Borealis AB v Geogas Trading SA

The Claimant (“Borealis”) claims damages from the Defendant (“Geogas”), arising out of the supply by Geogas to Borealis of some 5,200 mt of butane as feedstock for Borealis’ integrated olefin plant situated at Stenungsund, Sweden, in September 2003 (“the plant”). It is Borealis’ case that, in breach of contract, Geogas supplied butane (“the goods”) heavily contaminated with fluorides that cracked under normal processing conditions to produce, amongst other substances, hydrofluoric acid (“HF”) which, in turn, caused serious and extensive physical damage to the plant and equipment, together with consequential interruption to Borealis’ business.

It is common ground that the goods were contaminated with fluorides. It is further admitted by Geogas that it was in breach of contract – in that the contamination of the goods (by 2 methyl 2 fluoro propane, “2M2F”) placed Geogas in breach of an implied term of the contract to the effect that the goods must be of satisfactory quality. Geogas does not, however, concede that it was in breach of contract in other respects alleged by Borealis.

Against this background, the trial was essentially about causation, remoteness, mitigation and quantum.

Much of the factual history is common ground and can largely be taken from the relevant section of the most helpful List of Issues, prepared by counsel.

Parties: Borealis is a well known producer of polyethylene and polypropylene (“plastics”) for use in various applications with production sites all over the world. The plant includes an ethylene cracker complex that processes hydrocarbon feedstocks (such as butane and propane) to produce ethylene and propylene.

Geogas is a well-known trader in, inter alia, liquefied petroleum gas (“LPG”), including butane.

The contract: By an oral agreement, evidenced in and/or subsequently reduced into writing, entered into between Borealis and Geogas on 22^nd August, 2003, Geogas agreed to sell and Borealis agreed to buy 5,200 mt (5% more or less at Geogas’ option) of butane (i.e., the goods), at a price of US$259.50 per mt (“the contract”).

A firm called Norenergy Oilbrokers AS (“Norenergy”) acted as the parties’ broker. The documents evidencing the contract include Norenergy’s e-mail sent to the parties at 18.11 on the 22^nd August, 2003 confirming the transaction (“the 22^nd August e-mail”) and a further e-mail sent to the parties at 11.53 on the 25^th August, 2003 (“the 25th August e-mail”), setting out the detailed terms.

The 22^nd August e-mail stated “quality: field grade as per specs in separate e-mail”.

The 25^th August e-mail confirmed the sale of 5,200mt of “commercial butane to be delivered colder than zero deg c” on, inter alia, the following express terms:

“ quality: field grade with specs as follows:

c3 and lighter less than 1pct

nc4 80pct

ic4 14pct

butenes 3pct

c5 plus less than 2pct

sulphur less than 50ppm

oxygenates nil

water nil

copper corr 1 max

origin: us east coast

quantity: 5,200mt 5 pct molso

delivery: cif one safe port/berth stenungsund

during period 7-12^th september 2003

price: usd 259.50 mt

vessel: gt henning maersk….

inspection: at loadport to be inspected by independent inspector to be appointed and paid for by seller. At discharge port cargo to be inspected by independent inspector to be appointed and paid for by buyer ”

As is common ground, there were implied terms of the contract:

Delivery: The goods (in the event, 5,455.662 mt of butane) were sourced by Geogas from petrochemical facilities owned by Sunoco Inc. (“Sunoco”), situated at Marcus Hook, Pennsylvania. The goods were loaded onto the carrying vessel, the “Henning Maersk” (“the vessel”), on the 27^th and 28^th August, 2003 and the vessel arrived at Stenungsund on the 12^th September, 2003.

Discharge from the vessel into one of the plant’s underground caverns, number UC-731 (“the cavern”), commenced at about 13.20 the same day and was completed at about 00.58 on the 13^th September. At the time of discharge, the cavern already contained some 25,276mt of butane/propane feedstock mixture. As discharge took place, so fluids from the cavern were simultaneously fed to the plant (to the cracking furnaces) as is usual.

The contamination: As already noted, the goods were contaminated with a compound subsequently identified as 2M2F.

The incident: At about 20.00 on the 13^th September, 2003, a gas alarm in the disengaging basin (indicating contaminated seawater cooling medium returning to the sea) signalled the presence of hydrocarbon gas indicative of leaks in the cracker plant’s overhead heat exchangers. Subsequent investigations revealed that numerous tubes in the cracker’s primary and secondary heat exchangers sustained an extensive and severe corrosion attack in the hours leading up to the triggering of the gas alarm on the 13^th September – so much so that several of the tubes became perforated allowing (inter alia) ethylene to escape into the atmosphere.

Further matters not in dispute: A number of further matters were not in dispute:

General: I must next and in very simple terms, based on the excellent agreed note prepared by counsel, describe the working of the plant.

The Stenungsund site comprises a cracker complex, 3 polyethylene plants and an innovation centre. The plant uses naphtha, ethane, propane and butane as feedstock to produce ethylene and propylene which are in turn used to produce high-density polyethylene, low-density polyethylene and patented polyethylene products, primarily for pipe, wire and cable applications.

The raw materials are delivered by ship directly to Borealis’ harbour in Stenungsund and are stored in four underground “caverns” and one over ground ethylene tank.

Butane is stored in cavern UC-731 (i.e., the cavern), along with a proportion of propane (75:25 butane/propane) to create an LPG feed.

The cracker alone is a large plant consisting of approximately 950 kilometres of pipes, 20 distillation towers, reactors, some 400 pumps and 20,000 valves.

A process flow diagram of the relevant part of the plant for present purposes is attached herewith as Annexe I.

In essence, naphtha, ethane, propane and butane/propane (the LPG mix) in various quantities are fed from the caverns (and the over ground ethane tank) to 9 cracking furnaces (F-1601 A-G, V and X), where the feeds are mixed with steam, in a ratio of 70:30 hydrocarbon:steam, and heated to 850-870ºC. (It may be noted – though it does not effect this description - that on the date of the incident, 7-8 furnaces were operating.) The effect is to break up long chains of hydrocarbon molecules into shorter ones. The individual product streams from the furnaces are then cooled (quenched) individually, by heat exchangers (not the subject of these proceedings), before being mixed into a single product stream. That product stream is then transported through an overhead transfer line to the T-1651 primary fractionator tower. This is the plant’s first stage distillation tower where heavier components in the product stream are separated out from the lighter ones. The lighter components, consisting primarily of ethylene and steam, are taken from the top of T-1651 through overhead pipelines (the T-1651 overhead line) to a set of primary and secondary heat exchangers: E-1651 and E-1656.

The E-1651/1656 heat exchangers: E-1651/1656 are made up of 16 heat exchangers – E-1651 A-H and E-1656 A-H. They operate in pairs, each consisting of a primary exchanger sitting above a secondary exchanger: so, e.g., E-1651 A and E-1656 A. Further, they are divided into a North bank and a South bank, each bank consisting of 4 exchanger pairs. The South bank is made up of units A, C, E & G. The North bank is made up of units B, D, F and H.

The heat exchanger shells are made of carbon steel. The shells house heat exchanger tube bundles, consisting (in the case of the primary exchangers) of some 1,445 titanium tubes (i.e., ~ 8.8km of tubes). Cooling sea water flows through the titanium tubes whilst the product stream flows around the outside (i.e., the shell) of the tubes. Cooling water used in the plant as a whole is returned to the sea through two effluent pipes, known as the northern header and the southern header, which then merge before entering the disengaging basin (D-1101). The cooling water from the E-1651/1656 heat exchangers passes through the northern header. Alarms exist to detect the presence of hydrocarbons in the cooling water returning to the sea in both headers and in the disengaging basin.

The product stream from the T-1651 overhead line passes first into the primary E-1651 heat exchangers where it is cooled from about 110ºC to 58ºC and then on into the secondary E-1656 heat exchangers (sitting, as already described, below the primary exchangers). In the secondary heat exchangers, the product stream is further cooled to approximately 35ºC. The cooled product stream then flows downstream into a distillate drum (D-1681), where the hydrocarbon gas, hydrocarbon liquid and water are further separated from each other.

The pH of the water stream leaving D-1681 (sour water) is measured by an on-line meter (the pH meter) and also, three times a week, by the cracker laboratory. The measurements are recorded and, if the on-line pH meter readings drop below 5 (or increase above 7), a visible and audible alarm display in the cracker control room activates by means of a flashing white button which the operator acknowledges by pressing the button.

The feedstocks can contain compounds such as oxygenates that once cracked and dissolved in water create weak acids, e.g., carbonic acid. Over time, these acids can corrode carbon steel.

The plant is fitted with a system that provides for the manual injection of a neutralising agent, here ammonia, into the T-1651 overhead line. The original design purpose of the ammonia injection system was disputed between the parties; Geogas’ case was that it was intended to neutralise any acids, without discrimination, that might be present in the system; Borealis’ case was that it was to neutralise weak acids that form naturally in the system from the feed stocks used, including carbonic acid resulting from the presence of oxygenates in the feedstocks.

The unfolding of the incident can be seen having regard to a number of significant times which it is now convenient to bring together here (albeit that some have already been referred to):

To begin with, a word needs to be said as to the significant narrowing of the issues in the course of the trial.

I have already indicated the broad outline of the Borealis case; namely, that in breach of the contract, Geogas supplied heavily contaminated goods which caused serious and extensive physical damage to the plant and equipment, together with a variety of consequential losses.

Initially, amongst these consequential losses, the Borealis claim included the amount of €4,838,238 (some 38% of the total claim), in respect of alleged loss of profits in November 2003, said to have resulted from the damage to the heat exchangers in September 2003. Realistically, if I may say so, Ms. Buehrlen QC, for Borealis, did not ultimately pursue this head of claim, recognising the difficulties in establishing the requisite causative link.

At the start of the trial, Geogas relied on an extensive assortment of defences, including the following.

Realistically, with respect, Mr. Ashcroft, for Geogas, jettisoned these defences over the course of the trial, recognising the legal and factual difficulties they variously faced and, in at least one instance (the currency point) taking a simple pragmatic view.

As already recorded, Geogas conceded before the trial that it was in breach of an implied term of the contract that the goods must be of satisfactory quality. By the conclusion of the trial, both counsel were sensibly agreed that the question of whether there were further breaches of contract on the part of Geogas was academic.

In the event, by the time of closing submissions, Geogas’s case was, in a nutshell, as follows. First and foremost, Geogas relied on its “secondary case” (as it was known by way of shorthand throughout the trial); namely, the allegation that Borealis had failed to react appropriately, or at all, to the pH alarm (which sounded at 15.20 on the 12^th September). Conceptually, this case was advanced both in terms of causation and mitigation. Borealis’ acts or omissions following the sounding of the pH alarm had broken the chain of causation and Geogas denied liability for the resulting inflated losses, especially in respect of consequential loss; alternatively Geogas was not liable for avoidable losses flowing from a failure to mitigate. Secondly, Geogas raised two discrete arguments involving remoteness of damage. The first related to Borealis’ claim for loss of profits in September 2004 and the second to the unavailability of the cavern. Thirdly and (effectively or largely) in the alternative to the second of its remoteness arguments, Geogas submitted that the cavern was unavailable for too long; in that regard, Borealis had failed to mitigate its loss and, to the extent that it had so failed, Geogas denied liability. Fourthly, Geogas continued to maintain a miscellany of arguments as to the quantum of damage.

