E I Du Pont de Nemours & Co v Imperial Chemical Industries PLC

Federal Court of Australia (Full Court)|2007-10-22|Before: Bennett J, Branson J, Bennett JJ, Stone JJ

WEINBERG & STONE JJ 1 We have had the benefit of reading, in draft, the reasons for judgment prepared by Bennett J. We agree with those reasons, and with the orders proposed. We wish to add some observations of our own.

Introduction 2 By further amended statement of claim dated 17 October 2006 the applicant, E I Du Pont de Nemours and Company ("Du Pont"), sought injunctive relief against the first respondent, Imperial Chemical Industries Plc ("ICI"). ICI had earlier instituted a Divisional Application pursuant to s 79B of the Patents Act 1990 (Cth) ("the Act") in relation to a compound designed to produce a refrigerant that overcame the ozone depletion characteristics of earlier heat transfer fluids. 3 Du Pont had previously been successful in opposing ICI's Parent Application, filed on 6 October 1992, which was initially accepted by the Commissioner of Patents. However, that decision was set aside by Branson J: E I Du Pont De Nemours & Co v Imperial Chemical Industries PLC (2002) 54 IPR 304; [2002] FCA 230. Her Honour found that the Parent Application lacked novelty because it had been anticipated by an earlier patent known as the "Lubrizol Patent". Her decision was affirmed by a Full Court: Imperial Chemical Industries PLC v E I Dupont De Nemours & Co (2002) AIPC ¶91-818;[2002] FCAFC 264. 4 Du Pont argued that ICI's Divisional Application should similarly be rejected, essentially on the basis of various issue estoppels which were said to arise out of the earlier opposition proceedings. Gyles J, dismissed Du Pont's application for injunctive relief, holding that there were no issue estoppels, as claimed. His Honour found that there was no identity between any issue decided by Branson J, or indeed by the Full Court, and any issue which fell for determination in relation to the proposed Divisional Application: E I Du Pont De Nemours & Co v Imperial Chemical Industries PLC (2006) 70 IPR 274; [2006] FCA 1402. Du Pont now appeals against his Honour's judgment and orders.

Background 5 The Montreal Protocol on Substances that Deplete the Ozone Layer ("the Protocol") was adopted in 1987. Australia is a signatory to that Protocol. The Protocol recognised that the amount of ozone in the Earth's stratosphere was steadily declining. It further recognised that the production of certain chlorofluorocarbons ("CFCs") was a factor in ozone depletion. The Protocol established a worldwide schedule for phasing out those CFCs which were no longer protected by patents. This provided opportunities to develop and market new, and less harmful, compounds. 6 Refrigerants are substances that by undergoing a change in phase (liquid to gas and gas to liquid) release or absorb heat and thereby achieve a cooling effect. Originally, the vast majority of refrigerants were CFCs. At a later stage, and over time, CFCs were replaced by less damaging hydrochlorofluorocarbons ("HCFCs"). However, these too contained chlorine, and were still ozone depleting. 7 Later, hydrofluorocarbons ("HFCs"), which contain no chlorine, and therefore do not harm the ozone layer, came to be developed. The best known of these HFCs is described as R-134a. It should be noted that refrigerants are classified according to a numbering system, based upon the number of carbon, hydrogen, fluorine atoms that they contain. Customarily the prefix "R" precedes that number.

The earlier judgments 8 In order to understand the issues raised in this appeal, it is necessary to consider carefully the judgment of Branson J rejecting the Parent Application. Her Honour's judgment was delivered on 12 March 2002. It concerned two patent applications, numbers 658005 (the Parent Application) and 654176 (the separate patent application). Both patent applications involved refrigerant compositions to replace existing refrigerant compositions which contained CFCs. 9 Branson J observed (at [5]) that the basic components of a refrigeration system are a compressor, condenser, expansion device and evaporator. The cooling effect is achieved by the following steps: · a refrigerant is boiled in the evaporator, absorbing heat through its walls from the refrigerated space and its contents; · the refrigerant gas released by boiling the refrigerant is piped to the compressor where it is compressed to a high pressure. The hot gas then passes to the condenser. In the condenser it loses its heat through the wall to the atmosphere and, as it cools, condenses into liquid; and · the cooled liquid, still under pressure, is piped towards the evaporator, where the pressure is reduced. This lowers the boiling point of the liquid. The refrigerant, now under low pressure, starts to boil as it enters the evaporator. There, it absorbs heat, causing it to continue to boil until, by the time it reaches the other end, the heat which it has absorbed has vaporized the refrigerant. The vapour is then drawn back to the compressor, and the cycle repeated. 10 Both Du Pont and ICI responded to the Protocol by seeking ways to develop refrigerants which did not contain CFCs. ICI developed two compositions designed to achieve that objective. 11 The first of these was contained in the Parent Application. The specification in that application stated that: "The present invention relates generally to lubricants and more particularly to working fluid compositions contained in heat transfer devices which comprise the lubricant and a heat transfer fluid." 12 Claim 1 of the Parent Application was in the following form: "A working fluid composition comprising: (A) a heat transfer fluid comprising a mixture of at least two hydrofluoroalkanes selected from the group consisting of difluoromethane, 1,1,1,2‑tetrafluoroethane and pentafluoroethane; and (B) a lubricant which is at least partially soluble in each component of the heat transfer fluid said lubricant comprising one or more compounds of general formula: O | | R(O - C - R1)n | | wherein R is the hydrocarbon radical remaining after removing the hydroxyl groups from pentaerythritol, dipentaerythritol, tripentaerythritol, trimethylol ethane, trimethylol propane or neopentyl glycol, or the hydroxyl containing hydrocarbon radical remaining after removing a proportion of the hydroxyl groups from pentaerythritol, dipentaerythritol, tripentaerythritol, trimethylol ethane, trimethylol propane or neopentyl glycol; each R1 is, independently, H, a straight chain (linear) aliphatic hydrocarbyl group, a branched aliphatic hydrocarbyl group, or an aliphatic hydrocarbyl group (linear or branched) containing a carboxylic acid or carboxylic acid ester substituent, provided that at least one R1 group is a linear aliphatic hydrocarbyl group or a branched aliphatic hydrocarbyl group; and n is an integer." 13 Claim 1 was the only independent claim in the Parent Application. However, claim 4, a dependent claim, was in the following terms: "A working fluid composition as claimed in claim 1 wherein the heat transfer fluid (A) comprises a mixture of: (1) 1,1,1,2-tetrafluroethane; (2) difluoremethane; and (3) pentafluroethane." 14 It should be noted that 1,1,1,2-tetrafluoroethane is R-134a, difluoromethane is R-32, and pentafluoroethane is R-125. Because integer (A) of claim 1 refers to a mixture of "at least two" refrigerants selected from a group of three, that claim covered four separate mixtures of refrigerants, three binary (composed of two parts or elements) and one ternary (composed of three parts or elements) as follows: (a) R-32 and R-134a; (b) R-134a and R-125; (c) R-32 and R-125; and (d) R-32, R-134a and R-125. 15 The specification in the Parent Application referred to dichlorodifluoromethane, a compound generally known as R-12. It noted that R-12 was in the process of being replaced by R-134a. It further noted that previously, heat transfer devices had tended to use mineral oils as lubricants. These had worked well with R-12 because it contained chlorine substituents. Unfortunately, R-134a could not be used as a direct replacement. This was because it did not possess comparable boiling characteristics and thermal properties and therefore would not allow mineral oil to circulate around the heat transfer device. This meant that the use of R-134a did not allow proper lubrication of the compressor. 16 There then followed a detailed description of the various components that were to make up both the heat transfer fluid and the lubricant. On the topic of lubricants, the specification went on to state that suitable lubricants could be selected from those currently used with R-134a provided that the requirement of at least partial solubility was met. 17 After summarising the salient features of the Parent Application, Branson J turned to the question whether it was deprived of novelty by reason of the prior Lubrizol Patent (patent number 54480/90). Claim 1 of that patent was in the following terms: "A liquid composition comprising: (A) a major amount of at least one fluorine-containing hydrocarbon containing 1 or 2 carbon atoms; and (B) a minor amount of at least one soluble organic lubricant comprising at least one carboxylic ester of polyhydroxy compound containing at least 2 hydroxy groups and characterized by the general formula." 18 Integer (A) of claim 1 of the Lubrizol Patent included as a possibility a ternary mixture. Further, in disclosing a large range of HFCs, it referred to ternary mixtures although it did not mention R-32 or R-125. 19 Before Branson J, Du Pont opposed the grant of the Parent Application on a number of grounds. It argued that the working fluid composition:

