrees, temperatures outside of the claimed range would be required to initiate the polymerisation. The opponent referred to the evidence provided by Dr Evans in support of these submissions,[131] but I do not consider that these add much to the evidence provided based on known properties of this initiator. For example, Dr Moad recognised that there would be difficulties in using TBPO within the range claimed.[132]

[131] Evans 1 at [34] and [35].

[132] Moad 1 at [141].

157. Dr Wilson considered that increasing the amount of TBPO could achieve polymerisation at 85 degrees, although it would increase the time for completion of the polymerisation reaction.[133] Alternatively, he suggested addition of a small amount of reducing agent.[134]

[133] Wilson at [86].

[134] Wilson at [102].

158. Dr Evans noted that increasing the initiator would be contrary to the teaching of the specification that the amount of initiator should be less than 1%. He considered that this would risk residual initiator contaminating the polymer and would be a safety concern during the polymerisation since having large amounts of initiator sitting unreacted increases the risk of runaway. He also considered that while reducing agents were known to catalyse polymerisation of acrylates, the use of reducing agents to manage the initiation of TBPO is not common practice.[135]

[135] Evans 2 at [16] to [22].

159. On balance, the submissions of Dr Evans do not lead me to conclude that the TBPO would be entirely unsuitable in the temperature range defined and there is a plausible, if not commonly used, means of managing the initiation.

160. In short, the opposition is unsuccessful on the ground of section 40(2)(a).

Section 40(3) – claims not supported

161. Section 40(3) requires that the claims must be supported by matter disclosed in the specification. The test for support was provided by the delegate in CSR[136] as follows:

(i) Construe the claims to determine the scope of the invention;

(ii) Construe the description to determine the contribution to the art; and,

[136] CSR, supra.
(iii) Decide whether the claims are supported by the technical contribution to the art

162. There is a reasonable amount of overlap between the submissions made under the ground of section 40(2)(a) and this ground. The opponent made submissions on the following:

Claim 1 includes TBPO which would not promote polymerisation.
Claim 1 does not encompass all essential features.
Claim 1 is not limited to all initiator being added in step c).

163. I have dealt with the first issue under the ground of sufficiency. I do not see it necessary to reconsider this point under support, nor that it will give a different outcome.

164. The opponent’s submissions under the second point were that with the exception of the consistory statement, the only example of a procedure according to the claimed invention is given in Example 1.[137] In particular the opponent noted that the general procedure given in the example comprises an additional and separate heating step (step e below) following the addition of the initiator (step d):

a)Metering DMSO from DMSO storage tank to a reactor, then AN from AN storage tank to the reactor;
b)Purging the reactor with nitrogen;
c)Preheating the reactor and adding comonomers and RAFT agent into a reactor at above room temperature (25℃);
d)Heating up the reactor and then adding the initiator/catalyst at the desired temperature point of 40-85℃;
e)Starting the polymerisation for the time of 15 to 23 hours at the temperature of 60-80℃.
[137] Opponent’s submissions at [236].

165. The gist of the argument is that the procedure of Example 1 is tailored to the specific conditions and reagents, including AIBN. They argued that the temperature range includes lower values at which the initiator would not work. Accordingly, the use of an initiator such as AIBN at 40℃ would require a further heating stage to increase the temperature to where initiation would occur.[138]

[138] Opponent’s submissions at [239].

166. I do not find this submission persuasive. The submissions put forward by the opponent place significant emphasis on the example, but the second arm of the CSR approach to determining support requires a consideration of the specification as a whole. The first arm of the test requires the scope of the claim to be determined.

167. To that end, the scope of the claims is a method of synthesising specific PAN copolymers having particular molecular weight and PDI with specific RAFT agents, using a specific temperature range and mode of addition of reagents. I consider the contribution to the art, reading the specification as a whole, is consistent with this claimed invention. I acknowledge that the general conditions outlined in the example certainly include lower temperatures than optimal for the use of AIBN. However, the process is drafted in general terms and refers in that regard to the temperature being at a “desired temperature” within the range 40-85℃. The general conditions may well include heating regimes outside of those defined by the claims. Those simply would not fall within the scope of the claims. There is nothing that would lead me to consider that these outliers should be taken to constitute the technical contribution over what the specification as a whole indicates is the technical contribution.

168. As an aside, this, to a large extent, is a sufficiency consideration, and as noted above under my sufficiency consideration, if a person skilled in the art would recognise that a particular condition would not work then they would move on and use one that they would expect to work.

169. The third submission the opponent made is that the claim is not limited to where the initiator is added at step c). I have already dealt with this issue in the construction section above and see no need to reconsider it here.

170. In summary the claims are supported.

Section 40(3) – clarity of claims

171. Subsection 40(3) requires that the claims must be clear. A claim will lack clarity if a third party would be unable to ascertain whether an act would fall within the scope of the claim.[139] The question is therefore whether the issues identified by the opponent, even if an error, would be so unclear as to result in the person being unable to ascertain whether the act would fall within the scope of the claim.

[139] Monsanto Co v Commissioner of Patents (1974) 48 ALJR 59.

