Narellan Pools Pty Ltd v Huntsman Chemical Company Australia Pty Ltd

Narellan Pools Pty Ltd v Huntsman Chemical Company Australia Pty Ltd - [2010] FCA 267 - FCA 2010 case summary — Zoe

[7]

Background 1 The first applicant ('Narellan') is a fibreglass swimming pool manufacturer in Australia and, amongst other things, is in the business of researching, designing, manufacturing, marketing and selling fibreglass shells for use as swimming pools ('fibreglass pools'). The second applicant ('Narellan Franchise') trades as part of the Narellan group of companies and is responsible for, inter alia, the management of a franchise scheme for the sale and distribution of fibreglass pools manufactured by Narellan. 2 The respondent ('Huntsman') is in the business of manufacturing and selling, amongst other things, industrial resins. Huntsman has the use of vinyl ester technology developed by Ashland Inc ('Ashland'), a United States based chemical manufacturer, by virtue of a licence agreement entered into between Huntsman and Ashland in or about 1986. Many of the products developed by Huntsman are based on, even if only loosely, Ashland's formulations. 3 In or about October or November 2000, Narellan, through Mr Douglas Turner, approached Huntsman, through Mr Malcolm Holden, who was the New South Wales Sales Manager for Huntsman, as to the possibility of obtaining a low styrene emission vinyl ester resin for use in the production of its fibreglass pools as a back-up to Derakane, the Fiberglass International ('FGI') product, it normally used. Such resins are used in applying the 'tie layer' of a pool mould or what is also known as the corrosion barrier or the vinyl ester layer. It is used to wet out glass fibres in order to create a solid mass of fibreglass. Its application comes after the application of the gelcoat or cosmetic layer and before the application of the first structural or polyester layer. 4 Huntsman, through Mr Holden, indicated that it had a vinyl ester resin which would be suitable for such use and that laboratory tests for water resistance and failure had been completed. Mr Holden undertook to get a sample to Mr Turner. Approximately three weeks later, Huntsman delivered a tin containing 3 to 5 kg of Hetron 942/35 PAS resin ('Hetron 942') to Narellan to enable Mr Turner to conduct a hand laminate test ('the testing sample'). Mr Turner also used part of the testing sample to conduct production testing which lasted about 14 days. Relevantly, 'Hetron' refers to the brand name of the resin; '35' indicates the percentage of styrene in the resin; and PAS: 'P' stands for promoted; 'A' stands for amine boosted; and 'S' stands for summer grade - for use in summer. Hetron 942 is a vinyl ester resin made up of two other vinyl ester resins, Hetron 914 and Hetron 922, both of which are based on Ashland's technology. 5 In or about late November or December 2000, a sample batch of approximately 200 kg of Hetron 942 ('the production sample') was delivered by Huntsman to Narellan. Between 10 and 11 January 2001, Narellan manufactured two pools and one spa (nos. 7775, 7777 and 7781) using the production sample. The production of these three shells using the production sample was uneventful. 6 On 5 February 2001, Mr Turner, on behalf of Narellan, placed an order for three tonnes of Hetron 942. Pursuant to this order, on 26 February 2001, Huntsman manufactured 6.83 tonnes of Hetron 942 as Batch 053109001 ('Resin 942'). Huntsman supplied 3 tonnes of Resin 942 to Narellan on 2 March 2001 ('the March Supply'). The March Supply was exhausted by Narellan in the production of fibreglass pools. In September 2001, Huntsman (again through Mr Holden) offered to supply Narellan with the balance (3.83 tonnes) of the Resin 942. Narellan was informed that the balance had been re-tested and remained within Huntsman's specification limits. A further 3.83 tonnes was delivered to Narellan on 14 November 2001 ('the November Supply'). The November Supply was also exhausted by Narellan in the manufacture of fibreglass pools. The total quantity of Resin 942 supplied was used as resin in the tie layer of these pools; according to Narellan exclusively in the tie layer of 61 pools and, by reason of the mixing of Resin 942 with Derakane on changeover from or to Resin 942 during the manufacture of a pool shell, together with that other resin in the tie layer of 17 pools ('mixed resin pools'). 7 Huntsman has never supplied any of its Hetron 942 to any pool manufacturer before or since the deliveries to Narellan referred to in [5] and [6] above, nor has Hetron 942 been used in manufacturing swimming pools outside Narellan's premises. 8 In 2003, Narellan received a complaint regarding a fibreglass pool manufactured with Resin 942. Upon further investigation, Narellan discovered the pool had osmosis and was manifesting cracks, blisters and cobalt stains. Further complaints were received by Narellan relating to other pools manufactured with Resin 942 which were all exhibiting osmosis. Out of the 61 pools identified by Narellan as having been made exclusively with Resin 942 in the tie layer, at least 33 (8027, 8033, 8034, 8036, 8037, 8041, 8042, 8043, 8045, 8046, 8047, 8048, 8050, 8051, 8052, 8055, 8058, 8753, 8756, 8757, 8759, 8973, 8978, 8981, 8983, 8985, 9002, 9008, 9009, 9012, 9015, 9016, 9017) are said by Narellan to have failed due to the incidence of osmotic blistering. Such blistering occurs when water penetrates through the gel coat and mixes with water soluble materials found in the fibreglass layers beneath it. Huntsman maintains that evidence regarding the failure of pools 8033, 8037, 8048, 8052, 8055, 8985 and 9016 has not been adduced although, in the case of pool 8037, the consensus of the experts would seem to be that it has failed (Ex. 8; second conclave report, App 1, p 8; see [94] below). Out of the 17 pools identified by Narellan as being mixed resin pools, at least two (8980 and 8999) have failed.

[9]

The Claims and Pleadings 10 In their Amended Application ('AA') filed 17 October 2008 the applicants seek damages from Huntsman for breach of contract, negligence and for contravention of s 52 of the Trade Practices Act 1974 (Cth) ('TPA') as well as declaratory relief, that Huntsman has engaged in conduct towards the applicants which is misleading and deceptive. 11 On the same day, the applicants' senior counsel made it clear that the action in negligence was abandoned and so much is reflected in the applicants' Amended Statement of Claim ('ASC') also filed that day. The ASC also reflects at para 36 that the applicants rely on alleged contravention of s 53(a) of the TPA, as well as s 52; but that, contrary to the AA, only Narellan claims damages for breach of the two supply contracts. So much is consistent with the basis upon which, on 14 October 2008, pursuant to O 6 r 8 of the Rules, I granted leave to add Narellan Franchise to this proceeding, namely, that its claim was confined to that of a party who had suffered loss by Huntsman's alleged contraventions of the TPA. 12 The applicants, in paras 23 - 26 inclusive of the ASC, pleaded: '23. Before [Narellan] commenced using the Resin 942 in the manufacture of swimming pools, there were no failures or defects with the pools manufactured by [Narellan] in anyway related to the resin then utilised by [Narellan]. 24. Save for one instance in respect of a pool manufactured in 2000 (against a total of over 16,000 pools produced) defects of full osmosis have only occurred in the pools manufactured by [Narellan] using the Resin 942. 25. Upon the discontinuance of the use of the Resin 942 in production in January 2002, [Narellan] has experienced no failure due to osmosis in the product of fibreglass swimming pools. 26. At all material times, the only variable changed by [Narellan] in the production of swimming pools was the use of the Resin 942. There was no change in the methodology utilised in the production of swimming pools.' 13 In its Amended Defence ('AD'), Huntsman does not admit the allegations in paras 23, 25 and 26. Nor does it admit in answer to para 24, that Narellan has only suffered defects of 'full osmosis' save for one instance in respect of a pool manufactured in 2000 against a total of over 16,000 pools produced. It denies the balance of the allegation in para 24 and says, by Narellan's own records, pool failures have been reported in pools manufactured not using Resin 942. 14 Those matters which Huntsman did not admit were never put in issue either by way of leading evidence to the contrary, in cross-examination of Narellan's witnesses, in particular Narellan's managing director, Mr Christopher Meyer, or by way of submission. In short, they are not in dispute. What is disputed, and hence denied by Huntsman in response to para 24 of Narellan's ASC, is Narellan's allegation that defects of 'full osmosis' have only occurred in the pools manufactured by Narellan using Resin 942. According to Huntsman, Narellan's own records put lie to this.

[11]

The First Representations 16 In para 9 of the ASC it is pleaded that in or around November 2000, Huntsman represented to Narellan that: 'a. Hetron 942/35 epoxy vinyl ester resin would be suitable for use by [Narellan] in use in the construction of fibreglass swimming pools; b. Adequate laboratory tests for Hetron 942/35 epoxy vinyl ester resin had been completed.' 17 The first representations are alleged to have been made by Mr Holden for Huntsman to Mr Turner of Narellan in November 2000. The representation pleaded in para 9a is admitted by Huntsman. Huntsman, obliquely, put in issue Narellan's reliance on the representation but neither Mr Meyer nor Mr Turner were challenged on their evidence of reliance, nor was any evidence led by Huntsman to disprove that reliance. In my view, it is not open on the evidence before me to find that the representation in para 9a of the ASC, while admitted by Huntsman to have been made, was not relied on by Narellan. On the contrary, I find that it was relied on by Narellan. 18 The representation pleaded in para 9b is denied by Huntsman. 19 Mr Turner's evidence in chief on the para 9b representation was that Mr Holden had said to him: The laboratory tests for our resin for water resistance and failure had been done. This evidence was not challenged in cross-examination; indeed, he was not asked any questions on the subject. Moreover, Mr Holden was not called by Huntsman. Nevertheless, Huntsman submitted that this evidence does not go far enough; it does not amount to a representation that 'adequate laboratory tests … had been completed'. I cannot agree. Having regard to the primary function of the tie layer in which the resin was to be used - to prevent ingress of water (and any other chemicals dissolved in pool water) which could compromise the integrity of a fibreglass pool; to the context in which Mr Holden made the statement - to a prospective customer which would carry out non-laboratory testing of the product prior to placing a manufacturing order; and to Huntsman's admission that it made the representation in para 9a, I am of the view, and so find, that the words which Mr Turner deposed Mr Holden used, do amount to a representation that adequate laboratory tests were carried out and that the product passed those tests. 20 Huntsman put in issue whether Mr Turner, and through him, Narellan relied on the representation in para 9b of the ASC. First, it submitted that Mr Turner relied on his own testing to satisfy himself as to the suitability of the product. Second, that he, Mr Turner, was too experienced to rely on the sales pitch of Mr Holden. 21 Mr Turner's evidence as to the testing he carried out is contained in both of the affidavits (Ex 3) he swore for the purpose of this proceeding. In his affidavit of 5 December 2006 ('first affidavit'), he deposed: '13. I undertook the standard production testing, which is to test: a) Gel time. b) Wet out. c) Exertherm [sic] 14. The production and factory testing that was in place for the resins was as follows: a) Gel time - this is the time it takes for the chemical reaction to take place and the resin to turn to a jelly like state; b) Thixotropy - if the resin runs out of the glass fibres after sprayed and holds up on a vertical panel as it should c) Wetability - whether or not the resin wets out the glass fibres (however, the type of glass also plays a large part in this process); d) Cure time - the length of time it take [sic] from Gel time to Peak Exotherm (the highest temperature). e) Peak Exotherm - the highest temperature reached after the Gel time. 15. The handling characteristics of the resin was [sic] very good and I made the decision to use this vinyl ester as a back-up for FGI products. 16. After the testing was complete, I was satisfied the resin would be suitable and I placed the first order with Mal. 17. The factory and production resin testing that I carried out is ad hoc and only gives a guide as to how resins react while in production. Fit for purpose testing formulas come from the resin supplier's laboratory.' 22 In his affidavit of 2 October 2008 ('second affidavit'), Mr Turner deposed: '4. I refer to paragraph 12 of my first affidavit. What happened in relation to the sample was this: in about October or November 2000, Huntsman delivered a tin containing 3 to 5kgs of Hetron 942-35 PAS resin (Hetron 942) to Narellan Pools to enable me to conduct a hand laminate test. This test is conducted by: (i) Pouring the Hetron 942 into a container; (ii) Adding catalyst to the container; (iii) Pouring the mixture over a fibreglass base and rolling it out. The hand laminate test is a visual test, the purposes of which is to get a rough idea of whether the Hetron 942 coats the fiberglass [sic] and wets out the glass fibers [sic]. The Hetron 942 appeared to wet out the glass fibers [sic] and the laminate appeared to cure within a reasonable time. 5. I also used part of the 3 to 5kgs of Hetron 942 to conduct the production testing referred to in paragraphs 13 and 14 of my first affidavit. The production tests on the 3 to 5kgs of Hetron 942 lasted about 14 days. … 7. The factory and production resin testing gives a guide as to how resins initially react while in production. I did not chemically test the Sample Hetron 942. I did not have such facilities and, in any event, as noted in paragraph 11 of my first affidavit, Mr Holden had advised me that laboratory testing had been conducted and the Sample Hetron 942 satisfied those tests. 8. In February 2001, I again spoke with Mr Holden. To the best of my recollection, during that conversation, words to the following effect were said: Me: "Mal, I have tested the sample batch in production and it appears OK." Mr Holden: "It is a very good resin product and will be excellent for Narellan. Will you be ordering some?" Me: "Yes. I will send you over an order."' 23 On Huntsman's submission that Mr Turner relied on his own tests and not those of Huntsman, Mr Turner was not asked any question about the tests carried out by him. Nor was he asked any questions about what he said at para 7 of his second affidavit. Indeed, as Narellan submitted in support of a finding of reliance: 'Not a single question was asked of Mr Turner concerning this representation including whether or not he relied upon it … In those circumstances, it is difficult to see how a submission can be maintained that there was insufficient evidence of reliance …' 24 On Huntsman's submission that Mr Turner was too experienced to rely on the sales pitch and 'puffery' of Mr Holden, again it was not put to Mr Turner that his experience meant that he relied on nothing said by Mr Holden. 25 Consequently, as well as finding that the representation in para 9b of the ASC was made by Huntsman through Mr Holden to Narellan through Mr Turner, I am also of the view that the evidence supports a finding of reliance by Narellan on that representation, and I so find.