was not a patentable invention within the meaning of s 18(1)(a) of the Act in that it was not a "manner of manufacture" as understood in s 6 of the Statute of Monopolies;
lacked novelty when compared with the prior art base as it existed before the priority date of the claim: see s 18(1)(b)(i) of the Act;
lacked novelty by reason of the existence of the Lubrizol Patent;
lacked novelty by reason of other publications, which her Honour designated "Daikin Du Pont No 1" and "Du Pont No 2";
did not have a priority date of 11 October 1991, or alternatively 22 July 1992, as claimed, but rather a later priority date;
did not involve an "inventive step" within the meaning of s 18(1)(b)(ii) of the Act and therefore fell within the notion of "common general knowledge". In other words, the patent application failed by reason of "obviousness"; and
was not patentable because the complete specification failed to meet the requirements of s 40(2) of the Act. 20 Branson J concluded that Du Pont should succeed in its opposition to the Parent Application, but only by reason of grounds 3 and 7. Her Honour expressly rejected each of the other grounds. 21 Her Honour's decision in relation to ground 3 was based upon a series of findings. She concluded that a Mr Roger Symons was, as at the alleged priority date, 11 October 1991, a person skilled in a portion of the art to which the Parent Application related. Mr Symons agreed that the summary of the invention in the Lubrizol Patent would have disclosed to him in 1991 that integer (A) of the liquid composition claimed by that patent encompassed, at least, a single fluorine containing hydrocarbon, examples of which would have included R-134a, R-125, R-32 and R-143 and would also encompass blends of two, three, four or more HFCs. 22 Branson J was also satisfied that Professor John Ball was, in 1991, a person skilled in a portion of the art to which the Parent Application related. Professor Ball agreed that one of the things that the Lubrizol Patent would have taught him was that provided one had an ester lubricant that matched the formula of integer (B) of that patent, one would have a suitable lubricant to use with any mixtures of HFCs. 23 Branson J concluded that on the basis of her own reading of the Lubrizol Patent, and with the benefit of the evidence of Mr Symons and Professor Ball, the delegate had been clearly wrong in concluding that, as to integer (A) of the composition of the Lubrizol Patent "it is unlikely that a reader would consider that they had been taught mixtures of HFCs including R-32 or R-125 [by the Lubrizol Patent]". She added that integer (A) of the Lubrizol Patent disclosed, to a person skilled in the art to which the Parent Application related, a broad class of potential refrigerants, extending beyond those identified specifically in the preferred embodiments, with which the lubricant of integer (B) of the Lubrizol Patent might have been used. 24 These findings led Branson J to conclude that the Lubrizol Patent deprived the Parent Application of novelty. Her Honour noted that ICI had formally conceded that the Lubrizol Patent disclosed to persons skilled in the art lubricants within integer (B) of claim 1 of the Parent Application. 25 ICI challenged Branson J's judgment in favour of Du Pont by seeking leave to appeal. A Full Court comprising Heerey, Emmett and Dowsett JJ refused leave to appeal. One of the grounds upon which leave to appeal was sought complained that Branson J had erred in failing to hold that claim 4 of the Parent Application was not disclosed by the Lubrizol Patent. Put simply, ICI contended that the Lubrizol Patent did not disclose ternary refrigerant blends within claim 4. 26 In refusing leave the Full Court observed that there was no suggestion in the complete specification of the Parent Application that a liquid composition comprising a refrigerant that satisfied the description in integer (A), and a lubricant that satisfied integer (B), would be any more effective than a composition of the kind specifically contemplated by claim 1 of the Lubrizol Patent. In other words, the Parent Application did not involve the selection of something disclosed by the Lubrizol Patent by way of a further inventive step but was no more than a particular example of the invention disclosed by the Lubrizol Patent. The Full Court went on to say that it was clear therefore that a patent granted pursuant to the Parent Application would be revocable. Branson J had been correct to arrive at that conclusion. Accordingly, there would be no utility in granting leave to appeal from her Honour's orders. 27 It is necessary to say something further about the procedural history of this matter. On 28 March 2002, after Branson J had published her reasons for judgment in the first opposition proceedings, but before she had pronounced orders, ICI filed a Divisional Application from the Parent Application. Claim 1 of the Divisional Application was, at that stage, in identical terms to claim 1 of the Parent Application. 28 On 19 December 2002 an examination report issued on the Divisional Application. That report concluded that by reason of Branson J's judgment, the Divisional Application also lacked novelty. 29 On 2 April 2003 ICI proposed amendments to the Divisional Application, and in particular to claim 1. 30 On 11 August 2003 a delegate refused the Divisional Application and rejected the proposed amendments. He concluded that, having regard to Branson J's decision in relation to the Parent Application, the Divisional Application also lacked novelty. 31 Following that decision ICI appealed to the Federal Court pursuant to s 51 of the Act. ICI's appeal was heard by Crennan J. Her Honour's judgment records (at [23]) that claim 1 of the Divisional Application was proposed to be amended to provide: "A working fluid composition comprising: (A) a heat transfer fluid comprising a mixture of difluoromethane, 1,1,1,2-tretrafluoroethane and pentafluoroethane; and (B) a lubricant which is at least partially soluble in each component of the heat transfer fluid said lubricant comprising one or more compounds of general formula: (formula follows)" 32 Crennan J observed (at [24]) that the proposed amendment had the effect of narrowing the invention claimed to one specific ternary mixture of HFCs, namely R-32, R-134a and R-125. This was the same mixture as the subject matter of claim 4 of the Parent Application. 33 Du Pont sought leave to appear before her Honour as amicus curiae, but leave was refused. Her Honour allowed the appeal and set aside the delegate's decision. She ordered that the matter be remitted to the delegate to be reconsidered in accordance with her reasons for judgment: Imperial Chemical Industries Pty Ltd v Commissioner of Patents (2004) 213 ALR 399; [2004] FCA 1658. 34 As a result of her Honour's judgment, claim 1 of the Divisional Application was amended to bring it into line with claim 4 of the Parent Application. The Divisional Application as amended was subsequently accepted by the delegate. It was that decision which prompted the present, second opposition proceedings, brought by Du Pont which were heard and determined by Gyles J.

The decision of Gyles J 35 Before Gyles J, Du Pont submitted that the lubricant integer defined by the amended claim 1 of the Divisional Application was the same as the lubricant integer defined by claim 4 of the Parent Application. It submitted that what it termed the first issue estoppel arose out of ICI's concession before Branson J (which her Honour had accepted) that in October 1991 the Lubrizol Patent disclosed to persons skilled in the art lubricants within integer (B) of claim 1 of the Parent Application. 36 Du Pont further submitted that by reason of Branson J's findings at [64]-[69], [71] and [119] of her reasons for judgment, her Honour made an ultimate finding on a question of mixed fact and law to the effect that the Lubrizol Patent deprived the invention defined by the Parent Application of novelty. This was said to give rise to what it termed the second issue estoppel. 37 Du Pont next argued that by reason of the combined effect of her Honour's findings in the paragraphs set out above, she must have found that all of the claims in the Parent Application, including claim 4, were deprived of novelty by the Lubrizol Patent. This was said to give rise to the third issue estoppel. 38 Du Pont relied upon multiple variants of these points in relation to what it termed the fourth to tenth issue estoppels. However, it is unnecessary for present purposes to set them out. They all turn upon the proposition that her Honour's finding that the Lubrizol Patent deprived the Parent Application of novelty meant that claim 1, as amended in the Divisional Application, must also lack novelty. 39 Du Pont argued, before Gyles J, that Crennan J had erred in setting aside the delegate's decision. It submitted that the delegate had correctly found that the Divisional Application lacked novelty on the basis of Branson J's findings regarding the Parent Application. Alternatively, it argued that the judgment of the Full Court refusing leave to appeal from the decision of Branson J gave rise to the various issue estoppels pleaded. 40 Gyles J rejected Du Pont's claim and dismissed its notice of opposition. His Honour explained (at [2]) that there was: "… no identity between any issue decided by Branson J or the Full Court, on the one hand, and any issue which falls for determination in relation to this opposed Divisional Application, on the other". 41 His Honour's reasoning proceeded as follows. He noted (at [8]) that refrigerants were classified according to a particular numbering system. Difluoromethane was known as R-32, 1,1,1,2-tetrafluoroethane was known as R-134a, and pentafluoroethane was known as R-125. 42 He next summarised the specification of the Lubrizol Patent by reference to what Emmett J had said about it in earlier proceedings: ICI Chemicals & Polymers Ltd v Lubrizol Corporation Inc (1999) 45 IPR 577 at [35]-[45]; [1999] FCA 345. He noted that the delegate, who had approved the amended claim in the Divisional Application after Crennan J had delivered judgment, had said in his reasons for decision that it had been conceded that the lubricants of integer (B) in the Lubrizol patent were the same as those in the Parent Application. Therefore the only question was whether the anticipation disclosed a refrigerant integer (A) that was the same as that claimed in the Parent Application. As to that, the delegate had set out his reasons for concluding that the Lubrizol Patent did not establish lack of novelty. 43 In essence, his Honour found that the delegate had concluded that the Lubrizol Patent had referred explicitly to single HFCs, and especially to R-134a. However, the Lubrizol Patent had mentioned mixtures of HFCs which included compositions of the Parent Application. The critical question, in the delegate's mind was which mixtures were clearly disclosed, or formed part of the technical information that the citation made public. The delegate stated that he did not believe that mixtures of any fluorocarbons should be regarded as clearly disclosed. Those that were identified by their code number had not been specifically referred to in the description. Consequently it was unlikely that a reader would consider that they had been taught mixtures of HFCs that included R-32 or R-125. On that basis the delegate considered that the Lubrizol Patent did not disclose a mixture including either of those substances. 44 Since integer (A) had to have at least one of R-32 or R-125, the Lubrizol Patent did not disclose any compositions that would infringe the Parent Application. 45 In dealing with Du Pont's claim that Branson J's decision had created the various issue estoppels pleaded, Gyles J referred specifically to [68]-[71] and [119] of her Honour's judgment. His Honour also referred to Branson J's judgment in relation to the separate patent application that had been brought by ICI. That separate patent application involved refrigerants that were intended to function at far cooler temperatures than those the subject of the Parent Application. Branson J dealt with the separate patent application in conjunction with the opposition to the Parent Application, deciding both cases in the one judgment. 46 The relevant claims in the separate patent application were claims 1 and 10, which were limited to ternary or higher mixtures. Claim 1 of that separate patent application provided: "A refrigerant composition comprising: (I) a refrigerant comprising a ternary or higher mixture of: (a) tetrafluoroethane and/or heptafluoropropane; (b) difluoromethane and/or 1,1,1-trifluoroethane and optionally (c) pentafluoroethane: and (II) a lubricant comprising a polyalkylene glycol and/or an ester." 47 Claim 10 was in the following terms: "A refrigerant composition comprising: (I) a refrigerant comprising: (a) 1,1,1,2-tetrafluoroethane; (b) difluoromethane; and (c) pentafluoroethane; and (II) a lubricant comprising a polyalkylene glycol and/or an ester." 48 Gyles J observed that claim 10 of the separate patent application was indistinguishable from claim 4 in the Parent Application, at least as far as the refrigerant was concerned. His Honour added that the lubricant identified in claim 10 would fall within claim 1 of the Parent Application. 49 His Honour observed that Branson J had decided that the priority date of the separate patent application was earlier than that claimed by ICI. Moreover, ICI had conceded that, if that were the case, its application was invalid. However, Branson J had gone on to give separate consideration to the other grounds of opposition to the separate patent application. 50 Again, in relation to that separate patent application, it had been alleged before Branson J that the Lubrizol Patent anticipated the claims in question. Branson J concluded in relation to that ground that, for the reasons given in [68] dealing with the Parent Application, the delegate had erred in concluding that the Lubrizol Application did not disclose a mixture including R-32, R-125 or R-227ea. For that reason, her Honour found that the Lubrizol Application did disclose ternary and higher mixtures of HFCs. 51 Gyles J noted that Branson J went on to observe that it did not follow that the Lubrizol Application contained "clear and unmistakable directions to do what the patentee claims to have invented". Her Honour had referred to The General Tire & Rubber Company v The Firestone Tyre and Rubber Company Ltd [1972] RPC 457 at 486 per Sachs LJ, and concluded that it had not been clearly established that the Lubrizol Application deprived the claims, in relation to the separate patent application, of novelty. 52 After setting out several passages from the judgment of the Full Court refusing leave to appeal from Branson J's judgment, Gyles J turned to the Divisional Application. He referred to the judgment of Crennan J setting aside the delegate's finding that the invention defined by the claims in that application lacked novelty. He described the proposed amended claim as "similar in scope", though not in form, to claim 4 of the Parent Application, and also as "similar to claim 10 of Application 654176". 53 Gyles J referred to what Crennan J had said (at [45]) in relation to the question of issue estoppel: "It is relevant to note that it was sufficient for her Honour's disposal of the proceeding before her to find that claim 1 of the parent application lacked novelty over the Lubrizol patent. In making that ultimate finding of fact, it was not necessary for her Honour to deal with claim 4 of the parent application, that is the claim to the specific ternary mixture of HFCs which the applicant now claims. Furthermore, her Honour's finding that integer (A) of the Lubrizol patent encompassed ternary and higher mixtures of HFCs and her implicit finding that the Lubrizol patent taught mixtures of HFCs is not to be confused with, or taken to be equivalent to, a much narrower finding that integer (A), or the Lubrizol patent as a whole, taught (that is, disclosed for novelty purposes) the specific ternary mixture of HFCs which the applicant now claims. Her Honour made no such narrow finding and her judgment (at [144]) suggests to me that this was deliberate rather than constituting some inconsistency in her Honour's reasons. Such a narrow finding was unnecessary because it could not be said that the question of whether or not the Lubrizol patent taught the specific ternary mixture of HFCs now sought to be claimed "was a fact fundamental to the decision arrived at" as that phrase is explained by Dixon J in Blair v Curran (1939) 62 CLR 464 at 532. For me to now consider that narrow issue, on evidence which was not before her Honour, cannot impugn her Honour's decision in relation to a much wider question of anticipation." 54 Gyles J observed that in dealing with the effects of the Full Court's decision, Crennan J had stated (at [67]): "The refusal by the Full Court to grant leave to appeal in a brief decision in ICI v Dupont (especially at [18] and [20]) does not foreclose any question, arising on different evidence, as to whether the Lubrizol patent teaches the specific ternary compound of HFCs, which is the subject matter of this proceeding." 55 Finally, Gyles J referred to Crennan J's statements (at [83]-[85]): "In my judgment it is not correct to say that the Lubrizol patent taught the specific ternary mixture of HFCs, R-22, R-125 and R-134a. Accordingly, I would find the present application, as proposed to be amended before the delegate, novel over the Lubrizol patent. It can be noted this is consonant with the approach of Branson J to the "reverse infringement" test at [144] of her Honour's judgment in respect of ternary mixtures, which must have included the subject matter of claim 10 of 654176. There is no inconsistency between findings numbered (i)-(v) in [82] above and findings made by Branson J in the earlier decision. In my view her Honour did not make any express finding on the subject matter of finding numbered (vi) above which I would regard as both careful and deliberate given her Honour's findings at [144] of her judgment. Accordingly, there is no genuine inconsistency between the finding numbered (vi) in [82] above and her Honour's conclusions in respect of the parent application." 56 Gyles J ultimately concluded that the decision and reasoning of Crennan J were correct. He said that Branson J's finding that the Parent Application lacked novelty "was entirely based upon consideration of claim 1 in the light of the Lubrizol Patent". He found that the Full Court, on the leave application, had approached the matter in the same way. Branson J had noted that claim 1 was the only independent claim, and dealt with that claim. It followed that all dependent claims, including claim 4, fell once the principal claim was found to be invalid. 57 His Honour then observed that the fact that ICI might have sought to amend claim 4 was not relevant. It had not done so. What Crennan J had described as "the narrow question" had not arisen before, and therefore had not been considered by, Branson J. 58 In his Honour's terms (at [28]-[29]): "This explains the apparent contradiction between the approach to novelty taken in relation to Application 654176, on the one hand, and Application 658005 on the other, in the same set of reasons. When Branson J focused upon the refrigerant aspects of claim 4, she approached the issue in the same way and reached the same conclusion in substance as that later reached by Crennan J. If Du Pont's analysis were correct, the inconsistency between the two approaches by Branson J would be startling. Furthermore, it would be surprising if Branson J and the Full Court had dealt with the question of novelty of claim 4 of Application 658005 in the pre-emptory way that they respectively did if it had been considered as an independent claim. It is most unlikely that either would have disposed of the question without even noticing or discussing the issues which would clearly arise in that event, as exemplified by the decision of Branson J in relation to Application 654176 and Crennan J in relation to the Divisional Application in suit here. As there is no identity between any issue found by Branson J, confirmed by the Full Court, on the one hand, and any issue for determination in the present application, on the other, there can be no issue estoppel (Kuligowski v Metrobus (2004) 220 CLR 363 at [21], [40]-[54])."