It follows that the principal issues which remained at the end of the trial and with which I must deal, concerned causation, remoteness, mitigation and quantum. These may be summarised as follows:

This narrowing of the issues had certain consequences. First, on any view, Borealis was entitled to a substantial sum by way of damages. In the circumstances, the Borealis application at the conclusion of closing submissions for an interim payment was irresistible in principle; in the exercise of my discretion, I duly made an order for such a payment in the amount of €2.75 million. This judgment therefore deals with the balance of the sums in dispute.

Secondly, Borealis called a large number of factual witnesses and Geogas two such witnesses. Both parties called or relied upon a number of expert witnesses. It is unnecessary to list these witnesses here. Suffice to say that in the light of the narrowing of the issues, a not insignificant part of the evidence had ceased to be of practical relevance by the end of the trial. Such comments as I have on particular witnesses are best deferred until later and in the context of the issues which remain.

It is convenient at this stage, to express my thanks to both Ms. Buehrlen and Mr. Ashcroft, not only for their realism in narrowing the issues in dispute but also for the quality of their written and oral submissions – those thanks extending to the assistance given by Mr. King, Ms. Buehrlen’s junior, in the area of quantum.

Causation: I start with causation, extensively debated by counsel when addressing the Geogas secondary case (Issue (I)). The specific focus was the test for breaking the chain of causation. My approach, distilled from the submissions of counsel and the authorities cited, is summarised in the paragraphs which follow.

First, although an evidential burden rests on the defendant insofar as it contends that there was a break in the chain of causation, the legal burden of proof rests throughout on the claimant to prove that the defendant’s breach of contract caused its loss.

Secondly, in order to comprise a novus actus interveniens, so breaking the chain of causation, the conduct of the claimant “must constitute an event of such impact that it ‘obliterates’ the wrongdoing…” of the defendant: Clerk & Lindsell on Torts (19^th ed.), at para. 2-78. The same test applies in contract. For there to be a break in the chain of causation, the true cause of the loss must be the conduct of the claimant rather than the breach of contract on the part of the defendant; if the breach of contract by the defendant and the claimant’s subsequent conduct are concurrent causes, it must be unlikely that the chain of causation will be broken. In circumstances where the defendant’s breach of contract remains an effective cause of the loss, at least ordinarily, the chain of causation will not be broken: County Ltd v Girozentrale [1996] 3 All ER 834, at p. 849 b-c, per Beldam LJ and at pp. 857 f-g and 858 b-c, per Hobhouse LJ (as he then was). Other examples can be found in the area of shipping law. Where, in breach of charterparty, charterers order a vessel to proceed to an unsafe port, the conduct of the vessel’s master in obeying the order (placed as he well may be, on the horns of a dilemma) will be judged sympathetically, in context and will not lightly be treated as unreasonable: Compania Naviera Maropan v Bowaters (The “Stork”) [1955] 2 QB 68. But even negligent navigation following the charterer’s order to proceed to an unsafe port will not necessarily break the chain of causation: see, for example, The Polyglory [1977] 2 Lloyd’s Rep. 353, at p.366. Conversely, where the negligence of vessel X caused vessel Y to run aground, vessel X was not liable for such damages as were attributable to the subsequent, clearly separate and negligent re-floating of vessel Y: The “Spontaneity” [1962] 1 Lloyd’s Rep 460; the negligence of vessel X had ceased to be operative.

Thirdly, it is difficult to conceive that anything less than unreasonable conduct on the part of the claimant would be capable of breaking the chain of causation. It is, however, also plain that mere unreasonable conduct on a claimant’s part will not necessarily do so – for example where the defendant’s breach remains an effective cause of the loss, albeit in combination with the claimant’s failure to take reasonable precautions in its own interest: see, for example, County Ltd v Girozentrale, per Beldam LJ (loc cit). By its nature, reckless conduct by the claimant would or would ordinarily break the chain of causation, though there is no rule of law that only recklessness on the part of the claimant will do so: Lambert v Lewis [1982] AC 225, per Roskill LJ (as he then was) in the Court of Appeal, at p.252; County Ltd v Girozentrale (supra), per Hobhouse LJ at p. 857, more conveniently discussed below, when dealing with the claimant’s knowledge or lack of it.

Fourthly, the claimant’s state of knowledge at the time of and following the defendant’s breach of contract is likely to be a factor of very great significance. For the chain of causation to be broken, the claimant need not have knowledge of the legal niceties of the breach of contract; nor, as it seems to me, will the chain of causation only be broken if the claimant has actual knowledge that a breach of contract has occurred – otherwise there would be a premium on ignorance. However, the more the claimant has actual knowledge of the breach, of the dangerousness of the situation which has thus arisen and of the need to take appropriate remedial measures, the greater the likelihood that the chain of causation will be broken. Conversely, the less the claimant knows the more likely it is that only recklessness will suffice to break the chain of causation. Ready illustrations are furnished by the authorities:

Fifthly and ultimately, the question of whether there has been a break in the chain of causation is fact sensitive, involving as it does a practical inquiry into the circumstances of the defendant’s breach of contract and the claimant’s subsequent conduct. As Roskill LJ observed, in Lambert v Lewis (loc cit), it is “always a question of degree at what point the damage claimed for ceases to flow naturally and directly from the breach”. For my part, I respectfully agree with the observations of Evans-Lombe J, in Barings Plc v Coopers & Lybrand [2003] EWHC 1319 (Ch), at [838], where he expressed an unwillingness to accept any all-embracing test for what may constitute the breaking of the chain of causation:

“ …It seems to me that what will constitute such conduct is so fact-sensitive to the facts of any case where the issue arises that it is almost impossible to generalise. If one must do so, I would say that it must be some unreasonable conduct, not necessarily unforeseeable….., a new cause coming in and disturbing the sequence of events….., not necessarily reckless….., which may result from an accumulation of events which in sum have the effect of removing the negligence sued on as a cause….., which accumulation of events may take place over time….”

I would add too that while the authorities of course provide guidance, they are not to be read as statutes.

Remoteness: This topic can be taken very shortly indeed. First, because as already underlined it is raised by Geogas only in respect of two discrete areas of dispute. Secondly, because the “orthodox approach” (Hamblen J in Sylvia Shipping Co Ltd v Progress Bulk Carriers Ltd (The Sylvia) [2010] EWHC 542 (Comm); [2010] 2 Lloyd’s Rep. 81; at [20] et seq) based upon Hadley v Baxendale (1854) 9 Ex 341 and Czarnikow v Koufos (The Heron II) [1969] 1 AC 350 has been settled for so long. As succinctly formulated by Lord Reid in The Heron II, at pp. 382 - 3, was the loss of a kind which the defendant at the time of the contract ought to have realised was “not unlikely” to result from the breach - meaning “a degree of probability considerably less than an even chance but nevertheless not very unusual and easily foreseeable”? Thirdly, because if and to the extent that, with respect, the speeches of Lords Hoffmann and Hope in Transfield Inc v Mercator Shipping Inc (The Achilleas) [2008] UKHL 48; [2009] 1 AC 61 have added “assumption of responsibility” as an additional requirement of the remoteness rule in some cases (so that a loss even if “not unlikely to result” might still be too remote if the defendant had not assumed responsibility for that type of loss), the orthodox approach continues to apply in the “great majority of cases” - as Lord Hoffmann’s and Lord Hope’s observations themselves make clear: see, The Achilleas, at [9], [11] and [36]; see too, the valuable discussion to be found in Hamblen J’s judgment in The Sylvia (supra), at [20] – [51]. There are, in my judgment, no exceptional circumstances in the present case calling for specific consideration of the “assumption of responsibility” test. Fourthly, because, looking ahead, I cannot see that the application of Lord Hoffmann’s test (whether a party would reasonably be regarded as having assumed responsibility for the type of losses in question) to the issues in dispute here would realistically produce a different outcome to that flowing from the application of the orthodox approach. For present purposes, it is therefore unnecessary to say more as to the law on remoteness.

Mitigation: Mitigation was relevant both to the Geogas secondary case and to the discrete point as to the unavailability of the cavern. As it is so well settled and was not in dispute before me, the law here can likewise be taken very shortly. A convenient starting point is to be found in the judgment of Sir John Donaldson MR in The “Solholt” [1983] 1 Lloyd’s Rep. 605, at p. 608:

“ A plaintiff is under no duty to mitigate his loss, despite the habitual use by the lawyers of the phrase ‘duty to mitigate’. He is completely free to act as he judges to be in his best interests. On the other hand, a defendant is not liable for all loss suffered by the plaintiff in consequence of his so acting. A defendant is only liable for such part of the plaintiff’s loss as is properly to be regarded as caused by the defendant’s breach of duty. As Viscount Haldane, LC, put it in British Westinghouse Electric and Manufacturing Co. Ltd v Underground Electric Railways Co. of London Ltd., [1912] AC 673 at p.689:

‘The fundamental basis is thus compensation for pecuniary loss naturally flowing from the breach; but this first principle is qualified by a second, which imposes on a plaintiff the duty of taking all reasonable steps to mitigate the loss consequent on the breach, and debars him from claiming any part of the damage which is due to his neglect to take such steps.’

……

Whether a loss is avoidable by reasonable action on the part of the plaintiff is a question of fact not law. This was decided in Payzu Ltd v Saunders [1919] 2 KB 581.”

For present purposes, it is only necessary to add the following. First, the onus of proof on the issue of mitigation is on the defendant: McGregor on Damages (18^th ed.), at para. 7-019. Secondly, the conduct of the innocent party should not be “weighed in nice scales at the instance of the party whose breach of contract has occasioned the difficulty”: per Lord Macmillan in Banco de Portugal v Waterlow [1932] AC 452, at p. 506. Immediately thereafter (ibid), Lord Macmillan said this:

“ It is often easy after an emergency has passed to criticise the steps which have been taken to meet it, but such criticism does not come well from those who have themselves created the emergency. The law is satisfied if the party placed in a difficult situation by reason of the breach of a duty owed to him has acted reasonably in the adoption of remedial measures, and he will not be held disentitled to recover the cost of such measures merely because the party in breach can suggest that other measures less burdensome to him might have been taken.”