[12]

The Second Representations 26 In para 11 of the ASC it is pleaded that from on or around 5 February 2001 until the use of the Resin 942, Huntsman represented to Narellan that: 'a. the Hetron 942/35 epoxy vinyl ester resin which was to be (and from 26 February 2001 had been) manufactured and was proposed to be supplied ("Resin 942") was suitable for use by [Narellan] in the construction of fibreglass swimming pools. b. the Resin 942 proposed to be supplied to [Narellan] was the same as the Resin 942 supplied to [Narellan] in the Sample.' 27 It is alleged that the second representations were express or implied or partly express and partly implied; to the extent they were express, it is alleged that Mr Holden on behalf of Huntsman represented to Mr Turner on behalf of Narellan the second representations in the circumstances set out in para 8 of his second affidavit; to the extent they were implied, it is alleged that they were made by the offer of the supply in circumstances where Huntsman knew the purpose for which the Resin 942 was to be used and that it had previously supplied the Sample. Additionally it is alleged that the second representations were continuing representations as, until the time of the use of the Resin 942, Huntsman had not communicated any fact inconsistent with the second representations in circumstances where if such a fact existed Narellan was entitled to expect that such information would be conveyed to it. 28 Both representations in paras 11a and 11b are denied by Huntsman. 29 Huntsman accepted in its submissions that the representation in para 11a was 'almost identical' to that in para 9a, which it admitted having made. Moreover, the representation in para 11a is pleaded on the same facts as a continuing representation. As an express representation, reliance is placed on para 8 of Mr Turner's second affidavit (see [22] above). Mr Turner was not asked any questions concerning the conversation he had with Mr Holden in February 2001 deposed to in that paragraph and, as noted in [19] above, Mr Holden was not called by Huntsman. In these circumstances, I am unable to comprehend why I should not find that the representation was made, and I so find. 30 For the same reasons outlined in [17] above in relation to the representation in para 9a of the ASC, I also find that the representation in para 11a was relied upon by Narellan in deciding to order the March Supply. 31 There is no issue that the 'Sample' referred to in para 11b (and para 12b) of the ASC is a reference to both the testing sample (see [4] above) and the production sample (see [5] above). It is also common ground that there was no express representation in the form of a statement by Mr Holden to Mr Turner that the March Supply would be 'the same' as the Sample. 32 Huntsman submitted that because of the imprecision and ambiguity of the pleaded expression 'the same', the Court ought not find that Huntsman made any implicit representation other than that the resin supplied would, like the Sample, be Hetron 942 and not some other kind of vinyl ester resin. But such a concession, in my view, necessarily involves a finding that a representation was made that the constituent elements of the March Supply and the November Supply would be the same as the constituent elements of the Sample. Huntsman submitted that they were, by reference to a table, prepared from data taken from production sheets. As Narellan submitted, it is beyond dispute that what was delivered to Narellan by the March Supply and the November Supply was not what was stated on the production sheet for Hetron 942. The experts agreed that the Resin 942 contained 450 kg of unknown substance, which Huntsman could only speculate as being more Hetron 922 or 914 not recorded by the operator; and that it contained 2,000 kg of a substance with an infra-red signature consistent with terephthalate, although Huntsman consistently maintained that it was not open to the Court to find that it was delivered to Narellan containing that contaminant. A finding was also open on the evidence that the March Supply, as well as the November Supply which was admitted to be out of 'spec', was likely to be out of 'spec' at the time of the fabrication of the March 2001 pools. These are matters to which I return later in these reasons. For present purposes, it suffices that I find, contrary to Huntsman's AD (para 11.2), that the representation in para 11b was made. 33 I also find that the representation in para 11b was relied upon by Narellan. At para 13 of his second affidavit, Mr Turner deposed that: 'I would not have placed the orders for the supply of any of the Hetron 942 if I had of known that the Hetron 942 was not made from the same constituent elements as the Sample … It never occurred to me that this would not be the case as I thought this was the whole point of obtaining a sample.' He was not challenged on this evidence; indeed, he was not asked any question concerning the matter. Huntsman submitted that Mr Turner placed no reliance on the Sample as he did not chemically test it. However, Mr Turner did not test the Sample chemically because he relied on Mr Holden's representation that Huntsman had completed adequate laboratory testing; moreover, he did not have the facilities to chemically test the Sample.

[13]

The Third Representations 34 In para 12 of the ASC it is pleaded that from 19 September 2001 until the use of the Resin 942, Huntsman represented to Narellan that: 'a. the balance of the Resin 942 which had not been supplied was suitable for use by [Narellan] in the construction of fibreglass swimming pools; b. the balance of the Resin 942 proposed to be supplied to [Narellan] was the same as the Resin 942 supplied to [Narellan] in the Sample; c. the balance of the Resin 942 which had not been supplied had been retested and was still "in spec".' 35 It was alleged that the representations pleaded in paras 12a and 12c were partly express and implied; to the extent they were express, it is alleged that they were made in writing in a facsimile from Mr Holden to Mr Turner on 19 September 2001; to the extent they were implied, it is alleged that they were made by the offer of supply of the balance of the Resin 942 in circumstances where Huntsman knew the purpose for which the balance of the Resin 942 was to be used and that it had previously supplied the Sample and the March Supply for that purpose. It is alleged that the representation pleaded in para 12b was implied in that it was made by the offer of supply of the Resin 942 in circumstances where Huntsman knew the purpose for which the Resin 942 was to be used and that it had previously supplied the Sample. It is also alleged that the third representations were continuing representations as until the time of the use of the Resin 942, Huntsman had not communicated any fact inconsistent with the third representations in circumstances where if such a fact existed, Narellan was entitled to expect that such information would be conveyed to it. 36 The representation pleaded in para 12a is admitted by Huntsman. Huntsman made no substantive submission denying Narellan's reliance on the representation and for the same reasons outlined in [17] above in relation to the representation in para 9a of the ASC, I also find that the representation in para 12a was relied upon by Narellan in deciding to order the November Supply. 37 The representation pleaded in para 12b is denied by Huntsman. For the same reasons given in [32] above for my finding that, contrary to Huntsman's AD (para 11.2), that the representation in para 11b was made, I also find that, contrary to Huntsman's AD (para 12.2), the representation in para 12b was made. For the same reasons given in [33] above, I also find that the representation in para 12b was relied upon by Narellan. 38 In its AD, Huntsman answered Narellan's pleading in para 12c in the following terms: '[S]ave that it admits that by facsimile dated 19 September 2001 that Mal Holden of Huntsman stated that the 3.8MT balance of the resin had been retested and was in specification, denies the allegations contained in paragraph 12c.' In its written submissions, Huntsman admitted making a representation in the terms alleged in para 12c; it also admitted that it was not the case that it was 'in spec': 'Huntsman admits that it made a representation that the second batch of Hetron, supplied in November 2001, was 'in spec', when this was not the case.' 39 Moreover, in the face of Mr Turner's unchallenged evidence at para 14 of his second affidavit, where he deposed that: 'I would not have placed the order for the supply of the remaining Hetron 942 in November 2001 if I had known that the balance of the Hetron 942 had not been retested and was not still "in spec". Reliability and suitability of the resin was a key factor in any purchasing decision for materials to be used in the production process.' unsurprisingly, Huntsman made no submission that the representation was not relied upon by Narellan in deciding to order the November Supply. 40 Nevertheless, Huntsman argued that there was no causal link between that conduct and the loss suffered as: 1. There was no direct evidence that being out of 'spec' caused failures; 2. there was an equal spread of failures across both supplies suggesting that the cause was something else. I return to these matters later in the reasons.

[15]

Representations as to Future Matters 43 In para 35 of the ASC, the applicants pleaded that the representations in paras 9a, 11 and 12a of the ASC are representations as to future matters; that at the time of making them Huntsman did not have a reasonable basis for making them; and that the applicants rely on s 51A of the TPA. In their written submissions, the applicants confined the reference to para 11 as being a reference to para 11a only. 44 In its AD, Huntsman denied the allegations in para 35 of the ASC, but made no response in its written submissions to the applicants' reliance (in their written submissions) on s 51A with respect to the representations in paras 9a, 11a and 12a of the ASC. There can be no doubt that these representations are all representations with respect to future matters. In the case of paras 9a and 11a, the representations are made with respect to a product, Resin 942, to be manufactured in the future, for future use in the manufacture of fibreglass pools and, in the case of para 12a, the representation is made with respect to the suitability of Resin 942 for future use in the manufacture of fibreglass pools. 45 Section 51A of the TPA relevantly provides: '(1) For the purposes of this Division, where a corporation makes a representation with respect to any future matter (including the doing of, or the refusing to do, any act) and the corporation does not have reasonable grounds for making the representation, the representation shall be taken to be misleading. (2) For the purposes of the application of subsection (1) in relation to a proceeding concerning a representation made by a corporation with respect to any future matter, the corporation shall, unless it adduces evidence to the contrary, be deemed not to have had reasonable grounds for making the representation.' 46 In McGrath & Anor v Australian Naturalcare Products Pty Ltd (2008) 165 FCR 230 the Full Court drew together the authorities on s 51A(2). Emmett J stated at [44]: '[44] Under s 51A(1) of the Trade Practices Act, a representation is to be taken to be misleading if it is a representation with respect to any future matter and the maker of the representation does not have reasonable grounds for making the representation. Under s 51A(2), the maker of the representation with respect to any future matter is to be deemed not to have had reasonable grounds for making the representation unless it adduces evidence to the contrary. However, if evidence is adduced by a representor to the effect that the representor had reasonable grounds for making the representation, the deeming provision will not operate. Where the representor adduces such evidence, it is then a matter for the Court to determine, on the balance of probabilities in the ordinary way, whether or not the representor had reasonable grounds for making the representation.' See too Allsop J at [191] and [192] where his Honour analyses what Keane JA (with whom Williams JA and Atkinson J agreed) said in Downey & Anor v Carlson Hotels Asia Pacific Pty Ltd [2005] QCA 199 at [126] - [129]. 47 Narellan pointed out that, significantly, Huntsman had not asserted in its AD any reasonable ground or grounds to justify the making of the future representations that Hetron 942 and, more specifically Resin 942, would be suitable for use by Narellan in the construction of fibreglass pools, nor was any evidence lead in chief that specifically identified such grounds. According to Narellan, Huntsman simply made no attempt to engage on this issue and therefore these representations are presumed to be misleading: Ting & Anor v Blanche & Anor (1993) 118 ALR 543 at 552 - 553 (Hill J). In my view, that must be a correct application of the operation of the section in the circumstances. 48 Leaving aside this point, Narellan submitted that, in any event, in relation to each future representation the evidence established that there was in fact no reasonable ground for making that representation. Dealing with each in turn:

[23]

Contract 50 Narellan's pleading in contract was in the following terms: '37. Prior to, and at the time of making the contracts, [Narellan] expressly made known to [Huntsman] the particular purpose for which it would require Resin 942, namely the production of fibreglass swimming pools ("the purpose"), so as to show, as was the fact, that [Narellan] relied upon the skill and judgment of [Huntsman], and the goods were of a description which it was in the course of [Huntsman's] business to supply. 38. In the premises, it was a condition of the contracts implied by section 19(1) of the Sale of Goods Act 1923 (NSW) ("SGA") that the resin supplied would be fit for the purpose. 39. Further and in the alternative, the Resin 942 was bought by description from [Huntsman] who dealt in goods of that description. 40. In the premises, it was a condition of the contracts implied by section 19(2) of the SGA that the Resin 942 supplied would be of merchantable quality. 41. In breach of the contracts, the Resin 942 supplied was: a. not fit for the purpose; and/or b. not of merchantable quality.' 51 In its AD, Huntsman answered in the following terms: 'a. In answer to 37, [Huntsman]: (i) admits [Narellan] expressly made known to [Huntsman] the particular purpose for which it required resin 942, mainly production of fibreglass swimming pools; (ii) cannot admit the purpose for which such information was imparted; (iii) admits that the resin constituted goods of a description which it was in the course of [Huntsman's] business to supply; (iv) denies the balance of the allegations contained therein. b. In answer to 38, [Huntsman] admits that it was a condition of the contracts that the Hetron 942 supplied by [Huntsman] to [Narellan] would be reasonably fit for the purpose of being used in the production of fibreglass swimming pools, but otherwise denies the allegations in 38. c. In answer to 39, [Huntsman] admits that the Hetron 942 was bought by [Narellan] by description from [Huntsman] who dealt in goods of that description but otherwise denies the allegation contained in 39. d. In answer to 40, [Huntsman] admits that it was an implied condition of the contracts with [Narellan] that the Hetron 942 supplied would be of merchantable quality but otherwise denies the allegations contained in 40. e. [Huntsman] denies the allegation contained in 41.' I deal with the issues raised by these respective pleadings of Narellan and Huntsman at [182] below.

[25]

The Applicants' Case in Opening 53 In opening, it was contended on behalf of the applicants that although superficially complex, on close examination the liability phase of the proceeding was straight forward: in illustrating this proposition, reference was made in opening to 'nine key facts' which, when taken together, would be determinative of the case. In written submissions filed at the conclusion of the hearing, it was submitted that each of these 'key facts' had been proved (subject to slight variation of two of them to take into account developments subsequent to the opening) and remained determinative of all liability issues. They were put as follows: '1. Narellan has manufactured over 16,000 pools essentially without incident save for the pools manufactured with Resin 942. 2. Narellan's manufacturing processes were such "as to avoid incorrect resin application and/or contamination". 3. Huntsman has never supplied any of its Hetron 942/35 epoxy vinyl ester resin to any manufacturer before or since and only the Sample and the Resin 942 have ever been used in manufacturing swimming pools. 4. The Resin 942 (unlike the Sample) was all made in highly unusual and inadequately explained circumstances over three days. 5. The composition of the Resin 942 is uncertain because (a) "of an amount of 400 to 450kg that cannot be accounted for" and (b) because of a question as to the origin of components used. 6. The Sample (which was used successfully) "differed in composition from" the Resin 942 (which was used highly unsuccessfully). 7. A "signature" of a component consistent with terephthalate resin was present in the tie layers of affected pools; its origin is "unknown" and its presence in the tie layer is a contaminant and because of its location is unlikely to be present by reason of poor manufacturing technique. 8. The Resin 942 was supplied in two supplies and used (as resin in the tie layer): a. exclusively in 62 pools … b. together with another resin in 15 pools … 9. Of the pools that have failed to date all were pools in which Resin 942 was used in whole or in part.' 54 It goes without saying that Huntsman does not agree that the evidence establishes all of those 'key facts'. Moreover, Narellan submitted that the evidence supports a finding that, contrary to 'key fact' 8, Resin 942 was used exclusively in 61 pools and together with another resin in 17 mixed resin pools (see [8] above). The variances in the total number (78/77) and the component numbers (61/62 and 17/15) have no substantive significance to the issue of liability. 55 Narellan, in opening, then referred to a number of documents which were said to be 'the critical documents in the case' and 'give some flesh' to the 'key facts' in [53] above. These included: 1. A two page document dated 8 February 2001 (Ex 11: 25, 26) containing: 1.1 Approval for the manufacture of Hetron 942 PAS by Dr Durrant, the organic chemist employed by Huntsman; 1.2 A number of the characteristics or properties of Hetron 942 PAS including its storage life, retest time and retest properties; 1.3 Instructions as to its formulation for a total yield of 4,770 kg, and process details. 2. It is clear from the Revision No on the first page of this document that no manufacturing instructions had previously been issued for this product. 3. A document dated 16 February 2001 (Ex 11: 34) headed 'TEST INSTRUCTION' for Hetron 942 PAS containing certain specifications, including those for cone and plate viscosity, Brookfield viscosity, screen test, gel time and exotherm time. 4. A document dated 16 February 2001 (Ex 11: 103) headed 'FINISHED PRODUCT SPECIFICATION' for Hetron 942 PAS containing similar specifications. 5. The production sheet for the Resin 942 (batch 053109001) (Ex 11: 105) bearing date 26/02/01, bearing reference MI No. 53584 (the same as on the documents in (1), (2) and (3) above), the '-0' after the MI No. signifying that this is not a revision and the notation 'Narellan' in handwriting in the right hand top corner. My attention was drawn to the following aspects of this sheet: 5.1 The reference to the batch number for the Hetron 914 - 07001 - corresponds with the batch number of a production sheet dated 12/02/01 for Hetron 914 (Ex 11: 101). 5.2 There was no batch number specified for the Hetron 922 or the Aerosil 202. 5.3 The amounts of Hetron 914 (3,000 kg) and Hetron 922 (1,260 kg) and the amounts of the other components used correspond with a yield of 4,770 kg of Hetron 942 as shown on page 2 of the document in (1), yet only 3,000 kg of Hetron 942 was ordered, but 6,830 kg was produced. 5.4 The sheet bears the notation: 'Xtra adds put in : GC'. 5.5 The sheet also indicates that the following 'batch adjustments' were made: · 300 litres of styrene were added over and above the 300 litres specification for a yield of 4,770 kg of Hetron 942. · Four drums of Hetron 914 and three drums of Hetron 922 were added over and above the 3,000 kg of Hetron 914 and 1,260 kg of Hetron 922 specifications for a yield of 4,770 kg of Hetron 942. 6. It was common ground between the experts, Professor George for the applicants and Professor Shanks for Huntsman, that despite these batch adjustments, there was another 400 - 450 kg of material, to yield 6,830 kg of Resin 942, which simply cannot be accounted for. 6.1 During the course of these batch adjustments, the gel time, the cone and plate viscosity and the Brookfield viscosity were out of specification at various times. 6.2 The sheet also bears the notation 'Ready' and 'Passed by Technology'. 6.3 The initials of three operators appear on the sheet - PK, GC and TS - but none were called to give evidence. 7. A two page document headed 'TEST RESULT MAINTENANCE' of the Resin 942 (batch 053109001; spec code 53584) (Ex 11: 115, 116). It showed a 'DATE TESTED' of 27 February 2001. There followed, numbered (2) to (8), the results of various tests of the product and, significantly for present purposes, a failure in respect of (6), gel time test - 40 as against a specification range of 30 - 35. 8. A document headed 'NON ROUTINE TESTING REQUEST' for the Resin 942 (batch 053109001) remaining at Huntsman's premises after the March Supply (Ex 11: 106). According to the document the test was done on 6 May 2001. It showed that the Resin 942 was out of specification for Brookfield viscosity - 33 as against 'FPS SPEC' of 20 - 30. 9. A similar document to (6) (Ex 11: 110) to the Resin 942 (batch 053109001) date tested 6 October 2001. It showed that the Resin 942 was out of specification for cone and plate viscosity - 5.4 as against 'FPS SPEC' of 3.5 - 4.0 - as well as for Brookfield viscosity - 40 as against 'FPS SPEC' of 20 - 30. 10. A facsimile dated 19 September 2001 from Mr Holden of Huntsman to Mr Turner of Narellan (Ex 11: 153) which reads: 'Dear Doug, You may recall that back in March this year, Narellan Pools baught [sic] 3 haz boxes of HETRON 942/35 (vinyl ester with 35% styrene monomer content) summer grade at $5.00/kg FIS ex GST. When we made this batch of vinyl ester we had left over 3.8MT or 4 haz boxes and the product has been retested and is still in spec. Is it possible for Narellan to take the balance of this stock (400kg of which, would be set aside at no charge for the pending Polgran trial)? Our current price for this vinyl ester is $4.85/kg FIS ex GST. Best Regards, (Signature) Mal Holden' 11. A facsimile dated 17 October 2001 from Mr Holden to Mr Turner (Ex 11: 155) which reads: 'Dear Doug, Confirming, that we have 4 haz boxes or 3830kg of the above vinyl ester in stock left over from the March trial and our price is $4.85/kg FIS ex GST. Could you please consider taking 3 haz boxes asap and the remaining haz box we'll hold over for the Polgran trial. On the subject of the Polgran trial, I understand that Rick Wilson asked if you would be able to have the boom mounted to save time when both Barry Hutchison and Rick install your new Polgran Unit. Best Regards, (Signature) Mal Holden' 12. The same facsimile as in (9) (Ex 11: 157) bearing the notation 'Resent 12/11 with attached certificate of analysis'. 13. The certificate of analysis referred to in the notation on the document in (10) (Ex 11, 158) bearing the date Nov. 9, 2001 showing the results of the testing of the Resin 942 (batch 053109001) carried out on 27 February 2001 (not the results of the two subsequent tests carried out on 6 May and 6 October 2001). The results set out in this certificate are the same as the results recorded in the document in (5) above save that the gel time specifications are now shown as 30 - 40 rather than 30 - 35 resulting in a pass for a result of 40 rather than a failure.