THE ISSUES ON APPEAL TO THIS COURT 59 Du Pont submitted on appeal to this Court that Gyles J had erred in holding that no issue estoppel of the type pleaded had arisen. By necessity, it submitted that Crennan J, whose judgment Gyles J had followed, had also erred in so holding. 60 In relation to Branson J's decision, Du Pont relied upon four key propositions:

her Honour made a clear finding that the invention comprising claim 1 in the Parent Application, where integer (A) of that claim included the ternary combination of R-32 (difluoromethane), R-125 (pentafluoroethane) and R-134a (1,1,2,2-tetrafluoroethane), lacked novelty in light of the Lubrizol Patent;.
her Honour's reasons in relation to the separate patent application involved a different application and different proceedings. Accordingly, it was impermissible for both Crennan and Gyles JJ to have had regard to those reasons "for the purpose of seeking to cut down" the clear issue estoppel which arose on the face of Branson J's reasons in relation to the Parent Application;
in any event Branson J's remarks in relation to the novelty of the separate patent application were brief, and obiter dicta; and
even if, contrary to the second and third propositions above, it was permissible to have resort to Branson J's remarks in relation to the separate patent application, those remarks, properly understood, did not provide a sufficient basis for undermining the issue estoppel which otherwise arose. 61 In relation to the decision of the Full Court, Du Pont argued that a further issue estoppel arose by reason of the Full Court's affirmation of Branson J's judgment. 62 Du Pont argued, in support of its first key proposition, that Branson J's decision in relation to the Parent Application had proceeded on two bases. These were that: · ICI had conceded that the Lubrizol Patent disclosed lubricants within the Parent Application, but contended that it did not disclose the heat transfer fluid comprising the mixtures of refrigerants set out in the claims of the specifications in combination with those lubricants; and · claim 1 of the Parent Application was the only independent claim. 63 Du Pont submitted that both Crennan and Gyles JJ had erred in treating Branson J's decision on the Parent Application as being restricted to claim 1. However, even if her Honour's reasons were to be so understood, the issue estoppel would arise. 64 Du Pont noted that claim 1 of the Parent Application claimed a refrigerant component "comprising a mixture of at least two hydrofluoroalkanes selected from the group of" R-32, R-134a and R-125. Logically that meant that there could be only four mixtures:
a binary mixture of R-32 and R-134a;
a binary mixture of R-32 and R-125;
a binary mixture of R-125 and R-134a; and
a ternary mixture of R-32, R134a and R-125 (which is the refrigerant component of what, together with the lubricant integer, Du Pont described as the "Ternary Invention"). 65 Du Pont submitted that it was possible for Branson J to have found that claim 1 of the Parent Application lacked novelty in light of the Lubrizol Patent by:
making findings restricted to the binary combinations of the refrigerant and therefore finding, as a matter of drafting, that claim 1 (containing as it did the words "at least two") lacked novelty as drafted;
making findings restricted to ternary combinations of the refrigerant and therefore finding as a matter of drafting, that claim 1 lacked novelty as drafted; or
making a finding in relation to claim 1 which applied to both binary and ternary combinations which would mean that claim 1 failed in its entirety, and could not be saved by amendment. 66 Du Pont submitted that Branson J's reasons indicated that she had ultimately found that claim 1 lacked novelty on the basis of the third of these three possible sets of findings. It then submitted that the reasons given by Gyles J for rejecting its various claims of issue estoppel were misconceived. That was because his Honour had erroneously determined that the question of what was legally indispensable to Branson J's decision was to be answered by asking whether there was any possible way in which she could have made the orders that she did without needing to decide the novelty of the Ternary Invention. Having found that there was such a way, his Honour assumed that Branson J had in fact followed such a path. 67 Du Pont submitted that this reversed the natural inquiry. The correct approach was to determine the basis for Branson J's decision, as disclosed in her reasons for judgment. That was set out in [68] when she found, on the basis of her own reading of the Lubrizol Patent, and with the benefit of the evidence of Mr Symons and Professor Ball, that the delegate had been clearly wrong in concluding that it was unlikely that a reader would consider that they had been taught mixtures of HFCs including R-32 or R-125 by the Lubrizol Patent. 68 Du Pont noted that Branson J's conclusion referred to mixtures of HFCs, and was not restricted to binary mixtures. Moreover, her Honour in her reasons for judgment at [65]-[70] expressly indicated that she had decided the case on the basis of both binary and ternary combinations. Du Pont argued that if she had concluded that claim 1 was rendered not novel by the Lubrizol Patent only insofar as it related to binary inventions it would have been necessary for her to consider whether she could allow an amendment to be made which restricted the claim to the Ternary Invention, and thereby cured the defect. 69 Du Pont submitted that although it had put that precise argument to Gyles J, and it was reflected in paras 28F and 55A of its Further Amended Statement of Claim, his Honour had failed to address it in his reasons for judgment. Rather his Honour had found that there was no issue estoppel simply because he considered that Branson J had not necessarily made a decision on claim 4 of the Parent Application. Yet claim 4 was nothing more than claim 1, restricted to the ternary combination. 70 Du Pont contended that Gyles J had erred in making that finding. The whole of the Parent Application was in issue before Branson J. That meant that each and every claim in that application had been challenged. Her Honour's conclusion that the Lubrizol Patent deprived the Parent Application of novelty was expressed in general terms. It was not restricted to any particular claims. Had her Honour disposed of claim 4 solely on the basis that, as a matter of drafting, it was a dependent claim, she would have had to consider whether to allow an amendment to be made in order to save that claim. Her refusal to do so made it clear that dependence was not the basis on which claim 4 failed. 71 In elaboration of its second, third and fourth key propositions, Du Pont argued that whatever her Honour had intended in [144] in relation to the separate patent application, her remarks could not be used to restrict the ambit of her findings in relation to the Parent Application. In particular those remarks could not be used to restrict the earlier findings to binary combinations within integer (A). 72 Du Pont's arguments in support of this submission were highly elaborate, and, save for the first, need not be set out here. It argued that some 40 refrigerant mixtures fell within the separate patent application. These included 15 ternary mixtures. A combination of R-125, R-134a and R-32 formed one of those 15 ternary mixtures, whereas in the case of the Parent Application that combination was but one of four. Because it was one of so many, in the context of the separate patent application, Branson J may simply have overlooked the relevant ternary mixture in relation to that application.

