Chantiers de L'Atlantique v Gaztransport & Technigaz

From DMC
Jump to: navigation, search



Chantiers de L’Atlantique SA v Gaztransport & Technigaz SAS

English High Court (Commercial Court): Flaux J: [2011] EWHC 3383 (Comm): 20 December 2011

Mr Jonathan Hirst QC, Mr Alan Maclean QC and Mr Stephen Midwinter, instructed by Clyde & Co, for the Claimants, Chantiers de L’Atlantique SA (CAT)

Mr Toby Landau QC and Mr Iain Quirk, instructed by Orrick, Herrington & Sutcliffe (Europe) LLP, for the Defendants, Gaztransport & Technigaz SAS (GTT)



CAT applied under s.68(2)(g) of the Arbitration Act 1996 to set aside, on the ground of fraud, an arbitral award obtained by GTT. Flaux J granted an extension of time to make the application given the importance of the allegations raised. The Court held on the facts that there had been deliberate concealment by the winning party of adverse test results in relation to the viability of the LNG containment system in question, and in particular, that GTT’s head of research and development deliberately concealed in his evidence the existence of the first set of tests and their adverse results. This constituted fraud for the purpose of the 1996 Act. However, the award would not be set aside as the disclosure of the adverse results would not have affected the decision of the tribunal.

This note has been contributed by Ken T.C. Lee, LLB(Hons), PCLL (University of Hong Kong), BCL(Oxon) and Barrister-at-law in Hong Kong.


The Claimant, Chantiers de L’Atlantique SA (CAT), was a French shipbuilding company which specialized in building LNG carriers. The Respondent, Gaztransport & Technigaz SAS (GTT), specialized in the design of containment systems for LNG carriers.

Pursuant to a Technical Assistance and Licence Agreement (“the Agreement”) dated 17 December 2001, GTT licensed CAT to use its “CS1” LNG containment system in the construction of three LNG vessels. The system involved a number of layers between the LNG and the hull, and some of the layers were made up of more than one material (including Triplex panels) bound together by joints.

In November 2004, sea trials of one of the vessels took place and the LNG tank was cooled down. It was found that nitrogen gas (which was used for testing) was passing through the secondary membrane. This suggested a serious fault which needed to be resolved before the vessel could enter service. The parties agreed that GTT would carry out a series of laboratory tests in order to establish if there was an inherent difficulty in the bonding of the Triplex panels, and if so, how it could be resolved.

CAT supplied GTT with panels and glues used in the construction of the LNG tank. GTT used these materials to prepare “sandwich” samples used in the CS1 system, and subjected the samples to tests which simulated the stresses that would be encountered by the secondary barrier on board a vessel. The test results of this first batch of samples, known as “B0”, were totally unsatisfactory as there was failure in the bonding of the glue.

GTT then placed an urgent order directly with Hankuk, CAT’s supplier of Triplex panels, for the supply of Mark III panels. Sandwich samples were prepared with these Mark III panels (B1, B2, B3 and B5 samples), and tests were carried out on them. CAT was invited to, but did not, attend the tests. Only representatives of the classification society for the vessels were present. On 29 June 2005, the President of GTT wrote a letter to the President of CAT stating that the tests on the samples of the CS1 system gave results which were compliant with the calculations made by GTT and other experts with respect to the level at which they ruptured and the mode of rupture. It was further stated in the letter than there was no design error in the CS1 system. The results of the tests (in fact only the B1 tests) were enclosed in the letter in the form of a technical note headed, in French, “Samples CS1 Traction Test”.

In the meantime, the parties had been exploring improvements to the bonding process. This resulted in a “Technical Solution” signed on 5 July 2005. Before and after the Technical Solution, GTT carried out further tests using samples intending to be representative of the CS1 system, and the results of these tests were broadly satisfactory.

However, the Technical Solution worked imperfectly, and relations between the parties broke down. CAT commenced an ICC arbitration against GTT in July 2006 pursuant to the ICC procedural rules, seeking damages for the losses suffered as a consequence of the bonding failures. In order to succeed in its claim, which was governed by French law, CAT had to prove “gross fault” on the part of GTT.

During the course of the arbitration, CAT sought disclosure of the results of various tests carried out by GTT. GTT disclosed the results of tests B1, B2, B3 and B5, which supported its claim that there was no inherent problem in the CS1 system, but the results of the B0 tests were not disclosed. Requests were also made for various other test results and to reopen GTT’s data room, but CAT received no satisfactory reply.

In February 2009, the arbitral tribunal dismissed CAT’s claim on the ground that it could not establish the necessary criterion of “gross fault” as a matter of French law. Further, it was held that even if CAT could establish design fault, CAT could not satisfy the test imposed by French law as to the circumstances in which a licensor would be liable to a licensee for a design fault (“the Award”). In March 2009, CAT applied to the French court, challenging GTT’s execution of the Award.

Two weeks after the award was published, CAT received a tip-off from a disaffected employee of GTT and found that there were apparently non-explicable discrepancies in various test programmes. On 31 July 2009, shortly after the first of its expert reports was available, CAT took out the application to set aside the Award under s.68(2)(g) of the Arbitration Act 1996 on the ground of fraud. There was also an application to extend time as the 28-day period to challenge the Award under s.28(3) of the Arbitration Act 1996 had already expired. However, it was subsequently found that these discrepancies disclosed no fraud.