[26]

Mr Christopher Myer 56 Mr Christopher Meyer, the managing director of Narellan, gave evidence on its behalf. Four affidavits of Mr Meyer were read, the first sworn on 5 December 2006, the second on 11 August 2008 and the third on 23 October 2008. The first three were marked Ex 1. The fourth affidavit was sworn on 29 October 2008 and was marked Ex 4. Mr Meyer was cross-examined at some length over a wide range of matters the subject of his affidavits, but vigorously cross-examined on three particular subjects: (1) His evidence that seven Narellan production sheets - those for pools 8999, 9004, 9006, 9009, 9010, 9008 and 8975 - mistakenly identified the resin used in their tie layer as FCM/Derakane/SPV 6003 resin. (2) His evidence as to the sequence of production of a number of pools, particularly in relation to pools 9000, 9004 and 9005. (3) His evidence as to alterations that were made to Narellan's production sheets. 57 I was impressed by Mr Meyer as a witness both in the manner and content of his responses to questions put to him in cross-examination, some of which were long and because of that, tinged with ambiguity and uncertainty; and in his overall demeanour in dealing with questions which carried the 'barb' of an attack on his truthfulness and integrity. I have no hesitation in accepting him as a witness of truth according to his oath. 58 His evidence, that on the face of the production sheets for pools 9004, 9006, 9009, 9010, 9008 and 8974 there was a usage excess of FCM/Derakane/SPV 6003 resin and a usage shortfall of Resin 942, in paras (17) - (30) of his second affidavit sworn 11 Augusts 2008, was never put in issue and, in the absence of contrary evidence, I accept it as a sound basis for his conclusion that the six production sheets wrongly identified FCM/Derakane/SPV 6003 as the resin used in the tie layer of those pools rather than Resin 942; as I do that the production sheet for pool 8999 wrongly identified the tie layer resin as FCM/Derakane/SPV 6003 rather than as mixed resin. 59 I also accept as sound the basis upon which Mr Meyer, in his fourth affidavit of 29 October 2008, deduced the sequence of production of some ten pools, including 9000, 9004 and 9005. In order to arrive at the sequence, each production sheet was analysed for stated completion date, batching by colour, production book date, gelcoat application date and identity of each operator. Mr Meyer created a spread sheet (Ex 4: 30) bringing together these factors for all the pools involved to see more easily the information when collated together. He was cross-examined on this document and shown an alternative schedule of production sequence as an aide memoir. In effect, he was invited to agree that on the information available, contrary to his sequence, pool 8999 (before pool) actually came before pool 9007. However, the alternative schedule did not include a column for 'date completed' and, when all other factors were equal, this information was taken into account by Mr Meyer. In re-examination, Mr Meyer noted that pool 9007 left the factory on 19 January and 8999 left on 21 January, so Mr Meyer's original sequence was not undermined in any way by cross-examination. Moreover, it was confirmatory of the answer he gave to interrogatory 1A, filed 3 June 2008 (Ex 4: 9 - 19). 60 The other subject on which Mr Meyer was vigorously cross-examined was on the fact that some production sheets were altered or crossed out in places. Mr Meyer explained in cross-examination why production sheets were altered. Pools were re-allocated to different customers during production if circumstances changed for any reason. For sensible commercial reasons, details originally allocated on a production sheet would change so as to minimise delays in production and facilitate efficiency in the manufacturing process. 61 It was put to Mr Meyer that he had reconstructed the date on the production sheet for pool 9004 as the yellow copy was changed from 20 to 21 January 2002 and the white copy (which was copied in the applicants' bundle) was still dated 20 January. Mr Meyer did not resile from his explanation that his sequencing in the interrogatory was based on the yellow copies, being the carbon copy that remained in the production book (the white is removed and placed on the customer file). When the original production book was called for with the yellow copy for 9004, it is clear that the date on that sheet had been changed to 21 January (after removal of the white copy, which does not bear the alteration in pen). 62 Finally, on this subject, in cross-examination Mr Turner gave evidence that he wrote on a number of production sheets. Indeed, Mr Turner accepted that on occasion he had cause to make alterations to those production sheets. It was never put to Mr Turner that he fabricated entries on those production sheets to suit the convenience of Narellan's case. In short, I accept the evidence of Mr Meyer as to the reasons why alterations were made to the production sheets and why some alterations made to the yellow copy of the production sheets did not appear on the white copy torn from that book (compare Exs. B and 2). I reject out of hand any suggestion that Mr Meyer altered any copies to suit the convenience of Narellan's case.

[27]

Interlude 63 It is convenient at this stage to refer to a development which occurred on the morning of the fourth day of the hearing when Mr Meyer was being cross-examined on his fourth affidavit (Ex 4) sworn the day before. Senior counsel for Narellan informed the Court that a document had come into the hands of his instructing solicitors, through Professor Robert Shanks, the expert called on behalf of Huntsman, which indicated that Ashland had tested a cut out section of pool 9008 - one of the seven pools the production sheets for which, according to Mr Meyer's evidence, wrongly identified the resin used in the tie layer as FCM/Derakane/SPV 6003 and being one of three of such pools which had failed - and that the conclusion reached as a result of that testing was that the resin used in the tie layer of pool 9008 was Hetron 942. This document was not discovered nor, on this occasion, did senior counsel for Narellan suggest it should have been. However, he sought an adjournment until the following morning to enable his client to determine the course or courses it should pursue. The adjournment was not opposed by senior counsel for Huntsman who conceded that the newly discovered document suggests 'that where Mr Meyer has said a production record is wrong there is now some third party corroboration for it'. 64 When the hearing resumed the following morning, senior counsel for Narellan informed the Court that a number of other documents had been 'uncovered' which clearly demonstrated that Huntsman had not properly complied with its obligations of discovery. Whether that be right or not, the parties had agreed certain orders, compliance with which were designed to overcome perceived deficiencies in the discovery process to date, as well as to extend the timetable for the further hearing of the matter. On that occasion, I made the following orders: '1. Thesolicitor for [Huntsman] file and serve an affidavit by 4:00 pm on Monday, 3 November 2008 deposing to the matters identified by counsel for [Huntsman] at page 81 of the transcript of the proceedings on 28 November 2008. 2. The applicants have leave pursuant to Order 27A rule 2 to issue a subpoena to produce documents upon Professor Robert Shanks returnable at 9:00 am on 11 November 2008. 3. [Huntsman] have leave pursuant to Order 27A rule 2 to issue a subpoena to produce documents upon Professor Graeme George returnable at 9:00 am on 11 November 2008 4. The calls on the applicants' outstanding notices to produce dated 24 November 2008 and 30 November 2008 be adjourned to 9:00 am on 11 November 2008. 5. [Huntsman's] solicitors to notify the applicants' solicitors in writing by 7 November 2008 whether [Huntsman] has in its possession or control or is able to procure such documents from Ashland Inc. and/or Ashland Chemical Company specified in the schedule to this order. 6. [Huntsman] file and serve a verified list of documents giving discovery in accordance with Order 15 rule 2(3) by 7 November 2008 in relation to any further discoverable documents. 7. The proceedings be adjourned for further hearing with an estimate of three days to 10:15 am on 17 December 2008. 8. Liberty to restore on 2 days' notice. 9. Costs occasioned by the adjournment on 30 October 2008 and any costs thrown away by the further adjournment on 31 October 2008 and further discovery be reserved. SCHEDULE One copy of any document recording any communication between Ashland Inc. and/or Ashland Chemical Company, its servants and agents (Ashland) and [Huntsman], its servants and agents and/or Professor Robert Shanks, referring to: 1. Any resin supplied by [Huntsman] to [Narellan]. 2. Any test carried out by Ashland on the Resin 942 supplied by [Huntsman] to [Narellan]. 3. Any swimming pool manufactured by [Narellan]. 4. Any test carried out by Ashland on any swimming pool (or sample from such pool) manufactured by [Narellan]. 5. Discovery of documents between [Narellan] and [Huntsman] in these proceedings.' 65 As contemplated by the terms of the orders I made on 31 October 2008, the matter came back before me on 11 November, and again on 18 November 2008 to deal with issues arising out of compliance with those orders. 66 Professor Shanks produced various documents in response to the subpoena issued to him. Huntsman prepared a table of those documents so produced in respect of which it claimed legal professional privilege, in reliance upon which it denied Narellan access. 67 Huntsman produced various documents in response to Item 3 of Notice to Produce dated 22 October 2008, which required the production of - 'One copy of any document recording any communication between [Huntsman] its servants or agents and Ashland Inc and/or Ashland Chemical Company its servants or agents referring to: a. The Resin 942 and/or b. any resin supplied to [Narellan] by [Huntsman];' Huntsman also prepared a table of those documents so produced in respect of which it claimed legal professional privilege, in reliance upon which it denied Narellan access. 68 Huntsman also produced various documents in response to Item 1 of Notice to Produce dated 30 October 2008, which required the production of - '[O]ne copy of any document recording any communication between [Huntsman] its servants or agents (including its solicitors) and Professor Shanks and/or RMIT University referring to: a. the Resin 942; and/or b. any resin supplied to [Narellan]; and/or c. this proceeding and/or the evidence or proposed evidence of Professor Shanks; and/or d. communications between Professor Shanks and Ashland Inc. and/or Ashland Chemical Company or its servants or agents.' Huntsman also prepared a table of those documents so produced in respect of which it claimed legal professional privilege, in reliance upon which it denied Narellan access. 69 For its part, Narellan denied Huntsman's claims and contended further that if the relevant documents were, at the time they were brought into existence, privileged, that privilege had been waived. It sought access to all documents produced in response to the three mandatory processes. 70 On 20 November 2008 I heard argument on the access issue by reference to the following evidence: the tables referred to in [66] to [68] above (Ex 2A); the affidavit of Mr John Goulios sworn 18 November 2008 and annexures (Ex 1); the affidavit of Professor Shanks sworn 19 November 2008 and attachments (Ex 2); other documents tendered by Huntsman (Ex 3); the affidavit of Mr Noel Godfrey sworn 19 November 2008 (Ex 4); and various documents tendered by Narellan (Exs. A, B, C and D). Mr Goulios, Professor Shanks and Mr Godfrey, called on behalf of Narellan, were all cross-examined. I reserved my decision on the access issue. 71 On 15 December 2008, I granted Narellan access to the following documents: (1) The documents referred to as items 3 and 7 in the table of documents produced by Professor Shanks in respect of which privilege was claimed by Huntsman; (2) all the documents referred to in the table of documents produced by Huntsman, in response to para 3 of the Notice to Produce dated 22 October 2008, in respect of which privilege was claimed by Huntsman; and (3) the documents referred to as items 1, 3, 4, 5, 6, 11, 12, 16 and 21 in the table of documents produced by Huntsman, in response to para 1 of the Notice to Produce dated 30 October 2008, in respect of which privilege was claimed by Huntsman. 72 At the time, I indicated that I would publish my reasons for the orders I made at the time of pronouncing judgment in these proceedings; I did, however, proffer the following two observations: (1) First, that I was not satisfied that the evidence put forward and relied upon by Huntsman established that the documents in respect of which I granted Narellan access were privileged. (2) Second, that I was not satisfied that the evidence put forward and relied upon by Narellan established that Huntsman waived privilege in the communications contained in the other documents referred to in the tables, and Narellan did not put that privilege in issue. 73 On reflection, I have come to the view that these two observations say it all and I do not think there is any utility in detailing the underlying reasons for my lack of satisfaction, on either Huntsman's privilege claim, or on Narellan's waiver claim with respect to those documents to which Huntsman's privilege claim was not put in issue.