CONSIDERATION 73 In Blair v Curran (1939) 62 CLR 464 Dixon J formulated the doctrine of issue estoppel in the following terms (at 531-532): "A judicial determination directly involving an issue of fact or of law disposes once for all of the issue, so that it cannot afterwards be raised between the same parties or the privies. The estoppel covers only those matters which the prior judgment, decree or order necessarily established as the legal foundation or justification of its conclusion, whether that conclusion is that a money sum be recovered or that the doing of an act be commanded or be restrained or that rights be declared. The distinction between res judicata and issue estoppel is that in the first the very right or cause of action claimed or put in suit has in the former proceedings passed into judgment, so that it is merged and has no longer an independent existence, while in the second, for the purpose of some other claim or cause of action, a state of fact or law is alleged or denied the existence of which is a matter necessarily decided by the prior judgment, decreed or ordered." 74 His Honour went on to say (at 532): "Nothing but what is legally indispensable to the conclusion is thus finally closed or concluded. In matters of fact the issue estoppel is confined to those ultimate facts which form the ingredients in the cause of action, that is, the title to the right established. Where the conclusion is against the existence of a right or claim which in point of law depends upon a number of ingredients or ultimate facts the absence of any one of which would be enough to defeat the claim, the estoppel covers only the actual ground upon which the existence of the right was negatived. But in neither case is the estoppel confined to the final legal conclusion expressed in the judgment, decree or order… the judicial determination concludes, not merely as to the point actually decided, but as to a matter which it was necessary to decide and which was actually decided as the groundwork of the decision itself, though not then directly the point at issue. Matters cardinal to the latter claim or contention cannot be raised if to raise them is necessarily to assert that the former decision was erroneous." 75 It seems that the term "issue estoppel" was coined by Higgins J in his dissenting judgment in Hoysted v Federal Commissioner of Taxation (1921) 29 CLR 537. However, it is likely that the doctrine can be traced back as far as the Duchess of Kingston's Case (1776) 2 Smith LC (13th ed) 644. There are instances of a similar doctrine having been applied throughout the nineteenth century. See, for example, Outram v Morewood (1803) 3 East 346; 102 ER 630; Heston Overseers v St Bride Overseers (1853) 1 El & Bl 583; 118 ER 556; R v Inhabitants of the Township of Hartington Middle Quarter (1855) 4 El & Bl 780 at 794-797; 119 ER 288 at 293-294; and Routledge v Hislop (1860) 2 El & El 549; 121 ER 206. 76 Dixon J's formulation of the relevant principles governing issue estoppel was adopted by Fullagar J in Jackson v Goldsmith (1950) 81 CLR 446 at 466, although his Honour was in dissent in that case. The High Court has considered and approved Blair v Curran, or adopted the same approach to issue estoppel, on a number of occasions: see, for example, Queensland Trustees Ltd v Commissioner of Stamp Duties (Qld) (1956) 96 CLR 131 at 138 per Dixon CJ, 144 per Webb J, and 151 per Kitto and Taylor JJ; Ramsay v Pigram (1968) 118 CLR 271 at 276 per Barwick CJ; and Port of Melbourne Authority v Anshun Pty Ltd (1981) 147 CLR 589 at 597 per Gibbs CJ, Mason and Aickin JJ. 77 The concept of "issue estoppel", as developed in Australia, has been accepted in England: see Mills v Cooper [1967] 2 QB 459 per Diplock LJ; Carl Zeiss Stiftung v Rayner and Keeler Ltd (No 2) [1967] 1 AC 853; and Hunter v Chief Constable of the West Midlands Police [1982] AC 529. 78 In DSV Silo-und Verwaltungsgesellschaft mbH v Owners of The Sennar [1985] 2 All ER 104 Lord Brandon formulated the doctrine in the following terms (at [110]): "…in order to create an estoppel of that kind… three requirements have to be satisfied. The first requirement is that the judgment in the earlier action relied on as creating an estoppel must be (a) of a court of competent jurisdiction, (b) final and conclusive and (c) on the merits. The second requirement is that the parties… in the earlier action relied on as creating an estoppel and those in the later action in which that estoppel is raised as a bar must be the same. The third requirement is that the issue in the later action in which the estoppel is raised as a bar must be the same issue as that decided by the judgment in the earlier action." 79 The judgment of Branson J was plainly a final judgment of a competent tribunal making a judicial, and not an administrative, decision. Her Honour had jurisdiction to decide the issue conclusively between the parties and for all purposes. Accordingly, the first of Lord Brandon's three requirements for issue estoppel is met. 80 The parties to the second opposition proceedings were the same as those before Branson J. It follows that the second of Lord Brandon's three requirements was also met. 81 The critical issue in this appeal relates to the third requirement. The question is whether Branson J, in holding that the Lubrizol Patent deprived the Parent Application of novelty, effectively decided that the Divisional Application also lacked novelty. 82 Du Pont contends that both Branson J and the Full Court, by holding that the Parent Application must fail by reason of the Lubrizol Patent, necessarily decided that claim 4 of that application must also fail for that reason. ICI contends that claim 4, as a dependent claim had to fail once claim 1 was rejected. It argues that neither Branson J nor the Full Court found it necessary to consider whether claim 4, in isolation, lacked novelty. Accordingly, there is no issue estoppel. 83 Gyles J concluded that Branson J had rejected claim 4 because it was a dependent claim, and not because she considered it to lack novelty. His Honour noted that Crennan J had earlier come to the same conclusion. For that reason, he dismissed Du Pont's claim for injunctive relief. 84 It is true that Gyles J relied in part upon Branson J's reasoning in relation to the separate patent application. His Honour was concerned to interpret her Honour's reasons for judgment in relation to each patent in a way that avoided inconsistency. Crennan J had earlier approached the question of issue estoppel in the same way. 85 We can see no error in what Gyles J did. His Honour understood, and applied, the principles governing issue estoppel correctly. It is a singular feature of this case that Branson J did not consider claim 4 of the Parent Application, and whether it lacked novelty by reason of the Lubrizol Patent, separately from claim 1. She focused instead entirely upon claim 1 which she understood to be the only independent claim. 86 Du Pont brought this application for injunctive relief. It sought to invoke the doctrine of issue estoppel. It is well-established that a plea of issue estoppel must be clear and unambiguous as to the issue upon which the opponent is said to be estopped: Jackson v Goldsmith at 455 per Latham CJ and at 458 per McTiernan J. Du Pont was required to establish, before Gyles J, that one or more of the numerous forms of issue estoppel that it pleaded, and upon which it relied, had been made out. It had to demonstrate that it was an indispensable or fundamental aspect of Branson J's reasoning that claim 4 of the Parent Application, considered separately and apart from claim 1, also lacked novelty. Only then could it rely upon the fact that claim 1 of the Divisional Application as amended was in substance the same as claim 4 of the Parent Application. For the reasons comprehensively stated by Bennett J, Du Pont failed to discharge that onus. 87 It follows that the appeal should be dismissed, with costs. I certify that the preceding eighty-seven (87) numbered paragraphs are a true copy of the Reasons for Judgment herein of the Honourable Justices Weinberg and Stone .

NEW SOUTH WALES DISTRICT REGISTRY NSD 2296 OF 2006

ON APPEAL FROM A SINGLE JUDGE OF THE FEDERAL COURT OF AUSTRALIA

Procedural history 88 The appellant ('Du Pont') opposed two of the first respondent's ('ICI') patent applications, patent application No 685005 ('the 005 application') and patent application No 654176 ('the 176 application'). The appeals by Du Pont from the decision of the Commissioner of Patents ('the Commissioner') dismissing the two oppositions were heard together by Branson J, who delivered a single set of reasons for judgment on 12 March 2002 (E I Du Pont De Nemours & Co v Imperial Chemical Industries PLC (2002) 54 IPR 304; [2002] FCA 230). Justice Branson concluded that a prior patent ('the Lubrizol patent'), 'deprived the [005] application of novelty' (at [71]) but that it was not clearly established with respect to the 176 application that the Lubrizol patent 'deprives the claims of novelty' (at [144]). An application for leave to appeal Branson J's decision was dismissed by the Full Court (Imperial Chemical Industries PLC v E I Dupont De Nemours & Co (2002) AIPC ¶91-818; [2002] FCAFC 264). 89 ICI filed a divisional application of the 005 application ('the divisional') on 28 March 2002 after Branson J delivered reasons but before the making of orders. The divisional, as filed, was identical to the 005 application. The divisional was refused by a Delegate of the Commissioner ('the Delegate'), on the grounds that the decision of Branson J compelled refusal. ICI then sought to amend the divisional to delete the independent claim of the 005 application, claim 1, and to make the dependent claim 4 of the 005 application the new claim 1 of the divisional. The Delegate refused the amendment. On appeal from that decision (Imperial Chemical Industries Pty Ltd v Commissioner of Patents (2004) 213 ALR 399; [2004] FCA 1658), Crennan J considered the decisions of Branson J and of the Full Court on the application for leave to appeal Branson J's decision. Justice Crennan held that Branson J's decision in relation to the 005 application did not compel the refusal of the application for grant of the divisional or the application to amend the divisional (at [83]-[85]). Justice Crennan concluded that ICI should be permitted to amend the divisional to retain those dependent claims of the 005 application which, her Honour held, were not deprived of novelty by the Lubrizol patent and remitted the matter to the Commissioner (at [87]). Du Pont did not participate in the proceedings before Crennan J. 90 The Commissioner accepted the amended divisional. Du Pont has filed a notice of opposition to the divisional as amended. Du Pont contends that there are issue estoppels between Du Pont and ICI arising from Branson J's reasons for judgment and the orders made by her Honour and also from the reasons for judgment and orders of the Full Court on the application for leave to appeal. 91 As to Du Pont's contention, the primary judge held that there was no identity of any issue decided by Branson J and the Full Court on the one hand and any issue which falls for determination in relation to the opposition to the divisional on the other (E I Du Pont De Nemours & Co v Imperial Chemical Industries PLC (2006) 70 IPR 274; [2006] FCA 1402). Consequently, his Honour held that there was no issue estoppel (at [29]). Du Pont maintains that Branson J's findings on novelty with respect to the 005 application give rise to an issue estoppel in the opposition to the divisional that goes to the novelty of the invention so far as claimed in what is now claim 1 (previously dependent claim 4 of the 005 application).