172. The opponent made submissions on a number of issues under the ground of clarity.[140] I have dealt with each of these in turn.

The meaning of “affect”
[140] Opponent’s submissions at [247] to [262].

173. This issue was dealt with in the construction section above. The term is clear.

Errors in the formulae in claim 1

174. Formulae (I) and (II) contain errors in that the subscripts of the Z groups in each are absent, as shown below:

175. I do not consider that these errors result in a lack of clarity to the extent of the person being unable to ascertain whether the act would fall within the scope of the claim, and none of the declarants appear to have any difficulty in that regard. For example, Dr Moad considered these formulae to be identical to those provided in the description.[141] The error appears to be in the nature of an obvious error in which it is obvious that there is an error, but the skilled person would immediately recognise the correct meaning. While there is error in the claim, I do not consider this results in the scope of the claim being unclear.

[141] Moad 1 at [133].

176. The opponent also noted that in each of the groups Z₁, Z₃ and R₃, the definition of the group -CH₂(CH₂)₁₀CH₃ is redundant in view of the definition –(CH₂)_nCH₃ where n is 0-20. I do not consider this an issue that would result in any uncertainty as to the scope of the claims.

The meaning of PDI in Claim 1 and the “alternate” PDI in Claim 3 are not true alternatives

177. The opponent noted that the specification provided two definitions of PDI – one based on a measure of Mw/Mn and one based on Mz/Mw. I have discussed these in my background to the invention. The opponent considered that the claim was unclear as to which meaning was intended. I disagree. The latter measure is referred to as an alternative. I consider that in the absence of an indication to the contrary that the weight average measure Mw/Mn would be assumed.

178. In the case of Claim 3, the issue appears to be that these are not alternatives but are determined according to different methods and relate to different parameters of the polymer chains. There is no apparent dispute that the person skilled in the art would be able to determine these parameters for any given polymer. I understand the claim to define that the claim is defining the same matter in two different ways. Ultimately it is dependent on claim 1, which uses the weight average measure. I do not consider that the claim lacks clarity.

The meaning of molecular weight

179. The opponent submitted that the definition of molecular weight without specifying whether it is weight average, number average, size average or viscosity average renders the claim unclear. I disagree. In the absence of an indication otherwise, and consistent with the usage of the term throughout the specification,[142] I consider it would be read as the weight average (Mw).

Claim 4 is unclear in relation to the solvents
[142] Applicant’s submissions at [411].

180. The opponent noted the definition of the solvents in Claim 4 and in particular:

“wherein the solvent is selected from the group consisting of: dimethyl sulfoxide (DMSO), dimethyl formamide (DMF), and dimethyl acetamide (DMAc), a mixture of zinc chloride (ZnCl₂) and water, and a mixture of sodium thiocyanate (NaSCN) and water.”

181. The opponent argued that it is unclear whether this definition requires that dimethyl acetamide is used alone or combined with dimethyl formamide. This derives from the term “and” in the definition of DMAc. I consider it is clear that DMF and DMAc are alternatives.

Claim 6 is self-dependent

182. The opponent is correct in identifying that the claim is self-dependent. This alone does not render the claim unclear in scope. Nothing in evidence suggests that this results in a lack of clarity to the extent that a person could not ascertain whether or not an act would fall within the scope of the claim. In the first instance, I would understand the claim to be appended to Claim 5, but in any case, ultimately links the definition back to the corresponding definition in Claim 1. I do not consider this error renders scope of the claims unclear.

Claim 9 is redundant

183. The opponent noted that Claim 9 is redundant since it is ultimately dependent on Claim 1, which defines the same temperature range. I do not consider this a significant clarity issue.

Claim 14 has no antecedent

184. This issue appears to stem from the claim referring to a temperature range, whereas the claims on which it depends define no temperature range. I do not see where the issue lies in this regard. Claim 12 defines an oxidizing step as part of the manufacture of the carbon fibre. Claim 14 further characterises that step by defining the temperature. This is a common form of claim drafting and I see no clarity issue.

185. In summary I do not consider that the claims lack clarity.

Priority date

186. The opponent submitted that the claims are not entitled to the priority date claimed (23 December 2013), but instead should be afforded a priority of 16 March 2018. This is the date on which the claims were amended during prosecution. A consequence of this later priority date is that an additional citation becomes relevant for the ground of obviousness (D33).[143]

[143] Spörl et al., Carbon fibres prepared from tailored reversible-addition-fragmentation transfer copolymerization-derived poly(acrylonitrile)-co-poly(methylmethacrylate), Journal of Polymer Science, Part A: Polymer Chemistry 2014, 52, 1322-1333, published on-line 12 February 2014)

187. Section 102(1) is relevant here and provides that:

“(1) An amendment of a complete specification is not allowable if, as a result of the amendment, the specification would claim or disclose matter that extends beyond that disclosed in the following documents taken together:
(a) the complete specification as filed;
(b) other prescribed documents (if any).”