[30]

Dr Graham Durrant 77 Dr Graham Durrant, an organic chemist employed by Huntsman gave evidence on its behalf. Three affidavits of Dr Durrant were read, the first sworn on 14 July 2008, the second affirmed on 2 March 2009 and the third affirmed on 2 April 2009. They were collectively marked Ex F. Dr Durrant was cross-examined but his credit was not put in issue. 78 Dr Durrant was the person at Huntsman responsible for the development of Hetron 942 PAS for Narellan in late 2000 based on preliminary work he had done back in the middle of 1999 in developing a low styrene emission vinyl ester resin in connection with attempts to secure business from another swimming pool manufacturer, and based on Ashland's formulations for its low styrene emission vinyl ester resin called, Hetron 942/35. 79 In his affidavit of 14 July 2008 Mr Durrant made the following observations on what he called 'the Batch Adjustment Section of the Run Sheet (the bottom half)' (Ex 11:105) - see [55.7] above: '47 Ordinarily for a resin to be suitable for use in the manufacture of swimming pools, it requires a low Cone and Plate Viscosity, meaning that it is "thin" enough to run through a spray gun, and a high Brookfield Viscosity, meaning that it will be able sit without draining on a vertical mould. 48 The specification for the Cone and Plate Viscosity recorded on the Run Sheet is 3.5 to 4.0. The units of measurement of viscosity are Poise (P). The specification for the Brookfield Viscosity is 20 to 30. 49 There are seven columns of results recorded for both the Cone and Plate Viscosity and the Brookfield Viscosity. It is not until the last recorded results of 4.0 and 29 respectively that the Hetron942PAS was within specification for these properties as the first three Cone and Plate Viscosity Results and the first six Brookfield Viscosity results were too high. 50 The only way in which the Brookfield viscosity could have been lowered is if something with a lower Brookfield viscosity was added. I suspect that additional base resin, such as Hetron 914 or Hetron 922, was added and by oversight not recorded on the Run Sheet resulting in the unaccounted for 410 kg. Both Hetron 914 and Hetron 922 have a lower Brookfield viscosity than Hetron 942.' 80 And in cross-examination he gave the following evidence (T134): 'Yes, and you know Narellan contend that it had a manufacturing process which hadn't produced problems with osmotic blistering prior to the deliver of this material, save in one instance. You understand that - you will have to verbalise your answer, sorry?---Yes. And you understand that Narellan also contend that after the use of this material and in the period between the delivery of the two supplies, when they also used other resins, there was also no problem with osmotic blistering. You understand that?---That's what I've been told. And of course this product has never ever been used before or after, so there's no track record of it being used successfully, correct?--- No. And we have here, as you've conceded, a resin 942 which is uncertain in its composition as to 450 kilograms and there's some questions about the manufacturing process, correct?---Yes. Now, I know - and I'm not asking you to express any opinion about what is scientifically certain. I'm not asking you say definitely one way or the other what caused this problem. Do you understand?---Yes. What I'm asking is that when you draw, as a chemist, all these factors together, although you can't be certain, it's more likely than not that something in what was delivered as a result of this manufacturing process caused the problem. MR SIRTES: Well, I object to that, your Honour. Delivery is more likely than not - it's very difficult --- MR LEE: No. HIS HONOUR: Well, I'll allow the question. THE WITNESS: Based on those statistics it looks - probably, yes.' 81 Dr Durrant's evidence in cross-examination was that the variance between the amount that was proposed to be produced (4,770 kg), and what in fact was produced (6,830 kg)w was 'less than ideal', 'not good', '[substandard], if you like' and that even today, there was 'doubt' as to what precisely was in the 6,830 kg. His evidence was that it was important in and accepted by the industry that a customer of a corporation such as Huntsman would have regard to whether or not a product was 'in spec' in ascertaining whether or nor it purchased the product; that it was 'a flaw in [Huntsman's] system, which it [was] not in his power to correct', to quote a test-result which subsequent testing showed to be false. He conceded that this was 'not good'; and that to so quote when subsequent testing had shown the product to be 'out of spec' was 'misleading'.

[35]