Factual background of HFCs described and claimed 92 The 005 and 176 applications claim mixtures of hydrofluorocarbons ('HFCs'). 93 There is no dispute as to the identity of the HFCs described in the different specifications or as to the HFCs specifically identified in the claims. There is no dispute as to the numbers of combinations encompassed because of the description of the combinations or the different isomers of some of the HFCs. There is no need to elaborate on the precise combinations. The relevant HFCs were given individual designations, for example R-32. It is not necessary to consider the reasons for this. 94 The HFCs of primary interest are: · Difluoromethane - R-32, which has one carbon atom. · 1,1,1,2-tetrafluoroethane - R-134a, which has two carbon atoms. · Pentafluoroethane - R-125, which has two carbon atoms.

The 005 application 95 The specification of the 005 application noted that the refrigerant R-134a, which was generally replacing earlier refrigerants, had disadvantages as it does not possess appropriate boiling characteristics and thermal properties of other refrigerants. The invention as described was to a combination of HFCs of which R-134a was one. 96 Claim 1 was the only independent claim. Relevantly, it provided: 'A working fluid composition comprising: (A) a heat transfer fluid comprising a mixture of at least two hydrofluoroalkanes selected from the group consisting of [R-32], [R-134a] and [R-125].' 97 Hydrofluoroalkanes are HFCs. The claimed combination of claim 1 is, therefore, to a mixture of at least two HFCs, selected from a group consisting of three HFCs. This gives rise to four combinations: three binary and one ternary. The ternary combination is R-32, R-134a and R-125 ('the ternary mixture'). 98 Claim 4 of the 005 application was to the fluid composition of claim 1, comprising the ternary mixture.

The divisional 99 In the amended divisional, claim 1 claims the ternary mixture. Claims 2 to 5 claim working fluid compositions of claim 1 in specified proportions by weight.

The 176 application 100 Claim 1 of the 176 application claimed a refrigerant composition comprising, relevantly: '(1) a refrigerant comprising a ternary or higher mixture of: (a) tetrafluoroethane and/or heptafluoropropane; (b) difluoromethane and/or 1,1,1-trifluoroethane; and optionally (c) pentafluoroethane…' 101 Tetrafluoroethane, heptafluoropropane and trifluoroethane exist in isomeric forms. Examples were given in the specification of: · 1,1,1-trifluoroethane which is R-143a and has two carbon atoms. · 1,1,1,2-tetrafluoroethane which is R-134a and has two carbon atoms. · 1,1,2,2-tetrafluoroethane which is R-134 and has two carbon atoms. · 1,1,1,2,2,3,3-heptafluoropropane which is R-227ca and has three carbon atoms. · 1,1,1,2,3,3,3-heptafluoropropane which is R-227ea and has three carbon atoms. 102 Claim 1 included isomers of two of the five specified refrigerants. Accordingly, it described seven refrigerants, namely R-134a, R-134, R-227ca, R-227ea, R-32, R-143a and R-125. 103 Claim 1 provided for forty-four possible mixtures: · twenty-four possible ternary mixtures; · sixteen possible quaternary mixtures; and · four possible quinternary mixtures.

The Lubrizol patent 104 Integer A of claim 1 of the Lubrizol patent application is: '1. A liquid composition is described which comprises: (A) a major amount of at least one fluorine containing hydrocarbon containing one or two carbon atoms;…' 105 This encompasses HFCs containing one or two carbon atoms and binary, ternary and higher blends of any of the above. 106 There are in the Lubrizol patent: · fifty-five possible binary blends; and · one hundred and sixty-five possible ternary blends; · of the eleven HFCs containing one or two carbon atoms. 107 R-227, with three carbon atoms, does not fall within Integer A of the Lubrizol patent. Of the 44 possible mixtures of the 176 application claim 1, thirty-four do not overlap with the Lubrizol patent's Integer A, including those that contain R-227. Ten combinations do overlap with the mixtures of the Lubrizol patent. 108 There is no particular relevance to these numbers of possible blends within claim 1 of the Lubrizol patent, which were not adverted to by Branson J or the primary judge. They were not, apparently, put before their Honours. They do, however, exemplify the complexity of possible mixtures. 109 Branson J observed at [69] that the Lubrizol patent disclosed to the skilled reader a broad class of potential refrigerants extending beyond those identified specifically. While the three HFCs of the 005 application fell within the description of Integer A, neither R-32, nor R-125 are specifically mentioned in the Lubrizol patent. It is not in dispute that the Lubrizol patent disclosed lubricants within the 005 application.

The decision of Branson J 110 It is the decision of Branson J concerning the novelty of the claims of the 005 application and the 176 application that reveals what was decided in the oppositions to those applications. 111 It is not necessary for the purposes of this appeal to discuss the principles applicable to a determination of novelty. It is, however, relevant to consider the way in which her Honour considered the question of anticipation. Justice Branson set out the test to be applied as the reverse infringement test, that is 'whether the alleged anticipation would, if the patent were valid, constitute an infringement' (at [35], see Meyers Taylor Pty Ltd v Vicarr Industries Ltd (1977) 137 CLR 228 at 235 per Aickin J, cited by Lockhart J in R D Werner & Co Inc v Bailey Aluminium Products Pty Ltd (1989) 25 FCR 565 at 568). That test also requires that the prior publication contain 'clear and unmistakable directions to do what the patentee claims to have invented' (The General Tire & Rubber Company v The Firestone Tyre and Rubber Company Ltd [1972] RPC 457 at 486 per Sachs LJ). Her Honour referred to Nicaro Holdings Pty Ltd v Martin Engineering Co (1990) 91 ALR 513 and also to Sundberg J in MJA Scientifics International Pty Ltd v S C Johnson & Son Pty Ltd (1998) 43 IPR 287 at 303; [1998] FCA 1467, which citation bears repeating: 'Where the alleged anticipation does not disclose all the essential integers of the patent in suit, the fact that the skilled addressee could come from the alleged anticipation to the patent in suit without the exercise of inventive ingenuity in the light of common general knowledge does not establish lack of novelty.' 112 That is, her Honour was conscious of a need to consider whether, where the essential integers of a claim were not disclosed in the prior publication, there had been anticipation.