(1) The rival contentions: As already introduced, the Geogas secondary case is that Borealis failed to react appropriately, or at all, to the pH alarm which sounded at 15.20 on the 12^th September – some two hours after discharge of the goods had commenced. Borealis’ conduct broke the chain of causation between the admitted Geogas breach of contract and the loss and damage claimed by Borealis; alternatively, Borealis had failed to mitigate its loss. In broad outline and as presented in his closing submissions, Mr. Ashcroft argued that the operators’ failure to take prompt action in response to the pH alarm was unreasonable. The pH trend should have been monitored; such monitoring would have revealed a severe and unusual fall. The reliability of the pH meter should have been confirmed, ammonia should have been injected and laboratory samples obtained; that it was a Friday night was neither here nor there – the plant operated on a 24/7 basis. On the footing that, contrary to expectations, the injection of ammonia did not result in any significant increase in the pH level, plant management should have been involved. Once the matter had been properly considered, the linkage between the contamination and the goods, now forming part of the feedstock from the cavern, would have become clear. Urgent steps should then have been taken to stop using the LPG mix from the cavern; these should have been accomplished within 22 hours of the alarm sounding – so by 13.20 on the 13^th September – the contaminated feedstock from the cavern should no longer have been used. Accepting that the physical damage sustained by the plant was substantially suffered in the first 58 hours following discharge, Geogas should not be liable for more than 22/58 of that physical damage, or at any rate, Geogas’s liability for physical damage should be substantially reduced reflecting the time by when the feed should have been changed. Even if that were wrong – and by the conclusion of his submissions Mr. Ashcroft realistically accepted that the “meat” of the Geogas secondary case lay in reducing or eliminating its liability for consequential loss – had the feedstock been changed in a timely manner, the titanium tubes would not have been holed and the very sizeable claim for lost production in September 2003 would have been avoided.

Ms. Buehrlen’s response for Borealis was forthright. There had been no unreasonable conduct on the part of Borealis; even if there had been, it was not such as to break the chain of causation, nor had there been a failure to mitigate; in any event, the Geogas secondary case (if reasoned through) would not have avoided any loss, physical or consequential. Borealis did not know and had no reason to suspect that the goods had been contaminated with fluorides. The Geogas focus on the pH alarm failed to take into account the purpose of that alarm and Borealis’s operational experience. Having regard to the context, it would not have been unreasonable for the operators to wait until Monday 15^th September for the next scheduled laboratory samples – but of course by then, any concerns as to the pH alarm had been overtaken by events. There was moreover no or insufficient reason to suppose that, if steps had been taken in accordance with the Geogas secondary case, plant management would have concluded that the LPG feed from the cavern should be stopped. Even if the ammonia injection – though incapable of preventing damage to the primary heat exchangers - did not mask the damage being done by the HF in the contaminated feed, the more likely conclusion would have been that the drop in pH was attributable to the high level of oxygenates in the goods. In any event, the timeline contemplated by the Geogas secondary case was artificial and unrealistic. On the evidence, it would have required at least 12 hours to take the relevant furnaces off-line so as to change the feed; as it was known that titanium tubes had perforated by 20.00 on the 13^th September – the time when the gas alarm sounded – the Geogas secondary case went nowhere unless all the relevant decisions and conclusions had been reached by 08.00 on the 13^th September at the latest. Further, on the undisputed evidence that the titanium tubes needed to be replaced if 20% damaged, it could not be said that physical damage would have been avoided, even if the contaminated feed had been stopped as contended for by Geogas; there was not and could not be any assumption that damage was linear. Still further and as to consequential loss, the reasoning of the Geogas secondary case did not make good the case that the September 2003 production would not have been lost; had Borealis suspected the presence of an unusual and corrosive acid in the system, it was unlikely that full production would have gone ahead, regardless.

(2) The Geogas breach of contract: With a view to placing the Borealis reaction to the Geogas breach of contract in context it is appropriate first to focus on the breach of contract itself – albeit that the breach has been conceded. That concession, welcome, forensically wise and perhaps inevitable though it was, should not be allowed to deflect attention from the position in which Borealis found itself by reason of the breach.

The reality is that Geogas supplied goods which were heavily contaminated with fluorides. In accordance with the figures contended for by Borealis –Geogas advanced no positive case to the contrary – I find that the average level of contamination was 265 – 295 ppm fluoride compounds (wt:wt), corresponding to 1,060 – 1,180 ppm of the actual contaminant, 2M2F.

Samples of the contents of the cavern following discharge of the goods from the vessel showed levels of fluoride contamination up to 83ppm (wt:wt) and averaging 55ppm corresponding to 220ppm (wt:wt) of 2M2F. Prior to discharge of the goods into the cavern, the contents of the cavern contained levels of fluorides of 1ppm or less, i.e., at most trace levels.

The contract provided, expressly, that Geogas would supply Borealis with “field grade” butane. The evidence at trial differed as to whether “field grade” meant butane that had come directly from a field (so that “field grade” related to the source of the goods) or that it had the same quality or purity characteristics as butane sourced from a field (so that “field grade” related to the quality of the goods) – i.e., a high quality cargo free of contaminants. For present purposes it matters not which definition of “field grade” is adopted. Butane sourced from a field would not be contaminated with fluorides. “Field grade” butane sourced from a refinery plainly ought not to have been contaminated with fluorides. In any event, butane sourced from a refinery could only be contaminated with fluorides if the refinery was using an HF alkylation unit. On the evidence, of the 31 cargoes of butane sourced by Borealis from refineries in the period January – September 2003, only one (the “Stanlow cargo”) was sourced from a refinery using an HF alkylation unit. On the evidence of a number of witnesses called by Borealis (Messrs. Andersson, Petersen, Eerola and Harris) I further find that:

Borealis was further unaware that although the contract had provided for a water content of “nil”, the goods had in fact contained some 10mt of water at the load port. I find, with regret but no hesitation, that Mr. Gudefin – the Geogas trader whose trade this was and who gave oral evidence at trial – perhaps with others at Geogas, went to some lengths to conceal this fact from Borealis. Instead of claiming from Sunoco, the supplier to Geogas in respect of this water content, Geogas arranged for the removal of much of the water and went to the lengths of procuring the issue of new bills of lading for a reduced cargo quantity. It subsequently transpired, to my mind, revealingly, that the water had a pH of 2 – 2.2 Of this, Borealis was at all material times unaware but the evidence does not permit any conclusion as to whether Borealis’s actions would have been different had it known.

Borealis was aware of the levels of oxygenates in the goods, a matter of some significance and considered further below. The vessel’s tanks contained 225ppm, 135ppm, 200ppm and 130ppm respectively. The average oxygenates content of the goods overall was some 172.5ppm against a contractual specification of nil and Borealis’s internal specification limit of 200ppm. In the event, Borealis accepted the goods.

Queries as to the goods met with unhelpful responses from Geogas. So, on the 12^th September, 2003 (the morning of discharge), when Borealis expressed some concern as to the quality of the goods, Geogas replied, reassuringly, that the two previous cargoes on the vessel had been field grade products. An internal Geogas communication of the same day remarked “Luckily water was no[t] reported”. Subsequently, on the 24^th September, after Geogas had reserved its rights, Mr. Gudefin responded, inter alia, that Geogas’s suppliers confirmed “that nothing abnormal that could have contaminated the cargo happened during loading operations”. He remained, in my judgment and despite the answer he gave in cross-examination, most anxious to conceal the fact that there had been an unusual quantity of water in the cargo when first loaded.

Pulling the threads together, as at the 12^th September, 2003, Borealis was wholly unaware that the goods with which it had been supplied were contaminated with fluorides. Nor was there any reason why Borealis ought to have been aware that the goods might have been so contaminated. To my mind, this is an important factual consideration when addressing the Geogas secondary case.

(3) The pH alarm: Though there was some resistance from Geogas, there was overwhelming evidence that the essential purpose of the pH alarm was to protect downstream carbon steel plant and equipment from corroding weak acids (e.g., carbonic acid) naturally occurring in the product stream. This alarm system was not designed or intended to deal with an unexpected contaminant in the feed such as HF.

Mr. Harris is a Sales Director of Shaw Energy & Chemicals Group, a leading engineering and design contractor and the designer of the plant. Mr. Harris is also the Borealis account manager. Mr. Harris accepted in cross-examination that the pH alarm would detect the presence of any acid in the system and that the injection of ammonia – the anticipated response to a fall in pH below desirable levels - would (no doubt depending on quantities, proportions and the like) neutralise acids generally. As I understood his evidence, he did not, however, depart from the following observations in his witness statement:

“ The pH control system in process systems such as that existing at the …plant, is designed to counter organic acids that may form in the furnace outlet stream during normal operating conditions…..The pH control systems are not designed to counteract hydrofluoric acid as this is not an acid that one would expect to encounter in an ethylene plant….”

Mr. Hedvall was, in 2003, the Production Manager of the plant. His evidence was that the pH alarms were unreliable and, based on the information available to Borealis in 2003, there was no need for immediate action when the pH alarm sounded. As he put it:

“ ….this is a slow system so that there is no high priority on this alarm. …we know from experience that we can operate this system for a long period with low pH so there is no need for immediate response….”

Mr. Hedvall said in a witness statement that though, ordinarily, manual laboratory samples were not undertaken over the weekend, where pH remained low for “hours” (as here), he would have expected the operator to make a particular request for an additional manual sample on Sunday 14^th September; he surmised that no such request had been made because it was overtaken by events. The ammonia injection system had been designed for weak acids. Having regard to those acids, low pH was not seen as a problem.

“ Based on the experience, I mean, we have operated this plant since 1970 and low pH has never been a problem until September 2003. So based on the knowledge we had [in] 2003 I would say that the operators did not see low pH as a problem.”

Mr. Andersson was a control room shift supervisor at Borealis, from 1989 to 2007; he was not on duty on the 12^th September when the pH alarm was triggered (but came on duty subsequently). The tenor of Mr. Andersson’s evidence here was clear. The purposes of the pH alarm was to alert the operator to the pH being lower (and though he did not say it, higher) than recommended. However, a low pH was not harmful for the equipment and so was not a “big deal” for Borealis. They had operated with a low pH for a long time previously; in his view, there was no need to react promptly to the pH alarm sounding. In response to the suggestion that a strong acid had come into the system, he said this:

“ In our knowledge there was no strong acid. We didn’t know anything about that. We only knew …that the system is constructed for a slow build-up of acid in the process. This is a low capacity ammonia system. It has no big capacity to take care of anything larger than that.”

The pH alarm system was there to control the pH for the “carbon steel system”. The pH alarms were in any event unreliable so there was a “general practice” to wait for the regular testing undertaken by the laboratory on Mondays, Wednesdays and Fridays.

Prof. Davies was called by Borealis to give expert evidence in the discipline of chemical engineering. He had visited the plant. He pointed out that a fall in the pH level, even a rapid drop, while indicating the presence of an acid in the system would not tell the operator what type of acid it was. It was his opinion:

“ ….that the pH system….was designed and the introduction of ammonia was designed to neutralise acids that have formed from oxygenates that appear in the feed…..

Carbonic acid is a weak acid. So we are dealing with a system that has been principally designed to remove CO2 in the form of its weak acid in aqueous solution…..

So we have a system designed to remove weak acids…..

….downstream of the heat exchangers and before the compressors ….carbon steel is used and is attacked by weak acids. The reason why the ammonia system is there in the first place is to protect the downstream plant, it is as simple as that.

….the corrosion rate of …[these]…weak acids is very low. So there is no real urgency in adjusting the pH…..”