The Conclave Reports 90 The first conclave report, that of 11 August 2008, relevantly reads as follows: '2. Points of Agreement: 2.1 Blistering in Narellan Pools 2.1.1 There is a high incidence of blistering in pools where Hetron 942 has been used in the tie layer. 2.1.2 Blisters occur in the tie layer close to (within ~100µm) but not including the gelcoat (shown by microscopic examination of the pool cutouts taken around the blister) 2.1.3 Blisters form where there are glass fibres present in the tie layer 2.1.4 Blisters in a tie layer may form when there are water soluble materials (either present initially or formed by the action of water) and/or poorly bonded fibres 2.1.5 There is no evidence for blisters deep in the tie layer or in the structural laminate of any pool sections examined 2.1.6 The blisters examined are on the order of 1 cm in diameter. 2.1.7 The nature of the blistering is similar for all pool samples examined (ie. not dependent on gelcoat composition). 2.1.8 There is evidence that there is a layer of vinyl ester resin and wet-out glass fibres against the gelcoat in a region of tie layer away from the blisters. 2.2 Composition of tie layer in pools fabricated with Hetron 942 2.2.1 The tie layer of a blistered Narellan pool analyzed by Ashlands and independently by Prof George, by infra-red micro-spectrophotometry, exhibited a peak which gave evidence for the presence of a terephthalate resin as well as a vinyl ester resin within the tie layer. 2.2.2 This has been confirmed by spectral comparison with a spectrum of the known terephthalate resin in the structural layer. 2.2.3 Spectral subtraction has shown that this terephthalate resin was present in the tie layer of blistered pool 8036, made with Hetron 942 resin, and not pool 8031 which was not made with Hetron 942 resin. 2.2.4 The origin of the terephthalate resin in the tie layer is unknown. 2.2.5 Measurement of the relative amount of terephthalate compared to vinyl ester resin shows that within error the amount appears constant across the tie layer. 2.3 Specification and composition of Hetron 942 resin delivered to Narellan Pools 2.3.1 The composition of Hetron 942 resin, batch 05310911, is uncertain because of an amount of 400 to 450 kg that cannot be accounted for in the production (run) sheet. 2.3.2 Examination of several run sheets has indicated to us that it was normal practice to record only the final test data in the last column. 2.3.3 This practice appears to have been followed in the case of batch 053109001 of 26-Feb-01 so that the amount of 400 to 450 kg extra resin and/or styrene cannot be accounted for as the necessary extra material to achieve specification. 2.3.4 Examination of the production sheet shows that the gel time was probably not measured after the final addition of three drums of Hetron 914, but it is not possible to determine whether this is significant. 2.3.5 The gel time specification is agreed to be 30 to 40 min. 2.3.6 The resin supplied in the second delivery was out of specification for viscosity (both cone and plate, and Brookfield). 2.3.7 There is no significant difference between the incidences of failure (37 %) for the second delivery compared with the first delivery (35 %) of Hetron 942 resin. 2.3.8 Extra additives were included in the production (possibly after the addition of one drum and then two drums of Hetron resins, based on the proportion of additives added) to allow for the accelerator and stabiliser components. 2.3.9 The production batch of 26-Feb-01 differed in composition from the trial batch of 17-Nov-00, because of the extra resin and additives. 2.3.10 The thixotrope level in the production batch of 26-Feb-01, following addition of extra resin, approached the value in the trial batch of 17-Nov-00. 2.2.11 The differences in styrene concentration between the trial and production batches are not believed to be a factor in the formation of blisters. 2.4 Fabrication procedure for Pools at Narellan Pools 2.4.1 From separate site visits conducted by both experts, the pool fabrication procedure was such as to avoid incorrect resin application and/or contamination by having a separate work station for each layer of the pool construction process. 2.4.2 Spraying of resin and glass fibres and consolidation of resin and glass fibres was in accordance with expectations. Resin thickness was intermittently checked with a depth gauge. 2.4.3 The production records show an apparent wide difference between time of fabrication for different pools that could be construed as an indication of excessive time between depositing successive layers. However there is no correlation between these figures and incidence of failure. 2.4.4 We have been advised that the time taken between depositing the gelcoat and tie layer is less than 24 hours, but this is not recorded on the production sheets. 3. Points of disagreement In the following section, the points of disagreement are presented as separate position statements which give each expert's opinion on the evidence and any moderation of the points of agreement in Section 2 which is felt to be appropriate. 3.1 Position of Professor George. The evidence is presented in more detail in my Reports of 12 July 2004, 28 March, 2007 and 14 April, 2008. The important points and interpretation of the evidence from these reports are summarized below. 3.1.1 The statistics of failure of pools fabricated by Narellan Pools with Huntsman Hetron 942 resin Batch 053109001 point to a fundamental problem with the resin as supplied in the two deliveries of March and November 2001. The failure rate of these pools was 34% compared to an industry accepted standard achieved by Narellan of <1 % when using resin other than this batch of Hetron 942. 3.1.2 Three units (pools or spas) were fabricated with a Trial batch of 197 kg of Hetron 942 (Batch 238-034) manufactured by Huntsman in November 2000 and none of these have failed. In contrast, 77 units were manufactured with tie layer resin containing (either totally or partially) Hetron 942 resin of Batch 053109001 of 6830 kg. Of these, 28 units or 36% have failed by blistering of the tie layer. 3.1.3 Examination of production and Quality Assurance (QA) records for this batch of resin (053109001) point to it differing significantly from the trial batch of resin (238-034) in composition and compliance with the Huntsman specification for this resin. 3.1.4 The production sheet for Batch 053109001 shows that what was originally intended to be a batch of 4770 kg had become a batch of 6830kg through addition of extra resin and styrene because the resin could not readily be brought into viscosity specification. This may be linked to the contaminant discussed in 3.1.5 below, but was exacerbated by the fact that the Manufacturing Instruction No 53584 for batch 053109001 had a level of thixotrope (Aerosil 202) viscosity modifier that was 36% higher than used in the trial batch. The attempts to bring the resin into specification resulted in the addition of extra resin so the ratio of Hetron 922 to Hetron 914 changed as did the content of stabilizers, accelerator etc. (In spite of the addition of further additives, these were still ≥20% below what had been used in the trial batch). 3.1.5 Further, there is an amount of 400 to 450 kg of unknown material that has been added as part of this process of adjustment or was an unknown contaminant at the beginning of the batch. There is strong evidence that the Hetron 942 resin in the tie layers of pools fabricated by Narellan contains a contaminant which is a terephthalate resin. This is manufactured by Huntsman as a laminating resin and it may be the unknown extra material in 2.2.2 above. The fact that it is seen in uniform concentration through the tie layer of all pools analyzed by IR micro-spectrophotometry suggests that it has not arisen due to interpenetration from the laminate during fabrication, but was originally present in Batch 053109001. In addition, the levels of contaminant in all pools examined appears to be similar. 3.1.6 Inspection of the factory of Narellan Pools has shown that it is not possible for laminating resin to contaminate the tie layer resin. The Hazcons delivering the resin are sealed and tapped into directly. The work stations and spray equipment for the tie layer are separated and not used for laminating. 3.1.7 The QA records of Huntsman show that the resin as supplied by Huntsman to Narellan in the second delivery in November 2001 was outside viscosity specification and no adjustment had been made by them prior to delivery. Further, Huntsman advised Narellan that it was in specification and provided a QA sheet from the original production and not the QA data that had been generated on the resin actually delivered. 3.1.8 By using the Huntsman QA data for viscosity as measured on the resin batch 053109001 at manufacture and during storage, it may be determined that it drifted quite quickly out of specification such that it is likely to have been at the upper limit of, or just outside viscosity specification at the time that Narellan fabricated the first pools with this batch in March 2001. 3.1.9 The sum of these composition and performance differences between the trial batch 238-034 and the production batch 053109001 would have been to provide a material with different microscopic wetout properties when used to make pools by Narellan in the usual way. Further the contaminant will have compromised the integrity of the vinyl ester resin and its inherent resistance to hydrolysis (since a terephthalate resin has inherently lower hydrolytic resistance than a vinyl ester resin). 3.1.10 For the above reasons, the Hetron 942 resin batch 053109001 when used as a tie layer by Narellan Pools did not have the necessary quality or performance characterisitics [sic] to provide pools with resistance to osmotic blistering. 3.2. Position of Professor Shanks My observations, testing and interpretation is presented in my reports of 6 October 2004 and 15 July 2008. Information from these reports is included in the points of agreement and some further points are raised in this section. 3.2.1 Hetron 942 was a special low styrene content vinyl ester resin based on Ashland Hetron 942/35. Cone-plate and Brookfield viscosity measurements, performed by Huntsman Chemicals, aim to predict the performance in a spray gun and upon subsequent consolidation. The behaviour of Hetron 942 in spray application, with glass fibres, may be different from the behaviour of other vinyl ester or unsaturated polyester resins. 3.2.2 No comparable low styrene content vinyl ester resin is known to me to be available from Huntsman Chemicals or their competitors. I am not aware of Narellan Pools, previously or at the time of using Hetron 942, using an alternative low styrene content resin. 3.2.3 The trial batch was submitted to the normal quality control tests. I am unaware of any more rigorous performance testing being carried out by Huntsman Chemicals or Narellan. The basis on which the trial batch was accepted and upon which Narellan decided to proceed with order of a production batch has not been revealed. 3.2.4 The trial batch was used in preparation of three pools. There was no reported blistering. On the basis of the overall incidence of blistering of 36 % this would suggest that one pool of the three prepared from the trial batch should blister. The sample size of three pools, however, is insufficient to attribute significance to the observations of the three pools. 3.2.5 The gel coat and tie layer of sampled failed pools were readily separated, mainly along the interface. This indicates poor fibre wet-out by the resin or incomplete consolidation of the resin-fibre glass mixture in the vicinity of the gel coat-tie layer interface. 3.2.6 Incomplete cure of the Hetron 942 in the tie layer, prior to spraying the structural layer, was not supported by inspection nor observed in tests such as Barcol Hardness. 3.2.7 The source of contamination by terephthalate is unknown. The normal procedures of Huntsman Chemicals and Narellan Pools would avoid such contamination. The resin batch prepared prior to Hetron 942 PAS was Aropol 6433PAS that consists of Hetron FR992H, a vinyl ester resin, and an ortho-phthalate resin (Annex 7). Terephthalate resin was not present in the prior batch. Further, it is unknown whether an impurity of terephthalate would lead to blistering.' (Emphasis added) 91 What is significant about the first conclave report is that what appears under the heading 'Points of disagreement', are not points on which Professors George and Shanks necessarily disagree at all. For example, Professor George's conclusion at 3.1.10 is not put in issue by Professor Shanks and nothing in his evidence in chief or in cross-examination suggests otherwise. Again, what Professors Shanks says at 3.2.7 concerning the respective normal procedures of Huntsman and Narellan avoiding contamination, is not a point of disagreement; on the contrary, so far as Narellan is concerned, it is totally consistent with the Point of Agreement at 2.4.1. I raised this observation with Professor Shanks when I asked him: 'You seem to be like two ships passing in the night?' To which he replied: 'Yes, I see that too, your Honour, and I suppose, in reaching the points of agreement, we made a list of things where we tried to divide the issues into as many specific items as possible, but in recording points of disagreement, we went away and did those independently, and we didn't, sort of, come back together and look at the disagreements, and maybe another iteration might have said what you are saying.' I then said: 'So the label, Points of Disagreement, is really not a correct description?' To which he replied: 'Yes, well …' 92 There is no doubt that Professor Shanks' explanation as to what he and Professor George did after they had identified their points of agreement, that is, go off and independently write their own comments without any reference to what the other was writing and lump them under the head 'Points of disagreement', has led to the predicament that I raised with him. 93 There was a concession by Professor Shanks that there may be some disagreement between him on the one side and Professor George on the other, but he never attempted to identify what that disagreement was even though I gave him the opportunity to do so. 94 I have to say that in assessing the respective comments of the two experts under the head 'Points of disagreement', and I hasten to add that in saying what I am about to say is not a criticism of Professor Shanks nor the evidence he gave, but the evidence of Professor George was directed or fashioned in a way which was more utile to the task of resolution of the issues in dispute between the parties. Neither in his points of disagreement in the first conclave report nor in his oral evidence, in chief and in cross-examination, did Professor Shanks challenge or take issue with Professor George's conclusion at 3.1.10, nor the process of reasoning by which he got there. 95 The second conclave report, that of 3 March 2009, relevantly reads as follows: '2. Points of Agreement: 2.1 Identification of materials used in batch 053109001 of Hetron 942 resin produced by Huntsman Chemical Company on 26 February 2001. 2.1.1 The Production Sheet (No 105 of Applicants' Bundle of Documents) for the batch 053109001 of Hetron 942 does not reflect the final composition of the resin as analyzed in failed Narellan pools. 2.1.2 Infrared (FT-IR) analysis of the tie layer of the retained samples of pools fabricated by Narellan from batch 053109001 of Hetron 942 shows that it contains a component that is not consistent with the materials listed in the Production Sheet. 2.1.3 This analysis has been performed by Ashland Chemicals as well as separately by Professors Shanks and George and Shows a prominent signature at 730 cm-1 that is not present in vinyl ester resins (eg Hetron 922) manufactured by Huntsman Chemical Company or other suppliers of vinyl ester resins (eg Dow Chemical Company, manufacturer of Derakane resins). 2.1.4 Infrared analysis of the tie layer of pools fabricated by Narellan using Derakane resins (see Appendix 1) does not show this signature and this may be used to differentiate pools that have been manufactured using Derakane resin from those using Hetron 942 resin of batch 053109001. 2.1.5 The identification of the contaminant material with the infrared signature band of 730 cm-1 is uncertain. If it is assumed to be a terephthalate (which has a band in that position) the amount of terephthalate present would exceed the known level of contaminant (which has been previously agreed to be no greater than 450 kg). The estimated level of contaminant, if it is assumed to be a terephthalate resin, is around 30% which would be over 2000 kg. 2.1.6 If the contaminant originated during manufacture by Huntsman then the Production Sheet would have to be in gross error to the extent of mis-recording the addition of one major component such as Hetron 922 resin (which has no batch number recorded in the Production Sheet) and replacing it with a terephthalate resin. 2.1.7 The presence of the band at 732 cm-1 in Hetron 942 resin has also been detected in a test laminate (284-083-4) prepared by Huntsman for analysis by the Analytical Services and Technology section of Ashland Inc and shown in the report of July 24, 2008, page 31. 2.1.8 A question considered by the experts is whether the contaminant may be unambiguously identified as a terephthalate resin. A band at 731 cm-1 is also present in the infrared spectrum of one of the major starting materials in Hetron 922, bisphenol A. No batch number for the Hetron 922 was recorded in the production sheet for batch 053109001 of Hetron 942, so the details of the production of this batch of Hetron 922 resin are not known. A further example is that the resin Hetron 992H (which was in the batch preceding the production of batch 053109001) has a band at 738 cm-1 and is spectrally very similar to the Hetron 942 resin. However, consideration of the possible levels of contamination from either of these routes renders them extremely unlikely. 2.1.9 A component of the resin in the production sheet that has no batch number recorded against it is the hydrophobic thixotrope, Aerosil 202. In spite of a draft report of Professor Shanks that referred to this thixotrope being replaced with a hydrophilic thixotrope Wacker N20, there is no evidence that this was the case. The change of thixotrope from the hydrophobic Aerosil 202 to the hydrophilic Wacker N20 by Huntsman Chemical Company did not occur until 2002, after the date of manufacture of this batch of resin (affidavit of Dr G Durrant of 2 March 2009). 2.2 Sources of contamination of Hetron 942 resin 2.2.1 The experts considered further the possibility that the contamination could arise by infiltration of terephthalate resin from the structural layer due to incomplete cure of the tie layer before commencement of lay-up of the structural layer. This is not considered likely for the following reasons: • The amount of terephthalate that is required to migrate would result in secondary effects, in particular swelling of the tie layer and rendering of the boundary between the tie layer and structural layer diffuse, which was not seen. • It would be evident to the operator that the vinyl ester had not fully reacted. For interdiffusion of the amount of resin found in the tie layer (around 30%), this would require gelling of the tie layer to be severely delayed such that the operator could not walk on the structure to apply the laminate. • There is evidence that the concentration across the tie layer is approximately uniform (Experts' Joint Report of 11 August 2008, point 2.2.5) which would not be expected if there was diffusion of the terephthalate resin into the uncured tie layer. 2.2.2 Narellan has procedures in place to ensure that contamination cannot occur. These are dedicated workstations and pump-lines from the resin Hazcon containers. It is understood that these procedures were in place at the time of manufacture of the pools in question. 2.2.3 It is understood that Huntsman has procedures in place that isolate the reactor from all materials other than those being used in the resin manufacture so reducing the risk of contamination. 2.3 Role of contamination in blister formation 2.3.1 There can be other substances from the resin synthesis and formulation that are present in the Hetron 942 that lead to an osmotic cell provided certain conditions are met. This is widely accepted as a normal occurrence in osmotic blistering. 2.3.2 The primary process for osmotic cell formation requires a poorly wetted region of glass fibres that allows firstly condensation of moisture and secondly hydrolysis of the poly(ethylene-vinyl acetate) (EVA) sizing on the poorly wetted glass fibres to form acetic acid. 2.3.3 The subsequent process of black spot formation may involve physical and/or chemical processes which lead to the rupture of the gelcoat and release of cobalt salts. 2.3.4 There has been no analysis of blister fluids to determine if the terephthalate resin (which is suggested from our study as being present in the tie layer) has hydrolysed and so contributed to osmotic cell formation. 3. Position of Professor George 3.1 Appendix I summarizes the results of the infrared analysis of the tie layers of pools fabricated by Narellan. The number of pools analyzed now totals 17 and the analysis shows that: • the Hetron 942 resin was not a pure vinyl ester resin and contained an impurity which, while not unambiguously identified, is consistent with a terephthalate resin (Points of agreement 2.l .1 to 2.1.8) . • several of the pools recorded in the Narellan production sheets as containing a Derakane resin from FGI were in fact fabricated using batch 053109001 of Hetron 942 (Point of agreement 2.1.4). • all pools analyzed as containing this impurity have shown blistering (Particulars of failure provided by Marsden's Law Group, 20 February, 2009). 3.2 It is noted that batch 053109001 as produced by Huntsman Chemical Company on 26 February 2001 specifically for Narellan Pools differed substantially from the test batch No 238¬034 manufactured on 17 November 2000 that had been trialled by Narellan prior to the first order for Hetron 942. 3.3 None of the three units units fabricated with the test batch 238-034 have failed but the number of units fabricated with batch 053109001 that have shown blistering (based on the above particulars provided by Marsden's Law Group prior to this further conclave) is 35 which is a failure rate of greater than 40%. This failure rate has increased from the previous Conclave due to the report of a further 6 pools that have blistered. 3.4 The subjective nature of reporting of pool failures is such that the first date of blistering can never be unambiguously determined and statistics of the number of failed pools reflect only the minimum number that have failed. The failure rate expected in the pool industry is less than 1%. 3.5 Examination of the cross-section of blistered pools has confirmed the point of blistering to be in the tie layer within about 0.1mm of the gel coat, ie the region containing Hetron 942 resin and fibreglass. Microscopic examination is required to determine this and reports of Ashland Analytical Services and Prof Shanks that have reported failure "at the interface between the gelcoat and tie layer" have been inaccurate because they did not examine the failure region at high magnification. 3.6 The report of Ashland Analytical Services of July 24, 2008 (made available by Huntsman on 22 December, 2008) contains useful analytical information but the "Comments on Blisters" (Section 11, p3) suggest a lack of experience with osmotic blistering. In particular the view is put that the resin degradation observed may be related to "penetration of swimming pool chemicals". In osmotic blistering there is no penetration of any external agents, other than migration and condensation of water, and all chemical action is generated from within the resin/glass composite. 3.6 Retained samples of the pools in question (skimmer box cutouts) have been made available for examination and analysis. Delamination of the tie layer from the gelcoat in an unblistered region requires considerable force (impact cleavage with a blade) and microscopic examination of the fracture surface reveals that it always contains fibres that have pulled out of the tie layer resin and sections of gelcoat adhering to the tie layer, both of which indicate strong bonding between the two layers. This suggests that, in the absence of water exposure, the structure is sound which is consistent with appropriate manufacturing processes and the impurity or impurities are not interfering with the initial adhesion between the gelcoat and tie layer and between the fibreglass and resin in the tie layer. 3.7 Blistered regions have undergone debonding between the resin and fiberglass due to the chemical action of the blister fluids and it is impossible to determine either the strength of the original laminate or the extent of bonding between the fibers and resin by examination and testing of a blistered region alone. 3.8 It is important to note that the tie layer is the "corrosion barrier" that is designed to prevent blistering of the pool. For this reason premium grade resins such as pure vinyl esters are employed and the layer is richer in resin than the structural laminate that follows. 3.9 The substitution of a part of the vinyl ester resin with a terephthalate resin would render the system less resistant to hydrolytic attack by the acetic acid present in blister fluids. From the poor performance of the pools fabricated with a Hetron 942 tie layer in which blistering has occurred rapidly (within the first year of use as a pool in some cases) then the tie layer has failed in its purpose. This is consistent with a gross defect in the resin system as formulated. 3.10 The analysis of the production record for batch 053109001 and the Quality Assurance data performed immediately before the two separate deliveries of this batch, reveal that this material was possibly out of specification from the outset. It certainly contained foreign material not in the formulation; it had a viscosity profile close to the upper limit and was very different in composition from the material first trialled by Narellan. 3.11 Conclusion: The further studies and analysis since the first conclave have assisted in determining the true nature of the resin that was used in the tie layer of pools manufactured over the period when Hetron 942 had been purchased from Huntsman. Pools manufactured with Hetron 942 batch 053109001 all have an infrared signature that is consistent with a terephthalate impurity level of around 30% and this has enabled the resin to be distinguished from FGI (Derakane) resins that were used by Narellan when Hetron 942 was not used (eg between the two resin deliveries and between production runs). Blistering at a rate of over 40% of pools produced has occurred only from pools fabricated with Hetron 942 batch 053109001. When pools were made in the same way over the same period but using other tie layer resins (eg. Derakane resins from FGI) there have been no reports of blistering. The Hetron 942 resin batch 0531109001 was not fit for the purpose intended ie. a corrosion-resistant tie layer for fiberglass swimming pools. 4. Position of Professor Shanks 4.1 Hetron 942 was a special low styrene content vinyl ester resin based on Ashland Hetron 942/35. The production batch of Hetron 942/35 was used by Narellan in the manufacture of 82 pools; some pools were manufactured with a mixture of resin types, Hetron and Derakane, as containers were exchanged. Pools manufactured with Hetron were found to exhibit an extraneous peak in the infrared spectrum at 732 wavenumbers. This was identified by Ashland as being characteristic of a terephthalate resin. The source of the terephthalate contaminant was proposed by Prof. George to be due to an amount of 400-450 kg of material that was not recorded on the production sheet. This would amount to 6.7 % of the Hetron. The size of the peak at 732 wavenumbers relative to other peaks in the Hetron spectrum and by comparison with a spectrum of a terephthalate resin of the structural layer of failed pool test pieces looked to be much greater than 6.7 %. I used infrared spectroscopy to measure and calculate the content of terephthalate by several methods and an average of about 30 % was found. 4.2 A level of 30 % terephthalate in Hetron 942 will change the properties of the Hetron by more than expected of a low styrene content resin. The styrene content will not be as low as expected since the terephthalate resin will have a normal styrene content of 48-50 %, thereby increasing the styrene content of Hetron 942 to greater than the intended 32 %. The mixture of resins is likely to have unexpected application and structural properties. The likely application and structural properties are speculative since I do not know of any resin produced with this composition. Both terephthalate and vinyl ester resins are soluble in styrene so a homogeneous solution is expected. Some likely consequences of a mixture of vinyl ester and terephthalate resins are: (a) Spray viscosity different from expected, (b) Different cure rates between vinyl ester and terephthalate components, (c) Uneven solidification (as the resin passes from a liquid to a solid), (d) Cure can still be complete since all components can cure with the catalyst system used. Incomplete cure of the tie layer is not supported by observation prior to spraying structural layer, nor by a test such as Barcol Hardness. 4.3 The consistent location of blisters, near the gel coat-tie layer interface, the relatively small size of blisters (≤1 cm), and the distribution of pool failures over the period from manufacture until now, is consistent with intermittent defects (meaning poor fibre wet-out by the resin and incomplete consolidation of the resin-glass fibre composite) in the resin structure in the region of blistering. Some limited tests on removal of the gel coat from laminate samples showed that the gel coat readily separated from the tie layer. Measurement of the strength could not be performed. 4.4 Blisters were not formed deeper in the tie layer, indicating that the barrier properties of the tie-layer (also often called a barrier layer) were not compromised. The position of blisters and distribution of failures in the years between 2002-2006 are consistent with the description proposed in 2.3.3, rather than terephthalate hydrolysis which, being a chemical reaction involving a particular concentration of terephthalate, would be expected to continue evenly with increasing hydrolysis over time. 4.5 The formation of a blister occurs when water migrates into the laminate and condenses in a void. Soluble substances typically found in a resin (catalyst from resin manufacture - amine salts, accelerators - dimethylaniline, cobalt salts, peroxide catalyst residues, solvents for the additives) dissolve in the water and create an osmotic pressure. Poly(ethylene-co-vinyl acetate) (EVA) size coating from the glass fibres is hydrolysed by the water to give water-soluble acetic acid (vinegar) that can be smelled in blister fluids (though such fluids were not able to be collected from the blisters examined). The osmotic pressure created by dissolved substances in the condensed water causes physical rupture of the laminate giving an external bulge on the pool surface. When stretching of the resin structure exceeds its limiting strain, a crack is formed to the surface to relieve the pressure. Dissolved cobalt in the blister fluid is emitted from the crack and upon meeting the less acidic environment of the pool water forms black cobalt oxide, giving a black spot at the location of the blister. These are the necessary and sufficient factors for causing blister formation. Resin hydrolysis occurs progressively over a longer time due to continued presence of water and acidic (acetic acid) conditions in the blister region. Since the resin is a polymer, two adjacent hydrolysis reactions are required to form a small molecule, capable of being dissolved, from the polymer chain. 4.6 Conclusion: Formation of blisters is the result of intermittent defects in the structure promoted by an impurity of terephthalate resin. Terephthalate resins are manufactured by Huntsman. Narellen [sic] uses terephthalate resins in the structural layers of its pools. The source of the terephthalate resin impurity is not known. APPENDIX 1: Composition of tie layers of pools based on infrared analysis. The presence of a band at 730 cm-1 is common to pools fabricated with Hetron 942 resin batch 053109001 and may be used to distinguish them from those containing Derakane resin supplied by FGI, which do not contain this band. Pool Number Composition of tie layer Pool blistered or black spot reported (Yes/No) from IR analysis 8018 Derakane No 8020 Derakane No 8034 Hetron 942 Yes 8036 Hetron 942 Yes 8037 Hetron 942 Yes 8074 Derakane No 8075 Derakane No 8281 Derakane No 8753 Hetron 942 Yes 8914 Derakane No 8978 Hetron 942 Yes 8980 Hetron 942# Yes 8999 Hetron 942# Yes 9002 Hetron 942 Yes 9008 Hetron 942# Yes 9009 Hetron 942# Yes 9031 Derakane No