The 005 application 113 In the opposition, the Delegate, in a passage set out by Branson J at [40], held in effect, that the Lubrizol patent only disclosed those mixtures of HFCs specifically referred to or specifically disclosed in the description. As R-32 and R-125 were not specifically referred to either as individual components or as mixtures, the Delegate held that they had not been disclosed. That is, while the class of HFCs was disclosed, the Delegate considered that there had not been an enabling disclosure of the mixtures of claim 1 of the 005 application. 114 Justice Branson considered the disclosures of the Lubrizol patent as at 1991. This was for the purposes of both the 005 application and the 176 application. For the purposes of considering what Branson J decided, the Court was taken to her Honour's reasons, the notice of appeal before Branson J, submissions and the evidence of the witnesses on whom her Honour relied. That evidence was not before the primary judge but was referred to in the appeal, by consent. 115 The notice of appeal by Du Pont from the decision of the Delegate that was before Branson J extended to the invention as claimed in each and all of the claims of the 005 application and sought a finding of lack of novelty of each of those claims. There was reference in the submissions before Branson J to "all of the refrigerant mixtures" of the 005 application. There was also a submission that the claims other than claim 1 were anticipated by the Lubrizol patent, specifically claims 2, 3 and 4 which claimed subsets of the refrigerants of claim 1. There were no submissions directed to the novelty of claim 1 by reference to the subsets of refrigerants. ICI's submissions referred to the blends claimed and asserted that the Lubrizol patent did not "plant the flag" at a lubricant in combination with a blend of HFCs falling within the claims. 116 In Branson J's decision on the 005 application: · Only claim 1, the independent claim, was the subject of consideration. · There was no consideration of any dependent claim, including claim 4. · Branson J did not consider whether any claim could be saved by amendment but was under no obligation to do so in the absence of an application to amend. 117 Her Honour relied on the evidence of two skilled addressees, Mr Symons and Professor Ball, which was available for both applications. It is instructive to consider the precise words used in the cross-examination of those witnesses and the way in which her Honour expressed her conclusions based on their evidence. 118 Mr Symons' evidence was, relevantly (emphasis added): '[MR GLEESON:] If I could ask you to turn then to the refrigerant, component A, you see that it uses the words, "A major amount of at least one fluorine containing hydrocarbon containing one or two carbon atoms. You would have understood that as encompassing a situation, at least, where you had a single fluorine containing hydrocarbon?--- Yes. An example of that would have been R-134a? --- Yes. Another example would be 125? --- Yes. 32, 143, etcetera? --- Yes. When it uses the words "at least one", you would have understood that as saying apart from the cases of single or pure HFCs you can have more than one HFC in blend or mixture? --- Yes. So it would tell you at that time that you could have two HFCs in mixture? --- Yes. Or three or four or more? --- Yes. … Mr Symons at April 1991, I think you've agreed that the number of HFCs that you knew of or had heard of was something less than 10? --- Yes. Had you been reading this document in April 1991, what it would have taught you is that provided your ester lubricant met the formula you could combine that with any one of those, I'll call it 10 HFCs that you knew? --- Yes. And you could also combine such lubricants with any mixture of those, I'll call it, 10 HFCs? --- Yes.' 119 Mr Symons agreed that the words "at least one" would have led him to understand that there could be a mixture of two, three, four or more HFCs in the mixture. Mr Symons clearly stated that the Lubrizol patent taught the use of single HFCs and mixtures of two, three, four or more HFCs. He was not asked if the Lubrizol patent specifically disclosed the claimed combinations and in particular was not asked if the ternary combination of claim 4 was there disclosed. It is not clear that he was saying that it taught or disclosed the HFCs specifically identified in cross-examination or that they are single fluorine containing hydrocarbons that are, in fact, encompassed by the description in Integer A. The Lubrizol patent disclosed both single fluorine containing HFCs and blends. While examples of them would have included those of the 005 application, that is not to say that those examples were themselves disclosed or taught. In that regard, it is worth repeating that R-32 and R-125 were not mentioned in the Lubrizol patent and that R-143 was not referred to in the 005 application; the isomer, R-143a was claimed in the 176 application. It is also relevant, in considering what the general description in Integer A did disclose to Mr Symons that, as at April 1991, he only knew or had heard of fewer than ten HFCs. 120 There was no re-examination. 121 This raises a question of some ambiguity, arising in part by the use of the word "encompassing". Mr Symons' language was reflected in her Honour's reasons. Justice Branson at [65] drew from Mr Symons' evidence that the Lubrizol patent would have disclosed to him that component A of the liquid composition encompassed at least the examples which would have included R-134a, R-125, R-32 and R-143 and would also encompass blends of two, three, four or more HFCs. This could mean: the Lubrizol patent taught the class of single fluorine containing hydrocarbons and taught or disclosed all hydrocarbons that fell within the class. That is, the "teaching" for the purposes of novelty, was not restricted to those HFCs specifically disclosed but extended to any HFC within the class; or the Lubrizol patent taught the class as described in the Lubrizol patent. Included within the class were each of the examples described in the specification. It can be said that any HFC fell within that class but they were not taught or disclosed by the Lubrizol patent. 122 Professor Ball's evidence was, relevantly (emphasis added): '[MR GLEESON:] So it would cover at least the case of, for example, R-125 together with the ester lubricant of B?--- It would cover anything that has fluorine in one or two carbon…[atoms]. So, probably chlorine as well. But I just want to see if it includes at least R-125 together with an ester of formula B? --- I think so, yes. And similarly it includes 134a plus the ester? --- Yes. 32 plus the ester? --- Certainly. Because of the words "at least one" you would understand to mean that it also includes blends or mixtures of HFCs? --- Yes. Well, at least one, it might be at least one fluorine. No, I'm sorry, you're right, yes. So that amongst the many things this patent would've taught you one of them is provided you have an ester lubricant that matches the formula of B, whatever that means, then that would be a suitable lubricant to use with mixtures of HFCs? --- Yes. … So, my question Professor was, page 9, near the bottom, there was an explicit teaching that mixtures of fluorine containing hydrocarbons could be used, correct? --- Yes. You know that the particular examples that follow are different to the refrigerant blends of the ICI patents, don't you? --- Yes. You understand the document to teach that any mixtures of HFCs could be used with the lubricants of the invention? --- Yes.' 123 The questions answered by Professor Ball used the words "taught", "cover" and "includes". Professor Ball accepted that the Lubrizol patent would "cover" any compound with a fluorine in one or two carbon atoms and "includes" R-125, R-134a and R-32 as well as blends or mixtures of HFCs. It "taught" that any mixtures of HFCs could be used and Professor Ball would have expected that mixtures of HFCs known to him would work. At [67] her Honour said that Professor Ball's evidence addressed what he would have learned from the Lubrizol patent. That would have indicated a situation where there was a single HFC so that it would cover anything that has fluorine and one or two carbon atoms. It would include R-125 and similarly would include R-134a and R-32 respectively plus an ester. It also would have taught him that any mixtures of HFCs would be suitable. 124 From her own reading of the Lubrizol patent, Branson J concluded (at [18]) that the Delegate was wrong in concluding that it was unlikely that a reader would consider that they had been taught mixtures of HFCs including the two not specifically mentioned in the Lubrizol patent. Her Honour recognised that a single example of a blend of two HFCs in the Lubrizol patent does not teach the mixture of the 005 application which requires at least two HFCs out of four (at [69]). Her Honour relied upon the Lubrizol patent describing 'a major amount of at least one fluorine containing hydrocarbon containing one or two carbon atoms' and mixtures including binary and ternary mixtures (at [69]). This encompassed any mixtures of HFCs containing one or two carbon atoms. 125 There was no discussion by Branson J of the enabling disclosure of a specific composition or of mixtures containing HFCs not described in the Lubrizol patent. There was no discussion of claim 4, to the specific ternary combination. There was no discussion of any dependent claim, which included a claim to a refrigerator, or of the omnibus claim. The logic was based upon the breadth of a claim to a mixture of 'at least' two HFCs selected from the group. The conclusion was that the Lubrizol patent disclosed 'a broad class of potential refrigerants, extending beyond those identified specifically in the preferred embodiments, with which the lubricant of integer (B) might be used' (at [69], emphasis added). There was no finding of novelty that referred to specific binary or ternary combinations of the refrigerant.

The 176 application 126 There were two independent claims in the 176 application. 127 In the words of the Delegate, adopted by Branson J (at [122]): 'Integer I is a refrigerant that is itself a ternary or higher mixture of refrigerants. One of the components is identified as 'tetrafluoroethane', which can exist in two isometric forms: R-134 and R-134a. I consider that the generic name should be regarded as referring to either of the forms. However, I believe that tetrafluoroethane cannot make up two of the components of the mixture, ie. it is either R-134 or R-134a that is included. The refrigerants are presented as a list, which can be simplified to: (a) (R-134 or R-134a) and/or (R-227ea); (b) (R-32) and/or (R-143a); and optionally (c) (R-125). I consider that mixture must have three or more refrigerants selected such that there must be component (a) AND component (b) present, AND OPTIONALLY component (c). The refrigerant does not contain anything outside components (a), (b) and (c).' 128 The relevant integer of claim 1 comprised (in an inclusive sense): '1. A refrigerant composition comprising: (I) a refrigerant comprising a ternary or higher mixture of: (a) tetrafluoroethane and/or heptafluoropropane; (b) difluoromethane and/or 1,1,1-trifluorethane; and optionally (c ) pentafluoroethane; and (II) a lubricant comprising a polyalkylene glycol and/or an ester.' 129 Claim 10 of the 176 application was relevantly: 'A refrigerant composition comprising: (I) a refrigerant comprising: (a) 1,1,1,2-tetrafluoroethane [R-134a]; (b) difluoromethane [R-32]; and (c) pentafluoroethane [R-125];…' That is, claim 10 of the 176 application comprises the ternary mixture of claim 4 of the 005 application and claim 1 of the divisional. 130 The evidence of Mr Symons and Professor Ball was directed to the disclosures of the Lubrizol patent. ICI's submissions directed to the 176 application contained an assertion that the Lubrizol patent did not disclose the "ternary or higher mixture" of the particular refrigerants claimed in the 176 application. Du Pont argued that the Lubrizol patent deprived the claims of each of the 005 application and the 176 application of novelty. The evidence was therefore relevant to a consideration of the novelty of the invention the subject of the claims of both the 005 application and the 176 application, although it was considered by Branson J in the context of the 005 application. 131 The findings by Branson J on the novelty of the invention as claimed in the 176 application were obiter. However, in the context of a single set of reasons and evidence called on the way in which the skilled reader read the Lubrizol patent, the findings concerning the 176 application are of assistance in determining what was decided on the 005 application. 132 Justice Branson repeated at [143], in the context of the 176 application, that the Delegate had erred in concluding that, while the intellectual content of the disclosures of the Lubrizol patent included mixtures of HFCs, it did not disclose a mixture including R-32, R-125 or R-227ea, none of which were described. Her Honour noted the distinction drawn by the Delegate between "intellectual content" which is insufficient to deprive each composition of novelty and what the prior publication would make public to the instructed reader. Her Honour noted the Delegate's distinction between a disclosure that included compositions and a disclosure that, in fact, disclosed specific compositions. Her Honour repeated her conclusion reached in the 005 application that the Lubrizol patent does disclose ternary and higher mixtures of HFCs. However, her Honour continued at [144] to say that '[t]he above conclusion does not, however, lead inevitably to the conclusion that the Lubrizol [patent] contains "clear and unmistakable directions to do what the patentee claims to have invented"', citing General Tire at 486. 133 Her Honour found that, having regard to the 'highly particularised blends of refrigerants claim[ed]' it was not "clearly established" that the claims of the 176 application were deprived of novelty by the Lubrizol patent (at [144]). Again, her Honour referred to 'claims' in the plural. As was the case for the 005 application, there was no separate consideration of the individual claims. In particular, there was no consideration of or separate reference to claim 10 to the refrigerant comprising the ternary mixture. 134 That is, in the 005 application, without discussion of complexity her Honour held, in DuPont's submission, that the claim to the class of HFCs in the Lubrizol patent was a disclosure of each HFC and mixture of HFCs within that class (at [86]-[95]). Her Honour reiterated in considering the 005 patent application, that the disclosure of the class does disclose the mixtures that are included within the class. However, when directing her mind to the disclosure of specific mixtures which were so included for the purposes of the 176 application, her Honour held that specific blends were not disclosed (at [144]). There is no inconsistency in her Honour's reasons with respect to the 005 application and the 176 application if the decision on the 005 application was not directed to the questions of enabling disclosure of specific mixtures. These specific mixtures included the ternary mixture of dependent claim 4.