Pulling the threads together, on the evidence:

Mention has already been made of Mr. Lumley, an expert called by Geogas. Significantly, in cross-examination, he accepted a good deal of this evidence. As fairly summarised by Ms. Buehrlen:

“ The purpose of the pH system, ……the fact that the pH system was itself a low priority alarm, the fact that it was not designed to deal with contaminants in the feeds such as fluorides, ……the experience of how the plant operates were all accepted by Mr. Lumley as relevant factors when considering Borealis’ response to the pH alarm and drop in pH…. ”

Mr. Lumley did, however, place much emphasis on the following provision found in the Borealis Operations Handbook (“the Handbook”):

“The pH value must be maintained at 5.5 – 6.5. At a higher pH value emulsion problems may arise in D-1681, and at low PH’s corrosive attacks on the material increase.”

In the absence of other written procedures for the plant this provision was to be followed; it was not for the operators to go “off-piste” and ignore the pH alarm. The plant was to be operated within strict guidelines; it was dangerous to do otherwise.

With respect, I appreciate the force of Mr. Lumley’s reasoning on this part of his evidence. But I think it needs to be kept in perspective – a perspective provided by the context already outlined. The provision in the Handbook does not, for instance, say anything as to the necessary speed of response, should the pH level move outside the relevant parameters, or as to the prioritisation of the pH alarm. It is additionally of note that this provision cautions both against pH levels above and below the recommended levels; a matter to which I shall briefly return later.

For my part, therefore, I remain very much of the view that the purpose of the pH alarm and the operating experience of those at the plant are significant and material factors when considering the Geogas secondary case and its criticisms of Borealis’ conduct.

(4) The sounding of the pH alarm: Against the background already outlined, I come to the sounding of the pH alarm at 15.20 on the afternoon of the 12^th September.

The pH levels are a good place to start. It will be recollected that discharge of the goods commenced at 13.20 on the 12^th September. Between 12.00 and 14.00, the pH levels were a little above 6. However, between 15.00 and 20.00, the pH levels dropped considerably.

Surprisingly, in my view, Borealis did not call either of the operators who had been on duty when the pH alarm was triggered (or, for that matter, the supervisors then on duty). The best explanation was that neither of the operators any longer worked for Borealis. Even that explanation does not altogether satisfy concerns in this regard, in particular as to the reason why no witness statements were available. Mr. Ashcroft criticised Borealis with regard to the absence of the operators, in my judgment, rightly. He submitted, again rightly, that I should be cautious as to the explanations from others as to what was or might have been done or left undone by the operators; I agree. As canvassed at the trial, it also seems right to me not to speculate as to what the operators might have said, had they given evidence. I must proceed on the evidence available.

All this said, the lacuna left by the absence of the operators and supervisors (or any witness statements from them) is at least in good measure filled by the extremely helpful “Root Cause Failure Analysis Report”, an internal Borealis document, prepared after the incident and in its final form dated 18^th December, 2003 (“the Report”). The frankness of the Report in criticising various of Borealis’ own procedures is commendable and adds to its weight – such frankness doubtless aiding the manifest objective of the Report, namely, to avoid a recurrence. Furthermore, the Borealis lead investigator into the incident and (with a team of others) responsible for the preparation of the Report, was a Mr. Arvidsson. He did attend to give evidence. As both his evidence and the Report made clear, in the course of preparing the Report interviews were conducted, including of the operators. It follows that, in part, at least, the Report reflects what the operators were saying at the time.

Before turning to the substance of the report, I should mention two preliminary matters:

Coming to the substance of the Report but confining myself to matters relevant to the Geogas secondary case, the Executive Summary records that approximately 1 hour 20 minutes after discharging commenced, the pH of sour water in D-1681 “dropped rapidly” from 5.5 to a range of 3.5 – 3.7. The Executive Summary continued:

“ A pH alarm is present with the LLA set at 5.0. No actions were taken. There was no knowledge at production department or among operators about the potentially damaging effects of hydrofluoric acid on titanium. The alarm was later tested and found to be working.

The low pH condition persisted for approximately 59 hours.”

The Executive Summary recommended that:

“ Understanding about the potential corrosive effects of low pH, and the need to react to low pH alarm, have to be communicated to all operators.”

In the section of the Report entitled “Events and Conditions”, the drop in pH from 5.5 to a range of 3.5 – 3.7 was considered. Laboratory samples taken before and after the incident indicated that the pH meter was relatively accurate. While there was no alarm log for the pH alarm system, this passage of the Report was able to say the following:

“ No actions were taken to increase the pH (e.g., increase NH3). Causal Factor 4

◦

Historically the unit has run without NH3 with a pH of ~ 4 for extended periods without damage.

◦

In earlier years, the pH meter was unreliable and more emphasis was put on lab sampling of pH.

◦

Lab samples are not routinely taken during the weekend.

◦

……..

◦

There has been no experience to show that low pH is harmful to titanium. Causal Factor 5”

As Geogas sought to pray it in aid (discussed below), I should record the entry for 13.00 on the 18^th September. A request had been made by the investigation team and:

“ pH trend was noted for the first time”

Under the heading “Observations and Conclusions”, the Report has this to say:

“….pH is controlled by manually adjusting NH3 injection rates to the tower overhead. The target range for pH is 5.0 and 7.0. Historically the primary purpose for pH control has been to reduce corrosion of carbon steel in the tower overhead, piping and heat exchanger shells. Under normal operating conditions the primary corroding species is carbonic acid formed from a combination of steam condensate and CO2.

Since the bundle metallurgy was changed to titanium there have been a number of incidents when NH3 injection has been interrupted and pH levels have been allowed to fall without neutralization, in at least one case for a period of 2-3 weeks. The resulting overhead pH levels have typically fallen to around 4.6. It was determined that the corrosion rate of carbon steel under these conditions is relatively low and, during routine inspections, no visual indication of titanium has been observed as a consequence of intermittent operation below a pH of 5.0. Nevertheless, long term exposure of carbon steel to lower levels of pH has been deemed to be inadvisable. Until this incident, there have been no pH excursions below 5.0 during the past two years.

pH is monitored with an on-line pH meter located in the sour water distillate drum (D-1681). When the on-line pH meter was originally installed it was relatively unreliable. Laboratory samples are taken three times a week to verify pH meter readings. Comparison of recent laboratory measures of pH with the on-line meter has shown that the meter currently tracks pH reliably. However, appreciation of this improvement in reliability is not shared by all operators who are responsible for monitoring pH. ”

Later in the Report, there were these observations, linking the corrosive attack to the contaminated goods:

“ The drop in pH ….began approximately 1 hour after ….[unloading began]…of a shipment of N-butane from the Henning Maersk into the UC731 storage cavern. The cavern inlet from the dock is in close proximity to the unit feed pump suction line. Past experience has shown that the effects of new feed are typically observed in the unit in approximately 1 hour from the time that unloading begins. This observation is viewed as further confirmation that corrosion of the E-1651 and E-1656 is related to contaminants in the butane that was shipped on the Henning Maersk.”

As already highlighted, two “Causal Factors” are here relevant and are related. These were:

“ 4. No action was taken in response to dropping pH.

5.

Experience has never indicated that low pH is potentially damaging to Ti. ”

The Report noted the on-line measure of pH had fallen below the low level alarm of 5.0 and remained there for over 40 hours. After 30 hours at a pH of approximately 3.6, the first (gas) leak from the E-1651 exchangers was detected. The Report remarked that there was “no evidence that any attention was paid to this parameter” (i.e., the low level pH alarm). The Report went on to say:

“ The investigative team interviewed a number of operators and other members of the production staff to gain a better understanding of this oversight. Our objective was to determine the extent to which this could be explained by past operating experience with pH, lack of knowledge and understanding of the potential damaging effects of low pH, or simple failure to follow reasonable standard operating procedures.”

The Report indicated that the findings from these interviews had been set out in the Events and Conditions section – already recorded above.

The Report’s own conclusions in this regard were as follows:

“ ….operating experience for the past 20 years suggested that pH was a relatively unimportant operating parameter. It was certainly true that essentially no one in the production department was aware of the potentially damaging effects of hydrofluoric and/or hydrochloric acid on titanium or even for the potential for HC1 in the primary fractionator tower overhead.

This said, it is apparent that the low pH alarm was functional, yet no additional samples were sent to the lab to confirm that the low readings were real. Also, no adjustments were made to NH3 injections which may have mitigated the effects of the HC1.”

Finally, the Report dealt with Causal Factors 4 and 5 as follows:

“ Root cause category 2a) Lack of knowledge and skills –

Historical experience and lack of familiarity of the impact of low pH on the corrosion of titanium, resulted in little or no attention to pH levels below the alarm level. ”

To my mind, on the evidence before the Court, no action had been taken in respect of the pH alarm prior to the gas alarm sounding essentially for the reasons set out in the Report and noted above. (Understandably, after the gas alarms sounded, other concerns predominated.) Those reasons related to the experience of the plant and the purpose of the pH alarm. It is unsurprising that with an eye to the future – whatever the outcome of this litigation – the authors of the Report should recommend improved understanding of low pH, its possible consequences and the need to react to the low pH alarm.

For completeness, in the light of the Report’s conclusions, I accept that the low level pH alarm was in working order on the 12^th September. In his witness statement, Mr. Andersson described what happens when the alarm is triggered:

“ The low level pH alarm sounds when the pH in the D1681 distillate drum falls below pH5. When the alarm sounds the alarm indicator light on the panel above the process computer begins to flash. The operator acknowledges the alarm by pressing the flashing light button. Once acknowledged the alarm sound and flashing light will stop. However, the indicator light will stay lit until the pH returns above 5. ”

If only to stop the sound and flashing light, the probabilities are that an operator “acknowledged” the alarm by pressing the flashing light button. On the available evidence, the operator did no more than that, for the reasons already discussed. Against this background, I am unable, with respect, to attach to the 13.00 entry in the Report, under “Events and Conditions”, the significance for which Mr. Ashcroft contends. Even assuming that the particular entry is not simply referring to the work of the investigators, I do not think that any failure to “note” the pH trend advances the argument – it is no more than part of the overall picture already outlined.

(5) The high point of the Geogas secondary case: The high point of the Geogas secondary case may be seen as the level of similarity in the evidence of Mr. Harris, Mr. Lumley and Prof. Davies as to the reactions they would have expected from the Borealis operators and management to the drop in pH levels on the afternoon of the 12^th September. On the basis of this evidence in particular, Geogas pressed its secondary case; the Borealis reaction to the drop in pH levels was indefensible; it was unreasonable; it broke the chain of causation; had the reaction been as contemplated by these witnesses, physical damage and consequential loss would have been very substantially reduced.

Mr. Harris (to whom reference has already been made) said that the sudden drop in pH levels was “an indication that something was wrong”. On the assumption that there was nothing wrong with the measuring instrument, in his view, the sudden drop indicated to the operator that there was a strong acid in the system. If there was a strong acid in the system, then that called for an immediate laboratory analysis on the water in the system.