[36]

#: These pools are listed in Narellan production records as being fabricated with Derakane resin from FGI, but the infrared signature indicates they have been fabricated with Hetron 942, batch 053109001.' (Emphasis added) 96 In the second conclave report, the respective positions of Professor George (3.1. to 3.11) and Professor Shanks (4.1 to 4.6) outside their points of agreement (2.1 to 2.3), unlike the first conclave report, are not branded 'Points of disagreement'. On the other hand, like the first conclave report, the respective positions do not disclose substantive points of disagreement. This may be a function of the way in which the experts prepared their respective positions outside their points of agreement as explained by Professor Shanks; see [91] above. Whatever the reason, the conclusion reached by Professor George at 3.11 is not put in issue by Professor Shanks in the report, nor is it put in issue by Professor Shanks' evidence in chief or in cross-examination. Professor Shanks concludes that the Resin 942 is contaminated by a terephthalate resin with an estimated level of contaminant of around 30% (over 2,000 kg) (2.1.5; 4.1; 4.6) but that the source of the terephthalate resin impurity is unknown (4.6). Professor George does not seem as sure about the identity of the contaminant. He says that the impurity cannot be 'unambiguously identified' (3.1) but that it is 'consistent with a terephthalate resin' (3.1). Professor Shanks says the same thing in his agreement with Professor George at 2.1.5, namely, that '[t]he identification of the contaminant material with the infra-red signature band of 730 cm-1 is uncertain'. What is significant, but went unexplained, is that Professor Shanks seems to have changed his mind on the connection (if any) between a terephthalate resin and blistering. At (3.2.7) of the first conclave report he said: 'Further, it is unknown whether an impurity of terephthalate would lead to blistering', whereas at [4.6] of the second conclave report, he said: 'Formation of blisters is the result of intermittent defects in the structure promoted by an impurity of terephthalate resin'.

[39]

Professor Shanks 101 I was not as impressed with Professor Shanks. Professor Shanks was present in Court during the whole of the hearing and I gained the impression that he was more in Huntsman's camp than Professor George was in Narellan's camp. I make that observation because, and in the face, of what can only be described as a tirade of allegations made against Professor George in Huntsman's written submissions, many of which were unfounded. They are certainly not sourced in his cross-examination. 102 At 2.1.5 of the second conclave report, the following appears as a point of agreement between Professor George and Professor Shanks: 'The identification of the contaminant material with the infrared signature band of 730 cm-1 is uncertain. If it is assumed to be a terephthalate (which has a band in that position) the amount of terephthalate present would exceed the known level of contaminant (which has been previously agreed to be no greater than 450 kg). The estimated level of contaminant, if it is assumed to be a terephthalate resin, is around 30% which would be over 2000 kg.' Professor Shanks again referred to the estimated (30%) level of contaminant at 4.1 and 4.2 of the second conclave report. 103 In the course of his cross-examination Professor Shanks was asked a number of questions going to this subject which are informative and not sourced elsewhere: 'Would you dispute that if … the resin 942 was mixed with a terephthalate resin to give the amount of terephthalate resin at the 30 per cent that you're talking about, that the amount of the resin in the tie layer of the pools would have increased by around 2927 kilos?---Yes, true, yes. And from your understanding of the way in which these things operate, this would have given enough additional resin to produce another 27 pools more than the pools that appear to have been produced using the resin 942, correct?---Yes, on the assumption that it's 30 per cent. I think that's an over-estimate, but the amount of terephthalate is somewhere in that range. Yes?---The infra-red spectroscopy in the quantitative sense isn't so accurate as to be able to produce exactly that number, but I agree, yes, it would make many more pools. And we can tell from the analysis that you've done following on from the Ashlands report in July 2008, that we can tell what pools were made using the resin 942 by scientific analysis, correct?---Yes. If you had 2927 kilograms of resin that was supposed to be used in the manufacture of swimming pools that was unused, do you understand how one would go about disposing such a quantity of almost three tonnes of resin, what is required in order to do that?---I agree that it would be difficult. I know what's required, but --- It's a hazardous chemical?---Yes. And there would be all sorts of obligations to dispose of it in particular ways, wouldn't there, if it wasn't used in swimming pool manufacture?---Yes.'

[41]

Narellan's Written Submissions 105 The applicants relied upon the combination of a number of what are described as 'indisputable facts' which, when taken together, compel a finding that the cause of the osmotic blistering was either the supply of contaminated Resin 942 or that the Resin 942 was outside Huntsman's own specifications when delivered and therefore failed in its represented use in the manufacture of fiberglass pools. They also relied on the opinions of Professor George and Dr Durrant, Huntsman's organic chemist responsible for the development of Hetron 942 PAS for Narellan in late 2000. 106 In summary, the 'indisputable facts' relied on are: (1) Outside the pools manufactured using Resin 942, Narellan has manufactured over 16,000 pools using non Huntsman resin without incident save for one pool which suffered from osmotic blistering. The fact that so many pools have been manufactured by Narellan (before and since the use of the Resin 942) without any problems is irrefragable evidence of competent manufacturing processes. It is also compelling evidence as to the likely cause of the problem when the Resin 942 was used. Indeed this factor, along with the statistical failure of pools manufactured with Resin 942, are probably the most important tell tales in the proceeding and were very important factors in the analysis of Professor George. (2) Out of a total of 61 pools made using Resin 942 (not including mixed resin pools) at least 27 have failed due to osmotic blistering; on any view, a very substantial proportion of pools made with the use of Resin 942 have failed. (3) Narellan's manufacturing processes were such 'as to avoid incorrect resin application and/or contamination'. Further spraying of resin and glass fibres was in 'accordance with expectations'. The evidence admitted without objection or limitation was that Narellan's manufacturing processes remained constant after 2001. These processes are addressed below in detail given the apparent suggestion (never put directly to Mr Meyer or Mr Turner or to the two manufacturing operatives called by Narellan who were not cross-examined) that Narellan manufacturing processes may have been at fault. (4) Huntsman has never supplied Hetron 942/35 epoxy vinyl ester resin to any pool manufacturer before or since and developed this resin especially for use by Narellan. It was therefore truly a 'one off' product. (5) Examination of the tie layer of a sample of pools manufactured using Resin 942 reveals the presence of a contaminant, comprising approximately 30% which is consistent with it being a terephthalate resin by virtue of its infrared signature. This was constant across the tie layer of each sample examined suggesting that it did not arise due to interpenetration from the laminate/structural layer during fabrication. (6) The composition of the Resin 942 is uncertain because 'of an amount of 400 to 450 kg that cannot be accounted for in the production run sheet'. This phenomenon (which has never been explained by Huntsman) was unprecedented in Huntsman's experience of manufacturing chemicals and was admitted to comprise substandard practice. (7) The composition of the Resin 942 is uncertain because of a question as to the origin of the Hetron 922 used, in that no batch number is stated and so its provenance is unknown (8) The composition of the Resin 942 is uncertain because of a question as to the origin of the thixotrope used, in that no batch number is stated and so its provenance is unknown. (9) The composition of the Resin 942 is uncertain because of a question as to the use of the tank beforehand to manufacture Aropol 6433 PAS which contained Hetron FR992H and also a hydrophilic thixotrope, Aerosil 200. (10) The Sample (which was used successfully) 'differed in composition from' the Resin 942. (11) As to being out of specification for the March Supply, it can be seen from the test carried out on 26 February 2001 that the two viscosity test results were virtually at the maximum allowable under Huntsman's own specifications. Professor George concluded that this would have drifted (viscosity increases linearly with time) outside the specified Brookfield viscosity range by the time it was used in March. (12) For the November Supply, there was no doubt whatever that it was outside both viscosity parameters. 107 Some of these facts were then addressed by way of further submissions in greater detail:

[43]

I. Could the contaminant have been added by Narellan at any stage after each delivery of the Resin 942? 112 On 31 March 2009 the case, from the perspective of Huntsman, changed in what was described as 'quite a dramatic way'. It was asserted by senior counsel for Huntsman that: (1) There was 'some agreement between these two experts that there is about two tonnes of this material in the [Resin 942]'; (2) 'It really means that on one view of it what was in the product was not Hetron but something entirely different. Rather than being constituted largely by a vinyl ester resin there was a large amount of polyester resin in there'; and (3) 'What is the great mystery now, your Honour, is where this two tonnes came from. Now, your Honour, if it didn't come from us we say it came from them. And if it came from them then that changes the question of liability in this case quite substantially as, in my respectful submission, is probably a selfevidence [sic] proposition'. (Emphasis added.) 113 Narellan submitted that the evidence demonstrates that the contaminant must have been present in the Resin 942 when it was delivered to Narellan in March 2001 and then again in November 2001. The whole of the evidence makes it plain that it is (and was) wholly fantastic to suggest that it was added by Narellan either by accident or intention for the following reasons: (1) The Resin 942 was delivered to Narellan in three 1,000 kg Hazcons in March 2001 and later again in three 1,000 kg Hazcons plus a further Hazcon containing 830 kg of resin in November 2001. Each Hazcon has an internal bladder containing the resin making it impossible to use to mix other material in the container. (2) Given the uniformity of the contaminant in the samples tested, Narellan would have had to pour the three Hazcons into a mixing tank (or mix each individually into an empty drum) and blend it with terephthalate resin (assuming that the contaminant was terephthalate resin). The blending would need the use of an industrial mixer to attain the precise level of viscosity required to enable thinness to spray but thickness enough not to run off the mould. Narellan would then have to repeat that process again in November 2001 with exactly the same precision for the second delivery. This process would have had to have occurred with absolute precision on at least 14 occasions. The suggestion that this may have occurred is risible and even Professor Shanks conceded that such a scenario was unlikely. (3) Moreover, if the terephthalate was contained in Aropol (or other terephthalate) resin and Narellan mixed this in (being cheaper than Hetron 942) to 'cut' the resin, then there would have been much more than 6,830 kg for terephthalate to comprise 30% of the total. Therefore, many more (possibly up to 27) pools would have been able to made with the 'mixed' Resin 942, but the evidence shows that Derakane pools were made between the two shipments and also immediately after the second shipment. If it was not used to make more pools (and there would be no reason to mix it otherwise) then the excess would have not only been wasted but have been difficult to dispose given that it was a hazardous chemical. The apparent suggestion that there was some sort of commercial motivation for dilution disappears. It cannot be seriously suggested that it made sense for Narellan (on the Huntsman theory advanced on 31 March 2009 and in the late served evidence) wasting a huge amount of resin and unnecessarily incurring substantial waste disposal costs. (4) The terephthalate signature was not present in any Derakane made pool so if Narellan was engaging in such a diluting practice there would be no reason not to do it for both types of resin, yet they clearly did not. (5) The manufacturing methods adopted by Narellan in 2001 were such that the wrong layers could not be applied. Indeed both experts agreed that the fabrication procedure was such as to avoid incorrect resin application and/or contamination and that spraying of resin and glass and consolidation was in accordance with expectations. (6) Fabrication methods included the following: (i) Each layer was applied at a different dedicated workstation and the pool mould was physically moved to that workstation when ready for the next layer. Only resin for that layer was permitted at that station and was cross-checked by the operators beforehand; (ii) post 2000, Narellan installed underground resin lines in its new factory leading to one Hazcon per one gun to reduce risk of contamination; (iii) use of written production sheets for completion by operators after each layer to show resin used and depth of each layer applied across the mould; (iv) in addition to the gun operator there are up to three roller operators who hand roll the composite mixture to mix the resin and glass thoroughly and release any air and visually inspect the mixture and depth of the layer; and (v) in accordance with the Australian Standard 1838:1994 boiler testing and barcol testing is conducted on samples and skimmer box cut outs are preserved.

[44]

II. Could the contaminant have been added by Huntsman during manufacture of the Resin 942? - Possibility of Hetron FR 992H added to the Resin 942 instead of Hetron 922 114 Given the high percentage of the contaminant with the terephthalate signature within the tie layer of the tested samples made with Resin 942, perhaps the most plausible explanation is the mistaken use of Hetron FR992H for what should have been Hetron 922 in the making of Resin 942. 115 Following service by Huntsman of the new evidence shortly before resumption of the trial, it became clear that the batch of Resin 942 was in fact made over three daysand not one day (which had previously appeared to be the case on the face of the production sheet at Ex 11, p105). This evidence was read by Huntsman and was unchallenged by Narellan. The only other oral evidence on this topic came from Dr Durrant who confirmed that the operator's evidence would be the best source of the timing but that the production run had finished by 27 February. This necessarily means the production process commenced as early as 24 February. Furthermore, each of the operators worked individually over a number of separate shifts rather than together as a team. This gives rise to the following facts and inferences: (1) The prior batch produced on 24 February in the same tank was a batch of Aropol 6433 PAS which contained 2,200 kg of Hetron FR992H. FR992H 'has a band at 738cm-1 and is spectrally very similar to the Hetron 942 Resin'. (2) Operator Peter Knight said he regarded 922 and 992H as the same thing; 'Hetron 922 is a base vinyl ester resin. 992H is also a vinyl ester resin'. (3) Materials were delivered from the warehouse only twice weekly for manufacturing purposes, possibly less regularly. (4) Great store was made by Huntsman in evidence about the different coloured labels making it impossible to mistake one chemical for something else. However, both 922 and 992H were in green labelled drums with only one number difference in their respective titles. (5) In the batch of Resin 942, Mr Knight put in all the ingredients exceptthe 922 which had been added by Gary Carlton in a previous shift. Mr Carlton did not give evidence in these proceedings, so this scenario could not be tested further with him. (6) If it was FR992H that was added, then as this was a fire retardant resin, it was clearly not suited to application within swimming pools and it would be likely to have unexpected effects. 116 Professor George considered that this was a possible explanation and confirmed his view that it was not possible to say that the signature of 730cm-1 was without doubt, proof of the presence of terephthalate and that the precise nature of the contaminant was still an open question. 117 This issue was in fact the onlyarea of real disagreement between the two experts by the end of their evidence. The position of Professor George was that it was an open question as to the precise identity of the contaminant and it was possible for 992H to give the requisite signature, even though not precisely the same wavelength. Professor Shanks was of the view that it could not be 992H given the different wavelengths. The view of Professor George should be preferred for the following reasons: (1) Professor Shanks agreed at the second conclave that it was 'spectrally similar,' as critically did Dr Durrant. (2) Professor Shanks then said in further evidence in chief at T175.33 that in his view the wave numbers would need to be exactlythe same in order to conclude that the impurity was 992H. In cross examination, he then conceded that he was aware of a variation in wavelength of up to two numbers. (3) Professor Shanks was then shown the Ashland report of 24 July 2008 and accepted that this test provided a result of 734cm-1 for a Resin 942 sample from pool 8978 which was four wave numbers away from the 730 signature. Professor Shanks could simply not explain this variation. (4) Professor Shanks was quick to conclude the figure must be exactly730cm-1 and then just as quick to concede under cross examination minor variations. Where there is this residual conflict between the evidence of Professor George and Professor Shanks, the evidence of Professor George ought be preferred. This is not to say that Professor Shanks was a witness that should be disbelieved in all respects. For reasons set out below, Professor Shanks was in an unenviable position; he was not truly 'independent' because of his prior dealings with Huntsman. Objection was considered but not taken to the whole of his evidence because of two factors: first, the full extent of the relationship had not been revealed when objections were first to be notified and secondly, at least substantially, Professor Shanks conscientiously did engage in the process of narrowing disputes in the first conclave (and subsequently in the second conclave). Having noted this, the following should be borne in mind in considering the relatively minor conflict between the evidence of Professor George and Professor Shanks: (i) Professor George's evidence is consistent with the evidence of Dr Durrant the Chemist specifically authorised by Huntsman to speak on chemical matters; (ii) Professor Shanks was a less than happy choice as an 'independent' expert given his close connection with Huntsman, (reflected in the dealings with Huntsman on this very issue prior to his engagement as an expert); (iii) the nature of Professor Shanks' close pre-litigation contacts with Huntsman including his access to material that was being kept secret from Narellan including as to informers and other pool failures; (iv) the fact that Professor Shanks had a very long relation with Huntsman commencing with his employment in 1972; (v) the demeanour of Professor Shanks in the witness box where his evidence as to his consideration of the thixotrope issue was unsatisfactory, changed and became more definitive as the significance of the issue became apparent; and (vi) Professor Shanks' inexplicable conduct in holding on to and failing to disclose to counsel for Huntsman critically relevant (and discoverable) documents during the cross-examination of Mr Meyer (including documents which allowed him to identify with precision whether pools had been made by Resin 942).