Towards the end of 2010, CAT received an envelope from an anonymous whistleblower, containing an email which referred to the unsatisfactory results of the B0 tests. This was the first time CAT had heard of the B0 tests, and CAT eventually obtained copies of the report of the B0 test results. The application to set aside the Award was amended. The principal allegations of fraud were the concealment of the B0 test results, the fact that B1-B5 tests were conducted on Mark III panels and not CS1 panels, and that M. Chapot, the head of research and development of GTT, concealed both the B0 tests and the nature of B1-B5 tests when giving evidence before the tribunal. Other allegations included concealment of test results for the glue and other studies.


Flaux J dismissed CAT’s application to set aside the arbitral award.

Firstly, Flaux J noted that, in deciding whether time should be extended under s.80(5) of the Arbitration Act 1996, the court would consider various factors including the length of the delay, whether the party was acting reasonably in all the circumstances, whether the respondent would be prejudiced by the delay and the strength of the application (see Kalmneft v Glencore [2002] 1 Lloyd’s Rep 128). Although CAT had not issued its application immediately on receiving the tip-off two weeks after the award was published, it had acted reasonably by investigating the matter and in issuing its application after the first expert report had beome available. Given the importance of the allegations raised, Flaux J was reluctant to shut out the application on grounds of delay and an extension of time was granted.

Secondly, Flaux J noted that, in considering a challenge against an arbitral award on grounds of serious irregularity, the court must be satisfied that some form of unconscionable conduct had contributed in a substantial way to the obtaining of the award. Where the allegation was fraud in the production of evidence, the onus was on the applicant to make good the allegation by cogent evidence. The new evidence relied upon to demonstrate the fraud must be such that it could not have been obtained or produced at arbitration with reasonable diligence, and would have had an important influence on the result: see Double K Oil Products 1996 Limited v Neste Oil OYJ [2010] 1 Lloyd’s Rep 141; Elektrim SA v Vivendi Universal SA [2007] 1 Lloyd’s Rep 693.

Flaux J found on the facts that a deliberate decision had been taken by GTT management not to inform CAT about the B0 tests and their results, at least until further tests had been carried out. Internal emails of GTT showed that there was consternation and shock amongst senior management of GTT about the B0 test results, given the recurrence of unacceptable rupture in the tests. They were also in a hurry to carry out the test programme and get the Technical Solution signed. This was why GTT had ordered additional panels from Hankuk rather than from CAT. Flaux J rejected GTT’s explanation that there was contamination on the materials supplied by CAT, as it had not raised this problem with CAT before. Further, the evidence from the head of GTT’s laboratory was that he and possibly other employees were instructed by senior management not to say anything to CAT about the B0 test results.

The letter by the President of GTT in June 2005 continued the concealment and was deliberately misleading. CAT was justified in understanding from the letter that tests had been carried out on CS1 panels, and CAT had signed the Technical Solution as a result of the satisfactory test results. Although there was reference to “cooling down of sample B0-2” in an appendix to the Technical Solution, CAT could not be expected to pick up the existence of the B0 tests from the very small writing, let alone the whole history of the B0 tests.

There was no fraud in relation to the disclosure exercise. The Court noted that the arbitration here was conducted in accordance with civil law arbitration procedure. The rules for disclosure were based on the IBA rules, and there was no duty to disclose relevant documents. Flaux J found on the facts that, despite the unsatisfactory answers from GTT to the request for disclosure, CAT did not suspect deliberate concealment of the B0 test results by GTT. Rather, those drafting the answer to the requests were not party to any decision not to disclose the B0 results to CAT, and had honestly forgotten about the internal email on the result of B0 tests by the time of the disclosure exercise. The senior management did not knowingly allow misleading responses to be provided.

However, there was deliberate concealment by M. Chapot in his expert report and oral evidence to the tribunal of the existence of the B0 tests and their results and he had given the tribunal a false impression that B1-B5 tests were carried out on CS1 panels supplied by CAT. This went beyond the situation where a witness for one of the parties to an arbitration was lying to the tribunal and committed perjury (Elektrim SA v Vivendi Universal SA [2007] 1 Lloyd’s Rep 693). As M. Chapot was deputed to present GTT’s technical case to the tribunal, there was fraud by GTT as a party to the arbitration for the purpose of the Arbitration Act.

However, Flaux J held that even if the true position had been disclosed to the tribunal, it would probably not have made any difference to its decision. Tests conducted after the B0 tests demonstrated the ability of the CS1 system to withstand the stresses and the effectiveness of the bonding. Mark III panels were technically equivalent and representative of CS1 panels. CAT had also successfully installed the CS1 system after the Technical Solution and the vessels had been in service for a few years without any apparent problems. These pointed away from any design defect. Rather, the unsatisfactory outcome might have been due to the inexperience of the operators in applying the glue. There was also unchallenged evidence that the B0 tests were preliminary in nature. GTT had not carried out a traction test on such large scale samples before and the B0 test was used to set up the equipments. Had GTT explained to CAT about the teething problems in carrying out the tests in an honest and open way, CAT would probably still have signed the Technical Solution, albeit at a later time. Flaux J also held that even if CAT could establish design fault, CAT would not have satisfied the high test imposed by French law as to the circumstances in which a licensor would be liable to the licensee for a design fault, namely, that the technology was technically unusable or extremely difficult to use.

With respect to the other allegations of fraud in relation to tests on the glue, etc., the Court held that CAT was estopped from raising them, as they had already been raised before the French courts which had given judgment against CAT. This was so even if the foreign court adopted a different procedure or a different rigour in its consideration of the evidence, as the practice of English courts was not the only or even necessarily the best method of arriving at justice: see Yukos Capital v OJSC Rosneft Oil Company [2011] EWHC 1461 (Comm).