As I understood Mr. Lumley’s evidence, in its final form on this part of the case, his views may be summarised, by reference to a passage in one of his reports. Once the pH alarm had been triggered (at 15.20 on the 12^th September) the time line should have been as follows:

“ 1. The first step should have been to increase the ammonia flow to the maximum. This would have been a safe procedure with no negative consequences.

2.

Within 4 hours [so, 19.20], the pH signal could have been confirmed after drawing a sample from the plant, checking with pH indicator paper and confirming with the laboratory. Within this period as the ammonia had been increased to the maximum, the pH record should be monitored to check the effect of the additional supply

3.

A further allowance of 6 hours [01.20 on the 13^th] is made for Borealis to confirm the source of the acidity. Time is also required for discussions with the supervisor and management to obtain approval to change the feed. I regard this allowance as generous to Borealis, because…the operators, supervisors and management should quickly have appreciated that the HM cargo was the cause of the drop in pH.

4.

A further 12 hours is required [13.20 on the 13^th] to stop LPG and change over to alternative feed safely.

5.

Continue to operate the plant on alternative feed from 13.20 hours 13^th September….at full capacity, not 15^th September….

Therefore, I estimate, the contaminated feed through the heat exchangers should have ceased within 22 hours of the triggering of the pH alarm.”

(Times in italics in square brackets are added.)

It may be noted that in taking 12 hours as the time to stop the LPG and change to an alternative feed (step 4), Mr. Lumley relied on Mr. Hedvall’s evidence to this effect, rather than Mr. Andersson’s 16 hours.

Turning to Prof. Davies, he accepted that he did not know what Borealis’ procedures were and was surprised that (as he put it) he had not got to the bottom of the reasons why the operators had not acted in response to the pH alarm. He also said that one “wouldn’t be running…to the lab every time it went down below 5”. Subject to the caveat, twice underlined in the course of his cross-examination, that the “scenario” was “artificial”, Prof. Davies’ evidence as to timings and related matters on the 12^th – 13^th September was along the following lines:

(6) Discussion: Starting point: I start with the evidence characterised as the high point of the Geogas secondary case.

With respect, I am not sure that that part of Mr. Harris’ evidence relied upon by Geogas took the matter significantly further. I have no doubt that Mr. Harris was doing his best to assist the Court. It would, moreover, be difficult to quibble with his view that the sudden drop in pH levels gave “an indication that something was wrong”. But I am unable to accept his further answer that (assuming there was nothing wrong with the measuring instrument) the sudden drop in pH levels indicated that there was a strong acid in the system and – hence – the need for an immediate laboratory analysis. I accept in this regard the evidence of Prof. Davies that a drop in pH levels (assuming no instrument fault) indicated acid in the system but not what type of acid.

I turn to Mr. Lumley. His evidence came under severe attack from Borealis. Insofar as that attack included the suggestion that in part he had set out to mislead the Court, I should at once and in fairness to Mr. Lumley make it clear that I reject it. I do not at all think that Mr. Lumley was a dishonest witness. Nor was his evidence without a number of points of considerable force (one of which has already been mentioned). There were, however, a number of features of his evidence which do prompt very considerable caution when that evidence comes to be weighed; these features include the following:

Against the background of these criticisms, Mr. Lumley’s evidence, had it stood alone, would have furnished a weak foundation for the Geogas secondary case. In reality, the force of the Geogas secondary case, lay in the evidence of Prof. Davies, upon which Geogas placed understandable reliance.

Prof. Davies was a witness of manifest independence and, within his discipline (chemical or process engineering) undoubted expertise. To my mind, it was his evidence which gave the Geogas secondary case such credibility as it had.

Even, however, in connection with an expert of the calibre of Prof. Davies, it is necessary to bear in mind the following:

Unreasonable? Having regard to the evidence of Prof. Davies, all the more so, when considered in conjunction with the Report, I am readily satisfied that the Borealis reaction to the sudden and substantial drop in pH levels on the afternoon of the 12^th September left considerable room for improvement. On any view, this reaction fell well short of best practice. The recommendations in the Report were undoubtedly apposite.

But was the Borealis reaction unreasonable? I confess that I did not find this an easy question to resolve. But, on balance, on the whole of the evidence and notwithstanding Mr. Ashcroft’s advocacy, I do not think it was.

In my judgment, it is essential to keep uppermost in mind my earlier conclusion: namely, that Borealis did not know and could not reasonably have been expected to know of the kind of acid introduced into the system by the Geogas breach of contract and the nature of the threat it posed. Again, as already concluded, the pH system was not designed or intended to deal with acids of such a nature. It is correct that the sudden and considerable drop in pH and the triggering of the pH alarm indicated an acid in the system; but, strikingly, had it been an acid of the kind typically encountered by the plant, nothing untoward would have resulted from the operators waiting until the routine samples were taken on the Monday (the 15^th September). In the light of the experience of those at the plant, this would not have been a problem. That the operators could have done better, as indeed they did when pH dropped significantly after discharge of the Stanlow cargo (a cargo mentioned above), is one thing – and as already underlined amply justifies the conclusions in the Report. But, against this background, to characterise as unreasonable the operators’ failure to achieve best practice, seems to me a step too far. It should also be noted that operators and supervisors (including Mr. Andersson) on the shift/s subsequent to that when the pH levels first fell, reacted in the same manner as those on the initial shift. It might be said that the focus must then move (away from the initial operators and supervisors) to the training given to operators and others in the production department in the plant. If so, however, the same conclusion is reached: until the occurrence of this incident and in the light of experience hitherto, Borealis had no reason to anticipate that low pH would or could have consequences of this nature.

For my part, I conclude that it was not unreasonable of those at the plant not to take the steps on Friday 12^th September or Saturday 13^th September, canvassed in the evidence of Mr. Lumley and Prof. Davies when dealing with time lines. I am fortified in that conclusion and in preferring the evidence of those with practical experience of the plant, by Prof. Davies’ own concerns underlined by his description of the time lines “scenario” as “artificial”. In my judgment, it would not have been unreasonable to wait for routine sampling on Monday 15^th September, or, given Mr. Hedvall’s evidence (recounted earlier), at least Sunday 14^th September. I take no more time over the difference between the 14^th and 15^th September, as, in my view, it is academic; in any event, by the 14^th September, such questions as to sampling had been overtaken by events.

This conclusion is sufficient to dispose of the Geogas secondary case whether advanced in terms of a break in the chain of causation or as a failure to mitigate.

With specific regard to the question of mitigation, the linkage between the knowledge of a claimant and the reasonableness (or otherwise) of its conduct is plain. As illustrated by the judgment of Nolan LJ in Schering (supra), at p.16, the conclusion that there had in that case been a failure to mitigate was premised on the plaintiff’s:

“ …actual knowledge of the danger to which the defective heat sealer gave rise.”

There was no such knowledge in this case, arising from the drop in pH on the afternoon of the 12^th September – nor, in fairness to Borealis, ought there to have been.

If unreasonable, was the chain of causation broken? In case I am wrong and in any event out of deference to the arguments advanced, I go on to consider whether the Borealis reaction to the drop in pH levels and the triggering of the pH alarm, if unreasonable, broke the chain of causation. Here, without hesitation, I conclude it did not.

First, I did not understand the Geogas secondary case to be advanced on the basis that the Borealis reaction was reckless (as distinct from unreasonable). For the avoidance of doubt, any such case would have been hopeless. The absence of recklessness is itself a significant conclusion, favourable to Borealis, even if (as discussed above) it is not by itself sufficient to dispose of the Geogas secondary case.

Secondly, on the facts already found, while this is not a “horns of a dilemma” case, as in The Stork (supra), it is also far removed from the facts of Lambert v Lewis (supra) – where the defect in the coupling was, so to speak, staring the farmer in the face – and the facts of Schering (supra), where there was actual knowledge of the danger to which the defective heat sealer gave rise. Nor are the facts here close to those of The Spontaneity (supra), where the negligent re-floating was clearly separate, both in time and in terms of seamanship, from the earlier grounding. While Borealis did know of the rapid and substantial fall in pH, the facts here are closer to those of County Ltd v Girozentrale (supra); on the 12^th September, Borealis was entitled to start from the premise that it had been supplied with a high quality cargo free of contaminants, whereas the reality, unknown to Borealis, was very different.

Thirdly, the answer ultimately lies not in authority but in a fact sensitive inquiry. Approached as a question of fact and degree, I cannot begin to say that any failure/s on the part of Borealis, “obliterated” or destroyed the “causative potency” of the anterior Geogas breach of contract. It seems plain to me that the Geogas breach of contract remained, at the least, an effective cause of the Borealis loss and damage, regardless of any failure/s on the part of Borealis. Though Borealis knew or ought to have known of the presence of an acid in the system, it remained in the grip of an unknown breach giving rise to an unknown danger.

Would any loss have been avoided? The conclusion that the chain of causation was not broken is itself sufficient to dispose of the Geogas secondary case. In any event, however, I would have been unable to accept the Geogas secondary case because, reasoned through, I do not think the steps it advocated would have avoided any loss, physical or consequential. I proceed by way of an examination of those steps, as they broadly appear from the evidence of Mr. Lumley and Prof. Davies.

I consider, first, the adding of ammonia. I have no hesitation in preferring the evidence of Prof. Davies to Mr. Lumley; there was, simply, no or no proper basis for immediately increasing the flow of ammonia to the maximum; any additions would or would likely have been incremental. The fact that a pH of more than 7 was a concern would also have told against an immediate increase of the flow to the maximum. Thereafter, the point is to a degree double edged, for both parties. The more the addition of ammonia increased the pH, the more likely it would be that to some extent, the addition of ammonia masked the presence of the contaminants in the system. The less it did so, while underlining that the addition of ammonia by itself would not have achieved significant results, the more it would have prompted further consideration of the situation.

In my judgment, the addition of ammonia would not have had any significant effect in reducing the damage suffered – especially where the damage most mattered, in the area of the primary heat exchangers. Having regard to the evidence of Dr. Brisdon (an expert called by Borealis in the field of chemistry) and Prof. Davies, the relevant reaction depicting the addition of ammonia to hydrogen fluoride, producing ammonium fluoride is: HF + NH3 = NH4F; however, this reaction cannot take place at temperatures above 100ºC as ammonium fluoride disassociates at such temperatures. Accordingly, no additional amount of ammonia could have protected the primary heat exchangers (where the temperature of the water entering the exchangers under pressure is about 105ºC) from the corrosive effect of the HF. As Prof. Davies put it:

“ ….the HF cannot be neutralised at the entrance to the [primary] exchanger.

It doesn’t matter how much ammonia you put in because it is not going to work if it is unstable.

If that is the case you must always have the risk of corrosion at the entrance to the [primary] exchangers…if HF is present.”

In passing, it may be thought that this difficulty, by itself, serves to illustrate the fact that the pH alarm system was not designed for dealing with the problems posed by the introduction of an acid such as HF into the system.

If an addition of ammonia had “neutralised” HF it would have done so in the area of the secondary exchangers (where the temperature range was some 58 - 35º C), producing some reduction in damage – but also increasing to some extent the pH and, in so doing, masking the corrosive attack in the areas of the highest temperatures. That said, though the matter is to an extent speculative, on all the evidence, my inclination is that an increase in ammonia would not have significantly increased the pH level.