[48]

II Possibility of a hydrophilic thixotrope added to the Resin 942 instead of Aerosil 202 121 What is plain is that the use of a hydrophilic thixotrope in resins intended for use in the manufacturing of swimming pools is inappropriate. This was an issue in Tranquility Pools and Spas Pty Ltd v Huntsman Chemical Company Australia Pty Ltd [2008] NSWSC 58 where the court adopted the finding of the referee that a hydrophilic thixotrope present in Hetron 922 was responsible for the failure of pools manufactured with that resin. Dr Durrant agreed that there was a problem with such use of hydrophilic thixotrope. Professor Shanks agreed that if a hydrophilic thixotrope had been known to have been used in the Resin 942 then this would be an important consideration in these proceedings. Indeed in his draft report of 16 August 2004 Professor Shanks stated that Aerosil 200 (hydrophilic) 'will attract water more strongly and the clusters will provide a path for water to migrate through the resin'. Tellingly, Professor Shanks was asked by Huntsman to remove this reference from his final report. 122 In his affidavit of 2 March 2009, Dr Durrant said at [6] that prior to 2002, Huntsman only used hydrophobic fumed silicas (thixotropes) with vinyl ester resins. Although this statement is literally correct it is of no relevance whatever when closely examined. What it does not reveal is that Huntsman used hydrophilic thixotropes in otherproducts in February 2001 (including in this case on the same day in the same tank in the previous batch). 123 The production sheet for the prior batch reveals that Wacker N20 was originally specified and replaced by Aerosil 200 (both hydrophilic thixotropes). Further, the evidence of the operator Mr Knight was that the person adding the thixotrope did not load this material into the hopper, it was added by a person on the previous shift. The batch of Resin 942 does not have a batch number attached for the thixotrope and there is evidence of the use of a hydrophilic thixotrope in the previous batch (which on the evidence is the same day as the batch of Resin 942 commenced). 124 When these facts are considered - alongside the evidence of discussions on the issue between Huntsman and Professor Shanks - it is tolerably clear that this view was the initial theory of Professor Shanks after he was first engaged. At a meeting on 7 June 2004, the Narellan problem (as well as Tranquility and other issues) was discussed between Professor Shanks and representatives from Huntsman. In his handwritten note, Professor Shanks noted: 'Thix. N20 not A202 (hydrophobic), Wacker, hydrophilic'. Clearly discussion had taken place at that meeting concerning which thixotrope was used and hydrophilic thixotropes were discussed. On two separate occasions Professor Shanks agreed that this topic was discussed, but could not recall the conversation. This recollection 'evolved' as the significance of the point became evident. 125 On 16 August 2004, Professor Shanks prepared a draft report. Unlike the meeting where Tranquility was mentioned, this report was exclusivelyabout Narellan and the samples analysed were all Narellan pool samples. At page 8 of that report, Professor Shanks proceeded on an assumption that a hydrophilic thixotrope had been used and described in that report how it would attract water more strongly than a hydrophobic thixotrope. 126 That report was never discovered and was only obtained after argument on waiver of privilege. A further draft was then sent by Professor Shanks which has never been discovered or even revealed as a privileged document. The later (third) draft report of 24 September 2004 and the final version of 6 October 2004 contain no reference to thixotropes. Professor Shanks confirmed that he was expressly requested by Huntsman to remove this reference from his 'final' report - the only version he knew was being provided to Narellan. 127 Given that a hydrophilic thixotrope was used in the previous batch in the same tank on the same day as the Resin 942 commenced production, there was real scope for such a thixotrope to have been mistakenly added to the Resin 942 and that this was something clearly considered by Huntsman and Professor Shanks. Although it is not necessary in order to conclude the case on liability, there is sufficient evidence to also conclude on the balance of probabilities that a hydrophilic thixotrope was added to the Resin 942 which was responsible (as it was in the Tranquility matter) for osmotic blistering.

[50]

Huntsman's Written Submissions 131 Huntsman submitted that Narellan must prove that the representations/contractual terms that Hetron 942 was suitable for the manufacture of fibreglass pools were untrue/false. To do so, Narellan must establish on the balance of probabilities that the Resin 942 caused the pool failures. 132 Narellan's evidence fails to establish Resin 942 as the cause of the blistering. The Court must accept that the Resin 942 contained a contaminate (terephthalate) but is prevented from forming a conclusion on the evidence, and ought not conclude, that the source of the contaminant was Huntsman. 133 Only about 40% of the pools have failed. This fact cannot be reconciled with Narellan's case that the Resin 942 was inherently unsuitable for the construction of fibreglass pools. Sixty percent of the pools constructed with Resin 942 have been blister free since 2001/early 2002. 134 The main causation theory championed by Narellan is the absence of pool failures in Derakane pools, which is termed 'the Correlation Theory'. Huntsman acknowledges the superficial sheen of the Correlation Theory. At first blush the argument that pool blistering only ever occurred when using Resin 942 has a simplistic attraction. But the theory says nothing at all about how the blistering occurred or addresses why it has not occurred in the majority of cases. The Correlation Theory beguiles with its simplicity but in truth invites the Court to ignore the necessary complexity of establishing an actual causative link between Resin 942 and blister formation. (Emphasis added) 135 The only available evidence that the Court can consider in determining the question of liability is the expert evidence. Huntsman submitted that the Court ought dismiss the applicants' claims on the basis that they have failed to establish on the balance of probabilities that the Resin 942 was in any way defective or deficient. The Court ought find that the applicants have failed to exclude the manufacturing process adopted by Narellan when using this new and previously unused product in the fabrication of fibreglass pools. 136 In broad terms, Huntsman invited the Court to reject the applicants' claim for the following reasons: (1) Narellan must prove that the pool failures occurred because the Resin 942 was defective. No presumption is available that the Resin 942 was defective merely because of the existence of pool failures. Narellan must establish a mechanism of failure and this requires Narellan to exclude other possibilities such as the most obvious - production deficiencies which affected those pools that were not properly fabricated using Resin 942. (2) The fact that 60% of the pools have not suffered any failures strongly points to a manufacturing problem, not a constitutional deficiency with the Resin 942. (3) The predominant reason advanced by Professor George, being the existence of a contaminant (terephthalate) in the resin, does not explain the presence of the contaminant. If, as Huntsman submits, there is no evidence which allows the Court to find (at least to a 51% level of satisfaction) that any contaminant was introduced by Huntsman, then Huntsman cannot be blamed for any contaminant-related failure. (4) Professor George's theories are based upon a factual assumption that no Derakane pools have suffered blistering. The foundation for this view is tenuous and not the subject of any scientific verification. Indeed, what analysis does exist suggests that at least one failure did occur to a Derakane pool (Pool 9008). Moreover, Narellan has no plausible explanation as to how pool fabricators made fundamental errors in the Productions Sheets for (e.g.) Pools 9008 and 9009. (5) Narellan's production records are in such a state of disarray that the Court cannot accept the accuracy of those records. Narellan also manufactured the pools in such a way to give rise to cross-contamination as noted in an email (discussed below) sent by Professor George to Marsdens. This issue is relevant to the Derakane pools issue because a number of pools ostensibly manufactured with Derakane have failed - 'ostensibly' because a number of Production Sheets created by Narellan operators in 2001 and early 2002 recorded the use of Derakane forming the tie layer not Hetron (where Narellan now asserts Resin 942 was used). Narellan has conduced an ex post facto revision of those documents to eschew any suggestion that any pool containing Derakane has failed. Even a single Derakane pool failure would implode Narellan's case theory hence the great efforts that have been undertaken to avoid any such fact finding. The Court is, however, invited to find that there has been Derakane pool failures which precludes the Court accepting the Correlation Theory (that ought not in any event be accepted). (6) The Court ought find that no terephthalate was mixed into Resin 942 by Huntsman, thereby exonerating Huntsman from any causative rote in the blistering that has occurred insofar as the experts seemingly accept that such contamination may have played a role in the blistering process. (7) The arguments raised by Professor George that the Resin 942 was 'out of spec.' (specification) by the date of delivery of the second batch on November 2001 is a furphy because on Narellan's case the failures are equally distributed between the pools made in March 2001 and those made when the Resin 942 was 'out of spec.' in November 2001 so the argument offers no explanation for the March blistering and detracts from any argument that the November 2001 batch of Resin 942 caused pool blistering because it was not within specification. (8) The Court ought reject the implication of any term that the Resin 942 had a shelf-life for any lengthy period. Mr Turner, Narellan's Production Manager/Director, did not read the Certificate of Analysis he received in November 2001. Mr Turner made unwarranted, unjustified assumptions that the Resin 942 would be sound for six months. Even if the product was in specification for six months, it was stored outdoors over December and January 2001-2002 and then re-applied without any testing being undertaken. Huntsman gave no guarantee that the product would work if stored for months outdoors in the summer heat and then re-applied without any testing. (9) Huntsman cannot be liable for any pool failures to 'before' and 'after' pools (none of which have suffered damage to date). 137 Huntsman then embarked, by way of submission, on a variety of issues, but in relation to Professor George, in a fashion which can only be described as a tirade of allegations, in many cases unfounded and certainly not sourced in his cross-examination: (1) That the Court should not make any finding that the failure of any mixed resin pools was due to Resin 942; (2) that the Court should find that Narellan had failed to demonstrate, in accordance with the balance of probabilities, that only Resin 942 pools failed; (3) that the Court should find that pools 9008 and 9009 were made with Derakane; (4) a close analysis of Professor George's reports display the following: (i) he has advanced a large number of theories but fails to articulate any one of them as being more or less likely to have caused the blistering; (ii) the numerous reports take a 'mud-slinging approach' to Huntsman - if enough mud is thrown at least some of it may stick; (iii) despite agreeing during cross-examination that he could not attribute any causative effect to Correlation Theory (although not so described during cross-examination) Professor George is in truth a key apologist for the said theory. (5) despite eschewing the Correlation Theory during his cross-examination, Huntsman submitted that this simplistic and unscientific theory permeates throughout Professor George's theories and conclusions the essence of which is: there could be many factors that lead to blistering but at day's end there was no blistering in Derakane pools therefore the blistering must be caused by the Resin 942. It was also submitted that Professor George's approach was not that of a wholly impartial independent expert; (6) in preparing his report, Professor George: (i) devoted no analysis to the details of the fabrication process; (ii) made assumptions that the fabrication of pools with Derakane would be the same as the fabrication of Resin 942 pools, when they were different products that Narellan had no experience in applying; (iii) did not undertake any analysis (so far as his reports are concerned) about the ambient temperatures at which the resin was applied; (iv) did not seemingly consider the effect, if any, of gelcoat thickness on the presence or absence of blistering; (v) had no knowledge that the spray guns operated properly as to ensure it did not overly aerate the low styrene Resin 942; (vi) had no knowledge that temperatures in which the laminating occurred was maintained to produce an environment of more consistent heat and humidity temperatures (even though he accepted that he would have had to be satisfied that this was so in order to exclude fabrication as a cause of the pool failures); (vii) he had no knowledge of whether the apposite glass to resin ration was being adopted; (viii) he did not see any quality assurance procedure being in existence in 2001 (although he did treat the Barcol hardness testing and the skimmer box cut-outs as some evidence of quality control); (7) the failure to seriously consider fabrication issues and gloss over them without acknowledging their potential role in the creation of blisters is a major shortcoming in Professor George's analysis and even more so because fabrication-related problems are far more readily reconcilable with the fact that the majority of pools have not failed; (8) the Court must reconcile the competing expert opinions expressed by Professors George and Shanks. Clearly, these were gentlemen with very different demeanours. Professor Shanks was diffident and far more ill-at-ease with giving evidence than Professor George who was considerably more loquacious and confident with the rigours of giving evidence. In discounting form in favour of substance the Court is invited to prefer the opinions of Professor Shanks for two main reasons: (i) Professor Shanks' opinions held a consistency during the entire proceedings. He did not chop and change his opinions or approach the task of giving evidence with any particular agenda. The same, with respect, could not be confidently said about Professor George; (ii) at the hearing, Professor George inexplicably and speciously attempted to distance himself from the rigid view that he had formed, very early on, that the resin was contaminated with terephthalate. His pirouette on this issue ought properly undermine the Court's confidence in the reliability of his opinions and his impartiality as an expert; (9) the grab bag of minor theories advanced by Narellan to link the Resin 942 to the pool failures: (i) ought not be accepted as establishing any such causal link on the requisite evidential standard; (ii) do not establish or make good any of the three pleaded representations or the breach of contract claim; (iii) merely serve to demonstrate that Narellan have no predominant explanation for the pool failures but by firing a hale of bullets hope that one of more of the theories may hit their target.