The appeal to the Full Court 135 ICI's notice of appeal to the Full Court from the decision of Branson J referred to the novelty of the 005 application. In the application for leave to appeal, ICI referred to the mixtures of HFCs in the claims generally and the mixtures of the patent application, not restricted to claim 1. It described the holdings of Branson J in terms of the mixtures of HFC refrigerants of the group of R-32, R-134a and R-125. The notice of appeal also asserted error in [71] of her Honour's reasons in failing to hold that claim 4 was not disclosed by the Lubrizol patent. 136 The Full Court was cognisant of the fact that, in considering whether to grant leave to appeal, the patent applicant ICI would only have had one judicial consideration if leave were refused (at [7]). The Court referred to Branson J's conclusion that the 005 application lacked novelty over the Lubrizol patent and her Honour's finding that Integer A of the Lubrizol patent anticipated Integer A of the 005 application (at [17]). 137 There was no separate consideration of the dependent claims, nor was there reference to each mixture encompassed by the claims of the 005 application. 138 The Full Court acknowledged that the Lubrizol Patent described the invention in terms of 'at least one fluorine containing hydrocarbon containing one or two carbon atoms' (at [16]). It was satisfied that the hydrofluoroalkanes of claim 1 of the 005 application were each HFCs and that each contained fluorine and one or two carbon atoms. Therefore, each of the named HFCs of the 005 application was of the class 'encompassed by the description' in the Lubrizol patent (at [18]). The Full Court continued (at [18]) to say that 'the use of a liquid composition described in Claim 1 of [the 005 application] is clearly disclosed by the Lubrizol patent'. The Court concluded that the 005 application 'is no more than a particular example of the invention disclosed by the Lubrizol Patent' (at [19]). 139 At [19] the Full Court observed that there was no suggestion that the mixtures of the 005 application constituted a selection of the class of mixtures of HFCs. It followed that it was "crystal clear" that a patent granted on the 005 application would be revocable (at [20]). The Full Court did not discuss enabling disclosure of the scope of the claims or the specific combinations of the 005 application. It made no reference to the dependent claims, nor did it specifically refer to the binary or ternary mixtures. 140 Claim 1 of the 005 application, which encompassed mixtures of at least two fluorine containing hydrocarbons of one or two carbon atoms, comprised a mixture of at least one fluorine containing hydrocarbon containing one or two carbon atoms, as described in the Lubrizol patent. The Lubrizol patent did not descend to particular mixtures. Justice Branson had not descended to particular mixtures in considering the 005 application and neither did the Full Court.

The decision of the Delegate on the application to amend the divisional 141 On the application by ICI to amend the divisional as filed, the Delegate concluded at [17] and [18] that Branson J was aware that claim 1 of the 005 application and that the 176 application, in particular claim 10, included the possibility of a ternary mixture. The Delegate also concluded at [21] that, from her Honour's reasons, Branson J came to the conclusion that a person skilled in the art would have regarded the Lubrizol patent as providing clear and unmistakable directions to use any heat transfer fluid (or binary mixture) of the broad class described. This extended to all blends encompassed by the broad terminology, not just the preferred forms. However, more complex blends were not part of the teaching. 142 The Delegate posed the question: '[b]ut what of ternary blends, and in particular the specific ternary blend of the present application?' (at [22]). The Delegate observed that Branson J may not have appreciated that the single ternary blend of the 005 application was also a blend of the 176 application. The Delegate concluded that her Honour had decided that the Lubrizol patent contained clear and unmistakable directions to use the ternary blend of claim 1 of the divisional. The Delegate refused ICI's application to amend the divisional to mirror claim 4 of the 005 application.

The decision of Crennan J on the appeal from the Delegate's decision 143 On appeal from the decision of the Delegate on the application to amend the divisional, Crennan J considered the findings of Branson J. Du Pont sought leave to appear as amicus curiae but leave was refused. 144 At [45] Justice Crennan concluded that it was sufficient for Branson J to find that claim 1 of the 005 application lacked novelty. It had not been necessary to deal with claim 4; as a dependent claim, it "fell" with claim 1. Justice Crennan was of the view that Branson J did not make the narrow finding that Integer A of the Lubrizol patent, or the Lubrizol patent as a whole, taught or disclosed for novelty purposes the specific ternary mixture of claim 4. That "narrow" finding was unnecessary and not a fact fundamental to the decision (Blair v Curran (1939) 62 CLR 464 at 532 per Dixon J). Rather, the finding was that Integer A encompassed the ternary and higher mixtures of HFCs, the implicit finding being that Lubrizol taught mixtures of HFCs. 145 Justice Crennan discussed the "difficult" questions of anticipation that arise when a broad chemical claim in an alleged anticipatory document, in its terms, encompasses many, even thousands of compounds or whether such a broad claim discloses 'sub silentio' a particular compound from the broad class (at [64]). Her Honour concluded that the 'brief decision' of the Full Court on the application for leave to appeal the decision of Branson J 'does not foreclose any question, arising on different evidence, as to whether the Lubrizol patent teaches the specific ternary compound of HFCs' the subject matter of the proceeding before her Honour (at [67]). Justice Crennan determined, on the evidence before her as to the disclosures of the Lubrizol patent in the context of the specific mixtures which clarified those disclosures, that the Lubrizol patent did not teach the specific ternary mixture of claim 1 of the divisional, which was novel over Lubrizol (at [22]). Her Honour noted that this was consonant with Branson J's decision on the second patent application (at [84]). Her Honour held that Branson J's decision did not stand in the way of the proposed amendment (at [85]). 146 Justice Crennan was of the view (at [85]) that the finding of lack of novelty of the ternary mixture was not fundamental to Branson J's decision, as that phrase was explained by Dixon J in Blair and that, in view of her Honour's reasons at [144], the absence of such a finding was 'careful and deliberate' on the part of Branson J.

The decision of the primary Judge 147 The primary judge agreed with the decision and reasons of Crennan J. His Honour concluded at [28] that the decision of Branson J 'was entirely based upon consideration of claim 1 in the light of the Lubrizol patent and the Full Court, on the leave application, approached the matter in the same way'. Having found that claim 1 was anticipated, 'it followed' that the dependent claims fell. 148 The primary judge considered that the question of whether the specific ternary mixture was disclosed, the "narrow question" described by Crennan J, 'simply never arose' before Branson J (at [28]). His Honour observed that, when Branson J considered disclosure of specific mixtures not described in the Lubrizol patent in the context of the 176 application, the approach was the same and the conclusion the same in substance as that of Crennan J. If Branson J had directed attention to each of the claims of the 005 application and the 176 application, the reasoning would have led to a finding of anticipation of the claim to the ternary mixture by the Lubrizol patent when claimed in the 005 application but not in the 176 application. The primary judge observed at [28] that, if this were the case, the 'inconsistency between the two approaches by Branson J would be startling'. In addition, his Honour considered that it would be surprising if Branson J and the Full Court had dealt with the question of novelty of claim 4 of the 005 application 'in the pre-emptory way that they respectively did if it had been considered as an independent claim', without a discussion of the issues outlined by Crennan J and dealt with, shortly, by Branson J with respect to the 176 application. 149 Accordingly, the primary judge held that Du Pont failed to establish an issue estoppel with respect to the divisional (at [29]).

CONSIDERATION 150 Du Pont relies on the following propositions to establish its contention that there is an issue estoppel arising from Branson J's reasons for judgment:

The invention of claim 1, which includes the ternary invention, lacks novelty over the Lubrizol patent.
It is impermissible to have resort to the reasons relating to the 176 application.
In any event, the remarks concerning novelty of the claims of the 176 application were brief and obiter remarks.
If regard is had to the reasons concerning the 176 application, 'properly understood' they do not affect the reasons concerning the 005 application sufficiently to undermine the issue estoppel which otherwise arises.