Pulling these threads together, while the addition of ammonia would have been an entirely understandable step to take in response to a fall in pH, I conclude that it would not on its own have averted the physical damage suffered at the plant. Its contribution to the Geogas secondary case must be assessed on effectively the converse assumption; namely, that the addition of ammonia would not have had a significant effect in raising pH and so would have prompted further consideration as to the nature of the problem.

I turn next, therefore, to the position reached when it was ascertained that the pH readings were accurate and that despite the addition of ammonia up to the maximum there was no discernible improvement in pH levels. The conclusion would have been that there was an acid in the system and, logically, the inquiry would have been as to the source of the acid. I am not sure that the connection to the goods (and hence the feedstock to the cavern) would have been quite so straightforward and quick as argued by Geogas; nor for that matter do I think it would have been as complex as suggested by Borealis, in the light of the observations in the Report (set out above) as to the typical timescale for observing the effects of new feed.

If, however, a connection had been made between the low pH readings and the goods, then, in agreement with Ms. Buehrlen, I do think that considerable time would have been taken thereafter, while the acidity was attributed to the level of oxygenates in the goods. It is to be remembered that the weak acids normally present in the feeds and which the pH system is designed to detect are a consequence of the presence of oxygenates in the feeds. Both Prof. Davies and Mr. Lumley stated or agreed in cross-examination that it would have been reasonable for Borealis to have linked the drop in pH with the high oxygenates content of the goods. As already foreshadowed, Prof. Davies would have reached that conclusion himself. In re-examination, Mr. Lumley, reminded that the average oxygenates content of the four tanks of the vessel was less than the Borealis 200ppm acceptance criterion – and that therefore the cargo had been approved for unloading – retreated somewhat from his answers in cross-examination.

A clear perspective is necessary. In my view, Geogas was right to insist that the average level of oxygenates in the goods was within the Borealis acceptance parameters (not more than 200ppm) – otherwise discharge of the goods would likely not have taken place. That said, Borealis was right to underline that the oxygenates level was high - in that it was significantly above the norm of cargoes for feedstock deliveries received by Borealis at the plant over the period 14^th January - 12^th September, 2003. A study of the cargoes delivered over that period shows that the goods contained the second highest average level of oxygenates, exceeded only by the Stanlow cargo (referred to above). The Stanlow cargo had oxygenates ranging from 200 – 229ppm; the range for the goods was 130 – 225 ppm; the average of all deliveries over this period was 59ppm (including the Stanlow cargo and the goods) and 31ppm for the remaining 39 deliveries if the Stanlow cargo and the vessel’s cargo are excluded. Moreover, as canvassed earlier, the presence of these oxygenates in the goods disclosed a breach of contract.

I do not overlook Mr. Ashcroft’s submission, based on the reaction of witnesses from the plant (Messrs. Hedvall and Andersson) - that, as the drop in pH was not explicable by reference to normal operating conditions, it was unrealistic to postulate oxygenates as the likely explanation for the fall in pH. Nor do I overlook the data put forward by Mr. Ashcroft as suggesting that the pH fall – compared to other deliveries – could not be ascribed to oxygenates; in this regard, however, Mr. Ashcroft both accepted that he had not put this data to Prof. Davies and, by way of a qualification, that the data he relied upon could not be simplistically interpreted. I take all this into account. Having done so and stripping the matter of all hindsight, I am of the view that if the problem had been escalated up the Borealis hierarchy (as the Geogas secondary case assumes), so the question of oxygenates would indeed have loomed large. The presence of HF would not have occurred to anyone. They would instead have begun their consideration by reference to the available analyses. On those analyses, the presence of oxygenates would have been conspicuous. Suffice it is to say that some considerable time would likely have elapsed and, to my mind, quite properly, while this topic was canvassed. No doubt in time, the conclusion would have been reached that the explanation could not lie with oxygenates – but time is critical for the Geogas secondary case to have any practical impact.

It only needs to be kept in mind that for the feedstock to be changed decisions needed to be taken that the drop in pH was alarming; that the pH readings were accurate; that the reason for the drop in pH, which additions of ammonia had failed to reverse, was an acid in the system; that the acid was linked to the goods; that the explanation did not or could not be assumed to lie in oxygenates. All this would have involved the matter moving from the operators to the supervisors, via a special request to the laboratory and then further up the organisation chain to plant management. The decision to change feedstock was not one to be taken lightly.

For my part, I regard the notion that within 10 hours (or anything like it) of the pH alarm being triggered a decision would have been taken to change the feed as arbitrary and artificial in like measure. Bearing in mind that the titanium tubes needed replacing once 20% damaged, it can readily be seen that the Geogas secondary case (even if otherwise well-founded) was unlikely to avoid any or significant physical damage. Moreover, as perforations were present in the tubes by or at 20.00 on the 13^th September – hence the sounding of the gas alarm – and as the minimum time for changing the feedstock was 12 hours, it seems to me that the Geogas secondary case could not have avoided physical damage to the plant unless a final decision to change the feedstock had been taken by 08.00 on the 13^th September. But that timescale, for the reasons already given, I regard as unrealistic. Against this background, Mr. Ashcroft was, in my judgment, realistic when he accepted in his closing oral submissions that:

“…the meat of this secondary case….as the evidence has turned out, is not so much in relation to the costs of the physical repairs….”

The “real meat” of the Geogas secondary case, Mr. Ashcroft went on to contend, related to the substantial loss of profit claim for September 2003. I shall deal with that in a moment but I must first deal with the Borealis case as to E and G units which sustained (very limited) damage, having come on line at about 03.00 on the 15^th September. Here, Mr. Ashcroft submitted that if there was force in the Geogas secondary case other than in his time line argument, then such loss as related to these units would have been avoided; they ought not to have been exposed to contaminated feed at all. For the reasons already given, if I am right about Borealis’s conduct not being unreasonable and not breaking the chain of causation, this submission too must fail. But in any event, as it seems to me, if the matter is only to be considered once the gas alarm had sounded, it is unrealistic to castigate as unreasonable the steps taken by those at the plant after 20.00 on the 13^th September. No such case was developed on the evidence and, understandably, at that time, attention had shifted to other matters.

I come finally in this chapter to the submission that the September 2003 loss of profit claim should reasonably have been avoided. This argument was advanced by Mr. Ashcroft in these terms:

“ …..there is no reason to conclude that there were any holes in the heat exchanger tubes any significant time before the gas alarm was triggered. If the feeds had been changed before there were holes in the tubes, the probability …..is that production would have continued – using perfectly acceptable uncontaminated feeds which would cause no further damage to the tubes – for the balance of September…..

……..

If they were not leaking ….there would have been no reason to stop.”

Developing this submission, Mr. Ashcroft contended that the entire September 2003 loss of profit claim could have been avoided if he was right as to a change of feed within 22 hours of the triggering of the pH alarm (i.e., by 13.20 on the 13^th September). Even if I was against him on that but concluded that there should have been a change of feed by (say) 18.00 on the 13^th September, then he submitted that the same conclusion should be reached. It could not be assumed that the tubes were significantly holed before 20.00 on the 13^th September – otherwise the gas alarm would have sounded earlier. Once the feed had been switched, there was, he said, no reason to suppose that the tubes would have been perforated, at 20.00 on the 13^th September, or at all.

With respect, I am unable to accept this submission. First, for reasons already fully discussed, I am not persuaded that there was unreasonableness, still less unreasonableness breaking the chain of causation, prior to 20.00 on the 13^th September. Secondly, I think the case is speculative as to whether the perforations would have been avoided if there had been a change of feed close to 20.00 on the 13^th September. Thirdly and in any event, Ms. Buehrlen’s succinct response was as follows:

“ To suggest that the probable consequence of discovering that your plant has been subjected to a strong acid for 22 plus hours and to then continue at full production ignoring that fact, not even looking at your equipment to discover the level of damage….we say cannot possibly be the probable consequence of ceasing to use the LPG shortly before the perforations occur.

One could imagine the scenario in which Borealis were to conduct itself like that and what would be said if two weeks later there was suddenly a major incident arising out of the fact that these pipes have simply been allowed to stay in situ with nobody doing anything about it.”

Suffice to say that I entirely agree; notably, this case was not explored with the Borealis witnesses in cross-examination.

It follows that I am unable to accept that the steps proposed in the Geogas secondary case would have successfully avoided loss, whether physical or consequential. I accordingly dismiss the Geogas secondary case. It follows that Borealis is entitled to recover the September 2003 loss of profit, agreed at €2,902,237.

In the event, this is an Issue falling within a narrow compass and which can be taken very briefly indeed.

As ultimately put by Mr. Ashcroft in his written closing submissions, the Geogas case on remoteness was as follows:

By way of explanation as to the September 2004 loss of profits, this head of argument relates to the reinstalling of heat exchanger units E and G. It will be recollected that these units were brought on-line from 03.00 on the 15^th September, 2003. As a result of the incident, they were disconnected in March 2004 for inspection and were reinstalled – put neutrally – at a time when other works were done in September 2004. The argument here is purely one of remoteness; submissions that this loss is not recoverable by virtue of the Geogas secondary case have already failed; submissions that this loss is not recoverable because other works were in any event undertaken in September 2004 will be dealt with below. The amount in question is relatively small, some €72,000.

As to losses due to the non-availability of the cavern, these relate to two distinct, if not unrelated, heads of claim, amounting in total to €938,000 (agreed as a figure). The first, concerns the loss of use of the 32,000mt odd of butane stock that was in the cavern. The second, arises out of the inability to buy and crack butane after the existing butane stock had been used up, because of dealing with the heel of the contaminated product. The focus here is on remoteness only. The Geogas submissions as to a failure by Borealis to mitigate are dealt with below.

With great respect to Mr. Ashcroft, these valiant submissions lacked substance. As to the lost profits in September 2004, it was “perfectly likely”, as Ms. Buehrlen submitted, that if a contaminant was introduced into the plant that the equipment would need to be inspected and that production would be lost when taking such equipment off-line and bringing it back on-line. Whichever test of remoteness of damage is applicable (see above), Geogas seems to me to be plainly on the wrong side of the line. Moreover, as a question of remoteness, it cannot matter whether the E and G heat exchanger units were reinstalled in March or September 2004.

Turning to the losses due to the non-availability of the cavern, again, with respect, this does not seem to give rise to a question of remoteness at all – whichever test is applied. The introduction of goods heavily contaminated with fluorides was obviously likely to result in contamination of and damage to the pre-existing contents of the cavern. Insofar as Mr. Ashcroft suggested that this loss arose from Borealis’s inability to buy in replacement butane in September 2003 without loss, it seems to me that the point becomes no stronger; the type of loss was plainly likely; there is no requirement that liability depends on Geogas foreseeing or taking responsibility for its precise manifestation. Equally, in my judgment, the introduction of heavily contaminated goods was likely to result in the loss of use of the cavern while the contamination was dealt with. Even assuming in Mr. Ashcroft’s favour that, as he put it, the extent of the loss involved a “complex equation”, that is neither here nor there. For completeness, I do not think that The Forum Craftsman [1991] 1 Lloyd’s Rep. 81, to which Mr. Ashcroft referred, advanced the argument.