[51]

Terephthalate Contamination 138 Huntsman submitted that if the dominant scientific explanation of the pool failures is a gross contamination by terephthalate (at 30%), as the latest expert reports suggest, the evidence does not allow the Court to form any view as to the source of the contaminant. Each of the parties' evidence provides exculpatory explanations, no more plausible than the other, denying themselves as the source of the contamination. Industrial sabotage is just as possible as any other available explanation. 139 Narellan advanced reasons why the terephthalate could not have been added by it, thereby implicating Huntsman. Huntsman's evidence emphatically establishes that the contaminant could not have been added during the manufacturing stage, for a number of reasons: (1) As Narellan was at pains to demonstrate at the outset of the hearing, 450 kg of material is not accounted for on the Production Sheet of Batch 053109001. The balance of the 6,830 kg are wholly accounted for and Narellan has no basis to suggest otherwise. There is no avenue for terephthalate to constitute up to 30% of the mixture (i.e. over 2,000 kg); (2) the manufacturing process at Huntsman was careful and regulated albeit not perfect. Even though 450 kg of material has not been identified, this is not of great moment because the copious testing of the pool samples has never revealed that, other than the terephthalate, some other unaccounted for additive (not otherwise reflected in the Production Sheet) was mixed into the Hetron. Moreover, for the reasons advanced by the Huntsman technicians, there would be no ability to have mixed terephthalate into the mixture at the manufacturing stage; (3) the Production Sheet makes plain that no terephthalate was entered into the mixture in the quantities identified by the experts; (4) if around 2,000 kg of terephthalate was added to the mixture there would have been in excess of 8,000 kg of Hetron 942. There was not. Huntsman delivered only 6,830 kg of Resin 942. 140 Huntsman submitted that if the Court is unable to find that Huntsman created the contaminated product Narellan is unable to establish that any of the pool failures have been caused by Huntsman. The product used by Narellan was not Hetron 942 - it was an unauthorised hybrid resin that was part vinyl ester, part polyester resin, with its own properties (gel time, etc.) and characteristics. Huntsman made no representations about this formulation in the same way that Huntsman made no representations about any of the 'one before-one after' hybrid Derakane/Hetron pools. 141 Huntsman further submitted that Narellan's exculpatory explanations denying how the terephthalate contaminated the Hetron 942 is largely founded on the contention that in 2001 Narellan's manufacturing processes prevented cross-contamination taking place. Unlike Huntsman's premises, where raw materials were not stored anywhere near the manufacturing precinct, Narellan stored and used terephthalate in close proximity to its vinyl ester resin. The potential for such contamination was far more physically proximate in Narellan's premises than at Huntsman's. The calibre of evidence that suggests that in 2001 Narellan's factory premises were configured to make such cross-contamination impossible or unlikely is poor. 142 In conclusion on this point, Huntsman submitted that if, as the expert's agree, the terephthalate contamination may have been partially responsible for the blister formation and the Court cannot (and ought not) attribute any finding to Huntsman as the source of the contamination, it is not clear how Huntsman can be held liable for any damage caused by the use of a product it made no representations about or contracted to sell (i.e. the hybrid contaminated product).

[56]

What is the most satisfactory explanation of causation? 151 Huntsman submitted that if the pool failures are not the result of a gross (30%) contamination by terephthalate (as the experts now suggest), then the unshaken expert evidence of Professor Shanks is that the most likely explanation of the pool failures is that Narellan did not adjust its manufacturing method to accommodate a low styrene resin. 152 It is agreed that Narellan's manufacturing processes did not change and this may have been the precise cause of the problems. 153 It was a product that Professor Shanks believes may have had different rheological (i.e. physical) properties to Derakane, and may have needed to be treated differently. Professor Shanks says that Narellan had to recalibrate its manufacturing method to accommodate such a resin. 154 The theories advanced by Narellan are largely inconsistent with a defective product and more consistent with a fabrication and quality control problem associated with Narellan using a new product for the first time having never used a low styrene vinyl ester resin product. 155 The Court was invited to consider the lack of experience that Narellan had in working with the Hetron 942 product. The evidence on this topic was as follows: (1) Mr Meyer accepted, during cross-examination, that the spray gun operators were liable to error. He conceded that an operator put too much vinyl ester resin into one of the Seaspray 10 pools; (2) Narellan had a very lengthy and apparently close working relationship with FGI (who supplied Derakane). The form of the Production Sheets was prepared by Narellan in conjunction with FGI; (3) FGI had been supplying Narellan with vinyl ester resin for a number of years; (4) when Narellan ordered the low styrene Hetron 942 it had never before used a low styrene vinyl ester resin; (5) other than in March and November of 2001, Narellan otherwise used the FGI product; (6) when Jamie started spraying the vinyl ester resin in March 2001, that was the first experience he had in applying that product. 156 The Court will recall that in one of the Production Sheets it described the vinyl ester resin as 'Hetron Shit'. Of course, no definitive meaning can be attributed to this derogatory description but the spray gun operator was plainly not satisfied with the physical properties of Resin 942. It was only purchased, according to Mr Turner, as a back up for the principal use of Derakane. It was a product that no one at Narellan had any experience in working with.

[58]

Whether only Resin 942 Pools have failed? 158 My acceptance of Mr Myer's evidence as providing a sound basis (see [58] above) for his conclusions that the production sheets for pools 9004, 9006, 9009, 9010, 9008 and 8975 wrongly identify FCM/Derakane/SPV 6003 as the resin used in the tie layer of those pools rather than Huntsman's Resin 942; and that the production sheet for pool 8999 wrongly identifies the tie layer resin as FCM/Derakane/SPV 6003 rather than as mixed resin; as well as the consensus of the experts (Ex 8; second conclave report, pp 1, 2 and App 1, p 8; see [95] above) that contrary to Narellan's production records, pools 8980, 8999, 9008 and 9009 were manufactured with 'Hetron 942 (batch 053109001)', in other words, Resin 942 and not Derakane resin from FGI, leads me to find that pools 9004, 9006, 9009, 9010, 9008 and 8975 were manufactured with a tie layer of Resin 942 contrary to Narellan's production records; and that pools 8980 and 8999 were not manufactured with a tie layer of Derakane/SPV 6003 resin but, contrary to Narellan's production records, were mixed resin pools. 159 It is not in dispute, and I find, that of the eight pools referred to in [158] above, four (pools 8980, 8999, 9008 and 9009) have failed in the sense of manifesting osmotic blistering. 160 As noted in [8] above, Narellan has identified 61 pools as having been made exclusively with Resin 942. Apart from pools 9004, 9006, 9009, 9010, 9008 and 8975 this was not disputed by Huntsman and in the face of my first finding in [158] above, I find that all 61 pools identified by Narellan as having been made exclusively with Resin 942 were so made. 161 Again, as noted in [8] above, Narellan has identified 17 pools as being mixed resin pools. Apart from pools 8980 and 8999, this was not disputed by Huntsman and in the face of my second finding in [158] above, I find that all 17 pools identified by Narellan as mixed resin pools were mixed resin pools. 162 Again, as noted in [8] above, Narellan has identified 33 of the 61 pools referred to in [160] above as having failed due to the incidence of osmotic blistering. While Huntsman did not dispute the failure of 26 of these pools, it maintains that evidence regarding failure of seven pools (8033, 8037, 8048, 8052, 8055, 8985 and 9016) has not been adduced, although, in the case of pool 8037, the consensus of the experts would seem to be that it has failed. Accordingly, for the purposes only of the limited issue of liability, I find that at least 27 of the 61 pools referred to in [160] above have failed due to the incidence of osmotic blistering. 163 Again, as noted in [8] above, Narellan has identified two of the 17 pools referred to in [161] above as having failed due to the incidence of osmotic blistering. This was not disputed by Huntsman and accordingly I find that at least two of the 17 pools referred to in [161] above have failed due to the incidence of osmotic blistering. 164 The consensus of the experts (Ex 8; second conclave report at pp 1, 2; see [95] above) that infra-red (FT - IR) analysis of the tie layer of the retained samples of pools fabricated by Narellan from batch 053109001 of Hetron 942, performed by Ashland Chemicals as well as separately by Professors Shanks and George, shows a prominent signature at 730cm-1. The experts concur that infrared analysis of the tie layer of pools fabricated by Narellan using Derakane resins does not show the signature and this may be used to differentiate pools that have been manufactured using Derakane resin from those using Hetron 942 resin of batch 053109001 or, in other words, Resin 942. By recourse to this analysis, the experts concurred on the results set out in Appendix 1 on p 8 of the second conclave report and, in consequence, I find that none of the pools there listed which had a tie layer composed, from FT - IR analysis, of Derakane failed due to the incidence of osmotic blistering whereas all of the pools there listed which had a tie layer compound, from FT - IR analysis, of Resin 942 failed due to the incidence of osmotic blistering. 165 In the face of the findings in [158] and [164] above, I find that no pool with a tie layer manufactured with Derakane/SPV6003 resin has failed due to the incidence of osmotic blistering and that all pools that have failed to date due to the incidence of osmotic blistering have a tie layer manufactured with Resin 942 or are mixed resin pools. 166 On the basis of the foregoing findings, I find that to date 44% of pools which have a tie layer manufactured with Resin 942 have failed due to the incidence of osmotic blistering and that 12% of mixed resin pools have similarly failed.

[60]

A Fabrication Problem? 168 Huntsman submitted that the fact that 60% of the pools have not (yet) suffered any failures strongly points to a fabrication problem, not a constitutional deficiency with the Resin 942. This submission has no evidentiary foundation whatsoever. That aside, assuming a constitutional deficiency in the Resin 942 caused the pool failures, without knowing the precise constitutional deficiency which was the cause (and it is beyond argument that the Resin 942 had more than one potential constitutional deficiency), it is not possible to make any meaningful prediction as to what failure rate might be expected. Assuming a fabrication problem caused the pool failures, without knowing the precise fabrication problem which was the cause, it is not possible to make any meaningful prediction as to what failure rate might be expected. 169 If it is not common ground, then it is not disputed by Huntsman that Narellan's manufacturing record over many years is essentially without incident; it has manufactured many thousands of fibreglass pools which, save for the pools the subject of this proceeding, have invariably maintained their integrity. It was not put to any of Narellan's lay witnesses that the osmotic blistering was due to Narellan's manufacturing processes or techniques and it was common ground between the experts that Narellan's pool fabrication procedure was such 'as to avoid incorrect resin application and/or contamination' (see [90] above: [2.4] of first conclave report; and [95] above: [2.2.2] of the second conclave report). In the course of his cross-examination, Professor George was asked some questions going to Narellan's fabrication process (at T 39 (07.04.09)): 'The other important factor, that is, the fabrication process, was based upon a site visit on 1 March 2007 and assumptions that you have been asked to make about the quality of the fabrication process?‑‑‑From my instructions. Yes. In your reports, you have devoted no analysis, can I suggest to you, to considering the details of the fabrication process to assess how they may have contributed to the blistering?‑‑‑Correct. Your Honour, could I elaborate? Please do?‑‑‑When one is examining a set of data where people have moved from one resin to another, one makes an assumption that they have kept the same operating procedure, so that if you have the operators operating in the usual way, and we move from one resin to another then back to it again, and we only see the failure occurring when they have changed the resin, then it is a reasonable assumption that the fabrication procedure has not changed over that period. If there were some features of that resin that meant that the fabrication procedure that was usually used did not work, then yes, you may have a problem with the wet through and the wet out of the resin - of the glass by the resin. If you were using a different kind of vinyl ester resin, with different rheology, it may be necessary to recalibrate or reconsider the particular fabrication processes that you have deployed; correct?‑‑‑I can't comment on that.' 170 Professor George was cross-examined at some length over his alleged failure to devote any analysis in his reports to considering various aspects of the fabrication process. Some of these aspects are referred to in Huntsman's written submissions at [137(6)] above. The underlying premise of the questioning process was that because Resin 942 and Derakane are different products, it could not be assumed that the process of their application in the fabrication process should be the same; and Narellan had no experience of ever having applied Resin 942. This conveniently ignores the fact that Narellan had successfully applied the production sample to three shells. But more importantly, the falsity of the assumption has no evidentiary foundation. Neither Professor George nor Professor Shanks expressed a view, individually or in conclave, that suggested that Resin 942 called for different application procedures than Derakane in the fabrication process. Nor did Dr Durrant. Certainly nothing along those lines was ever conveyed by Huntsman to Narellan at any relevant time. 171 Having regard to the matters referred to in [168] to [170] above, I am satisfied that the fabrication procedures at Narellan were not the cause of the pool failures and that the cause lies elsewhere.

[61]

The Terephthalate Mystery 172 This matter has its provenance in the points of agreement of the experts at [2.1] of the second conclave report, in particular [2.1.1] to [2.1.6] (see [94] above). At [2.1.5] the experts agree: 'The estimated level of contaminant, if it is assumed to be a terephthalate resin, is around 30% which would be over 2000 kg.' 173 Huntsman's description of it as a 'mystery' goes not so much to the identity of the contaminant, although the experts agreed it was 'uncertain' ([2.1.5]), but to how it got there in the quantity it did. Huntsman contended that it was not open for the Court to find that Huntsman tipped it in and in the absence of such a finding, or a finding as to how the contaminant got there in the quantity it did, then whatever caused the blistering, it was not Hetron 942; it was something else. According to Huntsman, as a matter of legal analysis, it may be seen as some form of novus actus interveniens which breaks the claim of causation. 174 It is clear from a number of 'the critical documents in the case', referred to by Narellan in opening, in particular those referred to at [55(1), (3), (4) and (5)] above (Ex 11: 25, 26, 34, 103, 105) that the Resin 942 was not manufactured in accordance with Huntsman's own specifications as prepared and approved by Dr Durrant. These specifications contemplated a yield of 4,770 kg of Hetron 942 per batch comprising 3,000 kg of Hetron 914 and 1,260 kg of Hetron 922 as well as other components in their quantities there listed. Only 3,000 kg of Hetron 942 was ordered by Narellan so that if the Resin 942 had been manufactured according to Huntsman's own specifications to yield 4,770 kg of Hetron 942, that would have been more than sufficient to fill Narellan's order. No explanation was forthcoming from Huntsman as to why 6,830 kg of Resin 942 was produced nor, apart from what was, or was not, recorded on the production sheet for Resin 942, the identity of the components of the additional quantity of 2,053 kg. The production sheet records that an additional four drums of Hetron 914, an additional three drums of Hetron 922 and an additional 300 litres of styrene were added but it was common ground between the experts, Professors George and Shanks, that despite these batch adjustments, there was another 400 - 450 kg of material, to yield 6,830 kg of Resin 942, which simply cannot be accounted for. Moreover, having regard to other patent deficiencies in the recordings made on the production sheet for the Resin 942, e.g., the absence of batch numbers for the Hetron 922 and Aerosil 202, and in the absence of Huntsman calling any of the operators responsible for the manufacture of the Resin 942, in particular GC and PK, I have significant reservations about the extent to which reliance can be placed on the production sheet for the Resin 942 as to the identity of the batch adjustments which were made and which resulted in an over-yield of some 2,053 kg. My reservations in this regard assume a greater importance, and indeed significance, in the face of the agreement of the experts, Professors George and Shanks, in the second conclave report that pools manufactured with the Resin 942 all have an infra-red signature that is consistent with a terephthalate impurity level of around 30% (see [95] above), namely, 2,049 kg (cf., 2,053 kg) of the total yield of 6,830 kg. 175 Accepting the agreed view of the experts that: 'Narellan has procedures in place to ensure that contamination cannot occur' ([2.2.2] of second conclave report); and that: 'Huntsman has procedures in place … reducing the risk of contamination' ([2.2.3] of that report), the fact remains that for the reasons referred to in [174] above, the production sheet for the Resin 942 is sufficiently doubtful in terms of its reliability and accuracy, that in the absence of evidence from those operators responsible for making the batch adjustments and writing up the production sheet, it is open for me to find, and I so find, that it does not properly record batch adjustments of over 2,000 kg. 176 For these reasons, and accepting, as I do, Narellan's submissions set out at [113] above, I am satisfied, on the balance of probabilities, that the source of the contaminant in the quantity agreed by the experts is to be found in Huntsman's manufacturing process, rather than in some intervening act or default on the part of Narellan in the course of its fabrication procedures.

At a glance

Court

Decision date

Before

Source

Judgment (65 paragraphs)

Parties

Legislation Cited (2)

Cases Cited (4)

Cited By (2)