The first proposition 151 Du Pont's contention is as follows. Even if Branson J's decision on the 005 application were restricted to a consideration of claim 1, that claim referred to only four mixtures; three binary and one ternary. Properly construed, her Honour's reasons for finding that claim 1 lacked novelty over the Lubrizol patent applied to both binary and ternary combinations. It is impermissible to assume that her Honour did not consider the ternary combination and then to find a way to justify that assumption. Du Pont relies on the reference to mixtures of HFCs, not restricted to binary mixtures, the acknowledgment that Integer A of claim 1 of the Lubrizol patent encompassed blends of two, three or more HFCs and the fact that the 005 application claimed both binary mixtures and a ternary mixture. It also relies on Branson J's conclusion that the 005 application was deprived of novelty, that is the whole application, including claim 4 to the ternary mixture. Du Pont relies upon the absence of consideration of any amendment that might have saved claim 1 and the dependent claims, including claim 4, if only the binary mixtures were anticipated. 152 To found an issue estoppel, Du Pont needs to establish that Branson J not only found that some component of claim 1 of the 005 application was anticipated by the Lubrizol patent but also that the ternary mixture was so anticipated. 153 I agree, with respect, with what was said by Crennan J at [64]-[65] and [68]and by the primary judge at [28] concerning the difficult questions that arise when a prior publication discloses a class of compounds and a subsequent claim is to a particular member of the class. As already noted, the language of the experts and of Branson J contained ambiguities that needed clarification if this question were to be addressed. If Branson J had directed her mind to the subject matter of claim 4, there would have at least been a discussion of the matters raised by Crennan J. Although Branson J seems to have rejected the distinction between "intellectual content" or enabling disclosure and anticipation, her Honour did consider in the context of the 176 application, whether the disclosure of a class of HFCs gave 'clear and unmistakable directions' to particularised mixtures (at [144] citing General Tire at 486 per Sachs LJ). That is, her Honour did direct her mind to the broader question of disclosure of specific mixtures by a broad description of a claim when the issue arose. 154 The primary judge considered it surprising, if Branson J and the Full Court intended to deal with claim 4 of the 005 application or the claim to the ternary combination in claim 1, that it was not given more consideration (at [28]). As Crennan J pointed out, the question of a broad anticipatory claim and the extent to which it anticipates each of the possible members of a class is not necessarily straightforward and warrants attention (at [64]). When Branson J gave that issue attention in the 176 application (at [144]), her Honour came to the same conclusion in substance as that of Crennan J with respect to the anticipation by the Lubrizol patent of the corresponding claim in the divisional. This was commented upon by the primary judge who observed that the necessary inconsistency within her Honour's reasons, if Du Pont were correct, 'would be startling' (at [28]). 155 The question of "enabling disclosure", or in the words of the Delegate "technical disclosure", by a prior publication of compounds the subject of a later specific claim is not straightforward and is a matter that would be the subject of some discussion. In the present case, it cannot be said that the evidence was clear. It is not clear whether the witnesses, one of whom only knew of ten HFCs at the relevant time, would have considered the specific ternary mixture disclosed or taught by the general disclosure of the Lubrizol patent. This is not necessarily the same concept as agreeing that specific HFCs were "encompassed by" or were "included" in the general description. There is a difference in the assessment of novelty between a claim specifically to the ternary mixture (claim 4) and a claim (claim 1) to a class of compounds that includes the ternary combination. It is to be expected that it would have been the subject of some consideration, as it was for the 176 application. When Crennan J directed her attention to the disclosure of the ternary mixture of the 005 application, her Honour received further evidence and found that claim 4 was novel over the Lubrizol patent (at [83]). 156 It cannot be said that Branson J clearly decided that the prior disclosure of the class necessarily discloses each and every member of the class. That is apparent when the reasoning with respect to the 005 application is taken in isolation. There was no attention given to that issue. When the whole of the reasons are read, it is apparent that her Honour did not consider that this follows, as she found to the contrary with respect to the 176 application. The likely explanation is that her Honour did not direct her attention to the dependent claims of the 005 application as she did not need to do so. Those claims, in the absence of an application to amend, fell with claim 1. 157 The alternative explanation offered by the primary judge for the finding that the 005 application, that is all claims, were not novel was that, in a drafting sense, claim 4 as a dependent claim fell within claim 1 (at [28]). Du Pont says that there was no application to amend, which would have been expected if ICI wished to save the dependent claim. However, ICI took a separate course. ICI did not take the course of amendment and was not obliged to do so. Accordingly, Branson J was not obliged to consider amendments in the absence of any application. ICI took an alternative available course to amendment, to file the divisional. 158 It is apparent from the submissions before Branson J that Du Pont did press for a finding of lack of novelty of the invention the subject of each of the claims of the 005 application. It was put by Du Pont that claims 2, 3 and 4 were subsets of the refrigerants claimed in claim 1. Submissions were directed to all of the refrigerants of the 005 application. It is also most likely that the questions to Mr Symons and Professor Ball were intended to make that clear. However, it is not clear from Branson J's reasons that her Honour directed her attention to the subject matter of the dependent claims. In particular, it is hard to reconcile a specific finding of lack of novelty of claim 4 to the ternary mixture in the 005 application with a finding of novelty of claim 10 to the same ternary mixture in the 176 application. 159 In circumstances where it is not immediately apparent from the reasons, it is appropriate to consider not only what Branson J decided but also what was legally indispensable to her Honour's decision. An issue estoppel only covers the actual ground upon which the finding of novelty was made (Blair at 532). 160 As there is an alternative explanation to the finding of lack of novelty of the 005 application; a finding directed only to the independent claim 1, it cannot be said that Branson J's decision was directly upon the novelty of claim 4 as if it were an independent claim (Duchess of Kingston's Case (1776) 2 Smith LC (13th ed) 644 at 645). It may have been put in issue by Du Pont but it was not actually decided or found in the proceedings (Outram v Morewood (1803) 3 East 346 at 354; 102 ER 630 at 633). It was not necessarily decided by the finding that the whole of the 005 application lacked novelty, as claim 4 was dependent on claim 1 (R v The Inhabitants of the Township of Hartington Middle Quarter (1855) 4 El & Bl 780 at 794; 119 ER 288 at 293). It cannot be said that that there was a legally indispensable finding that the claim to the ternary mixture was not novel (Blair at 532). It was not necessary or indispensable for her Honour's conclusion of lack of novelty to find that each and every one of the mixtures of HFCs of claim 1 of the 005 application, individually the subject of dependent claims, was anticipated by the Lubrizol patent. A finding of lack of novelty of claim 4 was not in substance the ratio of, or fundamental to, the decision on the 005 application (Hoystead v Commission of Taxation [1926] AC 155 at 170) and was contrary to the decision on the 176 application, in the same set of reasons. 161 It is not necessary for the purposes of this appeal to make any observations on whether her Honour's apparent rejection of the concept of enabling disclosure was correct. The invention as claimed in claim 1 was anticipated and, in the absence of amendment, the dependent claims fell with it. That does not amount to a finding that the invention so far as claimed in claim 4 of the 005 application was anticipated by the Lubrizol patent. It was not 'necessarily and directly decided' (Ramsay v Pigram (1968) 118 CLR 271 at 276 per Barwick CJ) that the ternary combination was anticipated; a finding that anything within the scope of claim 1 was anticipated was sufficient for a finding of lack of novelty of claim 1. In the absence of amendment, it followed that the remaining claims fell and it could be said as did Branson J at [71] that the patent application lacked novelty. Her Honour did not express her conclusion in terms of, for example, 'the invention as claimed in each of the claims of the patent application' (cf s 18(1) of the Patents Act 1990 (Cth)). 162 Therefore it cannot be said, as Du Pont submits, that the ternary combination in claim 1 of the divisional was found to lack novelty over the Lubrizol patent. Du Pont's first proposition therefore does not establish its contention of an issue estoppel arising from Branson J's decision.

The reasoning on the 176 application: The second, third and fourth propositions 163 Du Pont submits that it is impermissible to consider what her Honour said about the 176 application in order to understand the finding of lack of novelty of the 005 application and to determine what was "legally indispensable" to that finding. Those remarks were, in any event, Du Pont submits, obiter dicta, and could not qualify an issue estoppel in relation to the 005 application. 164 If the remarks were to be taken into account, Du Pont draws attention to the differences between the subject matter of claim 1 of the 005 application and claim 1 of the 176 application. The number of refrigerant mixtures of claim 1 of the 176 application were forty-four in number, including thirty-four possible ternary mixtures. Du Pont submits that, while the mixture of R-125, R-134a and R-32 was present in both applications, this 'may simply have been overlooked by Branson J' in her Honour's consideration of the 176 application. 165 It is correct that the 176 application was a separate application but Branson J delivered a single set of reasons and the evidence given on the issue of the disclosures of the Lubrizol patent was given for the purposes of both applications. It is permissible to consider the totality of her Honour's reasons. Justice Branson herself made reference at [143] in considering the 176 application to the reasoning in respect of the 005 application. The reasoning on the 005 application was integrated with the reasoning on the 176 application. The reason why the remarks concerning the novelty of the 176 application were brief was, in part, because her Honour did not need to repeat what had been said about novelty in the context of the 005 application. Properly understood, Branson J's conclusion, that the Lubrizol patent did not deprive the claims of the 176 application of novelty (at [144]), recognised that specific ternary and higher mixtures of HFCs including the subject of claim 10, being the ternary mixture of claim 4 of the 005 application, were not disclosed in a novelty defeating sense in the prior publication. This is consonant with the finding that the 005 application was not novel, if it is accepted that her Honour only made a decision on the independent claim of that application, on which the remaining claims depended. 166 Accordingly, Du Pont's second, third and fourth propositions do not establish its contention that there is an issue estoppel arising from Branson J's reasons for judgment.

The fifth proposition 167 Du Pont raises a fifth proposition arising from the judgment of the Full Court: 5. The Full Court described Branson J's finding, that the use of a liquid composition described in claim 1 of the 005 application is clearly disclosed by the Lubrizol patent, as correct and attended with no doubt. 168 Du Pont relies on the finality of a decision refusing leave to a patent applicant to appeal from a decision upholding an opposition and the standard to be met by an opponent. Du Pont contends that the Full Court determined that the use of a liquid composition described in claim 1 of the 005 application was clearly disclosed by the Lubrizol patent. That finding was "legally indispensable" to the conclusion that leave to appeal should be refused because it was clear beyond doubt that there was no error on the part of Branson J and that any patent if granted would be revocable. Accordingly, Du Pont says, there is an issue estoppel and, as the ternary combination is described in claim 1, ICI is precluded from further agitating that issue. 169 I agree with the primary judge (at [28]) that the Full Court approached the matter in the same way as Branson J. 170 The Full Court did not need to consider the novelty of the dependent claims or the ternary mixture and did not do so. It is not necessary to decide whether a decision of the Full Court on an application for leave to appeal is available to found an issue estoppel. No such issue estoppel arises.

Conclusion 171 Du Pont submits that Branson J's finding was that all four of the combinations of claim 1 of the 005 application were anticipated by the Lubrizol patent and that it follows that the dependent claims to the individual mixtures were thereby also not novel. Du Pont says that this interpretation is preferable to that advanced by ICI, that it was sufficient that something within claim 1, the binary mixtures, were anticipated but not the ternary mixture. That interpretation may be open but it is not clear. Where a broad class is claimed and a later patent application claims specific members of the class, issues of enabling disclosure and selection would normally arise. The Full Court specifically noted that the issue of selection did not arise (at [19]). Justice Branson recognised the further complexity which arose in the 176 application. 172 Her Honour's decision on the 005 application may be consistent with the interpretation advanced by Du Pont. However, it is also consistent with the interpretation advanced by ICI. I agree with the primary judge at [29] that there is no identity between any issue found by Branson J and confirmed by the Full Court on one hand and any issue for determination in the opposition to the divisional on the other. Specifically, it has not been determined between the parties whether the Lubrizol patent deprives the claim to the ternary mixture of novelty. 173 It cannot be said that the issue estoppel sought to be raised is certain. As Du Pont recognised, this does not satisfy the necessary onus in this appeal to establish an issue estoppel. There is no issue estoppel. 174 The appeal should be dismissed with costs. Du Pont should also pay the Commissioner's costs of the appeal.

I certify that the preceding eighty-seven (87) numbered paragraphs are a true copy of the Reasons for Judgment herein of the Honourable Justice Bennett.

Counsel for the Appellant: Mr J T Gleeson SC, Mr W Flynn

Counsel for the First Respondent: Mr D K Catterns QC, Mr C Dimitriadis

Solicitor for the First Respondent: Phillips Ormonde & Fitzpatrick Lawyers

Counsel for the Second Respondent: Ms K J Howard SC

Solicitor for the Second Respondent: Australian Government Solicitor

Date of Hearing: 24 and 25 May 2007

Date of Judgment: 22 October 2007

Patents Act 1990(Cth)ss 18(1)(a), 18(1)(b)(i), (ii), 40(2), 51

At a glance

Court

Decision date

Before

Source

Judgment (35 paragraphs)

Parties

Legislation Cited (1)

Cases Cited (21)

Cited By (3)