No further elaboration is required. The remoteness defence fails.

In broad terms, Mr. Ashcroft’s attractively presented argument was that Borealis had unreasonably delayed replenishing the cavern and that Geogas should not be liable for losses resulting from such errors or failings on the part of Borealis. The Geogas case was that the cavern should have been ready for use by the 1^st February, 2004, or such intermediate date as the Court thought appropriate, whereas it was in fact only ready on the 21^st March, 2004. The amount in dispute was some €238,000 at most – and less if it should be held that there had been a failure to mitigate but with reference to a date later than the 1^st February, 2004.

In considering this criticism of Borealis it is necessary to keep well in mind what Borealis did do. The contents of the cavern were sold to Statoil on the 5^th January, 2004. Originally, this disposal was one of the main planks of the Geogas case that Borealis had failed to mitigate its loss; but, as foreshadowed, that challenge was later abandoned. The loading of the cargo pursuant to this contract was not ultimately completed until the 21^st January, 2004. One day before that, on the 20^th January, 2004, Borealis concluded a contract for the import of 3,300mt of butane (+/- 5%), at a price of US$327 per mt. This shipment was delivered on the 29^th January, 2004 (3,266mt to which 871mt of propane were added). Understandably, Borealis was concerned that the cavern contents (i.e., the heel after the disposal to Statoil) as diluted by this shipment were safe for cracking; vessel and cavern samples were accordingly taken and analysed (externally) by Saybolt. On the 10^th February, 2004, Borealis decided that at least another 4,000mt should be imported to further dilute the contents of the cavern before cracking. Some 14 days followed before, on the 24^th February, 2004, Borealis concluded two contracts. The first, a contract with BP, at US$315 per mt, produced a delivery on the 2^nd March of 5,018mt (to which 1,896mt of propane were added); vessel and cavern samples were again analysed. The second, a contract with Shell, at US$317 per mt, was for delivery of 20,600mt during the period 14^th – 18^th March. On the 9^th March, satisfied with tests of the fluoride content, Borealis proposed to start cracking but with one LPG furnace only; the use of one furnace was a matter of prudence but entailed a loss of profit as the best choice would have been to use two LPG furnaces. Independent advice at the time remained cautious and concerned. On the 12^th March, Borealis did proceed and re-started cracking, albeit using one LPG furnace only. On the 20^th/21^st March, Borealis started using two furnaces, so bringing this chapter to an end.

To my mind, working through this chronology is invaluable. First, even if the notion that Borealis should have bought in one single cargo sufficient by itself to dilute the heel appropriately had practical merit – without the benefit of hindsight – it was in no sense unreasonable not to have done so. On any view, such a purchasing strategy would have involved buying in a larger quantity of butane (whether 20,000mt plus or about 11,000mt of butane, depending on which variant of the Geogas case is under consideration) at a winter seasonal high – a matter evidenced even by the difference between the January and February purchase prices of the cargoes bought in. Such a strategy would have risked increasing the overall loss. Secondly, Borealis had been seared by the experience of the contaminated goods. If its subsequent approach disclosed caution, that was eminently understandable. Thirdly, against this background, I can see no proper ground – whether commercial or technical – for criticising the Borealis approach of replenishing the cavern in stages, having samples independently analysed as it proceeded. Fourthly, I can see nothing wrong with the timing of the first purchase, that contract having been entered into even before conclusion of the loading of the Statoil cargo. It follows that the Geogas submission that the cavern ought to have been available by the 1^st February is, with respect, entirely unrealistic. The mitigation argument, if it is to prevail at all, can only be for a significantly lesser period than that contended for.

Pausing there, I am not sure that these conclusions involve a rejection of Ms. Jago’s evidence – but, for reasons to which I shall come in a moment, if they do then I am not at all deterred. The key passage in Ms. Jago’s evidence seemed to me to be the following:

“ ….during the early part of January, they should have been looking at availabilities of butane to blend with what they could be fairly certain was going to be some heel left there. So by the time they were getting close to completion of lifting of the cargo to Statoil, they should have been close to purchasing a cargo of butane to blend in with what would have been left in the cavern…..

…..

….they would have to wait until they knew what the heel was before they could finally make the purchase, but they could still be determining what cargoes were available prior to that….. ”

With respect, given the timing of the first purchase of butane (the day before completion of the loading of the Statoil cargo), these observations hardly advance the Geogas case that Borealis failed in its duty to mitigate. They would only go somewhere if allied to other criticisms of the Borealis purchasing strategy which I have already rejected.

There are, however, with regret, other reservations I have as to Ms Jago’s evidence. Ms. Jago was called by Geogas as an expert in the fields of LPG trading or broking and as a market analyst. My first reservation goes to some (in fairness to Ms. Jago, by no means all) of her promotional material and other material in the case, which materially misstated her qualifications. So, Ms. Jago was in places described as “Dr. Jago”; but although Ms. Jago had been a doctoral student, she had not submitted her thesis and was not entitled to the title “Dr.”. Furthermore, she was also described as “Dr. Jago LLB” – but she did not have the LLB qualification either; she had a post-graduate diploma in law. Having regard to Mr. Ashcroft’s submissions, I am prepared to accept that these errors came about through lax oversight rather than that she had deliberately set out to misrepresent her qualifications. In the field of expert evidence, however, this remains a matter of grave concern and Ms. Jago must be in no doubt that there should be no recurrence. Secondly, despite the field in which she was called to give expert evidence, it transpired that Ms. Jago had never traded LPG or butane. This lack of experience did not inhibit Ms. Jago from expressing her opinion on a variety of issues which called for such experience. In the light of these reservations, even though Ms. Jago may well have had some experience in dealing with heels and blending, I confess that I would in any event have been unable to place any serious weight on her evidence, at least unless it was agreed or uncontradicted. In the event, for the reasons given earlier, I do not think that Ms. Jago’s evidence takes the present issue any further.

What remains of the Geogas case really comes down to two points: (1) a criticism that Borealis waited until the 10^th February before taking a decision to purchase additional cargo; (2) a failure thereafter to conclude any purchase contracts until the 24^th February.

Pausing here and as foreshadowed, it may be noted that this claim is necessarily restricted to a small sum only. The maximum period in issue would have been some three weeks – the 10-12 day period between discharging the first incoming cargo and determining to order more, coupled with the fortnight’s wait before concluding the second purchase contract (followed, it is to be remembered, by prompt delivery). I am, in any event, inclined to think that those two periods could not have been added together simplistically. It follows that on no view could Geogas realistically have expected the cavern to be ready before the 1^st March, 2004 and any successful claim for mitigation would have been for some reduced period, between a date later than the 1^st March and the 21^st March.

It is next pertinent to inquire as to the evidential basis for the Geogas case of a failure to mitigate – bearing in mind that on this issue, Geogas bears the burden of proof. Nothing in Ms. Jago’s evidence assists on these points. So far as concerns Mr. Lumley, he was asked in cross-examination about Borealis taking until the 12^th March to re-start cracking; to this, he answered:

“…they were slow, they could have been quicker.”

Asked next if he was saying that Borealis had been unreasonable, he gave this reply:

“ No, I wouldn’t say it was unreasonable.”

Plainly, that evidence provides no assistance to the Geogas case.

In fairness to Geogas and Mr. Lumley, it might be said that this area, in part at least, was outside his field of expertise. But be that as it may, it follows that the Geogas case is left somewhat bereft of evidential support.

The reality is that the Geogas case depends on gaps in or inferences to be drawn from the evidence called by Borealis. Mr. Eerola was the senior feedstock trader at Borealis. As to the wait until the 10^th February before deciding to purchase additional cargo, his somewhat speculative answer was that it was linked to the progress made with the first incoming shipment. As to the time taken until the 24^th February to conclude the next purchase contract, he frankly accepted that he could not remember but nonetheless maintained that the second cargo was ordered and received within the 3 week period ordinarily allowed by Borealis for sourcing feedstock.

Somewhat on balance, I am not persuaded by the Geogas case of a failure to mitigate. I do not think that the evidence makes good a case that Borealis acted unreasonably; the facts that there are do not speak for themselves; the inferences to be drawn from the gaps in the evidence do not point sufficiently cogently to unreasonable delay on the part of Borealis. It would not be right to be unduly precise in assessing every step taken by Borealis in dealing with the contaminated cavern and bringing it back into use. I keep in mind – in addition to the authority cited earlier - the trenchant observations of Lord Loreburn LC in Lodge Holes Colliery Company v Wednesbury Corporation [1908] AC 323, at p.325:

“ Now I think a Court of Justice ought to be very slow in countenancing any attempt by a wrong-doer to make captious objections to the methods by which those whom he has injured have sought to repair the injury. When a road is let down or land let down, those entitled to have it repaired find themselves saddled with a business which they did not seek, and for which they are not to blame. Errors of judgment may be committed in this as in other affairs of life. It would be intolerable if persons so situated could be called to account by the wrong-doer in a minute scrutiny of the expense, as though they were his agents, for any mistake or miscalculation, provided they act honestly and reasonably. In judging whether they have acted reasonably, I think a Court should be very indulgent and always bear in mind who was to blame.”

I further keep in mind that Borealis had no motive for not getting on with restoring the cavern into service; tellingly, the events in question took place in 2004; any recovery from Geogas will not be received until 2010. I think on this Issue, Mr. Lumley’s opinion captures the essence of the matter: the Borealis reaction was slow; it could have been quicker but it was not unreasonable. That is an end of the Geogas case of a failure on the part of Borealis to mitigate its losses in connection with the cavern.

Although the parties undertook a considerable amount of work to narrow the disputes as to Quantum, a number of issues remain. These are addressed here.

(1) Supplier repair costs: Borealis claims €1,332,451 in this regard.

Mr. Ashcroft invited me to deal with this head of claim on a “rough and ready” basis and argued for a 30% deduction or such other figure as the Court thought appropriate. First, Borealis had been over-inclusive in its allocation of costs to these repairs; there was a lack of underlying documentation to support this claim. Moreover, there was reason to be sceptical, having regard to the very large claim advanced by Borealis for lost profits in November 2003 and subsequently abandoned. Secondly, he relied on Mr. Lumley’s evidence for the proposition that Borealis had (unreasonably) failed to negotiate a 5-10% discount.

I think that there is much practical good sense in Mr. Ashcroft’s call for a rough and ready approach to be adopted. Whether it entitles Geogas to anything like the deduction for which he contended, may be another matter.