188. I note that the failure to meet the requirements of section 102(1) must be as a result of the amendment.

189. Section 114(1) of the Act provides that, if:

(a)a complete specification has been amended; and
(b)the amendment was not allowable under subsection 102(1); and
(c)as a result of the amendment, a claim of the amended specification claims an invention that:
(i)was not disclosed by the complete specification as filed in a manner that was clear enough and complete enough for the invention to be performed by a person skilled in the art, but
(ii)is disclosed in that manner by the amended specification;

the priority date of the claim must be determined under the regulations.

190. Regulation 3.14 provides for the priority date under subparagraph 114(1)(c)(ii) above as the date of filing of the amendment.

191. The requirements of subsection 102(1) were recently considered by Beach J in CSIRO.[144] One approach to this consideration is to ask “whether a skilled person would, upon looking at the amended specification, learn anything about the invention which he could not learn from the unamended specification”.[145] To this end, a comparison is made between the specification as proposed to be amended and the unamended specification to determine whether subject matter relevant to the invention has been added. This comparison is a strict one. Subject matter will be added unless it is “clearly and unambiguously disclosed in the application as filed”, including matter which may have been disclosed both explicitly and implicitly.

[144] Commonwealth Scientific and Industrial Research Organisation v BASF Plant Science GmbH [2020] FCA 328.

[145] Richardson-Vicks Inc’s Patent [1995] RPC 568 at [576].

192. The marked-up amendment dated 16 March 2018 is set out below. Deletions are marked as strike-through and additions as bold. Formulae have been omitted for sake of brevity.

“A method for synthesizing a polyacrylonitrile (PAN) polymer with a narrow molecular weight distribution, the method comprising:

(a) combining acrylonitrile (AN) monomer with a solvent, at least one co- monomer, and a thiocarbonylthio compound to form a solution;
(b) heating the solution to a temperature within the range of 40°C - 85°C ~~above 25^oC~~; and

(c) adding an initiator to the heated solution to affect polymerization reaction,

wherein polymerization is affected by controlled/living radical polymerization, in which the thiocarbonylthio compound functions as a Reversible Addition/Fragmentation Chain Transfer (RAFT) agent,
wherein the thiocarbonylthio compound is selected from the following structures: ...,
wherein the PAN polymer has a polydispersity index (PDI) of about 2 or less and a molecular weight within the range of 60 kg/mole to 500 kg/mole.”

193. The opponent submitted that there was no requirement in Claim 1 prior to amendment that the initiator be added to the “heated” solution of step c), nor that the PAN has any specific molecular weight. They argued that there is no indication in the complete specification as filed that there is an invention in the addition of initiator to the heated solution or that this feature has any technical significance. The addition of initiator to a heated solution is only disclosed in Example 1, and that step therefore would only be in the context, conditions and sequence of procedural steps in that example.[146] They argued that example 1 is a very specific example that is designed for very specific reagents, and that it would not be applicable to producing a copolymer having the desired molecular weight characteristics using the entire scope of reagents covered by Claim 1. For example, the claim covers solvents that would not work and initiators that would not promote polymerisation under the conditions defined.[147]

[146] Opponent’s submissions at [267] to [270].

[147] Opponent’s submission at [273].

194. The applicant provided the following table identifying where, in the specification as filed, the features referred to by the opponent were disclosed:[148]

[148] Applicant’s submissions at [337].

Claim Feature	Disclosure in Original Specification / Priority Document
heating the solution to a temperature within the range of 40°C - 85°C	Claims 1 and 10; Page 3; Example 1
adding an initiator to the heated solution to affect polymerization reaction	Claim 1; Page 3; Example 1
wherein the PAN polymer has a polydispersity index (PDI) of about 2 or less and a molecular weight within the range of 60 kg/mole to 500 kg/mole.	Claims 1 and 2; Page 3; Page 9; Example 1

195. The key consideration here is whether the skilled person would learn something from the specification as proposed to be amended that they would not learn from the specification as filed. As noted by the applicant, the temperature range of 40°C - 85°C and the molecular weight range of 60 kg/mol to 500 kg/mol were explicitly defined in claims 2 and 10 of the claims as filed, and also in the description. The specification as filed clearly disclosed the addition of an initiator to the heated solution, and if anything, the amendments acted to more clearly define this requirement.

196. The opponent’s submissions in this respect, as they were in relation to their submissions on section 40, are focussed on the example and do not take into account the teaching of the specification as a whole. The issues identified by the opponent concerning the solvents and initiators appear to relate to section 40, and if there is a section 40 deficiency, this was not introduced as a result of the amendment. In this regard I have already determined that the specification and claims meet the requirements of section 40.

197. In short, I do not consider that the opponent has made out their case in relation to the priority date of the claims. I therefore do not consider it necessary to further consider D33.

Conclusion

The opposition fails on all grounds.
The applicant sought a variation on costs. While oral submissions were made at the hearing, I consider it appropriate that the parties make further written submissions on costs, including reasons for any variation in costs. I therefore allow the parties 2 weeks from the date of this decision to make written submissions on costs.

Les McCaffery
Delegate of the Commissioner of Patents

Annex