To begin with, I am wholly unable to accept Mr. Lumley’s evidence that Borealis had unreasonably failed to negotiate a 5-10% discount. Mr. Lumley had said that Borealis was not to be given a blank cheque. It should have nibbled at the prices quoted by suppliers. From his experience, a discount of the order of 5-10% should have been negotiated. To my mind, Mr. Lumley’s evidence simply failed to grapple with the following considerations:

I turn to the question of allocation. In general, I accept the accuracy of the Borealis accounting system, under which costs were allocated to particular work orders. To my mind, the detailed material contained in the Scott Schedule, in its final form and as attached to the Borealis closing submissions (“the Scott Schedule”), was of considerable assistance in this regard. Accordingly, I can see no basis for any significant discount. That said, I do have regard to the abandoned claim for November 2003 profits. I also have in mind that in the course of her closing submissions, Ms. Buehrlen very properly conceded that she could not support some €55,000 of the claim. In works of this scale and nature it is very difficult to guard against any inaccuracies creeping in. I see no unfairness to Borealis if I make some limited allowance in this regard. Adopting Mr. Ashcroft’s rough and ready approach, I think that an award of €1,250,000 under this head of claim does justice to both parties.

(2) Personnel costs: Borealis claims €135,748 under this head of claim. It relates essentially to amounts paid to its own employees and managers in respect of the inspection of and repairs to the heat exchangers. This claim includes an amount of basic rate time where wages would have had to be paid to employees in any event – but as they were diverted from their ordinary duties, additional costs were incurred by way of payments to sub-contractors who fulfilled their duties. The claim has been restricted to the employees’ basic rate time given the difficulties of quantifying the increased costs relating to sub-contractors – and given that sub-contractors’ time is more costly than the time of the employees in question. The sum claimed here also includes a round sum in respect of management time.

As appears from the Scott Schedule, Geogas disputes this amount on the following grounds:

Developing these grounds orally, Mr. Ashcroft emphasised two matters in particular. First, he submitted that a significant part of the claim was for unproven sub-contractors’ costs, “dressed up as a claim for personnel costs”. Secondly, he contended that as the plant was focussed on the planned October turnaround, no inference could be drawn that if management and staff had not had to spend more time dealing with the heat exchanger damage, they would have been engaged on revenue-earning activities.

It was common ground that the relevant principles appear from the following passage in Aerospace Publishing v Thames Water Utilities [2007] EWCA Civ 3; [2007] Bus LR 726, per Wilson LJ at [86]:

“ I consider that the authorities establish the following propositions. (a) The fact and, if so, the extent of the diversion of staff time have to be properly established and, if in that regard evidence which it would have been reasonable for the claimant to adduce is not adduced, he is at risk of a finding that they have not been established. (b) The claimant also has to establish that the diversion caused significant disruption to its business. (c) Even though it may well be that strictly the claim should be cast in terms of a loss of revenue attributable to the diversion of staff time, nevertheless in the ordinary case, and unless the defendant can establish the contrary, it is reasonable for the court to infer from the disruption that, had their time not been thus diverted, staff would have applied to activities which would, directly or indirectly, have generated revenue for the claimant in an amount at least equal to the costs of employing them during that time.”

See too, Al Rawas v Pegasus Energy [2008] EWHC 617 (QB); [2009] 1 All ER 346, esp. at [22] – [23]; and 4 Eng Ltd v Harper [2008] EWHC 915 (Ch); [2009] Ch 91, esp. at [40].

Insofar as Geogas denies this claim in its entirely, it seems to me that its approach is, with respect, unrealistic. I am satisfied on the totality of the evidence helpfully summarised in both the Borealis opening and closing written submissions (and which need not be set out here), that there was significant and major business disruption flowing from the Geogas breach of contract. Such disruption inescapably consumed staff and management time which was thus lost. I think that any Geogas insistence on further detail risks a disproportionate approach to proof of damage. As Jack J observed in Al Rawas v Pegasus Energy (supra), at [23], in this area, “the application of common sense” may fill a gap in the evidence. I am also unable to accept Mr. Ashcroft’s submission that the inference, referred to by Wilson LJ under (c) in the passage of his judgment in Aerospace (supra) cited above, is negated because of the focus on the turnaround; that is to take altogether too narrow a view of revenue generating activities.

However, it is also the case that there is a good deal of the broad brush in this Borealis claim – as Borealis in effect itself avers. There are difficulties of proof relating to the correlation of the expense on sub-contractors and the additional work done by staff. There must be recognition of the fact that wages would have been incurred in any event. As Borealis itself acknowledges, the figure for management time is a round figure. Doing the best I can on the evidence, I am satisfied that Borealis suffered a loss in this regard of at least €75,000 and it is that sum I propose to award under this head.

(3) Own equipment used to effect repairs: Here, Borealis claims €18,630. The basis of the claim lies in Borealis’s use of its own equipment for 12 hours a day on 14 days to repair the heat exchangers. Borealis contends that this use gives rise to a recoverable loss, quantified as the same amount that an outside contractor would have charged by way of a daily rate for the equipment in question.

Geogas resists this claim, submitting that Borealis has not proved that costs were incurred or loss suffered. In any event - and though the amount of the claim is agreed as a reasonable figure for the hire of such equipment - the claim could only lie (if at all) for the cost of operating the equipment not the (notional) hire cost.

Both parties relied on the decision in The West Wales [1932] P 165. That case concerned the battleship HMS Nelson; she had been damaged in a collision and was put into a naval drydock for repairs. In assessing damages, the Registrar allowed nothing in respect of the use of the dock and cranes, on the ground that the Admiralty had suffered no pecuniary expenses or loss. Bateson J remitted the case to the Registrar for reassessment, holding that the ship had occupied the dock and used the cranes, to the exclusion of other vessels and some allowance was therefore to be made to the Admiralty in respect of these items. Bateson J said this, at p.168:

“ The learned registrar has allowed nothing for either of these items. I think it is clear that something must be allowed. What the figure may be is a matter for him to decide, but I do not think it can be right to say that the Admiralty have suffered no loss by giving up their dry dock and by using their cranes in doing these repairs. The Nelson occupied the dock to the exclusion of other ships, and made use of the cranes and so on for the purpose of docking and cleaning and coating the bottom, which would be necessary in connection with these repairs. Docks and cranes cannot be used without expense, and there must be some damage to the Admiralty from these matters. The learned registrar says that in respect of these there was not a pecuniary expenditure or loss and, therefore, these items are not allowable. I think that is wrong. I think there was pecuniary expenditure and loss. It may not be actual money paid out to any particular person, but these expensive appliances cannot be utilised without some loss.”

Mr. Ashcroft submitted that there was no evidential basis for a finding here that the use of the Borealis equipment could not take place without pecuniary expenditure and loss. I disagree. The logic of Bateson J’s reasoning is, with respect, inherently attractive, practical and, to my mind, applicable here – at least in the absence of evidence or a compelling inference to the contrary. The use of the equipment carries a cost. However, Mr. Ashcroft is plainly right in submitting that Borealis cannot recover for the daily rate it would have paid had it hired the equipment; that is simply not a loss it suffered. Borealis is accordingly restricted to the operating cost of the equipment in question; that cost cannot be equated with the cost of hiring the equipment – it must necessarily be significantly less. Doing the best I can, I assess the operating cost of the equipment as €6,000.

(4) Warehousing cost of corroded titanium tubes: €19,066. I confess to some regret that the parties did not find it possible to reach agreement on an item such as this. I take it summarily. Borealis was entitled to store these tubes while the litigation was pending; it is anything but fanciful to suppose that a point may have been taken by Geogas had Borealis not done so. I think this (external) cost is allowable. Geogas submitted that a deduction should be made for the scrap value of the tubes. There does come a time when common sense must prevail. I decline to investigate or to require the parties to investigate the scrap value of these tubes. I allow the claim in the full amount of €19,066 on terms that, on payment of this claim, Geogas shall be at liberty, if it so desires, to take delivery of the tubes and to dispose of them as it sees fit. In the order, a fallback provision should be made for Borealis to dispose of the tubes within a given period of time, should Geogas choose not to take delivery of the tubes.

(5) Spare parts: €19,420. This is a claim for consumables such as breathing masks, batteries, gloves, goggles, packing materials and so on. Geogas resists this claim on the basis that the cost of these items is not referable to its breach of contract. I adopt the same approach as I did in connection with the allocation question more generally. I think there is sufficient evidence of a linkage to justify the claim generally but that some allowance should be made for inevitable errors in allocation. I allow this claim but only in the amount of €15,000.

(6) Loss of profit – September 2004: The hard-fought and already much discussed issue which arises under this heading concerns a relatively small sum of money, €72,202, agreed as a figure, concerning loss of profit incurred in September 2004 when reinstalling heat exchangers E and G. As will be recollected, these heat exchangers were disconnected in March 2004 as a result of the incident. The issue arises because, over the same period in September 2004, Borealis effected repairs to other equipment at the plant.

Geogas contends that no loss of profit is recoverable in respect of this period, as, given the other works done to the acetylene reactor (unconnected with Geogas’s breach of contract), there would have been a similar loss of production – regardless of the reinstallation of the heat exchangers. Borealis contends that, as a matter of law, it is entitled to the entire period of lost production; that it did other works is neither here nor there. However, as a pragmatic matter, it is content to attribute the lost production over the 3 day period in question equally between the heat exchangers and the acetylene reactor.

In cross-examination, Mr. Eerola explained the position as to the various works as follows:

“ ….we had two parallel maintenance works going on at the site….both would have resulted …[in]…a production loss alone but since they were carried out in parallel the production loss that resulted was clearly less than in the case they would have been done at a different time individually.”

As it seems to me, on such evidence as there was: (1) Borealis needed to reinstall the exchangers at some time – and whenever it did so, production was likely to be lost; (2) the other works undertaken in September 2004 were conveniently undertaken then – but it was not necessary for Borealis to undertake those other works at that particular time.

These circumstances are such that guidance may usefully be obtained from a number of Admiralty (and commercial) cases, most simply summarised as follows. Let it be assumed that a ship is damaged in a collision. She is not immediately rendered unseaworthy but the repairs need to be undertaken at some convenient time. At the time when the collision repairs are undertaken, the owner takes the opportunity of undertaking other works, not immediately necessary. As I understand the authorities, the tortfeasor will be liable for the entire period of detention and the owner will not be obliged to give credit for the time required by the unrelated repairs. The position would be different if it was necessary to undertake the unrelated repairs at that time. See: the very helpful summary in McGregor on Damages (18^th ed.), at para. 32-037, together with the authorities there cited; see too, The Oinoussian Friendship [1987] 1 Lloyd’s Rep. 258, esp. at pp. 263-265.

In my judgment, the key to this issue is that whenever the heat exchangers were reinstalled there would, in all probability, be a loss of production. On the face of it, therefore, Borealis ought to be entitled to recover for the loss suffered when the reinstallation took place. No doubt if it had been necessary to do other works at the same time, then Geogas would have had a good answer to any such claim. But that was not this case; the evidence does not show that the other works Borealis undertook in September 2004 had to be done then. On that footing, Borealis is in principle entitled to recover for the lost production in September 2004. As, however, pragmatically (and attractively) Borealis has indicated that it is content to claim for only half the time in question, I shall allow the claim to such extent only, namely in the amount of €36,101.

I shall be grateful for the assistance of counsel in drawing up an appropriate order and as to all questions of costs.